235 Advancing the Understanding and Management of Groundwater Pollution with Arsenic and Other Geogenic Contaminants Using Geospatial Tools, Machine Learning, and Data Science, Part I

The growing trend of data aggregation and analysis, along with emerging technologies like machine learning, has significantly improved our understanding of groundwater basins, enabling better assessments of water quality for direct consumption and treatment in groundwater supply management.

Wednesday, 22 October 2025
8:00 AM - 12:00 PM
HBGCC, 210AB

An Integrated Data-Driven Strategy for Arsenic Mitigation in Bangladesh: Insights from 6.3 Million Tubewells and Hydrogeological Analytics

Paper number: 235-12
Program: Technical Sessions
Time: 10:55 AM – 11:10 AM

Authors

Abstract

Arsenic (As) contamination in groundwater presents a major public health crisis in Bangladesh, necessitating a robust, scientifically based approach to ensure safe drinking water safety. . This study introduces an Integrated Data-Driven Strategy (IDDS) designed to support stakeholders in identifying optimal tubewell locations that comply with World Health Organization (WHO) and national safety standards. The IDDS is underpinned by a rich, multi-source dataset, including the most comprehensive tubewell survey to date from the Arsenic Risk Reduction Project (ARRP), as well as hydrogeological, geological, and bore log data from collaborative efforts with agencies such as JICA-DPHE and ASMITAS.. A comprehensive requirements analysis, conducted in collaboration with key stakeholders including UNICEF, DPHE, and leading academic institutions, guided the design of a modular and scalable system architecture where  the system integrates advanced data processing, machine learning algorithms, and interactive visualization tools to support data-driven decision-making. Python was employed for efficient data extraction and preprocessing, while MySQL ensured secure and structured data storage. The user interface, developed in Power BI, offers an intuitive and interactive platform that allows stakeholders to explore As contamination patterns, assess risk zones, and evaluate targeted mitigation strategies in real time. Analysis of the ARRP dataset comprising over 6.3 million tubewells across 54 districts revealed a strong correlation between As contamination and well depth. Shallow wells (<300 ft) were the most affected, with over 1 million classified as unsafe. In contrast, intermediate (300–500 ft) and deep wells (>500 ft) exhibited significantly lower As levels, underscoring well depth as a key determinant of groundwater safety and targeted drilling as an effective mitigation strategy.

 A custom classification algorithm based on Euclidean distance was developed to address the challenges of a highly imbalanced dataset This algorithm, when combined with hydrogeological trend analysis derived from real-time telemetry sensor data, enhances the system’s ability to support nuanced, spatially-informed decision-making. The platform also incorporates detailed geological visualizations based on 340 bore logs, categorized into 13 distinct soil textures, further improving the correlation between subsurface characteristics and arsenic occurrence. In conclusion, the IDDS offers a major advancement in As contamination management by integrating large-scale datasets with machine learning, hydrogeological monitoring, and geological analysis. that  enables evidence-based decision-making to support sustainable water managementfor for safer drinking water supplies for communities across Bangladesh.

Geological Society of America Abstracts with Programs. Vol. 57, No. 6

doi: 10.1130/abs/2025AM-413493

© Copyright 2025 The Geological Society of America (GSA), all rights reserved. Permission is hereby granted to the author(s) of this abstract to reproduce and distribute it freely, for noncommercial purposes. Permission is hereby granted to any individual scientist to download a single copy of this electronic file and reproduce up to 20 paper copies for noncommercial purposes advancing science and education, including classroom use, providing all reproductions include the complete content shown here, including the author information. All other forms of reproduction and/or transmittal are prohibited without written permission from GSA Copyright Permissions.