Examination of the Standardized Precipitation Index for a Measure of Global Crop Losses by Extreme Wets and Dries

Climatic extremes associated with precipitation—the extreme wet and the extreme dry—are major causes of crop losses in the world, and then the study on the crop losses by the climatic extremes is one of the top priority to secure global food security in the changing world. For a pilot study, the standardized precipitation index (SPI) was examined as a key measure to investigate the climatic extreme impacts on global crop losses.The SPI was prepared from 1-month to 12-month due to its scales with the monthly precipitation data originated from WFDEI (Water and Global Change Forcing Data applied to ERA-Interim), and then compiled to an effective SPI to aggregate reasonable days for one month before crop harvest. The crop data was compiled to a yield anomaly to extract those climatic variabilities with the yearly crop yield data coming from GDHY (Global Dataset of Historical Yield). We found that: first, about one-fourth of the global harvested areas were statistically significant in the results of the linear regression analyses between the yield anomaly and the effective SPI in each climatic extreme; second, this significant areas were distributed according to the SPI scales which was the bin sizes of monthly precipitation mean for SPI calculation; finally, a global function of crop losses by the extreme wets could be estimated with the effective SPI like a global function of crop losses by the extreme dry suggested by our previous study. The findings would be useful informations to build a statistical simulation model forced by precipitation data for the past estimations and the future predictions of crop losses by the climatic extremes.