Stochastic generation of spatially coherent river discharge peaks for continental event-based flood risk assessment

Dirk Diederen,Sergiy Vorogushyn,Ben Gouldby,Ferdinand Diermanse,Ye Liu

doi:10.5194/nhess-19-1041-2019

Abstract

Abstract. We present a new method to generate spatially coherent river discharge peaks over multiple river basins, which can be used for continental event-based probabilistic flood risk assessment. We first extract extreme events from river discharge time series data over a large set of locations by applying new peak identification and peak-matching methods. Then we describe these events using the discharge peak at each location while accounting for the fact that the events do not affect all locations. Lastly we fit the state-of-the-art multivariate extreme value distribution to the discharge peaks and generate from the fitted model a large catalogue of spatially coherent synthetic event descriptors. We demonstrate the capability of this approach in capturing the statistical dependence over all considered locations. We also discuss the limitations of this approach and investigate the sensitivity of the outcome to various model parameters.

Highlights

Flood events cause a large amount of damage worldwide (Desai et al, 2015)
For Flood risk assessments (FRAs), a chain of models is applied, covering the entire risk cascade from hazardous extreme events down to flood damage or casualties resulting from inundation
We applied a new method of dynamic event identification where we aimed to capture discharge events in each major European river basin, after which we used a block-based time window method to merge them to spatially coherent, pan-European events

Summary

Introduction

Flood events cause a large amount of damage worldwide (Desai et al, 2015). Following the definition of risk (Field, 2012), put as the probability of damage, FRA requires an approximation of the risk curve under stationary climate conditions and a current distribution of asset values. For FRAs, a chain of models is applied, covering the entire risk cascade from hazardous extreme events down to flood damage or casualties resulting from inundation (e.g. expected annual damage, loss of life). The risk curve represents the probability of damage and is approximated by the evaluation of a comprehensive catalogue of hazard scenarios. To drive the chain of models, boundary forcing is required This typically comprises a large catalogue of synthetic forcing data, with models conditioned on observations

Objectives

Methods

Findings

Discussion

Conclusion