Answers are grounded in this paper and cite their sources.
Reliable assessment of groundwater quality (GWQ) is essential for sustainable water resources management in rapidly urbanizing regions. Long-term GWQ data (2001-2020) from Lucknow, Prayagraj and Varanasi district were analyzed for fifteen hydro-chemical parameters (pH, EC, TDS, TH, Mg, Na, Ca, K, Cl, SO 4 , CO 3 , HCO 3 , F, NO 3 , and B). To overcome the limitations of single-method weighting, four multi-criteria decision-making (MCDM) approaches Entropy, Analytic Hierarchy Process (AHP), Fuzzy-AHP and a newly developed Self Organizing Map-AHP (SOM-AHP) were employed for effective classification GWQ. The MCDM framework is widely used for long-term GWQ assessment. However, assigning appropriate weights to different parameters remains a significant challenge. SOM-AHP method addresses this limitation by employing unsupervised competitive learning to reduce the subjectivity and human bias associated with weight assignment. In spatio-temporal analysis a progressive deterioration of water quality observed over time; however, most parameters still remain within the permissible limits prescribed by Bureau of Indian Standards (BIS). MCDM provide a systematic approach to evaluate the best option on the basis of several criteria. It integrates multiple hydro chemical parameters into a single objective index using scientifically derived weights, enabling reliable classification of groundwater (GW). The Entropy method under classifies water quality due to statistically biased weight distribution, whereas AHP and F-AHP frequently overestimate. It is due to subjectivity in expert-based weighting. In contrast, the SOM-AHP hybrid approach produces a more balanced and realistic classification by capturing intrinsic, nonlinear relationships among hydro-chemical parameters and effectively representing the temporal deterioration of GW quality. SOM-AHP produces the most uniform weighting with lowest consistency ratio (CR) value 0.025 in comparison with AHP (0.064) and Fuzzy-AHP (0.068). SOM-AHP derive weights to reflect actual data structure rather than a predefined statistical or subjective rules by capturing the intrinsic nonlinear and spatial correlations among hydrochemical parameters through unsupervised learning. The proposed SOM-AHP demonstrates superior robustness and reliability for long-term GWQ monitoring. It provides a scientifically sound and realistic classification of water quality, supporting effective regional water management and policy planning. This approach is useful in the early detection of emerging water quality deterioration.
No takes yet. Share an insight, caveat, or question.
Kumar et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: