UMBC ebiquity

A Web Service Tool (SOAR) for the Dynamic Generation of L1 Grids of Coincident AIRS, AMSU and MODIS Satellite Sounding Radiance Data for Climate Studies

Authors: Milton Halem, Yelena Yesha, Curt Tilmes, David Chapman, Mitch Goldberg, and Lihang Zhou

Journal: EOS Transactions

Date: June 19, 2007

Abstract: Three decades of Earth remote sensing from NASA, NOAA and DOD operational and research satellites carrying successive generations of improved atmospheric sounder instruments have resulted in petabytes of radiance data with varying spatial and spectral resolutions being stored at different data archives in various data formats by the respective agencies. This evolution of sounders and the diversities of these archived data sets have led to data processing obstacles limiting the science community from readily accessing and analyzing such long-term climate data records. We address this problem by the development of a web based Service Oriented Atmospheric Radiance (SOAR) system built on the SOA paradigm that makes it practical for the science community to dynamically access, manipulate and generate long term records of L1 pre-gridded sounding radiances of coincident multi-sensor data for regions specified according to user chosen criteria. SOAR employs a modification of the standard Client Server interactions that allows users to represent themselves directly to the Process Server through their own web browsers. The browser uses AJAX to request Javascript libraries and DHTML interfaces that define the possible client interactions and communicates the SOAP messages to the Process server allowing for dynamic web dialogs with the user to take place on the fly. The Process Server is also connected to an underlying high performance compute cluster and storage system which provides much of the data processing capabilities required to service the client requests. The compute cluster employs optical communications to NOAA and NASA for accessing the data and under the governance of the Process Server invokes algorithms for on-demand spatial, temporal, and spectral gridding. Scientists can choose from a variety of statistical averaging techniques for compositing satellite observed sounder radiances from the AIRS, AMSU or MODIS instruments to form spatial-temporal grids for their respective studies. A range of scientific visualization and animation services are also provided for viewing the results of the user specified service requests. Results of gridding, visualization and animating services for compositing and convolving the AIRS and MODIS spectral sounding radiances will be presented. In addition, demonstrations of SOAR on demand visualizations and animations for subsetting multi-year high-resolution multi-instrument pre-gridded radiance fields will be presented.

Type: Article

Publisher: American Geophysical Union

Number: 25

Volume: 88

Google Scholar: search