Comprehensive Summary
This study investigates the ecological and human health impacts of soil contamination by potentially toxic elements (PTEs) in the Junín Lake basin, driven by agriculture, urbanization, and mining activities. Using remote sensing, machine learning algorithms, and land classification techniques, the researchers collected and analyzed 211 soil samples for heavy metals, metalloids, and trace elements. Severe contamination with arsenic (As), lead (Pb), cadmium (Cd), and zinc (Zn) was found, with concentrations surpassing environmental quality thresholds. Human health risk assessments revealed that As, Pb, and chromium (Cr) were present at levels associated with significant carcinogenic risk in both pediatric and adult populations. By leveraging AI-based machine learning and geospatial tools, the study accurately identified and mapped areas with elevated PTE concentrations, demonstrating the utility of data-driven approaches in environmental health monitoring.
Outcomes and Implications
These findings suggest that machine learning and remote sensing can be critical tools in the exact classification of contaminated areas and land cover types. They can help draw parallels between findings obtained from the environment and their impact on the ecological environment and human health. In order to preserve the central Peruvian Andes, more management interventions, regional coordination, and community-based monitoring utilizing local stakeholders will be necessary to protect public health in the basin.