Comparison of in situ and satellite-derived (MODIS-Aqua/Terra) methods for assessing temperatures on coral reefs

Thermal stress has been regarded as one of the most important parameters monitored on reefs to assess coral health, and therefore, it is important to have accurate temperature data for reef environments. Whereas most studies of thermal stress on reefs have relied on sea surface temperature (SST) data, recent findings have suggested that subsurface temperatures may differ considerably from those at the surface. We compared concomitant in situ and satellite-derived SST measurements on two different coral reefs off southern Belize. Seawater temperatures adjacent to corals at 3 to 5 m depth were measured every 10 to 30 min, from June 2002 to December 2007. An accompanying data set of seawater temperatures was collected every 10 min at 1, 3, 6, 9 and 15 m depth, from October 2006 to December 2007. Concurrent nighttime and daytime SST measurements from MODIS Aqua and Terra platforms with a spatial resolution of 4 km were obtained for comparative purposes. Results indicate a negative (cool) bias for remotely sensed SSTs when compared with in situ measurements for both satellites across these two locations. Compared with daytime values, nighttime satellite-derived SST measurements yielded larger negative biases and were less correlated with in situ measurements. Understanding these biases will not only provide a better evaluation of the thermal regime on individual reefs, but will also create opportunities for more precise temperature comparisons among coral reef environments. We stress the importance of "sea-truthing" and complementing satellite-derived SST readings with in situ data when measuring temperatures on coral reefs.