ABSTRACT
Weather index insurance offers rapid payouts to protect smallholder farmers against different kinds of weather shocks, such as droughts and floods. Ideally, an increased level of risk protection also promotes agricultural productivity. In many regions, in situ observations are scarce or not quality-controlled, leaving satellite data as the only reliable source of information for the design of insurance indices. Currently, most indices rely on satellite-derived estimations of precipitation or vegetation greenness/health. This study concentrates on an initial analysis of a multisensor satellite soil moisture product, which was generated within the Climate Change Initiative of the European Space Agency, to close the gap between atmospheric anomalies and their impact on the ground. Our findings show a better agreement between standardized seasonal soil moisture and vegetation anomalies than that between rainfall and vegetation anomalies over a test site in Central Senegal. Including a temporal lag between the response of vegetation to changes in moisture increases the correlation for both pairs, but it remains substantially higher for soil moisture. These findings suggest that satellite soil moisture (e.g., for rainfall validation) could be a particularly valuable dataset in index insurance, once an operational, quality-controlled, multisensor dataset becomes available.
