AUTHOR=Yang Li , Ziwen Zhang , Lin Xinhao , Hei Junmiao , Wang Yixiao , Zhang Ang 

TITLE=AI-powered approaches for enhancing remote sensing-based water contamination detection in ecological systems

JOURNAL=Frontiers in Environmental Science

VOLUME=Volume 13 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/environmental-science/articles/10.3389/fenvs.2025.1612658

DOI=10.3389/fenvs.2025.1612658

ISSN=2296-665X

ABSTRACT=IntroductionWater contamination poses a significant threat to both public health and ecosystems worldwide, leading to increased emphasis on developing robust detection and mitigation strategies. Traditional methods for monitoring water quality, such as manual sampling and basic chemical analysis, are limited in their ability to provide real-time data and often fail to detect contaminants in a timely manner. Recent advancements in artificial intelligence (AI) offer promising solutions to enhance water contamination detection, particularly by leveraging machine learning algorithms and sensor networks for continuous monitoring.MethodsThis paper presents a novel AI-powered approach for improving water contamination detection, which incorporates real-time data processing and predictive modeling to identify contamination events and optimize response strategies. We combine sensor data with advanced machine learning techniques to accurately predict contaminant concentrations and assess the effectiveness of various mitigation strategies in different water bodies.ResultsExperimental results across four benchmark datasets show that our model, AquaDynNet, achieves outstanding performance. Specifically, it achieves an accuracy of 90.75%, F1-score of 88.79, and AUC of 92.02 on the Terra Satellite dataset. On the Aquatic Toxicity dataset, the model obtains an accuracy of 92.58% and AUC of 94.13, and on the Water Quality dataset, it reaches an F1-score of 85.54 and AUC of 89.72. On the infrastructure-focused WaterNet dataset, it achieves 91.98% accuracy and AUC of 92.47.DiscussionThese results consistently demonstrate our model’s superior detection accuracy and robustness compared to baseline approaches. Furthermore, our approach is capable of providing actionable insights for policymakers and environmental agencies to mitigate the impacts of contamination on human health and aquatic ecosystems. This research addresses critical challenges in water quality management, offering a scalable and adaptable solution for addressing global water contamination issues.