Intercomparison of satellite-derived SST with logger data in the Caribbean-Implications for coral reef monitoring

Since the early 1980s measurements of Sea Surface Temperature (SST) derived from satellite-borne instruments have provided a wide range of global gridded products documenting changes in SST. However, there are many sources of uncertainty in these records and significant differences exist among them. One use of these products is identification of coral bleaching events, and the predictions of the impact of future warming on coral reefs. This relies on an understanding of how temperatures near reefs as recorded by SST products differ from the in-situ SST experienced by the corals. This difference is a combination of real spatio-temporal variations, inadequate in product resolution and errors in the products. This paper investigates the relationship between the local temperature measured in-situ by loggers at coral sites in the western tropical Atlantic and two high resolution satellite SST products. Using differences among ESA SST CCI v2.1 (CCI analysis SST), NOAA CoralTemp SST products and in-situ logger data from coral reefs, an assessment of the satellite products with focus on coral reef monitoring is carried out. Discrepancies between the two products can be large, especially in coastal areas and for the warmest and coldest months when there is a particular risk of bleaching. By comparison to the stable CCI analysis SST product, CoralTemp was found to overestimate the rise in SST by as much as 0.20 degrees C per decade. In almost all cases SSTs from CCI analysis SST were more consistent with temperatures measured near the corals than those from CoralTemp.