Probabilistic Ecological Risk Assessment of Synthetic Musk Compounds in the Han River Estuary: Integrating Multi-Year Monitoring and Explainable Machine Learning.

Synthetic musk compounds (SMCs) are emerging contaminants frequently detected in aquatic environments due to their widespread use in personal care and household products. This study investigated the occurrence, environmental drivers, and ecological risks of twelve SMCs in the Han River estuary and adjacent coastal waters, Korea, using three years of monitoring data (2020-2022). A total of 222 water samples from 13 monitoring stations were analyzed by GC-MS. 1,3,4,6,7,8-hexahydro-4,6,6,7,8,8-hexamethylcyclopenta-(g)-2-benzopyran (HHCB) was the dominant compound, followed by musk ketone (MK). Total SMC concentrations ranged from non-detectable levels to 157.6 μg/L, with the highest concentrations consistently observed at wastewater treatment facilities. XGBoost-SHAP analysis identified total nitrogen (TN) as the most influential predictor of SMC occurrence. This association reflects the shared wastewater origin of the two rather than any causal link, so that TN serves as an indicator of wastewater influence rather than of SMC concentration itself. Ecological risks were evaluated using Monte Carlo-based probabilistic hazard quotient (HQ) analysis. HHCB and MK exhibited the highest risks, with mean HQ values exceeding 1.0 at wastewater treatment facilities, indicating potential ecological concern. In contrast, most coastal sites showed low risk levels. The observed spatial patterns identified wastewater discharge zones as major hotspots of SMC exposure and ecological risk. These findings highlight the importance of wastewater-derived SMCs as emerging contaminants in estuarine environments and demonstrate the utility of integrating long-term monitoring, explainable machine learning, and probabilistic risk assessment for ecological risk characterization.