Chlorinated aliphatic hydrocarbons (CAHs) are widely used in agriculture and industries and have become one of the most common groundwater contaminations. With the excellent performance of the deep learning method in predicting, LSTM and XGBoost were used to forecast dichloroethene (DCE) concentrations in a pesticide-contaminated site undergoing natural attenuation. The input variables included BTEX, vinyl chloride (VC), and five water quality indicators. In this study, the predictive performances of long short-term memory (LSTM) and extreme gradient boosting (XGBoost) were compared, and the influences of variables on models’ performances were evaluated. The results indicated XGBoost was more likely to capture DCE variation and was robust in high values, while the LSTM model presented better accuracy for all wells. The well with higher DCE concentrations would lower the model’s accuracy, and its influence was more evident in XGBoost than LSTM. The explanation of the SHapley Additive exPlanations (SHAP) value of each variable indicated high consistency with the rules of biodegradation in the real environment. LSTM and XGBoost could predict DCE concentrations through only using water quality variables, and LSTM performed better than XGBoost.

As the exploitation and utilization of metal mines continue, issues concerning the groundwater ecological environment have become increasingly prominent, attracting significant attention. However, few studies have predicted the ecological risks associated with the attenuation of heavy metals in mining areas. This research focuses on a typical lead-zinc mine in Jiangxi Province, China. We employed a Self-Organizing Map (SOM) to analyze groundwater pollution clustering and integrated the Domenico model with Monte Carlo simulation to predict heavy metal attenuation. To assess present and long-term ecological risks, we applied the potential ecological risk index. The research findings indicate that: Mn, Fe, and As exhibited significant enrichment during periods, while Cu, Cd, As, and Al showed the highest coefficients of variation. According to the SOM-based clustering analysis, Zn and Pb show a strong correlation with Cd, while Cu, As, and Al are strongly correlated with Fe. The assessment of potential ecological risks reveals that single-factor evaluations for both periods mostly indicate slight ecological risks (E_rⁱ<30). Cd, As, Cu, and Mn pose very high ecological risks. The comprehensive ecological risk is very high around the tailings pond and the waste dump, with 11.11 % of the sites showing very high ecological risk. Combining the SOM output with the potential ecological risk analysis, cluster 1 includes sites identified as very high ecological risk points, with As being the key ecological risk factor. The ecological risk prediction based on Monte Carlo simulation indicates that the ecological risk of As will decrease by 75.6 % to 13,565, yet it persists at very high ecological risk. This study will offer critical insights for controlling heavy metal pollution in groundwater and for guiding ecological restoration in mining areas and their surroundings.

With the reform of China’s urbanization increasing in popularity, the security issues posed by urban groundwater, especially groundwater in industrial areas, have attracted scholars’ attention. This research aimed to predict and quantify the migration process of contaminants in a microconfined aquifer by conducting a groundwater contamination investigation in an abandoned chemical plant in the Jiangsu Province of China. First, data such as regional hydrogeological parameters and types of contaminants were obtained via hydrogeological drilling, groundwater well monitoring, pumping tests, and laboratory permeability tests, which helped identify the most serious pollution factor: chloroform. Then, a groundwater flow model was built using the Groundwater Modeling System (GMS) and verified using the general-purpose parameter estimation (PEST) package. In addition, based on the three-dimensional multi-species model for transport (MT3DMS) in GMS, a transport model was established. The results illustrate that the plume range of chloroform diffuses with water flow, but, because of its slow diffusion rate and inability to degrade naturally, the concentration of the contaminant has remained several times higher than the safety standard for a long time. The contaminant spread vertically to the soil layer above the microconfined aquifer under pressure, resulting in direct pollution. In addition, the contaminant in the microconfined aquifer is anticipated to migrate down to the clay layer and become enriched. However, the first confined aquifer has not been seriously polluted in the past 20 years. Finally, a sensitivity analysis of the parameters shows that groundwater contamination in the Yangtze delta region is greatly affected by precipitation recharge and hydraulic conductivity.

BTEX (benzene, toluene, ethylbenzene, and xylene), as characteristic pollutants in chemical plant sites, are widely present in the environment and pose a serious threat to the health and safety of nearby residents. Studying the spatial distribution characteristics and transport fluxes of BTEX in soil and air at contaminated sites and the health risks they pose to humans is of great significance for fine pollution control and environmental management. This study took a typical decommissioned steel plant as a case study. A total of 23 soil and air samples were collected from different locations to investigate the spatial distribution characteristics of BTEX in soil and air. The transport and fate of BTEX in soil and air were evaluated using the fugacity model, and finally, a human health risk assessment was conducted. The results indicate a relatively severe level of benzene pollution in both soil and air. The maximum exceedance factor of benzene in soil samples is 31.5, with the concentration exceedance depth at 1.5 m. The maximum concentration of benzene in air samples is 4.98 μg·m⁻³. Benzene, at 5.9% of the site, shows a low flux with negative values, while other components at various locations all exhibit a trend of transport from the soil phase to the atmospheric phase. Benzene is the pollutant that contributes the most to the transport flux from soil to air within the site. The coking area and sewage treatment area are key areas within the steel mill where BTEX accumulate easily in the soil. The non-carcinogenic risk values of the individual components of BTEX in the soil are below the acceptable risk level. However, the carcinogenic risk value of benzene in the children’s exposure scenario exceeds the carcinogenic risk level of 10⁻⁶. The carcinogenic risk range of various components of BTEX in the air is 2.63 × 10⁻⁶~3.88 × 10⁻⁵, with 28.6% of the locations exceeding the threshold of 10⁻⁶. The range of the total HI (hazard index) is 2.08 × 10⁻⁴~1.81 × 10⁻¹, all of which is below the safety threshold of 1. The results of this study will provide scientific support for the fine pollution control and environmental management of industrial contaminated sites with BTEX as their typical pollutants.

Soil heavy metals (HMs) pollution from mining activities poses a significant threat to environmental health and has attracted widespread attention. This study integrated eco-health risk assessment and identified driving factors for risks of HMs in soils at an open-pit lead-zinc mining area. Results indicated that Pb, As and Cu were the major contaminants, with 46.43%, 32.14% and 25.00% of soil samples exhibited moderate to severe pollution levels. Distinct from the pollution level, Cd total contents and effective available contents showed the highest ecological risk, with an average Er value of 180.06 (heavily polluted) and mean RAC value of 82.01(extreme pollution), respectively. Through the probabilistic health risk assessment, As was identified as the key pollutant posing health risks, contributing 98.6% and 97.8% of the Carcinogenic risks and 68.5% and 76.7% of the Non-carcinogenic risks for adults and children, respectively. Cd and As were identified as the primary pollutants posing potential ecological and human health risks. Redundancy and permutation feature importance analysis further demonstrated that electrical conductivity (EC) was a key predictor strongly associated with the ecological risk and health risk. This study distinguished the target priority pollutant between the ecological and health risks and identified the driving factors for risks, providing valuable insights into the pollution control and targeted soil management in mining areas.