Abstract
The oceanic cool skin effect is characterized by a temperature depression between the ocean's submillimeter thermal skin layer (SSTskin) and the underlying water (SSTdepth). This study evaluates the ability of the widely used Fairall et al. (1996), https://doi.org/10.1029/95jc03190 model (F96) to simulate the cool skin effect using surface forcing data from the Modern-Era Retrospective analysis for Research and Applications, Version 2 (MERRA-2), validated against high-accuracy infrared radiometric measurements from the Marine-Atmospheric Emitted Radiance Interferometer (M-AERI) and the Infrared Sea Surface Temperature Autonomous Radiometer (ISAR) collected during research cruises spanning diverse oceanic regions. Results show that F96 model simulations driven by MERRA-2 inputs generally match the observed skin effects, with mean biases of -0.03 to -0.04 K and standard deviations similar to 0.1 K, attributed primarily to uncertainties in humidity, longwave radiative flux, and wind speed. A 24-year global analysis (2000-2023) revealed a mean cool skin effect of approximately -0.2 K, with pronounced spatial and seasonal variability. Strongest cooling (-0.3 K or more) occurs in regions such as western boundary currents during colder months, while the Southern Ocean shows much weaker skin effects (-0.1 K) due to persistent high wind speeds and reduced net heat loss. These results suggest that the commonly applied constant correction of -0.17 K in satellite SSTskin retrievals does not accurately represent the true magnitude and variability of the cool skin effect. A more accurate correction scheme incorporating latitude and seasonal dependence is recommended to improve the accuracy of satellite-derived SSTskin fields, and better support climate model applications.