Evaluation of Retracking Algorithms over China and Adjacent Coastal Seas

Abstract

A major limitation of conventional altimetry is that most data in coastal seas are flagged as useless partly because of land contamination of altimeter return waveforms. The land effect on waveforms is analyzed at different coast terrains, shoreline shapes, and when the altimeter is approaching and leaving the land. Retracking algorithms (Ocean, Ice-2, OCOG (offset center of gravity), Threshold, and Beta5) are implemented and tested using Jason1 waveforms for 1 year (March 2006 to February 2007; cycle 155 to cycle 188) around China and adjacent coastal seas (14–45°N, 105–130°E). We use three methods to compare the retracking algorithms. First, the bias, root mean square, and standard deviation of sea level anomaly differences between ascending and descending passes are calculated at crossover points; Second, in situ sea surface height measurements from tide gauge stations are used to compare with the retracked altimetry sea surface height. Finally, in situ and altimetry significant wave height measurements are compared. We found that in coastal seas the OCOG algorithm provides more accurate results than others.