Adaptability and transferability of flood loss functions in  residential areas

Abstract. Flood loss modeling is an important component within flood risk assessments. Traditionally, stage-damage functions are used for the estimation of direct monetary damage to buildings. Although it is known that such functions are governed by large uncertainties, they are commonly applied – even in different geographical regions – without further validation, mainly due to the lack of real damage data. Until now, little research has been done to investigate the applicability and transferability of such damage models to other regions. In this study, the last severe flood event in the Austrian Lech Valley in 2005 was simulated to test the performance of various damage functions from different geographical regions in Central Europe for the residential sector. In addition to common stage-damage curves, new functions were derived from empirical flood loss data collected in the aftermath of recent flood events in neighboring Germany. Furthermore, a multi-parameter flood loss model for the residential sector was adapted to the study area and also evaluated with official damage data. The analysis reveals that flood loss functions derived from related and more similar regions perform considerably better than those from more heterogeneous data sets of different regions and flood events. While former loss functions estimate the observed damage well, the latter overestimate the reported loss clearly. To illustrate the effect of model choice on the resulting uncertainty of damage estimates, the current flood risk for residential areas was calculated. In the case of extreme events like the 300 yr flood, for example, the range of losses to residential buildings between the highest and the lowest estimates amounts to a factor of 18, in contrast to properly validated models with a factor of 2.3. Even if the risk analysis is only performed for residential areas, our results reveal evidently that a carefree model transfer in other geographical regions might be critical. Therefore, we conclude that loss models should at least be selected or derived from related regions with similar flood and building characteristics, as far as no model validation is possible. To further increase the general reliability of flood loss assessment in the future, more loss data and more comprehensive loss data for model development and validation are needed.

Download Full-text

Adaptability and transferability of flood loss functions in residential areas

Natural Hazards and Earth System Sciences Discussions ◽

10.5194/nhessd-1-3485-2013 ◽

2013 ◽

Vol 1 (4) ◽

pp. 3485-3527 ◽

Cited By ~ 1

Author(s):

H. Cammerer ◽

A. H. Thieken ◽

J. Lammel

Keyword(s):

Flood Risk ◽

Residential Buildings ◽

Model Development ◽

Loss Functions ◽

Residential Areas ◽

Damage Functions ◽

Residential Sector ◽

Geographical Regions ◽

Flood Loss ◽

Loss Data

Abstract. Flood loss modeling is an important component within flood risk assessments. Traditionally, stage-damage functions are used for the estimation of direct monetary damage to buildings. Although it is known that such functions are governed by large uncertainties, they are commonly applied – even in different geographical regions – without further validation, mainly due to the lack of data. Until now, little research has been done to investigate the applicability and transferability of such damage models to other regions. In this study, the last severe flood event in the Austrian Lech Valley in 2005 was simulated to test the performance of various damage functions for the residential sector. In addition to common stage-damage curves, new functions were derived from empirical flood loss data collected in the aftermath of recent flood events in the neighboring Germany. Furthermore, a multi-parameter flood loss model for the residential sector was adapted to the study area and also evaluated by official damage data. The analysis reveals that flood loss functions derived from related and homogenous regions perform considerably better than those from more heterogeneous datasets. To illustrate the effect of model choice on the resulting uncertainty of damage estimates, the current flood risk for residential areas was assessed. In case of extreme events like the 300 yr flood, for example, the range of losses to residential buildings between the highest and the lowest estimates amounts to a factor of 18, in contrast to properly validated models with a factor of 2.3. Even if the risk analysis is only performed for residential areas, more attention should be paid to flood loss assessments in future. To increase the reliability of damage modeling, more loss data for model development and validation are needed.

Download Full-text

Probabilistic Flood Loss Models for Companies

10.5194/egusphere-egu2020-7334 ◽

2020 ◽

Author(s):

Lukas Schoppa ◽

Tobias Sieg ◽

Kristin Vogel ◽

Gert Zöller ◽

Heidi Kreibich

Keyword(s):

Risk Assessment ◽

Flood Risk ◽

Loss Estimation ◽

Flood Risk Assessment ◽

Damage Functions ◽

Flood Loss ◽

Loss Data ◽

Multivariable Models ◽

Loss Models ◽

Estimation Models

Flood risk assessment strongly relies on accurate and reliable estimation of monetary flood loss. Conventionally, this involves univariable deterministic stage-damage functions. Recent advancements in the field promote the use of multivariable probabilistic loss estimation models which consider damage controlling variables beyond inundation depth. Although companies contribute significantly to total loss figures, multivariable probabilistic modeling approaches for companies are lacking. Scarce data and heterogeneity among companies impedes the development of novel company flood loss models.We present three multivariable flood loss estimation models for companies that intrinsically quantify prediction uncertainty. Based on object-level loss data (n=1306), we comparatively evaluate the predictive performance of Bayesian networks, Bayesian regression and random forest in relation to established stage-damage functions. The company loss data stems from four post-event surveys after major floods in Germany between 2002 and 2013 and comprises information on flood intensity, company characteristics and private precaution. We examine the performance of the candidate models separately for losses to building, equipment, and goods and stock. Plausibility checks show that the multivariable models are able to identify and reproduce essential relationships of the flood damage processes from the data. The comparison of the prediction capacity reveals that the proposed models outperform stage-damage functions clearly while differences among the multivariable models are small. Even though the presented models improve the accuracy of loss predictions, wide predictive distributions underline the necessity for the quantification of predictive uncertainty. This applies particularly to companies, for which the heterogeneity and variation in the loss data are more pronounced than for private households. Due to their probabilistic nature, the presented multivariable models contribute towards a transparent treatment of uncertainties in flood risk assessment.

Download Full-text

Estimation of flood loss functions by activity sectors in the Pamplona metropolitan area in Spain by using insurance flood direct damage data

10.5194/egusphere-egu2020-18894 ◽

2020 ◽

Author(s):

Luis Mediero ◽

Enrique Soriano ◽

David Santillán ◽

Luis Cueto-Felgueroso ◽

Luis Garrote

Keyword(s):

Metropolitan Area ◽

Hydrodynamic Model ◽

Loss Functions ◽

Regional Government ◽

Flow Depth ◽

Flood Events ◽

The Real ◽

Direct Damage ◽

Damage Data ◽

Water Depths

Flood risk assessment studies require information about direct damages that depend on several variables, such as water depth, water velocity, flood duration, activity sector and type of building, among others. However, loss functions are usually simplified through flow depth-direct damage curves. Direct flood damages driven by a given flood event can be estimated directly from such loss functions by using either known or estimated water depths.In his study, flow depth-direct damage curves are estimated for a set of activity sectors in the Pamplona metropolitan area located in the northern part of Spain, within the activities of the EIT Climate-KIC SAFERPLACES project. A dataset containing all flood direct damages in the Pamplona metropolitan area in the period 1996-2018 were supplied by the Spanish &#8216;Consorcio de Compensaci&#243;n de Seguros&#8217; (CSC), benefiting from the fact that CSC is the insurance company that covers all damages produced by natural hazards in Spain. Flood direct damages are classified by activity sectors and postal codes. In addition, observed streamflow data at a set of gauging sites in the Ulzama and Arga rivers were supplied by both the Ebro River Basin Authority and the Regional Government of Navarre. A set of seven flood events with both streamflow and direct damage data available were selected. Flood hydrographs in the Arga River at Pamplona were obtained with a temporal resolution of 15 minutes. The Regional Government of Navarre supplied the real flood extensions for a set of flood events. With such real flood extensions, the two-dimensional hydrodynamic IBER model was calibrated. Flood extensions and water depths with a spatial resolution of 1 m were estimated with the calibrated hydrodynamic model for the seven flood events. Combining the dataset of direct damages with standard flow depth-direct damage curves and with water depths simulated by the hydrodynamic model, flood depth-damage curves were estimated by municipalities and postal codes. Such curves were obtained for activity sectors, considering residential, commercial and industrial assets.AcknowledgmentsThis study was supported by the project SAFERPLACES funded by the EIT Climate-KIC. The authors also acknowledge the &#8216;Consorcio de Compensaci&#243;n de Seguros&#8217; for providing the flood direct damage dataset and the Regional Government of Navarre for providing the real flood extensions for given flood events.

Download Full-text

Do hazard maps mirror loss data? – A vulnerability assessment based on loss data and hazard maps

10.5194/egusphere-egu2020-19685 ◽

2020 ◽

Author(s):

Mirjam Mertin ◽

Mattia Brughelli ◽

Andreas Zischg ◽

Veronika Röthlisberger ◽

Matthias Schlögl ◽

...

Keyword(s):

Flood Risk ◽

Risk Mitigation ◽

Flood Hazard ◽

Flood Damage ◽

Hazard Maps ◽

Protection Measures ◽

Essential Information ◽

Flood Loss ◽

Loss Data ◽

First Results

Implementing effective flood risk strategies is an essential task for policy-makers which will gain in importance as flood losses are expected to increase due to socio-economic and climatic drivers in near future. Flood risk mitigation incorporates structural and non-structural measures such as the declaration of flood hazard zones, both of which are associated with high financial expenses. Essential information to ensure maximum effectiveness and cost efficiency of flood protection measures is provided by quantitative flood loss analyses based, for example, on data from insurance claims.This project aims to model the expected flood damage, thus the vulnerability to buildings by examining country-wide, empirical flood loss data of Switzerland of the past 35 years. The developed method includes several steps: First, the loss data are statistically analysed, second the spatial distribution of the loss data in the different hazard zones is assessed and third, vulnerability models for each hazard zone are developed including further parameters such as building values or building zones. A further objective is to provide an overview of possible methods which differ in complexity and data requirement and can be adapted for other applications outside of Switzerland. First results show that the extent of loss increases as the degree of hazard rises. In contrast, however, the number of damage events is highest in flood zones with a lower degree of hazard. Further possibilities how risk adaptation strategies can be supported or complemented by flood loss data are presented within this project.

Download Full-text

Influence of flood frequency on residential building losses

Natural Hazards and Earth System Science ◽

10.5194/nhess-10-2145-2010 ◽

2010 ◽

Vol 10 (10) ◽

pp. 2145-2159 ◽

Cited By ~ 72

Author(s):

F. Elmer ◽

A. H. Thieken ◽

I. Pech ◽

H. Kreibich

Keyword(s):

Residential Buildings ◽

Statistical Error ◽

Residential Building ◽

Water Levels ◽

Recurrence Interval ◽

Flood Events ◽

Flood Loss ◽

Loss Data ◽

Loss Models ◽

Loss Modelling

Abstract. For the purpose of flood risk analysis, reliable loss models are an indispensable need. The most common models use stage-damage functions relating damage to water depth. They are often derived from empirical flood loss data (i.e. loss data collected after a flood event). However, object specific loss data (e.g. losses of single residential buildings) from recent flood events in Germany showed higher average losses in less probable events, regardless of actual water level. Hence, models that were derived from such data tend to overestimate losses caused by more probable events. Therefore, it is the aim of the study to analyse the relation between flood damage and recurrence interval and to propose a method for considering recurrence interval in flood loss modelling. The survey was based on residential building loss data (n=2158) of recent flood events in 2002, 2005 and 2006 in Germany and on-site recurrence interval of the respective events. We discovered a highly significant positive correlation between loss extent and recurrence interval for classified water levels as well as increasing average losses for longer recurrence intervals within each class. The application of principal component analysis revealed the interrelation between factors that influence the damage extent directly or indirectly, and recurrence interval. No single factor or component could be identified that explained the influence of recurrence interval, which led to the conclusion that recurrence interval cannot substitute, but complement other damage influencing factors in flood loss modelling approaches. Finally, a method was developed to include recurrence interval in typical flood loss models and make them applicable to a wider range of flood events. Validation including statistical error analysis showed that the modified models improve loss estimates in comparison to traditional approaches. The proposed multi-parameter model FLEMOps+r performs particularly well.

Download Full-text

Augmenting a Sociohydrological Flood Risk Model for Companies with Process-oriented Loss Estimation

10.5194/egusphere-egu21-10064 ◽

2021 ◽

Author(s):

Lukas Schoppa ◽

Marlies Barendrecht ◽

Tobias Sieg ◽

Nivedita Sairam ◽

Heidi Kreibich

Keyword(s):

Flood Risk ◽

Risk Model ◽

Flood Damage ◽

Modeling Framework ◽

River Elbe ◽

Validation Experiment ◽

Flood Loss ◽

Loss Data ◽

Damage Process ◽

Process Oriented

Sociohydrological models are increasingly used in flood risk analysis to reveal and understand the temporal dynamics in coupled human-flood systems. While most sociohydrological flood risk models are stylized and describe hypothetical human-flood systems, very few recent case studies employ empirical data to investigate real world systems. The mathematical representation of flooding processes in these models is often simplistic and does not reflect the current state of knowledge. This is due to the intricacy of human-flood interactions and the lack of sufficient and suitable historical data.We augment an existing, parsimonious sociohydrological flood risk model with a process-oriented flood loss model to integrate better understanding of flood damage processes into a sociohydrological modeling framework. Using Bayesian inference, we simulate the co-evolution of the flood risk system for companies located at the river Elbe in Dresden, Germany, over the course of 120 years. We compare model versions with differently complex process description on the basis of their loss prediction accuracy and uncertainty. This allows for exploring the added value of (i) resolving the inundation and damage process with more detail and (ii) accounting for heterogeneity across economic sectors. Apart from historical sociohydrological data, the proposed, augmented model versions are informed by object-level loss data, inundation maps, and spatial data, enhancing the pool of information available to the model. A leave-one-out cross-validation experiment shows that the augmented model versions increase the precision and reduce the uncertainty of company flood loss predictions in Dresden. In addition, the augmented models provide reliable loss predictions even in the absence of extensive historical flood loss data.The demonstrated model augmentation concept is not limited to the flood damage process but could be transferred to other processes within the human-flood system. For instance, by incorporating a dedicated model from protection motivation theory that describes how flood awareness and preparedness change after the occurrence of a damaging flood event.

Download Full-text

INSYDE-BE: Adaptation of the INSYDE model to the Walloon Region (Belgium)

10.5194/nhess-2021-363 ◽

2021 ◽

Author(s):

Anna Rita Scorzini ◽

Benjamin Dewals ◽

Daniela Rodriguez Castro ◽

Pierre Archambeau ◽

Daniela Molinari

Keyword(s):

Flood Risk ◽

Damage Model ◽

Flood Damage ◽

Original Model ◽

Risk Assessments ◽

Damage Functions ◽

Residential Sector ◽

Damage Models ◽

Similarities And Differences ◽

Walloon Region

Abstract. The spatial transfer of flood damage models among regions and countries is a challenging but unavoidable approach, for performing flood risk assessments in data and model scarce regions. In these cases, similarities and differences between the contexts of application should be considered to obtain reliable damage estimations and, in some cases, the adaptation of the original model to the new conditions is required. This study exemplifies a replicable procedure for the adaptation to the Belgian context of a multi-variable, synthetic flood damage model for the residential sector originally developed for Italy (INSYDE). The study illustrates necessary amendments in model assumptions, especially regarding input default values for the hazard and building parameters and damage functions describing the modelled damage mechanisms.

Download Full-text

Flood Risk and Adaptation Strategies for Soybean Production Systems on the Flood-Prone Pampas under Climate Change

Agronomy ◽

10.3390/agronomy11061187 ◽

2021 ◽

Vol 11 (6) ◽

pp. 1187

Author(s):

Wouter Julius Smolenaars ◽

Spyridon Paparrizos ◽

Saskia Werners ◽

Fulco Ludwig

Keyword(s):

Climate Change ◽

Flood Risk ◽

Risk Perceptions ◽

Production Systems ◽

Adaptation Strategies ◽

Flood Events ◽

Climate Data ◽

Soybean Production ◽

Drying Trend ◽

Sowing Season

In recent decades, multiple flood events have had a devastating impact on soybean production in Argentina. Recent advances suggest that the frequency and intensity of destructive flood events on the Argentinian Pampas will increase under pressure from climate change. This paper provides bottom-up insight into the flood risk for soybean production systems under climate change and the suitability of adaptation strategies in two of the most flood-prone areas of the Pampas region. The flood risk perceptions of soybean producers were explored through interviews, translated into climatic indicators and then studied using a multi-model climate data analysis. Soybean producers perceived the present flood risk for rural accessibility to be of the highest concern, especially during the harvest and sowing seasons when heavy machinery needs to reach soybean lots. An analysis of climatic change projections found a rising trend in annual and harvest precipitation and a slight drying trend during the sowing season. This indicates that the flood risk for harvest accessibility may increase under climate change. Several adaptation strategies were identified that can systemically address flood risks, but these require collaborative action and cannot be undertaken by individual producers. The results suggest that if cooperative adaptation efforts are not made in the short term, the continued increase in flood risk may force soybean producers in the case study locations to shift away from soybean towards more robust land uses.

Download Full-text

Fragility estimation for global building classes using analysis of the Cambridge earthquake damage database (CEQID)

Bulletin of Earthquake Engineering ◽

10.1007/s10518-021-01178-x ◽

2021 ◽

Author(s):

Robin Spence ◽

Sandra Martínez-Cuevas ◽

Hannah Baker

Keyword(s):

Fragility Curves ◽

Resistance Index ◽

Earthquake Damage ◽

Ground Acceleration ◽

Damage Level ◽

Ground Motion Parameter ◽

Geographical Regions ◽

Concrete Building ◽

Damage Data ◽

Regional Comparisons

AbstractThis paper describes CEQID, a database of earthquake damage and casualty data assembled since the 1980s based on post-earthquake damage surveys conducted by a range of research groups. Following 2017–2019 updates, the database contains damage data for more than five million individual buildings in over 1000 survey locations following 79 severely damaging earthquakes worldwide. The building damage data for five broadly defined masonry and reinforced concrete building classes has been assembled and a uniform set of six damage levels assigned. Using estimated peak ground acceleration (PGA) for each survey location based on USGS Shakemap data, a set of lognormal fragility curves has been developed to estimate the probability of exceedance of each damage level for each class, and separate fragility curves for each of five geographical regions are presented. A revised set of fragility curves has also been prepared in which the bias in the curve resulting from the uncertainty in the ground motion parameter has been removed. The uncertainty in the fragility curves is evaluated and discussed and the curves are compared with those from other studies. A resistance index for each class of building is developed and cross-regional comparisons using this resistance index are presented.

Download Full-text

Minimal Building Flood Fragility and Loss Function Portfolio for Resilience Analysis at the Community Level

Water ◽

10.3390/w12082277 ◽

2020 ◽

Vol 12 (8) ◽

pp. 2277 ◽

Cited By ~ 1

Author(s):

Omar M. Nofal ◽

John W. van de Lindt

Keyword(s):

Decision Process ◽

Community Level ◽

Loss Functions ◽

Damage Functions ◽

Flood Vulnerability ◽

Flood Duration ◽

Propagation Of Uncertainty ◽

Flood Depth ◽

Level Model ◽

Building Loss

Current flood vulnerability analyses rely on deterministic methods (e.g., stage–damage functions) to quantify resulting damage and losses to the built environment. While such approaches have been used extensively by communities, they do not enable the propagation of uncertainty into a risk- or resilience-informed decision process. In this paper, a method that allows the development of building fragility and building loss functions is articulated and applied to develop an archetype portfolio that can be used to model buildings in a typical community. The typical single-variable flood vulnerability function, normally based on flood depth, is extended to a multi-variate flood vulnerability function, which is a function of both flood depth and flood duration, thereby creating fragility surfaces. The portfolio presented herein consists of 15 building archetypes that can serve to populate a community-level model to predict damage and resulting functionality from a scenario flood event. The prediction of damage and functionality of buildings within a community is the first step in developing risk-informed mitigation decisions to improve community resilience.

Download Full-text