Class-specific mechanistic regimes governing sediment–water partitioning of emerging contaminants. (IMAGE)
Caption
Class-specific mechanistic regimes governing sediment–water partitioning of emerging contaminants. Schematic overview of the study framework and key findings. A nationwide dataset comprising 5,085 paired sediment–water records was compiled across China's major river basins for three emerging contaminant classes: per- and polyfluoroalkyl substances (PFASs), antibiotics (ABs), and endocrine-disrupting chemicals (EDCs). Through a multi-branch multi-head attention (MB-MHA) machine learning architecture integrated with molecular dynamics simulations, the study identifies class-specific mechanistic regimes in sediment–water partitioning, quantified by the distribution coefficient log Kd. PFASs are governed by ion-mediated interfacial control, antibiotics are dominated by molecular properties, and EDCs exhibit synergistic multiscale regulation involving molecular, geochemical and basin-scale factors. The framework enables spatially explicit prediction and risk mapping, distinguishing high-log Kd zones where contaminants tend to accumulate in sediments from low-log Kd zones where they remain mobile in the water column. Integrating molecular interactions with geochemical and basin-scale controls reveals class-specific partitioning regimes and improves contaminant risk prediction at the basin scale.
Credit
Environmental Science and Ecotechnology
Usage Restrictions
Credit must be given to the creator.
License
CC BY