Title :
The Joint Ensemble Forecast System (JEFS) Experiment
Author :
Kuchera, Evan L. ; Cunningham, Jeffrey G. ; Rentschler, Scott A. ; Rugg, Steven A. ; Sittel, Matt ; Sestak, Michael ; Holt, Teddy ; Hansen, James
Author_Institution :
US Air Force Weather Agency (AFWA), Offutt AFB, NE, USA
Abstract :
The Joint Ensemble Forecast System (JEFS) Dedicated HPC Project Investment (DHPI) focuses on the creation and communication of environmental information in a timely, focused, useful, and reliable manner for the US Air Force and Navy. Currently, most environmental forecasts are created from a single best estimate analysis and forecast model. This deterministic process can provide imperfect results that result in large financial costs (extra fuel, weather damage or mission aborts). In this project, the ensemble-based approach of weather forecasting is used to improve forecast accuracy; reduce costs of defensive and offensive military scenarios impacted by inclement weather; and further the benefits of probabilistic forecasts to a variety of military applications. The project combines many model forecasts that start from different initial states of the atmosphere, that use different techniques for estimating how the atmosphere evolves over time, and that use different models provided by both the Navy (Fleet Numerical Meteorology and Oceanography Center) and Air Force (Air Force Weather Agency) numerical weather prediction centers to produce a joint ensemble. Ensemble member forecasts from two independently developed model systems are expected to better cover the range of potential atmospheric states, thus more completely representing the occurrences of greater or lesser forecast certainty. Experimental joint ensemble forecast products are already available on the AFWA web site (https://weather.afwa.af.mil/host_home/DNXM/JEFS/jefs.html), but operational products are not expected until late FY 2010 to early 2011. Most of the model development has been completed; ongoing work includes setting up real-time data exchanges across modeling centers, post processing to improve forecast reliability, and for production of mission specific products. Further work remains to educate decision makers about weather certainty-many expect a correct deterministic weather forecast and do - ot understand how to exploit probabilities. The additional computer resources provided by this DHPI have allowed more rapid development of the joint ensemble forecast system than would have occurred with the existing systems at AFWA and FNMOC.
Keywords :
atmospheric techniques; data assimilation; probability; weather forecasting; Air Force Weather Agency; Dedicated HPC Project Investment; Fleet Numerical Meteorology and Oceanography Center; JEFS experiment; Joint Mesoscale Ensemble; The Joint Ensemble Forecast System experiment; US Air Force; US Navy; cost reduction; data assimilation; decision making; defensive military scenarios; deterministic weather forecasting; environmental forecast; environmental information; forecast model; forecast reliability; inclement weather; military application; numerical weather prediction; offensive military scenarios; potential atmospheric state; probabilistic forecast; real-time data exchange; single best estimate analysis; weather certainty; Atmospheric modeling; Joints; Numerical models; Predictive models; Uncertainty; Weather forecasting;
Conference_Titel :
DoD High Performance Computing Modernization Program Users Group Conference (HPCMP-UGC), 2009
Conference_Location :
San Diego, CA
Print_ISBN :
978-1-4244-5768-7
DOI :
10.1109/HPCMP-UGC.2009.47