NASA’s Global Modeling and Assimilation Office (GMAO) uses the Goddard Earth Observing System Model, Version 5 Data Assimilation System (GEOS-5 DAS) to produce global numerical weather forecasts on a routine basis. GMAO forecasts play important roles in managing NASA’s fleet of science satellites and in researching the impact of new satellite observations.
In order to provide timely information about the state of the atmosphere for NASA instrument teams and researchers, the GMAO runs the GEOS-5 DAS four times each day in real time. For each forecast, it is necessary to provide accurate initial conditions that drive the GEOS-5 forecasts. To do this, the best estimate of the full, three-dimensional atmospheric state is determined by combining the latest observations and a short-term, 6-hour forecast—a process known as data assimilation. The GEOS-5 DAS assimilates more than 5 million observations during each 6-hour assimilation period.
These observations are assembled from a number of sources from around the globe, including NASA, NOAA, EUMETSAT (European Organization for the Exploitation of Meteorological Satellites), commercial airlines, the US Department of Defense, and many others. Similarly, each observation type has its own sampling characteristics. It can be seen in the animation how different observation types have different strategies. One of the main challenges of data assimilation is to understand how all these observations are alike, how they differ, and how they interact with each other.
Funding for the development of the GEOS-5 model and data assimilation system development comes from NASA's Modeling, Analysis, and Prediction Program and the NASA Weather Focus Area's contribution to the Joint Center for Satellite Data Assimilation.
The GEOS-5 DAS runs at the NASA Center for Climate Simulation, which is funded by NASA’s High-End Computing Program.