Bayesian calibration of a flood inundation model using spatial data

Abstract

[1] Bayesian theory of model calibration provides a coherent framework for distinguishing and encoding multiple sources of uncertainty in probabilistic predictions of flooding. This paper demonstrates the use of a Bayesian approach to computer model calibration, where the calibration data are in the form of spatial observations of flood extent. The Bayesian procedure involves generating posterior distributions of the flood model calibration parameters and observation error, as well as a Gaussian model inadequacy function, which represents the discrepancy between the best model predictions and reality. The approach is first illustrated with a simple didactic example and is then applied to a flood model of a reach of the river Thames in the UK. A predictive spatial distribution of flooding is generated for a flood of given severity.

Reproducing an extreme flood with uncertain post-event information

“…Bayesian techniques have been commonly applied in hydraulic and hydrological modelling (e.g. Hall et al, 2011;Renard et al, 2008) and can be used within a global sensitivity analysis (see summaries by Iooss andLemaître, 2015, andSarrazin et al, 2016) to assess the effect of each source of uncertainty on the output (e.g. Abily et al, 2016).…”

Enhancing river model set-up for 2-D dynamic flood modelling

“…In particular, uncertainty concerns the hydrological data, the hydraulic parameters, calibration and validation data, the governing equations describing the physical processes, the way to take into account man-made structures interacting with the flow and so on (among the most recent ones see Merwade et al, 2008b;Di Baldassarre and Montanari, 2009;Bales and Wagner, 2009;Di Baldassarre et al, 2010;Hall et al, 2011;Stephens et al, 2012;Warmink et al, 2011;Brandimarte and Kebede Woldeyes, 2013;Grimaldi et al, 2013;Domeneghetti et al, 2013;Dottori et al, 2013;Jung and Merwade, 2015).…”

Epistemic uncertainties and natural hazard risk assessment. 1. A review of different natural hazard areas

“…In some cases, historical flood mapping is available that can be used to condition hydraulic models of inundation and constrain the uncertainty in model predictions (Bates et al, 2014). Both Generalised Likelihood Uncertainty Estimation (GLUE; Aronica 30 et al, 1998;Romanowicz and Beven, 2003;Pappenberger et al, 2007; and more formal Bayesian methods (Romanowicz et al, 1996;Hall et al, 2011) have been used in this type of conditioning process (e.g. Figure 2; see also other examples in but the results will depend on how the Nat.…”