Storm-scale Data Assimilation and Ensemble Forecasts of 31 May 2013 El Reno, Oklahoma Tornado using Multi-function Phased Array Radar Observations

This study evaluates the impact of assimilating MPAR observations on ensemble probabilistic forecasts of 31 May 2013 El Reno Tornadic thunderstorm event. Known as the widest tornado in history, the EF-3 El Reno tornado took a complex path with rapidly changing speed and direction. Examination of the observations sampled by MPAR for nearly 4 hours with less than 1 minute interval reveals a fast-evolving storm with multiple-scale rotations including the low-level mesocyclone that relates to the El Reno tornado. The intensity and complexity of the supercell makes it an attractive case to evaluate the impacts of high-temporal-resolution MPAR observations on very short-term storm-scale ensemble prediction. A 36-member ensemble at 1-km horizontal grid spacing centered on El Reno, OK is initialized from the 3-km storm-scale WRF-DART ensemble system developed at the National Severe Storms Laboratory. Radial velocity and reflectivity observations from MPAR are assimilated continuously using the Center for Analysis and Prediction of Storms ( CAPS) four-dimensional asynchronous ensemble square-root filter (4DEnSRF) algorithm and ensemble forecasts are launched every 10 minutes using the WRF-ARW model. The goal is to assess how quickly and accurately the high-temporal-resolution MPAR observation assimilation initialize the storm into the model. The ensemble probabilistic forecasts of reflectivity, cold-pool and low-level rotations of the El Reno Tornado will be presented at the confererence.