Exploring Forecast Sensitivity through an Ensemble of Varying Land Surface Parameterizations, Soil Moisture, and Vegetation Characteristics

The NASA Short-term Prediction Research and Transition (SPoRT) Center supports the use of a variety of unique land surface data sets within the Weather Research and Forecast (WRF) modeling system.  The application of these experimental dataset are fostered through partnerships with NOAA/National Weather Service (NWS) Weather Forecast Offices (WFOs) who continue to operate local mesoscale models to understand the role of varying data sets and parameterizations on local weather simulations.  These efforts include distribution of a daily green vegetation fraction (GVF) product derived from the Visible and Infrared Imaging Radiometer Suite (VIIRS) at NOAA/NESDIS, and a higher-resolution depiction of soil moisture provided by a daily 3-km resolution simulation of the NASA Land Information System (LIS), forced by North American Land Data Assimilation-Phase 2 (NLDAS-2) and Multi-Radar Multi-Sensor gauge-correction radar precipitation estimates.  In previous studies, inclusion of these data sets within regional modeling frameworks have been demonstrated to have impacts on the prediction of surface temperatures and moisture, the timing and location of mesoscale circulations, and resulting impacts on the evolution of precipitation patterns, particularly for events with relatively weak synoptic-scale forcing.  With the increasing use of ensembles in regional severe weather forecasting, this study explores the impact of varying land surface parameterizations and initialization data sets on the evolution of a severe weather event from 31 March 2016 that occurred during the VORTEX-Southeast campaign, centered in North Alabama.  Simulations include six Global Ensemble Forecast System (GEFS) members with their corresponding surface fields, serving as a control set of runs.  Two other sets of six simulations consist of replacing the GEFS surface fields with the 3-km NASA LIS output, and another included both the NASA LIS and 4-km resolution VIIRS GVFs to improve representation of the land surface over the default monthly GVF climatology.  This presentation will compare and contrast various stability and severe weather parameters in the model, their evolution, and their prediction of afternoon and evening convection in comparison to ground-truth observations.