Air-sea turbulence statistics from synthetic aperture radar: An update

Abstract

A similarity theory based method for calculating turbulence statistics from synthetic aperture radar (SAR) derived 10 m neutral wind speed image sub-scenes was tested. We concentrated on the ability of the method to yield proper estimates of the ratio of reference height to Obukhov length (z/L) and kinematic buoyancy flux (B). The tropical ocean global atmosphere ‐ coupled ocean atmosphere response experiment (TOGA‐COARE) bulk flux algorithm, fed with buoy and mooring data corresponding in space and time to a particular sub-scene, provided a comparison data set. Thirteen case studies were examined covering a wide range of negative air‐sea temperature difference and wind speed. The statistics from the SAR method that we report herein are averages from a 3 × 3 sub-scene grid centered on a particular in situ station for each case study. The correlation coefficient for z/L is 0.77 and that for B is 0.95. The value of z/L produced by the SAR method, averaged over all 13 case studies, is ‐0.15. The standard deviation of z/L from the SAR method, averaged over all 13 case studies, is 0.12. The corresponding values for B are 0.046 m·s‐1K (average) and 0.034 m·s‐1K (standard deviation). The mean difference and root mean square difference (COARE results ‐ SAR results) for the z/L comparisons are ‐0.008 and 0.119, respectively, and the same statistics for the B comparisons are 0.030 m·s‐1K and 0.039 m·s‐1K. Potential sources of error and avenues for future research are briefly discussed.