@phdthesis{8665764,
  abstract     = {{Along with a worldwide growth of the human population, the aquatic environment on our planet is facing an ever increasing chemical input. With the aim of regulating chemical use and protecting both humans and the environment a number of regulatory frameworks for chemicals have been introduced in the European Union (EU). Yet, these regulations mainly focus on a limited list of priority pollutants that represent only a minor fraction of potentially water-emitted chemicals. In addition, even though there exist regulatory frameworks for the marine environment, risk assessment for the latter does not require ecotoxicity data for marine species. To partly address the herewith associated ecotoxicity data gap for marine species, in the present work, marine ecotoxicity data was generated for a total of 23 chemicals of emerging concern (CECs) with two species representing algae and crustacean (Chapter 4). While algae did show low sensitivity to all tested substances, relatively low effect concentrations on crustacean were found for 4 neonicotinoid insecticides. Acute and sub-chronic ecotoxicity data for crustacean was subsequently used together with existing literature data to derive Environmental Quality Standards (EQS) for the marine environment. Inclusion of the marine copepod data from this study led to a refinement of the EQS for clothianidin and thiamethoxam. An in-depth risk assessment for the Belgian part of the North Sea (BPNS) based on the derived EQS for the 4 neonicotinoid insecticides and their mixture (Chapter 4) resulted in an exceedance of predicted no-effect concentrations (PNECs) in the harbors of Ostend and Zeebrugge and a low margin of safety (MoS) for one coastal locations in front of each of these harbors.
Such derivation of EQS and risk assessments are time-intensive processes that require in-depth ecotoxicological and regulatory knowledge. Considering the huge amount of chemicals present in the marine environment, from a regulatory perspective, there are only two promising approaches to handle this complex task: i) automation of EQS/PNEC derivation and associated risk assessment on a substance-by-substance level or ii) moving from a single substance-based to a mixture-based risk assessment. Therefore, in Chapter 5 an automated calculation algorithm was developed and applied in a screening-level risk assessment for the BPNS. This screening-level marine risk assessment suggests to prioritize in future work Bisphenol A, certain herbicides, neonicotinoid insecticides and steroids for further ecotoxicological testing and/or refined PNEC calculation. Additionally, a comparison of grab sample and passive sampler-based risk assessment revealed no obvious differences between the two sampling methods. 
Although providing a useful tool for prioritization within the prevailing regulatory frameworks, a single-substance-based risk assessment bears the risk of neglecting interactive effects of chemicals. Yet, environmental risk assessment is meant to assess the real impact on ecosystems or species that are exposed to chemicals and for most of our waters this means simultaneous exposure to various chemicals. This indicates that there is a need for mixture-based risk assessment methods. To answer this need, we developed a novel method for passive sampler-based ecotoxicity testing of environmentally realistic chemical mixtures (ERCMs) in Chapter 6. This passive sampler-based method combines environmental sampling and ecotoxicity testing of chemical mixtures. During method development insights into the preservation of complex mixture samples were gained and the importance of a reduction of passive sampler extract handling and storage time were highlighted. With a relatively low sample enrichment of < 2 the developed method had one major drawback. 
This drawback was tackled in Chapter 7 where the previously developed method was modified to allow sample enrichment up to a relative enrichment factor (REF) of 44 as compared to environmentally realistic concentration levels. Further, the method was extended with a MoS assessment serving as indicator for potential risks in the BPNS. Here, margins of safety were found to be < 10 for 5 out of 8 samples from different sampling campaigns (SCs) and locations. According to current risk assessment procedures this suggests ecological risks for these locations since the lowest assessment factor (AF) in use is 10. This effect-based method addresses the lack of current environmental regulations that do not provide guidance on how to deal with mixtures of chemicals although simultaneous exposure to multiple chemicals is the prevailing scenario for aquatic organisms. Yet, a change in environmental regulation from single substance to mixture-based risk assessment is not to be expected in the near future since many of the current EU frameworks have only recently entered into force. Thus, to align our effect-based monitoring method with current risk assessment procedures, we recommend to extend the biotest battery with at least one crustacean and one fish biotest to comply with regulatory requirements.
In a first attempt to identify mixture toxicity driving chemicals, we applied multivariate statistics to find chemical concentration patterns in different speedisk extracts that might be associated with the toxicity observations (Chapter 7). Unfortunately, no clear patterns distinguishing between toxic and non-toxic samples could be identified based on 89 target personal care products (PCPs), pesticides and pharmaceuticals. Nevertheless, some chemicals like sodium diclofenac or naproxen, that had also been identified in the screening-level risk assessment to be potentially problematic substances, were found to be correlated with the first two principal components (PCs) that explained 55 % of the data inherent variability. Yet, in order to gain better insights into the effect-driving chemicals in a mixture including non-target chemical data is highly recommended. This would be another step forward from a single substance-based and priority pollutant-focused to an unbiased mixture-based risk assessment.}},
  author       = {{Moeris, Samuel}},
  isbn         = {{9789463573375}},
  keywords     = {{Ecotoxicology,Chemicals of emerging concern,Chemical mixtures,Mixture toxicity,Environmental Risk Assessment}},
  language     = {{eng}},
  pages        = {{xv, 198}},
  publisher    = {{Universiteit Gent. Faculteit Bio-ingenieurswetenschappen}},
  school       = {{Ghent University}},
  title        = {{An effect-based monitoring approach for environmental risk assessment of chemicals of emerging concern and complex chemical mixtures in the marine environment}},
  year         = {{2020}},
}