Chen, Jiaming: Deriving surface water level and discharge from improved nadir- and early SWOT altimetry observations. - Bonn, 2026. - Dissertation, Rheinische Friedrich-Wilhelms-Universität Bonn.
Online-Ausgabe in bonndoc: https://nbn-resolving.org/urn:nbn:de:hbz:5-91300
@phdthesis{handle:20.500.11811/14296,
urn: https://nbn-resolving.org/urn:nbn:de:hbz:5-91300,
doi: https://doi.org/10.48565/bonndoc-915,
author = {{Jiaming Chen}},
title = {Deriving surface water level and discharge from improved nadir- and early SWOT altimetry observations},
school = {Rheinische Friedrich-Wilhelms-Universität Bonn},
year = 2026,
month = jul,

note = {Reliable observations of river water levels and discharge are essential for water security, flood and drought risk management, and inland navigation. However, in-situ gauge networks are often sparse, unevenly distributed, and affected by delays in data availability. In particular, the declining number of terrestrial gauge measurements of water level and river discharge in main repositories, such as the Global Runoff data Center (GRDC) poses a major problem. At the same time, increasing hydrological variability is creating demand for observational systems that are robust and scalable across political borders and remote basins.
Satellite radar altimetry has evolved from its origins in oceanography into an important technique for monitoring inland water levels. Advances in Synthetic Aperture Radar (SAR) modes, including Delay-Doppler and Fully Focused SAR (FFSAR) processing, have significantly improved along-track resolution and measurement quality over hydrologically complex environments such as narrow and braided rivers. These techniques are now operationally implemented on missions including Sentinel-3 and Sentinel-6. Furthermore, the SWOT mission introduces wide-swath interferometric techniques, enabling two-dimensional mapping of water elevation, extent, and slope. These observations offer unprecedented potential for consistent, high-frequency monitoring of river systems from regional to global scales.
This thesis has two primary objectives. First, it focuses on the development, implementation, and assessment of SAR altimetry processing methods tailored to inland water bodies for the accurate estimation of water surface elevation and time-varying river slope. Second, it develops and evaluates a joint framework for estimating river discharge by combining multi-altimetry data from both nadir altimetry and SWOT, Sentinel-1 imagery, and supplementary information such as prior mean discharge or bathymetry. This integrated approach exploits complementary spatio-temporal sampling to advance hydrological monitoring across diverse river systems.
For the first objective, a refined frequency-domain Omega–Kappa FFSAR Level-1B processor is developed and implemented for Sentinel-3A, Sentinel-3B, and Sentinel-6A. The resulting FFSAR waveforms are then processed using the SAMOSA+ retracker, and the water level estimates are further improved by considering prior water masks and a constant slope derived from SWOT River Database (SWORD). In addition, an automated off-nadir processing method is introduced within the FFSAR radargram, enabling the retrieval of longitudinal water surface profiles and time-varying slopes in the cross-track direction beyond the limitations of nadir altimetry, thereby enabling spatiotemporal mapping of river profiles using nadir altimeters. Validation against in-situ data indicates that water levels derived from FFSAR exhibit improved accuracy compared to Hydroweb products, with off-nadir processing further reducing standard deviation of difference (STDD) to below 10 cm. Time-varying slope estimates achieve an accuracy of better than 1 cm/km.
As part of the second objective, four flow law parameter estimations (FLPEs) algorithms (Metro-Man, Momma, NeoBAM, and SIC4DVar) are evaluated under consistent preprocessing conditions to assess their performance in estimating unconstrained discharge from SWOT observations over European rivers. The biases are analyzed in relation to reach-scale depth characterization and Manning roughness parameterization. Furthermore, three complementary nadir-altimetry-based discharge estimation methods are developed to extend the temporal coverage of SWOT: (1) Manning-equation-based estimation combining FFSAR water levels with prior bathymetry; (2) rating curves calibrated using concurrent SWOT stage-discharge pairs and applied to FFSAR water levels; and (3) off-nadir discharge retrieval within the Confluence framework by integrating FFSAR river profiles and Sentinel-1-derived widths. Collectively, these approaches establish a consistent, multi-sensor discharge ensemble that advances reach-scale discharge estimation and improves applicability in sparsely gauged and ungauged basins.},

url = {https://hdl.handle.net/20.500.11811/14296}
}

The following license files are associated with this item:

InCopyright