Integrating open source nexus modelling (AI) and remote sensing in mapping water-related indicators of extreme poverty

A lack of access to clean and safe water and sanitation has commonly been seen as a proxy for poverty. The ongoing advancement in technological innovations presents new opportunities for mapping and identifying poverty hotspots using non-traditional indicators.

This study examined the potential of using non-traditional remotely sensed water-related indicators to calibrate and train a GPT-4o Vision AI model for the automated detection of poverty hotspots in the focus area. More than 400 high-resolution aerial images (7 cm and 30 cm resolution) were used in the study.

Available at: https://doi.org/10.55158/DEEPCFWP2