As we continue to see an increase in extreme precipitation anomalies during winter months across the United States, it is important to focus on how to simulate better these events. These events can cause substantial flooding that can severely erode soil and weaken energy-transmission infrastructure. One way to better understand and simulate these extreme precipitation events is to assess how resolution in regional climate models affects processes leading to extreme precipitation. During winter months, synoptic dynamics play an important role in extreme precipitation events and should be fairly well resolved in regional climate models.

This study focuses on extreme sub-daily precipitation in the Upper Mississippi River Valley during the months of December, January, and February (DJF). We analyze extreme, 6-hourly precipitation as simulated for the DOE FACETS program by RegCM4 and WRF using ERA-Interim boundary conditions for the period 2002 through 2012 at grid spacings of 12, 25 and 50 km. We compare composites of simulated 6-hourly extreme events with those occurring in NOAA Stage IV quantitative precipitation for the same period. In addition, we compare composite 6-hourly fields of 2m temperature, 500hPa geopotential height, 10m wind, and 2m specific humidity for the simulated events with composites of the same fields during Stage IV events, using the North American Regional Reanalysis. We evaluate how changing resolution in the two models affects the intensity of extreme events versus Stage IV events as well as how the circulation and thermodynamics of the events changes with resolution, highlighting factors are most sensitive to resolution and their impact on replicating observed behavior.