Statistical emulation of a tsunami model for sensitivity analysis and uncertainty quantification

Abstract

Abstract. Due to the catastrophic consequences of tsunamis, early warnings need to be issued quickly in order to mitigate the hazard. Additionally, there is a need to represent the uncertainty in the predictions of tsunami characteristics corresponding to the uncertain trigger features (e.g. either position, shape and speed of a landslide, or sea floor deformation associated with an earthquake). Unfortunately, computer models are expensive to run. This leads to significant delays in predictions and makes the uncertainty quantification impractical. Statistical emulators run almost instantaneously and may represent well the outputs of the computer model. In this paper, we use the outer product emulator to build a fast statistical surrogate of a landslide-generated tsunami computer model. This Bayesian framework enables us to build the emulator by combining prior knowledge of the computer model properties with a few carefully chosen model evaluations. The good performance of the emulator is validated using the leave-one-out method.

The influence of landslide shape and continental shelf on landslide generated tsunamis along a plane beach

“…The fluid domain in physical coordinates; η and σ are respectively the slide maximum vertical thickness and characteristic horizontal length, depending on the shape of the landslide. Sarri et al (2012).…”

The hydrodynamics of landslide tsunamis: current analytical models and future research directions

“…Here is where analytical models prove their importance. Based on the analytical model of Sammarco and Renzi (2008), Sarri et al (2012) built a fast statistical emulator which enables one to obtain tsunami estimates almost instantaneously. The Sarri et al (2012) emulator is capable of producing 1000 tsunami runup evaluations in as little as 83.9 seconds, providing accurate results quickly in situations where early warnings are necessary.…”

“…Based on the analytical model of Sammarco and Renzi (2008), Sarri et al (2012) built a fast statistical emulator which enables one to obtain tsunami estimates almost instantaneously. The Sarri et al (2012) emulator is capable of producing 1000 tsunami runup evaluations in as little as 83.9 seconds, providing accurate results quickly in situations where early warnings are necessary. From a modelling point of view, a limitation of the analytical models of Sammarco and Renzi (2008), Renzi and Sammarco (2010) and Renzi and Sammarco (2012) is the absence of bottom friction and slide deformation.…”

“…Indeed, modelling landslide tsunamis is intrinsically challenging as it involves a time-dependent bottom deformation in the same region where the waves are generated and propagate. In this paper, we shall review the analytical models of Sammarco and Renzi (2008), Renzi and Sammarco (2010) and Renzi and Sammarco (2012), which have become benchmark models for numerical codes (Romano et al 2013;Dias et al 2014;Stefanakis et al 2014) and statistical emulations (Sarri et al 2012) of landslide tsunamis. Analytical models provide an invaluable tool in the science of tsunami modelling, for they allow one to obtain a sound physical understanding of the wave dynamics and to carry out parametric investigations on the influence of the system main parameters on the maximum wave runup.…”

Multi-level emulation of a volcanic ash transport and dispersion model to quantify sensitivity to uncertain parameters

“…This enables the quantification of the impact of each simulator parameter on the prediction of the dispersion of volcanic ash. This approach has been used successfully in tsunami modelling (Sarri et al, 2012), simulating convective cloud (Johnson et al, 2015), aerosol modelling (Lee et al, 2011(Lee et al, , 2012(Lee et al, , 2013, galaxy formation and regional climate projections (Harris et al, 2010). In this study, the impact of the various simulator parameters can be assessed by their coefficients within the emulator.…”