Validity of Drought Indices as Drought Predictors in the South-Central United States

Monday, 14 December 2015
Poster Hall (Moscone South)
Robert V Rohli, Nazla Bushra, Nina Lam, Lei Zou, Volodymyr Mihunov, Margaret Reams and Jennifer Argote, Louisiana State University, Baton Rouge, LA, United States
Drought is among the most insidious types of natural disasters and can have tremendous economic and human health impacts. This research analyzes the relationship between two readily-accessible drought indices – the Palmer Drought Severity Index (PDSI) and Palmer Hydrologic Drought Index (PHDI) – and the damage incurred by such droughts in terms of monetary loss, over the 1975-2010 time period on monthly basis, for five states in the south-central U.S.A. Because drought damage in the Spatial Hazards Events and Losses Database for the United States (SHELDUSTM) is reported at the county level, statistical downscaling techniques were used to estimate the county-level PDSI and PHDI. Correlation analysis using the downscaled indices suggests that although relatively few months contain drought damage reports, in general drought indices can be useful predictors of drought damage at the monthly temporal scale extended to 12 months and at the county-wide spatial scale. The varying time lags between occurrence of drought and reporting of damage, perhaps due to varying resilience to drought intensity and duration by crop types across space, irrigation methods, and adaptation measures of the community to drought varies over space and time, are thought to contribute to weakened correlations. These results present a reminder of the complexities of anticipating the effects of drought but they contribute to the effort to improve our ability to mitigate the effects of incipient drought.