VALIDATION OF THE NWS MIAMI WEATHER RESEARCH AND FORECASTING SYSTEM MODEL FORECASTS DURING THE 2011 SOUTH FLORIDA CONVECTIVE SEASON

摘要

Advances over the past several years in high-resolution atmospheric weather modeling include theintroduction of the enhanced RUC13 (13km), which assimilates 3-d level III radar reflectivity, theavailability of NASA’s Short-term Prediction Research and Transition Center (SPoRT) highresolutionModerate Resolution Imaging Spectroradiometer (MODIS) Advanced MicrowaveScanning Radiometer (AMSR-E) Sea Surface Temperature (SST) composites (providing superiordetails across surrounding ocean areas), and NASA's high-resolution Land Information System(LIS) surface datasets. In order to study the impact of these improvements and the performance ofvarious local model configurations at WFO Miami, an experiment was conducted during the 2011convective season using the Weather and Research Forecasting (WRF) Environmental ModelingSystem (WRFEMS) released and maintained by the National Weather Service (NWS) Science andOperations Officer Science and Training Resource Coordinator (SOO STRC). The experimentconsisted of verifying and comparing against each other the performance of different modelconfigurations using Equitable Threat Scores (ETS), Areal Bias (AB) Scores, and Percent Correct(PC) scores for different precipitation thresholds. The focus of the analysis was on short termconvection. The model configurations consisted of a variety of local high-resolution WRFconfigurations and the RUC13. Emphasis was also given to the effect on the local WRF modelconfigurations of NASA’s SPoRT surface datasets as well as to the effect of using explicitconvection versus convective parameterization. The skill scores were computed using Stage IVgridded precipitation data as ground truth. This paper presents the results of this experiment.