Abstract

This PhD looks to address the growing demand for advanced ecotoxicological tools for effluent monitoring, in light of stricter water quality standards and increased monitoring requirements. Focusing on real-time effluent monitoring, the project ToxPrints aims to utilise the behavioural changes in aquatic organisms caused by chemical pollution surges. This industry PhD extends the collaboration with Viewpoint, co-creators of the ToxMate biomonitoring system, currently in use in French wastewater treatment plants (WWTP). The primary goal of ToxPrints is to discover "behavioural fingerprints," patterns in behavioural responses that can identify the nature of chemical pollution. The research involves three main objectives: characterising behavioural differences in response to chemical contamination for a single species, defining behavioural fingerprints for a set of substances at both single and combined species levels, and grouping these fingerprints to uncover links between contamination nature and behavioural alterations. Extensive data collection through ToxMate experiments was the first step in creating a behavioural database for over 50 micropollutants relevant to wastewater treatment. Field data used in the research comes from monitoring sites, and continuously enriches the database with naturally occurring chemical surge events. The subsequent data analysis is heavily based on functional data analysis (FDA), a rapidly developing field of applied maths suitable for time series classification with statistical techniques such as PCA and supervised and unsupervised machine learning algorithms. (As well as some deep learning techniques) The project ultimately aims to understand the structural properties linking groups of contaminants to look to compare them with in-silico toxicity prediction tools such as quantitative structure-activity relationship (QSAR) models used in risk assessment. We also look to enhance the identification of unknown chemical surges in complex media such as WWTP effluents, to move towards better water treatment and limit toxic impacts on the receiving environment.

Cite the thesis

George Ruck. A Data-Driven Exploration of Invertebrate Behavioural Fingerprinting for Advanced Ecotoxicological Monitoring in Effluent Environments. Ecology, environment. Université Claude Bernard - Lyon I, 2024. English. ⟨NNT : 2024LYO10049⟩. ⟨tel-05181187⟩

Access manuscript on HAL thèses

Publication

• George Ruck, Alexandre Decamps, Jean–Baptiste Aubin, Hervé Quéau, Laura Garnero, et al.. Avoidance behaviour of aquatic macroinvertebrates for real–time detection of micropollutant surge in wastewater effluents. Water Research, 2023, 242, pp.120228. ⟨10.1016/j.watres.2023.120228⟩. ⟨hal–04372023⟩