T62 Groundwater Pollution with Arsenic and Other Geogenic Legacy Contaminants, Related Health Risks and Management II: Advancing Learning through Geospatial Tools, Machine Learning, Data Science, and Artificial Intelligence

Sunday, 22 September 2024
1:30 PM - 5:30 PM
211A

DEVELOPING A CLIMATE-RESILIENT DECISION SUPPORT SYSTEM FOR SUSTAINABLE WATER SOLUTIONS FOR DRINKING WATER SAFETY IN BANGLADESH

Paper number: T62-11
Program: Technical Sessions
Time: 3:45 PM – 4:00 PM

Authors

Abstract

In recent decades, reliance on groundwater has increased significantly for water supplies for drinking, agriculture, and industrial use. The complexity of groundwater systems, require in-depth knowledge of subsurface conditions to address the challenges for sustainable water management. The present study addresses the critical public health issue of arsenic (As) contamination in Bangladesh’s groundwater by developing a Decision Support System (DSS).The DSS integrates extensive datasets, including arsenic screening, geological, and hydrogeological data, alongside inputs from local stakeholders. The system employs advanced data processing, machine learning algorithms, and interactive visualizations to provide a robust tool for stakeholders to identify suitable tubewell locations that comply with WHO (As < 10 μg/L) and Bangladesh (As ≤ 50 μg/L) standards. Additionally, real-time hydrogeological monitoring and detailed geological visualizations further support the system’s analytical depth. We collected data from the ARRP project, covering 6.37million tubewells in 54 districts in Bangladesh. Rigorous preprocessing was applied to manage outliers and missing values, ensuring data accuracy. We developed custom statistical models using Euclidean distances to predict arsenic levels and classify tubewells. The geological analysis included borelog data from 279 wells by JICA-DPHE and 61 from ASMITAS digital groundwater platform. After preprocessing, the datasets were integrated into Power BI with DENEB for customized visualizations, showcasing 13 soil textures like clay, sand, silt, and gravel. The hydrogeological analysis used data from 19telemetric sensors across six districts to monitor water level, conductivity, temperature, and battery voltage. Data collection was automated using a Python script interacting with a third-party API. Outliers were managed through polynomial curve fitting and rolling mean calculations, ensuring data consistency. The DSS supports informed decision-making through dynamic visualization and comprehensive insights into groundwater conditions. Despite data imbalance challenges, custom algorithms and trend analysis enhanced predictive capabilities, making the DSS a scalable solution to mitigate public health impacts and promote sustainable water management for healthier future for Bangladesh.

Geological Society of America Abstracts with Programs. Vol. 56, No. 5

doi: 10.1130/abs/2024AM-402876

© Copyright 2024 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.