Prediction Modeling and Mapping of Groundwater Fluoride Contamination throughout India

• DOI: 10.1021/acs.est.8b01679
• Publication Date: 2018-07-27T13:23:28Z
• Authors: Joel E. Podgorski; Pawan Labhasetwar; Dipankar Saha; Michael Berg
• Keywords: Skeletal fluorosis; Tamil; Groundwater Contamination

For about the past eight decades, high concentrations of naturally occurring fluoride have been detected in groundwater different parts India. The chronic consumption is recognized to cause dental and skeletal fluorosis. We used random forest machine-learning algorithm model a data set 12 600 from throughout India along with spatially continuous predictor variables predominantly geology, climate, soil parameters. Despite only surface parameters being available describe subsurface phenomenon, this has produced highly accurate prediction map exceeding 1.5 mg/L at 1 km resolution country. most affected areas are northwestern states/territories Delhi, Gujarat, Haryana, Punjab, Rajasthan southern states Andhra Pradesh, Karnataka, Tamil Nadu, Telangana. total number people risk fluorosis due predicted be around 120 million, or 9% population. This based on rural populations accounts for average rates nonmanaged sources. new hazard maps can by authorities conjunction detailed utilization information prioritize need mitigation measures.