A machine learning based approach for estimating site-specific partition coefficient Kd of organic compounds: Application to nonionic pesticides

The partitioning coefficient Kd for a specific compound and location is not only a key input parameter of fate and transport models, but also critical in estimating the safety environmental concentration threshold. In order to reduce the uncertainty caused by non-linear interactions among environmental factors, machine learning based models for predicting Kd were developed in this work based on literature datasets of nonionic pesticides including molecular descriptors, soil properties, and experimental settings. The equilibrium concentration (Ce) values were specifically included for the reason that a varied range of Kd corresponding to a given Ce occurred in a real environment. By transforming 466 isotherms reported in the literature, 2618 paired equilibrium concentrations of liquid-solid (Ce-Qe) data points were obtained. Results of SHapley Additive exPlanations revealed that soil organic carbon, Ce, and cavity formation were the most important. The distance-based applicability domain analysis was conducted for the 27 most frequently used pesticides with 15952 pieces of soil information from the HWSD-China dataset by setting three Ce scenarios (i.e., 10, 100, and 1000 μg L-1). It was revealed the groups of compounds showing log Kd < 0.06 and log Kd > 1.19 were composed mostly of those with log Kow of -0.800 and 5.50, respectively. When log Kd varied between 0.100 and 1.00, it was impacted by interactions among soil types, molecular descriptors, and Ce comprehensively, which accounted for 55% of the total 2618 calculations. It could be concluded that site-specific models developed in this work are necessary and practicable for the environmental risk assessment and management of nonionic organic compounds.