For centuries, water supply and sanitation have been used as indicators of poverty levels in Tanzania, and the link between the two is likely to strengthen due to population growth coupled with the effects of climate change. However, increasing technological development and modernization have unlocked new, affordable and faster ways of measuring poverty levels using non-traditional indicators. This research, carried out by Global Water Partnership Tanzania, uses non-traditional water-related indicators, such as overhead water storage tanks, proximity to wastewater ponds, payment of water bills, and proximity to water bodies and landfill, to detect extreme poverty in urban areas. The research will focus on the coast sub-basin of the Wami Ruvu Basin and data will be used to train generative AI to develop synthetic spatial distributions of poverty severity and the associated water-related indicators in the area. The intention is to develop a scalable model which will have the capacity to provide freely available, accessible, and ready to use data on poverty distribution to rapidly inform evidence-based policy decisions and interventions.
This research project is supported by DEEP challenge fund
