Flood Inundation Modeling Using the Surface Water Ocean Topography Mission
- 1School of Geographical Sciences University of Bristol Bristol UK
- 2Fathom Clifton Heights Bristol UK
- 3School of Mathematics University of Bristol Bristol UK
Abstract
Flood risk mapping requires detailed river bathymetry, yet these data are unavailable for most rivers. We propose a new algorithm to estimate river bathymetry from multitemporal observations of water height by the Surface Water and Ocean Topography (SWOT) mission. The robustness of bathymetry estimates were evaluated against detailed in situ data on a narrow river (40–70 m width) that is often below SWOT's mission specification. Typical SWOT errors, not always flagged by the SWOT processing algorithm, could severely impact bathymetry estimates. However, a simple “Tulip” shaped misfit function could largely mitigate for these and was especially important along narrow sections of the river. SWOT River Single Pass node (∼200 m reaches) height errors against in situ gauge levels with stringent quality filtering were 0.16 m, approaching the mission accuracy requirements for much larger 1 km 2 waterbodies. Derived channel bed elevations were physically plausible (∼1 m error) when estimated at a spatial resolution below half the river backwater length. A flood model built from SWOT‐derived bathymetry had commensurate accuracy to a model based on high‐resolution sonar bathymetry. For flooding in 2000, the model had a Critical Success Index of 0.88 against airborne SAR and absolute differences from gauged levels ranging from 0.14 to 0.31 m across the event.
Related
- Evaluating the performance of the surface water and ocean topography (SWOT) mission for monitoring reservoirs across the conterminous United States
- Coupled Physics-Based and Machine Learning Streamflow Modeling with High-Resolution Flood Inundation and Impact Assessment for Hurricane Florence (2018): A Case Study of the Neuse River at Goldsboro, North Carolina
- Threshold Sensitivity and Reference Uncertainty in Sentinel-1 Inundation Mapping: The July 2021 Mahad Flood
- Hydrologically enhanced machine learning framework for urban flood inundation mapping using multi-sensor remote sensing data: a case study of Mumbai, India
- WarpFlow v1.0: A Single-Source CPU/GPU Shallow-Water Solver for Operational City-Scale Urban Flood Modeling
BibTeX
@article{neal2026flood,
title = {{Flood Inundation Modeling Using the Surface Water Ocean Topography Mission}},
author = {Jeffrey Charles Neal and Stephen Chuter and Leanne Archer and Laurence Paul Hawker and Paul Bates and James Thomas Steven Savage and Hamish Wilkinson and J Rougier},
journal = {Water Resources Research},
year = {2026},
month = jul,
doi = {10.1029/2025wr041527},
url = {https://doi.org/10.1029/2025wr041527},
note = {CC BY},
}
RIS
TY - JOUR
TI - Flood Inundation Modeling Using the Surface Water Ocean Topography Mission
AU - Jeffrey Charles Neal
AU - Stephen Chuter
AU - Leanne Archer
AU - Laurence Paul Hawker
AU - Paul Bates
AU - James Thomas Steven Savage
AU - Hamish Wilkinson
AU - J Rougier
PY - 2026
DA - 2026/07/01
JO - Water Resources Research
DO - 10.1029/2025wr041527
UR - https://doi.org/10.1029/2025wr041527
AB - Flood risk mapping requires detailed river bathymetry, yet these data are unavailable for most rivers. We propose a new algorithm to estimate river bathymetry from multitemporal observations of water height by the Surface Water and Ocean Topography (SWOT) mission. The robustness of bathymetry estimates were evaluated against detailed in situ data on a narrow river (40–70 m width) that is often below SWOT's mission specification. Typical SWOT errors, not always flagged by the SWOT processing algorithm, could severely impact bathymetry estimates. However, a simple “Tulip” shaped misfit function could largely mitigate for these and was especially important along narrow sections of the river. SWOT River Single Pass node (∼200 m reaches) height errors against in situ gauge levels with stringent quality filtering were 0.16 m, approaching the mission accuracy requirements for much larger 1 km 2 waterbodies. Derived channel bed elevations were physically plausible (∼1 m error) when estimated at a spatial resolution below half the river backwater length. A flood model built from SWOT‐derived bathymetry had commensurate accuracy to a model based on high‐resolution sonar bathymetry. For flooding in 2000, the model had a Critical Success Index of 0.88 against airborne SAR and absolute differences from gauged levels ranging from 0.14 to 0.31 m across the event.
SN - 0043-1397
SN - 1944-7973
ER -