Geostationary sea surface temperature product validation and retrieval methodology

The National Oceanic and Atmospheric Administration's Office of Satellite Data Processing and Distribution are generating operational sea surface temperature (SST) retrievals from GOES-9, 10 and 12 satellite imagers. The direct regression based algorithms have been replaced by retrieval schemes based on radiative transfer modeling (RTM). The satellites are situated at longitudes 154.5 E, 135 W and 75 W respectively, thus allowing the acquisition of high temporal SST retrievals from 30 degrees west to 100 degrees east. Combined with data from the Meteosat Second Generation (MSG) SEVIRI instrument, there is now the capability to determine the diurnal cycle of SST throughout most of the world's oceans. This is an important step in the production of a global high resolution SST analysis that combines polar and geostationary observations.

The GOES-12 satellite imager presents a particular challenge because it has only two channels (3.9 and 11 µm) available to generate SSTs. The former is difficult to use during the day because of solar contributions to the signal that derive from surface reflection and atmospheric scattering. The current scheme for GOES-12 consists of: (1) a new cloud masking methodology based on a probabilistic (Bayesian) approach to detection by comparing thermal infrared and visible data with prior estimates clear-sky values derived from fast radiative transfer calculations using NWP data as input; (2) screening of areas of significant sunglint; and (3) adjustment of the 3.9 µm brightness temperature for solar contamination (from surface and atmospheric scattering) prior to SST retrieval. Early validation results indicate that daytime GOES-12 SST retrievals are comparable in accuracy to those from GOES-9 and 10.

The GOES-SST products generated from these algorithms include regional hourly and 3-hourly hemispheric imagery, 24 hour merged composites and a combined POES/GOES 10-km resolution demonstration SST analysis. Future plans include the merging of data from the EUMETSAT MSG satellite and an International Geostationary SST Workshop on Global Algorithm Development.