Groundwater contamination risk assessment is an important basis to support the protection of the groundwater ecological environment. In this research, the groundwater contamination risk in typical karst areas in China was evaluated by PLEIK model (P: protective cover; L: land use; E: epikarst development; I: infiltration conditions; K: karst development) and classification and discrimination methods were used to assess groundwater vulnerability and pollution source load; the water quality index method was used to assess the status of groundwater contamination in the research area. The results show that groundwater vulnerability values in the research area range from 3.04 to 7.79, and most areas have low groundwater vulnerability. Groundwater pollution loading indexes, in the region of the pollution sources, gathered numerical up-water quality status evaluation that shows that most of the regional groundwater quality in the research area has good groundwater at present. The groundwater contamination risk assessment results show that the groundwater pollution risk is mainly at a very low level in most areas of the study area, but the groundwater pollution risk is higher in the areas where the pollution sources gather. The result reveals that the regional groundwater contamination risk level for regional groundwater ecological environment protection provides a theoretical basis for policy making.

Accurately identifying groundwater contamination sites is vital for groundwater protection and restoration. This study aims to use a machine learning (ML) approach to identify groundwater contamination sites with total petroleum hydrocarbons (TPH) as target contaminants in a case study of gas stations in China. Firstly, six classical ML algorithms, including logistic regression, decision tree, gradient boosting decision tree (GBDT), random forest, multi-layer perceptron, and support vector machine, were applied to develop the identification models of TPH-contaminated groundwater with 40 features and the performances were compared. The comparison results showed that the GBDT model achieves the best prediction performance, with F1 score of 1 and AUC value of 1. Next, Bayesian optimization optimized GBDT (BO-GBDT) was conducted to further decrease the training time from 19,125 s to 513 s while maintaining the same prediction performance (F1 score = 1, AUC = 1). Finally, Shapley additive explanations (SHAP) analysis was performed on the BO-GBDT model. The SHAP results displayed that the critical feature variables in the BO-GBDT model include wind, population, evaporation, total potassium in the soil, precipitation, and leakage accident. This study demonstrated that BO-GBDT is one satisfactory model to identify groundwater TPH-contamination at gas stations. The method proposed in this study has the potential to be applied to other types of groundwater contamination sites.

Hydrochemical data were utilized in this study to elucidate the hydrogeochemical characteristics and genesis of high-arsenic groundwater in the North China Plain, employing both traditional hydrogeochemical analysis and unsupervised machine learning techniques. The findings indicate that the predominant hydrochemical types of groundwater in the study area are HCO₃-Ca·Na and SO₄·Cl-Na·Ca. The primary mechanism influencing groundwater chemistry has been identified as rock weathering. The unsupervised machine learning framework incorporates various methods, such as principal component analysis (PCA), non-negative matrix factorization (NMF), machine learning models (gradient boosting trees and random forests), and cluster analysis to explore the characteristics and genesis of groundwater hydrochemical types within the study area. This study demonstrated that the formation mechanism of high-arsenic groundwater results from multiple interacting factors.

Per- and polyfluoroalkyl substances (PFASs) presented in landfill leachate can be transported to groundwater and soil, posing risks to human health in food and water. However, the occurrence characteristics of PFASs in landfill leachate and the influence paths of abiotic and biotic factors have not been fully discussed. Our study found that the detection rate of PFASs in the leachate samples was 100%; ΣPFASs ranged from 1328 ng/L to 37,447 ng/L; and the average ΣPFASs was 9593 ng/L. Most of the physical and chemical indexes in the landfill leachate showed a significant positive correlation with PFASs concentration, with the main physical and chemical factors including TOC, sodium ion, chromium ion, and chloride ion. Moreover, dissolved organic matter had a very important influence on the environmental behavior of PFASs. The degree of dissolved organic matter (DOM) humification promoted the transportation of PFBA and PFBS from the landfill waste to leachate, while microbial DOM inhibited the transportation of ∑₁₃ PFCAs. The microorganisms in the leachate were mainly Firmicutes, Campylobacterota and Proteobacteria, among which there was a negative correlation between PFBS and Proteobacteria, indicating that PFBS was highly toxic to Proteobacteria and would inhibit the growth of Proteobacteria in leachate. Firmicutes and Campylobacterota had little influence on PFASs. However, Synergistota, and Halanaerobiaeota, which had a low abundance, both positively correlated with the various PFASs. This result may imply that these rare microphyla are the main microphyla driving the transformation of PFASs in leachate. Microorganisms in leachate indirectly affected the occurrence of PFASs, mainly by influencing the environmental factors in leachate. Therefore, abiotic factors are important factors affecting PFASs in the landfill leachate. In summary, PFASs pollution management in landfills should be enhanced by regulating abiotic factors to control PFASs in leachate.