What's Happening at PSD
Related News Links
Upcoming Events

PSD Researchers Receive ASCR Leadership Computing Challenge Award

DOE logo
DOE logo
Cray XT4 supercomputer cluster (Franklin), Photo credit: NERSC

July 6, 2010

Researchers from the Earth System Research Laboratory (ESRL) have been selected to receive a 2010 ASCR Leadership Computing Challenge (ALCC) award. Tom Hamill and Jeff Whitaker of ESRL's Physical Sciences Division have been awarded 14,500,000 processor hours to use Department of Energy (DOE) high-performance computational resources for creating a next-generation multi-decadal "reforecast" data set. This data set is a first step in developing new, more accurate weather forecast applications for renewable energy such as several-days lead and even week-2 forecasts of the potential for wind-energy generation and incoming solar radiation, as well as heavy precipitation and streamflow into reservoirs generating electrical power.

The DOE Office of Advanced Scientific Computing Research (ASCR) supports the federal government's largest and most active computer science research effort. Federal and university researchers, suppliers, and companies, have used system software and software tools ASCR develops to capitalize on the capabilities of high-performance computers; allowing researchers to simulate and predict complex physical, chemical and biological phenomena. The ALCC program allocates up to 30% of the computational resources at the National Energy Research Scientific Computing Center (NERSC) and the Leadership Computing Facilities at Argonne and Oak Ridge for special situations of interest to advance clean energy and understanding the Earth's climate, for national emergencies, or for broadening the community of researchers capable of using leadership computing resources.

Reforecasts are retrospective forecasts of the weather, performed using a fixed version of a forecast model. If performed over a long enough training period, the statistics of past forecast errors can be used to correct real time forecasts with the same model. This technique has been shown to greatly improve the skill of probabilistic forecasts, increasing their value to a wide range of customers. This award of supercomputing resources will allow NOAA to more quickly and efficiently develop a range of new and more accurate probabilistic forecast products for renewable energy applications.

Contact: Tom Hamill More Information: