Comparison of hydrological and vegetation remote sensing datasets as proxies for rainfed maize yield in Malawi

Weather Index-based Insurances (WIIs) have emerged as a promising risk coping mechanism to compensate for weather-induced damage rainfed agriculture. Remote sensing may provide cost-effective information capable of discriminating the weather spatial variability thus reducing basis risk, i.e., mismatch between weather-based index triggering insurance payout and actual experienced by farmers, which is often one causes hindering wide implementation WIIs. In this work we assess indices based on remote datasets are best proxy indicators maize yield in Malawi. We analyse (district scale) temporal (monthly) correlations historical data several including Climate Hazards group Infrared Precipitation with Stations (CHIRPS) dataset, ESA CCI Soil Moisture combined dataset (version 4.2), Evaporative Stress Index (ESI) from Atmosphere-Land Exchange Inversion model (ALEXI), MOD13Q1 Normalized Difference Vegetation (NDVI) Enhanced (EVI). With respect previous literature, exploits crop at sub-national level allows us correlation hydro-meteorological vegetation variables higher resolution than what commonly done (i.e., national using FAO statistics) ultimately explore issues related WII risk. Results show that satellite high variability, making it difficult identify unique same time simple effective entire country. Precipitation, particularly standardized March precipitation anomaly, has highest (with Pearson values 0.55), Central South moisture NDVI do not add much value anticipating district scale. From methodological perspective, our shows indexes identified by: i) considering fine resolution, whenever possible; ii) accounting vulnerability different growing stages water-stress; iii) distinguishing water scarce abundant events.