Endpoint characterisation modelling for marine

cleansing agents, nanomaterials; J. Schoenfeld, Federal Environment ... Although standardised test and evaluation methods are the basis for comparable...

0 downloads 41 Views 2MB Size
Downloaded from orbit.dtu.dk on: Feb 09, 2020

Endpoint characterisation modelling for marine eutrophication in LCIA

Cosme, Nuno Miguel Dias; Larsen, Henrik Fred; Hauschild, Michael Zwicky Published in: Abstract book - SETAC Europe 23rd Annual Meeting

Publication date: 2013 Document Version Publisher's PDF, also known as Version of record Link back to DTU Orbit

Citation (APA): Cosme, N. M. D., Larsen, H. F., & Hauschild, M. Z. (2013). Endpoint characterisation modelling for marine eutrophication in LCIA. In Abstract book - SETAC Europe 23rd Annual Meeting (pp. 54-55)

General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.  Users may download and print one copy of any publication from the public portal for the purpose of private study or research.  You may not further distribute the material or use it for any profit-making activity or commercial gain  You may freely distribute the URL identifying the publication in the public portal If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

1 Tailored risk assessment of pharmaceuticals – a regulatory point of view J. Bachmann, German Environment Agency UBA; A. Hein, GKSS Research Centre Geesthacht / Pharmaceuticals, washing and cleansing agents, nanomaterials; J. Schoenfeld, Federal Environment Agency. The European authorization of human and veterinary pharmaceutical products requires the assessment of possible environmental risks before market authorisation. The procedure to assess the environmental risks is defined by several European guidelines and “question & answer” documents by the European Medicines Agency (EMA). Although standardised test and evaluation methods are the basis for comparable assessment of active pharmaceutical ingredients, there is the impression that in regulatory context well established but antiquated methods and approaches are used. Established and validated methods are the basic prerequisite for planning and legal reliability and therefore not every scientific approach can be included in regulatory risk assessment immediately. Providing that knowledge about specific mechanisms of action of the active ingredients of pharmaceuticals is available the standard guidelines allow already a tailored approach. From an agency point of view, candidate groups suitable for tailored approaches are e.g. steroidal hormones resp. their analogs, progestins, anti cancer drugs with anti-androgenic mechanism of action or parasiticides for pasture animal treatments. These are examples of substance groups with specific mechanisms of action. Modified fish life stage tests or fish sexual development tests can be used as adequate methods to evaluate the relevant effects caused by the test substances in the aquatic environment. So far these tailored approaches are a “case-bycase” decision The German Federal Environment Agency develops a decision strategy for tailored environmental risk assessment for a more consistent evaluation and a reliable assessment of potential environmental risks. However, as this is not the standard procedure, applicants are asked to contact the respective authority before the formal procedure of application is started to prevent unnecessary time conflicts and potential refusing of a market authorization. The presentation will deal with the mentioned area of conflict from a regulatory point of view. Examples of tailored approaches and a possible decision strategy will be presented. 2 Spatial distribution of pharmaceutical residues after environmental discharge: Evaluation of occurence and degradation in the Gironde estuary, France y. aminot, UMR CNRS 5805 EPOCLPTC; P. Pardon, H. Budzinski, University of Bordeaux / UMR CNRS 5805 EPOC-LPTC. In regards to the growing concern on pharmaceuticals in water bodies, assessing the impact of urban discharge on the environment is the key point. Urban effluents are actually considered as the main contributors to the presence of pharmaceuticals in the water compartment in such a way that full scale studies have to characterize both urban effluent and the receiving environment. Once discharged, pharmaceuticals can undergo sorption to suspended matter, biotic or abiotic degradation. These processes can modify bioavailability or toxicity by generating transformation products. The objectives of this work are both to characterize a urban effluent and its receiving environment and to investigate the fate of selected pharmaceuticals after discharge. In this context, the Garonne, a 647 km river located in the South-West of France drains a 56 000 km² catchment area and is impacted in its estuarian part by the effluents of Bordeaux city, the sixth most populated urban area in France. After treatment in the two main Waste Water Treatment Plants (WWTP) of Bordeaux city, wastewaters of more than 700 000 population-equivalent reach the Garonne with an average 320 dilution rate (2011 data). The ability of the Garonne riverwater - and especially its Maximum Turbidity Zone (MTZ) - to enhance or inhibit micropollutant degradation has to be investigated for a complete impact assessment. 53 pharmaceuticals including some metabolites were selected for this study. In order to qualify the variability of the pharmaceuticals intakes, a preliminary characterization of the Bordeaux main WWTP effluents has been carried out. The range of degradability in the WWTP was evaluated through the removal rates. In addition, the absence of seasonality was highlighted on the target compounds. Environmental monitoring has then been performed on 6 sites along the estuarian

SETAC Europe 23rd Annual Meeting

Garonne river for one year. The fate and persistance in the receiving water body was evaluated through three different methods: a field load calculation, longitudinal attenuation on a low-flow pilot river and on batch experiments at the lab scale. The results obtained in this study clearly demonstrate the different behaviours of the compounds from the most persistant to the most labile ones, likely to generate transformation products. 3 Bioconcentration and metabolization of pharmaceuticals in marine mussel C. Boillot, Université Montpellier 1; M. Bueno, University Montpellier 1; M. Bachelot, H. Fenet, Université Montpellier 1; D. Munaron, IFREMER; C. Casellas; E. Gomez, Université de MontpellierCNRS Hydrosciences Montpellier UMR / Sc. Environnement & Sante Publique. The availability of organic contaminants to biota in aquatic environment is assessed firstly through bioaccumulation. Pharmaceuticals accumulate in freshwater organisms as it has been shown for some of them, mainly in fish species, but also in other model species such as invertebrates and algae. Despite those factors are species-dependent, very few is known in nontarget organisms. In marine organisms, pharmaceuticals have been detected -1 with concentration ranging from ng g d.w. to some thousand ng g-1 d.w. Other hand, it is generally accepted that substances with octanolwater partition coefficient (log Kow) equal to or higher than 3 have the potential to bioaccumulate in biological tissues; but other factors such as metabolism and the uptake and depuration kinetics must also be taken into consideration. However, studies on pharmaceuticals bioaccumulation in nontarget, marine organisms, determining kinetics for uptake and depuration and taking into consideration metabolites are very scarce. The aim of this work was the determination of bioconcentration, the kinetic constants of uptake and depuration of selected pharmaceutical in Mytilus galloprovincialis. The selected molecules were two benzodiazepines (BZD): diazepam (DZP) and tetrazepam (TZP), and the antiepileptic drug carbamazepine (CBZ). Furthermore, two major metabolites of CBZ were investigated in mussels tissues.Two extraction and analytical methods were developed for determine the concentration of selected pharmaceutical and the two major metabolites of CBZ at low concentration in mussel tissues. The first one for BZD, consisted of a microwave-assisted extraction and an analysis on a GCMS. The second, for CBZ and the metabolites, was a pressurized liquid extraction and a LCMSMS analysis. The calulated BCFs of the selected pharmaceuticals were 2.2, 50.7 and 99.3 L kg-1 d.w for respectively CBZ, DZP and TZP and increased with logKow (2.3, 2.8 and 3.2, respectively). This relation was not linear suggesting involvement of other mecanisms than lipohily in bioconcentration. DZP was detected as TZP metabolite. Furthermore, metabolites of CBZ were -1 below the detectable concentrations of 0.7 ng g d.w. in mussels. According to our results, previous described effects of CBZ in mussel should arrive without major CBZ bioconcentration.This late result deserves further research as mussels metabolism should conduct to other more relevant metabolites than those, humans, investigated in this study. 4 Ecological risk assessment of veterinary medicines applied in Asian aquaculture A. Rico, Wageningen University / Aquatic Ecology and Water Quality Management Group; P.J. Van den Brink, Wageningen University and Research Centre. During the last decades, the Asian aquaculture sector has experienced an unprecedented increase and nowadays accounts for about the 90% of the global aquaculture production by volume. Veterinary medicines (i.e., antimicrobials, fungicides, anthelmintics and other parasiticides) have been occasionally used in Asian aquaculture to treat and prevent disease outbreaks and as growth promoters, posing a potential risk for surrounding aquatic ecosystems. Although several studies have qualitatively described the potential environmental consequences of the use of veterinary pharmaceuticals and other chemicals in Asian aquaculture, none of them has been able to quantitatively assess the environmental exposure and their potential ecological risks. The main objectives of the present study were: i) to quantitatively assess the current use and risks of veterinary medicines for several aquaculture scenarios of Asia, ii) to prioritize compounds and scenarios that should be targeted in monitoring

1

programs, and iii) to identify research needs for performing refined risk assessment studies. In this study, we analysed recently developed databases on veterinary medicines use and aquaculture management practices. These databases were built through interviews performed between 2011 and 2012 in more than 250 and 1600 grow-out farms (for chemical use and management practices, respectively) from Bangladesh, Thailand, Vietnam and China. The collected data was used to build risk assessment scenarios for several aquaculture species (i.e., tilapia, Penaeid shrimps, Pangasius catfish and Macrobrachium prawn). Risk assessments were performed by using the ERA-AQUA model (www.eraaqua.wur.nl), a newly developed modelling tool designed to evaluate the environmental exposure and risks of veterinary medicines applied in pond aquaculture. The model output was evaluated regarding the percentage of the applied dose that is released through effluent discharges, the concentration dynamics of veterinary medicines in the aquaculture effluents, and the acute and chronic risk quotients for primary producers, invertebrates and fish in surrounding aquatic ecosystems. Preliminary results of this study suggest that the Pangasius catfish farms of Vietnam followed by the shrimp farms of China constitute possible hot-spots for environmental pollution due to the intensity of the aquaculture production and considerable discharge of chemical residues into surrounding aquatic ecosystems. 5 Reducing uncertainty for PECs in coastal zone: case study on carbamazepine, oxcarbazepine and their metabolitesF.H. Hélène , UMR Hydrosciences / UFR des Sciences Pharmaceutiques; A. Lauren, UMR Hydrosciences, Université Montpellier I; V. Alice, m. dominique, F. Annie, IFREMER; M. Olivier, CHU Lapeyronie / Laboratoire de Pharmacologie Médicale et Toxicologie; B. Helene, Université Bordeaux 1-UMR EPOC; c. claude, G. elena, UMR Hydrosciences, Université Montpellier I. To assess pharmaceutical contamination in surface waters at a national and regional scale, TM different geo-referenced watershed models PhATE and GREAT-ER were used. These models incorporate spatial and temporal characteristics of the receiving environment allowing more accurate and realistic PECs than those calculated according to EMEA guidelines methodology. In coastal zone, models including the diffusion are scarce; however, this interface between the land and the sea should also been affected by drugs contamination. Thus, the aim of the present project is to propose a model to assess pharmaceuticals concentrations in a coastal zone receiving effluents from a WTP through a submarine outfall. For this purpose, pharmaceuticals, carbamazepine (Cbz), oxcarbazepine (OxCz) and their main metabolites were selected. In effluents, PECs calculation took into account in addition to the common wastewater fate, the initial Cbz and OxCz prescriptions, their metabolisms and their excretion rates. The obtained PECs results were similar to the measured concentrations (MECs) in the wastewater effluent. In the coastal zone, PECs were calculated taking into account the diffusion in the coastal zone through the submarine outfall using the hydrodynamic numeric model MARS 3D. PEC values in coastal zone were in agreement with MECs, measured using the passive samplers polar organic chemical integrative sampler implemented around the submarine outfall. The combined approach considering both detailed local consumption of parent compounds and their human metabolisms, in association with the diffusion estimation in coastal water allowed a reliable estimation of the pharmaceutical PEC for parent compounds and metabolites. MARS 3D model allows a refined assessment of concentrations in seawater influenced by specific mechanisms occurring. It can be further applied to others molecules, including elimination half-lives. 6 Review of a five year study on pharmaceuticals in the Irish aquatic environment B. Quinn, Irish Centre for Environmental Toxicology ICET / Galway-Mayo Institute of Technology; W. Schmidt, Irish Centre for Environmental Toxicology; D. Sheehan, University College Cork; G. McEneff, Dublin City University; L. Barron, King's College London; B. Kelleher, Dublin City University; B. Paull, University of Tasmania. The presence of pharmacutical compounds in the aquatic environment is a research area of growing concern, particularly with the current review of annex X of the WFD and the

SETAC Europe 23rd Annual Meeting

identification of new potential priority substances including 17?estradiol (E2), 17?-ethinylestradiol (EE2) and diclofenac. In 2008 the Irish Environmental Protection Agency funded a five year study into the assessment and potential human impact of exposure to novel environmental contaminants (pharmaceuticals) on marine and freshwater bivalves. This multidisiplinary project was divided into the two distinct but inter-related areas involving (a) chemical analysis of various pharmaceutical compounds in municipal effluents and receiving environments and their ability to bioconcentrate / bioaccumulate; (b) biological analysis of the potential effects (both acute and sub-lethal) of these novel pollutants on animals exposed in the aquatic environment and development of new endpoints to measure these effects. The five pharmaceuticals investigated (diclofenac, mefanamic acid, gemfibrozil, trimethoprim and carbamazepine) were all detected in high ng/L low µg/L range in both municipal effluents and the marine receiving environment. Both Mytilus spp and D. polymorpha showed expression of biomarkers of stress (glutathione transferase) and damage (lipid peroxidation, DNA damage) at environmentally relevant concentrations (1 µg/L) of gemfibrozil & diclofenac. Proteomic analysis revealed a significant impact on proteins invovled in energy metabolism, oxidative stress response, protein folding and immune responses. An oxidative stress effect was also confirmed by the biomarker response. Steroid levels were measurable in mussels using human immunoassay technology and were significantly effected following exposed to EE2. Human diagnostic techniques showed a significant impact on liver and kidney function from Mytilus and fish exposed to diclofenac. From this work it is apparent that pharmaceuticals are entering the Irish aquatic environment from municipal effluent and have the potential to negatively impact on aquatic species. 7 Population and seasonal variations in the zebra mussel transcriptome: Physiological changes and environmental inputs B. Pina, IDAEACSIC / Environmental Chemistry; B. Campos, IDAEACSIC / Environmental Toxicology; M. Faria, Unicersidad de Aveiro / Dept. Biologia; M. Casado, Institute of Environmental Assessment and Water Research (IDAEA-CSIC); M. Blazquez, CSIC; A. Torreblanca, University of Valencia / Department of Animal Biology; M. Fernandez-San Juan, CSIC; S. Lacorte, IDAEACSIC / Environmental Chemistry; A. Navarro, CSIC; C. Barata, CSIC / Environmetal Chemistry. The zebra mussel Dreissena polymorpha is a Caspian Sea bivalve that colonized freshwater bodies worldwide during the XX century. To analyze the impact of seasonal and environmental variations on the physiology and metabolism of this invasive species, we developed a custom microarray using 4057 publicly available DNA sequences from Dreissena and other related genera. Seasonal variations of transcriptome profiles were analyzed using half-body samples from a relatively clean site (Riba-Roja, low Ebro River, N.E. Spain), at three different stages of the annual cycle: Pre-spawning (February), spawning (June), and gonad resorption (September). Transcripts from a total of 745 unique sequences showed significant changes among these three groups of samples. Functional characterization of these transcripts based on their closest known homologues showed that genes involved in stress defense (oxidative and infection) were overrepresented in September, whereas genes related to reproductive functions were overrepresented in the spawning and pre-spawning periods. Transcriptomic analyses of gill samples from six populations in the low Ebro showed a clear differentiation between upstream and downstream populations. These differences correlated positively with heavy metals (Cu, Cd, Mn, Ni, Zn) loads, and negatively with indexes of general fitness, like lipid content or tissue condition index. No physiological or reproductive effects were linked to mercury and organochlorine compounds, present at the most downstream Ebro sites due to the activity of a chlor-alkali factory operating in Flix. We concluded that heavy metals were the main stressors for these zebra mussel populations and that this species is resilient to other pollutants proved to be extremely toxic to vertebrates. Results also show that zebra mussel populations having very low genetic differentiation may acclimate to different levels and forms of pollution through modulations in their transcriptomic profile. Our studies may help to identify developmental stages and environmental conditions at

2

which the organism is more vulnerable for future control strategies. 8 Comparative genomics in zinc toxicity: a multi species approach to determine common modes of action T. De Boer, Vrije Universiteit / Ecological Sciences; B. Nota, VU University medical center Amsterdam; M. Castro Ferreira, R. Vooijs, H. Welle, J. Legler, N. van Straalen, VU University Amsterdam; M.J. Amorim, Universidade de Aveiro / Department of Biology and CESAM; D. Roelofs, Vrije University Amsterdam. Heavy metal contamination of soil may pose health and hazard risks to ecosystems and human health alike and due to anthropogenic activity these metals have been contaminating the environment since Roman times. We are interested in the effect that zinc, which is both a contaminating heavy metal as well as an essential element, has on gene expression patterns in ecologically important animals. Here we exposed two relevant ecotoxicological test species (Folsomia candida and Enchytraeus crypticus) to two concentrations of zinc (EC10 and EC50 on reproduction) for four different exposure times (1, 2, 3 and 4 days) in order to investigate the similarities as well as the differences in transcriptomic responses to this heavy metal (comparative genomics). For the single exposure Enchytraeus showed, in general, a stronger response in significantly differentially regulated genes at longer exposure times except for the 4 day exposure which showed a smaller response in differentially regulated genes as compared to the 3 day exposure. A possible explanation for this might that at the 4th day these animals have fully activated their metal excretion system and are able to regulate their internal zinc concentration better. Although the comparative analysis is still work in progress we feel that comparative genomics as a technique might shed light on common responses in multiple species in order to develop sets of genes that may act as biomarkers and fit into pathways that can be used to develop Adverse Outcome Pathways (AOP). These AOP’s make it then possible to integrate transcriptomics into environmental risk assessment 9 Tracking of novel potential biomarkers in Porcellionides pruinosus (Isopoda) exposed to nickel: a transcriptomic approach N.G. Ferreira, CESAM Universidade de Aveiro / Departamento De Biologia & CESAM; R. Morgado, University of Aveiro / Department of Biology and CESAM; M. Novo, L. Cunha, Cardiff University; A.M. Soares, Universidade de Aveiro / Department of Biology and CESAM Centre for Environmental and Marine Studies; P. Kille, Cardiff University; S. Loureiro, Universidade de Aveiro / Biology. Isopods represent keystone sentinel detritivorous, widely exploited for ecotoxicological assessment. This study describes development and deployment of an ecotoxicogenomic tool kit that significantly enhances the utility of Porcellionides pruinosus as a toxicological model and underpins mechanistic interpretation. It describes the assembly and comprehensive annotation of the transcriptome coupled with global transcript analysis of a dose course of nickel exposure. We further dissect this temporal response to nickel exposure within critical functional pathways using QPCR on specific targeted genes. P. pruinosus has proved highly tolerant to metal contaminated environments that may reflect adaptions at various levels of biological organization. Moreover, there is a lack of functional genetic information for isopods, with the majority of data representing phylogenetic mtDNA markers. It was for these two reasons that this species was chosen as a candidate for our current analyses. Furthermore, there is little information as to the mechanistic toxicology underlying the isopods response to nickel exposure. In order to establish a transcriptome representing a mixture of different life-stages from genotyped individuals representing juveniles, adults and pregnant females was pooled out. The resulting material was used to generate 17.5Million 50bp paired-end sequences using an Illumina Hi-Seq platform. An assembly was performed using various pipelines and a consensus transcriptome was generated representing more than 35K contigs and then annotated by homology to sequences represented in Genbank. Genoptyped adults were then exposed for a period of 96h to nickel, in the concentrations of 50mg/kg soil (maximum allowable limit within the Canadian EQG) and 250 mg/kg soil, a realistic concentration found

SETAC Europe 23rd Annual Meeting

within the natural soils. Quality validated samples of total whole body RNA were prepared from each exposure. RNAseq was then performed on a pool of three independent individuals from each dose together with controls More than 15M reads for each sample were mapped to the consensus transcriptome and the Reads Per Killo-base Million was calculated and used to profiles differential gene expression. The study is the first full body descriptive transcriptome for isopods, and shows the global transcriptomics that these organisms undergoes when exposed to nickel. Besides these two important features it also presents a target pathway based analysis. 10 Looking for biomarkers of metal exposure in complex environment by transcriptome analysis in the aquatic plant Elodea nuttallii C. Cosio, Geneva University. To gain insight into the impact of metal contamination on ecosystems, the potential of various tests on metal accumulating organisms is explored. Previous studies suggest that macrophytes might participate in bioaccumulation and biomagnification of toxic metals in aquatic environment. Rooted macrophytes are well-suited to be used in ecotoxicological tests because they are exposed to both water and sediments. Amongst several macrophytes collected in an Hg contaminated reservoir in Romania, Elodea nuttallii showed a high organic and inorganic Hg accumulation and was then further studied in the laboratory. E. nuttallii is a rooted macrophyte native to Northern America, but as a neophyte, has spread throughout the world. It is able to accumulate large amounts of metals and is therefore a good candidate for the aim of this study. Recently developed genomics tools have a promising potential to identify early biomarkers of exposure to multiple toxicants. In the present work we used transcriptome analysis (RNA-seq) of Elodea nuttallii to identify biomarkers of metal exposure. RNA-seq allowed identification of genes affected by Hg and Cd exposure and also unraveled plant response to the toxic metal: a change in energy/reserve metabolism caused by the inhibition of photosynthesis, and an adaptation of homeostasis networks to control accumulation of Hg. Data were validated by RT-qPCR and selected genes were further tested as biomarkers. Samples exposed in the field and to natural contaminated sediments clustered well with samples exposed to low metal concentrations under laboratory conditions. Moreover, genes responded to low realistic concentrations in a dose-dependent manner. Our data suggest that this plant and/or this approach could be useful to develop new tests for water and sediment quality assessment. 11 Proteogenomics, a straightforward approach for protein discovery in unsequenced test organisms in ecotoxicology J. Trapp, Irstea / Laboratoire d’écotoxicologie; O. Geffard, Irstea; G. Imbert, CEA / Laboratoire de détection et de caractérisation des agents du risque environnemental; J. Gaillard, A. Davin, CEA / Laboratoire de Biochimie des Systemes Perturbés; a. chaumot, Irstea; J. Armengaud, CEA / Laboratoire de Biochimie des Systemes Perturbés. In ecotoxicology, proteomics allows new insights into the molecular mechanisms related to stress response. The interpretation of tandem mass spectra recorded for thousands of peptides generated from the whole proteome proteolysis requires a comprehensive protein sequence database. Unfortunately, for most species used in ecotoxicology, such informations are missing. Thus, many studies experienced difficulties in terms of protein identification. Here, we proposed an approach for quickly identifying specific proteins of the freshwater crustacean Gammarus fossarum, an ecotoxicology relevant species, by the alliance of genomic and proteomic (i.e. proteogenomic). As for animals only a small portion of the genome encodes proteins, we used next-generation sequencing technologies to sequence mRNA. We caracterized the transcriptome of key physiological organs: the female and male reproductive system, the cephalon and the hepatopancreatic caecum. For improving structural annotation, this database was used to interpret protein data obtained by high-throughput proteomics on the four tissues. By Illumina sequencing method, we obtained a mRNA sequences dataset comprising 218,574 contigs, with a median read length of 799 bp, of which 127,332 were functionnally annoted. A total of 1,033,282 MS/MS spectra was recorded during our shotgun proteomic procedure

3

and assigned with the RNA-seq derived database, resulting in the identification of 48,633 peptide sequences and 1,624 proteins. Besides the identification of ubiquitous or key metabolic proteins, some well known as biomarkers, the functional annotation of numerous proteins resulted quite scare. Indeed, 202 proteins could be considered as orphans and 325 proteins as hypothetical proteins, with among them 122 closely related to protein sequences predicted from the Daphnia pulex genome annotation. Or, genes uniquely found in the Daphnia lineage have been shown to be amongst the most ecoresponsive. Thus, our dataset is of utmost importance for a comprehensive understanding of the molecular mechanisms related to stress response. The approach developed here for identifying specific proteins in test organism in ecotoxicology is proved to be straightforward. Also, in addition to being the first large scale identification of genes and proteins for G. fossarum, our results will help to provide useful insights into key physiological functions, especially for neuroendocrine regulation after an in-depth ecophysiological characterization. 12 Identification of metabolic pathways in Daphnia magna exposed to SSRI’s using transcriptomic, immunocytochemistry and physiological responses. B. Campos, IDAEACSIC / Environmental Toxicology; N. Garcia-Reyero, Mississippi State University; L. Escalon, Enginner research and development center, US Army; T. Habib, BTS; H. Dircksen, University of Stockholm; S. Tsakovski, University of Sofia; C. Rivetti, B. Pina, R. Tauler, C. Barata, IDAEA-CSIC. Assessing the risks of long term exposure to low doses of pharmaceuticals is an unmet reasearch need, particularly for substances that may act as neuronal disruptors in non-target species. Selective Serotonin Reuptake Inhibitors (SSRIs) are used worlwide to treat depression, and act by blocking the reuptake of serotonin in the nerve synapses, thus increasing its concentration in the synaptic cleft and thererefore stimulating serotoninergic neurons. In this study we explore the hypothesis that SSRIs can affect D. magna juvenile developmental rates and offspring production by disrupting serotonin activity, known to regulate carbohydrate and oxidative metabolic pathways. To test this hipotesys, D. magna juveniles and adults were exposed to SSRIs (fluoxetine and fluvoxamine) and its transcriptome was analysed using microarrays, then validated by qPCR. Immunocitochemistry assays were also performed on control and exposed adults brains, using an antiserotonin antibody. Microarray analysis was performed using hierarchical clustering, PLS, GO and Kegg.This allowed to identify over 250 differentially expressed genes in SSRI treated individuals. Many of these genes belong to the metabolism of nucleotides, serotonin and lipids and to the Krebbs cycle, while biochemical and respirometry assays indicated that exposure to SSRI increased oxidative metabolism and the catabolism of carbohydrates. These findings are in line with reproduction and survival assays that showed how females exposed to SSRIs increase the number of offspring but have lower survival under anoxia. Immunocytochemistry assays showed for the first time the existence of serotoninergic neurons in the brain ganglia of D. magna and that serononin activity is increased in SSRI exposed females. These results support the hypothesis that SSRIs disrupte serotonin activity deregulating carbohydrate and aerobic oxidative metabolism allowing the animals to catalyse carbohydrate reserves more efficiently but at a cost of being more susceptible to anoxic conditions, which can have important consequences in nature. These findings indicated that the use of transcriptomic responses combined with biochemical, physiological and toxicological endpoints in D. magna have a great potential to elucidate novel mechanisms of action of emerging pollutants to aquatic biota. 13 A large-scale field radio-tracking study investigating unintentional acute effects of pesticides on Perdix perdix in agricultural landscape F. Millot, ONCFS; A. Decors, E. Bro, ONCFS (National Game & Hunting Institute). Detrimental effects of pesticides, on wildlife have been little investigated in intensive cereal farmlands at field and landscape scale. We conducted a large-scale radiotracking study to examine this topic on grey partridges, an emblematic bird of cereal ecosystems that is likely to be highly exposed to

SETAC Europe 23rd Annual Meeting

pesticides. During 2010 and 2011, more than 500 wild birds have been tagged on 12 sites (14000 ha) typical of intensive cereal farming landscape of North-Central France. The location of each bird was reported on a mobile GIS twice a day from early March to late August. Carcasses of dead birds were collected less than 12 hours after the death.When possible, a necropsy was carried out and organs were collected for toxicological analyses. At the same time we mapped all habitat features on GIS and made a farmer survey to list every intervention that they carried out on each of their field plots (ie pesticide used, date of application, dose used, etc.). Thus, crossing bird habitat use (ie radio-tracking data) and farming practices data allowed us to determine for every bird its exposure to pesticides in time and space. We focused toxicological analyses on pesticides for which the birds were exposed during the last ten days before their deaths. Such analyses were made for 98 dead birds. 282 analyses screened more than15 chemical families such as pyrethroides, triazoles and strobilurins (see table 1). In addition we compared the exposure of pesticides in space and time of dead vs. surviving birds. We will detail these results at the SETAC meeting and discuss the impact on grey partridge survival rate, through direct (ie lethal intoxication ) vs. indirect (ie sublethal intoxication that could make birds to be more sensitive to other mortality causes such as predation) effects. Chemical group Number of analysis % of cases higher than the detection limit pyrethroids 97 24% triazoles 87 0% Strobilurins 39 0% others 58 9% Table 1 : chemical groups searched, number of analysis and % of cases for which the we detected a concentration higher than the detection limit \n Our field dataset (farming practices, complex pesticide exposure patterns) could be used in a further step in mesocosm studies to test in controlled conditions the impact of various realistic exposure scenarios as well as in individualbased or life-cycle models 14 Trace elements in seabirds from French Southern Lands C.V. Zecchin Cipro, Université de La Rochelle / LIENSs; Y. Cherel, Centre d’Etudes Biologiques de Chizé, UPR 1934 du Centre National de la Recherche Scientifique; P. Bocher, F. Caurant, Université de La Rochelle / LIENSs; P. Miramand, Université de La Rochelle; P. Bustamante, Université de La Rochelleq / LIENSs. The French Southern Lands are breeding site for more than 40 seabird species, many of them long lived and situated at the top of the food web. These seabirds are therefore prone to bioaccumulation and biomagnification of contaminants, both organic and inorganic ones. Among the latter, cadmium (Cd) and mercury (Hg), are non-essential elements, potentially toxic. Marine life can be exposed to these elements from anthropogenic and natural sources, and exposure is governed by various factors, including foraging location and trophic position. Sample collection (n=134, 26 species, n?6 for 10 of them) occurred from October/1998 to March/2000 at Kerguelen, Amsterdam and Crozet Archipelagos, Southern Indian Ocean. All birds used in this study died accidentally either killed by station lights/skuas or as a result of bycatch in longline vessels in these areas. Liver, muscle and kidney tissues were analysed for Cd, Cu, Se, Hg and Zn. Results were majorly, but not exclusively, driven by the different diets of the species: zooplankton-eating birds showed, in a general way, lower Hg concentrations when compared to fish/squid-eating ones and a Se:Hg molar ratio always superior to 1. Hg is mainly due to myctophids in diet, because its biomethylation in the water column renders Hg bioavailable to mesopelagic organisms. Cd, by its turn, is more related to cephalopod consumption but also to limpets’. Nevertheless, zooplankton-feeders may also present elevated Cd levels as a result of preying on the pelagic amphipod Themisto gaudichaudii, which has exceptionally high Cd levels in the region. Correlation analyses showed significant inter and intra-tissue results, notably between Se and Hg (due to Hg detoxification process) and within these, liver ones emphasised its importance as long term storage organ. In an analogous manner, Cd and Zn coaccumulation is seen in the kidney. Renal Cd and hepatic Hg presented a stratification that stresses differences not only in ecological niches, but in detoxification mechanisms as well. The present work brings up detailed data that is unpublished for the area or absolutely scarce to many of the sampled species, most of them of high conservation concern. The work also

4

indirectly confirms ecological data, specially due to the results linked to feeding ecology and Hg detoxification due to moulting schemes, even though some questions about the subject remain, particularly the ones in regard to the non-breeding period of these birds. 15 Species Sensitivity Distribution (SSD) and SPEAR – Validation of predictions by empirical data with pesticide mixturesS. Jesenska, Masaryk University / Faculty of Science, RECETOX; L. Blaha, Masaryk University, Faculty of Science, RECETOX / Research Centre for Toxic Compounds in the Environment (RECETOX); M. Liess, UFZ Helmholtz Centre for Environmental Research / Dept. system ecotoxicology; R.B. Schaefer, M.A. Beketov, UFZ - Helmholtz Centre for Environmental Research. Species sensitivity distribution (SSD) is statistical method broadly used in the ecotoxicological risk assessment. Originally it has been used for prospective risk assessment of single substances but nowadays it is becoming more important also in the retrospective risk assessment of mixtures SSD. Validation of SSD is thus of high importance. SPEAR is bioindicator index, which is pesticide based on biological traits responsive to the effects of pesticides and postcontamination recovery. It was shown to be specific for pesticide stress in streams and rivers and its relevance and specificity has been proved by a number of studies. The aim of this study is to compare the outcomes of both methods, and to validate the use of SSD method in retrospective risk assessment of pesticide mixtures using empirical SPEAR data and concentrations of 25 pesticides from Germany, France and Finland. The thresholds used for the comparison were SPEAR = 45% (LOEC level) and msPAF = 0.05 (corresponds pesticides to HC5). First results shows statistically significant correlations (Pearson’s R, p< 0.01) between msPAF values and SPEAR . pesticides Comparisons of the thresholds established for the SSD and SPEAR approaches show that the SSD based on the acute toxicity data underestimate the actual effects. The use of chronic data, based on the frequently applied acute-to chronic ratio of 10, improved the agreement between the two methods. However, the SPEAR index is still pestcidies more sensitive and a higher acute-to chronic ratio would be required to match the thresholds from both methods. According to results, the present study suggests that field data should be used more frequently for validation of the toxicological predictive methods. 16 Pesticides reduce biodiversity – Losses in stream invertebrate taxa in Europe and Australia M.A. Beketov, UFZ - Helmholtz Centre for Environmental Research / Department of System Ecotoxicology; B.J. Kefford, University of Technology Sydney (UTS) / Centre for Environmental Sustainability, School of the Environment,; R.B. Schafer, University of Koblenz-Landau / Quantitative Landscape Ecology, Institute for Environmental Sciences; M. Liess, UFZ Center for Environmental Research / Department of System Ecotoxicology. The biodiversity crisis is one of the greatest challenges facing humanity, but our understanding of the roles of biodiversity loss drivers remains limited. Thus, despite the general concern and immense amount of studies, it remains unknown whether, to what degree, and at what concentrations modern agricultural pesticides cause regional-scale species losses. We analyzed the effects of pesticides on the taxa richness of stream invertebrates in Europe and Australia by applying rarefaction and richness estimators. Analyzing the sets with quantified pesticide concentrations, we show that pesticides produced statistically significant effects on both the regional species and family richness for both Europe and Australia, with losses in taxa up to 42% of the recorded taxonomic pools. Furthermore, the effects in Europe were detected at concentrations that current legislation considers environmentally protective. These results show that currently used pesticides do have pronounced effects on the regional biodiversity, and development of more protective standards and mitigation measures, as well as new largescale-oriented approaches in the ecological risk assessment of contaminants is urgently needed. For details see the upcoming publication: Beketov M.A., Kefford B.J., Schäfer R.B., Liess M. Pesticides reduce biodiversity – Losses in stream invertebrate taxa in Europe and Australia. Submitted.

SETAC Europe 23rd Annual Meeting

17 Multichemical pollutants’ exposure on Mediterranean (Iberian) rivers: occurrence, risk and effects on the aquatic ecosystemM. Kuzmanovic; A. Ginebreda, M. Petrovic, IDAEA-CSIC; I. Munoz, UB; s. sabater, ICRA; D. Barcelo, IIQABCSIC. The increasing contamination of aquatic systems with numerous man-made and natural chemicals is one of the key environmental problems. Although most of these chemicals are present at low concentrations, many of them may pose ecotoxicological concerns especially when occurring as components of chemical mixtures. In the EU there are more than 100000 registered chemicals listed by EINECS of which many are in daily use. One approach for identifying potentially dangerous compounds is screening of the environment for a large set of chemicals together with an assessment of the potential toxicity of the observed concentrations, which can be done by use of measured or predicted effect concentrations for standard test species. Moreover, combined biological and chemicalanalytical approaches provide an important progress in identification of those toxicants that are relevant for site-specific risk and towards an estimation of the portion of an effect that can be explained by analyzed chemicals. In this study we assessed risk for almost 200 compounds of possible concern and estimated their relationships with 7 relevant biological descriptors along 77 sites located in 4 Iberian river basins (Llobregat, Ebro, Jucar andGuadalquivir). Comprehensive data set was gained from the ongoing multidisciplinary SCARCE project. There were 3 objectives of this study. First, to identify pollutants which are relevant to each specific site in terms of their occurrence and hazard to the environment. Second, estimate the ecotoxicological risk associated to the simultaneous occurrence of many pollutants, considering both their sum and complexity and to include the statistical distribution of data as tool for interpretation of results. And third, to estimate the relationship between chemical and ecological status, correlating the ecotoxicological risk with some representative biological variables. 18 The influence of suspended particles on the effects of pulse exposure of pyrethroids on aquatic biota J.J. Rasmussen, National Environmental Research Institute / Department of Bioscience; B. Kronvang, Aarhus University; B. Strobel, H. Hansen, Copenhagen University. The use of synthetic pyrethroids has raised much concern in the preceding decades because of their high toxicity to especially nontarget freshwater macroinvertebrates. The sorption of strongly lipophilic insecticides, such as pyrethroids, to sediments has however been shown to reduce their bioavailability (and therefore toxicity) to sediment dwelling macroinvertebrates by up to several orders of magnitudes. The reduced ecotoxicity of particle associated pyrethroids to sediment dwelling macroinvertebrates has promoted a general belief among managers and stakeholders that the increased concentrations of suspended particles during storm water runoff probably mitigate some of the effects of pyrethroids in aquatic ecosystems. We studied the pulse exposure effects of lambda-cyhalothrin (LC) on locomotor behaviour and mortality of the freshwater amphipod Gammarus pulex. Locomotor behaviour was monitored during exposure and on each of 7 postexposure days, and mortality was registered prior to each video tracking sequence. The exposure scenarios were conducted in the presence of humic acid or montmorillonite or without microparticles. Three concentrations microparticles (5 - 125 mg/L) and five concentrations of LC (0.032 - 3.2 ?g/L) were used in a full cross factorial design. The addition of humic acid particles to the LC treatments slightly reduced the acute behavioural stress and reduced the fraction of individuals with unnaturally low locomotor activity during the postexposure period. The obtained LC50 concentrations for combined LC and humic acid treatments were increased two to fourfold compared to the pure LC treatments indicating a relatively low mitigating effect probably due to reduced bioavailability. Unexpectedly, the addition of the clay mineral montmorillonite produced the opposite and slightly synergistic effect prompting an increased behavioural stress response during exposure and an increased fraction of individuals with unnaturally low locomotor activity during the postexposure period. The obtained LC50 concentrations in the combined LC and montmorillonite treatments were decreased by a factor two compared to the pure LC treatments indicating that the presence of montmorillonite increases the

5

bioavailabiltiy of LC to G. pulex. These findings imply that the presence of suspended particles during pyrethroid exposure in the field may not mitigate the effects on aquatic biota. 19 Direct and indirect effects of fungicides on a leaf-shredding invertebrate J.P. Zubrod, N. Koksharova, P. Baudy, M. Konschak, K. Englert, R. Schulz, Institute for Environmental Sciences, University of Koblenz-Landau; M. Bundschuh, Department of Aquatic Sciences and Assessment, Swedish University of Agricultural Sciences. Decomposer-detritivore-systems are considered responsible for the leaf litter breakdown in streams – a fundamental ecosystem function. Microbial decomposers (especially fungi) are of great importance in this process as they alter the leaves (i.e. conditioning) resulting in a higher food quality for detritivorous macroinvertebrates (i.e. shredders). Consequently, shredders may be affected two-fold by fungicides, i.e. by direct, toxic and indirect, food-quality related (due to an alteration of the leaf-associated microbial community) effects. The present study investigates both pathways using the model shredder Gammarus fossarum and several organic (azoxystrobin, carbendazim, cyprodinil, quinoxyfen and tebuconazole) and inorganic (three copper-based substances and sulphur) fungicides. All organic fungicides resulted in direct, toxic effects on gammarids’ feeding activity, while sulphur was ineffective. Furthermore, indirect effects of fungicides were assessed using food-choice experiments. Gammarids preferred leaves conditioned (i.e. microbially colonized and altered) in the presence of inorganic fungicides over the control. However, they preferred control leaves over leaves conditioned in the presence of organic fungicides. This difference may be triggered by a broad-spectrum activity of inorganic fungicides, affecting also leaf-associated bacteria. The resulting lower competitive pressure may have promoted fungal growth potentially resulting in a higher nutritional value of leaf material. On the contrary, quality of leaves exposed to organic fungicides may be reduced due to their intended specificity to fungi. The underlying mechanisms for the observed indirect effects on Gammarus will be further elucidated by analyses of the leaf-associated microbial communities. Nonetheless, it is obvious that organic and inorganic fungicides may affect decomposerdetritivore-systems and thus leaf litter breakdown in different ways, which should be considered in environmental risk assessment of fungicides. 20 Effects of food starvation on digestive activities in Gammarus fossarum l. charron, UFR Sciences Exactes et Naturelles; O. Geffard, A. Chaumot, R. Coulaud, Irstea, UR MALY, Laboratoire d'écotoxicologie; A. Geffard, Université de Reims ChampagneArdenne / Sciences Exactes et Naturelles; O. Dedourge-Geffard, Université de Reims Champagne-Ardenne / UFR Sciences Exactes et Naturelles. Among freshwater species intensively used in ecotoxicology, Gammarus fossarum is a relevant test species. Many subindividual and individual responses are studied in gammarids in order to monitor and predict effects of chemical water quality on aquatic life. In this framework, energetic metabolism variations appear to be a good predictive tool to detect physiological disturbance of organisms linked to ecosystem quality. Among biological processes involving energy uptake, digestive enzymes play a key role. Some papers clearly demonstrated that the activity of these enzymes may be disturbed by pollutants. Consequently digestive enzymes can be proposed as relevant indicators of health organisms. In this way, the monitoring of digestive responses was recently successfully applied with caged organisms for the diagnosis of aquatic system. However, confounding factors may highly influence activity levels of these enzymes, consequently the impact of these parameters should be known and taken into account for an accurate interpretation of activity levels in term of toxicity. The use of control organisms for experiments allows preventing the impact of these biotic factors. However, some other factors as the amount of available food or the food eaten by organisms cannot be easily controlled. The present study aimed to show the influence of food availability on digestive enzyme activities (amylase and trypsin) on both gender of the freshwater amphipod Gammarus fossarum, throughout their reproductive cycle. Organisms were exposed to 3 levels of food

SETAC Europe 23rd Annual Meeting

available. First data showed a decrease of amylase activity correlated with the food availability. In contrast, for trypsin no effect of starvation was observed on enzyme activity for the both gender. In female, an interaction effect between moult cycle and level food available was also observed on digestive enzyme activities. So male appears as a better biological matrix for using digestive enzymes in biomonitoring survey. Moreover results indicated in starved female a delay in reproductive cycle compared to ad libitum fed organisms which demonstrated the diet stress effect on physiological responses. The digestive capacity inhibition may induce changes in energy allocation. These results concerning digestive capacities (both gender) and reproductive cycle (female) will be discussed in regard to energy reserve measurements and reproductive outcomes (fertility and fecundity). 21 Modeling physiological and chemical factors controlling lead (Pb) bioaccumulation in Gammarus pulex: Application from laboratory to field studies N. urien, J.D. Lebrun, O. Geffard, Irstea; E. Uher, HBAN; C. Gourlay-France, Irstea. Bioaccumulation is a good indicator of the metal exposure of aquatic organisms and also enables to integrate the effect of water chemistry on metal bioavailability, fraction expected to be toxic for biota. However, the link between contamination of waters and those in organisms is complex. Indeed, bioaccumulation depends on both the physiological abilities of organism to regulate contaminant, the metal considered and chemical factors of water, which can change metal uptake. The development of bioaccumulation model as a tool for quantifying metal bioavailability constitutes promising approaches for understanding and predicting metal impacts. Pb is considered within the EU Water Framework Directive as dangerous substance because it is both non-biodegradable in aquatic media and nonessential for organisms. However, few data are available about Pb bioaccumulation in aquatic organisms.\nThe present study aims at modeling the waterborne bioaccumulation of Pb and the effects of chemical characteristics of water on this process in a ubiquitous amphipod in Europe, Gammarus pulex. Then predictive quality of the model will be assessed using environmental dataset.\nIn mesocosms, gammarids were exposed to several dissolved Pb concentrations (1 to 100 ?g.L-1) so as to establish toxicokinetics and assess their capacity to regulate metal over time. Results show a significant accumulation of Pb for each tested concentration. Thus, accumulation (ku) and elimination (ke) rate constants were determined.\nThen, gammarids were exposed to 10 ?g.L-1 Pb in mesocosms with various concentrations of Na+, Mg2+ or Ca2+. Results show no significant effects of Na+ or Mg2+ on Pb uptake. However, a protective effect of Ca2+ on Pb bioaccumulation was observed. This protective effect was taken into account in the bioaccumulation model, as the apparent accumulation rate constant (ku’) expressed as a function of Ca2+.\nPb bioaccumulation measured in native gammarids collected from 24 sites of Seine watershed, all different in terms of contamination and water chemistry, were compared to those predicted. Results show that forecasts better match measured data when the model includes Ca2+. So, these results prove the significance of considering water chemistry for assessing metal bioavailability and to improve the predictions of the bioaccumulation model. As a perspective, prediction quality will be assessed by using field datasets especially from native and transplanted gammarids. 22 Toxicokinetic-toxicodynamic modeling explains differences in sensitivity among invertebrate species A. Nyman, K. Schirmer, R. Ashauer, Eawag. Species respond to chemical stress in different ways and interspecies variation in sensitivity causes large uncertainty in chemical risk assessment. But why do species differ in their sensitivity to toxicants? Generally it is assumed that closely related species have similar ways to deal with toxicants mostly due to their similarities in morphology and physiology. Calibrating toxicokinetic-toxicodynamic (TKTD) models is one way to investigate this question - toxicokinetics describe the chemical uptake, biotransformation, distribution and elimination while toxicodynamics describe the processes causing the toxicity. By comparing differences in TKTD model parameters, we can estimate whether the differences in sensitivity are caused by toxicokinetics (bioaccumulation, activation) or toxicodynamics

6

(presence of sites of toxic action). In this study, we investigated three aquatic invertebrate species: Gammarus pulex, Gammarus fossarum and Lymnaea stagnalis. The two Gammaridae species are related (family level) but the fresh water snail L. stagnalis represents a different phylum amongst animals. We used three pesticides: diazinon, imidacloprid and propiconazole. To study toxicokinetics (TK), we performed metabolite screening experiments, toxicokinetic experiments (measuring internal and external chemical concentrations over time) and chemical distribution analyses using quantitative whole body autoradiography (QWBA). Toxicodynamic (TD) tests included acute and pulsed toxicity experiments. Data collected were used to calibrate the General Unified Threshold model of Survival (GUTS). Our results show that even the closely related species, G. pulex and G. fossarum, differ in their sensitivity to diazinon: the 4-day LC50 value was roughly 5 times higher for G. fossarum. This can be explained by toxicokinetics because the simulated concentration of the metabolite diazoxon, which causes the toxic effects, in LC50 exposure concentrations is the same for both species. Lymnaea stagnalis was more than 200 times less sensitive to diazinon than G. fossarum. It did not activate diazinon to diazoxon as much as Gammaridae but this does not fully explain the differences in sensitivity among the species: in the snail internal diazoxon concentrations in the LC50 exposure exceeds those of Gammaridae species. Instead, the differences are explained by chemical distribution in the organisms as well as likely differences in biochemical and physiological sensitivity to this neurotoxin. 23 Using the simplified DEB model to analyse ecotoxicity data - a case study with the pond snail Lymnaea stagnalisE. Zimmer, Vrije Universiteit Amsterdam; V. Ducrot, INRA / Ecotoxicology and Quality of Aquatic Ecosystems; A. Agatz, University of York; T. Jager, Vrije Universiteit / Dept. of Theoretical Biology. Dynamic Energy Budget (DEB) models offer the possibility to link effects on several endpoints simultaneously. The model captures how the energy that is taken up from food is translated into the main processes of life, such as growth, reproduction, maturation, and maintenance. Toxic effects are incorporated as changes in the parameter values that capture those fluxes. Using the full DEB model can provide a consistent explanation of the effects on all endpoints simultaneously, however, it is very data intense. In this study, we discuss the benefits and limitations of using a simplified DEB model which is less data intense. The model was applied to analyse life-cycle trait values in the pond snail Lymnaea stagnalis exposed to pulses of diquat, a non-selective herbicide. A hormetic effect pattern was found in juvenile feeding and time to sexual maturity. Although all snails ate the same amount of lettuce according to their size, they grew differently. The snails exposed to 5 ?g/L grew faster and reached maturity earlier than the control, while the snails exposed to 10 ?g/L grew slower and reached maturity later than the control. We observed a dose-dependent degradation of the lettuce that was used as food source for the snails. To investigate the link between these findings, we used the simplified DEB model and analysed the growth and reproduction data simultaneously. We found that the observed effect patterns can be explained by effects on assimilation only, with assimilation enhanced by low doses of diquat, and decreased by higher exposure. Higher assimilation lead to faster growth, with differences in growth explaining the shift in time to sexual maturity. By applying the model to the toxicity data, the importance of the interaction between the lettuce and the herbicide was revealed. Effects of food stress vs toxic stress were disentangled even though assimilation efficiency was not measured during the test. When using a food source that is similar to the target of the toxicant (e.g., a plant and a herbicide), one should be careful when interpreting the test results. To this end, DEB models are an excellent tool to investigate the dynamic interactions between toxic effects and food quality and quantity on multiple endpoints. 24 Accounting for inter-individual differences between animals improves the evaluation of toxic effects A. Barsi, INRA / UMR ESE; T. Jager, Vrije Universiteit / Dept. of Theoretical Biology; L.L. Lagadic, INRA / UMR INRAAgrocampus Ouest Ecology and

SETAC Europe 23rd Annual Meeting

Ecosystem Health; V. Ducrot, INRA / Ecotoxicology and Quality of Aquatic Ecosystems. When analyzing toxicological test data it is assumed that individual animals in the test are not unique but equal regarding their physiology and life-cycle traits. When modelling toxic effects, a single model curve is often fitted to the toxicity data. The same set of physiological and toxicological model parameters applies to each individual. However, the use of such an approach may introduce a bias in data analysis because even genetically identical animals differ in their physiology and thus in their responses towards toxicants. In our study, we used a simplified Dynamic Energy Budget (DEB) model and a novel statistical approach to investigate how accounting for inter-individual variablity in toxicity test leads to more accurate estimation of metabolic and toxicity parameters, and thus to improved evaluation of toxic effects. We combined experimental and modelling approaches. We designed and performed partial life-cycle toxicity tests on juveniles and adults of the pond snail Lymnaea stagnalis. Snails were exposed in isolation to a range of acetone concentrations during 56 days. Survival and clutch production were followed every day, while shell lengths were measured once a week. Obtained results indicated greater toxicity of acetone on juveniles than on adults. Toxic effects were further modelled by using simplified DEB model where we accounted for inter-individual differences between animals. Firstly, we applied a single set of model parameters to fit the model to the obtained data on control and exposed snails. In the second step we fitted model to the experimental data for each individual in the control group. A distribution of model parameters (allocation fratcion to the soma and specific assimilation rate) that drive growth and reproduction processes of control snails was obtained. These paremeters were negatively correlated and showed considerable differences between individuals. The third step will be to use the parameter distribution obtained from control snails to constrain fits for growth and reproduction of each individual in the treatment. Starting toxicity parameters will be those obtained in the first step. In this way, we expect to achieve better estimations of physiological and toxicity parameters (e.g. narower confidence intervals) for the exposed snails, which will be great improvement of the evaluation of toxicity test data. Keywords: Dynamic Energy Budget – Toxic effects – Lymnaea stagnalis 25 Time-resolved target and non target screening with LC-MS/MS and LC-HRMS techniques as a tool for water quality monitoring. m. schluesener, T. Ternes, Federal Institute of Hydrology. The increasing worldwide contamination of water bodies with thousands of industrial and natural chemical compounds is one of the key environmental problems facing humanity. Although most of these compounds are present at low concentrations, many of them raise considerable toxicological concerns, particularly when present as components of complex mixtures [1]. More as 100,000 chemicals are registered in the European Union, on which 30,000 up to 70,000 are daily used (EINECS, European Inventory of Existing Commercial Chemical Substances). Significant sources of the entry of chemical substances to the water bodies are closely coupled with anthropogenic activities e.g. run-off of chemicals from agriculturally used areas and urban areas, the introduction of municipal and industrial waste water to receiving waters. Furthermore, by accidents or even through spills and upset condition in technical systems a unintentional discharge of chemicals into the water bodies can occur. To identify the chemicals in water bodies, a reliable water quality monitoring strategy has to be established. Therefore, two contrary approaches can be discussed: (1) target analysis and (2) non-target screening. The target analysis is based on the knowledge of the pollutant and an established analytical method. The chemical substance can be identified and quantified with a high precision in the environment, the generated data is reliable and comparable to datasets from other laboratories. However, unknown substances are not determined. Accidents or spills can not be detected with this procedure if the substance is not included in the target list. The second approach is non-target screening. This technique is based on high resolution mass spectrometry (HRMS) coupled with liquid chromatography, together with a mass-spectrometrical databases a powerful tool for the identification of substances in the environment. In this study we present target screening data of some pharmaceuticals

7

from seven years of monitoring with 28 days composite samples at the Rhine near Koblenz and discuss the temporal trends. This data is compared to a non-target approach on one day composite samples. There, data is processed using principal component analysis (PCA) combined with a mass-spectrometry database search to identify unknown substances in the river Rhine. 26 Universal Solid Phase Extraction method suitable for laboratory and field analysis prior to toxicity assessment V. Kokkali; B. Bajema, A. Berg, R. Bosch, W. van Delft, Vitens. 1.Introduction The improved quality of surface water has lead to the integration of a pretreatment step for concentrating the samples prior to toxicity assessment [1].This study aims to develop a universal Solid Phase Extraction (SPE) method usable in the laboratory and on field attached to toxicity biomonitors.This method should ensure that the selected 257 compounds give comparable recoveries when received from both the SPE advanced laboratory method and the SPE field method.3 experiments were performed.The first one involved online SPE using Symbiosis™ (Spark,The Netherlands) tandem to Q-TOF liquid chromatography-high resolution-mass spectrometry analysis (LC-HR-MS) (Agilent,The Netherlands), the second one dealt with offline SPE using Symbiosis™ and the final one involved the application of the SPE unit of TOXconltol (microLAN, The Netherlands).The extracts were analysed using Q-TOF LC-HR-MS. 2.Materials and methods 10ml (two experiments) or 1L (third experiment) of drinking water samples were spiked with 257 compounds and concentrated using Symbiosis or the SPE unit of TOXcontrol with Oasis® HLB cartridge material. The eluting solvent was 80% acetonitrile for the online experiment and 100% methanol for the offline ones. 3 runs were performed for each experiment and the extracts were analysed with Q-TOF LC-HR-MS in triplicates. 3.Results and discussion The recoveries of 237 polar compounds measured with QTOF LC-HR-MS–positive mode were comparable for offline SPE using Symbiosis and the SPE unit of TOXcontrol although the sample volume was 10ml and 1L (Fig 1). Furthermore, the average recovery of all the tested compounds was higher for the online experiment (84%) than for the other two whose procedure was more prone to human error. In particular, the average recovery ranged at the same level for both offline experiments with either Symbiosis (70%) or SPE unit of TOXcontrol (68%). 4.Conclusions In conclusion, the obtained results demonstrated that the SPE unit of TOXcontrol can be applied for concentrating high volumes of samples spiked with a wide range of polar compounds with the same efficiency as advanced instruments. Consequently, the SPE offline method can be used in the field prior to toxicity assessment for producing high quality alarms. 5.References [1] Penders E.J.M.2011. Development of aquatic biomonitoring models for surface waters used for drinking water supply. Ph.D. Thesis,Wageningen University,The Netherlands. 27 Comprehensive Target and Suspect Screening for Insecticides/Fungicides in Surface Waters by Liquid Chromatography High Resolution Mass SpectrometryJ. Hollender, Eawag / Dept Environmental Chemistry; C. Moschet, Eawag / Uchem; A. Piazzoli, ETH; H. Singer, Eawag. The use of pesticides in agriculture inevitably leads to losses of residues into surface waters. Many studies investigated the occurrence of either herbicides or legacy insecticides (organochlorines, organophosphates) in surface waters. On the other hand, fungicides (e.g., anilino pyrimidines, azoles) and modern insecticides (e.g., diacyl hydrazines, neonicotinoids) have rarely been measured in rivers although they can show ecotoxicological effects in the low ng/L range. Here, we present a combined target and suspect approach for surface water screening to measure all organic synthetic insecticides and fungicides registered in Switzerland. The method consists of an offline solid phase extraction (SPE) using a mixed-bed multilayer cartridge (enables the enrichment of a broad substance spectra) followed by liquid chromatography with electrospray ionisation and high-resolution mass spectrometry (LC-ESI-HRMS/MS; Thermo Fischer QExactive). Twenty insecticides, 23 fungicides and 8 metabolites were selected for a quantitative target method based on consumption amounts, physical-chemical properties and predicted fate

SETAC Europe 23rd Annual Meeting

behaviour. For the target method, normalized collision energies were 2 optimized and characteristic MS fragments were derived. Absolute and relative recoveries of most target analytes lay between 80-120% and LOQs were below 2.5 ng/L for 75% of analytes. The exact masses (m/z + + [M+H] , [M+Na] , [M-H] ) of the remaining insecticides, fungicides and selected metabolites were screened in environmental samples. The peaks were studied further if a meaningful peak shape as well as isotopic pattern was visible, and if the peak was missing in the blank sample. A 2 targeted MS spectra was compared with library spectra (Thermo Fisher Library Manager 2.0). If possible a standard was purchased for unequivocal confirmation. With this combined target and suspect approach, 84% of all registered insecticides and fungicides were covered (remaining substances were pyrethroids/dithiocarbamates which are not ionisable with ESI). Field measurements in 5 agriculturally influenced rivers showed low insecticide and fungicide concentrations (90% of all detections below 20 ng/L and 50 ng/L, respectively). In addition to the target approach, 3 insecticides/-metabolites and 15 fungicides were found with the suspect screening. A non-selective enrichment step together with high-resolution mass spectrometry allowed the screening of nearly all registered insecticides and fungicides. 28 Transformation of non-steroid anti-inflammatory drugs during sewage treatment E. Larsson, Lund University; S. al-Hamimi, Lund University / Centre for Analysis and Synthesis; J. Jonsson, Lund University / Department of Chemistry. Non-steroid anti-inflammatory drugs (NSAIDs) are frequently occuring in water both entering and exiting sewage treatment plants (STPs) as well as in recipients due to the consumption and later urinary/bile extraction by humans. However, a major part of these compounds are excreted from the human body and thus entering the STPs as their phase I or phase II metabolites. Several studies also indicate that the major removal mechanism for the parent compounds during sewage treatment is biological transformation. Thus, significant amounts of transformation products might also be formed during the treatment and released into the recipient. However, few studies have so far been performed regarding the presence, formation and release of specific transformation products, mainly due to the analytical challenges associated with such studies. In this work we have developed a hollow fiber liquid phase microextration method combined with LC-MS/MS analysis to simultaneously extract the NSAIDs ketoprofen, naproxen, diclofenac and ibuprofen together with six of their known degradation products (o-desmethoxynaproxen, 1hydroxyibuprofen, 2-hydroxyibuprofen, carboxyibuprofen, 4´hydroxydiclofenac and N-(2,6-dichlorophenyl)anthranilic acid) from sewage water. Enrichment is obtained in 5 h and the method provides excellent clean-up resulting in extracts which can be directly injected onto LC-MS/MS. So far the method has been applied on influent water from a Swedish STP and showed the presence of o-desmethoxynaproxen, carboxyibuprofen, 1-hydroxyibuprofen and 2-hydroxyibuprofen in amounts at the same level or exceeding those of their parent compounds. Analysis of water exiting the primary and biological treatment as well as effluent water will also be analyzed and the final aim is to calculate a mass balance for all compounds and quantify their presence, formation and relase during sewage treatment which is planned to be finished in early 2013. 29 HPLC-QqLIT-MS2 multiresidue trace analysis of sulfonamide antibiotics and their metabolites in soils and sewage sludgeS. Diaz-Cruz, IDAEACSIC / Environmental Chemistry; M. Garcia-Galan, CSIC / Environmental Chemistry; D. Barcelo, IIQABCSIC. Several studies have demonstrated the widespread presence of sulfonamides (SAs) in the environment and identified their entrance pathways [1]. The occurrence of their main metabolites, the acetylated conjugates, has become evident recently, especially in environmental waters [2]. These metabolites, once excreted, can deconjugate and revert back to the original parent drug [3]. Therefore, the inclusion of these metabolites in environmental occurrence surveys for SAs is key to obtain more complete and reliable information on the actual levels, avoiding potential underestimations. The lack of sensitive enough methodologies for the sulfonamide analysis in solid matrices and the need to consider

8

the presence of their metabolites promoted this study. The present work describes the development, validation and a practical application of an automated analytical method based on pressurized liquid extraction (PLE) followed by solid-phase extraction-liquid chromatography–tandem mass spectrometry (SPE-LC–MS/MS) for the simultaneous determination of 22 SAs and acetylated metabolites in sewage sludge and soil samples. Matrix effects were evaluated, and different internal standards were used for quantification. The recovery efficiencies were found to be 60%130% for almost all SAs in both matrices and at two spike levels. The -1 MLODs achieved were in the range 0.01–4.19 ng g . The method was applied to evaluate the occurrence of the target SAs in sewage sludge and soil samples taken in wastewater treatment plants and agricultural areas. Results confirmed the wide presence of SAs in both matrices. -1 Maximum concentrations corresponded to sulfamethazine (139.2 ng g -1 and 8.53 ng g for sewage sludge and soils respectively). The acetylated metabolites were detected in sewage sludge and in soils. To the author´s knowledge, it is the first time that acetylated metabolites of SAs have been identified in sewage sludge. Acknowledgements We acknowledge the Spanish Ministry of Economy and Competitiveness (SCARCE project, Consolider Ingenio 2010 CSD2009-00065). References S.C. Kim, K. Carlson, Environ Sci Technol 41 (2007) 50. M.J. García-Galán, T. Garrido, J. Fraile, A. Ginebreda, M.S. Díaz-Cruz, D. Barceló, Anal Bioanal Chem 399 (2010) 795. M.J. García-Galán, T. Frömel, J. Müller, M. Peschka, T. Knepper, S. Dïaz-Cruz, D. Barceló, Anal Bioanal Chem 402 (2012) 2885. 30 Environmental fate of nitramines and nitrosamines released as degradation products from post combustion CO2-capture plants L. Sorensen, SINTEF Materials and Chemistry / Marine Environmental Technology; O. Brakstad, K. Zahlsen, E. da Silva, A. Booth, SINTEF Materials and Chemistry. Recently there has been increased awareness of the need to understand and control emissions from post-combustion CO capture (PCCC) plants. Among the large number of possible and 2 identified degradation products that may form in amine based PCCC plants, nitramines (R NNO ) and nitrosamines (R NNO) are of 2 2 2 particular concern, due to their carcinogenic potency. In this study, the hydrolytic and photolytic degradation rates of a suite of four nitramines and five nitrosamines are reported in freshwater. In addition, the potential of these compound groups to degrade by microbial action under oxic and anoxic conditions in freshwater and lake sediment is also reported. As nitrosamines and nitramines will be released in very low concentrations from a PCCC plant during normal operation, a major focus of the project has been to develop suitable analytical methods for working with these contaminants at environmentally relevant trace levels. For this purpose, chemical analysis were performed by LCMS/MS QqQ using deuterated internal standards for quantification. Of the nine studied compounds, seven showed no hydrolytic degradation at any pH. Hydrolytic stability of a range of nitramines and nitrosamines has also previously been observed. In the current study, two compounds derived from the amine piperazine (1-nitropiperazine and nitrosopiperazine) showed degradation, but only at pH 7. Results from a rate determining study of these compounds at three different temperatures will be presented. Preliminary tests have shown that nitrosamines in surface waters will degrade rapidly (within days) when exposed to natural sunlight, whilst nitramines are photostable. Results from an experimental study aimed at determining the degradation rates of the four nitrosamines NDELA, NDMA, NMOR and NPz exposed to artificial sunlight will be presented. Previous studies have indicated that the biodegradation of nitrosamines in freshwater is enhanced greatly by the presence of a hydroxyl group, such as for n-nitrosodiethanolamine (NDELA). It is expected that the same trend will be observed for the nitramines. Two freshwater biodegradable compounds, NDELA and the nitramine 2-nitroaminoethanol (MEA-NO ), will be tested alongside 2 two less biodegradable compounds, n-nitrosodimethylamine (NDMA) and n-dimethylnitramine (DMNA), to determine whether or not microbial degradation in the water column and the lake sediment are comparable for these compounds. 31 Abandoned uranium mine-induced effects in caged roach: a

SETAC Europe 23rd Annual Meeting

multiparametric approach for the evaluation of in situ metal toxicity B. Gagnaire; A. Bado-Nilles, Université de Reims/INERIS; S. Betoulle, Université de Reims; R. Amara, E. Kerambrun, Université du Littoral Côte d’Opale; X. Denier, C. Minier, Universite du Havre; I. Cavalie, C. Adam-Guillermin, IRSN; A. Le Guernic, IRSN/INERIS; W. Sanchez, INERIS. In the frame of the characterization of environmental risk assessment linked to abandoned uranium mines in the Limousin region (France) and in the context of the DEVIL program (Development and validation of fish biomarkers for the implementation of environmental legislation) led by INERIS (National institute for industry environment and risks), a study has been conducted on the effects of uranium mining releases on roach health status. The first part of the work was the search of potential sites for the study. Based on information from IRSN (National institute for radioprotection and nuclear safety) and from the owner of old mines, two private pounds were identified: one presenting uranium contamination, and one upstream the mining zone presenting no uranium in water. Roaches were therefore caged during one month on these two pounds. Biological sampling have been performed at the beginning of the experiment, and 2 and 4 weeks after, in order to measure several parameters: immune parameters followed by flow cytometry (phagocytosis, ROS production, apoptosis, lyosomes), genotoxicity (comet assay), oxidative stress (glutathione, lipoperoxidation), biotransformation (EROD, GST, ABC transporter), neurotoxicity (AChE), endocrine disruption (vitellogenin) and physiological parameters (ADN/ARN ratio, stomach contents, otoliths, condition index, sex ratio). Physico-chemical parameters of water and sediments (cations, anions, metals, radioactive elements) were also followed and bioaccumulation of metals in several organs (muscle, liver, kidney, gonad and gill) was measured. The results showed on the uranium contaminated site a water contamination of iron, aluminium, barium and manganese. Aluminium and iron were also present in the water of the uranium-free site. The sediments from the uranium contaminated site showed high levels of radioactive elements coming from the disintegration chain of uranium. Preliminary biological results indicated a stimulation of immune parameters, an increase of oxidative stress and a decrease of AChE in fish coming from the uranium contaminated site compared to the uranium-free site. The whole results, using integrated index, will allow the establishment of roach health status in the context of pluri-metallic mining release pollution. These data will strengthen the research for the improvement of methods and tools needed for the environmental risk assessment associated to radioactive substances in the environment. 32 Evolution of aquatic ecology at the abandoned Silvermines site, Co. Tipperary, Ireland. S. Barnes, Golder Associates UK Ltd; K. Harrington, B. Balding, A. Murray, Golder Associates Ireland / Town Centre House, Dublin Road. Abstract The Silvermines area of North Tipperary in Ireland has experienced mining activity for over a thousand years, with lead, zinc, copper, barite and sulphur being extracted from a series of Carboniferous limestones. The last mine to operate in the area was closed in 1993. A legacy of contamination and derelict surface structures remain. Following the death of cattle in the vicinity in 1999, a governmental Inter-Agency Group was established to conduct an investigation. A ‘Silvermines Rehabilitation Project’ resulted, and included proposed development aimed at containing mine wastes derived from problematic sites across the area and to ultimately help reduce emissions to the water environment. The Project involved an evaluation of aquatic ecology spanning the termination of mining activity. Biological water quality data since the 1970’s from the Kilmastulla and Yellow Rivers area of the Shannon catchment have been sourced and supplemented by update survey, and spot stream water chemistry data. Historical mining activities have rendered an extremely varied habitat and water quality distribution within a relatively small 2 study area of approximately 13 km . Q-values from the Environmental Protection Agency (EPA) Q-rating (biotic indices) methodology ranged between 1 and 5 during the mining period. Q-values have been fairly consistent at individual locations between consecutive observations; but there is limited evidence that significantly impacted waters persist at distance downstream. In the 20 years since mine closure, locations in

9

proximity or down stream to mining activities have often shown some natural improvement in River Quality Classification, but slight to moderate pollution persists. Upstream samples typically have high diversity and a frequency of Group A and B taxa including pollution sensitive mayflies and stoneflies. Impacts have severly reduced overall macroinvertebrate diversity and eliminate sensitive fauna. A moderate to strong inverse correlation (r) exists between Q-values and the spot zinc concentrations (-0.73) determined, which reflect an inverse correlation of -0.64 between zinc concentration and the total number of taxa. A habitat enhancement programme is vital to rehabilitate the study area. It is essential remedial works, not yet implemented, do not result in future release of mine tailings and/or excessive sediment loads. 33 Mine Dewatering into a Tropical Ecosystem J. Woodworth, GHD; A. Sawicki, Vista Gold Australia; A. Brandis, Envirotech Monitoring. Vista Gold Australia Pty Ltd (Vista Gold) proposes to reestablish and operate the Mt Todd Gold Mine, located 55 kilometres north of Katherine and 250 kilometres south of Darwin in the Northern Territory, Australia. The site was previously mined for gold in the 1990s with mining operations ceasing in the early 2000s. Mining infrastructure such as open pit, tailings dam, waste rock dump, retention ponds and remains of processing facilities remain on-site. The site has many surface water bodies (retention ponds or RPs), some of which contain water with high metal and low pH levels that have the potential to overflow during the wet season. The site needs to be dewatered prior to mining commencing. The poor water quality on site combined with the tropical wet and dry seasons required a site specific ecotoxicological program integrated with environmental monitoring to allow the site to be dewatered and provide maximum protection to the receiving ecosystem. Currently Vista Gold is treating 10.5 GL in the Batman Pit (RP 3) to reduce the high cadmium, copper, lead, nickel and zinc concentrations to allow increased discharge from the site during the 2012/2013 and following wet seasons. Ecotoxicological analysis of the treated water has enabled the correct volume to be calculated for discharge to meet the 80% species protection values (ANZECC 2000) at a specified point on the Edith River as stated in the Waste Discharge Licence for the site. Trigger values at the 80% species protection level have been applied for a two year period only to allow site dewatering and prevent catastrophic uncontrolled discharge from the site. Community concern has been shown regarding the use of the 80% values as the discharge from the mine site enters the Edith River which is a tributary of the Daly River, a world renowned barramundi fishing location. Aquatic organisms caught in the Edith River also provide a traditional source of food for the indigenous population. Ecotoxicity testing and monitoring programs have helped to alleviate the community concerns to some extent as no significant impacts have been detected in macroinvertebrate and sediment quality studies. This presentation will discuss the application and results from ecotoxicological analysis using tropical species bioassays and the integration of a macroinvertebrate monitoring program designed to ensure that the downstream ecosystem is not adversely impacted by the mine dewatering. 34 The environmental legacy from centuries of metals mining in south central Scotland I.W. Oliver, Keele University / Physical Geographical Sciences; F. Naysmith, Scottish Environment Protection Agency / Operations; A.M. Moorhouse, L.M. Wyatt, I.A. Watson, the Coal Authority; J. Redshaw, . Mining for lead (Pb) and other metals in the Leadhills area of S. central Scotland dates back to the 1200s. The long mining history has left its mark and an environmental legacy. 1 Elevated levels of Pb have been noted in the soils of the area . Fish monitoring in the 1980s in one stream, the Glengonnar Water, found trout with blackened tails (an indicator of chronic Pb poisoning) while 2 analysis of fish tissues revealed elevated Pb concentrations . Monitoring 3 of the Glengonnar Water under the Water Framework Directive has also revealed it to persistently fail environmental quality standards (EQS) set for cadmium (Cd), zinc (Zn) and Pb in water. The local environment is thus affected by its mining legacy and this influences public perceptions. To inform decisions on potential future remediation work, the Coal Authority, with support from SEPA, investigated the

SETAC Europe 23rd Annual Meeting

extent of contamination in the soils, sediments and water of the Glengonnar catchment to pinpoint the locations where metal inputs occur. A preliminary examination of another stream, the Wanlock Water, in a neighbouring catchment that had not previously been assessed for metal contamination was also conducted. Mine drainage and stream water assessments revealed that while Zn and Cd were entering Glengonnar Water as dissolved metals directly from mine drainage points, Pb was entering via diffuse inputs (i.e. run-off bearing particulate Pb from soils and surface depositions). Pb concentrations reached 4.8% (w/w) in soil and 10% (w/w) in stream sediments. Whilst having no statutory standing in the UK, sediment quality guidelines 4 proposed elsewhere suggest that metal concentrations are present here at potential effect levels and thus ecological impacts beyond those previously noted for fish may have occurred. Water build-up behind blockages within old mine adits were also identified as posing a potential flood risk. For the Wanlock, preliminary analyses indicate that it too is likely to exceed EQS for Pb and Cd in water. The work highlights the mining legacy in the area and the future work required to develop remediation options. 1 Towers et al. 2006 Scotland’s soil resource- current state and threats. 2 ClydeRiverPurification Board 1984 Concentrations of trace metals in water and brown trout from Glengonnar Water. 3 Directive 2000/60/EC of the European Parliament: The Water Framework Directive. 4 Council of Ministers of the Environment (Canada) 2002 Canadian Sediment Quality Guidelines 35 Venus Mine Site, Yukon, Canada: Acid Mine Drainage Assessment and Mitigation M. Dodd, Royal Roads University / School of Environment & Sustainability; W. Liebau, Aboriginal Affairs and Northern Development Canada / Waste Manangement. Venus Mine is a former silver mine located on the at the base of Montana Mountain along the South Klondike Highway, 20 km south of Carcross, Yukon, Canada. The mine produced approximately 84,000 tonnes of ore over its lifetime generating about 50,000 tonnes of waste rock which was piled along the edge of the mountain near adits and shafts. The results of acid base accounting test in this study indicated that the waste rock contained patches that were oxidized and had acid pH while the remaining unoxidized materials had the potential to become net acid generating. Arsenic, Cd, Pb, and zinc in effluents from adits and waste rock piles were very high. The elevated levels were attributed to metal leaching from the waste rock since laboratory leach test showed elevated levels in the leachate. The migration of As, Sb, Cd, Pb, Ag and Zn towards Tagish Lake was occurring however metal concentrations in the lake were comparable to background levels due to attenuation by natural onsite vegetation. Despite the indication of potential net acid generation of the waste rock the effluents were alkaline (pH 7.99 – 8.71) while SO 24 and Fe concentrations were relatively low. It therefore appeared any acid generation within the waste rock was occurring at a rate that was not exceeding the available neutralization potential. Remedial options to prevent further oxidation and ensure the neutralization potential of the waste rocks were not exceeded were investigated. The use of biosolids amendments was unsuccessful since that yielded increased metal leachate concentrations. Underwater storage and covering the waste rock in place were also considered undesirable due to the steepness of the slopes. The use of vegetative cover was deemed the most feasible based on field observations and laboratory data. Natural vegetation at the Venus Mine site could be used to minimize the effects of acid mine drainage and associated elevated metal concentrations in effluents from the adits and waste rock. This could be accomplished by diverting the effluents and any surface run-off into an identified vegetated area onsite. Drainage ditches could also be constructed to intercept and divert water from waste rock piles in addition to the use of vegetative cover. Ongoing studies include investigations into metals uptake into the onsite vegetation and laboratory bench scale studies on soil amendments to improve neutralization and re-vegetation. 36 When is clean “clean enough”? Integrating ecological function with remediation goals for the historic Britannia Mine.B.G. Wernick, Golder Associates Ltd. The former Britannia Mine operated near Vancouver, Canada from 1904 to 1974, and at its peak was the largest

10

copper mine in the British Commonwealth. During its 70 years of operation it generated more than 40 million tonnes of tailings, much of which was deposited in the adjacent marine environment of Howe Sound and used as fill along the Britannia Beach shoreline. The acidgenerating tailings and upland former mine workings leached dissolved copper and zinc into Britannia Creek (which drains to Howe Sound) as well as directly into Howe Sound until recently when the provincial government began an ambitious remediation program intended to intercept, collect and treat water-borne metals discharging to the environment. Monitoring of water quality and intertidal ecology is providing information confirming that environmental conditions are improving along a majority of the shoreline and that residual impacts appear to be localized around areas of historical infrastructure. A remediation program or risk manager for a contaminated site needs to know “how clean is clean enough?” The objective of the initial remediation activities was to address severe, acutely lethal conditions that posed a mortality hazard to out-migrating juvenile salmon and this objective has been achieved. As site characterization continues and additional remediation is undertaken, realistic remediation goals need to be defined. In the case of the Britannia Mine, the return of the shoreline to pre-development conditions is likely not possible as the long-term operations of the mine resulted in significant filling and replacement of the natural shoreline with engineered surfaces (e.g., rip rap). Moreover, the construction of a transportation corridor adjacent to the shoreline will limit the use of intrusive remediation techniques. “Resource conservation objectives” (RCOs) have thus been articulated to help direct remediation efforts. The RCOs consist of a description of important shoreline ecological functions and/or features as well as acceptable levels of those features that are based on: 1) the results of the monitoring program which provide an understanding of the ecological conditions of near-field and reference areas, including natural variability; 2) the broader mandate of environmental agencies to maintain and/or or restore a healthy productive ecosystem; and 3) the local community’s desire for environmental improvement but not at all costs. 37 Use of environmental science and scientific results by ECHA: present and future J.V. Tarazona, C. AJAO, D. APE, R. CESNAITIS, B. DILHAC, D. KNIGHT, H. PIHA, H. SCHIMMELPFENNIG, ECHA. The European Chemicals Agency (ECHA) is the driving force among regulatory authorities in implementing the EU's groundbreaking chemicals legislation. The use of the best available scientific knowledge, within the limits of the regulatory context, is an essential element in all REACH processes. This is partially addressed through the ECHA Scientific Committees and complemented with internal horizontal and targeted scientific activities. As the Agency is entering into the consolidation phase and has new regulatory roles in the field of Biocides and PIC regulations, its strategy for addressing the scientific challenges is being adapted. This contribution will present a general overview on the current ECHA practices for monitoring and examining the regulatory implications of knowledge progress in environmental sciences, and for ensuring that updated states of the science regarding environmental fate and ecotoxicity of chemicals can be applied for substantiating ECHA decisions, opinions, recommendations and guidance, giving concrete examples on particularly relevant topics. In addition, we will present our approach to ensure that ECHA can play an increased and significant role of the scientific community in Europe and world-wide. With this we aim at a wider contribution of ECHA in the identification of scientific needs for environmental hazard and risk assessment, and at channelling ECHA contributions to the challenge of transforming research results into regulatory scientific knowledge, ready for supporting decision-making processes. ECHA Topical Scientific Workshops are a key element of this development, and the first one, already in 2013, will address the risk assessment for sediment. 38 Missing links in EU chemical risk management of consumer products L. Molander, Stockholm University / Department of Applied Environmental Science (ITM); M. Breitholtz, S. Bejgarn, A. Sobek, C. Ruden, Stockholm Univeristy / Department of Applied

SETAC Europe 23rd Annual Meeting

Environmental Science (ITM). The aim of this study is to investigate the interlinks between EU legislations concerning chemicals in consumer articles and cosmetic products with regard to how they share and integrate information useful for assessing and reducing negative environmental impacts. This contribution consists of two case studies, including (1) an anlysis of the extent to which the regulation of hazardous substances in articles under REACH is coherent with the restrictions under the Water Framework Directive (WFD), and (2) an investgation of whether publicly available data on PBT properties of sunscreen UV-filters would lead to hazard classifications under the CLP regulation, as they are now exempted from CLP. The coherence analysis shows that the majority (22 of 24) of the substances or groups of substances that are prioritized for phase-out under the WFD are allowed to be used in most articles according to REACH. The incoherencies in which substances are prioritized and restricted between REACH and the WFD constitute a possible obstacle for reaching environmental goals as set out by the WFD. For 13 of 26 UV-filters a classification as hazardous to the aquatic environment could be made based on publicly available data, or was already made by industry. Two sunscreen products were also evaluated and both could be classified as hazardous to the aquatic environment according to the CLP criteria for mixtures.The aquatic hazard classifications of UV-filters in sunscreens and sunscreen products derived from publicly available data currently fall between different regulatory frameworks. Such missing links hamper important information from being disseminated. In conclusion, information that is made available, e.g. via the REACH registration database, should be utilized to a greater extent in order to increase coherence between different regulatory frameworks. This would more effectively reduce environmental risks and enable a better management of these risks. 39 UVCB : Which data entries for unknown substances ? P. Garrigues, University of Bordeaux / Institute for Molecular Sciences; F. Lafaye, University of Lyon / ENTPE/UMR CNRS EVS; N. Leca, University of Bordeaux / Institute for Molecular Sciences. The categorization of UVCB substance has been introduced into the REACH regulation. UVCB is a substance of unknown or variable composition, complex reaction products or biological materials. Since characterization of UVCB is difficult due to its complex composition, the production of data could be waived or be replaced by more specific analytical techniques. The lack of some data or the production of data under specific techniques will lead to generate sets of data that may not really fit with the analytical data obtained with well defined substances. The presentation will be focused on the characterisation and the identifiers of UVCB, required for the registration of chemicals, the first step of REACH procedure. Some difficulties appear in this first phase and others will certainly be stressed in the future in the evaluation phase.Regard to analytical data, spectroscopic methods must be produced when providing information on the composition of UVCB substance (UV/Vis, Infra-red, nuclear magnetic resonance or mass spectrum). Some of these techniques appear as non appropriate for characterization of some UVCB substances for several reasons. Most of the spectroscopic techniques are well suited for analysing individual chemical products or very simple complex mixtures. When several dozens of constituents are present, the resulting spectra will provide average broad-banded spectra without any clear identification. This is particularly true for UV/Vis spectra. In that presentation, several case studies will be presented on identification of constituents some complex UVCB (natural products, oil refinery products). The results will be discussed toward the requirement of the REACH regulation annex 6. 40 Have Sediment Quality Guidelines Improved Since the SETAC 2002 Pellston Workshop? R.J. Wenning, ENVIRON International Corporation; K.M. Leung, The University of Hong Kong; G.E. Batley, CSIRO Land and Water / Centre for Environmental Contaminants Research; T.S. Bridges, US Army Corps of Engineers / USERDC-WESEM-D; S. Hall, ENVIRON International Corp; G. Merrington, Environment Agency. Historically, the significance of contaminant concentrations in sediments has been assessed by making comparisons

11

between measured concentrations and numerical sediment quality (SSD) is a key tool to assess the ecotoxicological threat of pollutants to guidelines (SQGs). Based on such comparisons, the potential risks, or biodiversity. SSD is widely used by environmental regulatory bodies to hazard of sediment-bound contaminants can be estimated and decisions define safe levels for toxicants. SSD is a representation of the sensitivity made to reduce perceived threats to aquatic life. The derivation and of a community to a pollutant as a probability distribution. Toxicity scientific validity of different approaches used to derive SQGs was experiments provide critical effect concentrations (CECs) for single reviewed in 2002 at a SETAC Pellston Workshop (Wenning et al. 2005). species, which are used to characterize this distribution. Very often, At that time, limitations were identified, specific research activities CECs cannot be estimated precisely, because the real values lie outside recommended, and cautions issued on the misuse of SQGs. Here we the range of the tested concentrations, or the uncertainty is very large. consider the progress made to advance scientific knowledge about the Then only a lower or a higher bound for the real CEC values is known, relationship between sediment chemistry and impacts on aquatic or the interval including them. Such data are called censored data. The ecosystems. Advances have occurred, but translation into improved general practice is to discard left/right censored data and to consider the ecological risk assessments has been less rapid and remains problematic. centre of the interval censored data as a pointwise value. However, this In ecotoxicology, a limitation was the suite of bioassays; however, the severely undermines the reliability of any conclusion drawn from the number of bioassays has expanded beyond amphipods to include SSD analysis. Therefore, we present a new statistical web-tool to copepods, benthic algae, bivalves, polychaetes and snails. Ecogenomics properly include censored data in SSD using maximum likelihood has revolutionized ecological assessments allowing thousands of taxa to methods. MOSAIC_SSD is based on a general R package, fitdistrplus, be compared to the traditional macrofauna used in standard ecological which can be used to fit an SSD, encompassing all the functionalities of surveys. The recognition that major drivers of sediment contaminant existing software such as CADDIS SSD Generator, ETX or BurrliOZ bioavailability are grain size and organic carbon content led to [5]. Package fitdistrplus allows fitting any user-defined distribution development of the sediment biotic ligand model and new approaches to through a variety of methods and computes confidence intervals using deriving SQGs that incorporate these factors. Weight of evidence bootstrap. The possibility to include not only left/right, but also interval approaches have been embraced in California, Australia and elsewhere, censored data using a frequentist framework is not available in any of lessening reliance on expert judgment. Improved numerical guidelines the aforementioned software, and is generally not put into practice in the are now available for several key contaminants. Problematic, however, literature. We will demonstrate how discarding censored data leads to a remains lack of clarity about contaminated sediments referred to as biased estimation of the SSD, and illustrate the versatility of ‘transition zone’ conditions not easily identified as either “contaminated” or MOSAIC_SSD. By lowering the barrier to using sophisticated “clean”. The challenge remains determining the role contaminated statistiscal tools, our web-tool offers a convenient and user-friendly sediment conditions play in supporting either a ‘healthy’ (but possibly interface to perform a statistically sound SDD estimation without altered), or ‘degraded’ ecosystem. Although understanding of the worrying about technicalities. It is easily accessible within the familiar ecotoxicology of contaminants has improved, research on organism and framework of a webpage, providing a turnkey alternative for species responses to contaminant exposure is less advanced. In fact, non- ecotoxicologists. Moreover, the fitdistrplus R-package allows for many chemical stressors have complicated interpretations about causal more refinements, such as fitting through moment matching or relationships between contaminant concentrations in sediments and minimum distance estimation, as well as comparing goodness-of-fit biological responses. statistics for non-censored data. Together, MOSAIC_SSD and fitdistrplus should satisfy the needs of both casual and advanced users in 41 Chesar – ECHA’s Chemical Safety Assessment tool c. team, h. ecotoxicology. magaud, A. Ahrens, r. bernasconi, e. nogueiro, r. gilioli, z. gabor, s. frattini, ECHA. The Chemical Safety Assessment and Reporting tool, 43 A novel approach to allocation of by-products in background Chesar, has been developed by the European Chemicals Agency systems and databases B. Weidema, The ecoinvent centre;G. (ECHA), aiming at supporting companies in carrying out their Chemical Wernet, ecoinvent Centre. The problem of allocation is a long-standing Safety Assessment (CSA), preparing their Chemical Safety Report issue in Life Cycle Assessment (LCA). The question of recycling (CSR) and finally generating the Exposure Scenarios to be annexed to represents an especially difficult question, as physical or economic the Safety Data Sheet for communication to downstream users. Chesar allocation properties can be difficult to determine in such cases. The provides a structured workflow: it guides the users to assess a substance ecoinvent database has so far used 3 distinct approaches to deal with bystarting from the hazard information and conclusion available in a products, recyclable materials and wastes. Recyclable materials were IUCLID 5 dataset, through the description of the uses of the substance simply handled using the cut-off approach. This presentation describes a and the definition of the existing operational conditions and risk new approach considered for the next version of the ecoinvent database, management measures (OC/RMM). It calculates or accommodates the called Allocation at the point of Substitution (ApoS), that removes the results of the environment and human health exposure estimations for cut-off approach and handles all by-products consistently. In ApoS, any quantitative assessments. Exposure scenarios are generated and risk allocation is deferred to a point after any necessary treatments of bycharacterisation (quantitative and/or qualitative) is performed. The products, e.g. in recycling. A process system can be combined to create Chesar library enables the re-use of pre-defined CSA elements like use a larger system that only stops when by-product treatment has resulted maps, conditions of use and SpERCs (specific environmental release in a marketable product. This does not remove the need for allocation categories) that can be exchanged easily among users. Chesar increases completely, but it does avoid the need for allocation factors within the efficiency in conducting the CSAs (re-use of CSA elements, treatment systems. In ApoS, any by-product requiring treatment is communication with IUCLID) and provides consistency between CSAs moved to be a negative input in the producer and a negative output in and the information communicated downstream. Presentation the treatment activity, comparable to a treatment service input. The bypreference: posterpresentation Session Risk assessment, regulation, product of the treatment is merged into the producer to result in a system policy and public perception Keywords: CSA, CSR, exposure scenario, where the treatment activity is no longer multi-functional. Thus, (Co-)authors: Hélène Magaud – ECHA, Andreas Ahrens – ECHA, allocation within the treatment activity between the function of a waste Roberta Bernasconi – ECHA, Stefano Frattini – ECHA, Zsolt Gabor – removal service and a useful by-product is avoided and the distinction ECHA - Roberto Gilioli - ECHA, Eugénia Nogueiro – ECHA. between individual activities is maintained, a feature critical for regionalized impact assessment. The resulting datasets are further 42 MOSAIC_SSD: A new web-tool to include censored data in SSD adapted to consolidate the resulting datasets. Using ApoS, treatment byG. Kon Kam King, Université Claude Bernard Lyon / Laboratoire de products are available to be included in other product systems carrying Biométrie et Biologie Evolutive; P. Veber, Université Claude Bernard the share of the burdens from the previous life cycles. Such assessments Lyon 1 / Laboratoire de Biométrie et Biologie Evolutive; S. Charles, were not possible in previous versions of the ecoinvent database due to University Lyon / Laboratory of Biometry and Evolutive Biology; M. the cut-off approach. The presentation shows results of such Delignette-Muller, VetAgro Sup. The species sensitivity distribution comparisons. ApoS is a novel approach to determining the impacts of by-

SETAC Europe 23rd Annual Meeting

12

products on a database-wide level, and allows the comparison of treatment products with comparable virgin materials. Therefore, it can offer a critical new perspective on process systems. 44 Methodology applied to the background analysis of energy data to be considered for the European Reference Life Cycle Database (ELCD) M. Recchioni, EC JRC IES; D. Garrain, CIEMAT / Energy Dpt. - Energy Systems Analysis Unit; S. Fazio, EC JRC - IES; C. De la rua, CIEMAT; F. Mathieux, EC JRC - IES; Y. Lechon, CIEMAT / Energy Dpt. - Energy Systems Analysis Unit. In the Integrated Product Policy Communication of 2003, the European Commission recognised LCA as “the best framework for assessing the potential environmental impacts of products”. Since then, life cycle approaches were further strengthened in EU policies through the Sustainable Production and Consumption/Sustainable Industry Policy Action Plan Communications that encompasses various policies. Within this context there is an urgent “need to improve data availability and quality worldwide by internationally cooperating on LCA data and methods”. To address this situation, the European Reference Life-Cycle Database (ELCD) has been developed by the EC’s Joint Research Centre (DG JRC). Within the ELCD, several energy-related data are provided, being energy a major input for almost all the environmental analyses of products or processes. The present paper will provide the description of the methodology that will be used for the analysis of the quality of energy data for European rd markets that are available in 3 party life cycle databases and from authoritative sources that are, or could be, used in the context of the ELCD. The methodology has been developed by a joint project with the cooperation of the EC JRC and the Energy Systems Analysis (ASE) Unit of CIEMAT (Public Research Centre for Energy, Environment and Technology).The methodology was developed and tested on the most relevant energy datasets for the EU context; however it can be widely applied for the quality evaluation of energy data, even if provided by rd other sources. Basing on the above mentioned principles, some 3 party life cycle databases have been selected, to be compared with ELCD database.The proposed approach has been based on the quality indicators developed within the ILCD handbook that have been further developed to facilitate their use in the analysis of energy systems. The quality indicators have been refined in order to identify key aspects that are involved in both quality and methodological aspects for energy related LCI datasets.The quality of each dataset can be estimated for each indicator and then, compared across the different databases/source. Results will point out the findings and/or recommendations in order to improve the data quality as regards the established criteria. A template including quality parameters, rating, and suggestions for improving (e.g. sources or studies to be included) will be compiled for each analysed dataset. 45 Global Sensitivity Analysis: a tool to analyse LCA variability of energy systems P. Padey, EDF Mines ParisTech; D. Le Boulch; R. Girard, Mines ParisTech; I. Blanc, Mines ParisTech / Centre for Energy and Processes. Policy makers are nowadays debating about the future electricity mixes that should be deployed. The environmental impacts of electricity generation systems is one of the central issue for this debate. They have been widely assessed over the past decades, in particular with the LCA approach. Several literature reviews have shown the large variability associated with these results. It leads sometimes policy makers to consider LCA as an inconclusive method. Improving the understanding of the LCA results variability origins is a key issue to extend the use of LCA as a decision support tool. One approach to adress variability are sensitivity analysis (SA). However, when dealing with environmental impact assessment, most SAs remain at a local level or evaluate the variation of the input parameters one factor at a time. These approaches only partially reflect the LCA results variability; indeed, it does not consider the full range of input parameters interval and their probability distribution. To overcome these limitations, Global Sensitivity Analysis (GSA) approach has been developped in statistics. It enables apportioning the results variability of a model to its different input parameter variability, by varying all of them simultaneously according to their probability distributions. This link between result

SETAC Europe 23rd Annual Meeting

variability and parameter variability is quantitatively evaluated by the calculation of the so called Sobol indices. While it has been applied in only a few analyses in the field of environmental impact assessment, this statistical tool is yet to be embedded in LCA methodology. Thereby, this paper aims at proposing a method to implement GSA in the LCA field to adress the results variability issue related to energy pathways. We applied the proposed methodology for assessing the carbon footprint variability for two renewable systems: the photovoltaic (PV) electricity and the wind power electricity pathway in Europe. Geographical parameters (irradiation and wind profile) are found to be the parameters inducing the highest sensitivity while technical parameters (PV efficiencies for example) and methodological parameters (life time of the system for example) are of the same order of importance. Such ranking is of interest for decision makers to understand the relation between the carbon footprint and its sources of variability. Applying the GSA approach in the LCA field enables a better understanding of the environmental impact variability. 46 How to fill data gaps in life cycle inventories? A case study on CO2 emissions from coal electricity Z. Steinmann, Radboud University Nijmegen / Departmenf of Environmental Science; M. Hauck, Radboud University Nijmegen; A. Schipper, Radboud University; M.A. Huijbregts, Department of Environmental Science; A. Venkatesh, R. Karuppiah, ExxonMobil Research and Engineering Company. Life cycle assessment (LCA) is increasingly being used for decision-making purposes. Approximations and assumptions are often made if appropriate data are not readily available. These proxies may introduce error or uncertainty into the results, leading decision makers to implement suboptimal or deleterious actions. Electricity constitutes one of the largest sources of environmental impacts for many products and processes. Approximations of power generation data inventories using ‘typical’ values from other contexts (for example, assuming Indian power plants have impacts similar to U.S. power plants) may significantly influence an assessment of environmental impacts over a product’s life cycle. We employed a multiple linear regression and a locally weighted linear regression approach to estimate the greenhouse gas emissions associated with power generation in the absence of reported data. Both approaches are built on reported emissions in selected nations and technological information reported in the World Electric Power Plants (WEPP) database. Emissions data were collected independently from governmental databases and reports and other public literature for 446 coal power plants, primarily in the U.S., Australia, India, the EU and South Africa; approximately 90% of these plants (401) were used for model training, and the remainder (45) were used for model validation. WEPP reports data on over 3000 coal power plant characteristics across the world, including steam pressure, type of fuel, age and capacity of plant; however, no data on efficiencies, net generation or emissions are included. Our framework hypothesizes that emissions from coal power plants can be explained by a subset of factors reported in the WEPP database as well as the GDP per capita of the resident nations of those plants. We then applied these models to all coal power plants in the WEPP database to predict CO emissions per kWh of electricity they 2 generate. We propose that this framework may be employed to assess “missing” data in LCAs of other products and processes. 47 Exploring parameter uncertainty and seasonal variability for human health intake fractions in life cycle assessment. R. Manneh, Chemical Engineering; M. Margni, Ecole Polytechnique de Montreal / Department of Mathematical and Industrial Engineering; R.K. Rosenbaum, Technical University of Denmark / Management Engineering; L. Deschenes, Ecole Polytechnique de Montreal / Genie Chimique. The intake fraction (iF) is used to calculate the potential human health burden of exposure to chemical emissions. It combines environmental persistence, direct and indirect exposure via bioaccumulation in the food chain. It takes the emission instead of the receptor perspective and is therefore a very useful tool in life cycle and comparative chemical screening assessments. However, it yields uncertainty from parameter uncertainty, spatial variability and model and scenario uncertainty. The objectives of this study are to i) carry an

13

importance analysis to determine the parameters that contribute the most to iF uncertainty and identify where to focus research efforts and ii)perform a seasonal scenario analysis on the most important parameters to determine the relevance of seasonal differentiation when assessing iFs. The organic chemicals included in Canada’s national toxic pollutant inventory and a set of 32 organic chemicals that represent the variability of physico-chemical properties were selected to calculate the contribution to iF uncertainty for emissions to air and water using the Canadian nonspatial version of the IMPACT 2002 fate and exposure model. The model includes 11 physico-chemical properties and 105 landscape and exposure parameters. Using importance and cluster analyses, it was determined that for emissions to air, the parameters that were the most important were rainfall rate, the chemical half-life in air and water, soil area and the octanol/water partition coefficient (K ). OW For emissions to water, the important parameters were Henry’s constant, the half-life in air and water and the K . The important parameters OW were then varied to define two scenarios: the summer and winter seasons. iFs for the set of 32 organic chemicals were calculated for both seasons for the Canadian and global context. For Canadian air emissions, results indicated that iFs for winter emissions could be up to one to two orders of magnitude higher than summer iFs (and model default average temperature, 25?C). For Canadian water emissions, results showed that iFs for both summer and winter conditions were, in general, closer to each other with outliers within one order of magnitude to iFs calculated at 25?C. Results also indicated that seasonal variability was of lesser importance when assessing iFs within a global context. Because the ranking between chemicals was maintained, it is concluded that seasonal variability is not relevant within a comparative context. 48 Towards an improved accuracy of Input / Output environmental assessments using a general equilibrium economic modelA. Some, CIRAIG, Ecole Polytechnique de Montreal; T. Dandres, CIRAIG; C. Gaudreault, R. Samson, CIRAIG, Ecole Polytechnique de Montreal. Input/Output tables describe monetary trades between sectors among a whole economy. Since the 90’s some of them have been included a sattelite matrix that links emitted substances to the economic sectors. I/O analyses are therefore able to model life cycle emissions of a major change occuring in a large scale system. These I/O analyses however have been used in case studies that aim only at a few economic sectors. The structure of the I/O tables yet relies on the interdependance among sectors. Studying a major economic change while changing only the demand for a few sectors may therefore not consider every affected sector and ultimately underestimate the impacts. GTAP[1] is a computable general equilibrium model that relies on national I/O tables to assess the variations due to a major change in the economy. It therefore seems interesting to couple economic data from GTAP with emissions from an I/O database to obtain a proper LCA for the whole economy of a given region. Recent developments in GTAP have also enabled to model direct and indirect land use changes and the related greenhouse gas emissions using IPCC conversion factors. In this study, GTAP model has been run for two scenarios: (1) a baseline scenario modeling rises in population and gross domestic product (GDP) by 2020 assuming a business as usual situation; (2) a biofuel scenario, assuming the same evolution of population and GDP but also including the policy of the USA and of the UE regarding biofuels by 2020.Economic production variations given by GTAP have then been coupled with the emissions of an I/O database. The impact assessment has been done using the Impact 2002+ method. Land use change impacts have been calculated coupling GTAP land use data and IPCC factors. The results show that both scenarios have almost the same economic impacts with the exception of agriculture where the biofuel scenario causes more land use change. Therefore environmental impacts are rather the same excepted for greenhouse gas emissions where the baseline scenario emits more than the biofuel one even when land use change is included. Using GTAP to calculate the economic data that are then coupled directly with environmental I/O tables enables a more complete LCA compared to classical I/O analysis. The inclusion of the emissions due to land use change considers the indirects impacts in the final results and therefore improves the completeness of the method.


SETAC Europe 23rd Annual Meeting

[1] Global Trade Analysis Project 49 Challenges of modelling the environmental exposure of nanopesticides S. Beulke, FERA / Food and Environmental Safety Programme; S. Monteiro, FERA; M. Kah, University of Vienna; K. Tiede, FERA; T. Hofmann, University of Vienna. The knowledge of the mechanisms controlling pesticide behaviour is extensive. Models incorporating these mechanisms can match the environmental exposure of conventional pesticides reasonably well. Such models are now well established tools to estimate the potential for pesticides to leach to groundwater and surface water within the regulatory framework. Standard experiments are performed to derive the model input parameters. These studies are usually only undertaken with the active substance and not with the formulated product. It is assumed that the intact formulation exists in the environment only for a short time and there is no effect on the long-term behaviour of the active. But this view has recently been challenged. Novel formulations of pesticides are continuously developed to enhance efficacy and optimise delivery to the target. Nano-pesticides have increased in importance as the potential of nano-materials in various disciplines has been recognised. Nanopesticides could potentially pose new risks to humans or the environment and it is unclear whether the current regulatory framework is adequate for the evaluation of these products. A literature review was carried out [1] to identify possible impacts of the properties on nanopesticides on environmental fate, and evaluate the suitability of current experimental protocols and exposure assessment procedures within the EU regulatory context. This was followed by experimental work and modelling. The sorption, degradation and leaching behaviour of free and nano-formulated active ingredients was investigated. The leaching of the nano-formulated pesticide was simulated using conventional input parameters and modelling tools and the results were compared with the experimental data. Alternative modelling techniques were explored. [1] Kah M, Tiede K, Beulke S, Hofmann T. 2012. Nano-pesticides: state of knowledge, environmental fate and exposure modeling. Critical Reviews of Environmental Science and Technology. DOI:10.1080/10643389.2012.671750. 50 Volatilization of pesticides from the bare soil surface- modelling of the humidity effect M. Schneider, Analytical Environmental Chemistry; K. Goss, Eawag / AUC. Volatilization of pesticides from soils under conditions dryer then the wilting point can be significantly influenced by sorption to hydrated mineral surfaces. This sorption process strongly depends on the hydration status of the mineral surfaces and their available surface area. Indeed established pesticide fate models considering volatilization do include the humidty effect by assuming an increase of sorption under conditions dryer then the wilting point. However they do not explicitly consider the hydrated mineral surfaces as an independent sorption compartment and therefore cannot correctly cover the moisture effect on volatilization. For this work we integrated our knowledge on the sorption of organic compounds to the mineral surfaces and its dependence on the humidity state of the soil into a simple volatilization model. The model considers availability and contribution of the mineral surfaces as a function of their hydration status, described via the equilibrium relative humidity in the pore space of the soil. The resulting model was tested with experimental volatilization data for the herbicide triallate from a bare soil surface in a wind tunnel experiment under various well defined humidity conditions. A sensitivity study was performed to estimate the influence of the different input parameters. The model captures the general trend of the volatilization under different humidity scenarios and shows the required degree of sensitivity to humidity changes in the soil. The results reveal that it is essential to have high quality input data for K the min/air , available specific surface area SSA, the penetration depth of the applied pesticide solution and the humidity conditions in the soil. The model approach presented here in combination with an improved description of the humidity conditions under dry conditions can be integrated into existing volatilization models that already work well for humid conditions but still lack the mechanistically based description of the volatilization process under dry conditions. However the rather accurate

14

information on the humidity state of the soil that were extracted here from the well defined wind tunnel experiments might be very challenging to gain from field situations. 51 Science Behind the Plant Uptake Factor in Leaching Models W. Schmitt, Bayer CropScience AG / Environmental Modelling; R. Sur, Bayer CropScience LP / Environmental Safety; G. Goerlitz, Bayer CropScience AG / Environmental Safety. Leaching models used to predict groundwater concentrations of pesticides and their metabolites consider uptake into plants as one dissipation process. For this the focus is on the disappearance of material from soil. Previous experimental investigations of uptake and distribution of chemicals in plants, on the other hand, primarily focused on the emergence of the substances in the plants because they were basically performed for understanding the emergence of plant residues or for estimating bioavailability in case of pesticides. For this reason quantitative measures are established describing absorption of substances by roots and the transport into shoots, i.e. root concentration factor (RCF) and transpiration stream concentration factor (TSCF). However, the plant uptake factor (PUF) implemented in leaching models was differently named and it was never specified how to measure or estimate its value. Unfortunately in some cases it was also called TSCF. That use of non-zero PUF values is increasingly challenged by regulatory agencies may be caused by uncertainties raised simply by missing or misleading definitions of terms used in model descriptions, guidances and publications. Since the transport by the transpiration stream through the plant is a chromatographic process, the concentration in this stream depends on location and time. While the temporal behavior is described in the literature, the spatial aspect was widely ignored so far. We applied a mechanistic biokinetic model for plants that describes uptake and distribution on the basis of physicochemical properties, i.e. logKow and molecular weight, for simulating the fate of the substances. Thus it was possible to systematically investigate the dependence of the different measures characterizing transport processes in the plant-soil system on different factors. One important result is that the TSCF and the PUF show opposing time dependence and merge only after considerable time, which depends on the lipophilicity of the substance. Moreover the simulations support the experimental finding that plant uptake factors are close to one, almost independent of the compound properties. It is concluded that the TSCF does not represent the PUF required for leaching calculations. Moreover, simulation results confirm recent experimental findings that the default value for the PUF of 0.5, mentioned in the current guidance, is conservative for a very high percentage of substances. 52 Risk Envelope Approach: Applicable for PEC groundwater calculations? D. Nickisch, Efate & Modelling; N. Seiterle-Winn, . The idea of the risk envelope approach as presented in SANCO/11244/2011 indicates that in the environmental fate area the supported uses of a product can be grouped taking into account certain criteria (e.g. crop, application rate). So the assessment can be targeted at the worst case group for a specific assessment covering all other supported uses. Estimation of the predicted environmental concentration in groundwater is driven by a variety of factors e.g. application rate, degradation rate or leaching behavior. Leaching into soil of a substance is a very complex process dependent on various factors related to soil type and weather conditions that are taken into account within the models. Thus options for a reasonable grouping based on the above mentioned parameters may be limited. Therefore, we tested an envelope approach for parent substances using the model FOCUS PEARL in two different ways: a) based on substance parameters and b) based on crop Substance based calculations: We assumed that degradation of a substance and its sorption behavior in combination with the application rate are the key drivers of leaching. We calculated several combinations of DT (10 – 100 d), K (50 – 500 L/kg) and application rates (10 – 50 OC 1000 g/ha) for the FOCUS scenario Kremsmünster. Crop based calculations: For the four FOCUS substances (A-D) we calculated all scenarios and crops as well as a bare soil scenario with the same application dates used for the single crops. The applicaton rate was set to

SETAC Europe 23rd Annual Meeting

1 x 1000 g/ha neglecting crop interception. Results showed that application to bare soil does not represent the worst case and that no worst case crop could be defined. Only trends could be shown, e.g. that crops like winter wheat are more conservative than others which can probably be traced back rather to application timing than to the crop scenario itself. Nevertheless, substances with a K > 500 L/kg were OC below 0.1 µg/L for every calculated combination. Based on our results it is questionable if a worst case crop or location can be defined. However, it seems reasonable that the influence of the K on the leaching OC process is strong enough to define a K criterion so that a time OC consuming risk assessments for substances above this trigger is not necessary. This assumption is supported by a literature review of EFSA conclusions from 2005 – 2012. 53 GERDA: Risk assessment for pesticide inputs into surface waters via surface runoff, erosion and drainage in GermanyM. Bach, University of Giessen; D. Grossmann, German Federal Environment Agency (UBA); D. Guerniche, RLP Agrosience; U. Hommen, Fraunhofer IME; M. Kaier, German Federal Environment Agency (UBA); M. Klein, Fraunhofer IME; R. Kubiak, RLP Agroscience; A. Mueller, German Federal Environment Agency (UBA); T.G. Preuss, RWTH Aachen University / Institute for Environmental Research; S. Reichenberger, Footways SAS / FOOTWAYS S.A.S.; K. Thomas, M. Trapp, Agroscience RLP. The regulation (EC) No 1107/2009 increases the need for harmonized rules and conditions for the authorisation of plant protection products (PPP). The German national registration procedure currently uses the model EXPOSIT 3.0 beta to evaluate the surface water exposure and risk from PPP input with runoff, erosion, and drainage. The German Federal Environment Agency (UBA) has launched a project to harmonise the German national exposure and risk assessment procedure for surface waters with the FOCUS surface water scenario approach. The new approach shall comprise the following elements. The models MACRO and PRZM are regarded as sufficiently accurate for the estimation of pesticide inputs into surface waters by runoff, erosion and drainage on the field scale. Since the inter-annual variability of the weather is by far greater than the variance between different climate scenarios it is essential to simulate at least 20-year longterm weather records. PRZM and MACRO simulation runs calculated long-term PECsw time series for numerous soil/climate/crop/substancecombinations in Germany. For each of these time series ecotoxicologically relevant exposure endpoints were derived from Monte Carlo analysis with a Toxicokinetic/Toxicodynamic (TKTD) model. The PECsw time series were grouped and ranked according to the calculated exposure endpoints to create area-weighted cumulative distribution functions for Germany. This approach defines a set of national soil/climate-combinations, which assure the worst-case scenario requirement for a substance with given properties. With the parameterisation for these scenarios PRZM and MARCO simulate the runoff, erosion, and drainage losses of the respective substance from the field. In tier 1 subsequently the model STEPS-1234 is used to calculate PEC time series in the receiving water body. A higher tier approach should respect elementary properties of stream systems such as losses from all fields hydrologically connected to the recieveing stream within a catchment as well as hydrodynamic and geomorphological dispersion. Further-more soil conservation measures should be implemented as appropriate risk mitigation options. The features and pre-defined national soil-climate scenarios will be imple-men-ted in a German runoff, erosion and drainage risk approach for PPP. The new approach shall harmonize the German authorisation with the established FOCUS procedure but allows to consider the specific agro-environmental conditions in Germany. 54 Soil/climate scenarios for modelling pesticide inputs into surface waters via surface runoff, erosion and drainage in Germany S. Reichenberger, Footways SAS / FOOTWAYS S.A.S.; M. Bach, University Gießen / Institute of Landscape Ecology and Resources Management; D. Grossmann, German Federal Environment Agency; D. Guerniche, RLP Agroscience / Institute of Agroecology; U. Hommen, Fraunhofer Institute of Molecular Biology and Applied Ecology; M.

15

Kaiser, German Federal Environment Agency; M. Klein, Fraunhofer Institute of Molecular Biology and Applied Ecology; R. Kubiak, RLP Agroscience / Institute of Agroecology; A. Mueller, German Federal Environment Agency; T. Preuss, RWTH Aachen University / Institute for Environmental Research; K. Thomas, M. Trapp, RLP Agroscience / Institute of Agroecology. The new EU regulation 1107/2009 requires a harmonization of the various national pesticide exposure and risk assessment approaches between EU member states. The German Federal Environment Agency (UBA) therefore launched a project (FKZ 3711 63 427) to harmonize the German national exposure and risk assessment procedure for surface waters with the FOCUS approach used at the EU level. Within the project the new national exposure assessment tool GERDA (German Erosion, Runoff and Drainage Assessment)is being developed, which will take into account the full range of agro-pedoclimatic conditions in Germany. The agricultural area of Germany (arable and permanent crops) has been classified into 12 climatic zones. Moreover, the German 1:1000000 soil map has been translated into FOOTPRINT Soil Types (FST). Climate zones and soil map were intersected and the areas of the resulting 1153 climate/FST combinations determined. In the next step, soil/climate combinations representing a realistic worst cases for Germany were identified. For this aim, 93312 long-term drainage simulations were performed with MACRO 5.2 for all 324 occurring combinations of climate and drained FSTs in Germany. Similarly, 415080 surface runoff and erosion simulations were performed with PRZM forall 1153 occurring FST/climate combinations. From these results long-term hourly time series of Predicted Environmental Concentrations in surface water (PECsw) were generated using the model STEP-3, for a standard surface water body. A Toxicokinetic/Toxicodynamic (TKTD) model was employed to estimate generic combinations of exposure endpoints which can be used to define realistic worst case exposure profiles for several compound/species combinations for all relevant species in aquatic risk assessment. The PECsw time series were grouped and ranked according to the values of the selected exposure endpoints, and area-weighted cumulative distribution functions (CDF) were generated. The obtained CDFs allow, based on the substance properties, crop and application pattern to be modelled, the identification of a set of soil/climate combinations with a defined worst-case-ness for the surface water risk assessment. This constitutes a major improvement in comparison to FOCUS, where representativity and worst-case-ness are only vaguely known. 55 World café interactive session R. Owen, . This session will start with a world café interactive session, in which also video clips of short vox pops will be showed, where people will describe their own thoughts on responsible innovation. This will be complemented with te results of a short questionnaireon Responsible Innovation 56 Adaptive Governance for Responsible Innovation J. Petts, . Emerging technologies create the possibility of novel materials, new life forms and opportunity not offered in nature; indeed deliberately intervening in the ‘natural’. But opportunity is accompanied not only by data and model uncertainty, but often by ambiguity as well as sheer ignorance. Further, innovation is a complex nonlinear and collective process enacted over varying but often very long timescales, involving multiple actors and at both local and global scales. Against this background, calls for a new and, importantly, adaptive framework for the governance of innovation have been emerging. The new governance framework implies a move away from a reliance on the top-down riskbased, regulatory approach to one that attempts to set the parameters of a system within which people and institutions behave. This paper will consider the components of this framework addressing important issues of responsibility and accountability and particularly the relative roles of soft law and cooperative approaches not least in the early stages of research and innovation. The discussion will consider the fundamental principles of effective governance and the necessary tools to ensure anticipation, reflection, openness and responsiveness. It will speak directly to the implications for science and scientists. 57 A Framework for Responsible Innovation

SETAC Europe 23rd Annual Meeting

R. Owen, . I will

describe a framework for responsible innovation developed for the UK Research Councils. I will first briefly describe the foundations and rationale for responsible innovation and important commitments emerging at an EU policy level. I will argue that the departure point and first challenge for responsible innovation is not ‘what are the risks?’ – important though this is - but ‘what kind of future do we want science and innovation to bring into the world?’ - and how this can be defined in a democratic, equitable and sustainable way. This will require reflection primarily on the very visions, purposes and underlying motivations for innovation and the values these are anchored in. Acknowledging the place and limits of both market choice and regulatory governance underpinned by risk characterisation, I will then argue that responsible innovation must also meet the challenge of how to proceed (innovate, conduct science) responsibly under conditions of ignorance and uncertainty. I will develop a framework around integrated dimensions of anticipation, reflection, deliberation and (importantly) responsiveness (personal, institutional, political) that attempts to address these challenges before highlighting practical experiments in the areas of environmental nanoscience and geoengineering. In doing so I will emphasise the collective nature of responsibility and the need to enlarge, reframe and even redefine role responsibilities, including those of the SETAC community itself. 58 Industrial discharges of pharmaceuticals induce adverse effects in wild fish O. Cardoso, INERIS / Unité d'ecotoxicologie in vitro et in vivo; O. Palluel, E. Chadili, INERIS; S. Betoulle, Univ Reims; W. Sremski, ONEMA; B. Piccini, J. Porcher, W. Sanchez, INERIS. Recent evidences reveal high concentrations of a large number of active pharmaceutical effluents (APIs) in effluents from pharmaceutical manufactures and in receiving aquatic ecosystems. Laboratory experiments reveal adverse effects in aquatic organisms exposed to these industrial effluents. However, no data is available on adverse effects induced by industrial pharmaceutical discharges in wild aquatic organisms. The aim of the present work was to assess wild fish health disturbance downstream from industrial effluents. For this purpose, a set of biochemical and histological biomarkers was measured in fish sampled upstream and downstream from several API manufactures selected among more than 200 industrial sites involved in French pharmaceutical production. Firstly, between 2008 and 2012, gudgeons were electrofished in four sites located upstream and downstream from the discharge of pharmaceutical factory involved in steroid synthesis. In all sampled fish, a set of biomarkers including endocrine disruption endpoint (vitellogenin and gonad histology) was measured. Sex-ratio and fish assemblage were analyzed to identify potential population disturbance. Strong signs of endocrine disruption including vitellogenin induction, intersex and male-biased sex-ratio were observed at several kilometres downstream from the industrial effluent. Secondly, to characterise the occurence of adverse effects induced by API industrial discharges in wild fish, a monitoring campaign was performed around 4 pharmaceutical factories including a manufacture of veterinary pharmaceuticals. First observations confirm that effluents from API manufactures induce adverse effects in wild fish from several locations. Results presented in this study confirm that effluents from pharmaceutical manufactures contain active compounds able to disturb the health of wild fish and populations. This conclusion argues for an increase of knowledge related to the occurence of these effects around the world and for the deployment of specific monitoring approaches around pharmaceutical factories This work was supported by the French Ministry for Ecology and Sustainable Development (MEDDE-Program 181). 59 Enantiomerism of medicinal products – a new paradigm in environmental risk assessment B. Kasprzyk-Hordern, University of Bath / Chemical and Biological Sciences; J.P. Bagnall, University of Bath; D.R. Baker, University of Huddersfield; S.E. Evans, University of Bath. Chiral pharmacologically active compounds (PACs) are environmental pollutants. They enter the environment mainly through insufficiently treated sewage, waste effluents from manufacturing processes, runoff and sludge. They are bioactive, ubiquitous and

16

persistent with synergistic properties. Surprisingly, the environmental fate and effects of PACs are assessed without taking into consideration their enantiomeric forms (this is despite existing knowledge on enantiomer dependant toxicity of PACs to humans, taking thalidomide as a flagship example). Such an approach leads to an underestimation of toxicity of PACs, incorrect environmental risk assessment, and direct risk to the environment and human health, as PACs are likely to be present in the environment in their non-racemic forms(in the case of single enantiomer PACs racemisation in-vivo can also take place). This presentation aims to discuss: The phenomenon of chirality in the context of fate and effect of chiral pharmacologically active compounds (PACs) in the environment Monitoring of several wastewater treatment plants and receiving waters in the UK to study stereoselective biodegradation of PACs Stereoselective biodegradation of PACs in river microcosms Implications of enantiomer-dependant fate of PACs for prodecures applied in environmental risk assessment 60 How does sewage effluent exposure affect the pharmacokinetics of non-steroidal anti-inflammatory drugs (NSAIDs) in fish?L. Gunnarsson, Goteborg University / Department of Infectious Diseases, Institute of Biomedicine; J. Fick, Umea University / Department of Chemistry; A. Grans, M. Axelsson, University of Gothenburg; J.D. Larsson, The Sahlgrenska Academy at the University of Gothe / Department of Infectious Diseases, Institute of Biomedicine. We have previously performed a meta-analysis comparing bioconcentration factors (BCFs) of several non-steroidal anti-inflammatory drugs (NSAIDs). This analysis suggests a considerably higher BCF in fish exposed to treated sewage effluents (containing NSAIDs and other chemicals) than in fish exposed in the laboratory to a single NSAID in pure water. Basing environmental risk assessments of NSAIDs on traditional exposure studies in the laboratory may therefore underestimate their bioconcentration potential and consequently their toxicity. The aim of this study is to disentangle if the observed differences in bioconcentration are due to differences in uptake, tissue distribution, metabolism and/or excretion. We have therefore performed pharmacokinetics studies with single-dose i.v. administered ketoprofen (1mg/kg) in rainbow trout either kept in clean water or exposed to treated sewage effluent. The dorsal aorta was cannulated to allow repeated sampling of blood and the i.v. route of exposure eliminates potential differences between the groups due to uptake. The results show that the plasma elimination half life (t ?) of ketoprofen in a fish kept 1/2 in clean water is more similar to the t ? of mammalian and bird 1/2 species than to the much longer half life observed in another poikilothermic vertebrate (Iguana iguana). The pharmacokinetic data from fish exposed to treated sewage effluent will be analyzed shortly. Liver homogenate from trout kept in either clean water or exposed to sewage effluent will be analyzed for its potential to metabolize ketoprofen. An increased understanding of how complex mixtures can affect pharmacokinetics and consequently the toxicity of pharmaceuticals has the potential to direct improvements of environmental risk assessment procedures of pharmaceuticals. 61 The Effects of Anti-Androgenic Pharmaceuticals on Oestrogen Induced Feminisation in Model Fish Species C. Green, J. Brian, Brunel University / Institute for the Environment; R. Williams, Centre for Ecology and Hydrology; M. Scholze, S. Jobling, Brunel University / Institute for the Environment. The widespread feminisation of wild fish inhabiting UK rivers, characterised by ova-testes intersex and abnormally high concentrations of vitellogenin in males, has long been linked to the steroid oestrogens. However, in recent years concern has also grown over the presence of widespread anti-androgenic activity in sewage effluents and it has been hypothesised that anti-androgens may significantly impact feminisation in both singular and co-exposure to steroid oestrogens. This study aimed to test this hypothesis through exposure of model fish species to environmentally relevant concentrations of the anti-androgenic pharmaceuticals bicalutamide and cyproterone acetate to assess feminisation endpoints including vitellogenin and intersex induction. Predictive hydrological modelling was used to assess their concentrations in UK river catchments

SETAC Europe 23rd Annual Meeting

following outflow from sewage treatment works using LowFlows 2000 WQX, which ranged from 0-149ng/L and 0-62ng/L of bicalutamide and cyproterone acetate respectively. Based on modelled data, fathead minnow (Pimephales promelas) were exposed to a mixture of bicalutamide (100ng/L) and cyproterone acetate (70ng/L) as a single treatment and in combination with environmentally relevant levels of steroid oestrogens: 17?-oestradiol (2.1ng/L), oestrone (9.6ng/L) and 17?ethinylestradiol (0.3ng/L) for 14 days to assess vitellogenin induction. In a second experiment, Japanese medaka(Oryzias latipes) were exposed from embryos for 98 days to mixtures of anti-androgens and oestrogens, both alone and in combination. Two sets of concentrations were investigated, representative of sewage effluent (bicalutamide 100ng/L; cyproterone acetate 70ng/L; steroid oestrogen mixture 18ng/L combined oestradiol equivalent EEQ) and river concentrations (bicalutamide 47ng/L; cyproterone acetate 30ng/L; steroid oestrogen mixture at 8ng/L EEQ), to assess the impact on intersex. Preliminary data has found no evidence of a siginificant effect of the anti-androgenic pharmaceuticals bicalutamide and cyproterone acetate on either vitellogenin or intersex induction in the two model fish species at environmentally relevant concentrations alone or in combination with steroid oestrogens. Further analysis of ovarian cavities and fibrotic tissue induced in male gonads as well as the intersex severity under each treatment is required. 62 Predicting the impact of endocrine disrupters on the environmental health: an ecosystem model. L. Clouzot, Universite Laval / Département de génie civil et de génie des eaux; M. Paterson, A. Dupuis, P. Blanchfield, M. Rennie, Fisheries and Oceans Canada; K. Kidd, Canadian River Institute; P.A. Vanrolleghem, modelEAU Université Laval / Département de génie civil et de génie des eaux. Literature indicates an increasing concern about the release of pharmaceuticals, especially endocrine disrupters, through the effluent of wastewater treatment plants. Endocrine disruption has been measured in aquatic environments across the world. However, experimental approaches to characterize the ecological impact are costly and timeconsuming and thus, the consequences of endocrine disruption on whole ecosystems remain unclear. Mechanistic models can help understand the impact of such contaminants on aquatic environments and assess their ecological risk. This study takes up the challenge to develop an ecosystem model that considers endocrine disruption in fish and the consequences on the whole ecosystem through ecological interactions, i.e feeding and competition. The experimental data used to develop the model come from a multi-year whole-ecosystem study performed at the Experimental Lake Area (ON, Canada): (i) two years of reference data (ii) three years of exposure to environmentally-relevant concentrations of the synthetic hormone 17?-ethinylestradiol (EE2) and (iii) five years of recovery. EE2 is widely used in human contraceptives and it was chosen for this study because it is one of the most widespread and potent endocrine disrupters. During the experimental study, endocrine disruption was observed in the fish species with a collapse of fathead minnow after the second year of EE2 addition. The ecoystem model that is being developed is an object-oriented model based on simplified AQUATOX equations for the species naturally present in the experimental lake (benthic invertebrates, phyto- and zooplankton, fish). The novelty of the study is to add appropriate equations for endocrine disruption. Two fish classes are used in the model: juveniles and adults. The developed model can simulate endocrine disruption based on (i) an increase of gamete mortality, (ii) a decrease of gamete production or (iii) an increase of fish mortality. For example, if the sex ratio is different from 50:50 (female:male) gamete production will decrease, less juveniles will be recruited and thus, the fish population will decrease. In addition to modeling the direct effects of EE2 on fish, the ecosystem model also considers the lake dynamics. For example, two stratified layers are modeled (epilimnion and hypolimnion) with different biological and physico-chemical compositions. 63 Bioaccumulation potential of pharmaceuticals in the aquatic food chain - pseudo-persistence versus lipophilicity A. Zenker, University of Appl Sc Northwestern Switzerland / Institute of Ecopreneurship; S. Joerg, University of Applied Sciences and Arts

17

Northwestern Switzerland / School of Life Sciences. Even with the best technology currently available, waste water treatment plants (WWTPs) cannot completely remove micropollutants, in particular drug residues. Pharmaceuticals have therefore been continuously discharged to water bodies worldwide, leading to high concentrations of active ingredients and their transformation products in surface waters downstream of WWTPs. On the other hand to deduce mechanisms of toxic action or to evaluate the toxicity of mixtures, and to interpret field data on bioaccumulated toxicants the tissue residue approach has been taken into account most recently. Therefore, we have focused in our study on a multi-residue analysis of ionic and lipophilic pharmaceuticals from different trophic levels of the aquatic food chain (fish prey, fish, and fish eating birds). Determined pharmaceuticals were the frequently detected ibuprofen (an analgetic), diclofenac (an analgetic), mefenamic acid (an anti-inflammatory drug), sulfamethoxazole (a bacteriostatic antibiotic), atenolol (a beta blocker), diphenhydramine (an antihistamine), diltiazem (a calcium channel blocker), carbamazepine (an anticonvulsant), fluoxetine (an antidepressant) and its main metabolite norfluoxetine. Since our preliminary results have revealed exclusively residues of norfluoxetine, fluoxetine, diphenhydramine and carbamazepine in fish and fish prey the list of analytes were extended to other antidepressants (sertraline and citalopram) but also pharmaceuticals like gemfibrozil, fenofibrate and naproxen and metabolites of diclofenac (2-[2(chlorophenyl)amino]benzaldehyde) and ibuprofen (2-[4-(2-hydroxy-2methylpropyl)phenyl]propionic acid), which have been in parts detected in surface water. Trace analysis of compounds was performed in a multiresidue method based on HPLC coupled to mass spectrometry (LC-MS) whereby identification and quantification of pharmaceuticals in biota were arranged in MS/MS mode. MRM transitions were classified in a different elution time window to increase measurement sensitivity. Eleven out of 22 pharmaceuticals measured, as well as their transformation products, could be detected in biota samples. The antidepressant fluoxetine, its main metabolite norfluoxetine and diphenhydramine were predominant. 64 Evaluating effects of a multi-generation pollution on Caenorhabditis elegans’ populationB. Goussen, INERIS / Unit METO; R. Beaudouin, INERIS; F. Parisot, M. Dutilleul, A. BuissetGoussen, J. Bonzom, IRSN / PRP-ENV/SERIS/LECO; A.R. Pery, INERIS / Unit METO. The assessment of toxic effects at biologically and ecologically relevant scales is an important challenge in ecosystem protection. Indeed, in most time, stressors impact populations over longterm. The selection pressure exerted by a pollutant is known to amplify the phenomenon of natural selection and could lead to evolutionary changes across generations. It is therefore important to study the evolutionary response of a population submitted to a long term stress. Regarding this background, we assessed the evolution of two populations (control and exposed to 1.1 mM of the heavy radiotoxic metal, uranium) of the ubiquitous nematode Caenorhabditis elegans submitted to a long-term exposure to uranium. The experimentation was conducted over 16 generations and life history traits (growth, reproduction and survival) as well as dose-response evolution were assessed. These parameters were followed daily on individuals extracted from the populations and exposed to a range of concentration (from 0 to 1.2 mM U). Our experiment showed an increase of adverse effects as a function of uranium concentration. Indeed the NOEC for reproduction and growth traits were respectively of 0.5 mM U and 0.9 mM U. Moreover, reproduction and growth were respectively reduced by over 60% and 20% for individual exposed at 1.1 mM U. This reduction remained constant throughout the generations. We also pointed out the appearance of genetics differentiations on reproduction traits throughout the generations. This differentiation, observed from generation 3, showed us that the total egg-laying of the uranium population was significantly decreased compared with the control population. In contrast, no differentiations were highlighted on growth traits. Our results confirm the importance of studying environmental risks related to pollutant through multi-generational studies in order to capture effects that may appear after several generation of exposition.

SETAC Europe 23rd Annual Meeting

65 Multi-generational exposure of Folsomia candida to Cd – organisms size distribution, reproduction and gene expression C.M. Pereira, S.I. Gomes, A.M. Soares, M.J. Amorim, University of Aveiro / Departement of Biology. Because contamination can be historical or due to long term scenarios organisms may be exposed throughout many generations. Due to the required efford, few studies are available on multi-generational effects. The aim of this study was to assess the effects of multigenerational exposure of Folsomia candida to Cadmium [reproduction EC10 and EC50]. Results are presented untill F35, to our knowledge no study has been performed along so many generations in soil invertebrates. Measured endpoints were survival, reproduction, size and the expression level of the metallothionein coding gene. Further, toxicity was monitored at generations 6, 10, 26 and 34. Continuous exposure to Cd (EC10 and EC50) caused an improvement in reproduction (number of juveniles) performance until F6, after which it started to decline and for EC10 exposure failed at the 13th generation, whereas for the EC50 exposure it continued at lower performance. Changes in reproduction occur simultaneously with changes in the size of produced juveniles. The organisms seem to be responding via Mt gene expression, which is concentration and generation dependent. Organisms are acclimated to the stress but do not seem to become resistant, given that once in clean soil the Mt gene is less expressed. Curiously, the higher tolerance observed in the population exposed to the EC50 compared to the EC10 could be related to the fact that higher Mt levels confered larger detoxification and tolerance/longevity to Cd exposure. Effects also indicate that it is not possible to extrapolate effects from generation to generation which could be important to consider in a risk assessment framework. Keywords: adaptation dynamic, Population effects; Collembola. 66 Linking DNA damage and effects on life history in a multigenerational exposure of Daphnia magna to uraniumD. Plaire, IRSN; J. bourdineaud, Université Bordeaux 1; L. GarciaSanchez, IRSN; J. Poggiale, Universite de la Mediterranee; F. Alonzo, . Chronic effects of waterborne depleted uranium (U) were previously studied under laboratory conditions on successive generations of Daphnia magna. Results showed that U effects on physiology and life history increase across generations. Observed reduction in somatic growth and reproduction was analysed using the DEBtox approach suggesting that U primarily affects assimilation. This mode of action was confirmed by measurements of assimilation reduction and observations of histological alteration of the digestive epithelium. However the mechanisms involved in the transgenerational increase in sensitivity remained unknown, results pointing the egg as a potentially sensitive life stage. The present study aimed to evaluate how U alters DNA and affects physiology and life history of Daphnia magna. We investigated chronic effects of U on two successive generations, with different objectives: 1) to obtain a better estimation of DEB parameters -1 at concentrations lower than 10 ug L 2) to evaluate how exposure to U during the egg stage may explain the increase in severity observed between a F0 generation (exposed from the neonate stage) and a F1 generation (exposed from the egg stage). 3) to determine concentrations at which DNA damage become significant with a special emphasis on linking perturbations between the molecular and the individual levels. 4) to elucidate the mechanism causing the increase in effect severity over successive generations. Daphnids were exposed to waterborne uranium at concentrations ranging from 2 to 50 ug L 67 Long term effects of an early exposure to PAHs on zebrafish behavioural responses C. Vignet, IFREMER; K. LE MENACH, LPTC; S. PEAN, iIFREMER; L. LYPHOUT, D. LEGUAY, M. BEGOUT, IFREMER; X. COUSIN, INRA. Polycyclic aromatic hydrocarbons (PAH) emission in the environment is constantly increasing with human activity. In this study, zebrafish embryos were exposed to a mixture of three PAH molecules dosed at environmental concentrations based on values measured in the Seine Estuary (France) a highly polluted site. Eggs were collected from AB zebrafish strain and were placed at 4 hours post fertilization (hpf) until 96 hph in a 3 cm diameter Petri dish containing 3 g reference sediment either plain

18

(control) or spiked with Benzo[a]pyrene, Pyrene and Phenanthrene at concentration 600 ng/g, 1600 ng/g and 2100 ng/g respectively. Thereafter, larvae were transferred in clean water and raised until adulthood. Behavioural tests like locomotion during 24-hrs, dark sudden change challenge and T-maze were performed at adults’ stage (F0) and sudden dark change on larvae from F0 adults. In adults, during night period, contaminated fish were significantly less active than control fish. Contaminated F1 larvae were more active when light was on than control and less active when the light was off. Contamination using environmentally relevant concentration during the very first stage of development with a PAH mixture of 3 molecules on zebrafish induced behavioural effects measured at the adult stage. Further, effect was transmitted to the next generation and behavioural responses of F1 larvae were different to that of F1 control larvae. Keyword: fish, locomotion activity, oil 68 Co-tolerance of oxidative stress in Orchesella cincta (Collembola) populations genetically adapted to heavy metal stressD. Roelofs, Vrije Universiteit / Inst. of Ecological Science; T. Brand, VU University / Inst. of Ecological Science; N.M. van Straalen, Vrije Universiteit Amsterdam / Inst. of Ecological Science. Field-selected tolerance to heavy metals has been reported for Orchesella cincta populations occurring at metal-contaminated mining sites. This tolerance is correlated with heritable increase of metal excretion efficiency and overexpression of the metallothionein gene. A transcriptomicstudy showed that population-specific responses were identified for genes involved in 2+ carbohydrate metabolic processes, Ca dependent stress signaling and cuticle synthesis. It is well known that heavy metals such as Cadmium exert oxidative stress. Here we asked the question whether metaltolerant O. cincta populations exert co-tolerance to oxidative stress. To that end, we exposed metal-tolerant and reference O. cincta populations to paraquat, which induces oxidative stress. Survival during paraquat exposure confirmed that the resistance of the tolerant populations to stress caused by cadmium exposure, contributed to increased tolerance against oxidative stress induced by paraquat. A number of relevant genes were selected to study differential expression due to paraquat at the genetic level. Metallothionein and CYP450s showed highly similar transcriptional responses when paraquat exposed animals were compared to Cd exposed animals. However, genes involved in stressactivated protein kinase signalling differently regulated between paraquat exposed andimals and Cd-exposed animals. We conclude that tolerance to oxidative stress and heavy metals is very similar at the phnotypic level, but deviates significantly at the transcriptional level. Furthermore, the data suggest that tolerant phenotypes exert a broader range tolerance to withstand additional stressors. 69 Phenotypic evolution in response to chemical dose S. Gerber, Eawag; F. Pomati, Aquatic Ecology. The resilience of biological systems to stress depends on the capacity of constituting organisms to adapt and evolve. Aquatic organisms are increasingly exposed to mixture of emerging contaminants. Natural populations may acclimate physiologically or evolve resistance in response to this novel pollution. Which levels of exposure result in evolution of resistance traits and whether physiological acclimation plays a leading role is not well understood. While it appears that resistance to anthropogenic pollutants in microalgae and cyanobacteria arises at lethal exposure levels as a result of selection on standing genetic diversity, phytoplankton are able to survive in adverse habitats as a result of physiological acclimation, which is supported by generally plastic phenotypes. Here we investigated the dose-dependency of changes in population abundance and phenotypic traits in toxic and non-toxic Microcystis aeruginosa (cyanobacteria) strains as a response to a biocide of emerging concern, triclosan. We explored if induced phenotypic change corresponds to an evolutionary adaptation and therefore was maintained across generations, and whether trait changes could be reverted to the original phenotype after a period of detoxification and recovery. The non-toxic mutant appeared to be more resistant to the chemical and showed an overall different phenotype, with larger cell size. Significant observable changes in cellular phenotype occurred for the wild type (WT) at IC50

SETAC Europe 23rd Annual Meeting

levels and above. Stressed WT cells were phenotypically more similar to unstressed mutant cultures than to control cultures of their own strain. The non-toxic mutant showed less phenotypic plasticity and a phenotype resembling the stressed one in the WT. Changes in the WT phenotype, acquired after exposure to triclosan, were maintained after cultures were released from chemical stress, suggesting a form of phenotypic inheritance after few generations such as epigenetic effects. We provide original findings that advocate for a more careful consideration of sublethal doses and environmentally relevant exposure scenarios in the risk assessment of chemicals, since physiological and evolutionary processes, e.g. phenotypic transitions, can interact and occur at low levels of exposure to anthropogenic chemicals. 70 Simulating fate of pesticides and potential mortalities of Asellus aquaticus in the water network of an agricultural landscape M. Ter Horst, Alterra; A. Focks, Wageningen UR / Mathematics/Computer Science; E. van den Berg, Alterra. For aquatic ecological risk assessment possible ecotoxicological effects on aquatic organisms are considered on an edge of field scale in the lower tiers of the European pesticide regulation framework. Upscaling exposure simulations to the landscape scale improves the realism of the risk assessments. The CASCADE model has been developed to assess the fate of pesticides in the water network at the landscape scale. The pesticide fate module of CASCADE is based on the TOXSWA model, that calculates the behaviour of a pesticide in a single watercourse and includes the processes of transport, transformation, volatilisation and sorption. The model couples single watercourses into a system of communicating watercourses. At junctions all incoming pesticide mass fluxes are distributed proportionally between the outgoing water fluxes. The present research version model simulates fate processes in the water layer only; accordingly there is no interaction with sediment. 2 CASCADE was parameterised for a 10 km Dutch polder area, the Drentse Veenkoloniën. Exposure calculations were done for 15 applications of 0.005 kg/ha lambda-cyhalothrin in seed potatoes, starting st from May 1 with intervals of 7 days. Spray drift was the only entry route of pesticide in the water courses and was fixed to 5%. Only part of the water courses in the catchment received spray drift. The half-life in water was set to 1 d. Daily maximum concentrations per numerical compartment were coupled to a dose response relation to calculate the potential for chemical induced mortality of Asellus aquaticus. The dose response relation used an LC50 value of 48 ng/L. It was found that decrease and increase of mortality corresponded to respectively decrease and increase in concentration. The effect of either polluted or non polluted tribituaries on pesticide concentrations in water and the corresponding mortalities were clearly visible. Despite the fast dissipation in the water layer, pesticide applications caused mortality at the outlet in two km distance from the application area. On a landscape scale the connections of the water courses within the catchment and the pesticide application pattern on a spatial scale are important additional drivers next to those already known for edge of field scales (i.e. toxicity, water volume, hydraulic residence time, dissipation). The spatiotemporal explicit exposure patterns could directly be used as input for population-level effect modelling. 71 Evaluation of a spatially explicit model for predicting exposure of bats to soil-associated metals B.V. Hernout, University of York / Environment; A. Boxall, University of York / Environment Department. Bats may be exposed to metals by consuming insects that have accumulated soil-associated metals. Food chain models can be used to estimate potential risks of metals to bat health. A recent modelling framework has been developed to predict metal risks to bats in England. However, due to a lack of data, this model has not been evaluated against experimental measurements. This study was therefore performed to develop data on concentrations of metals in soils, soil invertebrate and bat samples in order to test different steps in the modelling framework. While correlations between model outputs and measurements were obtained for some components of the model and some metals, for the majority of metals and model steps, no correlations were obtained. These results point to the fact that we need a much better understand of

19

spatial metal concentrations, metals uptake into insect orders and bioaccessibility and regulation in the bat in order to adequately predict risks. 72 Validation of mechanistic effect models for ecological risk assessments: A practical approach J. Augusiak, Wageningen UR / Environmental Sciences - AEW; V. Grimm, Helmholtz Centre for Environmental Research UFZ / Department of. Ecological Modeling; P.J. van den Brink, Alterra and Wageningen University. Common European environmental risk assessments for pesticides focus on standardized procedures that inherently induce limitations to a full understanding of adverse effects on non-target species populations and ecosystems. Ecological effect models have long been identified as useful tools to extrapolate limited experimental findings to more realistic conditions, such as larger temporal or spatial scales, or higher levels of biological organization, which can make a risk assessment ecologically more relevant. However, ecological effect models are not often used or accepted in regulatory risk assessments due to doubts about whether a given model represents the real world sufficiently well. Overall model credibility and a lack of validation have been raised as causes of doubt from regulators. Additionally, usually different educational backgrounds and levels of experience among involved stakeholder groups often induce time-demanding discussions concerning functional terminology and methodology. In order to support a better understanding of applied modelling and model evaluation techniques, we reviewed a number of publications that addressed model quality assessment and validation efforts across various scientific fields with a close connection to policy and decision making. We found that some terms were used with rather strong consistency whilst others caused conflict and were used interchangeably with varyingly broad background considerations. Aside from identifying a suitable system of terminology, we classified several methods that can be employed for a majority of existing modelling approaches. The classification was based on the degree to which a given method addresses underlying uncertainties but also how generally applicable the method is to different modelling approaches, e.g. stochastic vs deterministic modelling We found, that in order to introduce ecological effect models as valuable decision support tools, particularly more complex models such as spatially explicit ones, more comprehensive quality assessment tools and protocols need to be developed and established. Such protocols should contain a glossary to enable an easier communication between involved stakeholder groups. Additionally, following a concise documentation scheme that includes explicit descriptions of model uncertainties and applied evaluation methods would facilitate a ready illustration of how a given model’s output information relate to the question addressed by the model. 73 An individual-based vertical distribution model of Folsomia candida (Collembola) in an agricultural soil columnV. Roeben, RWTH Aachen University Institute for Environmental Research / Environmental Safety; F. Scherr, Bayer CropScience AG; G. Goerlitz, Bayer CropScience AG / Environmental Safety. Folsomia candida is an ubiquitary soil inhabitant often referred to as the “Standard Soil Arthropod” and is part of the regulatory framework of pesticide risk assessment. With respect to risk assessment it is crucial to know where and when to protect an organism but still little is known on the vertical dispersal and seasonal fluctuations of collembolan communities in agricultural landscapes. To overcome this lack of knowledge, ecological modelling offers a powerful tool. This work presents a vertical distribution model of the soil-dwelling collembolan Folsomia candida in an agricultural soil column to demonstrate the effect of variations in environmental parameters on the population. In addition, the model features the option to evaluate the effect of a pesticide application. The model is implemented in the modelling environment of NetLogo and is constructed based on literature data on the biology and life-history trait of F.candida. It represents a vertical soil column with heterogeneous layers of 1 cm. The environmental parameters and concentrations of pesticide in the soil column are supplied through outputs of PEARL calculations. The movement patterns of F. candida observed in laboratory studies are reproduced by the model. Temperature and

SETAC Europe 23rd Annual Meeting

organic matter content proved to have a major influence on the movement and therefore on the vertical distribution of the population. In the model the population shows response to a pesticide application considering the population density and the reproductive success. On the basis of current literature, the model is able to simulate the possible spatial and temporal distribution of a Folsomia candida population in an agricultural field and emphasizes the effect of fluctuating environmental parameters, such as temperature on the population. In the next step the model will be validated and verified by field studies and/or terrestrial ecosystem models (TMEs). 74 Consequences of choice of landscape scale for a spatial modelling ERA C.J. Topping, bioscience; W. Schmitt, Bayer CropScience AG / Environmental Modelling; M. Ebeling, Environmental Safety; T. Schad, Bayer CropScience AG; G. Goerlitz, Bayer CropScience AG / Environmental Safety. The use of modelling for environmental risk assessment is becoming more commonplace, but this raises questions of exactly how to best use these models. There is also a greater awareness that spatial and temporal scale are important in determining the impacts of toxicants in the environment. This awareness comes with an increasing cost of complexity in the risk assessment. Here we address the question of the consequence of landscape scale and structure on a spatial modelling ERA for a small mammal non-target species. The analysis was carried out using the field vole (Microtus agrestis) model of the ALMaSS, a simulation system designed to evaluate the impact of management on wildlife populations. Danish 10 x 10 km landscapes were simulated with orchards covering 2.5% of the total landscape area. These orchards were subject to a single yearly pesticide application. To ensure a clear effect, the ficticious pesticide was assumed to have either a direct lethal effect above a NOEC for all voles exposed, or a severe endocrine disruption effect. A number of scenarios were evaluated assuming different landscape configurations with both pesticide types. A spatially structured ERA was carried out by considering the impact of the pesticide against a baseline where no pesticide was applied, and making this evaluation on the whole landscape, then subdividing the landscape into 4, 16, and 64 equal sized grid squares and repeating the analysis per grid square. Different topographies predicted different impacts and recovery with an interaction between pesticide type and landscape structure. When carrying out the spatially structured ERA, the size of the landscape grid evaluated had a clear impact on the variability of the result. Smaller scale landscapes increased the variability around the predicted impact. This variation showed that the conclusion reached by an ERA could be very misleading if only a single small area was assessed. There was no one simple landscape metric that could explain this variation, and general rules may be difficult to find due to differential species ecology and behaviour. In conclusion, the larger the landscape scale the lower the variability of the ERA result. If simple landscapes are used then it is important to consider a wide range of spatial scenarios to avoid bias. If large scale landscapes are used then this requires a re-focussing of the ERA to the landscape level population and greater ecological realism. 75 Effects of a non-mitigated exposure to a mixture of pesticides used in a fruit and tuber crop on the recovery of aquatic invertebrates. P.J. van den Brink, Alterra and Wageningen University; M. Zorn, Board for the Authorisation of Plant Protection Products and Biocides (Ctgb); T. Brock, Alterra; E. Roex, DELTARES; T. Van der Linden, RIVM, National Institute for Public Health and the Environment; R. Luttik; A. Focks, Wageningen UR / Mathematics/Computer Science. The risk assessment of pesticides in the legislative process is based on active substance by active substance or preparation by preparation. In addition the risk analysis may be based not only on the expected ecological effect but also on the potential for ecological recovery. However, implementing the recovery principle in the aquatic risk assessment of individual compounds could be unrealistic if in most crops pesticides are repeatedly applied during the growing season or multiple pesticides are applied at the same time. The aim of the conducted work was to assess whether recovery times increase when a population of a vulnerable aquatic invertebrate is exposed to the

20

exposure profile of one pesticide or to exposure profiles of multiple pesticides. The evaluated two sets of mixture of pesticides is representative for pesticides packages that are used on a fruit and tuber crop in The Netherlands. Exposure concentrations were predicted using the FOCUS step three modelling framework and the Dutch drainage ditch scenario. Recovery times were assessed using the MASTEP individual based population model. MASTEP is an individual-based model where Asellids are individually simulated, and some of the most basic model parameters such as time of proliferation and natural mortality are varied stochastically between the individuals. We simulated the population dynamics and pesticide effects in a Monte Carlo style, by varying the sensitivities of the simulated population by generating random LC50 values from the arthropod species sensitivity distribution of the respective pesticide. For the tuber scenario lambdacyhalothrin resulted in long-term effects, while fluazinam hardly resulted in effects. This absence of interaction of effects of the mixture of chemicals used in the tuber scenario on effect size and recovery is also shown by the distribution of the recovery times. In the fruit scenario the applications of all three chemicals alone already resulted in large effect sizes just after exposure The recovery times, however, are not higher for the mixture compared to those associated with exposure to the individual compounds The conclusion is that exposure to the evaluated mixtures of chemicals that are representative for the tuber and fruit scenarios does not lead to longer recovery times than when exposed to the individual compounds alone. 76 Dynamic modeling of toxicokinetics of short- and long-term pulsed copper in tilapia W. Chen, National Taiwan University / Dept Bioenviron Sys Eng; J. Tsai, China Medical University / Graduate Institute of Ecology and Evolutionary Biology; C. Liao, National Taiwan University / Department of Bioenvironmental Systems Engineering. The variations in toxicant concentration can generate either pulsed or fluctuating exposures in aquatic ecosystems. Aquatic organisms living in aquatic systems would positively experience pulsed/fluctuating contaminant exposures. Moreover, the organ-specific copper (Cu) accumulations in fish are critical indicators for Cu exposure. The purpose of this study was to combine mechanistic and statistical models-based data from short- and long-term pulsed Cu accumulation of tilapia (Oreochromis mossambicus) to examine the organ specific toxicokinetics. The accumulation, distribution dynamics, and correlation between organs Cu burden and that of whole body were also investigated. Results found that the majority of organs in response to short-term pulsed exposure had higher bioconcentration factor than those of long-term pulsed exposure, expect blood, muscle, and carcass. The pulsed Cu burdens estimated by the physiologically-based pharmacokinetic (PBPK) models were consistent with the experimental data. The experimental results indicated that muscle plays the main distributive dynamics of Cu burden (34.63–50.38%) in the initial exposure time, whereas liver is the most contribution tissue (18.77–64.83%) during the exposure periods. Based on the mutlivariateanalyses, this study found the similarity of organ accumulation property for pulsed or non-pulsed periods in tilapia occurred in acute pulsed Cu exposure. The positive correlations were found between Cu burdens of whole body and liver in acute (r=0.748) and chronic exposures (r=0.607). This study provided a first stage that the integrated models contribute to better realizing the fundamental toxicokinetic processes. It may also provide the opportunity to construct a preferable understanding of dynamics relationship between organ Cu concentration and adverse effect for biomonitroting and optimizing ecotoxicological risk assessment in the realistic situations. 77 The role of feeding habits in controlling Hg bioaccumulation and biomagnification in freshwater tilapia R. WANG, College of Environmental Science and Technology; W. Wang, Hong Kong University Sci Technol. The bioaccumulation and biomagnification of mercury (Hg), especially methylmercury (MeHg), is commonly found in aquatic organisms. The mercury level in fish is not only related to its surrounding water conditions but also the feeding habit as well, and predatory fish always has high Hg level because MeHg could be

SETAC Europe 23rd Annual Meeting

biomagnified through trophic transfer. Whether the biomagnification potential (in terms of trophic transfer factor, TTF) could be influenced by feeding habit is an intriguing but unknown question. To explore this issue, we conducted a 40-days Hg (both inorganic and organic form) accumulation experiment, by feeding freshwater tilapia with three types of Hg labeled food (aquatic grass, freshwater shrimps, commercial pellets) at a certain ingestion rate. The results showed great differences between inorganic Hg and MeHg accumulation patterns. Interestingly, the determined TTF varied significantly with food type for Hg(II), but not for MeHg, which could be well explained by variation of assimilation efficiency (one important parameter in the energetic-based biokinetic model, could be determined by using radioisotope technique). Moreover, a three-month field study was designed to investigate the influences of feeding condition to Hg accumulation process in tilapia under local conditions, by comparing the growth and Hg accumulation process of fish feeding on additional commercial food and natural food. These results again showed the important role of feeding habit in controlling the Hg bioaccumulation and biomagnification process in fish. 78 A benchmarking method to measure dietary absorption efficiency of chemicals by fish R. Xiao, Stockholm University / Applied Environmental Science; M. Adolfsson-Erici; M.S. McLachlan, Stockholm University; M. MacLeod, ITM Stockhom University / Department of applied environmental science. Understanding dietary absorption efficiency (E ) for different organic chemicals in fish is of D critical importance from both a scientific and a regulatory point of view. However, reported fish E for well-studied chemicals are highly D variable. We applied an internal chemical benchmarking method to measure E of 15 chemicals. 2,2’,5,6’-tetrachlorobiphenyl (PCB53) and D decabromodiphenyl ethane (DBDPE) were selected as an absorbable and a non-absorbable benchmark chemical to reduce the uncertainty and variability in each individual fish response and to correct for bias that results from incomplete collection of feces from different tests. Due to the lack of standard E measurement, the performance of the D benchmarking method was evaluated in recovery statistics of the test chemicals. Our results showed that, after benchmarking, the recovery statistics of the test chemicals was significantly increased and variability of recovery of the test chemical was reduced. The benchmarking method has considerable promise to reduce variability in measurments of fish E . \n D

79 Matching Metal Pollution with Bioavailability, Bioaccumulation and Biomarkers Response in Fish Resident in Neotropical Estuaries I. Souza; I. Duarte, Universidade Federal do Espirito Santo / Departamento de Ciências Biológicas; M.N. Fernandes, Univeridade Federal de Sao Carlos / Ciencias Fisiologicas; J.B. Fernandes, Universidade Federal de São Carlos / Quimica; M. Bonomo, Universidade Federal do Espirito Santo / Departamento de Ciências Biológicas; V.C. Cavicchioli Azevedo, Instituto de Ciências Biológicas; N. Pimentel, Univeridade Federal de Sao Carlos / Departamento de Ciências Fisiológicas; C. Pereira, Universidade de Santa Cecilia; M. Monferran, Universidad Nacional de Cordoba; S. Matsumoto, Universidade Federal do Espirito Santo; D.A. Wunderlin, Universidad Nacional de Cordoba; L. Rocha, Universidade Federal do Espirito Santo / Departamento de Ciências Biológicas; M. Morozesk, Universidade Federal do Espirito Santo. Two neotropical estuaries, affected by different anthropogenic factors, were studied. We report levels of metals in water and sediment, spatial and temporal changes as well as their influence on genetic, biochemical and morphological biomarkers in the native fish Centropomus parallelus. Multivariate data treatment (discriminant and procrustes analyses) showed both spatial and temporal differences between studied areas throughout studied matrixes (water, sediment and fish). High concentrations of metals in sediments did not match metal bioaccumulation in fish. Conversely, we observed better correlation between intersticial water and levels of metals in fish. Morphological and enzymatic biomarkers reflected the condition of fish but genetic biomarkers did not. So far, current results highlight the need of an integrated approach to evaluate the influence of estuary pollution on native fish, considering swimming and feeding habits. Multivariate

21

statistics present a quite novel approach to integrate data from different origin allocated in the same estuary. Financial support: Prefeitura de Vitória/ES, FAPESP, CNPq 80 Temporal changes in Stable Isotope Analysis in Baltic herring (Clupea harengus): an explanation for observed PCDD/F concentrations? A. Miller, Naturhistoriska Riksmuseet / Contaminant Research Department; J. Hedman, Policy Instruments for Natural Resources, Swedish Environmental Protection Agency; A. Bignert, Swedish Museum of Natural History; M. Kiljunen, University of Jyvaskyla; J. Duberg, E. Gorokhova, Stockholm University. Extensive measures have been taken to reduce dibenzo–p–dioxin (PCDD) and dibenzofuran (PCDF) emissions within the EU. However, despite continual decreases in dioxin air emissions and the Baltic Sea abiotic environment (water and sediments), temporal trends in PCDD/Fs (l.w. and w.w.) in herring from various sites within the Baltic Sea region have shown inconsistencies in decreases or no significant decreases. This suggests that other factor(s) may be preventing herring PCDD/F concentrations from responding more quickly to decreased emissions. Diet is an important determinant of PCDD/F body burden, due to bioaccumulation and biomagnification of these chemicals in the food chain. Here, we use long-term trends of stable isotope signal to examine whether changes in trophic level may explain the observed PCDD/F concentrations in Baltic herring. Changes in herring nitrogen isotope 15 signature (? N) using muscle tissue of autumn-caught fish were evaluated in relation to the baseline represented by zooplankton (adult 15 copepods Acartia sp. and Eurytemora affinis). ? N in herring varied from 8.4 to 11.0 (mean 9.6 ± 0.6), and increased by ~ 3‰, whereas 15 zooplankton ? N showed a simultaneous decrease of the same magnitude. This implies a change of more than one step in herring trophic level over the last 30 years, assuming a fractionation factor of 3.2‰. Due to the bioaccumulation and biomagnification properties of PCDD/Fs, the increase in trophic level observed in Baltic herring from the southern Bothnian Sea over time is one plausible reason for the observed inconsistent or absent decreases in these contaminants at various sites throughout the Baltic Sea region. However, to be certain this is not just a local phenomenon, SIA should be conducted using herring from different areas together with dietary studies. KEYWORDS: Baltic herring, stable isotope analysis, dioxins, PCDD/Fs 81 Effect of food provisioning on persistent organic pollutant bioamplification in Chinook salmon larvae J.M. Daley, Great Lakes Institute of Environmental Research; T.A. Leadley, University of Windsor / Great Lakes Institute for Environmental Research; T.E. Pitcher, University of Windsor / Department of Biological Sciences; K.G. Drouillard, University of Windsor / Great Lakes Institute for Environmental Research. Fall spawning pacific salmon provision large amounts of yolk to their eggs to allow survival of larvae during under the ice winter conditions. This yolk provisioning leads to maternal offloading of persistent organic pollutants (POPs) to eggs and larvae. Previous research has shown that Chinook salmon larvae exhibit limited capacity to eliminate POPs during the cold water period resulting in bioamplification of POP residues. This study compared POPs bioamplification in Chinook salmon larvae under a high food provisioning treatment and a non-fed treatment to test whether or not food availability attenuates POPs bioamplification via growth dilution. Results demonstrate that larvae in the food provisioning treatment did not gain weight until after day 129. Between hatching and day 129, fed and non-fed treatments exhibited similar decreases in whole body lipid content, negligible POPs elimination and POPs bioamplification factors approaching 1.5. By day 184 of the study, POPs bioamplification factors in the non-fed treatment were as high as 5.3 across chemicals but ranged from non-detectable to approaching 1 in the fed group. This study demonstrates that POPs bioamplification occurs in Chinook salmon larvae even under ideal rearing conditions but peaks after day 129, following which growth dilution can attenuate bioamplification relative to starved individuals. While bioamplification in a real population would likely be captured between the two extremes of maximum food provisioning and the starved group, other stressors could interact with

SETAC Europe 23rd Annual Meeting

the toxicokinetics of POPs. Based on the observations from the current study, it is predicted that multiple stressors such as resource limitation, habitat quality and climate change could potentially push bioamplification factors closer to the non-fed state observed in the present research. 82 Screening organic contaminants in tree cores using comprehensive two-dimensional gas chromatography coupled with time-of-flight mass spectrometry C. Gauchotte-Lindsay, University of Glasgow / Infrastructure and Environment; A. Cooper, University of Strathclyde / Department of Civil and Environmental Engineering; H. Roussel, ADEME / Urban Brownfield and Polluted Sites; M. Chalot, Université de Franche-Comté / UMR 6249 Laboratoire Chronoenvironnement; L.A. McGregor, R. Kalin, University of Strathclyde / Department of Civil and Environmental Engineering; J. Balouet, Environment International. Pollution investigation by trees (PIT) is an international research programme funded by the French agency for environment and energy management (ADEME) to develop the use of phytoscreening and dendrochemical applications at polluted sites. The term phytoscreening was coined to describe the use of plants to characterise polluted sites. Outermost rings of tree cores are microsampled and tested for sap-transported contaminants such as volatiles, polycyclic aromatic hydrocarbons (PAHs), heavy metals, polychlorinated biphenyls (PCBS) and dioxins. Phytoscreening allows semi-quantitative delineation and mapping of plumes in a non-invasive, low cost and rapid manner. Comprehensive two-dimensional gas chromatography coupled with time-of-flight mass spectrometry (GCxGC-TOFMS) is a chromatographic technique with a resolution power an order of magnitude greater than that of classic gas chromatographic techniques. Two columns with different selectivity are placed in series and permit planar resolution of the separation. The use of a time of flight mass spectrometer as a detector allows molecular identification of the eluted compounds. Because of its high resolution power, GCxGC-TOFMS has been employed for the characterisation of complex environmental samples in soils, sediments and waters. As part of the PIT research programme, we are developing a one-step screening method for organic contaminants in tree core samples using GCxGCTOFMS. Organic compounds are extracted from the tree core using pressurised liquid extraction, a faster, greener option to common Soxhlet methods and extracts are analysed by GCxGC-TOFMS without further treatment thereby minimising losses. We present here preliminary results and method development stages establishing the potential of the technique for phytoscreening of PAHs and PCBs. Additionally, through the analysis of a series of Arcolors by GCxGC-TOFMS, we show that multivariate data analysis methods such as principal components analysis further improve the technique by potentially making source identification possible. 83 Phytoforensics Analytics: Delineating Solvents and Explosives through Novel Plant Sampling and Analysis J.G. Burken, Missouri University of Science and Technology / Civil Architectural and Environmental; M.A. Limmer, Missouri University of Science and Technology / Civil, Architectural, Environmental Engineering; Y. Yuan, Missouri University of Science and Technology / Environmental Engineering; H. Shi, Missouri University of Science & Technology / Chemistry; Y. Ma, R. Mu, Missouri University of Science and Technology / Chemistry; C. Balouet, Environment International. Harnessing tremendous energy of evapotranspiration, plants extract nutrients and water from their surrounding environment and concurrently accumulate contaminants. Novel and innovative analytic approaches were developed to access this data and provide information on toxic compounds distribution and exposure easily, rapidly, inexpensively and with little impact to ecology or private property. Fugitive contaminants are difficult to detect in groundwater and the potential risk assessment and remediation efficacy relies on accurate delineation of contaminated areas. Methods: Novel centrifugation, solvent-free extraction techniques were developed to collect tissue fluids, i.e. sap, mechanically. Collected fluid was filtered and injected directly. LC-MS-MS and IC analytic methods were developed to

22

analyze sap for PETN, HMX, RDX, TNT, DNAN and Perchlorate. Method detection limits (MDLs) are the lowest reported for many compounds, down to 20 ng/l. Volatileorganic compounds (VOCs) analysis was done by collecting tree cores, followed by headspace solid phase microextraction (SPME) sampling and GC-?ECD/FID or GC-MS analysis. This method requires no processing and can reach ng/l levels for sap concentrations. In-planta SPME and direct GC-MS analysis methods were developed to analyze contaminants in trees. Solvent-free sampling methods were applied for a variety of monocot and dicot species. Results compared favourably to traditional solvent extractions, which are expensive, time consuming and use hazardous solvents.Linear relationships for in-planta sap and pore water concentrations for HMX, RDX, and Perchlorate confirmed the successful phytoforensic methods development for explosives and enegertics. The methods were applied at a perchlorate field site to delineate the groundwater plume, also confirmed with groundwater sampling, revealing potential for using native plant species in phytoforensics. In-planta SPME and real time/infield GC-MS analysis of VOCs was developed, and using specially designed ports, repetitious sampling of the same tree over years offers a novel long term monitoring approach. Detection limits as low as 0.5 ng/l (parts per trillion) in the xylem tissue were reached. Plumes have been delineated in many settings including urbanized areas, where sampling landscaping trees on personal properties indicated presence of indoor contaminants (vapor intrusion potential), thereby directly assessing exposure pathways to protect human health. 84 Leaching of additives by artificial rain fall under laboratory conditions J. Eltfeld, Instrumenatal Analytical Chemistry; T. Schmidt, University of Duisburg-Essen. New thermal regulations in Germany have resulted that the outer shells of buildings have to be more equipped with polymer materials as in the past. This has the consequence that insulation layers are increasingly equipped with additives: biocides, fungicides, flame retardants, and other organic substances. Forecasts predict that, due to climate change, heavy rainfall coupled with prolonged dry periods will increase. Particularly heavy rainfall events lead to increased leaching of these substances from building facades and thus make their entry more likely to surface water, as there is currently no treatment of this materials provided in the treatment plants. In this project, the mass flows of selected markers (diuron, octylisothiazolone, terbutryn, TCPP,…) was estimated in order to make predictions about future developments. In addition, it was also assessed whether currently or in the future these substances are expected to exceed the thresholds in waters. Aim of the laboratory experiment was to determine which kind of insulation material is leading to the release of organic trace elements and in which extent they were exposed into the environment. The experiment was performed under realistic rain conditions, wherefore four differently equipped facades (1 x 1 m) were irrigated. The runoff water was collected in 1 Litre bottles each and was enriched by solid phase extraction for further GC-MS analysis. Based on the obtained data, water pollution will be estimated for urban areas. This will also be verified by spot sampling from rivers of these particular areas. Additionally it will quantitatively be examined whether "first flush events" were more relevant as the following rain. These results will be compared with the investigation of a large-scale basin. From this work, existing toxicological data and still to be collected dose-response relationships will be estimated for the ecological risk of individual substances and mixtures, which are valued according to the Water Framework Directive. 85 Uranium distribution among cytosolic proteins of Danio rerio gills after different exposure conditions G. Bucher, IRSN / L2BT; S. Mounicou, LCABIE; O. Simon, IRSN; L. Fevrier, IRSN / LRE; R. Lobinski, LCABIE; S. Frelon, IRSN. Uranium (U), found in a wide range of concentrations in aquatic ecosystems, was shown to potentially exhibit toxicity against aquatic organisms, but all mechanisms have not been elucidated yet. To better understand intake, incorporation, storage and elimination mechanisms and then potentially elucidating its associated toxicity, the study of U compartmentalisation and more specifically the identification of U target biomolecules is a key step.

SETAC Europe 23rd Annual Meeting

Indeed U, mainly uranyl cation in aerobic biological matrices, is known to be likely to bind to proteins, then inducing a potential competition with endogenous essential metals. Thus, U speciation in cytosol of cells is of great importance to identify metal toxic fraction as this compartment, containing hydrosoluble proteins, is known to have an important role in the toxicokinetics and toxicodynamics of metals. Therefore, this study focused (i) on the development of non-denaturing and highly sensitive hyphenated analytical techniques (SEC-ICP SFMS presented here) and (ii) on their application to better describe U speciation in cytosols. The ecotoxicological objective was to investigate the influence of different waterborne exposure conditions to U -1 -1 (environmental 20 ug L (c ), incidental-like 250 ug L (c )) on 20 250 its distribution among cytosolic proteins in Danio rerio gills, the latter being one of the first barriers against dissolved pollutants. Consequences on endogenous essential metals distribution in the cytosol have also been studied. Results show that U accumulation in exposed zerbafish gills was significantly different from control (c ) and rather depended 0 on contamination level than on duration. At the subcellular level, a larger extent of U was accessible to cytosolic proteins in c compared 250 to c exposure (i.e 32 vs. 24%) and cytosolic Zn burden increased 20 noticeably in the case of c exposure while Fe and Cu burden 250 remained unchanged. SEC-ICP SFMS chromatographic patterns of gill cytosols were similar for all conditions but U distribution within the pool of proteins varied from c and c to c exposure with a larger 0 20 250 extent of U bound to high molecular weight proteins. Co-elution of U with some Fe, Cu and Zn peaks let suppose a possible competition for endogeneous metal binding proteins. Results confirmed U accumulation on proteins of gill cytosols and provided new information (i) to identify the U toxic fraction and (ii) on potential interaction and competition with endogenous essential metals. 86 Determination of selenium species and some other elements in wild and fish farm trout (Oncorhynchus mykiss) and seabass (Dicentrarchus labrax) U. Kristan, Jozef Stefan Institute / Environmental Sciences; V. Stibilj, Jozef Stefan Institute. Fish are an important source of protein, micro- and macro elements, fatty acids and fat-soluble vitamins and therefore beneficial to human health, because of their high nutritional value. Selenium (Se) as an essential trace element plays a fundamental role in human health, as a component of several major metabolic pathways including thyroid hormone metabolism, antioxidant defence systems and immune function [1]. Its determination and speciation is of strong importance, since in humans, selenium has one of the narrowest ranges between dietary deficiency and toxic levels. The recommended adequate intakes of Se published by DACH are 30 70 µg/day for adults [2]. Nevertheless, some fish species may contain + significant levels of mono methylmercury (MeHg ), polychlorinated biphenyls (PCBs), dioxins and other contaminants. These substances bioconcentrate in the aquatic food chain in such way, that levels are generally highest in older, larger, predatory fish and marine mammals [3]. In our study we included two main fish species that represent the most commonly consumed fresh fishes in Slovenia; trout (Oncorhynchus mykiss) and seabass (Dicentrarchus labrax) bothfrom different locations (Slovenia, Italy, Greece) and environments (fish grown up in fish farms and wild fishes). Total Se concentrations were measured by HG-AFS [4], speciation in soluble fraction after enzymatic extraction was performed by HPLC-ICP-MS, while other elements were determined after microwave digestion by ICP-MS. Content of toxic elements, such as cadmium (Cd) and lead (Pb) were under the detection limit (3 and 23 ng/g respectively) regardless fish species. Se concentrations were more variable; in trout concentrations ranged from 120 to 211 ng/g on wet mass basis while in seabass Se concentrations varied regarding fish origin and breeding method. In seabass, which were grown up in fish farms Se concentrations were in the same range as in trout, whereas seabass caught from natural environment have significantly higher concentrations around 400 ng/g. Keywords: fish, ICP-MS, selenium, speciation 87 Rapid measurement of petroleum hydrocarbons in field soils using mid-infrared diffuse reflectance spectroscopy L. Janik, S.

23

Forrester, J. SorianoDisla, CSIRO Land and Water; M. McLaughlin, CSIROUniversity of Adelaide; G. Webster, R. Stewart, Ziltek Pty Ltd. Contamination of soil by crude oil and refined oil products (total petroleum hydrocarbons – TPH) is a worldwide problem, and the quick and inexpensive analysis of soils and sediments contaminated with these materials would significantly aid remediation and risk assessment activities. Diffuse reflectance infrared Fourier transform spectroscopy, with partial least-squares (PLS) regression, was used to develop calibration models for the non-destructive and economical prediction of TPH concentrations in soils. Standard soils were spiked with known concentrations of TPH and spectral response of the spiked sampled examined. Soils were also collected from different contaminated sites in south eastern Australia known to be contaminated with TPH. A number of selected near-infrared (NIR) and mid-infrared (MIR) frequency ranges were tested for prediction of TPH concentrations. The aliphatic alkyl stretching vibration region was found to be the most sensitive to -1 TPH concentration; NIR frequencies in the 4500-4100 cm region and -1 MIR frequencies at 3000-4600 cm . In particular, the MIR range -1 included two specific alkyl peaks, one at 2950 cm and the other near -1 2730 cm , both shown to have a strong correlation with TPH at low and high TPH concentrations respectively. These peaks were considered to be either weak or absent in natural soil organic matter relative to the -1 usual –CH region at 2930-2850 cm . Using these frequencies, the 2 technique is therefore has the potential to be a rapid and accurate nondestructive method to determine concentrations of TPH in contaminated soils and can be adaptated for in-field applications. 88 Practical Quantification of Ecosystem Services for Environmental Decision-Making in Mine Development J. Nicolette, ENVIRON International Corporation; S. Deacon, ENVIRON UK Ltd; F. Colombo; N. Eury, ENVIRON UK Ltd; R. Wenning, M. Rockel, ENVIRON International Corporation. The incorporation of ecosystem services into international guidance has been increasing since the Millennium Ecosystem Assessment. This includes the incorporation of ecosystem services into the development of environmental and social impact assessments (ESIAs) for large scale extractive industries. In particular, the mining sector is influenced by the multilateral banks and private lenders who have adopted the Equator principles. As an example, the International Finance Corporation (IFC) has incorporated ecosystem services into large-scale impact assessment considerations with Performance Standard 6 on Biodiversity Conservation and Sustainable Management of Living Natural Resources. Although ecosystem services have been incorporated into developing guidance, there is considerable effort being conducted to develop methodologies and approaches to quantify ecosystem services to demonstrate environmental sustainability, stewardship and biodiversity. Many of these methods attempt to place a monetary value on ecosystem services when in fact, those estimates contain significant uncertainty. The aim of this paper is to demonstrate an approach that does not rely on the valuation of all the ecosystem services provided by a habitat, but a subset of representative measures in order to provide a cost-effective approach to develop information sufficient for environmental-decisionmaking. In addition, the approach does not necessarily require monetary valuation of services. The proposed approach will support developing guidance under the IFC and a multitude of other frameworks. In understanding the need to quantify ecosystem services, it is important to understand how that information will be used and what components of valuation are necessary. Ecosystem service quantification for impact assessment originally arose out of the United States natural resource damage assessment (NRDA) process. These tools and approaches have been used on small scale environmental issues but also have merit in evaluating ecosystem services associated with large scale ecosystems. These tools and approaches are adapted within a framework known as net ecosystem service analysis (NESA). We will demonstrate, via a large scale case study, how environmental economics models using monetary and non-monetary metrics, have been used to quantify ecosystem service losses and gains from a practical standpoint facilitating decisions by various stakeholders directly applicable to the mining industry.

SETAC Europe 23rd Annual Meeting

89 Deriving a Water Quality Standard for Iron from Field Evidence A. Peters, WCA Environment Ltd; B.J. Adams, Rio Tinto / Product Stewardship; P. Simpson, WCA Environment Ltd; P. Whitehouse, Environment Agency. Recent studies on the ecotoxicity of iron to fish, invertebrates and algae have shown a clear effect of water chemistry on iron toxicity. As such, water chemistry should ideally be taken into account when establishing quality standards for iron. Field data can be an important line of evidence in setting quality standards alongside conventional laboratory ecotoxicity data. This study aims to take account of water chemistry conditions on the ecotoxicity of iron in assessing whether or not impacts are observed on ecological communities in real field conditions. Previous analyses of the effects of iron on aquatic communities have not taken account of the effect of water chemistry on the ecotoxicity of iron. A limiting function model was fitted describing the ecological quality, expressed relative to a reference condition, as a function of each of the three measures of iron exposure. A statistically significant decline in ecological quality was observed in each of the different measures of iron exposure, although different thresholds were derived for each exposure metric. In each of these analyses the sites with the highest iron exposures have reduced ecological quality, irrespective of whether the exposure is expressed as total iron or as one of the forms of “effective iron”. This supports the assertion that the models developed by Iron Platform for predicting the effects of iron on three different trophic levels are taking account of the principal factors which influence iron ecotoxicity. 90 The use of macroinvertebrates to characterise neutral mine drainage and river flow variability P. Byrne, Liverpool John Moores University; P. Wood, Loughborough University; J. Gunn, University of Birmingham. Contamination of aquatic environments by acid mine drainage has potentially far reaching implications for the quality of aquatic ecosystems. However, little attention has been given to sites where the mine drainage is typically circum-neutral (6 > pH < 8). Previous research on a river receiving neutral mine drainage indicated that aquatic macroinvertebrate assemblages in metal-mine impacted regions are frequently deemed to be in relatively good condition when using standard bio-monitoring indices, despite severe metal contamination of bed sediments and river water. However, the application of multivariate analysis techniques has facilitated the identification of differences in macroinvertebrate community structure in mining impacted and un-impacted reaches of the river associated with metal mine contaminants. It is hypothesised that macroinvertebrate biomonitoring indices fail to identify impacts of metal contamination at the community level because they either seek to identify impacts of a specific contaminant or are dependent on a model community response to a given stress, and neither of these responses are typical of neutral mine drainage. In this paper we examine how the macroinvetebrate community and biotic indices respond to changes in dissolved zinc concentrations and river discharge in the River Lathkill, Derbyshire, associated with the partial blockage by collapse of an underground drainage sough. Analysis of macroinvertebrate biomonitoring indices of water quality indicated moderate improvements in both the Biological Monitoring Working Party Score and Average Score Per-Taxon. In contrast, the Lotic Invertebrate index for Flow Evaluation score (LIFE), which provides a measure of the aquatic macrovertebrate community response to changes in the river flow regime displayed no significant change. The use of multivariate analysis techniques examinaing the macroinvertebrate community (Detrended Correspondence Analysis) and both flow and water quality parameters (Canonical Correapodence Analysis) allowed the identifcation of the biotic and abiotic drivers of communty structure. The resuts indicate that caution is required when interpreting the results of macroinverterbatre biomonitoring indices in regions affected by historic metalifourous mining activity, particularly when the drainage is circum-neutral. Where biotic indices are unresponsive, multivariate analysis may provide an alternative method to examine changes in community strucure. 91 Using tissue residues to predict ecological effects of metal mining

24

on macroinvertebrate communities M. De Jonge, University of Antwerp / Biology; E. Tipping, Centre for Ecology and Hydrology / Lancaster Environment Centre; S. Lofts, Centre for Ecology Hydrology / Shore Section; L. Bervoets, University of Antwerp / Dept. of Biology, Systemic Physiological and Ecotoxicological Research group; R. Blust, University of Antwerp / Systemic Physiological and Ecotoxicological Research, Department of Biology. The present study investigated whether tissue concentrations can be used to predict metal-induced effects on aquatic invertebrate communities. Total dissolved metal levels and four invertebrate taxa (Leuctra sp., Simuliidae, Rhithrogena sp. and Perlodidae) were sampled in 36 headwater streams located in the northwest part of England. Using the River Invertebrate Prediction and Classification System (RIVPACS) taxonomic completeness of invertebrate communities was assessed. Quantile regression was used to th relate tissue concentrations to a maximum (90 quantile) ecological response. The current abstract presents the results for Zn, which was one of the metals showing significant impact. A significant negative relation was observed between RIVPACS-calculated taxonomic completeness and Zn accumulation in Simuliidae. Using the quantile regression model, accumulated Zn concentrations in Simuliidae corresponding to taxonomic completeness associated with 0%, 20%, 50% and 100% -1 decrease could be calculated (4.4, 14.8, 30.3 and 56 µmol g Zn respectively). Based on invertebrate tissue concentrations dissolved metal concentrations representing safe levels for the protection of invertebrate communities could be derived, resulting in 0.26, 9.2, 76 and 463 µM dissolved Zn, which are generally in good agreement with current international environmental quality standards (EQS) and fieldbased results from literature. Overall, the present study concludes that accumulated metal concentrations in aquatic invertebrates can be used to 1) predict metal-induced ecological effects and 2) to derive safe metal levels for the protection of aquatic invertebrate communities. 92 Toxicity and bioaccumulation assessment with the oligochaete Tubifex tubifex in the Nalón River catchment: impact of abandoned metal mines in Asturias L. Mendez, P. Rodriguez, UPV/EHU / Zoology and Cell Biology; M. Martinez-Madrid, University of Pais Vasco / Genetcs, Anthropologie and Animal Physiology. Abandoned mine sites are a great environmental problem. In central Asturias (North Spain), two main focuses of metal mine activities were active until the early 1970s, one in Texeo Mine in Riosa district (Cu mines), and a mine group in the districts of Mieres and Pola de Lena (Hg/As mines). The main objective of this work was to assess sediment toxicity and the bioaccumulation (metal tissue residues) in the aquatic oligochaete Tubifex tubifex exposed to field sediment chronic bioassays. Twenty-six sites were studied in the Nalón Catchment, including reference and nonreference sites from the water surveillance nets in Spain. The 28-day T. tubifex chronic bioassay included survival, reproduction and total growth rate endpoints. Sediment metal concentration and sediment related toxicity and metal bioaccumulation in worms were analyzed using multivariate techniques. Sites from Hg/As mines were severely affected by high levels of both metals, reaching to a maximum sediment -1 concentration of 5320.9 µg As g dw at Los Rueldos site (LRu) and -1 312.5 µg Hg g dw at La Peña-El Tarronal (LPET). In Texeo Cu Mine -1 area, LLamo (LL) reached 115.2 µg Cu g dw sediment concentration. Regarding toxicity 5 different groups were distinguished showing a marked toxicity gradient (ANOSIM, R= 0.987). Eleven test sites were ordered within the same groups than reference sites and thus assessed as Non-toxic, including Texeo copper mine sites. Four sites were assessed as Potentially toxic, and 5 sites from Hg/As Mines with a great impairment on survival (< 20%) and almost no reproduction were assessed as Toxic. Evaluation through Reference Condition Approach using probability elipses, on a database of 61 reference sites in northern Spain, agrees with nMDS classification approach. Due to high worm mortality observed in 4 sites from Hg/As mining areas, worms from only 22 sites were analyzed for metal tissue residues. Two groups (ANOSIM R= 0.987) were distinguished, one group included sites with low metal tissue residues, whereas the other included 3 sites with -1 extremely high values of As and Zn (up to 2166.7 and 2498.1 µg g dw, respectively). Sites from the TexeoCu Mines exhibited low values of Cu

SETAC Europe 23rd Annual Meeting

tissue residue, indicating scarce Cu bioavailability and explaining low chronic toxicological effects observed. A great environmental problem can be inferred from Hg/As mines, and an extensive Evaluation of Risk Assessment (ERA) should be conducted in the area to undertake remediation plans. 93 Effects of metal mixtures on fish and benthic macroinvertebrate populations in a mining-impacted stream: Railroad Creek, Washington, USA C.B. Meyer; J.S. Meyer, ARCADIS; R.A. Hummell, MWH; W. Adams, Rio Tinto. Railroad Creek is a highgradient tributary to Lake Chelan, at 1981 to 335 m asl in north-central Washington State in the USA. The Holden Mine, which produced Ag, Au, Cu, and Zn ore from the late 1800s until 1957, is located adjacent to the creek, approximately 16 km upstream from the lake. Concentrations of Al, Cd, Cu, Fe, and Zn have historically increased as Railroad Creek flows past the mine portal and several waste rock and tailings piles, but remediation efforts are in progress to remove metal and acid rock drainage inputs to the stream. Rainbow trout (Oncorhynchus mykiss) and westslope cutthroat trout (O. lewisi clarki) and their hybrids inhabit the entire creek and are the only fish that have been observed upstream of a set of barrier falls approximately 2.5 km from the lake. We combined fisheries, benthic macroinvertebrate, and physical-habitat data with historical and current water chemistry data to evaluate the relationships of population densities to habitat characteristics and metal concentrations in the creek [after collapsing the metals data into two principal components (PCs)]. The five strongest predictors of trout density were, in order of significance (sign of correlation in parentheses): PC1 “metals from tailings piles” (-), proximity to lake (-), proportion of pools (-), PC2 “metals from mine portal” (-), and density of rootwads (+); additionally, impacts that might be attributed to physical effects of iron-oxide deposition extended downstream from the tailings piles. Historical benthic macroinvertebrate data suggest impacts to trout food sources also extended along and downstream of the mine site, but recent data suggest impacts to macroinvertebrates are now spatially limited to areas immediately adjacent and just downstream of the tailings piles. Ferricrete has cemented some of the macroinvertebrate habitat, decreasing habitat quality. These models suggest improvement of the stream substrate and cover (e.g., adding rubble/boulders [substrate >150 mm, which was also related to trout density] and rootwads, and decreasing iron oxide precipitates and ferricrete on the stream bed) might provide benefits in a restoration project to improve trout habitat in Railroad Creek. 94 Associating risk to chemicals with both natural and anthropogenic sources S.M. Mudge, Exponent International Limited; P.C. DeLeo, American Cleaning Institute; S.E. Belanger, The Procter Gamble Company / Environmental Stewardship and Sustainability Organization; R.A. Stackhouse, Sasol North America Inc / Global Product Safety and Sustainability. Fatty alcohols and their derivatives are used as surfactants in many detergents and personal care products. These high production volume chemicals have been assessed under REACH and similar global chemical management programs with data requirements driven to understand environmental sources, fate and effects of these compounds. European and US regulatory bodies have indicated further interest in the whole category of long chained fatty alcohols given their widespread use in consumer products. Their usage generally results in a down-the-drain disposal pathway that leads to wastewater treatment plants (WWTPs). Recent work using stable isotopes has demonstrated that the chemicals entering the WWTPs are degraded within the system and the fatty alcohols in the liquid effluents are not the same ones that entered in the influent. Likewise, analysis of fatty alcohols in the sediments of the receiving (fresh and salt) waters shows that these are dominated by terrestrial plant derived compounds, and in situ production by bacteria and algae. PNECs for the alcohols are derived for all chain lengths through empirical and toxicity modelling whereas for many alcohol derivatives (surfactants) PNECs are derived using mesocosms and Species Sensitivity Distributions. Measured environmental concentrations are available for a range of locations including the sediments from rivers, soils and coastal environments. By

25

using the stable isotopic signatures and chain length profiles, it is possible to determine the risk ratios for each location and for each source thereby apportioning risk to the different compartments. Since the vast majority (>99%) of the fatty alcohols measured in the sediments are from natural sources, the anthropogenic signal and associated risks are very small.

regulators there is no way around defining precautionary policy targets for the use of PBTs.

97 HC5s from Taxonomically Structured Hierarchical Species Sensitivity Distributions P. Craig, Durham university / Mathematical Sciences; M. Galay Burgos, ECETOC / Environmental Sciences Manager; P. Chapman, Tecsolve; M. Hamer, Syngenta; A. Hart, The 95 Assessment of inorganic UVCBs under REACH F. Iaccino, K. Food and Environment Research Agency; G. Hickey, Durham university Arijs, Arche consulting; K. Lacasse, ECI; F. Verdonck, Arche / Mathematical Sciences; S. Marshall, Unilever; O. Price, Unilever / consulting; D. Vetter, EBRC; H. Waeterschoot, Eurometaux; K. Colworth Science Park; W. Roelofs, The Food and Environment Delbeke, ECI; V. Verougstraete, Eurometaux; C. Braibant, EPMF. Research Agency. One approach to deriving the predicted no-effect Complex inorganic substances containing varying amounts of metals, concentration for a chemical is to use a species sensitivity distribution metal compounds and/or minerals, which may occur naturally (e.g. (SSD) model to estimate the hazardous concentration affecting p% of mineral ores) or be manufactured during the various refining streams of species (HCp), where p is usually 5. Many questions have been raised the metal and mineral industry (e.g. inorganic intermediates) represent about both principles and application of SSDs but the concept has an important fraction of the substances manufactured within the metal nevertheless been found to be useful. In refining the SSD approach, industry (e.g. fluxes, slags, …) and used for a wide range of applications. several statistical issues need to be addressed. These include: interInorganic UVCBs have unique characteristics (e.g. chemical speciation species correlation; tendencies of particular species to one or other end and valence bioavailability) that should be considered when assessing of the sensitivity distribution; and inter-test variation. Attempts have their risks. A specific approach has been developed to identify the been made at addressing each of these issues on its own. Addressing inorganic UVCB substances meeting the REACH requirements and them collectively requires multivariate statistical modelling.\n\nWe appropriately run the risk assessment. Specific data as published in the present a Bayesian hierarchical model of variability and uncertainty for ECHA website and data from consortia have been collected and used to sensitivities of species to a chemical undergoing assessment and for a assess the overall risk of the complex substance without having to database of relevant test results for other chemicals. The Bayesian perform unnecessary testing on the UVCBs. In view of the range of approach has several advantages over traditional non-Bayesian statistical complex inorganic UVCBs, a vast amount of testing would be required methodology aimed primarily at analysing experimental data. It can and the results would not be predictive due to the temporal and spatial incorporate both data and other information such as expert judgements uniqueness of the UVCBs. Eurometaux is working to develop ad hoc or results of meta-analyses. It provides a collective description of guidelines and testing them on specific industry cases. uncertainty for all components of a model, a coherent mechanism for revising uncertainty when additional data become available, and a 96 A socio-economic framework for decision-making on chemicals' decision-making framework which addresses both uncertainty and authorisation: The case of PBT chemicals S. Gabbert, Wageningen utility.\n\nOur model generalises the single chemical random-sampling University / Social Sciences; M. Scheringer, ETH Zuerich / Institute for model proposed by Aldenberg and Jaworska (2000) and addresses the Chemical and Bioengineering; C.A. Ng, ETH Zurich / Institute for issues raised above. It models inter-species correlation by building Chemical and Bioengeneering. Under the European chemicals’ species tendencies and sensitivities hierarchically, based on the legislation, REACH, “substances of very high concern” (SVHC) that have taxonomic classification of species. Taxonomically-related structure been included in Annex XIV can only remain on the market if the seems natural and makes the model a better description of the available European Commission has granted an authorisation. Companies who data but means that it is necessary also to specify a taxonomic scenario: apply for an authorisation must submit a socio-economic assessment the taxonomic structure of the community being protected by the HC5. documenting that “social benefits” from using the chemical outweigh The HC5 is then scenario-specific, being the 5th percentile of sensitivity “negative impacts”. We focus on PBT chemicals, an important subgroup to the chemical for species in the scenario.\n\nWe illustrate the of chemicals for which authorisation might be requested. Assuming that application of the hierarchical model to an existing database of test data the authorisation process requires performing such impact assessment and present open-source software developed to implement the raises the problem of how coherent decisions on the authorisation of hierarchical SSD approach. The software allows a user to estimate the PBT chemicals can be adopted. This, clearly, depends on how benefits HC5 for a chemical based on test data provided by the user and on the from the use of chemicals are weighed against impacts, in particular user’s choice of scenario. The user can also explore various related environmental impacts.\nWe develop a socio-economic framework that aspects of uncertainty. provides conceptual guidance for regulatory decision-making on chemicals’ authorisation. The framework includes cost-benefit analysis 98 Trophic Magnification and the new provisions for and cost-effectiveness analysis as key socio-economic tools that have bioaccumulation in Annex XIII of REACH –a regulatory point of become widely applied in many different domains. For both tools we view W. Drost; J. Ackermann, U. Joehncke, Federal Environment develop a methodological setup for the authorisation of PBT chemicals Agency. Bioaccumulative substances are of concern because substances by assuming that environmental impacts can be modelled as a stock which accumulate in biota are slowly eliminated and thus, can be pollution problem. Furthermore, we survey information and data retained in the organism for a longer time period, they can provoke requirements for their application, and we discuss decision rules for PBT adverse effects at low external concentrations and over a greater time chemicals’ authorisation. Taking hexabromocyclododecane (HBCDD) as scale and can be transferred to the food web. The assessment of the an illustrative example, we examine the implications from CBA and bioaccumulation potential is a key aspect of the European REACH CEA for authorisation of PBTs and discuss implications for short- and regulation in terms of identifying substances which are persistent, long-term policy-making.\nThe CBA model illustrates that the decision bioaccumulative and toxic (PBT) or very persistent and very on granting or refusing an authorisation, and the optimal timing for bioaccumulative (vPvB). The identification of PBT and vBvP removing a chemical from the market, depends on the emission path and substances is based on a set of criteria described in Annex XIII of the the shape of the pollution cost function. An empirical application of a REACH regulation. The assessment of bioaccumulation is usually based CBA approach seems, however, intractable as long as the relationship on distinct cut-off values such as bioconcentration factors (BCF) with between risks and environmental impacts cannot be predicted in a BCF>2000/5000, required for classification as “bioaccumulative (B)” or sufficiently reliable way for all relevant ecosystems. Applying CEA “very bioaccumulative (vB). The standardised bioaccumulation test with requires to define policy targets for PBT use, for example in terms of fish (OECD305) is used to achieve these BCF-values and reflects a maximum tolerable PBT scores. Our results, therefore, emphasise that chemical equilibrium between water and organism, i.e. fish. coherent and sound decision-making requires further in depth research Additionally, since the amendment of Annex XIII of the REACH on assessing the risks of the use of PBT chemicals. In addition, for regulation, alternative test results such as bioaccumulation factors (BAF)

SETAC Europe 23rd Annual Meeting

26

and biomagnifications (BMF) as well as trophic magnification (TMF) derived from field studies may be considered for the assessment using a weight-of-evidence approach. The amendment of Annex XIII and the consideration of other endpoints apart from the BCF help to classifiy substances as bioaccumulative where biomagnifaction is clearly shown in various field studies but the BCF is below the trigger value. However, apart from benefits TMFs and BMFs derived from field studies have limitations. These will be discussed from a regulatory point of view. 99 Mining of the terrestrial toxicity data in the REACH database A. Kapanen, D. Vesentini, ECHA; M. Sobanska, European Chemical Agency; R. Cesnaitis, European Chemicals Agency; P. Karamertzanis, J.V. Tarazona, ECHA. The REACH Regulation (EC 1907/2006) foresees the collection and evaluation of information on industrial chemicals to ensure the protection of human health and the environment. The REACH registration database contains information for over 7 600 substances in 30 000 registration dossiers. For the most of these substances the information on environmental fate and ecotoxicological endpoints is available in the database. REACH information requirements in relation to the effects on terrestrial organisms encompass three trophic levels, invertebrates, plants and microorganisms, and the study of both long and short term exposure. The Predicted No Effect Concentration (PNEC) for the terrestrial compartment is calculated on the basis of the available hazard information on terrestrial toxicity. There is possibility also to calculate a surrogate PNEC based on aquatic toxicity data, using the Equilibrium Partitioning Method (EPM). Based on PNEC and the Predicted Environmental Concentration (PEC) for the substance, a risk for the terrestrial compartment can be characterised and risk management measures can be applied if needed. The mining of the REACH registration database for data relating to terrestrial toxicity has been performed to gain an understanding about the type and quality of data available and the extent of the terrestrial toxicity knowledge used by industry for ensuring safe uses. We will present an overview of the terrestrial data in the REACH database with focus on the toxicity to soil organisms and PNECs. In addition we will discuss sensitivity and representativeness of specific terrestrial end points. We will focus on species diversity needed for the evaluation of long term terrestrial plant toxicity, the importance of soil microbial toxicity assessment and selection criteria for long term tests on soil dwelling invertebrates in more details. Based on the data provided, it was possible to identify the most critical areas were further information could improve the scientifically-justified and consistent assessment of terrestrial risks. Consequently, the importance for more extensive information on terrestrial toxicity to support decision making processes will be discussed. 100 A systematic comparison of IMPACT World+ with other existing life cycle impact assessment methodologies C. Bulle, CIRAIG Polytechnique Montreal / Chemical Engineering; G. Bourgault, CIRAIG; M. Margni, Ecole Polytechnique de Montreal / Department of Mathematical and Industrial Engineering; P. Lesage, CIRAIG / CIRAIG; O. Jolliet, University of Michigan / School of Public Health. This work aims at systematically compare the newly developed IMPACT World+ life cycle impact assessment (LCIA) methodology with ReCiPe and IMPACT 2002+ methodologies by characterizing the 4000 accumulated datasets of the ecoinvent 2.2 database. As comparing CF between methods does not give a sense of how the differences play out in actual LCA applications, this research project proposes to develop a systematic evaluation approach allowing a pairwise comparison of characterized inventories by two different LCIA methodologies. The methodology is applied to compare IMPACT World+ to established methods. Each ecoinvent v2.2 cradle-to-gate inventory (or accumulated dataset) is characterised for one production unit at mid-point and at endpoint levels using the following LCIA methodologies : a) IMPACT World+, the global default version, in which CFs are weighted averages of the regionalised CFs using emissions proxy as weighting factors; b) IMPACT World+, 6 continental versions, in which characterization factors are weighted averages of the finest resolution scale regionalised CFs within each continent using emissions proxys as weighting factors;

SETAC Europe 23rd Annual Meeting

c) ReCiPe and d) IMPACT 2002+ A contribution analysis by midpoint impact category to the respective damage categories human health and ecosystem quality has been performed in order to identify if recuring patterns are observed across the entire database and if they depend on the LCIA methodology. This “category fingerprints” were used to determine systematic differences of behaviour between methodologies. The modeling choices and the main contributing elementary flows underlying those differences were identified and analyzed to establish our level of confidence in those influent modeling choices and to determine if those differences are to be considered as the consequence of an improvement in the assessment or if they are reflecting model weaknesses which still has to be addressed or errors that have to be corrected. The present study allowed a better understanding of the behaviour of IMPACT World+ methodology, but also of the two other LCIA methodologies which have been compared, ReCiPe and IMPACT 2002+. An indepth understanding of the most influent underlying modelling assumptions was allowed by this systematic comparison, putting in light both the improvements brought by IMPACT World+ and its remaining weaknesses in order to better adress them. 101 Improving the spatial scale of impact assessment: analyzing and applying the regionalized LC-IMPACT methodsC. Mutel, S. Hellweg, ETH Zurich / Institute of Environmental Engineering. Regionalized life cycle impact assessment has recently been an productive area of research, and the LC-IMPACT project will produce a number of regionalized characterization factor (CF) maps. However, serious questions about regionalized life cycle assessment (LCA) are still unanswered, such as the best way to choose the spatial scale of regionalized CF maps and the use of raster data in LCA calculations. Regionalized method developers use the most detailed input data available, but this does not mean that the published maps of CFs should be on this highly detailed scale. Rasters with high levels of detail present computational challenges (hundreds of thousands of individual cell and large file sizes); simplified CF maps are easier to use in LCA calculations, easier to interpret, and provide important information to inventory developers about the needed spatial scale of inventory data. The choice of spatial scale should be based on the maximization of some objective criteria. The minimization of spatial autocorrelation is one such approach. We examine regionalized characterization maps from the LC-IMPACT project for noise, emission of metals, forestry, freshwater eutrophication, acidification, and impact on wetlands from freshwater consumption. As a first step, we examine and discuss the similarities and differences in spatial pattern and range for the various methods. We improve the spatial autocorrelation algorithm by the inclusion of spatial relationships in the discretization process, and apply this approach to calculate a best spatial scale for each regionalized method. We discuss the different approaches to discretization, and quantify their effect on outputs from the minimization of spatial autocorrelation algorithm. The choice of spatial scale is important in understanding, interpreting, and applying regionalized impact assessment methods correctly and with minimal uncertainty. The analysis and comparison of multiple regionalized CF maps from the LC-IMPACT project shows the applicability of the spatial autocorrelation algorithm. In some cases, however, the spatial autocorrelation algorithm cannot help when the input data is already highly aggregated or pixelated. In this case, interpolation may help in building smoother CF surfaces. Analysis and comparison of multiple regionalized methods also increases our understanding of regionalized LCA. 102 Towards specific archetypes for the impact assessment of chemicals B. Ciuffo, European Commission Joint Research Centre / Institute for Environment and Sustainability; S. Sala, Joint Research Centre European Commission / Sustainability Assessment Unit Institute of Environment and Sustainability. Emission of chemicals is increasing over years and the related impacts are greatly influenced by spatial differentiation. Chemicals are usually emitted locally but, due to their physical-chemical properties and persistence, may exert both local and global impact. Besides, variability of environmental parameters of the emission compartment may affect the fate and the exposure up to

27

orders of magnitude of difference. Several spatially distributed fate and transport models of chemicals, have been therefore developed at various resolutions. These models allow assessing the distribution and fate of chemicals in the environment after their emissions, on the basis of chemical (viz. physical chemical) and landscape related properties. Unfortunately, most of the models adopted so far are mainly simplistic box models in which the concept of spatial differentiation is based on the scale/resolution. This approach, however, can reduce the uncertainty of the assessment to the extent in which the “box” is really representative of the removal/transport processes. Since this is not necessarily the case, in the present work the authors try to propose a novel approach in which emission archetypes are defined on the basis of the quantitative results achieved by means of a global sensitivity analysis of a complex spatially resolved model. In particular, global sensitivity analysis techniques have been applied to the MAPPE with two aims: to assess the variability in removal rates, focusing on the relative influence of substance properties and of environmental characteristics; and to support the development of chemical specific and compartment specific archetypes. Results strongly call for a “scenario-oriented” approach in the archetype definition. In particular given the underlying variability in the behaviour of chemicals, archetypes should be: (i) compartment specific; (ii) chemical specific; and (iii) target specific. In particular, from the results of the sensitivity analysis the authors attempted defining climate-based archetypes, testing their suitability with respect to the more classical geographical approaches (e.g. continents or countries). Results confirm their expectations, even though, the necessity for further work is also acknowledged. 103 Considering local variability in lignocellulosic feedstock for 2nd generation biofuel LCA thanks to agro-ecosystem modellingK. Dufosse, INRA / Environement and Arable Crops; B. Gabrielle, AgroParisTech; J. Drouet, INRA. Context Agriculture represents 10 to 15% of GHG emissions in France and 65% of N O emissions (GWP = 2 298). Owing to these facts and new regulations (RED) it is essential to carefully develop, evaluate and utilise methods to assess the nd environmental balance of lignocellulosic crops used as 2 generation (G2) biofuel feedstock. However, the literature has pointed out the lack of common methodology for G2 biofuel LCA, which could counterbalance the general results of net reduction in GHG emissions. They especially put stress on the lack of consideration for land use change and use of the generic factors used to calculate greenhouse gases (GHG) emissions, strongly bound with local pedo-climatic conditions and technological options (especially N application rates). A multi-scale approach was required to fulfill a complete LCA. This approach was used to estimate direct N O emissions, downstream indirect N O 2 2 emissions and emissions due to land use change. Materials and methods Firstly, the agro-ecosystem model CERES-EGC was developed and used to simulate crop growth and related C and N cycles. Simulations were carried out at plot (i.e. a few hectares) and regional scales (i.e. hundreds of hectares) for a wide range of crops including G2 lignocellulosic feedstock. In future work, it will be combined with prospective scenarios of land use change towards lignocellulosic crops at regional scale. Secondly, downstream indirect emissions will be simulated from the NitroScape model, which accounts for hydrological and atmospheric transfers of reactive nitrogen between landscape elements (e.g. plots, farm buildings). This model typically works at a scale of a few (20-30) square kilometers. Results and discussion First results on an example region (Picardy, France) showed ranges of yields -1 -1 and direct N O emissions for Miscanthus (14.3 to 20.5 t DM ha yr 2 -1 -1 and 0.18 – 0.38 kg N O-N ha yr ) depending on the selected crop 2 management strategies and local pedoclimatic conditions. Future work will present LCA simulated at farm gate for a mixture of feedstock and integrated at regional scales. Conclusions This multi-scale method helps at integrating local variability in estimates of GHG emission, an important step in biofuel LCA. Moreover, it will be applied in the French project FUTUROL to assess the sustainability of a proposed bioethanol production and supplied by a mixture of feedstock (annual and perennial crops, especially Miscanthus, and crop residues).

SETAC Europe 23rd Annual Meeting

104 Opening Pandora’s box: Uncertainty propagation in life cycle impact assessment P. Lesage, CIRAIG / CIRAIG; G. Bourgault, R. Samson, Ecole Polytechnique de Montreal / CIRAIG; M. Margni, Ecole Polytechnique de Montreal / Department of Mathematical and Industrial Engineering. Uncertainty calculation is receiving more and more attention in Life Cycle Impact Assessment (LCIA). For the first time, statistical distributions are being reported for characterisation factors (CF). This uncertainty is based on statistical distribution on the input parameters, which is then propagated through the models. Due to strong non-linear equations in the impact models, unexpected phenomenon occur that have not been reported so far in the LCA field. The case of a the model used to estimate direct human health impact of water consumption in IMPACT World+ is presented. One parameter in the model, y, is expressed as a function of x, with f(x) being a sigmoidal (sshaped) function, and the input parameter x being lognormally distributed. This type of model is common in LCIA. The quantitative example shows that the probability distribution function (PDF) of the output parameter y is a bimodal that could be approximated with a beta distribution, far from the lognormal distribution of input parameter x. More importantly, if a deterministic value for y had been calculated using the arithmetic mean of x, it would have been far from the observed mean of the function f(x). In fact, in the case of non-monotonic functions, it is even possible that the deterministic result of the model do not fall within the 95% confidence interval calculated by the Monte Carlo. This raises many important questions for LCA as a whole. Should we avoid calculating impact scores with deterministic values? Are the current tools for uncertainty assessment in LCA appropriate for this kind of phenomenon? The purpose of this intervention is to raise awareness about the complexity of the issues at hand and to urge people working in LCA to consult with mathematicians or statisticians to gain better insights about the implication of uncertainty in input parameters of strongly non-linear systems. 105 The first steps towards simplifying the Northern Zone groundwater requirements A. Gimsing, The Danish Environmental Protection Agency / Pesticides and Gentechnology; F. Stenemo, Geosigma AB; J. Asp, G. Czub, J. Ostgren, Swedish Chemicals Agency; U. Bukss, State Plant Protection Service; R. Holten, P. Mulder, The Norwegian Food Safety Authority; D. Kavaliauskaite, The State Plant Service under the Ministry of Agriculture; R. Silvo, A. Sari, Finnish Safety and Chemicals Agency; J. Raukas, Estonian Agricultural Board. Regulation EC 1107/2009 concerning the placing of plant protection products on the market in the EU entered into force on 14 June2011. Acentral aspect in the new regulation is worksharing and harmonization within and among the three European zones. In the Northern Zone there are different national requirements for assessing the risk of leaching of active substances and metabolites to groundwater. No harmonisation on this point has yet been achieved and the Member States are reluctant to accept assessments on the basis of the other Member States requirements. For the core assessment simulations with the following FOCUS models and scenarios are currently required, as described in the Northern zone guidance document: - PEARL with the Jokioinen scenario (Finnish and Latvian approach) - PELMO Hamburg or MACRO Danish scenarios Karup and Langvad with Danish input parameters - MACRO with Norwegian scenarios Rustad og Heia MACRO with Swedish scenarios Önnestad, Krusenberg and Näsbygård \nDue to strict deadlines in the regulation EC 1107/2009, it is imperative to minimize the work load for both companies and regulators. A first step towards harmonisation has been to compare the different models and approaches to assess how much the model predictions differ from each other and how they are influenced by external preconditions. A project has been carried out with the purpose of running the different models with hypothetical substances (different combinations of halflives and sorption properties) and comparing the model predictions with the aim of finding a worst-case ranking order of the models to allow for a stepwise approach to groundwater modelling. The results of this project will be presented and the conclusions and perspectives for the Northern Zone guidance will be discussed. Session: Modelling of chemical fate and exposure in the context of pesticide and biocide

28

regulations Keywords: Groundwater modelling, pesticides, Northern zone, authorisation Presentation preference: Platform and poster 106 Modelling in Support of an Extended Groundwater Monitoring Study in the EU P. Sweeney, Syngenta; G. Hoogewegg, S. Zelonis, Waterborne Environmental; P. Hendley, Syngenta Crop Protection Inc; S. Hayes, Syngenta. The groundwater assessment for agrochemicals in the EU has become increasingly hard to pass due to changing parameterisation of models. Registrants have increasingly resorted to monitoring studies to provide higher-tier support for groundwater assessments in addition to modelling. The data available to identify candidate sites for monitoring in the EU are of variable quality and this limits the ability of registrants to identify credible areas to monitor in locations where there is no existing network of monitoring wells. We show how modleling can be used with the data available at a consistent resolution for the EU 27 with reference to an example of a weakly sorbed metabolite of a maize herbicide. Modelling is used from the initial stages of defining a conceptual model of leaching - identifying likely travel times to groundwater and identifying environmental parameters controlling movement of the substance to groundwater - to providing estimates of leaching using a spatial model. Mass flux is the most relevant quantity to estimate and compare likely leaching of particular soil/weather combinations and we propose that a 10km x 10km grid cell is a relevant spatial unit to consider leaching, given the spatial ncertainty inherent in pan-European datasets such as CAPRI. Calculated median mass flux at a European level is combined with estimates of shallow groundwater (defined as < 10m) and cropping density to provide candidate regions that can be sampled at random as part of a statistically robust monitoring program relevant to leaching across the EU. Comparison with field data show the effectiveness of this approach in identifying high density maize-growing regions overlying shallow groundwater. Finally we show how the sites selected for monitoring can be placed in context of vulnerability for the EU. 107 Characterizing pesticide dissipation in food crops P. Fantke, Technical University of Denmark; R. Juraske, ETH Zurich; O. Jolliet, University of Michigan / School of Public Health. Ingestion of residues via consumption of food crops is the predominant exposure route of the general population toward pesticides. However, pesticide dissipation in crops constitutes a main source of uncertainty in estimating residues in harvested crop parts and subsequent human exposure. Nevertheless, dissipation is a key mechanism in models assessing pesticide distribution in the crop-environment and the magnitude of residues in harvest. We provide a consistent framework for characterizing pesticide dissipation in food crops for use in modeling approaches applied in health risk and impact assessment. We collected 4,482 unique dissipation half-lives for 341 substances applied to 182 different crop species and fully characterize these data by describing their variance, distribution and uncertainty as well as by identifying the influence of substance, crop and environmental characteristics. We obtain an overall geo-mean half-life over all data points of 3.9 days with 95% of all halflives falling within the range between 0.6 and 29 days. Uncertainty in predicting a substance-specific geo-mean half-life varies with varying numbers of available data points with the highest uncertainty associated to pesticides with less than seven reported half-lives. Temperature in air was identified to have a significant influence on dissipation kinetics. We, hence, provide estimated half-lives for a default temperature of 20°C, while introducing a correction term for deviating temperature conditions. Diffusive exchange processes also have a significant influence on pesticide dissipation, wherever these processes dominate dissipation rates compared to degradation. In these cases, we recommend not to use measured dissipation half-lives as basis for estimating degradation, which is recommended in cases, where degradation is dominating. We are currently testing the regression to predict degradation half-lives in crops. By providing mean degradation half-lives at 20°C for more than 300 pesticides, we reduce uncertainty and improve assumptions in current practice of health risk and impact assessments.

108 Characterizing exposure of bystanders and residents to pesticides applied in agricultural fields M. Ryberg, P. Fantke, Technical University of Denmark; R.K. Rosenbaum, Technical University of Denmark / Management Engineering. Humans are exposed to agricultural pesticides via different pathways. Bystanders and residents living near agricultural fields, in particular, are potentially exposed to pesticides primarily via inhalation. However, bystander/resident exposure has not yet been considered in life cycle impact assessment (LCIA), even though bystander/resident exposure is expected to contribute significantly to overall human exposure to pesticides. Therefore, we aim at quantifying human exposure of bystanders/residents to agricultural pesticides applied under realistic field conditions. We start from a pulse application, of which a certain fraction is subsequently lost to air. We thereby build upon an existing model for quantifying pesticide emissions from field applications. The model will calculate the fraction from wind drift and volatilization leaving the field, based on the quantity of pesticide applied to the field. From the emission, the concentration near the receptor – either bystanders or residents living near the field – will be modelled as a function of the distance to the field. Human exposure will furthermore be depending on the duration of the exposure and the inhalation rate. Hence, the exposure differs between bystanders and residents due to different activity patterns. Based on this, intake fractions and – after combination with respective effect information – characterization factors will be derived. Because the impact only affects a fraction of the total population, the results will be normalized, for the characterization to be used together with other exposure pathways where the total population is included. Bystander and resident exposure is expected to be in the same range as exposure via food consumption and is furthermore expected to be higher than exposure of the general public via exposure to environmental emissions drifting far beyond the treated field. Hence, it is necessary to include this exposure into current LCIA methodologies for pesticides to provide a holistic view of the impacts related to pesticides use. 109 Comparison between a priori regulatory predicted concentrations and measured ones from monitoring: the French case A. Duboisset, A. Boivin, A. Conrad, V. Poulsen, ANSES. In France, risk assessment of surface water contamination required for pesticide authorization is conducted with European FOCUS (FOrum for the Co-ordination of pesticide fate models and their Use) exposure models. Recent publications have raised potential concern about the reliability of the FOCUS predictions compared to the real concentrations from monitoring. In the present communication, based on a dataset of 28 active substances recently assessed (national re-registration process), a comparison of FOCUS step 1 - 4 PECsw to measured concentrations from SOeS (Observation and Statistics Office of the French Ministry for Ecology) monitoring network was performed. It does not show any overall underestimation of the FOCUS sw concentrations compared to the measured concentrations. This comparison offers the opportunity to remind that the FOCUS modelling approach is not aimed at describing pesticide transport at specific field or location (i.e. scenarios) after which they are named. For French risk assessment, all the FOCUS sw scenarios used together are considered, as covering the national variety of agro-pedo-climatic conditions, in order to identify acceptable and/or unacceptable risks for aquatic organisms, including mitigation measures when appropriate. However, the actual reliability and sensitivity of the FOCUSsw tools used for regulatory purposes is not straightforward. They have indeed to reflect complex and transient processes (e.g. runoff due to heavy rainfalls). The high temporal variability of the processes involved remains a key issue to be faced for regulatory modelling (e.g. the choice of application date or the time period and resolution to consider for climatic data). At the light of results and Anses experience of regulatory modelling, suggestions for improvement are made. Updating FOCUS surface water in order to produce more accurate PECsw by taking into account several years of repeated applications would be welcome. The new PECsw to be account for risk assessment would have to be defined according to protection goals. 110 Using modelling and GIS analyses to identify appropriate

SETAC Europe 23rd Annual Meeting

29

measures for product stewardship of bentazoneB. Jene, T. Haering, BASF SE. The registration process for plant protection products in the EU according to Regulation 1107/2009 is very stringent and protective. This assures a high level of safety for groundwater regarding the entry of pesticides via leaching after application according to good agricultural practice. Nevertheless, extreme or unusual hydrogeological situations can be relevant at local scale, leading in exceptional cases to exceedances of the drinking water threshold 0.1 µg/L for mobile pesticides such as bentazone. In order to further increase the margin of safety and reduce the number of situations where leaching of bentazone can occur, product stewardship measures have been implemented on the basis of analyses using efate models. The modelling analysis includes local as well as geographical aspects. The sensitivity of the leaching of bentazone to factors such as application time and rate in combination with the type of crop and the respective growth stage was tested under relevant pedo-climatic boundary conditions. The sensitivity of bentazone leaching as a function of site specific parameters such as soil organic carbon content, which determines the actual sorption, and the soil hydrology was also tested. Furthermore, the soil profile depth, the travel distance through the unsaturated zone, was analysed with respect to its effect on the leaching of bentazone. The most important influencing factors on the leaching of bentazone were the application timing, organic carbon content and the depth of the unsaturated soil profile. It was found that the total application rate has an overproportional effect on the predicted leaching of bentazone. The sensitivity analysis therefore resulted in a maximum application rate of 1.0 kg active substance per ha according to product stewardship. As a further important product stewardship measure no autumn or winter applications of bentazone containing products are supported. Furthermore bentazone should not be applied in soils with an organic carbon content below 1.0%. Finally it was specified that bentazone containing products should not be applied to areas with shallow groundwater at less than 1 m depth or shallow soils above karst groundwater aquifers, where water can easily flow through cracks fissures or caves and rapidly reach the groundwater. Spatial analysis using geographical data was carried out to identify areas with shallow soils and karst geology or areas where shallow groundwater can be expected. 111 Can Life Cycle Assessment contribute to strengthen Responsible Research and Innovation? P. Masoni, ENEA / Protezione e Sviluppo dell'Ambiente e del Teritori. Responsible research and innovation (RRI) is a novel approach to develop a legitimate, inclusive, and transparent decision-making process, taking into account wider social, ethical and environmental issues and assessing potential impact and potential for unintended consequences of innovative products and technologies. Life Cycle Assessment can provide relevant contributions to RRI. Life cycle approach is generally considered the best way to assess potential impacts avoiding possible problem shifting. Research in LCA is addressing the question of how better to assess potential impacts in a perspective way with: dynamic LCA, scenario analysis, spatially differentiated LCA, risk based LCA, environmental input-output based LCA (EIO-LCA) and hybrid LCA. Moreover, LCA is broadening the scope of the analysis to include not only environmental but also economic and social aspects (life cycle costing and social life cycle assessment), and broadening the object of the analysis from product systems to technologies, industrial sectors up to the whole economy. Recently, a novel framework for the life cycle sustainability analysis (LCSA) has been proposed, linking life cycle sustainability questions to knowledge needed for addressing them, identifying available knowledge and related models, knowledge gaps and defining research programmes to fill these gaps. In addition, LCSA deepens current LCA to also include other than just technological relations, e.g. physical relations (including limitations in available resources and land), economic and behavioural relations, etc. Moreover, as part of deepening, normative aspects can be explicitly incorporated. Concisely, LCSA is a trans-disciplinary integration framework of models: it integrates empirical facts, scientific knowledge and normative positions, values and rules, in a coherent way. LCSA could therefore

SETAC Europe 23rd Annual Meeting

contribute to strengthen RRI. 112 Good Science, Bad Researchers and Ugly Politics: Reflecting on Ethics in an Ecotoxicology Laboratory F. Wickson, . In aiming to develop “responsible research and innovation”, the Research Council of Norway is currently pursuing a model in which scholars from the social sciences and/or humanities are “integrated” into scientific or technological research projects. This presentation will reflect on my own experience working as an integrated scholar embedded in ecotoxicology laboratories. Ecotoxicology plays a significant role in shaping our understandings of environmental harm and influences how new and emerging technologies are understood, managed, and governed within the industries that are developing them, the policy circles that are regulating them, and the societies that are using them. Ecotoxicology research is therefore a highly relevant site for an integration that seeks to further responsible innovation. The ecotoxicology laboratories that I have been involved with are studying the controversial fields of biotechnology and nanotechnology and in this presentation I will discuss some of the ethical issues that have emerged during our collaboration. These include: 1) Good Science: the role of international standards in defining quality in science for policy; 2) Bad Researchers: factors that currently inhibit honest practice; and 3) Ugly Politics: how the interface between science and politics is generating ethical dilemmas. In concluding the presentation I will open for a discussion about the extent to which this type of socio-technical integration can in fact facilitate the development of responsible innovation. 113

J. van den Hoven, TU Delft - Centre for Ethics and Technology.

114 Plenary panel discussion 115 Questions

R. Owen, .

R. Owen, .

116 Development of a plant toxicity test using species native to the Canadian boreal wetlands M.J. Moody, Saskatchewan Research Council / Environment Division; R.P. Scroggins, Environment Canada / Biological Methods. Development of a plant toxicity test specific to wetland habitats of the boreal eco-zone is a priority recognized by governments and industry inCanada. Hundreds of accidental spills of crude and produced water from oil and gas production pipelines into boreal wetlands acrossWestern Canadahappen each year. Relevant testing tools are needed to quantify biological impacts from pipeline spills or poor waste treatment practises which lead to releases of high levels of salts and hydrocarbons into wetland environments. Candidate test species native to three wetland habitat types (bogs, fens and shallow water marshes) were chosen to offer an ecologically relevant test battery. Achieving consistent germination and growth in preliminary tests led to adoption of suitable test durations and measurements of growth in substrates and waters collected from clean reference sites. As a first method validation step, the sensitivity of the candidate plant species exposed to produced water brine was measured. The next step was to test samples from wetland locations impacted by brine salts and/or petroleum hydrocarbons from pipeline spills or former land treatment sites. The goal of the research and validation efforts is development of a plant testing method using ecologically relevant species that are capable of quantifying the impacts on vulnerable wetland habitats. 117 Effects of salinity and freeze temperatures in Enchytraeus albidus – increased freeze tolerance and changing impact on toxicity A.P. Silva, University / Biology; M. Holmstrup, Aarhus University / Department of Bioscience; M.J. Amorim, Universidade de Aveiro / Department of Biology and CESAM. Enchytraeus albidus is a freezetolerant enchytraeid found in diverse habitats, ranging from supralittoral to terrestrial ecosystems and spanning temperate and arctic regions. Therefore, this species is not only frequently exposed to salinity and temperature fluctuations as well as pollutants. Freeze-tolerance of E. albidus is known but the effect of other natural stressors and/or

30

chemicals in this winter-survival strategy is still poorly understood. We studied the effect of salinity and/or nonylphenol (as an example of an organic pollutant) on the freeze-tolerance of E. albidus, when exposed to different temperature regimes. Several endpoints were addressed, including survival, reproduction, physiological and biochemical parameters. Pre-acclimation to even modest salinities of soil water improved worms’ survival to freezing at low temperature considerably, mainly due to increase in osmolality, decrease in melting point and water content, depletion of glycogen reserves and accumulation of glucose. These physiological and biochemical readjustments led to a lower internal ice fraction during freezing and membrane adjustments, that were crucial to guarantee the survival of worms in extreme temperatures. The presence of Nonylphenol affected negatively the survival of E. albidus, particularly under combined effect with constant frost temperatures or daily temperature fluctuations (freeze-thaw cycles), the latter being the worst. Glycogen was significantly depleted in worms exposed to higher concentrations of nonylphenol in the soil, with more prominence in worms exposed to the combined effect with constant frost temperatures. Lower glucose levels in worms exposed to the combined effect of nonylphenol and daily freeze-thaw cycles were probably due to its use as energy source during thawing processes. Higher glucose levels in worms exposed to the combined effect of nonylphenol and constant frost-temperatures may act as cryoprotectant, which can also explain their better overall survival compared with worms exposed to nonylphenol combined with freeze-thaw cycles. The present results are highly relevant, not only in terms of physiology of invertebrates but also in the ecotoxicology field, where the level of information outside temperate regions and controlled conditions is virtually absent.

regarding the trait-classification of collembolans, the effects of carbofuran on the abundance of local morphospecies (within the family Isotomidae) possessing the same combination of traits scores as F. candida, and the survival of latter species is also alike. As this was only a case-study, further studies, including other pesticide classes, groups of organisms and climate regions are urgent to confirm or not this similarity as well as to enlarge the database on pesticide toxicity under non-temperate conditions.

119 Applying laboratory and in situ ecotoxicological and ecological tools in ERA in the tropics: the experience on a metal contaminated site in Brazil J.C. Niemeyer, Rua Prof Sabino Silva / Department of Life Sciences; M. Moreira dos Santos, IMARCMA Instituto do Mar / IMAR-CMA, Dept. of Life Sciences; R. Ribeiro, Universidade de Coimbra / IMAR-CMA, Dept. of Life Sciences; E.M. da Silva, Federal University of Bahia / Institute of Biology, Department of Botany; J. Sousa, University of Coimbra / IMAR-CMA, Dept. of Life Sciences; S. Chelinho , University of Coimbra. This study was carried out in a metal contaminated area in Brazil, with the major goal to further contribute to the application of a tiered ecological risk assessment (ERA) framework to tropical environments, evaluating the feasibility and usefulness of different assessment tools to be used in different tiers within a Triad approach.This presentation is focused on the ecotoxicological and ecological tools in complementing the chemical characterization of the site. Avoidance behavior tests with Eisenia andrei and Folsomia candida were included in the Ecotoxicological LoE of tier 1, while soil basal respiration, in situ bait lamina test and evaluation of vegetation cover were included in the Ecological LoE of this tier. At tier 2, the Ecotoxicological LoE integrated the results on growth and biomass of plants, Avena sativa and Brassica rapa; the reproduction of the 118 Sensitivity of non-temperate versus standard-test species to oligochaete species Eisenia andrei and Enchytraeus crypticus, and of pesticides: a case-study with carbofuran and Collembola in Brazil the collembolan Folsomia candida. In the Ecological LoE of tier 2, plant S. Chelinho, IMAR CMA / Department of Zoology; X. Domene, litter decomposition (litter bag test), ecological surveys on plant CREAF / Universitat Autònoma de Barcelona; T. Natal da Luz, community and surface dwelling invertebrates (using pitfall traps) and University of Coimbra / Department of Life Sciences, University of conventional microbial parameters were carried out. A sensitivity Coimbra; P. Andres, CREAF - Centre de Recerca Ecològica i Aplicacions analysis was conducted taking into account not only the ability of each Forestals / Universitat Autònoma de Barcelona; I. Lopes, University of parameter to detect differences between contaminated and nonAveiro / CESAM; E.L. Espindola, Universidade de São Paulo - USP / contaminated points (outside the area), but also their ability to detect a Escola de Engenharia de São Carlos - EESC; R. Ribeiro, Universidade gradient of contamination, and the time necessary to obtain the de Coimbra / IMAR-CMA, Dept. of Life Sciences; P. Sousa, IMARparameter.The ability of the avoidance tests to detect toxicity within a CMA / Department of Life Sciences, University of Coimbra. In the short test period and at low costs makes them suitable for use in decision present contribution, the effects of carbofuran applications on a tropical processes. In general, the risk values pointed by them in tier 1 were collembolan community and also on the standard collembolan Folsomia confirmed in tier 2 with the reproduction tests. Oligochaete species were candida were assessed. The main goal was to compare the sensitivity of the most sensitive. The high sensitivity of feeding activity of soil fauna both collembolan community and F. candida towards this insecticide. and its practicability make the bait-lamina test a definitive parameter to The endpoints evaluated were changes in richness and abundance of be included in the ecological LoE of tier 1. Litter bags showed a high Collembola families and mortality and reproduction of F. candida. After sensitivity to contamination and derived habitat disruption, but presented field contamination of an agricultural soil from Brazil, a gradient of a low capacity to differentiate the level of contamination. Soil basal carbofuran dilutions was prepared. Soil cores were taken from the respiration and microbial biomass carbon were the most promising respective uncontaminated surrounding areas and the Colembola microbial parameters in distinguish the level of soil contamination. Soil community of three cores was extracted directly to the test soil. After fauna structural parameters were not able to detect contamination extracting the collembolans to the test soil, these were incubated under gradients, but they were able to differentiate the sites inside and outside laboratory conditions for 4 weeks, after which the mesofauna was the smelter area. extracted again. The organisms were assorted into 5 families and also 120 Development and Application of Toxicity Tests for Plants and classified in different morphotypes, according to five morphological traits related their life-form (namely: ocelli, furca, antenna, pigmentation Soil Invertebrates of Canada’s Subarctic Taiga Plains Eco Zone G. Leighton-Boyce, WorleyParsons / Infrastructure and Environment; M.J. and the presence of hairs and scales). In parallel, the treated soil samples Moody, Saskatchewan Research Council / Environment Division; J. were also used to conduct a standardized reproduction test with F. Battigelli, Stantec; C. Fraser, Environment Canada / Ecotoxicology and candida. Results showed that the community of collembolans was Wildlife Health Division; J. Princz, Environment Canada; R.P. negatively affected by soil contamination and a dose-response pattern Scroggins, Environment Canada / Biological Methods. In Canada, was observed along the contamination gradient, with a lower abundance generic, Canada-wide soil quality guidelines protective of plants and and taxonomic diversity relatively to the control. With respect F. invertebrates (the soil eco-contact pathway) are available for use in the candida, very low concentrations of carbofuran (0.460 mg/kg) caused assessment and remediation of contaminated sites. These guidelines the concurrent impairment of both survival and reproduction of F. were generally developed using toxicity testing on agronomic plant candida. For this particular soil community, the derived toxicity species and earthworms. Guidelines for petroleum hydrocarbons (PHCs) parameters suggest that the intrinsic sensitivity of local Collembola were typically developed using fresh crude. The eco-contact pathway is community (EC50 for decrease in abundance= 0.061 mg/kg) to commonly included for soils from surface to 3 m depth. In sub-arctic carbofuran was rather similar to the one found for Folsomia candida in regions with intermittent permafrost, endemic plant and invertebrate the same test-soil (LC50 for survival = 0.057 mg/kg). Furthermore,

SETAC Europe 23rd Annual Meeting

31

species may have a different sensitivity to contaminants than agronomic species used in the generic guidelines. Distribution of soil invertebrates within the soil profile and the rooting depth of plants in sub-arctic regions may also differ from the distribution within temperate regions. Also, at some sites with historic impacts, or sites where soils have been bioremediated, the objective may be the management of weathered or aged as opposed to fresh PHC impacts to soils. A collaborative research project between industry and government was initiated in 2009 with the overall objective of developing site-specific remediation guidelines for a range of contaminants for the eco-contact pathway that are protective of native plants and invertebrates found at sites located in the Canadian subarctic. For plants, method development has progressed through initial species selection from published surveys, germination, growth and test duration trials, assessment of the need for fertilizers, and an evaluation of the applicability of using the standard Environment Canada test climate or a northern climate. For the invertebrate tests, a literature search found very few surveys of soil mesofauna in Canada’s north. A field investigation of the distribution, density and diversity of soil invertebrates, and of plant rooting depths, through the soil profile from surface to either permafrost or 3 m was completed. Bulk soil samples were also collected from the study area for extraction and culturing of soil invertebrates, with a goal of establishing laboratory cultures to be used in future toxicity tests. Results of the plant method development and of the field invertebrate survey and culturing will be presented and implications for the depth to which the eco-contact pathway may be applicable in sub-arctic regions will be discussed. 121 Assessing microbial toxicity of heavy metals using flow cytometry S. Kim, Konkuk University;Y. An, Konkuk University / Department of Environmental Sciences. The effects of heavy metals on microbial growth were investigated on liquid media, and the colony forming unit (CFU) assay and colony area evaluation were performed on agar media. The Escherichia coli and Bacillus subtilis were chosen as a test species. In order to assess the live cell ability and size distribution, we used the Fluorescence Activated Cell sorting (FACs) and calcein acetoxymethyl ester (CAM) as a fluorescent dye of live cell. On agar media, the colony area evaluation shows higher sensitivity than CFU assay. Also, Cu and Ni induced the early developmental inhibition of E. coli and B. subtilis on liquid media. On flow cytometer analysis, live cell ability, side and forward scattering of cells were measured. The live cell abilities of exposure groups were decreased, and the cell size and granularity show lower than control group. This phenomenon can be related with small colony formation and growth inhibition. Our results suggested that effects of heavy metal may be assay-dependent, because heavy metal can change the cell morphology characterization. This subject is supported by Korea Ministry of Environment as the GAIA project (2012000540011). 122 Unexpected dose response of repeated annual veterinary drug applications on microbial diversity and functions in an agricultural soil E. Topp, Agricultural and AgriFood Canada; A. Scott, L. Sabourin, Agriculture and Agri-Food Canada; P. Grenni, National Research Council CNR / Water Research Institute; A. Barra Caracciolo, National Research Council / Water Research Institute; R. Marti, Agriculture and Agri-Food Canada. Use of antibiotics as growth promoting agents in livestock production contributes to the increasingly worrisome development of antibiotic resistance. In order to evaluate the long term impacts of antibiotic exposure on soil microbial populations, a series of field plots were established in 1999 that have since received annual applications of a mixture of sulfamethazine, tylosin and chlortetracycline at concentrations (0, 0.1, 1.0 and 10 mg/kg soil) bracketing that which would result from an annual application of manure from medicated swine. Soils were evaluated for drug persistence, herbicide and crop residue decomposition, distribution of major bacterial groups by FISH analysis, abundance of Sul1 by qPCR, and tolerance to the antibiotics by PICT. Sulfamethazine and tylosin were biodegraded significantly more rapidly in the soils historically exposed to 10 mg/kg drugs than in control soils. Most endpoints were significantly affected by historical drug exposure at the lower and

SETAC Europe 23rd Annual Meeting

intermediate drug exposure concentrations, compared to untreated and highest level of exposure. We speculate that the evolution of a biodegradative population enriched at the highest exposure concentration results in reduced exposure of the overall soil population to the drugs. 123 Fate and transport of chlortetracycline and resistance genes in the environment after land application of swine manureS. Joy, Civil Engineering Dept / Civil Engineering Dept.; S. Bartelt-Hunt, X. Li, University of Nebraska-Lincoln; D. Snow, Water Science Laboratory; J. Gilley, USDA-ARS; D. Marx, University of Nebraska-Lincoln / Statistics Department. The Centre for Disease Control and Prevention says antimicrobial resistance (AMR) is one of the world’s most pressing public health issues. One source of antimicrobials to the environment is animal agriculture. Antimicrobials are used for growth promotion, prophylaxis, and for disease treatment within concentrated animal feed operations (CAFOs). Antimicrobials allow resistant bacteria to proliferate in manure applied as fertilizer, and therefore antimicrobials and AMR genes can contaminate surface and ground water. This is studied through monitoring the quantities of antimicrobials and AMR genes in fresh and aged swine wastes. The fate and transport of antimicrobials, AMR genes in soil and surface runoff is quantified after land applications of surface applied and injected swine manure. Rainfall simulations were conducted after manure containing CTC was applied by broadcast, incorporated, or injected, and runoff samples were collected. Samples were analysed for CTC in liquid and solid phases through liquid chromatography tandem mass spectrometry or high pressure liquid chromatograph, respectively. Published qPCR protocols were followed to determine a concentration of AMR genes/ml of runoff for tetQ, tetX, and 16s. Both CTC and AMR genes decreased in runoff throughout the three day experiment. Our results showed despite of some difference in the runoff profiles, the AMR gene runoff profiles were similar for the three application methods. Because the experiment only lasted three days, it is likely that the indigenous soil microbes had not developed substantial levels of resistance and the resistant bacteria in the runoffs were largely from the original manure. The concentrations of antimicrobials in the runoff indicate that manure injection may reduce transport of antimicrobials compared with broadcast and incorporation. The transport of antimicrobials is also affected by the timing of rainfall, as lower antimicrobial concentrations were observed in runoff after the third runoff event. 124 Multiple factors govern the effects of pharmaceutical antibiotics in structured field soil R. Reichel; S. Thiele-Bruhn, Soil Science, University of Trier. When pharmaceutical antibiotics reach soils via contaminated excreta, adverse effects on biota are expected. Numerous studies showed that soil microbial biomass, functions and structural diversity are affected after spiking soils with antibiotics. Effects on soil biota are routinely determined by standardized laboratory tests using homogenized soil, artificially spiked with the antibiotic compound in the presence of a nutrient substrate. In practice, manure exhibits varying molecular and microbial composition. Consequently, manure from medicated livestock represents a mixture of nutrient substrates, microbial inoculum and antibiotic chemicals including metabolites. This mixture is added to structured field soil that is characterized by heterogeneous microcompartments such as macroaggregates, earthworm channels and the rhizosphere. The consequences of such mixed influencing factors on the effects of antibiotic in microcompartments of structured soil are still poorly investigated. These aspects were evaluated by determining microbial measures, e.g. PLFA and DGGE patterns or enzyme activities, using topsoil from a Luvisol in several experiments from laboratory to field scale. Sulfadiazine and difloxacine were added to soil as test compounds\nranging from artificial spiking to soil amendment with slurry from medicated pigs. Results showed that antibiotics already altered the molecular and microbial composition of the slurry, which blurred a precise differentiation between responses to slurry composition or to the excreted antibiotic compounds. Manure borne microorganisms were shown to survive for weeks in soil with likely consequences for the indigenous soil microflora. Moreover,

32

antibiotic-related effects in soil microcompartments commonly were different compared to the corresponding bulk soil. Hence, the evaluation of veterinary antibiotics’ effects in soil has also to consider these mixed effects on manure and soil when assessing the environmental relevance of antibiotics.

and iii) the incubation conditions (anaerobic/aerobic) on the effects of the antibiotics in plant tests. The platform presentation presents the experimental results of the research project with a special focus on the methodology and the influence of the test design on the variability of the results.

125 Effects of soil properties on the uptake of pharmaceuticals into earthworms L. Carter, University of York / Environment; J.J. Ryan, GlaxoSmithKline; A. Boxall, University of York / Environment Department. Pharmaceuticals are being increasingly detected in soils. This is primarily due to the land application of sewage sludge and reclaimed wastewater containing high levels of pharmaceuticals which are then transferred to soil. Once in soil, there is the potential for pharmaceuticals to be taken up by soil dwelling organisms however relatively little is known regarding terrestrial invertebrates. This study was therefore performed to better understand pharmaceutical uptake into earthworms (Eisenia fetida) and to evaluate effects of soil properties on the uptake of pharmaceuticals. Earthworms were exposed to soils spiked 14 with either C labelled fluoxetine, carbamazepine or orlistat (studies with an additional drug, diclofenac, are in progress) for 21 days (uptake phase) followed by a 21 day depuration phase. Soil and pore water was also sampled during the uptake period. Samples were then extracted and analysed by liquid scintillation counting. Internal worm, soil and pore water pH measurements were made to observe any pH changes that occurred during the exposure. All pharmaceuticals were taken up by E. fetida. Fluoxetine and carabamazepine were completely eliminated from the worm in the depuration phase whilst traces of orlistat remained. Slight pH changes were observed in the pore water and soil in the orlistat study and for fluoxetine the internal pH of the worm was higher during the uptake phase then decreased to pH values comparable to control in the depuration phase. Between the five different soil types Bioconcentration factors (BCFs) were very similar for carbamazepine whereas large differences in BCFs were observed (30.3-114.9) for orlistat. For fluoxetine BCFs ranged between 16 and 21.5 in the different soils. As fluoxetine is a basic drug you would expect an increase in BCFs as the pH increases as there would be more nonionised species with higher hydrophobicity than ionised species at pH values closer to the pKa. However this relationship is not shown; suggesting soil properties other than pH can influence uptake into worms. This research shows earthworms can accumulate pharmaceuticals if they are present in soils at evironmentally relevant concentrations and that for some drugs soil properties can influence the degree of uptake into the worm and hence BCFs. Soil, pore water and earthworm data will be combined; with the ultimate aim to develop models to improve the assessment of the risks posed by pharmaceuticals in the terrestrial environment.

127 Biotransformation products of ibuprofen in soil – a new view on the relevance of non-extractable residues K.M. Nowak, RWTH Aachen University / Institute for Environmental Research Biology V; C. Girardi, A. Miltner, UFZ - Helmholtz-Centre for Environmental Research / Environmental Biotechnology; A. Schaeffer, RWTH Aachen University / Institute for Environmental Research; M. Kaestner, UFZ Helmholtz-Centre for Environmental Research / Environmental Biotechnology. Ibuprofen was reported to degrade quickly in soils and sediments with the significant formation of mineralisation products and non-extractable residues (NER). NER formed during biodegradation of organic contaminants in soil are considered to be a result of various physical and chemical interactions of a parent contaminant and / or its primary metabolites with soil organic matter (SOM). To date, their chemical composition is still unknown, therefore it is speculated that these compounds immobilised as NER may pose a risk for environmental and human health after their release from SOM. However, in the case of biodegradable organic compounds, NER may be biogenic and can contain microbial biomass components, for example fatty acids (FA) and amino acids (AA). After cell death, these biometabolisation products are subsequently incorporated into nonliving SOM, where they are stabilised ultimately forming hardly extractable residues of biogenic origin. We investigated biodegradation 13 13 of C -ibuprofen, in particular the formation of C-labelled FA and 6 13 13 AA and their fate in soil over 90 days. C-FA and C-AA in the living microbial biomass fraction initially increased, thereafter their contents decreased and a continuous incorporation of these biomolecules into the non-living SOM pool was observed. At the end, nearly all NER were biogenic as they contained only natural microbial biomass compounds. This can be relevant also for other biodegradable contaminants; therefore, the possible biogenic NER formation needs to be considered in the assessment of the potential risks of the readily biodegradable contaminants in soil for the environment.

126 Veterinary Antibiotics in Terrestrial Plant Tests – Effects of a more realistic exposure way via manure M. Simon, Fraunhofe IME; M. Herrchen, Fraunhofer Institute for Molecular Biology and Applied Ecology (IME); B. Foerster, ECT Oekotoxikologie GmbH (ECT); N. Graf, J. Roembke, ECT Oekotoxikologie GmbH; U. Kuehnen, I. Ebert, Federal Environment Agency. In the frame of a research project initiated and funded by the German Federal Environmental Agency (UBA), a special terrestrial plant test for veterinary pharmaceuticals – especially antibiotics - with a more realistic exposure scenario via manure application was developed. The research project comprehends: i) development of methods of preparation, acclimatization, incubation, and application of manure in a plant test, ii) tests investigating necessary technical background (e.g. suitable plant species, suitable manure concentration), iii) tests according to the OECD 208 standard test tesign and modified test designs, considering an application of the test substance via manure. To ensure a significant evidence of the studies regarding universal validity, main tests were conducted with six plant species and eight replicates. The studies were conducted with pig and cattle manure and two representative veterinary antibiotics. The test design considers additional effects of manure to the test substance (e.g. adsorption) as well as transformation/metabolization of the test substance in manure by investigating the influence of i) the duration of anerobic incubation in manure, ii) the way of application,

SETAC Europe 23rd Annual Meeting

128 Genetic variation of transcriptomic expression in Lymnaea stagnalis A. Bouetard, INRA; C. Hoede, INRA / BIA and Genotoul Bioinformatics; A. Besnard, INRA / UMR ESE 0985; T. Pecot, Ohio State University / Comprehensive Cancer Centre; M. Collinet, INRA; L.L. Lagadic, INRA / UMR INRAAgrocampus Ouest Ecology and Ecosystem Health; M. Coutellec, INRA / Aquatic Ecotoxicology, UMR ESE. Population response to stress may have a genetic component which, if additive, is the basis for adaptive evolution to local conditions. Apart from monogenic resistance, adaptive processes have been traditionally investigated through phenotypes at quantitative traits. However, more elementary responses may also entail a heritable component. This is true for gene expression, which results from various molecular interactions. We investigated the evolutionary potential of transcriptomic expression induced by a pro-oxidant herbicide, diquat, using lines from four natural populations of the pond snail Lymnaea stagnalis. Populations stemmed from contrasted environments (close to vs distant from agricultural zones), which allowed testing the influence of historical exposure to putative environmental stressors. These populations were significantly differentiated at neutral genetic markers and showed significant genetic divergence at several life history traits. Previous results on diquat molecular effects on L. stagnalis suggested (1) the occurrence of various responsive pathways, and (2) that induced transcriptional overexpression may not translate into detectable functional changes (enzyme activity). In the laboratory, F1 individuals were exposed during five hours to diquat vs control conditions. RNA was extracted from hepatic tissue. RNAseq analysis was based on highthroughput sequencing (Illumina Hiseq2000) of 16 cDNA libraries, i.e., two replicates (each based on three different families) × four populations × two exposure conditions (diquat, control). Read assembly strategy was based on a meta-assembly using Abyss and MIRA, after a preprocessing

33

to extend short-read length by overlapping paired-end reads using FLASH. Then all initial reads were mapped against all contigs using bwa. Filtering contigs on which less than 1/1000000 reads mapped led to 48321 contigs, 26958 of which had a blast hit against Swissprot, Refseq-Prot or Refseq-RNA. Diquat-induced differential expression is currently compared across genetic origins using DESeq package. Functional annotation is used to identify and compare responsive molecular pathways (KEGG). Results will be discussed in the light of diquat early toxicological impact and of evolutionary potential of transcriptomic responses. Family and population variance in life history traits will be linked to constitutive transcriptomic expression, to draw a more integrative interpretation. 129 Micro-evolutionary response in a natural Daphnia magna population under Cu and Zn stress J. Hochmuth, Environmental Toxicology and Aquatic Ecology; C. Janssen, University of Ghent / Laboratory of Environmental Toxicology and Aquatic Ecology; K. De Schamphelaere, Ghent University UGent / Environmental Toxicology and Aquatic Ecology. A 10 week experimental evolution study was carried out under semi-field conditions to test for micro-evolutionary effects in a natural Daphnia magna population exposed to a control, 2 Cu, and 3 Zn concentrations. We investigated if the long-term exposures to Cu or Zn resulted in a higher organism fitness compared to that in the control exposures and in the original (= start) population. At the end of the microevolution experiment a life-table experiment was initiated with clones from the start population, a control population and metal exposed populations to assess if metal acclimation had occurred in the exposed populations. The populations experimentally evolved at 180µg Cu/L and 760µg Zn /L had significantly higher reproduction at the corresponding concentrations than the lower metal or control treatments, i.e. evidence that metal acclimation had occurred. After 4 months of culturing under control conditions, thus eliminating any acclimation history, an additional life-table experiment was conducted to determine whether metal adaptation, measured as an increase in mean population fitness occurred. We observed a significantly higher total reproduction at 760µg Zn /L and at 180µg Cu/L in the respective long-term exposed populations compared to the long-term control exposed population and the start population. In long-term exposure to 760µg Zn/L acclimation and adaptation to the metal had enabled the population density to recover, matching that of the control, despite an initial reduction of 75% of the clones. Under long-term exposure to 180µg Cu/L, however, despite lesser initial mortality (50%), acclimation and adaptation effects were not sufficient to lead to a full recovery of the population denity. Our results confirm that micro-evolution can occur after only a few generations but that adaptation in itself is not a guarantee for a complete recovery of the population density. 130 Adaptation to pollutants through modifications of sensitivities and life-history traits: a case study based on nine field Gammarus populations. a. vigneron; O. Geffard, a. chaumot, Irstea. The question of evolutionary processes in ecotoxicology becomes substantial in order to propose relevant ecological risk assessment. Actually, it is well admitted that pollutants can be strong selective agents which can induced adaptation in exposed populations. Our study analyses two possible ways of adaptation: first, the evolution of resistance and second, the evolution of life-history patterns. This second point is rarely tackled in ecotoxicological issues, contrary to other fields of ecology dealing with stress adaptation. To answer these questions we adopted a retrospective approach based on the comparison of sensitivities and traits of exposed and unexposed field populations. To provide representative results, the choice has been made to work on an ecologically and ecotoxicologically relevant species, the freshwater amphipod, Gammarus fossarum. Particular attention was also paid to the design of the study through the sampling effort with the selection of nine field populations spread out at a regional scale. The exposition of organisms from these populations to cadmium, and the measure of their life-history traits in the field and in common garden in the lab, show that organisms are not only able to respond to pollutants by increased tolerance, but also by life-history adaptation. Nevertheless, our results

SETAC Europe 23rd Annual Meeting

also demonstrate that adaptation is not automatic and seems to depend on contamination strength. In addition, these phenomenon are probably not out of unexpected outcomes since they can induced fitness costs, and changes in population dynamics can affect higher ecological scales. 131 Population genomics reveals adaptive variation and a potential path to environmental forecasting J.R. Shaw, Indiana University / The School of Public and Environmental Affairs and The Center for Genomics and Bioinformatics; J.K. Colbourne, University of Birmingham / School of Biosciences. Daphnia, or the water flea, is a sentinel species of freshwater ecosystems. Their populations are defined by the boundaries of ponds and lakes, are sensitive to modern toxicants in the environment, and thus are used to assess the ecological impact of environmental change. Their short generation time, large brood sizes, and ease of laboratory and field manipulation have assured Daphnia’s importance for setting regulatory standards by environmental protection agencies, testing chemical safety, monitoring water quality, and as a model for environmental genomics research. A hallmark of the genome sequence is a large number of duplicated genes that are most responsive to ecological challenges and are specific to the Daphnia lineage. In this study, we take advantage of maturing genomics tools to understand the molecular basis for evolved tolerance to toxic levels of certain metals. We also test the adaptive significance of Daphnia’s genome structure. Natural populations that have faced severe chemical challenges for over a century of industrial iron/ore smelting demonstrate evolved tolerance to cadmium. Other reference populations that have no history of chemical stress are clearly harmed by metal exposure, showing slower growth rates, lower fecundity and higher mortality. By measuring the distribution of copy number variants (CNV) and interrogating differential expression of 31,000 annotated genes from sampled populations across chemical conditions and through time, this study provides new insights into the functional interactions between genome structure and environment. We observe allele specific copy number increases in both extant and extinct populations living along a steep metal cline. We identify a large number of CNV, including the metal detoxication protein metallothionein that strongly correlate and are predictive of phenotypic differences between populations. These studies begin to quantitatively link genomic variation with individual fitness and population-level outcomes, and both benefit from and contribute to the Daphnia Genomics Consortium. 132 What genomic mechanisms cause a cadmium-adapted Daphnia pulex isolate to be more tolerant to cyanobacterial stress than a nonadapted isolate? D. De Coninck, Ghent University / Laboratory of Environmental Toxicolgy & Aquatic Ecology; J. Asselman, Ghent University / Laboratory of Environmental Toxicology; S. Glaholt, Indiana University; J.K. Colbourne, University of Birmingham / School of Biosciences; C. Janssen, University of Ghent / Laboratory of Environmental Toxicology and Aquatic Ecology; J.R. Shaw, Indiana University / The School of Public and Environmental Affairs and The Center for Genomics and Bioinformatics; K. De Schamphelaere, Ghent University UGent / Environmental Toxicology and Aquatic Ecology. Natural populations are characterized by genetic variability which may allow populations to adapt to a stressor. As a consequence of this genetic adaptation, populations may become more (co-tolerant) or less tolerant (cost-of-tolerance) to other stressors. In aquatic ecosystems, stressors rarely occur isolated and the number of stressors are predicted to increase with global climate change (e.g. cyanobacteria). The aim of this study is to unravel the genomic basis of the observed differences in Microcystis aeruginosa (MC, a cyanobacterium) tolerance among cadmium (Cd) adapted and Cd non-adapted isolates of Daphnia pulex to gain insight into the genomic basis of the co-tolerance of Cd-adapted D. pulex to cyanobacterial stress. Gene-expression profiles of a Cd adapted and a Cd non-adapted isolate in response to MC were characterized using a micro-array platform. Functional enrichment analysis of pathways, based on KEGG reference pathway maps was performed. In addition, gene copy numbers for different isolates from Cd adapted populations and Cd non-adapted populations were determined by hybridizing gDNA to the microarray. In total, 9 pathways and 14

34

paralogous gene familes were found to be significantly enriched. Among the top most significantly enriched pathways that were regulated by MC in different ways in both isolates were the oxidative phosphorylation, ribosomes, proteasome, sucrose and starch metabolism and linoleic acid metabolism pathways. All pathways were mostly only differentially regulated in the Cd non-adapted isolate, but not in the Cd adapted isolate which suggest an overall lower stress experience in the Cd-adapted isolate. We could attribute this to (i) lower accumulation of toxins than in the Cd non-adapted isolate and (ii) higher naïve basal expression of genes related to these pathways as a consequence of its adaptation to Cd. We could not confirm that copy number variation correlated with the higher naïve basal expression. This study showed possible genomic mechanisms of co-tolerance of Cd-adapted D. pulex populations to cyanobacterial stress. As micro-evolutionary changes in populations and their consequences (such as co-tolerance) for the toxicity of unrelated stressors gain more and more interest in risk assessment, it is benificial to understand the mechanisms that drive these consequences. E.g., if we understand mechanisms of co-tolerance we can perhaps ultimately predict which combinations of stressors may show co-tolerance. 133 Epigenetic programming of disease and the role of developmental exposure to environmental contaminantsJ. Kamstra, VU University Amsterdam; L. Bastos Sales, P. Cenijn, T. Hamers , J. Legler, VU University Amsterdam / Institute for Environmental Studies. Epigenetic programming of disease and the role of developmental exposure to environmental contaminants Jorke Kamstra, Liana Bastos Sales, Peter Cenijn, Timo Hamers and Juliette Legler, 1Institute for Environmental Studies, VU University Amsterdam, The Netherlands, [email protected] A growing body of literature has demonstrated the crucial importance of epigenetics in gene regulation. Research shows that environmental factors, including exposure to contaminants, can alter epigenetic control of gene expression, with important implications for development and susceptibility to disease. In our laboratory, we study the effects of developmental exposure to endocrine disrupting chemicals (EDCs) on the latent onset of diseases such as obesity, using in vitro mouse and zebrafish models. In this study, we focus on the effects of EDC exposure on in vitro adipocyte differentiation, and investigate underlying changes in global and gene-specific DNA methylation. To this end, murine 3T3L1 pre-adipocyte cells were exposed to EDCs during differentiation, and multiple gene targets involved in the adipocyte differentiation pathway were assessed with QPCR. Global DNA methylation was analyzed with HPLC and specific DNA methylation was analyzed on promoter regions of PPAR?2 and Leptin with Methylation Sensitive High Resolution Melting Analysis (MS-HRM). Our results show that exposure to EDCs can alter adipocyte differentiation in vitro, which is accompanied by changes in (global) DNA methylation. Novel effects of the brominated flame retardant BDE-47 on adipocyte differentiation and methylation of key genes involved in adipogenesis have been found. 134 A modelling approach to characterize sub-lethal responses of Daphnia magna populations to chemical exposure in the presence of environmental stressors F. Gabsi, RWTH Aachen University / Institute for Environmental Research; A. Schaeffer, RWTH Aachen University / Chair of Environmental Biology and Chemodynamics; T.G. Preuss, RWTH Aachen University / Institute for Environmental Research. Changes in population responses to chemicals in the presence of environmental stressors are reported in the literature for several organisms. Yet, the current risk assessment (RA) methodology is based on laboratory tests where chemicals’ effects are assessed independently of natural stressors. It is a challenging task to account for these factors experimentally because of the testing efforts required and the difficulty in simulating real scenarios at a laboratory scale. Also, joint chemical and environmental stresses cannot be assessed based on knowledge on their separate effects since interactions don’t always lead to additive effects. The ‘Virtual Ecologist’ approach has been suggested as a method to circumvent data limitations. It relies on the use of powerful models that allow testing complex, realistic scenarios. Individual-based models (IBMs) relate the measured toxicity on individuals to populations. They

SETAC Europe 23rd Annual Meeting

also integrate various environmental factors, allowing for a mechanistic understanding of ecological impacts on populations. These features make them powerful ‘‘virtual laboratories’’ for testing diverse hypotheses on population properties. In this study, we use a developed and validated IBM for Daphnia magna as a virtual laboratory to explore potential interactive effects of chemical and non-chemical stressors on populations and determine how the ability of populations to cope with sub-lethal effects is affected by the presence of environmental stressors. In the model simulations, constant exposure to toxicants affects solely or in combination, the daphniid's reproduction, growth, filtration rate or survival. Environmental factors include predation (Chaoborus crystallinus), competition or food level. Model results revealed a strong implication of environmental stressors in determining population sensitivity to chemicals. Equal toxicity levels had different impacts on population size when integrating environmental stressors. Interactions between chemical and non-chemical stressors manifested through additive, synergetic or antagonistic effects. We conclude that population resilience cannot be only attributed to chemicals’ effects. Ignoring environmental stressors might lead to non-realistic estimations of chemicals’ risks to populations. Integrating validated models into the current RA procedure is urgently needed. 135 Impacts of imidacloprid on individual performance and population dynamics of Daphnia magna A. Agatz, T.A. Cole, University of York; E. Zimmer, Vrije Universiteit Amsterdam; T.G. Preuss, RWTH Aachen University / Institute for Environmental Research; C.D. Brown, University of York / Environment Department. Various effects of xenobiotics on aquatic organisms might not be caused directly by the compound, but rather arise from adaptation of the organism to stress invoked by feeding inhibition during exposure. We demonstrate how multiple lines of evidence linked by understanding the ecology of the organism are necessary to elucidate xenobiotic impacts along the effect cascade (feeding, growth, maturation, reproduction and survival) and implications of these effects for population dynamics. Experiments were conducted to identify effects of a one-week pulse of imidacloprid on all endpoints of the effect cascade for Daphnia magna under surplus and reduced food availability. Concentrations inhibiting feeding by 5, 50 and 95% after one day of exposure were 0.19, 1.83 and 8.70 mg/L, respectively. Surplus food availability after inhibition allowed recovery following growth inhibition of up to 53±11%. Limited food availability provokes a loss of recovery potential even when feeding inhibition did not exceed 5%. A shift in the distribution of energy reserves towards reproduction rather than growth was the driving factor, resulting in changed reproduction after exposure to all concentrations tested; increased or decreased reproduction occurred depending on the intensity of effect on feeding. We determined that all effects beyond feeding depression were secondary using the individual based Daphnia magna population model IDamP as a virtual laboratory. Additionally, we assessed the effect of similar pulses to populations, applying imidacloprid at different developmental stages (different food supply due to intra-specific competition); and investigated the implications for response to subsequent stress comprising a one-day exposure to carbaryl. Inhibition of feeding by 3% switched the population from negative to positive growth resulting in significantly increased population abundance. Inhibition by 97% reduced the total abundance by 56±7% within three days and thus transferred the populations from a phase of food limitation for individuals to a phase of high food availability. This transfer occurred from feeding inhibition not causing mortality when tested at the individual level and increased the sensitivity to carbaryl; resulting in a reduction of the population by 53±14%; this impact was four times stronger than that to populations not inhibited in their feeding. 136 Effects of pesticide exposure on zooplankton dormant egg bank dynamics: from laboratory to mesocosm studiesS. Navis, Laboratory of Aquatic Ecology Evolutionary Biology KU Leuven / Laboratory of Aquatic Ecology, Evolution and Conservation; A. Waterkeyn, L. De Meester, L. Brendonck, KU Leuven. Many aquatic invertebrate species produce dormant eggs to survive unfavourable

35

environmental conditions. These dormant eggs accumulate in the sediment to form a dormant egg bank, from which only a fraction hatches during each growing season. When environmental conditions fluctuate, egg banks can function as a reservoir of species and genetic diversity. Through this benthic-pelagic coupling, events in the dormant phase can affect the active, aquatic phase and vice versa. Despite its importance in ecological and evolutionary processes, dormant egg bank dynamics are rarely included in zooplankton population and community studies. Daphnia magna, a well established model organism and standard test species in ecotoxicology, reproduces by cyclical parthenogenesis, where environmental cues associated with unfavourable conditions trigger the sexual production of dormant eggs. Standard ecotoxicity tests with Daphnia (OECD TG 202, 211) generally focus on the effects of chemicals on the asexual part of the reproduction cycle. However, there is almost no information available on the effects of pollution on dormant eggs of D. magna, or ondormant egg bank dynamics in general. To get more insight into the acute and chronic effects of pesticides on D. magna dormant eggs (ephippia), we have conducted two series of laboratory experiments. In a first series we focused on the effects of pesticides on embryonic development and hatching characteristics of the dormant eggs. In the second experiment we tested whether exposure to pesticides, during the time the eggs were also exposed to hatching cues, could have long term effects on survival and life history characteristics of the hatched neonates. In addition, to better understand the environmental relevance of these findings, we have conducted a two-year outdoor mesocosm experiment, in which the long-term effects of repeated pesticide exposure on both the active and dormant component of zooplankton communities were studied. Our results show that, depending on their mode of action, pesticides can have severe negative effects on hatching characteristics of D. magna dormant eggs (ephippia), as well as on survival, growth and reproduction of the hatched neonates. This indicates that, in addition to inducing mortality of active individuals, pesticides can affect zooplankton communities by altering hatching dynamics and life history traits of hatched individuals. 137 Feeding activity and oxidative stress in Daphnia magna S.M. Furuhagen, B. Liewenborg, Stockholm University / Department of applied environmental science; M. Breitholtz, Department of applied environmental science; E. Gorokhova, Stockholm University / Department of applied environmental science. Caloric intake is an important factor influencing oxidative status, as increased caloric intake leads to enhanced metabolic rates and generation of reactive oxygen species. As biomarkers of oxidative stress are frequently used as indicators of exposure and toxicological effects, it is important to establish the connection between these biomarkers and basic physiological rates, such as feeding and metabolism. The aim of this study was to deliniate the effects of feeding on anti-oxidative capacity and oxidative damage, measured as lipid peroxidation, in Daphnia magna from the effects attributed to toxicity. The pesticide lindane was used to test whether toxic exposure affects feeding rate and the relationships between feeding activity and oxidative biomarkers. Results show that feeding rate has a significant positive effect on protein content, which in turn is positively correlated to both anti-oxidative capacity, measured as ORAC (oxygen radical absorbance capacity), and lipid peroxidation, assayed as TBARS (thiobarbituric acid reactive substances). Lindane exposure had a significant negative effect on feeding rate, however did not alter the relationship between feeding rate and protein content. Moreover, lindane negatively affected the relationship between ORAC and protein content, whereas the relationship between TBARS and protein was unaffected by the toxicant. The ratio ORAC/protein can thus be used as a biomarker of anti-oxidant response to toxic stress. In contrast, TBARS/protein ratio is rather an indicator of altered feeding rates than of toxic effects. The results show that oxidative response to toxic substances is largely mediated via alterations in protein synthesis, which, in turn, is positively related to feeding activity. Therefore, feeding activity and protein content have to be considered when the effects of toxic substances on these biomarkers are evaluated. Otherwise, biomarkers normalized to protein content may be erroneously interpreted as toxicity effects

SETAC Europe 23rd Annual Meeting

whereas in fact they reflect only alterations in feeding. 138 The effect of temperature on cadmium kinetics in Folsomia candida (Collembola) D.M. Jevtic, Institute of Environmental Sciences Jagiellonian Univeristy / Ecotoxicology and Stress Ecology; J.B. Schmidt, Roskilde University / Department of Environmental, Social and Spatial Change; V.E. Forbes, University of Nebraska Lincoln / School of Biological Sciences; R. Laskowski, Jagiellonian University / Ecotoxicology Stress Ecology Group. The importance of temperature in (eco)toxicological tests has been long recognized. Still, ecological risk assessment (ERA) protocols are based on tests conducted in one constant temperature. This approach lacks ecological realism and possibly leads to erroneous conclusions on effects of temperature in natural environments. Understanding toxicokinetic processes and applying toxicokinetic models to the analysis of (eco)toxicity data is one of the most useful approaches in linking exposure concentrations to the effects of toxicants. In this study individuals of the springtail Folsomia candida were exposed to cadmium-spiked OECD soil (100 mg/kg dry wt) under five different temperature regimes – two fluctuating and three constant temperatures. Fluctuating regimes represented daily fluctuations – high (from 9°C to 26.5°C) and moderate (from 15°C to 25°C). Constant temperatures corresponded to 15°C, 19.5°C and 25°C. The constant temperature of 19.5°C was chosen based on equalizing organisms’ daily energy budgets with fluctuating regimes, taking into account that metabolic activity is not linearly related to temperature. The time-dependent internal concentration of cadmium was measured at ten time points during the exposure phase (28 days) and at four time points upon transfer to an uncontaminated soil (14 days). Accumulation (k ) a and elimination rate constants (k ) were estimated by fitting the time e course of internal concentrations to a one compartment first-order toxicokinetic model. Our results indicate that cadmium kinetics in fluctuating environments follows a different pattern compared to the corresponding constant temperature. Accumulation and elimination rates were lower in fluctuating regimes compared to constant temperatures. In contrast, the body burden was higher in fluctuating temperatures, suggesting that ecotoxicological tests conducted at a constant temperature may not adequately predict bioaccumulation. In order to reduce the uncertainty in current ERA approaches, methods have to be revised – either by measuring toxic effects under naturally varying temperature conditions, or by developing and applying appropriate mechanistic effect models able to accurately predict toxic effects in different temperature regimes. 139 Plastic as a carrier of POPs to aquatic organisms. A model analysis. A.A. Koelmans, Wageningen University / Environment; E. Besseling, A. Wegner, Wageningen University; E. Foekema, Wageningen IMARES. It has been hypothesised that persistent organic pollutants (POPs) in microplastic may pose a risk to aquatic organisms. Here, we present a conceptual model for bioaccumulation of POPs including uptake from water, food and ingested plastic. The model accounts for dilution of exposure concentration by sorption of POPs to plastic (POP 'dilution'), increased bioaccumulation by ingestion of plastic containing POPs ('carrier'), and decreased bioaccumulation by ingestion of clean plastic ('cleaning'). A dynamically modelled absorption efficiency from plastic is proposed, which is calculated from particle size, POP polymer diffusivities, the time variable gradient between plastic and organism POP concentrations, and gut retention time. The model is parameterised for the lugworm Arenicola marina and evaluated against recently published polychlorinated biphenyl (PCB) bioaccumulation data for this species from laboratory bioassays with polystyrene microplastic. Further scenarios include polyethylene microplastic, nano-sized plastic and open marine systems. Implications for species with longer food retention times are discussed. Model analysis shows that plastic with low affinity for POPs like polystyrene will have small negative effects on bioaccumulation, governed by dilution. However, for stronger sorbents like polyethylene, the dilution, carrier and cleaning mechanism are relevant. In closed systems as in laboratory bioassays, dilution and cleaning dominate, leading to decreased bioaccumulation. However, in open marine systems, dilution

36

will be marginal and increased uptake of plastic bound POPs is predicted. 140 Characteristic of spatiotemporal distribution of microplastics in surface microlayer in southern coast of South Koreay. Song, Oil and POPs research group; N. Heo, M. Jang, Korea Institute of Ocean Science and Technology; S. Hong, Korea Institute of Ocean Science and Technology / Oil and POPs research group; G. Han, Korea Institute of Ocean Science and Technology; W. Shim, Korea Institute of Ocean Science and Technology / Oil and POP Research Group. Microplastics, less than 1 mm in size, have been recently recognized as marine pollutants of significant concern due to their persistence, ubiquity, toxic potential, and their ability to act as vectors for transfer of absorbed and additive toxic chemicals to marine organisms. Floating of plastics on sea surface is affected by the density of the plastic materials. For example, polyethylene, polypropylene and expanded polystyrene which are generally less dense than sea water is likely to become floating debris. There is a microlayer in the sea surface formed by the surface tension of water with thickness of 1 mm. Sea surface microlayer is a habitat of a variety of life and accumulates light particles as well as pollutants deposited from the atmosphere and buoyant from water column. So the place is used for researching of pollutants. Considering the size and specific gravity of the microplastics, they are also expected to be accumulated within the microlayer. In this study, microplastic debris was quantitatively determined in surface microlayer at 21 stations in southern coast of South Korea in May (dry season) and July (rainy season), 2012, which is receiving Nakdong River discharge. The microplastics in the microlayer samples were extracted using the surface tension. The specially fitted 2 mm mesh sieve was dipped into the sea surface for 100 times, then the trapped water within mesh space was collected in the range of 2.2-2.8 L per site in the stainless steel tray and transferred to the 1 L polyethylene bottle. In the laboratory, the microlayer sample was filtered, using a glass fiber filter and counted using a dissecting microscope. Microplastic abundances in microlayer are in the range of 50-110 particles/L in May and 55-132 particles/L in July. Even if the average of microplastics (110±45 particles/L) in May was higher than those (132±106 particles/L) in July, there is not a statistical signification. Among four categories of microplastics (fragment, fiber, sheet and spherule), fragment type accounted for 80% in May and 98% in July. Small size microplastics down to 0-100 mm class are dominant in all the samples. Microplastics are relatively abundant at stations near shore in comparison with offshore stations. Abundances of microplastics in microlayer were folders or orders of magnitude higher than those in top 20 cm surface water collected by filtering (0.7 mm pore), a hand net (50 mm mesh) and a manta trawl net (330 mm). 141 Sorption, desorption and bioavailability of persistent organic pollutants by microplastics in the marine environmentA. Bakir, University of Plymouth Enterprise ltd / Science and engineering; S. Wright, Biosciences; S.J. Rowland, University of Plymouth / SoGEES; T.S. Galloway, University of Exeter / Biosciences Department; R.C. Thompson, The Marine Institute, University of Plymouth / School of Biological Sciences. Microplastics are small fragments of marine debris. Such fragments now appear to be widespread in the marine environment and have been reported at the sea surface, on shorelines and on the sea bed. Microplastics have been identified as particles less than 5 mm in diameter, but fragments much smaller (< 20 mm) than this are widely reported including pieces of nylon, polystyrene, polyethylene and PVC. It has been suggested that microplastics present potential mechanisms for the transport of persistent organic pollutants (POPs) and the release of chemical additives from plastics, to organisms. Unplasticised PVC (uPVC) and ultra high molecular weight polyethylene (UHMW PE), in the size range 200 to 250 mm, were investigated for their potential to sorb DDT, phenanthrene (Phe), PFOA and DEHP in seawater. Sorption equilibrium times were determined over 15 days and most contaminants reached equilibrium onto plastic in 24-48 hours. Equilibrium distribution coefficients (K ) were used to d represent sorption capacity of plastic for the pollutants under

SETAC Europe 23rd Annual Meeting

investigation. Desorption kinetics were investigated in seawater and using a gut surfactant to represent gut conditions of marine organisms. Desorption of contaminants from plastic was faster in sodium taurocholate than in seawater but is slower than desorption of organic contaminants from natural sediments. A bioavailability model was also proposed to predict contaminants concentration in the tissues of a range of marine organisms. 142 Exploring the effects of microscopic plastic particles in ecologically-important benthic invertebratesS. Wright, Biosciences; A. Bakir, University of Plymouth Enterprise ltd / Science and engineering; S.J. Rowland, University of Plymouth / School of Geography; R.C. Thompson, University of Plymouth / Science and engineering; T.S. Galloway, University of Exeter / Biosciences Department. Plastic debris at the micro-scale is a widespread element of marine litter. Microplastics have accumulated in oceans and sediments worldwide from low densities to localized ‘hotspots’. Since they occupy the same size fraction as sediment grains and some plankton, microplastics may be ingested by low trophic fauna adopting indiscriminate feeding strategies, with uncertain consequences for the health of the organism. The current work aims to determine the extent to which microplastics affect the behaviour and health of ecologically important invertebrate species. We are using an integrated approach combining sublethal toxicological measurements at the cellular and physiological level alongside behavioural responses to assess the shortand long-term implications of microplastics in vivo. We are also considering the effects on growth to assess population-level impacts. At present, we have attained data regarding the low and high density impacts of virgin microplastics on the sediment-dwelling polychaete Arenicola marina, representative of environmental and worst-case scenarios. Significant effects have been found at the cellular level. We now hope to quantify the impacts of different types of plastics, varying in shape and polymer-type. In addition, we are assessing the capacity for microplastics to facilitate contaminant transfer to organisms and whether this occurs at levels capable of causing harm. The results will offer insight into the potential impacts of marine litter on the fundamental physiological processes in these important components of marine food webs, indicating the level of risk microplastic debris poses to the marine environment. 143 Micro-CT as a novel and innovative technique in microplastic research L. Van Cauwenberghe, Ghent University / Laboratory of Environmental Toxicology and Aquatic Ecology; M. Dierick, Ghent University / Department of Physics and Astronomy; C. Janssen, University of Ghent / Laboratory of Environmental Toxicology and Aquatic Ecology. Microplastics, the degradation product of larger plastic debris, are accumulating in marine habitats worldwide. Given the small dimensions of microplastics (< 1mm) these particles can potentially be ingested by a wide array of marine animals. Especially marine invertebrates, commonly not affected by the larger marine debris, are threatened by these microplastics. Data on the ingestion, and subsequent translocation to the tissues, of ingested microplastics in invertebrates is, however, scarce. This is, amongst others, due to methodological challenges associated with the detection of these minuscule plastic particles. Traditional microscopic techniques, such as light microscopy of histological sections, are commonly confronted with distortions due to intensive specimen preparation and difficulties in reconstructing samples. This invasive technique will consequently lead to the loss of key information. These issues can be overcome with the help of CT-scanning. Here, X-rays are used to create cross-sections of a sample from which a virtual model can than be recreated. Especially micro-CT provides a very high resolution (up to 1µm for small samples) and hence high accuracy. Additionally, this high spatial resolution can be obtained without destruction of the sample and without invasive sample preparation. In this study, we explored the use of micro-CT as a novel technique in microplastic research. In particular, we looked into micro-CT techniques to identify the transport of ingested microplastics. Using radiopaque microspheres, i.e. microspheres containing barium sulphate (polyethylene 1.5 g/cc 10-20µm), we visualised transport in the

37

intestine after ingestion of the microspheres. Using this technique we assessed the possible translocation of ingested micrpoplastics through the gut wall and into the tissues of exposed animals. This research is currently ongoing. 144 “F53B” an overlooked PFC produced in China for 40 years c. harman, NIVA / Environmental Contaminants; J. Huang, S. Wang, Y. Yang, Y. Ge, Tsinghua University; T. Larssen, NIVA. Perfluorooctane sulfonate (PFOS) is a well studied compound due to concerns about its toxicity and persistance in the environment which have led to various national and international controls, for example inclusion in the Stockholm Convention. Whilst PFOS has largely been replaced by less toxic alternatives, concern has focussed towards China where production persists. However another very similar compound, which is almost unreported in the scientific literature has been produced and used as a mist surpressant in the chromeplating industry in China for over 40 years. Synthesized in 1970s this compound, locally called F-53B is the potassium salt of a one-chlorinated polyfluoroethanesulfonic acid. Its chemical formula is C ClF KO S and it has only this year received a 8 16 4 CAS No. (73606-19-6). The amount of F-53B produced and used in China is quite comparable with PFOS, and may be increasing since international controls placed on PFOS are a problem that China must face in the coming future. However, to our knowledge there are no reports concening the existence of F-53B, its presence in the environment or its toxicity. Thus the aim of this study was to provide the first data in this regard. Acute toxicity was determined using Zebrafish (Brachydanio rerio) according to OCED guidelines. The 72h-LC50 was found to be 17.72 mg/L, and 96h-LC50 was 12.18 mg/L. Grab samples were taken at several points above and below a municipal wastewater treatment plant situated on the Oujiang River, Wenzhou city, which receives (treated) wastewater from the electroplating industry and samples were also taken at treatment plant directly associated with the industrial site. Levels in the treatment plant at the industrial site were -1 unsurprisingly very high, in the order of 10s of µg L and both PFOS and F-53B were found in the receiving waters near the treatment plant in -1 similar amounts to each other (10s of ng L ). In conclusion F-53B shows similar properties to PFOS in terms of toxicity, it is present in the environment and has likely been there for some decades. Results will be further discussed in terms of on-going degradation and bio-accumulation studies. There is an urgent need for more information concerning this PFC. 145 Stockholm Arlanda Airport as a Source of Per- and Polyfluoroalkyl Substances to Water, Sediment and FishL. Ahrens, Swedish University of Agricultural Sciences SLU / Dept of Aquatic Sciences and Assessment; K. Norstrom, T. Viktor, IVL Swedish Environmental Research Institute; S. Josefsson, SLU / Dept. of Aquatic Sciences and Assessment. Per- and polyfluoroalkyl substances (PFASs) have been used in a variety of consumer and industrial applications such as aqueous fire fighting foams (AFFFs). AFFFs have been used at fire training facilities at airports and oil refineries since the 1970s and are a potential source of PFASs in the nearby environment. In this study, PFASs were measured in water, sediment and fish near a fire training facility at Stockholm Arlanda Airport in Sweden. Samples were analyzed for 11 PFASs including C –C perfluoroalkyl carboxylates 6 11 (PFCAs), C , C , C perfluoroalkyl sulfonates (PFSAs), 6 8 10 perfluorooctanesulfonamide (PFOSA) and 6:2 fluorotelomer sulfonate (FTS). The concentration levels and pattern of PFASs showed a high variety depending on their spatial distribution and matrices (i.e., water, sediment, fish). However, perfluorooctane sulfonate (PFOS) was the dominant compound in all compartments with a contribution of 40% for surface water, 86% for sediment and 98% for European perch (Perca fluviatilis). Highest PFAS concentrations were found in water at a ditch close to the fire training facility with about 4000 ng/L for ?PFASs. The PFAS concentrations decreased continuously from the fire training facility to lake Mälaren. During the transport, PFASs can partition to sediment or bioaccumulate in the food chain. The highest sediment/water partition coefficients (log K ) were found for PFOS d (2.13) and PFOSA (2.99). Similarly, PFOS and PFOSA had also the

SETAC Europe 23rd Annual Meeting

highest bioconcentration factors (BCF) with, on average, 3320 and 2310 for muscle tissue, respectively. PFAS concentrations in the water phase at a nearby lake did not show a decreasing trend over the last years which indicate that Stockholm Arlanda Airport may be an important source for long term contamination of the nearby environment with PFASs. 146 Occurrence and Concentrations of Contaminants of Emerging Concern in Wastewater Matrices across United StatesK. Dasu, US Environmental Protection Agency National Risk Management Research Laboratory / Department of Agronomy, Crop, Soil and Environmental Sciences; M.A. Mills, US EPA; L. Zintek, US Environmental Protection Agency / Department of Chemistry; W. Brashear, PTS; K. Tadele, B. Crone, Student services, USEPA. Contaminants of emerging concern (CECs) are detected globally in wastewater treatment plant (WWTP) matrices such as effluents and biosolids. Wastewater matrices may play an important role in the transport of these contaminants into the environment. Some of the CEC are environmentally persistent, and are shown to have endocrine disrupting effects and other harmful effects to the aquatic ecosystems and human health. Hence, there is a growing concern about the potential risks associated with their presence during water reuse and the landapplication of biosolids. Once land-applied, the emerging contaminants present in biosolids can enter the surface and ground waters or carry over to the plants grown on such soils and this increases the potential exposure of humans and aquatic ecosystems to these compounds. The main objective of the current study is to quantify the occurrence and concentrations of CECs and their transformation products and/or precursors in treated wastewater effluents and biosolids. Differences in treatment performance will be evaluated based on the concentrations of contaminants detected in biosolids from different origin (i.e. aerobic or anaerobic digestion, presence of additives and other factors). Classes of chemicals to be monitored include perfluoroalkylated substances and their precursors, alkylphenol ethoxylates and alkylphenols, steroid hormones and pharmaceuticals and personal care products. Effluents and biosolid grab samples were collected from 9 wastewater treatment plants around the United States. Biosolids from 3 plants have undergone aerobic digestion and 6 plants have undergone anaerobic digestion treatment process. The effluent and biosolid samples are extracted separately for different classes of contaminants and are analyzed on UPLC/MS/MS and GC/MS/MS for different classes of analytes. The data will aid in understanding the fate of CECs in wastewater matrices and the results will provide insightful information for the risk management of these chemicals in the environment. From the current study, the data on the occurrence and concentrations of different CECs and their precursors or metabolites in effluents and biosolids from 9 WWTPs will be discussed in detail. 147 Uptake of perfluorinated alkyl acids in crops via land applied biosolids: Field and greenhouse studies A.C. Blaine, Colorado School of Mines / Civil & Environmental Engineering; C.D. Rich, L. Kudryk, Colorado School of Mines; L.S. Hundal, Metropolitan Water Reclamation District of Greater Chicago; C. Lau, M.A. Mills, U.S. EPA / Office of Research and Development; K.M. Harris, U.S. EPA / Region 5; C.P. Higgins, Colorado School of Mines / Civil & Environmental Engineering. Perfluoroalkyl acids (PFAAs) consist of a carbon backbone saturated with fluorine atoms in place of hydrogen. Of particular concern are PFAAs that contain either a carboxylate or sulfonate group, resulting in strong surfactant behavior. Their dual lipophobic and hydrophobic nature makes them both oil and water repellent, giving them a myriad of applications in both industrial and consumer settings. Many PFAAs are also environmentally persistent, bioaccumulative, and toxic making them a high priority contaminant of emerging concern. The presence of PFAAs in municipal biosolids has been well documented. Biosolids, like animal manures, are rich in both plant nutrients and organic matter and are commonly used as a fertilizer in crop production. As a consequence, concerns have arisen about the potential uptake and subsequent bioaccumulation of PFAAs into crops grown in biosolids-amended soils. Previous studies have documented

38

the potential for bioaccumulation of PFAAs, particularly perfluorooctane sulfonate (PFOS) and perfluorooctanoate (PFOA) into food crops, while other field studies have documented the transfer of these and other PFAAs from industrially-contaminated biosolidsamended soils into grass. Additional literature has documented the transfer of PFAAs from spiked hydroponic systems into lettuce. The present study included both field and greenhouse experiments to evaluate the potential for PFAA uptake and bioaccumulation into crops grown in biosolids-amended soils. Results from this study demonstrate that lettuce, tomato, zucchini, and corn plants can uptake PFAAs, and that bioaccumulation is dependent on chain length, crop type, and soil properties. These results may have important implications with respect to the potential routes of PFAA exposure in humans. Disclaimer: This abstract does not necessarily reflect U.S. EPA policy. 148 Uptake of perfluorinated alkyl acids by hydroponically and field grown crops S. Felizeter, University of Amsterdam; M.S. McLachlan, Stockholm University; H. Juerling, J. Mueller, Fraunhofer Institute; P. de Voogt, University of Amsterdam / IBED. Perfluorinated alkyl acids (PFAAs) are bioaccumulative persistent, organic pollutants (POPs), which can be detected ubiquitously in the environment. PFAAs pose a risk to human health due to accumulation in the food chain. The occurrence of PFAAs in animals, such as fish, birds and mammals including humans is fairly well documented, but little can be found in the literature about crops or plants in general. Also, most studies focus just on the two main compounds perfluorooctane sulfonate (PFOS) and perfluorooctanoic acid (PFOA). Humans are possibly exposed to PFAAs through consumption of vegetables and other plant-related food items. The objective of this study is to understand the accumulation process of PFAAs in crops. In greenhouse experiments the uptake of Cabbage and Lettuce as leafy vegetables were tested to prove or disprove this hypothesis. Furthermore Tomato and Zucchini were analyzed to test for accumulation in fruits. These four plants will be tested using hydroponical culturing conditions and employing varying concentrations of PFAAs. Uptake factors were calculated for 14 different PFAAs. While short-chained PFAAs show higher Uptake factors in the Foliage than in the roots, longer-chained PFAAs predominantly accumulate in the roots. Furthermore, the uptake behavior of PFAAs cannot be described with existing uptake models. In a field experiment lettuce, radish, peas and corn were grown in lysimeters in 4 different concentrations of spiked soil to have a comparison to the greenhouse experiment. The results of the concentrations in the different parts of the plants show a different pattern than in the greenhouse experiment with higher concentrations in the foliage part for most of the compounds. 149 Polyfluorinated chemical exposure in blood from mothers and infants of the Canadian Healthy Infant Longitudinal Development Study C.S. Wong, University of Winnipeg / Richardson College for the Environment; C. McConkey, M. Loewen, A. Becker, University of Manitoba. Concentrations of 17 perfluorinated compounds (PFCs) were measured in the blood of volunteers from Winnipeg, Manitoba, Canada, of the Canadian Healthy Infant Longitudinal Development (CHILD) study, to assess exposure levels and chemical distributions, and potential correlations of these with maternal and home characteristics from CHILD. Chemical measurements of PFCs were done for plasma from 414 women during pregnancy, 247 women one year after delivery, and 50 infants at the time of birth via cord blood. Mean levels of perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS) were 1.1 and 2.6 ng/mL, respectively, in prenatal plasma, and 0.7 and 2.2 ng/mL, respectively, in postnatal plasma. These were similar to that observed in other human monitoring studies done elsewhere in the world, and were generally lower than those studies done earlier in the last decade or so, indicating possible decreases in PFOA and PFOS from decreasing primary use. Perfluoropentanoic acid (PFPA) had median concentrations of 0.5 ng/mL. This chemical was not detected in earlier studies, indicating increased exposure to this short-chain PFC used in recent manufacturing. Most perfluorocarboxylic and perfluorosulfonic acids were lower in matched maternal postnatal samples compared to that in prenatal blood (n = 161), except for PFPA for which the opposite

SETAC Europe 23rd Annual Meeting

was observed. Cord plasma concentrations of PFCs were generally lower than that of maternal plasma (e.g., mean 0.7 and 1.0 ng/mL for PFOA and PFOS, respectively), with increasing maternal transfer for perfluorocarboxylic acids with shorter chain length. Principal components analysis indicated that there was a correlation between known PFC half-lives and the age of the mothers and the number of babies they had delivered, as well as for greater maternal transfer for postnatal samples and for mothers that had breastfed their infants for longer periods of time. There was no correlation between incidence of wheezing in infants and PFC levels, except for PFOA, nor evidence for linkages between plasma levels of PFCs and various developmental measures (e.g., birth length, ponderal index, head circumference, gestational age). No significant correlations were observed between PFC plasma levels and home characteristics (e.g., carpeting, home renovations, new furniture), suggesting exposure to cohort participants was ubiquitous. 150 The necessity of disinfected toilet brushes – Information policy of companies on disinfectants in private homes S. Wieck, . The use of disinfectants in private homes is intensely debated among scientists. It is widely accepted that disinfectants should only be used by private persons if there is a medical indication (Bockmühl, 2011). These might be infectious diseases of family members or if people in need of care are living in the household. However, companies are promoting disinfectants for several areas of use apart from medical indications, e.g. disinfection of toilet brushes or hair rollers (Colgate-Palmolive, 2012). These uses have no beneficial effects for consumers and lead to a needless exposure of consumers to the active substances and unnecessary releases of chemicals to the environment. This causes several problems for the human health and the environment. In the project websites of 14 companies were analyzed with content analyses with regard to the information provided for consumers and whether they abode by the regulations concerning the advertisement of biocides. The results were compared to results of similar analyses of websites of German authorities engaged in the authorization of biocides. The analyses show that the information provided by the company websites is not enough to fulfill the information needs of consumers. Details on risks caused by disinfectants are missing and not all companies abide the regulations on advertisement. In the following prosecutions by the responsible federal state authorities even more violations were detected. The websites of the German authorities informed the consumer comprehensively. However, the texts often seemed hard to understand for laymen. The company websites do not educate the consumers properly about the use and the risks of disinfectants; other stakeholders involved in consumer education have to undertake this task. The German authorities provide a wide range of relevant information but the comprehensibility has to be improved to be understandable for interested consumers. Eventually, stricter regulations on the advertisement of disinfectants should be considered to support consumers in their decisions of buying and using disinfectants. This poster does not necessarily reflect the opinion or the policies of the German Federal Environment Agency. Bockmühl, Dirk: Welche Zukunft haben antimikrobielle Produkte im Haushalt? In: Hygiene + Medizin Bd. 36 (2011), Nr. 1&2, S. 12–18 Colgate-Palmolive. 2012. Tipps für Verbraucher. DanKlorix. [Citation from: 23. 11 2012.] http://www.danklorix.de/tipps_desinfektion.php. 151 Experience with Risk Communication in different Estuaries and Lagoons I. Stresius, Fachhochschule Lübeck / Laboratory of Urban Water and Waste Management; S. Heise, Hamburg University of Applied Sciences; M. Grottker, University of Applied Science Lübeck. Management and communication of environmental risks in complex natural systems like estuaries and lagoons implicates an enormous challenge for decisionmaker, stakeholder, experts and the general public. Stakeholders and decisionmakers often have to contend with multidisciplinary, multidimensional processes in systems with long time delays, positive feedbacks, non linear cause-and effect relations and uncertainties. For effective management and communication of complex tasks a deep understanding of the system and the mental models of all

39

participants of the process is required. For sustainable planning and management of complex natural systems a integration of existing scientific knowledge is of vital importance. To communicate scientific results and information about risks to non-scientists in a way that brings benefits for everybody is the ultimate challenge. In the Interreg IVBproject diPol the communication and management tool SIMACLIM was developed and used in four case sides in Oslo, Gothenburg, Copenhagen and Hamburg in several stakeholder workshops. SIMACLIM provides assistance to regional and national decision makers to illustrate the multiple impacts on ecological quality, to evaluate consequences of potential measures and to adapt their management to the changing th climate. In the project ARCH, a framework 7 programme, SIMACLIM is used to employ a true participatory process with stakeholders including policy makers, authorities, NGO’s and businesses to formulate realistic strategies towards sustainable lagoon management. Communication between science and and stakeholder, policy and the general public as the beneficiary of scientific results needs to be a multiy-way process. Scientists must listen to the public or other "consumers" of their scientific results like for example risk messages to get to know the knowledge level of their audience, to find the right language and to find out what kind of information is needed. An evaluation of the effectiveness of communication should guide the process. 152 How does the public react when scientists disagree? Scientific consensus and risk communication A. Hunka, A. Palmqvist, Roskilde University / Department of Environmental Social and Spatial Change; V.E. Forbes, University of Nebraska Lincoln / School of Biological Sciences. The way people perceive risks is an important element of risk communication, that has to be considered by media professionals and food safety authorities alike. Non-experts in risk assessment have to rely on the expertise of others. Therefore, the role of scientific consensus and trust in the judgement of experts seem to lie at the core of risk perception and successful risk communication with the general public. Yet, studies on the role of scientific consensus in th perceiving risk are scarce. On the 19 of September, a study on long term toxicity of herbicide-tolerant, genetically modified maize NK603 diet in Sprague Dawley rats was published online by the journal of Food and Chemical Toxicology. The results, illustrated by pictures of malformed rat females quickly found their way into leading popular internet media and triggered a heated debate within the scientific community, among European food risk authorities, and anti-GMO campaigners. So far the majority of the scientific critics have agreed that Seralini et al’s study displayed some serious methodological flaws, while the first EFSA review of the study showed that Seralini et al.’s methods did not meet OECD standards. In our study we focus on public reactions to scientific controversy surrounding the study of Seralini’s team, using this example as a case study. Our aim is to link perceived risk from GM organisms with the role of scientific consensus in risk communication. We employed a mixed-method approach, combining thematic analysis with a survey. We have collected materials from web pages on Englishspeaking internet sites and launched a web-based survey to study the general public’s response to the GMO controversy. Our preliminary results indicate that the lack of scientific consensus is the main and very serious factor undermining trust both in food safety authorities and the scientific community in general. On the other hand, trust does not only apply to the information source, but to communication content as well. Many studies on risk perception show a very strong bias towards negative information occurring regardless of trust in the information source. The different tone of media reports in the U.S. and Europe indicate that initial attitudes towards GMO play a crucial role in risk communication. Finally, media also fine-tune their messages to readers’ reactions, reinforcing initial beliefs of the public. Scientific consensus appears to be a promising way to break this vicious circle. 153 Discussing environmental issues from different disciplinary perspectives. The first step for public communication on science. F. Lafaye, Université de Lyon / ENTPE; P. Garrigues, University of Bordeaux / Institute for Molecular Sciences; N. Léca, Université

SETAC Europe 23rd Annual Meeting

Montesquieu Bordeaux 4 / CRDEI. Environmental questions require the commitment of various research fields and disciplines (chemistry, toxicology, anthropology, laws). Most of the times, these disciplines remain within their own research field and when put together, the results look like split up. But dialog with the public needs to answer the questions globally and in a practical way. What are the needs for an integrated approach which allows everybody to understand scientific procedure and results ? Interdisciplinarity implies to take time and to get committed both the disciplines and the researchers . Beyong the risks taken to spread their research fields, the researchers must also undertake a deep dialog based on a strong willing to understand and to respect each other. 154 Plenary discussion: How can we communicate to improve our outreach? P. Calow, University of Nebraska Lincoln / ORED. Environmental sciences are at the heart of what people affect in their daily lifes: environmental quality, safe food, clean air, fresh water - and hence crucial for sound public health. Why aren't we in their daily minds? How can we get there? Why should we want to? Moderated by Peter Calow and together with the presenters and external guests we want to discuss the current and future state of science and risk communication in environmental sciences. Panel members are: - Valery Forbes, University of Nebraska Lincoln - Ursula Klaschka, Hochschule Ulm - John Redshaw, SEPA - Birgit Sokull-Kluettgen, European Commission JRC - Ken MacDonald, BBC Science Correspondent 155 Biomonitoring of an iron mine in the Arctic Circle using field transplanted mussels S. Brooks, NIVA / Ecotoxicology and Risk Assessment; C. Harman, Norwegian Institute for Water Research (NIVA); M. Hultman, Norwegian Institute for Water Research; K. Macrae; J. Berg, NIVA. The blue mussel (Mytilus spp.) has been used to assess the potential biological effects of the discharge from the Sydvaranger mine, which releases its tailings into Bøkfjorden at Kirkenes in the north of Norway. Sensitive health biomarkers and metal bioaccumulation were measured in mussels caged at varying distances from the discharge outlet in Bøkfjorden. The biomarkers used include: Stress on stress; condition index; cellular energy allocation; micronuclei formation; lysosomal membrane stability by neutral red retention; basophilic cell volume; and neutral lipid accumulation. The individual biomarkers were integrated using the integrated biological response index (IBR/n). The accumulation of Fe was significantly higher in mussels located closer to the discharge outlet, indicating that these mussels had been exposed to the mine effluent. The IBR/n results were in good agreement with the location of the mussels in relation to the distance from the discharge outlet. The biomarker responses were not severe but did exhibit effects for some of the biomarkers resulting in a higher IBR/n in the mussels closest to the outlet. However, based on the individual biomarker responses the overall biological effects measured in the exposed mussels were considered low. 157 Using WHAM-FTOX to assess the toxicity of mining-impacted freshwater systems S. Lofts, Centre for Ecology Hydrology / Shore Section; E. Tipping, NERC Centre for Ecology and Hydrology; A. Stockdale, University of Mancheser; S. Ormerod, Cardiff University; W. Clements, Colorado State University; R. Blust, University of Antwerp. Freshwater systems impacted by metal mining are predominantly characterized by contamination with multiple metals and in many cases low pH. Tools to predict the toxicity of such systems therefore need to account for the toxicity of metals and acidity, and should also account for bioavailability of metals. The WHAM-F model, described here, TOX is designed to perform this task. The model assumes that toxicity results from the binding of metals and protons to receptor sites on freshwater organisms, and that the binding of these ions to humic acid correlates to the organismal binding. Effects on field communities are quantified by fitting the model to data using piecewise quantile regression. The model has been parameterised using datasets from the UK, Japan and the United States. The parameterised model may be used to predict how water chemistry changes over time (resulting, for example, from remediation actions) may be expected to affect field communities.

40

158 Bioavailability and toxicity of metals to Crassostrea gigas larvae in an estuary affected by mining C.B. Braungardt, Plymouth University / School of Geography, Earth and Environmental Sciences; C. Money, AstraZeneca / Brixham Environmental Laboratory; P. Worsfold, Plymouth University / School of Geography, Earth and Environmental Sciences; E. Achterberg, National Oceanographic Center. Proposed environmental quality standards (EQS) for metals in estuarine waters require renewed attention to two concepts: (i) bioavailability and (ii) the dynamic nature of the water body. Detailed understanding of estuarine processes that affect metal speciation and hence bioavailability, at appropriate spatial and temporal resolution, is required in order to set up appropriate monitoring strategies for EQS compliance. In this study, we deployed the Voltammetric In situ Profiler (VIP) for 10 h in the Fal estuary (SW England) to determine the truly dissolved ( The OEL bioassay response to samples taken in Percuil Creek, located near the mouth of the estuary, was low (up to 11% PNR, n=3) and varied little between samples taken at different times during the survey and indicated that the concentrations of dynamic Cu (3.8-7.0 -1 -1 nmol L ), Cd (0.13-0.34 nmol L ) and other contaminants had limited toxic effects. In contrast, total metal concentrations of Cu (5.0-88 nmol -1 -1 L ) and Cd (0.68-2.3 nmol L ) at Restronguet Point (RP) were elevated, and maxima of both metals were observed just after low water. The increase of Cu around low water occurred mainly in the colloidal phase (68%), while over 88% of Cd was potentially bioavailable at low water. The bioassay response to waters from Restronguet Point (RP) ranged from a minimum of -4% PNR for a sample taken near the time of -1 high water to +93% PNR at low water. Dynamic Cu (6.1-28 nmol L ) -1 and Cd (0.24-2.0 nmol L ) concentrations for the corresponding survey show a similar pattern as the bioassay, indicating that the increase in both, sources of metals and toxicity to the OEL were located upstream of RP, probalby in the Carnon River, which receives acid drainage from an extensive historic mining area. This study shows the spatial and temporal variability of contaminant concentrations, speciation and combined toxicity in estuaries, thus emphasising the need for careful consideration of when, where and how often samples are to be taken for compliance monitoring of the new EU Water Framework Directive EQS.\n\n 159 Utilizing Eisenia andrei to assess the ecotoxicity of platinum mine tailings disposal facilities M. Maboeta, North West University. South Africa is an important platinum mining country which results in environmental impacts due to the construction of tailing disposal facilities (TDFs). It is unclear what the effects of ageing are on the ecotoxicity of TDFs and whether it increases or decreases over time. The aim of this study was to determine the ecotoxicity of differently aged TDFs by investigating earthworm (Eisenia andrei) responses viz. growth, reproduction, neutral red retention times (NRRT) and tissue metal concentrations. Further, to evaluate the status of these in terms of a geoaccumulation index (I ), pollution index (PI) and integrated geo pollution index (IPI). Results indicated that earthworms showed reduced reproductive success (hatchlings per cocoon) and decreased NRRT in all the sites. Juveniles per cocoon between all of the different treatment groups were; control (2.83±0.54) > Site 2 (20 years old; 1.83±0.27) > Sites 1 and 3 (40 years old; 1.06±0.15 and 6 years old; 0.88±0.39). This -1 might be ascribed to the elevated levels of Cr (±200 – 1166 µg g ) and Ni -1 (±100 – 316 µg g ) in all of the sites. Earthworms did not bioaccumulate metals with bioconcentration factors for all the different treatments < 0.01. Studies like these could be useful when establishing a ranking of TDFs in the future to provide legislative institutions with an indication of the environmental liabilities of platinum mines. 160 Sphalerite in agricultural soils: implications for Cd contamination of crops T. Robson, C. Braungardt, Plymouth University / Biogeochemistry Research Centre; J. Rieuwerts, P. Worsfold, Plymouth University. Aerially and fluvially distributed fine mineral particles produced by mining operations can behave as vectors for toxic trace elements (e.g. Cd) in surface environments. Limited data exists on the short-term stability of these particles in agricultural soils

SETAC Europe 23rd Annual Meeting

and the associated risk of producing contaminated crops. Sphalerite (ZnS) is abundant, widely distributed and generally contains 0.2-1 wt. % Cd. The principal intake route for non-smokers is dietary and Cd is readily taken up by plants. This study investigated whether ZnS weathering rates are relevant to crop growth and evaluated the bioavailability of the weathering products to wheat. Acid-neutral soil (Tamar Valley, UK) spiked with ground ZnS (< 63 µm, 0.92 wt. % Cd) at 0.1 wt. % was incubated under laboratory conditions for 7, 30, 90, 180, 270 and 365 days, after which samples were freeze-dried prior to aqueous extraction: 0.1 M EDTA (Cd ) and 0.01 M CaCl EDTA 2 (Cd ). Extracts were filtered (0.45 µm) and analysed for dissolved CaCl2 Cd/Zn (ICP-MS/OES). Analogous incubations were planted with spring wheat seedlings after 180 days incubation. At maturity, plant tissues were freeze-dried, ground and digested (50% v/v HNO ) before tissue 3 Cd concentrations were determined (ICP-MS). Extractable Cd in spiked soils deviated from those in control soils. A linear relationship between Cd and incubation duration indicated steady-state dissolution EDTA -1 -1 occurred (2.81 ± 0.30 nmol Cd kg soil day ), resulting in 1.23 ± 0.26 atm. % ZnS dissolution after 365 days. Tissue Cd concentrations of -1 wheat from spiked soil (24.8 µmol kg ) were a factor of ? 70 of plants -1 grown in the control soil (0.35 µmol kg ). Flux calculations estimate 3.7 ± 0.6 % and 35 ± 5 % of the Cd was taken up into seeds and EDTA stems, respectively. Total uptake was an order of magnitude greater than the predicted Cd pool, suggesting it was readily replenished as Cd CaCl2 -1 was absorbed by the plants. A Cd ingestion rate of 1.25 µmol Cd kg -1 BW month was estimated for consumers of the contaminated wheat seed, 5.6 times current WHO guideline. ZnS undergoes slow, steady dissolution in acid-neutral soils and releases highly bioavailable Cd. Consequently the impact of ZnS soil contamination is long-lived. Wheat grown in ZnS-contaminated soils can accumulate hazardous Cd concentrations in the edible tissues; therefore populations subsiding on crops grown in ZnS-affected soils may be at risk of chronic Cd poisoning. 161 Ecosystem services in risk assessment and -management of pesticides J. Van Wensem, TCB. The Ecosystem services (ES) concept can be used to place pesticide use and the concomitant agricultural management practices in a sustainability context, to evaluate the ecological, economical and societal aspects. Ecosystem services are the benefits people obtain from ecosystems. The ES concept is mainly used to strengthen the position of natural resources in decision-making. ES are highly relevant for agriculture as provision of food, feed and fibre is a major ecosystem service, both in term of wellbeing as in economical terms, and is supported by many other ES. Land is a limited resource, and productive land even more. There is scarcity of land for other uses too, and competition with for instance recreation, water regulation and biodiversity. Especially in densely populated countries there is pressure on agricultural (and urban) land to also deliver other ES. Changes in the EU common agricultural policy may lead to a focus on land sharing, e.g. multiple ES provision by agricultural land. At present the ES concept is in use in risk assessment and risk management of pesticides. The European Food Safety Authority published in 2010 a scientific opinion in which it was proposed to use the ES concept to specify general protection goals mentioned in the new Regulation No 1107/2009 and other relevant regulation(2010). The derivation of these specific protection goals by means of the ES concept allowed a more systematic way of inventorying the taxonomic groups that are important for delivering a suite of ES by agricultural landscapes, and was a onetime exercise. EFSA is currently working on a number of guidance documents (revisions) that build on this opinion. Many diagnostic risk assessment studies describe effects of pesticides on ES in the field, often by using indicators for ES. However, the link with ES is not always made clear, even when there is one. A promising development is the use of the ES concept in agricultural management scenarios, in which pesticide application is part of the management. Some studies use ES valuation (in monetary terms) to determine the optimal scenario. The results of the diagnostic risk assessment studies and the management scenario studies with ES might be of great use in pesticide regulation, despite the present lack of guidance how to apply the ES concept in

41

these areas. 162 Examining new environmental data requirements under Regulation 1107/2009 – Anses opinion V. Poulsen, ANSES. Current data requirements for pesticides, for active substances and plant protection products, are based on Regulations 544/2011 and 545/2011, respectively. Discussions at EU level occurred during the last few years, and final drafts were discussed in July 2012. For environmental fate and ecotoxicology, a couple of new data requirements were added to the previous one. These new requirements and additional tests are intended to increase environmental safety in the framework of pesticide use. Some of them are also intended to have a better understanding on pesticide properties, but to Anses’ opinion, not all of them could be used for risk assessment purposes. Moreover, new guidance documents or update of the existing ones are necessary to develop risk assessment schemes including the new studies in a couple of areas, such as the risk assessment for vertebrates other than birds and mammals, or long range transport. 163 Selection of an additional test species besides Daphnia magna for the effect evaluation of insecticides to invertebrates under Regulation 1107/2009 M. Daam, DPPF/CEER; L. Santos, M.J. Cerejeira, Technical University of Lisbon; T. Brock, Alterra. Lower-tier risk assessments of pesticides for aquatic invertebrates in the EU have been based on routine testing of the crustacean Daphnia magna and (in certain cases) the insect larvae Chironomus spp. An important change in the proposed update under the new Regulation 1107/2009 for the aquatic testing is the inclusion of an acute test for a second aquatic invertebrate (i.e., besides Daphnia) as a basic data requirement for insecticides and substances with insecticidal activity. Recently, Brock and Van Wijngaarden [2012.Environ. Sci. Pollut.Res. 19:3610-3618] evaluated these new requirements by comparing the threshold concentrations for treatment related effects of 31 insecticides, as derived from aquatic micro-/mesocosm tests with the predictive value of the European Tier-1 acute effect assessment on basis of laboratory toxicity tests with Daphnia magna, Chironomus spp., Americamysis bahia and Gammarus pulex. They concluded that the Tier-1 procedure on basis of acute toxicity data (EC50/100) for the combination of Daphnia and A. bahia and/or Chironomus overall is protective to pulsed insecticide exposures in micro-/mesocosms. The number of compounds that can be evaluated with model ecosystem studies and the number of populations of (potentially sensitive) species that are present in these test systems is by definition limited (e.g. species for a specific lentic or lotic community only). To obtain another line of evidence, a different approach was followed in the present study to evaluate the predictive value of the Tier 1 approach on basis the core data-set. To this end, we compiled toxicity data from single species tests evaluating insecticides with aquatic invertebrates and subsequently compared these with that of Daphnia magna, with or without an additional invertebrate species (C. riparius or A. bahia), by applying the assessment factors as set in the first-tier risk assessment in the EU. Implications and remaining uncertainties for the risk assessment of aquatic invertebrates under the new data requirements are discussed. 164 OECD harmonized international guidance and ecoregion crosswalk for pesticide terrestrial field dissipation studies (Part 1) M. Mitchell, Health Canada / Pest Management Regulatory Agency; R. Gangaraju, Health Canada; M. Shamim, Office of Pesticide Programs, US-EPA; I. Nicholson, Health Canada / Pest Management Regulatory Agency; M. Egsmose, EFSA European Food Safety Authority; B. Grenier, OECD. Data on pesticide terrestrial field dissipation/accumulation (TFD) are required by regulatory agencies worldwide for the registration of pest control products. Use ofstudies conducted at foreign sites can strengthen regulatory decisions and reduce burdens for both the regulated and regulatory communities. Studies conducted at foreign sites are generally acceptable if the study objectives, protocol and methodology are harmonized and the study sites represent local use conditions. A chemical is expected to behave similarly in ecoregions that are based on similar soils and climate.

SETAC Europe 23rd Annual Meeting

Neither internationally harmonized TFD guidance or Europe-North American ecoregion comparison maps are currently available. Harmonized TFD guidance (led by the US Environmental Protection Agency) and an ecoregion crosswalk (led by Health Canada) are therefore, being developed under OECD project. Guidance for TFD Studies: In North America, the objective of a TFD study is to determine the fate and behaviour of a chemical when it is used according to label directions in representative use areas and all factors of transformation and transport are acting together. In the European Union (EU), the objective is to determine the DegT and persistence under conditions 50 where surface losses such as runoff, photolysis and volatilization are minimised. The proposed methodology is based on a conceptual model and modular approach. Guidance is provided on a basic study to determine persistence, residue carryover, transformation, leaching, formation and decline of transformation products, DegT , and routes of 50 dissipation, with additional guidance on modules to determine volatilization, surface runoff, leaching to ground water, etc. Ecoregion crosswalk: The second component of the project is the development of ENASGIS (Europe North America Soil Geographic Information for Pesticide Studies). This model identifies similar ecoregions between Europe and North America which will facilitate a study conducted in Europe to be considered by the North American regulators and vice versa. In Europe, FOCUS guidance is used to facilitate use of overseas TFD studies for regulatory purposes in EU. In addition the model permits selection of field sites based on concerns identified by the conceptual model. The use of field studies conducted at foreign sites for national and global joint reviews would strengthen regulatory decisions and reduce regulatory burdens for both registrants and regulators. 165 OECD Harmonized International Guidance and ecoregion Crosswalk for Pesticide Terrestrial Field Dissipation Studies (Part2) m.t. shamim, USEPA / OCSPPOPPUSEPA; R. Gangaraju, Health Canada / EAD PMRA; M. Egsmose, EFSA European Food Safety Authority. The OECD ecoregion crosswalk project was initiated to develop harmonized guidance for conducting terrestrial field dissipation (TFD) studies and to develop an ecoregion crosswalk across North America and Europe. The TFD studies assess the transformation, transport, and fate of pesticides under representative actual use conditions. They also validate and/or refine the relative importance of selected modules of dissipation such as soil abiotic/biotic processes and the processes of leaching, volatilization, run-off, plant uptakes, DegT50 and others. The major objective of this project is to determine if studies conducted at a specific site in North America or Europe can be used across international borders based on the similarity of the ecoregion in North America and Europe. US-EPA has the lead for harmonization of the guidance for terrestrial field dissipation studies while PMRA-Canada has the lead for developing the ecoregion crosswalk component. EFSA is co-lead for both projects. This includes development of the GIS-based ENASGIS (Europe North America Soil Geographic Information for Pesticide Studies) model that is based on the understanding that sitespecific conditions such as soil and climate play major role in determining the fate and behaviour of a pesticide in addition to vegetation, pesticide intrinsic properties and the use pattern. These environmental variables can be categorized into “ecoregions” and the fate and behaviour of a chemical is expected to be similar in ecoregions with similar environmental variables. This presentation will focus on the recommendations from the OECD ecoregion crosswalk workshop held in Ottawa, Canada in March 2011 to solicit input from experts on issues related to harmonization of terrestrial field dissipation guidance and construction of ecoregion crosswalk model, ENASGIS. It will also provide the status of the OECD project and the progress made in finalizing the draft guidance document and updating the ecoregion crosswalk model. Note: The content of this presentation does not necessarily represent the official views of the OECD or of the governments of its member countries and EFSA. 166 Quantifying Soil Surface Photolysis under Conditions Simulating Water Movement in the Field: A New Laboratory Test Design L.H. Hand, Syngenta Limited / Product Metabolism; C.

42

Nichols, S. Kuet, R. Oliver, Syngenta / Product Metabolism; C.M. Harbourt, Waterborne Environmental, Inc.. New guidance on the conduct of terrestrial field dissipation studies in Europe require that surface processes, such as photolysis, are excluded by incorporation of the active ingredient into the bulk soil. The role of surface processes must now be addressed separately, through higher tier studies are required to better understand the contribution of surface processes under field conditions. Since minimal light penetration occurs below the top 1-2 mm of the soil surface, it is often assumed that, once a compound has moved out of this “photolytic zone” following a rainfall event, no further photolysis can occur. However, as both downward and upward water movement occurs under field conditions, there is the potential for relatively mobile compounds to return to the surface, prolonging effective exposure to UV light and increasing the potential contribution of photolysis. To test this hypothesis, a novel test system was designed to assess the contribution of photolysis, under more realistic conditions, to the dissipation of a new herbicide, bicyclopyrone. This compound dissipates rapidly in field dissipation studies but microbial soil metabolism is relatively slow. Thin layer soil photolysis studies have shown susceptibility to photolysis; however its contribution to dissipation in the field could not be established from these studies. Soil cores were taken from three US field study sites and treated with 14 C-bicyclopyrone. The radioactivity was moved away from the surface using a simulated rainfall event and the cores were incubated under a Xenon-arc lamp with continuous provision of moisture from below. The light intensity, light/dark cycle and soil surface temperature were set to mimic the US field study conditions as closely as possible. Cores were removed at intervals, sectioned to different depths and analysed for total radioactivity and remaining bicyclopyrone and its photodegradates. After only 2 days, most of the radioactivity had returned to the soil surface. Significantly more degradation was observed in the irradiated samples than in a parallel dark control sample. The degradation rates in the new core study were very similar to those observed in both the thin layer photolysis study and the field studies and was significantly faster than the regulatory soil metabolism studies, indicating that photolysis can be a significant process under field conditions, even after initial movement out of the photolytic zone by rainfall. 167 Development of country specific weighting factors and external cost of environment K. Murakami, N. Itsubo, Tokyo City University; K. Kuriyama, Kyoto University; K. Yoshida, Nagasaki University; K. Tokimatsu, Advanced Industrial Science and Technology. LIME, an advanced life cycle impact assessment method based on endpoint modeling, has been developed as part of the LCA national project of Japan. One of the aims of LIME is to develop the weighting methodology, which enable us to integrate various environmental impacts that are used for life-cycle impact assessment (LCIA) and facilitates the interpretation of environmental information, such as in the selection of products. This project has now reached to the third phase with the additional aims, one of which is to update the weighting factors from the national average (LIME2) to the global scale (LIME3). We report our preliminary results obtained by pilot survey at five countries: Japan; China; Vietnam; South Africa; Kenya. The calculated results can be used to develop integration factors in LIME3, enabling us to express LCIA results as a single index, such as external cost. 168 Economic valuation of environmental impacts under REACH possibilities and limitations N.M. Deleebeeck, Arcadis Belgium / REACH & Product Stewardship Services; H. Descamps, Arcadis Belgium; J.K. Verhoeven, M. Beekman, D. Sijm, National Institute for Public Health and the Environment (RIVM); S. Bogaert, Arcadis Belgium. The European REACH Regulation foresees procedures for Restriction and Authorisation of chemicals that (may) give rise to unacceptable risks to human health and/or the environment. Under both procedures, a socio-economic analysis must/may be performed to weigh the costs and benefits to society of different scenarios (e.g., discontinued use versus continuation of use, continuation of use versus use of an alternative substance or technology) in terms of economic, social, human health and environmental impacts. The outcome of the socio-

SETAC Europe 23rd Annual Meeting

economic analysis is taken into account when the European Commission takes its decision on the continuity of use. Ideally, all costs and benefits are compared in a cost-benefit analysis using the same denominator (e.g., Euros). Therefore, it needs to be investigated to what extent human health and environmental impacts resulting from production and use of a REACH substance can be monetised. Monetisation of environmental impacts seems to be more challenging compared to monetisation of human health impacts, as generally, much more uncertainties need to be dealt with. Currently, no guidance documents have been issued by the European Chemicals Agency on this specific topic. Therefore, RIVM asked Arcadis Belgium to investigate the applicability of existing methods for economic valuation (revealed preference, cost-based, and stated preference methods) for monetisation of environmental impacts under REACH. A thorough evaluation was done of pros and cons, data needs, data availability (e.g., for benefit transfer), compatibility of the different methods with typical outcomes of an environmental impact assessment, etc. This resulted in a conceptual framework that should assist the user in 1) deciding whether or not it is possible/useful to monetise the environmental impacts, and 2) selecting the most suitable method when it is decided to proceed to the step of monetisation. Consequently, several illustrations were elaborated for further testing the applicability, possibilities, and limitations of the preferred methods. Based on the outcome of this exercise, data gaps were identified and recommendations were made for future research. 169 Impact-pathway approach and metals: monetisation of external costs of cadmium emissions to soil from agricultural fertilizerM. Pizzol, Aalborg University / Development and Planning; J. Smart, Griffith University / Griffith School of Environment; M. Thomsen, Aarhus University. We applied the Impact-Pathway Approach (IPA) for the monetary evaluation of health-related impacts of Cadmium (Cd) emitted to soil via agricultural application of phosphorous-rich fertilizers in Denmark. Due to the high persistency of Cd in soil, and high soil-toplant transfer rates, humans may be exposed to Cd through their diet, with adverse health impacts. We used the Simplified Fate and Speciation model (SFSM) to calculate Cd increase in soil for different future scenarios of agricultural application of fertilizer to Danish soil. Based on soil-plant bio-concentration factors and Danish dietary intake rates, we determined human exposure. We used updated dose-response functions (DRF) linking lifetime Cd intake with the probability of developing Cdinduced renal disease and osteoporosis. These impacts are converted into monetary values by using the EU standard value of VOLY (40000€) adjusted for quality of life experience. Total costs over 100 years are then annualized and discounted at 3% to present value to obtain the external costs. Results show that discounted external costs of Cd are of ?3800 [€/kg Cd] for sludge application, and ?4100 [€/kg Cd] for mineral fertilizer, due to the lower Cd content of sludge in kg dry matter. When P content is considered, costs are of ?0.092 [€/kg P] for sludge and ?0.089 [€/kg P] for mineral fertilizer, due to the higher content of P in the latter. Since the method is under testing, such values are considered as only preliminary. The pilot study is a unique example of monetization for a previously uninvestigated impact pathway. It allows exploring the IPA approach beyond its classic applications (e.g. airborne non-persistent contaminants) and to address critical issues such as uncertainty assessment, long-term modeling perspective and discounting. The study provides useful insights to Life Cycle Impact Assessment (LCIA). Most LCIA methods apply a similar emission-to-impact causal chain approach, and the same models can be used in IPA/LCIA. Although IPA is applied to a single contaminant, IPA is more detailed than LCIA concerning the definition of temporal and spatial conditions. The innovative modeling of Cd fate in soil and the identified DRF could also be applied, for example, to improve LCIA of metals. 170 Spatially explicit characterization factors for damage costs of soil erosion due to agricultural land occupation on a global scale R. Van Zelm, Radboud University; M. van der Velde, IIASA; T. Koellner, University of Bayreuth; M. Nunez, IRTA; M. Obersteiner, IIASA; E. Schmid, University of Natural Resources and Life Sciences; M.A. Huijbregts, Department of Environmental Science. A main cause

43

of increased soil erosion is agricultural land use, as the cultivation (irrigation, ploughing, soil processing etc.) of crops contributes to increased soil loss, which on its turn degrades arable land. The objective of this work is to provide spatially explicit characterization factors (CFs) for erosion regulation due to crop cultivation for the world. CFs are provided for land occupation by cassava, rapeseed, sunflower, sugarcane, soybean, corn, and wheat for three levels of agricultural intensification defined by fertilizer and irrigation uses. Because different management of the crops results in different crop yields, the characterization factors express the damage caused per kg of crop and 2 not per m of land use, which has been more common up to now. The damage is in terms of extra onsite (water and nutrient) and offsite (problems caused by the soil particles) costs (in US $/kg crop) due to soil loss caused by crop production. To simulate erosion rates, a global implementation of the biophysically based agro-environmental simulation model EPIC (Environmental Policy Integrated Climate) was employed. EPIC is driven by spatially explicit information on climate, weather, topography, soil and agricultural management practices and operates on a 30 arcmin scale. Rangeland and trees were simulated with EPIC around the world as an estimation of the potential natural vegetation when there would be no crop cultivation. On- and offsite erosion costs in 2012 were estimated to be 13 US $ per ton of eroded soil. Largest CFs were found around and below the equator. For some grids, the potential natural vegetation showed larger erosion rates than crop cultivation, hence negative characterization factors were derived. When aggregating to country level (yield-weighted), the negative factors showed no influence due to the insignificant yields in these grids compared to the overall country yield. Larger CFs are obtained when management without input of nutrients or irrigation is performed. Small yields obtained without fertilizer and irrigation generally lead to higher costs per kg crop due to erosion than when fertilizer and irrigation are applied. CFs for the 7 crop types vary up to 6 orders of magnitude from each other, with soybean cultivation generally leading to the lowest CFs per country and rapeseed to the largest CFs. The high variability in CFs per crop, grid, and input scenario show the importance of the applied method. 171 Assessing the impacts of abiotic resource use: application to electric vehicles batteries V. De Bruille, CIRAIG École Polytechnique de Montréal / Chemical Engineering; C. Bulle, CIRAIG Polytechnique Montreal / Chemical Engineering; T. Dandres, CIRAIG; C. Gaudreault, NCASI; O. Jolliet, University of Michigan / School of Public Health; R. Samson, Ecole Polytechnique de Montreal / Department of Chemical Engineering. In this study, a new assessment method is proposed to assess the impacts from resources use, as none of the previously existing methods is considered mature enough to be recommended in the ILCD handbook. This new approach allows impact monetisation based on resource functionality, future abundance and the financial consequences of the use of resources. This allows considering an effect factor expressing the extent to which the extraction of a certain amount of abiotic resource forces a future user to pay more to preserve the function provided by the resource as it becomes scarcer. It is translated by the difference between the cost for a resource to fulfil a given functionality today and that of the back-up alternative. In specific, this project aims at including the main affected users and at defining both the increase in cost to avoid functionality loss and substitution costs. The environmental impact is obtained by multiplying the effect factor by a competition factor, which is a material competition scarcity index (MACSI) varying between 0% and 100% and defined to assess the competition. The MACSI is based on the “remaining years of availability”: a ratio between resource consumption and available known stock, representing the years left before total dissipation of a resource at current dissipation rate. Resource dissipation is the amount used from which recycling and reuse is subtracted. This choice was made considering that resource will remain (partly-) functionally equivalent, even after the end of a product life through reuse or recycling. The approach was illustrated by applying it to metallic resources used in electrical vehicles batteries to calculate the impact on natural resources depletion. Results have been compared with the impacts obtained with

SETAC Europe 23rd Annual Meeting

other LCIA methodologies to put in perspective advantages and drawbacks of each of them and see to what extend results obtained are consistent when applying different approaches showing different degrees of sophistication. Inconsistencies between methodologies are put into light as the ranking between contributors to the impact differ from one methodology to the other. This indicates the need for clarification on how resource depletion should be assessed in LCA and for the development a coherent framework for that assessment. 172 Monitoring herbicide toxicity in rivers of the Great Barrier Reef catchments using benthic diatom community changesR.J. Wood, P. Depresle, S. Mitrovic, University of Technology, Sydney; R. Prasad, S. Choy, Department of Science, Information Technology, Innovation and the Arts; B.J. Kefford, University of Technology Sydney / Department of Environmental Science. The extensive use of herbicides in agricultural regions of the Great Barrier Reef (GBR) catchment area in North-east Australia, has resulted in the widespread contamination of its waterways. Routine chemical monitoring within GBR catchments has regularly detected concentrations of herbicides over the recommended trigger values for ecological protection. Freshwater ecosystems are at greatest risk of herbicide toxicity due to their proximity to the source of herbicide application in the environment. Sensitive monitoring tools are needed to be able to determine the ecological impacts of herbicide toxicity. This study aims to develop a new method for monitoring herbicide toxicity in rivers using benthic diatom community changes. Despite the importance of benthic diatoms as primary producers and as potential bioindicators, there have been few studies on their sensitivity to herbicides. This study will address the current knowledge gap, by examining the sensitivity of freshwater diatoms to herbicide exposure through rapid toxicity testing. The effect of pre-exposure and light intensity on the relative sensitivity of diatoms genera was examined for the herbicides; Atrazine, Simazine, Hexazinone, Tebuthiuron, Diuron, MCPA, Glyphosate, 2,4-D. Preliminary results indicate the following diatom genera sensitivity ranking: Navicula > Achnanthidium > Synedra > Cymbella > Gomphonema. Diatom sensitivity rankings remained consistent between the herbicides tested. The sensitivity data presented was used to establish a new biomonitoring index, the SPEcies At Risk (SPEAR) index, which uses this toxicity data to determine the fraction of the abundance of sensitive taxa in a community at particular sites. The new index will be used to determine if the abundance of benthic diatom genera senstive to herbicides declines across a herbicide expsosure gradient consisting of 14 sites within the GBR catchment. The results of this project to date will be assessed in terms of developing a new SPEAR index that uses benthic diatoms as an indicator to provide a cost effective and ecologically relevant method for the detection and assessment of herbicide impacts in rivers flowing into the GBR. 173 Zoning and spatiotemporal evolution of PAHs, described combining sediment concentration, baseline toxicity, bioaccumulation in fish and TEC for fish E. Rojo-Nieto, CactymarUniversity of Cadiz / Department of Environmental Technologies; J. Perales, CACYTMAR-University of Cadiz / Department of Environmental Technologies. In this work, fate and effects of PAHs in the Bay of Algeciras, a semi-enclosed coastal zone subject to an intense industrialization, have been studied. This Bay has suffered a chronic anthropogenic pressure, due to urban (five urban areas with more than 250,000 inhabitants) industrial activities (petrochemical and metallurgical industry), and to the intense maritime traffic (Algeciras Harbor is ranked among the most important ports of the world). In previous studies occurrence and levels of PAHs in sediment were studied, and a zoning of the bay according to contamination level and source was proposed. Additionally, sources, transport and fate of these compounds in sediments were studied obtaining a spatiotemporal evolution. The aim of this work was to study if different parameters related to biota, as Bioaccumulation in feral fish, Baseline Toxicity for marine organisms (described by the chemical activity) and Toxicity Equivalent Concentration for fish of PAHs found in sediments (TECs), could provide a similar zoning and spatiotemporal

44

evolution that produced by sediment concentrations, to define PAHs pollution in the study area. The results obtained show that the study of the total concentration, the interstitial concentration and the chemical activity of PAHs, in marine sediments from chronically polluted environments (and the translation of them to other parameters, through well-defined factors), provides fairly accurate knowledge about the distribution and spatiotemporal evolution of the environmental risks associated to their presence. However, since in natural environments rarely the environmental compartments are in total equilibrium and since the biological effects and bioaccumulation are influenced by several factors such as the metabolization of the compounds under study, the analysis of biotic compartment can not be completely ignored, being this compartment essential to define, among others, site/species-specific BSAFs. 174 Toward a tailored ERA tool for mosquito control practices in France M. Roucaute, INRA / UMR ESE; L.L. Lagadic, INRA / UMR INRAAgrocampus Ouest Ecology and Ecosystem Health; T. Caquet, INRA / UMR ESE 0985. In metropolitan France, mosquito control mainly targets larval stages and, in most cases, this means spraying insecticides directly into natural aquatic ecosystems. This obviously raises the question of the risk of such practices for non-target organisms especially in Natura 2000 protected areas where the European legislation now requires the implementation of incidence studies of human activities. Therefore, there is a need for tools that are able to detect potential environmental effects of biocides used for mosquito control. Indices such as SPEAR (SPEcies At Risk) aim at detecting effects of the exposure to chemicals (e.g., pesticides) on invertebrate communities in streams. They do not make assumption about the pattern of exposure and rely on a generic vulnerability profile of species toward chemicals. In the context of properly managed mosquito control, the nature, rates and strategies of application of authorised products are known. It is therefore possible to assess the real exposure pattern to larvicides. Based on such knowledge, it is feasible to develop an ERA tool specifically designed for detecting the impacts of mosquito control on the most vulnerable part of aquatic biota. We applied this approach to the case of two insecticides used for larval mosquito control: Bti (Bacillus thuringiensis var. israelensis), the only currently authorized larvicide in France, and S-methoprene, a candidate compound. Based on published laboratory acute toxicity data for both compounds, we ranked the species found in the samples from a past environmental survey according to their level of sensitivity. The assessment of the potential effect of treatments was then performed by comparing toxicity thresholds to predicted larvicide environmental concentrations based on currently used application rates. Such an approach may allow the a priori assessment of the risk of mosquito control practices on the invertebrate community of a given ecosystem. It may also be used to focus long term surveys on the most vulnerable part of this community. The rationale of the method, the results of species sensitivity ranking and a comparison of the outcomes of this approach with the results of field invertebrate community studies are presented. 175 Using observational field data for risk assessment of nutrients in freshwaters L.B. Azevedo, Radboud University Nijmegen / Department of Environmental Science; R. Van Zelm, Radboud University; R.S. Leuven, Radboud University Nijmegen / Department of Environmental Science; J.A. Hendriks, Radboud University Nijmegen / Department of Environmental Sciences; M.A. Huijbregts, Department of Environmental Science. We present a framework in which observational field data can be used to derive concentration-response relationships of the effects of nitrogen (N) and phosphorus (P) on species richness. To attain that, we first gathered data on individual species occurrences and N and P levels in temperate lakes and streams. We included autotrophs and heterotrophs as the two main species groups in our analysis. Second, we determined the species richness along TP and NO levels separately and used log-logistic regressions to calculate 3 the potentially not occurring fraction (PNOF) of freshwater species along the two stressors. Finally, we tested if the species richness effects could be explained by concentration addition of combined N and P

SETAC Europe 23rd Annual Meeting

stresses. At SETAC, we will (1) illustrate how logistic regressions illustrating the effects of nutrient stress to species richness can be obtained from observational field data; (2) test if concentration addition of N and P explains our empirical data; and (3) provide ecological grounds for the difference between effects of N and P, in lakes and streams, and for autotrophs and heterotrophs. 176 Competition between an indigenous and invasive snail (Physidae) under impact of additional stressors (temperature, fungicide) R. Mueller, LOEWE Biodiversity and Climate Research Centre / Aquatic Organisms and Ecosystems; A. Seeland, Goethe University Frankfurt Main; J. Albrand, Helmholtz Centre for Environmental Research GmbH - UFZ; D. Jilani, Goethe University Frankfurt; J. Oehlmann, Johann Wolfgang GoetheUniversitat Frankfurt am / Aquatic Ecotoxicology. Aquatic species are highly influenced by abiotic and biotic factors such as temperature and pollution (e.g. pesticides) or intra- and interspecific competition. Moreover, the ongoing process of climate change enforces these impacts and leads to far-reaching shifts in the aquatic environment. Beyond, these disturbed environments may favour invasive species. Within this context, the aim of the present study was to investigate the potential of the invasive snail Physella acuta to displace the indigenous Physa fontinalis out of their natural habitat under a multistressor situation. To investigate the temperature-dependent sensitivity of these pulmonates towards the fungicide pyrimethanil, a juvenile growth test (dose-response-design) at optimal and suboptimal temperatures (15°C, 20°C and 25°C) was conducted. Furthermore, the competitive relationship between both species was studied directly with juveniles at a sublethal concentration -1 (0.1 mg pyrimethanil L ) and 20°C using a response-surface-design. Both stressors, temperature and the fungicide pyrimethanil, caused singular and interactive effects on both species. However, the growth of the indigenous P. fontinalis was more severely influenced by pyrimethanil compared to the invasive P. acuta, especially at higher temperatures. The direct competitive situation depicts that P. acuta points to a lower mortality rate and faster development than P. fontinalis. Our results clearly show that P. acuta has a clear competitive advantage compared to P. fontinalis, in particular with regard to progressive climate and environmental changes. Therefore it can be concluded that P. fontinalis will be most likely displaced from its habitats by P. acuta with increasing frequency. 177 Combined exposure of Chlamydomonas reinhardtii to copper and solar-simulated radiation G. Cheloni, Aquatic Biogeochemiostry and Ecotoxicology, Institute F.-A. Forel, Earth and Environmental Sciences, Faculty of Sciences; E. Marti, V.I. Slaveykova, University of Geneva / Aquatic Biogeochemiostry and Ecotoxicology, Institute F.-A. Forel, Earth and Environmental Sciences, Faculty of Sciences. In natural aquatic environments photosynthetic organisms experience variable levels of light composed of photosynthetically active radiation (PAR) and UV radiation. High PAR intensities have a direct effect on algal photosynthesis, while high UV radiations, besides being harmful for the organisms, may alter the biogeochemistry of redox sensitive metals and thus their chemical speciation and bioavailability. However in laboratory toxicity tests these organisms are grown under artificial light which consist of only PAR at very low intensities, variations in light intensity and their effect on both organism physiology and metal bioavailability are not taken into account. Considering the paucity of information concerning the metal toxicity in complex aquatic environments, in this work we explored the combined effect of solarsimulated light and copper on the green alga Chlamydomonas reinhardtii. C. reinhardtii cells were exposed to two different free -8 -6 copper ions concentrations (6x10 and 6x10 M) representative for slightly and heavily contaminated water bodies. Three different light conditions were tested: 1) a low PAR typical for laboratory tests, 2) a solar-simulated light (PAR+UV) with spectral distribution and intensity comparable to what measurable at midday in a winter clear day at mid latitude in Europe and 3) a solar-simulated light (PAR+UV) with increased UV radiation, aimed to simulate future changes in UV scenarios. Short term effects of copper on algal cells under different

45

light conditions were determined using flow cytometry in order to evaluate changes in algal population features at single cell level and, using different fluorescent dyes, changes in membrane permeability, oxidative stress and lipid peroxidation. Combined exposure to copper and solar-simulated light revealed a synergic effect with increased damages compared to those obtained in cultures exposed to copper or solar simulated radiation alone. Cells irradiated with increased UV radiation presented an enhanced membrane permeability and experienced stronger oxidative stress than cultures exposed to laboratory light conditions. The membrane alteration potentially resulted in an increased copper bioavailability explaining in part the synergic effect observed. 178 Validation of the iSQ (invertebrate soil quality) Chip for Ecological Effects Assessment of Soil Quality Management G. Chen; T. De Boer, Vrije Universiteit / Ecological Sciences; A. Vakhrusheva, Vrije University Amsterdam; N.M. van Straalen, Vrije Universiteit Amsterdam / Inst. of Ecological Science; D. Roelofs, Vrije University Amsterdam. Environmental pollution is a worldwide problem, and the ability of soil to function properly is threatened by soil contamination. Transcriptomic tools in ecologically relevant organisms, such as sprintails, enhance classical eccotoxicological research by presenting more descriptive mode of action of soil pollutant. To study ecological responses to soil conditions, an invertebrate Soil Quality (iSQ) Chip (74k gene clusters) for the springtail Folsomia candida were developed and applied. Animals were exposed field samples collected from river Dommel area in Netherlands, then reproduction rate of juveniles and gene expression profiles were measured. We identified different transcription profiles in different field soil samples, and acquired specific responded genes and metabolic pathways for each soil samples. Also, according to diverse expression patterns among different soil samples, some genes (e.g. abc transporter genes) may potentially be a good indicator to screen potentially polluted sites, in order to identify the ones that need further ecotoxicological investigation. Combining with soil chemical analysis, we discovered some biotic stress factors may disturb traditional eccotoxicological test. This study shows that in order to apply transcriptomic analysis tool to soil quality management, more insight is needed to understand the complexity of field samples and better utilize classification method to predict contamination in soil samples. Key words: soil; ecotoxicogenomics; microarray; springtails. 179 Unraveling mode of actions and toxin profiles with high throughput microarrays: a case study in Daphnia exposed to different cyanobacterial stressors. J. Asselman, Ghent University / Laboratory of Environmental Toxicology; M. Pfrender, J. Lopez, Notre Dame University; J.R. Shaw, Indiana University / The School of Public and Environmental Affairs and The Center for Genomics and Bioinformatics; K. De Schamphelaere, Ghent University UGent / Environmental Toxicology and Aquatic Ecology. Over the last decade, molecular technologies have evolved into robust high throughput platforms available for many scientists in a wide variety of disciplines. Yet, knowledge of stress responses and mechanisms of toxicity across stressors remains limited. Therefore, we have studied stress response to 6 distinct cyanobacterial species in Daphnia pulex. By studying effects on gene level with high throughput microarray technology and effects on organism level through standard ecotoxicology methods for a group of stressors simultaneously, this study aimed at identifying and unraveling stress response patterns and mechanisms of toxicity. Furthermore, by first considering all 6 species as cyanobacterial stress in general without distinction, it was possible to identify mechanisms involved in general cyanobacterial stress response. Four day old Juveniles of Daphnia pulex were exposed to a diet contaminated with cyanobacteria for a period of ten days. The experiment consisted of six different treatments, representing six different species of cyanobacteria. Afterwards, RNA was extracted, reverse transcribed to cDNA and hybridized to a two color whole transcriptome microarray.Image data was processed and normalized in R using Limma. Data analysis revealed six pathways involved in a general cyanobacterial stress response. These could be divided in three groups of mechanisms. First, pathways

SETAC Europe 23rd Annual Meeting

involving detoxification (Cytochrome P450 and Glutathione) were significantly induced in organisms exposed to cyanobacterial stress. This could potentially be a response to the toxins produced by the cyanobacteria as these pathways are known detoxification steps for the cyanobacterial toxins. Second, pathways involved in steroid and poly unsaturated fatty acid synthesis, were also significantly induced. This may potentially be linked to the lack of steroid and PUFA’s in cyanobacteria in contrast to green algae as a nutrient source. Third, we observed a significant inducement of genes in pathways related to the energy metabolism (ubiquinone cofactor and carbohydrate metabolism). By combining expression of exposures to different cyanobacteria, it was possible to gain insights in mechanisms involved in cyanobacterial stress response that could be potentially related to effects at the organism level. Finally, species specific stress response for each cyanobacteria can be identified by contrasting the expression profiles with these mechanisms of general cyanobacterial stress. 180 Dose-dependent effects induced in Arabidopsis thaliana after uranium (U) exposure at pH 7.5 E. Saenen, SCKCEN / Biosphere Impact Studies; n. Horemans, Belgean Nuclear Research Centre SCKCEN; N. Vanhoudt, Belgian Nuclear Research Centre (SCK-CEN); H. Vandenhove, Belgian Nuclear Research Centre, SCK-CEN / Biosphere Impact Studies; G. Biermans, SCKCEN; M. Van Hees, Belgian Nucrlear research centre (SCK-CEN); J. Wannijn, Belgian Nuclear Research Centre (SCK-CEN); J. Vangronsveld, A. Cuypers, Universiteit Hasselt. To evaluate the environmental impact of U contamination, it is important to unravel the mechanisms by which plants respond to U stress. It was already shown that U exposure at pH 5.5 can disrupt the cellular redox balance and induce oxidative stress related responses in Arabidopsis thaliana plants. However, U speciation strongly depends on environmental pH. Since U uptake and hence its toxicity is influenced by U speciation, it is important to investigate the effects of U at different pHs. This study aimed to investigate dosedependent effects of U at pH 7.5. A. thaliana plants were exposed to a U concentration range at pH 7.5 during 3 days. U concentration and fresh weight were analysed. Enzymes and metabolites of the antioxidative defence system were analysed on protein level. Effects at molecular level were analysed by real-time PCR. A significant decrease in root and shoot fresh weight was observed after U exposure. Together with an increased dry weight, this can be an indication that the plants are stressed. In the leaves, the antioxidative capacity is increased by an increased concentration of total and reduced ascorbate (Asc), a metabolite important in the antioxidative defence. In the roots however, the redox status could not be maintained. Here, a decrease in reduced Asc and an increase in dehydroascorbate was found, indicating that the roots are damaged after exposure to U. By analysing enzyme activities in the roots, increased activities in Asc peroxidase (APX) and guaiacol peroxidase (GPX) were found. Together with the changes in Asc concentrations, the increased APX activity can be an indication for the importance of Asc under U-stress. GPX in turn is important in cell wall lignification. The increased GPX activity can indicate an enhanced lignification of the root endodermis. This can be a response to increase cell wall binding sites in the roots and hence limit the transport of U to the shoot. By analysing the U responses at molecular level, the involvement of microRNA398b/c was observed. MIR398b/c binds to mRNA of CSD1/2 and hence, a decreased expression of the latter genes was observed. However, CSDs are indispensable, and an increase in root FSD1 expression was observed, that probably guaranteed sufficient ROS-detoxification capacity. In conclusion, elevated U-concentrations at pH 7.5 can cause important morphological, physiological and biochemical effects in A. thaliana seedlings. 181 The informative value of transcriptomics in combination with the fish embryo test M. Fenske, Fraunhofer Gesellschaft IME / Institute for Molecular Biology and Appl. Ecology; E. Muth-Koehne, Fraunhofer Institute for Molecular Biology and Applied Ecology IME / Department of Ecotoxicology; V. Schiller, Fraunhofer IME / Institute for Molecular Biotechnology (Biology VII); V. Delov, RWTH Aachen University / Institute for Molecular Biotechnology (Biology VII); A.

46

Wichmann, H. Hollert, RWTH Aachen University / Institute for Environmental Research (Biology V); R. Kriehuber, Jülich Research Centre / Radiation Biology; M. Hecker, University of Saskatchewan / School of the Environment & Sustainability; C. Schaefers, FraunhoferInstitut / Department of Ecotoxicology. In aquatic ecotoxicology, the fish embryo toxicity test FET can be considered as one of the most promising animal alternative methods. Internationally, considerable endeavours have been made to bring the FET to acceptance as a valid alternative method to the fish acute toxicity test (OECD203). However, numerous studies have repeatedly shown that the FET can deliver more than just acute toxicity data, in particular when molecular endpoints are included. Notably postgenomic approaches, like transcriptomics using microarrays or qPCR, and fluorencent microscopybased morphometirical analyses, are well suited to evaluate subacute or teratogenic effects mechanistically due to their high sensitivity and information rich data. In a proof-of-principle approach, we conducted gene expression studies with zebrafish and medaka embryos after exposure (for 48 hpf to 120 hpf and 7 days, respectively) to different groups of chemicals. For each chemical, we conducted a standard FET (48h), and for the insecticides also extended FETs (120h) in 96-well plates. Embryos exposed to a selection of the tested substances were subjected to whole-genome expression analysis (using the Agilent© Zebrafish Gene Expression Microarrays) or quantitative PCR. To exclude non-specific stress responses, embryos were exposed to concentrations at EC , EC /2 and in some cases at EC level for the 10 10 20 gene expression experiments. Estrogenic and anti-androgenic compounds (ethinylestradiol-EE , flutamide, genistein, bisphenol A2 BPA, linuron, prochloraz) were studied in both species. Thirteen insecticides of known acute adult fish toxicity, representing four different modes of action were tested only in zebrafish. Insecticides classified as ‘energy metabolism modulators’ by IRAC displayed a higher number of differentially expressed genes than those classified by ‘nerve action’. Intersections revealed that insecticides with functionally similar MoAs share a higher number of differentially expressed genes than those with dissimilar ones. Functional analysis showed that regulated genes relate to molecular functions, which are linked to the morphological effects and the given MoA of an insecticide. For the EDCs, responsive pathways comprising endocrine relevant pathways were found in both species and for all chemicals. An estrogenic mode of action was clearly discernible, and the transcriptome of all tested substances showed patterns indicating either a predominantly estrogenic or antiandrogenic activity. 182 Assessment of adverse effects of environmental contaminants in complex mixture using physiological, transcriptional and metabolic endpoints in fish H.R. Habibi, University of Calgary / Biological Sciences; A. Zare, Department of Biological Sciences; J. Jordan, University of Calgary / Biological Sciences; S. Henderson, Biological Sciences; A. Weljie, L.J. Jackson, University of Calgary / Biological Sciences. Environmental contaminants including natural and synthetic steroids, organic compounds and pharmaceuticals were detected at various sites sampled along the Southern Rivers in Canada. The nature contaminants were different depending on the location and extent of agriculture activities, upstream and downstream of wastewater treatment plants. Trace amounts of pharmaceuticals, phthalate esters, and all nonylphenol ethoxylates were observed in receiving river waters of Southern Alberta. In field studies, we used longnose dace as an abundant local species to assess health impact of contaminants in Southern Alberta Rivers. A significant increase in female to male adult ratio was observed in longnose dace caught down stream of certain municipalities along the Oldman River correlating with high levels of vitellogenin expression in male longnose dace. These observations suggested severe endocrine disruption of gonadal development likely due to presence of compounds with estrogen-like activities. To investigate the effects of chemicals detected, we performed controlled laboratory experiments in which fish in aquaria were exposed to the same concentrations of a selected number of chemicals detected in the river system, individually and as mixtures. We used various biological end points, including morphological abnormalities, transcriptomics and metabolomics in

SETAC Europe 23rd Annual Meeting

samples obtained from liver, gonads and brain of fish. The results demonstrate that different chemicals present in the Oldman River disrupt the gene expression profile and metabolism of the liver, ovary, testis, and brain. Significant changes were observed in gene expression profile as well as dysregulation of amino acid, lipid, energy, carbohydrate, nucleotide and cofactor/vitamin metabolism. The effects of mixture of contaminants were significantly different from the individual treatments in the liver and testis. The results provide novel information on the effect of contaminants individually and in mixture on global metabolism dysregulation in fish and a framework for better understanding of the metabolic pathways affected by environmental contaminants in fish. The results are important to assess biological impact of contaminants and more accurate risk assessment associated with complex mixture of contaminants in our environment. Funded by grants from NSERC of Canada. 184 A study of the effects of Cr(VI) on the earthworm Eisenia andrei: from molecular/cellular changes to whole organism responses S. Sforzini, Universita Del Piemonte Orientale Amadeo Avogadro / Department of Science and Technological Innovation; M. Boeri, S. Olivieri, A. Viarengo, University of Piemonte Orientale. The aim of this investigation was the evaluation of the quality of Cr(VI)contaminated soils using the earthworm Eisenia andrei. A set of physiological parameters at different levels of functional complexity was assessed in diverse cells and tissues, and in the whole organism. Field soil contaminated by a relatively high concentration of Cr(VI) (175 mg/kg) was investigated. Relevant biological changes were observed in worms exposed for 28 days to contaminated soil, without resulting in mortality. No effect was found in the number of juveniles hatched from each cocoon, although the number of cocoons and total offspring was reduced. The exposure to polluted soil caused significant changes in lysosomal biomarkers evaluated in the cells of the chloragogenous tissue i.e. an increase in lipofuscin and neutral lipid content as well as in the lysosome/cytoplasm volume ratio, a biomarker of tissue damage. A drastic decrease of lysosomal membrane stability was measured in the immunocompetent coelomocytes and significant genotoxic effects (DNA damage and micronuclei frequency) were observed in the same cells. The perturbation of the worm immune system was also demonstrated by an increase in lysozyme activity in the coelomic fluid. Based on these results, we investigated the potential toxicity/genotoxicy of soils contaminated by lower environmental levels of Cr(VI). E. andrei was exposed for 28 days to Cr(VI) at concentrations of 1, 2, 15 mg/kg spiked in standard soil. The selected Cr(VI) concentrations fall within the limits for soils set by Italian law. In addition to biomarkers employed in the first part of this study, we 2+ utilized a novel method to assess the activity of Ca -ATPase on frozen tissue sections. Moreover, in the coelomocytes, the oxidative DNA damage was also determined by the use of Fpg enzyme. Significant changes in different biomarkers were observed in earthworms exposed to Cr(VI) spiked soils, with greater effects at the highest concentrations. 2+ There was a significant decrease in the activity of Ca -ATPase in the intestinal epithelium, as well as a relevant increase of oxidative DNA damage in worms exposed to soils containing the lowest level of Cr(VI). Overall, the data demonstrated that E. andrei is a suitable model organism for ecological risk assessment and that the use of sensitive sublethal parameters is important for demonstrating the vulnerability level reached by the biotic resources in terrestrial ecosystems. 185 Assessing chemical effects on earthworm physiology: an energy budget-driven agent based modelling approachA.S. Johnston, University of Reading / Biological Sciences; M. Hodson, University of York; P. Thorbek, Syngenta / Environmetal Safety; T. Alvarez, EcoRisk Solutions Ltd / Dept. of Ecological Sciences; R. Sibly, University of Reading / Biological Sciences. Earthworms are significant contributors to the ecosystem services provided by soils. Consequently, they are established model organisms for environmental risk assessments of plant protection products (PPPs). If PPPs are deemed to pose an unacceptable risk to earthworm populations, based on laboratory studies for acute mortality or chronic effects on growth or reproduction in lower tier risk

47

assessments, risk has to be further investigated at a higher tier using field trials. Field trials are expensive, time-consuming and variable, providing coarse information about population-level effects. Mechanistic population modelling of organism responses to PPPs has the potential to act as a refinement option in risk assessment, using lower tier results to make ecological extrapolations to the population level. Here, we evaluate the ability of energy budget-driven agent based models (ABMs) to serve as methods in current risk assessments for soil invertebrates. As multiple physical and chemical stressors have biological effects on all organisms in nature, such methods combining biological detail with environmental exposure are useful in aiding ecotoxicological inferences. We develop an initial step towards generating a useful tool for predictive toxicology, by investigating the ability of the model to replicate sublethal effects of different chemicals on individual physiology (e.g. maintenance, growth, reproduction). Energy budget models offer a robust method for predicting physiology by relating individual life cycle processes to one another through energy and mass conservation. Organisms uptake resources from their environment and expend assimilated energy on maintenance, growth and reproduction. The allocation of energy to these different subsystems depends on a combination of environment- and organism-specific conditions. Using typical lower tier risk assessment results from the literature, we incorporate toxicity data by a simple curve-fitting exercise to translate individual endpoints into metabolic modes of action. We predict the effects of two chemicals, copper oxychloride and chlorpyrifos, on the earthworm Eisenia fetida. Good model fits to the data show the simple energy budget ABM to be capable of making detailed predictions of interacting chemical and environmental stress at the individual level. Limiting food conditions had a particularly antagonistic impact on toxicity effects at the sub-individual level. 186 Hazards of anthropic contaminated soils: what could we learn from environmental genotoxicity markers?S. Lemiere, University of Lille / Nord; A. Deram, University of Lille / LSVF/ILIS University of Lille 2; F. Bernard, PRES Univ Lille Nord de France / LGCgE Univ Lille1 and LSVF Univ Lille2; S. Dumez, PRES Univ Lille Nord de France / LSVF Univ Lille2; F. Nesslany, A. Platel, PRES Univ Lille Nord de France / Toxicology Lab - Pasteur Institute of Lille; M. Delattre, PRES Univ Lille Nord de France / Univ Lille1; D. Cuny, Univ Lille Nord de France / LSVF Univ Lille2; F. Vandenbulcke, PRES Univ Lille Nord de France / LGCgE Univ Lille1. It is now well-accepted than hazard and risk assessment of anthropic contaminated soils may lead to erroneous diagnostics if only analytical chemical results are taken in account. Biological responses should be considered in decision-making regarding risk evaluation and strategies of remediation or management of contaminated soils and wastelands. Among these biological responses, environmental genotoxicity has to be regarded particularly. Study of formerly or recently contaminated soil genotoxicity is complex, since we have to consider: (1) the environmental availability, environmental and toxicological bioavailabilities of present contaminants, influenced by soil factors, ageing phenomena..., their concomitant presences (2) at low concentrations and then (3) their potential toxic interactions. For such purpose, methods of classical genetic toxicology do not seem fully satisfactory and we decide to implement environmental genotoxicity markers. The comet assay, a technique allowing the evaluation of DNA damage (single and double strand breaks, alkali-labile sites) in a cellular population, was developed for example on cœlomocytes of the biological model largely used in soil ecotoxicology, Eisenia fetida. In a first experiment, we conduct in vivo short-term exposures using eight well-characterised soils originating from two different areas in the North of France, (1) from the vicinity of a former smelter and (2) from a suburb zone close to a battery plant still in activity. In both contexts, soils were sampled along a contamination gradient. Little DNA breaks are observed after exposures to these anthropic contaminated soils. The rare positive responses are not obtained with the most contaminated soil. These results raise the question of the suitability of this environmental marker to assess genotoxic hazard in soils. So, we conduct in vivo short to mid-term exposures (3, 10 and 56 days) using an urban-surrounding soil spiked

SETAC Europe 23rd Annual Meeting

with cadmium or/and lead at concentrations found in soils sampled from the former smelter impacted area. Observed genotoxicity results, as well as metal bioaccumulation, after these exposures will be detailed and discussed. The main conclusion of this work is that our marker of environmental genotoxicity is useful and relevant for environmental hazard assessment of anthropic contaminated soil. 187 Soil Phototrophic microorganisms as sensitive indicators for the assessment of herbicide stress O. CROUZET, INRA / soil ecotoxicology; J. Wiszniowski, Silesian University of Technology / Environmental Biotechnology Department; F. Bonnemoy, C. Mallet, LMGE, Université Blaise Pascal (UMR-CNRS). This study aims to investigate dose-response effects of an herbicide on the soil phototrophic microbial communities, and more particularly cyanobacteria and microalgae, by a microcosm approach. Pure mesotrione (active ingredient), ® belonging to triketone herbicides, and Callisto (commercial formulation) were spread at different rates on soil microcosm surfaces. Chlorophylls in soils were quantified to assess photosynthetic biomass and genetic structure and diversity of the cyanobacterial community was investigated by a group-specific polymerase chain reaction followed by denaturing gradient gel electrophoresis (PCR-DGGE). Dose-dependent responses were evidenced for both functional and structural parameters. No effect was detected in the 1×AR (one fold recommended application rate) treated soils, irrespective of the herbicide formulation. At 10×AR ® (ten folds recommended application rate), only Callisto treatment induced significant decreases of photosynthetic biomass and shifts on structural parameters. At 100×AR (one hundred folds recommended ® application rate) both pure mesotrione and Callisto had strong negative impacts on soil chlorophyll concentrations and cyanobacterial genetic ® structure and diversity. At both 10×AR and 100×AR, Callisto induced significant stronger effects than pure mesotrione. Also, the indicators of photosynthetic biomass responded (within 7 days) more quickly to herbicide stress, than structural parameters of cyanobacterial communities (within 14 days). This study underlined the relevance of soil phototrophic microbial communities to develop indicators for herbicide risk assessment. It also showed, the usefulness of molecular tool to improve the knowledge about the biodiversity of these unrecognize microorganisms in terrestrial ecosystems. 188 The enzymatic functional stability: a suitable approach to detect terrestrial ecotoxicity of metals on tolerant microbial communities I. Lessard; S. Sauve, Environmental Chemistry Laboratory / Chemistry; L. Deschenes, CIRAIG Polytechnique Montreal. A soil microbial community exposed to metals over a long period can acclimate or adapt and some of its biogeochemical functions may become metal-tolerant. This tolerance can increase the vulnerability of the microbial community to additional soil disturbances and in turn, decrease the soil microbial functional stability and ecological resilience. It is hypothesized that the functional stability assessment can detect an ecotoxic effect that would not be found using common bioindicators. This concept is timely and well-suited in the current context of multi-stressed ecosystems and global assessment of soil health. The objective of this study is to determine whether the functional stability of soil enzymes assessed by the RSSI, Relative Soil Stability Index, provides complementary information to the routine enzymatic assessment of metal-contaminated soils. The RSSI and the activity of four enzymes (arylsulfatase, phosphatase, protease and urease) were obtained for 20 Zn-contaminated field soils after a heat disturbance simulating severe drought (desiccation at 60?C for 24hrs). Among the most meaningful results obtained, the RSSIs of two enzymes were highlighted: they were negatively correlated to the labile Zn concentrations of the field soils (partial inhibition model: 2 2 phosphatase R =0.39 and EC =1160 µg/kg and protease R =0.42 and 50 EC =365 µg/kg), while their enzymatic activities were not correlated to 50 2 2 it (same model: phosphatase R =0.05 and protease R =0.02). These findings suggest that assessing functional stability by RSSI leads to different ecotoxicological information than enzymatic activity assessment and may be used as a complementary approach to assess metal toxicity on a metal-tolerant soil community.

48

189 Relative bioavailability of tropical volcanic soil-bound chlordecone in farm animals: laying hens, piglets and lambsC. Jondreville, URAFPA; S. Jurjanz, C. Bouveret, S. Lerch, Universite de Lorraine / URAFPA; M. Lesueur-Jannoyer, CIRAD / PRAM; H. Archimede, M. Mahieu, INRA / URZ; C. Feidt, G. Rychen, Universite de Lorraine / URAFPA. Chlordecone (CLD) is a chlorinated polycyclic ketone pesticide used from 1971 to 1993 in French West Indies to fight against banana black weevil. The former use of this organochlorine insecticide has resulted in long-term pollution of soils, sediments and of related food chains. CLD may be transferred into products from farm animals reared outdoors, through involuntary polluted soil ingestion. However, due to different properties of clays, tropical volcanic soils display variable capacities of pollutant retention: CLD is less available to plants and more persistent in andosol than in nitisol. In order to assess the risk of animal products’ contamination by soil-bound CLD, the impact of soil type on CLD bioavailability to animals has to be established. It was assessed in three farm animal species differing for their digestive tract properties (laying hens, piglets and lambs), through relative bioavailability (RBA) studies. Thus, the response of CLD ingestion through andosol and through nitisol was compared to the response obtained with CLD ingestion through oil, taken as a reference matrix. Our hypotheses were that i) CLD would be less available in soils than in oil, ii) CLD would be less available in andosol than in nitisol and iii) RBA in soils may differ between animal species. The deposition of CLD in egg yolk (hens), in liver (piglets) and in serum (lambs) was measured in individually housed animals fed graded levels of CLD from polluted andosol, nitisol or spiked oil. Hens, piglets and lambs were exposed to CLD during 28, 14 and 15 days, respectively. For each animal species, the concentration of CLD in target tissue linearly increased with the amount of ingested CLD within each ingested matrix (P< 0.001). However, the responses to andosol-diets, nitisol-diets and oil-diets could not be differentiated (P>0.1), indicating that CLD was equally bioavailable, irrespective of the matrix. The current experiments clearly show that, for the three animal species, soil does not modulate CLD availability. Therefore, involuntary ingestion of polluted soils by farm animals must be considered as potential contributor to the risk of contamination of animal products. In addition, accounting for the higher CLD concentration in andosol compared to nitisol, raising animals on andosol would result in a higher risk of products’ contamination with CLD than on nitisol. 190 Genotoxic impact of oil sands process-affected water to rainbow trout: what is the role of naphthenic acids?E. Lacaze; A. Devaux, Université de Lyon-ENTPE; F. Gagne, Emerging Methods Environment Canada / Emerging Methods. Oil sands exploitation has raised major environmental concerns, particularly regarding the presence of naphthenic acids found at high concentration in the oil sands process-affected water (OSPW). Naphthenic acids have been proved to have toxic effects on a variety of organisms including plants, fish, and mammals. However, their role in the toxicity of OSPW is not well known. Along the Athabasca River in Alberta, Canada, extraction of oil 2 sands occupies almost 600 km of boreal forest and generates a huge quantity of wastewater, released in tailing ponds. In a previous study, Athabasca water extracts in the vicinity of an oil sand extraction site were able to induce DNA damage in trout hepatocytes. Whereas naphthenic acids are suspected to be responsible for the largest part of toxicity in tailing ponds, nothing is known concerning their potential genotoxicity. The purpose of this study was to evaluate the genotoxic impact of oil sands process-affected water (OSPW) and OSPW-related contaminants to an indigenous fish species in order to investigate the role of naphthenic acids. For this purpose, rainbow trout hepatocytes were exposed in vitro to extracts of OSPW (oil sands processed water, obtained by a laboratory bitumen extraction), oil sands leachate water (OSLW, mimicking lixiviation of the shore) and environmentally relevant concentrations of naphthenic acids mixture, synthetic naphthenic acids found in tailing ponds, and PAHs. DNA damage was assessed by the Fpg-modified Comet assay. Genotoxic impact was observed in hepatocytes exposed to OSPW and OSLW extracts in a dose-dependant manner, and genotoxicity was more marked for OSPW

SETAC Europe 23rd Annual Meeting

than for OSLW. Furthermore, exposure to a mixture of naphthenic acids at concentration lower than those measured in runoff water has resulted in higher genotoxic impact compared to benzo(a)pyrene exposure at concentration found in tailing ponds. These results proved that the genotoxicity of OSPW could also be due to naphthenic acids. High concentrations of naphthenic acids in runoff and ground waters in the vicinity of oil-sands extraction (24 and 50 mg/L respectively), could represent a clear ecotoxicological risk. Further research has to be conducted to understand the relationship between the structure of naphthenic acids and their genotoxicity, to finally gain knowledge about the potential reclamation of tailing ponds. 191 Chemical characterization and treatment by enhanced coagulation of dissolved organic matter in biochemical-treated effluent of textile wastewater Y. Lu, Peking University / Department of Environmental Engineering, School of Environment and Energy, Peking University Shenzhen Graduate School; S. WU, Beijing Gaia Technology Center Co. Ltd.; L. Zhao, School of Environment and Energy, Peking University Shenzhen Graduate School; J. NI, Peking University / Department of Environmental Engineering. Performed with regular physicochemical-biochemical combined treatment process, textile wastewater effluent was treated by enhanced coagulation with a new kind of coagulant. Dissolved organic matter (DOM) in biochemicaltreated effluent of textile wastewater as well as effluent treated by enhanced coagulation were isolated and enriched by extraction. Both DOM isolates were characterized by Gas Chromatography-Mass Spectrum (GC-MS) and both effluents by Excitation Emission Matrix (EEM). Results showed that biochemical-treated textile wastewater effluent mainly contained hydrophilic organic matters: triethylene glycol, tributyl phosphate and phthalates, and hydrophobic organic matters: amino acids and humic substances. And after enhanced coagulation treatment, most of the dissolved organic matter was removed from the biochemical-treated textile wastewater effluent. 192 Removal of hexavalent chromium from aqueous matrices using plantain (Musa paradisiaca l.) peel biomass B.O. Opeolu, Cape Peninsula University of Technology / Faculty of Applied Sciences; O.S. Fatoki, Cape Peninsula University of Technology; O. Olatunji, Cape Peninsula University of Technology / Department of Chemistry. Chromium (VI) has found application in metallurgical, tanning, textile 6+ and paint industries. Hexavalent chromium (Cr ) is very toxic and mutagenic when inhaled. It is also one of the substances whose use is restricted by the European Restriction of Hazardous substances 6+ Directive. The use of Cr in various industrial processes has led to its presence in industrial wastewaters and consequently, surface waters. Abatement methods such as chemical precipitation and reverse osmosis are often expensive when available in the developing world. Furthermore, most conventional removal methods have been associated with the generation of secondary wastes such as sludge. This work therefore aimed at assessing the removal capacity of plantain peel for 6+ Cr from aqueous solutions. Parameters studied include effects of contact time, PH, temperature, weight, particle size, and shaking period. Physico-chemical characterization of the biomass (plantain peel) was 6+ carried out and residual Cr measured using atomic absorption spectrophotometry. The optimal contact time, concentration, o temperature and weight for adsorption were 90 min, 100 mg/L, 50 C 6+ and 0.4g respectively. Cr removal increased with shaking time and 6+ carboxyl and hydroxyl groups seem to be responsible for Cr binding. Percentage removal from paint and textile effluents was over 90%. Plantain peel biomass may therefore be an innovative alternative to imported expensive synthetic resins in Africa. 193 Nutrients, pharmaceuticals, and antibiotic resistance genes in municipal wastewater after wetland treatment: A case study at Grand Marais, MB, Canada J. Anderson, Department of Chemistry; J.C. Carlson, University of Winnipeg / Richardson College for the Environment; J.E. Low, University of Winnipeg / MSc student; J. Challis, University of Winnipeg; C.S. Wong, University of Winnipeg / Richardson College for the Environment; C.W. Knapp, University of

49

Strathclyde; M.L. Hanson, University of Manitoba / Department of Environment and Geography. The discharge of complex mixtures of nutrients, organic micropollutants and microbes bearing antibiotic resistance genes (ARGs) from treated municipal wastewater into freshwater systems are global concerns for human health and aquatic organisms. Use of lagoon treatment systems for managing rural wastewaters is common across the North America. In the rural community of Grand Marais, Manitoba, Canada, wastewater is treated passively in sewage lagoons followed by passage through a treatment wetland prior to release into surface waters. Using this facility as a model system, the aim of this study was to assess the presence of nutrients, micropollutants, and ARGs in lagoon outputs and their potential removal by the wetland. Nutrients, organic micropollutants (i.e., pesticides, pharmaceuticals), and standard water quality parameters in the system were characterized prior to and following lagoon discharge events into the treatment wetland over in 2012 and the quality of the water assessed by comparisons to regulatory guidelines where they exist and through hazard quotients. Quantitative PCR was used to measure the abundances of ten ARGs, as well as 16S-rRNA, in order to evaluate the effectiveness of the wetland in removing ARGs. As expected, concentrations of nitrogen and phosphorus species were greatest in the lagoon and declined with movement through the treatment system. Pharmaceutical and agricultural chemicals were detected at concentrations in the ng/L range and predictably, concentrations spiked downstream of the lagoon following discharge. Only atrazine and carbamazepine were shown to be significantly attenuated by processes within the wetland. Hazard quotients calculated for micropollutants of interest indicated minimal toxicological risk to aquatic biota. There was no significant removal of ARGs in the wetland and the results suggest that up to 2.7% of the bacterial population in this system may have genes imparting antibiotic resistance. The results of this study indicate that while the treatment wetland may effectively attenuate excess nutrients and remove some micropollutants and bacteria, it does not specifically target ARGs for removal. Additional studies would be beneficial to determine whether upgrades to extend retention time or alter plant community structure within the wetland would optimize removal of micropollutants and ARGs. 194 First determination of diclofenac, sulfamethoxazole and their nitration transformation products in wastewaters and evaluation of their acute toxicity V. Osorio; J. Sanchis, IDAEA-CSIC; J. Abad, IQAC-CSIC; A. Ginebreda, M. Farre, S. Perez, D. Barcelo, IDAEACSIC. Diclofenac (DCF) is a non-steroidal drug with anti-inflammatory and analgesic effects, widely used for treatment of rheumatic diseases and pain relief. Sulfamethoxazole (SMX) is a sulfonamide bacteriostatic antibiotic extensively used in both veterinary and human medicine. These pharmaceuticals are of environmental concern due to frequent detection in monitoring surveys on sewage-impacted surface waters associated with high consumption rates and low removal efficiencies during conventional activated sludge treatment in wastewater treatment plants (WWTPs). DCF is proposed to be included in the priority list of organic pollutants in surface waters. Despite this, the overall fate of these compounds has as of today been considered only marginally and little is known about the biotransformation processes taking place in the aeration tank of the activated sludge treatment. The formation of two transformation products (TPs), nitroso-DCF and nitro-DCF during conventional treatment, was observed. Nitration of DCF and SMX was also reported in aquifers under denitrifying conditions. Despite natural attenuation of these compounds, it is assumed that they can behave as pseudo-persistent compounds simply because of their continuous entry into the aquatic environment via wastewater (WW) effluents. The generation of nitrated derivatives in the WWTPs is a highly relevant matter of concern because of their potential toxicityThis research aimed at investigating still uncovered aspects of the fate of DCF in the aquatic environment by determining the occurrence of the TPs, formed in the WWTP, in influent, effluent and also in surface waters samples. In addition, the potential ecotoxicological effects of DCF, nitro-DCF, nitroso-DCF, SMX and nitro-SMX on the aquatic organisms Daphnia magna and Vibrio fischeri were assessed. For that purpose, one

SETAC Europe 23rd Annual Meeting

objective was to develop a quantitative analytical methodology for trace determination of compounds targeted, based on off-line solid-phase extraction (SPE) followed by LC/ESI-MS/MS analysis on a hybrid quadrupole-linear ion trap (QqLIT) instrument. 195 Continuous monitoring of turbidity and conductivity in wastewater network: an easy tool to assess the pollution load discharged into receiving water T. Bersinger, LCABIE IPREM; G. Bareille, LCABIE IPREM UMR 5254; T. Pigot, IPREM ECP UMR 5254; I. Le Hecho, LCABIE IPREM UMR 5254. Sanitation represents a major threat for the aquatic environment since it is collecting a wide variety of sources from both domestic and industrial activities and it is continuously discharging various polluted effluents. Urban stormwater runoff which is considered as transient source may also contain high levels of pollutants discharged directly in receiving streams by Combined Sewer Overflow (CSO). Continuous monitoring of no treated wastewater via CSO becomes a key issue for management of sanitation and protection of receiving environment. Monitoring based only on samples being technically and economically unrealistic it is thus necessary to turn to an indirect measure of pollutant parameters. This can be done via the measurement of physical parameters such as turbidity and conductivity. In such a context we have studied the urban catchment of Pau urban area (south west France) which is about 50 km² with about 150 000 inhabitants. 4 rain gauges, 40 flowmeters, 3 turbidimeters and conductimeters were installed in all the study area to monitor the sanitation and particularly CSO discharges. After intensive and various samples periods it has been possible to correlate safely in one hand global parameters such as Chemical Oxygen Demand (COD) and Total Suspended Solid (TSS) to turbidity and in the other hand conductivity and total nitrogen. These correlations will be first presented. The sensors were placed at the entrance of the wastewater treatment plant (WWTP) and main CSOs. Continuous recording of turbidity and conductivity have been achieved during one year. Use of acquired data allows a real time evaluation of: - Pollutant fluxes directly discharge by CSO and treated by WWTP - Input of sanitation on receiving streams - Micro pollutant fluxes (metals and hydrocarbons aromatic polycyclic) mainly associated to suspended solids This continuous monitoring is innovative for wastewater network management. This allows continuous estimation of discharged wastewater concentrations into the receiving environment. The sanitation manager can then resize and optimally manage the treatment and wastewater storage.\n 196 Is it possible to increase bioavailability but not environmental risk in PAH bioremediation? J. Ortega-Calvo, Instituto de Recursos Naturales y Agrobiologia / Agroquimica y Conservacion del Suelo; M. Tejeda-Agredano, Instituto de Recursos Naturales y Agrobiología de; C. Jimenez Sanchez, Inst de Recursos Naturales y Agrobiol de Sevilla; L. Congiu, European Food Safety Authority; R. Sungthong, Department of Agrochemistry and Soil Conservation; M. Cantos, Instituto de Recursos Naturales y Agrobiología de Sevilla (IRNAS-CSIC). The current poor predictability of end points associated with the bioremediation of polycyclic aromatic hydrocarbons (PAHs) is a large limitation when evaluating its viability for treating contaminated soils and sediments. This fact is usually caused by a poor bioavailability of the chemicals for biological destruction by microorganisms. Due to their partitioning into sorbents and NAPLs, the chemicals may exhibit only weak chemical activity gradients that promote their uptake and transformation by active microbial cells. Hence, the biodegradation rates may reflect the dependencies of restricted phase exchanges, and, as a result, a priori biodegradable PAHs may present longer persistence. Biodegradation may be effective in reducing risks, but recalcitrant pollutants, left behind by bioremediation, may still follow exposure routes specific to humans and ecological targets. Therefore, innovative methods are needed to increase bioavailability of these pollutants for an enhanced bioremediation performance. This overview contribution will examine our recent research efforts in this issue, performed at IRNAS-CSIC, Spain, including surfactant applications to promote biodegradation in soils exhibiting a slow-desorption profile (Environ. Sci. Technol.

50

45:3019-3026, 2011), targeted fertilization of free-oil phases or NAPLs (Environ. Sci. Technol. 45:1074-1081, 2011), modulating deposition and motility of degraders in porous media (Environ. Sci. Technol. 46:67906797, 2012), and promoting bioavailability with plants (Soil Biol. Biochem. In press, 2012). The integration of these strategies into practical remediation protocols would be beneficial to the bioremediation industry, as well as improve the quality of the environment. 197 Classification and modelling of non-extractable residue (NER) formation from xenobiotics in soil – a synthesisM. Kästner, Helmholtz Centre for Environmental Research UFZ / Department of Environmental Biotechnology; K. Nowak, RWTH Aachen University / Institute for Environmental Research; A. Miltner, Helmholtz-Centre for Environmental Research / Department of Environmental Biotechnology; A. Schaeffer, RWTH Aachen University / Chair of Environmental Biology and Chemodynamics; S. Trapp, Danmark Tekniske Universitet. This presentation provides a comprehensive overview about the formation of non-extractable residues (NER) from organic pesticides and contaminants in soil and tries classifying the different types. Anthropogenic organic chemicals are deliberately (e.g. pesticides) or unintentionally (e.g. polyaromatic hydrocarbons [PAH], chlorinated solvents, pharmaceuticals) released in major amounts to nearly all compartments of the environment. Soils and sediments as complex matrices provide a wide variety of binding sites and are the major sinks for these compounds. Many of the xenobiotics entering soil undergo turnover processes and can be volatilised, leached to the groundwater, degraded by microorganisms or taken up and enriched by living organisms. Xenobiotic NER may be derived from parent compounds and primary metabolites that are sequestered (sorbed or entrapped) within the soil organic matter (type I NER) or can be covalently bound (type II NER). Especially type I NER may pose a considerably environmental risk of potential release. However, NER resulting from productive biodegradation, which means the conversion of carbon (or nitrogen) from the compounds into microbial biomass molecules during microbial degradation (type III, bioNER), do not pose any risk. Experimental and analytical approaches to clearly distinguish between the types are provided and a model to prospectively estimate their fate in soil is proposed. 198 Hexachlorobenzene-contamination of sediments of the Upper Rhine River – Impact of clay minerals on the deviation of KOC values L. Boehm, Justus Liebig University Giessen / Institute of Soil Science and Soil Conservation; T. Pohlert, Federal Institute of Hydrology (BfG); R. Duering, Justus Liebig University Giessen / Institute of Soil Science and Soil Conservation. Due to historical emissions of hexachlorobenzene (HCB), the sediments of the ponds of Upper Rhine River in southern Germany remain highly contaminated -1 (up to 7000 µg kg ). HCB is a bioaccumulative, toxic, and persistent hydrophobic organic pollutant (HOC), and is mainly bound to organic matter (OM). OM (quantified as total organic carbon, TOC) and its content is mostly considered to be the significant parameter controlling the extent of sorption. Therefore, sorption coefficients K of samples are d specified as K values, normalized to their TOC content. Recently, the OC influence of inorganic components on HOC sorption increasingly has been identified as relevant. Knowledge of the specific HCB-sorbents in the different fractions is of importance, because their characteristics control sorption strength, which is an important factor for remobilization as well as bioavailability of HCB. The mineral composition of sediments (measured by X-ray Diffraction, XRD) and its influence on the amount of sorption is presented, as well as composition of and sorption to sediment fractions. Sorption and desorption isotherms and partition coefficients (K , K , K ) for sediments, sediment fractions, and d OC des reference minerals were determined in batch equilibrium studies by solid-phase microextraction (SPME) coupled to GC-MS. Sorption studies resulted in log K values of sediments from 2.8 to 3.5 and log d K values from 4.5 to 5.0. K values of the sediments yield similar OC d K values, except for two highly contaminated sites, where the OC amount of sorbed HCB could not be explained solely by TOC nor by

SETAC Europe 23rd Annual Meeting

particle size distribution and specific surface area. However, this deviation can be explained by sorption to the mineral fractions. Sorption studies with clay suspensions resulted in very high log K values from d 2.9 to 3.5. The pattern of sorption intensity measured in clay suspensions matched the pattern of deviation for K values of total OC sediment fractions. Hence, sorption of HCB to minerals plays a significant role in sorption processes, in particular due to their high occurrence. Results on physicochemical properties of sorbents contribute to a better understanding of the environmental behaviour and pathways. Knowledge of sorption strength of specific adsorbents as well as desorption processes and hysteresis facilitates the location of contaminated sites which function as key sources of HCB remobilization and should therefore be prioritised in remediation purpose. 199 Sorption capacity of agricultural hormones to different particle size fractions of a natural aquatic sedimentJ. Sangster, University of Nebraska / Civil Engineering; D.D. Snow, University of Nebraska / Associate Professor; S. Bartelt-Hunt, University of Nebraska-Lincoln / Civil Engineering. Increasing evidence indicates that agricultural production constitutes a significant source of steroidogenic compounds to aquatic systems. While several studies have explored the fate and transport of steroidogenic compounds in agricultural systems, relatively little is known about the effects of soil or sediment particle sizes on steroid fate in the environment. This study explores the partitioning of agricultural hormones, estradiol (E2), estrone (E1), progesterone (Pr), testosterone (T), and 17?-trenbolone (Tb), to different size fractions of a natural sediment and evaluates how competition between particle size fractions effect the distribution of hormones in the sorbed phase. Experiments were performed using a natural sediment fractionated using both sieve and deposition methods. Five fractions of the sediment were used in sorption experiments to determine the sorption capacity of the fractions for five common agricultural hormones.Based on the results of sorption capacity experiments, several trends are evident from the data. All hormones evaluated exhibit significantly higher sorption capacity in the clay and collloid fractions when compared to the sand and silt fractions. This may be at least partially attributed to an increased surface area within the smaller particle fractions. Hormone sorption to colloids is not significantly different from sorption capacity of the clay fraction for E2, E1, and Pr. However, in the case of T, colloids exhibited a lower sorption capacity than that of the clay fraction. The estrogens, E2 and E1, sorbed at comparable levels for both the sand and silt fractions. In contrast, Pr and T, in the sand and silt fractions behave significantly differently. Results for Tb have not been evaluated as of the time of this abstract submission, nor have the results of sorption with competeion between partcle size fractions. The data from this study may help in our understanding of steroid transport and bioavailability in aquatic systems. 200 The uptake of PPCPs in curcubits and the influence of soil factors on bioavailability M. Williams, CSIRO / Land and Water; E. Harris, Cranfield University; S. Martin, J. Kirby, R. Kookana, CSIRO. Relatively few terrestrial studies have been undertaken to assess the potential impacts of PPCPs in the terrestrial environment, with those that have demonstrating that uptake of PPCPs in plants from either contaminated soil or irrigation water can occur. The aim of this study was to assess the uptake of four PPCPs, the human pharmaceuticals carbamazepine (CBZ), propranolol (PRL) and fluoxetine (FLX) and the personal care product triclosan (TCS), in the curcubit zucchini (Curcubita pepo spp. pepo var. Black Beauty) at nominal initial soil concentration of 5 mg/kg. C. pepo was grown in two different soils (Inman and Tepko) over a 28 day period within a growth chamber. Tepko soil was further modified through the addition of 0.5% w/w biochar derived from wheat and eucalyptus, respectively, to assess its impact on uptake.. In all soil treatments, there was a large extent of uptake of CBZ in C. pepo, with bioconcentration factors (BCFs) ranging from 1.8±0.7 to 13.5±0.2. Conversely, there was comparatively little uptake of PRL or FLX with respective BCFs ranging from 0.008±0.001 to 0.1±0.01 (PRL) and 0.014±0.004 to 0.02±0.01 (FLX). TCS was not detected in plant residues from any of the soil treatments. Monitoring of soil concentrations over the 28 d period suggested that TCS was possibly

51

transformed, with a decrease in its soil concentrations from 20 to 43% of original values. Porewater concentrations of the PPCPs were also monitored over the 28 d C. pepo growing period, with TCS and FLX unable to be detected in any porewater samples extracted from the soil treatments. In contrast, high levels of CBZ were measured in porewater, corresponding with K values ranging from 1.5±0.1 to 15±0.7 L/kg. The d K values measured for PRL were considerably higher and ranged from d 62±28 to 753±197 L/kg. For both CBZ and PRL, addition of both wheat and eucalypt-derived biochar reduced the extent of uptake relative to the unamended Tepko soil, although uptake in the amended soils was still comparable with the Inman soil. Dissipation of TCS in biochar-amended soil occurred but was found to occur at approximately half the rate of that in the unamended soil. Work is currently being undertaken to adapt a stable isotope technique, commonly used in for the assessment of inorganic contaminants, to measure and predict the bioavailable fraction of the PPCPs present in the soil. 201 Atrazine degradation in a Brazilian Acrisol under no-tillage d. dick, Federal University of Rio Grande do Sul / PhysicalChemistry; D. Barbosa, Federal University of Rio Grande do Sul / Soil Science Department; P. Burauel, Forschunszentrum Jülich / Institute of Chemistry and Dynamics of the Geosphere, IBG 3; A. Lelickens, Institute of Chemistry and Dynamics of the Geosphere, IBG 3 - FZJülich. The pattern of ATZ degradation (mineralization and extractable residues) in a Brazilian Acrisol cropped with maize under no-tillage was evaluated, focusing on the roles of the ATZ application history and of the cover straw on these processes. Incubation experiments were conducted for 85 days with soil samples from an Acrisol (0-10 cm) located in southern Brazil. Four treatments were used: Microcosm 1, cultivated soil since 1991 and with history of ATZ.application; Microcosm 2, native soil from an adjacent area; Microcosm 3, same soil as Microcosm 1, covered with straw (Avena sativa); Microcosm 4, similar to Microcosm 1. A 14 spiking solution was prepared using technical-grade and C-ATZ in ethanol and applied to the Microcosms. For monitoring of water and ASE extractable radioactivity soil samples were analyzed at 0, 9, 16, 30, 14 and 85 days of incubation. The C activity was measured by liquid scintillation counter. Cultivated soil presented higher ATZ mineralization (86%) than native soil (10%) by the end of incubation. 14 The water-extractable C activity decreased from 69 to 1% and from 62% to 20% in Microcosms 1 and 2, respectively. The amount of ASE14 extractable C ranged from 8 to 17% in Microcosm 1 while in Microcosm 2 it increased from 15 to 51%. Besides indicating the lesser persistence of ATZ in the cultivated soil, these results suggest that ATZ was transferred to a soil compartment less available to its mobilization 14 during the incubation. The amount of non-extractable C-residues were lower than 0.08%. In Microcosm 3, that included cover straw, the mineralization rate was slower than in Microcosm 4 (soil without straw) th 14 and at the 85 day the mineralized C ATZ amount reached 30% and 14 85 % respectively. The water extractable C-residues in Microcosm 3 th were higher in the straw than in the soil until the 30 day of incubation th and reached 9.2% in the soil and 7% in the straw at the 85 day. In the 14 soil of Microcosm 3, the amount of ASE-extractable C increased from 14 2 to 13% and that of non-extractable C-residues from 0,9 to 8% along 14 the incubation. The water extractable C-residues in the straw increased with time and were greater than in the respective soil. Our results corroborates the accelerated degradation of ATZ in cultivated soils. The cover straw, frequently found in Brazilian agricultural soils under no-tillage represents an important sorbent for ATZ, reducing its mineralization. 202 First report of pyrethroid insecticides in birds: A case study on Doñana National Park (Spain) C. Corcellas, IDAEA / Química Ambiental; A. Andreu, EBD-CSIC; F. Sergio, F. Hiraldo, EBD-CSIC / Applied Biology; E. Eljarrat, D. Barcelo, IDAEA-CSIC / Química Ambiental. In last decades the usage of pyrethroids has increased widely in the indoor as household insecticides, insect-control products, pet shampoos and lice treatments, and in the outdoor as agricultural pesticides and for pest control. Because of that, they are almost ubiquitous and, as long as they are dumped continuously, they will be

SETAC Europe 23rd Annual Meeting

always present in the environment. Even when their toxicity was supposed low for non-target organisms, it is known the high toxicity in fish. Despite the assumption that these insecticides are converted to nontoxic metabolites by hydrolysis in mammals, recent studies showed their presence in dolphins or even in human breast milk. These studies points out a potential bioaccumulation of these insecticides. The aim of this work was studied the presence of 13 common pyrethroids in infertile bird eggs of different species from different trophic level, collected at the Doñana National Park, a protected reserve in the South of Spain, of high environmental interest because all migratory Afro-European birds cross through it. For the first time, pyrethroids were detected in all egg samples analysed, at concentration levels ranging between 0.75 and 25 ng/g lw. These results confirmed the bioaccumulation potential of these insecticides in biota samples, corroborating the previous findings in marine mammals and humans. Of the 13 investigated pyrethroids, resmethrin and fluvalinate were usually under the limit of quantification. However, cypermethrin and permethrin were the more ubiquous pyrethroids, followed by bifenthrin and tetramethrin. As regards the contamination of each bird specie analysed, the less contaminated corresponded to glossy ibis, whereas white storks and black kite were the most contaminated ones. These results could indicate that pyrethroids bioconcentrate along the food chain, and that because the lower levels corresponded to the species of lower trophic level, while the highest concentrations were found in the higher trophic level species. Finally, enantiomeric characterization was carried out in order to determine potential enantiomeric-specific accumulation of pyrethroids. This study seems to point out some enantiomeric-selective accumulation. This is the case of enantiomeric-selective accumulation of the enantiomer II of trans-tetramethrin observed in egg bird samples analysed. 203 Bird communities in chlorpyrifos-treated citrus in Spain: 3 years intensive field monitoring of species-diversity, abundance & reproduction 2010-12 R. Dittrich, Tier Solutions GmbH / Wildlife Ecology; B. Giessing; S. Norman, Ridgeway Eco; B. Striffler, Tier3 Solutions GmbH; C. Wolf, Tier Solutions GmbH. The organophosphate insecticide chlorpyrifos (CP) has been the foundation of red scale control in citrus cultivation in Spain for decades. Chlorpyrifos has a relatively high acute toxicity to birds in lab studies so the regulators require intensive field evidence to demonstrate safety to birds. Therefore the long-term response of the bird community to CP has been studied in citrus for 3 yrs (2010-2012; 2.4 kg/ha) culminating in probably the largest bird study ever done on a pesticide. A comprehensive approach was followed which highlights the structure of the bird community over consecutive years. This allows measurement and ranking of any factors influencing the local bird community, including the application of CP itself. The bird communities of ten CP treated citrus orchards in the Xativá valley (Valencia province, Spain) were studied during the breeding seasons from 2010 to 2012. Following common practice a spray of chlorpyrifos (2400 g a.i./ha) was done on all ten sites between end of May and mid-June. Methods applied were regular bird trapping with mist nets in all sites and years. Trapped birds were sexed, ringed and checked for signs of reproduction. Also systematic nest searching was conducted by periodical check of every tree over the breeding season. Active nests were monitored until the young birds had fledged or the nesting failed. The bird community found in CP treated citrus orchards was diverse and abundant with13,511 birds of 82 species trapped in three years. The total number of birds trapped and the species composition varied between the ten sites. Also the nest density varied between sites, in particular between one site (site 4) and the other nine sites (due to difference in surrounding habitats). The average rate of successful nests was 32 % of all active nests in 2010-2012. The breeding success and density of nests depends on the presence of predators and pruning activities. The re-trapping of individuals year to year indicates long term survival of adults in CP-treated orchards. Overall, no clear effects at the population and community level could be attributed to CP. 204 The use of monitoring results of plant protection products in surface water in Product Stewardship and PPP authorisation R.

52

de Werd, WageningenURApplied Plant Research / Applied Plant Research; R. Kruijne, WageningenUR - Alterra; W. Tamis, Leiden University - Institute of Environmental Sciences. Exceedances of environmental quality standards (EQS) for plant protection products (PPP) in Dutch surface water need to be further reduced. A procedure for the use of PPP monitoring results for ERPs and a feedback to PPP authorisation has been developed and tested. It is based on responsibility for the registration holder and obligatory measures if necessary. It consists of: 1) identification and ranking of problematic substances 2) analysis of plausible causes for the exceedances and the composition of an ERP 3) feedback to the board of PPP authorisation and ministries involved Yearly substance selection and ranking is carried out, based on the last three years of monitoring results for PPP in the Dutch Pesticides’ Atlas. Substances exceeding the EQS in the Water Framework Directive (WFD) water bodies are selected. WFD priority substances are ranked highest. The remaining selected substances are given points based on water body type, exceedance degree and percentage of monitoring locations with exceedances. Registration holders can be requested a causal analysis and ERP, triggered by: 1) a prolongation request for an existing registration, 2) a new registration request for a substance already on the Dutch market, or 3) top position in the list. The causal analysis searches for plausible relations between specific applications and emission pathways on the one hand, and standard exceedances on the other hand. First a fact sheet is compiled, covering i) substance properties (ii) authorised use (iii) usage statistics; (iv) indicators for the emission to surface water & crop maps, and (v) monitoring results. Next, experts are consulted, using predefined questions referring to the factsheet. If after interpretation the plausible causes are identified, the protocol manager draws the final conclusions. Alternatively, additional fact finding and expert consultation is carried out for certain topics or regions. The final conclusions are the fundament for the ERP. Causal analysis and ERP are reported to the responsible authorities. If plausible that authorised use (applying Good Agricultural Practice) led to standard exceedances, and the ERP is expected to be insufficient, there may be obligatory consequences for the registration. Implementation of this methodology is expected to contribute to the reduction of standard exceedances, whilst respecting the registration holders’ responsibility. The standardised methodology facilitates transparency and substantiated mitigation measures. 205 National Pesticide Screening for the Selection of Relevant Compounds for Improved Future Monitoring StrategiesC. Moschet, Eawag / Uchem; A. Piazzoli, ETH; I. Wittmer, H. Singer, C. Stamm, Eawag; M. Junghans, I. Werner, Oekotoxzentrum; J. Hollender, Eawag. Monitoring and risk assessment of agricultural and urban pesticides are critical issues. In most countries, pesticide monitoring focuses on herbicides and is not carried out systematically. The goal of this study was to define relevant compounds from all pesticide classes on a national scale which are recommended to be monitored in future activities. The study consisted of a systematic, step-wise procedure with a combined theoretical, analytical and ecotoxicological approach including the screening of nearly all pesticides registered in Switzerland. In a first step, all registered pesticides were ranked according to sales numbers, organic carbon to water partition coefficients and degradation rates in soil. The 156 substances with the highest ranking (occurrence in surface water likely) were selected for a quantitative target screening using high-resolution mass spectrometry. The pesticides with lower rankings were qualitatively screened in the samples (suspect screening). All substances were measured in 5 catchments across Switzerland with different cropping pattern between March and July 2012. The analytical method covered 88% of all registered organic pesticides at low detection levels. From the 156 target substances, 97 were detected in at least one sample; another 13 substances were detected in the suspect screening. Results showed that 95-percentile herbicide concentrations were on average 2.3 times higher than fungicide concentrations and even 6.9 times higher than insecticide concentrations. The ecotoxicological risk assessment showed that in a few samples some substances already exceeded their environmental quality standard (EQS). From these results, Switzerland-relevant pesticides were selected. It was shown that

SETAC Europe 23rd Annual Meeting

besides substances already monitored (e.g., azoxystrobin, isoproturon), less known substances (e.g., thiamethoxam, piperonyl butoxid) were also found to be relevant. Whereas some substances were only important in catchments with a specific cropping pattern, others were found throughout the catchments. As expected, pesticides detected depended on land use e.g., highest insecticide concentrations were found in catchments with high vegetable or orchard density. Risk estimation of combined exposure allowed identification of critical catchments where risk mitigation measures are necessary. This integrated approach provides stakeholders with important information in the planning of future monitoring and risk assessment programs. 206 Monitoring effects of pesticides on pollinators – a review of methods and outcomes by the ICPPR working groupA. Alix, Dow Agrosciences / Risk Management; A. Bartels, AGES; C. Garrido, IBACON; C. Hart, PMRA; K. Knauer, Bundesamt für Landwirtschaft BL; J. LAPORTE, Syngenta; B. Maurizi, Ministry of Agriculture; C. MAUS, Bayer Ag; M. Miles, Dow Agrosciences; C. Schneider, BASF; H. Thompson, FERA; J. Wassenberg, CTBG. Monitoring studies, in the context of the environmental assessment of Plant Protection Products (PPP) or pesticides, aim at getting feedback regarding the fate and/or effects of active substances and/or their relevant degradation products in/on the environment, when PPP are used under realistic conditions for crop protection. These studies complement the risk assessment performed in application of Regulation 1107/2009/EC and previous texts, which aims at identifying the conditions of exposure of organisms in the environment, the conditions of occurrence of risks if necessary and propose appropriate risk mitigation measures. In this context, monitoring studies are required to complement the risk assessment in addressing possible uncertainties that cannot be fully addressed through field studies for time/space scale reasons. Monitoring studies are also a way to validate or adjust the risk mitigation measures that may have been recommended as a condition of approval of the product. Monitoring effects of pesticides on honey bees has been getting more importance over the last years and has been recommended along with approval decisions for some some active substances. However there exists no harmonized guidance on monitoring methodology and this raises legitimate questions about how to use the generated data in support of decision making as well as on possibilities for extrapolations. This presentation illustrates the work undertaken by the International Commission on Plant-Pollinator Relationship to review existing monitoring of the effects of pesticides on managed and wild bees and propose guidance on good monitoring practices. 207 Guideline and practical tools for a sustainable use of Plant Protection Products within the framework of the Sustainable Use Directive M. Calliera; F. Berta; P. Meriggi; R. Rossi; T. Galassi; F. Mazzini; A. Bernard; R. Bassi; A. Di Guardo; A. Marchis; E. Capri, . In 2008-2009,a survey in the Emilia Romagna region of north Italy collected information on the farm use of Plant Protection Products (PPPs) and evaluated whether the provisions of the Directive 2009/128/EC for the Sustainable Use of Pesticides (SUD) are applicable. It was concluded that the provisions of the Directive can be implemented, even if some gaps need to be filled and also the behaviour of farmers needs to be improved. Moreover, it was observed that all stages in the use of PPPs on farms could generate risks for the operator and/or the environment. One of the Directive recommendations is to promote training for operators and to adopt good agronomic practices in order to improve sustainable use of PPPs. The findings were used to develop a Guideline for Sustainable Use of PPPs to help the user in identifying the flaws in current practices at farm level as well as their corresponding corrective actions. The Guidelines are accompanied by free on-line software (www.agricoltura-responsabile.it) to be used as a diagnostic tool as well as to provide recommendations for improvements. The project had a stepwise approach: Step 1: Preliminary investigation and survey on a sample of 100 farms. The main objective was to identify the potential environmental and human health pressures of pesticide use at farm level, starting from pesticide delivery to the farm through to the final disposal of the packing materials. Step 2: Based on

53

the data collected in the survey and their analysis, the project participants decided to develop an operational, sustainable-use guideline to be used for training and raising awareness among professionals involved with pesticides. The objectives are to collect recommendations for responsible, safe and sustainable use of these products and at the same time to provide tools to evaluate how they are managed. Step 3: To fulfil the latter objective, the guidelines were equipped with a check list and a free on-line software to assist the user in the analysis of their agricultural practices and to identify the critical points in the pesticide management, the critical issues on the farm, taking into account both the structural aspects of the farm that the behavior of the farmer so as to prevent environmental contamination and to assure a high standard of safety for operators. The tools can be considered both as a support to train technicians and as an operational guide to provide an innovative assistance service with the purpose of improving the safe, responsible and sustainable use of pesticides. The approach adopted, taking into account the variability in farm structure, cropping pattern, risk attitude and economic availability, is not an instrument to identify the most suitable protection strategy for a given crop in a given period, but to help professional users to improve their practices in managing PPPs on farms and to make the most appropriate choices leading to reduced environmental and human risk, without compromising the profitability of agricultural production and food standards. 208 Climate change related impacts and benefits from wood extraction for bioenergy on a global scale P. Muchada, Radboud University Nijmegen; R. Van Zelm, Radboud University; M. van der Velde, IIASA; G. Kindermann, IIASA / Ecosystem Services and Management Program; M.A. Huijbregts, Department of Environmental Science. A current challenge in impact assessments of bioenergy is how to derive a scientifically valid greenhouse gas (GHG) life cycle balance as there are many processes that lead to GHG emissions during the life cycle of the biofuel. Common practice in life cycle impact assessment (LCIA) of forestry and wood has been to assume that any carbon dioxide (CO ) emission related to biomass combustion equals the 2 amount of CO absorbed, thus assuming a carbon neutral system with 2 no climate change impacts. This carbon neutrality assumption has been challenged because it ignores the damage caused by the CO during the 2 time it spends in the atmosphere and when a change in forest management occurs. In a study that addresses some recurring issues in LCIA namely land use, spatial variability, implications of value choices, and endpoint damage, we developed and applied a method that estimates biogenic climate change related impacts of global wood extraction for biofuel. We modeled changes in global forest carbon stock due to an increase in wood extraction with the global forestry model G4M on a 0.5°x0.5° scale. Resultant carbon emissions lead to climate change related damage to ecosystem quality and human health. Grid-specific and country-specific characterisation factors (CF) were derived as 2 disappeared fraction of species over space and time (PDF?m ?year) and disability-adjusted life years (DALYs) per unit of wood biomass 3 harvested (m wood). CFs were derived for three sceanrio’s, coinciding with three cultural perspectives, to show differences in time horizon, species protection, discount rate, and age weighting. Our results for the hierarchist scenario show that the wood biomass extraction for biofuel -3 -2 -3 has an effect of between -8.5?10 and 1.5?10 DALY?m wood on 3 4 2 -3 human health and between -9?10 and +2?10 PDF?m ?yr?m wood on ecosystem quality, depending on the extraction region. The negative values indicate climate change benefits obtained due to a change in rotation time of the forest ultimately leading to more carbon uptake. As a case study, we compared the life cycle impacts of energy production via wood with coal to determine the significance of including biogenic CO emissions due to changes in forest management in LCIA. Results 2 for the bioenergy production case study show that current life cycle assessment (LCA) methods underestimate, and in some cases overestimate the (climate change) impacts of wood used as a biofuel. 209 Inclusion of Climatic Tipping Potential in LCA S. Jorgensen, M.Z. Hauschild, Technical University of Denmark / Department for Management Engineering; P.H. Nielsen, novozymes a/s. Much

SETAC Europe 23rd Annual Meeting

attention is today given to the climate change driven by release of anthropogenic greenhouse gases (GHGs) to the atmosphere. Often this impact is only considered in terms of global warming potential (GWP) most commonly by applying the GWP . However, this metric does 100 not take into account the need for staying below certain climatic target levels, in order to avoid crossing critical and possibly irreversible climatic tipping points. Some suggestions have been made as to include a target level in climate change impact assessment, such as the global temperature potential, but with the consequence of disregarding impacts beyond that target level. The climatic tipping potential (CTP) presented here is suggested as a supplementary impact category to the GWP, treating the issue of climatic tipping potential relative to a specific target time. The CTP is expressed as the impact from a GHG emission compared to how much impact can still occur before reaching the given climatic target level. The latter is considered the ‘capacity’ of the climate system with respect to the target level. The capacity is diminished over time due to increasing atmospheric GHG levels, meaning that the CTP impact of GHGs increases as a function of closeness to the climatic target level at the time of emission. The relative impact of three GHGs, CO , CH and N O are explored and the results show that the relative 2 4 2 impact of the gasses is also affected by the time of GHG emission relative to the target time. The calculated CTP values of the different GHGs at release times from present untill the target time enable direct application of this metric in LCA. The inclusion of the CTP as additional impact category in LCA could enable better damage modeling and improve the value of LCA as a tool for decision support for climate change mitigation. Furthermore, the CTP impact category will be very useful for representing the climate change mitigating impact of temporary carbon storage in an analysed product system. 210 Endpoint characterisation modelling for marine eutrophication in LCIA N. Cosme, Technical University of Denmark DTU / DTUMAN QSA; H. Larsen, Danish Road Directorate / Research and Development; M.Z. Hauschild, Technical University of Denmark (DTU) / Dept. of Management Engineering. Marine eutrophication processes include the excessive growth of phytoplankton biomass in response to increased availability of nitrogen (N) in the photic zone of marine coastal waters. The eventual degradation of this biomass results in oxygen consumption in bottom waters by bacterial respiration. The excessive depletion of dissolved oxygen (DO) presents a potential impact to ecology, economy, and water quality. Land based human interventions are increasing the N loadings to marine coastal systems and overrunning their natural capacity to absorb N. Marine eutrophication is an impact category of LCIA which is still lacking a sound methodology to link midpoint and endpoint indicators and an overall model to assess the potential impact of the over-enrichment of marine ecosystems by N. Within the EU FP7 project LC-IMPACT, a model was built for the estimation of endpoint CF for marine eutrophication: Fate Factors (FF) were estimated for both airborne and waterborne N emissions based on modelling of river-N and marine-N losses to deliver the midpoint category indicator (i.e. increase of N in the marine compartment); Exposure Factors (XF) were estimated based on the Redfield ratio and a stoichiometric conversion of N in the photic zone to DO consumption in the benthic habitat. Finally, Effect Factors (EF) were estimated by applying the statistical Distribution of the Species Sensitivity (SSD) to hypoxia delivering the endpoint category indicator (i.e. species diversity loss). The product of the three factors, FF•XF•EF delivers the CF with the desired damage dimension, i.e. 3 (PAF•)[m •d/kg], to be applied to the emitted amount of N [kg]. The Large Marine Ecosystem (LME) biogeographical classification system was used to define the receiving spatial units. The model provides CFs for the “N to air” and “N to freshwater” inventory flows at a country-to-LME (214), country (143), region/continent (11), and global level. The proposed methodology covers all the processes of relevance to the marine eutrophication phenomenon and delivers CFs at a considerably high geographic applicability, good environmental relevance and reproducibility. The sensitivity and uncertainty analyses identified key issues for data quality improvement, namely PP datasets and N-export splitting rules for multiple receiving LME. Further improvements should

54

address the inclusion of spatial differentiation of N-losses within the fate models. 211 Characterisation factors for the impact category human noise S. Cucurachi, Institute of Environmental SciencesCML / Industrial Ecology; R. Heijungs, Institute of Environmental Sciences - CML. Leiden University. In 2010, the European Commission’s LCA handbook tried to set the standard for the best practise in LCA. The handbook suggested methodologies to be applied for different impact categories. Even though the handbook recognised the importance of the impacts caused by noise and the necessity of increasing the modelling effort of this impact category, no recommended approach was presented. As part of the LC-IMPACT project, we have developed a framework for the assessment of noise impacts on humans. This contribution presents the results of the operationalization of the proposed methodology. Characterisation factors (CFs) for human noise are presented at different levels of spatial differentiation, as well as temporal and frequency differentiation. The example of sound emissions is useful to discuss the right level of differentiation that is needed both at the LCI, as well as at the LCIA modelling phases. The influence of the input factors to the variance of the output is discussed, and may provide a useful insight of the way the variance of the output can be tested using techniques of global sensitivity analysis. Uncertainty and data availability are also taken into account when the transition from midpoint human noise to endpoint human health is described. \n 212 Land use in LCA: Global characterization factors based on regional and global species extinction L. de Baan, ETH Zurich IED / NSSI; M. Curran, C. Mutel, Institute of Environmental Engineering, ETH Zurich; S. Hellweg, ETH Zurich / Institute of Environmental Engineering; T. Koellner, University of Bayreuth. Large areas of our land surface are strongly modified by humans for agriculture, industry and settlements. This has major impacts on ecosystems and biodiversity. However, no consensus exists on how these impacts could be quantified within Life Cycle Assessment (LCA). Here, we present a globally applicable approach how to assess the impacts of land use on biodiversity on regional and global scales. We model the potential regional species loss due to all accumulated land use activities within all global WWF ecoregions and use this as a basis for calculating characterization factors for life cycle impacts assessments. We distinguish between potentially reversible impacts (i.e. regionally extinct, non-endemic species) to calculate land occupation and transformation impacts and irreversible impacts (i.e. global extinction of endemic species) to calculate permanent impacts. Species loss is calculated based on the matrix calibrated species-area relationship. We calculate characterization factors based on multiple species groups, such as mammals, birds, amphibians, reptiles and plants. Finally, we analyze and quantify the uncertainties of the characterization factors applying Monte Carlo simulations. Characterization factors for four land use types were calculated: agricultural land, pasture, forestry and urban land. In the ecoregions which have been largely converted to human use, and thus only little undisturbed habitat remains, the impacts are highest. For all three impact types, the results range several orders of magnitude across ecoregions. In general, the results show high uncertainties. Application and testing of the characterization factors in case studies will show their robustness and applicability. Currently, data availability limits a more refined land use classification. However, in future better data availability might allow including a refined classification and which might reduce some of the uncertainty. 213 Assessing impacts on biodiversity from land use in LCA with global datasets, exemplified with kiwifruit production in New Zealand O. Michelsen, NTNU / Industrial Ecology Programme; C.R. Coelho, Auckland Council / Geospatial Services Delivery. The question on how to include impacts on biodiversity from land use and land use changes (LULUC) in LCA has been discussed for decades and at present there is still no widely accepted method for assessing changes in quality in terms of biodiversity, partly as a consequence of lack of data. Here a method making use of globally available datasets is presented. The idea

SETAC Europe 23rd Annual Meeting

is to combine data on ecosystem vulnerability and ecosystem scarcity on ecoregion level, dividing the world in 825 ecoregions. Ecosystem scarcity is a measure on the rareness of an ecosystem and ecosystem vulnerability is a measure on the present condition of the ecosystem type on a global scale. These data are combined with data on naturalness which indicates how much of the biodiversity is still present in an area given the present land use activities. Impacts from land use is assessed as changes in quality × ha × year, so the value 1 indicates a complete removal of biodiversity from one ha for one year. The method is tested on a case study on kiwifruit production on 10 different localities in New Zealand. The sensitivity of a number of assumptions is also tested. The results show a range in the impact from 0.054 to 0.141 for the production of 1,000 tray equivalents (approximately 3600 kg green kiwifruit). The results are sensitive to some of the assumptions made. Since ecosystem scarcity and ecosystem vulnerability are given values in the range from 0 to 1, they must be normalized to a reference value. The choice of a global or national reference value highly influences the results. National data on percentage of natural vegetation left is also tested as an alternative to global data on ecosystem vulnerability. The range of the scores was not changed, but the rank of some of the areas changed. This shows that present data on conservation status on ecoregion level may be too coarse to give a realistic picture of the actual situation on site, but at the same time the method with using local data incorporated using GIS modelling shows how this can be overcome. As a conclusion the results obtained using globally available data show how this method is applicable to incorporate impacts on biodiversity from LULUC in LCA, capturing much of the essence in biodiversity valuation through vulnerability and scarcity. The only information needed to apply the method is geographic location to identify the relevant ecoregion and productivity to identify land requirement. 214 Inhaled toluene from printed matter indoors: Complementary results from Life Cycle Assessment and Risk Assessment T. Walser, R. Juraske, Institute of Environmental Engineering, ETH Zurich; E. Demou, College of Medical, Veterinary and Life Sciences, University of Glasgow; S. Hellweg, ETH Zurich / Institute of Environmental Engineering. Toluene is the most important solvent used in rotogravure printing. The pronounced presence of toluene indoors partly results from the high use frequency of printed matter together with lower air volumes and air exchange rates compared to outdoors. In homes, more people are exposed to toluene emissions from printed matter, albeit toluene concentrations are lower than in printing rooms. If inhaled, toluene can cause dizziness, mild throat, and eye irritation over short term, and impaired cognitive functions over long term. While several studies investigated indoor concentrations of toluene in printing halls, knowledge of the consequences to human health from toluene emissions from printed matter downstream in the life cycle of magazines is still scarce. Therefore, the aim of this study was to quantify negative impacts on human health caused by toluene emissions from rotogravure printed magazines over the entire life cycle, with a particular focus on indoor environments. We used a one-box indoor model for the estimation of toluene concentrations in printing facilities and newspaper stands with large ventilation rates and multiple emission sources as well as in the small residential rooms with low ventilation rates and single emission sources. We validated the model with measurements from literature. Effect factors were taken from the USEtox database. Health impacts resulting from indoor exposure were quantified and then compared to the total human health impacts of a magazine, calculated with Life Cycle Assessment (LCA). These results allowed a relative quantification of human health impacts on a population basis. Then, the absolute individual health risk in the indoor compartments was determined with Risk Assessment (RA). The results showed that one box models provide an accurate estimation of toluene concentrations indoors for the different indoor characteristics. While workers are exposed to significant toluene concentrations in printing facilities (3), 3 the concentrations in homes are rather low (< 0.4 mg/m ). However, inhaled toluene at home becomes the dominant contribution to the total impacts from toluene of printed matter if assessed with LCA, because of the large number of people exposed. With the significant contribution of

55

the toluene exposure indoors to the total life cycle impact of a magazine (44%), we demonstrated that the indoor compartment requires particular attention and that RA and LCA results provide complementary information. 215 Volatile Chlorinated Solvents from Consumer Products: Emissions to Exposures W.J. Doucette, T. Wetzel, Utah State University / Utah Water Research Laboratory; E. Dettenmaier, K. Gorder, Hill Air Force Base. Volatile organic compounds (VOCs), including many with documented short- and long-term adverse health effects, can enter indoor environments through internal (i.e. consumer products, building materials) and external sources (i.e. vapor intrusion from contaminated groundwater). Indoor air concentrations of VOCs vary widely, but concentrations of most VOCs are consistently higher indoors than outdoors. Many consumer products such as adhesives and cleaning solvents contain volatile chlorinated organic compounds (cVOCs) such as trichloroethene (TCE) tetrachloroethene (PCE) that are also the focus of soil and groundwater cleanups in the USA and Europe. In this study, emissions rates of volatile chlorinated solvents from several consumer products (opened and unopened) including adhesives (PCE), cleaning solvents (TCE), household cleaning products (CCl ) 4 and molded plastic objects (1,2-DCA) were measured using a flowthrough laboratory chamber approach. Measured emission rates varied from 0.001 to 20 ug/min depending on the product and CVOC. Subsequent screening-level calculations suggested that the emissions from some these items could lead to indoor concentrations high enough to be of regulatory concern. To verify this, several controlled release experiments were conducted in an actual residence. Consumer products were introduced into a single room of the test residence and indoor air concentrations of CVOCs were measured throughout the house over time using sorbent tube sampling followed by thermal desorption gas chromatography/mass spectrometry (GC/MS) and/or a portable GC/MS. Measured indoor air concentrations of PCE and TCE and peaked within the “source” room and became relatively uniform throughout the residence within several hours after being circulated through the heating, ventilation, and air conditioning (HVAC) system. Both opened and unopened consumer products emitted signifiant amounts of chlorinated solvents that could confound VI investigations. Initial indoor air concentrations of chlorinated solvents were highest in the room where the consumer products were introduced but were quickly and uniformly distributed throughout the three level house when circulated with the HVAC system. The measured emission rates could be used along with a simple box model to estimate exposure concentrations after “mixing” with the HVAC system. 216 The Fate of PBDEs in The Use and Waste Phases Following Changes of PBDE Levels in Products; An SFA ApplicationA. Golnoush, University of Toronto / Geography; A. Buser, ETH Zurich; M.L. Diamond, University of Toronto / Department of Earth Sciences. Brominated flame retardants (BFRs) are synthetic additives which are widely used to reduce the flammability in consumer products. Due to their persistent, bioaccumulative and toxic (PBT) properties, some representative of BFRs such as polybrominated diphenyl ethers (PBDEs) were regulated or phased out in Europe and North America. Despite the global decline in the use of PBDEs, the large inventory of flame retarded products will continue to contribute to emissions of these compounds into indoor and then outdoor environments. Further, "old" PBDEs may return in new consumer products fabricated with recycled materials. In this study, a substance flow analysis (SFA) was used to characterize the flow of PBDEs from product inventory to the surrounding environment. This system is built upon a detailed dynamic inventory of flame retardant products for North America (US and Canada). The results show the changing PBDE inventory over time in the use and waste phases as a result of changes in PBDE levels in consumer products. The outcome of this study can be served as a rational basis for planning systematic action and measures to reduce of the use and human exposure to these substances. 217 Evaluation of the Stockholm Multimedia URban Fate (SMURF)

SETAC Europe 23rd Annual Meeting

model – application to BDE 209 and two phthalate esters A.P. Palm Cousins, Natural Resources and Environmental Effects; T. Holmgren, Umea University / Department of Chemistry; M. Remberger, IVL Swedish Environmental Research Institute / Natural Resources and Environmental Effects. The indoor environment has been proposed to play an important role for release of SVOCs to urban air as a result of emissions from consumer products. Cousins (2012) recently studied the contribution of emissions indoors to outdoor pollution using an indoorinclusive multimedia fate model (SMURF). The aim of this study was to evaluate the SMURF model using realistic emission estimates and real monitoring data and through uncertainty analysis. The SMURF model is a steady state, 8-compartment multimedia fate model parameterized for the municipality of Stockholm. The model incorporates a novel indoor module representing the indoor environment of Stockholm. Evaluation was conducted using emission estimates of DINP, DEHP and BDE 209 based on data from literature and from new emission tests studying the emissions of DINP from PVC materials. Measurements in urban air and ventilation outlets were performed and used in combination with literature data for comparison with model predicted concentrations. Predicted concentrations agreed well with monitoring data for phthalates, but BDE 209 concentrations were a factor of 100 too low, indicating that current emission estimates for BDE 209 do not capture all the important sources. Emission to indoor air, background concentration and ventilation rate were influential inputs for the variance in the target outputs for all substances. The wet removal rate contributed significantly to the variability in dust concentration. The indoor vertical deposition contributed to the variance in all target outputs for BDE 209, reflecting the capacity of BDE 209 to partition to particles, thus emphasizing that particle behaviour is central for the fate of BDE 209. Cousins, A.P. 2012 The effect of the indoor environment on the fate of organic chemicals in the urban landscape. Science of The Total Environment. 438(0): p. 233-241. 218 Prioritizing consumer product categories based on exposure estimates A. Ernstoff, School of Public Health, University of Michigan; P. Fantke, Technical University of Denmark; O. Jolliet, University of Michigan / School of Public Health. Data on exposure to consumer products and the chemicals therein is limited and costly to obtain. To prioritize mitigation efforts on chemicals and consumer product categories of concern, we must systematically predict dominant consumer product exposure pathways for a range of chemicals even when empirical data are absent. In this study we present a risk assessment framework for modeling cumulative exposure to consumer products with the end goal of identifying the contribution of each exposure pathway. We investigate a case study of parabens (methyl-, ethyl-, butyl-, and propyl-) in dermally applied cosmetics. Dermal exposure to cosmetics is often not considered in risk assessment and evaluation of cumulative exposure with respect to population scale in vivo chemical concentrations. To validate our predictitions of consumer exposure we compare modeled results with in vivo US National Health and Nutrition Examination Survey NHANES data, and provide insight to develop a consistent and accurate evaluation of chemical exposures through various consumer product uses. Results suggest our method may aid predicting consumer exposure (specifically the 95%-ile of consumer exposure for women) to chemicals, like parabens, found in cosmetics. Estimated doses for daily use range from 0.001 to 0.04 mg/kg/day, and urine loads, extrapolated from NHANES in vivo data, are at comparible magnitude. Paraben-specific predicted doses correlate best with NHANES 95%-ile for women especially when considering the 2 maximum likely concentration of each respective paraben (R =0.60). Most importantly, predicted doses best correlate with urinary loads (95%-ile for women) when adjusting the dose by the relative occurence 2 in paraben-containing cosmetics (R ?0.97) across ranges of likely concentrations within the products. Using an advanced skin permeability model to estimate systemic absorption through skin does not improve prediction of urinary excretion. As parabens are also present in some foods, oral exposure must also be considered; however, is likely negligible in comparison to dermal exposure to cosmetics. Urinary doses of CP-related chemicals are orders of magnitude higher than exposure to

56

many environmental contaminants, this considered with our findings suggests dermal exposure to cosmetics is an exposure route category which must be prioritized within the overall framework for calculating human exposure to chemicals in consumer products. 219 Comparison of generic modelling of emissions from different material sources T. Holmgren, Umea University / National CBRN Defence Centre; P. Andersson, Umea University; P. Haglund, Department of Chemistry. Building materials and consumer goods contain organic additives that may be released to indoor air and reach the environment. In this study, 2 generic models were developed and compared with experimental data. Three cases was studied: emissions of di-isononyl phthalate (DINP) from vinyl flooring to air, triphenyl phosphate (TPP) from LCD-TVs to air, and leaching of triisobutyl phosphate (TiBP) and tributyl phosphate (TBP) from concrete to water. 3 Emissions to air were studied using a 1m emission chamber. Four vinyl floors and a 32-inch LCD-TV were tested; the TV at various conditions (standby, on, 40°C, and 60°C). Exhaust air was sampled on PUFs for 10 days, test object removed, and chamber heated to recover residues. Concrete leaching tests were made using 9 concrete cylinders, 3 with TIBP, 3 with TBP, and 3 without additive. Water was circulated around cylinders for 16 weeks. Samples (daily/weekly) were extracted with hexane and analysed with GC-MS and GC-FPD. The generic model used to predict emissions used an Abraham solubility model to estimate partition, the Piringer equation to predict diffusion and the Chilton–Colburn analogy to estimate convective mass transfer. Emission factors were calculated using a numeric solution and a simplified model that do not consider material diffusion resistance. When additive concentration is high and additive volatility low the depletion of the material surface is insignificant and diffusion in the material of low importance. This was clearly the case for the LCD, Figure 1, in which TPP are added at high concentration to the front layer of the LCDscreen. It was, however, not the case for the vinyl materials for which the simplified model overestimated the emissions. For such thick materials the material resistance to mass transfer had to be considered and the detailed model used. For concrete, the prediction models did not fit experimental data. In hardened cement, capillary pores and cracks are formed that disturbs the cement structure. Experiments showed an initial 2 2 release of 4200 µg TiBP /m h, which rapidly dropped to 10 µg/m h. The initial release could be explained through a porosity model and the lower long time release through diffusion theory. 220 An overview on emerging contaminants

J. Readman, .

221 The integration of chemistry with toxicological testing in European systems M. Lamoree, Chemistry & Biology. 222 Effects of multistressors and bioinvasions in estuaries and coastal areas: implications for ecological risk assessmentL. Guilhermino, CIIMAR University of Porto. 223 Alternative flame retardants in the coastal and marine environment R. Ebinghaus, Helmholtz-Zentrum Geesthacht. 224 What Defines Nanomaterials? - An examination of the size dependence of physicochemical properties critical to environmental risk assessment C.D. Hassinger, ARCADISUS; K. Sellers, ARCADIS; E.A. Bleeker, RIVM; W.J. Peijnenburg, RIVM / Laboratory for Ecological Risk Assessment; D. Sijm, National Institute for Public Health and the Environment (RIVM). Efforts to assess the risks from nanomaterials implicitly incorporate a definition of what it means to be “nano”. Members of the scientific community and regulators around the world commonly use a definition of 1-100 nanometers (nm) in size. However, uncertainties about this definition persist, with some defining nanoparticles as sizes up to 2000 nm. This presentation will describe, based on a comprehensive literature review, what is known about the size dependence of the physical/chemical properties critical to environmental risk assessment (the known knowns) and current data gaps (known unknowns). As the size of a particle decreases into the

SETAC Europe 23rd Annual Meeting

‘nano’ size range some physicochemical properties may significantly change. Therefore, within the context of the European Commission’s (EC) definition of “nanomaterial” and the data needs under the Registration, Evaluation, Authorisation and Restriction of Chemical substances (REACH) regulation (EC 1907/2006), this research sought to determine which physicochemical properties make a nanomaterial different from a ‘conventional’ material and at what size these properties are changed to ‘nanospecific’ properties. The literature review focused on: surface morphology, crystalline structure, water solubility, reactivity, and photocatalytic reactivity. Relative few papers were identified investigating size-related changes in surface morphology. With respect to crystallinity, which can affect reactivity, several aspects of metals and metal oxides (e.g., lattice contraction/expansion, phase transformation) may vary with particle size although no simple conclusions can be drawn regarding size dependence. Some experimental data show that the solubility of metal and metal oxides and sulphides can increase at nanoscale; additionally, the increased rate of solubility is well known. Reactivity can also increase with decreasing particle size. Maximum catalytic activity of nanoparticles occurs at sizes well below 100 nm (1520 nm), with a sharp change in reactivity below approximately 5 nm in some cases. Photocatalytic reactivity of certain metal oxides and sulphides also increases with decreasing particle size, in some cases changing at a particle size of approximately 5-10 nm. These studies reviewed primarily investigated nanoparticles 225 Mechanistic interpretation of single-particle ICP-MS data G. Cornelis, University of Gothenburg / Department of Chemistry and Molecular Biology; M. Hassellov, Goteborg University / Chemistry Dept. Progress in (Eco)toxicological research on fate and effects of engineered nanoparticles (ENP) is hampered by a lack of sensitive methods that can measure particle number-based size distributions (PSDs) of ENP in environmental matrices at realistically low concentrations. Single particle inductively coupled plasma mass spectrometry (spICP-MS) is a promising technique in this respect as it offers element specific measurements of ENP that generate ion bursts in the plasma, the magnitude of which corresponds to their size and the number of which corresponds to the frequency of particles, thus allowing to calculate PSDs. However, the lower size limit of ENP size measurements is limited mainly by lack of a process-based data analysis tools. The current study proposes an alternative approach, where precise knowledge of dissolved and background signals in ICP-MS is used to deconvolute it from spICP-MS. Well-characterised gold ENP of 10, 30 and 60 nm were analysed using spICP-MS and the data was analysed using known algorithsm and the algorithm proposed in this study. It is proposed that much more accurate PSD can thus be obtained, especially for relatively small ENPs. A mechanisticallty based spICP-Ms signal analysis algorithm was thus developed that was able to deconvolute dissolved signals from ENP signals in a combined dataset, but in the case of small ENP, there was no statistically signifcant difference between dissolved and ENP signals that could be used to distinguish the signals, effectively imposing a size limit to spICP-MS determinations of PSDs and concentrations of ENP in the environment. 226 Fate of nanosilver in lake mesocosms C.D. Metcalfe, L. Furtado, Trent University / Environmental & Resource Studies; J. Fischer; M. Hoque, H. Hintelmann, Trent University; D. Mitrano, J. Ranville, Colorado School of Mines; B. Cheever, M. Xenopoulos, P. Frost, Trent University. Silver nanoparticles (AgNPs) are widely used in textiles and in household products as antimicrobial additives. AgNPs may enter aquatic ecosystems via discharges of municipal and industrial wastewater. AgNPs undergo transformations in aquatic environments that can alter the toxicity to aquatic organisms, In a study conducted as part of the Lake Ecosystem Nanosilver (LENS) project, we studied the transformations and fate of AgNPs in 12 mesocosms (2 m diameter x 2 m deep) installed in a soft water lake in Ontario, Canada. AgNPs with a mean diameter of 50 nm and capped with either PVP or citrate were purchased as a suspension in deionized water from NanoComposix, CA, USA and this material was added to mesocosms either as incremental ("drip") additions every second day for a 70 day period, or as a single

57

("plug") addition followed by a monitoring period of 50 days. Total Ag in the drip mesocosms increased over time and reached about 80% of the target nominal concentrations of 4, 16 and 80 ppb by the end of the experiment. About 40% of Ag was retained on a 0.22 um filter, indicating some hetero- and homo-aggregation. There was no sighificant difference between the fate of the PVP and citrate capped AgNP in the drip mesocosms. In the plug experiment with PVP capped AgNP at a nominal concentration of 80 ppb, the half life for total Ag in the two mesosocms was apperoximately 20 days. Analysis of the size of the AgNPs in the plug mesocosms using single particle ICP-MS indicated that the mean particle size declined over time from approximately 50 nm to approximately 35 nm. ICP-MS analysis of the filtrate after ultrafiltration indicated that the cocnentrations of dissolved silver (dAg) in the mesocosms were low, but the results were highly variable. Estimates of concentrations of dAg using passive sampling with diffusive gradient in thin film (DGT) devices deployed in the 80 ppb drip treatments indicated that dAg concentrations were in the range of 14 ppb. Overall, this study showed that the AgNPs were relatively stable in suspension in the mesocosms, which could be due to the low ionic strength and the moderate levels of dissolved organic material present in the lake water. 227 Release of TiO2 Nanoparticles from Sunscreens into Surface Waters A. Gondikas, University of Vienna / Environmental Geosciences; F. Von der Kammer, Vienna University / Department for Environmental Geosciences; R. Reed, Colorado School of Mines / Department of Chemistry and Geochemistry; J.F. Ranville, Colorado School of Mines / Chemistry and Geochemistry; T. Hofmann, University of Vienna / Environmental Geosciences. The natural cycle of information flow between risk assessment studies (RAS) and research and development (R&D), which is vital for the production of ‘environmentally responsible’ engineered nanomaterials (ENM) is currently lacking appropriate momentum. One of the major reasons is the scientific gap between ideal lab-scale studies and realistic environmental conditions (e.g. complex media, transformations of ENM, realistic concentrations, time-dependent exposure, and relation to background materials) that introduce uncertainties in RAS. In order to reduce these uncertainties, it is necessary to develop methods that detect and characterize ENM in natural environments. In this work, we attempted to address the issues of detecting ENM release (TiO from 2 sunscreens) in a complex matrix (lake water), at environmentally relevant concentrations, distinguish them from natural background, while taking into account the time-dependent nature of the release. By collecting samples of suspended matter and sediments we quantified TiO concentrations in the Alte Donau Lake (Vienna, Austria) that is 2 heavily used for bathing activities during the summer season. The Alte Donau Lake can serve as a ‘low background’ water body with high bathing activity that is prone to release of TiO ENM from sunscreens. 2 Indeed, an increase of Ti concentrations was monitored during the bathing season in the sediments, SPM, and the total water samples. However, distinguishing between engineered and naturally occurring TiO nanoparticles is an additional major challenge. In an attempt to 2 address this issue, we compared titanium to aluminium concentrations. The ratio of Ti/Al is heavily weighted towards the denominator for natural soils, while the opposite is true for the majority of sunscreens products, where the mineral particle consists of a TiO core and only a 2 thin layer of aluminium oxides on the surface. We measured total Ti and Al content in the sediment fractions and SPM collected at several time points spanning from June to October of 2012. The Ti/Al ratio appeared to increase during the bathing season of July and August, following changes of temperature during this time. Additionally, in SEM investigations of the filtered SPM some TiO particles could be found 2 with EDX signatures similar to particles from sunscreen products. 228 Filling the gaps in nanomaterial exposure assessment heteroaggregation and fate modelling in surface watersA. Praetorius, ETH Zurich / Institute for Chemical and Bioengineering; M. Scheringer, ETH Zuerich / Institute for Chemical and Bioengineering; J. Labille, J. Bottero, CEREGE, Aix-Marseille Université, CNRS; K.

SETAC Europe 23rd Annual Meeting

Hungerbuehler, ETH Zurich / Institute for Chemical and Bioengineering. Environmental fate models serve as an important tool to obtain predicted environmental concentration (PEC) values and enable a proactive risk assessment of engineered nanomaterials (ENMs). For reliable model predictions, a thorough understanding of the dominant processes affecting the fate of ENMs is of great importance. At currently expected environmental concentrations of ENMs, heteroaggregation of ENMs with naturally occurring suspended particulate matter (SPM) is probably much more important than homoaggregation as a factor determining the environmental fate of ENMs. However, to date, very little research has been done to elucidate and quantify the heteroaggregation of ENMs with SPM under realistic conditions. We here present a new method for measuring the heteroaggregation of TiO nanoparticles (NPs) with SPM and for the 2 determination of heteroaggregation attachment efficiencies (? ). het-agg The heteroaggregation kinetics of SiO particles (0.5 mm) and TiO 2 2 NPs (5-20 nm) were followed by laser diffraction and ? values het-agg for determined for different environmentally relevant solution conditions by fitting the aggregation curves with a Smoluchowski-based model. Our results show that ? increases with increasing salt het-agg concentration (NaCl and CaCl ), whereas the presence of humic acid 2 stabilizes the particles against heteroaggregation. The obtained ?het-agg values were implemented in an environmental fate model predicting the transformation and transport behaviour of TiO NPs in a river system. 2 The modelling results show that most TiO NPs accumulate in the 2 sediment compartment, due to heteroaggregation with SPM and subsequent sedimentation, but a small fraction of NPs remains in the water compartment. The combination of measured ? values and het-agg an environmental fate model enables predictions of the environmental fate of TiO NPs and thereby fills important gaps for a realistic 2 exposure assessment of ENMs. 229 Why testing stones in mesocosms? C. Schaefers, FraunhoferInstitut / Department of Ecotoxicology; B. Knopf, Fraunhofer IME - Institute for Molecular Biology and Applied Ecology; H. Ruedel, Fraunhofer IME Institute for Molecular Biology and Applied Ecology / Molecular Biology & Applied Ecology; K. Ebke, MESOCOSM GmbH Institut für Gewässerschutz; U. Hommen, Fraunhofer IME. Within a large research activity on the ecological risk assessment of slag material from copper production used as armour stones, the potential leaching of metals and its effects on the aquatic community in outdoor mesocosms was monitored over one year. The study was conducted in 25 stainless steel enclosures of 2 m³ volume, installed in a large artificial pond with an established aquatic community. Five enclosures served as controls including 25 g/L basanite crushed stone fines and 100 g/L basanite armour stones as natural reference material. In eight enclosures 3.25, 6.25, 12.5 and 25 g/L iron silicate crushed fine stones (two replicates each) and in 12 enclosures 12.5, 25, 50 and 100 g/L iron silicate armour stones ( three replicates) were introduced. In each enclosure the appropriate amount of basanite stone fines and stones were add to achieve the same amount stone fines and stones in all enclosures including the controls. During the study the concentrations of metals in the water and the sediment was measured while metals in biota were measured at the end of the study after one year of exposure. For the effect assessment the development of the populations of algae, macrophytes, zooplankton and macroinvertebrates was monitored. Cu, Ni, Zn, Mn and Fe concentration in the water increased, related to the amount of introduced iron silicate, The maximum Cu concentrations found at the highest stone treatment level were 14 µg/L in March 2010 while the highest concentration in the crushed fines enclosures were found to be close to 13 µg/L 7 days after introduction of the test items. After one year of exposure, Cu concentration in the water decreased down to 3 and 5 µg/L in in highest crushed fines and stone treatment level, respectively. No dose related increase of metals could be found in the sediment. In biota Cu concentration increased up to a factor of 5 compared to the controls while other metals showed usually no or a smaller increase in biota. No indication of biomagnification in the food chain was found. Up to 12.5 g crushed stone fines / L or 50 g stones / L no long-term or pronounced effects on the bio-coenosis were observed.

58

At 25 g crushed stone fines / L or 100 g stones / L, effects on algae, macrophytes and insects over more than 8 weeks or at the end of the study could not be excluded. Thus, 12.5 g crushed stone fines / L (1:80) or 50 g armour stones / L (1:20) are considered ecologically acceptable amounts of iron silicate slags in this study. 230 Linking structural, functional and behavioural endpoints to detect effects on Gammarus roeseli after repeated insecticide pulses S. Mohr, Umweltbundesamt / IV 2.5; R. Berghahn, Federal Environmental Agency / Field Station Marienfelde; R. Boettger, M. Feibicke, Umweltbundesamt. Gammarids play a key role in freshwater lotic ecosystems. A population decrease or breakdown e.g. due pesticides may therefore have severe consequences for stream ecosystems. Streams may experience repeated pesticide pulses, which may be of short duration but have high pesticide concentrations. This may not lead to death of organisms but to reduced health, which is hard to detect with routine sampling methods. A stream mesocosm study was conducted in order to investigate the effects of short and repeated pulses of the neurotoxic insecticide imidacloprid on macroinvertebrate communities with special focus on behavioral and functional endpoints of the amphipod Gammarus roeseli. Prior the study, laboratory toxicity tests were run with this species to determine the optimal sublethal pulse concentrations for the mesocosm study. In the mesocosm experiment, weekly 12 h pulses of 12 µg/L of the insecticide imidacloprid were set 3 times in 4 stream mesocosms in 2 series, one in spring and one in summer. Another 4 mesocosms served as controls. Prior to each pulse series, the mesocosms were equipped with sediment and water, and stocked with macroinvertebrates from an uncontaminated reference stream using straw bags as attraction devices. Within the investigation period of 11 weeks, gammarids were sampled regularly for the following endpoints: drift, activity, live counts, abundance and shredder activity of G. roeseli. In the toxicity tests, the EC50 (96 h) for G. roeseli varied considerable from 1.9 to 129.5 µg/L imidacloprid depending on the different experimental conditions. Besides effects of food and age, the origin of the test species taken from a field population in autumn or spring had strong effects on the sensitivity of G. roeseli. In the stream mesocosm study, abundance data and shredding activity did not indicate negative pulse effects, but strong effects were found for the endpoints drift and activity. Under more natural conditions including fish predation and higher flow velocity inactive or impaired gammarids may drift out of the system or be eaten by fish. A combination of different sublethal endpoints such as drift effects may provide a realistic view of what may actually happen in the field and help to predict detrimental effects for ecosystems. 231 Extensive toxicity testing with the neonicotenoid imidacloprid shows a high chronic toxicity to mayfly larvae I. Roessink, Alterra; L.B. Merga, Wageningen University; H. Zweers, Alterra, Wageningen UR; P. van den Brink, AlterraWageningen UR / (a) Aquatic Ecology and Water Quality Management Group; (b) Alterra. Although solely used in the terrestrial environment, the neonicotenoid insecticide imidacloprid also finds its way to surface waters, for instance via spray drift, leaching and run off. Most of the research to its effects on aquatic non-target invertebrates comprises acute tests and chronic data is scares. A knowledge gap, since chronic exposure may be expected given the persistence of imidacloprid. The current study was designed to generate both acute and chronic toxicity data for a range of freshwater invertebrates. Both acute 96h and chronic 28d toxicity tests have been performed using glass jars placed in a water bath. Tested species comprised Cloeon dipterum, Caenis horaria, Chaoborus obscuripes, Plea minuttissima, Sialis spp., Asellus aquaticus, Gammarus pulex, Trichoptera spp., Notonecta spp., and Corixidae spp (the latter three only tested in acute tests). Both immobilisation and mortality parameters were scored and EC and LC parameters were calculated. These 50 50 values were further used to construct Species Sensitivity Distributions (SSD). The acute tests showed that mayflies (C. dipterum and C. horaria) and caddisflies (Trichoptera spp.) are the most sensitive species tested. While Notonecta spp. and Sialis spp. did not show any treatmentrelated mortality in the acute experiments at all. Chronically, both

SETAC Europe 23rd Annual Meeting

mayfly species (C. dipterum and C. horaria) were the showed a much higher sensitivity compared to the other species tested. Acute to chronic ratios were highest for C. obscuripes and the two mayfly species, and overall ranged from 4 to 85, suggesting that it is not recommended to use acute data to assess the effects of long-term exposure to imidacloprid. 232 Plant Species Sensitivity Distributions for ozone exposure T.v. van Goethem, Radboud University / Environmental Sciences; L.B. Azevedo, Radboud University Nijmegen / Department of Environmental Science; R. Van Zelm, Radboud University; F. Hayes, Environment Centre Wales / Centre for Ecology and Hydrology; M. Ashmore, University of York / Stockholm Environment Institute; M.A. Huijbregts, Department of Environmental Science. Current critical levels commonly used in environmental policy assessment for ozone exposure to natural vegetation are based on relationships between ozone concentrations and effects such as yield loss and biomass reduction. These levels are based on sensitive but ecological relevant species. These species, and corresponding critical levels, are used as indicators for species groups to determine risk for larger ecological units. For seminatural plant communities, with the large range of species present, an approach based on a single indicator species ignores the wide range of sensitivity across all the component species. To date, an approach which gives the affected fraction of a species assemblage due to ozone exposure is lacking in risk assessment for semi-natural vegetation. Therefore, Species Sensitivity Distributions (SSD), representing a cumulative stressor-response distribution for species within a plant community, were derived for damage to natural vegetation caused by ozone exposure. SSDs were constructed for three species groups, i.e. trees, annual grassland and perennial grassland species, using speciesspecific exposure-response data. The SSDs were applied in two ways. First, critical levels were calculated for each species group and compared to current critical levels for ozone exposure. Second, spatially explicit estimates of the potentially affected fraction of plant species in Northwestern Europe were calculated based on ambient ozone concentrations in 2010. We found that the SSD-based critical levels were lower than the current critical levels for ozone exposure, with conventional critical levels for ozone relating to 8-20% affected plant species. The potentially affected fractions based on ambient ozone concentrations in 2010 indicate that in some regions potentially 13% of the perennial grassland species and 30% of annual grassland species have growth reductions of at least 10%. Furthermore, the mean values of the SSDs were significantly lower for annual than for perennial grassland species. This indicates that annual grassland species, as a species assemblage, are more sensitive to ozone than perennial grassland species. Our study shows that the SSD concept can be successfully applied to derive critical ozone levels as well as to estimate the affected fraction of a community along specific ozone gradients. 233 Deriving species sensitivity distributions (SSDs) from species distribution models (SDMs) A.M. Schipper; L. Posthuma, RIVM / Lab. for Ecological Risk Assessment; D. De Zwart, RIVM / Centre for Sustainability, Environment and Health; M.A. Huijbregts, Department of Environmental Science. Species sensitivity distributions (SSDs) are commonly used to describe the inter-species variation in sensitivity to an environmental stressor. SSDs are typically established based on laboratory experiments where a few test species are exposed to various levels of a single stressor, usually a toxicant, under otherwise optimal conditions. These tests typically rely on a few easily cultivable species that may be absent in the field site of concern. More ecologically relevant SSDs may be obtained by establishing field-based SSDs (fSSDs) from monitoring data, thus covering the actual species pool. However, unless the stressor of concern is the main, dominant stressor that influences species occurrence or abundance, such field-based response curves may be flawed by the confounding influences of other stressors. In the present study we aim at deriving f-SSDs from singlespecies distribution models (SDMs) that predict the occurrence of species under multiple environmental factors. We used a monitoring data set comprising 103 benthic macroinvertebrate taxa and 12 abiotic

59

variables collected at 349 sites in tidally influenced fresh water bodies in the west of The Netherlands. For each taxon in the database we constructed a distribution model relating the probability of occurrence to the 12 abiotic variables. To translate the predicted probabilities of occurrence into presence or absence, we identified taxon-specific thresholds of occurrence. Subsequently, we used the SDMs and thresholds to predict the presence-absence of each taxon as function of each single abiotic variable. This was done by varying the level of the abiotic variable of concern across its range while keeping the other abiotic variables at optimal levels, i.e., levels maximizing the predicted probability of occurrence of the taxon of concern. Finally, we aggregated the presence-absence for all taxa across the range of the abiotic variable, thus establishing an f-SSD showing the potentially not occurring fraction (PNOF) of taxa in relation to each abiotic variable of concern. Our preliminary results indicate the feasibility of establishing SSDs by applying multiple regression techniques on field data, thus yielding f-SSDs for multiple stressors based on actual species pools. 234 Effects of herbicides and fertilizer on the plant community of field margins J. Schmitz, Inst for Environmental Sciences / Institute for Environmental Sciences; C. Bruehl, University of Landau Institute for Environmental Sciences. The risk assessment of herbicides aims to protect non-target plants in off-crop habitats such as field margins from adverse effects of pesticides. Therefore, short-term phytotoxicity tests with single and annual plant species are performed in young development stages in greenhouse experiments. Although testing guidelines allow using non-crop species, the standard risk assessment uses crop plants for phytotoxicity testing even though non-crop species (annual and perennial species) are to be protected in field margins. Due to the short test duration of 21-28 days effects on reproduction cannot be detected with these test methods. Furthermore, the phytotoxicity tests are performed under standardized greenhouse conditions that differ markedly from field conditions (e.g. intra- and interspecific competition for resources). The present study was undertaken to investigate the effects of the misplacement of pesticides and fertilizer on the plant community of field margins. The study was a perennial field study, which started in 2010 and ended in 2012. The applications of the treatments and their application sequences mimicked the field management of winter wheat fields with their recommended agrochemical products and application rates. The applied fertiliser and pesticide concentrations were consistent with their inputs (drift+overspray) in the first meter of a field margin directly adjacent to a field under Good Agricultural Practices. To detect the effects of the applications, vegetation assessments were performed in May and June 2010 to 2012. Additionally, the seed production of selected species (Ranunculus acris, Lathyrus pratensis, Vicia sepium, and Rumex acetosa) was assessed in 2012. The plant density of the four species was significantly affected by the fertilizer and herbicide applications. The plant density of R. acris and L. pratensis was affected stronger in the fertilizer treatments than the herbicide treatment and the plant density of R. acetosa and V. sepium was similar affected through the fertilizer and herbicide treatment. However, the treatment combination of fertilzer and herbicide resulted in additive effects. In addition, the herbicide treatment reduced the seed production of R. acris, L. pratensis and V. sepium. The experiment showed that the effects in the field are complex, interaction effects between agrochemicals (e.g. herbicides and fertilizer) can occur and are certainly important for the sensitivity of species to agrochemicals. 235 Does soil type influence the toxicity of silver to plants? K. Langdon, CSIRO Land and Water; M. McLaughlin, CSIROUniversity of Adelaide; J. Kirby, CSIRO; G. Merrington, Environment Agency. Due to the extensive use of silver (Ag) in a wide range of consumer products, there in an increased interest in understanding the potential risks associated with this element in various environmental + compartments. Although Ag in its ionic form (Ag ) has been shown to exhibit toxicity to aquatic organisms at low solution concentrations, it is likely that in the terrestrial environment interactions with a soil matrix will result in increased sorption and binding of Ag, making it less

SETAC Europe 23rd Annual Meeting

bioavailable to soil-dwelling organisms. Research into on ecotoxicity of other metals has shown that soil properties can modify metal toxicity, with soil pH, organic carbon (OC), clay content and cation exchange capacity (CEC) shown to have the greatest influence in mitigating toxicity. The aim of this study was to examine the toxicity of Ag to plants in a range of soils with varying properties to determine if differing soil properties influence the toxicity of Ag in soils. A standard five day root elongation test was used, with barley as a test organism. Increasing concentrations of Ag significantly inhibited barley root elongation to varying degrees in the soils. EC50 values were found to vary over three orders of magnitude from approximately 80 to 1000 mg Ag/kg. Toxicity thresholds were found to be (positively) related to the soil OC content and CEC. These results indicate the need to consider soil properties when determining the predicted no effect concentrations for Ag in the terrestrial environment. 236 Plant toxicity in Pb-salt spiked soils: unraveling effects of acidification, salt stress and ageing reactions E.E. Smolders, Katholieke Universiteit Leuven; K. Cheyns, KU Leuven. This study was set-up to quantify the differences in lead (Pb) toxicity between 2+ Pb spiked soils and field contaminated soils and to identify factors involved Yield of wheat seedling was unaffected by total Pb (up to 14,000 mg Pb/kg) in field contaminated soils. In contrast, yield declined 2+ by factors 2 or more in corresponding soils freshly spiked with Pb 2+ salts at similar total soil Pb concentration. Toxicity of Pb -salts in soil (21-8700 mg Pb/kg) to tomato and barley plants was stepwise reduced by leaching with pH correction and by five years ageing after spiking. Shoot-phosphorus (P) concentration strikingly explained the differences in toxicity among soils and treatments. Bioassays and soil analysis show that Pb toxicity in freshly spiked, unleached soils is primarily confounded by salinity stress and acidification. In addition, ageing processes do reduce Pb toxicity, likely due to reduced Pb induced P deficiency for plants. Soil leaching and pH correction after spiking are practical methods that increase the field relevance of test soils in laboratory assays. 237 Single Effects of UV radiation to Folsomia candida D.N. Nunes Cardoso, CESAM University of Aveiro. At the same time that organisms are exposed to chemical contaminants, they can also experience a large range of environmental fluctuations such as drought and flood conditions, temperature and pH or even UV radiation increments. In this study we have carried out the exposure of the collembolan Folsomia candida to different doses of UV radiation in order to attain for differences in exposure type effects. F. candida was highly sensitive to UV radiation, affecting its reproduction and survival. Two situations were tested when the organisms were exposed to UV radiation: direct (Plaster) and indirect (Lufa 2.2) exposure to the radiation. Additionaly, we tested the influence of UV radiation to soil organisms in compacted soil. When radiation was applied directly, at the highest UV doses all exposed collembolans died. Curiously, the reproduction of collembolans was higher at the highest doses when they received indirect radiation. Also, with our results, we proved the negative importance of soil compaction on the soil fauna regarding the protection it can also provide to this kind of organisms, when evaluating UV radiation effects. Furthermore, UV radiation influenced the capacity of egg’s hatching when their exposed to direct UV radiation. At high doses of UV, the number of eggs hatched was much lower than in control situations. With those bioassays, we proved that UV radiation influence the survival and reproduction of the organisms and the protection that soil give to organisms when they are exposed to UV radiation. 238 Long-term sorption of Mn(II) by viable and dead Shewanella putrefaciens: speciation & bioavailability relevanceN. Chubar, Glasgow Caledonian University / School of Engineering and Built Environment; C. Avramut, Leiden University Medical Center / Department of Molecular Cell Biology. The interfacial processes at the microbial cell walls (due to variety of functional groups and metabolic activity of living cells) influence the cycling, speciation and

60

bioavailability of xenobiotics and nutrients. Additionally, these processes influence the composition of surface and groundwater, acid mine drainage and formation of soils and minerals. Taken together, these processes are of great importance for the development of bioremediation strategies. Although many researchers have evaluated metal sorption in viable and non-viable bacteria, forming conclusions about their similar adsorptive behaviour for short periods of contact time this is the first work where the difference between the adsorptive properties of viable and dead cells is clearly shown, due to the application of a longer contact time. Manganese is one of the most important redox chemical elements in the environment; it is a nutrient in smaller amounts and a contaminant at high concentrations, particularly at mining. The data reported here are the first insight into the ability of viable cells to remove metal ions over a one-month period. The long2+ term ability of the bacteria to remove Mn over one month and an accelerated removal of this metal after 4 days is due to the formation of inorganic precipitates of Mn(II) which are a function of the contact time, metal loading, temperature and a total number of the microbial cells. This is the first study demonstrating the ability of the initially viable bacteria S. putrefaciens to synthesise manganese phosphate, and also showing how the other experimental condition can direct the living cells to form the other precipitate(s), such as manganese(II) carbonate. Dissolved organic substances released by viable S. putrefaciens did not 2+ complex Mn , at least over 14 days. However, in the presence of Mn(II), an increased concentration of the released dissolved intracellular inorganic phosphate and the production of dissolved organic substances containing phosphate groups are possible reasons of the formation of the new inorganic Mn-containing phase, which is Mn(II) phosphate. 239 The effect of soil pH on cadmium toxicokinetics in the springtail Folsomia candida M.M. Ardestani, Ecological Science; C. van Gestel, Vrije Universiteit Amsterdam / Ecological Science. Heavy metals should be taken up by the organism to show toxic effects. Soil organisms are exposed to metals mainly through pore water. Therefore, availability of metals in the porewater is important factor which determines in what extent metals could reach their targets in the organisms’ body. Bioavailability is a dynamic process, which is affected by physical-chemical soil properties including soil pH. Taking into account the effect of time might better explain the effect of soil properties on metal bioavailability in soil. We therefore performed a toxicokinetics study to assess the effect of soil pH on the uptake and elimination kinetics of cadmium in the springtail Folsomia candida. Animals were exposed to three cadmium concentrations (0, 5, and 20 µg -1 Cd g dry soil) in LUFA 2.2 soil, using three pH levels (4.5, 5.5, and 6.5). Cadmium concentrations in the animals were measured by sampling the animals at day 0, 0.25, 1, 2, 4, 7, 10, 14 and 21 for the uptake phase and at the same sampling time points for the elimination phase after the animals were transferred to clean LUFA 2.2 soil. A firstorder one-compartment model was fitted to the internal cadmium concentration data for each pH level and exposure concentration. At the beginning and the end of uptake and elimination phases, total and water or 0.01 M CaCl extractable concentrations of cadmium were measured 2 in the soil as well as concentrations in the pore water. Extractable and porewater concentrations decreased with increasing pH. The results showed a fast uptake of cadmium in the animals, with uptake rate -1 constants ranging between 0.33 to 1.00 g g -1 d based on soil animal total soil concentrations followed by a fast elimination with elimination -1 rate constants between 0.15 to 0.45 d . Steady state was reached after approximately in three weeks of exposure. Cadmium uptake rate constants based on total soil concentrations were independent of soil pH and so were steady state concentrations in the earthworms. Uptake rate constants related to extractable and porewater concentrations, however, increased with increasing pH, which could be explained from the absence of proton competition at high pH. These findings support the biotic ligand model approach. 240 Suitability of flow cytometric analysis of primary cultures of coelomocytes of Eisenia fetida for toxicity testing A. Irizar Loibide, University of the Basque country UPVEHU / Department of Zoology

SETAC Europe 23rd Annual Meeting

and Animal Cell Biology; C. Rivas, University of the Basque Country; F. Goni, J. Etxebarria, GAIKER Tecnological Centre. IK4 Research Alliance.; I. Marigomez, University of the Basque Country; M. Soto, University of Basque Country / Zoology and Animal Cell Dynamic. Soil pollution is a very common problem and the assessment of its effects on soil ecosystems has become a priority issue. Recently, primary cultures of earthworm immune cells, coelomocytes, have been used for in vitro toxicological research following the 3R principle, and the suitability of coelomocyte primary cultures for toxicity assessment was demostrated. The autofluorescence of riboflavin molecules present in eleocytes, a subpopulation of coelomocytes, is a very useful characteristic for toxicity evaluation, since it has been seen that it decreases after exposure of earthworms to chemical stress (Plytycz et al., 2009). Several works have investigated the changes of autofluorescence levels after exposure of earthworms to contaminated soils with flow cytometry, since this technique allows the rapid and precise quantification of autofluorescent cells and can also give information about the chanhes in the fluorescense signal of the eleocytes. In the present work, coelomocyte primary cultures have been exposed in two different manners to a model metal, cadmium, in order to study its effects on coelomocytes and on their different subpopulations (the autofluorescent eleocytes and the nonautofluorescent amoebocytes) and to prove the suitability of the flow cytometric analysis of in vitro exposed coelomocyte primary cultures for toxicity assessment. The results show that primary cultures were sensitive to Cd exposure although the stabilization of cells in culture medium for 24h caused a decrease in their sensitivity. Among coelomocyte subpopulations, Cd caused toxicity in both studied cell populations, but viability of amoebocytes was affected at higher Cd concentrations than eleocytes, indicationg different sensitivity levels. The comparison of those in vitro exposures and the pararel in vivo exposures showed an increased toxicity in the first ones. It can be concluded that primary cultures of coelomocytes exposed in vitro to contaminantsare suitable for flow cytometric analysis and toxicity assessment since. both coelomocytes subpopulations were sensitive to the presence of Cd, and their mortality was dose-dependent. However, the stabilization period before the exposure could cause a loss of sensitivity. The higher sensitivity of in vitro exposures made them suitable for toxicological studies. 241 Activated Sludge, Respiration Inhibition Tests; Experiences with the Updated OECD 209 Test Guideline C. Mead, Harlan Laboratories Ltd / Ecotoxicology dept; P. Roulstone, Harlan Laboratories Ltd. The OECD 209 Test Guideline describes a method to determine the effects of a substance on the respiration rate of activated sludge micro-organisms. The results obtained can be used to assess the effect that the test substance will have on a sewage treatment facility and / or an indicator of potential inhibitory effects in a biodegradation test. In July 2010 an updated Test Guideline was published which introduced replication of the test vessels and the possibility of determining inhibition of both carbon and ammonium oxidation rates. As a result of these changes to the Test Guideline it was increasingly difficult to conduct a ‘full’ study during a standard working day when using the suggested sequential preparation of test vessels. Here we describe some of the alternatives available for conducting ‘full’ studies and present a summary of the studies that have been conducted over the last 12 months to illustrate the applicability of these alternatives. 242 Comparing concentrations of micropollutants and toxicological effect thresholds in wastewater-impacted riversR. Bloch, Helmholtz Centre for Environmental Research GmbH / UFZ; C. Hug, Helmholtz Centre for Environmental Research / Effect-Directed Analysis; P.C. Von der Ohe, UFZ Helmholtz Centre for Environmental Research; M. Krauss, Helmholt Centre for Environmental Research / Effect-Directed Analysis; T. Schulze, W. Brack, Helmholtz Centre for Environmental Research UFZ / Effect-Directed Analysis. Wastewater discharges result in ubiquitous contamination of surface waters by a large number of polar organic chemicals, which stem from households and industrial uses. These micropollutants are only incompletely

61

removed by conventional wastewater treatment plants (WTPs). Although occurring in rather low concentrations, micropollutants might cause adverse effects on aquatic organisms. A target screening for 120 compounds was done in 31 surface water samples from the catchments of the rivers Saale and Mulde in Germany. Compounds were chosen from different compound classes, such as pharmaceuticals, industrial chemicals, pesticides, biocides, and personal care products based on likely or known occurrence in the environment. A total of 82 target compounds were found in at least one sample. Among them, 34 compounds were detected at more than 50% of the sites. For the evaluation of the toxicological impact on the aquatic environment, predicted no-effect concentrations (PNECs) were derived for all compounds, using acute and chronic toxicity data. An exceedance of the PNEC was registered for 7% of positive detections. This illustrates that environmental concentrations of organic micropollutants, discharged by WTPs, can easily reach and exceed toxicological effect thresholds. Pharmaceuticals and pesticides are compound groups that are likely to cause effects on aquatic organisms and therefore monitoring programs have to be extended to address all relevant substances. 243 Effluent Toxicity Testing in Ireland (1983-2013) r. hernan, Shannon Aquatic Toxicity Laboratory; K. O Rourke, Enterprise Ireland / Shannon Aquatic Toxicty Laboratory. Effluent Toxicity Testing in 1 1 1 Ireland (1983-2013) Robert Hernan , Kathleen O’Rourke Shannon Aquatic Toxicity Laboratory, Enterprise Ireland, Shannon Town Centre, Co.Clare, Ireland \nE-mail contact: [email protected] InIrelandthe EPA, as part of their IPPC programme, regulate industrial effluent discharges. In the case of complex (nonsimple) effluents, the EPA requires a combination of toxicity testing and the traditional substance by substance analysis. Approximately 25% of all licenses have a toxicity clause. Toxicity testing is carried out on a final effluent sample to either marine or freshwater fish/crustacean/algae/bacteria. Discharges to WWTP also require a respirometry test. The EPA has assigned a toxicity emission limit value of no more than 10 toxic units (TU) i.e. E(L) < 10% vol./vol. to 50 most effluent discharges. In 2011, 90% of the samples tested were within the 10 TU limit. Between 1991 and 2011, greater than 70% effluents were within the 10 TU limit however between 1985 and 1990, greater than 50% effluents were over the 10 TU limit. In 2011, S. costatum and T. battagliai were the most sensitive test species with fish the least sensitive. In general we have found that the fish tests, rainbow trout and turbot, are the least sensitive test species. Other species including Lemna minor and Crangon crangon have also exhibited low sensitivity to effluents. The marine alga, Skeletonema costatum, is the most sensitive. In 2011, 35% of effluent samples tested discharged to freshwaters, 34% to WWTPs and 31% to coastal waters. In recent years increasing numbers of discharges are going to tertiary WWTPs. Domestic wastewaters are licenced by the EPA through Wastewater Authorisation (WWDA’S) since 2008. Approximately 20% of large wastewater treatment plants (population equivalent >200) require toxicity testing. Initial testing shows these are mostly non-toxic. Keywords: toxicity testing, wastewater effluents, toxic units, species sensitivity 244 In situ reprotoxic effects of urban and hospital effluent on the mudsnail P. antipodarum: role of biofilm and water columnM. Gust, Irstea; A. Berlioz-Barbier, CNRS; A. Bouchez, INRA; J. Labanowski, C. CREN, CNRS; J. GARRIC, Irstea. Municipal effluents are well-known contributors of pollution in aquatic ecosystems. Bellecombe (Sipibel, Haute-Savoie, France) Waste-Water Treatment Plant (WWTP) treats urban effluents, but also the effluent of the newly established hospital. Treated effluents are discharged directly in the Arve River. Previous reports (unpublished results) showed that chemicals (pharmaceuticals and endocrine disrupters) were released in the river by the effluents, and found directly in water and also in biofilms. Potamopyrgus antipodarum was shown to be a sensitive and adapted organism to assess the toxicity of effluents. It is able to feed naturally with biofilm, in submerged cages. Thus the contamination occurs both via water, and the ingested biofilm. The aims of this study

SETAC Europe 23rd Annual Meeting

were: (i) to assess the reproductive effects of the exposure to the effluent discharge contaminated both by urban and hospitals pollutants, (ii), to assess the contamination in water, biofilm and snails, and (iii) to assess wether a link existed between the contamination of the different compartiments and the observed biological effects (trophic transfert). Adults snails were placed at one upstream and two downstream sites, during 6 weeks in September-October 2012. Mortality was checked weekly, and reproduction in snails (number of embryos in the brood pouch) was assessed after 42 days. Energy status was quantified in snails as well as reproduction-related biomarkers. Chemcatchers were used to measure the chemical contamination of the water. Biofilm was collected at the end of the exposure to assess its diversity, and its load in various chemicals. Measurements of chemicals was also performed in snails at the end of the exposure. The reproduction in upstream snails was in the upper range of physiological values. Nonetheless, the effluent induced a significant increase of reproduction, which disappeared in the farther downstream site. As dietary causes seem to be excluded, a toxic cause is suggested. This will be further investigated when the contamination of biofilm (snail feeding), water and snails will be available. The possible trophic transfert of contaminants will be assessed. 245 In situ reproductive bioassay with caged Gammarus fossarum (Crustacea): a tool to diagnose toxicity of wastewater effluentsC. Lopes, Université Lyon UMR CNRS; O. Geffard, Irstea; B. Xuereb, University of Le Havre, Laboratoire d'Ecotoxicologie-Milieux Aquatiques (UPRES EA3222); R. Coulaud, G. Jubeaux, H. Queau, A. Francois, A. Chaumot, Irstea. Monitoring the adverse effects of environmental contaminants, and especially wastewater effluents, on the reproduction of species from diverse taxa of biodiversity is an ongoing challenge in ecotoxicology. In this context, a chronic sub-lethal toxicity test was developed on Gammarus fossarum, a keystone species of European freshwater ecosystems, to detect adverse outcomes (such as endocrine disruption) of the exposure to chemical compounds. Moreover, protocols were developed for in situ bioassays with transplanted caged G. fossarum to diagnose toxicity in freshwater systems. Such field bioassays permit to control biotic factors known to influence endpoints measurements, but the impact of abiotic factors like water temperature and hardness can not be controlled. In previous laboratory ecophysiological studies, we shown that the moulting cycle of G. fossarum only depends on temperature. A temperature-dependent model was thus developed to describe the moulting development of G. fossarum according to water temperature. The aim of our presentation will be to show how we can use such a model to take into account the effects of temperature in the interpretation of responses measured during in situ assays. Based on ex situ and in situ case studies, we will firstly show the improvement of the the temperature-dependent model by transplanting gammarids throughout four seasons in two reference stations. Secondly, to illustrate the relevance of this model in interpreting the moulting development modulation of G. fossarum exposed to wastewater effluents, we will show the results from: 1- ex situ exposure performed to assess toxicity reduction by alternative tertiary treatments in pilot WWTP; 2- in situ exposure upstream and downstream from three WWTP effluent outputs discharged in RhôneAlpes Rivers (France). Our study underlined the relevance of formalizing the impact of confounding factors on the responses of ecophysiological tests in order to have a good interpretation of in situ bioassays. Furthermore, the reproductive bioassay with caged G. fossarum appeared as a relevant tool to diagnose toxicity of wastewater effluents with potential adverse effect for reproductive processes within receiving ecosystems. 246 Organic pollutants from wastewater treatment plants affect macroinvertebrate communities K. Bunzel, Helmholtz Centre for Environmental Research UFZ / Department Bioenergy/System Ecotoxicology; M. Kattwinkel, Helmholtz Centre for Environmental Research UFZ / System Analysis, Integrated Assessment and Modelling; M. Liess, UFZ Center for Environmental Research / Department of System Ecotoxicology. More than 10 years after the adoption of the EU Water Framework Directive, the majority of German

62

rivers and streams are still at risk of failing to achieve the good ecological and chemical status by 2015. Pesticide contamination of streams is a major stressor for stream ecosystem health preventing the achievement of a good status. Emission from point sources such as municipal wastewater treatment plants (WWTPs) is an important pathway by which pesticides enter surface waters. However, to date, no studies have focused on the ecological effects of pesticide-contaminated WWTP effluent on macroinvertebrate communities. On the basis of governmental monitoring data of 247 sites in Hesse, Germany, we identified insecticidal long-term effects on the structure of the macroinvertebrate community up to 3 km downstream of WWTPs. The effects were quantified using the trait-based SPEAR index, pesticides which has been shown to be an effective tool for identifying community effects of pesticides (insecticidal toxicity). According to the ecological status class boundaries of SPEAR proposed by Beketov et al. pesticides (2009), a total of 75% of the sites with a WWTP within 3 km upstream had a moderate to bad ecological status, compared to only 31% of the sites with no WWTP within 3 km upstream. In addition, based on the German Saprobic Index, we revealed that WWTPs are still an important source of oxygen-depleting organic pollution, despite the extensive technological improvements in wastewater management over several centuries. According to the German WFD classification system, a total of 51% of the sites with a WWTP within 3 km upstream had a medium to bad ecological status. In contrast, 91% of the sites with no WWTP within 3 km upstream had a good or even high ecological status. For pesticide management, our findings emphasize the need to take municipal WWTPs into consideration in the management of river basins under the EU Water Framework Directive to achieve good ecological and chemical status for European streams and rivers. 247 Active materials for the in-situ stabilization of contaminated soil S.E. Hale, Norwegian Geotechnical Institute; J. Jensen, Aarhus University; L. Jakob, Alfred Wegener Institute for Polar and Marine Research; P. Oleszczuk, Maria Curie-Sk?odowska University; T. Hartnik, Climate and Pollution Agency; T. Henriksen, Lindum Ressurs og Gjenvinning; G. Okkenhaug, Norwegian Geotechnical Institute; P. Meynet, D. Werner, R. Davenport, Newcastle University; G. Breedveld, G. Cornelissen, Norwegian Geotechnical Institute. The project “Active materials for the in-situ stabilization of contaminated soil” was carried out in order to investigate how polycyclic aromatic hydrocarbon (PAH) and heavy metal contaminated soils could be remediated using materials capable of stabilizing these pollutants in-situ. Field tests were carried out in collaboration with a waste treatment company, in which activated carbon was added to stabilize PAHs and Fe(OH) powder/limestone and 3 0 Fe granulate were used to stabilise metals. A great variety of end points were used to assess treatment effectiveness, including available pollutant concentrations, effects on microbes, invertebrates and plants. In addition the toxicity of the amendments themselves was tested. The field plots for the PAH and metal contaminated soils were constructed as lysimeter cells. For the PAH contaminated soil, three plots were constructed with buffer areas of about 5 m; one reference, one PAC amended and one GAC amended (containing 2 % AC). For the heavy metal contaminated 0 soil, Fe(OH) powder/limestone and Fe granulate were mixed at 2 and 3 4 % in to a soil taken from near a shooting field at Steinsjøen. The available PAH concentrations measured with passive samplers were reduced 93 % and 76 %, 17 and 28 months after PAC amendment, compared to 84 % and 69 % for GAC. PAC had a negative effect on plant growth while the GAC increased the growth rate of plants. PAC was toxic to earthworms, demonstrated by a significant weight loss, while the results for GAC were less clear. Reproduction of F. candida was significantly increased when 2 and 5 % GAC and biochar were used. In avoidance tests, E. fetida showed no avoidance of the different amendments. After three years, total microbial cell counts and respiration rates were highest in the GAC amended soil. In the metal contaminated soil the porewater was analyzed for pH, EC, Pb, Cu, Zn, Fe, Ca, Mg, Sb(III)/Sb(V) and DOC. A good efficiency of Sb and Pb retention in the treated soil for both amendments was seen. These results show activated carbon was successful in reducing the leaching of organic contaminants from soils, and iron-based sorbents, were

SETAC Europe 23rd Annual Meeting

successful in reducing antimony leaching from soils. This use of this method to stabilize contaminated soil is now being offered by the industry partner, Lindum Resources and Recycling, in this project. 248 Bioturbation affects on natural attenuation and in-situ remediation with thin layer AC applicationD. Lin, Y. Cho, Stanford University / Department of Civil and Environmental Engineering; D. Werner, Newcastle University / Civil Eng. and Geosciences; R.G. Luthy, Stanford University / Department of Civil and Environmental Engineering. Capping hydrophobic organic carbon (HOCs) contaminated sediment with various clean materials, including sand and activated carbon and clay mixtures are effective in reducing sediment-to-water flux of HOCs and porewater concentrations in the bioactive layer. Natural deposition of clean sediments may also be effective in burying older contaminated sediment, although it is less clear how effective natural deposition is for capping contaminated sediment. This is especially the case in the presence of bioturbation, since natural sedimentation is usually slow (typically 1 cm/year) and bioturbation may continually re-mix clean and contaminated sediment. These processes were studied in microcosm experiments using DDTcontaminated sediment (2 mg/kg dw ?DDT) from Lake Maggiore, Italy. The microcosms consisted of an underlying control contaminated sediment with three different treatment methods with thin caps: 1) control, with no capping 2) natural deposition, comprising of a control sediment sublayer with a thin layer (0.5 mm) of background sediment cap (0.03 mg/kg dw ?DDT) and 3) in-situ treatment, comprising of a control sediment sublayer with a thin layer (0.3 cm) of virgin activated carbon cap. For each sediment treatment method, we conducted a set of microcosm experiments with Lumbriculus variegatus and microcosms without worms. The thin-layer sediment cap was only effective in reducing the sediment-to-water flux when bioturbation was not present, while the in-situ treatment AC cap was effective in reducing the ?DDT flux even in the presence of bioturbation by 94 ± 4% compared to the control sediment with bioturbation. The 28-day bioaccumulation study showed that the thin-layer AC treatment method reduced bioaccumulation in the worms by 66 ± 33% compared to the control. Potential ecotoxicological effects in lipid contents and survival rates were not observed in our study. Porewater concentration profile measurements at 0.5 cm resolution showed bioturbation creates a more uniform concentration profile by increasing the surface porewater concentration. These experimental results show thin activated carbon placement above sediment can be effective in reducing the short-term flux of contaminants from the sediment, even in the presence of bioturbation. 249 Biochars as a Climate Change Mitigation Technology: The Implications for Bioavailability and Efficacy of Pesticides R.S. Kookana, S. Martin, CSIRO Land and Water; S. Nag, Central Inland Fisheries Research Institute. Biochar is an emerging technology worldwide for mitigation of climate change. However, little attention has been paid towards potential unintended consequences of such a practice, especially on the bioavailability and efficacy of pesticides. The objective of this study was to establish to what an extent the application of biochar to soil can influence the bioavailability of soil-applied pesticides to biota and adversely affect their efficacy. We evaluated the ability of several biochars in sorbing pesticides through batch sorptiondesorption experiments. Sorption-desorption behaviour was also compared between soils freshly ammended with biochar versus those with biochar aged under field conditions for two years. We also carried out a glasshouse study to evaluate a wheat straw biochar produced at 450° C for its ability to influence bioavailability and persistence of two commonly used herbicides with different mode of actions (atrazine and trifluralin) in two contrasting soils. Sorption was found to increase with increasing biochar contents in soil and even small amounts of biochar (0.1%) made a significant difference in not only sorption but also their desorption behaviour. In presence of 0.5 or 1% biochar in soils, the sorption was essentially non-reversible suggesting sequestration of the sorbed pesticide. The study showed that due to ageing (e.g. via organomineral interactions) in two years under field conditions, biochars can

63

lose their extraordinary sorption capability. The efficacies of both herbicides were reduced in the presence of biochar in soil, but the magnitude of effect was dependent on the herbicide and soil type. In the presence of 1% biochar in soils, the dose of atrazine required for 50% weed control needed to be increased by about 3.5 times in Calcarosol -1 -1 (from 1.17 kg ha to 4.16 kg ha ) and 2.5 fold in the Ferrosol (from -1 -1 1.45 kg ha to 3.48 kg ha ). The corresponding increase was much smaller in the case of trifluralin. The effect was strongly dependent on the mode of action of the herbicide. Persistence increased in biochar amended soils, indicating the reduced bioavailability of herbicides. The study concluded that freshly applied biochars to soils can significantly reduce the bioavailability and efficacy of soil applied pesticides to the extent that up to 3-4 times the normal recommended rate may need to be applied for an effective control or weed/pest in a soil amended with 10 tons of biochar per ha.

removed in the bioreactor significantly exceeded the percentage desorbed from untreated soil, suggesting that the contaminants in lessaccessible domains were susceptible to biodegradation. Toxicity of solvent extracts of the whole soil slightly decreased after biological treatment. However, genotoxicity of solvent extracts of the whole soil slightly increased after biological treatment. Both toxicity and genotoxicity of the desorbable constituents in the soil decreased after treatment, suggesting that any genotoxic constituents that may have formed during treatment were primarily associated with less-accessible domains in the soil. If the desorbable fraction of a contaminant approximates bioavailability to ecological or human receptors, this finding suggests that biological treatment may correspond to a decrease in overall risk.

252 Evidence of limited exposure originating from desorption resistant PAHs in black carbon. V. Gouliarmou, Aarhus University 250 In situ activated carbon amendment reduces bioaccumulation in Science and Technology Faculty / Environmental Chemistry and aquatic food chains D. Kupryianchyk, M. Rakowska, Wageningen Microbiology; E. Christiansen, Aarhus University / Department of University; I. Roessink, Altera Wageningen; T. Grotenhuis, A.A. Environmental Science; K.J. James, University of Saskatchewan / Koelmans, Wageningen University. In situ activated carbon (AC) Toxicology Group and Soil Science; R. Peters, Soil Science Department amendment is a new direction in contaminated sediment management, / Toxicology Group and Department of Soil Science; S.D. Siciliano, yet its effectiveness and safety have never been tested on the level of University of Saskatchewan / Toxicology Group and Department of Soil entire food chains including fish. Here we tested the effects of three Science; P. Mayer, Technical University of Denmark / Department of different AC treatments on hydrophobic organic chemical (HOC) Environmental Engineering. Bioremediation of historically concentrations in pore water, macrophytes, benthic invertebrates, contaminated soils often ends up with a desorption resistant fraction of zooplankton and fish (Leuciscus idus melanotus). AC treatments were HOCs that is very difficult to remove even after intense treatment [1]. mixing with powdered AC (PAC), mixing with granular AC (GAC), and Additionally, during the last decades various black carbon sorbent addition/removal of GAC (sediment “stripping”). AC sediment treatments amendments have been applied for soil remediation [2]. In this case, the resulted in a significant decrease in HOC concentrations in pore water, aim is not to remove contamination but to eliminate or at least reduce macrophytes, benthic invertebrates, zooplankton, and fish. In 6 months, the exposure pathway between the contaminants and the receptor. This PAC treatment caused a reduction of accumulation of approach takes advantage of strong sorption for the immobilization of polychlorobiphenyls (PCB) in fish by a factor of 20 bringing pollutant the contaminants. However, the crucial question remains whether levels below toxic thresholds. All AC treatments supported growth of desorption resistant contaminants indeed give rise to no or at least very fish, but growth was inhibited in the PAC treatment, which was limited risk and exposure. In the current study we incubated wood soot explained from reduced availability of zooplankton. in contaminant traps for more than one year in order to remove the readily desorbing PAH fraction from the soot. The exposure originating 251 Bioavailability of Genotoxic Contaminants in PAHfrom the remaining desorption resistant PAHs was then studied in terms Contaminated Soil Before and After Biological Treatment J. Hu, of (1) chemical activity, (2) in vitro oral bioaccessibility and (3) uptake The Dow Chemical Company / Department of Environmental Sciences in juvenile swine. The working hypothesis was that: 1) due to desorption & Engineering; M. Aitken, University of North Carolina / resistance, the reduction in exposure is higher than the reduction in total Environmental Sciences and Engineering; A. Adrion, University of PAH concentration and 2) limited exposure originates from the North Carolina; D. Shea, North Carolina State University / Department desorption resistant PAHs. All measurements confirmed limited of Biology; J. Nakamura, University of North Carolina at Chapel Hill / exposure originating from the desorption resistant PAHs in black carbon Department of Environmental Sciences & Engineering. Field(soot). The present study strongly supports the underlying assumption of contaminated soil from a former manufactured-gas plant site was treated bioavailability research, that desorption resistance gives rise to limited in a laboratory-scale, aerobic, slurry-phase bioreactor. The bioreactor exposure. These findings further support that bioremediation and sorbent was operated in a semi-continuous (draw and fill) manner, by replacing amendment can be used as efficient exposure reduction measures, even one-fifth of the reactor slurry with untreated soil every week. if they do not succeed to remove the desorption resistant contaminants. Desorbability and biodegradability of 14 polycyclic aromatic References 1. Reichenberg, F.; Karlson, U. G.; Gustafsson, O.; Long, S. hydrocarbons (PAHs) and four oxygenated PAH metabolites (oxyM.; Pritchard, P. H.; Mayer, P. 2010 Low accessibility and chemical PAHs) in the soil were investigated throughout a treatment cycle. Of the activity of PAHs restrict bioremediation and risk of exposure in a four oxy-PAHs analyzed, the most prevalent was 9,10-anthracene dione manufactured gas plant soil Environ. Pollut. 158: 1214-1220 2. Hilber, (9,10-anthraquinone, or AQ). Desorbability was determined using a I.; Bucheli, T. D. 2010 Activated carbon amendment to remediate mixed-function HLB resin or Tenax beads in dialysis tubing suspended contaminated sediments and soils: A review. Global NEST 12: 305-317 in the soil slurry. Toxicity and genotoxicity of the whole soil and the desorbable fractions were determined by DNA damage response 253 Indirect effect of an insecticide on an ecosystem service: analysis with the DT40 chicken B-lymphocyte isogenic cell line and its pollination by moths M. Hahn, Institute for Environmental Sciences; -/DNA-repair-deficient mutant Rad54 , respectively. Biological C. Bruehl, University of Landau Institute for Environmental Sciences. treatment significantly removed PAHs and oxy-PAHs, and the Pollination is an essential ecosystem service in terrestrial habitats. In desorbability of both PAHs and oxy-PAHs decreased throughout the agro-ecosystems, field margins can provide habitat for reproduction and one-week bioreactor treatment cycle. Collectively, oxy-PAHs were more foraging for several species including many pollinators. However, desorbable and biodegradable than the PAHs. When comparing oxyinsecticide inputs of adjacent fields via overspray and spray drift might PAHs with their parent PAHs, we observed that 9,10affect insect pollinators in field margins and, hence, be detrimental to phenanthrenequinone and AQ were more desorbable and biodegradable pollination services for crops and wild plants. We focused on the than the parent compounds phenanthrene and anthracene, respectively; 9- indirect effects of a field margin-relevant concentration of an insecticide fluorenone was less desorbable and biodegradable than the parent (lambda-Cyhalothrin, formulation: Karate Zeon) on moths and the compound fluorene; and benz[a]anthracene-7,12-quinone was more pollination services provided by them. As study organism we used the desorbable but less biodegradable than the parent compound White Campion (Silene latifolia), a dioecious plant species specialized benz[a]anthracene. For both PAHs and oxy-PAHs, the percentage on nocturnal moth pollination. In our study, half of the test plants (6

SETAC Europe 23rd Annual Meeting

64

plants, 36 flowers) were sprayed with the insecticide in an application rate consistent with the pesticide input due to overspray and spray drift in 1m wide field margins (30% of the field application rate). Such small field margins are common landscape features in German agroecosystems. The treated plants and untreated control plants (6 plants, 34 flowers) were exposed to natural pollination during one night in a semifield design. Two weeks later, seed production of the flowers was compared between treated and untreated plants. Furthermore, it was assessed if the flowers were used for oviposition by a specialized moth pollinator (Hadena bicruris). Our results showed that flowers treated with the insecticide were less likely to be pollinated and used as oviposition sites compared to untreated flowers. It is known that Lepidoptera in juvenile as well as adult development stages are sensitive towards low rates of pesticides. The reduced pollination and oviposition observed in this study were possibly caused by repellence of moths due to the insecticide treatment. Since field margins constitute habitats for several Lepidoptera species in agricultural landscapes, inputs of insecticides might not only be detrimental to pests but also to non-target species and the ecosystem-services provided by them. Repellence is so far not included in a risk assessment approach; however, it might affect population size of pollinators in agro-ecosystems dramatically. 254 Ecological models in chemical risk assessment – Recommendations of the SETAC workshop MODELINKU. Hommen, Fraunhofer IME; A. Alix, Dow Agrosciences / Risk Management; D. Auteri, Auteri; P. Carpentier, ANSES / Direction du Vegetal et de l'Environnement; P. Dohmen, BASF SE / Landw. Versuchsstation, APD/RO; V. Ducrot, INRA / Ecotoxicology and Quality of Aquatic Ecosystems; V.E. Forbes, University of Nebraska Lincoln / School of Biological Sciences; V. Grimm, Helmholtz Centre for Environmental Research UFZ / Department of. Ecological Modeling; T.G. Preuss, RWTH Aachen University / Institute for Environmental Research; M. Reed, Chemicals Regulation Directorate, HSE; W. Schmitt, Bayer CropScience AG / Environmental Modelling; P. Thorbek, Syngenta / Environmetal Safety; L. Wendt-Rasch, KEMI. Recently EFSA has published a framework based on an ecosystem services approach for deriving specific protection goals for environmental risk assessment of pesticides. Within this framework ecological modelling is identified as a promising tool to link the results of ecotoxicological studies to such specific protection goals because it can facilitate extrapolation from standard test endpoints to higher levels of biological organization and can explore the influence of various kinds of ecological complexity on the degree of risk. However, currently there are no recommendations on which models are suitable or how to apply them to risk assessments. Therefore, the general objective of the SETAC Europe technical workshop MODELINK is to provide guidance for when and how to apply ecological models to regulatory risk assessment. Such guidance is essential because risk assessment questions, data and protection goals vary across chemicals, ecosystems and species and consequently the question of how to apply models to risk assessments is not trivial. MODELINK focusses on the risk assessment for plant protection products. Approximately 60 experts from diverse backgrounds and representing the tripartite structure of SETAC participate upon invitation. The workshop is built on two meetings over 3 – 4 days in October 2012 and April 2013 and a homework period in between to work within breakout groups on case study reports covering risk assessments without and with ecological models for birds and mammals, fish, soil invertebrates, terrestrial arthropods, aquatic invertebrates and macrophytes. These reports will be discussed and refined during the second meeting and serve as the basis to derive general recommendations on the use of effect models in pesticide risk assessment which will be presented. 255 When Pests Pose Greater Risks to Ecosystem Services than Pesticides: A Case Study D.P. Kreutzweiser, S. Capell, Canadian Forest Service / Natural Resources Canada; D. Nisbet, P. Sibley, University of Guelph / School of Environmental Sciences; P. Hazlett, Canadian Forest Service / Natural Resources Canada; T. Scarr, Ontario Ministry of Natural Resources / Forest Health and Silviculture. The

SETAC Europe 23rd Annual Meeting

exotic invasive insect, emerald ash borer (EAB), is rapidly spreading through easternNorth Americaand causing extensive mortality of ash trees. Many of these trees are in riparian (shoreline) areas that provide critical refuge habitats, movement corridors, and other ecological services including the support of a rich biodiversity and the protection of aquatic ecosystem health. We conducted microcosm experiments to test effects of systemic insecticides for control of EAB on non-target, leafdecomposing organisms. We then designed field studies to compare a no-intervention option, to determine impacts of the insect pest on riparian forests and some of the ecosystem services they provide. Preliminary results indicate that EAB causes rapid and extensive mortality of all ash tree species in riparian areas. When ash trees compose about 30% or more of riparian forests, this mortality causes large and sudden canopy openings. Light penetration to forest floors is measurably increased, forest floor vegetation proliferates, incursions by invasive plants are about doubled, and nitrogen cycling in riparian soils is increased by about 4 times above baseline rates. Across our riparian plots, ash is always among the top 4 tree species contributing leaf litter to forest floors and adjacent water bodies, with an average contribution of 20% and ranging to 45%. Ash litter inputs have a distinct seasonal trend; always among the earliest inputs. Among litter from the 6 most common riparian trees, ash is preferred (decomposed) by aquatic invertebrates as first or second choice in selection microcosms. Further endpoints, including invertebrate communities on in-situ leaf packs with or without ash, are being assessed and will be reported. We show that risks posed by effective insecticides are quantifiable and manageable. Conversely, risks posed by the insect, if left unchecked, appear to threaten the ecological function of riparian forests in terms of canopy cover and shading, forest plant biodiversity, soil nitrogen cycling, leaf litter inputs, and the food web dynamics of adjacent water bodies, probably for several decades at least. 256 Ecosystem Services in Pesticide Regulation: Soil Treatments used in Tomato Production in Italy S. Deacon, ENVIRON UK Ltd; A. Alix, Dow Agrosciences / Risk Management; G. Quadri, ENVIRON Italy; E. Tescari, M. Miles, Dow Agrosciences; P. Burston, J. Nicolette, ENVIRON; M. Rockel, ENVIRON International Corporation. This project delivered a case study to inform pesticide regulation at Member State level where an ecosystem services approach can inform risk and integrated pest management (IPM) decisions under the European Plant Protection Products Regulation 1107/2009 and the Sustainable Use Directive 2009/128/EC. The sustainable use of pesticides calls for a holistic approach to be taken in decision making. An ecosystem services framework has been developed and is applied to a range of soil treatments for the control of nematodes in tomato cultivation and other salad crops in southern Italy. Farmers rely on soil sterilisation treatments as an important component of a pest control management strategy. The use of chemical (1,3-dichloropropene (1,3-D)), physical (solarisation) and biological (microbial) treatments limit the root damage to tomatoes caused by nematodes, greatly effecting the growth and yield of tomatoes and the income earned by farmers at harvest time. Surveys of tomato growers in the Puglia and Sicily regions of Italy were undertaken to gather information on the socio-economics and the management practices of tomato cultivation. A spatial model was developed to evaluate the use of a range of soil treatments in tomato farms in the two regions. Specifically, the results of the evaluation were used to determine: (a) changes in ecosystem services in the absence of nematicide use; (b) the influence of other pest management options in the absence of, or in combination with, nematicides; (c) the effect on socio-economics and the environment posed by the discontinued use of nematicides and the strategies that might be developed to minimise changes and enhance current ecosystem services for future generations. 257 Climate change and species community responses between north and southern Europe V.B. Oliveira; J.J. Scott-Fordsmand, Aarhus University / Department of Bioscience – Terrestrial Ecology; A.M. Soares, Universidade de Aveiro / Department of Biology and CESAM - Centre for Environmental and Marine Studies; M.J. Amorim, Universidade de Aveiro / Department of Biology and CESAM. Climate

65

change may have a severe impact on terrestrial ecosystems. Further, effects are expected to be more severe in the southern Europe than north. It is important to study regional variations occurring between soil species communities (e.g. soil types) and the associated services/functions as well as to anticipate the level of change that temperature may induce. Here, we studied species community changes and function using soil mesocosms - Soil Multispecies System (SMS) – assessing 1) the effect of a temperature range resembling northern European (10-14-19-23ºC) conditions; 2) the effect of a temperature range resembling southern European (19-23-26-29ºC) conditions and 3) the interaction between temperature and Cu for both 1 and 2. Further, 1 year field experiments were run on the respective countries and field conditions. SMSs contained six species representative of different functional groups and were sampled along three exposure periods (28, 61, 84 days). OM breakdown and feeding activities were also assessed. Soils were from two different regions in Europe (Portugal and Denmark). Higher temperature (19-29ºC) caused a higher impact in species community than lower (10-23ºC). Effect of Cu was observed at 10-23ºC experiments whereas at 19-29ºC the high temperature overruled the Cu toxicity. Differences between historical Cu contamination and freshly Cu spiked soil indicate large differences in bioavailability as verified by the measured total concentration (ca. 1000 and 100 mg/kg respectively). Major changes in communities occurred during the growth phase (28 days at SMS tests and late spring in field), after which, at stable phases, changes in populations dynamics will be more reflecting actual species interactions and matured ecosystem status being both less resilient to changes and better fitted to the exposure scenarios (Cu and temperature). Ecosystem services, e.g. biodiversity, agriculture and associated OM turnover, will be impacted. Effects are more pronounced at more extreme low (10ºC) and high temperatures (29ºC) and the momentum feeding activity at different time intervals will better describe such changes compared to OM breakdown cumulated along time. Effects on species community and ecosystems were less predictable at higher temperature conditions. This is partly related with differences in species optimum curves, more skewed at the higher end, hence smaller differences will impact very rapidly. 258 Ecosystem services within sustainable soil management and area development using the Triple-D approach S. Keuning; M.H. Wagelmans, Bioclear. The definition of Ecosystem Services was first formalised in the Millenium Ecosystem Assessment [1] although scientists have been working on the concept for decades. A wealth of information is available with the professionals that are working at the scientific part of ecosystem services but also with the professionals within area development, urban development, soil remediation and nature development and control. However, communication between all different expertises keeps hampering, even within organisations, which slows down or even puts area development to a stop. How can soil management, soil and area functions and spacial redevelopment be connected? How can we link air quality, biodiversity, water and soil? How can different and even opposing interests be weighed and considered? That needs a more-dimensional consideration framework. The ecosystem services approach combines two worlds that apparently are difficult to reconcile. Economy and ecology. Profit and sustainability. Whealth and environment. Exploiting and protecting our environment in one approach, that acknowledges and connects all these differences. This was the base for the development of the Triple-D approach that subsequently will be further described and presented. The strengths and opportunities of the Triple D-approach will be illustrated in three examples in which the approach was used to investigate the area in which the ambitions needed to be realized, with ecosystemservices and the natural capital as starting point. 259 Linking of exposure and effects in higher tiers of the assessment of the risk of pesticides to honey bees as proposed in the EFSA guidance document J. Boesten, Alterra / ERA team; A. Rortais, EFSA / Emerging Risk Unit; F. Streissl, C. Szentes, EFSA. In the EFSA guidance proposal the assessment of the risk of pesticides to honey bees consists of two components, i.e. the effect assessment and

SETAC Europe 23rd Annual Meeting

the exposure assessment, which have to be adequately linked. In the past five years considerable experience has been gained with adequate linking for the effect assessments of aquatic organisms which can be used for the risk assessment of honey bees. It is proposed to base the effect assessment primarily on the concentration in nectar and pollen entering the hive and to combine these concentrations with consumption rates of nectar and pollen by adult bees and larvae (later also contributions of guttation water and other water sources may be considered). Furthermore it is proposed to base the field-exposure assessment on a 90th percentile of the exposure concentration considering a well-defined statistical population of hives, i.e. all hives at edges of treated fields in the area of use of the substance. The 90th percentile concentration entering the hive may as a first approximation be combined with 90th percentile rates of consumption of nectar and pollen to achieve an overall 90th percentile pesticide intake rate. There are two types of exposure assessment in such a linked risk assessment procedure: (A) the exposure of the hive in the field resulting from the use of the pesticide in agriculture (resulting in the 90th percentile concentration described above), and (B) the exposure of the hive in the higher-tier effect studies. So (A) assesses the reality in agriculture whereas (B) assesses what happened in the regulatory studies. In the past, the concentrations in nectar and pollen in the hives in the regulatory studies were usually not measured; instead it was assumed that exposure was adequate e.g. because the hive was located at the edge of a treated oil seed rape field. However, it is questionable whether this is a defensible approach to achieve at least a 90th percentile exposure. Therefore we recommend to measure in higher-tier semi-field and field studies always the course of time of the concentrations in nectar and pollen entering the hive. For adequate linking it is needed that the exposure in the higher-tier studies is higher than the endpoint of the field-exposure assessment. Otherwise the outcome of such a higher-tier study may be that there are no unacceptable effects at a concentration that is lower than the endpoint of the exposure assessment which is a non-relevant result for the risk assessment. 260 Higher tier refinements for birds and mammals under EFSA 2009: a regulator's perspective J. O\'Leary Quinn , Chemicals Regulation Directorate / Ecotoxicology. EFSA 2009 lays out the European approach to assessing the risk to birds and mammals for pesticide active substances. It has now been in use in practice for about three years allowing regulators to gain practical experience of higher tier refinements which have been successfully used in practice. The Chemical Regulations Directorate (CRD), the regulatory body for the United Kingdom, highlights refinement steps based on our direct experience and which have been successfully used, as well as some of the challenges that can arise. For the acute risk assessment for several active substances body burden modelling has been proposed however EFSA (2009) highlights that further research is needed first. Also where acute data are available for multiple species a geometric mean of the toxicity end point may be taken in certain circumstances. Field trials may also provide additional information to support the risk assessment but such studies need careful and detailed examination and this approach has had very limited use. For the reproductive risk, the most frequent refinements are the degradation time used for both single and multiple application as well as use of the use of relevant focal species to the crop together with an appropriate food intake rate for that species. For birds it may also be possible to identify certain periods of the year when a reproductive risk assessment is not required as the birds are not breeding, additionally the UK has some recent research on breeding phases for UK species which potentially could be used to with the phase specific approach. Another important approach for each higher tier assessment is to consider the uncertainties and level of conservatism of the risk assessment relative to the protection goals. 261 Optimizing experimental setup and sampling techniques to assess pesticide effects on macroinvertebrates in aquatic mesocosm studies T. Strauss, RWTH Aachen University / Research Institute Gaiac; M. Hammers-Wirtz, Research Institute gaiac / Research Institute Gaiac; S. Norman, Ridgeway Eco; U. Hommen, Fraunhofer IME.

66

Aquatic mesocosm studies are often used as a higher tier tool in the environmental risk assessment of plant protection products. To minimize the uncertainty in the extrapolation to the field situation it is necessary that vulnerable species are present in the test systems in sufficiently high numbers to allow the detection of effects caused by the test item. Here, vulnerability is related to the intrinsic (toxicological) sensitivity to the test item and the ecological sensitivity with respect to the species life cycle and recovery potential. An additional factor influencing species vulnerability, the exposure, is given by the experimental setup. For insecticides in particular, emerging insects such as mayflies and further arthropods such as isopods and amphipods are considered as ecologically relevant and representative vulnerable species to be included in such a study. In this poster, we will present results from an outdoor mesocosm study where the design was optimized for these groups of macroinvertebrates. Stainless steel tables with a sediment layer introduced into the mesocosms provided a shallow zone with planted macrophytes as a habitat for macroinvertebrates. Colonization of the mesocosms came from the introduced sediment taken from a natural pond, water samples from ponds or untreated mesocosms and, in the case of flying insects, via the air. In addition, approximately 250 individuals of Asellus aquaticus (Isopoda) and Crangonyx pseudogracilis (Amphipoda), respectively, were introduced in each mesocosm and supported by supplying additional food (Populus leaves). The macroinvertebrates were intensively sampled to increase the sampling efficiency and thus to reduce the sampling error. Emerging insects were sampled by two traps, one installed above the shallow macrophyte area and the other above the deeper water column area. The emergence rates of the two traps were compared with respect to abundance and community composition. The efficiencies of different types of epibenthic and floating non-destructive sampling devices were analysed for Cloeon dipterum (Ephemeroptera), Asellus aquaticus and Crangonyx pseudogracilis. 262 An evaluation of five years experience in off-field NTA testing F.M. Bakker, Mitox Consultants; S. Aldershof, Bioresearch Promotion. Spray drift of plant protection products occurs and effects on ecosystem services potential (ESP) inherent to non-target areas (NTA’s) must be assessed. Bakker and Miles (2007) proposed a field study design for an ERA on arthropod communities present in the NTA (also see de Jong et al. 2010). These communities provide a suite of services, all of which require stable and complex food webs for sustainable delivery. The ESP of NTA’s is thus in nature a biodiversity issue and measurements of changes in biodiversity on a regulatory relevant time scale are key to risk assessment. The time factor is important because it provides insight in the resilience of the ecosystem (the service provider) and thus in the acceptability of effects. Historically, the analysis uses PRC-techniques (van den Brink and Ter Braak, 1998, 1999) where changes in relative abundance can be conveniently plotted over time and relative to a control. With this paper we evaluate five years of experience in off-field testing with particular reference to the following questions: (1) Were faunistic studies in an off-field environment needed? The alternative are in-field studies, therefore we compare both the biodiversity and the sensitivity of the NTA community to an in-field situation. (2) Was the test design sufficiently precise to find dose-response relationships and derive relevant endpoints such as the NOEAER? Some selected but anonymized cases will be presented to illustrate this point. (3) Was the test design sufficiently robust to buffer against effect dilution (e.g. immigration) and yet still realistic for the habitat type under study? Time-response patterns following disruption were analysed. (4) Were habitats sufficiently diverse to capture spatial heterogeneities in exposure that may be expected to occur in true off-field exposure scenarios? Vegetation surveys, their analysis and their relation to the arthropod community will be presented to provide a first insight. 263 Guidance Document of EFSA’s PPR Panel on tiered effect assessment for plant protection products in edge-of-field surface waters A. Aagaard, M. Arena, S.K. Bopp, J. Boesten,T. Brock, M. Klein, M. Liess, R. Luttik, D. Pickford, A. Tiktak, L. Wendt-Rasch, Members of the Working Group Aquatic Ecotoxicology of the PPR

SETAC Europe 23rd Annual Meeting

Panel of the European Food Safety Authority (EFSA). This presentation aims to provide scientific background information on the aquatic effect assessment scheme developed by EFSA for authorisation of plant protection products (and their metabolites) in the EU. It has its focus on experimental approaches. Furthermore, the appropriate linking between exposure and effect assessment estimates is described, as well as the possible use of non-testing methods (in particular for metabolites). The Guidance Document (GD) describes first the specific protection goals (SPG) for aquatic organisms. The overall aim of the SPG is to protect aquatic plants and animals at the population level in surface water. However, the SPG selected for aquatic vertebrates aims at protection at the individual level. To protect populations of aquatic organisms, effect assessment schemes are developed that enable derivation of Regulatory Acceptable Concentrations (RACs). In the aquatic effect assessment scheme a distinction is made between acute and chronic effects and the following tiers are developed, viz: Tier 1 effect assessment based on toxicity data for standard test species Tier 2 effect assessment based on laboratory toxicity data for both standard and additional test species, i.e. the geomean (Tier 2A), the species sensitivity distribution (Tier 2B) and the refined exposure laboratory toxicity test (Tier 2C) approaches. Tier 3 effect assessment based on model ecosystem (micro-/mesocosm) studies. 264 Apple orchards & chlorpyrifos: 1st year results of UK field monitoring study on species-diversity, abundance & reproduction of birds & mammals 2012 F. Sotti, Tier3 Solutions GmbH; B. Giessing; S. Norman, Dow Agrosciences; C. Wolf, Tier Solutions GmbH; G. Weyman, Makhteshim Agan (UK) Ltd. The aim of this longterm field monitoring study is to provide data on species-diversity, abundance & reproduction of birds & mammals in 10 large apple orchards in Herefordshire, UK. All 10 orchards were sprayed with chlorpyrifos in 2012, as normal. The work in 2012 was done from start of April to end of July, and will continue in ‘13 & ’14. The methods were: bird-ringing (mist-netting), nest searching & monitoring (open nests & natural cavities), visual surveys, predator surveys & mammal trapping. In total, 1166 birds (931 adults, 223 juveniles & 12 nestlings) of 37 species were trapped; the most abundant were great tit (274), blue tit (236), robin (172) & blackbird (107). During visual surveys, 636 individuals of 33 species were observed; most frequent were wood pigeon (114), blue tit (95), blackbird (78) & great tit (69). Concerning breeding, a total number of 123 active nests of 20 species, were found & monitored. Among them, 45 open-nests of 12 species were located in the orchards & their immediate surroundings, mainly of blackbird (19) and wood pigeon (5). 962 tree- cavities were found in 4 orchards (with older trees). Active nests were in 68 cavities, mainly of blue tit (30) & great tit (18). Predators were recorded & motion-sensitive cameras were used to include nocturnal species. 448 observations of potential predators were noted, mainly carrion crows, common buzzards, domestic dogs & badgers. For mammals, 809 individuals of 7 species were trapped. The most common were bank vole (334), wood mouse (272) & yellownecked mouse (177). Despite only 20% of traps being in the off-crop habitat within a few metres of the orchards, the majority of captures (89.5%) occurred in these traps (7 species). In contrast, within the orchards only wood mice & one individual bank vole were trapped. Overall, the results showed a diverse community of birds. Notably, the most abundant species trapped (great tit & blue tit) were small insectivores, the group highlighted in EU pesticide assessment schemes as being at greatest risk. For mammals, in this landscape, the off-crop appeared to be much more attractive than inside the orchard. This is logical due to cover from predators by long grass & food availability. The results are valuable for plans in 2013, including use of nest boxes. 265 Development of characterization factor to assess biodiversity damage due to salinity increase in a coastal wetland M. Amores Barrero, Universitat Rovira i Virgili / Chemical engineering department; F. Verones, Institute of Environmental Engineering (IfU); C.E. Raptis, Institute of Environmental Engineering; R. Juraske, S. Pfister, F. Stoessel, ETH Zurich; A. Assumpcio, IRTA; F. Castells, Universitat Rovira i Virgili / Chemical engineering department; S. Hellweg, ETH Zurich / Institute of Environmental Engineering. Coastal wetlands are

67

among the most productive, valuable, and yet most threatened ecosystems in the world. They provide a critical interface between terrestrial and marine environments, where fresh water and salt water are often mixed. Coastal wetlands in arid and semi-arid zones experience periods of increasing salinity as a consequence of high evaporative conditions, variability of inflows, impacts of human pressure and their proximity to the sea. In this paper, a new method has been conducted close to our case study, so in order to develop a methodology for salinity impacts authors selected a coastal Spanish wetland called “Albufera de Adra” as a case study. It is located in a semi-arid region (South-East of Spain), where agricultural activities require substantial irrigation and areas with native vegetation and fauna are restricted to some small patches and wetlands. This wetland is situated closer to the sea and it is predominantly fed by groundwater. This work derived the first Characterization Factor (CF) for salinity impacts in a coastal wetland defined as the change in the Potentially Affected Fraction (PAF) of species due to a change in salinity and extraction of groundwater for crop irrigation. A Life Cycle Impact Assessment method was developed to evaluate the environmental impacts associated with salinity on biodiversity in this Spanish coastal wetland. This CF for the salinity impact is defined as the change in the PAF of species due to a change in salinity which is caused by the changed of groundwater for irrigation in the vicinity of the Albufera de Adra. The developed CF consists of a fate 3 (FF) and an effect factor (EF) and it indicates a “potential loss of 0.32 m 3 -1 ecosystem” for a water consumption rate of 1 m yr . The FF was calculated from seasonal water and salt balances of the wetland Albufera de Adra and the EF was obtained from the fitted curve of PAF of native wetland species due to salinity. In order to test the applicability of the CF, an assessment of water consumption of greenhouse crops in the area (mainly 86% of groundwater) was conducted as a case study. Results converted into ecosystem quality damage using the ReCiPe method were compared to other categories. Keywords: life cycle assessment, salinity, ecotoxicity, coastal wetland. Session: Life cycle analysis (LCA) and sustainability Presentation preference: platform presentation 266 Developing an assessment methodology for the impact of consumptive water use on the biodiversity of wetlands of international importance F. Verones, Institute of Environmental Engineering (IfU); S. Pfister, ETH Zurich; D. Saner; C. Rondinini, Università di Roma La Sapienza / Department of Animal and Human Biology; D. Baisero, Università di Roma La Sapienza; S. Hellweg, ETH Zurich / Institute of Environmental Engineering. Wetlands harbor large varieties of species, but they also belong to the world’s most threatened ecosystems. Half of their global area was lost during the last century. So far no approach exists in life cycle impact assessment which acknowledges the vulnerability and importance of wetlands and provides impact factors for this kind of ecosystems on a global scale. We use 1184 inland wetlands which are currently designated as sites of international importance under the Ramsar Convention for developing regionalized Fate Factors (FF) and Effect Factors (EF) for consumptive water use. We distinguish between surface water-fed and groundwaterfed wetlands and between surface water and groundwater consumption. FFs quantify the change of wetland area caused per m3 water consumed and EFs quantify the “global species”-equivalents which are potentially lost for each change in wetland area. Since we are assessing the impact per wetland area, our assessment methodology is compatible with impact assessment methodologies of land use impacts. EFs are calculated for waterbirds, non-residential birds and water-dependent mammals. Characterization factors (CFs) which are derived with the FFs and EFs of the individual wetlands are distributed over each wetland’s individual catchment. In areas where catchments overlap, the CFs are summed up, since obviously several wetlands will be damaged from consumptive water use at this location. Thus global, spatially differentiated maps of impacts from both consumptive surface water and groundwater use are derived for waterbirds, non-residential birds and water-dependent mammals. A case study of rose production in Kenya is calculated in order to assess the impacts of both consumptive surface and groundwater use in the region around Lake Naivasha, which is one of Kenya’s designated wetlands of international importance. The case

SETAC Europe 23rd Annual Meeting

study highlights the importance of incorporating these impacts into life cycle assessment, since impacts of consumptive water use proved to be in the same order of magnitude like agricultural land occupation and several orders of magnitude more important than freshwater eutrophication or terrestrial acidification. 267 Towards including regionalized soil-water consumption in LCA M. Núñez , UR050, Laboratoire de Biotechnologie de l’Environnement; S. Pfister, ETH Zurich; P. Roux, Irstea; A. Anton, IRTA. This research aimed to address in life cycle assessment (LCA) the impact pathway of soil-water consumption, also denoted as green water consumption in the literature, which is of great importance for dry lands. The developed approach consists on quantifying, at the inventory stage, the net change in soil-water consumption under a specific human land use compared to the natural reference situation. The potential natural vegetation (PNV) was selected as the reference situation. To allow the approach to be applied, we estimated water consumption of PNV as a function of location for global dry lands of the world. Two methods were combined to estimate soil-water consumption of PNV: an empirical approach based on annual precipitation and potential evapotranspiration data, and another approach based on remote sensing models for actual evapotranspiration of natural areas. The combination of both approaches enables to obtain more robust results. Water consumption of the regionalized natural reference situation was derived on three spatial scales of aggregation: 10 arcmin grid cells, ecoregions (501 units) and biomes (14 units). This facilitates inclusion of soil-water consumption in LCA for different detail of the geographic information available. The method is intended to be used in contexts of rain-fed agriculture and rainwater harvesting, which includes direct soil moisture uptake by plants and rainwater harvested and reused in production systems. The approach provides a basis for impact assessment of soilwater consumption. Specific characterisation factors and impact assessment model have not been proposed so far. The life cycle inventory framework and the associated mapped results are already usable for several situations of land use and land use-changes. Impacts of soil-water consumption should be subsequently assessed by characterisation factors (CFs) for water consumption on the impact assessment (LCIA) stage. The approach is of high relevance for wateruse-dependent land uses such as agriculture and forestry. 268 Linkage with land use and freshwater availability in the context of water footprint M. MOTOSHITA, National Institute of Advanced Industrial Sci and; K. TAHARA, National Institute of Advanced Industrial Science and Technology; A. INABA, Kogakuin University. Water footprint is a tool for assesing freshwater avalability loss caused by artificial activities. Water consumption as a result of direct withdrawal from freshwater resources is basically discussed and focused in many present cases. However, freshwater availability seems to be affected by not only withdrawal of water but also artificial land use. Land cover change will result in the infiltration of rain water to soil and groundwater. The change of infiltration amount of rain water will vary directly the amount of groundwater and indirectly the amount of surface water through changing freshwater outflow from ground to surface freshwater resources. While the linkage between water and land use have been potentially pointed out in some studies [1][2], quantitative evidence has not been provided yet. The focus of this study Is to assess the effect of land use on water availability and clarify the significance of land use in the context of water footprint. Water deprivation intensities 3 2 for each artificial land use type could be calculated as -0.0065 [m /m ] 3 2 3 2 for planted forest, 0156 [m /m ] for paddy field, 0.0161 [m /m ] for 3 2 arable land, 0.452 [m /m ] for urban area, on average in Japan. Average 3 2 freshwater resource availability in Japan is calculated as 1.23 [m /m ] by deviding total amount of available freshwater resource with total land area. Thus, in case of land use as urban area, the loss of freshwater availability caused by land use is accounted as almost 37% of freshwater availablitiy per land area on average. While water consumption by withdrawal is still significant in many industries, water deprivation by land use is more dominant in some industries. If the availability loss of water by land use were taken into account in water footprint, the value

68

of water footprint will be 14 times (at the maximum) more than that calculated by conventional concept (only considring withdrawal-based consumption). Water deprivation by land use affects on the freshwater availability and the effect on water footprint may be accounted up to 14 times. In addtion, large regional difference could be found in the intensity of water deprivation by land use. The interaction between the amount of availability loss by land use and regional impact factor may affect on the results of water footprinting. 269 Diving into water scarcity indicators: model components, choices and uncertainties A. Boulay, CIRAIG École Polytechnique de Montréal / Chemical engineering department; S. Pfister, ETH Zurich; C. Bulle, CIRAIG Polytechnique Montreal / Chemical Engineering; M. Margni, Ecole Polytechnique de Montreal / Department of Mathematical and Industrial Engineering. This paper presents preliminary results of the third deliverable of the WULCA project (Water Use in LCA) of the UNEP/SETAC Life Cycle Initiative performing a quantitative comparison of the available water impact assessment methods in order to enhance the understanding of differences, similarities, modeling choices, data used and uncertainty related to the choice of model. This paper aims to 1) compare selected methods addressing water scarcity as a midpoint indicator by identifying and quantifying the key elements and modeling choices driving their results, and 2) assess individual regional model uncertainty and propose recommendations towards a consensual indicator. Results from water scarcity methods are first normalized using a reference flow of a world 3 average water m and then compared using the Spearman rank correlation coefficient and the Gini mean difference coefficient. The former expresses the consistency of model response for different flows and the latter expresses the mean difference between two model results for the same flows. Similar models have high rank correlation and low mean difference. These tests are also performed for specific modeling choices to assess their relative influence: source of data, source of water (surface water vs. groundwater), consideration of temporal variation, use of consumption or withdrawal based scarcity index, choice of regional resolution scale and of aggregation methods. Results show that the type and scale of regional resolution, the source of data used and the consideration of consumed water to assess scarcity have a significant effect on the results. However to enlighten these choices, questions need to be answered as to which data source is the most representative, how should scarcity be defined when using consumption-to-availability ratios and what is the most relevant spatial resolution. Choices to include source of water or seasonal variation could be limited only to more sensitive regions in the world as they are time and data consuming to apply and will not add to the discrimination power of the indicators in most regions of the world. 270 Assessing cosmetic products life cycle water use impacts focused on water pollution S. Vionnet, Quantis; I.M. Francke, Natura Cosméticos; S. Humbert, Home. Water use impact assessment, or water footprint, methodologies are quickly evolving on one hand. On the other hand, companies need to measure water footprints of their products in order to better manage their impacts on water resources. Water use impact assessment methodologies within the frame of life cycle assessment will answer this need, although there is no concensus on the method to use yet among the various options. In this work we explored methodologies addressing the degradative use of water. This is particularly a challenge for cosmetic products that incorporate complex ingredients. Key parameters of those ingredients, like human toxicity or degradability, are usually difficult to find resulting in difficulty to apply the methodologies. We assessed five methodologies that adress degradative use of water: Grey water footprint (GWF), Critical Dilution Volume (CDV), Water Impact Index (WIIX), USEtox ecotoxicity and Ridoutt and Pfister 2012 methodology. The endpoints methods used for calculating this latter were eutrophication (ReCiPe), acidification (IMPACT2002+) and ecotoxicity (USEtox). Two products of a cosmetic company called Natura were assessed: a facial cream and a body oil made out of 32 and 10 different ingredients respectively. The assessement emcompassed the entire life cycle of both products, from

SETAC Europe 23rd Annual Meeting

cradle to grave. The functional unit were defined regarding the use of one entire packaged product. Each methodology and associated results were then assessed in term of science, communication and deployment criterias in order for Natura to select the more sound methodology to deploy its water strategy. The results showed that each methodology had different life cycle stages contributing at various extent to the impacts. The parameters considered by each method (toxicty, degradability, local water stress index, amount of water consumed, etc) are first responsible for the differences between methodologies results. The data gaps and uncertainty was the second source of variability in results. Overall the Ridoutt and Pfister 2012 methodology allows for the best 3 communication (with units in m -eq), is based on the best science available (including USEtox methodology) and is the most comprehensive in term of pollution issues. Its applicability shortcomings are possible to overcome in the future, although it will require new development from the scientific community. 271 Spatially resolved approach for linking emission to human exposure: a case study for Europe S. Sala, Joint Research Centre European Commission / Sustainability Assessment Unit - Institute of Environment and Sustainability; B. Ciuffo, Joint Research Centre / Sustainability Assessment Unit - Institute of Environment and Sustainability; D. Marinov, JRC EC / Institue for Environment and Sustainability; M. Trombetti, Joint Research Centre / Institute of Environment and Sustainability. Spatial differentiation is a topic of increasing interest within ecological risk assessment (ERA) and Life Cycle Impact Assessment (LCIA). Consequently, multimedia models are increasingly adopted for supporting the analysis of the elements affecting the spatial variability of ecotoxicity and human toxicity impacts. A key issue to be addressed in the impact assessent methods, models and corresponding impact factors is the level of spatial detail required and uncertainties related to the use of generic impact factors when the exact location of the activities is unknown. For human exposure via inhalation, we run the spatially resolved multimedia model MAPPE Europe (1*1km), in order to assess the relative influence of scale and input parameters in calculation of Intake Factors (iFs) under different emission scenarios, such as: (1) a multiple uniform emission coming from highly populated and industrialised areas. The amount of emission in each European country has been scaled up by the population density. In these runs, two substances have been evaluated: 1,1,2,2 tetrachloroethane - representative of highly volatile chemicals; Nitrobenzene - representative of multimedia chemicals; acephate – representative of hydrophilic chemicals; (2) a point source emission coming only from a single country (Spain, Luxemburg and Slovenia); (3) a multiple non uniform emission, accounting for real industrial emission of 1,2 dichloroethane (representative of volatile chemicals), as reported in the European Pollutant Release and Transfer Register (EPRTR), from the countries where these emissions are reported and then from Belgium only. The results were, then, compared to those of a nested box model (USEtox) in order to address the uncertainty related to the adoption of a non-spatially resolved model. The iFs for the different tested substances vary of 1 order of magnitude amongst countries in Europe and the values are mainly affected by the physical chemical properties of the substances and by the population density . The location of the emission affect the results, as for example, considering the case of a single emission per country, emitting the same quantity from Luxemburg and Slovenia implies a factor of 2 of differences. Comparing the results with USEtox, as average, the iFs at continental scale are one order of magnitude higher whereas at urban scale two orders of magnitude higher than in USEtox. 272 Predicting honeybee exposure to pesticides from vapour drift using a combined pesticide emission and atmospheric transport model T.S. Geoghegan, K.J. Hageman, University of Otago / Chemistry; M. Scheringer, ETH Zuerich / Institute for Chemical and Bioengineering. Modelling pesticide emissions to air is important for understanding pesticide vapour drift and its potential effects on nontarget environments. Several environmental fate models have been developed for pesticides but few use plant or species-specific descriptors

69

to describe volatilisation from crops. As a chemical’s environmental fate is a function of its physical-chemical properties and the nature of the environmental media in which it resides, both factors should be considered when estimating pesticide volatilisation. This work proposes a new way to estimate pesticide emissions from a sprayed field using soil-air and plant-air partition coefficients. It then links those emissions to potential effects in the near-field non-target environment using risk assessment methods and honeybees as an indicator species. This risk assessment approach is unique in that the emphasis is on the effects caused by the process of vapour drift rather than by particular pesticides. The objectives of this work were to (1) develop a model for estimating pesticide emissions from agricultural fields that considers the interactions of pesticides with soil and plants (including plant size and surface characteristics), (2) combine emissions data with a chemical transport model to determine downwind concentrations of pesticides in air and plant phases and (3) use modelled results to calculate risk quotients for honeybees exposed to volatilised pesticides in a non-target area, thereby assessing the importance of vapour drift as an environmental exposure mechanism. Volatilisation emissions were calculated using mass balance and flux equations, and had a good 2 correlation (R =0.8) with literature values. Downwind air concentrations were calculated from persistence and characteristic travel distance data determined using the Small World model. Pesticide concentrations in plants in the non-target area were estimated from deposition flux to plant surfaces. Honeybee exposure was calculated for respiratory, contact and oral routes which were combined to give total daily exposure. Preliminary results indicate that, for some pesticides, vapour drift alone could contribute enough pesticide to a non-target area to cause an adverse affect on foraging honeybees; while for other pesticides it is an important contributing factor to environmental contamination. Therefore vapour drift should be considered in traditional risk assessments along with other environmental exposure mechanisms. 273 Evaluating the environmental fate of short-chain chlorinated paraffins (SCCPs) in the Nordic environment using a dynamic multimedia model I.S. Krogseth, Norwegian Institute for Air Research; K. Breivik, Norwegian Inst for Air Research; J.A. Arnot, ARC Arnot Research Consulting / Department of Physical Environmental Science; M. Schlabach, Norwegian Inst. for Air Research. Short chain chlorinated paraffins (SCCPs), also called polychlorinated n-alkanes, are mixtures of compounds of molecular formula C H Cl containing 10-13 carbon atoms and usually 30x 2x+2-y y 70 % degree of chlorination. They have a range of industrial applications, but there are concerns regarding their potential for environmental persistence, bioaccumulation, adverse effects and longrange transport. However, the environmental fate and exposures of SCCPs to humans and ecological receptors is poorly understood. The technical mixtures consist of thousands of isomers, which make analysis and modeling of these compounds very challenging. The goal of this study was to evaluate the overall understanding of the link between emissions of SCCPs, environmental levels and human exposure in the Nordic environment, and to identify the more critical knowledge gaps. Additionally, we wanted to assess the extent of expected variation of environmental fate within the complex mixture of SCCPs, and to evaluate the merits and limitations of using an average property estimate for the whole group. The 46 theoretically possible formula groups of SCCPs were represented by one isomer each, and physical-chemical properties were estimated. Realistic emission scenarios for sum SCCPs in the Nordic countries were estimated using a high-throughput screening method and data on chemical usage of SCCPs in the Nordic countries. The physical-chemical property and emission estimates were used to parameterize an integrated, non-steady state environmental fate and bioaccumulation multimedia model (CoZMoMAN), and the model results were compared to measured levels in the Nordic environment, as well as to predictions based on the “average” SCCP properties used in the EU Risk assessment report of SCCPs. Preliminary results show that the EU “average” SCCP does not capture the large range in possible environmental behaviors. In general the model underestimated

SETAC Europe 23rd Annual Meeting

concentrations for all compartments compared to measured levels, implying that the emission estimate was underestimated. Results will be presented and discussed with emphasis on the more critical research needs with respect to the overall fate and exposure of SCCPs. 274 Highly dynamic multimedia fate models in risk assessment of contaminated sites: why is “one fits all” unsuited? M. Morselli, University of Insubria / Department of Science and High Technology; G. Porto, A. Biasiolo, Copernico S.r.l.; A. Di Guardo, University of Insubria / Department of Science and High Technology. Subsurface contamination is one of the environmental problems of most concern for human health and environment. The most acknowledged references offering guidelines for the managements of sites affected by this type of contamination are the American Society for Testing and Materials (ASTM) Risk Based Corrective Action (RBCA) standards for evaluating petroleum release sites and chemical release sites. In these documents, the use of modelling tools is encouraged. However, even the most used modelling tools, such as RBCA Toolkit for Chemical Releases, consider a static environment, characterized by environmental properties which are constant in time. This often leads to extremely conservative predictions. In order to understand the potential improvement deriving from a more accurate description of the environment, a new dynamic multimedia fate model (SoilREM) was developed. SoilREM is based on the site-specific dynamic model SoilPlus, in which two atmospheric layers interact with a litter/soil multi-layered system. The height and wind speed of the atmospheric layers, representing the planetary boundary layer (PBL) and the atmospheric residual layer, are computed on an hourly basis starting from standard meteorological observations, in order to reflect the high variability which is usually observed in this part of troposphere. The development of SoilREM implied the nesting of 3 SoilPlus units in order to simulate different spatial scales, the smallest of which can represent a specific outdoor point as well as a building located within the contaminated site. The simulations with SoilREM were performed for a semi-urban site located near Milan (Northern Italy) for two selected chemicals (benzene and phenanthrene); results showed that air chemical concentrations directly respond to the shortterm variations occurring in the lower troposphere (i.e., PBL height and wind speed oscillations). A comparison with the results produced by RBCA Toolkit for Chemical Releases confirmed the excessively conservative nature of the latter approach. For this reason, a dynamic model such as SoilREM could be useful for the assessment of the occurrence and magnitude of exposure peaks and could provide vital information for performing adequate monitoring campaigns in contaminated sites. 275 Development of a realistic environmental scenario for a dynamic bioaccumulation model of organic contaminants in plant biomass E. Terzaghi, University of Insubria Como / Dept. of Science and High Technology; M. Morselli, University of Insubria / Department of Science and High Technology; B. Cerabolini, University of Insubria; A. Di Guardo, University of Insubria / Department of Science and High Technology. The approach currently employed for environmental risk assessment need to be improved in the next few decades since it lacks of ecological realism. This goal can be achieved through the development of a number of more realistic scenarios which consider temporal and spatial variability of environmental parameters and their adoption in dynamic multimedia fate model capable to predict time and space variable concentrations.\nModels which consider variability in exposure concentrations and in environmental and compartment parameters are scarce; additionally, even fewer models include a dynamic vegetation compartment.\nThe aim of the present work was to develop a fully dynamic scenario that consider the variability of exposure concentrations, meteorological and compartment parameters and to adopt this scenario in a dynamic bioaccumulation model (SoilPlusVeg) to predict the temporal uptake and release (on hourly basis) of some Polycyclic Aromatic Hydrocarbons in a mixed broadleaf wood located in Northern Italy.\nSoilPlusVeg is a multimedia fate model based on the fugacity approach which incorporates a double layered dynamic atmosphere compartment which varies in height on hourly basis, a multi-

70

layered soil (bare or covered by up to three litter horizons) and a vegetation compartment. The vegetation compartment can be composed of a mono-specific or multi-specific forest canopy which includes roots, stem and leaves. While a preliminary version was presented before, now the model is improved considering a dynamic LAI (Leaf Area Index) and SLA (Specific Leaf Area) to obtain a leaf biomass that change with time.\nAir and leaf concentrations of 16 PAHs and two ecological parameters (LAI and SLA) measured weekly in a small broadleaf wood located in Northern Italy were used to developed the realistic scenario. Hourly or daily values of meteorological parameters were instead provided by the Regional Environmental Protection Agency (REPA).\nSimulations were performed to evaluate the role of air concentration variations, meteorological and compartment parameters variability in influencing the change of foliar biomass concentration with time. Comparison of measured vs predicted leaf concentrations produced results within the same order of magnitude. A preliminary sensitivity analysis showed the relevance of a number of parameters which need to be calibrated in order to obtain a better fit. 276 Using passive samplers to assess sources, transport and fate of legacy and emerging organic pollutants in a U.S. North East coast estuary R. Lohmann, University of Rhode Island / Graduate School of Oceanography. The emergence of passive samplers in recent years has made the sampling of hydrophobic organic contaminants in water and porewater easy and reliable. A group of volunteer citizen-scientists was recruited to deploy passive samplers in the Narragansett Bay (NB) watershed, greatly expanding the study area. Freely dissolved concentrations of emerging contaminants (triclosans, alkylphenols, and PBDEs) as well as legacy contaminants (OCPs) were determined. Triclosan, previously undetected in the Bay was detected in the NB watershed suggesting releases in coastal waters. Tetra, penta, and hexa PBDEs were consistently detected in the NB watershed at -1 concentrations ranging from 100’s to 1000’s of fg L , with higher maximum concentrations than previously reported within the NB. Lastly, OCPs such as hexachlorobenzene, trans-chlordane, cischlordane, t-nonachlor, p,p’-DDE and methoxychlor were widely detected within the watershed. Methyl triclosan and the PBDEs showed higher concentrations in the Providence River and the head of the bay where human influence is greatest and flushing is lowest. Alkylphenols and OCPs in contrast, did not show the same pattern of distribution suggesting that there is substantial release from historically contaminated sediments. Passive samplers are suited for deployment by citizen-scientists and can help raise awareness of pollutants and identify spatial/temporal trends and potential point sources. 277 Endocrine disrupting chemicals and antibacterial agents in the coastal environment Z. Xie, HelmholtzZentrum Geesthacht. 278 Contaminants of emerging concern in the Norwegian marine environment K. Thomas, NIVA. 279 Panel discussion: Future issues - contaminants in estuarine and coastal environments J. Readman, . 280 Intestinal uptake of copper nanoparticles in rainbow trout using an in vitro intestinal perfusion, involves nystatin and vanadatedependent pathways R.D. Handy, Ecotoxicology Research and Innovation Centre; H. Soliman, University of Plymouth. The production of manufactured nanoparticles is increasing, making their release into the environment inevitable. There is an emerging literature on the toxic effects of nanoparticles in a variety of aquatic organisms, although the uptake mechanisms for nanoparticles are not entirely understood. In this study, whole gut sacs from rainbow trout were used to identify the regions of gut involved in Cu NP absorption. Exposure of whole gut -1 sacs to 1 mg l Cu as CuSO or Cu NP in the luminal solution caused 4 Cu accumulation mostly in the mucosa (62% or more) rather than the underlying muscularis. Tissue total Cu concentrations indicated that Cu was predominantly absorbed in the mid and hind intestine, regardless of the nano form. Perfused intestines were used to investigate details of the

SETAC Europe 23rd Annual Meeting

uptake mechanisms and showed good viability criteria including low LDH leak (-1) and steady perfusate flow. The accumulation of Cu from CuSO was higher than that from Cu NP in both mid and hind intestine. 4 Serosal applications of the P-type ATPase inhibitor vanadate (100 µmol l1 ) caused 3.1 and 5.35 fold decreases in net Cu uptake from copper sulphate and copper nanoparticles, respectively. Also, additions of 120 u -1 ml nystatin caused a 1.9 and 3.75 fold decline in net Cu flux to serosal compartment from CuSO and Cu NP, respectively. The present study 4 shows that copper as CuSO or Cu NP are mostly absorbed across the 4 mid and hind intestine, with the uptake rate of Cu from Cu NPs being especially sensitive to inhibitors. The uptake mechanisms involve a vanadate sensitive pathway, as well as a nystatin–sensitive endocytosis across the mucosal membrane. 281 Uptake kinetics of silver nanoparticles and silver nitrate by Pseudokirchneriella subcapitata related to Ag speciation in solution F. Ribeiro, University of Aveiro CESAM / department of Biology & CESAM; J. Gallego, University of Gothenburg / Department of Chemistry and Molecular Biology; C. van Gestel, Vrije Universiteit Amsterdam / Ecological Science; E.V. Soares, Chemical Engineering; A.M. Soares, Universidade de Aveiro / Department of Biology and CESAM - Centre for Environmental and Marine Studies; S. Loureiro, Universidade de Aveiro / Biology. Freshwater algae plays an important role in lakes and ponds as primary producers, representing a great fraction of the diet of zooplankton feeding on suspended particulate matter. A change in the productivity and/or population structure of algae has the potential to affect the entire food chain of a lake, and therefore it is important to consider such organism as a key species for toxicological assessments. This study assessed the uptake kinetics of AgNP (3-8nm, AMEPOX) and silver nitrate (AgNO from Sigma Aldrich, 99% purity 3 CAS 7761-88-8) in the green algae Pseudokirchneriella subcapitata, in order to relate internal concentrations in the algae with Ag speciation in the colloidal suspension to which they were exposed. To describe the bioconcentration of silver in P. subcapitata over time, a first order kinetics model was used to calculate uptake and elimination rate constants (k and k ). Toxicokinetcs were calculated relating Ag uptake 1 2 to each of the different colloidal fractions that were analysed. To take into account the decrease of silver concentrations in the different fractions during the uptake phase, a decay rate constant was estimated based on first order kinetics. This decay rate constant was included in the first order uptake kinetics model. Bioconcentration Factors (BCF´s) were calculated as for each exposure, being K (uptake constant) and 1 K (elimination constant) Considering the non-filtered fraction, both 2 uptake and elimination were fitted to the kinetics model. Assuming that K is more related to exposure and bioavailability, we therefore will 1 center our conclusions based on these values. Even though the BCF values look similar for both concentrations of AgNO , we obtained 3 different values of K meaning that at 15 ?g/L of AgNO , algae were 1 3 able to uptake silver faster than at 30 ?g/L. For silver nanoparticles, the picture looks quite dissimilar from AgNO since K for both 3 1 concentrations are almost identical to each other. This suggests that silver nanoparticle uptake by algae is not a concentration-dependent parameter, whereas it can be considered an Ag-speciation dependent parameter. 282 Bioaccumulation and Toxicity of Metal NanoParticles to Sediment-Dwelling Organisms H. Selck, Roskilde University; G.T. Banta, Roskilde University / Environmental, Social and Spatial Change; V.E. Forbes, University of Nebraska-Lincoln / School of Biological Sciences. Nanotechnology is expanding rapidly, thus increasing the release of engineered nano-particles (NPs) to the aquatic environment. Sediment is likely to be the environmental compartment that is most exposed to metal-bearing NPs, and organisms that feed on sediment are likely to be particularly at risk from metal-NP exposure. Due to their increased surface area and reactivity, NPs may be more bioavailable and toxic than their macroscale counterparts. However, very little is known about the environmental impact of NPs and whether they behave similarly in the environment compared to their macroscale counterparts, and these factors make it difficult to perform environmental risk

71

assessments and to set quality standards for NPs. Some of the questions we are facing are whether nano-specific properties lead to unexpected and new biological effects? If so, is the potential for such effects diminished when NPs are introduced to a complex environment such as the sediment compartment? Acknowledging that we are not yet able to characterize NPs once they have been introduced to the sediment compartment, we used a comparative approach to assess toxicity and bioavailability of metal NPs. This presentation will show that by using realistic exposure scenarios, applying different well-characterized forms and sizes of metals to the sediment compartment, and examining subsequent bioavailability and toxicity, we found evidence that bioavailability was affected by the metal form introduced even though metal inevitably will change speciation once it enters the sediment compartment. The effect and importance of metal form added to sediment for metal bioavailability and subsequent toxicity is not straightforward, however, and depends both on biological factors, such as digestive complexity, and physical-chemical factors affecting the fate and speciation of metals in the sediment compartment. To better understand these mechanisms we need to focus future research efforts on developing methods to characterize nanomaterials in complex environments, including sediment, soil and diet, that are robust, reliable and yet relatively easy to implement in a standard laboratory. 283 ZnO nanoparticle fate and effects on Folsomia candida in soil P. Waalewijn-Kool, VU University; M. Diez Ortiz, Centre for Ecology and Hydrology; C. van Gestel, VU University. In this contribution we will provide an overview of the short-term and long-term fate and effects of coated and uncoated ZnO-NP on the springtail Folsomia candida. Also, the effect of soil pH on the toxicity of ZnO-NP will be shown for a field soil amended to three different pH levels. In a 28-day toxicity test, springtail reproduction was reduced in a dose-dependent manner and EC values of 1964, 1591 and 298 mg Zn/kg were 50 estimated for uncoated ZnO-NP, non-nano ZnO and ZnCl , 2 respectively. Coated ZnO-NP (EC = 873 mg Zn/kg) was more toxic 50 than uncoated ZnO-NP. Based on porewater Zn concentration EC50 values for all Zn forms were in the same range (7.94-16.9 mg Zn/l), suggesting that the zinc ions released from the ZnO rather than the particles themselves are repsonsible for the observed toxic effects. ZnONP dissolution and related toxicity continued changing for one year equilibration of spiked soil. Our toxicity tests showed that an equilibration period of three months or longer is able to reduce toxicity F. candida, except for coated ZnO-NP which showed reduced for toxicity only after 12 months equilibration. ZnO addition caused Zn porewater concentrations to steadily increase with time for at least one year. Already after three months, porewater concentrations peaked at intermediate concentrations for ZnO-NP (67.1 mg Zn/l after 12 months) and non-nano ZnO (66.5 mg Zn/l), while for coated ZnO-NP such a clear peak was only seen after 12 months (36.6 mg Zn/l). Dose-related increases in soil pH may explain decreased soluble Zn levels due to fixation of Zn released from ZnO at higher soil concentrations. For the pH amended soils, the effect on reproduction decreased with increasing soil pH for all three Zn forms, with 28-d EC values of 553, 1481 and 50 3233 mg Zn/kg for uncoated ZnO-NP and 331, 732 and 1174 mg Zn/kg for ZnCl , at pH 4.5, 5.9 and 7.2, respectively. EC values based on 2 50 porewater Zn concentrations increased with increasing pH for uncoated ZnO-NP from 4.77 to 18.5 mg Zn/l, while for ZnCl no consistent pH2 related trend in EC values was found (21.0-63.3 mg Zn/l). In 50 comparison with ZnCl , uncoated ZnO-NP toxicity is lower based on 2 total Zn concentrations and higher based on porewater Zn concentrations. This indicates that toxicity of uncoated ZnO-NP can not be solely attributed to the dissolved Zn fraction. The fact that different Zn forms do affect soil pH differently may contribute to differences in toxicity. 284 Bioavailability and toxicity of sediment-bound nanoparticles T. Ramskov, Roskilde University / Department of Environmental, Social and Spatial Change; H. Selck, Roskilde University; G. Banta, Roskilde University / Department of Environmental, Social and Spatial Change; D. Berhanu, Natural History Museum; S.K. Misra, E. Valsami-Jones,

SETAC Europe 23rd Annual Meeting

Natural History Museum, University of Birmingham; V.E. Forbes, University of Nebraska Lincoln / School of Biological Sciences. Due to the widespread applications of metal bearing nanoparticles (NPs), such as copper oxide (CuO), in our daily life, both human and environmental exposure is likely to increase. Once released into the aquatic environment, CuO NPs are likely to agglomerate and/or aggregate, whereby it is expected that they will partition to sediments. Depositfeeding organisms may be at particular risk of exposure to CuO NPs. Thus, in a number of studies we have used an environmentally realistic exposure scenario by examining the fate and effects of metal oxide NPs added to the sediment compartment. The overall aim of the studies performed has been to explore the relative importance of metal form (aqueous vs. NP) and shape (rods, spheres, spindles) added to sediment for subsequent bioaccumulation and effects in the deposit-feeding snail, Potamopyrgus antipodarum. Our results show that both particle form and shape have an important influence on the toxicity and bioaccumulation of Cu in P. antipodarum, and that correlations between these parameters are not straightforward. Besides summarizing highlights from the performed studies, we will also focus on how sediment exposure can increase our understanding of the behavior of nanoparticles in the environment, and thus how increasing ecological relevance in ecotoxicological studies can add value to environmental risk assessment of NPs. 285 Environmental conditions alter the ecotoxicity of silver nanoparticles to Daphnia magna F. Seitz, Inst for Environmental Sciences / Institute for Environmental Sciences; R.R. Rosenfeldt, Uniersity of KoblenzLandau Institute for Environmental Sciences / Institute for Environmental Sciences; K. Storm, Universitiy of KoblenzLandau / Institute for Environmental Sciences; R. Schulz, University of KoblenzLandau / Institute for Environmental Sciences; M. Bundschuh, Department of Aquatic Sciences and Assessment Swedish University of Agricultural Sciences / Department of Aquatic Sciences and Assessment. Products containing silver nanoparticles (nAg), e.g. textiles, are increasingly used. Hence, nAg may finally enter aquatic environments, where they potentially pose a risk to wildlife. For this reason, several studies investigated the aquatic toxicity of different nAg products, frequently focusing on implications for the waterflea Daphnia magna. However, there is only limitted information about ecotoxicological effects of nAg when considering varying environmental conditions, such as pH or dissolved organic carbon (DOC) levels. Therefore, the present study assessed the 48-h acute toxicity of nAg products with varying particle size (citrate coated; 20, 60, 100 nm) and silver nitrate, under different pH (6 and 8) and DOC (0 and 8 mg TOC/L) levels for D. magna. Independent of pH and DOC level, the results clearly indicated a size dependent toxicity of nAg. This was displayed by up to 10 fold increased 48-h EC when comparing 2050 and 100-nm nAg. Moreover, the toxicity of nAg increased with decreasing pH (~3 times) and DOC level (up to ~1.3 times). Silver nitrate, in contrast, displayed – regardless of pH and DOC level – a 48-h EC of 2.07±0.83 µg/L. The findings of the present study hint at the 50 pathway of nAg toxicity, which has been controversially discussed in literature previousely. As the total amount of silver ions in the media strongly depends on the present pH, particle coating (citrate or DOC) as well as on the actual partilce size, most likely ions play the major role in nAg toxicity to D. magna. However, this hypothesis needs to be further assessed by chemical analysis. In conclusion, the present study underpins the importance of environemental parameters for the fate and ecotoxicity of nanoparticles, which should be further investigated in future. 286 Body Residues and Responses of the Midge Chironomus riparius to Sediment Associated FullereneC60 G.C. WaissiLeinonen, University of Eastern Finland; K. Pakarinen; S. Bold, University of Eastern Finland; E.J. Petersen, National Institute of Standards Technology; J. Akkanen, University of Eastern Finland / Department of Biology; M.T. Leppanen, Finnish Environment Institute / Department of Biology; J.V. Kukkonen, University of Jyvaskyla / Biological and Environmental Science. The research of nanoparticles

72

toxicity to organisms has faced dilemma of appropriate testing methods and concepts to deal with them. Common ecotoxicity test do not encounter well to nanoparticles due to their unique chemical characters and thus may cause misleading conclusions. In this study a benthic invertebrate Chironomus riparius larvae were exposed to fullerene (nC ) masses in sediment using an environmentally realistic method. 60 By simulating a sensitive exposure route for C. riparius feeding habits larvae body residues were analyzed using spectrophotometric measurements and gut content investigated with transmission electron microscopy. Exposure conditions were created to test short-term effects after exposure for 10 d consisting instars from first to fourth. In addition to 10d standard exposure, body residues of fourth instar larvae were investigated also after 12 and 24 h exposure times. Results showed clear evidence that extraction method was able to use and perform when studying body residues of C. riparius and indicated that body residues were lower after 15 days than 10 days. This might stem from change in behavior of larval feeding during the development time or possibly after 10 d time larvae have reached the steady state. It was also shown that after short exposures (12 and 24 h) larvae were rapidly packed by fullerenes. Fullerene exposure might have an effect on development time which could be taken as a possible response of stressed larvae. 287 Nanomicex – Ecotoxicity of pristine and modified pigment nanoparticles B.K. Gaiser, School of Life Sciences; J.H. Kinross, D.M. Brown, A. Kermanizadeh, V. Stone, Heriot-Watt University; T.F. Fernandes, HeriotWatt University / School of Life Sciences. A variety of nanoparticles are used as pigments in paints, inks and. Exposure to these particles can be experienced in an occupational setting or by the end user. Nanoparticulate pigments can also reach the environment through waste, spills, abrasion and wash-off. Nanomicex is a recently launched FP7 project combining partners from industry and academia, in which seven different nano-sized pigments (Ag, TiO , ZnO, 2 CoAl O , Cd/Se QDs, Al O and Fe O ) manufactured by the 2 4 2 3 3 4 industrial partners are investigated for their toxic effects in the environment and on human cell lines. The aim of the project is hazard assessment of nanoparticles and the modification of any deemed hazards following those tests in order to reduce adverse effects in the environment and human cells, while preserving the properties which make the particles valuable as paints and dyes. In addition, the project seeks to establish structure-activity relationships and to investigate realistic exposure scenarios. A variety of endpoints including mortality/lethality and sub-lethal effects and mechanisms of action are investigated in Daphnia magna, Lumbriculus variegatus and Pseudokirchneriella subcapitata as environmental model organisms, and in human A549, C3A and J774 cells (alveolar epithelial cells, hepatocytes and macrophages). The pilot studies carried out to date indicate that the TiO NPs studied were of low toxicity to the models 2 and endpoints studied. Further tests will be conducted to verify these results. In contrast, ZnO NPs were found to exert a level of toxicity and therefore further studies will be conducted after NP surface modifications. The collaboration with the industrial partners, the combined information on human and environmental hazards, as well as the facilities to modify particles to keep them suitable for their intended application while minimising the hazard to environmental and public health, make Nanomicex a very promising project for investigating realistic steps towards safe, environmentally compatible use of nanomaterials. 288 Interactions of CdSe/ZnS quantum dots with the ciliated protozoa Tetrahymena thermophilaM. Mortimer, Aquatic Biogeochemistry and Ecotoxicology, Institute F.-A. Forel, Faculty of Sciences; A. Kahru, National Institute of Chemical Physics and Biophysics / Laboratory of Environmental Toxicology; V. Slaveykova, University of Geneva / Aquatic Biogeochemistry and Ecotoxicology, Institute F.-A. Forel, Faculty of Sciences. Understanding the environmental impact of engineered nanomaterials has become an active area of research over the past years due to their possible entrance to ecosystems, including the food chain. Research in this area is still insufficient for a complete environmental risk assessment. In the current

SETAC Europe 23rd Annual Meeting

study carboxyl functionalized CdSe/ZnS core shell quantum dots (QDs) were studied for their cellular uptake, localization and clearance. To consider both the cellular and organism level effects, the current study employed a unicellular eukaryote, freshwater protozoan Tetrahymena thermophila as a model organism. The QDs were not found to be cytotoxic to T. thermophila at concentrations up to 80 nM, which could be explained by the low solubility of the studied QDs: the dissolved fraction of Cd in the exposure medium was negligible and the concentration of dissolved Zn remained below toxic concentrations to T. thermophila. As evidenced by the optical microscopy imaging the investigated QDs, with an average hydrodynamic diameter of 12 nm, were readily phagocytosed by the protozoan upon the exposure to QDs in HEPES buffer. Flow cytometry analysis showed that the food vacuoles of T. thermophila became filled with QDs in time and dose dependent manner. The maximum filling of the food vacuoles with QDs was detected after 2-h exposure, while after 24-h exposure the tendency towards the clearance of the food vacuoles was detected. The process was shown to be dose dependent – the cells exposed to lower concentrations of QDs were able to excrete the content of the food vacuoles more efficiently. After replacing the QD-containing medium with the QD-free buffer about 50% of the protozoan cells had cleared their food vacuoles of QDs after 20 h. Confocal microscopy studies of QD-exposed T. thermophila revealed that phagocytosis was not the only pathway of entry for QD in the cell. In addition to larger aggregates of QDs in the food vacuoles also smaller-sized fluorescent particles were detected in the cells, suggesting QD internalization through cell membrane by alternative mechanisms (e.g. endocytosis). The intracellular QDs localized outside the food vacuoles were not subject to cellular clearance. The current study highlights that inherently non-toxic NPs may pose possible risks by accumulating in the cells and becoming bioavailable for the organisms at higher trophic levels. 289 Multiwalled carbon nanotubes affect long term recolonization and structure of benthic communities I. Velzeboer, Wageningen University; K. Peeters, VITO; A.A. Koelmans, Wageningen University / Environment. Whereas for traditional pollutants knowledge evolved from studying effects in the laboratory to studying effects in situ, assessment of in situ effects for ENPs still is in its infancy. Most ecotoxicity studies focus on single species laboratory tests. This kind of tests is less realistic with respect to multiple exposure routes, dynamic particle mixing, ‘fouling’ and aging of ENPs, or ecological processes like community effects or recolonization. Consequently, assessment of in situ effects is urgently needed because laboratory studies can be expected to be poor predictors of effects under field conditions, especially for ENPs. Aim of this study was to assess long term community effects of macroinvertebrates exposed to sediment contaminated with multiwalled carbon nanotubes (MWCNTs). This study focussed on the influence of MWCNT contaminated sediment on the recolonization by benthic macroinvertebrates in a long term controlled and replicated field experiment (15 months). MWCNT treatment effects were assessed and quantitatively compared to community composition after 3 months and to similar treatment effects of another carbonaceous material (activated carbon; AC) which was applied simultaneously in parallel systems in the framework of another study. A remarkable finding of this recolonization study is that MWCNT contaminated sediment showed significant effects on the benthic community structure after 15 months. After 3 months, effects were not (yet) significant. A dose of 0.0002 – 0.2 % MWCNTs explained 9.9% of the community composition (biodiversity and species abundance, p=0.012). The field-relevant lowest dose tested also caused significant effects (p=0.020). For the first time this shows that despite aging and burial, MWCNTs affect the structure of natural communities. Exposure to AC showed a comparable impact at a 50x higher dose, which suggests that the MWCNTs were about 50 times more potent than AC. 290 Effects of nanoparticles of TiO2 on food depletion and lifehistory responses of Daphnia magna. C. Barata, CSIC / Environmetal Chemistry; B. Campos, IDAEACSIC / Environmental Toxicology; C. Rivetti, CSIC; P. Rosenkranz, INIA; J. Navas, INIA -

73

Madrid. The extent to which different forms of nanoparticles of titanium dioxide (nano-TiO2) aggregated with microalgae, decreased food levels and hence impaired growth, reproduction and fitness of Daphnia magna individuals were studied. Treatments included three different types of nano-TiO2 differing in their coating or cristalline structure but of similar primary size (20 nm) plus a micron-sized bulk material, two exposure levels (1, 10 mg/l) and two food ration levels of the microalgae Chlorella vulgaris that included an non limiting (1.5 µg C/ml) and a limiting one (0.3 µg C/L). Effects were assessed using standardized chronic tests and assays that maximized food depletion in the water column under semi-static and re-suspension conditions. Results indicated that the high ion levels in culture medium lead to the aggregation of nanoparticles followed by particle destabilization. Nanoparticle aggregates interacted with the algae cells, forming clusters. Large TiO2-algae agglomerates settled readily depleting dramatically the concentration of edible food for D. magna. At limiting food ratios food depletion by nanoparticle aggregation had dramatic effects on reproduction and fitness of exposed D. magna at 1 mg/L irrespectively of the particle form. At high food rations effects were only observed for one of the nano-TiO2, P25, at high exposure levels (10 mg/l) under both semi-static and particle re-suspension conditions, which suggest that P25 effects were mediated by clogging the gut and hence diminishing food acquisition. These results indicate that nano-TiO2 may affect the transfer of energy throughout the planktonic aquatic food webs increasing the settlement of edible particles from the water column. 291 Integrating bioaccessibility and community analyses into ecological risk assessment of metal contaminated sites: A shooting range case study S.R. Bowman, The Ohio State University / Evolution Ecology and Organismal Biology; J.L. Bryant, The Ohio State University / Department of Entomology; R.P. Lanno, Ohio State University / Department of Evolution, Ecology, and Organismal Biology. Although lead (Pb) occurs naturally, anthropogenic activities have caused Pb to be elevated beyond normal background concentrations in some areas. Since metal bioavailability varies with soil modifying characteristics and species-specific uptake and elimination, it is sometimes necessary to use site-specific measures such as bioaccessibility estimates and analysis of biotic communities when evaluating these areas. Our study site is a private shooting range located in central Ohio, USA. Study areas include the main shotfall zone of a trap and skeet range (2899 ±1228 mg/kg soil Pb), an area where Pb pellets were excavated (Fall 2009; 628 ±44 mg/kg soil Pb), and a reference area (1356 ±86 mg/kg soil Pb) that does not directly receive spent shot. Our first objective was to determine how ground beetle (Carabidae) community structure (i.e., diversity and abundance) changes along a Pb gradient. Ground beetles were collected from 2008 to 2012 using pitfall traps in shotfall, reference, and Pb extracted areas. Fortyfive species were collected during the first two years. In 2009, overall abundance was greater in the reference area than shotfall. Overall species richness did not differ between sites; however, an indicator species analysis suggested that six species were indicators of the reference site. With additional years of sampling (and a more robust dataset), we expect that we will be able to understand temporal and spatial trends in ground beetle communities at our study site. Our second objective was to adapt in vitro bioaccessibility assays to small mammal physiology so that we can better estimate exposure to Pb from ingestion of soil and prey. Preliminary results suggest that soil bioaccessibility is 5-10% of the total Pb in the field soils. For earthworms (Eisenia fetida) exposed to field soils for 30 days, bioaccessible Pb ranged from < 1% - 4% of the total earthworm Pb. Our next step is to field collect invertebrates, shrews (Blarina brevicauda) and voles (Microtus pennsylvanicus). We will analyze gut content and incidental soil ingestion of shrews and voles from the shooting range and control sites. We will use diet composition, incidental soil ingestion and bioaccessibility estimates to adjust exposure estimates at our field site. We anticipate that results from our study will provide evidence for the usefulness of IVG analyses and invertebrate community analyses in sitespecific ecological risk assessment of metals.

SETAC Europe 23rd Annual Meeting

292 Application of the OECD 307 study to assess the persistence of Gas to Liquid (GTL) Fuel G. Whale, Shell Health; C. Bayliss, Harlan Laboratories Ltd.; S. Forbes, Shell Global Solutions (UK); J. Dawick, Shell Health; C. Mead, Harlan Laboratories Ltd.. The main study currently used to assess the biodegradation of chemicals in soil is the OECD Guideline 307 for testing chemicals: Aerobic - Anaerobic Transformation in Soil (April 2002). This test was originally designed to provide degradation rate data for crop protection products but this is now being undertaken to provide data for other ‘chemicals’ under the EU Reach regulations. Many do not believe the current guidelines are suitable to assess the fate of complex substances. For example, the current approach recommended by CONCAWE (Conservation of Clean Air and Water in Europe) for complex hydrocarbon substances is to model persistence of the constituent hydrocarbon blocks. However as part of the registration of a new substance the European Chemicals Agency (ECHA) have stipulated to the registrants that an OECD 307 study should be undertaken to determine potential persistent hydrocarbon components of a Gas to Liquid (GTL) Fuel. This has proved challenging and in the course of our investigations a number of lessons have been learnt. In this presentation we provide an overview of some of the limitations and issues encountered with the conduct of the OECD 307 test, steps which could be taken to improve the test and finally the relevance of such tests to assess the fate of complex hydrocarbon substances in soil. 293 Closing the bottleneck: evaluation of higher tier options for refinement of endpoints obtained by Collembola and Hypoaspis reproduction tests P. Kabouw; S. Mastitsky, BASF / Crop Protection - Ecotoxicology; S. Royer, BASF SE, Global Ecotoxicology, Crop protection; F. Staab, M. Bergtold, BASF SE. The registration process of plant protection products includes tests on soil organisms such as Collembola and Hypoaspis. When refining endpoints obtained by Collembola OECD 232 and Hypoaspis OECD 226 reproduction tests risk assessors are limited by the test systems they can deploy. To overcome this bottleneck we evaluate previously published test systems for refinement of Collembola/Hypoaspisendpoints. Test systems we evaluate are aged residue tests, multi-generation tests, and RAMAS ® population models. All methodologies were evaluated based on the following criteria; simplicity in lower tiers, a higher level of realism in higher tiers, and the potential of these methods to address recovery. Weaknesses and strengths of the three methodologies will be demonstrated using case studies. Aged residue tests with Collembola/Hypoaspis evaluate the potential for external recovery. This approach is similar to aged residues tests conducted with NTAs. The outcome of the test system is dependent on how soils are aged and on the properties of the test compound. Soils can be aged by considering different environmental factors. Multigeneration tests are similar to aged residues test. However, in contrast to aged residue tests that consider external recovery, multigeneration tests consider internal recovery through reproduction. In this test system potential effects that are transferred over generations can be assessed. Both aged residues and multigeneration tests experimentally examine long term toxicity and take recovery into consideration. In both test systems ageing of treated soil leads to decreased toxicity only when the toxic effect is caused by readily degradable formulation additives or active substances. The RAMAS ® model projects population development based on generated data by OECD 226/232 tests. Such a modeling exercise can, therefore, be conducted without performing additional experimental work. At the same time multiple generations are considered increasing realism and taking recovery from reproduction into consideration. The projection assumes constant and homogeneous exposure of Collembola/Hypoaspis to the potential stressor. We conclude that the above mentioned refinement options are a suitable approach to bring more realism in risk assessment for soil organisms. 294 Collembolan two-generation study within a tiered soil risk assessment G. Ernst, Bayer CropScience / Ecotoxicology; U. Frommholz, U. Menke, P. Neumann, Bayer CropScience. An increasing conservatism in the tier 1 risk assessment (RA) of plant protection

74

products (PPP) for soil organisms is expected due to changing requirements in exposure modeling (e.g. reduced crop interception rate, PEC calculation for 0-1 cm soil layer). However, the exposure of PPP in soil is often limited in time because of various degradation and dissipation processes of the product in soil. In some cases this leads to a fast reduction in toxicity. Thus, the risk, e.g. for Collembola, might be overestimated in a highly conservative tier 1 RA with substances degrading fast. Therefore, a Collembolan two-generation study is proposed as an intermediate tiered laboratory test system in order to assess the inherent recovery potential for Collembola from the treated area/population itself. Juvenile individuals hatched in a Collembolan st reproduction test after 28 d (OECD 232; 1 bioassay) will be transferred nd into a 2 consecutive bioassay containing soil with aged test substance st that was treated together with the soil from the 1 bioassay. This procedure aims to assess if populations which were initially impacted in st the 1 bioassay (e.g. at or above the LOEC) are not impacted in their nd reproduction performance in the 2 bioassay and thus able to build up a new population. A dose response test was performed with Lindane EC150 (emulsifiable concentrate; 150 g/L) following the proposed st nd design. The 1 bioassay leads to a NOEC of 0.1 mg/kg and after the 2 bioassay a NOEC of 0.25 mg/kg was determined indicating only a moderate reduction in toxicity to collembola. Furthermore, a doseresponse Terrestrial Model Ecosystem (TME) is available as well with Lindane EC150 (Scholz-Starke et al., submitted) considering application rates between 24 and 2400 g Lindane/ha. The collembola community was shown to be not even initially impacted at a rate of 240 g Lindane/ha. Comparing the NOEC from the two-generation study with the application rate being safe for the Collembola community in the TME the conservatism of the proposed intermediate tiered RA approach is assessed. A risk (TER 0.68) would still be indicated at an exposure of 240 g Lindane/ha (NOER in TME) even with an a priori TER trigger value of 1. The consideration of the potential for inherent recovery in an intermediate tiered risk assessment approach is a valuable tool to fill the gap between an overly conservative tier 1 and higher tier Collembola risk assessment for substances dissipating fast. 295 The chemical quality of urban soils in Glasgow, UK, with reference to anthropogenic impacts and current toxicologicallybased soil guideline values. F. Fordyce, Geochemical Baselines and Medical Geology; S. Nice, B. Lister, B. O Dochartaigh, British Geological Survey. Until recently systematic data on the chemical quality of urban soils was lacking in many countries as traditional soil survey programmes focussed on rural environments. The advent of environmental protection legislation in the UK in the 1990s has driven the need for urban soil quality information to ensure healthy and sustainable environments. Since 1992, the British Geological Survey (BGS) has completed soil surveys in 27 UK cities under the Geochemical Baseline Survey of the Environment (G-BASE) programme. This included a survey of Glasgow and surrounding rural areas to link to a wider geoenvironmental assessment: the BGS Clyde and Glasgow Urban Super Project (CUSP). The survey provides an overview of land quality in Glasgow and is based upon the collection of 2 2and 4 per km in rural and urban areas respectively. The samples are analysed for total concentrations of c.50 chemical elements by X-ray fluorescence spectrometry. Glasgow is Scotland’s largest city and was a th major industrial centre until the mid 20 century. Much of this industry has now declined and the city is undergoing regeneration. However, the legacy of the industrial past remains. Results for 1381 urban and 241 rural soils reveal that concentrations of many metals are elevated in urban soils relative to the rural background regardless of the geological parent material. Elements that are commonly associated with anthropogenic pollution such as Pb, Sb and Sn show greatest enrichment in urban versus rural soils (2.6 – 3.3 times, based on median values). -1 -1 Median topsoil Cr (108 mg kg ) and Ni (47 mg kg ) concentrations in Glasgow are higher than in many other UK cities due to the presence of volcanic bedrock and history of metal processing in the city. In terms of toxicologically-based soil quality assessments, with the exception of Cr, only a small proportion of soils exceed the current UK human Contaminated Land Exposure Assessment (CLEA) residential Soil

SETAC Europe 23rd Annual Meeting

Guideline Values despite the city’s industrial heritage – Cr (22%), Pb (5%), As, Cd, Ni (2%) and Se (0%). However, a much greater proporton exceed the proposed UK Ecotoxicity Soil Screeing Values – Cr (100%), Ni (94%), Zn (86%), Pb (30%), Cu (17%) and Cd (4%). The G-BASE data allow these thresholds to be evaluated in terms of typical soil element abundances to aid ecotoxicological assessments and inform policy. 296 Results from a Workshop on Ecological Soil Levels—Next Steps in the R.S. Wentsel, Exponent / Ecological Sciences; A. Fairbrother, Exponent Inc / EcoSciences. This paper presents the results from a workshop to develop a process for establishing ecological soil clean-up values (EcoSCVs) in the U.S. The goal of the workshop was to leverage advances from research conducted in support of REACH and the Australian risk assessment approaches by providing regulators with methods and processes to incorporate bioavailability, normalize toxicity thresholds, address food-web issues, and incorporate background concentrations. These recent major terrestrial research projects have significantly advanced our understanding of the behaviour and toxicity of metals in soils. Large data sets were developed that are useful for risk assessment of metals in soil environments, and were used by workshop participants as case studies in the development of the ecological standards for soils. Manuscripts from the workshop discussed bioavailability adjustments based on pH, cation exchange capacity, and organic carbon; application of leaching and aging factors; and consideration of the source and form of metals in the soil. Incorporation of soil microbial processes and a path forward for wildlife toxicity reference dose and food-chain modeling were also discussed in prepared manuscripts. In addition, one of the workgroups described the processes needed to gain regulatory acceptance as a directive or guidance by North American state and federal governments. 297 Simulation of age-dependent survival of Daphnia magna following time-varying exposure of pentachlorophenol and dimethoate Y. Wang, C.D. Brown, University of York / Environment Department. Toxicokinetic-toxicodynamic models (TKTD models) have been recommended as a promising tool to simulate the time-course of toxic effects at organism level and even beyond test conditions over time. Wider use is partially restricted at present because there has not been extensive application of the approach to the key risk assessment species. This study consisted of TK and TD experiments for pentachlorophenol and dimethoate. Each experiment included three independent tests which were started with different age groups of Daphnia magna (i.e. less than 24-hour old neonates, 7-day old juveniles, and 14-day old adults). The dry weight of Daphnia was used as a measure of the growth of the organism. A TKTD model was established to simulate effects of time-varying exposure with correction for growth dilution. This is the first detailed study of using a TKTD modelling approach to simulate the effect of time-varying exposure of pesticides on the age-dependent survival of Daphnia magna. This study demonstrated that: i) growth greatly influenced uptake and elimination of the two compounds, causing a large reduction in BCF for greater than 1 order of magnitude between neonates and adults; ii) sensitivity of Daphnia to the two compounds varied by up to a factor of 2 between different life stages with LC s increasing and decreasing with age for 50 pentachlorophenol and dimethoate, respectively; iii) the TKTD model gave a good simulation of the survival of Daphnia both for the TD experiments and an independent validation test with long-term timevarying exposure. TKTD models do not currently account for changes in sensitivity between life stages, so this needs to be incorporated to develop fully-functional models for use in chronic risk assessment. Key words: time-varying exposure, TKTD models, model evaluation 298 A method to predict and understand fish survival under dynamic chemical stress using standard ecotoxicity dataR. Ashauer, University of York / Environment; P. Thorbek, Syngenta / Environmetal Safety; J. Warinton, Syngenta; J.R. Wheeler, Syngenta Ltd; S. Maund, Sygenta Crop Protection AG. We present a method to predict survival of fish under exposure to fluctuating concentrations and

75

repeated pulses of a chemical stressor. The method is based on toxicokinetic-toxicodynamic (TKTD) modelling using the General Unified Threshold model of Survival (GUTS) and calibrated using the raw data from standard fish acute toxicity tests. The model was validated by predicting fry survival in a fish early life stage test. Application of the model was demonstrated by using surface water exposure patterns (FOCUS-SW) as model input and predicting the survival of fish over 485 days. Exposure patterns were also multiplied with a factor of five and ten to achieve higher exposure concentrations for fish survival predictions. Further we quantified how far the exposure profiles were below the onset of mortality by finding the corresponding exposure multiplication factor for each scenario. We calculated organism recovery times as additional characteristic of toxicity as well as number of peaks, interval length between peaks and mean duration as additional characteristics of the exposure pattern. We also calculated which of the exposure patterns had the smallest and largest inherent potential toxicity, respectively. Sensitivity of the model to parameter changes depends on the exposure pattern and also differs between GUTS-IT (individual tolerance) and GUTS-SD (stochastic death). Possible uses of the additional information gained from modelling to inform risk assessment are discussed. 299 Role of ecological modelling in soil risk assessment: Folsomia candida model supports the definition of eco(toxico)logically relevant depths F. Scherr, Bayer CropScience AG; K. Hammel, Bayer CropScience AG / Environmental Safety; G. Ernst, Bayer CropScience / Ecotoxicology; V. Roeben, RWTH Aachen University Institute for Environmental Research / Environmental Safety; G. Goerlitz, Bayer CropScience AG / Environmental Safety. After application of plant protection product (PPP) to a field crop, soil organisms are exposed to a pronounced concentration gradient. Conversely in laboratory studies soil organisms are exposed to a homogeneous distribution of the test substance in the medium. In a typical risk assessment (RA) the effect concentrations observed in the laboratory test are then compared with a predicted environmental concentration (PEC) averaged over the so called eco(toxico)logically relevant soil depth ERD. In a recent opinion the EFSA PPR panel has proposed scenarios and models to calculate PEC in soil as a function of time and soil depth. While EFSA experts also suggest to base risk assessment on species dependent ERDs, e.g. 1 cm for collembolans, little is actually known on the vertical distribution of soil organisms and how these interact with a realistic concentration profile of a PPP. Consequently an imbalance exists and the refinement brought into the RA on the exposure side is not yet matched by refinements on the effect side, e.g. the derivation of the ERDs for soil inhabitants. In this study we employ an individual based, ecological model of F. candida to illustrate how biological variations of soil inhabitants and environmental parameters as well as the distribution of the pesticide in the soil profile can influence the effects on a population of soil organisms. Furthermore we propose how the ecological model may be used in adequate determination of ERDs for soil risk assessment. Results from a range of compounds on the reproduction performance of F. candida s under different enviornmental conditions were gathered. The results were evaluated to derive the appropriate ERDs by comparing the effects of relaistic concentration profiles with homogeneous concentration profiles. The model performance was validate by simulating the findings of the TME study. Results show the importance of environmental variations for the vertical distribution of F. candida. Calculations with a range of compounds clearly show that realistic concentration profiles are relevant in order to define the appropriate ERD, which deviates from the 1 - 5 cm currently proposed. The results from the present study show that ecological modelling has the potential to support the definitions of risk assessment schemes as proposed by EFSA. The presented ecological model helps to derive an appropriate ERD for tier 1 in the area of soil risk assessment. 300 Soil organisms exposed to plant protection products – analysis of exposure and effects over different soil layers A. Toschki; U. Hommen, M. Klein, Fraunhofer IME; W. Koenig, S. Pieper, Federal

SETAC Europe 23rd Annual Meeting

Environment Agency; C. Possberg, Institute for Environmental Research RWTH; J. Roembke, ECT Oecotoxikology; M. Ross-Nickoll, A. Schaeffer, Institute for Environmental Research RWTH; A. Scheffczyk, ECT Oecotoxikology; B. Schmidt, B. Scholz-Starke, Institute for Environmental Research RWTH; M. Hammers-Wirtz, gaiac, Research Institute for Ecosystem Analysis and Assessment. It has been questioned in the last years whether the risk assessment of Plant Protection Products (PPP) in soil addresses properly the relationship between pesticide exposure and effects on soil organisms, especially with respect to the time distribution of the PPP over time in relation to the distribution of different species in the soil layer. To understand this relationship, a study with outdoor Terrestrial Model Ecosystems (TME) using intact soil cores was conducted, analysing the distribution of two pesticides and their effects on soil organisms (collembolans, oribatid mites, enchytraeids, earthworms) at the same time. Two insecticides were tested in 68 TME (Ø 467 mm, height 400 mm), in two concentrations each; untreated TME served as controls. Different soil layers (0-2.5 cm, 2.5-5 cm, 5-10 cm, 10-20 cm) were analysed separately for PPP concentrations and species abundance and community composition over 12 months. The results of chemical analyses for the different soil layers show how the pesticides move through the soil column. The abundance of soil organisms in the different soil layers over time will be given for the different treatments and concentrations. Univariate statistical analyses for single species as well as multivariate analyses were used to assess effects of the investigated PPP on the soil organism community. The results of this experiment allows a direct linkage of exposure and effects on soil organisms under realistic environmental conditions. The relevance of the results for refinement of exposure calculation and risk assessment for soil organisms will be discussed. 301 Development of archetype exposure scenarios for use in risk assessment in Asia O. Price, Unilever / Colworth Science Park; A. Franco, Unilever; O. Jolliet, University of Michigan / School of Public Health; P. van den Brink, AlterraWageningen UR / (a) Aquatic Ecology and Water Quality Management Group; (b) Alterra; C.M. Holmes, Waterborne Environmental Inc. Environmental risk assessment frameworks used in Europe for regulatory purposes are protective of ecosystems but lack environmental realism. Such assessments fail to address the ecological consequences of environmental contaminants on ecosystem structure and function via the simplistic use of default parameters and assessment factors. They fail to provide a mechanistic framework to assess ecological consequences at high hierarchical levels of ecological organisation (e.g. communities), which makes the usefulness of molecular approaches (e.g. as early warning systems) difficult to incorporate within risk-based decision making frameworks. There is a clear need to develop mechanistic exposure and ecological scenarios to enable risk assessors to ascertain spatial and temporal ecological consequences of single and multiple stressors (including chemical mixtures) on ecosystem health. Our research programme aims at developing archetype scenarios for use in risk assessment, including the development of models to: provide spatially resolved emission estimates, predict the fate and elimination of chemicals in wastewater treatment plants, predict chemicals transport and distribution using multiscale multimedia fate models, and link them with water quality models. The ultimate aims are to integrate these exposure models within a single platform to (1) assess spatially and temporally resolved environmental concentrations, (2) compare the fate of these stressors in multiple stressed environments, (3) develop archetype scenarios for use in prospective risk assessment and (4) integrate ecological effects models. This presentation gives an overview of the project, an update on each model and initial results on their integration within a single platform, currently applied to aid the identification of relevant exposure scenarios in Asia. We present the initial results of the integration of these models into one platform to assess source to receptor impacts for two test chemicals that are widely used in home and personal care products across Asia. We present the initial development of a suite of exposure scenarios that try to capture the variability of scenarios across Asia for use in a prospective risk assessment framework. The exposure

76

models developed to date still need further refinement and there is a need to update them to better address chemicals with a wide range of physicochemical properties (e.g. ionisables) to better assess site-specific bioavailability. 302 Acute risk of organic compounds in European river basins E. Malaj, Helmholtz Centre for Environmental ResearchUFZ / Department of Effect-Directed Analysis; P.C. Von der Ohe, UFZ Helmholtz Centre for Environmental Research; M. Grote, EDF RD / National Hydraulics and Environment Laboratory; R. Schaefer, University Koblenz-Landau. Compliance with environmental quality standards for 33 priority substances is required from Water Framework Directive to achieve at least a good chemical status of European water bodies, while all other compounds which could affect freshwater ecosystems should be identified as river basin specific pollutants. Our aim was to estimate the ecological risk for freshwater ecosystems from organic compounds on the European level, as well as to identify the chemicals of highest concern. We evaluated a total of 5208 sites distributed over 29 countries and 38 major European river basins, for a total of 239 compounds. Potential risk was assessed for the green algae Pseudokirchneriella subcapitata, the invertebrate Daphnia magna, and the fish Pimephales promelas using experimental and predicted acute toxicity data. Phthalates, PAHs and BFRs were the chemical groups most frequently detected, while pesticides were the most toxic group for the standard test organisms. Furthermore, 56% of the most toxic compounds (eg. dichlorovos or acenaphthene) were not listed as priority pollutants. We found that the ecological risk depends on the number of toxic compounds present in the monitored sites. Although potentially underestimated, organic compounds, especially pesticides, pose a threat to the ecological status for a high proportion of monitored sites in Europe, where species at 1/6 of these sites are likely to be acutely affected. To our knowledge, this analysis represents the most thorough study on the spatial distribution of the ecological risks from toxicants on the continental scale. 303 Multidisciplinary research for the design of “green” plasticizers V.V. Yargeau, McGill University / Chemical Engineering; M. Maric, R.L. Leask, R. Bernard, McGill University. Phthalates are high production volume chemicals, used to make poly vinyl chloride flexible and easy to process. The phthalates in many consumer products can leach out into the environment. Consequently, human exposure to phthalates is widespread. Animal studies have shown that phthalates act as endocrine disruptors at environmentally relevant concentrations; in utero exposures result in abnormal development of the male reproductive system. Epidemiological studies have reported an association between phthalate exposure during pregnancy and an elevated incidence of hypospadias in the offspring. Plasticizers are needed for the production of an immense range of consumer products. The challenge is to develop “green” plasticizers to replace phthalates in poly vinyl chloride. Green plasticizers should not be toxic and, ideally, should be biodegradable so that they do not accumulate in the environment. We have established a multidiscipline-based methodology for the design, synthesis and testing of alternative plasticizers. The chemical engineers on the multidisciplinary team have designed and synthesized potential plasticizers. They have tested many of these candidates for their mechanical and rheological properties and have elucidated the biodegradation pathway of many of these compounds. This has allowed us to identify lead compounds within series of compounds from different classes of plasticizers, such as dibenzoates, succinates and poly(caprolactone) diols. In parallel, the toxicologists have used cell-based assays to determine if our candidate plasticizers affect cells and future testing is currently on-going using animal studies. Members of the team are also investigating the ethical and legal implications of testing novel green chemicals prior to their release into the marketplace. This concerted research effort provides essential information with respect to the identification of safer new “green” replacement plasticizers and to knowledge translation strategies to make this information accessible to regulators, policy makers and the public.

SETAC Europe 23rd Annual Meeting

304 Comparative assessment of the toxicity of "Green" household products J.E. Weinstein, The Citadel / Department of Biology; J.A. Miller; A.D. Gray, Department of Biology. Although it is generally assumed that green household products contain individual ingredients that are less toxic and/or more degradable than conventional formulations, little research on this topic has been conducted. This study examined the toxicity of several green household products—before and after degradation treatments—using either larvae of the estuarine daggerblade grass shrimp, Palaemonetes pugio, or adults of the freshwater cladoceran, Daphnia magna. Seven green product formulations (Seventh Generation Dishwashing Gel, Green Works AllPurpose Cleaner, Green Works Dish Detergent, Earth Options Insect Killer, Tom’s Mouthwash, Martha Stewart Bathroom Cleaner, and Seventh Generation Laundry Detergent) were compared to that of categorically equivalent products using 48-hour toxicity tests. Of the seven categories of household products tested, in only two cases were the green products less toxic than either of the conventional formulations. In three cases, there was no difference in the toxicity between the green products and the conventional formulations. And, in one case, the green product was more toxic than both tested conventional products. Following a biodegradation treatment involving activated sludge, none of the green products became less toxic. In five cases, the biodegradation treatment did not significantly alter product toxicity. And, in one case, the green product became more toxic after the biodegradation treatment. By contrast, the biodegradation treatment decreased toxicity of the conventional product formulations in eight cases, increased toxicity in five cases, and had no significant effect on toxicity in one case. Following a photodegradation treatment for a subset of these green products, in only one case did the green product become less toxic. Photodegradation did not significantly alter toxicity of the other three tested green products. This research demonstrates that green household product formulations are not necessarily less toxic and/or more degradable than their conventional counterparts. These results also suggest that the toxicity and degradability of end product formulations need to be considered in the overall framework for green product evaluation. 305 Designing non-PBT chemicals by QSARINS P. Gramatica, University of Insubria / QSAR Res. Unit Environ. Chem. Ecotox./Dep.Structural & Functional Biology; N. Chirico, University of Insubria / QSAR Research Unit/Department of Theoretical and Applied Sciences; E. Papa, QSAR Res. Unit Environ. Chem/Dep. Structural Functional Biology. The chemicals that are jointly Persistent, Bioaccumulative and Toxic (PBTs) are substances of very high concern (SVHC) and subject to an authorization step in the European REACH regulation, which includes plans for safer substitutions of recognized hazardous compounds. The limited availability of experimental data necessary for the hazard/risk assessment of chemicals and the expected high costs have increased the interest for alternative predictive in silico methods, such as Quantitative Structure–Activity (Property) Relationships (QSA(P)R). In the Green Chemistry approach the design of a safe molecule is the earliest phase of the long process of placement on the market new, safe substances. Molecular modeling approaches, such as QSAR, can be successfully applied in "safe Chemical Design". A structurally-based approach was proposed by our research group for a holistic screening of potential PBTs in the environment. The PBT Index, obtained by combining Persistence, bioconcentration and toxicity data into a Principal Component Analysis, was defined to rank compounds according to their cumulative PBT behaviour. A simple, robustand externally predictive QSPR model, based on fourmolecular descriptors, was developed for the PBT Index. This model is ahazard screening tool for the early identification and prioritizationof not yet known PBTs, only on the basis of their molecular structure, and is now implemented in the new software QSARINS. QSARINS was recently developed by our research group for the development and validation of MLR-OLS QSAR models. The PBT index model, implemented in QSARINS, can be applied to any chemical, without experimental data and not yet recognized as PBTs or even not yet synthesized, to verify the potential PBT behaviour, just starting from the chemical structure. The software

77

verifies, by the Insubria graph, if a new hypothesized compound is in or out the structural applicability domain of the PBT Index model, in order to be confident only on the interpolated predictions. Finally some case studies will be presented regarding how to screen the PBT potential behaviour of existing and new chemicals. This structure-based approach responds to two levels of action in relation to the management, according to REACH regulation, of chemicals of highest concern: a) the need for tools for identification and prioritization, and b) the design and production of safer alternatives, according to the green chemistry philosophy of "benign by design". 306 The Chemistry Scoring Index (CSI): A Hazard-Based Scoring and Ranking Tool for Chemicals and Products Used in the Oil and Gas Industry T.A. Verslycke, T. Bowers, K. Reid, Gradient; S. Thakali, URS; J. Sanders, D. Tuck, Halliburton Energy Services, Inc.. A large portfolio of chemicals and products is needed to meet the wide range of performance requirements of the oil and gas industry. Included within this portfolio are chemicals that exhibit a variety of potential hazards to human health, safety, and the environment. These hazards have historically been managed by the industry through various measures that effectively limit exposure to these chemicals. Nevertheless, the oil and gas industry is under increased scrutiny from regulators, environmental groups, the public, and other stakeholders for its use of chemicals. In response, industry is increasingly incorporating "greener" products and practices, but is struggling to define and quantify what exactly constitutes "green" in the absence of a universally-accepted definition. As one of the world's largest providers of services to the oil and gas industry, Halliburton Energy Services, Inc. recently developed a tool for scoring and ranking hazards of products in its portfolio. The Chemistry Scoring Index (CSI) is ultimately intended to be a globally implementable tool that comprehensively scores and ranks hazards to human health, safety, and the environment for products used in oil and gas operations. CSI scores are assigned to products designed for the same use (e.g., surfactants, catalysts) on the basis of product composition, intrinsic hazard properties, and data availability for each product component. As such, products with a lower CSI score within a product use group are considered to have a lower intrinsic hazard compared to other products within the same use group. The CSI provides a powerful tool to evaluate relative product hazards; to review and assess product portfolios; and to aid in the formulation of safer products. 307 Environmentally compatible flame retardants: The ENFIRO approach P. Leonards, VU University Institute for Environmental Studies / Chemistry & Biology; S. Brandsma, IVM institute for environmental studies / Faculteit der aard- en levenswetenschappen; H. Hendriks, Utrecht University; N. Jonkers, IVAM; J. Parsons, University of Amsterdam / IBED; S.L. Waaijers, University of AmsterdamIBED Institute; R. Westerink, Utrecht University; C. de Wit, Stockholm University; J. de Boer, VU University, Institute for Environmental Studies. Some brominated flame retardants (BFRs) have unintended negative effects on the environment and human health. Some of them show a strong bioaccumulation in aquatic and terrestrial food chains, some are very persistent, and some show serious toxicological effects such as endocrine disruption. During the last decade, an increasing number of reports have presented evidence of these negative effects caused by some BFRs. A number of BFRs (in particular polybrominated diphenyl ethers (PBDE’s), hexabromocyclododecane (HBCD) and tetrabromobisphenol-A (TBBP-A)) can be found in increasing concentrations in the human food chain, human tissues and breast milk. Less toxic alternatives appear to be available already but comprehensive information on their possible toxicological effects are lacking. The European Commission-funded project ENFIRO investigates the substitution options for some BFRs and compares the hazard, exposure, fire performance and application of the alternatives versus the BFRs. In addition, a risk assessment and comparative life cycle assessment was carried out. The current paper shows the main outcomes of ENFIRO and shows the ENFIRO approach “how to substitute chemicals which are environmentally compatible”. Information on the hazards, exposure,

SETAC Europe 23rd Annual Meeting

leaching behavior, emission to air, and life cycle assessment of the alternative halogen free flame retardants (HFFRs) compared to the BFRs are discussed. ENFIRO showed that viable alternative flame retardants are available. Some HFFRs show less risk for the environment and human health, and show similar fire performance and technical application capabilities as BFRs. The lower risk is mainly due to the lower hazards of the HFFRs, and probably not due to a lower exposure. In conclusion, for the substitution of chemicals a complete substitution cycle is needed: technical/application performances, hazard, exposure, and impact assessments. Such an assessment can only be performed with a group of experts from different disciplines (material experts, toxicologist, chemist, life-cycle experts etc). 308 What is the industry doing to develop safe flame retardants based on phosphorus, nitrogen and inorganic chemistries? A. Beard, Clariant GmbH / BU Additives - Flame Retardants; U. Wietschorke, WTC Consulting; K. Spriestersbach, . The environmental behaviour of flame retardants has been studied since the 1990ies. Mainly brominated flame retardants were found in many environmental compartments up to human milk, and some individual substance were found to have toxic properties. Therefore, alternatives have been demanded by authorities, NGOs and equipment manufacturers. The producers of alternatives have formed an industry association for phosphorus, inorganic and nitrogen flame retardants (pinfa) in order to better inform flame retardant users and the public about technical options, the properties of their products and develop flame retardants with an improved health and environmental profile. pinfa has run a pilot project with GreenScreen, an assessment tool for users of chemicals to determine the environmental and health profile of halogen free flame retardants. In addition, the pinfa participated in the EU-funded collaborative research project ENFIRO (www.enfiro.eu , EU research project FP7: 226563) which aimed at evaluating alternatives to the major established brominated flame retardants. In order to make the evaluation fully comprehensive, it was decided compare also material and fire performance as well as attempt a life cycle assessment of a reference product containing halogen free versus brominated flame retardants. About a dozen halogen free flame retardants were studied representing a large variety of applications, from engineering plastics, printed circuit boards, encapsulants to textile and intumescent coatings. A large group of the studied flame retardants were found to have a good environmental and health profile. For several years now, the US-EPA has been carrying out a series of projects related to the environmental assessment of alternative flame retardants, the “design for environment” projects on flame retardants for printed wiring boards and alternatives to decabromo diphenylethers and hexabromocyclododecane (HBCD) which are all expected to be finalized in 2013. pinfa members have actively engaged in the process and industry’s view on these hazard focussed assessments will be presented. 309 SimpleTreat evolution: applicability to ionisable chemicals A. Franco, T. Gouin, Unilever / Safety and Environmental Assurance Centre; O. Price, Unilever / Colworth Science Park; J. Struijs; D. van de Meent, RIVM / Institute of Wetland and Water Research. SimpleTreat is the sewage treatment plant (STP) model implemented in the EU framework for the environmental risk assessment of chemicals under different regulations(e.g. REACH regulation, Biocidal Products Directive). Although the previous version (SimpleTreat 3.1) was adapted to describe ionisation, the fate of organic ions was limited to the unbound aqueous phase, which seriously restricted the applicability domain. A recent model update aimed at enlarging (and defining) the applicability domain with respect to ionisable organics is presented here. SimpleTreat 3.2 includes improved estimates of the sludge-water partition coefficient normalized to organic carbon (K ) of monovalent OC acids and bases from the octanol-water partition coefficient (K ), the OW dissociation constant (pK ) and the pH. An evaluation study was carried a out with ten test chemicals, specifically selected to challenge the applicability domain and to cover a wide range of functionalities and physical-chemical properties. The comparison of modelling results with monitoring data for activated sludge STPs collected from the literature

78

lend confidence to the use of SimpleTreat 3.2 with respect to monovalent ionics and neutral organics. The accuracy of the new sludge K regressions is acceptable for monovalent acid, but is lower for OC bases, for which measured sludge K is highly recommended, until OC better estimation methods become available. It was also shown that the approach remains unsatisfactory for ionic surfactants and organic ligands, for which measured K is necessary. This may limit the OC applicability of SimpleTreat using a basic input dataset, as available from regulatory registration data. 310 Ionogenic organic compound sorption models: Advancing qualitative concepts for quantitative descriptionsA. Mackay, University of Connecticut / Environmental Engineering Program; D. Vasudevan, Bowdoin College / Department of Chemistry. Suitable models are not yet available to predict the sorption of ionogenic organic compounds to soils because they do not account for contributions of cation exchange, surface complexation and cation bridging sorption mechanisms. Two elements must be addressed to advance qualitative understanding of these mechanisms to quantitative models: (i) an ability to quantify the abundance or availability of appropriate receptor sites, and (ii) knowledge of the influence of ionogenic organic compound structure on such interactions. A promising approach that addresses both of these challenges is the use of mechanism-specific probe compounds with structure based sorption models. Sorption of mechanism-specific probes (e.g. for cation exchange and for surface complexation) provides a implicit measure of the site availability and baseline sorption. Structure-based sorption models for cation exchange and surface complexation, respectively, would enable probe sorption measures to be extrapolated to other compounds that can participate in those sorption mechanisms. Existing sorption studies highlight consistent trends with structure in the sorption of organic compounds by cation exchange or surface complexation; however, more extensive observations with consistent experimental conditions (pH, ionic strength, etc) are needed to develop mathematical tools to describe structure trends in ionogenic organic compound sorption.

paradigm. Predictive hazard measures that depend on the K to OW describe bioaccumulation potential or on the narcotic LC as a proxy 50 for toxic effect can miss substantial activity by these chemicals. Perfluorinated alkyl acids (PFAAs) are some of the best studied yet still quite poorly understood examples. Although PFOS is now included in the Stockholm Convention, the mechanisms behind the bioaccumulation of perfluorinated chemicals still require elucidation. This represents a worrisome lag in our understanding given that intensive industrial production of a wide variety of fluorine chemistries continues to increase. Here, we explore what is known about the properties of PFAAs and how they relate to observed bioaccumulation patterns in fish. We focus in particular on the acid dissociation constant, pK , and the a (measured) protein-water distribution coefficients for bovine serum albumin (K ). We then propose a new albumin-binding model in BSA-W rainbow trout, Oncorhynchus mykiss, and compare its performace in predicting BCF for C6-C12 PFAAs to a model based on equilibrium partitioning of only the neutral fraction. Our model evaluations indicate that the neutral fraction alone cannot account for PFAA bioavailability; some facilitation exists for the ions as well. More comprehensive studies are needed to measure K for PFAAs along a wide range of chain A lengths. More also needs to be known about the nature and distribution of PFAA-binding proteins in fish, particularly if we wish to investigate observed tissue distributions, which show marked preferences for some tissues (e.g. liver, kidneys) over others (adipose, muscle) and cannot be fully explained using our current knowledge. Our specific binding model holds promise for the prediction of PFAA bioconcentration and as a way to investigate gaps in our knowledge about these important chemicals.

313 Modelling the accumulation and internal tissue distribution of PCBs and their hydroxylated metabolites in polar bears (Ursus maritimus) J. Armitage, University of Toronto Scarborough / Department of Physical and Environmental Sciences; J.A. Arnot, ARC Arnot Research Consulting / Department of Physical Environmental Science; F. Wania, University of Toronto at Scarborough / Department of Physical & Environmental Sciences. It is well established that neutral 311 Improving risk assessment modeling of the soil sorption affinity hydrophobic organic contaminants such as polychlorinated biphenyls of cationic organic compounds s.T. droge, Utrecht University / (PCBs) exhibit the highest wet weight concentrations in adipose tissue IRAS; K. Goss, Eawag / AUC. Using a consistent experimental data set followed by other tissues with relatively high total lipid content. Recent for 70 bases, we examined how the soil properties and molecular biomonitoring studies have also shown that hydroxylated PCB structure of organic cations affects the sorption affinity to natural soil. metabolites (i.e., OH-PCBs) exhibit a different internal tissue Current risk assessment models are not suitable to predict the sorption distribution with higher wet weight concentrations in blood and liver affinity of organic cations because they are not sufficiently focused on compared to adipose. While interactions with plasma proteins may be an the ion-exchange sorption processes that occur in soil. Under fixed important factor explaining these observations, we suggest that another medium conditions (pH, ionic composition), we used a HPLC method to important factor to consider is that OH-PCBs are ionogenic (i.e., test whether specific sorption coefficients to natural organic matter dissociate in vivo and hence are present in a neutral and charged form). (NOM, micronized pahokee peat) and clay (illite) can be summed to The physical-chemical properties of hexachlorobenzene (HCB, neutral predict sorption coefficients to soil (eurosoils). From the two Eurosoils, organic) and pentachlorophenol (PCP, weak acid, pKa ~ 4.7) provide a we derived a sorption model based on specific sorption coefficients to useful analogy when considering PCBs and OH-PCBs. Here we apply a NOM and clay. The observed differences in ion-exchange affinities for recently developed Physiologically-Based Pharmacokinetic model organic cations are not explained by calculated differences in octanol(PBPK) for ionogenic organic chemicals (IOCs) to simultaneously water distribution coefficients (Kow), as recently proposed in risk model the accumulation and internal tissue distribution of a set of PCBs assessment procedures. We show that in “NOM rich / clay poor” soil the and their OH- metabolites. The model distinguishes between neutral sorption is dominated by NOM, whereas in the “NOM poor / clay rich” (storage) lipids, zwitterionic (neutral) phospholipids, plasma proteins soil, sorption to clay needs to be accounted for besides NOM, otherwise and other non-lipid organic matter (NLOM). As there is a relationship sorption is strongly underpredicted up to 3 orders of magnitude. between degree of chlorination and estimated pKa (as # of chlorines ?, pKa ?, degree of ionization at pH 7.4 ?), it is important to recognize that 312 Do we need to include protein interactions in risk assessment all OH-PCBs do not distribute internally in the same manner. Besides models? The case of perfluorinated alkyl acids. C.A. Ng, ETH whole blood, OH-PCBs exhibiting a high degree of ionization (i.e., Zurich / Institute for Chemical and Bioengeneering; K. Hungerbuehler, predominantly charged in vivo) tend to favour tissues with elevated ETH Zurich / Institute for Chemical and Bioengineering. phospholipid content such as the liver whereas OH-PCBs exhibiting a Environmental risk assessment as applied in the context of REACH was low degree of ionization (i.e., predominantly neutral) behave like the largely developed with neutral organic chemicals in mind, and is wellparent compound (and hence exhibit the highest wet weight suited to assess risk from traditional legacy chemicals (e.g. PCBs). concentrations in adipose tissue). Assumptions regarding However, new classes of chemicals with very different properties have biotransformation (Phase I and II processing) are also influential. While since gained a wider share of the market. Some, once regarded solely the current model application is focused on PCBs and OH-PCBs, the within specialized contexts (e.g. pharmaceuticals), are now recognized findings have implications for the distribution of the hydroxylated as important environmental contaminants. Ionic and ionogenic metabolites of other POPs such as OH-PBDEs (i.e., biotransformation chemicals are particularly ill-suited to the ‘neutral-centric’ risk assessment products of polybrominated diphenyl ethers).

SETAC Europe 23rd Annual Meeting

79

314 Recovery of populations and communities after pesticide contamination M. Kattwinkel, Helmholtz Centre for Environmental Research UFZ / System Analysis, Integrated Assessment and Modelling; M. Liess, UFZ Center for Environmental Research / Department of System Ecotoxicology; J. Roembke, ECT Oekotoxikologie GmbH. Assessing population recovery from toxicant effects plays a crucial role for the protection of biodiversity and ecosystems. However, little is known on the recovery potential of different taxa and the information is scattered in the scientific literature. To enable an informed assessment and to aid the revision of the Guidance Documents we address the following questions: (1) What scientific information is available on population and community recovery after contamination? (2) What are the main drivers of recovery that can be used for field level estimations based on laboratory or semifield experiments? (3) What is missing in the current procedures for an effective risk assessment? We evaluated to recovery potential of populations and communities after pesticide contamination by an expanded literature review. Information dealing mainly with experimental studies were retrieved for the following taxa groups: mammals, birds, soil invertebrates, non-target arthropods (NTAs), bees, soil microorganisms, reptiles, terrestrial plants, fish, aquatic invertebrates, amphibians, aquatic microorganisms, algae, and aquatic plants. Although the investigated literature was very divers and the possibility for direct comparisons was limited, some general conclusions are drawn: Based on the results for aquatic invertebrates, most taxa recover within five generation times. The absolute time of internal recovery depends strongly on the reproduction capacity. Migration from uncontaminated areas is a main driver for external recovery. In many studies including such re-colonization sources recovery occurred within one generation. If recovery from external sources is assumed, it has to be ensured that the magnitude of re-colonization is a realistic estimation in landscapes heavily influenced by agriculture. Environmental stress generally acts in addition or synergistically to pesticide stress and hence recovery has to be evaluated within the ecological context. This is especially true for endangered species that are under particular stress. Indirect effects based on competition and predation can play an important role on effect magnitude and recovery time. 315 Long-term incubation of Nereis virens in metal-spiked sediment: behavioural, biochemical, cellular and genotoxic responses. G. Watson, Institute of Marine Sciences; J. Pini, University of Portsmouth. Nereis virens is an ecologically and commercially important temperate polychaete of intertidal soft sediment and an ideal species to investigate long term effects of metals. Using a spiked sediment approach juveniles are being incubated for 9 months in environmentally relevant concentrations of sediment (5 kg per box) spiked with copper, zinc and copper and zinc together at low (copper: 70 -1 -1 -1 mg kg , zinc: 200 mg kg ; medium: copper 120 mg kg , zinc: 270 -1 -1 -1 mg kg and high: copper 575 mg kg , zinc: 1160 mg kg ) concentrations. These concentrations are based on an extensive sampling regime of sediment, pore water and worms from seven UK sites with different levels of copper and zinc contamination. Worms are being fed during the experiment, but maintained under ambient conditions (temperature and photoperiod) in a flow-through seawater system. Throughout the experimental period feeding rates and out-of-burrow activity levels will be recorded using video with boxes sacrificially sampled at 3, 6 and 9 months. Using BCR sequential extraction, labile (bioavailable) amounts of metals in the sediment will be assessed in addition to pore water and tissue concentrations. The effects of these metals will be measured using a suite of biochemical, cellular, genotoxic and molecular endpoints. Scope for growth and metabolic changes (e.g. lipid, protein and carbohydrates levels) will assess the trade-offs between growth and responses to metal toxicity. Metallothionein concentrations, acetylcholinesterase activity, lysosomal membrane stability, micronucleus and comet assays and expression levels of key genes will all be used to build a ‘global’ picture of the effects of these metals on this species. It is expected that this study will show for the first time the relationship between biomarkers at different levels of

SETAC Europe 23rd Annual Meeting

organisation and enable their relative sensitivity to be ascertained from a long term study. Ultimately this work will show how these individual responses directly link to the ecology and population of benthic polychaetes. 316 Statistical Analysis of Fish Early Lifestage Experiments J.W. Green, DuPont / Applied Statistics Group. OECD TG 210 for fish early lifestage experiments has been updated recently. As part of that update, many GLP studies of freshwater and marine species have been reanalyzed under proposed revised guidelines and extensive computer modeling of both NOEC and ECx approaches have been done based on those studies. Responses to be analyzed under TG210 are size (length and weight), egg hatching success, larval mortality and abnormalities, first and last day of hatch, and first and last day of swim-up. It is the purpose of this presentation to present the results of these analyses and computer modeling studies, including specific hypothesis tests and regression models. It will be shown that regression provides good estimates of EC20 and EC30 in 75-90% of studies for size, hatch, and larval mortality, dependent on study conditions, but fails in 10-25% of studies for each of those responses. EC10 estimates are less often reliable, especially for hatching and mortality. NOECs are shown to be protective in most studies for all responses except egg hatch and larval mortality, where the power to detect a 20% effect is less than 80% for the most variable species or data. Guidance will be given on statistical tests for each response and models for ECx estimation, as well as criteria by which to evaluate the quality of ECx estimates and NOEC determinations. 317 Can we use kinetic data to estimate chronic effect levels for exposure situations different from laboratory EDC testing?M. Teigeler, Fraunhofer IME / Ecotoxicology Dept.; C. Schaefers, FraunhoferInstitut / Department of Ecotoxicology. In our study, two fish life cycle tests were performed using an anti-estrogen, considering both flow-through and static peak exposure design. Effect threshold concentrations on population relevant endpoints of both test designs were compared. In a further step, we tried to derive a linkage between exposure and resulting effects using a kinetic approach. The calculation was based on data on lipophility and time weighted average concentrations taken from the static Fish Full Life Cycle (FFLC) test. The aim was to validate the relation of the exposure linked effect levels via kinetics. Two life cycle tests were performed using zebrafish (Danio rerio). As a test substance the selective estrogen receptor modulator Fulvestrant was chosen. In a static test three life stages of zebrafish (fertilized eggs, juveniles, spawning adults) were exposed simultaneously to a single substance peak. A second test was performed under continuous flow through design. For estimation of effect levels related to decreasing test substance concentrations we used a kinetic approach for which different assumptions were made. The fertilisation success was found to be the most sensitive population relevant endpoint in the static FFLC. Fertility was recorded daily, thus, data was available on effect duration and the timepoint of highest effect intensity. A daily uptake as well as the daily depuration was calculated considering the exponentially decreasing concentrations. The geometric mean of the substance concentrations in fish of each day in relation to the previous day were determined. The depuration constant k2 was calculated using data on lipophility of the substance. The uptake constant k1 was determined by iterative fitting of the concentration curve against the observed effect levels taken from the static FFLC and from a pre-test. Using the kinetic approach, a LOEC was estimated to be around 16 µg/L, the effect threshold was determined to be approx. 11 µg/L. Based on these findings, the concentration range for the flow through study could be outlined (40, 13, 4.0 and 1.3 µg/L). The kinetic approach can be considerred as a useful tool to assess effect levels after short term exposure based e.g time weighted average concentrations. A broad range of applications is conceivable. Beside the assessment of short term exposure with plant protection products, also the assessment of effects on fish passing short term peaks of substance e.g. from sewage treatment plant effluents, is possible.

80

318 Investigating population relevance of the effects observed in standard rat reproduction studiesS. Mastitsky, BASF / Crop Protection - Ecotoxicology; M. Ebeling, Environmental Safety; N. Kreling, BASF SE / Crop Protection - Ecotoxicology; P. Thorbek, Syngenta / Environmetal Safety; W. Schmitt, Bayer CropScience AG / Environmental Modelling. Wild mammal reproductive risk assessment is typically based on results from rat reproduction studies conducted for toxicological risk assessment. Since such toxicological studies are primarily conducted with the purpose of ensuring a high level of individual protection for humans, the NOAEL from these studies is sometimes based on effects on endpoints of questionable relevance for wild mammals, where the protection goal is the sustainability of populations whereas the fate of single individuals plays a minor role. In regulatory practice, it is particularly challenging to assess a threshold that should be considered as ecotoxicologically relevant for wild mammal reproductive risk assessment when it comes to effects on body weight of dams and/or offspring. In order to contribute to the development of scientifically sound decision criteria for such cases, we report an evaluation of regulatory guideline rat reproduction studies on a number of active substances of various classes and modes of action. Statistical analysis was conducted to identify systematic relationships between the body weight effects and reproductive performance parameters deemed of relevance for population sustainability. In addition to a better understanding of the population relevance of sublethal effects in general, the results of our study could also be helpful when developing and applying population models, such as those investigated during the MODELINK workshop. 319 Using individual based modeling to quantify the importance of sub-lethal effects on population level - a case study for Nitocra spinipes T. Bui, Institute for Environmental Research Biology / Chair of Environmental Biology and Chemistry; E. Lundstrom, Stockholm University / Dept. of Applied Environmental Science; M. Breitholtz, Department of applied environmental science; A. Schaeffer, RWTH Aachen University / Chair of Environmental Biology and Chemodynamics; T.G. Preuss, RWTH Aachen University / Institute for Environmental Research. In order to protect populations of organisms, identifying important endpoints for ecotoxicological risk assessment is necessary. Therefore, the relevance of sub-lethal effects on population level should be assessed. Current ecotoxicological tests usually measure on individual level and consider each endpoint separately. However, statistically significant effects observed on individuals or a group of individuals are not necessarily relevant on population level. This aspect is essential because the goal of ecotoxicological risk assessment is to protect populations and communities. Chemicals can also have combined effects, resulting in much difficulty to detect effects at population level in the laboratory or in the field. In order to assess these problems and study the relevance of sub-lethal endpoints, individual based population modeling was considered to extrapolate effects detected on individuals to population level. We used an individual based model for Nitocra spinipes, a harpacticoid copepod, to investigate several adverse effects in combination with food dependent life cycle parameters and intra-specific competition on food. Model simulations showed the inhibition of the nauplius development to have a stronger adverse effect on the survival of a population than an increased overall mortality, requiring higher food levels for the population to maintain the same chance of survival. A detailed analysis of the population dynamics and population structure revealed the reason for this to be (1) higher amplitudes within the population fluctuations and (2) a shift towards and higher percentage of nauplii in the population. Hence, this study demonstrated the importance and relevance of sub-lethal endpoints for the survival of a population based on modeling results. The potential of those endpoints on population level should therefore be considered in to risk assessment. 320 Regionalised Global Assessment of Health Impacts of NonToxic Air Pollutants based on TM5-FASST P. Preiss, econcept AG / Institute for Energy Economics and the Rational Use of Energy (IER); R. Van Dingenen, F. Dentener, Joint Research Centre - Ispra / Institute

SETAC Europe 23rd Annual Meeting

for Environment and Sustainability, Climate Change Unite. This paper presents new globally differentiated intake fractions (iF) and characterization factors (CF) based on TM5-FASST source-receptormatrices (SRM). The assessment of impact to human health includes effects of a) emission of primary particulate matter (PPM) and b) emission of the precursor pollutants NMVOC, NOx, NH and SO2 3 regarding the reactants ozone (i.e. so called ground level or photochemical ozone) and secondary inorganic aerosols (SIA). Human health damages need to be assessed in a spatially resolved context in order to increase the accuracy of impact assessment results [1]. Research on LCIA for transboundary pollutants have been reviewed for example by [2], suggesting that spatially differentiated fate modelling can be crucial. Based on results of chemical transport models (CTM) such as described in [3], [4], [5] impacts to human health have been expressed per unit of emission of the above mentioned pollutants already in the past. The present work aims to fill the gap of consistent global modelling by developing an approach to derive globally applicable and spatially explicit values. The advantages and disadvantages of global modelling based on a global CTM are explained by providing a comparison between model results with a different degree of spatial resolution and sophistication. A novel approach using the TM5-FASST model has been applied to derive globally applicable and spatially explicit impacts of the emission of the transboundary pollutants (PPM, SO , NOx, NH , NMVOC). In this context, the TM5-FASST model 2 3 allowed to assess the whole globe as source and receptor area. This is a clear advantage of global models compared to only regional (continental) models. Moreover, sensitivity assessment regarding assumptions of the effects of different qualities of particulate matter and region specific sensitivity of receptors are possible. For the application within LCA it is necessary to adjust the results to archetypal conditions. It is desirable to rerun the global CTM models in order to better reflect the demands of LCIA. These are for example, a higher resolution of receptor areas, smaller marginal change of emissions, industry and sector specific model runs, etc.. 321 Fate process modelling in LCI: improving inventory quality or double counting? T.J. Dijkman, Technical University of Denmark / DTU Management Engineering; M. Birkved, M.Z. Hauschild, Technical University of Denmark. In order to reliably determine environmental impacts, LCIA models require good inventories of flows to and from the technosphere as well as accurate characterization factors. Calculations of characterization factors require, amongst others, an understanding of the fate of chemicals that are emitted to the environment. However, fate processes not only occur in the ecosphere. Before emission, i.e. in the technosphere, chemicals can undergo a variety of fate processes. These processes determine a chemical’s pathway before it crosses the border between technosphere and ecosphere, and should therefore be considered in the LCI. The aim of this presentation is to demonstrate the relevance of fate modelling in LCI and the necessity of choosing appropriate technosphere boundaries, using pesticide emissions as an example. Pesticide emissions from agricultural land were modelled using 2 approaches, after which freshwater ecotoxicity and human toxicity impacts were calculated using USEtox characterization factors. The first approach uses PestLCI, a LCI model to calculate pesticide emissions to air, surface water and groundwater applicable under European circumstances. PestLCI defines the technosphere as the field up to 1 meter of depth and the air column above it, up to 100 meters above the soil. Only when a pesticide crosses these system boundaries it is considered an emission. Within this technosphere a number of processes is modelled. Some of these processes (for example volatilization and runoff) lead to emissions to the environment. Other processes (such as degradation) result in removal of the pesticide from the system. Current LCI practice, the second approach, usually ignores these processes. It is assumed that all pesticide is emitted to soil. The pesticide hence enters the ecosphere upon application. Comparing the toxicity impacts found using the PestLCI approach shows a clear difference with the currently prevalent approach, with the current approach being more conservative. The observed variation can be explained by the difference in system boundaries on the one hand and

81

the difference in emission routes and their characterization factors on the other hand. Concluding, fate modelling is a useful tool to improve the inventory quality, thus reducing uncertainty. Moreover, in the case of PestLCI, emissions can be spatially differentiated. Combined with spatially resolved characterization factors, LCIA can reach a higher degree of accuracy. 322 Spatial differentiation for toxic emissions in LCA: How well the (nested) USEtox model mimic a truly spatially differentiated model? A. Kounina, Quantis EPFL; M. Margni, Ecole Polytechnique de Montreal / Department of Mathematical and Industrial Engineering; S. Shaked, University of Michigan / Applied Physics and School of Public Health; C. Bulle, CIRAIG Polytechnique Montreal / Chemical Engineering; O. Jolliet, University of Michigan / School of Public Health. Multimedia and multi-pathways exposure models have been recognized being suited characterization models to generate characterization factors for ecotoxicity and human toxicity impact categories in LCA. Given that human and eco-toxicity impacts are highly influenced by the emission location, recent multimedia and multipathway model developed spatially differentiated capabilities to reduce model uncertainty and improve accuracy and therefore confidence in LCA results. This work aims at selecting an appropriate model and spatial resolution to evaluate freshwater ecotoxicity and human toxicity impact through freshwater pathways. We evaluated inter-continental variation by comparing a developed nested USEtox model based on the sub-continental versus a fully connected IMPACT World model. Variation of fate and intake fraction represents up to 2 orders of magnitude among the 17 zones assessed with IMPACT World and USEtox model. Difference in continent specific landscape and population parameters is a key step to improve model relevance. Differences between the two models are explained by choices in model algorithms, e.g., modeling of freshwater outflow and volatilization algorithm. The model architecture of the surrounding of the continental box generally does not have a significant influence on results with the exception of volatile and persistent pollutant both in air and water. Intracontinental variation was analyzed on a finer resolution with two consistent versions of the same model: IMPACT Europe single zone model adopting the typical multimedia approach without spatial distinction, and IMPACT Europe spatial model accounting for spatial differentiation through 135 land zones (watersheds). Results showed that a-spatial models might overestimate the chemical fate and characterization factors for freshwater ecotoxicity up to a factor 5 when compared to a spatially differentiated model for unknown emission location. When the emission location is known, a spatially differentiated model can improve the model accuracy about 2-3 orders of magnitude in the case of IMPACT Europe model, because of its ability to accurately predict the freshwater residence time to the sea depending on the emission location. We then demonstrated that a developed archetype model based on water residence time to the sea can mimic aquatic fate and intake fractions results with up to a factor 5 of difference with results of the spatial model, whilst minimizing model sophistication. 323 Development of characterization factors for metals in 7 EU water archetypes Y. Dong, Section of Quanlitative Sustainability Assessment, DTU Management Engineering; M.Z. Hauschild, Technical University of Denmark / Department of Management Engineering. Toxicity potential of most metals in the freshwater are estimated in current life cycle impact assessment (LCIA) models without taking the spatial differentiated speciation behavior of the metals into consideration. Using a novel approach developed by Gandhi and Diamond (Gandhi et al. 2010), new characterization factors (CF) representing freshwater ecotoxicity potentials are calculated for metals (e.g. Cr, Be and Ba) in 7 EU water types, taking into account the influence of speciation behavior on metal bioavailability and metal fate in seven different EU water types.\nUSEtox is used to model the fate of the metals, WHAM 7.0 is used to model the metal speciation, Kd values and bioavailability, while the Free Ion Activity Model (FIAM) is used to model the ecotoxicity effect. The resulting archetype-specific CFs show up to ~4 orders of magnitude difference for Cr and Be. This indicates

SETAC Europe 23rd Annual Meeting

that the toxicity potential of these two metals is strongly dependent on differences in water chemistry. In comparison, Ba shows a constant bioavailability ratio and toxicity effect across the modeled water chemistries. Thus CFs are strongly correlated with fate, which results in a more narrow range of CFs. The differences in water chemistry not only changes the absolute values of the CFs for the individual metals, but also their ranking in terms of freshwater ecotoxicity potential, illustrating the relevance of taking water chemistry into account when modeling metal ecotoxicity potential in LCIA. In order to support LCIA in the frequent situation where no information is available of the specific water type into which the metal emission occurs, site generic average factors are also calculated and different approaches to averaging across archetypes are investigated and discussed. Reference:\nGandhi, N., M. L. Diamond, D. van de Meent, M. A. J. Huijbregts, Wjgm Peijnenburg, and J. Guinee. 2010. New Method for Calculating Comparative Toxicity Potential of Cationic Metals in Freshwater: Application to Copper, Nickel, and Zinc. Environmental Science & Technology 44 (13):51955201. Key words: metal, characterization factor, life cycle impact assessment (LCIA), ecotoxicity 324 Considering temporal variability for the characterization of metals aquatic ecotoxicity impacts in LCA F. Lebailly, CIRAIG, Ecole Polytechnique de Montreal; A. Levasseur, CIRAIG École Polytechnique de Montréal / Chemical Engineering; R. Samson, Ecole Polytechnique de Montreal; L. Deschenes, Ecole Polytechnique de Montreal / Genie Chimique. LCA is a recognized tool to assess potential environmental impacts of products or services, and researchers are undergoing efforts to reduce the uncertainty of outcomes. Temporal variability is a major source of uncertainty associated to both inventory and impact characterization phases. The ecotoxicity impact category is particularly sensitive to temporal aspects because characterization deals with contaminants behaviour in the environment that depends on dynamic processes. Moreover, impacts are integrated over an infinite time horizon so that long lived pollutants such as metals have very high characterization factors (CFs) compared to others. Time-related issues regarding metals toxicity are often raised and CFs are sometimes qualified as “interim” factors. In this paper, dynamic CFs are developed in order to assess aquatic ecotoxicity impacts of metals and point out relevant temporal information. The method following a dynamic LCA approach provides the distribution of impacts over time and is based on inventory temporal disaggregation and time-dependent characterization. CFs have been generated introducing time in fate calculation so that metal mass in freshwater is known for each time step. All data are provided by the USEtox model. Both bioavailability and metal forms of emission can be integrated in the calculations. Two types of characterization factors are obtained. While instantaneous CFs provide the impact of metals at the time “t” following a pulse emission, cumulative CFs provide the total time-integrated impact since the pulse emission occurred. Results demonstrate that CFs strongly depends on the metal itself and on the emission compartment as these two variables determine maximal mass and also persistence time in freshwater compartment. Cumulative CFs converge to the traditional steady state values of USEtox for each metal but time horizon seems to be an important issue as metals can require thousand of years to produce their total impact. The use of traditional values might be questionned as they correspond to extended time perspectives that are not necessarly relevant in the context of LCA and differ from one impact category to another. Relative impact of metals compared to organic substances should also decrease as it was expected. Dynamic characterization, coupled to speciation and temporal inventory, improves interpretation and has an important influence on decision-holder conclusions. 325 How policy makers use TMF N. Eckbo; B. Nordboe, Norwegian Climate and Pollution Agency. The aim of the presentation is to show how policy makers use trophic magnification factors (TMFs) in risk assessment. In addition, the talk will address uncertainties when using TMF in monitoring programs, and our needs in future development of the TMF. For policy makers, TMFs is a useful tool for assessing the biomagnification potential of different environmental

82

pollutants. The Climate and Pollution Agency (Klif) is responsible for monitoring contaminants in the Norwegian environment. The primary goal is to get an overview of the state of the environment, in relation to Norwegian environmental goals and international obligations. Recently, biomagnification factors were included in some of the monitoring programs as a part of the risk assessment of contaminants, e.g. the monitoring of Norwegian lakes and the Norwegian coast line. Three new monitoring programs will also include TMFs. The monitoring results are used in national and international environmental management negotiations. By comparing the TMFs of well-known chemicals with TMFs of new contaminants with unknown chemical properties, the TMF may be a tool to identify B-properties of emerging contaminants. Our monitoring also results in time trends of TMFs, hence providing knowledge on how TMFs may change over time. 326 Eutrophication may lead to lower mercury concentrations in aquatic food webs of East African lakes A. Poste, Norwegian Institute for Water Research; D.C. Muir, Environment Canada / Aquatic Ecosystem Protection Research Division; S. Guildford, R. Hecky, University of Minnesota-Duluth. Much of what we know about mercury in aquatic food webs is based on extensive study of temperate systems with low to intermediate primary productivity, and the behaviour of this contaminant in eutrophic tropical systems lacks the same level of understanding. The primary objectives of this study were twofold: 1) to characterize accumulation and biomagnification of Hg at 8 East African study sites; and 2) to explore Hg dynamics in hypereutrophic lakes where phytoplankton productivity is high year-round. The study sites ranged in size from crater lakes to East Africa's great lakes (e.g. Lakes Albert and Victoria) and ranged in productivity from mesotrophic to hypereutrophic. Comprehensive water, plankton and fish samples were collected between 2007-2009 and were analyzed for both THg as well as stable carbon and nitrogen isotopic ratios. We found that THg concentrations in fish were generally low, and THg trophic magnification factors (TMFs) ranged from 1.88 (in hypereutrophic Lake George) to 5.6 (in mesotrophic Lake Nkuruba). We found that TMFs were significantly higher in hypereutrophic lakes than in meso- and eutrophic lakes, and found that log(THg) vs. trophic level regression slopes were negatively related to chlorophyll a concentrations. Furthermore, our results indicated that phytoplankton accumulate relatively less THg per unit biomass in higher productivity systems. These observations suggest that growth and/or biomass dilution may act to reduce Hg concentrations at the food web base, while growth dilution at consumer trophic levels may act to reduce Hg biomagnification rates. THg concentrations in fish appeared to be determined, at least in part, by processes at the base of the food web, with year-round high phytoplankton biomass and growth rates reducing the potential for high THg in fish in these productive tropical lakes. 327 Controls on the Trophic Magnification Factor of Organic Chemicals in Aquatic Foodwebs M. MacLeod, ITM Stockhom University / Department of applied environmental science; J.A. Arnot, ARC Arnot Research Consulting / Department of Physical Environmental Science; K. Borga, Norwegian Institute for Water Research; M.S. McLachlan, Stockholm University. Measured trophic magnification factor (TMF) greater than 1 has been proposed as the toptier of a framework for assessing the bioaccumulation potential of chemicals. TMF greater than 1 will be observed when the average biomagnification factor (BMF) between trophic levels of a food web is greater than 1. Therefore, the TMF can be viewed as a field-measured BMF where the measurements have been conducted – and thereby averaged – across several trophic levels. We modeled BMFs using the Arnot & Gobas fish bioaccumulation model with the goal of identifying the processes that control BMF, and thus TMF, for different chemicals. Our modeling approach considered a broad range of combinations of log K from –2 to 12, and a range of rate constants for whole-body OW -1 biotransformation by fish of 0.00001 – 10000 day . Modeled BMFs and the dominant processes that remove chemicals from fish over the entire range of chemical properties were determined. The model identifies -1 chemicals with biotransformation rate constants less than 0.1 day and

SETAC Europe 23rd Annual Meeting

(4.5 < log K < 7.5) as having BMF > 1, and thus having the OW potential to have TMF >1. The nature of the process that controls BMF has implications for the variability in TMFs that should be expected in the field. Near the lower log K boundary gill elimination is the OW controlling process. At the upper boundary of biotransformation rate constant the potential variability between species and between systems is poorly characterized. Finally, at the upper boundary of log K there OW is clearly large variability between species and between aquatic systems in the growth rate of fish and other biota. Therefore, our modeling results indicate that we should expect high variability in TMF values determined in different systems for chemicals with log K above 7, OW with lower values in systems where growth rates are higher. We discuss the implications of these findings for the use of TMF in regulatory assessments of bioaccumulation potential conducted at national and international levels. 328 Trophic dilution of cyclic volatile methylsiloxanes (cVMS) in the pelagic marine food web of Tokyo Bay, Japan D.E. Powell, Dow Corning Corporation / Health & Environmental Sciences; N. Suganuma, Silicone Industry Association of Japan (SIAJ); S. Ushioka, Environmental Contol Center (ECC); K. Kobayashi, T. Nakamura, K. Ninomiya, M. Itai, Silicone Industry Association of Japan (SIAJ); K.B. Woodburn, Dow Corning Corporation / Health & Environmental Sciences. Trophic transfer of three cyclic volatile methylsiloxanes (cVMS) and two polychlorinated biphenyl (PCB) congeners was evaluated in the pelagic marine food web of Tokyo Bay, Japan. The cVMS materials were octamethylcyclotetrasiloxane (D4; CAS No. 55667-2), decamethylcyclopentasiloxane (D5; CAS No. 541-02-6) and dodecamethylcyclohexasiloxane (D6; CAS No. 540-97-6). The PCB congeners were 2,3,4,5,2’,4’,5’-heptachlorobiphenyl (PCB-180; CAS 35065-29-3) and 2,4,5,2’,4’,5’-hexachlorobiphenyl (PCB-153; CAS 3506527-1), which are "legacy" chemicals known to bioaccumulate in aquatic organisms and biomagnify in aquatic food webs. The PCB congeners were used as a benchmark chemical (PCB-180) to calibrate the food web and as a reference chemical (PCB-153) to validate the results. The interrelationship between BSAF and TMF was applied to the field data to minimize bias and uncertainty associated with the nitrogen-15 enrichment factor (used to estimate relative trophic level position), omnivorous feeding, and migration, and to correct for variable exposure across concentration gradients. The field data was used to derive other measures of bioaccumulation (e.g., BCF, BAF, BMF) and associated properties (e.g. concentrations in water, food web partition coefficients, water-sediment fugacity ratios) directly from the natural food web. When not benchmarked against PCB-180 or corrected for variable exposure across concentration gradients, the mean TMF for PCB-153 2 was 2.2 (95%CI=1.5 to 3.1) with a coefficient of correlation (r ) of 61%. When benchmarked against PCB-180 (TMF=4.0) and corrected for variable exposure, the mean TMF for PCB-153 was 3.8 (95%CI=2.6 2 2 to 6.1) with an r of 86%. The increase in r from 61% to 86% indicated that correction for variable exposure was appropriate. When benchmarked against PCB-180 (TMF=4.0) and corrected for variable exposure, mean TMF values for the cVMS materials ranged from 0.8 to 2 0.9 (95%CI=0.5 to 1.3) with r values ranging from 1.9% to 8.5%. In contrast to PCB-153, correction for variable exposure had minimal impact on TMF for cVMS. Probabilistic uncertainty analyses demonstrated with greater than 80% certainty that TMF values for cVMS were less than 1.0 for the pelagic food web. These results for Tokyo Bay are comparable to results observed for other pelagic and demersal food webs in marine and freshwaters, and further demonstrate that cVMS materials do not biomagnify in aqautic environments. 329 Revisiting TMFs for cVMS and legacy pollutants in Norwegian lakes K. Borga, E. Fjeld, Norwegian Institute for Water Research; A.H. Kierkegaard, M.S. McLachlan, Stockholm University. A recent study of the food web biomagnification of cyclic volatile methyl siloxanes in the pelagic food web of Lake Mjøsa, Norway, found surprisingly high levels of decamethylcyclopentasiloxane (D5 CAS no. 541?02?6) with trophic magnification factors (TMF) significantly higher than 1. There are presently few studies of food web magnification of

83

cVMS. The two other available studies of cVMS TMF have demonstrated a lack of food web biomagnification in benthic associated food webs, with TMFs below 1 To address this difference between studies, and to increase our understanding of the cVMS food web biomagnification, their potential sources and differences between lakes, the pelagic food web of Lake Mjøsa was revisited (resampled) JulySeptember 2012, in addition to a comparable lake (Lake Randsfjorden) and a reference site (Lake Femunden), far from any known sources of contaminants. Results from the present study support both the high concentration in Lake Mjøsa and food web biomagnification for D5 in pelagic food webs of Norwegian Lakes. The levels in the control lake were low and only measured in sediments, brown trout and one arctic charr, and almost all samples were below the limit of quantification except for a few of the brown trout that were just above. The factors affecting TMFs of D5 in the different lakes will be discussed along with comparison to benchmark chemicals such as PCBs, and potential sources of D5 to these ecosystems. 330 What is the cause of biomagnification of PCBs and organochlorine pesticides in the aquatic environment?M.T. Jonker, Utrecht University; S.A. van der Heijden, Institute for Risk Assessment Sciences / Institute for Risk Assessment Sciences. For decades, researchers have observed concentrations of hydrophobic organic chemicals (HOCs), such as polychlorinated biphenyls (PCBs) and organochlorine pesticides, to increase throughout food chains. The increase in concentrations of these chemicals with trophic level is referred to as “biomagnification” or “trophic magnification”. The phenomenon is not completely understood, but the general perception is that biomagnification is predominantly caused by slow elimination kinetics of hydrophobic chemicals that cannot keep up with the ongoing intake via food ingestion. In this presentation, this view will be challenged by showing that lipid-normalized bioaccumulation of a series of PCBs and organochlorine pesticides varies among different (nonmammalian) species by as much as almost one log unit and tends to increase with trophic level. Adjusting biomagnification factors (BMFs) from the literature for these different lipid affinities seems to level out the factors to unity (i.e., no or little increase in concentrations with trophic level). It is not the intention of this presentation to invalidate the existing paradigm of slow elimination kinetics, but we will discuss an alternative cause of biomagnification, based on partition experiments with different organism constituents (lipids, proteins). This hypothetical cause basically questions the existence of a true causative relationship between HOC concentrations and trophic position. 331 Food, the fish gap and multi-trophic aquaculture SAMS.

K. Black,

332 Alternative feeds for Atlantic salmon (Salmo salar) using novel plant-based meals and oils to replace fishmeal & oil G. Bell, Institute of Aquaculture University of Stirling Scotland. 333 Sea lice and their relevance to the sustainability of salmon aquaculture C. Todd, University of St Andrews / Biology. 334 Ensuring environmental sustainability of sea lice treatments J. McHenery, NOVARTIS Animal Vaccines. 335 Salmon culture as a source of POPs and metals to the Scottish marine environment? C. Robinson, Scottish Government. 336 Discussion Life Sciences. 337 Conclusions Life Sciences.

T.F. Fernandes, HeriotWatt University / School of T.F. Fernandes, HeriotWatt University / School of

338 Comparing the sensitivity of Lemna spp. to herbicides and plant growth inhibitors with that of non-standard macrophytes. A. Arenas, Wageningen University; M. Daam, DPPF/CEER; G. Arts,

SETAC Europe 23rd Annual Meeting

Alterra Wageningen University and Research Centre / Environmental Risk Assessment. There has been an increasing concern in the last few years regarding the representativeness of the standard macrophyte test species Lemna gibba and Lemna minor for the sensitivity of other aquatic macrophytes. Differences in life history,growth strategy, exposure route or sensitivity to chemicals with specific modes ofaction (MoA) could be some of the reasons for this. These differences are currently being discussed in the light of the revision of the Aquatic Guidance Document in Europe. This document recognizes that additional testing might be required on other macrophyte species for modes of action or exposure routes for which Lemna spp. is not sensitive. However, it is still unclear for which MoA additional macrophyte testing is needed and which are the most sensitive plant species or growth forms for these MoA. Our research aimed at addressing these issues in the context of the update of the Aquatic Guidance Document. Aquatic macrophyte EC50 values were selected for synthetic herbicides and plant growth inhibitors from the (US?EPA) ECOTOX database. Due to low data availability, the relative tolerance (Trel) approach was used to enable the comparison of macrophyte toxicity data grouped by MoA. Subsequently, Species Sensitivity Distributions were constructed to evaluate differences in sensitivity between Lemna spp. and other aquatic macrophytes. Algae toxicity data were also included in the final sensitivity analysis. Our results showed that the percentage (%) of species with a Trel value < 1 (i.e. greater sensitivity than Lemna spp.) was 66.78, 49.2, 45.5, 35.4 and 22 for synthetic auxins, ALS inhibitor (Metsulfuron methyl), microtubule inhibitors, PS II inhibitors and cell synthesis inhibitors, respectively. However, when a safety factor of 10 was applied to the toxicity value of Lemna, the % of unprotected species (i.e. Trel < 0.1) was 10.3 for synthetic auxins and below 5 for the other MoA. For the majority of the compounds, algal toxicity values were higher than those of Lemna spp. This indicates that aquatic macrophytes with higher sensitivity to herbicides than Lemna are not protected by algae testing. Implications for the risk assessment of herbicides and indications for future research are discussed. 339 Aquatic macrophytes – does their sensitivity depend on their systematic position (monocotyledons / dicotyledons) H. Christl, Tier Solutions GmbH. Many selective herbicides act specifically on certain terrestrial macrophytes, often either on monocotyledons (i.e. Liliopsida) or on dicotyledons (i.e. Magnoliopsida and Rosopsida). The current revised data requirements suggest to test further aquatic macrophytes also based on their systematic position. However there is indication in the literature that in contrast to terrestrial macrophytes the systematic position of aquatic macrophytes appears not to define their relative sensitivity. To verify or falsify these contradicting hypotheses, available macrophyte data collated by the AMDAP SSD group and evaluated in Giddings (2010) have been further analysed based on the taxonomic position of the assessed data. One problem of all analyses is that available parameters vary data sets (i.e. one study may have only assessed shoot length, another study provided several endpoints based further variables. The original SSD analysis always used the lowest endpoint of a given study / species. While this take reflects the prevailing regulatory approach (always use the lowest endpoint), the lower tail of these distributions is more affected by the type and number of parameters assessed (see above). A central toxicity point per species is considered to be more robust for a comparison of the relative sensitivity of different taxonomic groups. The outcomes of analyses either based on the lowest toxicity estimate or on the central points are presented and compared, and in addition individual SSD distributions for monocots and dicots are presented for several active substances, including the most extreme cases (with either monocots or dicots being more or less sensitive than the other). The main finding was that compared to the scatter of endpoints within a taxonomic group the differences between the two taxonomic groups were generally small. While this is only a preliminary evaluation that must be expanded once further aquatic macrophyte data are available for further active substances, the currently available information casts some doubt on the need of additional aquatic macrophyte testing based on the systematic

84

position of the terrestrial target plants. 340 Are threshold derived from specific species sensitivity distributions protective for benthic diatom assemblage? The case of a four herbicides mixture F. Larras; V. Gregorio, UNIL Anthropole; B. Montuelle, A. Bouchez, INRA; N. Chevre, Img / Faculty of Geosciences and Environment. Pollution of aquatic ecosystems is characterized by mixture of chemicals such as herbicides that are rejected by watershed runoff and urban discharge. Photosystem II inhibitors as atrazine, terbutryn, diuron and isoproturon are still found in European surface waters even if some are banned or restricted. These herbicides affect primary producers such as benthic diatoms which represent a considerable proportion of the fixed biomass. In the light of diatoms ecological significance and due to their sensitivity to herbicides, evaluating the risk of mixture of herbicides is therefore of great importance. Species sensitivity distributions (SSD) are used in risk assessment to derive protective thresholds (Hazardous Concentration, HC) from data obtained in laboratory. However, SSD are often based on model species because of the lack of data on environmental species. To assess if SSD based on environmental species are more adapted for herbicide mixture risk assessment, we compared the protectiveness of HC derived from generic SSD (G-SSD) built on literature sensitivity data, and specific SSD (S-SSD) built on the sensitivity of benthic diatoms typical of biofilm diversity in Lake Geneva. We built one SSSD and one G-SSD for atrazine, terbutryn, diuron and isoproturon. On G-SSDs, diatoms species were more resistant while chlorophytes and cyanobacteria were sensitive except for terbutryn. As a result, G-HC 5 thresholds were lower than S-HC . Then, we combined the 4 G-SSDs 5 together, the 4 S-SSDs together and concentration addition model (CA), which predicts mixture toxicity when components share the same mode of action, for assessing the risk of mixture. We obtained one G-SSD mix and one S-SSD for a quaternary mixture and we derived HC mix 5mix thresholds from each curve. In parallel, we exposed an artificial diatom community of 11 species to the same quaternary mixture to assess its sensitivity and compare it to both HC . G-HC over protected 5mix 5mix the artificial community while S-HC was under protective. 5mix Heterotrophic diatoms appeared more resistant in benthic bioassays, suggesting that other parameters, for example concurrence, may influence their sensitivity to PSII inhibitors when species grow in benthic community. In the light of our results, it seems essential to represent as possible the whole diversity of species targeted in S-SSD to prevent any underestimation of protective thresholds for herbicide risk assessment for phytobenthos. 341 Growth recovery of Lemna gibba and Lemna minor from a 7d diuron exposure M. Burns, Hortsys Research Unit, CIRAD;M.L. Hanson, University of Manitoba / Department of Environment and Geography; A. Crossan, I. Kennedy, Faculty of Agriculture and Environment, University of Sydney,. In agricultural catchments, aquatic ecosystems can experience periodic exposure to varying concentrations of pesticides. After pulsed exposure and possible adverse impact, populations within an ecosystem that are not extirpated may recover. This paper investigates the potential for recovery of two duckweed species (Lemna minor and L. gibba) following a laboratory exposure to the herbicide diuron for seven days. Recovery potential was assessed for exposed plants by transferring duckweed to clean media and monitoring over seven days. Population growth and biomass production were measured in both exposure and recovery treatments. There was significant inhibition from the exposure to diuron under the initial 7d -1 exposure treatment (EC = 59 and 52 µg L ; for L. minor and L. gibba 50 frond numbers, respectively). Both duckweed species were able to recover their growth from diuron inhibition levels that were not significantly different from the control treatments after seven days. Specifically, the greatest exposure concentrations at which recovery were observed for L. minor and L. gibba were 60 and 208 µg L-1, respectively. These species were able to recover from exposure concentrations that are typically deemed significant in ecological risk assessments for diuron. These results suggest that in the case of exposure to this herbicide, a level of ecosystem resilience is definable

SETAC Europe 23rd Annual Meeting

and should be considered in the assessment of ecological risk of diuron occurring in catchments. 342 Lines of evidence for establishing an aquatic level of concern for the herbicide atrazine in North American surface watersR. Brain, Syngenta Crop Protection Inc / Department of Environmental Risk Characterization; A.J. Hosmer, Syngenta Crop Protection Inc; D. Campbell, Syngenta Crop Protection LLC. Atrazine, a triazine herbicide registered primarily for the control of broadleaf weeds in corn and sorghum, is estimated to be the second most heavily applied herbicide in the U.S. (~35 million kgs annually). Approximately 75% of the field corn acreage grown in the U.S. is treated with atrazine, which has been detected in surface waters from watersheds with vulnerable landscape characteristics as indicated by the Atrazine Ecological Monitoring Project (AEMP). The United States Environmental Protection Agency (EPA) is in the process of evaluating an aquatic level of concern (LOC) for atrazine. The Agency’s approach for establishing an LOC involves calibrating a growth-based primary producer model using a set of ~40 micro/mesocosm (cosm) studies, which is highly sensitive to subjective binary cosm classification, delineated as ‘effect’ (scored as 1) and ‘no effect’ (scored as 0). A recent Scientific Advisory Panel (SAP) convened by the EPA indicated that most of the existing cosm studies with purported effects identified at atrazine concentrations less than 30 ?g/L have design weaknesses rendering interpretation of results and scoring difficult and subjective. An independent cosm evaluation, which was reviewed by the SAP and designated as appropriate, was used to derive a 60-d LOC of 25 ug/L using the EPA model. Multiple additional lines of evidence support an LOC of ?20 ug/L for 60 d, including a) the micro/mesocosm calibration dataset, b) species sensitivity distributions (SSDs) of the underlying laboratory toxicity data for primary producers used to populate the Agency’s model, c) rapid recovery of primary producers from atrazine exposure in singlespecies laboratory tests and evaluations of native in-stream communities, d) two years of biomonitoring of algal communities in the Midwestern U.S., demonstrating no observed direct effects attributable to atrazine on community dynamics (e.g. species richness, composition, diversity etc.). These lines of evidence will be presented to provide context for an aquatic LOC for atrazine. 343 The Effects of Mixtures of Herbicides in Use in Thailand on Lemna minor R. Tagun, Environment. The aim of this study was to test the interactions of herbicide mixtures atrazine, 2,4-D, alachlor and paraquat on Lemna minor. Two mixture exposure scenarios were investigated: exposure to mixtures in combination; and exposure to different pesticides in sequence. For the combined exposure studies, the effects of single compounds and a range of binary mixtures of 2,4-D and atrazine and alachlor and paraquat on the growth of L. minor. The resulting data were analysed to determine EC50 values for the single substances and to concentration combinations causing a 50% inhibition of growth. For sequential studies, four treatments was assessed: atrazine followed by 2,4-D; 2,4-D then atrazine; alachlor then paraquat; paraquat then alachlor. The interactions were assessed at a number of concentrations and a simple study design was adopted, where plants were exposed to the first pesticide for 3.5 d (50% of the study duration) and then removed and exposed to the second pesticide for the remainder of the study. In order to determine the nature of any interaction, observed effects of the sequential exposures to the different compounds were compared to predicted effects obtained using a simple model. For the chemical analysis, atrazine and 2, 4-D was analysed by HPLC while alachlor and paraquat were determined by ELISA test kits. For the binary mixture, atrazine and 2,4D indicates that these substances interact antagonistically. In contrast, alachlor and paraquat, the resulting isobole indicates a possible synergistic effect. In term of sequential exposure, when L. minor were exposed to atrazine and then 2,4-D the effect of the mixture was lower than predicted using the simple model. On the other hand, when plants were exposed to 2,4-D and then atrazine, an enhancement of the predicted toxicity was seen. When plants were exposed to paraquat followed by alachlor, toxicity was lower than predicted but the effects of exposure to alachlor then paraquat seemed to

85

be highly concentration dependent with some mixture concentrations causing enhanced effects and others reduced effects .These results show that interactions of different pesticides can occur when organisms are exposed to mixtures in combination or sequentially. The interactive effects vary according to the modes of action of the pesticides studied, the sequence of exposure and the concentrations tested. \n\n 344 Clotting time recovery and tissue residues following cessation of exposure to the anticoagulant rodenticide diphacinone in Eastern screech-owls B.A. Rattner, USGSPatuxent Wildlife Research Ctr / USGS; R.S. Lazarus, USGS Patuxent Wildlife Research Center / USGS; K.E. Horak; S.F. Volker, U.S. Department of Agriculture / Animal and Plant Health Inspeciton Service, National Wildlife Research Center; D.A. Goldade, N. Hoffman, U.S. Department of Agriculture / Animal and Plant Health Inspection Service, National Wildlife Research Center; J.J. Johnston, U.S. Department of Agriculture / Food Safety and Inspection Service. Anticoagulant rodenticides are used for the control of vertebrate pests in urban and suburban settings, agriculture and in island restoration projects. New regulatory restrictions have been placed on the use of some second-generation anticoagulant rodenticides (SGARs) in the United States, and in some situations this action may be offset by expanded use of first-generation compounds (FGARs). We have demonstrated that the FGAR diphacinone (DPN) evokes overt signs of intoxication and lethality in raptors at exposure doses that are 20 to 30 times lower than reported for traditionally used wildlife test species (mallard, Anas platyrhynchos and Northern bobwhite, Colinus virginianus). Sublethal exposure of American kestrels (Falco sparverius) and Eastern screech-owls (Megascops asio) resulted in prolonged clotting time, reduced hematocrit, and/or gross and histological evidence of hemorrhage at doses as low as 0.16 mg DPN/kg bwt/day. Our most recent study examined clotting time, hematocrit and DPN liver and kidney residues in owls fed a diet of 10 ppm DPN for up to 7 days followed by untreated diet for up to 21 days. By day 3 of DPN exposure, Russell’s viper venom time (RVVT) was prolonged, and by day 7 of DPN exposure, both RVVT and prothrombin time were prolonged and there was evidence of anemia in a few individuals. Upon termination of DPN exposure, coagulopathy and anemia were resolved to baseline values within 1 to 4 days. Surprisingly, DPN residues were consistently greater in kidney than in liver tissue (e.g., DPN on day 7 of exposure was 5.52 ?g/g ww kidney versus 0.96 ?g/g ww liver). Postexposure concentrations decreased rapidly within 24 hours; within 1 week liver and kidney values were < 0.3 ?g/g ww, and within 3 weeks values were < 0.1 ?g/g ww. The terminal phase half-lives of DPN in liver and kidney were 7.8 days and 4.7 days, respectively. Both FGAR and SGAR exposure monitoring of free-ranging raptors has principally utilized liver tissue, but the present findings suggest that future monitoring efforts should also quantify concentrations in kidney. These data are being used to develop an adverse outcome pathway for anticoagulant rodenticides in avian species. In addition, our findings demonstrate that low level dietary exposure to DPN can evoke toxicity in raptors in a matter of days, but once exposure is terminated, recovery can occur rapidly. 345 Does resistance to second generation anticoagulant rodenticides (SGARs) in rats increase secondary exposure of predators?R.F. Shore, CEH Lancaster; L.A. Walker, Centre for Ecology & Hydrology; A.D. Buckle, University of Reading; J.S. Chaplow, Centre for Ecology & Hydrology; L.J. Daniells; N.R. Llewellyn, M.G. Pereira, Centre for Ecology Hydrology; E. Potter, Centre for Ecology and Hydrology. Second-generation anticoagulant rodenticides (SGARs) are a key tool used in Britain and many other countries globally to control commensal rodents. However, use of SGARs is associated with widespread exposure of non-target wildlife through primary and secondary exposure. In Britain, long-term monitoring of SGAR exposure in a sentinel small mammal predator, the barn owl Tyto alba, has indicated an increase in secondary exposure to SGARs over the last 30 years. One potential reason is the spread of resistance in rats to difenacoum and bromadiolone. Barn owls generally only eat rats occasionally but are likely to be affected indirectly because resistance in rat populations

SETAC Europe 23rd Annual Meeting

leads to subsequent greater deployment of bromadiolone and difenacoum outdoors in often ineffectual attempts to control populations. This is likely to enhance exposure in non-target rodents that are then eaten by owls. We hypothesised that barn owls from counties in Britain where SGAR resistance [in rats] has been documented would experience greater exposure than owls from elsewhere. We quantified the extent of exposure to SGARs in barn owls that died between 2000 and 2011, reporting (i) the proportion of owls with detectable (? 0.025 ?g/g wet weight) liver SGAR residues and (ii) the magnitude of those liver residues. We found that barn owls were more likely to have detectable liver summed SGAR residues if they were from counties where resistance to SGARs in rats has been detected (Z = 2.36, P< 0.02). The magnitude of summed SGAR residues in birds with detectable residues was also higher in owls from counties where resistance has been detected (F = 5.21, P< 0.05). This is the first 1,184 study, as far as we are aware, to demonstrate that resistance in rats is associated with an increase in SGAR exposure, ands potentially effects, even in predators that rarely eat rats. Our results emphsise the need to manage resistance in rats and/or ineffectual use of SGARs in resistance areas to limit exposure of non-target wildlife. 346 Intentional and accidental poisoning of wild and domestic animals in Spain R. Mateo, UCLMCSIC / Instituto de Investigacion en Recursos Cinegeticos; I.S. Sanchez-Barbudo, P.R. Camarero, U.C.L.M.-C.S.I.C.. In this study we have analyzed of 1,157 suspected cases of poisoning of wild and domestic animals in the natural environment (1,800 animals and 340 baits) from different Spanish regions during the period 2004-2010. We detected 41.2% of positive cases (40.8% of animals and 52.6% of baits). In domestic carnivores, detection of toxic reached 71.4%, indicating its usefulness as sentinels of the use of poison in the environment. In those animals positive for toxicological analysis, 78.3% have been considered as intentional poisonings. The diurnal raptors were most affected by poisoning (43.6% of positives), followed by carnivorous mammals (27.1%). The most frequently detected toxicants were anticholinesterase insecticides (baits/animals: 80.4%/65.8%), followed by anticoagulant rodenticides (5%/19.6%), strychnine (2.2%/6.5%) and arsenic (4.5%/2.3%). The differences observed between regions underlines the dominance in the use of strychnine in Asturias, anticoagulant rodenticides in Castilla y Leon, organophosphate insecticides in Aragon, carbamate insecticides in Castilla-La Mancha and Madrid, and the emergence of other poisons, such as ?-chloralose or barbiturates, in Catalonia. In summary, 82.3% of intentional poisonings were due to anticholinesterase pesticides and 85.5% of accidental anticoagulant rodenticides. Future regulations of pesticides and biocides should take into account the risk of illegal use in the preparation of poisoned baits which involves the marketing of formulations with high richness of active ingredients with low LD . 50

347 Compliance with the ban of lead ammunition in a Mediterranean wetland, the Ebro deltaN. Vallverdu-Coll, Instituto de Investigacion en Recursos Cinegeticos; M.E. Ortiz-Santaliestra, J. Rodriguez-Estival, A. Lopez-Antia, M. Martinez-Haro, M.A. Taggart, Instituto de Investigación en Recursos Cinegéticos (IREC); R. Guitart, Universitat Autònoma de Barcelona; R. Mateo, Instituto de Investigación en Recursos Cinegéticos (IREC). The ingestion of lead (Pb) shot used for hunting continues being the main cause of Pb poisoning in waterfowl. In the Ebro delta (Spain), protected wetlands are surrounded by rice fields where waterbirds feed, and where Pb ammunition is still allowed. High Pb shot densities in sediments, and in turn high ingestion prevalence in waterfowl, have been detected. The use of Pb ammunition and the accumulation of Pb by birds after shot ingestion may pose a risk for human health due to consumption of contaminated meat. We assessed the degree of compliance with the ban on Pb ammunition in the Ebro delta wetlands, and studied the effect of the ban on the prevalence of Pb shot ingestion in waterbirds and on Pb levels in game meat. Waterfowl carcasses were collected and X-rayed from hunting bags (2007-2011) to determine the percentage of Pb and non-toxic embedded shot. Concentrations of Pb were analyzed in livers and muscles. In addition, gizzards were collected from hunting bags (2007-2012) and

86

examined to determine the percentage of Pb shot ingestion. During the first study season minimum hunter compliance, estimated as the percentage of waterbirds having only embedded steel shot (the non-toxic alternative) was 48.75%, while 26.88% of birds had only embedded Pb shot (minimum hunter noncompliance). These values changed in the subsequent seasons to 68.95% and 1.13%, respectively. The little compliance detected during the first study season led local administration to notify hunters that a total prohibition of hunting in protected wetlands would enter into force if the prohibition was not observed. Pb ingestion prevalence in 2007-2008 (28.6%) was not different from the pre-ban value (30.2%), but decreased significantly to values below 17.9% in the following seasons. Birds continue ingesting Pb shot at a relative high proportion, although their prohibition slowly contributes to reduce prevalence of ingestion. Pb muscle concentrations decreased significantly after the ban, in spite of which most species present individuals with Pb liver and muscle concentrations over the maximum safety limits. Whereas muscle Pb levels were determined by the presence of both ingested and embedded shot (all p< 0.001), liver levels largely depended on ingested shot (p< 0.001). Thus, besides restrictions in Pb ammunition use, additional mechanisms to reduce Pb ingestion prevalence in waterfowl are necessary to reduce risks for human consumers. 348 Environmentally relevant concentrations of an antidepressant alter physiology and behaviour in wild birds T. Bean, University of York / Environment; A. Boxall, University of York; J. Lane, The Food and Environment Research Agency; K. Herborn, University of Glasgow; K. Arnold, University of York. A growing and ageing population has led to increased pharmaceutical usage. Following ingestion, pharmaceuticals may not be completely metabolised so can be excreted to the wastewater system. Bird and bat species foraging on the invertebrates living on wastewater treatment works, and fields fertilised with sewage sludge, may therefore be exposed to a mixture of pharmaceuticals. Little is known about the risks to wildlife posed by environmentally relevant levels of pharmaceuticals. In this project, we investigated the effects of environmentally relevant concentrations of the antidepressant fluoxetine on wild caught starlings. Wild caught starlings housed in outdoor aviaries were allocated to either the treatment or the control group. To mimic foraging exposure, doses were injected into waxworms and hand fed to birds. We assayed relevant behaviours based on fluoxetine’s effects on humans: lethargy was measured from general activity levels of birds and changes in anxiety were assessed via two behavioural responses to standardised stressors (exploration in a novel environment and boldness following human disturbance). Behaviour was measured during 2 trials over 2 days. After 16 weeks of treatment, all birds were more active during trials than at baseline but treated birds were significantly more active than controls, contrary to predictions. Controls showed low levels of exploration in the novel environment in trial 1 which increased in trial 2 as they became habituated to the new cage. In contrast, fluoxetine treated birds showed high risk exploratory activity in both trials. Similarly post-treatment, all birds became bolder but fluoxetine treated birds were relatively bold in both trials and showed considerably less habituation over the two days than pre-treatment. Moreover, at the end of the experiment, moving to the test cage caused a rise in stress hormones of controls which dropped after 24 hours. However, for the fluoxetine treated birds, stress hormones were not as elevated following the move to the test cage and stayed low for 24 hours. Thus, ingestion of low concentrations of fluoxetine may induce a ‘therapeutic’ effect in birds apparently reducing their anxiety levels. Environmentally relevant levels of fluoxetine could potentially reduce stress responsiveness of starlings, increasing the risks posed by predation and other dangers. 349 Biomonitorization of birds under recovery: closing the gap between risk assessement and wildlife management C.S. Santos, University of Aveiro / department of Biology & CESAM; M.S. Monteiro, Aveiro University / Department of Biology; A.M. Soares, Universidade de Aveiro / Department of Biology and CESAM - Centre for Environmental and Marine Studies; S. Loureiro, Universidade de

SETAC Europe 23rd Annual Meeting

Aveiro / Biology. Birds, including waterbirds, have been used as bioindicators of environmental quality in a broad range of ecosystems. Amongst other attributes, their conspicuousness and sensibility to environmental contaminants makes them key species in environmental biomonitorization. Over the past years a significant part of the Portuguese biomonitoring studies has focused on organisms at lower trophic levels (e.g. crustacean and mollusks), but failed to address contaminants’ effects upon organisms at higher trophic levels such as mammals or birds. The present study aims were to: (i) assess the exposure of Portuguese birds to environmental contaminants, in particular birds accepted for rehabilitation in wildlife recovery centres, (ii) clarify if these factors could lead to birds’ illness and influence their recovery, and (iii) investigate the usefulness of ecotoxicological tools in the monitorization and recovery processes of birds. In order to address these issues, markers of neurotoxic and genotoxic exposure were assessed in aquatic birds from the orders Ciconiiformes and Charadriiformes. In the first part of this study, the analysis of neurotoxic markers, the reactivation of cholinesterase (ChE) activity on brain of the yellow-legged gull (Larus michahellis) was tested. In the second part of this work, the assessment of genotoxic effects, the frequency of micronucleus and other nuclear abnormalities was analyzed in erythrocytes of the white stork (Ciconia ciconia), grey heron (Ardea cinerea), purple heron (Ardea purpurea) and the little egret (Egretta garzetta). ChE activity in brain of L. michahellis was restored at a significant extent (increase in 50%), suggesting a previous exposure of these individuals to anti-ChE agents. High levels of genotoxic damage were also observed in the species of Ciconiiformes studied, with values varying significantly between years and geographical origins (p< 0.05). These results suggest that Portuguese birds might be exposed to different levels of environmental contamination and that this contamination may impair birds’ health. The use of ecotoxicological tools seems, therefore, to be a promising approach to monitor bird’s recovery as it will probably allow the screening for early physiological signs of toxicity, enabling a more insightful evaluation of birds’ health condition. Moreover, the use of these types of biomarkers may allow monitoring the potential rehabilitation of these birds. 350 Environmental media induced changes of surface functionalisation of two titanium dioxide nanomaterialsC. Nickel, Institute of Energy and Envionmental Technology eV IUTA / Air Quality & Sustainable Nanotechnology; B. Hellack, Institute of Energy and Environmental Technology e.V. - IUTA / Air Quality & Sustainable Nanotechnology; A. Nogowski, F. Babick, M. Stintz, Technical University Dresden / Institute of Process Engineering and Environmental Technology; T.A. Kuhlbusch, Institute of Energy and Environmental Technology e.V. - IUTA / Air Quality & Sustainable Nanotechnology. Titanium dioxide nanomaterials are already used in different products like in sunscreens. For their application in sunscreens the TiO nanoparticles are coated to avoid adverse effects of the TiO 2 2 core. The aim of the investigation presented was to determine the possible degradation or change of the surface functionalisation of two different nanomaterials, UV Titan M262 and UV Titan M212 under varying conditions. Any change of the coating may affect the fate and behaviour of the nanomaterial in the environment which is very important for transport, bioavailability and risk assessment. Both tested materials are coated with aluminium oxide to avoid light induced generation of reactive oxygen species at the TiO surface. A further 2 coating with dimethicone (UV Titan M262) or glycerol (UV Titan M212) is applied to ensure hydrophobic and hydrophilic behaviour, respectively. Possible changes of the surface coating were studied by determination of the coating material in the supernatant and by the behaviour of the nanomaterials when environmental parameters like pH, CaCl or humic acid (HA) concentrations were varied. Changes in the 2 behaviour were used as indicators for possible aging of the coating. After suspension of the materials in deionised water both materials showed a comparable behaviour with a point of zero charge around pH 8.5. The analyses of the supernatant revealed that 86% ± 12% of the glycerol and 88% ± 8% of the dimethicone coating was removed from the surface. However, no release of the aluminium oxide coating was

87

observed, which agrees with the result that both materials did not show intrinsic hydroxyl radical generation even after severe treatment. The variation of CaCl concentration resulted in a destabilization of the 2 suspension due to reduced double layer thickness, whereas HA resulted in an increased stabilization of the suspension, by electrostatic and steric repulsion. After HA adsorption both materials showed a negative zeta potential and no point of zero charge was observed (pH 4.5-10). Based on the results of this study it was shown that HA has a large stabilizing effect for the studied nanomaterials and tested conditions, but in natural environments additional conditions with opposite behaviour can occur which affected the fate and behaviour. Acknowledgement: This work was sponsored by the Environmental Agency of Germany (UBA) 351 Occurrence, distribution and bioaccumulation of UV filters in water, sediment and fish from four Iberian river basinsP. Gago Ferrero, IDAEACSIC / dep. of environmental chemistry; S. Diaz-Cruz, IDAEACSIC / Environmental Chemistry; D. Barcelo, IIQABCSIC. Awareness regarding the harmful effects of sun-light on the human skin has recently increased the production and consumption of sunscreens. These products contain organic and inorganic ultraviolet filters (UV F) that absorb, reflect and/or scatter UV radiation, therefore preventing skin diseases such as skin cancer. UV filters are active ingredients added to sunscreens and to personal care products used daily, such as cosmetics, skin creams, body lotions, soaps, hair sprays, hair dyes, and shampoos. The concentrations used in these products are regulated and in general do not exceed 10% due to the fact that some UV filters show toxic effects. The increased production and use of UV F has led to increased and continuous inputs into the environment and this fact makes UV F to be considered environmental pseudo-persistent pollutants. These compounds and their active metabolites may enter the environment as complex mixtures through direct and indirect pathways. The direct sources are the washing off during bathing activities (sea, lake, river and swimming pool) as well as industrial wastewater discharges. Indirect inputs are related to urban wastewater discharges (showering, clothes washing and urine excretion) and via wastewater treatment plants. Aquatic pollution is particularly troublesome because aquatic biota is captive to continual life-cycle, multigenerational exposure. Despite several studies performed during the last decade on this topic, the knowledge on their occurrence, fate and effects in the environment remains highly fragmentary. Therefore, an important effort in this field is required to provide a global view of the risks for the ecosystems. The aim of this work was to develop analytical methodology and apply it to assess the occurrence of UV filters in surface waters and sediments collected along four Iberian rivers basins with Mediterranean regimen: Ebro, Llobregat,Guadalquivirand Júcar, all of them subjected to intensive anthropogenic activities. Beyond obtaining occurrence data, this study aimed also to assess the bioaccumulation and potential biomagnification processes through the trophic web. Individuals of different fish species collected along the four river basins, in the same locations as water and sediment, were also analyzed and evaluated. 352 Micropollutant concentrations in sewage effluent from McMurdo Station and Scott Base and surrounding coastal waters, Ross Island, Antarctica S.K. Gaw, University of Canterbury / Chemistry; P. Emnet, University of Canterbury; G. Northcott, Plant and Food Research; L. Graham, University of Canterbury. Antarctica is considered to be one of the last untouched wilderness areas on earth and as such is the focus of intense scientific investigation. The growing influx of science staff in the spring/summer season has increased focus on reducing the environmental impacts of scientific research activities in Antarctica. Many of the research stations are located in coastal areas into which they discharge sewage. Only 63% of the permanent bases and 31% of the summer stations have any kind of sewage treatment. In addition some research parties undertaking extended periods of fieldwork are allowed to dispose of raw sewage via tidal cracks in the sea ice. Sewage discharges are a recognised source of organic micropollutants entering the environment. Personal care products including soaps, sunscreens and toothpaste are a key source of organic micropollutants entering sewage treatment plants. A pilot study was

SETAC Europe 23rd Annual Meeting

undertaken to determine if micropollutants from personal care products were present in treated sewage effluents from Antarctic Research Stations and were being released into Antarctic coastal waters. Wastewater samples from two Antarctic research stations, McMurdo Station (USA) and Scott Base (New Zealand) and seawater samples from the surrounding coastline were analysed for a suite of organic micropollutants. Organic micropollutants were detected in the WWTP effluents as well as at all of the sea water sampling locations including the reference site located up current of the research stations. Detected compounds include paraben preservatives, triclosan, octylphenol, bisphenol A, UV-filters, and the hormone estrone. Target analytes were detected in both wastewater and seawater at concentrations similar to those reported in temperate environments with higher population densities. The results of current research investigating temporal trends in micropollutant concentrations in the WWTP effluents and a more extensive spatial survey of micropollutants in Erebus Bay will also be presented. The environmental fate of the detected compounds will be discussed in terms of Antarctica’s unique environmental conditions. 353 Can organic contaminants in lake sediments bioaccumulate in Daphnia resting eggs? A.C. Chiaia-Hernandez, Eawag / Environmental Chemistry; T. Hollingshaus, M. Moest, P. Spaak, Eawag; J. Hollender, Eawag / Dept Environmental Chemistry. Chemical contaminants can affect aquatic organisms such as natural Daphnia populations either by direct toxin uptake, or indirectly by e.g. ingestion of contaminated algae. In addition, chemicals in the sediment may bioaccumulate in Daphnia diapausing eggs (ephippia) influencing their fitness and hatching abilities. Since most probably not a single chemical has caused Daphnia to adapt but the combined sum of different compounds, it is important to screen for a broad range of organic contaminants. Sediments are excellent archives of environmental contamination if the chemicals persist over time and mostly under anaerobic conditions. after an appropriate extraction and LC combined with high resolution Orbitrap mass spectrometry (HR-MS), we screened for a broad range of organic compounds, including many PCPs in sediment samples. The results show that sediments can be integrators in time and space for emerging contaminants providing history of chemical deposition. In Lake Greifensee, the musk fragrance tonalide and the biocides triclosan and terbutryn have a similar input pattern with a maximum concentration in the late 1970s follow by a rapid decline. Their concentrations correlate with the total phosphorus concentration in the lake, which indicates an input to surface waters via waste water treatment plant (WWTP) discharge. Early layers from the lakes show the presence of PCPs such as octocrylene, and the transformation product of galaxolide (galaxolidone). In addition, suspect screening aproaches allowed the detection of suspected compounds like the biocide triclocarban. This is a significant advantage compared to water samples, where usually no historical samples are available and emerging contaminants have not been studied in the past. The target analysis combined with the suspect screening of possible contaminants using HRMS provide a comprehensive picture of the overall contamination pattern, and PCP was found as an important class of the contamination. Base in our outcomes, a group of chemicals were selected to study their bioaccumulation and effect on ephippia. The results show that ephippia can take up contaminants from the pore water in the sediment and that the BCF for the neutral compounds correlate with the log K The ow. findings provide information about the past and present contamination of lakes as well as insights to understand the impact of habitat alteration on Daphnia and their reaction to contaminants. 354 Metabolism of benzophenone-2 in novel in vitro and in vivo zebrafish bio-assays V. LE FOL, INERISINRA TOXALIM / In vitro and in vivo ecotoxicology; S. AIT-AISSA, B. PICCINI, INERIS, National Institute of industrial environment and risk assessment / In vitro and in vivo ecotoxicology; A. HILLENWECK, INRA UMR 1331 TOXALIM, research centre in food toxicology / Metabolism of Xenobiotics; E. MAILLOT-MARECHAL, INERIS, National Institute of industrial environment and risk assessment / In vitro and in vivo ecotoxicology; E. PERDU, INRA UMR 1331 TOXALIM, research

88

centre in food toxicology / Metabolism of Xenobiotics; L. DEBRAUWER, INRA UMR 1331 TOXALIM, research centre in food toxicology / Axiom-Metatoul; F. BRION, INERIS, National Institute of industrial environment and risk assessment / In vitro and in vivo toxicology; D. ZALKO, INRA UMR 1331 TOXALIM, research centre in food toxicology / Metabolism of Xenobiotics. Contamination of aquatic environment by xeno-estrogens present in personal care products causes raising concerns. Benzophenones are used as UV filters in sunscreens. They can be detected in rivers, lakes, sewages and sediments with increasing concentrations during the summer season. A number of reports have demonstrated that rivers contamination with xeno-estrogens can result in hormonal dysfunctions in aquatic organisms, which have been characterized through histopathological changes at the level of gonads, developmental disorders and reproductive inability. In fish, benzophenone-2 (BP2) was reported to cause alterations such as the induction of plasmatic vitellogenin and decrease in spermatozoa production (males), as well as reduced spawning (females). Several bioassays, including aquatic vertebrate models, have been designed over the last years with the aim to characterize the estrogeno-mimetic potential of chemicals. However, the characterization of the biotransformation capability of these biological models which allows to take into account bio-activation/detoxification processes in effect assessment, is rarely reported. We have recently developed in vitro bio-assays based on the expression of zebrafish estrogenic receptors in an hepatic cell line (ZFL), and an in vivo zebrafish assay based on the expression of the Green Fluorescent Protein (GFP) under the control of the cyp19a1b promoter (an estrogeno-regulated gene located in the brain). These bioassays were used to investigate the estrogenic potency of benzophenone derivatives. Contrary to other chemicals, BP2 elicited a different response depending on the bio-assay we used. We hypothesized that this difference could rely on a different metabolic capability expressed by our models (i.e. detoxification/bio-activation ratio), which would explain the observed differences in the respective estrogenic responses. To examine this hypothesis, the in vitro and in vivo zebrafish systems were exposed to BP2 at various concentrations, using a tritium-labeled 3 molecule ( H-BP2). BP2 metabolism was explored using radio-HPLC (metabolite profiling) and metabolite characterization was investigated using biochemical tools and high resolution mass spectrometry (HRMS). 355 Determination of methyl siloxanes in soils and biota samples from the Antarctic region M. Farre, J. Sanchis, C. Galban-Malagon, A. Cabrerizo, J. Dachs, D. Barcelo, IDAEA-CSIC. Low molecular weight cyclic (cVMS) and linear (lVMS) volatile methyl siloxanes are a class of synthetic compounds, which belong the class of silicones. These compounds are very stable and are used in a plethora of applications, in particular en personal care products. cVMS are used as carriers/thinners for the more viscous siloxanes that are meant to remain on treated surfaces. In industry applications, are used as lubricants, slips, hydraulic, transmission fluids, moisturizers, anti-foaming agents and plasticizer in silicone joint sealants in the construction sector. In consumer products, are used on textiles and in polishes for its water repellent property and as an antifoaming agent in washing powder. Also in food, dimethyl siloxanes are used as antifoaming agents in the production of beer, jam, juices, deep-frying fats and oils and as anti-clotting agents in powdered food. Many of these compounds are also used in personal care products including cosmetics, deodorants and soaps among others (Horii and Kannan, 2008; Foster, 2008; Lewis, 1999; Lodén et al., 2011; Mills and Showell, 2004). These compounds are also present in the industrial releases associated with the manufacture of high molecular weight silicon polymers. As a result of their wide use, siloxanes are presumably spread into the environment (~10×106 kg year-1 in the US) according to Dow Corning estimation (Navea et al., 2011), by both via point sources and via diffuse sources and may be found everywhere in the environment. Recent studies have suggested that siloxanes may have direct or indirect toxic effects on various biological processes. Consequently, a number of cVMS are currently under review for priority pollutant classification in North America and Europe. Therefore, the occurrence, fate and behaviour of linear and cyclic methyl siloxanes should be assessed and characterized in the environment. In this work

SETAC Europe 23rd Annual Meeting

we have investigated the presence of eight compounds (hexamethylcyclotrisiloxane (D3), octamethylcyclotetrasiloxane (D4), Decamethylcyclopentasiloxane (D5), dodecamethylcyclohexasiloxane (D6), octamethyltrisiloxane (MDM), decamethyltetrasiloxane (MD2M), dodecamethylpentasiloxane (MD3M), and tetradecamethylhexasiloxane (MD4M)) in soils, phytoplankton and krill from the Antarctic region. 356 Background and future prospects for the Global Monitoring Plan under the SC R. Guardans, MARM. The historical background and future prospects for the GMP under the SC are briefly described with particular attention to the need of international stable and effective monitoring strategies to establish and maintain QA/QC standards in the sampling, analytical and data management procedures and to make effective use in science and policy making of the best available data 357 Air Measurements under the GAPS Network in Support of the Global Monitoring Plan of the Stockholm Convention on POPsT. Harner, Environment Canada / Atmospheric Science & Technology Directorate; L. Ahrens, Swedish University of Agricultural Sciences SLU / Dept of Aquatic Sciences and Assessment; S. Lee, Environment Canada / Atmospheric Science and Technology Directorate; E. Sverko, Environment Canada / Science and Technology; J.K. Schuster, Atmospheric Science & Technology Directorate; J. Charland, Environment Canada. The Global Atmospheric Passive Sampling (GAPS) Network has been operating since 2004 at more than 50 sites measuring persistent organic pollutants (POPs) in air. Data from GAPS made a significant contribution to the first Global Monitoring Plan (GMP1) report that was accepted by Parties at COP4 of the Stockholm Convention on POPs in 2009. The first GMP highlighted challenges for future reporting that included: i.) addressing the lack of data in some regions through capacity building, technology transfer; and implementation of new programs; ii.) interpreting air monitoring data in an integrated approach that includes models, emissions information, meteorology, and the context of a changing environment; and iii.) generating data on the next generation of priority chemicals to inform risk assessment. Since GMP1, the GAPS network has generated data on new classes of chemicals inter alia alternative flame retardants, per- and polyfluoroalkyl substances, siloxanes, penta- and hexachlorobenzene. In some cases, these measurements required modifications to the conventional passive samplers used under GAPS and/or improved characterization of existing samplers. For instance, the sorbentimpregnated polyurethane foam (PUF) or SIP disk samplers were developed for capturing a broader range of target compounds that includes the more volatile chemicals; and recent field calibration of the PUF disk sampler indicates that both gas- and particle-phase chemicals are captured at similar sampling rates indicating that the sampling chamber does not discriminate particle-phase chemicals. The GAPS Network has also worked with partners in the GRULAC (Group of Latin American and Caribbean countries) region to generate data for dioxins/furans in air across the region that will contribute to GMP2, due in 2015. Starting in January 2011, PUF disk samplers were deployed for 4 consecutive 6 month periods, at up to 14 sites in the GRULAC region and samples were analyzed for a suite of 17 PCDD/Fs. Highest levels of PCDD/Fs are associated with urban sites and the agricultural site, where biomass burning may be a factor. The air profile is dominated by OCDD, hepta-DD and several PCDFs. Concentrations of PCDD/Fs in air at background sites are about an order of magnitude lower compared to the source-type sites. Temporal trends are also available for some sites for the first 3 consecutive sampling periods and exhibit similar levels and profiles of PCDD/Fs at each site, indicating consistency in source contributions. 358 Persistent organic compounds in ambient air of Africa J. Klanova, P. Kukucka, Masaryk University / RECETOX; P. Pribylova, J. Boruvkova, Masaryk University; R. Prokes, Masaryk University / RECETOX, Faculty of Science. Application of persistent organic pollutants (POPs) in the African countries is closely connected to pesticides. They were used in the agricultural production of food crops such as maize, sorghum and millet as well as cash crops for export such

89

as cocoa, rubber, cotton and timber. They were applied also for disease vector control, especially for mosquito (malaria) and tsetse fly (trypanosomiases). POP pesticides have been generally imported and not produced in Africa, but pesticide formulation plants exist in many countries. DDT, endosulfan, chlordane, lindane (HCH), heptachlor, toxaphene, HCB and aldrin were identified as most frequently used pesticides. The most serious problem the African region faces is the issue of stocks and reservoirs of obsolete discarded and banned POP pesticides. In addition, banned pesticides continue to be used in practice without any control of the authorities. The air samples from 13 sampling sites in 12 countries were collected during two years of the African study. Sampling sites ranged from continental through rural to urban backgrounds. Levels of various chemicals ranged over several orders of magnitude. Congo, Sudan and Senegal had the highest levels of PCDDs/Fs (up to 262, 120 and 80 pg/sample, i.e. 0.45, 0.2 and 0.13 3 pg/m , of 17 EPA congeners, respectively) while they never exceeded 1 pg/sample at Mount Kenya. The highest concentrations of PCBs were 3 found in Sudan (4.2 ng/sample, 21 pg/m ) and highest concentrations of 3 PBDEs in Sudan (14.3 ng/sample, 24 pg/m ), Congo (5.8 ng/sample, 10 3 pg/m ), Canary Islands and Mauritius (both up to 4 ng/sample, 7 3 pg/m ). For pesticides, the highest levels of DDTs were determined in 3 Ethiopia (78 ng/sample, 390 pg/m ), HCHs in Togo (2.4 ng/sample, 12 3 3 pg/m ), and endosulfane in Ethiopia (75 ng/sample, 120 pg/m ). This study is an important contribution to our current knowledge on the levels and distribution of POPs in the African continent. As such, it also significantly contributes to successfull realization of the Global Monitoring Plan, an important tool for effectiveness evaluation of the Stockholm Convention. 359 POPs in the Oceans – spatial, temporal trends and air-water exchange R. Lohmann, University of Rhode Island / Graduate School of Oceanography; J. Klanova, Masaryk University / RECETOX. The fate of persistent organic pollutants (POPs) in the open Oceans is under debate. In 1989/1990, a first global study covering the world’s oceans reported polychlorinated biphenyls (PCBs) and other POPs mostly being taken up via air-to-water exchange. This was backed up by theoretical work that predicted air-water exchange (via net deposition) to be the main delivery pathway for PCBs into the Atlantic Ocean. The initial transfer of PCBs into the mixed surface ocean layer would lead to removal fluxes to deep oceans and sediments, probably by coupling to the ‘biological’ pump. Shelf sediments were identified as major repositories of PCBs, but the exact pathway of how the PCBs reached the sediments was not investigated. Removal of PCBs by deep water formation seems to be of regional importance, but represents only a small fraction of total PCB losses from the oceans. Most previously published ship-based transects that reported on POPs and their cycling were performed on North-South transects on the eastern side of the Atlantic Ocean. Yet the sampling regime was invariably affected by continental emissions off Europe and Africa, making extrapolations across the entire Atlantic Ocean difficult and fraught with uncertainties. In addition, oceans are not homogenously well-mixed water bodies, but display strong gradients in temperature, productivity, and influx of terrestrial materials, including POPs. Numerous currents move water around and induce mixing both horizontally and, less efficiently, vertically. Recently, cruises that covered both sides of the Atlantic Ocean transect have shown important variations in concentrations as a function of ocean currents or river plumes, especially for perfluorinated compounds, chlorinated pesticides, PCBs and to a lesser degree also for polycyclic aromatic hydrocarbons (PAHs). Air-water exchange gradients for PCBs and OCPs (but not for PAHs) suggest that the major oligotrophic oceans display net volatilization. Similar results were also observed in the Pacific Ocean. As was previously concluded for PCBs, there is little indication that concentrations of PAHs in the surface water of the Atlantic Ocean have changed strongly over the last decade. Overall, the picture of POPs in the global oceans is complicated, but more details are emerging which support a scenario in which major parts of the oceans are returning POPs to the atmosphere. 360 Biogeochemical controls on the remobilization and reservoirs of

SETAC Europe 23rd Annual Meeting

polychlorinated biphenyls (PCBs) in AntarcticaA. Cabrerizo, IDAEA-CSIC / Department of Environmental Chemistry; J. Dachs, IDAEACSIC / Environmental Chemistry; D. Barcelo, IIQABCSIC; K.C. Jones, Lancaster University / Lancaster Environment Centre. After decades of primary emissions, reservoirs of persistent organic pollutants (POPs) have accumulated in soils and snow/ice in Polar Regions. Even though once considered a final sink for these chemicals, several studies have shown evidence that historical burdens of POPs are currently being remobilized from retreating glaciers/ice cover in the Arctic and Antarctica. This remobilization may be enhanced under climatic change scenarios (changes in temperature, retreating ice,…) once primary emissions have decreased due to regulation actions, increasing their availability for exchange with the atmosphere, and therefore increasing the ecosystem’s exposure to POPs. While an increase of 1 ºC in ambient temperature due to climate change would increase current Antarctic atmospheric inventories of PCBs by 21-45 %, a concurrent increase of 0.5% SOM would counteract the influence of warming by reducing the POP fugacity in soil. A 1 ºC increase in Antarctic temperatures will induce an increase of the soil-vegetation organic carbon and POP pools by 25%, becoming a net sink of POPs, and trapping up to 70 times more POPs than those remobilized to the atmosphere. Here we show, for the first time, field measurements proving that polychlorinated biphenyls (PCBs) are currently re-volatilizing from Antarctic snow/ice and soils. This remobilization is driven not only by temperature and ice-cover changes, but also by changes in soil organic matter (SOM) and other biogeochemical processes. 361 WHO/UNEP-coordinated global surveys on concentrations of Persistent Organic Pollutants (POPs) in human milkR. Malisch, K. Kypke, A. Kotz, State Institute for Chemical and Veterinary Analysis of Food; A. Tritscher , WHO GEMS / Food Programme, Food Safety and Zoonoses, World Health Organization; K. Magulova , United Nations Environment Programme (UNEP), Stockholm Convention Secretariat, International Environment House; H. Fiedler, UNEPDTIE Chemicals Branch. Since 1987 the World Health Organization (WHO) has carried out global surveys on the concentrations of polychlorinated dibenzo-p-dioxins (PCDDs), dibenzofurans (PCDFs) and biphenyls (PCBs) in human milk. The first WHO-coordinated exposure study took place in 1987-1988, the second round in 1992-1993 and the third round in 2000-2003. These studies fulfil important requirements for biomonitoring as requested by the Stockholm Convention on Persistent Organic Pollutants (signed in 2001 and entered into force in 2004). The objective of this Stockholm Convention is to protect human health and environment from POPs by reducing or eliminating their releases into the environment. For the evaluation of the effectiveness, human milk was chosen as one of the two core matrices to be monitored under the Stockholm Convention. Therefore, it was agreed expand the studies for inclusion of the Stockholm Convention POPs and to perform the WHO human milk surveys in close collaboration with United Nations Environment Programme (UNEP) and the Stockholm Convention Secretariat. Thus, the fourth round in 2005–2007 and the since 2008 following continuous studies were organized as joint WHO/UNEP studies. A comprehensive protocol was developed for collection of representative samples, handling and analysis samples. 66 countries participated between 2000 and 2011 submitting 84 pooled (= mixed representative) samples. Large global and regional differences with respect to contamination of human milk with different POPs were found. Time trends can be derived for countries with repeated participation over time. Selected results will be presented to give an overview of the 1 complex picture. References Environmental Health Series No 34 (1989): Levels of PCBs, PCDDs, and PCDFs in breast milk, WHO 2 Regional Office for Europe, Copenhagen, Denmark. Environmental Health in Europe No. 3 (1996): Levels of PCBs, PCDDs and PCDFs in human milk: Second round of WHO-coordinated exposure study, WHO 3 Regional Office for Europe, Copenhagen, Denmark WHO (2000) Quality Assessment of Levels of PCBs, PCDDs and PCDFs in Human Milk and Blood Plasma – fourth round of WHO-coordinated study (2000), EUR/00/5020352, WHO Regional Office for Europe, Copenhagen, Denmark. 4 Rainer Malisch and FX Rolaf van Leeuwen

90

(2003) Results of the WHO-coordinated exposure study on the levels of PCBs, PCDDs and PCDFs in human milk, Organohalogen Compounds 64: 140 – 143 5 Malisch, R. and van Leeuwen, F.X.R. (2002) Third round of WHO-coordinated exposure study: Analysis of PCDDs, PCDFs and PCBs in human milk, Organohalogen Compounds, Vol. 56, 317-320 6 WHO protocol Fouth WHO-Coordinate Survey of Human Milk for Persistent Organic Pollutants in Cooperation with UNEP Guidelines for Developing a National Protocol, Revised 1 October 2007, Department of Food Safety and Zoonoses, World Health Organization, Geneva 7 http://www.who.int/foodsafety/chem/POPprotocol.pdf UNEP, Status report on the human milk survey conducted jointly by the Secretariat of the Stockholm Convention and the World Health Organization, Conference of the Parties of the Stockholm Convention on Persistent Organic Pollutants, Fourth meeting, Geneva, 4–8 May 2009, 8 UNEP/POPS/COP.4/INF/31 Rainer Malisch, Karin Kypke, Alexander Kotz, Kerstin Wahl, Nicole Bitomsky, Gerald Moy, Seongsoo Park, Angelika Tritscher, Seoung Yong Lee, Katarina Magulova and Heidelore Fiedler (2010) WHO/UNEP-coordinated exposure study (2008-2009) on levels of Persistent Organic Pollutants (POPs) in human milk with regard to the Global Monitoring Plan, Organohalogen 9 Compounds 72: 1766-1769 Rainer Malisch, Karin Kypke, Alexander Kotz, Kerstin Wahl, Nicole Bitomsky, Gerald Moy, Seongsoo Park, Angelika Tritscher, Seoung Yong Lee, Katarina Magulova and Heidelore Fiedler (2011) Levels of HCH, endosulfan, chlordane, nonachlor, S-421, musk xylene and musk ketone in human milk of WHO/UNEP-coordinated exposure study for determination of POPs (2008-2010) with regard to the Global Monitoring Plan, Organohalogen 10 Compounds 73:1023-1025 Hoogenboom LAP, Talidda A, Reeuwijk NM, Traag WA, Kotz A, Malisch R, Moy G, Tritscher A, Lee SL, Magulova K, Fiedler H (2011) Are Elevated Dioxin Levels in Milk of Women from Ivory Coast and Congo due to the use of Clay During Pregnancy? Organohalogen Compounds 73:1026-1029 362 Bioanalytical tools to measure endocrine active compounds F.D. Leusch, Griffith University Smart Water Research Centre / School of Environment; A. Hebert, Veolia Environnement Recherche et Innovation SNC; M. Schriks, KWR Watercycle Research Institute. Chemicals present in the environment have been conclusively linked to endocrine dysfunction in exposed wildlife. While the evidence from wildlife is overwhelming and conclusive, the evidence for human effect is less so. Nevertheless, the precautionary principle would dictate that we take endocrine disruption as a potentially serious threat to human life even at low concentrations, and it is important to understand the contribution from water and its relevance. Methods that rely on biological activity are finding increasing utility as screening tools, because the chemical nature of complex water samples may be unknown and/or difficult to quantify/identify, and the biological method may be the best (or only) indicator of biological activity. Due to the complex nature of the endocrine system it is important to assess multiple endpoints when investigating endocrine active compounds. Much of the work on endocrine disruption from exposure to environmental contaminants has been carried out on the estrogenic axis. In the last decade substantial effort has been put into the development of bioassays to assess the estrogenicity of various waters. But the endocrine system is not simply about estrogenic activity, and several other hormonal systems play a crucial role in the maintenance of homeostasis, sexual development, metabolism, growth and behaviour. Substantially less information is available on those other endocrine endpoints, although it is becoming clear that these pathways can also be disrupted by exposure to environmental contaminants. The Global Water Research Coalition (GWRC) project “bioanalytical tools to analyse hormonal activity in environmental waters“ continues and expands on previous GWRC efforts to develop and validate methods to measure estrogenic activity in water to include a range of substantially less well-studied endocrine endpoints. Our review specifically focused on androgens, progestagens, steroidogenesis, glucocorticoids, thyroid hormone, PPAR and retinoid acid receptors, and collected information on: production, function and molecular and cellular mode of action of the different hormones; overview of chemicals identified as EDCs for each particular endpoint;

SETAC Europe 23rd Annual Meeting

review of in vitro methods currently available for each endpoint; current status of interassay comparisons and interlaboratory validations; reported activity in environmental waters; and an expert evaluation of the different in vitro methods currently available. 363 Transcriptomics and/or functional genomics tools for effect based analysis of environmental water S. Van der Linden, B. Pieterse, Biodetection Systems BV; R. Cavill, Maastricht University / Department of Toxicogenomics; B. Brouwer, VU University Amsterdam / Department of Animal Ecology; B. van der Burg, Biodetection Systems BV. Many chemicals that are intentionally or unintentionally produced finally end up in the aquatic environment, where they may build up or can be degraded to a multitude of unknown breakdown products. As some of these compounds can withstand environmental degradation or sometimes even water purification treatment steps, animals and humans may run the risk of continuous exposure to complex mixtures of unknown composition and unknown effects. Chemical analysis alone will not suffice to determine the presence of (compounds with) adverse effects. Partly because it is complicated to - cost effectively - identify all components present in complex mixtures like an environmental extract, but also because knowledge regarding their biological effects is largely lacking. One way to tackle this problem is by using in vitro bioassays that respond specifically to groups of compounds with a known mode of action. By focusing on essential pathways that are known to be linked to adverse effects (i.e. adverse outcome pathways (AOPs), the presence of compounds that can disrupt these key AOPs can be determined. We have developed a panel of bioassays focusing on several AOPs that are related to different aspects of human and animal health, most importantly pathways related to endocrine disruption and genotoxicity. These allow for a fast quantification of a multitude of effects in environmental samples. A more holistic way of analyzing the presence of adverse effects qualitatively, might be by measuring the effects on a multitude of genes simultaneously, using transcriptomics tools. The gene expression pattern, e.g. analyzed by using microarrays, may be indicative of the types of activity present in the sample. We have analyzed a range of pure compounds and environmental samples for activity on a large set of nuclear (hormone) receptors. Simultaneously, we have analyzed gene expression pattern from cells that are exposed to the same set of samples, using microarrays for quantification of the genes expressed. An overview of these analyses on a range of pathways will be presented and the results from reporter gene assays and gene expression analysis will be compared. As many of AOPs are thought to be of key importance for human and animal health and many of these pathways are relatively conserved, specific effects detected on these pathways are thought to be of high relevance for hazard and quality assessment of environmental water and related matrices. 364 Endocrine Disruption Potentials and Related Mechanisms of Several Organophosphate Flame RetardantsX. Liu, Seoul National University / School of Public Health; K. Ji, Seoul National University; H. Moon, Hanyang University / Marine Environment Research Division; K. Choi, Seoul National University / School of Public Health. Organophosphate flame retardants (OPFRs) have been widely used as alternatives to polybrominated diphenyl ethers (PBDE). However, knowledge on their potential toxicities is limited. In this study, cell lines and a zebrafish (Danio rerio) model were employed to investigate the endocrine disruption potency and underlying mechanisms of several OPFRs. Firstly, two human cell lines (H295R and MVLN) were employed to screen the endocrine disrupting potentials of six OPFRs, i.e., tris-(2-chlorethyl) phosphate (TCEP), tris-2chloroisopropyl phosphate (TCPP), tris-(1,3-dichloro-2-propyl) phosphate (TDCPP), tris-(2-butoxyethyl) phosphate (TBEP), triphenyl phosphate (TPP), and tricresylphosphate (TCP). By all six OPFRs, both 17?-estradiol (E2) and testosterone (T) concentrations increased in H295R cells, and transcription of major steroidogenic genes was upregulated but sulfotransferase genes were downregulated. In MVLN cells, TDCPP, TPP, and TCP acted as ER antagonists. Then, a 21 d reproduction test was conducted to study their effects on reproduction

91

performances. In this test, adult fish pairs were exposed to TDCPP or TPP (0, 0.04, 0.2, and 1.0 mg/L) for 21 d. The results showed that fecundities were significantly decreased, which were accompanied with significant increases of plasma E2, vitellogenin (VTG), E2/T and E2/11KT, and decreases of T and 11-KT. Altered transcription of genes along hypothalamus-pituitary-gonad (HPG) axis was sex-dependent. Finally, a long term exposure was conducted to assess developmental toxicity and cross-talks among the HPG, hypothalamus-pituitary-thyroid (HPT) and hypothalamic-pituitary-adrenal (HPA) axes. In this test, fish eggs were exposed to 0.005, 0.05 and 0.5 mg/L TDCPP or TPP for 120 d. After exposure, conditional factor (CF), gonad somatic index (GSI), and liver somatic index (LSI) were significantly decreased. In females, exposure to TDCPP or TPP led to increase of cortisol, follicle stimulating hormone (FSH), luteinizing hormone (LH), thyroxine (T4), triiodothyronine (T3), and E2, and lesser concentration of 11-KT. In males, decreased plasma cortisol, FSH, LH, T4, T3, T, and 11-KT were observed. Transcription of genes along HPG, HPA and HPT axis were mostly up-regulated in female, while down-regulated in males. The results of a series of experiments showed that several OPFRs are endocrine disrupting even at levels environmentally occurring. Ecological implications of these observations deserve further investigation. 365 Glucocorticoid-like compounds detected in wastewater and river water samples from the Czech Republic and SwitzerlandP. Macikova, Masaryk University / Faculty of Science, RECETOX; A.A. Ammann, K.J. Groh, K. Schirmer, M.J. Suter, Eawag. Many contaminants found in surface waters and wastewater can disturb endocrine functions, for example via interaction with the endogenous hormone receptors. Glucocorticoids are among the hormones that regulate key physiological processes such as immune response, energy metabolism and stress response. Disruption of glucocorticoid signaling has been associated with several adverse effects and diseases including teratogenicity, obesity, type 2 diabetes or inflammatory and autoimmune diseases. A number of chemicals known to interfere with glucocorticohormone signaling pathway were detected in river water and wastewater extracts using HPLC-MS/MS. Pure compounds as well as extracts of environmental samples were further assessed for their ability to interfere with glucocorticoid receptor using GR-CALUX assay. Relative potencies of pure compounds to elicit GR-mediated activity were calculated compared to dexamethasone (Dex) as the standard. Glucocorticoid-like activity was detected in untreated hospital wastewater effluents as well as in treated municipal WWTP effluents, and reached up to 542 ng Dex-equivalents/L (hospital WW), resp. 37 ng Dex-EQ/L (WWTP effluent). GR-mediated activity of river water samples was very low and could not have been calculated from the bioassay, however, concentration of GCs in rivers were not negligible. Furthemore, anti-glucocorticoid activity was found in some river water samples suggesting that glucocorticoid-like effects of the whole mixture could be masked. 366 Identification of Thyroid Hormone-Disrupting Compounds in Polar Bear Plasma by Effect-Directed Analysis (EDA) E. Simon, IVM VU University; E. Lie, The Norwegian School of Veterinary Science (NVH); K. Loken, The Norwegian School of Veterinary Science; J. Bytingsvik, B.M. Jenssen, Norwegian University of Science and Technology; M. van Velzen, J. de Boer, T. Hamers, M. Lamoree, Inst. for Environmental Studies (IVM) - VU University. Recent studies demonstrated that in addition to other endocrine systems the thyroid hormone (TH) system is also vulnerable for environmental contaminants. TH-disrupting compounds can bioaccumulate in adipose tissue and blood of top predators and are capable of competitive binding to transthyretin (TTR), which is one of the transport proteins of the TH, thyroxine (T4). Effect-Directed Analysis (EDA) was performed on polar bear plasma samples to identify the compounds responsible for the elevated TH-disrupting potency measured in the radioligand T4*-TTR binding assay. A set of initially target analyzed TTR-binding compounds explained ~40% of the measured activities. In order to detect the compounds causing the remaining activity, a generic

SETAC Europe 23rd Annual Meeting

identification strategy was developed. The extracts were analyzed by high resolution LC-ToF-MS. Library-based identification was applied to the full-scan data screening for accurate mass- and isotope pattern-match between the compound lists and the data files. The libraries were compiled based on different selection criteria, such as TTR-binding-, blood accumulating potency and environmental occurence of the compounds. Out of the tentatively identified 23 suspects nonylphenol was confirmed chemically by evaluating its chromatographic and mass spectrometric behaviour after simultaneous injection of its analytical standard and the sample extracts on the LC-TOF-MS. Then, Isotope Cluster Analysis (ICA) was applied to the original LC-ToF-MS data enabling specific screening of halogenated isotope patterns. This strategy led us to the successful identification and confirmation of three (di-)hydroxylated octachlorinated biphenyls. As no pure standards were available, these compounds were successfully synthesized from their corresponding methoxylated form prior to analytical and toxicological confirmation. All analytically confirmed, identified compounds (i.e. nonylphenol and (di)OH-octaCBs) showed TTR-binding potency in the bioassay and could explained another ~35% of the total measured TTRbinding activities. However, the contribution to the total measured activity of nonylphenol compared to the (di)OH-octaCBs is negligible due to its weaker TTR-binding affinity its plasma concentration level is remarkable (2.5-6.2 µg/L). A major but very trivial obstacle for the confirmation of toxicants that is the often limited availability of pure standards that are required for the confirmation of the identified compounds. 367 Can PNEC for estrogenic in vitro assays be derived? B. Jarosova, Masaryk University / Faculty of Science, RECETOX; K. Hilscherova, Masaryk University Faculty of Science RECETOX / Faculty of Science, RECETOX; L. Blaha, Masaryk University, Faculty of Science, RECETOX / Research Centre for Toxic Compounds in the Environment (RECETOX). Since in vivo potential of estrogenic substances differs from their in vitro potential, results of in vitro bioassays alone cannot be directly used in environmental assessment. However, estrogenic in vitro bioassays are so sensitive, relatively inexpensive and rapid that they would be a very good tool for environmental risk assessment if we could better interpret their results. Therefore we decided to review the literature and gather information about: i) compounds which were responsible for measured estrogenic activity in European waste waters ii) relative in vitro potency (relative to standard estrogen (17?estradiol, E2)) of these substances iii) the range of ratios in which these substances might occur in waste water samples and iv) in vivo derived PNECs for these substances. Based on this review, we would like in the platform presentation to derive possible ranges of PNECs for different in vitro bioassays and discuss uncertainties of their possible use. 368 Which are the synergists? And are they of any importance from a risk assessment perspective? N. Cedergreen, Royal Veterinary Agricultural University / Department of Plant and Environmental Sciences; M.A. Bjergager, University of Copenhagen / Faculty of Science. The past decades of research have shown that joint toxicity of non-interacting chemical mixtures can be well predicted by the reference models of Concentration Addition and Independent Action. Despite this, little has yet been done to incorporate this knowledge in legislation and risk assessment of chemical mixtures. One reason often given is the concern of synergy. Synergy is defined as chemical interactions leading to larger biological effects than those predicted by a reference model. The question is, however, how often significant synergism actually occurs under natural conditions? And how severe it is? And which chemicals are likely to induce synergy? To answer these questions we reviewed more than 1000 chemical combinations reported in the litterature, picking out the studies where synergy was well described and the ratio between the predicted and observed effect concentrations, the Synergy Ratio (SR), was larger than 2. We focused on four well investigated groups of chemicals likely to co-occur in the environment: pesticides, phamaceuticals, metals and antifoulants. For the mixtures of pesticides, pharmaceuticals and metals, the frequency of synergy was

92

remarcably stable around 5%. The average synergy factor for the pesticides and pharmaceuticals was 5.01±4.30 (n=50), whereas for the metals it was 2.89±1.05 (n = 4) and for the antifoulants it was 6.98±8.18 (n = 27). For the 50 synergistic pesticide combinations either azole fungicides, organophosphates or cabamate insecticides were involved. These groups of pesticides are known to interact with either P450 monooxygenases or esterases known to be involved in the degradation of xenobiotics. The synergy could therefore be explained by toxicokinetic interactions in all cases. The few metal combinations showing synergy did not reveal any specific trends in terms of synergists. The modes of action of the antifoulants are very diverse, and no pattern relating which of the compounds acted as synergists could be distinguished. Approximately 95% of the tested mixtures show additivity or antagony. This is an important take-home message, when it comes to risk assessment. For the synergistic pesticide mixtures, synergy is not occuring randomly but seem to be associated with pesticides already known to interact with the metabolisation of other pesticides. Hence, a more focussed regulation for these compounds could be made by for example adding extra safetyfactors. 369 Independent, additive, synergistic and complex response behaviour of microorganisms on toxic mixtures characterized by micro fluid segment technique J. cao, Ilmenau University of Technology / Inst. for Chemistry and Biotechnology; A. Knauer, Ilmenau University of Technology; m. köhler , University / Phys ChemMicro Reaction Technology. The determination of dose/response functions for combination of drugs in different concentrations is expensive and time-consuming by use of conventional methods. The micro fluid segment technique allows the complete screening of twodimensional concentration spaces at a very low consumption of chemicals. Here, it is shown that this technique is well suited for distinguishing different characteristics of combinatorial effects. Segments were formed by mixing the cell suspension with the three effector solutions and additional buffer. Fluid segments with single volumes of about 0.5 µL have been formed. An automated evaluation of toxic effects can easily by realized by multi-channel micro flow-through photometry and fluorimetry. This arrangement allows the measuring of cell density by optical transmission as well as the detection of coloured cell populations and the measurement of autofluorescence or under the addition of fluorescent sensor particles. The investigation of the behaviour of E. Coli and Chlorella vulgaris on combinations of different frequently applied druges (antibiotics and substances for blood pressure control), food components, a widely used herbicide as well as metallic nanoparticles resulted in very different characteristics of cellular response: Independent, additive, synergistic as well as complex response behaviour of microorganisms on the different binary and ternary mixtures have been found. In particular it was found that the toxicity of silver nitrate solution and silver nanoparticles on chlorella is comparable for the pure substances, but leads to a significant different behaviour in case of combination with atrazine. A strong activation of the growth of E. coli in the presence of ampicilline was found in dependence of the concentrations of caffeine and the blood pressure drug. The investigations show that all principle types of combinatorial behaviour can be observed in microtoxicological screenings by the applied microfluidic method. In addition, it was found that non-monotonous functions have to be expected and that complex response pattern can be found including cellular growth behind the critical thresholds of concentration of a single substance if binary or ternary mixtures are applied. It can be concluded that there is an urgent need for the use and further development of micro fluidic techniques in order to realize such studies with a reasonable effort. 370 How low can you go? M.A. Bjergager, University of Copenhagen / Faculty of Science; K. Norgaard, Roskilde University; A. Kretschmann, University of Copenhagen / Faculty of Science; P. Mayer, Aarhus University / Department of Environmental Science; N. Cedergreen, Royal Veterinary Agricultural University / Department of Plant and Environmental Sciences. Several studies have shown synergistic interaction between the azole fungicides and pyrethroid

SETAC Europe 23rd Annual Meeting

insecticides towards aquatic and terrestrial organisms both in controlled laboratory and in outdoor microcosm studies. Both types of pesticides are among the most widely used groups of pesticides and may occur in the environment together. We could expect, however, that there is a lower threshold concentration below which the azole fungicides do not enhance the toxic effect of the pyrethroids. To test this hypothesis, laboratory experiments with Daphnia magna were carried out testing the synergising potential of the three azole fungicides propiconazole, epoxiconazole and prochloraz on the pyrethroid insecticide alphacypermethrin. Two types of experiments were conducted. In the first experiments, full concentration-response setups with alpha-cypermethrin were conducted with six varying constant concentrations of each of the fungicides. Pesticide concentrations in these setups could, however, not be confirmed analytically, hence, a second approach using passive dosing techniques was tested. In this setup, a dilution series ranging from 1.56 µg/L to 200 µg/L fungicide was tested in combination with two fixed alpha-cypermethrin concentrations of 0.01 µg/L and 0.1 µg/L corresponding to EC and EC , respectively. All pesticides were also 7 50 assessed singly and untreated controls were included. The results showed that there was indeed a threshold for the three fungicides, below which less than two-fold synergy occured. Prochloraz was observed to decrease the alpha-cypermethrin EC more than a factor of two at 50 concentrations of 34±14 ug/L and above. At the concentrations tested, prochloraz in itself is not expected to have any significant acute effect at D. magna as its EC -value has previously been estimated to 3500 ug/L. 50 Propiconazole and epoxiconzole showed similar synergising patterns as prochloraz. To be able to discuss the results in an environmental context we need measured and not only nominal concentrations. The passive dosing experiments including chemical analyses of the test media will add to the information about the threshold concentration. Based on these experiments, the range of the threshold values found for the three fungicides will be discussed in relation to fungicide concentrations measured in the environment. Also the consequences of test system choice for the evaluation of synergy will be discussed. 371 Non-toxic chemicals become toxic in a mixture – “solubility addition” of solid hydrophobic chemicals increases exposure P. Mayer, Technical University of Denmark / Department of Environmental Engineering; K.E. Smith, Aarhus University / Department of Environmental Science; S.N. Norgaard Schmidt, Aarhus University / Environmental Chemistry and Microbiology; N. Dom, University of Antwerp / Biology; R. Blust, University of Antwerp / Systemic Physiological and Ecotoxicological Research, Department of Biology. This study addresses the question whether hydrophobic organic chemicals that exert no toxicity at their solubility limit (saturation) can form a toxic mixture. Spiking methods generally do not allow testing exactly at saturation without introducing microcrystals. Passive dosing was thus applied to test the acute toxicity of several high melting point polycyclic aromatic hydrocarbons (PAHs) and their mixtures at the respective saturation levels to Daphnia magna. Anthracene, chrysene and benzo(a)pyrene exerted no or only limited acute toxicity (0-20%), whereas binary and tertiary mixtures of these resulted in significant acute toxicity (70-88%). It is in such cases not possible to describe and predict mixture toxicity using the toxic unit approach, since there are no toxicity data to calculate the toxic units. The toxicity of five PAHs and their mixtures were instead fitted with one (sum) chemical activity-response curve indicating a similar mode of toxic action (i.e., concentration addition). The effective chemical activity (Ea-50) of 0.029 and the effective concentration on a lipid basis (EC , eq.-50) of 95.7 mM were well within the range for baseline lipid toxicity. These results raise questions about the focus of risk assessment schemes and toxicity testing guidelines on individual substances, since apparently non-toxic chemicals might become toxic in a mixture. 372 Neuroprobabilistic Integrated Risk Index of Chemical Aquatic Pollution (IRICAP): Case studies in Spanish river basinsf. fabrega, Universitat Rovira i Virgili / Chemical engineering department; s. kumar; A. Ginebreda, IDAEACSIC; M. Schuhmacher, Rovira i Virgili University; J. Domingo, Universitat Rovira i Virgili; M. Nadal,

93

University Rovira i Virgili. Because of technical advances, an increasing number of chemicals are found in the environment in general, and in aquatic compartments in particular. Those substances create a notable concern since many of them may pose an important risk for the human health. Because of this large number of pollutants, reliable prioritization methods are clearly necessary. One of the most extended methods to prioritize the hazard of different compounds is related to their properties of persistence, bioaccumulation and toxicity, also known as PBT. Furthermore, these data can be clustered by using Selforganizing Maps (SOM), a special kind of neural networks with the capability to cluster data depending on PBT similarities. In previous works, a SOM-based Hazard Index (HI) was successfully applied in soils, being afterwards improved taking into account neuroprobabilistic aspects. The objective of the present study was to develop a neuroprobabilistic Integrated Risk Index of Chemical Aquatic Pollution (IRICAP), whose aim is to evaluate the human health risk associated to the exposure of chemical mixtures contained in river water. For the IRICAP development, a new HI was created by considering nearly 200 compounds, including pharmaceutical products, illicit drugs, estrogens, detergents, pesticides, perfluorinated compounds, endocrine disruptors, and persistent organic pollutants. According to the HI outcomes, the most hazardous compounds were perfluorinated compounds, pesticides, and illicit drugs. In the second step, a neuroprobabilistic IRICAP was developed and applied to four river basins in Spain: Llobregat, Ebro, Jucar and Guadalquivir. An IRICAP value was calculated in those sampling sites where monitoring data were available. For that purpose, the HI corresponding to each chemical was multiplied by the concentration of that pollutant, and the results of all substances were aggregated. Subsequently, a risk profile along each river basin was obtained. Finally, a Pareto distribution was applied to prioritize those chemicals with the most significant incidence on the IRICAP. Therefore, those pollutants posing the highest risk were identified, making easier the implementation by regulatory organizations. 373 Retrospective RA for almost 400 analytes – challenges and lessons learned from applying current assessment schemes for mixture toxicity under the WFD M. Junghans, Swiss Centre for Applied Ecotoxicology EAWAG EPF / Ecotox Centre; P. Kunz; R. Gauch, Oekotoxzentrum / EPFL; S. von Arb, Swiss Centre for Applied Ecotoxicology Eawag-EPFL; C. Moschet, Eawag / Uchem; A. Piazzoli, ETH; I. Wittmer, Eawag; H. Singer, Eawag - aquatic research; C. Stamm, Eawag; J. Hollender, Eawag / Dept Environmental Chemistry; I. Werner, Swiss Centre for Applied Ecotoxicology / Department of Anatomy, Physiology and Cell Biology. Modern chemical analytics are able to detect increasing numbers of chemical substances in a single run. Measurement campaigns in surface waters can yield anaytical results for several hundred substances per sample. For assessing the chemical status of a water body under the Water Framework Directive (WFD) based on such measurements, only single substances are currently evaluated for their compliance with Environmental Quality Standards (EQSs). With more than 100 substances detected in a single sample however, the need for an assessment of the combined risk of these substances becomes pressing. Recently, a decision tree for prioritising samples regarding mixture risk assessment was proposed, that combines the mixture toxicity framework proposed by the WHO with the decision tree for mixture toxicity assessment proposed by EU scientific committees. This decision tree was applied to analytical results from a recent measurement campaign in Swiss surface waters, in which 379 organic substances were analysed by high resolution mass spectrometry. These analytes comprised biocides, plant-protection products, pharmaceuticals and their transformation products as well as some substances from other applications. The number of substances quantified in 34 individual samples ranged from 42 to 138 (median = 105). Of these substances 214 were detected in at least one sample. Most EQS could be found for plant protection products. For transformation products as well as for pharmaceuticals the EQS availability was generally poor. The samples are classified into 3 classes: I) a risk quotient (RQ=measured concentration/EQS) greater than 1 already for single substances in the sample II) no risk identified and III) risk

SETAC Europe 23rd Annual Meeting

identified only if combined risk is assessed. Subsequently, risk assessment is further refined for samples in classes I and III, by analyzing the risk for the different taxonomic groups. The study shows that the bottlenecks of mixture risk assessment under the WFD are: identification of mixture components and availability of EQSs and/or toxicity data especially for pharmaceuticals and transformation products. In addition the presence of the insecticide synergist piperonyl butoxid in 7 samples prooved to be a challenge for mixture risk assessments. 374 What if Science follows Policies? B. Cheze, IFP Energies nouvelles; L. Patouillard, IFP Energies nouvelles / Economics and environmental evaluation department. One of the drivers for the development of advanced biofuels (i.e. biomass-based fuels produced from lignocellulosic materials (G2) and microalgae (G3)) is the will of policy makers to increase the share of biofuels in the transport sector. For instance, the United States (US) and the European Union (EU) have recently developed regulations inciting biofuel production. Both regulations set minimum thresholds for life cycle greenhouse gas (GHG) emission savings compared to fossil fuel as the most important environmental criteria that should be met to be eligible as sustainable biofuel. Despite a substantial literature on advanced biofuel, there is no consensus about their Life Cycle Analysis (LCA) GHG emission benefits compared to fossil fuels and the fact that they can meet legal GHG emission targets remains unclear. Conversely, the extent to which literature that meets those targets are more likely to be published has to be examinated. This paper investigates the potential influence of biofuel policies on LCA GHG emission results of advanced biofuels by identifying publication biases thanks to specific set of analytical techniques (Funnel graphs and Galbraith plots). Then an harmonisation is performed using specific econometrics method (meta-regression analysis), corrected from publication biases. The purpose of this harmonisation is to identify and quantify key factors that influence LCA GHG emission results of advanced biofuels in order to be more conclusive on its real performances and to support policy makers. This quantitative literature review is based on 47 LCA studies providing 593 observations. Results reveal the existence of an asymmetrical publication bias of North American (NA) vs. EU studies, highlighting the influence of public policy designs on the very scientific publication process. After identifying the main determinants of the variability of advanced biofuel LCA GHG emission results, we perform harmonisation correcting from identified biaises. Our results indicates a hierarchy between advanced biofuels. Mean value and Confidence Interval of LCA GHG emissions weighted by the influence of its main drivers for BtL, Ethanol and G3 biofuel are estimated to 19.5 [16.7;22.2], 19.7 [17.4;22.0] and 60.0 [43.3;76.7] gCO eq/MJ of biofuel 2 respectively. Even if this range of values is lower than fossil reference (about 83.8 in gCO eq/MJ), only G2 Ethanol and BtL do comply with 2 the GHG emission reduction thresholds defined in the US and EU regulations. 375 Geopolitical Implications of Life Cycle Assessment G. Sonnemann, . Life Cycle Assessment (LCA) has been developed as a tool for the systematic evaluation of the environmental impacts of a product through all stages of its life cycle. By addressing both the emissions to the environment and the resource use of a product system LCA has a number of geopolitical implications. On the one side it accounts for the emissions covered under the Multilateral Environmental Agreements that are legal instruments for environmental protection, governed by international law. On the other hand, it accounts for resource uses like water, land & soil, fossil fuels and minerals, including the so-called critical materials, and in this way it is also used as a risk assessment tool, addressing among other issues geopolitical aspects in the supply chain. Emissions related to the UN Framework Convention on Climate Change are taken into account by assessing the Global Warming Potential. The Ozone Layer Depletion Potential assesses ozone-depleting chemicals covered under the Vienna Convention. Chemicals included in the Stockholm and Rotterdam Conventions are part of various Life Cycle Toxicity Impact Models such as USEtox. These Conventions can assist companies in indentifying the emissions of

94

which a sound management is a globally agreed priority. The challenge of ensuring the Earth’s ecosystem services that more are more companies are trying to take adequately into account has also geopolitical implications and is in the hands of the Convention on Biological Diversity. With regard to resource uses the increased use of geographic information in life cycle assessment allows to better address biodiversity loss and to map the supply chain of companies for their products and hence to identify potential future supply constraints for instance for water due to risks related to changing weather patterns and for minerals due to the risk of conflicts. The further development of advanced LCA methods using geographic information system will facilitate a growth in using LCA for this type of applications. Hence, overall, it can be expected that LCA, which was developed as a tool for assessing environmental impacts, will be used increasingly as an instrument to address geopolitical aspects and to identify related risks throughout the life cycle of a product. 376 UNEP Capacity Building Approach towards Mainstreaming Life Cycle Thinking in the Emerging and Rapidly Growing Economies A. Quiros, ECO GLOBAL ALCALA; S. Valdivia, UNEP / Secretariat ILCI; T. Mermer, UNEP. The main limiting factor in the developing world to advance more LCA based policies and applications is the lack of necessary capacity to understand, develop and use life cycle (LC) based approaches towards their use for strategic purposes. Thus, capacity building activities for the implementation of recommended policies and tools by the public and private sectors are fundamental. The 3-year project on ‘Enabling companies and consumers to benefit from information on life cycle environmental performance of products choices’ started in March 2012 and is being implemented by UNEP. Its aim is to promote quantification and assessment methods of LC impacts and a management system and their use and to facilitate the agreement on product sustainability communication principles. The project will be implemented in Latin America, Africa and Asia through the following: a) Capacity building on LC based quantification and assessment tools; b) Capacity building for the development of life-cycle data for the main export commodities and products; c) Pilot cases on LCM in business and governments in three countries; and d) consensus building and dissemination of internationally recognised principles for communication. The target groups include: i) governments to create the relevant enabling policy framework, iii) standard-setters to participate in international negotiations, iii) business and industry to implement such tools; and iii) academics and other training institutions to continue building capacity. 377 The role of consequential LCA for private and public decisionmaking: status-quo and perspectives E. Benetto, CRP Henri Tudor / Resource Centre for Environmental Technologies (CRTE); M. Guiton, Resource Centre for Environmental Technologies (CRTE); A. Marvuglia, CRP Henri Tudor / Cork Constraint Computation Centre; I. Vázquez-Rowe, CRP Henri Tudor / Resource Centre for Environmental Technologies CRTE; E. Igos, S. Rege, CRP Henri Tudor / Resource Centre for Environmental Technologies (CRTE); E. Popovici, Public Research Centre Henri Tudor / Resource Centre for Environmental Technologies (CRTE); J. Garcia, SCORE LCA. Since a decade, consequential life cycle assessment (cLCA) approach is preconized to comprehensively assess the consequences, in terms of induced environmental impacts, of human driven actions, most often related to policy or strategic decisions. cLCA has been defined as “the attempt to describe how environmentally relevant flows will change in response to possible decisions” or “how micro-economic decisions do influence macro-economic scenarios”. While in principle cLCA adopts a completely different perspective than conventional (attributional) LCA, in practice, the type of questions which can be answered by cLCA, as well as the operational approach, data needs, limitations, uncertainties and finally its practicability remain dramatically unclear to most of practitioners in industry and to policy makers. At the current stage of development of cLCA, there is clearly a gap between academic research on one side and stakeholders needs and understanding on the other side. The SCORE LCA association, including leading industry players (EDF,

SETAC Europe 23rd Annual Meeting

GDF SUEZ, Renault, Saint-Gobain, Total, Veolia) and the French Environmental Protection Agency (ADEME), together with the CRP Henri Tudor initiated a research study aiming at contributing to fill these gaps. The first step of the study was to draw a comprehensive and independent status quo of cLCA, as compared to the other LCA perspective, based on a literature review and on the position of cLCA experts interviewed via a detailed questionnaire. The aim of the statusquo was to position the different LCA perspectives around a number of criteria (pertaining e.g. to the decision context, the modelling approaches, ...) and thus highlighting methodological inconsistencies, gaps and further research challenges. The executive synthesis and overall conclusions are currently being harmonized. However, it can already be inferred from this study that cLCA is certainly a meaningful concept to address large scale decision processes in the studied sectors (at industrial and policy level) but, at the present state of development, lacks of operability and harmonization. The main research challenges to be addressed are the explicit definition of the decision context (and its boundaries), the position of the change(s) to be analysed along the time horizon, the availability, transparency and reliability of consequential inventory data and finally the communication of the results, eventually in relation to aLCAs of the same product system. 378 Application of three independent consequential LCA approaches to the agricultural sector in LuxembourgI. VázquezRowe, CRP Henri Tudor / Resource Centre for Environmental Technologies CRTE; A. Marvuglia, CRP Henri Tudor / Cork Constraint Computation Centre; J. Thenie, A. Haurie, ORDECSYS; S. Rege, E. Benetto, CRP Henri Tudor / Resource Centre for Environmental Technologies (CRTE). Consequential LCA (C-LCA) is widely used to link micro-economic actions with macro-economic consequences, by identifying the marginal suppliers and technologies prone to be affected by large scale changes in the demand of a specific product. Moreover, it has shown to be useful to monitor environmental consequences linked to bioenergy production. However, detecting the marginal technologies subject to suffer indirect effects due to changes in the production system is not an easy task. Hence, to provide a C-LCA assessment, researchers have combined LCA with different econometric models, by adapting the principle of C-LCA to the specific characteristics of the analysed system. For instance, some studies suggest the use of dynamic economic models to identify possible consequences occurring at the agricultural stage of a given production system. On the contrary, other perspectives advocate for developing a simplified model of the related effects in CLCA to obtain manageable systems, based on default assumptions and decision nodes.\nThe main aim of this study is to assess one single case study linked to biocrops cultivation using three different C-LCA model approaches. Firstly, an economic modelling tool, the partial equilibrium (PE) model is integrated with C-LCA to detect changes in land cultivation based on the farmers’ revenue maximization. This approach considers the agricultural sector as a closed system, allowing detailed representations of agricultural and land use restrictions. Secondly, another PE model perspective is proposed, which aims at minimizing a total adaptation cost (so called opportunity cost) to satisfy a given new demand of domestically produced biofuel. Finally, the consequential system delimitation for agricultural LCA approach, which seeks expert criteria to model the economic assumptions, was modelled.\nThe chosen case study aims at assessing the environmental changes in the agricultural sector in Luxembourg due to a predicted increase in maize cultivation to produce biomethane (projected based on the 2020 target set by the Luxembourgish Renewable Energy Action Plan). Obtained results are used to perform a discussion on the strengths and weaknesses of applying the different approaches, as well as identifying the consequences they may engender in the bioenergy policy. Discussion will also focus on the main drawbacks and assumptions that were needed in each approach to adapt their characteristics to those of the selected case study. 379 What is the right pathway to be sustainable? Case of biofuels and bioproducts in Europe S. Belboom, Department of Chemical Engineering - Processes and Sustainable Development; A. Leonard,

95

University of Liege / Departement of Chemical Engineering - Processes and Sustainable Development. Due to subsidies and the European promotion of biofuels, crops areas used for the production of bioethanol were increased through years inducing direct and indirect land use change. A lot of studies about the sustainability of biofuels were carried out and results are mitigated depending on methodological assumptions and used feedstock. Biofuels for transportation is the most common application of the bioethanol but the chemical field is also easily accessible. Indeed, bioethanol can be catalytically dehydrated to obtain bioethylene which is a monomer allowing the production of a multitude of plastics. This study presents the comparison of two applications of the same product, bioethanol, using the LCA methodology with the ReCiPe 2008 method. Bioethanol is produced in Europe, using sugar beet or wheat, cultivated on the already available land. No direct or indirect land use change was taken into account. This study tends to evaluate the best use of the already accessible crops, not the expansion of the market. The final products taken into account are the biofuel E5 (5% of bioethanol in volume) and the High Density PolyEthylene. Agricultural areas in Western Europe are limited and then it is very important to dedicate their use to the best application, i.e. with the highest gain for climate change and fossil fuel reduction. The functional unit is then the use of one hectare of cultivated land. Results are shown as the difference between the impact of the life cycle of the amount of fossil fuels that can be replaced by the cultivation of one hectare of land and the impact of the life cycle of their agricultural equivalent. The comparison for climate change and fossil fuel depletion categories between both crops gives an advantage to sugar beet when the comparison is performed by hectare. This is due to the lowest yield of wheat per hectare. When comparing both applications, the gain for climate change and for fossil fuel is higher in both cases when using HDPE. This study showed, with a short example, that LCA methodology can be a powerful tool to assess the best application of the same product. In our case, the policy used for the promotion of biofuels permits to achieve a reduction of GHG emissions and resources consumptions but there is a more performant application which is the chemical use. As it was already the case for the waste field, a hierarchy of the use of land and agricultural products should be developed. 380 Metals-induced temperature sensitivity in natural waters: Will chronic pollution amplify the effects of global warming on aquatic organisms? M.L. Brooks, Aquatic Ecology Environmental Chemistry Lab / Dept of Zoology; T. Hallman, Oregon State University Department of Fisheries and Wildlife. Evaluating the combined impacts of climate change and other anthropogenic disturbances on aquatic organisms is a tremendous ecological challenge. Metabolic oxygen demand from pollutants (e.g. elimination, sequestration) compounds the elevated metabolism of ectotherms at higher temperatures. Animals stressed by commonly observed pollution levels will have lower pejus temperatures (T ). Temperatures beyond T p p inhibit growth because costs from cellular oxygen demand (maximal aerobic efficiency) exceed oxygen availability. Methods. To consider risks of range contractions, Geographic Information System (GIS) coverages of the eastern United States (USA) combined current summer temperatures, current legally impaired waters (US EPA 303(d) listing), o and 2.5 C temperature projections (IPCC A2 scenario). In outdoor mesocosms and filtered lake water, we measured energetics (e.g. growth rates, size at age), metals bioaccumulation, and body condition of Cope’s grey tree frogs (Hyla chrysoscelis) from Gosner stage 25 to 42: 3 o temperatures: ambient, +1.5, +2.5 C x, 6 levels of Cd, Cu, and Pb mixtures x 4 replicates = 72 aquaria. Mixtures were bioavailable cumulative criterion units (BCCU)—the sum of free ionic activity per the criterion value as free ionic activity. Thus, BCCU are multiples of the protective criterion; herein 3.7 to 30 BCCU—10 BCCU is below USA drinking water standards. The 3.7 BCCU “control” was collected within a nature preserve. Adult frogs from the control and 30 BCCU treatments underwent 20-d, post-metamorphosis feeding trials. Results. In waters >10-fold BCCU. In >10 BCCU, temperature explained 44% of variance and tissue Pb explained another 18%. Initial size at metamorphosis did not differ between 3.7 and 30 BCCU metamorphs. After 20-d feeding

SETAC Europe 23rd Annual Meeting

o

trials, body condition of adults experiencing 30 BCCU and +2.5 C as larvae declined by 200% relative to those raised in 30 BCCU without thermal stress as larvae. Control adults did not significantly differ among treatments. Implications. We consistently observed lower T as p altered growth for animals stressed by metals >10 BCCU. Thus, animals had different temperature tolerances depending on water quality. Data indicate that pollution will compound the energetic costs of global warming, causing range contractions among aquatic organisms in the geographic center of the Midwestern USA. 381 Effects of competition and repeated exposure to a pesticide on populations of Daphnia magna I. Dolciotti, UFZ Leipzig / System Ecotoxicology; K. Foit, UFZ Helmholtz Centre for Environmental Research; M. Liess, UFZ Center for Environmental Research / Department of System Ecotoxicology. In the evaluation of the toxicity of substances on aquatic non target animals, test organisms are exposed either to a short contamination of 24 or 48 hours or to a chronic one in which the test system is renewed every day to maintain a constant concentration.However, standard toxicity tests do not represent entirely the way aquatic organisms are exposed to contaminants in situ. In the field we rather observe sequential pulses of contaminants due to drift or run off. Large amounts of pesticide are applied to agriculture area each year with much of it washed into adjacent water bodies. Moreover populations in a community context experience intra and interspecific competition, interactions that influence the way a population is shaped and likely also the way a population respond to a contaminant. In our experiment we studied how repeated insecticide pulses affect populations of Daphnia magna in the presence of the important ecological factor competition. Populations of Daphnia magna were exposed to two pulses of the insecticide pirimicarb (3, 10, 24 µg/L). Each pulse was followed by a recovery period of 28 days. In half of the testsystems, mosquito larvae (Culex pipiens) were added to exert a strong competition pressure. Population abundance and biomass were quantified by image analysis. Daphnia magna showed an acute mortality up to 80% at the highest treatment concentration while Culex pipiens larvae were insensitive at all concentration levels. The acute mortality of Daphnia was not increasing from the first to the second pulse. The sufficiency of recovery time after each pulse was depending on the presence of competition: without Culex larvae, populations of Daphnia recovered fast and abundance repeatedly reached control level within the recovery time of 28 days. With Culex larvae, populations of Daphnia failed to recover, after the second pulse and one population became extinct. Our experiment revealed that competition might delay the recovery of Daphnia magna populations, a common test organism in ecotoxicology that is known to recover fast. We conclude that a high competition pressure combined with repeated pulses of pesticides might explain unexpected long-term effects observed in the field. 382 Effects of silver nanoparticles on the functioning of estuarine bacterial communities V. Echavarri, HeriotWatt University / Centre for Marine Biodiversity and Biotechnology, School of Life Sciences; L. Paterson, Heriot-Watt University / School of Engineering and Physical Sciences; T.J. Aspray, Heriot-Watt University / School of Life Sciences; M. Hartl, HeriotWatt University / Centre for Marine Biodiversity and Biotechnology, School of Life Sciences. The rise of nanotechnology has led to an increase in the manufacturing and use of new materials that, at the nano-scale, exhibit different characteristics compared to their respective bulk material. Nano-silver is one of these materials, and, owing to its antimicrobial properties, is being incorporated in a wide variety of health and personal care products. As much of the silver is disposed of through domestic waste water, the question arises whether accumulation in the receiving estuarine environment could negatively affect the functioning of resident bacterial communities that play an important role in biogeochemical processes, providing important ecosystems services. The fate of AgNPs and their impact on highly dynamic estuarine ecosystems is still unclear. In the present study a microcosm approach was established in order to model the estuarine conditions to carry out a more realistic study of the effects of AgNPs on bacterial communities. It focused on the effects of AgNPs on bacterial

96

community functioning in water and sediments. The viability and abundance of bacteria was analysed using plate counts. Effects on the community metabolic profile was assessed based on the utilisation of different substrates and oxygen uptake. To assess the effects on the water quality, inorganic nitrogen and chemical oxygen demand (COD) were monitored. The results obtained revealed a significant reduction in the bacterial abundance was observed owing to the exposure to AgNPs which was reflected on the oxygen uptake rate. However the quality of the water was not affected based on the analysis of COD, nitrate and nitrite. The AgNPs used exhibited greater toxicity than those used in similar studies. This toxicity depended on their physicochemical characteristics, their interactions with the components of the environment in which they were released and the organism targeted. Understanding these aspects is essential for meaningful environmental risk assessment and regulation of an emerging technology. For this reason the AgNPs used in the exposures have been characterized under different salinity gradients. We conclude that, even though a negative impact on the abundance and viability of bacteria and also in the oxygen uptake was detected, this did not lead to a decrease in the water quality. This approach could enable the investigation of the acute toxicity of similar nanoparticles on estuarine and coastal bacterial communities in a cost-effective way. 383 Microecosystems "bryophyte-microorganisms": a good tool for ecosystemic ecotoxicology C. Meyer, Laboratory of ChronoEnvironment; D. GILBERT, N. BERNARD, University of Franche-Comte / Laboratory of Chrono-Environment. A new challenge for modern society is to understand the effect of pollutants on ecological integrity of ecosystems. This concept is used in several legal agreements on environmental protection and is defined by the ability of an ecosystem to support and maintain a community of organisms that as species composition, diversity and functional organisation (Parrish et al., 2003). Because the study of the impact of pollutants on ecosystems is complicated, microecosystems "bryophyte-microorganisms" could be good realistic models to understand the impacts of contaminants on ecosystems. Bryosystems constitute whole ecosystems integrating aboveand belowground linkages that can be used in the study of environmental pollution on biodiversity and ecosystem functioning. For many years, we highlighted the potential of these microecosystems to assist in the study of the response of the microbial communities' structure, functional groups and the species composition of top predators to environmental disturbances i.e. particulate, particulate + NO and 2 PAH atmospheric pollutions in situ and under controlled conditions. Our results showed that atmospheric contaminants have a direct effect on microbial species (decrease of biomass) and indicated that indirect effects exist through modifications of the interactions between microbial communities or between microbial communities and mosses. Among predators, testate amoebae (TA) are more affected (decrease of diversity or biomass of TA with superior trophic level) suggesting a decrease of the pressure of predation on lower trophic level, that is confirmed by the modification of the microbial communities' structure. Bryosystems are a good tool to evaluate the impacts of atmospheric pollutants on community ecology and it is now important to well understand the effects of these pollutants on microecosystem ecology (fluxes of energy and material among functional groups and their environment) and on food web ecology (structure and dynamics of species' feeding relationships). Thus, the use of such microecosystems could be reveal central and relevant information on anthropogenic disturbances at ecosystem scale. 384 Comparison of abundance and trait responses of aquatic macroinvertebrate exposed to fungicide or hydrocarbon mixture in stream and pond mesocosms Y. Bayona, INRAAgrocampus Ouest / UMR 985; M. Roucaute, INRA / UMR ESE; K. Cailleaud, Total Petrochemicals France / PERL; A. Basseres, TotalFinaElf; L.L. Lagadic, INRA / UMR INRAAgrocampus Ouest Ecology and Ecosystem Health; T. Caquet, INRA / UMR ESE 0985. Complex responses may be observed at the community level following exposure to chemicals due to a combination of direct and indirect effects affecting structural features

SETAC Europe 23rd Annual Meeting

and functional processes. The nature and intensity of these effects depend on various factors including the characteristics of the ecosystem and the traits of exposed taxa. Community recovery may be more rapid in lotic than in lentic systems due to drifting of individuals from upstream. Conversely, lentic invertebrate species may be more adapted to stressful conditions such as low dissolved oxygen concentration. To address all these effects Ecological Risk Assessment (ERA) has to include endpoints that are more closely linked to ecosystem functioning than taxa abundances such as those involving biological and ecological traits. The purpose of the present study was to compare the effects of a dithiocarbamate fungicide (thiram) and a hydrocarbon mixture on pond and stream mesocosms. In both types of mesocosm, nominal exposure concentrations were of 35 and 170 µg/L for thiram whereas nominal hydrocarbon mixture concentrations were 0.01, 0.4, 2 and 20 mg/L. All treatments were performed in duplicates and four systems were kept as controls. Flow-through outdoor streams (length: 40 m, depth: 50 cm) were continuously treated for 3 weeks. Outdoor pond mesocosms (7.5 3 m ) were treated weekly for 4 weeks. The exposure period was followed by a 2- or 10-month recovery phase for stream and pond mesocosm, respectively.Macroinvertebrates were sampled using artificial substrates weekly or every 3 weeks in stream and pond mesocosms, respectively. Alder litter breakdown rate was measured using litter bags. Changes in the structure of the invertebrate community were analyzed by the PRC method and the ones in the trait structure were analyzed using the RLQ method. Hydrocarbon effect on communities was assessed by abundance and biological traits of macroinvertebrates and by litter breakdown rate, with recovery in most case. In hydrocarbon exposed mesocosm, effects only occur for concentrations above the toxic threshold assessed in MSDS. Fungicide impacted all measured endpoints without recovery in streams whereas only effect on biological traits was assessed in pond mesocosms. Finally, recovery depends of mesocosms and type of chemicals. Both differences in term of impact measurement and recovery potential (structure and traits) in dynamic and static mesocosm will be discussed in presentation. 385 Ionic silver can directly and indirectly impairs the functioning of detritus-based aquatic ecosystems J. Arce Funck, Laboratoire Interdisciplinaire des Environnements Continentaux; V. Felten, H. Clivot, F. Guerold, M. Danger, Universite de Lorraine / Laboratoire Interdisciplinaire des Environnements Continentaux. Silver (Ag) is a very potent biocide found in natural water at concentrations ranging -1 from 0.03 to 500 ng.L , but it has insufficiently been studied compared to other metals. The increasingly use of Ag nanoparticles can raise the risks for aquatic system as they can salt out ionic Ag, potentially affecting the organisms and in turn the whole ecosystem functioning. Forested headwater streams rely mainly on the decomposition of allochthonous organic matter as a source of nutrient and energy. In these ecosystems, initial stages of detritus decomposition are mainly ensured by aquatic hyphomycetes, followed by the degradation mediated by detritivores invertebrate (e.g. Gammarus fossarum). The successful detritus processing by both trophic levels is essential for the optimal functioning of detritus-based ecosystems. While the interactions between leaf litter, decomposers, and detritivores can be impaired by some metals, very little is known about Ag impact. We conducted two microcosm experiments to evaluate: (1) Ag effect on fungal decomposition process, and (2) its indirect impact on G. fossarum feeding rate, but also (3) G. fossarum Ag LC50s (according to gender and development status) and (4) Ag effect on G. fossarum physiological and behavioural biomarkers. Decomposition process, fungal community structure and gammarids feeding rate were significantly impacted only -1 at the highest [Ag] (100 µg.L ). 96h-LC50s indicated that juvenile -1 gammarids are extremely sensitive (1.01 µg Ag.L ). Strong physiological and behavioural disturbances were registered on male G. -1 fossarum, as soon as 96h-exposure to 0.5 µg Ag.L . Compared to control, Ag exposure lead to higher mortality, Ag bioconcentration, SeGpx activity and lipid peroxydation, but also to strong iono-/osmo+ regulation disruption (lower haemolymph osmolality, [Na ] and [Cl ]) and behaviour disturbance (lower ventilation and locomotion). This study shows that Ag could significantly affect the functioning of

97

forested headwater streams. In natural water, it seems unrealistic that dissolved Ag directly affects microbial detritus processing (significant -1 effect only at 100 µg Ag.L ) but realistic [Ag] had direct effects on the physiology, the behaviour and the survival of a key detritivore, suggesting a probable indirect negative impact on detritus processing at realistic concentrations. The release of ionic Ag from nanoparticles increasingly used should thus certainly deserve more attention in the next few years. 386 Detection, Monitoring, and Control of Tributyltin - An Almost Complete Success Story P. Matthiessen, Independent Consultant. 387 Top three myths about TBT

M. Waldock, Cefas.

388 Not running afoul with antifouling paints - the next generation of protection K. Long, Regulatory Compliance Ltd / Bilston Glen Business Centre. 389 Panel discussion

E.M. Mihaich, .

390 Estrogens in the aquatic environment: From science to regulation J.P. Sumpter, Brunel University. 391 Aquatic environments and endocrine activity: The regulatory approach in Japan T. Iguchi, National Institute for Basic Biology / Molecular Environmental Endocrinology. 392 Endocrine active chemicals in the environment - can we determine acceptable risks? S. Marshall, Unilever / SEAC. 393 Panel discussion

E.M. Mihaich, .

394 Effects of Carbo-Iron nanomaterial on the metabolome of the chlorophyte Scenedesmus vacuolatus F. Sans Piche, Helmholtz Centre for Envrionmental Research – UFZ / Dept. Bioanalytical Ecotoxicology; O. Frank, W. Busch, Helmholtz - Centre for Envrionmental Research – UFZ; D. Kühnel, HelmholtzCentre for Environmental Research; M. Schmitt-Jansen, UFZ Helmholz Ctre Environm Research / Dept. Bioanalytical Ecotoxicology. Carbo-Iron nanomaterial has been developed as resource-saving and effective groundwater and sewage remediation technology.Before usage of CarboIron nanomaterial in the environment, their ecotoxicological potential should be assessed. Therefore, the aim of this study was to assess the effects of Carbo-Iron on the metabolome of the unicellular microalga Scenedesmusvacuolatus by applying a metabolomics approach modified after Kluender et al. (2009). Next to modifications of the protocol for testing nanoparticles, two types of materials, Carbo-Iron (CI) and activated carbon particles (AC) were tested. The results revealed that the particles do not interfere with the metabolomics protocol. The two particles tested showed comparable decrease of light intensities during exposure (shading) and growth inhibition of algae. Principal component analysis performed on polar metabolite extracts revealed concentrationdependent metabolic changes induced by both particles and a relatively lower amount of most metabolites compared to controls but did not show a difference between CI and AC. Sugars such as sucrose, ribose and mannose decreased indicating a disturbance of the energy metabolism of the cells. The results will be discussed in perspective of the biochemical changes induced by CI beyond a shading effect. 395 Effects of iron nanoparticles on Physcomitrella patens (Hedw.) Bruch & Schimp. L. Canivet, University of Lille; P. Dubot, 2MCMC – ICMPE UMR 7182; G. Garcon, R. Courtecuisse, University of Lille 2; F. Denayer, Universite de Lille Droit Sante. With the emergence of nanotechnology and the increasing use of nanoparticles, a question emerges: “what are the effects of this nanoparticles on ecosystems?” To try to answer to this question, the objective of our study was to evaluate the genotoxic and cytotoxic effects and the modulation of gene expression in a bryophyte, Physcomitrella patens, (Hedw.) Bruch & Schimp. exposed to iron engineered nanoparticles. Nanoparticles can be

SETAC Europe 23rd Annual Meeting

dispersed and transported by wind over long distances. Plants are the first exposed to nanoparticles. To date, plants are widely used as biological model to study impacts of pollution on ecosystems and studies on nanotoxicity in plants develop in recent years. But no study has been made on the impact of nanoparticles on bryophytes. However, bryophytes are specific organisms that could be used to evaluate the impact of atmospheric pollutants on the environment. So, to study the impact of nanoparticles on ecosystems, we interested to these effects on Physcomitrella patens. Moreover, Physcomitrella patens has a particular interest because its genome has been completely sequenced. Then, the chemical composition of nanoparticles plays an important role in their ecotoxicity. The novel properties of nanoparticles such as their smaller size, large surface area and higher reactivity are a major point of concern. So, an essential step was the characterization of iron nanoparticles in their pristine state but also in their dispersion medium. In our study, we exposed plants of Physcomitrella patens to ironengineering nanoparticles, dispersed by atmospheric way, under the most realistic conditions. Before studying the effects of iron-engineering nanoparticles, we verified the uptake capabilities of these nanoparticles in bryophyte cells by confocal microscopy. Then, a wide range of chemical and biological variable was analysed to determine the nanoparticles effects: the viability, the cytotoxicity, and the genotoxicity. Moreover, to explain the nanoparticles effects further upstream, we studied the modulation of the expression of some genes involved in oxidative stress. Perspective, with best knowledge of engineering nanoparticles impacts, we will continue this project by focusing on environmental nanoparticles collected on an industrial site, whose characterization will be as complete as possible. Key words: Physcomitrella patens, nanoparticles, ecotoxicity, realistic conditions 396 Effect of a set of different TiO2 and ZrO2 (nano)particles in soils S.I. Gomes, University of Aveiro / department of Biology & CESAM; N. Pinna, University of Aveiro / Department of Chemistry & CICECO; J.J. Scott-Fordsmand, Aarhus University / Department of Bioscience – Terrestrial Ecology; M.J. Amorim, Universidade de Aveiro / Department of Biology and CESAM. Titanium dioxide (TiO ) particles 2 is one of the most used nanomaterials wordlwide, used e.g. in sun lotion and tooth paste. It is also used because its photocatalitical activity enhanced by UV is being explored for diverse applications. Zirconium dioxide nanoparticles (ZrO -NPs) are also widely used, e.g. in catalysis 2 and ceramics. Given this, the release of these NPs in the enviroment will occur and organisms may be exposed. However, little is known about the effects of these particles on terrestrial organisms. Enchytraeids (Oligochaeta) are important members of the soil community, there is a standard (OECD, 2004) test-protocol for these organisms, and they are able to survive both in soils and water, which makes them important and usefull organisms for comparing exposure routes. In this study we assessed the effects - on the enchytraeid Enchytraeus crypticus (Oligochaeta) - of i) 4 different TiO NPs (with various sizes, coatings 2 and crystal structures) in comparison with bulk TiO , and ii) one ZrO 2 2 NPs in comparison with the respective bulk (micro) and ZrCl (salt). 4 Effects were assessed via soil exposure (OECD, 2004) at survival and reproduction level and via reconstituted water and soil extracts. The effects of UV irradiation on TiO toxicty were measured by the 2 combined exposure of the organisms to several TiO concentrations and 2 UV irradiation (tested to be non-toxic alone). TiO and ZrO 2 2 nanoparticles did not affect E. crypticus survival and affected reproduction only to a imited extent, when exposed in soil. An approach with possible increased bioavailability (water exposure) revealed similar results for survival. Realistic worse case scenario of UV exposure did not potentiate (increase) the toxicity of TiO -NPs. Keywords: 2 nanoparticles, Ti, Zr, Ecotoxicity, soil organisms 397 ZnO nanoparticle toxicity at different soil pHs – survival and biomass change in the isopod Porcellionides pruinosusP.d. Tourinho, University of Aveiro / department of Biology & CESAM; C. van Gestel, Vrije Universiteit Amsterdam / Ecological Science; A.M. Soares, Universidade de Aveiro / Department of Biology and CESAM Centre for Environmental and Marine Studies; S. Loureiro,

98

Universidade de Aveiro / Biology. Environmental conditions are well known to affect the stability of nanoparticles (NPs) in soils. When it comes to soil properties, pH is one of the most important factors as it can change NP surface charge and zeta potential. The aim of this study was to evaluate the effects of soil pH on the toxicity of ZnO NPs to the terrestrial isopod Porcellionides pruinosus, using a natural soil with three different pH values and a natural standard soil (Lufa 2.2 soil). The toxicity test was also conducted with other two Zn forms: a microsize ZnO (non-nano ZnO) and ZnCl . A forest soil was collected and 2 amended with calcium carbonate to adjust pH to approximately 4.5 (Soil 1), 5.9 (Soil 2), and 7.3 (Soil 3). The Lufa 2.2 soil was also used in the experiment with a pH of 5.1. Soils were spiked at nominal concentrations of 250, 500, 1000, 2000 and 4000 mg Zn/kg dry soil. Adult isopods were exposed individually, in 10 replicates per treatment and a control for each pH set. Animals were feed with alder leaves ad libitum. After 14 days, survival, feeding inhibition and biomass gained were determined. LC50 and EC50 on biomass gained were calculated with a logistic curve. LC50 values were compared by a generalized likelihood ratio test. Zinc body content in isopods were analysed by a two-way analysis of variance (ANOVA). Soil pH, calcium and zinc levels in pore water changed in a dose-related manner, showing a great difference between soils spiked with ZnO particles (ZnO NPs and nonnano ZnO) and ZnCl . LC50 values differed significantly between soils 2 only for ZnO NPs and ZnCl . Food consumption decreased in isopods 2 exposed to ZnO NPs and non-nano ZnO only in Soil 3. For ZnCl , 2 consumption ratio decreased in all tested soils. Effects on biomass (EC50) showed a wide range across soils, with higher values in Soil 2, suggesting that soil properties do affect the Zn toxicity in a similar manner for all Zn forms. Zinc body content in the isopods was affected by Zn concentration in soil for ZnO NPs and ZnO non-nano, but did not differ between soils. For ZnCl , zinc body content was significantly 2 affected by both Zn soil concentration and soil type. These results indicate that isopods exposed to ZnO NPs were able to store similar Zn quantities independent on the soil, but still the effects occurred at different soil concentrations. Toxicity outcomes were quite comparable between Zn forms, indicating that ZnO NPs toxicity could not be caused 2+ only by Zn ions from NPs dissolution. 398 Transcriptome analysis of zebrafish embryos exposed to dendrimers E. Oliveira, IDAEA CSIC / Environmental Chemistry; M. Casado, Institute of Environmental Assessment and Water Research (IDAEA-CSIC) / Environmental Chemistry; A.M. Soares, Universidade de Aveiro / Department of Biology and CESAM - Centre for Environmental and Marine Studies; J. Navas, INIA - Madrid; C. Barata, CSIC / Environmetal Chemistry; B. Pina, IDAEACSIC / Environmental Chemistry. The recent aplication of dendrimers into nanomedicine field for diagnostic imaging, drug delivery and gene transfection results in an increased release of those biomaterials into the environment. Therefore, the development of appropriate methodologies for their monitoring and evaluation of their potencial toxicity becomes increasingly important. Poly(amidoanime) dendrimers have a well-defined, mono dispersive and stable molecular architure, composed of an ethylenediamine core initiator and successive layers of radially repeating units attached to the core and outer surface of terminal functional units. Dendrimers are envisioned as attractive drug carrier systems, largely considered nontoxic, although there is little information about their health risks in humans and other vertebrates. We present here a toxicological and transcriptomic approach using zebrafish as a model organism to the study of PAMAM third-generation (G3) and fouth-generation (G4), NH2 terminated, water soluble dendrimers. The data reveals similar toxicity profiles for both molecules, and suggests that the nervous system may be particularly sensitive to these molecules. In addition, a more generic foreign body reaction pattern appeared also activated, suggesting the potential for immflamatory and/or allegenic responses. The development of a system to monitoring toxicity of newly developed macromolecules is of great interest to assess the possible effects to developing vertebrates, including humans. New regulations may be introduced in terms of safety and efficacy of these new products, since they affect important metabolic pathways concerned to de development

SETAC Europe 23rd Annual Meeting

and also immunological system. 399 Modelling plant uptake of organic compounds using molecular properties M. Limmer, G. Morrison, Missouri University of Science and Technology; J.G. Burken, Missouri University of Science and Technology / Civil Architectural and Environmental. Uptake of xenobiotics such as pharmaceuticals by plants has received recent attention. Traditional models to explain plant uptake of organic compounds have utilized octanol-water partitioning (log K ) to OW explain empirically the transpiration stream concentration factor (TSCF). However, these single-parameter models vary between researchers and provide little insight into the fundamental mechanisms underlying plant uptake of organic chemicals. This presentation will describe recent, fundamental work by our lab to develop rule-based and predictive models that can better explain the variability in measured TSCFs for a wide variety of compounds such as pharmaceuticals and pesticides. We found that compounds translocated by plants followed Lipinski's Rule of 5. (His “Rule of Five” states such a compound generally has 5 or fewer hydrogen bond donors, 10 or fewer hydrogen bond donors, a molecular mass of less than 500 Daltons and a log octanolwater partitioning coefficient (log K ) of less than 5.) A series OW resistance model of symplastic and apoplastic transport of contaminants by plants was developed to quantify a compound's ability to enter a plant using molecular properties such as polar surface area, molecular volume and hydrophobicity. These models indicate that a wide variety of compounds, such as pharmaceuticals, could enter plant roots and be translocated above-ground. 400 Uptake of organic compounds by submerged, freshwater, sediment-rooted macrophytes N. Diepens, Wageningen University; A. De Agustin Camacho, Alterra; G. Arts, Alterra Wageningen University and Research Centre / Environmental Risk Assessment; A.A. Koelmans, Wageningen University / Environment. Historically, plantchemical interaction was mainly focused on terrestrial and emergent plants while submerged aquatic macrophytes received less attention. Furthermore, only the water phase was considered. The AMRAP (Aquatic Macrophyte Risk Assessment) workshop identified a lack of knowledge about the relative importance of sediment exposure for the uptake of toxicants by sediment-rooted macrophytes. Translocation of contaminants from sediment into the water column via the macrophytes may be a significant pathway and might be important in hazard assessment. Despite the fact that several simulation models have been developed, none has considered all relevant aquatic compartments including sediment, water and plants. The aim of the current work is to provide estimates of parameters that drive uptake, translocation and elimination of organic compounds in sediment-rooted, submerged aquatic macrophytes, and to apply these parameters in chemical uptake models for macrophytes. The approach includes the generation of experimental datasets in combination with modelling of the processes mentioned above. The results from the previous water only uptake studies can be used as extra data in the model and interpretation of the data in an ecological context. Optimum growth tests were done using Myriophyllum spicatum and Elodea canadensis following the AMRAP set-up. These species differ in their traits such as leaf area and transport mechanisms. Furthermore, uptake, elimination and translocation of sediment-spiked chemicals via roots and shoots was studied during 28 day sediment toxicity tests. The spiked chemical included a mixture of PCBs and an insecticide (chlorfyrifos). For Elodea canadensis total dry weight did not markably differ among the OM treatments. For Myriophyllum spicatum biomass showed a tendency to be higher at lower OM percentages (Fig. 1). The same pattern was observed in the total length of the Myriophyllum plants. The higher nutrient concentrations of N:P 3:1 and 6:1 showed a higher Chlorophyl-a content in the water layer (more algae), less growth in macrophyte shoots, less side shoots and no or less developed macrophyte roots in the sediment. Growth on sediment without nutrients was lowest, with the other nutrient levels in between. Furthermore, results of the the uptake, translocation and elimination of PCBs and chlorpyrifos by the aquatic macrophytes will be presented.

99

401 Hg bioaccumulation in shoots of the macrophyte Elodea nuttallii: an opportunity for phytoremediation or a risk for the trophic chain? C. Cosio, Geneva University. Previous studies suggest that macrophytes might participate in bioaccumulation and biomagnification of toxic mercury (Hg) in aquatic environment. Hg bioaccumulation and uptake mechanisms in macrophytes need therefore to be studied. Amongst several macrophytes collected in an Hg contaminated reservoir in Romania, Elodea nuttallii showed a high organic and inorganic Hg accumulation and was then further studied in the laboratory. Tolerance and accumulation of Hg this plant was also high in the microcosm. Basipetal transport of inorganic Hg was predominant, whereas acropetal transport of methyl-Hg was observed with apparently negligible methylation or demethylation in planta. Hg concentrations were higher in roots>leaves>stems and in top>middle>bottom of shoots. In shoots, more than 60% Hg was found intracellularly where it is believed to be highly available to predators. Accumulation of inorganic and methyl-Hg in shoots was highly reduced + by cold, death and by competition with Cu for inorganic Hg. Hg in E. nuttallii shoots seems to mainly originate from the water column, but methyl-Hg can also be remobilized from the sediments and might drive in part its entry in the food web. At the cellular level, uptake of Hg into the cell sap of shoots was linked to the metabolism, in particular to copper transporters. The present work highlights an important breakthrough in our understanding of Hg accumulation and biomagnifications: the remobilization of methyl-Hg from sediments to aquatic plants and differences in uptake mechanisms of inorganic and methyl-Hg in a macrophyte. Our results revealed that shoots of E. nuttallii are extremely tolerant to Hg and show huge bioaccumulation ability. Both might be interesting for phytoremediation of contaminated water. Based on densities previously reported in the field (350-2800 g -3 DW m ) and concentration of Hg found in Babeni reservoir, it can be -3 estimated that 0.7 to 5.6 mg Hg m could be removed with these plants. On the other hand, the Hg uptake by shoots might be an important pathway of Hg from the water to organisms feeding on the macrophytes. In conclusion, the presence of these plants should not be overlooked in contaminated sites. 402 Use of the Lemna minor growth inhibition test to study dose dependent effects of uranium in plants. n. Horemans, Belgean Nuclear Research Centre SCKCEN; M. Van Hees, Belgian Nucrlear research centre (SCK-CEN); A. Van Hoeck, belgiean Nuclear Research Centre (SCK-CEN); H. Vandenhove, Belgian Nuclear Research Centre, SCK-CEN / Biosphere Impact Studies. Anthropogenic activities can lead to an enhancement of radionuclides and radioactive radiation in the environment. To assess the impact of uranium on plants, a growth inhibition test was set up in which Lemna minor was exposed to different uranium concentrations ranging from 0.05 µM up to 500 µM for 7 days. The test was performed as described in the OECD guidelines (OECD 221, 2006) with some modifications towards the growth medium. The growth medium was selected so that uranium was predominantly present as soluble uranyl kations (modelled using Geochemist’s WorkbenchÒ) and so that it was sufficiently rich to sustain the growth of the Lemna plants. For the different media tested it was shown that adapted K-medium as described by Cedergreen et al. (2007 Envionm Toxic Chem 26:149-156) with low phosphate concentrations (0.05mg/L) best fulfilled the constraints for growth and uranium bioavailability. Uptake of uranium into the plants increased linearly with uranium dose. Growth of the Lemna plants was analysed as the total area of the plant fronds and the number of fronds. A clear discrepancy between these two endpoints assessed was noticed. For the frond area the dose response curve followed a typical sigmoidal shape with an EC around 75 µM. For the number of fronds, however, growth 50 inhibition stabilised between 20 and 100 µM uranium indicating the number of new fronds appearing was stable but new fronds were smaller. Above 100 µM growth rate inhibition based on frond number caught up with that expressed on frond area. Following the pH of the medium during the growth inhibition test it was noticed that plants were able to increase the pH with more than 2 pH units. As uranium

SETAC Europe 23rd Annual Meeting

speciation and hence its toxicity strongly depends on the pH experiments with and without MES (2mM) as buffer were compared. It was shown that in the presence of this MES concentration the pH maximally increased with 0,3 to 0,5 pH unit indeed resulting in a higher U toxicity (EC50 shifted to about 30µM). These results clearly indicate the dependency of the toxicity of uranium on the presence the uranyl kations which are the most prevalent uranium species at lower pH levels. 403 Developing good practice on the management of aquatic plants in watercourses – taking a risk-based approach O. Tarrant, Environment Agency / Flood Risk. In many watercourses the control of aquatic plants is an essential part of their effective management. However, we feel that there is currently insufficient guidance to support this good practice approach to the management of aquatic plants and riparian vegetation in and alongside watercourses. Over the last two decades – since the last published guidance document on this subject there has been significant advances in our understanding of the relative effectiveness of the techniques for aquatic plant control alongside significant changes in legislation particularly, the need to comply with the Water Framework Directive (WFD). Also there have been many calls from the operational flood risk management authorities, including the Internal Drainage Boards (IDBs) for further research and guidance particularly, to provide advice on herbicide application. We are aiming to synthesise the latest research to develop good practice guidance on the management of aquatic plants in, and vegetation alongside watercourses, through the comparison of a number of management techniques in different watercourses. This work will ensure that those who manage aquatic plants and vegetation in and alongside watercourses can carry out their work in a way that is cost effective, minimises negative impacts on the environment and is compliant with the governing legislation. The use of herbicide for aquatic plant management will be a strong research focus and will be placed within the wider context of the other available plant management techniques. The project is in its early stages so is yet to report in full. However, this paper offers is a timely update for the Environmental Toxicology and Chemistry community on the objectives and progress with this work. Session: Risk assessment, regulation, policy and public perception Key words: aquatic plants, flood, management, herbicide Presentation preference: Platform 404 FLOATING PENNYWORT IN THE RIVER SOAR, LEICESTERSHIRE P. Harding; r. Brunt, Environment Agency. The invasive non-native floating pennywort (Hydrocotyle ranunculoides) first appeared in the River Soar near Leicester City Centre in 2004, and has continued to spread downstream through Leicestershire and into Nottinghamshire. It has now become established in the River Trent, with the potential to spread more widely in the catchment. Monitoring by fixed-point and aerial photography has enabled semi-quantitative comparison of pennywort beds to track the invasion and assess the impact of control measures and natural factors. Efforts to manage the invasion in the River Soar have focussed on sustainable management through partnership in line with the GB Framework Strategy for invasive non-native species. Early management was largely by mechanical removal, which proved successful but resource-intensive. Using this approach Leicester City Council has progressively removed the plant in Leicester, demonstrating that topdown eradication is achievable. Downstream of Leicester the Canal & Rivers Trust (formerly British Waterways) and the Environment Agency have progressively replaced mechanical control with herbicide use, and this has proved both effective and cost-effective. Growth of floating pennywort is influenced significantly by temperature and flow. In a mild winter rafts may remain intact and consequently become displaced as large masses, potentially posing an increased risk of flooding. Conversely, severe air frosts cause fragmentation of the beds into small pieces which wash downstream as thousands of potentially viable ‘cuttings’. This scenario may have led to the sudden appearance of floating pennywort in the River Trent in 2009. Monitoring in 2010 did not indicate growth suppression following a cold winter, but exceptionally low temperatures in December 2010 appeared to eliminate

100

floating pennywort from several parts of the river until 2012. 405 Nanosilver effects in marine ectotherms: linking low and high informational levels to assess adaptation under realistic conditions F. Dondero, DiSIT; I. Saggese, University of Piemonte Orientale Amedeo Avogadro / DiSIT; V. Palmeri, Università degli Studi di Palermo. The rising use of nanomaterials in several applications including health care products and domestic uses- is posing new ecotoxicological concerns also for the marine environment. Here we propose a framework of analysis for the assessment of biological effects of engineered nano-objects (ENO) in the marine ectothermic organism Mytilus galloprovincialis, a filter feeding mollusk with great ability to accumulate trace contaminants. The framework included high and low order level effect determination, spanning from bioaccumulation, molecular and biochemical responses, bioenergetics, life trait history and fitness. Different nano-silver preparations (5 nm and 50 nm) were selected and further used in acute, chronic and microcosm exposures aimed to assess ecotoxicological endopoints and mechanistic effects of nano-silver. We provided short-term toxicity endpoints using actual doses for silver in marine bivalves; long-term microcosm study under realistic conditions; mechanistic effects of (nano)-silver linking low and high informational levels. In conclusion, or data underpin the hypothesis that the environmental risk due to silver released from engineered nanomaterial tends to be underestimated. 406 Antibacterial effect of Ag-nanoparticles: modulation of the dissolution at the particle-cell interface A. Kahru, Environmental Toxicology; O. Bondarenko, National Institute of Chemical Physics and Biophysics; A. Ivask, Nat Inst of Chemical Physics and Biophysics / Environmental Toxicoloy. Silver nanoparticles (AgNPs) are currently used as broad-spectrum antimicrobials in over 300 consumer products including cosmetics, clothing, detergents, dietary supplements, water filters and children’s toys. It is generally acknowledged that the toxic properties of AgNPs to various aquatic organisms and bacteria are largely determined by their dissolution., i.e., release of toxic Ag-ions. The effective concentration of released Ag-ions, however, depends on various biotic and abiotic factors and is difficult to predict. In this study we thoroughly characterized three types of AgNPs with different coatings (uncoated AgNPs (nAg); casein-coated colloidal AgNPs (nAgCol) and polyvinylpyrrolidone-coated AgNPs (nAg-PVP) for their size, stability, z-potential, dissolution and antibacterial efficiency. Using three different techniques to determine the dissolution of AgNPs, we showed that the antibacterial effects of these AgNPs correlated with their dissolution. Specifically, using six bacterial strains (Bacillus subtilis, Staphylococcus aureus, Escherichia coli, Pseudomonas fluorescens, P. putida and P. aeruginosa) we demonstrated that the minimal inhibitory concentration, MIC; mg Ag/l, for uncoated AgNPs towards all bacteria was >100, for nAg-PVP from 20-40 (for S. aureus >100), for nAg-Col from 5 (P. aeruginosa) to 100, for AgNO from 2.5 to 40. To all 3 bacteria most inhibitory was AgNO (100% soluble), followed by nAg3 PVP (solubilized fraction 2?8%), nAg-Col (1.8?4%) and, finally, uncoated nAg (practically insoluble). The obtained results, however, revealed the specific role of particles in enhancement of toxicity of Agions. Specifically, the nanoparticulate formulations of silver improved the delivery of Ag-ions into bacterial cells whereas the efficiency was strain-specific, being highest for human pathogen P. aeruginosa intrinsically resistant to various conventional antibiotics. The separation of bacterial cells from Ag-NPs by particle-impermeable membrane (cutoff 20 kDa, ~4 nm) remarkably reduced the toxicity of Ag NPs, confirming the importance of direct contact between Ag-NPs and bacterial cells for enhanced delivery of Ag-ions. We assume that in the close vicinity of cell envelope the direct cell-particle contact lead to the enhanced solubilisation of AgNPs facilitating the enhanced internalisation of Ag-ions and we showed that this effect could not be always predicted from the conventional dissolution studies. 407 Species specific toxicity of copper nanoparticles: responses explained by particles or ions? M.G. Vijver, CML Leiden University; J. Hua, L. Song, Leiden University; W.J. Peijnenburg,

SETAC Europe 23rd Annual Meeting

RIVM / Laboratory for Ecological Risk Assessment. Various studies showed that CuNPs can cause a diversity of toxic effects to biological systems. CuNPs showed a size- and concentration-dependent toxicity. It is recently known that the toxicity caused by micro copper is lower than the toxicity of CuNPs and the toxicity caused by copper ions in CuNPs media and the toxicity of copper oxide NPs cannot be simply explained by Cu ions released to the cell medium. However, little attention has been paid to species-specific NP toxicity and only a limited number of studies have quantified the toxicity contribution of the particle form of NPs and released ions to the total toxicity of particle suspensions for varying NP-sizes. Therefore in this contribution, the basic research question was: what is the relative toxic contribution of Cu ions and particles inducing toxic effects in different aquatic organisms? We present experimental data on dose-responses of different sized copper nanoparticles among different cell lines. And we have preliminary findings on data for different fresh water invertebrates and on zebrafish development. Our results show that both chemical and physical characteristics of NPs contribute to their toxicity to living cell lines and that sensitivity is species-specific. For mammalian cell lines both the ion and particle forms of CuNPs in suspensions contribute significantly to total toxicity, while particulate forms are most important for piscine cell lines. Experiments using zebrafish developments (preliminary data) showed that all sizes of Cu suspensions had inhibitory effects on hatching success, behavioural responses, and malformations. Different kinds of abnormalities were observed in zebrafish embryos depending on the size of Cu suspensions. MP colloids induced 77% three-combined malformations at lower 0.5 mg/L nominal concentrations. This study also revealed that CuNPs contributed to the toxicity of copper suspensions mainly (71%-100%), contribution of CuNPs increased with decreasing concentration. Therefore our study shows that species sensitivity is likely to be found and that the relative contribution of ions versus particles differs among organisms groups. 408 Titanium dioxide nanoparticles increase sensitivity in the next generation of the water flea Daphnia magnaM. Bundschuh, Department of Aquatic Sciences and Assessment Swedish University of Agricultural Sciences / Department of Aquatic Sciences and Assessment; F. Seitz, Inst for Environmental Sciences / Institute for Environmental Sciences; R.R. Rosenfeldt, Uniersity of KoblenzLandau Institute for Environmental Sciences / Institute for Environmental Sciences; R. Schulz, University of KoblenzLandau / Institute for Environmental Sciences. The nanoparticle industry is expected to become a trillion dollar business in the near future. Therefore, the unintentional introduction of nanoparticles into the environment is increasingly likely. However, currently applied risk-assessment practices require further adaptation to accommodate the intrinsic nature of engineered nanoparticles. Combining a chronic flow-through exposure system with subsequent acute toxicity tests for the standard test organism Daphnia magna, we found that juvenile offspring of adults that were previously exposed to titanium dioxide nanoparticles exhibit a significantly increased sensitivity to titanium dioxide nanoparticles compared with the offspring of unexposed adults, as displayed by lower 96 h-EC values. This observation is particularly remarkable because 50 adults exhibited no differences among treatments in terms of typically assessed endpoints, such as sensitivity, number of offspring, or energy reserves. Hence, the present study suggests that ecotoxicological research requires further development to include the assessment of the environmental risks of nanoparticles for the next and hence not directly exposed generation, which is currently not included in standard test protocols. 409 Extrapolating population level effects from nanoparticle toxicodynamics A. Gergs, Roskilde University / Department of Environmental Social and Spatial Change; M. Pettitt, University of Birmingham, School of Geography, Earth and Environmental Sciences,; T. Dalkvist, Eawag / Department of Environmental Toxicology; T.G. Preuss, RWTH Aachen University / Institute for Environmental Research; R. Ashauer, University of York / Environment; E. ValsamiJones, Natural History Museum, University of Birmingham; A.

101

Palmqvist, Roskilde University / Department of Environmental Social and Spatial Change. Lethal and sub-lethal toxic effects are commonly determined in standard laboratory tests on the level of individual organisms or the sub-individual level. What these (sub-) individual-level effects mean at higher levels of biological organisation remains an open question. For bridging the gap, toxicodynamics as observed on the individual level have to be extrapolated to population level, which is the primary protection goal for environmental risk assessment. Toxicants differ in their biochemical and physiological modes of action which may translate into specific temporal patterns of growth and reproduction in individual organisms. In the present study we therefore tested the ultimate consequences of different theoretical physiological modes of action for population dynamics by means of an individual based population model (IBM). Within the IBM, assimilation of food and subsequent allocation of energy to growth, reproduction and survival in individual organisms is modelled based on formulations provided by the dynamic energy budget (DEB) theory. In response to a toxicant, energy allocation patterns change in particular ways according to the toxicant’s physiological mode(s) of action. Similar to real life, modeled population dynamics emerge from individual life histories, toxicodynamics and intraspecific competition for food and space. To unravel the physiological mode(s) of action and investigate lethal effects of silver nanoparticles (Ag-NP) and titanium dioxide nanoparticles (TiO -NP) in 2 Daphnia magna, we conduced a literature search on acute and chronic effects caused by the different nanoparticles and parameterized the DEB model accordingly. In order to simulate lethal effects we additionally implemented the general unified threshold model of survival (GUTS) and parameterized the model based on Ag-NP data available in the literature. Finally, by application of the IBM, we explored population level responses of sub-lethal and lethal effects found for individual D. magna exposed to Ag-NPs an TiO -NPs. We will discuss the 2 implications that different chemical modes of action bear for environmental risk assessment as well as the current potential for use of mechanistic effect models in toxicity assessment of engineered nanoparticles. 410 Toxicity of tungsten carbide nanoparticles with cobalt doping (WC-Co) investigated as a binary mixture of xenobiotics S. Reynaldi, Dept Agronomia Medellin / Cs Agronomias; D. Kuehnel, Helmholtz Centre for Environmental Research / Bioanalytical Ecotoxicology; R. Altenburger, UFC Centre for Environmental Research / Department of Bioanalytical Ecotoxicology. Tungsten carbide nanoparticles (WC) with cobalt doping (WC-Co) were considered as a mixture of xenobiotics, and its toxicity was investigated in Daphnia magna. After 72 and 96h exposure, observed toxicity of WCCo was compared with predictions from concentration addition (CA) and independent action (IA) models. These models predict mixture toxicity based on effect concentrations (CA) or effects (IA) of individual components of mixtures. WC and cobalt released from cobalt chloride (CoCl ·6H O) were considered individual components of WC-Co. 2 2 After 72h, observed EC for WC-Co was 0.18 mM, and predicted 50 EC was 0.66 mM and 0.72 mM for CA and IA, respectively. After 50 96h, observed EC for WC-Co was 0.11 mM, and predicted EC by 50 50 CA was 0.58 mM and by IA was 0.49 mM. Thus, the observed EC 50 clearly exceeded those predicted by models. Moreover, the differences between observed and predicted EC became larger from 72 to 96h 50 exposure. WC-Co toxicity may result from a synergistic interaction between individual components and this interaction may increase over exposure time. 411 A DNA metabarcoding approach to understand trophic transfers of pollutants S. Drouhot, UMR 6249 Chronoenvironnement; C. Tougard, University of Montpellier II / CNRS / UMR 5554 Institut des Sciences de l’Evolution de Montpellier; F. Pompanon, University Joseph Fourier / CNRS / UMR 5553 Laboratoire d’Ecologie Alpine; O. Faure, Ecole Nationale Superieure des Mines de Saint-Etienne Centre SPIN / CNRS / UMR 5600 Environnement-VilleSociété; C. Druart, University of Franche-Comte / CNRS / UMR 6249 Chrono-environnement; D. Rioux, F. Boyer, University Joseph Fourier /

SETAC Europe 23rd Annual Meeting

CNRS / UMR 5553 Laboratoire d’Ecologie Alpine; A. Prudent, A. Goydadin, N. Capelli, P. Giraudoux, R. Scheifler, F. Raoul, University of Franche-Comte / CNRS / UMR 6249 Chrono-environnement. Wildlife is considered to be mainly exposed to environmental contaminants via oral route. Food web analysis is therefore crucial in environmental risk assessment. In this context, diet information is commonly obtained from literature and little attention has been given to site-specific considerations (habitat, season…), which might considerably affect diet composition and thus wildlife exposure assessment. By associating next-generation sequencing and DNA barcoding techniques, the metabarcoding molecular approach of diet analysis could achieve a better taxonomic identification of food items with lower time investment than traditional micro- and macro-histological observations of food remains. In this context, the aim of this study is to evaluate the use of this new molecular method to investigate the diet of wild small mammals exposed to arsenic and eventually to better understand and model its transfers. On a partially remediated former gold mine in southern France, small mammals were captured in spring and autumn in zones differing by their remediation treatment and in a control site. Botanical surveys were also performed on the polluted site. DNA was extracted from stomach content (n=96) and faeces (n=19) and then amplified with 3 primers sets, allowing to get plants and invertebrates (molluscs, arthropods and earthworms) DNA. Finally, amplified DNA was sequenced on next-generation sequencer. Preliminary results on plant DNA showed that 95% of the sequences were at least identified to family level and among them 13% were identified up to species level. This attests the possibility to reach a precise taxonomic level with this molecular method. Stomach contents and faeces from a given small mammal specimen gave complementary information, probably because they correspond to different meals. The results also suggested different food patterns among small mammal species, probably related to their foraging ecology. Comparisons between plant taxa identified in the diet and local botanical survey showed some mismatches, which may be related to small mammal mobility during their foraging activity. As a conclusion, the recent DNA metabarcoding is a promising approach to relate local food web structure to resource availability and thus to better understand pollutant trophic transfer in ecosystems. 412 Spatial and Temporal Patterns and Trends of Perfluorinated Carboxylates and Sulfonates in Seabird eggs from the Pacific Coast 1 of Canada J.E. Elliott, Environment Canada. John E. Elliott , Kyle 2 2 1 1 H. Elliott , Melanie F. Guigueno , Laurie K. Wilson , Sandi Lee , 3 1 2 Francois Cyr Environment Canada, Delta, BC, Canada; University 3 of Manitoba, Winnipeg, MB, Canada; Environment Canada, Ottawa, ON, Canada There is a continuing need to monitor contamination of the marine environment by xenobiotic compounds, particularly those which are persistent and accumulate in food chains. Eggs of marine birds have proven to be an efficient and effective means of measuring and tracking substances, such as persistent organic pollutants (POPs) and mercury which are transferred from the female bird to the egg via yolk lipids or proteins. Here we report and discuss data from long term monitoring of POPs and mercury in seabird eggs from the northeast Pacific. For this program, the marine system was divided, and representative species selected. The nearshore subsurface is monitored using two cormorant, Phalacrocorax, species, auritus and pelagicus, both feed on a variety of benthic and pelagic fish. The offshore subsurface is monitored using the rhinoceros auklet, Cerorhinca monocerata, a feeder mainly on small pelagic fishes, with the offshore surface species, the Leach’s stormpetrel, Oceanodroma leucorhoa, which feeds mainly on surface plankton and larval fishes. At three breeding colonies each along the Pacific coast of Canada and at four year intervals 15 eggs are collected and analyzed as five pools of 3 eggs each. Among the chemicals measured in this long term study are the perfluorinated carboxylates and sulfonates. Data from a recent retrospective study, using archived samples collected from 1990 to 2011, shows, as reported for more polluted environments, that PFOS (perfluorooctane sulfonate) increased in continental shelf ranging auklet eggs until the late 1990s and have declined since then. In contrast, another compound, PFUdA (perfluoroundecanoate) increased steadily in eggs of both near and

102

offshore species. Stable isotopes will be used to examine the possible role of dietary variation, possibly related to marine regime shifts, in variation in contaminant levels in these monitored seabirds. 413 Lifetime PCB 153 bioaccumulation and pharmacokinetics in pilot whales: Bayesian population PBPK modelling and Markov chain Monte Carlo simulations L. Weijs, University of Antwerp Dept Biology; D. Tibax, University of Antwerp; A. Roach, Office of Environment and Heritage; R. Yang, Colorado State University; R. McDougall, AEgis Technologies Group; M. Lyons, Colorado State University; C. Housand, AEgis Technologies Group; T. Manning, J. Chapman, K. Edge, Office of Environment and Heritage; D. Pemberton, Dept of Primary Industries, Parks, Water and Environment; A. Covaci, University of Antwerp / Toxicological Centre; R. Blust, University of Antwerp / Systemic Physiological and Ecotoxicological Research, Department of Biology. Physiologically based pharmacokinetic (PBPK) models are mathematical representations of the ADME processes of chemicals in organisms. The ‘experiments’ involved in PBPK modeling can be minimally- or non-invasive; thus, this approach is of particular interest for assessing pollution in wild marine mammals since all toxicological experiments are prohibited or ethically undesirable. PBPK models for chemicals in marine mammals have previously been parameterized by adopting those from other organisms and validated using results measured in tissue samples from dead, stranded animals. However, such an approach entails a fair amount of model uncertainty. Bayesian methods and Markov chain Monte Carlo (MCMC) simulations are currently the best statistical approaches to address this uncertainty. The goal of the present study was therefore to develop a PBPK model for lifetime bioaccumulation of PCB 153, one of the most persistent PCBs in marine mammals, in male long-finned pilot whales and to evaluate that model using the Bayesian approach and MCMC simulations. The structural PBPK model consisted of 2 flow-limited compartments: blubber and the ‘rest of the body’. Physiological and chemical parameters in this model were primarily taken from the literature. In the statistical model, a Bayesian approach with MCMC analysis, was applied in order to evaluate or update the parameters taken from the literature (priors) with regard to pilot whale data taken from our own analyses. The model predicts a steep increase in PCB 153 level from birth until the age of about 1.5 years followed by a rapid decline until the age of 3 years due to the growth dilution effect and a further slow decline caused by other factors such as decreasing PCB 153 levels in the environment. This method starts with estimated or known parameters from the literature (priors) and uses the available dataset to derive new parameter values (posteriors). These posteriors can accordingly be used further as “priors” as soon as new datasets are available resulting in more updated parameters. Such an approach provides progressively more robust parameters so that the updated parameters are better representing the bioaccumulation in populations in the wild. Therefore, this study is the first step towards the production of reliable parameter values for pilot whales. 414 Squamous epithelial proliferation as a biomarker of exposure to AhR-active compounds for ecological monitoring using wild mink as a sentinel species M.J. Zwiernik, Michigan State University / Animal Science Vet Med; S.D. Fitzgerald, Michigan State University / Department of Pathobiology and Diagnostic Investigation; J.L. Newsted, Cardno Entrix; J.N. Moore, U.S. Fish and Wildlife Service / Ecological Services; J.E. Link, E.M. Koppel, S.J. Bursian, Michigan State University / Wildlife Toxicology Laboratory, Department of Animal Science. Squamous epithelial proliferation (SEP) in the mandible and maxilla of wild mink has been proposed as a sentinel biomarker for mammalian wildlife exposure to dioxin-like compounds (DLCs) including polychlorinated dibenzo-p-dioxins (PCDDs), polychlorinated dibenzofurans (PCDFs), and polychlorinated biphenyls (PCBs). The proposal originates from the fact that mink are very sensitive to the effects of DLCs, exquisitely sensitive to DLC induced SEP, they are widely distributed throughout Europe and the Americas, and can be used in controlled laboratory exposure studies. The SEP biomarker is abnormal tissue consisting of nests and cords of epithelial cells that

SETAC Europe 23rd Annual Meeting

extend into the alveolar bone causing skull and jawbone osteolysis. Its occurrence is easily and inexpensively identified histologically at lesser severity or grossly at great severity. Herein we provide additional support for the use of SEP as a sentinel endpoint in wild mink for the assessment of risk from DLC exposure The additional support includes a review of the literature pertaining to gingival and alveolar bone abnormalities in mammals exposed to DLC as well as a summary of recently preformed laboratory mink dosing studies utilizing individual DCL compounds or environmental mixtures of DLC compounds. In short, SEP and SEP symptoms have been noted in a wide range of mammalian species exposed to DLC, the incidence and severity of SEP and SEP associated symptoms, increases in severity with increased DLC dose, and in controlled studies jaw lesions always occurred at doses lesser than those for which adverse individual or reproductive health effects were observed. Given that the assessment of the ecological risk of DLC is often a complex and expensive undertaking, the availability of a low cost, easily observed biomarker that can be used to screen for biologically significant exposures of DLCs in individuals or populations would be of great utility for environmental monitoring, source identification, and effectiveness of remedial activities. 415 Influence of individual parameters, habitat diversity, and trace metal contamination on tick-borne disease prevalence in wild wood mice N. TETE, Laboratoire Chrono-environnement; E. AFONSO, C. FRITSCH, R. SCHEIFLER, University of Franche-Comte / Laboratoire Chrono-environnement. Recent studies have shown that parasitic infections in aquatic animals may be related to environmental pollution. However, to our knowledge, this link has received little attention in terrestrial ecosystems. Thus, the aim of this study was to test the hypothesis that pollution by trace metals (TM) influences prevalence to bacterial infection transmitted by vectors in a rodent species. The field work took place around the site of an ancient lead (Pb) and zinc (Zn) smelter. After more than 100 years of activity, this area is drastically polluted, mainly by three TM: cadmium (Cd), Pb, and Zn. Categories of environmental contamination (low, medium, high) were determined for both Cd and Pb. During the autumn 2010, 227 wood mice were trapped along a pollution gradient. Four bacteria were studied by PCR: Borrelia spp., Coxiella burnetii, Anaplasma phagocytophilum, and Bartonella spp. To study the influence of the habitat on bacteria prevalence a Shannon habitat diversity index was calculated around the trapping site of each animal. Over the four bacteria tested, Borrelia spp. was not detected; C. burnetii was detected in only one individual (0.4%), A. phagocytophilum in 22 individuals (9.7%), and Bartonella spp. in 125 individuals (55.1%). No differences in A. phagocytophilum and Bartonella spp. prevalence were observed nor depending on age (logistic regression, LRT p = 0.372 and p = 0.622, respectively), neither depending on gender (LRT p = 0.678 and p = 0.461, respectively). A. phagocytophilum prevalence were positively influenced by the habitat diversity (LRT, p = 0.008) while Bartonella spp. were not (LRT, p = 0.569). Once the habitat diversity taken into account, A. phagocytophilum prevalence was positively though marginally influenced by Cd contamination in the environment (LRT, p = 0.054) and,to a lower extent, by Pb contamination (LRT, p = 0.087). However, Bartonella spp. prevalence was not influenced by Cd and Pb levels of contamination (LRT, p = 0.434 and p = 0.569, respectively). None of Cd or Pb internal concentrations in the liver or in kidneys in had any influence on these two bacteria prevalence. Due to the possible synergistic effects of pollution and parasitism on wildlife conservation, further studies based on such a systemic approach including pollution and other stressors should be developed in ecotoxicology. 416 Long time trends of concentrations of POPs and Mercury in Scottish wild birds M.G. Pereira, Centre for Ecology Hydrology; E. Potter, Centre for Ecology and Hydrology; L. Walker, Centre for Ecology Hydrology; J. Crosse, Centre for Ecology and Hydrology; R.F. Shore, CEH Lancaster. The Predatory Bird Monitoring Scheme (PBMS) is the umbrella project that encompasses long-term contaminant monitoring work in avian predators in theUK. The PBMS monitors sentinel avian species to detect and quantify current and emerging

103

chemical threats. It also provides scientific evidence on how chemical risk varies over time and space. This may occur due to market-led or regulatory changes in chemical use and may also be associated with larger-scale phenomena, such as global environmental change. The PBMS has monitored long-term trends in pollution in the eggs of a range of terrestrial and marine feeding predatory birds from Scotland. Some (merlin- Falco columbarius, gannet- Morus bassanus) have been studied as sentinels of terrestrial and marine contamination, respectively,while others (golden eagle Aquila chrysaetos, white-tailed sea eagle- Haliaeetus albicilla) have been monitored because of their high conservation status. In this paper we report long-term trends in concentrations of old and new POPs and mercury (Hg) in the eggs of these species. From this long term data sets we conclude that: Concentrations of PCBs and Hg did not show a consistent temporal trend for all the species studied and the legacy PCBs do not show a significant temporal decline in all species as may have been expected given the long-term prohibition of their use. PBDE concentrations in gannet eggs declined after the legislation banning the use of these compounds was introduced. Although it is not certain that DDE or PCBs were a contributory cause of the reproductive failure observed in whitetailed eagles fromScotland, the residues in some eggs were of a magnitude that may be expected to have adverse effects. 417 High-resolution fractionation of complete GC chromatograms for Effect Directed Analysis J. Kool, VU University / BioMolecular Analysis group. Effect-directed analysis (EDA) is an important tool in studying the biological effect of individual compounds in a mixture, e.g., toxic compounds in the aquatic environmental. At present, for environmental EDA, complex environmental samples showing toxic activity are fractionated with liquid chromatography (LC) into a limited number of fractions. Each fraction is tested for toxic activity in a bioassay. Further fractionation steps may be necessary to find and identify the toxic compounds. Although post-column fractionation by capturing an LC eluent is straightforward and LC fractionation setups are widespread, many compound classes relevant in environmental EDA are preferably separated and analyzed by gas chromatography (GC) rather than LC, e.g., polycyclic aromatic hydrocarbons (PAHs), many pesticides, and mineral oils. Moreover, many methodologies are certified in this area are GC rather than LC methods and considered the standard. Fractionation of compounds after GC is not often performed because of the complicated setups required, and GC fractionation only allows automated collection of a very limited number of fractions, usually only 6, utilizing complicated cryo-trapping or adsorbent collection vessels. As GC offers a very high resolution in the separation of compound classes like non-polar organics and volatiles, a short collection time per fraction (second range) is necessary for fractionation after GC without losing (too much of) the separation efficiency. In this presentation, we describe the development, optimization, validation, and application of a novel GC fractionation technology that allows complete GC chromatograms to be fractionated in high yield and high resolution. The number of fractions collected is drastically increased over the currently available technology. The GC fractionation technology developed was applied to environmental extracts which were successfully analyzed with GC combined with high-resolution EDA. Complete GC chromatograms were collected in second range fractions prior to straightforward EDA. Endocrine disrupting pollutants in these mixtures analyzed were identified as bioactives towards the dioxin receptor by using a mammalian cellular gene reporter assay as readout. For this, another novelty was introduced in which cells were seeded and grown post-fractionation, directly onto the collected fractions, which additionally eliminates handling steps and circumvents compound dilution. 418 Hormone receptor-affinity extraction coupled to high-resolution mass spectrometry for activity-directed analysis of EDCs in complex mixtures L. Ferguson, Pratt School of Engineering / Department of Civil and Environmental Engineering; K.A. Stencel, Duke University / Nicholas School of the Environment; G.J. Getzinger, Duke University / Civil and Env Engineering; L.K. Shaw, University of South Carolina /

SETAC Europe 23rd Annual Meeting

Department of Chemistry & Biochemistry. We report the development of a new method, hormone receptor-affinity extraction, for isolating xenoestrogenic environmental contaminants from complex water and wastewater samples prior to analysis by high resolution mass spectrometry. This technique utilizes the specificity of hexahistidinetagged, recombinant human estrogen receptor (alpha isoform) ligand binding domain (ER?-LBD) to bind trace estrogenic compounds in solution prior to co-purification of the ER?-LBD-xenoestrogen complex using immobilized metal-affinity chromatography. The method reduces sample complexity and enriches for estrogen receptor-relevant endocrine disruptors. Xenoestrogens in municipal wastewater effluent and in surface waters impacted by irrigation runoff from land-application of wastewater effluent were isolated by receptor-affinity extraction and quantified by HPLC coupled to high-resolution mass spectrometry using stable-isotope dilution. Synthetic xenoestrogens (nonylphenol, bisphenol A, and octylphenol) were detected in water and wastewater at -1 concentrations up to approximately 350 ng·L , while biogenic estrogens (17?-estradiol, estrone, and estriol) were present in these -1 samples at lower concentrations (typically below 10 ng·L ). The pharmaceutical contraceptive estrogen 17?-ethynylestradiol was not measured in water or wastewater samples above detection limits (< 1 -1 ng·L ). Qualitative analysis of non-target xenoestrogens in wastewater receptor-affinity extracts was performed by HPLC-QTOF and HPLCOrbitrap MS/MS, and the weak xenoestrogen 2-(2-(4nonylphenoxy)ethoxy)acetic acid (NP2EC) was identified by accurate mass measurement and high resolution MS/MS. Identification of this compound (an oxidative degradation product of nonylphenol polyethoxylate surfactants) in estrogen-receptor affinity isolates of wastewater effluent demonstrates the utility of this method for characterizing trace xenoestrogens in complex mixtures without prior knowledge of their identities. 419 Identification and confirmation of contaminants in marine waters using passive samplers, high resolution fractionation and mass spectrometry P. Booij, P. Leonards, Institute for Environmental Studies; P. de Voogt, S. Sjollema, University of Amsterdam; D. Vethaak, DELTARES; M. Lamoree, Institute for Environmental Studies. Anthropogenic contaminants and natural toxins in estuarine and coastal ecosystems can affect primary production at the basis of the marine food chain. These chemical stressors are hypothesized to disturb the photosynthesis of phytoplankton and, therefore, the carrying capacity of coastal and estuarine waters. In our study we aim to identify and confirm the contaminants that have an effect on algae in order to quantify the toxic pressure in the Dutch coastal and estuarine waters. To identify contaminants Effect-Directed Analysis was performed with passive samplers (POCIS and silicone rubber sheets) at three estuarine and coastal locations in The Netherlands. The passive sampler extracts were tested for toxicity on algae using the Pulse Amplitude Modulation (PAM) fluorescence assay. To reduce the complexity of the extracts high resolution fractionation in 96 well plates was performed. The well plate was directly used to perform the PAM assay after evaporation of the LC solvent. In general, there was a significant overlap in active fractions between the POCIS and silicone sheets for all three locations. This is an indication that both passive samplers are able to sample the same types of compounds that are responsible for the effect on algae. Compounds were identified using a stepwise approach. A selection of compounds was purchased and analytically confirmed by using LC-ESI-MicrOTOF-MS. Confirmed compounds were quantified with LC-ESI-MicrOTOF, and the PAM assay was performed to determine the effect concentrations on the ?PSII. Twentysix compounds could be identified. Five of these compounds (all pesticides) could be confirmed for retention time and response in the PAM assay. These five pesticides are probably the main contributors to the photosystem II inhibition in microalgae for the Dutch coastal waters. Acknowledgement - This project is financed by DELTARES, NL. 420 Effect-Directed Analysis approach (EDA) for the identification of active compounds in pig manure C. Gardia Parege, EPOCLPTC / EPOC-LPTC UMR 5805 CNRS; M. Devier, University of Bordeaux /

104

UMR 5805 CNRS EPOC-LPTC; G. Hernandez Raquet, INRA / LISBP INSA UMR INSA/CNRS 5504 - UMR INSA/INRA 792; P. Balaguer, University of Montpellier 1 / INSERM U896; H. Budzinski, University of Bordeaux. Animal breeding activities are identified as a major source of hormones in environment. In addition to veterinary drugs used in feed and water to improve the growth performance or to prevent infection in intensive animal production, animals may also be in contact with many endocrine disrupting compounds (EDCs) as PAHs, dioxins, phtalates, detergents coming from farm activities. These EDCs will be concentrated in manure. Considering this and the use of manure as agricultural fertilizer, manure is considered as potential source of contamination. In this study, the assessment of the potential endocrine disrupting activity of pig manure was performed using in vitro bioassays. In order to isolate the active chemicals and to identify them, Effect-Directed Analysis (EDA) method was used. This approach combining bioassays, fractionation procedures and chemical analytical methods (LC-MS/MS, LC-QToF) aims to establish the cause-effect relationships by sequential reduction of the complexity of environmental mixtures to individual toxicants. Pig manure was extracted by Accelerated Solvent Extraction (ASE). Then, the crude extract was fractionated using sequential procedure based on a preliminary fractionation by Solid-Phase Extraction (SPE) followed by a finest fractionation by Reverse Phase – High Performance Liquid Chromatography (RP-HPLC). Each fraction was tested using 2 cell lines based on luciferase gene under the control of estrogen receptor a and aryl hydrocarbon receptorallowing the specific and integrative detection of a wide range of active chemicals. The most active fractions were identified. On the basis of RP-HPLC fractionation calibration, targeted chemical analyses were performed on several fractions but the detected compounds poorly explained the observed biological activities. So for all selected active fractions, compounds responsible for these activities remained unknown. The identification of active compounds was performed using a LC-HRMS system (LC-QToF) and several drugs, metabolic drugs and pesticides were found in several fractions. These results demonstrated the usefulness of EDA-based strategy for the identification of active compounds in environnemental complex samples when mass balance analysis (MBA) approach was not satisfactory. 421 Identification of polar mutagenic contaminants in river water: Improving candidate structure selectionM. Krauss, Helmholt Centre for Environmental Research / Effect-Directed Analysis; C.M. Gallampois, Linköping University / Department of Clinical and Experimental Medicine; A. Bahlmann, HelmholtzZentrum fur Umweltforschung UFZ; E. Schymanski, Eawag Swiss Federal Institute of Aquatic Science; U. Nadin, Helmholtz centre for environmental research - UFZ; W. Brack, Helmholtz Centre for Environmental Research UFZ / Effect-Directed Analysis. To identify mutagenic compounds, effect-directed analysis (EDA) using a combination of biotesting, fractionation and mass spectrometry (MS)-based structure elucidation has proved to be a valuable approach. A major challenge in the workflow is the assignment of a candidate structure to the molecular formula derived from MS data. Our objective was to improve candidate selection on the basis of liquid chromatography-high resolution tandem mass spectrometry. Blue rayon passive sampler extracts from a location at the Elbe river were separated into acidic, neutral and basic fractions and fractionated further by LC using a sequence of two different columns. The Ames fluctuation assay with strain TA98 was used to screen samples for mutagenic compounds with and without metabolic activation. Fractions showing the highest mutagenicity were analyzed using a high resolution/high accuracy LTQ Orbitrap MS and additionally tested using the YG1024 and YG1041 strains. High mutagenicity was found predominantly in the neutral and acidic fractions. For one selected sub-fraction of the neutral fraction showing the highest mutagenicity, the effects were about five times higher in the Ames test using strains YG1041 and YG1024 than for strain TA98 (all with metabolic activation), suggesting that aromatic amines contributed considerably to the total mutagenicity. In this subfraction 17 peaks of interest could be detected and a molecular formula assigned to these. Application of MetFrag involving candidate selection via a comparison

SETAC Europe 23rd Annual Meeting

of measured vs. predicted spectra enabled us to propose a list of 92 candidate structures. After comparison of retention times measured with those predicted using a linear solvation energy relationship approach, the list of candidates was decreased to 22. As aromatic amines were likely responsible for the mutagenic effects, we performed a mutagenicity prediction of the 46 aromatic amine candidates on the basis of the stability of corresponding nitrenium ions as ultimate electrophiles attacking the DNA. This approach identified nine aromatic amine candidates as likely mutagens. In combining the results of these two methods, only for two candidates the predictions of retention and mutagenicity were positive. This work showed that the predictions of retention, MS fragmentation and mutagenicity can be valuable tools in structure elucidation by significantly reducing the candidate structure list. 422 Characterization of the toxic constituents of oil sands process affected water by Orbitrap-MS A. dos Santos Pereira, University of Alberta / Division of Analytical and Environmental Toxicology; S.B. Wiseman, University of Saskatchewan / Toxicology Centre; R. Mankidy, University of Saskatchewan / Toxicology; Y. He, H. Alharbi, J.P. Giesy, University of Saskatchewan / Toxicology Centre; J.W. Martin, University of Alberta / Laboratory Medicine and Pathology. Oil sands deposits in Northern Alberta, Canada, represent the world’s third largest crude oil deposit (in the form of bitumen), estimated at 176.8 billion barrels. Environmental challenges in the surface mining oil sands industry arise from the fact that the bitumen is separated from sand and fine clay using alkaline hot water, resulting in large volumes of toxic oil sands process affected water (OSPW). This OSPW must be contained in tailings ponds due to its toxicity, and a zero-discharge policy. Tailings ponds are growing in volume and in number, and in 2011 covered 2 approximately 170 km despite heavy recycling of OSPW into the extraction process. OSPW contains a very complex mixture of organic compounds and is acutely and chronically toxic to diverse organisms including zooplankton, phytoplankton, invertebrates, fishes, mammals, birds and plants.The acute toxicity of OSPW has been attributed to a complex mixture of dissolved organic acids, however the specific chemicals that contribute to the acute toxicity, let alone the chronic, reproductive, or developmental toxicity, remain unknown. Approximately 2000 organic compound groups were detected in the OSPW fractions, including compounds matching the formulae: C H O x y z (z = 1 to 6); C H O N (z = 0 to 4); C H O S (z = 1 to 4) and x y z x y z C H O NS (z = 1 to 2). The O species (C H O ) were the more x y z 2 x y 2 abundant species detected in the neutral fraction under both ionization conditions; and this fraction had the greatest estrogenicity and embryotoxicity among all primary fractions (neutral, acid and basic). The O species detected in negative mode (Figure 1A) are 2 predominantly of classic naphthenic acids (NAs), however the analysis in positive mode showed the presence of an additional homologous family of non-acid O species (probably hydroxy-keto compounds) 2 constituting 100 different species ranging from 10 to 25 carbon atoms and with double bound equivalents (DBE) ranging from 2 to 16. These new O species should also be considered as candidate molecules for the 2 adverse effects of OSPW. 423 Identifying new Persistent Organic Pollutants under the Stockholm Convention M. Scheringer, ETH Zuerich / Institute for Chemical and Bioengineering; S. Strempel, S. Hukari, C.A. Ng, ETH Zurich; M. Blepp, Öko-Institut; K. Hungerbuhler, ETH Zurich. The Stockholm Convention on Persistent Organic Pollutants (POPs) is an important Multilateral Environmental Agreement (MEA) that helps to identify POPs and to regulate them at the global level. As of 2012, 22 POPs are regulated under the Convention. These POPs are listed in Annex A (elimination; 18 chemicals), B (restriction; two chemicals), or C (unintentional production; five chemicals); three chemicals are listed in both Annex A and C. Before a chemical can be regulated under the Convention, it is assessed by the POP Review Committee (POPRC). In this assessment, a potential POP candidate is first evaluated against screening criteria defined in Annex D of the Convention; these criteria cover persistence (P), bioaccumulation potential (B), toxicity (T), and

105

long-range transport (LRT). When the POPRC has agreed that a chemical fulfills the Annex D criteria, the next steps are a fuller evaluation of the environmental hazard and risk caused by the chemical according to Annex E (risk profile) and a socio-economic assessment (Annex F). For each chemical, the entire process takes several years. Yet, there are many thousands of unevaluated chemicals on the market and, therefore, a key question for the future identification and regulation of POPs under the Stockholm Convention is how many potential POPs exist and how they can be identified. To address this question, we compiled a list of 93 000 chemicals specified by CAS registry number and SMILES code. For each of the chemicals on this list, we collected data on biodegradation half-life, bioconcentration and bioaccumulation factor (BCF, BAF), acute and chronic toxicity in aquatic species, and degradation half-life for reaction with OH radicals. These data include both measured and estimated values; because measured data are scarce, most data points had to be estimated from the chemical structure. Next, we compared the property data to the Annex D criteria: 60 days for the degradation half-life in water, 5000 for BCF and BAF, and 2 days for the degradation half-life in air. For acute toxicity, we used the threshold defined for PBT assessment under the European chemicals regulation, REACH (0.1 mg/L). We found 574 chemicals that exceed all four thresholds and can be considered potential POPs; most of these chemicals are halogenated. We further prioritized the chemicals by means of production volume and found 12 potential POPs that are highproduction volume chemicals or have been registered under REACH.

The extent of this accumulation correlates directly with its ever-more prevalent use. Limited toxicological information is available to assess the long-term implications for health or the environment of the HBCDD contamination. HBCDD is an endocrine disruptor in animals. In 2008 The European Chemicals Agency identified HBCDD as 1 of 14 substances of “Very High Concern”, and in September 2010 HBCDD was added to REACH’s Authorization List. In February 2011 HBCDD was selected to be phased out by EU REACH Regulation before 2015. In October 2010, 2011 and 2012 the POP Review Committee under the Stockholm Convention has assessed the risks from HBCDD, and ithad concluded that HBCDD fulfills the criteria of a persistent organic pollutant (POP), and the committee recommended a global ban of HBCDD use (Annex A). In January 2012 HBCDD it was among 15 new priority substances proposed by the European Commission to be regulated by the EU Water Framework Directive This presentation will provide a succinct up-to date overview of HBCDD’s properties, uses, regulation and discusses the risks associated with its prevalence in our homes and immediate environment.

424 Analytical challenges for new and candidate POPs J. de Boer, VU University / IVM; H.A. Leslie, Institute for Environmental Studies VU Amsterdam. The Stockholm Convention has listed eighteen persistent organic pollutants (POPs) for elimination (Annex A), two for restriction (Annex B) and five for reduction of unintentional production (Annex C). Several of these POPs consist of mixtures and all are halogenated aromatic or aliphatic hydrocarbons. A Global Monitoring Program (GMP) has been designed to follow the concentrations of these POPs in the environment and in human milk and blood. All countries that have signed the Convention are obliged to contribute to the GMP. However, many countries have no or little experience in analyzing POPs. Apart from instrumentation such as gas (GC) and liquid chromatography (LC) and mass spectrometry (MS), trained staff is essential for a successful monitoring program per country. Therefore, United Nations Environment Program (UNEP) is investing in capacity building. Key activities are a bi-ennial interlaboratory study, on-site training of laboratory staff, provision of consumables, instruction films and several specific training activities using expert laboratories. Recent results of these training programs show several common factors that slow down progress in participating laboratories, for example: a lack of daily routine in analyzing POPs, bureaucracy (which slows down ordering processes), limited access to the scientific literature, no maintenance of instruments, power failures, dust in laboratories, etc. On top of that, there are the usual challenges of developing analytical methods for existing and candidate POPs. Difficulties are particularly met for the analysis of perfluorinated alkyl substances, brominated diphenylethers and toxaphene. No doubt, chlorinated paraffins will cause serious problems when they would be listed as POPs. There is insufficient capacity for the analysis of chlroinated dioxins, furans and dioxin-like PCBs. The first bi-ennial interlaboratory assessment on POPs showed substantial discrepancies between laboratory results, at least comparable to the results obatined by European and NorthAmerican laboratories in the mid-1980s.

426 Recent decreases of some perfluoroalkyl substances and brominated flame retardants in the ArcticK. Vorkamp; F.F. Riget, R. Bossi, Aarhus University; C. Sonne, R. Dietz, Aarhus University / Department of Arctic Environment. Documentation of the ubiquitous presence of Polybrominated diphenyl ethers (PBDEs) in the environment and in humans has led to bans of the technical products PentaBDE and OctaBDE through the Stockholm Convention on Persistent Organic Pollutants (POPs). The perfluoroalkyl substances (PFAS) PFOS and perfluoroctane sulfonyl fluorid are included in Annex B (restriction) of the Stockholm Convention. PBDEs and PFAS are systematically monitored in Greenland biota as part of the AMAP Monitoring and Assessment Programme (AMAP), including ringed seals from East and West Greenland and polar bears from East Greenland. The most recent international AMAP assessment concluded that levels of PFOS and other PFAS increased in ringed seal and polar bear from Greenland. With regard to PBDEs, the international AMAP assessment showed a significant increase of BDE-47 in ringed seals from West Greenland, but no trend was reported for PBDEs in ringed seals from East Greenland. The updated temporal trends of PFOS in West Greenland and East Greenland ringed seals showed a marked increase from 2003 to 2006, followed by a sharp decrease. In polar bears, the PFOS concentrations were considerably higher than in ringed seals, however, the temporal trend of PFOS in polar bears showed a similar pattern to that in ringed. This recent decrease is different from the conclusion in the latest AMAP assessment and might indicate effects of international regulations. In juvenile ringed seals from West Greenland, BDE-47 is the only BDE congener consistently above detection limits. BDE-47 continues to increase in blubber of these ringed seals, at an annual rate of 4.8%. No time trend is apparent in ringed seals from East Greenland: While a relatively constant concentration is observed in juvenile seals, concentrations decrease in adult seals, although not significantly. These results confirm the latest international AMAP assessment. If global regulations under the Stockholm Convention have the effect on decreasing levels in Central East Greenland, the current significant trend in Central West Greenland should be temporary. In subadult polar bears, significant increases were found BDE-100 and BDE-153, at average annual rates of 5.0%. BDE47, BDE-99 and ?PBDE, however, showed a significant non-linear trend and peaks in 2000 and 2004, respectively, which agrees with indications of declines in ringed seals from the same area.

425 Hexabromocyclododecane (HBCD) - A Hazardous Flame Retardant Used in Polystyrene Insulation ProductsA.A. Jensen, NIPSECT. Hexabromocyclododecane (HBCDD or HBCD) is a brominated flame retardant mainly used in polystyrene-based building insulation products (EPS and XPS) but is also used in electronics as a substitute for banned flame retardants. This lipophilic and persistent organic pollutant accumulates in natural organisms and magnifies in the food chain, leading to progressively increasing background levels in human tissues and in wildlife, especially of the most stable a-isomer.

427 Prioritization, screening and identification of organosilicon contaminants in the environment A.H. Kierkegaard, M.S. McLachlan, Stockholm University; K. Breivik, Norwegian Inst for Air Research; J.A. Arnot, ARC Arnot Research Consulting / Department of Physical Environmental Science; F. Wania, University of Toronto at Scarborough / Department of Physical & Environmental Sciences. A mass balance model of chemical fate and bioaccumulation in the environment was used to rank 287 high- and low-production volume organosilicon compounds for their concentration in the environment and

SETAC Europe 23rd Annual Meeting

106

in top predators. Key physical chemical properties of each chemical were estimated using quantitative structure-activity relationships (QSARs) and a total emission estimate of each chemical was made using information, which included amounts entering commerce and emission factors. Based on the model predicted concentrations in air, sediment and human tissue, chemicals were selected for screening through environmental sampling and analysis. Known environmental organosilicon contaminants such as the cyclic and linear volatile methyl siloxanes (VMS) were excluded as well as structures subject to rapid hydrolysis, a feature which was not taken into account in the model simulations because of current limitations in predicting hydrolysis halflives with QSARs. Analytical standards were only commercially available for half of the remaining 30 organosilicon compounds. Ten of these were not stable in solution, which left 5 organosilicon compounds eligible for environmental screening. These were tetrakis(trimethylsilyloxy)silane, phenyl-tris(trimethylsiloxy)silane, trifluoropropyltrimethylcyclotrisiloxane, trifluoropropylmethylcyclotetrasiloxane and tetraphenyltrisiloxane. Four of these chemicals were identified in sewage sludge, in sediment from Stockholm harbor, and in Stockholm ambient air samples. The trifluoropropyl-substituted siloxanes were analysed with UPLC-MS/MS, the others with GC-MS. Trifluoropropyltrimethylcyclotrisiloxane was solely detected as its corresponding linear diol. To date it is unclear whether the diol is present in the environment as such or formed during extraction or cleanup. The concentrations of the chemicals ranged from -3 -1 pg m in air up to ng g d.w. in sewage sludge, which are orders of magnitude below the levels of cyclic VMS (such as D5) in the same matrices. 428 Forensic Techniques to Age Date Human Exposure to PCBs D. Megson, Plymouth University; G. O'Sullivan, Department of Environmental Science, Mount Royal University; S. Comber, P. Worsfold, M. Lohan, School of Geography Earth and Environmental Science, Plymouth University. Polychlorinated Biphenyls (PCBs), a group of 209 chlorinated organic compounds, were first synthesised for industrial purposes in 1929. They were widely used until the 1980s when their use was phased out due to environmental and human health risks. Due to the persistence and toxicity of PCBs their concentrations in the US population are routinely monitored within the United States National Health and Nutrition Examination Survey (NHANES). This work presents a multivariate statistical analysis of the NHANES 200304 dataset to identify steady state (persistent in the body) and episodic (identified in humans transiently) congeners from background concentrations of PCBs present within the US population. These results were used in conjunction with other reviews on PCB metabolism to identify how the structure of PCBs relate to rates of biotransformation in humans. Congeners with chlorine atoms in the 2,5- and 2,3,6- positions appear to be more susceptible to biotransformation whereas congeners with chlorine bonds in the 2,3,4- 2,4,5- 3,4,5- and 2,3,4,5- positions appear to be more persistent. This work identifies a subset of 23 key PCBs which have the best potential to age date human exposure to PCBs. A summary of on-going method development using GC-GC-MS to accurately quantify these 23 key congeners will also be presented. 429 EDC testing in the future: Exploring roles of pathway-based in silico, in vitro and in vivo methods G.T. Ankley, US EPA / National Health and Environmental Effects Research Laboratory; D.L. Villeneuve, US EPA / Mid-Continent Ecology Division; T. Hutchinson, School of Biological Sciences Plymouth University. Current methods for screening, testing and monitoring endocrine-disrupting chemicals (EDCs) rely substantially upon moderate- to long-term assays that can, in some instances, require significant resources, including numbers of animals. Recent developments in the areas of short-term in vivo tests/endpoints (including genomics), in vitro assays (e.g., high throughput technologies), and computational biology (including bioinformatics) could provide the basis for efficiently detecting chemicals with the potential to perturb specific biological pathways within the hypothalamic-pituitary-gonadal and -thyroidal axes. Adverse outcome pathways (AOPs) offer a useful organizational framework for

SETAC Europe 23rd Annual Meeting

identifying and implementing these types of approaches. Pathway-based in vivo, in vitro and computational approaches, while not yet capable of wholly replacing longer-term whole-organism tests, nonetheless could fill roles related to screening/prioritization of large numbers of test chemicals for endocrine activities, or rapid broad-scale environmental monitoring of EDCs. This presentation will provide an overview of the AOP framework in the context of EDCs, and explore some of the newer pathway-based technologies being considered for EDC testing and monitoring, including their potential advantages and drawbacks. 430 Mechanism-based testing strategy using in vitro approaches for identification of thyroid hormone disrupting chemicalsA. Murk, Wageningen Agricultural University / Dept. of Toxicology. The thyroid hormone (TH) system is responsible for several important physiological processes, including regulation of energy metabolism, growth and differentiation, development and maintenance of brain function, thermoregulation, osmo-regulation, and axis of regulation of other endocrine systems, sexual behaviour and fertility, cardiovascular function. Therefore, strategies are being developed to identify TH disrupting (THD) chemicals (THDC). Information on THD potency of chemicals still is derived from animal studies. For the majority of chemicals, however, this information is hardly available. It is unlikely that animal experiments will be performed for all THD relevant chemicals in the near future for ethical, financial and practical reasons. In addition, typical animal experiments often do not provide information on the mechanism of action of THDC, making it harder to extrapolate results across species. Relevant effects may not be identified in animal studies when the effects are delayed, life stage specific, not assessed by the experimental paradigm (e.g., behaviour) or only occur when an organism has to adapt to environmental factors by modulating TH levels. Therefore, in vitro and in silico alternatives to identify THDC and quantify their potency are needed. THDC have many potential mechanisms of action, including altered hormone production, transport, metabolism, receptor activation and disruption of several feed-back mechanisms. In vitro assays are available for many of these endpoints, and the application of modern ‘-omics’ technologies, applicable for in vivo studies can help to reveal relevant and possibly new endpoints for inclusion in a targeted THDC in vitro test battery. An international group of experts in the areas of thyroid endocrinology, toxicology of endocrine disruption, high-throughput screening, computational biology, and regulatory affairs has reviewed the state of science for 1) known mechanisms for THD plus examples of THDC and 2) in vitro THD tests currently available or under development related to these mechanisms. Based on this scientific review, the panel has recommended a battery of test methods to be able to classify chemicals as of less or high concern for further hazard and risk assessment for THD. Research gaps are identified to be able to optimise and validate the targeted THD in vitro test battery for a mechanism-based strategy to decide to opt out or to proceed with further testing for THD. 431 Müllerian duct differentiation – a sensitive target for endocrine disrupters in amphibians C. Berg, Dept of Environmental Toxicology. The present project aims to develop methods to investigate developmental and reproductive toxicity of endocrine disrupters using the model frog Xenopus tropicalis test system. Here I review our findings regarding developmental reproductive toxicity involving disruption of Müllerian duct differentiation. Müllerian ducts are the precursors of the female reproductive tract in most vertebrates but they are absent in teleost fish. Developmental reproductive effects were characterised using two types of endocrine disrupters present in the aquatic environment; progestins and estrogens. The estrogen ethynylestradiol EE is a well studied reproductive toxicant. Progestins 2 (synthetic progesterone) were recently shown to present a risk to female reproduction in fish and amphibians. Progestins are extensively used in human and veterinary medicine and they are ubiquitous in surface and ground waters. Xenopus tropicalis were exposed to EE2 or the progestin levonorgestrel (LNG) via the water during the entire larval period. Nominal exposure concentrations were verified by chemical analysis at regular time intervals. Exposure was discontinued at metamorphosis,

107

and a subset of frogs was raised to adult age for fertility evaluation i.e. gametogenesis, fecundity, fertility success, mating behaviour, and sperm quality. We found that low environmental concentrations of ethynylestradiol EE (as low as 1.8 ng/L) caused infertility by disrupting 2 the differentiation of Müllerian ducts and testicles. Developmental levonorgestrel exposure inhibited oogenesis and caused a complete lack of developed oviducts rendering the females sterile. No effects of levonorgestrel were seen in the males. In summary, 1) testicular and Müllerian duct differentiation are targets for developmental estrogen exposure 2) egg development and Müllerian duct differentiation are sensitive targets for developmental progestin exposure, 3) development of the female reproductive system is more susceptible to progestin toxicity than the male system. In conclusion, Müllerian duct differentiation is susceptible to disruption by different types of endocrine disrupters resulting in severe reproductive toxicity. 432 Development and validation of OECD test guidelines on mollusc reproductive toxicity tests V. Ducrot, INRA / Ecotoxicology and Quality of Aquatic Ecosystems; D. Azam, INRA; R. Brown, AstraZeneca; S. Charles, University Lyon / Laboratory of Biometry and Evolutive Biology; H. Holbech, University of Southern Denmark / Department of Biology; T. Hutchinson, School of Biological Sciences Plymouth University; L.L. Lagadic, INRA / UMR INRAAgrocampus Ouest Ecology and Ecosystem Health; J. Oehlmann, Johann Wolfgang GoetheUniversitat Frankfurt am / Aquatic Ecotoxicology; L. Weltje, BASF SE / Agricultural Centre; P. Matthiessen, Independent Consultant. Validated guidelines in line with the OECD Conceptual Framework for the Testing and Assessment of Endocrine Disrupting Chemicals (EDTA) have been developed for rodents, amphibians, fish, aquatic insects and crustaceans. Only aquatic arthropods have been considered in this test battery although the comparison of endpoints relevant for reproduction in invertebrates often shows a much higher sensitivity in molluscs vs. e.g. daphnids. The OECD test guideline programme has thus been extended to cover reproduction effects of chemicals in molluscs. Existing mollusc toxicity test protocols have been reviewed in an OECD Detailed Review Paper that identifies two relevant candidate species for developing freshwater tests: Potamopyrgus antipodarum and Lymnaea stagnalis. However, this review did not clarify which toxicity test design/conditions are the most appropriate for chemicals assessment. Therefore, a mollusc reproduction test guideline will be developed describing partial- and full- life-cycle test protocols in these species, so as to propose a balanced suite of apical mollusc toxicity tests applicable for the assessment of any type of chemical, including endocrine disruptors, as level 4 and 5 assays of the EDTA Framework. The guideline project is led by a consortium of experts (Germany/UnitedKingdom/France/Denmark) from academia, industry and government stakeholders. To date, expert knowledge has been gathered and formed the basis of draft standard operating procedures (SOPs) for the culture and test implementation of the partial life-cycle tests in both species. Prevalidation of these SOPs consisted in two ring-tests involving 10 partner labs with different level of expertise in mollusc tests. Effects of cadmium on the individual reproductive outputs were studied over 28-d in P. antipodarum and over 56-d in L. stagnalis. Pre-validation tests were successful in both species. Indeed, both species could be successfully cultured at the lab. Toxicity tests provided consistent results among laboratories for each species (e.g. homogenous and nonsignificantly different NOECs, LOECs and EC values for most 50 partner labs). The draft SOPs will be optimized (e.g. statistical tests showed that the duration of the test with L. stagnalis can be shortened to from 56-d to 35-d) and further tested for various types of reproductive toxicants including endocrine disruptors. A broad scale ring-test will be conducted in 2014 for both species based upon the consolidated SOPs. 433 Use of vitellogenin as a biomarker for estrogenic effect: What is the baseline? J.E. Morthorst, University of Southern Denmark / Department of Biology; K. Mathiesen, University of Southern Denmark / Institute of Biology; H. Holbech, K.L. Pedersen, P. Bjerregaard, University of Southern Denmark / Department of Biology. Vitellogenin is used as a biomarker for estrogenic effect in laboratory-based

SETAC Europe 23rd Annual Meeting

regulatory test systems OECD TGs 229, 230 and 234 and vitellogenin levels in the plasma of juvenile or male specimens of various fish species have also been used for monitoring estrogenic impacts in freshwater and estuarine environments, especially in association with discharges from waste water treatment plants. The use of any biomarker requires knowledge about the baseline levels in an uncontaminated environment and the present investigation was initiated to identify baseline vitellogenin levels in juvenile brown trout Salmo trutta. Juvenile brown trout were caught by electro-fishing from 15 sites in 12 streams of Funen, Denmark, during late September and early October 2010. The fish were brought to the laboratory, blood samples were taken and plasma vitellogenin concentrations were determined by means of a direct non-competitive sandwich ELISA. Vitellogenin concentrations in the plasma of the 135 male, juvenile brown trout caught during the investigation were all below 52 ng/ml and vitellogenin concentrations in the juvenile, female brown trout were generally one order of magnitude higher. The results indicate that baseline vitellogenin levels in juvenile, male brown trout in an uncontaminated environment are below 100 ng/ml. This also indicates that most of the high values obtained in earlier investigations (2000 to 2004) investigations could be attributed to estrogenic exposure. Six of the streams investigated in 2004 were revisited in the 2010 study and vitellogenin concentrations had decreased. Between 2004 and 2010 some discharges (via septic tanks) from scattered houses in the open land known to discharge estrogenic activity had been removed and this may have reduced the addition of estrogenicity to the streams. 434 Impact of climate change on a risk assessment for intersex in fish due to steroid estrogens V. Keller, Centre for Ecology Hydrology; P. Lloyd, Wallingford Hydrosolutions Ltd; R. Williams, Centre for Ecology and Hydrology. The impact of climate on the natural environment has been a reason for concern for many scientists across the globe. Although water quantity and water resources have been the focus of many studies over the past years, water quality raised less attention. In England and Wales, steroid estrogens, namely estrone (E ), 1 estradiol (E ) and ethinylestradiol (EE ) were identified as being the 2 2 main chemicals causing intersex in male fish. A national risk assessment is already available for intersex in fish arising from these estrogens under current flow conditions. This study, presents to our knowledge, the first set of national catchment-based risk assessments for steroid estrogen under future flow scenarios. A geographically referenced model was used to predict concentrations in surface waters across England and Wales for E , E and EE . The river flows were perturbed 1 2 2 using 3 climate change scenarios for the 2050’s defined by the 2009 UK Climate Projections (UKCP09). These climate change scenarios were chosen to represent a selection of possible changes: ranging from a relatively dry scenario to the wettest scenario available. The effects of demographic changes on estrogen consumption and population served by sewage treatment works were also included by using population projections for the UK in 2050. These predicted concentrations were then combined into estradiol equivalent (E eq) and compared to known 2 biological effect levels to assess the risk of endocrine disruption across England and Wales. This risk was then mapped in order to identify hotspots and quantify how the risk could change in the future compared to the current situation. For the 2050s, depending on the climate scenario selected, between 51 and 54% of the total river length modelled is predicted to be at no risk from endocrine disruption ([E eq] < 1 ng/l). 2 A significant proportion of reaches are predicted to be at risk (1 ? [E 2 eq] < 10 ng/l): between 43 and 45%, and there are between 3 and 4% of reaches estimated to be at high risk ([E eq] ? 10 ng/l). Compared to 2 the present situation (no risk: 61%, at risk: 38, and high risk: 1%), this study indicates the possibility of an increased future risk of endocrine disruption in particular within the high risk category where fish intersex is likely to occur. This study provides a spatial overview of this possible change in risks and may provide regulators and policy makers useful information to prepare for this potential risk. 435 Formulation additives in the environmental risk assessment of biocidal products A. Coors, ECT Oekotoxicologie GmbH; F. Sacher,

108

DVGW-Technologiezentrum Wasser; U. Schoknecht, BAM Federal Institute for Materials Research and Testing; B. Weisbrod, ECT Oekotoxikologie GmbH; A. Kehrer, Federal Environment Agency. Biocidal products often contain a wide variety and considerable number of different formulation additives in addition to one or more active substances. According to the Biocidal Products Directive (BPD) and the Biocidal Product Regulation (BPR), which will replace the current BPD in September 2013, an environmental risk assessment must be performed not only for active substances, but also for any substance of concern contained in a biocidal product. These substances must not be seen isolated, but possible cumulative and synergistic effects shall be taken into account according to the new regulation. In this context, the following question will be addressed in the presentation: Is it sufficient to consider hazardous and dangerous substances together with the active substances in a theoretical mixture toxicity model in order to obtain a toxicity estimate for the product? Theoretical and experimental investigations were conducted here with three wood preservative products and their technical eluates in order to verify if their toxicity can be correctly predicted by the concept of concentration addition (CA). The obtained results demonstrate that formulation additives as well as transformation products of active substances can significantly contribute to or even dominate the toxicity of wood preservatives. For some products and eluates, the consideration of the labelled hazardous components in addition to the active substances was sufficient to predict theoretically by CA the toxicity with a deviation of less than factor 2 from the observed toxicity for a number of typical regulatory endpoints, both acute and long-term. This supports for formulated products the reliability of a theoretical hazard assessment. However, the case of other products and eluates demonstrated that this approach is only reliable if indeed all relevant product ingredients are considered. 436 The UK’s approach to addressing the toxicity of mixed active substance plant protection products M. Reed, Chemicals Regulation Directorate, HSE; J. O'Leary-Quinn, Chemicals Regulation Directorate. For Plant Protection Products (PPPs) containing more than one active substance (a.s.) the combined effects of simultaneous exposure to all the a.s. and the influence of any co-formulants in a PPP should be considered. The approach varies between areas, partly depending on whether formulation data are usually available. It is important that any approach does not result in increased vertebrate testing, so the approach taken for birds and mammals is different to the approach taken for other groups (the requirement for fish formulation testing is set and is based on the sensitivity of the different aquatic groups). For groups other than birds and mammals the acute or shortterm risk is usually assessed via the use of formulation studies. As regards reproductive risk from PPPs it is considered unlikely that the a.s. and all the co-formulants will remain intact and hence that the longterm risk will primarily come from the a.s.. The long-term riskto aquatic life from spray drift of PPPs that contain more than one a.s. is assessed using data on the individual a.s.. This is based on the assumption that the formulation will break down into its component a.s. once it enters the water. For drainflow the potential additive effects should be considered if the a.s. are likely to co-occur. The fate and behaviour of the individual a.s. will determine the degree of cooccurrence and should be considered in the risk assessment. It should also be considered whether the risk is being driven by one a.s.. If all the a.s. within the formulation pass the risk assessment with a margin of safety then the need for further assessment will be limited. For birds and mammals the first step is to determine if one a.s. in the PPP drives the risk. The size of the toxicity exposure ratio for each a.s. in the PPP can be considered. Where it is necessary to calculate the acute toxicity of the PPP the initial approach is based on the toxicity and the proportions of the each a.s. in the PPP and using an additive toxicity approach as described in EFSA 2009 (based on the ‘Finney equation’). CRD currently considers the reproductive risk by firstly looking at the basis of the long term end point for each a.s.. If the NOEC in the reproductive study is based on different parameters then it is considered that the effects are unlikely to be additive and the long term risk from the mixed active PPP can be considered by addressing each active substance independently.

SETAC Europe 23rd Annual Meeting

437 Tiered Mixture Risk Assessment With Modes of Toxic Action: Improving Pesticide Risk Assessment C. Borgert, Applied Pharmacology Toxicology Inc; L.S. McCarty, LS McCarty Scientific Research Consulting; A. Weyers, Bayer CropScience / BCS-D-EnSaETX. The use of simple, highly conservative approaches in risk assessment (RA) can focus effort on mixtures of concern, but additional tiered approaches considering mode of action (MoA) are needed for future mixture regulations and to close knowledge/data gaps. We examined an empirical demonstration offered as justification for applying simple concentration addition (CA) to pesticide mixtures in the aquatic environment irrespective of their toxic MoA. Predicted environmental concentrations (PEC) for 25 herbicides were compared the observed 24h algal toxicity of 23- and 25- chemical mixtures to an estimate derived by applying CA and IA to the algal EC50s for the individual chemicals, irrespective of their pesticidal MoAs. The analysis concluded: "The toxicity of the tested mixture showed good predictability by CA." and "IA slightly underestimated the actual mixture toxicity." Our analysis indicates that the comparison between CA and IA is flawed as a single MoA (Photosystem II) comprises 80% of the toxicity estimate and the experiment fails to test CA against IA based on the usual understanding of their application to chemicals with similar and dissimilar MoAs, respectively. A tiered approach allows a more detailed RA analysis and better fits the data. In the first tier, both a "de minimus" and the CA toxic unit approach used in the herbicide approach above are employed. As this mixture data fails both Tier 1 assessments, it is necessary to move to a Tier 2 assessment. For Tier 2, the CA potency-weighted sum of the components with the most common mode/mechanism of action strongly suggests that the a single MoA - photosystem II inhibition - dominates the algal toxicity of this mixture of herbicides and also explains why CA approximated the casespecific algae toxicity tests carried out. Also, using a critical body residue (CBR) evaluation it appears unlikely that baseline neutral narcosis is contributing to the herbicide mixture toxicity. Plant protection products (PPP) have specific non-narcotic MoAs and mixture narcosis is unlikely at low exposures. However, this is a useful Tier 2 check, ruling out, in this case, an alternative explanation. MoA is vital for estimating mixture risk assessment and possible for data rich substances such as PPP. A tiered approach to regulatory RA based on concepts that better explain the data is preferred over those assuming CA for all chemicals irrespective of exposure concentrations or MoA. 438 Predictive mixture toxicity assessment of pesticides in Swedish surface waters T. Backhaus, M. Gustavsson, University of Gothenburg / Dep. of Biological and Environmental Sciences. Several studies have shown that the combined effect of a pesticide mixture can give rise to higher effects exceeding that of each individual compound. Even if all pesticides are present at levels lower than their no effect concentration (NOEC) the mixture effect might become substantial. We have applied a range of methods based on Concentration Addition to pesticide monitoring data from surface waters in Swedish agricultural areas. A database with data from 6 different locations covering 751 individual weekly samples and a total of roughly 60 000 data provided the input for the study. Ecotoxicological data for each monitored pesticide were extracted from the US EPA database AQUIRE and the EU database eChemPortal. The compiled data comprise results from studies with algae, crustaceans and fish and exposure times of 1-4 days. ECOSAR was used to predict toxicity when experimental data were unavailable. Results indicated that algae are frequently more at risk than crustaceans, which on average are slightly more at risk than fish. There are however individual sampling events where fish and crustaceans have been subjected to a higher mixture toxicity than algae. The average mixture risk quotient is 0.2, ranging from 0.6 to 0.036. Using the typical AF of 1000 would lead in most cases to a mixture risk quotient exceeding 1, which would call for follow-up studies. In more than 60% of the cases the estimated mixture risk exceeded the average risk of an individual pesticide by more than a factor 100. Pesticide risks are expected to fluctuate over seasons, following agricultural activities, with a dip towards the end of the year. However, while in some years almost

109

no seasonal dependence of pesticide risks was observed (e.g. in 2002 and 2003), an increase in pesticide risk over the year was observed in 2004 and 2008. 439 Environmental Risk Assessment of Pesticide Mixtures – Lessons learned from a project on regulatory implementationT. Frische, Federal Environment Agency UBA / Section IV 1.3; R. Altenburger, UFZ Helmholtz Centre for Environmental Research; T. Backhaus, Goteborg University; A. Coors, ECT Oekotoxikologie GmbH; M. Faust, Faust & Backhaus Environmental Consulting GbR; D. Frein, Federal Environment Agency (UBA); D. Zitzkat, UFZ Helmholtz Centre for Environmental Research. A paradigm shift is underway in the environmental risk assessment of chemicals leaving behind the traditional “individual substance approach”. For plant protection products (PPP), the new regulation 1107/2009 thus requires in article 29 that “interaction between the active substance, safeners, synergists and coformulants shall be taken into account” in the evaluation and authorisation. This explicitly refers to marketed PPP, which are by origin technical mixtures containing one to several active ingredients plus several co-formulants. Moreover, common agricultural practice comprises also the application of two or more PPP simultaneously (tankmixtures) as well as the repeated application of several PPP during the growing season (serial applications). Despite this well-justified concern for exposure of non-target organisms towards pesticide mixtures in the environment, the associated risks are largely not considered in risk assessment so far. Obviously, todays` lack of sufficiently specific technical guidance is a major obstacle for a widely accepted and consistent implementation of mixture risk assessment under regulation 1107/2009. Against this background, the research project “Ecotoxicological combined effects from chemical mixtures” was commissioned by the German UBA in 2009 addressing the relevance and means to account for mixture toxicity within the environmental risk assessment in the authorization process for PPP (and biocide products). Major outcomes of the project are presented in order to illustrate the lessons learned for the on-going regulatory implementation of mixture risk assessment: (i) The data for performing mixture risk assessment are available (i.e. comparatively data rich situation for PPP). (ii) The methodology for component-based predictive approaches - especially Concentration Addition - is available and seems to provide a reliable tool. (iii) The impact for regulatory decision making appears to be rather modest (i.e. not overly conservative outcomes are to be expected). (iv) Common principles are necessary to define coherent regulatory settings for a mixture risk assessment. (v) Tiered approaches are the way forward to deal with e.g. heterogenous data (i.e. a three-tiered approach allowing for efficient use of the available data and resources will be discussed) (vi) Some questions remain (i.e. regarding indicators for synergism, higher levels of biological complexity, chronic toxicity, realistic exposure assumptions). 440 Life Cycle Assessment of bio-based ethanol produced from different agricultural feedstock I. Muñoz , Unilever / Safety and Environmental Assurance Centre SEAC; K. Flury, N. Jungbluth, ESU Services; L. Mila-i-Canals, G. Rigarlsford, H. King, Unilever. Introduction Bio-based products are often considered sustainable due to their renewable nature. However the environmental performance of products needs to be assessed considering a life cycle perspective to get a complete picture of potential benefits and tradeoffs. In this article we present an attributional life cycle assessment (LCA) of a global commodity, ethanol, produced from different feedstock and geographical origin. The aim is to understand the main drivers for environmental impacts in the production of bio-based ethanol as well as its relative performance compared to a fossil-based alternative. Methods Ethanol production is assessed from cradle-to-gate, although end-of-life emissions are also included in order to allow a full comparison of greenhouse-gas (GHG) emissions, assuming degradation of ethanol once emitted to air from household and personal care products. The functional unit is 1 kg ethanol, produced from the following combinations of feedstock and region: maize grain in USA, maize stover in USA, sugarcane in North- East of Brazil, sugarcane in Centre-South of Brazil,

SETAC Europe 23rd Annual Meeting

sugar beet in France and wheat in France. As a reference, ethanol produced from ethylene in Western Europe is used. Six impact categories from the ReCiPe assessment method are considered, along with seven novel impact categories on biodiversity and ecosystem services (BES). Results and discussion GHG emissions per kg biobased ethanol range from 0.7 to 1.5 kg CO -eq per kg ethanol, and from 2 1.3 to 2 kg per kg if emissions at end of life are included. Fossil-based ethanol involves GHG emissions of 1.3 kg CO -eq per kg from cradle2 to-gate and 3.7 kg CO -eq per kg if end-of-life is included. Maize stover 2 in USA and sugar beet in France appear as the ones with lowest GHG emissions, although when other impact categories are considered tradeoffs are encountered. BES impact indicators show a clear preference for fossil-based ethanol. The sensitivity analyses performed showed how certain methodological choices (allocation rules, land use change accounting, land biomes), as well as some scenario choices (sugarcane harvest method, maize drying) affect the environmental performance of bio-based ethanol. Also, the uncertainty assessment showed that results for the bio-based alternatives often overlap, making it difficult to tell whether they are significantly different. 441 Integrated assessment of prospective scenarios of energy demand by combining economic equilibrium models and LCA E. Igos, CRP Henri Tudor / Resource Centre for Environmental Technologies (CRTE); B. Rugani, University of Siena / Resource Centre for Environmental Technologies (CRTE); S. Rege, E. Benetto, CRP Henri Tudor / Resource Centre for Environmental Technologies (CRTE); L. Drouet, D. Zachary, Public Research Centre Henri Tudor / Resource Centre for Environmental Technologies (CRTE); T. Haas, F. Adam, Service Centrale de la Statistique et des Etudes Economique (STATEC). The energy sector at country level is very sensitive both from a political, economical and environmental viewpoint, due to the high dependency from the availability and cost of fossil fuels. Initiated by the Public Research Centre Henri Tudor, the Luxembourgish statistics institute STATEC and the Luxembourgish Ministry of Sustainable Development and Infrastructures, the project LUXEN (Integrated assessment of future energy scenarios for Luxembourg) aims at assessing the economic and environmental consequences of possible future scenarios of energy supply consumption in Luxembourg. LUXEN has been carried out in two phases: (1) assessment of future energy supply and demand and their consequences on the Luxembourgish economy, and (2) evaluation of related environmental impacts. The first step has been performed by coupling the Energy Technology Environment Model (ETEM) of Luxembourg with the LUXembourg General Equilibrium Model (LUXGEM). Based on the results of this economic- and energy-based coupling, the environmental impacts related to the entire economy were evaluated. Input-output tables in timeseries (from 2005 to 2009) for Luxembourg were extended with environmental satellite accounts both at the domestic production and import levels. Specific EE-IO tables were thus created and further disaggregated within the energy sector to meet the purposes of LUXEN. Indeed, the energy sector (i.e. production and use) within the use table for imports as well as the domestic I/O is replaced by the results of ETEM. The corresponding technologies were adapted from the Ecoinvent database (efficiency ratios and emissions of combustion from ETEM), distinguishing imported or domestically produced inputs and outputs. The results showed that the future energy demand will be mainly covered by additional imports of liquid fuel, gas and electricity, which represent the largest part of the environmental impacts. Transport is the most demanding sector (using liquid fuels), due to the dominance of tertiary activity in Luxembourg and the high number of commuters. These results could steer policy proposals since they reflect forecasts linked to two sustainability pillars at the national scale: economy and environment. The outcomes could be also used to identify future marginal energy supplier to answer to an additional demand (e.g. implementation of electric vehicles) and therefore to assess the environmental impacts using a consequential approach 442 LCA as a decision support tool in the urban water system of Aveiro (Portugal) A. Dias, D. Lemos, University of Aveiro; X.

110

Gabarrell, Universitat Autònoma de Barcelona; L. Arroja, University of Aveiro. The objective of this study is to assess the environmental impacts associated with the urban water system of Aveiro municipality (Portugal) using Life Cycle Assessment (LCA), in order to identify the stages and processes with the higher environmental impacts and to propose improvement scenarios. The whole water system was considered, including the following stages: water catchment and treatment, water distribution and sewage, wastewater treatment and water administration. The production of chemicals, fuels and electricity consumed in theses stages is also considered, as well as the environmental impacts from deposition both in landfill and agricultural soils of the waste generated in wastewater treatment. The transportation of chemicals and fuel consumption by the vehicle fleet of the administration are also considered, as well as the production of ducts used in water distribution and sewage collection. Data on the inputs and outputs of the water system stages refer to the year 2008 and were mostly obtained directly from the entities responsible for their management. The impact assessment method used is the hierarchist approach of ReCiPe. The results show that the stage of water catchment and treatment was the most relevant for the majority of the impact categories, because most of the electricity consumption happened in this stage. For marine eutrophication and marine ecotoxicity, the stage of wastewater treatment carried alone almost all the burden, due to the release of nitrogen and phosphorus in the sea. Moreover, no stage could be regarded as irrelevant (including water administration). The environmental performance of the system could be improved by adopting some measures, such as installation of units for nitrogen and phosphorus removal in the wastewater treatment stage, reduction of electricity demand by improving the efficiency of pumps and other electrical equipment, decrease of water losses in the water distribution system, reduction of water demand, decrease of water inputs to the sewage collection system, and use of solar or other green energy for the pumping systems and for the other equipment that consume electricity. The proposed improvement measures are a basis for the decision making process regarding future investments by local authorities towards environmental sustainability of the urban water system. 443 Environmental assessment of electricity scenarios with Life Cycle Assessment T. LARBI, Mines ParisTech; I. Blanc, Mines ParisTech / Centre for Energy and Processes. The environmental impacts of existing electricity generation systems have already been assessed with Life Cycle Assessment (LCA) studies. However environmental impacts assessment of scenarios is very rarely evaluated through a life cycle perspective partly because of the complex parameters it involves, such as the temporal technology variation and the spatial dependency per country. Moreover some studies only analyze the effect of limited indicators variation. Considering these parameters is however crucial when steering and evaluating policy options with regards to the possible environmental consequences. The main objective of this paper is to present the methodology undertaken to perform an LCA for energy scenarios through the analysis of environmental impacts. Such methodology has been applied to a specific scenario, a ”Renewable” scenario and results are analysed for Austria. This scenario has been developped within EnerGEO European project. The methodology relies on the following assumptions and steps: The share of energy pathways for electricity production have been considered in line with the scenario definition. The current country-specific technical parameters (efficiency, lifetime, etc...) of each energy pathway have been modeled. The temporal evolution of future energy performances have been considered. LCA for each energy pathway has been performed for each year based on ecoinvent 2.2 database. We apply this methodology to evaluate the impacts on climate change and human health of our renewable scenario (100% RE) for Austria. In this scenario, electricity generation in Austria increases from 87 TWh in 2000 to 137 TWh in 2050. The renewable energy sources share, mainly wind and biomass, increases from 50.34 % to reach 80 % for the same period. Oppposite trends are to be found : a substantial decrease of GHG by 75 % due to the decline of fossil energy sources share and a major increase by tenfold of DALYs due to biomass combustion. These first

SETAC Europe 23rd Annual Meeting

results are based accounting for (1) restricted data corresponding to the foreseen future technological performances for this scenario and (2) unknown shares of sub-categories for some energy pathways such as biomass where we had to sort between wood, agricultural crops, biogas...etc.. We therefore propose to customize ecoinvent 2.2 data with additional data from other sources: IEA and NEEDS. With this strategy the unavailability restriction is outstripped. 444 Database and IT infrastructure for Life Cycle Inventories of Bioenergy in Germany L. Schebek, Technische Universitaet Darmstadt / Industrial Material Cycles; W. Poganietz, Karlsruhe Institute of Technology; A. Ciroth, GreendeltaTC GmbH; C. Duepmeier, Karlsruhe Institute for Technology; L. Eltrop, University of Stuttgart; S. Simon, German Aerospace Center; T. Targiel, Wuppertal Institute for Climate, Environment and Energy; H. Wagner, RuhrUniversitaet Bochum; T. Zschunke, Hochschule Zittau/Goerlitz. Bioenergy is an important part of Climate Policy in Germany. Within the German Sustainable Biomass Strategy, the assessment of existing and novel technologies on a life cycle approach has been acknowledged as important contribution to achieve policy goals and the “need of an adequate database” has been identified as one focus of the funding program of the Ministry of Environment (BMU). The project BioEnergieDat has been carried out by a consortium of German research organizations from 2010 – 2012 and developed a comprehensive database and IT infrastructure for bioenergy process chains for German framework conditions. The database comprehends data sets for processes and process chains of selected bioenergy technologies and energy carriers for today’s state of technology as well as for the projected state of technology of 2020/2013. Based on methodological specifications worked out as part of the project, the unit modules may be combined in a flexible way within different process chains and applications. This concept of a modular data supply was enabled on the technical side by a web-based Open Source IT infrastructure. The methodological specifications are based on the idea of a methodological core that specifies fundamental aspects to be supported by all data sets. Datasets for technologies in 2020/2030 are based on a learning curve approach complemented by an analysis of expected market shares of various technologies. The IT infrastructure includes data storage for unit modules, aggregated processes and complete product models and a calculation tool for LCA modelling. As to data storage, a database application based on Internet technology allows local data storage in the modelling tool, but also shared data storage in the central database. In order to ensure continuous provision, up-dating and support of the database, a host organization is being implemented which takes over the operational responsibilities for the database and will be active in future revision, extension of the database and the IT infrastructure. Summing up, the BioEnergieDat database provides core data sets for bioenergy technologies and energy carriers in Germany to be used in support of climate and technology policy as well as further LCA-based applications. The web-based IT infrastructure enables new possibilities for data exchange between different tools, methods and user environments, connecting the field of bioenergy to surrounding fields of application. 445 Adaptation of the LCA framework to support land use planning policies E. Loiseau; P. Roux, Irstea; G. Junqua, Ecole des Mines d'Alès; P. Maurel, V. Bellon-Maurel, Irstea. Since the implementation of the European Directive on strategic environmental assessment (2001), local authorities are in charge of carrying studies on the environmental impacts of land use planning policies. However, they are facing with a lack of standardized approaches to perform such assessement. Methodological developments are therefore needed for the environmental assessment of spatial planning policies adopted on a territory. In order to achieve this, different kinds of tools and methods can be used. Among them, LCA has been identified as a promising tool as it has the ability to avoid burden shifting between environmental impacts as well as between life cycle stages. Yet, no study which performs the environmental assessement of a territory as a whole has been reported. The aim of this presentation is twofold. Firstly, the main

111

methodological bottlenecks which can partlty explain the lack of applications of LCA to territories have been identified and discussed, i.e., (i) functional unit definition, (ii) boundary selection, (iii) data collection and (iv) the refinement of LCIA indicators in order to provide useful information for land planning policies. Secondly, for each bottleneck, proposals have been made to overcome it and to provide a general framework adapted to land planning issues. One of the main adaptations relies on the functional unit definition. The territory (defined by its geographical boundaries) and its related land planning scenario will now be considered as the reference flow. The adapted framework will hence deliver two kinds of indicators (calculated outputs), the environmental impacts and a set of services provided by the territory. This adapted framework offers a new perspective on the issue of the environmental assessment of land planning policies by proposing a global approach which include all the activities located on a territory, avoiding burden shifting between territories, prioritizing environmental issues and identifying the most polluting activities on a territorial context which could need a complemetary analysis (e.g., environmental risk analysis). 446 Practical aspects in water quality monitoring of pesticides in streams; autosamplers and GamTox M.D. Koster, Amt für Umwelt Thurgau / Gewaesserqualitaet. In areas of intensive agricultural land use, monitoring water quality in flowing waterways is often expensive and frustrating. The strategy that the Water Quality Board of Canton Thurgau (Switzerland) has developed for their many small streams involves ecotoxicological testing in situ with local Gammarids, pesticide analysis with autosamplers, and benthic fauna indices, besides the usual water chemistry. This presentation aims to show results of small-scale intensive monitoring of a stream in a fruit growing area. It also introduces our passive monitoring device for collective water samples, and our experiences with the caged Gammarus GamTox toxicity test. In this case-study, highest pesticide input came from the drainage connected to the farm yard, despite protection measures and Good Agricultural Practice. Key Words: pesticides, Gammarus, stream toxicity test 447 Calculation of community responses to pesticide exposures from mesocosm data: comparison between SPEAR and PRCA. Focks, Wageningen UR / Mathematics/Computer Science; R. Bhatta, A. Stegerman, Wageningen University / Aquatic Ecology and Water Quality Management; P.J. van den Brink, Alterra and Wageningen University. Aquatic higher tier testing of pesticides using mesocosm experiments is regularly performed as part of the registration procedure. The increase of the ecological relevance of mesocosm tests takes it toll: often complex response patterns are the result of such experimental approaches. To serve as decision criterion for the registration of pesticides, these complex response patterns need to be aggregated, mostly into a concentration where no community effects of a pesticide could be observed in comparison to controls (NOEC ). We community compared two techniques for determining NOEC values, the community Principal Response Curves (PRC) and the SPEcies At Risk (SPEAR) indicators. Mesocosm data was used for three substances: chlorpyrifos(CPF), fluazinam (FLU) and ?-cyhalothrin (CYH). Numbers of different macroinvetebrate species were counted at the given sampling dates in control mesocosms and cosms exposed to different treatment levels. The PRC analysis was performed using CANOCO (version 4.5). The SPEAR calculation was performed as pesticide described in the literature. To adjust the species sensitivities, LC50 values were taken from the ECOTOX database (http:// cfpub.epa.gov/ecotox) and specific sensitivities were calculated. NOEC calculations were performed for each sampling date commnunity by applying the Williams test on the community data set of that particular sampling date by using the two methods. For the CPF data set, the PRC detected most NOEC values at lower concentrations and at more time points than SPEAR. For the FLU data set, PRC detected NOEC values throughout at much lower concentrations community (table 2), probably because the sensitivity ranking by SPEAR is done based on data dominated by organophosphates which does not represent

SETAC Europe 23rd Annual Meeting

the sensitive towards the fungicide very well. An increase of species at risk over time was indicated by the SPEAR index for this compound (temporal dynamics not shown here). For the CYH data set, the results indicated always lower or equal detection levels for PRC when compared to SPEAR. In our evaluation of the sensitivities of PRC and SPEAR to detect the effects of pesticide exposure from mesocosm data it appeared clearly that PRC detected lower NOEC valuesin community all five evaluated mesocosm experiments. Our results indicate that PRC is applicable with less effort and provide a higher level of detection of effects of pesticides on aquatic communities from the evaluated mesocosm experiments. 448 Impact of polluted sediment resuspension events on chemical and biological water quality guidelines E. Prygiel, Université Lille; P. Superville, L. Lesven, D. Dumoulin, B. Ouddane, University Lille1; O. Geffard, A. Francois, Irstea; J. Prygiel, Agence de l'Eau ArtoisPicardie and University Lille1; G. Billon, University Lille1. In northern France, numerous rivers have been channeled to allow navigation. th th During the 19 and the 20 centuries, the Nord Pas-de-Calais region was an important industrial area where the main activities were coal extraction, heavy industry and textile industry. From these activities, a strong pollution of soils and sediments by metals like Pb, Zn, Cd, Hg, Cr…was inherited and some sites are even considered as among the most polluted in Europe. Although industrial emissions have been dropping drastically these last ten years along with the shutting down of polluting factories (such as that of Metaleurop located near the city of Douai), historical sediment pollution still persist in rivers. In some canals, the navigation has been given up so that the resuspension events are rather limited, but they still occur from time to time during river maintenance, restoration or strong raining occasions. In the largest canals, fluvial transport tend to increase with barges able to contain thousands tons of merchandises. As a consequence, resuspension events of contaminated sediments occur recurrently. The impact of sediment resuspension on the water quality has been studied: (i) at high frequency by voltammetry; (ii) by using integrative chemical sensors (DGT, Diffusive Gradients in Thin films) deployed several days in situ; and (iii) biologically with caging experiments. For that purpose, the crustacean Gammarus fossarum was selected to act as a bioindicator of the water quality. Three canals were selected for this study: the Deûle River, navigated with highly polluted sediments, the Scarpe River, highly polluted but without navigation and the Sensée River, navigated with relatively low pollution level towards trace metals. Preliminary ecotoxicological data shown that only one of measured life traits (feeding rate) was slightly inhibited in the Deûle River. For the other markers studied (molting, oocyte growth, fecundity and fertility), no significant difference has been observed between the 3 rivers. In the other hand, the trace metal dynamic composition of waters, measured on line by voltammetry, shows relevant daily cycle with higher concentrations daytime in the Deûle River while high labile trace metal fractions were accumulated on DGT. These apparent differences between chemical water quality in the Deûle River and caged organisms responses will be discussed as well as the consequences on the WFD ecological and chemical status assessment. 449 Behavioural, physiological and biochemical markers in damselfly larvae as a tool for assessing effects of accumulated metal mixtures N. Van Praet, Antwerp university / Department of Biology; M. De Jonge, University of Antwerp / Biology; R. Blust, University of Antwerp / Systemic Physiological and Ecotoxicological Research, Department of Biology; R. Stoks, University of Leuven / Department of Biology; L. Bervoets, University of Antwerp / Dept. of Biology, Systemic Physiological and Ecotoxicological Research group. The present study investigated relationships between accumulated metals with behavioural, physiological and biochemical endpoints in the damselfly Ischnura elegans. Concentrations of accumulated metals were analyzed in larvae of the damselfly Ischnura elegans, captured from seven different Flemish ponds. Furthermore different physiological (energy availability), biochemical (acetylcholinesterase (AChE) and glutathione-S-transferase (GST)) and behavioural endpoints (locomotory

112

activity and the feeding rate) were measured. Accumulated metal levels significantly differed among ponds, both expressed separately or as metal load. The accumulated metal load influenced some of the measured endpoints on different levels of biological organisation. For all measured levels of organisation pond differences are found but only for GST and energy availability this could be partly predicted by the measured mixed accumulated concentrations, expressed as metal load, whereas non of the behavioural endpoints could be predicted by the accumulated metal load. Moreover the major prediction occurs by GST followed by energy availability. By using biochemical and physiological endpoints the effect of accumulated trace metal mixtures could be investigated under natural field conditions, which can ultimately lead to the construction of more accurate biota standards including various metals mixtures. 450 Trace element bioaccumulation and effects in indigenous brown trout (Salmo trutta) and caged brown trout eggs exposed to shooting range run off L.S. Heier, UMB / Centre for Environmental Radioactivity (CERAD CoE); E. Mariussen, Norwegian Defence Research Establishment; H. Teien, Norwegian University of Life Sciences (UMB) / Centre for Environmental Radioactivity (CERAD CoE); B. Salbu, Norwegian University of Life Sciences (UMB) / Department of Plant and Environmental Sciences; B.O. Rosseland, Norwegian University of Life Sciences / Department of Ecology and Natural Resource Management. Concern has been raised regarding release of munitions related contaminants in firing ranges such as metals/metalloids and explosives (e.g trinitrotoluene). In some Norwegian military firing ranges there is a legacy of metal deposits which has led to elevated metal and metalloid concentrations in lakes and streams, mostly lead (Pb), copper (Cu), zinc (Zn) and antimony (Sb). The objective of the present study was to investigate the metal accumulation and potential negative effects in brown trout (Salmo trutta) inhabiting a lake affected by run off from ammunition deposits, as well as to investigate the survival and metal accumulation of caged trout eggs in the outlet stream. A lake nearby, not affected by metal contamination from ammunition, was used as a reference. Indigenous brown trout were caught by gillnets. Accumulation of trace elements were determined in gills, blood, liver, kidney, brain, muscle and brain tissue. Determination of delta-aminolevulinic acid dehydratas (ALA-D) were done in red blood cells. Brown trout eggs were dry fertilized and immeditaty placed in cages in the outlet streams of the lakes. The eggs were checked for mortality and development and some eggs were collected for metal accumulation at 4 different times during the exposure, after approximately 110 day degrees (dd) (eggs), 260 dd (eyed egg), 480 dd (at hatching) and 680 dd (alevins). Water samples for determination of metal concentrations, total organic carbon and major cations and anions were collected at each sampling. The brown trout in the lake had very high bioaccumulation of Pb, and the highest accumulation was observed in the kidney, bones and gills. In addition to high concentrations of Pb there is also an additional stress in periods with low pH and elevated levels of Cu. There was a strong inhibition of ALA-D in the trout confirming exposure to high Pb levels, which may cause negative adverse effects. Trout eggs exposed in the outlet stream accumulated Pb, Cu and Sb, and higher mortality was observed after hatching. The high Pb concentrations along with additional stressors such as Cu and low pH could possibly cause lower reproduction of fish in the lake, explaining the lack of young yearclasses during the test fishing. 451 POPs in catfish (Silurus glandis): bioaccumulation and effects S. Lacorte, M. Valenzuela, S. Fabregat, M. Faria, IDAEA-CSIC / Environmental Chemistry; C. Barata, CSIC / Environmetal Chemistry. Fish have the ability to accumulate Persistent Organic Pollutants (POPs) through the diet and become good indicators of environmental pollution. In areas impacted by historic and continuous releases of POPs, unexpected high levels of POPs may affect fish condition, behavior, reproduction, and may lead to serious consequences for the development of the species, reinforcing the necessity of a better knowledge on the presence of POPs in sensitive and impacted areas which are refuges for

SETAC Europe 23rd Annual Meeting

numerous wildlife species. In the small city of Flix, there is a chloroalkali industrial area which historically has produced organochlorinated compounds such as PCBs and OC pesticides and has resulted in the contamination of river waters and sediments in this area and downstream to the Ebro delta [1,2]. The aim of this study is to study the bioaccumulation and biological effects of organochlorine pesticides and polychlorinated biphenyls in the centran European catfish (Silurus glandis) sampled across a pollution gradient of a chlor-alkali superfun site located in the low Ebro River, Spain. Integration of analytical tools and toxicological tools were used to determine the levels and effects of high levels of contaminants in a species which is both invasive and with high gastronomic value. 452 Ecological risk assessment of atrazine in North American surface waters K.R. Solomon, University