Satellite-Based System for Groundwater Arsenic Contamination Detection

Overview

Arsenic is a toxic carcinogen that may be present at high levels in groundwater. Over 100 million people worldwide are estimated to be exposed to arsenic through drinking water, with studies linking long-term exposure to cancers, neurological disorders, and cardiovascular diseases. Monitoring heavy metal contamination in groundwater is therefore crucial for mitigating its impact, especially for developing countries. Conventional lab detection methods are time-consuming, expensive, and require extensive fieldwork, with remote sensing offering a potential solution. Previous studies successfully used remote sensing to predict soil concentration; there is also a strong correlation between soil arsenic concentrations and groundwater contamination. With this knowledge, this study explores the potential of using hyperspectral satellite data and machine learning models to directly predict groundwater arsenic contamination. Hyperspectral datasets of the Western United States were integrated with recorded arsenic levels of groundwater wells. Then, different combinations of machine learning techniques were tested to draw predictive relationships, each demonstrating high predictive accuracy and suggesting overall feasibility. The most accurate combination features SMOGN-Based Data Augmentation, Genetic Algorithm, Second Derivative Transformation, and Random Forest, with results indicating a strong correlation (R2 ≈ 0.91, RMSE= 0.089) between satellite signatures and groundwater contamination. Future work will refine model accuracy and assess applicability across different geographic regions. By leveraging distant remote sensing technology and intelligent analysis, this study aims to provide a scalable approach to combat a global health concern.

Competition history

• AJAS 2026 Category not listed

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Browse more like this

• AJAS None projects
• Every AJAS 2026 project