How much drought is ‘just right’?Spatial Differences in ‘Optimal Drought Severity’ for Drought Tolerant Maize

The potential of agricultural biotechnology to produce seed traits that reduce yield risk generates widespread excitement. Hopes are high that in an era of tightening water constraints and climate change drought tolerant (DT) crop varieties will stabilize food production and allow for greater adaptation to changing production conditions. Although the DT agenda has united the public and private agricultural research system around concerns that seem to span rich and poor countries alike, the underlying crop-drought relationships differ substantially over space. We devise a methodology for characterizing these spatial differences in ‘optimal drought severity’ across two locations in Africa (Ethiopia and Mali) and one in the U.S. (Wisconsin) and analyze these differences to infer implications for the diffusion and impact of DT crop varieties. Drought-conditioned yield distributions for non-DT maize provide the benchmark against which we assess relative DT benefits in each setting. We construct a distribution of expected relative DT benefits as the product of the site-specific relative DT benefits distribution by rainfall and the site-specific rainfall distribution. We discuss how spatial differences in these expected relative DT benefit distributions may affect farmer decision making and welfare as well as agricultural adaptation to climate change.