Unravelling integrated groundwater management in pollution-prone agricultural cities: a synergistic approach combining probabilistic risk, source apportionment and artificial intelligence

Abstract

Groundwater is vital for agricultural cities, but intensive farming and fertilizer use have increased contamination risks, particularly for non-carcinogenic health hazards. This study reveals the sources of contaminants in groundwater, their health impacts, and targeted strategies in such cities. The study analyzed 115 groundwater samples, with the main groundwater chemical type being HCO₃-Na·Ca. Significant exceedances were found in Mg²⁺, HCO₃⁻, F⁻, total hardness (TH), and Mn, with HCO₃⁻ and Mg²⁺ surpassing standards in nearly all samples. The average Comprehensive Environmental Water Quality Index (CEWQI) was 100.68, indicating that overall groundwater quality in the study area is good. High-quality water is mainly found near reservoirs and rivers, while urban and eastern regions have relatively poorer water quality. The proportion of groundwater unsuitable for drinking is low. Monte Carlo risk assessments revealed that F⁻ and NO₃⁻ pose non-carcinogenic risks to both adults and children, with NO₃⁻ presenting a higher potential health risk. The Positive Matrix Factorization (PMF) model identified that groundwater pollution primarily results from natural geological processes and human activities, with agriculture being the major anthropogenic factor. AI-based zoning strategies highlighted industrial areas and high-fluoride zones as critical areas requiring enhanced prevention and control measures.