Item

Abstract

Agriculture is an important economic pillar of many countries in Africa, and also it is also one of the most vulnerable industries to climate change. Increases in extreme events such as floods and droughts will seriously threaten the survival of many stallholder farmers in Africa. In this context, the Global Center on Adaptation, together with the African Development Bank and the World Bank, has launched a series of projects to strengthen infrastructure construction, establish adaptation measures, and improve climate resilience in African countries. However, due to the data scarcity in, for example, climate, meteorology, physical geography and agricultural data, the implementation of these projects also faces some challenges. For instance, when promoting inclusive agricultural insurance, due to the incompleteness of agronomic data and the poor quality of meteorological data, it is hard to accurately estimate the basic risk, which makes the insurance product risky and high in cost. For this motivation, this study will take Chad as an example, to demonstrate the use of disparate data sources to make up for the defects of relevant data. By combining remote sensing, historical observation and topographic data and other data, using hierarchical Bayesian models and artificial intelligence technology, it is possible to more accurately identify and predict the climate risks, so as to improve the sustainability of inclusive agricultural insurance. Ultimately, the capacity of climate adaptation measures will be enhanced through the use of disparate data sources.