Answers are grounded in this paper and cite their sources.
Accurate assessment of river pollution risk under changing climatic conditions is essential for sustainable water resources management and environmental protection. This study presents a climate-informed approach that integrates machine-learning-generated extreme hydroclimatic scenarios with a physically based hydrodynamic model and a Lagrangian particle tracking scheme to characterise contaminant transport under extreme flow conditions. An extreme generative adversarial network is employed to generate synthetic extreme precipitation scenarios, which are subsequently translated into hydrological forcing for a calibrated and validated two-dimensional TELEMAC-2D model. The simulated hydrodynamic fields are then coupled with a Lagrangian particle tracking model to investigate contaminant transport across an ensemble of 100 independent realisations of the 100-year flood scenario, corresponding to a 1% annual exceedance probability. The proposed method is applied to the Martil River basin in northern Mediterranean Morocco, where pronounced hydroclimatic variability, combined with increasing anthropogenic pressures, poses significant challenges to the protection of water quality. The probabilistic results reveal pronounced spatial heterogeneity in contaminant arrival across the analysed area. Specifically, 69.58% of the area exhibits an arrival probability below 0.2, whereas 19.38% is reached in more than half of the simulated realisations and 11.25% in more than 80% of them. The analysis further identifies a recurrent downstream transport corridor, characterised by consistently high particle arrival probabilities, while transport into other floodplain areas is more strongly dependent on the characteristics of individual flood realisations. These findings highlight the importance of an ensemble-based probabilistic procedure for distinguishing persistent contaminant transport pathways from those that emerge only under particular extreme flood conditions. The proposed approach provides a quantitative basis for identifying river reaches and floodplain areas that may warrant priority attention in environmental monitoring and flood-response planning. More broadly, it offers a transferable modelling approach for characterising the spatial variability and likelihood of contaminant transport under synthetic extreme hydrological conditions, while providing a foundation for subsequent investigations incorporating additional pollutant processes and sources of uncertainty.
No takes yet. Share an insight, caveat, or question.
Belhajjam et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: