Predicting sorption of organic pollutants on soils with interpretable machine learning.

Environ Pollut

School of Resources and Geoscience, China University of Mining and Technology, Xuzhou, 221116, China. Electronic address:

Published: October 2025


Category Ranking

98%

Total Visits

921

Avg Visit Duration

2 minutes

Citations

20

Article Abstract

The sorption of organic pollutants (OPs) on soils plays a critical role in determining the environmental fate and transport of these compounds, which has been extensively studied. However, the complex nonlinear relationships between adsorption capacity and multiple influencing factors, as well as the relative contributions of these factors to adsorption behavior, remain inadequately understood. This study develops five machine learning (ML) models-support vector machine (SVM), deep neural networks (DNN), extreme gradient boosting (XGBT), random forest (RF), and gradient boosting decision tree (GBDT)-using a dataset of 352 data points from previous studies to predict OPs sorption on soils based on multiple factors. Shapley additive interpretation (SHAP) is applied to perform interpretability analysis based on the model exhibiting superior performance. Additionally, the distribution map of the sorption capacities of 12 OPs across mainland China is generated using the interpretable ML model. The results indicate that the XGBT model demonstrates superior performance, achieving a coefficient of determination of 0.952 and a root mean square error of 0.103 for the testing dataset. Interpretability analysis reveals that the electronic effects (E) of OPs and soil organic matter (SOM) content are the most influential factors. This finding underscores the dominant roles of π-π interactions and hydrophobic partitioning in the sorption mechanisms. The distribution map indicates that high sorption capacities are predominantly located in southern and southwestern regions, correlating with reduced environmental risks. This study presents a novel interpretable ML framework for predicting OPs adsorption potential and offers valuable insights into the mechanisms governing OPs sorption on soils. Furthermore, this framework supports the environmental management applications in risk assessment, land remediation strategies planning, and soil protection policy.

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.envpol.2025.126665DOI Listing

Publication Analysis

Top Keywords

sorption organic
8
organic pollutants
8
machine learning
8
gradient boosting
8
ops sorption
8
sorption soils
8
interpretability analysis
8
superior performance
8
distribution map
8
sorption capacities
8

Similar Publications

Adsorption-desorption behavior of difenoconazole onto soils: Kinetics, isotherms, and influencing factors.

Pestic Biochem Physiol

November 2025

National Engineering Laboratory for Efficient Utilization of Soil and Fertilizer Resources, Key Laboratory of Agricultural Environment in Universities of Shandong, College of Resources and Environment, Shandong Agricultural University, 61 Daizong Road, Tai'an 271018, PR China. Electronic address: wj

Difenoconazole (DFC) is a commonly used triazole fungicide known for its high efficiency and environmental persistence. A thorough understanding of its environmental behavior, particularly sorption in soil, is critical to obtain a comprehensive assessment of the ecological risk of DFC. In this study, three soils with distinct physicochemical properties (brown soil, cinnamon soil, and fluvo-aquic soil) were used to elucidate the adsorption mechanisms of DFC on soil.

View Article and Find Full Text PDF

Microplastics (MPs) are emerging vectors for hydrophobic organic pollutants, including polycyclic aromatic hydrocarbons (PAHs), in aquatic environments. Due to their high surface area and sorption potential, MPs can enhance the environmental persistence and bioavailability of toxic compounds, posing potential risks to both aquatic organisms and human health. This study investigates the distribution, sorption behavior, and effects on pollutant transport, distribution, and exposure pathways of PAHs-contaminated microplastics in two major Romanian rivers: the Prahova and Ialomita.

View Article and Find Full Text PDF

Porous metal-organic polyhedra (MOPs) have strong covalent and coordinate bonds that define the intrinsic pore of the cage. The intermolecular interactions between cages tend to be weaker, such that they rearrange during the solvent exchange process preceding gas sorption measurements. The reduction in crystal size that this often causes limits the availability of structural data that could enable understanding of observed gas uptake.

View Article and Find Full Text PDF

The transport of micropollutants through soil is, inter alia, largely influenced by their interaction with humic acids (HAs). As chemically complex carbon molecules, HAs make part of natural organic matter and play a significant role in the retention of micropollutants in the environment. This study examines the interactions of pH-dependent HA fractions with metazachlor, paracetamol, and caffeine, using the surface plasmon resonance imaging (SPRi) method.

View Article and Find Full Text PDF

Investigating chemical composition at the sea-air interface: A subsampling approach for marine surface microlayer analysis.

Sci Total Environ

September 2025

Faculty of Civil and Environmental Engineering, Technion, Israel. Electronic address:

The marine surface microlayer (SML) is distinct from the subsurface water by physical, chemical and biological properties. Being the interface, the SML regulates mass and energy transfer between the ocean and the overlying atmosphere. Given the wide surface area covered by oceans, even small change in flux may have a significant global impact.

View Article and Find Full Text PDF