Developing A Quantitative Measure of Convective Forcing To Evaluate High Resolution Rapid Refresh Ensemble (hrrre) Variance

Abstract

Hazardous weather events have the greatest impact when they are not accurately forecasted. The quest for advanced lead times of accurate forecasts has motivated the improvement of numerical weather prediction models. Stemming from successful hourly updated forecasting by the High Resolution Rapid Refresh (HRRR) model, the High Resolution Rapid Refresh Ensemble (HRRRE) utilizes ensemble data assimilation for improved convective scale forecasts. The HRRRE was run in real-time experimentally during the Spring 2016 Hazardous Weather Testbed Experimental Forecast Program. HRRRE and other Convective Allowing Model guidance skill can vary widely in different weather regimes. Convective forcing is hypothesized to influence forecast skill and ensemble variance. Understanding the correlation between forcing and ensemble skill/variance has the potential to enhance the future HRRRE ensemble design. To study this relationship, an objective measure of convective forcing is required. This study developed the Reflectivity Convective Forcing Categorization (RCFC), a quantitative method to categorize convective forcing using Multi-Radar Multi-Sensor (MRMS) composite reflectivity observations. Both reflectivity coverage and rate of change of reflectivity were examined for the months of May and June 2016 utilizing RCFC. Several events exemplifying strong and weak forcing regimes were qualitatively analyzed using Storm Prediction Center mesoscale analyses, upper air maps, and surface analyses, to verify the RCFC method. HRRRE skill and variance for reflectivity and surface observations will be presented for strongly and weakly forced events. Results of the RCFC method will aid improvement of the future HRRRE ensemble design.