Verification of High-Resolution RAMS Forecasts over East-Central Florida during the 1999 and 2000 Summer Months

Abstract

This paper presents an objective and subjective verification of a high-resolution configuration of the Regional Atmospheric Modeling System (RAMS) over east-central Florida during the 1999 and 2000 summer months. Centered on the Cape Canaveral Air Force Station (CCAFS), the innermost nested grid of RAMS has a horizontal grid spacing of 1.25 km, thereby providing forecasts capable of modeling finescale phenomena such as ocean and river breezes, and convection. The RAMS is run operationally at CCAFS within the Eastern Range Dispersion Assessment System (ERDAS), in order to provide emergency response guidance during space operations. ERDAS uses RAMS wind and temperature fields for input into ERDAS diffusion algorithms; therefore, the accuracy of dispersion predictions is highly dependent on the accuracy of RAMS forecasts. The most substantial error in RAMS over east-central Florida is a surface-based cold temperature bias, primarily during the daylight hours. At the Shuttle Landing Facility, the RAMS point error statistics are not substantially different than the National Centers for Environment Prediction Eta Model; however, an objective evaluation consisting of only point error statistics cannot adequately determine the added value of a high-resolution model configuration. Thus, results from a subjective evaluation of the RAMS forecast sea breeze and thunderstorm initiation on the 1.25-km grid are also presented. According to the subjective verification of the Florida east coast sea breeze, the RAMS categorical and skill scores exceeded that of the Eta Model predictions in most instances. The RAMS skill scores in predicting thunderstorm initiation are much lower than the sea-breeze evaluation scores, likely resulting from the lack of a sophisticated data assimilation scheme in the current operational configuration.