A novel framework for groundwater quality evaluation in industrial zone using unsupervised machine learning methods
Environmental Geochemistry and Health · 2026
DOI: 10.1007/s10653-026-03052-8 ↗Kurz gefasst
Fachartikel in der Zeitschrift Environmental Geochemistry and Health, 2026. Verfasst von Mukesh Panneerselvam und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Civil Engineering, M.Kumarasamy College of Engineering, Karur, 639113, India. mukeshssoft@gmail.com. Hauptthemen laut MEDLINE: Groundwater, Environmental Monitoring, Water Quality, Unsupervised Machine Learning. Die Arbeit nennt 55 Literaturangaben.
Abstract (Originalsprache)
Population density, urbanization and industrialization are the rapidly increasing and it's causes severe damages to freshwater ecosystem. The present study aims to assess the quality of groundwater using integrated methods in the industrial zone of South India. The integrated assessment of groundwater using unsupervised machine learning method associated human health risk assessment are identified as research gap in the study region. We collected 55 samples based on groundwater availability, population density, and industrial activity in summer and winter seasons. The study found that calcium-chloride and mixed calcium-magnesium-chloride types of water are the dominating category in both seasons. The groundwater pollution index (GPI) and piper trilinear and gibbs diagram indicate that evaporation, water-rock interface, and other anthropogenic activities are the dominating factors in groundwater chemistry. Analysis using the entropy water quality index (EWQI) and human health risk estimation confirmed that nitrate is the key parameter affecting groundwater sustainability. Unsupervised machine learning techniques were applied to evaluate groundwater chemistry. The principal component analysis (PCA) results revealed that seasonal variation is influenced by mineral dissolution, rainwater recharge, and anthropogenic activities. The hierarchical cluster analysis (HCA) results shows that a tight cluster forms among pH, Na, K, NO3, and F, suggesting that processes such as agricultural inputs from fertilizer use, synthetic pesticides, and natural geochemical interactions control ion exchange. The k-means clustering yielded three clusters: low-salinity fresh water, moderately mineralized water undergoing geochemical alteration, and high-hardness water indicative of specific geogenic and anthropogenic influences in both seasons. Overall, the integrated assessment of groundwater revealed that the water-rock interface, evaporation, and synthetic fertilizer use in agricultural fields are significant factors controlling groundwater quality. The key findings of this study provides clear knowledge about the nature of groundwater, influencing factors and help to improve the water resources management strategies in investigation zone.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Department of Civil Engineering, M.Kumarasamy College of Engineering, Karur, 639113, India. mukeshssoft@gmail.com.
- Department of Civil Engineering, University College of Engineering- BIT Campus, Anna University, Tiruchirappalli, 620024, India.
- Department of Pharmaceutical Technology, University College of Engineering- BIT Campus, Anna University, Tiruchirappalli, 620024, India. slaxmanvel@aubit.edu.in.
Zitieren
APA
Panneerselvam, M., Govindan, V., & Sakthivel, L. P. (2026). A novel framework for groundwater quality evaluation in industrial zone using unsupervised machine learning methods. Environmental Geochemistry and Health. https://doi.org/10.1007/s10653-026-03052-8
Vancouver
Panneerselvam M, Govindan V, Sakthivel LP. A novel framework for groundwater quality evaluation in industrial zone using unsupervised machine learning methods. Environmental Geochemistry and Health. 2026. doi:10.1007/s10653-026-03052-8
BibTeX
@article{Panneerselvam2026,
author = {Panneerselvam, Mukesh and Govindan, Venkatesan and Sakthivel, Lakshmana Prabu},
title = {A novel framework for groundwater quality evaluation in industrial zone using unsupervised machine learning methods},
journal = {Environmental Geochemistry and Health},
year = {2026},
doi = {10.1007/s10653-026-03052-8}
}
EUVIMED