Hot
Items 2007 Experimental forecasts could predict severe weather a week in advance
NSSL scientists are partnering with others to develop forecasts that could predict severe weather a week in advance. The effort is part of the 2007 NOAA Hazardous Weather Testbed Spring Experiment currently underway in Norman, OK and is called the Experimental Forecast Program (EFP).
Studies have shown people are more likely to seek shelter if they have been made aware of the severe weather threat prior to the issuance of a warning, such as through a severe weather watch. Accurate severe weather watches are dependent on forecasters being able to properly predict not only where and when severe thunderstorms will develop and how they will evolve over the next 4-8 hours, but also what kinds of storms are most likely to occur – single-cell, multi-cell, or supercell. Each poses a different type of threat.
The EFP is addressing this issue by testing and evaluating a number of daily, real-time Weather Research and Forecasting models (WRF) over domains covering much of the continental U.S. One highlight of the experiment is a high-resolution WRF ensemble called Storm Scale Ensemble Forecast (SSEF). SSEF represents an unprecedented real-time computational achievement and is part of a three-year project that will test and refine a convection-allowing ensemble to provide probabilistic guidance on high impact convective weather events. SSEF will do this by quantifying aspects of uncertainty in the forecast product and offering insights about a possible range of solutions.
Another ensemble model under evaluation will be a mesoscale model that assimilates hourly surface data. Since a key aspect in forecasting convective initiation is the specification and prediction of the mesoscale environment, accurate hour-by-hour analyses are an important component of the forecast process. Teams of forecasters and researchers will evaluate this ensemble and compare the output with the hourly SPC mesoscale analyses.
Several collaborators are providing valuable research and computing resources, some of which are available for the first time, allowing for significant improvements in precision. Single models are being provided by the NOAA Environmental Prediction Center and the National Center for Atmospheric Research. In addition, the Center for Analysis and Prediction of Storms at the University of Oklahoma, as part of the NSF Program Linked Environments for Atmospheric Discovery (LEAD), will provide an ensemble of high-resolution models run at the Pittsburgh Supercomputing Center. High speed Internet access links these resources by National LambdaRail to the HWT in conjunction with OneNet and a use of resources provided by Cisco Systems.
5/28/07