Enhanced Sea Surface Height Estimation With Interference Rejection Using High-Frequency Fully Focused SAR Altimetry Data Over Island Areas

The fully focused synthetic aperture radar (FFSAR) altimetry technology has greatly enhanced the accuracy of coastal sea surface height (SSH). However, due to its pulse-limited mode in the cross-track direction, FFSAR is still susceptible to land interference. Additionally, the spatiotemporal resolution of in situ data limits the evaluation of altimetric SSH in both coastal and open ocean regions. This study proposes a waveform interference detection and removal algorithm (WIDA), which utilizes waveform differencing and azimuthal constant false alarm rate (CFAR) techniques to identify and reject interference signals in island areas. Tide gauge data, particularly a seamless and high-resolution airborne gravimetric quasi-geoid (QG) model, are employed to validate the performance of WIDA, providing a new perspective for the continuous and consistent evaluation of SSH profiles. Validation of five years of data from six Sentinel-3A/B tracks over the Xisha Islands in the South China Sea demonstrates that FFSAR waveforms retracked by WIDA (FFSAR_WIDA), with 80-Hz sampling, significantly outperform FFSAR_coastal retracker for SAR altimetry (CORAL), unfocused SAR (UFSAR), and UFSAR_WIDA within 5 km of the tide gauge station. The standard deviations (SDs) for FFSAR_WIDA, FFSAR_CORAL, UFSAR, and UFSAR_WIDA are 0.153, 0.192, 0.367, and 0.173 m, respectively. Moreover, the evaluation with the QG also reveals a marked improvement over FFSAR_CORAL and UFSAR within 5 km from the islands. The spatial distribution assessment shows that as the distance from the island decreases, the SD for FFSAR_CORAL (UFSAR) increases from 0.010 m (0.011 m) at 15 km to 0.061 m (0.089 m) at 3 km, while FFSAR_WIDA (UFSAR_WIDA) remains consistently below 0.011 m (0.014 m).