CREATION OF A PLATFORM-INDEPENDENT VERSION OF THE UW-CIMSS GEOSTATIONARY , HIGH-DENSITY WIND DERIVATION ALGORITHM

Timothy L. Olander,C. Velden,C. Spinoso
Abstract:In response to the widespread use of geostationary satellite-derived winds in the meteorological community, a portable, platform-independent version of the UW-CIMSS winds processing algorithm has been developed. The incentive for this development derives from the desire to allow flexibility to satellite wind processing and research centers around the world to be able to locally create site-specific data sets tailored to suit their forecast, NWP or research needs using the latest advances and techniques. The prototype version of the CIMSS winds algorithm must be executed within a McIDAS environment, utilizing specific McIDAS formatted satellite data and model forecast “guess” fields. The revised algorithm will run within either McIDAS or UNIX shell environments, allowing utilization and processing of other geostationary satellite data and model guess field formats in addition to McIDAS. This revised edition is currently being implemented and evaluated at the U.S. Naval Fleet Numerical Meteorology and Oceanography Center, and the Air Force Global Weather Center. Transfer of the new software to NOAA/NESDIS for operational testing and evaluation is also planned. The new CIMSS winds processing algorithm will soon be released and available to other interested data processing centers or research institutes. The updated algorithm includes the latest post-processing quality-control module that combines the objective-editor/recursive filter (RF) analysis developed at CIMSS with the quality indicator (QI) scheme developed by EUMETSAT. Future versions of the algorithm will continue to be advanced and supported by UW-CIMSS. Further expansion of I/O interfacing will allow easier portability to various operating systems utilizing a wide variety of input data sets. A generic X-Windows and/or IDL graphical display capability will be added to allow easier data entry, and visualization tools now supplied with the McIDAS version of the code.
What problem does this paper attempt to address?