Assessing potential toxic metal threats in tea growing soils of India with soil health indices and machine learning technologies

Abstract

This study explores the impact of potentially toxic metals (PTMs) contamination in Indian tea-growing soils on ecosystems, soil quality, and human health using machine learning and statistical analysis. A total of 148 surface soil samples were collected from: Terai, Dooars, Darjeeling, and Jorhat, which are significant regions for tea cultivation in India. Results showed that the PTM concentrations in total soil, bioavailable fraction (DTPA), and tea leaves were significantly higher than the acceptable limits in Zone 1. Positive Matrix Factorization (PMF), correlation analysis and Self-Organizing Map (SOM) were applied to identify primary pollution sources and contributors. A geospatial model was developed to forecast the PTMs distribution patterns in tea-growing zones. The Free Ion Activity Model (FIAM) assessed PTM solubility in contaminated soils and transfer to tea leaves, with Hazard Quotient values (FIAM-HQ < 0.5) moderately exceeding the safety threshold for Cr, Ni, Pb, Cd, and Cu. The severity adjustment margin of exposure (SAMOE) results indicate that PTM-contaminated tea poses a moderate (concern level 4: 0.01–0.1) health risk to humans for Cr: 0.045 and Ni: 0.062 except Pb, Cu, and Cd. Monte Carlo simulation (MCS) indicated that total carcinogenic risk (TCR) is significant, with children being the most affected compared to adults, as shown by Sensitivity analysis. This underscores the importance of addressing TCR, especially for children. This study underscores the ecological and health risks posed by PTM contamination in tea-growing soils, highlighting the need for targeted mitigation strategies to safeguard both soil quality and human health.