An overview of the 2010 NOAA Hazardous Weather Testbed spring forecasting experiment

The NOAA Hazardous Weather Testbed conducts annual Spring Forecasting Experiments organized by the Storm Prediction Center (SPC) and National Severe Storms Laboratory (NSSL) to test and evaluate emerging scientific concepts and technologies for improved analysis and prediction of hazardous mesoscale weather. A primary goal is to accelerate the transfer of promising new tools from research to operations, through the use of intensive realtime experimental forecasting and evaluation activities conducted during the spring and early summer convective storm period. From May 17 through June 18, 2010, more than 70 participants including operational forecasters, research scientists, academic faculty, graduate students, and administrators from numerous organizations across the US participated in the HWT, focusing on application of cutting edge numerical weather prediction systems to address high impact convective weather forecasting challenges.

This year, in addition to the traditional HWT focus on severe convective storms producing tornadoes, damaging wind gusts, and large hail, collaborations with other NCEP Centers were established within the HWT to help address a wider range of convective weather hazards. The Hydrometeorological Prediction Center (HPC) led an initial effort to explore high resolution model forecasts of precipitation and excessive rainfall associated with warm season convection, and the Aviation Weather Center (AWC) tested and evaluated new forecasting tools to improve thunderstorm forecasts for aviation. The three convective forecasting components operated simultaneously within the HWT with structured forecast and evaluation activities occurring each day. The weekly participants rotated through each component on a daily basis to broaden their understanding and gain unique perspectives on the different thunderstorm forecasting challenges. The experiment also provided an opportunity to strengthen collaborations between the severe weather, aviation, and QPF communities by identifying shared thunderstorm forecast challenges, and helped to enhance awareness of operational issues related to convective forecast consistency.

High resolution modeling systems were contributed by NSSL, the University of Oklahoma's Center for Analysis and Prediction of Storms (CAPS) working with the University of Tennessee National Institute for Computational Sciences, the NOAA/National Centers for Environmental Prediction Environmental Modeling Center (NCEP/EMC), NOAA/Earth System Research Laboratory Global Systems Division (ESRL/GSD), and the National Center for Atmospheric Research (NCAR). Automated thunderstorm guidance products were also provided by the NWS/Meteorological Development Laboratory (MDL) and MIT/Lincoln Lab.

A cornerstone of the experiment was a 4 km grid length, 26 member multi-model, multi-physics, multi-initial condition Storm-Scale Ensemble Forecast (SSEF) system produced by CAPS to help quantify uncertainty associated with convective-scale predictability. There were also multiple deterministic convection-allowing WRF forecasting systems configured with 1-4 km grid spacing run over CONUS geographic areas. All models were initialized at 00 UTC and integrated forward in time to at least 30 hrs; the NCAR and EMC models were also updated at 12 UTC. In addition, the GSD High Resolution Rapid Refresh (HRRR) model was run hourly out to 15 hours and tested for its capability to provide frequent short-term update information on convective storms. Innovative WRF model fields, such as simulated satellite imagery and explicit total lightning from microphysics parameters, were contributed by several NOAA/NESDIS Cooperative Institutes (CIRA, CIMSS, and SPoRT). In addition to the subjective evaluations, traditional and new object-based objective verification measures produced by the Developmental Testbed Center were utilized to assess their value in convective scale model verification, and to compare with the subjective evaluations.

Examples of the forecasting and evaluation activities and relevant findings will be discussed. Various aspects of the specific experimental models are found in a number of companion papers submitted to the conference.