Sea surface temperature trends from satellite and in situ data

Monitoring long-term changes in sea surface temperature (SST) requires a well-defined global SST climatology and careful generation and averaging of SST anomalies. In this study, an SST climatology based on Pathfinder Advanced Very High Resolution Radiometer (AVHRR) data has been developed along with a method to evaluate its usefulness in climate change studies. The method of evaluation is based on a statistical comparison between the climatology and individual in situ observations from both the 1994 World Ocean Atlas (WOA94) and the Comprehensive Ocean Atmosphere Data Set (COADS). The SST climatology was created at a resolution of 9.28 km using both day and night satellite fields generated with the Version 1 Pathfinder algorithm and cloud-masking procedures, plus an erosion filter that provides additional cloud-masking in the vicinity of cloud-edges. Using the statistical comparison method, this satellite climatology is found to generally better limit noise in SST anomalies than traditional in situ climatologies, except in the tropics where the Pathfinder algorithm faces difficulties arising from high levels of water vapor, atmospheric aerosols, and persistent cloudiness. The Pathfinder SST climatology and individual WOA94 and COADS anomalies are used to characterize global and regional changes in ocean surface temperature since 1942. Anomaly trends are computed using a traditional five-degree latitude-longitude binning technique as well as a new method which uses climatological temperature classes to group the individual anomalies. Linear trends in the data-rich period between 1960 and 1990 calculated separately from these two independent data sets confirm that global SST warmed at a rate of approximately 0.10$\sp\circ$C decade$\sp{-1}$ in that period. The results of this study also verify regional trends such as cooling in the North Atlantic and generally larger warming in the Southern Hemisphere. This study also reveals that trends in the recent, well-sampled period are insensitive to the averaging technique, but that significant differences are found when the trends are extended further back in time into the more data-sparse periods. Regardless of the binning technique or SST data set used, however, all global SST warming trends are found to equal or exceed the results of previous studies