Author(s): Ignacio Ojer-Garcia; Adrian Navas-Montilla; Sergio Martinez-Aranda; Pilar Garcia-Navarro
Linked Author(s): Pilar García-Navarro
Keywords: Remote sensing; Sentinel; Machine learning; Model calibration; Fluvial simulation
Abstract: This work explores the idea to parameterize two-dimensional hydrodynamic models (2D-SWE) using satellite imagery as a primary data source rather than for post-simulation validation only. Since model accuracy strongly depends on input parameters such as Manning’s roughness coefficient, the study states the premise of this coefficient being formulated as a function of Sentinel spectral bands. This hypothetical dependence is evaluated through a simplified test case and the use of Machine Learning techniques for its calibration process.
Year: 2026