This study focused on the Chao River and Baimaguan River located upstream of the Miyun Reservoir in Miyun District (Beijing, China). Soil and sediment samples were collected from the river and drainage basin. Total nitrogen, total phosphorus, and six potentially toxic elements including cadmium, zinc, lead, chromium, arsenic, and copper, were analyzed in terms of concentration, potential ecological risk, and human health risk. The average concentrations of the six potentially toxic elements were all below the soil environmental quality standards for China. Cadmium was the most serious pollutant in both soils and sediments, at 2.58 and 3.40 times its background values. The contents of Cd and Pb were very closely related (p < 0.01) to total nitrogen concentrations in both soil and sediment samples. The potential ecological risks posed by Cd in the Chao and Baimaguan River soils were considerable and moderate, respectively. The historical iron ore mining and agricultural activity were identified as the primary sources of potentially toxic element pollution of soil and sediment in the Chao-Bai River in Miyun District. Human health risk assessment indicated that non-carcinogenic risks all fell below threshold values. The total carcinogenic risks due to Cr and As were within the acceptable range for both adults and children. This conclusion provides a scientific basis for the control of potentially toxic element pollution and environmental protection of the Miyun Reservoir in Beijing.

Soil contamination by potentially toxic elements (PTEs) poses substantial risks to the environment and human health. Traditional investigational methods are often inadequate for large-scale assessments because they are time-consuming, costly, and have a limited accuracy. Machine learning (ML) techniques have emerged as promising tools in environmental studies because of their superiority in processing high-dimensional and unstructured data. However, critical evaluations of contemporary ML applications and methods in PTEs content, distribution, and identification remain scarce. To address this research gap, this study reviews applications of ML to soil PTEs contamination including content prediction, spatial distribution, source identification, and other related tasks. Hyperspectral data combined with ML methods can predict the content of PTEs in large-scale areas at a low cost. In addition, ML algorithms that integrate environmental covariates offer superior performance in spatial predictions compared with traditional geostatistical methods. Moreover, ML techniques incorporated with receptor models provide important advances in the quantitative identification and apportioning of PTE sources, thereby supporting effective environmental management and risk assessment. Based on the frequency of the variables used, we propose that soil pH, soil organic matter (SOM), industrial activities, soil texture, and other relevant factors are key environmental variables that enhance the accuracy of predictions regarding the spatial distribution and source identification of PTEs. From these findings, ML techniques, through their powerful data processing capabilities, provide new perspectives and tools for the efficient assessment and management of soil PTEs contamination.