Tracking the Effects of Global Climate Change
Point a new video camera at children playing soccer or dancing in a show, and the scene is often a blur. Adjust the focus, and then a clear picture comes into view.
Climate scientists have long had a similar—although far more daunting—experience when trying to model Earth's atmosphere. The supercomputers and modeling techniques they use have not been powerful enough to yield fine-detailed images like those taken by satellites.
This year, NASA climate modeling took a dramatic step towards matching satellite resolution with the combination of an expanded supercomputer at NASA's Goddard Space Flight Center in Greenbelt, Md., and an innovative global atmospheric model.
GEOS-5-modeled clouds at 3.5-kilometer global resolution on January 2, 2009 as a strong blast of arctic air moves off the east coast of the United States, producing a series of cumulus cloud streets forming over the ocean, parallel to the flow of the low-level winds.
Goddard recently added 8,256 current-generation Intel Xeon 5500 series processors to its Discover supercomputer as two integrated IBM iDataPlex "scalable units." Using a 20 gigabit-per-second InfiniBand network, Goddard technical staff linked the two scalable units to host climate simulations with higher resolution and longer time duration than possible using processors installed at Goddard just last year.
Within hours of the linkage, Discover's newest processors were running the Goddard Earth Observing System Model, Version 5 (GEOS-5) from the Global Modeling and Assimilation Office (GMAO). Researchers used a special "cubed-sphere" version of GEOS-5 that covers the globe with nearly identically sized grid boxes, within which the model solves mathematical equations representing the atmosphere's dynamics and physics. The cubed-sphere GEOS-5 ran at resolutions up to 3.5 kilometers, allowing the formation of numerous cloud types within a global simulation.
"We see convective clusters in the tropics, the open and closed cell structure of the marine stratocumulus layer, and unprecedentedly detailed mid-latitude cyclone structure for a global model," said William Putman, lead of the Advanced Software Technology Group in Goddard's Software Integration and Visualization Office. With 8,256 highly efficient processors performing in tandem, Discover enabled simulating regional climate details at the highest resolutions to date with a global climate model. Visualizations of the results closely resemble imagery from the latest satellites.
By exploring global climate at regional scales NASA researchers are able to better understand the influence of regional features such as organized tropical convection, hurricanes, flooding, and drought, and the effects these features have on the global circulation, as well as how these features might be altered in a changing climate.
Simulations by GEOS-5 and other NASA climate models work in concert with the agency's Earth-observing satellite missions to advance scientific understanding and meet societal needs. Beginning this month, Discover is hosting NASA's modeling contributions to the Intergovernmental Panel on Climate Change (IPCC), the leading scientific organization for assessing climate change. The GMAO and the Goddard Institute for Space Studies in New York City will perform simulations for the IPCC's Fifth Assessment Report, which is due to be published in 2014.
More information:
+ Global Non-Hydrostatic Modeling with the NASA GEOS-5 Atmospheric Model (PDF-580 KB)
Contact information:
William Putman
Lead, Advanced Software Technology Group
Software Integration and Visualization Office
NASA Goddard Space Flight Center
William.M.Putman@nasa.gov
(240) 778-5697
