Can Moisture and Temperature Errors Within the Inflow Region be used to Predict MCS Displacement Errors in WRF?

Nocturnal Mesoscale Convective Systems (MCS) frequently affect the Great Plains in the summer, and a contributing factor to their initiation and evolution is the Great Plains low level jet (LLJ). Moisture brought in by the LLJ plays a key role in the formation and sustenance of MCSs. Thus, how well models depict specific humidity, as well as equivalent potential temperature, is likely to play a role in how accurately they forecast MCSs. In this study, the Weather Research and Forecasting (WRF) model is used to examine the relationship between displacement errors for the initiating MCSs and moisture and temperature errors that are present up to three hours beforehand upstream of the MCSs in WRF model simulations. A total of 18 cases are examined – nine type A cases (anticyclonic flow aloft) representing weakly forced synoptic regimes, and nine type C cases (cyclonic flow aloft) for strongly forced synoptic regimes. Rapid Update Cycle (RUC) analyses were used to represent observations, and all analyses were focused in 3 layers, 250 - 750 m AGL, 1000 - 1500 m AGL, and at 250 – 1500 m AGL. The WRF was configured to use the Thompson microphysics scheme and the Yonsei University (YSU) planetary boundary layer scheme. A significant correlation was found for Type C cases between the displacement error distances for the initiating MCSs in the WRF model, and the average moisture errors within subdomains of 125 km by 125 km centered near and upstream from where the MCSs initiated. The displacements were farther in the downstream direction the more negative the moisture errors. The correlations were present at all times examined, suggesting that forecasters might be able to anticipate how to adjust a WRF forecast based on the errors seen a few hours prior to MCS initiation. Type A cases did not correlate as well, suggesting a more complex dependence on a variety of factors in their initiation. More detailed case studies will be discussed for the Type A events to better explain differences in the correlations compared to Type C events.