HR: 11:05h
AN: IN32A-04 [Abstracts]
TI: Science Enabling Applications of Gridded Radiances and Products
AU: * Goldberg, M
EM: mitch.goldberg@noaa.gov
AF: NOAA/NESDIS
Office of Research and Applications, 5200 Auth Road, Room 712, Camp Springs, MD 20746
AU: Wolf, W
EM: Walter.Wolf@noaa.gov
AF: NOAA/NESDIS
Office of Research and Applications, 5200 Auth Road, Room 712, Camp Springs, MD 20746
AU: Zhou, L
EM: Lihang.Zhou@noaa.gov
AF: NOAA/NESDIS
Office of Research and Applications, 5200 Auth Road, Room 712, Camp Springs, MD 20746
AB:
New generations of hyperspectral sounders and imagers are not only providing vastly improved information to monitor, assess
and predict the Earth's environment, they also provide tremendous volumes of data to manage. Key management
challenges must include data processing, distribution, archive and utilization. At the NOAA/NESDIS Office of Research and
Applications, we have started to address the challenge of utilizing high volume satellite by thinning observations and
developing gridded datasets from the observations made from the NASA AIRS, AMSU and MODIS instrument. We have developed
techniques for intelligent thinning of AIRS data for numerical weather prediction, by selecting the clearest AIRS 14 km field
of view within a 3 x 3 array. The selection uses high spatial resolution 1 km MODIS data which are spatially convolved to
the AIRS field of view. The MODIS cloud masks and AIRS cloud tests are used to select the clearest.
During the real-time processing the data are thinned and gridded to support monitoring, validation and scientific studies.
Products from AIRS, which includes profiles of temperature, water vapor and ozone and cloud-corrected infrared radiances for
more than 2000 channels, are derived from a single AIRS/AMSU field of regard, which is a 3 x 3 array of AIRS footprints (each
with a 14 km spatial resolution) collocated with a single AMSU footprint (42 km). One of our key gridded dataset is a
daily 3 x 3 latitude/longitude projection which contains the nearest AIRS/AMSU field of regard with respect to the center of
the 3 x 3 lat/lon grid. This particular gridded dataset is 1/40 the size of the full resolution data. This gridded
dataset is the type of product request that can be used to support algorithm validation and improvements. It also provides
for a very economical approach for reprocessing, testing and improving algorithms for climate studies without having to
reprocess the full resolution data stored at the DAAC. For example, on a single CPU workstation, all the AIRS derived
products can be derived from a single year of gridded data in 5 days. This relatively short turnaround time, which can be
reduced considerably to 3 hours by using a cluster of 40 pc G5processors, allows for repeated reprocessing at the PIs home
institution before substantial investments are made to reprocess the full resolution data sets archived at the DAAC. In
other words, do not reprocess the full resolution data until the science community have tested and selected the optimal
algorithm on the gridded data. Development and applications of gridded radiances and products will be discussed. The
applications can be provided as part of a web-based service.
DE: 0434 Data sets
SC: Earth and Space Science Informatics [IN]
MN: Fall Meeting 2005