Sea surface temperature has a long history of measurement and is intimately linked with climate change. Data from the National Oceanic and Atmospheric Administration's (NOAA) polar orbiting satellites have been used to attempt global mapping of sea surface temperature for the past 17 years. Remotely sensed sea surface temperature retrievals became considerably more accurate following the launch of the AVHRR in 1978. Multichannel sea surface temperatures (MCSST) have been computed operationally from AVHRR, by NOAA's National Environmental Satellite, Data and Information Service (NESDIS), since late 1981 (McClain et at., 1985). Although the coefficients in the algorithms used to derive MCSST change with each new spacecraft, they are tuned with respect to global sets of close (temporal and spatial) match-ups with drifting buoys, thus improving continuity from spacecraft to spacecraft. Procedures are applied to remove atmospheric effects (Griggs, 1985) and to verify cloud-free conditions (McClain et at., 1985). The MCSST data set on this set of CD-ROMs was produced from the NOAA NESDIS MCSST retrievals, by the University of Miami, School of Marine and Atmospheric Sciences (UM/RSMAS) and by the Physical Oceanography Distributed Active Archive Center (PO.DAAC) at the Jet Propulsion Laboratory (JPL). The MCSST values (daytime and nighttime are binned separately) were binned into a global, equal area grid which is 1024 lines by 2048 samples. For each grid point, the average of all MCSST measurements for one week was computed. Open areas were interpolated using an interative Laplacian relaxation technique. The JPL PO.DAAC computed monthly averaged MCSST fields from the UM/RSMAS weekly fields, and them converted the fields to Hierarchical Data Format (HDF) to allow for portability of the CD-ROM set between different operating systems.