Validation of Special Sensor Microwave/Imager Ocean Surface Wind Retrievals in Equatorial Regions.
Abstract
The Fleet Numerical Meteorology and Oceanography Center (FNMOC) has the charter to provide Special Sensor Microwave/Imager(SSM/I) data to DOD and NOAA users. This tasking has led to new methods for processing SSM/I data being developed to improve NAVY SSM/I products, in particular the ability to remotely sense ocean surface winds. Currently, alternative SSM/I ocean surface wind speed algorithms include 'physical' or 'statistical' methods. Typically "physical" retrievals require additional data. e.g., cloud liquid water, along with SSM/I brightness temperatures while statistical methods are stand alone algorithms based on brightness temperature only. In this study four candidate wind retrieval methods proposed at the SSM/I Algorithm Symposium (June 1993) for implementation at FNMOC are examined. Limitations of the SSM/I calibration/validation data set to the mid-latitude region prompted the requirement to develop a tropical data set for evaluation of alternative algorithms. Comparison of SSM/I wind retrieval methods reveal neural networks display a high wind speed bias for winds above 11 m/s and a low wind speed bias for winds below 3 mi. The current FNMOC operational algorithm may overestimates wind speeds when water vapor is greater than 50 kg/m2. Partitioning of SSM/l retrieved wind speeds according to accuracy is by accomplished when using brightness temperature received at 37 GHZ. (MM)
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 1994
- Accession Number
- ADA293750
Entities
People
- Elton G. Sayward
Organizations
- Naval Postgraduate School