Forecaster Decision Support Environment

Abstract

The Forecaster Decision Support Environment (FDSE) project comprises a set of activities designed to provide NWS forecasters more efficient techniques so that they may spend more time on Impact-based Decision Support Services (IDSS). During its first year of development, the project has focused on three areas: a gridded forecast monitor so forecasters may quickly assess and compare the current forecast to observations and numerical guidance, new tools to assist the forecaster when maintaining the gridded forecast in the short-term (0-12 hours), and new set of ensemble access and manipulation capabilities that will offer a much richer data set, as well as advanced tools, that calculate ensemble-based probabilities of events and conditions. The gridded forecast monitor automatically compares forecasts to observations and other forecasts and allows forecasters to quickly evaluate at a glance the state of the current gridded forecast. Additional displays such as histograms and scatter plots let forecasters more deeply investigate the details concerning where discrepancies exist so that they may more easily diagnose the cause. The short-term forecast tool provides an interface with which forecasters may identify observational and/or numerical guidance data sets along with weighting factors for each. The result is a set of blended forecast grids that represent a consensus solution composed of each selected observed or forecast data set. The third component of FDSE is the addition of tools to access and operate on ensembles of numerical weather predictions. With proper workstation tools, forecasters can combine and blend these forecasts into a result that not only expresses a most likely scenario, but alternative scenarios and estimates of event probabilities, such as overnight freezing or winds over 25 kts. Since ensemble modeling generates an extremely high volume of data, far too large to communicate in real time to each WFO, we are investigating an approach for extracting and transmitting small, customized subsets of large, remote data sets via OGC-compliant web services. This presentation will discuss our accomplishments over the past year and provide a view of future work.