Spurious precipitation in the Weather Research and Forecasting model (WRF) was analyzed to determine its impact on future forecasts of Mesoscale Convective Systems (MCSs) in the model. Cases were initially identified using output from the National Severe Storms Laboratory (NSSL) WRF model on days where spurious precipitation occurred during the early daytime hours preceding an MCS that was poorly forecasted within the model. These cases were then simulated using a WRF model configuration that was similar to the NSSL WRF model in order to perform sensitivity tests to better understand the role of several atmospheric processes. Additionally, an investigation into WRF Planetary Boundary Layer (PBL) and microphysics schemes sensitivities in producing spurious or false alarm MCSs was performed. These tests involved using model simulations from the NOAA Hazardous Weather Testbed 2010 and 2011 spring experiment datasets.

Four sensitivity tests compared with a control run were performed to examine the role spurious precipitation has in altering the PBL within the WRF model prior to initiation of a MCS. One sensitivity test removed all moisture fields except vapor in order to remove the effects spurious precipitation had on the PBL. A second test removed the latent heating effects caused by the microphysics scheme during the period of spurious precipitation. A third test removed the impacts of spurious precipitation on soil moisture and vegetation content. A final test removed only cloud radiation effects without eliminating clouds like in the vapor-only sensitivity. After spurious precipitation ended in the model runs all sensitivity test runs were halted and returned to their original state and run for completion. Equitable Threat Score (ETS) analyses, the factor separation method, and a qualitative analysis were performed in order to determine the impact of each of the sensitivities, and the possible roles played by the various processes in the erroneous forecast of the MCS. It was found that among the sensitivities, the removal of cold pool effects (latent heating) and the removal of soil moisture effects (soil moisture) had the largest positive influence in affecting the main MCS forecast when spurious precipitation occurred. There were no conclusive patterns found in thermodynamic variables examined among the Hazardous Weather Testbed WRF ensemble members that reduced the probability of a false alarm MCS.