Assessing direct flood damages using open data in diverse urban environments

Paprotny, Dominik

Schröter, Kai

Bagli, Stefano

Castellarin, Attilio

Essenfelder, Arthur H.

Mazzoli, Paolo

Mediero, Luis

Morales-Nápoles, Oswaldo

Mysiak, Jaroslav

Kreibich, Heidi

2021-02-19T15:07:02Z

2021-02-19T15:07:02Z

2021

Flood vulnerability

Probabilistic methods

Flood exposure

Local flood risk assessments require integration of many detailed data sources from the public and private sector. In the framework of the EIT Climate-KIC Demonstrator project “SaferPLACES”, we explore how openly available datasets can be harnessed to both reanalyse past flood losses and estimate potential present and future flood damages. Three different cases studies were selected: Cologne in Germany, Rimini in Italy, and Pamplona in Spain. Each city has different size, economic structure and is subject to different types of flood events, namely fluvial, pluvial, and coastal. Here, we concentrate on methods for extracting exposure and predicting vulnerability for the residential and commercial sectors with openly available and crowd-sourced spatial datasets and public statistical data. Exposure is quantified at building level, covering residential and commercial assets. Further, vulnerability is calculated by Bayesian Network-based probabilistic models for residential and commercial sectors created on the basis of postdisaster household and company surveys. Finally, we use flood compensation data from a major flood in Pamplona in 2013 and analyse whether the cascade of models is able to recreate flood damages from a particular event. Flood risk estimates for Rimini are also shown to highlight the application of the project’s model chain.

angol

könyvfejezet

application/pdf

http://hdl.handle.net/10890/15145