Identification of hydroclimate subregions for seasonal drought monitoring in the US Great Plains

摘要

Identification of subregions that share similar historical drought variability provides useful information for drought monitoring, mitigation planning, and resource allocation. This study examined space-time historical drought variability for the Great Plains spanning from 1901 to 2015 by using rotated Empirical Orthogonal Functions (rEOFs). The Standardized Precipitation-Evapotranspiration Index (SPEI) on a three-month timescale was utilized to examine spatial and temporal changes in agricultural drought. We propose a new procedure for identifying the number of rEOFs to be selected for reconstructing subregions. Drought event intensities of moderate, severe, and extreme categories increased in recent years although the number of drought events decreased. Seasonal rEOFs demonstrated that 9-12 subregions were adequate to explain a significant proportion of the original variability in the Great Plains. The time series for each subregion was highly correlated to the original SPEI data and reflected the seasonal meteorological processes that drive drought variability. Several significant wetting trends were found, and there was statistical evidence that drought and wetting event severities had increased for a few subregions. Summer drought has become more variable across space and time, indicating that a more diverse set of resources and strategies might be needed to mitigate impacts of spatially-variable drought and wetting events in coming decades. Winter season drought has become less variable, indicating that perhaps resources could be consolidated when dealing with impacts on a larger scale; however, less variability implies that drought and wetting events may occur across larger regions of the Great Plains during a given season.