WindSat ocean vector wind retrievals are retrieved in near-real-time and operationally assimilated into the numerical weather prediction (NWP) global models of the U.S. Navy, the U.K. Met Office and NOAA NCEP. These retrievals are produced using the Naval Research Laboratory (NRL) retrieval algorithm. The retrieval output includes binary data quality flags and relative measures of retrieval accuracy such as chi-squared tests and retrieval error covariances. Currently users of WindSat data only use the data quality information to screen the data. Previous evaluations of the WindSat retrievals have also primarily used the data used the data quality information to screen the data.

Quality flags are provided in the WindSat data for several conditions that can degrade retrieval performance. These include rain, radio frequency interference, land or sea ice contamination and missing brightness temperature data. Relative measures of data quality based on land-water fraction, cloud liquid water path, retrieval error covariance and chi-squared tests can also be considered.

We will use wind analyses from NWP models and retrievals from other satellite data to investigate the correlation of WindSat retrieval performance with the data quality information provided in the WindSat data. The goal of this effort is to provide better guidance to WindSat data users. Improved utilization of the data quality information can increase the value of the WindSat data. This work will also be used to guide future development of the WindSat algorithms.