Were global numerical weather prediction systems capable of forecasting the extreme Colorado rainfall of 9–16 September 2013?

Abstract

AbstractFrom 9–16 September 2013 significant portions of Colorado experienced extreme precipitation and flooding resulting in large socioeconomic damages and fatalities. Here we investigate the ability of eight global state‐of‐the‐art numerical weather prediction systems to forecast rainfall during the event. Forecasts were analyzed from initializations at 12 UTC 5 September to 12 UTC 12 September to determine when, and how well, the event was captured. Ensemble mean rainfall patterns initialized on 5 September (roughly 4+ day lead time) did not forecast the event's persistent nature; conversely, forecasts initialized on 9 September captured the rainfall patterns reasonably well, although with incorrect rainfall values. Accumulated rainfall forecasts improved when the region considered increased from a 0.5° area centered over Boulder to the entire state of Colorado. We conclude that the models provided guidance indicating a significant period of rainfall in Colorado from 9 September 2013, although not necessarily in the correct locations.