Abstract's details
Water level retrieval from Delay Doppler altimetry
CoAuthors
Event: 2019 Ocean Surface Topography Science Team Meeting
Session: Science IV: Altimetry for Cryosphere and Hydrology
Presentation type: Type Poster
Contribution: not provided
Abstract:
River discharge is a key variable to quantify the water cycle, its flux and stocks at different scales. Water level from classical and Delay Doppler altimeter measurements allow the computation of river discharge, once additional information and/or assumptions are made. Radar altimetry will become an imaging sensor in the SWOT mission, which is expected to significantly improve the spatial mapping and provide directly water level, river width and surface slope.
It has been shown that CA altimetry monitors about 15% of the global lake volume variation and large rivers with the accuracy being increased with the development of new sensors and retrackers. Root mean square errors (RMSE) range between a few cm and 70 cm today for large rivers, depending on the CA altimeter, lake area and shape, surroundings and retracking method.
In this paper, we revise these limits for Sentinel-3A data in the Rhine and Po European rivers, where simultaneous in-situ data are available for validations. The performances of standard products available within Copernicus (from OCOG and SAMOSA2 retrackers) are evaluated and compared to the GPOD ESA products (from the SAMOSA+ retracker and from the new SAMOSA++ retracker).
The results show a good agreement with in situ measurements for the Rhine and Po rivers, of width about 300 meters, with a root-mean-square errors (RMSEs) between 0.10 m and 0.30 m. Results are a clear improvement with respect to standard Jason-3 data and to values obtained from CryoSat-2. The GPOD results are also superior to the Copernicus Sentinel-3A results at all stations. We conclude that the Delay Doppler altimetry mode on Sentinel-3A improves the estimation of river heights on medium size rivers, especially when enhanced retrackers are applied. The new virtual gauges are further used to derive river discharge at these locations.
It has been shown that CA altimetry monitors about 15% of the global lake volume variation and large rivers with the accuracy being increased with the development of new sensors and retrackers. Root mean square errors (RMSE) range between a few cm and 70 cm today for large rivers, depending on the CA altimeter, lake area and shape, surroundings and retracking method.
In this paper, we revise these limits for Sentinel-3A data in the Rhine and Po European rivers, where simultaneous in-situ data are available for validations. The performances of standard products available within Copernicus (from OCOG and SAMOSA2 retrackers) are evaluated and compared to the GPOD ESA products (from the SAMOSA+ retracker and from the new SAMOSA++ retracker).
The results show a good agreement with in situ measurements for the Rhine and Po rivers, of width about 300 meters, with a root-mean-square errors (RMSEs) between 0.10 m and 0.30 m. Results are a clear improvement with respect to standard Jason-3 data and to values obtained from CryoSat-2. The GPOD results are also superior to the Copernicus Sentinel-3A results at all stations. We conclude that the Delay Doppler altimetry mode on Sentinel-3A improves the estimation of river heights on medium size rivers, especially when enhanced retrackers are applied. The new virtual gauges are further used to derive river discharge at these locations.