RiwiSAR-SWH: A data-driven method for estimating significant wave height using Sentinel-3 SAR altimetry
Event: 2022 Ocean Surface Topography Science Team Meeting
Session: Coastal Altimetry
Presentation type: Type Poster
Contribution: PDF file
More than 600 million people (about 10% of the world's population) live in coastal areas that are less than 10 m above sea level. Despite the urgent need to monitor coastal waters, in-situ measuring stations including wave buoys around the world do not provide sufficient insight into coastal water level variations, and in particular, they cannot provide sufficient information on one of the essential properties of water surfaces, namely the Significant Wave Height (SWH). Satellite altimetry plays an increasingly important role, especially after operating in Synthetic Aperture Radar (SAR) mode. However, due to the complexity of the coastal water surfaces, the performance of the satellite altimeters over the coastal area falls behind the open ocean surfaces. In addition, the well-known direct relationship between waveform rise time and SWH does not hold for SAR waveforms due to a different processing scheme. This study proposes a data-driven method to determine SWH using the Sentinel-3 data for both oceanic and coastal zones. For this purpose, we propose a method based on the rise time and the width of a waveform, called RiwiSAR-SWH (rise time width model for SAR-SWH), which is free from the complexity of the SAR physical model and estimates SWH over the coastal area and open ocean in a relatively straightforward manner. We have employed our method over different regions in the coastal zone of the North Sea. The results are validated against in-situ buoy data and compared with SWH estimates from SAMOSA+, SAMOSA++ and the Sentinel-3 Ocean retracker. The validation shows that the proposed method can determine SWH with accuracy ranging from 0.25 m to 0.91 m for different locations in the North Sea. Moreover, we obtain reliable SWH to within 1 km from the coast, which is an improvement of more than 40% compared to existing methodologies.