Exploring a Missed Convection Initiation Forecast by Assimilating GOES-16 Brightness Temperatures and WSR-88D Observations

Prediction of convection initiation (CI) remains challenging, as it depends upon accurate forecasts of both the pre-convective environment and the processes responsible for lifting a parcel to its level of free convection. Numerous cases have been documented where convection allowing models (CAMs) missed CI that did take place or predicted CI that ultimately did not occur. Here, we present a case where the operational High-Resolution Rapid Refresh (HRRR) model predicted a supercell late in the day on 28 April 2019 near Dodge City, Kansas, that was not observed. To explore the ability of novel GOES-16 and WSR-88D observations to improve this forecast, a 40-member ensemble is created using the HRRR configuration of the WRF-ARW model. The ensemble uses 20 members from the 0000 UTC Global Ensemble Forecast System (GEFS) and another 20 members from the 0600 UTC GEFS on 28 April, in which all members are then centered around the 1200 UTC HRRR analysis. We will assimilate GOES-16 brightness temperatures and WSR-88D observations of convective boundary layer (CBL) depth and boundary layer radial velocities to improve the environmental conditions on this day and hopefully eliminate the predicted CI in southern Kansas that was not observed. Results from the ensemble data assimilation experiments will be compared with each other and observations.