Validation of microwave sea surface temperature measurements for climate purposes

Abstract

A comparison is provided between recent SST observations obtained from the TRMM Microwave Imager (TMI) with the commonly used Reynolds and Smith SST analysis and in situ data that highlights the significant value added by the microwave SST observations above what can be obtained from infrared technology. The largest benefit of the microwave technology clearly comes from the unprecedented near all-weather sampling of ocean phenomena that yields measurements of ocean SST without the heavy smoothing in space and time that is traditionally being applied to close data gaps & resulting from clouds. This comparison of the TMI fields with the Reynolds and Smith analysis results in a mean offset of 0.l8°C and a standard deviation (STD) difference of 0.54°C over the latitude range ±36°. Regionally STD differences are found between both SST fields that reach 1°C or more, a number that is significantly larger than the error specification of either dataset alone. STD differences are obtained between Reynolds and Smith and in situ data of 0.55°C as compared to only 0.45°C from the TMI fields. Moreover, those uncertainties are time dependent and spatially varying, revealing a clear seasonal cycle with amplitudes varying by 0.35°C in the Reynolds and Smith SST fields around 25° latitude on the annual cycle in both hemispheres. Both TMI and the Reynolds and Smith analyses show temporal trends relative to in situ observations that differ in sign and amplitude.