New Supercomputer to Boost Weather Forecasting

(Oct. 24) -- Jokes about inaccurate weather forecasts are as popular as jokes about doughnut shops and the police, but advances in satellite data and computers have dramatically improved climate science in the past 50 years.

And there will soon be a brand new supercomputer available to government scientists aimed at improving forecasts, whether just a few hours out or decades into the future.

The National Oceanic and Atmospheric Administration's recent ribbon-cutting ceremony in Fairmont, W.Va., marked the future home of the NOAA Environmental Security Computing Center (NESCC), a $27.6 million, 54,000-square-foot facility expected to be fully operational by fall 2011.

NESCC will be used to improve the accuracy of existing computer models and to develop new systems.

"NOAA is committed to developing climate models that are scientifically credible, well-vetted and capable of making predictions and projections of Earth's climate variations and change over a broad range of time scales," V. Ramaswamy, director of NOAA's Geophysical Fluid Dynamics Laboratory, told AOL News. "The issues to be addressed include the seasonal-to-decadal-to-multidecadal time scales and extremes, with the models being underpinned by the advancements in scientific knowledge and making use of NOAA and other observations."

In addition to developing these long-term climate models, existing weather forecast models will undergo improvements, said Stephen Lord, director of NOAA Environmental Modeling Center, adding that the new computer facility "will be extremely useful in developing new systems."

Current systems include the GFS (Global Forecast System) and the CFS (Climate Forecast System).

The GFS is a computer forecast model that generates global forecasts four times a day and is depended upon by meteorologists to generate detailed, specific location forecasts (temperature, precipitation, wind, etc.) for time periods ranging from the next couple of hours to the next couple of weeks. Lord said that the "development of advanced versions of the GFS data assimilation and forecast model will be tested for future implementation" with the new supercomputer.

Once an improved GFS is made available to weather forecasters, short-range forecasts will immediately improve.

The CFS is a longer range model that combines data from both the oceans and the atmosphere to create much more general forecasts months into the future, such as where it will be warmer and wetter than normal next month or cooler and and drier than normal next summer. Historically, most weather forecast models were based only on atmospheric data, not ocean data, and this coupled approach is expected to become an increasingly important tool in generating the type of longer-range weather forecasts that the public and business sector are beginning to expect.

The next version of the CFS will be made available to forecasters before NESCC becomes operational, but research and development done at NESCC will be included in a version to be implemented in five or six years. As this model improves, long-range forecasts will gradually become more specific, accurate and dependable.

Potential new systems will be researched and developed using the new supercomputer, including those focusing on climate factors such as El Nino and La Nina -- weather phenomena that often have a pronounced effect on global weather.