Paper Reviewed
Ganguli, P. and Ganguly, A.R. 2016. Robustness of Meteorological Droughts in Dynamically Downscaled Climate Simulations. Journal of the American Water Resources Association 52: 138-167.

Noting that "the impacts of climate change on drought attributes continue to be debated in the scientific community, even as multiple regions, globally and in the United States, experience severe droughts," Ganguli and Ganguly (2016) describe how they examined "the robustness of a suite of regional climate models (RCMs)" -- which make up part of the North American Regional Climate Change Assessment Program (NARCCAP) -- "in simulating meteorological droughts and associated metrics in present-day climate (1971-2003) over the conterminous United States (CONUS)." And what did they learn by so doing?

The two researchers report that the studied models (1) "fail to simulate significant drying trends over the Southwest and West," that (2) "reanalysis-based NARCCAP runs underestimate the observed drought frequency overall, with the exception of the Southwest," where (3) "all models including their multi-model ensembles overestimate observed drought frequency." In addition, they report that the models (4) "underestimate persistence in the drought-affected areas over the Southwest and West-North Central regions." And they add that (5) "global climate model-driven NARCCAP ensembles tend to overestimate regional drought frequencies."

Also of note is the fact that Ganguli and Ganguly report that the studied models (6) "exhibit considerable uncertainties while reproducing meteorological drought statistics," as evidenced by (7) "a general lack of agreement in the Hurst exponent, which in turn controls drought persistence." And, therefore, they conclude that (8) "water resources managers need to be aware of the limitations of current climate models," while (9) "regional climate modelers may want to fine-tune their parameters to address impact-relevant metrics."