Detecting surface geostrophic currents using wavelet filter from satellite geodesy

Abstract

According to the features of spatial spectrum of the dynamic ocean topography (DOT), wavelet filter is proposed to reduce short-wavelength and noise signals in DOT. The surface geostrophic currents calculated from the DOT models filtered by wavelet filter in global and Kuroshio regions show more detailed information than those from the DOT models filtered by Gaussian filter. Based on a satellite gravity field model (CG01C) and a gravity field model (EGM96), combining an altimetry-derived mean sea surface height model (KMSS04), two mean DOT models are estimated. The short-wavelength and noise signals of these two DOT models are removed by using wavelet filter, and the DOT models associated global mean surface geostrophic current fields are calculated separately. Comparison of the surface geostrophic currents from CG01C and EGM96 model in global, Kuroshio and equatorial Pacific regions with that from oceanography, and comparison of influences of the two gravity models errors on the precision of the surface geostrophic currents velocity show that the accuracy of CG01C model has been greatly improved over pre-existing models at long wavelengths. At large and middle scale, the surface geostrophic current from satellite gravity and satellite altimetry agrees well with that from oceanography, which indicates that ocean currents detected by satellite measurement have reached relatively high precision.