Toward Improved Validation of Satellite Sea Surface Skin Temperature Measurements for Climate Research

A poor validation strategy will compromise the quality of satellite-derived sea surface temperature (SST) products because confidence limits cannot be quantified. This paper addresses the question of how to provide the best operational strategy to validate satellite-derived skin sea surface temperature (SST skin ) measurements. High quality in situ observations obtained using different state-of-the-art infrared radiometer systems are used to characterize the relationship between the SST skin , the subsurface SST at depth (SST depth ), and the surface wind speed. Data are presented for different oceans and seasons. These data indicate that above a wind speed of approximately 6 m s 1 the relationship between the SST skin and SST depth , is well characterized for both day-and nighttime conditions by a cool bias of 0.17 0.07 K rms. At lower wind speeds, stratification of the upperocean layers during the day may complicate the relationship, while at night a cooler skin is normally observed. Based on these observations, a long-term global satellite SST skin validation strategy is proposed. Emphasis is placed on the use of autonomous, ship-of-opportunity radiometer systems for areas characterized by prevailing low-wind speed conditions. For areas characterized by higher wind speed regimes, well-calibrated, qualitycontrolled, ship and buoy SST depth observations, corrected for a cool skin bias, should also be used. It is foreseen that SST depth data will provide the majority of in situ validation data required for operational satellite SST validation. We test the strategy using SST skin observations from the Along Track Scanning Radiometer, which are shown to be accurate to approximately 0.2 K in the tropical Pacific Ocean, and using measurements from the Advanced Very High Resolution Radiometer. We note that this strategy provides for robust retrospective calibration and validation of satellite SST data and a means to compare and compile in a meaningful and consistent fashion similar datasets. A better understanding of the spatial and temporal variability of thermal stratification of the upper-ocean layers during low-wind speed conditions is fundamental to improvements in SST validation and development of multisensor satellite SST products.