The Improved Retrieval of Coastal Sea Surface Heights by Retracking Modified Radar Altimetry Waveforms

Measuring sea surface height (SSH) using satellite altimetry in coastal ( from coasts) and shallow water region has long been a challenge since the radar altimeter waveforms are often contaminated by complex coastal topography and do not conform to theoretical Brown waveform shapes. The land contamination or surface variation due to ocean dynamics induce spurious peaks in altimeter waveforms that deviate from Brown's theoretical model as the altimeter footprint approaches or leaves the shoreline. These spurious peaks should be mitigated to minimize the error in the determination of the leading edge and associated track offset in the waveform retracking process. Here, we introduce a novel algorithm to modify coastal waveforms (0.5-7 km from coasts, using 20 Hz altimetry data), thus improving coastal data coverage and accuracy. We apply our processing algorithm and use various retrackers to compare retrieved coastal SSHs in four study regions in North America, using both Envisat and Jason-2 altimetry. The retrieved altimetry data in the 1-7 km coastal zone indicate that the 20% Threshold retracker with modified waveform has a RMSE of 21 cm as compared with in situ tide gauge data, which corresponds to a 63% improvement in accuracy compared to the use of the original deep-ocean waveform retracker.