Obtaining Accurate Sea Surface Skin Temperature from Multiple Data Sources during the CASPER-East Field Campaign

Accurate sea surface temperatures (SSTs), also referred to as the sea surface skin temperature (SSST), are crucial for marine atmospheric surface layer (MASL) and air sea interaction studies. Uncertainties in SST result in erroneous estimates of heat and water vapor exchange across the air sea interface as well as misinterpretation of surface evaporation duct properties. Thus, it is fundamentally important to obtain well-calibrated SSTs for the Coupled Air Sea Processes and Electromagnetic ducting Research (CASPER) project. SSTs are not always readily available and most vessels measure bulk water temperatures from the ship intake at one to several meters below the ocean surface. Because of the presence of two physical processes in the upper ocean: (1) the cool skin effect caused by the combined cooling effects of net longwave radiation, the sensible heat and latent heat flux at the air-sea interface and (2) the warm layer effect as a result of the solar radiation flux convergence in the top layers of the ocean, even the most accurate subsurface temperature measured by conventional contact thermometers at intake depth, normally referred to as the bulk water temperature, cannot provide accurate SST. It is therefore necessary to derive the surface skin temperature from the available bulk water temperature measurements when the bulk temperature is the only source of temperature near the air-sea interface. During CASPER-East field campaign (offshore Duck, NC, October-November 2015), an absolute radiometric skin temperature probe, the integrated infrared SST autonomous radiometer (ISAR) was deployed on the R/V Sharp. The ISAR is a self-calibrating instrument capable of measuring in situ SSST to an accuracy of 0.1 K. We used these SSST data to make corrections to the bulk SST (BSST) from the ship's intake at 1 m below the waterline. This dataset was most complete with minimum missing data, compared to SSST from ISAR. BSST was corrected based on the SSST through a two-step correction method. This comparison revealed a wind speed dependence of the difference in SST (BSST- SSST; ΔSST) in two wind speed regimes. At low winds (< 4 m/s), ΔSST also shows diurnal variation. We first separated ΔSST values based on wind speed as low wind (< 4 m/s) and high wind (> 4 m/s) cases. A quadratic fit is found to be optimum for the low wind cases and a linear fit worked better for high wind cases. Corrections are done for BSST using these wind speed dependent functions. Diurnal corrections are done for low wind cases only by fitting a 4th degree polynomial function to the ΔSST. Comparisons show the corrected BSST matches well with ISAR measured SSST. Further assessment of the cool skin/warm layer effect is also performed using the method embedded in the COARE surface flux parameterization algorithm.