Geospatial AI approaches for domestic and agricultural water demand management: a systematic literature review

Le résumé fourni par la source

Water is a vital resource for sustaining life, advancing human progress, and protecting ecosystems. However, climate change, persistent over-extraction, and land use changes have led to the depletion of water resources, land subsidence, and exhaustion of groundwater reserves. To address these challenges, effective water demand management (WDM) is essential for the efficient use of precious water supplies. In recent years, there has been growing interest in leveraging Geographic Information Systems (GIS), Remote Sensing (RS) data, and Artificial Intelligence (AI) for advanced WDM. Despite this growing interest, few studies have critically reviewed the opportunities for integrating fused GIS/RS datasets with AI techniques in WDM. This systematic review of 119 studies focuses on water-use patterns in two key WDM areas: residential consumption and agricultural irrigation. It highlights the role of advanced technologies in supporting conservation efforts. This review evaluates recent AI-based geospatial approaches, including computer vision techniques, combining GIS, remote sensing, and machine learning for managing water demand in both residential and agricultural sectors. These areas are emphasized due to their high consumption, data availability, and potential for impactful demand-side interventions. The review aims to synthesize current knowledge, identify research gaps, and guide the development of scalable, intelligent WDM strategies using AI-enabled geospatial tools.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.