Marine gravity anomaly from satellite altimetry: a comparison of methods over shallow waters

Abstract

Gravity anomalies over shallow waters are useful in many geodetic and geophysical applications. This work compares three methods of gravity anomaly derivation from altimetry over shallow waters near Taiwan: (1) compute gravity anomalies by LSC using along-track, differenced geoidal heights and height slopes, (2) compute gravity anomalies by least-squares collocation (LSC) using altimeter-derived along-track deflections of vertical (DOV), and (3) grid along-track deflections of vertical by LSC and then compute gravity anomalies by the inverse Vening Meinesz formula. A nonlinear filter with outlier rejection is applied to along-track data. We used altimeter data from Seasat, Geosat, ERS-1, ERS-2 and TOPEX/POSEDION missions. Retracked ERS-1 waveforms are shown to improve the accuracy of estimated gravity anomalies. For the three methods, the RMS differences between altimetry-derived gravity anomalies and shipborne gravity anomalies are 9.96 (differenced height) and 10.26 (height slope), 10.44 and 10.73 mgals, respectively. The RMS differences between shipborne gravity anomalies with gravity anomalies from retracked and non-retracked ERS-1 waveforms are 11.63 and 14.74 mgals, indicating retracking can improve the accuracy.