Great Plains Drought in Simulations of the Twentieth Century

Abstract

Coupled global circulation models (CGCMs) have been widely used to explore potential future climate change. Before these climate projections can be trusted, the ability of the models to simulate present-day climate must be assessed. This study evaluates the ability of three CGCMs that participated in the Fourth Assessment Report of the Intergovernmental Panel on Climate Change to simulate long-term drought over the Great Plains region with the same frequency and intensity as was observed during the twentieth century. The three models evaluated in this study are the Geophysical Fluid Dynamics Laboratory Coupled Model, version 2.0 (CM2.0); the National Centers for Atmospheric Research Community Climate System Model, version 3 (CCSM3); and third climate configuration of the Met Office Unified Model (HadCM3). The models are shown to capture the broad features of the climatology of the Great Plains, with maximum precipitation occurring in early summer, as observed. However, all of the models overestimate annual precipitation rates. Also, in CCSM3, precipitation and evapotranspiration experience unrealistic decreases between the months of June and August. Long-term droughts are found in each simulation of the twentieth century that are comparable in duration, severity, and spatial extent as has been observed. However, the processes found to be associated with simulated long-term droughts vary among the models. In both CM2.0 and HadCM3, low-frequency variations in Great Plains precipitation are found to correspond with low-frequency variations in tropical Pacific SSTs. In CCSM3, on the other hand, there appears to be no significant correlation between tropical Pacific SST variability and Great Plains precipitation. Strong land?atmosphere coupling in CCSM3 may explain the persistence of long-term droughts in this model.