Extracting Ocean Surface Currents from SAR: MCC and Doppler Centroid Tracking
暂无分享,去创建一个
σ 0 (d B ) MCC_SAR Currents – Filtered 20080417: 1804-1834 GMT The primary ocean backscattering mechanism for SAR is Bragg scattering through small-scale surface capillary waves. Ocean surface slicks cause damping of surface waves through “Marangoni damping”, which in turn damps the backscatter in SAR intensity images. C-band SAR instruments aboard Envisat (ASAR) and ERS2 (AMI-SAR) offer a unique possibility of observing temporal changes in sub-mesoscale to mesoscale ocean surface features due to 30-minute time-lag in their orbits during certain portions of their missions coincidence. The Maximum Cross-Correlation (MCC) method (Emery et al., 1986) is applied here to Envisat-ERS2 30-minute lag image pairs to generate ocean surface currents at 3 km resolution. The MCC_SAR currents are compared separately with CODAR current fields of 6 km and 2 km resolution (Kim et al., 2008). Some sample plots of re-gridded vectors are shown below, along with the histograms of residuals (u, v, and direction). As a preliminary statistical analysis, the residual histograms for both the 2 km and 6 km CODAR data are tending towards Gaussian, which is an indication that differences between CODAR and MCC_SAR currents are due to random errors. The means of the u and v residuals, however, are not zero and show a mean +ve value; this suggests that the MCC_SAR vectors show an overall larger magnitude than the CODAR vectors.
[1] Bertrand Chapron,et al. Retrieval of Sea Surface Range Velocities From Envisat ASAR Doppler Centroid Measurements , 2011, IEEE Transactions on Geoscience and Remote Sensing.
[2] Eric Terrill,et al. Mapping surface currents from HF radar radial velocity measurements using optimal interpolation , 2008 .
[3] W. Emery,et al. An objective method for computing advective surface velocities from sequential infrared satellite images , 1986 .