Machine Learning-Based Prediction Of Groundwater Fluoride Contamination And Dental Fluorosis Risk In Rural Areas
DOI:
https://doi.org/10.29284/wympek12Keywords:
Fluoride Contamination, Dental Fluorosis, Machine Learning, Water Quality Prediction, Rural Drinking Water, SDG-3 & 6.Abstract
Groundwater fluoride contamination remains a major public health concern in rural regions where communities depend on untreated aquifers for drinking water, hindering progress toward Sustainable Development Goals 3 and 6. This study examines groundwater fluoride dynamics and their association with dental fluorosis across ten rural blocks of Unnao district, Uttar Pradesh, India. A five-year dataset (2021–2025) containing physicochemical water-quality parameters and dental health indicators was analysed, and future conditions were forecast for 2026–2030. Supervised machine-learning models, including regression and multi-class classification with ensemble techniques, were applied to estimate fluoride concentration (mg/L), predict fluorosis prevalence (%), and classify disease severity. Data preprocessing, temporal and stratified data splitting, and feature-selection procedures ensured robust model development. The HistGradientBoosting regressor achieved the best regression performance (RMSE = 0.089; R² = 0.985), while the XGBoost classifier produced the highest classification accuracy (0.986) and F1-score (0.985). Forecasts indicate increasing fluoride levels and fluorosis risk, highlighting the need for targeted mitigation and improved rural water-management strategies.
References
(1) Arora, V., Singh, R., & Tiwari, S. (2025). Hybrid ANN–XGBoost framework for fluoride risk prediction in groundwater of Uttar Pradesh, India. Groundwater for Sustainable Development, 28, 100634.doi: 10.1016/j.gsd.2025.100634
(2) Banerjee, P., Karmakar, S., & Mukherjee, A. (2025). Hybrid time-series–machine-learning forecasting of groundwater contaminants in eastern India. Journal of Hydrology, 620, 129572.doi: 10.1016/j.jhydrol.2023.129572
(3) Dubey, M., Yadav, A., & Chauhan, R. (2025). Multi-state fluoride prediction using CatBoost and feature-importance analysis. Environmental Science and Pollution Research, 32(12), 18755–18770.doi: 10.1007/s11356-024-33121-1
(4) Verma, R., Singh, P., & Chauhan, A. (2025). Spatial analysis of dental fluorosis and groundwater quality in Uttar Pradesh. Environmental Monitoring and Assessment, 197, 411.doi: 10.1007/s10661-025-13354-y
(5) Ghosh, P., Prasad, R., & Suresh, S. (2024). Machine-learning-based modelling of groundwater fluoride in central India using XGBoost. Environmental Advances, 16, 100371.doi: 10.1016/j.envadv.2024.100371
(6) Jayalakshmi, P., Ramesh, T., & Kumar, V. (2024). SVM modelling of fluoride contamination in Karnataka. Water Supply, 24(1), 112–123.doi: 10.2166/ws.2023.310
(7) Mwangi, P., Oiro, S., & Cheruiyot, S. (2024). ML prediction of dental fluorosis in Kenyan Rift Valley communities. Science of the Total Environment, 903, 166855.
doi: 10.1016/j.scitotenv.2023.166855
(8) Sinha, R., Tripathi, S., & Verma, A. (2024). Dental fluorosis mapping in rural Chhattisgarh. Environmental Geochemistry and Health, 46(1), 95–109.doi: 10.1007/s10653-023-01637-4
(9) Bhatia, A., Reddy, D. V., & Srinivas, Y. (2023). XGBoost-based fluorosis classification in Telangana. Environmental Monitoring and Assessment, 195(3), 299.doi: 10.1007/s10661-023-11049-8
(10) Kumar, N., Bhattacharya, P., & Shankar, H. (2023). Histogram-based gradient boosting for groundwater prediction. Journal of Water and Health, 21(2), 311–325.doi: 10.2166/wh.2023.128
(11) Prakash, K., Yadav, R., & Singh, D. (2023). Fluoride contamination in rural Bihar. Archives of Environmental Protection, 49(2), 72–85.doi: 10.24425/aep.2023.147337
(12) Rathore, A., Choudhary, M., & Goyal, R. (2023). Dental fluorosis prevalence in rural Rajasthan. Journal of Clinical Paediatric Dentistry, 47(2), 155–162.doi: 10.22514/jocpd.2023.046
(13) Yadav, M., Dudi, R., & Singh, S. (2023). Random-forest modelling of fluoride in arid Rajasthan. Environmental Earth Sciences, 82(4), 109.doi: 10.1007/s12665-022-10804-4
(14) Gupta, V., Sharma, A., & Mishra, P. (2022). Dental fluorosis among schoolchildren in UP. Indian Journal of Public Health Research & Development, 13(2), 88–95.
(15) Patel, D., Verma, S., & Tripathi, R. (2022). ML classification of dental fluorosis using RF & XGBoost. Environmental Health Insights, 16, 1–10.doi: 10.1177/11786302221108856
(16) Singh, P., Gautam, S., & Singh, R. (2022). Hybrid ML–time-series model for groundwater. Water Resources Management, 36, 441–457.doi: 10.1007/s11269-021-03013-7
(17) Wang, J., Li, Q., & Sun, R. (2022). Deep-learning fluorosis classification. Chemosphere, 301, 134654.doi: 10.1016/j.chemosphere.2022.134654
(18) Zhang, Y., Zhou, L., & Lin, H. (2022). Hybrid ML–geochemical model for fluoride. Journal of Hydrology, 608, 127646.doi: 10.1016/j.jhydrol.2022.127646
(19) UP Jal Nigam (2022). State-Level Water Quality Monitoring Report. Lucknow.
(20) Adimalla, N., Qian, H., & Li, P. (2021). ML prediction of groundwater fluoride. Environmental Geochemistry and Health, 43(8), 3097–3114.doi: 10.1007/s10653-021-00865-w
(21) Kumar, R., Singh, S., & Prakash, A. (2021). Hydrogeochemical modelling & ML prediction. Environmental Monitoring and Assessment, 193, 788.doi: 10.1007/s10661-021-09590-x
(22) Mandal, S., Roy, S., & Mukherjee, S. (2021). Dental fluorosis burden in West Bengal. Clinical Epidemiology and Global Health, 11, 100730.doi: 10.1016/j.cegh.2021.100730
(23) Martínez, J., Díaz, L., & Ortega, A. (2021). Predicting fluoride in Mexican aquifers. Hydrological Processes, 35(4), e14088.doi: 10.1002/hyp.14088
(24) Rai, A., Singh, R., & Das, S. (2021). ARIMA-based forecasting of groundwater contaminants. Journal of Environmental Management, 289, 112523.doi: 10.1016/j.jenvman.2021.112523
(25) Sinha, A., Roy, D., & Kumar, V. (2021). Multivariate modelling of fluoride in Bihar. Environmental Challenges, 4, 100087.doi: 10.1016/j.envc.2021.100087
(26) Sharma, R., Gupta, P., & Jaiswal, S. (2021). Random forest groundwater fluoride model. Groundwater for Sustainable Development, 15, 100650.doi: 10.1016/j.gsd.2021.100650
(27) UNICEF/WHO JMP (2021). Progress on Household Drinking Water, Sanitation and Hygiene.
(28) CGWB (2021). Groundwater Resource Assessment Report.
(29) Reddy, D. V., et al. (2020). Hydrogeochemical controls on fluoride contamination. Environmental Earth Sciences, 79(3), 1–14.doi: 10.1007/s12665-019-8734-y
(30) Hassani, H., Silva, E., & Gupta, R. (2020). Tree-based ensemble methods for environmental prediction. Environmental Modelling & Software, 132, 104792.doi: 10.1016/j.envsoft.2020.104792
(31) Jha, S., Singh, P., & Gautam, S. (2020). Fluoride in Indo-Gangetic aquifers. Environmental Earth Sciences, 79, 356.doi: 10.1007/s12665-020-09072-1
(32) NPPCF (2020). District Fluorosis Surveillance Report.
(33) CGWB (2020). Report on Groundwater Quality in India.
(34) Government of India (2019). Jal Jeevan Mission Baseline Household Survey.
(35) Reddy, D.V., Nagabhushanam, P., & Rao, T. (2019). Fluoride contamination pathways. Environmental Earth Sciences, 78, 119.doi: 10.1007/s12665-019-8140-5
(36) Susheela, A.K., & Kumar, A. (2019). Health impacts of endemic fluorosis. Indian Journal of Medical Research, 150(3), 303–315.doi: 10.4103/ijmr.IJMR_1035_18
(37) MoHFW (2019). National Programme for Prevention and Control of Fluorosis: Annual Report.
(38) Saxena, V., Ahmed, S., & Srivastava, S. (2018). Hydrogeochemistry of fluoride. Environmental Monitoring and Assessment, 190(5), 281.doi: 10.1007/s10661-018-6644-3
(39) WHO (2017). Guidelines for Drinking-Water Quality (4th ed.).
(40) Kundu, A., Mandal, P., & Sharma, R. (2017). Dental fluorosis epidemiology in India. International Journal of Environmental Health Research, 27(3), 220–229.doi: 10.1080/09603123.2017.1319315
(41) Liu, C., Wang, Y., & Sun, Z. (2017). Hydro-geochemical controls in northern China. Science of the Total Environment, 593–594, 189–199.doi: 10.1016/j.scitotenv.2017.03.135
(42) Rango, T., et al. (2016). Geochemical controls in Rift Valley aquifers. Hydrogeology Journal, 24, 1681–1698.doi: 10.1007/s10040-016-1424-6
(43) Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. ACM SIGKDD, 785–794.doi: 10.1145/2939672.2939785
(44) Rango, T. et al. (2015). Fluoride exposure in East African Rift Valley. Science of the Total Environment, 532, 144–158.doi: 10.1016/j.scitotenv.2015.05.004
(45) Sharma, S., Yadav, M., & Pandey, B. (2015). Controls of fluoride in arid aquifers. Journal of Hydrology, 523, 515–527.doi: 10.1016/j.jhydrol.2015.02.005
(46) Dean, H.T., Elvove, E., & McKay, F. (2015). Classic fluorosis studies. International Journal of Paediatric Dentistry, 25(4), 241–251.doi: 10.1111/ipd.12179
(47) Gupta, S. et al. (2014). Fluorosis impact on children in India. Journal of Clinical and Diagnostic Research, 8(12), 01–05.
(48) BIS (2012). Indian Standard Drinking Water Specification IS 10500:2012. Bureau of Indian Standards.
(49) Rango, T., Kravchenko, J., & Atlaw, B. (2012). Fluoride mobility in East African Rift. Science of the Total Environment, 435–436, 130–141.doi: 10.1016/j.scitotenv.2012.07.019 (50) Chae, G.T., Yun, S.T., Kwon, M.J., Mayer, B., & Kim, K.H. (2007). Fluorine geochemistry in South Korea. Applied Geochemistry, 22(4), 683–708.doi: 10.1016/j.apgeochem.2006.11.013
(50) Dean, H.T. (1942). The Investigation of Physiological Effects by the Epidemiological Method. In Fluorine and Dental Health.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
This work is licensed under a Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
