Author(s): L. Cea; M. Bermudez; J. Puertas; I. Fraga; S. Coquerez

Linked Author(s): Luis Cea Gómez, María Bermúdez Pita, Jerónimo Puertas Agudo, Ignacio Fraga Cadórniga

Keywords: No keywords

Abstract: Two dimensional shallow water models have demonstrated good capabilities for flood inundation mapping in urban areas. However, even if High Performance Computing techniques have greatly decreased the computational time needed to run a 2D inundation model, this approach remains unsuitable for applications as real time forecasting or uncertainty propagation in a Monte Carlo context, which requires the evaluation of hundreds or thousands of model runs. For these applications there remains a need for fast urban inundation models. In this paper we propose and compare the application of different linear, non-linear and non-parametric regression techniques as surrogate models of the 2D shallow water equations (SWE) applied to flood inundation mapping in ungauged urban areas. A coastal urban area is used as a test case. The case is specially challenging since the spatial distribution of water depth in the study area depends on the flow discharge in three different tributaries as well as on the tidal level and range.

DOI:

Year: 2016