Prediction Model of Soil Heavy Metal Content Based on Particle Swarm Algorithm Optimized Neural Network

Modelling of soil environmental quality and early warning of integrated ecological risk

The knowledge of dynamic trend in soil heavy metal contamination and associated risk is important for soil pollution prevention, safe utilization and early warning of soil environmental quality and ecological risk. In this study, a modified integrated risk index (NIRI) was adopted to evaluate ecological risk in agricultural soil in Wenzhou with 70 samples, which is located in the southeast of China. In addition, two scenarios with different metal fluxes (optimistic and default scenario) were constructed to predict future dynamic trend of metal concentrations. Results showed the agricultural soil was mainly contaminated by Cd and Pb. The NIRI indicated moderate to considerable risk in most sites and Cd posed the greatest contribution to NIRI value. Besides, higher risk was determined in paddy soil than that in vegetable. Scenario simulation results revealed general declining trend in optimistic scenario while increasing trend in default scenario for metal concentration. However, exceedance varied with prediction period, soil types and metals. Ecological risk probability showed similar trend with metal concentration, indicating significant shift to higher risk level in default scenario while insignificant decrease in optimistic scenario. The proposed scenario simulation results provide reference to support soil quality improvement and risk management.

Source-specific probabilistic contamination risk and health risk assessment of soil heavy metals in a typical ancient mining area

Heavy metal pollution (HMP) from mining operations severely threatens soil ecosystems and human health. Identifying the sources of soil heavy metals (HMs) and assessing source-specific risks are critical for developing effective risk mitigation strategies. In this study, a combination of methodologies including PMF, Monte Carlo analysis, soil pollution risk index, and a human health risk assessment model were utilized to investigate soil HM risks in a typical ancient mining area in Daye City, China, considering both environmental pollution and human health impacts. Cu emerged as the most significant soil pollution risk, whereas As posing the highest health risk. About 48.44 % of the multi-element integrated soil pollution risk has escalated to the heavy level. Furthermore, around 22.42 % of the non-carcinogenic risk (NCR) and 9.53 % of the carcinogenic risk (CR) exceeded unacceptable thresholds (THI > 1 for NCR and TCR > 1E-4 for CR). The PMF model identified four distinct sources: the smelting industry, traffic emissions, a combination of agricultural and natural factors, and mining activities. The mixed agricultural and natural source significantly impacted health risks, contributing 42.17 % to NCR and 53.88 % to CR, followed by the mining source, contributing 31.67 % to NCR and 24.07 % to CR. Interestingly, the mining source contributed the highest soil pollution risk at 42.45 %, while the mixed agricultural and natural source exhibited the lowest at 16.33 %. Furthermore, the study explored source-specific risk components by evaluating the contributions of different sources to specific elements. The mining source was identified as the focus for soil HMP control, followed by the mixed agricultural and natural source. Overall, this study provided an in-depth analysis of soil heavy metal risks in mining areas from the source apportionment perspective, which broadened the research framework of soil heavy metal source analysis and risk assessment, potentially providing scientific guidance for managing regional soil HMP.