Evaluation of Real-Time WRF Forecasts Over the Mid-Atlantic

TerpWRF, a real-time application of the Weather Research and Forecasting Model (WRF) running at the University of Maryland, College Park has been running since 2015. The main goals for TerpWRF are to provide real-time forecast guidance as well as provide a testbed for students to learn more about numerical weather prediction and model verification. TerpWRF uses the WRF-ARW model and is cold-started four times daily from the National Centers for Environmental Prediction Global Forecast System analysis.  TerpWRF utilizes a 12-km resolution outer domain and a 4-km inner domain centered over Maryland. Different configurations of the model physics and resolution were tested to optimize the potential increase in forecast skill over the local area within reasonable computational time. Results from both qualitative and quantitative verification metrics that were used to determine the performance of TerpWRF will be presented.  A qualitative evaluation from a convective flooding rain event on July 30, 2016 in the local area was used for a case study to help test the different model configurations. The model evaluation tools (MET) from the Developmental Testbed Center was used for quantitative verification of TerpWRF as well as to provide a comparison to other models for the Maryland area.