Flood Risk Change A Complexity Perspective Andreas Paul Zischg

Flood Risk Change A Complexity Perspective
Andreas Paul Zischg download
https://ebookbell.com/product/flood-risk-change-a-complexity-
perspective-andreas-paul-zischg-46882478
Explore and download more ebooks at ebookbell.com

Here are some recommended products that we believe you will be
interested in. You can click the link to download.
Climate Change Adaptation Flood Risk And Beyond State Of Play In The
Sciencepolicyaction Nexus Massimiliano Granceri Bradaschia
https://ebookbell.com/product/climate-change-adaptation-flood-risk-
and-beyond-state-of-play-in-the-sciencepolicyaction-nexus-
massimiliano-granceri-bradaschia-59562888
Creating Resilient Futures Integrating Disaster Risk Reduction
Sustainable Development Goals And Climate Change Adaptation Agendas
Stephen Flood
https://ebookbell.com/product/creating-resilient-futures-integrating-
disaster-risk-reduction-sustainable-development-goals-and-climate-
change-adaptation-agendas-stephen-flood-38469518
Creating Resilient Futures Integrating Disaster Risk Reduction
Sustainable Development Goals Andclimate Change Adaptation Agendas
Stephen Flood
https://ebookbell.com/product/creating-resilient-futures-integrating-
disaster-risk-reduction-sustainable-development-goals-andclimate-
change-adaptation-agendas-stephen-flood-38579690
Changes In Flood Risk In Europe Zbigniew Kundzewicz
https://ebookbell.com/product/changes-in-flood-risk-in-europe-
zbigniew-kundzewicz-4421480

Climate Change Risks And Food Security In Bangladesh Earthscan Climate
Winston Yu
https://ebookbell.com/product/climate-change-risks-and-food-security-
in-bangladesh-earthscan-climate-winston-yu-2329268
Chelsea Market Cookbook 100 Recipes From New Yorks Premier Indoor Food
Hall Di 1 Ban May
https://ebookbell.com/product/chelsea-market-cookbook-100-recipes-
from-new-yorks-premier-indoor-food-hall-di-1-ban-may-11890368
Flood Risk Management 1st Edition Oliver P Chin
https://ebookbell.com/product/flood-risk-management-1st-edition-
oliver-p-chin-51356328
Flood Risk Assessment And Management How To Specify Hydrological Loads
Their Consequences And Uncertainties 1st Edition Andreas H Schumann
Auth
https://ebookbell.com/product/flood-risk-assessment-and-management-
how-to-specify-hydrological-loads-their-consequences-and-
uncertainties-1st-edition-andreas-h-schumann-auth-2047450
Flood Risk The Holistic Perspective 1st Edition Zoran Vojinovic
https://ebookbell.com/product/flood-risk-the-holistic-perspective-1st-
edition-zoran-vojinovic-36496398

This page intentionally left blank

Flood Risk Change
A Complexity Perspective
Andreas Paul Zischg
University of Bern, Bern, Switzerland

Elsevier
Radarweg 29, PO Box 211, 1000 AE Amsterdam, Netherlands
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom
50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States
Copyright Ó 2023 Elsevier Inc. All rights reserved.
No part of this publication may be reproduced or transmitted in any form or by any means,
electronic or mechanical, including photocopying, recording, or any information storage
and retrieval system, without permission in writing from the publisher. Details on how to
seek permission, further information about the Publisher’s permissions policies and our
arrangements with organizations such as the Copyright Clearance Center and the Copyright
Licensing Agency, can be found at our website: www.elsevier.com/permissions.
This book and the individual contributions contained in it are protected under copyright by
the Publisher (other than as may be noted herein).
Notices
Knowledge and best practice in this ﬁeld are constantly changing. As new research and
experience broaden our understanding, changes in research methods, professional
practices, or medical treatment may become necessary.
Practitioners and researchers must always rely on their own experience and knowledge in
evaluating and using any information, methods, compounds, or experiments described
herein. In using such information or methods they should be mindful of their own safety
and the safety of others, including parties for whom they have a professional responsibility.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or
editors, assume any liability for any injury and/or damage to persons or property as a matter
of products liability, negligence or otherwise, or from any use or operation of any methods,
products, instructions, or ideas contained in the material herein.
ISBN: 978-0-12-822011-5
For information on all Elsevier publications visit our website
at https://www.elsevier.com/books-and-journals
Publisher: Candice G. Janco
Acquisitions Editor: Amy M. Shapiro
Editorial Project Manager: Joshua Mearns
Production Project Manager: Kumar Anbazhagan
Cover Designer: Greg Harris
Typeset by TNQ Technologies

To my family, Heike, Jule, and Linus

Contents
Preface xi
1. Introduction 1
References 7
2. Key drivers of flood risk change 9
Principles of flood risk analysis 11
Environmental changes 17
Environmental changes in the upstream catchment 17
Dams, reservoirs, and regulated lakes 17
Land use changes at catchment scale 19
Changes in the glacial and periglacial environment 20
Natural climate variability 20
Environmental changes in the floodplain 21
Changes in river morphology at reach scale 21
Climate changes 24
Socioeconomic changes 25
Coevolution of key drivers of change 26
References 30
3. Disentangling drivers of change 37
Analysis and modeling framework 38
Simulations driven by global and regional climate models 41
Sensitivity analysis and scenario-neutral approaches 43
The storyline approach 44
The retromodel experiment approach 46
Quantifying the isolated effects of drivers of change
with model experiments 47
The Kander river deviation of 1714 AD and its effects on
hydrological change 47
Analyzing the cumulative effects of river training measures
on flood impacts 75
Assessing changes to flood hazard maps due to climatic changes 104
Changes in sediment supply for torrential floods 108
vii

Analyzing spatiotemporal dynamics of fluvial erosion and
riverbed aggradation 112
Analyzing changes in flood exposure 119
Analyzing the effects of land use planning on flood exposure 125
Modeling the effects of changing vulnerability on flood risk change 128
Analyzing the effects of feedback and time lags on flood
risk change 134
Feedback effects of fluvial erosion and riverbed aggradation on
flood risk change 134
Feedback effects of landscape shaping flood events on
hydrological change 141
Feedback effects of flood events on the initiation of flood
protection projects 146
The societal memory effect on house construction activities in
flood-prone areas 148
Time lags in the implementation of flood risk management
measures 151
Other aspects that are contributing to the complexity of flood
risk management 153
Acknowledgments 155
References 155
4. Rivers and floodplains as complex adaptive systems? 167
Characteristics of complex systems 167
Numerosity and diversity 168
Order and disorder 169
Nested structure, modularity, and multiple scales 169
Feedback 170
Nonlinearity 171
Nonequilibrium, fluctuations, episodical changes 172
Emergence, self-organization, self-regulation 172
History and memory, path dependency 173
Robustness 173
Adaptive behavior 174
Uncertainty 175
Long-term evolution of floodplains seen from a complex systems
perspective 175
Complexity of flood risk change 177
Trajectories and pathways of flood risk evolution 178
Sensitivity to climatic changesdshort-term future pathways 179
References 184
5. Modeling spatiotemporal dynamics of flood
risk change 187
Examples of flood risk change analysis 188
The tool “Flood memory” 189
viii Contents

The tool “Flood damage potential” 190
The tool “Flood damage simulator” 198
The tool “Flood risk dynamics”dflood risk change at centennial
scale 214
The tool “Flood dynamics”dflood impact variability at hourly
scales 225
Decision-making leads to the emergence of spatial patterns of
flood risk change 242
Monitoring flood risk change 247
Modeling framework for considering complexity in the
analysis of flood risk change 247
Development of coupled component models 251
Model coupler 251
Development of model components 255
Data sources 258
Perspectives of coupled component models in hydrology 263
Context for hydrology 263
Hydrology as a context 263
Feedbacks: unifying top-down and bottom-up modeling approaches 264
Acknowledgments 264
References 264
6. Confronting complexity in flood risk management 273
A new perspective on flood risks 273
Lessons learnt from the Kander river deviation 279
Levers for controlling flood risk in the long term 283
Implications for flood risk research and management 286
References 292
Index 297
Contents ix

Preface
The book emerged from two different experiences, from practical experience in
moderating participative planning processes in flood risk management and from
the results of my research in flood risk. I worked for several years as a consultant
for public administrations that are responsible for flood protection and water
resources management. Planning and implementing flood protection measures
require to involve persons from different sectors and disciplines. Flood protec-
tion measures need space to enlarge rivers. Therefore, land ownersdin many
cases the agricultural sectordare stakeholders in the planning process. More-
over, stakeholders from environmental protection, hydropower production,
tourism, local recreation, monument preservation, city planning, landscape
ecology, archeology, landscape esthetics, water resources management, and
individual residents may be involved in the planning process. This results in a
diversity of perspectives on the topic or on the project and in a variety of interests
and expectations for the planning process. Balancing the interest conflicts or
target conflicts in participatory planning processes and finding a common sense
over the possible solutions with the group to solve the problems are one of the
key challenges for experts in flood risk management. Discussions in such diverse
groups shed light on the interdependencies between the different sectors and
problems. Simple solutions that may solve one single isolated problem affect
many other problems, positively or negatively, or may result in unintended
consequences or new problems. Therefore, possible solutions for the flood risk
management problem have to satisfy also the expectations of other planning
targets. This makes such planning processes complex and very challenging.
However, when adopting a complexity perspective on this topic, we can
approach the problems with a positive attitude. Instead of being an obstacle for
implementing fast-forward and simplistic solutions for a narrowly defined
problem, diversity becomes the key for finding sustainable and robust solutions. I
experienced many times that the inclusion of multiple stakeholders and their
perspectives in the planning process led to holistic solutions. In many cases, the
commonly elaborated solutions for flood risk management solved also other
problems that are connected with the flood risk management problem. The
foundations of complex systems research can provide the basis for a new
thinking and thus provide help in tackling complex and wicked problems.
With each flood protection project, we adapt our environment and landscape
to our needs. Together with natural changes in the environment and climate,
xi

these human adaptations shape the relationships between humans and their
environment. All three components of flood riskdhazard, exposure, and
vulnerabilitydare changing. This makes that flood risk is changing dynamically
in space and time. A successful and sustainable adaptation to future flood hazards
is only possible if we consider the main drivers of changing risks. We must know
where and how to intervene for keeping flood risks constant in time, or even
reducing it, also under increasing hazards, increasing exposure, and increasing
vulnerability. Again, the complexity perspective supports the analysis of the
dynamic nature of flood risks. Complex systems science provides theoretical
foundations for analyzing change, numerosity, feedbacks, nonlinear behavior,
emergence, system behavior and system sensitivity, and deep uncertainty.
The book addresses two types of readerships. Experts and engineers working
in flood risk management may find some ideas of how to consider and tackle
complexity in their projects. Scientists may find ideas and concepts for analyzing
flood risk change. In Chapter 1, the topic of flood risk change is introduced.
Chapter 2 summarizes and explains the key drivers of flood risk change.
Chapter 3 highlights approaches to show how the complex interactions between
the coevolving drivers of flood risk change can be disentangled and how the
isolated effects of single drivers of change on overall flood risk evolution can be
quantified. Chapter 4 discusses the question if rivers and their floodplains can be
interpreted as complex adaptive systems. Chapter 5 describes examples of model
experiments for analyzing the spatiotemporal dynamics of flood risk change. The
last chapter summarizes the conclusions of the model experiments by looking at
the topic of flood risk change from the perspective of complex adaptive systems.
The book cover tells the story of changing flood risk in a mountainous river
catchment. A previous flood event triggered new flood protection measures, in
this case the heightening of the lateral protective walls. Flood risk was lowered
for a time, but the question is how long it will take that the following increase in
exposure and vulnerability, as well as the increase in hazard intensity and
probability due to climatic changes, will increase flood risk again to preadap-
tation levels. The cover shows also that there are limits to adaptation by a further
heightening of the lateral protective walls in the future. The book is a first step
into the analysis of the dynamic nature of flood risks. It follows a systemic
approachdincluding environmental, socioeconomic, and sociotechnical
factorsdfor modeling and managing flood risk change. Readers will get a more
complete picture of the topic for understanding the complexity of flood risk
change, both from human and natural causes of flooding.
xii Preface

Chapter 1
Introduction
Flood risk change: A complexity perspective. This title covers two main topics
of this book. The first and main topic is flood risk change. Floods are still one
of the most damaging natural hazards, accounting for the majority of all
economic losses from natural hazards worldwide (UNISDR: Making Devel-
opment Sustainable: The Future of Disaster Risk Management, 2015). Risk
resulting from floods is defined as a function of the probability of a flood event
or scenario and its related extent of damage. However, as we are entering in
the era of “Great Acceleration” (Steffen et al., 2015), socioeconomy and the
human environment are changing rapidly. With these global changes, every
single factor that contributes to flood risks is changing. This results in a highly
dynamic change in flood risks.
Looking at how flood risk is changing or evolving in space and time re-
quires a paradigm change. Most of the problems we are currently dealing with
result from solutions of problems of the past. Many decisions were taken in the
past without foreseeing possible future development paths. The built envi-
ronment created will last for a long life cycle, and it shapes the adaptation
options of future generation. Flood risk managers are still planning structural
flood protection measures under a very static perspective. When conducting a
risk analysis, we become a snapshot in time and space. We do not know the
trajectory and the change rate of the past flood risk evolution. Even less, we
can foresee future developments of flood risk.
For this, the second topic of the book title opens a new perspective on flood
risks. Instead of considering flood risk as static in time and space, the book
aims at showing how flood risk can change. The book intends to summarize
the way toward a new understanding of flood risks. This comprehends to look
at how the drivers of risk, the factors of risk, and the resulting risks are
changing. The changes of the single risk factorsdhazard, exposure, and
vulnerabilitydare not independent from each other. Changes in flood hazards
are influencing and triggering changes in hazard and vulnerability. Therefore,
we can look at flood risk change with a systemic perspective. In analyzing the
dynamic changes of flood risk, we can observe many features of complexity,
namely the emergence of patterns, feedback mechanisms, numerosity and
diversity, order and disorder, nested structures, modularity, multiple scales,
hierarchies, nonlinearity, episodical changes, history and memory, path de-
pendency, and adaptive behavior. These characteristics of complex systems
Flood Risk Change. https://doi.org/10.1016/B978-0-12-822011-5.00004-1
Copyright © 2023 Elsevier Inc. All rights reserved. 1

can be observed in the evolution of coupled human and natural systems and
must be considered in flood risk management. In this book, I therefore want to
look at changing flood risks from the complexity perspective. The book is a
primer for analyzing and modeling flood risk change. It paves a way toward
the development of modeling frameworks that are able to consider selected
aspects of complexity. This is absolutely needed to avoid unintended conse-
quences of decisions in current flood risk management in the long-term
perspective. The book should give an outlook on how new-generation
modeling approaches can represent the complex behaviors of floodplains
and coupled human and natural systems in a highly dynamic environment.
These tools and methods should help flood risk management practice to tackle
with increasingly wicked problems.
Flood risks are now increasingly being analyzed from a dynamic rather
than from a static perspective (Mazzorana et al., 2012; Merz et al., 2010).
Several studies have addressed changes in natural risks over recent decades
and centuries (Himmelsbach et al., 2015; Hufschmidt et al., 2005; Paprotny
et al., 2018), and research on climate change and its impacts has focused on
future changes in risks (Alfieri et al., 2016, 2017; Arnell & Gosling, 2016;
Dottori et al., 2018; Hirabayashi et al., 2013). However, most studies focused
solely on the future increase in flood hazard. Only few studies consider both
the impacts of climatic changes to river flows and the future dynamics in the
elements at risk (Bouwer et al., 2010; Jongman et al., 2012; Liu et al., 2015;
Löschner et al., 2017; Winsemius et al., 2016). Closer examinations of the
spatiotemporal dynamics and the actual rate of change are rather rare.
Knowledge about hazardous processes and their impacts, as well as about the
trajectories of flood risk changes, is essential for the sustainable management
of flood risks.
Several intertwined natural and anthropogenic drivers influence the
spatiotemporal evolution of flood risk. In this book, the following drivers of
flood risk change that are related to environmental changes are considered:
l Floods are either caused by direct rainfall on the floodplain (pluvial floods
and surface water floods) or rainfall on river catchments resulting in
catchment outflow. The latter causes floods in downstream floodplains
(riverine floods and lake floods). Consequently, changes in flood processes,
i.e., changes in frequency and magnitude of floods in a floodplain, are
determined by changing precipitation.
l In mountainous areas, flood hazards are influenced by sediment transport
and deposition processes and debris flows. Debris flows are influenced by
environmental changes, such as melting of glaciers and permafrost, or
changes in weathering processes and mass movements.
l River morphology changes over time, including natural and gradual
changes in the river morphology, or disruptive changes by flood events. An
important aspect of river morphology changes is anthropogenic
2 Flood Risk Change

interventions, which are relevant drivers of flood risk in a floodplain, for
example, the construction of flood defenses such as levees and dams or
river restoration projects. However, the construction of levees as flood
protection measures in one floodplain can have adverse effects in the
downstream floodplains and can result in flooding trade-offs between up-
stream and downstream floodplains.
Beside changes in the natural environment, flood risk is also evolving due
to changes in the exposed elements at risk and their vulnerability. From this
aspect, the following drivers of change are considered here:
l The increase in the elements at risk change due to socioeconomic devel-
opment. The growth of settlements and thus the increase of residential
buildings are related to population growth.
l Infrastructure is increasing in parallel with population growth. This has
wider impacts on the socioeconomic system. For example, in economically
active areas, floodplains are increasingly occupied by production facilities,
as these require relatively flat areas for their construction. With economic
development, the elements at risk and the infrastructure in floodplains are
increasing both in terms of quantity and monetary value.
l Increasing values at risk compete with opposing drivers of flood risk
reduction measures implemented by individuals and the public. Hence,
changes in exposure and vulnerability are influenced by governmental
interventions and regulations and by the actions of individuals.
The built environment in floodplains, whether the settlement area or the
river channel, is subject to changes and coevolutionary dynamics in both so-
ciety and nature (Di Baldassarre et al., 2013, 2015; Fuchs et al., 2017). As
Vitousek et al. (1997) postulated, the human impact on nature is now
considerably larger than at any point in history. This is true for the floodplains,
as humans are shaping landscapes with the built environment. These impacts
of society on nature influences pathways of flood risk change. The spatio-
temporal development of these drivers of change in flood risk leads to diffi-
culties in predicting future flood risk. Consequently, recent studies have
extended the framework of risk analysis toward a spatiotemporal framework as
drivers for flood risk changes are varying in space and time (Ahmad &
Simonovic, 2013; Zischg et al., 2018). This book goes one step further and
shows examples of model experiments that analyze flood risk change, the
drivers of flood risk change, or features of complexity. We look at flood risk
from a dynamic perspective. The reader will become hands-on examples of
how we can analyze past changes in flood risk, of how we can monitor
changing risks, and of how we can model flood risk changes by considering
complex interactions between the drivers of change.
Chapter 2 summarizes and explains the key drivers of flood risk change.
The chapter begins with definitions of flood risks and explains how flood risk
Introduction Chapter | 1 3

can be calculated from the risk factors hazard, exposure, and vulnerability. It
describes how systems can be delimited and represented in risk analyses. The
focus is laid on river systems and floodplains. In these areas, flood hazards
meet the anthropogenic elements at risk and their vulnerability. River systems
and floodplains evolve in time. The chapter describes the main factors that
shape the long-term evolution of floodplains and of flood risk. It provides an
overview of all factors that are relevant for the increase or decrease of flood
risks, ranging from environmental change, climatic change, and socioeco-
nomic change. After describing the isolated effects of the single drivers of
change, Chapter 2 ends with a description of coevolution of risk factors.
Moreover, an insight into spatiotemporal dynamics of flood risk change is
given by explaining where changes in the system components occur and where
these become relevant for flood risk change. Herein, upstreamedownstream
and a river basineriver reach relationship as well as legacy effects in time and
space are explained.
Chapter 3 highlights theoretical and methodological approaches to show
how the complex interactions between the coevolving drivers of flood risk
change can be disentangled and how the isolated effects of single drivers of
change on overall flood risk evolution can be quantified. A special focus is laid
on the potential of model experiments for analyzing flood risk change. Thus,
principles of frameworks for analyzing and modeling flood risk change are
explained. The chapter introduces several examples of model experiments that
aim at disentangling the effects of selected drivers of change to overall flood
risk evolution. A special example shows how flood risks changed within the
past 300 years after the early geoengineering project of the Kander River
deviation in the Canton of Bern, Switzerland. This case study shows how a
cascade of unintended consequences for the society can be caused by a human
intervention in the natural river system. Moreover, this case study also points
out how difficult a long-term perspective is to keep on flood risk management.
The technological advancement of the industrial revolution totally changed the
framework of society’s values that underline decision-making in flood risk
management. The presented case studies can be used to compare the effects of
environmental changes and the effects of human interventions on flood risk
evolution. It becomes clear that future flood risk evolution can radically be
shaped by human adaptation and intervention. One example clearly shows that
the implementation of flood prevention measures remarkably drives flood risk
reduction. The model experiment bases mainly on counterfactual simulations,
i.e., comparing the present-day situation with alternative pathways of natural
and societal development. Some of the case studies exemplify how negative
feedback effects will contribute to flood risk reduction although climatic
changes might increase the intensity and frequency of extreme precipitation
events. Herein, sediment supply or sediment deficit determines flood risk
change in mountain areas. In the last part, Chapter 3 shows analyses on the
effects of land use planning and time lags in the land use regulation on
4 Flood Risk Change

emerging patterns of risk hot spots in the long term. Moreover, it presents
examples of the effects of the declining societal memory of past flood events
on flood exposure and vulnerability.
Chapter 4 argues if the long-term evolution of rivers and their floodplains
can be interpreted as complex adaptive systems. To this end, the main features
and characteristics of complex systems are summarized. This chapter sheds
light on the nature of the coupled human and natural systems of floodplains
and identifies which characteristics of complex systems can be observed on
floodplains. The reader will get an answer to the question if flood risk in
floodplains is a complicated issue or if floodplains can be considered as a
complex system or as a complex adaptive system; or, if rivers and floodplains
are nowadays to be considered as sociotechnical systems rather than natural
systems. These answers help the reader to understand the emergent phenom-
ena in the long-term development of floodplains. Examples of past flood
events will be discussed, and they will be shown how these events influenced
policy, technology, and risk awareness. Vice versa, examples will show how
past technological innovations shaped rivers and floodplains. The chapter ends
with an outlook on possible future trajectories and pathways of flood risk
evolution and with a discussion of the sensitivity of floodplains to climatic
changes.
Chapter 5 describes examples of model experiments and analyses of
spatiotemporal dynamics of flood risk change. It explains five interactive web
tools that have been developed by the Mobiliar Lab for Natural Risks of the
Oeschger Center for Climate Change Research of the University of Bern in
Switzerland. The first tool supports citizen science that aims at collecting
geolocalized images and photographs of flood events. This growing database
can be visualized and queried in a web-based map application. It contributes to
keeping the societal flood memory alive. The second tool describes an inter-
active web-mapping application that estimates the nation-wide exposure of
buildings, residents, work places, and critical infrastructure to floods. It con-
tributes to the understanding of the difference between a hazard-centered and
an exposure-centered visualization of flood risks. The third tool goes one step
further and provides an interactive model experiment to test the sensitivity of
flood damages to urban densification, the implementation of property-level
flood protection measures, and structural flood protection measures in the
river channel. This tool aims to show how one desired effect of urban planning,
namely to prioritize the internal densification of cities by developing unused or
abandoned spaces against urban sprawl, potentially can result in an unintended
increase of flood risks. It furthermore shows how existing flood risks can be
reduced by implementing flood prevention measures by different stakeholders.
Fourth, a model experiment is described and implemented in an interactive
web-based application that allows to assess the trajectory of flood risk change
in a floodplain in the past two centuries. The user can test counterfactual
scenarios of urban development and of alternative flood risk management

strategies. Moreover, the user can test the effects of different drivers of change
to the overall flood risk evolution. This tool contributes to the understanding of
the dynamic nature of flood risks in space and time. The fifth tool shows
storylines of extreme rainfall events and their impacts. The user of this
interactive web application can navigate through the spatiotemporal evolve-
ment of a flood event with a time slider and zoom in and out at different scales.
This tool visualizes direct and indirect effects of flood events at an hourly
timescale. It helps to understand the complex patterns of rainfall events over a
complex mountain topography, how the impacts of the event on society are
evolving in space and time, namely from upstream to downstream, and how
the spatial footprint of the flood impacts follows the rainfall pattern with a time
lag. Chapter 5 is also describing a model experiment that shows how the
philosophical school that is behind decision-making in flood risk management
can lead to the emergence of patterns in flood risk hot spots. It shows how the
consideration or nonconsiderations of social justice aspects in priority setting
can potentially guide the distribution of public funds for investments in flood
protection toward remote and economically disadvantaged or toward
economically advantaged regions. Finally, the last example describes the
operationalization of an information system for monitoring flood risk change.
Chapter 5 summarizes the principles of model coupling for analyzing and
modeling flood risk change. It discusses the advantages or limitations of
coupled component models. These models have a potential for considering
features of complex systems such as feedbacks, adaptive behavior, structural
change, regime change, and nonlinear behavior. The chapter closes with a
summary of the lessons learnt during the development of several coupled
component models and the implementation of the model experiments that
were described in the previous chapters. It closes with a perspective on the use
of coupled component models in hydrology and in the simulation of coupled
human and natural systems.
The last chapter summarizes the conclusions of the model experiments and
analyses described in this book by looking at the topic from the perspective of
complex adaptive systems. It discusses recommendations of how to consider
and implement some thoughts of complex systems science into flood risk
management practice and in the development of next-generation simulation
models. This aims at enabling modeler’s and decision-makers to analyze the
big trends or future pathways of flood risk evolution by considering human
behavior in simulations or at least by coupling models for simulating human
behavior with models for simulating natural processes. This chapter shows
how flood risk management can be enhanced with the perspective of complex
systems science. It will show how urgently we need to consider the complex
interactions between the drivers of change and complexity in flood risk ana-
lyses and decision-making in flood risk management. Finally, the chapter
outlines a complexity perspective for governing flood risk change in the 21st
century.
6 Flood Risk Change

With all this, the book scraps the surface of a new perspective and a new set
of methods to tackle the complex problems of the Anthropocene and to
confront deep uncertainties.
References
Ahmad, S. S., & Simonovic, S. P. (2013). Spatial and temporal analysis of urban flood risk
assessment. Urban Water Journal, 10(1), 26e49. https://doi.org/10.1080/
1573062X.2012.690437
Alfieri, L., Bisselink, B., Dottori, F., Naumann, G., de Roo, A., Salamon, P., Wyser, K., &
Feyen, L. (2017). Global projections of river flood risk in a warmer world. Earth’s Future,
5(2), 171e182. https://doi.org/10.1002/2016EF000485
Alfieri, L., Feyen, L., & Di Baldassarre, G. (2016). Increasing flood risk under climate change: A
pan-European assessment of the benefits of four adaptation strategies. Climatic Change,
136(3e4), 507e521. https://doi.org/10.1007/s10584-016-1641-1
Arnell, N. W., & Gosling, S. N. (2016). The impacts of climate change on river flood risk at the
global scale. Climatic Change, 134(3), 387e401. https://doi.org/10.1007/s10584-014-1084-5
Bouwer, L. M., Bubeck, P., & Aerts, J. C. J. H. (2010). Changes in future flood risk due to climate
and development in a Dutch polder area. Global Environmental Change, 20(3), 463e471.
https://doi.org/10.1016/j.gloenvcha.2010.04.002
Di Baldassarre, G., Kooy, M., Kemerink, J. S., & Brandimarte, L. (2013). Towards understanding
the dynamic behaviour of floodplains as human-water systems. Hydrology and Earth System
Sciences, 17(8), 3235e3244. https://doi.org/10.5194/hess-17-3235-2013
Di Baldassarre, G., Viglione, A., Carr, G., Kuil, L., Yan, K., Brandimarte, L., & Blöschl, G. (2015).
DebatesdPerspectives on socio-hydrology: Capturing feedbacks between physical and social
processes. Water Resources Research, 51(6), 4770e4781. https://doi.org/10.1002/
2014WR016416
Dottori, F., Szewczyk, W., Ciscar, J. C., Zhao, F., Alfieri, L., Hirabayashi, Y., Bianchi, A.,
Mongelli, I., Frieler, K., Betts, R. A., & Feyen, L. (2018). Increased human and economic
losses from river flooding with anthropogenic warming. Nature Climate Change, 8(9),
781e786. https://doi.org/10.1038/s41558-018-0257-z
Fuchs, S., Röthlisberger, V., Thaler, T., Zischg, A., & Keiler, M. (2017). Natural hazard man-
agement from a coevolutionary perspective: Exposure and policy response in the European
Alps. Annals of the American Association of Geographers, 107(2), 382e392. https://doi.org/
10.1080/24694452.2016.1235494
Himmelsbach, I., Glaser, R., Schoenbein, J., Riemann, D., & Martin, B. (2015). Flood risk along
the upper Rhine since AD 1480. Hydrology and Earth System Sciences Discussions, 12,
177e211. https://doi.org/10.5194/hessd-12-177-2015
Hirabayashi, Y., Mahendran, R., Koirala, S., Konoshima, L., Yamazaki, D., Watanabe, S., Kim, H.,
& Kanae, S. (2013). Global flood risk under climate change. Nature Climate Change, 3(9),
816e821. https://doi.org/10.1038/nclimate1911
Hufschmidt, G., Crozier, M., & Glade, T. (2005). Evolution of natural risk: Research framework
and perspectives. Natural Hazards and Earth System Science, 5(3), 375e387. https://doi.org/
10.5194/nhess-5-375-2005
Jongman, B., Ward, P. J., & Aerts, J. C. J. H. (2012). Global exposure to river and coastal flooding:
Long term trends and changes. Global Environmental Change, 22(4), 823e835. https://
doi.org/10.1016/j.gloenvcha.2012.07.004

Liu, J., Hertel, T. W., Diffenbaugh, N. S., Delgado, M. S., & Ashfaq, M. (2015). Future property
damage from flooding: Sensitivities to economy and climate change. Climatic Change, 132(4),
741e749. https://doi.org/10.1007/s10584-015-1478-z
Löschner, L., Herrnegger, M., Apperl, B., Senoner, T., Seher, W., & Nachtnebel, H. P. (2017).
Flood risk, climate change and settlement development: A micro-scale assessment of Austrian
municipalities. Regional Environmental Change, 17(2), 311e322. https://doi.org/10.1007/
s10113-016-1009-0
Mazzorana, B., Levaggi, L., Keiler, M., & Fuchs, S. (2012). Towards dynamics in flood risk
assessment. Natural Hazards and Earth System Science, 12(11), 3571e3587. https://doi.org/
10.5194/nhess-12-3571-2012
Merz, B., Hall, J., Disse, M., & Schumann, A. (2010). Fluvial flood risk management in a changing
world. Natural Hazards and Earth System Science, 10(3), 509e527. https://doi.org/10.5194/
nhess-10-509-2010
Paprotny, D., Sebastian, A., Morales-Nápoles, O., & Jonkman, S. N. (2018). Trends in flood losses
in Europe over the past 150 years. Nature Communications, 9(1). https://doi.org/10.1038/
s41467-018-04253-1
Steffen, W., Broadgate, W., Deutsch, L., Gaffney, O., & Ludwig, C. (2015). The trajectory of the
anthropocene: The great acceleration. Anthropocene Review, 2(1), 81e98. https://doi.org/
10.1177/2053019614564785
UNISDR. (2015). Making development sustainable: The future of disaster risk management. In
Global assessment report on disaster risk reduction, 4.2015. United Nations.
Vitousek, P. M., Mooney, H. A., Lubchenco, J., & Melillo, J. M. (1997). Human domination of
Earth’s ecosystems. Science, 277(5325), 494e499. https://doi.org/10.1126/
science.277.5325.494
Winsemius, H. C., Aerts, J. C. J. H., van Beek, L. P. H., Bierkens, M. F. P., Bouwman, A.,
Jongman, B., Kwadijk, J. C. J., Ligtvoet, W., Lucas, P. L., van Vuuren, D. P., & Ward, P. J.
(2016). Global drivers of future river flood risk. Nature Climate Change, 6(4), 381e385.
https://doi.org/10.1038/nclimate2893
Zischg, A. P., Hofer, P., Mosimann, M., Röthlisberger, V., Ramirez, J. A., Keiler, M., &
Weingartner, R. (2018). Flood risk (d)evolution: Disentangling key drivers of flood risk change
with a retro-model experiment. Science of the Total Environment, 639, 195e207. https://
doi.org/10.1016/j.scitotenv.2018.05.056
8 Flood Risk Change

Chapter 2
Key drivers of flood risk change
The term flood risk is here defined according the general concept of disaster
risk of the Intergovernmental Panel on Climate Change IPCC (IPCC, 2012).
Disaster risk is defined as the “likelihood over a specified time period of severe
alterations in the normal functioning of a community or a society due to
hazardous physical events interacting with vulnerable social conditions,
leading to widespread adverse human, material, economic, or environmental
effects that require immediate emergency response to satisfy critical human
needs and that may require external support for recovery.” Thus, risk requires a
potential hazard (H) and the “presence of people; livelihoods; environmental
services and resources; infrastructure; or economic, social, or cultural assets in
places that could be adversely affected.” This is defined as exposure (E).
Moreover, exposed people, communities, and infrastructure must be vulner-
able to the hazard. Vulnerability (V) is the “propensity or predisposition of
exposed people or infrastructure to be adversely affected” (IPCC, 2012). Risk
is composed of these three main components. Thus, risk is a function of
hazard, exposure, and vulnerability (UNISDR, 2015) and is often expressed in
Eq. (2.1)
R ¼ H E V (2.1)
These three main components of risk are not independent. Exposure is
connected with the magnitude and probability of the hazardous event. The
higher the magnitude of the hazard, the higher is the number of exposed
population or infrastructure. However, this interaction is nonlinear and de-
pends on many factors, which will be discussed below. The vulnerability of
exposed assets also depends on the frequency and magnitude of the hazard as
well as on the characteristics of the exposed objects, and thus there is an
interaction between these components of risk. The main reason of this inter-
dependency is mostly due to the effects of social learning. Buildings and
infrastructure built near rivers with frequent floods have often a reduced
vulnerability because they have been adapted to the hazard after previous
events (Elmer et al., 2010; Kienzler et al., 2015). The consideration of three
main components in defining risk opens the possibility to consider human
interventions for risk reduction. These interventions can range from the
installation of concrete hydraulic engineering structures to measures for
avoiding urbanization in hazardous areas by land use planning or building

flood-proof infrastructure with reduced vulnerability. The first type of in-
terventions targets at reducing the hazard H, either in terms of reducing fre-
quency or magnitude of flood events. The second type of intervention targets at
reducing exposure E, and the third type of interventions aims at reducing
vulnerability V.
The calculation of risk with or without risk management measures leads to
the assessment and quantification of the efficacy of these measures in terms of
risk reduction. The comparison of the benefits with the costs of the risk
reduction measures allows to analyze the efficiency of risk reduction mea-
sures. However, this requires the monetization of risks, i.e., quantifying the
risk and the reduced risk after the implementation of risk reduction measures
in monetary units. In natural sciences and engineering, risk is quantified as a
measure of uncertainty based on the concept of probability. Thus, risk is the
probability of a loss within a certain time period. As the loss depends on the
magnitude and frequency of the hazard triggering the loss, the probability of
occurrence of a hazard event or scenario with a certain magnitude is taken as
the probability of the loss. Thus, event risk or scenario risk can be written as
shown in Eq. (2.2) (Oberndorfer et al., 2020).
Ri;j ¼ f pj; pi;j; Ai;j; vi;j

(2.2)
Where Ri,j is the risk dependent on object i and scenario j, pj is the probability
of defined scenario j, pi,j is the probability of exposure of object i to scenario j,
Ai is the value of object i affected by scenario j, and vi,j is the vulnerability of
object i in dependence on scenario j. The probability of scenario j is the
probability of occurrence, mostly simplified by the number of how often the
event of this magnitude potentially occurs within a specified time period. This
is also termed as the return period of an event. If a hazard event of a certain
magnitude is expected to occur once in 100 years, it is called a “1-in-100 years
event” or a “100 year event.” If the value of object i is expressed in monetary
terms and the vulnerability is a dimensionless degree of the object suffering
loss, the risk function is a multiplication and results in the probability of a
damage event (Eq. 2.3).
Ri;j ¼ pj pi;j Ai;j vi;j (2.3)
The term “loss” here defines the quantifiable aspects of harm and damage
that result from a hazard event. In this way, the scalar value of risk includes
both probability and the severity of the impact, but in a way that keeps hazard,
exposure, and vulnerability distinct and gives rise to a single dimension
(Rougier et al., 2013). The term “damage” is used for explicitly focusing on
the monetized impacts of hazard events and thus is a subset of the overall loss.
If multiple hazard scenarios are considered in the quantification of risk, risk
or the mean expected lossdin many cases in a 1-year-intervaldis the sum of
the product of each possible loss and its probability. The losses of several
10 Flood Risk Change

hazard scenarios with increasing magnitudes and decreasing occurrence
probability form the loss exceedance probability curve (Fig. 2.1). In natural
hazards management, this function is often termed as the “hazard footprint”
function. The footprint function aggregates the spatial overlay between the
footprint(s) of the hazard scenario(s) and the values at risk with their object-
specific vulnerabilities. The form and shape of the footprint function show
how the loss increases with event magnitude. Fig. 2.1 shows an example of a
flood hazard footprint function in terms of exposed population. If risk is
derived from the exceedance probability curve or from the footprint function,
risk is calculated as the mean loss from the area under the curve. From the risk
change perspective, it becomes obvious that risk quantification must include
the time period for which the footprint function and the assessed risk value are
valid.
Principles of flood risk analysis
As this book deals mainly with flood risks in mountainous or hilly terrain, the
procedure of analyzing flood risk is outlined schematically in the following
section. The concept is described from a “top-down” perspective, by following
the course of the water through the landscape.
The main starting point is the definition of the hazard domain and the
spatial delimitation of the system that is considered in the risk analysis. The
FIGURE 2.1 Hazard footprint. An example of a hazard footprint function showing the
relationship between increasing flood magnitude and increasing flood exposure. Modified from
Zischg, A. P., Bermúdez, M. (2020). Mapping the sensitivity of population exposure to changes
in flood magnitude: Prospective application from local to global scale. Frontiers in Earth Science,
8. https://doi.org/10.3389/feart.2020.534735.
Key drivers of flood risk change Chapter | 2 11

first step is the definition of the hazard process. In the case of floods, there are
several possibilities to consider. Probably the simplest flood hazard in terms of
direct causal effect from heavy rainfall to consider in flood risk analysis is
flooding from infiltration excess overflow, i.e., surface runoff leading to sur-
face water floods. Floods by surface runoff result from rainfall events that
exceed the infiltration capacity of the soil. In urban environments with a high
share of sealed surface and limited sewer capacity, the threshold of rainfall
intensity and sum for triggering surface water floods are lower. This process is
also termed as pluvial floods. In this case, the spatial delimitation of the system
equals the area for which the risk must be calculated (area of interest) or for
which risk reduction measures are planned plus the (small) catchments from
which surface water is flowing into the area of interest (Fig. 2.2, left). In case
of riverine floods, the spatial delimitation of the system considered for risk
analysis is the floodplain where the settlements or infrastructure are located
(area of interest) plus the upstream watershed or river basin, which contributes
to the discharge in the river that is triggering flood events in the area of interest
(Fig. 2.2, right). In larger river basins, there can be several areas of interest if
more than one floodplain exists. In this case, the areas of interest for risk
analysis are nested within the river basin.
In summary, the system delimitation comprises the type of hazard
considered in the study, the area of interest where the risk has to be analyzed,
and the upstream areas where the hazard processes originate from and form the
boundary condition of the inner system. The inner system can be defined as an
open system. Hazards can originate and occur within the system itself and can
enter into the system from outside at the system boundary. The outer system
delimitation, i.e., the upstream catchment area, is thus needed for modeling the
hazard in case of floods.
Besides surface water floods and riverine floods, flood hazards are also
influenced by other flood processes such as sediment transport, lateral erosion,
large wood transport and deposition, as well as debris flow. Especially in
FIGURE 2.2 System delimitation. Spatial delimitation of the system considered in flood risk
analysis. Left: the area of interest for risk analysis plus the contributing areas for surface water
floods. Right: the area of interest plus the upstream catchment. Background map from
map.geo.admin.ch.

mountainous areas, floods and their impacts on infrastructure are likely
dominated by sediment transport and deposition. Thus, the definition of the
system for which the risk analysis is valid contains a clear statement for which
type of hazard the risk analysis is valid. This is part of the system delimitation.
This explicit definition of the hazard processes considered and the explicit
exclusion of the hazard types that are not considered are necessary for the
interpretability of the results. Especially if the risk analysis is termed as “flood
risk analysis,” misunderstandings or misinterpretations can result from unclear
definitions of the flood process type. Another requirement of the system de-
limitation is a clear statement of which type of exposure is considered in the
risk analysis. Is the risk assessment valid for people, buildings, or infra-
structure only? Are mobile assets as, for example, cars moving around
considered in the analysis? Depending on the type of asset considered, the
system can be spatially limited, but it is open to input from outside. One
example of an open system is road traffic entering and exiting the floodplains.
In most cases of flood risk analysis, the area of interest (inner system) is a
floodplain or a set of floodplains within a river basin. In floodplains, the main
factors of flood risk, the flood process, and the values at risk intersect spatially.
From a physical perspective, floodplains are defined as areas of land adjacent
to and formed by flowing water in times of floods. From a socioeconomic
perspective, floodplains provide land for settlement, infrastructure, and other
human activities. In the following, the term “floodplain” is used synonymously
for the area of interest for flood risk analysis if not otherwise specified.
Once the type of hazard and exposure is defined and the system is spatially
delimited, the main hazard input parameters for estimating the probability and
magnitude of the hazard events and scenarios are determined. The calculation
of flood risk requires a set of events or scenarios with increasing magnitude.
This is the so-called “event set.” In an ideal case, the case study area has a long
record of well-documented flood events. The spatial footprints of these past
flood events can be reconstructed (Bomers et al., 2019) and overlaid with the
actual or historic dataset of exposure, e.g., houses or traffic infrastructure. If
the occurrence probability of the reconstructed flood events can be determined,
the historic event set leads to the loss footprint function of the system under
investigation. Eventually, risk can be quantified from this footprint function.
However, the event set is very small in many cases, due to the inherent low
probability of occurrence of extreme events. In this situation, the event set
must be developed by means of simulation models or by a cascade of simu-
lation models. The first step for determining the event set for surface water
floods is to analyze the characteristics of rainfall events (Bernet et al., 2019).
The meteorological data from nearby located meteorological stations is
analyzed with extreme value statistics. An intensity-duration function is
derived from the data and a set of rainfall scenarios with increasing magnitude
is derived from the extreme value statistics. This set of rainfall scenarios then
is used as an input for modeling surface water floods. The hydrodynamic

simulation models use the rainfall upon the digital elevation model within the
spatial delimitation of the system and convert the rainfall into surface runoff.
The water fluxes over the digital elevation model are computed with the
shallow water equations or with simplified formulas (Bernet et al., 2018). The
output of these hydrodynamic simulation models is a map of the water depths
and flow velocities for selected time steps during the time interval of the
simulation for each rainfall scenario. In case of modeling surface water floods,
the simulation time (temporal system delimitation) equals the duration of the
rainfall event.
In contrast to modeling surface water floods, the modeling of riverine
floods requires an additional modeling step. A hydrological model is needed to
convert the rainfall scenarios over the catchment upstream of the area of in-
terest for the risk analysis (inner system) into water discharge at the interface
between the catchment outlet and the outer boundary of the inner system
(Felder et al., 2018). The water coming from the catchment and flowing into
the area of interest for risk analysis is given as an input into the hydrodynamic
model (Felder et al., 2017). In this case, the hydrodynamic model computes
the water fluxes within the inner system from the upper boundary condition to
the lower boundary condition. Advanced hydrodynamic models are able to
simulate both flood processes in parallel, surface water floods, and riverine
floods, e.g., LISFLOOD-FP (Bates De Roo, 2000). The output is the same
as for the simulation of surface water floods, namely a set of maps showing
water depths and flow velocities for selected time steps during the time interval
of the simulation. However, the simulation time period must be extended to
consider the time lag between rainfall in the upper catchment and outflow from
the catchment and the time lag between the inflow of water into the floodplain
and the outflow from the floodplain. The higher the time lag is, the larger is the
size of the catchment, and the length of the river reaches through the system.
Thus, the simulation time (temporal system delimitation) remarkably exceeds
the duration of the rainfall event.
In case of riverine floods, the event set, or the set of scenarios, must not
necessarily be derived from rainfall characteristics over the catchment. If
hydrological measurements are available nearby the area at interest, the time
series of the discharge measurements can be analyzed with extreme values
statistics. The occurrence probability of flood peaks of a certain magnitude
can be derived directly from these statistics. However, the hydrodynamic
simulations of the inundation processes require a hydrograph for the whole
length of the simulation period. Thus, if flood probability is derived from
discharge time series, the form of the hydrograph must also be derived from
the time series (Zischg et al., 2018). Each flood scenario in the event set must
be described by the flood peak and the flood volume, i.e., by the shape of the
flood hydrograph. This hydrograph is then used as an input for the hydro-
dynamic model to simulate inundation dynamics and to create flood maps.
The hydrological model to convert the rainfall event set in a corresponding

set of flood hydrographs is not needed in this case. The spatial delimitation of
the system is therefore restricted to the spatial extent of the inner system, i.e.,
the area of interest for the risk analysis. Moreover, the temporal delimitation
of the system can also be restricted to the time period of the flood event in the
hydrograph plus the time needed for water traveling through the floodplain
within the area of interest.
In summary, the model cascade from rainfall to inundation enables to
correlate each rainfall event or flood event of the event set with a related flood
map. Thus, the probability of occurrence of the input event or scenario cor-
responds to the occurrence probability of the flood event, expressed in a hazard
footprint. The hazard footprint is described by the extent of the inundated area,
the inundation depths, and the flow velocities.
In mountainous areas, flood processes are often influenced by sediment
transport and large wood dynamics. Thus, flood maps differ if these processes
are considered in hazard mapping or not. The effects of sediment transport
dynamics on flood maps can be complex and have consequences for the
calculation of damages. A flood event with sediment transport can lead to
aggradation of the riverbed, thus leading to a decrease of the transport capacity
of the river channel during the flood event. This increases the volume of water
exiting the river channel and thus the extent of the flooded areas in the
floodplain. Sediment deposited in the floodplain can also lead to higher
damages to infrastructure and houses, at least by increasing the cleaning costs
after the flood event (Fuchs et al., 2007). The vulnerability of houses to floods
is also higher if sediment is involved or deposited around and within the
houses. Another example of increasing hazard is the transport of large wood
coming from the upstream catchment. Wood pieces come from flooded forest
areas and lateral erosion and are transported downstream by the river. At
bridge piers, the wood logs can lead to occlusion of the river channel below the
bridge and thus to a decreasing of the discharge capacity of this cross section
(Ruiz-Villanueva et al., 2014). This in turns leads to an increase of the outflow
of water from the river channel into the floodplain and an increase of the
flooded area. If these processes have to be expected, the flood hazard maps in
many cases have a higher extent as the flood maps of the same occurrence
probability that are based on plain water simulations only. The consideration
of fluvial sediment transport and large wood dynamics in the elaboration of the
hazard maps requires the adaptation of the spatial delimitation of the system to
be analyzed because the sources of sediment and large wood transport often
originate from river reaches upstream of the areas of interest (Mazzorana et al.,
2009).
The next step in this top-down model cascade is the flood impact modeling
part. The complexity of the impact modeling increases with the kind of assets
and values at risk considered in the analysis. The simplest analysis setup is the
spatial overlay of the flood maps with houses and the derivation of monetary
damages from the estimated repairing costs. The houses are fixed in space and

do not move around as population does. The consideration of population in the
risk analysis increases the level of complexity. First, population is mobile and
leaves or enters the area of interest. During floods, people move out of the
flooded area and relief units move in. The same is valid for cars. The
complexity of impact modeling further increases with the consideration of
adaptation measures, e.g., the effects of early warning systems on prepared-
ness, or when cascading impacts have to be expected. Examples of cascading
impacts are the physical damage to single parts of the electric grid infra-
structure leading to a blackout and to subsequent cascading impacts to the
economic activities. Another example is the cascading impact if roads are
effected widely. Working places and production facilities are not only affected
directly by the flood but can also be indirectly affected by road closures that
hamper work forces to commute from home to their working place.
Impact modeling is mostly conducted by means of geographical infor-
mation systems (GIS) because they have the ability to combine the outcomes
of the flood models with geodata and other data on socioeconomic activities.
The impact analysis is done on a spatially explicit way and the damages or
impacts calculated for each object (value at risk) are aggregated at the level of
the area of interest. This results in a scalar value of risk for the floodplain.
However, the spatiotemporal dynamics of the flood impacts can also be
analyzed. Fig. 2.3 shows schematically a model cascade from rainfall to flood
impacts.
In summary, there are many factors that influence flood risk and flood risk
change. Most of these factors change over time at different timescales and
change in space. Thus, the assumption of stationarity and constant conditions
FIGURE 2.3 Model cascade. Model cascade from rainfall to flood impacts.

in the analysis of flood risk is to be discussed. In the following, the main
drivers of flood risk change are summarized.
Environmental changes
As stated above, floods are either caused directly by rainfall falling onto the
system under investigation (pluvial floods and surface water floods, for
example) or by falling onto river catchments resulting in a catchment outflow,
with or without sediment and large woods. The latter causes floods in down-
stream floodplains (riverine floods and lake floods). Thus, the boundary con-
dition of floods in floodplains can either be rainfall, river flow, or both.
Consequently, changes in flood processes, i.e., changes in the frequency and
magnitude of floods, are determined by these external influencing factors. This
opens a number of freedom degrees in changing the hazard footprint. In the
following, plausible factors of change in the environment and in society are
listed that are possibly influencing flood risk. The list is ordered systematically
by the distinction between changes in the upstream catchments that are rele-
vant for the changing the boundary conditions of the risk analysis and changes
in the narrow area of interest.
Environmental changes in the upstream catchment
Dams, reservoirs, and regulated lakes
The most obvious factor that is altering the hydrology of a river basin is the
construction of dams and reservoirs. Dams are constructed for temporarily
holding back the water flow and store water in the reservoir. Reservoirs can
have multiple purposes, ranging from hydropower and electricity production to
drought management and flood control. The effect of an upstream reservoir on
downstream risk is given by the retention effect of the reservoir. A reservoir
with the purpose for flood risk reduction is constructed to store the volume of
water during the peak flow exceeding the discharge capacity of the down-
stream river reaches. After the rainfall event, the reservoir is steadily releasing
the stored flood water over a longer period. This leads to an attenuation of the
flood peak downstream. Depending on the storage volume, the water level in
the reservoir must be lowered before the rainfall event to increase the storage
capacity for the inflow flood volume. This requires the establishment and
functioning of weather forecast and early warning systems. In many river
basins, multiple reservoirs exist along the river course. In these river basins,
the hydrological regime is remarkably altered. Flood attenuation is one of the
main purposes for many of the existing dams and for some major dams of the
world currently under construction (ICOLD International Commission on
Large Dams, 2021). In total, 7450 large reservoirs worldwide are in operation
for the purpose of flood control.

Reservoirs for other purposes as well as regulated lakes also store water
during rainfall events und thus have similar effects on flood frequency and
magnitude at downstream river reaches. However, the effects of these reser-
voirs are not as optimized for flood control as reservoirs built for flood control.
The attenuation effect of a reservoir or a regulated lake is measured by
comparing the flood frequency curve of the inflowing and the outflowing flood
hydrograph. This requires discharge measurements upstream and downstream
of the reservoir, which are in many cases commercially sensitive and not
available. A simplified method for estimating the retention effect is to
calculate the flood attenuation index. This is the ratio of dam outflow to inflow
at the flood peak (Zhao et al., 2020). The analysis of the effects of reservoirs
on hydrology is analyzed by inserting reservoir operation rules in hydrological
models (Fleischmann et al., 2019; Meng et al., 2019). Boulange et al. (2021)
estimate that dams reduce the number of flood-exposed population down-
stream of dams by 20% in a climate change scenario if assuming population as
constant.
A similar flood attenuation effect arises from natural dams developed by
large landslides, debris flows, or rock avalanches that obstruct the main river
course of a valley bottom. Until the river erodes the deposits and enlarges the
discharge capacity through the deposits, a natural dam delimits the flow
through the deposits. In case of floods, a share of the flood water is held back
by the natural dam. This leads to backwater flooding of the area upstream of
the natural dam and to a reduction of the flood peak downstream. Thus, their
effect on flood risk is twofold: flood risk in downstream areas may decrease
while in upstream areas risk may increase.
An indirect effect of dams and reservoirs on flood risk is the alteration of
sediment transport. Dams are holding back sediment fluxes from upstream
catchments. Sediment transport downstream of dams is limited, except during
operation phases intended for emptying the reservoir from sediment. Thus,
long periods (decades) without sediment input from upstream change river
morphology and vegetation. Sediment transport capacity increases and thus
riverbed erosion tends to increase. This results in increasing conveyance ca-
pacity downstream and thus in reduced probability of flooding in the flood-
plain. Another effect of flow regulation by reservoir operation is a change in
vegetation in and near the riverbed. The decreasing frequency of high-water
situations leads to the growth of forests and dense vegetation. This has a
negative effect from the flood risk perspective as dense vegetation increases
the surface friction, lowers the flow velocity, and ultimately increases water
levels during flood situations. This eventually increases the probability of
flooding in the floodplain. Moreover, the presence of trees in flood areas in-
creases the availability of large wood recruitment.
In summary, dams and reservoirs have in many cases a flood attenuation
effect and thus tend to reduce flood risks downstream of the facility. In rare
cases, dams of reservoirs may collapse or may be operated erroneously and

trigger widespread flooding downstream with catastrophic consequences
(Larrauri Lall, 2020). This residual risk will increase in the near feature
because the majority of existing dams are reaching their lifetime (Perera et al.,
2021).
Land use changes at catchment scale
The outflow from a river basin or a mountainous catchment during or after a
rainfall event is varying with a high number of factors related to the land cover
and land use. The study of the influence of environmental conditions on
catchment outflow is a wide field of research in the discipline of hydrology and
covers multiple spatial and temporal scales. Vegetation cover (type of vege-
tation, areal extent, and canopy density) and especially forest cover markedly
influence interception during rainfall. A higher forest cover intercepts a rela-
tively higher share of the rainfall and reduces surface runoff. Thus, affores-
tation and reforestation lead to a reduced flood peak. Prior to a rainfall event,
vegetation cover controls basin evapotranspiration and thus controls ante-
cedent conditions of the catchment, e.g., soil moisture and soil saturation.
Moreover, forests and natural vegetation are related with natural soils that are
not compacted by anthropogenic activities such as some agricultural practices.
Therefore, soils in forested areas have a higher infiltration capacity than soils
in agricultural land and catchments with a higher share of forested areas have
lower flood peaks than catchments with a high share of areas with compacted
soils or bare land (Schilling et al., 2014). Forested and reforested areas reduce
surface runoff and eventually flood peak but also sediment and nutrient load of
rivers (Ouyang et al., 2013). As surface runoff increases over agricultural land
and especially over temporarily bare soils, soil erosion increases too, and thus
fine sediment is directed toward the water courses and ultimately to the river
network.
The intensification of agricultural land does not only lead to soil
compaction (Alaoui et al., 2018), but comprises also investments into drainage
systems. This infrastructure drains the water from agricultural land in flood-
plains and transports the water toward the main river course. Depending on the
timing of the flood peak during a rainfall event, this additional water from the
floodplains can alter the flood peak in the river if synchronized with the peak
flow of the main river. The drainage system decreases the retention effect of
the floodplains, and a higher share of the rainfall is directly converted into
runoff within the time period of the rainfall event. The transformation of
wetlands to agricultural areas has a similar effect on flood peak as drainage
systems (O’Connell et al., 2007). The retention capacity is lowered and the
drainage from these transformed areas is increased (Gulbin et al., 2019). The
study of Javaheri and Babbar-Sebens (2014) shows that wetlands can reduce
the peak flow in downstream floodplains up to 42%, flood areas up to 55%, and
maximum flow velocity up to 15%. Even more important than these land use

changes for altering flood peaks is the transformation of agricultural land to
settlement and infrastructure with sealed soils (Monk et al., 2019). Sealed soils
totally inhibit the infiltration of rainfall water and thus reduce the storage
capacity of the soil layer. Moreover, sealed surfaces have low surface frictions.
Surface runoff is therefore fast on these surfaces and catchments with a high
share of sealed areas react fast.
As land use has changed significantly in the last centuries, nonstationarity
in the processes of surface runoff generation and hydrological catchment
characteristics must be expected widely. However, the impact of land use
changes on stream flow and floods is complex and studies often obtain con-
tradictory results (Rogger, Chirico, et al., 2017).
Changes in the glacial and periglacial environment
Mountain permafrost is characterized by ice-rich sediments and quaternary
deposits. The ice layers below the surface inhibit infiltration into deeper parts
of the substrate and thus rainfall on permafrost leads to rapid reactions of high-
alpine catchments. Locally, also potentially very deep groundwater storage
volumes as, for example, debris slope below rock walls and deep sedimentary
deposits can be impermeable because of the existence of thin ice layers.
Warming of permafrost thus eliminates the ice layers in the long term, i.e.,
with a long delay and potentially increases the storage capacities of high alpine
catchments (Rogger, Agnoletti, et al., 2017). However, permafrost thawing
will decrease slope stabilities and thus led to increased landslide activities.
These processes provide an increased amount of material for debris flow and
sediment transport in high alpine torrents.
Natural climate variability
Flood frequency varies naturally at decadal and centennial time scales. The
variability of flood occurrence is related with the long-term variability of the
global atmospheric circulation patterns. Paleoflood reconstructions show that
periods with higher flood occurrence in the region of the European Alps
coincide with warmer summer temperatures (Glur et al., 2013) and lower with
cooler summer temperatures. Increased flood frequency is correlated with
latitudinal shifts of Atlantic and Mediterranean storm tracks. A known influ-
ence of climate variability on flood occurrence are the El Niño and La Niña
events, the extremes of ENSO climate variability (Emerton et al., 2017). These
events influence river flow and flooding at the global scale. In Central Europe,
a high number of floods occurred during and at the end of the little ice in the
19th century (Brönnimann et al., 2019). An increased frequency of flood-prone
weather types for the European Alps has been found in the mid-19th century
and a decreased frequency in the postwar period, consistent with a climate
reconstruction that shows changes in the cyclonic flow over Western Europe.

Also in other parts of Europe, a relation between flood frequency and climate
variability has been found (e.g, Ward et al., 2008; Wirth et al., 2013). Periods
with a cooler climate are periods with advancing glaciers, a more extensive
snow cover during winter, and a higher potential for snow melt influenced
floods. Volcanic eruptions can also have a temporal influence on flood
occurrence (Rössler Brönnimann, 2018). Schmocker-Fackel and Naef
(2010) analyzed flood-rich and flood-poor periods in Switzerland and found
that periods with reduced flood occurrence coincide with periods of minimum
solar activity. An overview of regions where flood occurrence is actually
increasing or decreasing in Europe is given by Blöschl et al. (2017).
Environmental changes in the floodplain
Environmental changes in the floodplain itself, i.e., in most cases the area of
interest for risk analysis, alter both surface water floods and riverine floods.
Changes in the floodplain can be gradual or abrupt. They can have natural
causes but also anthropogenic causes, ranging from changes in the river
morphodynamics and flood regime, to changes in the adjacent vegetation or
disruptive changes by flood events, for example, by levee failures. Anthro-
pogenic interventions are more or less the most relevant driver of flood risk in
a floodplain; that is, the construction of flood defenses such as levees and dams
or river restoration projects. Furthermore, the construction of levees as flood
protection measures in one floodplain can have adverse effects in downstream
floodplains and thus, result in trade-offs between upstream and downstream
floodplains. Moreover, floodplains can be affected by land subsidence due to
drainage or groundwater extraction. This results in increasing flood hazards
and consequently, increasing flood risk (Carisi et al., 2017). In the following,
an overview of environmental changes in the floodplain is given.
Changes in river morphology at reach scale
Themorphologyoftheriverisimportantforfloodriskasitdeterminesthedischargeca-
pacityorconveyancecapacity.Thehigherthetransportcapacity,theloweristheprob-
abilityofinundationintheadjacentriverfloodplain(Fig.2.4).Thus,floodoccurrence
probabilityinthefloodplainisdeterminedmainlyberiverwidthandriverdepth,i.e.,the
geometryoftherivercrosssections.
Water flows and sediment supply and transport are inherently unsteady
(Church Ferguson, 2015). Thus, natural rivers change their channel
morphology periodically. These changes are mostly driven by lateral erosion
and riverbed erosion. Especially flood events can change the river morphology
because of the high shear stress on riverbed and riverbanks. Besides riverbed
erosion and aggradation during and after flood events, also adjacent vegetation
is changed (Guan et al., 2016). Bank erosion leads to the elimination of dense
bank vegetation and the hydrodynamic force on vegetated river islands can

lead to the elimination of vegetation within the river channel. This lowers
hydraulic friction and leads to an increased flow velocity. Riverbank vegeta-
tion and riparian forests have an effect on bank stabilization and the flow
regime (Abernethy Rutherfurd, 2001). Riparian vegetation increases stream
bank resistance and critical shear stress (Pollen Simon, 2005) and reduces
flow velocity at the stream bank interface by increasing channel roughness
provoking flow energy dissipation through plant deformation. One approach to
implement the mechanical effects of roots on bank erosion in modeling is by
adapting critical shear stress. Further, the degree of protection depends on the
location, root depth, and species composition (Slater et al., 2019) showed that
river morphology adjusts to climate variability. In addition, river morphology
determines also the water surface elevation in case of floods. The substrate of
the river channel determines friction and with this it determines flow velocity
and water surface levels. A coarser surface has a higher friction. This lowers
flow velocity and heightens water surface elevation. Vice versa, a smooth
surface increases flow velocity. Another factor that determines flow conditions
in the river channel is the slope.
An important driver of morphology change during floods is the failure and
breach of lateral levees. The local change of water flow and sediment at levee
breaches triggers erosion and sedimentation processes within the river channel
and in the adjacent floodplain. This change obviously modifies the inundation
pattern in the floodplain. Flood models used for the elaboration of flood
inundation maps rarely consider levee breaches. In this case, the water outflow
from the levee breach potentially increases the flooded area. Thus, risk anal-
ysis based on flood maps of floodplains with river levees that do not consider
levee breaches underestimates risks.
More relevant than natural causes of changing river morphology are
anthropogenic causes (Gregory, 2006). In Europe, rivers have been modified
since more than 300 years. The reasons for actively modifying river
morphology range from making rivers navigable, making use of hydropower
generation, to flood protection. More recently, the main motivation for
FIGURE 2.4 River morphology. The effect of river channel morphology changes for flood risk
change. Fluvial erosion amplifies the river conveyance capacity and reduces inundation probability
in the floodplain (left). Bed aggradation reduces the river conveyance capacity and amplifies
inundation probability (right). Modified from Slater, L.J., Singer, M.B., Kirchner, J.W. (2015).
Hydrologic versus geomorphic drivers of trends in flood hazard. Geophysical Research Letters,
42(2), 370e376. https://doi.org/10.1002/2014GL062482.

changing river morphology is river revitalization. These anthropogenic in-
terventions are more or less the most relevant driver of flood risk change.
The intention of flood risk reduction by river engineering is to enhance
river conveyance capacity. The more water can flow within the river channel,
the lesser is the occurrence probability of inundation in the floodplain. This
aim can be reached by the construction of lateral levees. These constructions
enhance conveyance capacity by heightening the levels of the lateral shores.
Thus, the cross-sectional area is increased. However, the heightening of the
lateral riverbanks is limited, especially in urban areas. Higher lateral dams
require the heightening of the access roads to bridges too with consequences
for city planning. Thus, another option is to artificially deepen the riverbed by
excavation. This option is limited because it influences the stability of the
riverbanks and foundations of existing flood defenses. A third option is to
enlarge the river channel to extend conveyance capacity. This is the most ideal
situation, but it is usually limited because the lateral borders of rivers in urban
areas are occupied by houses or road infrastructure. All three options require a
balancing as they can have unintended consequences. Widening the river too
much can lead to a decrease of sediment transport capacity and to the
aggradation of the riverbed. This is eventually reducing conveyance capacity
(Hall et al., 2014). A narrow river channel tends to incise if sediments supply
from upstream is limited. The increasing of conveyance capacity has also
another effect for flood risk reduction: a wider river has not only a lower
probability of inundation in the floodplain but inundation depths are also lower
on average because a higher share of the water remains in the channel in
comparison to narrower rivers. Structural flood defenses such as lateral levees
have an effect downstream. If the inundation of the floodplain is prevented
upstream, more water enters the downstream floodplain. This potentially en-
hances peak discharge downstream. Hence, the effects of river engineering
works must be analyzed at larger scale, e.g., at river basin scale (Seher
Löschner, 2018).
Traditional engineering solutions against flooding and lateral mobility of
river channels (e.g., straightening, embankment, channel dredging, bank
protection) have strongly altered basic functions of hydrosystems, such as the
sediment, water, and nutrients exchanges between the channel and the
floodplain, or the longitudinal continuity of sediment transport along stream
networks (Arnaud et al., 2019; Bravard et al., 1999; Gilvear, 1999). It became
also clear that the physical status of rivers has been negatively impacted by
other strong human pressures, such as gravel mining and
hydroelectricity (Best, 2019). Many human interventions in the river system
today are river restoration projects aiming at improving the ecological
quality of rivers by dam or dike removal, riverbed widening, artificial sedi-
ment replenishment, among others (Wohl et al., 2015). Today, floodplain
restoration and giving more space to rivers are actively promoted in flood risk
management. These goals of restoring or enhancing catchment and river

processes that have been affected by human interventions are also aiming at
reducing flood hazards. These activities are summarized by the term “natural
flood management” (Dadson et al., 2017).
The draining of floodplains in the past centuries has a long-term effect on
the soil subsidence. The drying of former wetlands leads to the compaction of
the whole soil layer and thus leads to shrinking. The subsidence of the soil
layer in the floodplain in comparison a more stable riverbed increases flood
inundation depths and thus can alter flood risk in the long term. Groundwater
extraction has similar effects on soil subsidence and consequently on flood risk
change (Carisi et al., 2017).
The analysis of the effects of morphological changes on flood risks requires
to study the environmental changes in the entire catchment in an integrated
way. Only with this perspective, local changes in river morphology and eco-
systems can be evaluated in context with upstream catchment processes
(Molnar et al., 2002).
Climate changes
Flood processes heavily depend on rainfall. Since rainfall is an important
parameter of global and regional climate and climate is expected to change, a
change in rainfall intensity and frequency, especially extreme rainfall events, is
also expected to alter flood processes (IPCC, 2012). The most important aspect
of a warming climate is the increase of water in the atmosphere. According to
the ClausiuseClapeyron relation, saturation humidity will increase with higher
temperature. This implies that extreme moisture transport will increase in a
warmer climate. This can lead to more intensive rainfall or a higher frequency
of extreme rainfall events. In mountainous areas, a warming climate influences
the altitude level of the 0
C temperature line. The upward shift of this
parameter determines the form of precipitation. Above this line, precipitation
falls as snow and (more or less) below this line precipitation is falling as rain.
The upward shift of the snow line extends the areas where precipitation is
falling as rain and thus increases the contributing areas of flood processes. It is
expected that this change is also influencing the probability of rainfall on snow
cover (Rössler et al., 2014). A warmer atmosphere in mountainous areas
potentially leads to the intensification of convective processes and to a higher
rainfall intensity during thunderstorms. Flood peaks tend also to shift
seasonally (Blöschl et al., 2017). In upstream catchments, climate change will
influence soil erosion processes (Li Fang, 2016). This potentially causes the
formation of gully erosion and an increase of sediment transport in down-
stream river reaches. The increase of temperatures and changing precipitation
patterns affect glacial and periglacial areas in the mountains. Glaciers are
shrinking and permafrost ice is melting. This leads to an increase of exposed
sediments that might become susceptible for erosion and triggering debris
flows (Keiler et al., 2010).

Climate change is also expected to change atmospheric circulation pat-
terns. This might lead to changes in regional weather types. This changes
spatiotemporal patterns of the rainfall and can play an important role for flood
runoff generation (Martius et al., 2020). Depending on the topology of the
river network, the timing of stream flow accumulation along the river network
driven by spatially distributed rainfall influences peak discharge in the indi-
vidual river reaches (Zischg, Felder, Mosimann, et al., 2018; Pattison et al.,
2014; Zoccatelli et al., 2011). Hence, the synchronization of peak flows in a
river network may change with changing rainfall patterns. Flood assessment at
the catchment scale is dependent on spaceetime variability of rainfall within a
catchment, hydrological (soil and vegetation) processes leading to local runoff
formation, and the propagation through the river network.
Socioeconomic changes
Besides changes in the natural environment, flood risk is also changing due to
variations in the exposed elements at risk and their vulnerability. The elements
at risk increase with socioeconomic development (Elmer et al., 2012). This
increase in exposure directly enters the risk formula (Eq. 2.3) and thus has a
direct effect on flood risk change. Settlement growth and thus the increase of
residential buildings are related to population growth. Studies have shown that
the number of houses that are potentially exposed to flood processes increased
by a factor of 7 in the past century (Fuchs et al., 2015). These are long-term
changes of population exposure that are shown by the growth of houses.
However, population growth increases also the mobility of people and there-
fore short-term exposure to hazards. By traveling, people temporarily pass
hazardous areas that are in many cases avoided for constructing houses.
Hence, fluctuation of the exposure of people and mobile assets is an important
factor in analyzing natural risks (Keiler et al., 2005).
Infrastructure is increasing in parallel with population growth. This has
wider impacts on the socioeconomic system. For example, in economically
active areas, floodplains are increasingly occupied by production facilities, as
these require relatively flat areas for their construction. With economic
development, the elements at risk and the infrastructure in floodplains are
increasing both in terms of quantity and monetary value.
Only if population and infrastructure exposed to floods are vulnerable, a
flood event results in flood damages. The state of susceptibility to harm from
exposure to stresses associated with environmental and social change and from
the absence of capacity to adapt mirrors regional socioeconomic characteris-
tics and varies in space and time (Adger, 2006). In flood risk research, physical
vulnerability is distinguished from social vulnerability. Physical vulnerability
is defined as the degree of damage. The damage is expressed as a share of the
total value of an object. The values range from 0 to 1, where 0 means no
vulnerability and 1 total damage. Social vulnerability describes the ability or

inability of a society to cope with the adverse impacts of a hazard (Cutter,
2003). Both physical and social vulnerability change over time. The experi-
ence of flood events often results in a lower social vulnerability against the
subsequent flood events. This is the effect of social learning. Homeowners see
what can happen and reduce the physical vulnerability of their houses by
installing object protection measures (Elmer et al., 2010; Kreibich et al., 2011;
Thieken et al., 2007). In relation to the increase in gross domestic product and
in population, fatalities and losses do not increase to the same extent and are
thus relatively decreasing (Jongman et al., 2015). However, vulnerability can
also increase over time. An example of an increasing physical vulnerability is
the installation of home automation techniques in houses that result in high
damages in case of floods. These damages due to new characteristics of the
houses are additional to the other structural damages. Increasing mobility and
the dependency of the economy on traffic, e.g., due to the optimization of
supply chains, lead to an increasing social vulnerability. Highly interconnected
industrial production processes that are spatially distributed are more vulner-
able than single production facilities. The more infrastructure is connected, the
higher are the probabilities of cascading effects of flood on the functioning of
infrastructure (Wang et al., 2018). This interconnectedness of infrastructure is
increasing vulnerability. Hence, vulnerability changes with the technological
advancement. New technologies allow increasing efficiency and reduce
redundancy what in case of widespread flood leads to outage of parts of the
production system. These outages cascade through the whole socioeconomic
system. Moreover, physical vulnerability changes with legislation. In countries
with a strong land use planning sector, the construction of houses in hazard
zones is prohibited or is only possible with constraints (Röthlisberger et al.,
2017). These constraints aim at reducing vulnerability, e.g., the obligation to
construct new buildings in a flood save way. Last, but not the least, vulnera-
bility reduction is target of flood risk management strategies, and thus it
changes with a widespread implementation of these strategies.
Coevolution of key drivers of change
In the previous chapters, we have seen that many of the components of the risk
formula (Eq. 2.3) change over time. Some changes occur at a large timescale
(decades to centuries), whereas others occur at shorter timescales (daily to
yearly). The built environment in floodplains, whether the settlement area or the
river channel, is subject to changes and coevolutionary dynamics in both society
and nature (Di Baldassarre et al., 2013). As Vitousek et al. (1997) postulated, the
human impact on nature is now considerably larger than at any point in history.
This is also true for the floodplains, as humans are shaping landscapes with the
built environment, including the rivers in the Anthroposphere by modifying
their courses. These impacts of society on nature influence future risk pathways.
The spatiotemporal development of these drivers of change in flood risk leads to

difficulties in predicting future flood risk. Consequently, recent studies have
extended the framework of risk analysis toward a spatiotemporal framework as
drivers for flood risk changes are varying in space and time (Ahmad
Simonovic, 2013; Zischg, Hofer, Mosimann, et al., 2018).
To improve the understanding of flood risk change, the drivers of flood risk
change are described from a coevolutionary perspective. This widens the
explanatory power of multiple drivers in hazard dynamics, exposure dynamics,
and vulnerability dynamics. The coevolutionary framework provides a
guideline for analyzing and explaining the linkage between these three main
factors of flood risk (Fuchs et al., 2017). Coevolution includes two or more
interdependently evolving systems (Gual Norgaard, 2010). Two systems
coevolve when their change in space or time depends on each other. The aim
of this framework is to analyze and understand the intertwined changes within
the different interacting systems, where coevolutionary dynamics are path-
dependent (Kallis, 2007). These dynamics include social adaptation to envi-
ronmental change and the shaping of the environment by social adaptation. A
central theme inherent to the coevolutionary perspective is the analysis of
human adaptation to catastrophic flood events. The analysis of coevolution of
the drivers of flood risk changes opens the opportunity to analyze the feedback
mechanisms, i.e., to analyze how each driver of change influences others (Roe,
2009). The interaction must not necessarily endure over the whole period but
can be reactive and abrupt. Coevolution can consist of slow and fast dynamics
and can be variable in space and time.
One of the most postulated coevolutionary dynamics is the increase of
exposure to extreme floods with the increase of flood protection measures that
reduce frequent flooding. For example, Ferdous et al. (2020) observe a higher
growth of population density with higher levels of flood protection. This is
sometime termed as the “levee paradox” (White, 1945). Thus, human adap-
tation coevolves with changes in flood frequency. Herein, the interaction is a
two-way feedback. A flood event of a certain magnitude may initiate the
installation of flood protection measures against flood events of this magni-
tude. Subsequently, only flood events markedly exceeding this magnitude will
cause flood damages and trigger new adaptation (Sivapalan Blöschl, 2015).
If the impacts of flood events are large enough to influence not only the local
level but are relevant also for national or international societies, governments
at higher hierarchy may adapt their legislative framework, subsidy policies,
and risk management policies (Lane, 2014). This has feedback effects to
adaptation at community and individual levels.
It can also be that coevolutionary dynamics in the drivers of flood risk
change have opposing trends. For example, Boccard (2021) observed two
distinct trends in flood impacts: Fatalities from flood events are falling
worldwide while financial losses are rising. Increasing values at risk compete
with opposing drivers of flood risk reduction measures implemented by in-
dividuals and the public. Hence, changes in exposure and vulnerability are

influenced by governmental interventions and regulations and by the actions of
individuals. Increasing values at risk also coevolve with the observed increase
of insured damages. Barredo (2009) thus proposed to normalize the time series
of flood damages with the increasing values at risk.
The physical drivers of flood risk change are also intertwined, involving a
multitude of interlinked processes. Catchments respond to changes in land use
and climate in a nonlinear way. Climate, vegetation, and soil coevolve and
result in landscape changes and runoff characteristics of catchments (Gaál
et al., 2012). Changes of vegetation cover and type cause changes of hydro-
logical processes affecting flooding by changes of interception, evapotrans-
piration, and infiltration (Runyan et al., 2012). Ultimately, changes in
vegetation influence soil moisture and thus antecedent conditions of catch-
ments. Similar coevolutionary dynamics arise with human modifications of
land cover in catchments. Deforestation or afforestation also change the runoff
properties of the catchment. A coevolution is also expected with the shrinking
of glaciers and the melting of subsurface permafrost ice and the arising of new
source areas for debris flows in high mountain catchments.
Davenport et al. (2021) showed that change in precipitation coevolves with
changes in damages of pluvial floods. More extreme rainfall does not only
increase surface runoff and flooding but also may increase erosion and land-
slides. Perdigão and Blöschl (2014) showed that changes of spatial patterns of
rainfall markedly cause changes in flood risks.
The influence of climate change, as well as that of human interventions on
nature, is being studied comprehensively and in a differentiated manner with
regard to future flooding scenarios. The contribution of humans to changes in the
sediment budget of the world’s largest rivers was investigated on a global scale
by Syvitski et al. (2005). Human-induced soil erosion increases the sediment
load in rivers. At the same time, less sediment is transported to the coasts due to
reservoir effects caused by humans (e.g., large dam projects). However, changes
in soil erosion with an influence on the bedload balance are not only caused by
humans, but also to a large extent by climate change. There are direct and in-
direct influences of climate change on soil erosion with associated changes in the
sediment content of rivers, which can cause both an increase in sediment load
and a decrease, depending on the climatic conditions (Li Fang, 2016). Yang
et al. (2018) studied the changes in sediment load in the Yellow River in China
and concluded that it has decreased by 60% since 1997 and that climatic
changes, in addition to human influence, have contributed a large part to this. An
opposite effect was researched by Dibike et al. (2018) in Canada. The future
sediment load was modeled with the inclusion of climatic changes in the Lower
Athabasca River. The results indicate that an overall increase of up to 50% in
sediment load in the river as well as in the deposition area is expected for the
next 60 years. Increased sediment load with regard to an increase in extreme
events was investigated by Tu et al. (2020). They concluded that floods will

increase especially during events with a 200-year return period, resulting in more
sediment being transported and deposited.
According to Naylor et al. (2017), river geometry dynamically adapts to
changing flow regimes, which can lead to extreme events altering channel ca-
pacity and floodplain geometry, thereby changing the risk of future flooding.
This form of feedback means that extreme events that occurred in the past
represent a modified flood risk for future events due to the geomorphological
response (Croke et al., 2015), and Coulthard and Van De Wiel (2007) discuss
these morphological changes of river morphology by river bank failures as a
manifestation of self-organized criticality. On the one hand, it is known that
high-magnitude flood events can erode, transport, and deposit large amounts of
sediment, which can reshape the river system and have a corresponding effect
on channel capacity (the cross-sectional area of the channel) and thus on flow
velocity. On the other hand, it has been shown that increasing flooding is also
caused by ongoing geomorphological changes (sedimentation in the channel),
which gradually reduce channel capacity. The influence of river morphology
and interactions within the riverbed on flood behavior has long been under-
estimated (Raven et al., 2010). Lane et al. (2007) criticize the lack of consid-
eration of geomorphological processes in flood risk management. It shows how
important it is to consider the natural processes of a river. Lateral erosion, for
example, can cause high sediment inputs and thus influence river transport and
increase flood risk. According to their calculations, a sediment input during
16 months can cause a short-term bed elevation, which leads to about half of the
increase in flood area in an event with a return period of about 1e5 years. Slater
et al. (2015) complement this research by showing the importance of the
coevolution of river morphology change and river conveyance capacity. On
average, a 1% decrease in channel velocity or flow cross-sectional area resulted
in a 2% increase in flood hazard frequency. These results suggest that shifts in
sediment flow, grain size, and/or vegetation may play a significant role in
controlling flood hazard frequency. Slater (2016) examined the findings from
the United States for England and Wales. She concluded that a 10% decrease (or
increase) in channel capacity would result in an increase (or decrease) in flood
frequency of about 1.5 days per year on average. The widespread increase in the
frequency of flood events was amplified by both hydrological and geomor-
phological effects.
Important sediment contributors to rivers are tributaries. Picking up on this
theme, Bressan et al. (2020) showed that both river geometry and runoff input
from tributaries are key factors determining the magnitude and duration of
shear stress. Larger tributaries can keep the water depth higher and thus in-
crease the shear stress. Smaller tributaries can add clearer water and thus
reduce sediment concentrations. Failure to include tributaries or the use of
incorrect hydrographs leads to a large under- or overestimation of sediment
volumes.

In summary, there are many possibilities how one driver of flood risk
change interacts with other drivers of change. The effects of these intertwined
drivers on flood risk have to be disentangled before being analyzed in an
isolated way. In the next chapters, methods for disentangling the effects of
single drivers on flood risk change are shown.
References
Abernethy, B., Rutherfurd, I. D. (2001). The distribution and strength of riparian tree roots in
relation to riverbank reinforcement. Hydrological Processes, 15, 63e79. https://doi.org/
10.1002/hyp.152
Adger, W. N. (2006). Vulnerability. Global Environmental Change, 16(3), 268e281. https://
doi.org/10.1016/j.gloenvcha.2006.02.006
Ahmad, S. S., Simonovic, S. P. (2013). Spatial and temporal analysis of urban flood risk
assessment. Urban Water Journal, 10(1), 26e49. https://doi.org/10.1080/
1573062X.2012.690437
Alaoui, A., Rogger, M., Peth, S., Blöschl, G. (2018). Does soil compaction increase floods? A
review. Journal of Hydrology, 557, 631e642. https://doi.org/10.1016/j.jhydrol.2017.12.052
Arnaud, F., Schmitt, L., Johnstone, K., Rollet, A. J., Piégay, H. (2019). Engineering impacts on
the Upper Rhine channel and floodplain over two centuries. Geomorphology, 330, 13e27.
https://doi.org/10.1016/j.geomorph.2019.01.004
Barredo, J. I. (2009). Normalised flood losses in Europe: 1970-2006. Natural Hazards and Earth
System Science, 9(1), 97e104. https://doi.org/10.5194/nhess-9-97-2009
Bates, P. D., De Roo, A. P. J. (2000). A simple raster-based model for flood inundation
simulation. Journal of Hydrology, 236(1e2), 54e77. https://doi.org/10.1016/S0022-1694(00)
00278-X
Bernet, D. B., Trefalt, S., Martius, O., Weingartner, R., Mosimann, M., Röthlisberger, V.,
Zischg, A. P. (2019). Characterizing precipitation events leading to surface water flood
damage over large regions of complex terrain. Environmental Research Letters, 14(6). https://
doi.org/10.1088/1748-9326/ab127c
Bernet, D. B., Zischg, A. P., Prasuhn, V., Weingartner, R. (2018). Modeling the extent of surface
water floods in rural areas: Lessons learned from the application of various uncalibrated
models. Environmental Modelling and Software, 109, 134e151. https://doi.org/10.1016/
j.envsoft.2018.08.005
Best, J. (2019). Anthropogenic stresses on the world’s big rivers. Nature Geoscience, 12(1), 7e21.
https://doi.org/10.1038/s41561-018-0262-x
Blöschl, G., Hall, J., Parajka, J., Perdigão, R. A. P., Merz, B., Arheimer, B., Aronica, G. T.,
Bilibashi, A., Bonacci, O., Borga, M.,
Canjevac, I., Castellarin, A., Chirico, G. B., Claps, P.,
Fiala, K., Frolova, N., Gorbachova, L., Gül, A., Hannaford, J., .
Zivkovi
c, N. (2017).
Changing climate shifts timing of European floods. Science, 357(6351), 588e590. https://doi.
org/10.1126/science.aan2506.
Boccard, N. (2021). Analysis of trends in disaster risk. International Journal of Disaster Risk
Reduction, 53. https://doi.org/10.1016/j.ijdrr.2020.101989
Bomers, A., Schielen, R. M. J., Hulscher, S. J. M. H. (2019). Application of a lower-fidelity
surrogate hydraulic model for historic flood reconstruction. Environmental Modelling and
Software, 117, 223e236. https://doi.org/10.1016/j.envsoft.2019.03.019

Boulange, J., Hanasaki, N., Yamazaki, D., Pokhrel, Y. (2021). Role of dams in reducing global
flood exposure under climate change. Nature Communications, 12(1). https://doi.org/10.1038/
s41467-020-20704-0
Bravard, J. P., Landon, N., Peiry, J. L., Piégay, H. (1999). Principles of engineering geo-
morphology for managing channel erosion and bedload transport, examples from French
rivers. Geomorphology, 31(1e4), 291e311. https://doi.org/10.1016/S0169-555X(99)00091-4
Bressan, F., Mantilla, R., Schilling, K. E., Palmer, J. A., Weber, L. (2020). Hydrologic-hydraulic
modeling of sediment transport along the main stem of a watershed: Role of tributaries and
channel geometry. Hydrological Sciences Journal, 65(2), 183e199. https://doi.org/10.1080/
02626667.2019.1687897
Brönnimann, S., Frigerio, L., Schwander, M., Rohrer, M., Stucki, P., Franke, J. (2019). Causes of
increased flood frequency in central Europe in the 19th century. Climate of the Past, 15(4),
1395e1409. https://doi.org/10.5194/cp-15-1395-2019
Carisi, F., Domeneghetti, A., Gaeta, M. G., Castellarin, A. (2017). Is anthropogenic land sub-
sidence a possible driver of riverine flood-hazard dynamics? A case study in Ravenna, Italy.
Hydrological Sciences Journal, 62(15), 2440e2455. https://doi.org/10.1080/
02626667.2017.1390315
Church, M., Ferguson, R. I. (2015). Morphodynamics: Rivers beyond steady state. Water Re-
sources Research, 51(4), 1883e1897. https://doi.org/10.1002/2014WR016862
Coulthard, T. J., Van De Wiel, M. J. (2007). Quantifying fluvial non linearity and finding self
organized criticality? Insights from simulations of river basin evolution. Geomorphology,
91(3e4), 216e235. https://doi.org/10.1016/j.geomorph.2007.04.011
Croke, J., Denham, R., Thompson, C., Grove, J. (2015). Evidence of self-organized criticality in
riverbank mass failures: A matter of perspective? Earth Surface Processes and Landforms,
40(7), 953e964. https://doi.org/10.1002/esp.3688
Cutter, S. L. (2003). The vulnerability of science and the science of vulnerability. Annals of the
Association of American Geographers, 93(1), 1e12. https://doi.org/10.1111/1467-8306.93101
Dadson, S. J., Hall, J. W., Murgatroyd, A., Acreman, M., Bates, P., Beven, K., Heathwaite, L.,
Holden, J., Holman, I. P., Lane, S. N., O’Connell, E., Penning-Rowsell, E., Reynard, N.,
Sear, D., Thorne, C., Wilby, R. (2017). A restatement of the natural science evidence
concerning catchment-based “natural” flood management in the UK. Proceedings of the Royal
Society A: Mathematical, Physical and Engineering Sciences, 473(2199). https://doi.org/
10.1098/rspa.2016.0706
Davenport, F., Burke, M., Diffenbaugh, N. S. (2021). Contribution of historical precipitation
change to US flood damages. Proceedings of the National Academy of Sciences, 118(4).
https://doi.org/10.1073/pnas.2017524118. e2017524118.
Di Baldassarre, G., Kooy, M., Kemerink, J. S., Brandimarte, L. (2013). Towards understanding
the dynamic behaviour of floodplains as human-water systems. Hydrology and Earth System
Sciences, 17(8), 3235e3244. https://doi.org/10.5194/hess-17-3235-2013
Dibike, Y., Shakibaeinia, A., Eum, H. I., Prowse, T., Droppo, I. (2018). Effects of projected
climate on the hydrodynamic and sediment transport regime of the lower Athabasca River in
Alberta, Canada. River Research and Applications, 34(5), 417e429. https://doi.org/10.1002/
rra.3273
Elmer, F., Hoymann, J., Düthmann, D., Vorogushyn, S., Kreibich, H. (2012). Drivers of flood
risk change in residential areas. Natural Hazards and Earth System Science, 12(5),
1641e1657. https://doi.org/10.5194/nhess-12-1641-2012

Elmer, F., Thieken, A. H., Pech, I., Kreibich, H. (2010). Influence of flood frequency on resi-
dential building losses. Natural Hazards and Earth System Science, 10(10), 2145e2159.
https://doi.org/10.5194/nhess-10-2145-2010
Emerton, R., Cloke, H. L., Stephens, E. M., Zsoter, E., Woolnough, S. J., Pappenberger, F.
(2017). Complex picture for likelihood of ENSO-driven flood hazard. Nature Communica-
tions, 8. https://doi.org/10.1038/ncomms14796
Felder, G., Gómez-Navarro, J. J., Zischg, A. P., Raible, C. C., Röthlisberger, V., Bozhinova, D.,
Martius, O., Weingartner, R. (2018). From global circulation to local flood loss: Coupling
models across the scales. Science of the Total Environment, 635, 1225e1239. https://doi.org/
10.1016/j.scitotenv.2018.04.170
Felder, G., Zischg, A., Weingartner, R. (2017). The effect of coupling hydrologic and hydro-
dynamic models on probable maximum flood estimation. Journal of Hydrology, 550,
157e165. https://doi.org/10.1016/j.jhydrol.2017.04.052
Ferdous, M. R., Di Baldassarre, G., Brandimarte, L., Wesselink, A. (2020). The interplay be-
tween structural flood protection, population density, and flood mortality along the Jamuna
River, Bangladesh. Regional Environmental Change, 20(1). https://doi.org/10.1007/s10113-
020-01600-1
Fleischmann, A., Collischonn, W., Paiva, R., Tucci, C. E. (2019). Modeling the role of reservoirs
versus floodplains on large-scale river hydrodynamics. Natural Hazards, 99(2), 1075e1104.
https://doi.org/10.1007/s11069-019-03797-9
Fuchs, S., Heiss, K., Hübl, J. (2007). Towards an empirical vulnerability function for use in
debris flow risk assessment. Natural Hazards and Earth System Science, 7(5), 495e506.
Fuchs, S., Keiler, M., Zischg, A. (2015). A spatiotemporal multi-hazard exposure assessment
based on property data. Natural Hazards and Earth System Sciences, 15(9), 2127e2142.
Fuchs, S., Röthlisberger, V., Thaler, T., Zischg, A., Keiler, M. (2017). Natural hazard man-
agement from a coevolutionary perspective: Exposure and policy response in the European
Alps. Annals of the American Association of Geographers, 107(2), 382e392. https://doi.org/
10.1080/24694452.2016.1235494
Gaál, L., Szolgay, J., Kohnová, S., Parajka, J., Merz, R., Viglione, A., Blöschl, G. (2012). Flood
timescales: Understanding the interplay of climate and catchment processes through
comparative hydrology. Water Resources Research, 48(4). https://doi.org/10.1029/
2011WR011509
Gilvear, D. J. (1999). Fluvial geomorphology and river engineering: Future roles utilizing a fluvial
hydrosystems framework. In Geomorphology (Vol. 31, pp. 229e245). https://doi.org/10.1016/
S0169-555X(99)00086-0, 1e4.
Glur, L., Wirth, S. B., Büntgen, U., Gilli, A., Haug, G. H., Schär, C., Beer, J., Anselmetti, F. S.
(2013). Frequent floods in the European Alps coincide with cooler periods of the past 2500
years. Scientific Reports, 3. https://doi.org/10.1038/srep02770
Gregory, K. J. (2006). The human role in changing river channels. Geomorphology, 79(3e4),
172e191. https://doi.org/10.1016/j.geomorph.2006.06.018
Gual, M. A., Norgaard, R. B. (2010). Bridging ecological and social systems coevolution: A
review and proposal. Ecological Economics, 69(4), 707e717. https://doi.org/10.1016/
j.ecolecon.2008.07.020
Guan, M., Carrivick, J. L., Wright, N. G., Sleigh, P. A., Staines, K. E. H. (2016). Quantifying the
combined effects of multiple extreme floods on river channel geometry and on flood hazards.
Journal of Hydrology, 538, 256e268. https://doi.org/10.1016/j.jhydrol.2016.04.004

Gulbin, S., Kirilenko, A. P., Kharel, G., Zhang, X. (2019). Wetland loss impact on long term
flood risks in a closed watershed. Environmental Science and Policy, 94, 112e122. https://
doi.org/10.1016/j.envsci.2018.12.032
Hall, J., Arheimer, B., Borga, M., Brázdil, R., Claps, P., Kiss, A., Kjeldsen, T. R., Kriau
ci
unien_
e, J.,
Kundzewicz, Z. W., Lang, M., Llasat, M. C., Macdonald, N., McIntyre, N., Mediero, L., Merz,
B., Merz, R., Molnar, P., Montanari, A., Neuhold, C., . Blöschl, G. (2014). Understanding
flood regime changes in Europe: A state-of-the-art assessment. Hydrology and Earth System
Sciences, 18(7), 2735e2772. https://doi.org/10.5194/hess-18-2735-2014.
ICOLD International Commission on Large Dams. (2021). World register of dams. https://www.
icold-cigb.org/GB/world_register/general_synthesis.asp.
IPCC. (2012). Managing the risks of extreme events and disasters to advance climate change
adaptation: Special report of the Intergovernmental Panel on Climate Change.
Javaheri, A., Babbar-Sebens, M. (2014). On comparison of peak flow reductions, flood inun-
dation maps, and velocity maps in evaluating effects of restored wetlands on channel flooding.
Ecological Engineering, 73, 132e145. https://doi.org/10.1016/j.ecoleng.2014.09.021
Jongman, B., Winsemius, H. C., Aerts, J. C. J. H., Coughlan de Perez, E., van Aalst, M. K.,
Kron, W., Ward, P. J. (2015). Declining vulnerability to river floods and the global benefits
of adaptation. Proceedings of the National Academy of Sciences, 112(18), E2271eE2280.
https://doi.org/10.1073/pnas.1414439112
Kallis, G. (2007). When is it coevolution? Ecological Economics, 62(1), 1e6. https://doi.org/
10.1016/j.ecolecon.2006.12.016
Keiler, M., Knight, J., Harrison, S. (2010). Climate change and geomorphological hazards in the
eastern European Alps. In Philosophical transactions of the royal society A: Mathematical,
physical and engineering sciences (Vol. 368, pp. 2461e2479). Royal Society. https://doi.org/
10.1098/rsta.2010.0047, 1919.
Keiler, M., Zischg, A., Fuchs, S., Hama, M., Stötter, J. (2005). Avalanche related damage po-
tential- Changes of persons and mobile values since the mid-twentieth century, case study
Galtür. Natural Hazards and Earth System Science, 5(1), 49e58. https://doi.org/10.5194/
nhess-5-49-2005
Kienzler, S., Pech, I., Kreibich, H., Müller, M., Thieken, A. H. (2015). After the extreme flood in
2002: Changes in preparedness, response and recovery of flood-affected residents in Germany
between 2005 and 2011. Natural Hazards and Earth System Sciences, 15(3), 505e526. https://
doi.org/10.5194/nhess-15-505-2015
Kreibich, H., Seifert, I., Thieken, A. H., Lindquist, E., Wagner, K., Merz, B. (2011). Recent
changes in flood preparedness of private households and businesses in Germany. Regional
Environmental Change, 11(1), 59e71. https://doi.org/10.1007/s10113-010-0119-3
Lane, S. N. (2014). Acting, predicting and intervening in a socio-hydrological world. Hydrology
and Earth System Sciences, 18(3), 927e952. https://doi.org/10.5194/hess-18-927-2014
Lane, S. N., Tayefi, V., Reid, S. C., Yu, D., Hardy, R. J. (2007). Interactions between sediment
delivery, channel change, climate change and flood risk in a temperate upland environment.
Earth Surface Processes and Landforms, 32(3), 429e446. https://doi.org/10.1002/esp.1404
Larrauri, P. C., Lall, U. P. (2020). Assessing the exposure of critical infrastructure and other
assets to the climate induced failure of aging dams in the.
Li, Z., Fang, H. (2016). Impacts of climate change on water erosion: A review. Earth-Science
Reviews, 163, 94e117. https://doi.org/10.1016/j.earscirev.2016.10.004
Martius, O., Viviroli, D., Rössler, O., Zischg, A., Röthlisberger, V., Aemisegger, F., Schulte, L.,
Stuber, M., Pena, J. C., Ruiz-Villanueva, V., Molnar, P. (2020). Understanding flood trig-
gering mechanisms and flood risk changes.

Mazzorana, B., Zischg, A., Largiader, A., Hübl, J. (2009). Hazard index maps for woody
material recruitment and transport in alpine catchments. Natural Hazards and Earth System
Science, 9(1), 197e209. https://doi.org/10.5194/nhess-9-197-2009
Meng, X., Chang, J., Wang, X., Wang, Y., Wang, Z. (2019). Flood control operation coupled
with risk assessment for cascade reservoirs. Journal of Hydrology, 572, 543e555. https://
doi.org/10.1016/j.jhydrol.2019.03.055
Molnar, P., Burlando, P., Ruf, W. (2002). Integrated catchment assessment of riverine landscape
dynamics. Aquatic Sciences, 64(2), 129e140. https://doi.org/10.1007/s00027-002-8061-1
Monk, W. A., Compson, Z. G., Choung, C. B., Korbel, K. L., Rideout, N. K., Baird, D. J. (2019).
Urbanisation of floodplain ecosystems: Weight-of-evidence and network meta-analysis
elucidate multiple stressor pathways. Science of the Total Environment, 684, 741e752.
https://doi.org/10.1016/j.scitotenv.2019.02.253
Naylor, L. A., Spencer, T., Lane, S. N., Darby, S. E., Magilligan, F. J., Macklin, M. G., Möller, I.
(2017). Stormy geomorphology: Geomorphic contributions in an age of climate extremes.
Earth Surface Processes and Landforms, 42(1), 166e190. https://doi.org/10.1002/esp.4062
Oberndorfer, S., Sander, P., Fuchs, S. (2020). Multi-hazard risk assessment for roads: Proba-
bilistic versus deterministic approaches. Natural Hazards and Earth System Sciences, 20(11),
3135e3160. https://doi.org/10.5194/nhess-20-3135-2020
O’Connell, E., Ewen, J., O’Donnell, G., Quinn, P. (2007). Is there a link between agricultural
land-use management and flooding? Hydrology and Earth System Sciences, 11(1), 96e107.
https://doi.org/10.5194/hess-11-96-2007
Ouyang, Y., Leininger, T. D., Moran, M. (2013). Impacts of reforestation upon sediment load
and water outflow in the Lower Yazoo River Watershed, Mississippi. Ecological Engineering,
61, 394e406. https://doi.org/10.1016/j.ecoleng.2013.09.057
Pattison, I., Lane, S. N., Hardy, R. J., Reaney, S. M. (2014). The role of tributary relative timing
and sequencing in controlling large floods. Water Resources Research, 50(7), 5444e5458.
https://doi.org/10.1002/2013WR014067
Perdigão, R. A. P., Blöschl, G. (2014). Spatiotemporal flood sensitivity to annual precipitation:
Evidence for landscape-climate coevolution. Water Resources Research, 50(7), 5492e5509.
https://doi.org/10.1002/2014WR015365
Perera, D., Smakhtin, V., Williams, S., North, T., Curry, A. (2021). Ageing water storage
infrastructure: An emerging global risk. UNU-INWEH report series (Vol. 11).
Pollen, N., Simon, A. (2005). Estimating the mechanical effects of riparian vegetation on stream
bank stability using a fiber bundle model. Water Resources Research, 41(7), 1e11. https://
doi.org/10.1029/2004WR003801
Raven, E. K., Lane, S. N., Bracken, L. J. (2010). Understanding sediment transfer and
morphological change for managing upland gravel-bed rivers. Progress in Physical Geogra-
phy, 34(1), 23e45. https://doi.org/10.1177/0309133309355631
Roe, G. (2009). Feedbacks, timescales, and seeing red. Annual Review of Earth and Planetary
Sciences, 37, 93e115. https://doi.org/10.1146/annurev.earth.061008.134734
Rogger, M., Agnoletti, M., Alaoui, A., Bathurst, J. C., Bodner, G., Borga, M., Chaplot, V., Gallart,
F., Glatzel, G., Hall, J., Holden, J., Holko, L., Horn, R., Kiss, A., Kohnová, S., Leitinger, G.,
Lennartz, B., Parajka, J., Perdigão, R., . Blöschl, G. (2017). Land use change impacts on
floods at the catchment scale: Challenges and opportunities for future research. Water Re-
sources Research, 53(7), 5209e5219. https://doi.org/10.1002/2017WR020723.
Rogger, M., Chirico, G. B., Hausmann, H., Krainer, K., Brückl, E., Stadler, P., Blöschl, G.
(2017). Impact of mountain permafrost on flow path and runoff response in a high alpine

catchment. Water Resources Research, 53(2), 1288e1308. https://doi.org/10.1002/
2016WR019341
Rössler, O., Brönnimann, S. (2018). The effect of the Tambora eruption on Swiss flood gen-
eration in 1816/1817. Science of the Total Environment, 627, 1218e1227. https://doi.org/
10.1016/j.scitotenv.2018.01.254
Rössler, O., Froidevaux, P., Börst, U., Rickli, R., Martius, O., Weingartner, R. (2014). Retro-
spective analysis of a nonforecasted rain-on-snow flood in the Alps-A matter of model lim-
itations or unpredictable nature? Hydrology and Earth System Sciences, 18(6), 2265e2285.
https://doi.org/10.5194/hess-18-2265-2014
Röthlisberger, V., Zischg, A. P., Keiler, M. (2017). Identifying spatial clusters of flood exposure
to support decision making in risk management. Science of the Total Environment, 598,
593e603. https://doi.org/10.1016/j.scitotenv.2017.03.216
Rougier, J. (2013). Quantifying hazard losses. In J. Rougier, R. S. J. Sparks, L. Hill (Eds.), Risk
and uncertainty assessment for natural hazards (pp. 19e39). Cambridge University Press.
Ruiz-Villanueva, V., Bodoque, J. M., Dı́ez-Herrero, A., Bladé, E. (2014). Large wood transport
as significant influence on flood risk in a mountain village. Natural Hazards, 74(2), 967e987.
https://doi.org/10.1007/s11069-014-1222-4
Runyan, C. W., D’Odorico, P., Lawrence, D. (2012). Physical and biological feedbacks of
deforestation. Reviews of Geophysics, 50(4). https://doi.org/10.1029/2012RG000394
Schilling, K. E., Gassman, P. W., Kling, C. L., Campbell, T., Jha, M. K., Wolter, C. F.,
Arnold, J. G. (2014). The potential for agricultural land use change to reduce flood risk in a
large watershed. Hydrological Processes, 28(8), 3314e3325. https://doi.org/10.1002/hyp.9865
Schmocker-Fackel, P., Naef, F. (2010). Changes in flood frequencies in Switzerland since 1500.
Hydrology and Earth System Sciences, 14(8), 1581e1594. https://doi.org/10.5194/hess-14-
1581-2010
Seher, W., Löschner, L. (2018). Balancing upstreamedownstream interests in flood risk man-
agement: Experiences from a catchment-based approach in Austria. Journal of Flood Risk
Management, 11(1), 56e65. https://doi.org/10.1111/jfr3.12266. Blackwell Publishing Inc.
Sivapalan, M., Blöschl, G. (2015). Time scale interactions and the coevolution of humans and
water. Water Resources Research, 51(9), 6988e7022. https://doi.org/10.1002/2015WR017896
Slater, L. J. (2016). To what extent have changes in channel capacity contributed to flood hazard
trends in England and Wales? Earth Surface Processes and Landforms, 41(8), 1115e1128.
https://doi.org/10.1002/esp.3927
Slater, L. J., Khouakhi, A., Wilby, R. L. (2019). River channel conveyance capacity adjusts to
modes of climate variability. Scientific Reports, 9(1). https://doi.org/10.1038/s41598-019-
48782-1
Slater, L. J., Singer, M. B., Kirchner, J. W. (2015). Hydrologic versus geomorphic drivers of
trends in flood hazard. Geophysical Research Letters, 42(2), 370e376. https://doi.org/
10.1002/2014GL062482
Syvitski, J. P. M., Vörösmarty, C. J., Kettner, A. J., Green, P. (2005). Impact of humans on the
flux of terrestrial sediment to the global coastal ocean. Science, 308(5720), 376e380. https://
doi.org/10.1126/science.1109454
Thieken, A. H., Kreibich, H., Müller, M., Merz, B. (2007). Coping with floods: Preparedness,
response and recovery of flood-affected residents in Germany in 2002. Hydrological Sciences
Journal, 52(5), 1016e1037. https://doi.org/10.1623/hysj.52.5.1016
Tu, T., Ercan, A., Carr, K. J., Kavvas, M. L., Trinh, T., Ishida, K., Nosacka, J., Brown, K. (2020).
Coupling hydroclimate-hydraulic-sedimentation models to estimate flood inundation and

sediment transport during extreme flood events under a changing climate. Science of the Total
Environment, 740. https://doi.org/10.1016/j.scitotenv.2020.140117
UNISDR. (2015). Making development sustainable: The future of disaster risk management. In
Global assessment report on disaster risk reduction (Vol. 4)United Nations, 2015.
Vitousek, P. M., Mooney, H. A., Lubchenco, J., Melillo, J. M. (1997). Human domination of
Earth’s ecosystems. Science, 277(5325), 494e499. https://doi.org/10.1126/
science.277.5325.494
Wang, W., Yang, S., Hu, F., Stanley, H. E., He, S., Shi, M. (2018). An approach for cascading
effects within critical infrastructure systems. Physica A: Statistical Mechanics and Its Ap-
plications, 510, 164e177. https://doi.org/10.1016/j.physa.2018.06.129
Ward, P. J., Renssen, H., Aerts, J. C. J. H., Van Balen, R. T., Vandenberghe, J. (2008). Strong
increases in flood frequency and discharge of the River Meuse over the late Holocene: Impacts
of long-term anthropogenic land use change and climate variability. Hydrology and Earth
System Sciences, 12(1), 159e175. https://doi.org/10.5194/hess-12-159-2008
White, G. (1945). Human adjustment to floods. USA: Department of Geography, University of
Chicago.
Wirth, S. B., Glur, L., Gilli, A., Anselmetti, F. S. (2013). Holocene flood frequency across the
Central Alps - solar forcing and evidence for variations in North Atlantic atmospheric cir-
culation. Quaternary Science Reviews, 80, 112e128. https://doi.org/10.1016/
j.quascirev.2013.09.002
Wohl, E., Lane, S. N., Wilcox, A. C. (2015). The science and practice of river restoration. Water
Resources Research, 51(8), 5974e5997. https://doi.org/10.1002/2014WR016874
Yang, X., Sun, W., Li, P., Mu, X., Gao, P., Zhao, G. (2018). Reduced sediment transport in the
Chinese Loess Plateau due to climate change and human activities. Science of the Total
Environment, 642, 591e600. https://doi.org/10.1016/j.scitotenv.2018.06.061
Zhao, G., Bates, P., Neal, J. (2020). The impact of dams on design floods in the conterminous
US. Water Resources Research, 56(3). https://doi.org/10.1029/2019WR025380
Zischg, A. P., Felder, G., Mosimann, M., Röthlisberger, V., Weingartner, R. (2018). Extending
coupled hydrological-hydraulic model chains with a surrogate model for the estimation of
flood losses. Environmental Modelling and Software, 108, 174e185. https://doi.org/10.1016/
j.envsoft.2018.08.009
Zischg, A. P., Felder, G., Weingartner, R., Quinn, N., Coxon, G., Neal, J., Freer, J., Bates, P.
(2018). Effects of variability in probable maximum precipitation patterns on flood losses.
Hydrology and Earth System Sciences, 22(5), 2759e2773. https://doi.org/10.5194/hess-22-
2759-2018
Zischg, A. P., Hofer, P., Mosimann, M., Röthlisberger, V., Ramirez, J. A., Keiler, M.,
Weingartner, R. (2018). Flood risk (d)evolution: Disentangling key drivers of flood risk change
with a retro-model experiment. Science of the Total Environment, 639, 195e207. https://
doi.org/10.1016/j.scitotenv.2018.05.056
Zoccatelli, D., Borga, M., Viglione, A., Chirico, G. B., Blöschl, G. (2011). Spatial moments of
catchment rainfall: Rainfall spatial organisation, basin morphology, and flood response. Hy-
drology and Earth System Sciences, 15(12), 3767e3783. https://doi.org/10.5194/hess-15-
3767-2011

Chapter 3
Disentangling drivers of
change
Flood risk is composed of multiple factors, summarized by the three main
factors: hazard, exposure, and vulnerability. These three main factors of risk
vary in space and time. The changes of these main risk factors are influenced
by a high number of drivers of change. As the previous chapter has shown,
drivers of change comprise changes in the environment as well as changes in
society. The drivers of change are furthermore interconnected, they coevolve
with each other. Therefore, causes and effects of flood risk change are not
always clearly distinguishable by statistical inference methods. An example is
an expected correlation between the number of houses constructed annually in
a floodplain and the increase in flood risk. Flood risk is not increasing linearly
with the number of exposed houses but with the number of exposed houses
that are vulnerable to flooding and with the exposure to different hazard in-
tensities. If a land use regulation foresees the obligatory installation of object
protection measures such as sealing the openings of the house to avoid water
entering into the house up to a specific flood water level, the relationship
between the number of newly constructed houses and the flood risk in the
floodplain is nonlinear. Moreover, statistical inference methods require spe-
cific data. In this example, the increase of flood risk in a growing settlement
area is not directly deductible from data, e.g., from house registers or from
insurance claim data.
Moreover, the main aim of analyzing flood risk change is to learn about
how the different factors drive the increase or decrease of flood risk. Knowing
the key drivers of risk and their changes in space and time enables decision-
makers to target the most promising levers for flood risk reduction. Thus,
flood risk reduction measures can be targeted at the most influencing driver of
change. The key drivers of change heavily depend on local context. While in
some floodplains, settlement growth might be the most important driver of
change, climate change, or other environmental changes might outpace soci-
etal drivers of flood risk change. The request to learn about the most influ-
encing driver of change requires to look at the different drivers of change
individually. This requires a method that enables the analysis of the effects of
single risk drivers on overall flood risk in an isolated mode. The in-
terdependencies between the drivers of change must be artificially eliminated

Random documents with unrelated
content Scribd suggests to you:

he sat down and wrote an extremely diverting and trenchant little
book entitled “The Vampires of London.” Herein the methods of
usury were exposed in a fierce light. This, however, the wily Jew
might have forgiven. What he could never forgive was the ridicule
which the gallant officer threw on his ménage. He had invited his
customer to accept the hospitality of his home, and now the secrets
of that home were held up to public ridicule and contempt. The
writer had not spared the members of the family. The very children
of Israel were sacrificed on the altar of John’s vengeance. The
allurements of Rachael, the schemes of “blear-eyed Leah,” were set
forth with fiendish particularity.
The trial came off at the Old Bailey, and the prosecutor was
represented by a rising barrister called Mr. Hardinge Giffard. That
rising young barrister has, in so far as the Bar is concerned, risen
and set many a day ago. He is now Lord Halsbury. The jury found
for the persecuted Hebrew. The Hon. John was sentenced to certain
months in gaol as a first-class misdemeanant, and ordered to pay a
heavy fine. Defendants in cases of the kind were not so closely
watched in those days as they are in the present year of grace, and
when Mr. Colborne was called upon to receive sentence he was
nowhere to be found. Having a very clear notion of the sort of
verdict the jury would give, he had skipped over to France earlier in
the day.
John had carried with him across the Channel a new and enlarged
edition of “The Vampires,” and he at once set about issuing copies
by post to advertisers desiring to acquire a work about which the
trial had set all the town talking. To stop this fresh persecution,
plaintiff was willing to accept any sort of terms in reason. All that
Mr. Colborne desired was liberty to return to his native land, to
obtain cancellation of the excessive interest on his bills, and to live
thenceforth in peace with all men. His friends were enabled to
arrange terms on this basis, and John was free to prosecute those
schemes for improving the condition of his fellow-man to which he
purposed to devote his energies. His schemes were fated to “gang

agley.” He joined the Egyptian army, and died in action. It was
probably the kind of death he would have wished, for, however he
may have proved wanting in other qualities, no one ever doubted his
high courage.
Chinery, in his club promotions, aimed at higher game. He had
served as Consul-General in a West African State, was a member of
the Reform and the Devonshire, was a convinced Liberal, and had a
wonderfully good connection. Owing to these circumstances, he was
able to muster a much stronger committee than others who had
started before him in the club industry. His first venture was the
Empire Club. For this establishment he had acquired what the
auctioneers call “eligible” premises. He got a lease of the house in
Grafton Street, Piccadilly, which had been the last home of Lord
Brougham. Men like the late (and great) Marquis of Dufferin
became members. Viscount Bury was President of the club. A large
membership, including many leading colonials, was assured. The
management was reliable, the cellar unimpeachable, the house
dinner (always presided over by a colonial Governor-General or some
other potentate interested in our overseas Empire) became a
welcome feature, and a long spell of prosperity seemed to be ahead
of us. But our hopes did not reach fruition. Something went wrong
with the accounts, and the Empire closed its doors.
The festive Chinery, in no whit discouraged, started on fresh
promotions. None of them achieved the brilliant reputation of his
original venture, and Chinery himself died a broken man.
At one time I belonged to a club called the Wanderers, in
compliment, I suppose, to the Travellers, which was nearly
opposite. The club-house occupied the corner, on the other side of
Pall Mall, corresponding to that of the Athenæum. This was a
comfortable and well-found establishment. Tod Heatley, the wine-
merchant, was supposed to be interested in it; but it passed through
many vicissitudes, and went under many names, till it was eventually
devoted to more profitable purposes. Although the Wanderers had
always other and higher pretensions, it was essentially a Bohemian

club. A mixture of such pretensions with such actualities should be
foredoomed to failure. In clubland the Wanderers was known as
“The Home for Lost Dogs.”
Chief among the genuine Bohemian clubs is the Savage Club, whose
home is on the Adelphi Terrace. Although the Bohemianism of this
famous club is mainly traditional, it preserves the good custom of
general communication among members, and encourages that spirit
of playful geniality which is inseparable from the idea of
Bohemianism. But the Savage Club of to-day is a very different
thing from the same association as I knew it in 1870. This, indeed,
will be admitted by the official historian of the club, Mr. Aaron
Watson, whose admirable monograph on the Savage leaves nothing
for any future writer to tell concerning the genesis and early
struggles of the Savages.
I was a guest at the Savage on about half a dozen occasions in early
years, and I once passed a few hours with Christie Murray in its new
and more abiding home.
It was on a dull November day, and Pat Macdonald and I were
walking westward from Fleet Street. We had taken Covent Garden
on our way. “Let’s see if there’s anybody in the Savage Club,” he
said casually, as we left the central avenue of the market, under the
shadow of St. Paul’s, of the convent garden. To me the invitation
was delightful. Often I had heard of the celebrated resort of actors,
authors, and musicians. With the rest of the world, I had become
impressed with the idea that election to this coterie was extremely
difficult. I had read with much interest the first issue of “The
Savage Club Papers,” and it came upon me as a surprise that my
friend Macdonald, whose contributions to literature were of the most
tenuous character, should be a member, and that he should hold his
membership so lightly.
Soon I discovered the reason, and this, by the way, is a rather
interesting morsel of history which has escaped the vigilant eye of
Mr. Aaron Watson. In those early and unsophisticated days, when a

man was put up for membership at the Savage, he was given the
run of the club until the date of the next election; and some men are
by nature such excellent company that a club existing above all
other things for congenial companionship will be apt to regard the
claims of the professionally unqualified candidate as above those of
the highly qualified man who happens to be a dull dog. This month
of probation afforded the good fellow—“the clubbable man” of Dr.
Johnson—the opportunity of asserting his claims; and although the
committee was bound by its first rule, which provided that only men
professionally connected with literature, the drama, or the arts,
should be eligible, when they got the chance of electing a man of
Macdonald’s erudition, humour, and powers of conversation, they
were not likely to give that chance away. It was a strange rule, but
it worked well. In those days there was no place in a club forced to
forgather in a single room for men who could not talk well and laugh
loudly.
Under the guidance of my friend, I crossed to the right through the
inevitable slush and vegetable refuse, and we were soon mounting
the steps that led to Evans’s Hotel. With the celebrated Supper-
Room beneath the hotel I was already acquainted, but I had never
before visited the hotel. Nor did I for a moment imagine that the
club which occupied so large a place in my fancy and my esteem
occupied rooms on licensed premises. The Savage Club was in
possession of the room on the left of the hall as you entered the
hotel. It had originally been the coffee-room, and was one of the
principal apartments in the building. Evans’s Hotel is now the
National Sporting Club. It was first the Falstaff, and to fit it for its
new purposes considerable structural alterations were necessary,
including a small private theatre, now abolished, but the lines of the
old home of the Savages can still be made out.
There were very few members present on the occasion of this first
visit of mine, and I was reminded of the omnipresence of the legal
profession on finding that two of them were barristers. One was Mr.
Jonas Levy, Chairman of the London, Brighton, and South Coast

Railway; and the other Mr. Hume Williams—not the K.C. and
Recorder of Norwich, but the father of that learned gentleman.
Another of those present was Henry S. Leigh, the author of “The
Carols of Cockayne”—a gentleman whom I came to know intimately.
He had the bitterest tongue and sweetest nature of any man I ever
met. The arrangements of the room testified to the simplicity of
taste observed by those primitive Savages. On the tables that lined
the walls were laid out clay pipes of the shape and size with which
we associate the name “churchwarden,” and I observed that Leigh
was drinking beer out of a pewter pot. There are no pewter pots in
the Savage Club nowadays, but neither are there any Leighs.
Whether it was the deadly dulness of the autumn afternoon or my
own lack of responsiveness, or whether it was that I had cherished
exaggerated expectations, or whether it was the result of a
conspiracy of all these causes, I cannot say, but my first visit to the
Savage was a disappointment and a disillusion. A year or more went
by before I was afforded an opportunity of reviewing my earlier
impressions. This time I had no cause to complain of the quality of
the entertainment. “Jimmy” Albery, who had recently made his
name with “Two Roses”; H. S. Leigh; E. A. Sothern; George Honey,
the actor; Arthur Boyd Houghton, the artist; and Andrew Halliday,
the author and journalist-dramatist, were among those present. My
earlier impressions were at once erased. Never had I been thrown
into the society of a number of grown men where such a spirit of
fun, of camaraderie, of irresponsibility, and of the joy of life,
prevailed and sparkled. They talked in the spirit of schoolboys, but
with the point of seasoned wits. It was altogether a delightful
experience.
It was at the Savage Club that I first saw the game of poker played.
The game had been introduced by some Americans who enjoyed the
privileges of corresponding membership in respect of their
connection with the Lotus Club, New York. It was shortly made
taboo by a ukase of the Portland and Turf Clubs, and disappeared
from the card-rooms of all the West End clubs. I have always

thought this rather a pity. Poker is one of the best games to be got
out of a pack. It calls into exercise other faculties beside memory,
judgment, skill, and a nice knowledge of the value of cards. You
want to be a bit of a physiognomist. Your own expression should be
under control, and your manner absolutely inscrutable. It is in
respect of their natural endowment in these qualities that the
Yankees make such good poker-players. I became greatly interested
in the game, and it was indirectly through my instrumentality that its
rules were first published in this country. General Schenk drew up
the enactments governing the science of the pastime, at the request
of Lady Waldegrave. Lady Waldegrave had them set up in type at
Strawberry Hill. She had a few dozen copies printed for the use of
her acquaintances. I became the proud possessor of one of these
copies. A friend of mine—or perhaps I should say a gentleman
whom up to that time I had regarded as a friend—induced me to
lend him the brochure to settle some dispute which had arisen
between certain correspondents on his paper; for my friend was a
rather distinguished writer on the sporting press. I never saw that
book again, but to my intense surprise and chagrin I found the
whole of the Strawberry Hill rules published in the columns of my
friend’s paper, with their place of origin given, and Lady
Waldegrave’s authority cited.
The transaction did more harm to the gentleman who had betrayed
my confidence than it did to me. In those days an act of the kind
would be generally reprobated. Dog did not eat dog when Plancus
was Consul. Nowadays I am given to understand that it would be
regarded as a bit of smart journalism.
As I write, the memory of that first game of draw-poker comes
vividly back to me, and, singular as it may seem to you, it comes
back to an accompaniment of music. It was night, and in the
supper-room below and at the back the little pale-faced choristers in
their Eton suits were singing glees for Paddy Green’s customers.
These vocal exercises were resented by grumpy members of the
club, but to me distance enhanced the beauty of the singing, and I

never hear poker mentioned now, such is the strange influence of
the association of ideas, that I do not instantly hear the far-away
voices of boys singing:
“Oh, who will o’er the downs with me—
Oh, who will with me ride?
Oh, who will up and follow me
To win a blooming bride?”
Poor words, perhaps; set to old-fashioned glee music, no doubt;
introducing in the last line a word rendered vulgar by a merciless
modernity, admitted. But, Lord! how sweet the memory of them
comes back to me over the years—how inexpressibly sweet, yet how
incalculably sad! for nothing but the haunting memory is left. My
contemporaries of that time have, nearly all of them, satisfied their
curiosity concerning the Great Secret. The pale-faced choir boys
have grown to manhood, developing, perhaps, into “fat and greasy
citizens.” Only the song remains.
Baker Green, editor of the Morning Post, was a member of the
Savage at a somewhat later date. He was a great hulking figure of a
man, with a terrible mordant humour of his own, and a devilish
solemn manner of stating the most absurd propositions. His
monocle was as inseparable from him as that of Sir Squire Bancroft.
His peculiar style of humour may be best illustrated anecdotically.
A member who loomed large in the life of the club in the days when
the Imperial Institute was being nursed into life was Somers Vine.
In respect of his services rendered to the Institute the excellent man
received the honour of knighthood. It is to be feared that Baker
Green had no great liking for Sir Somers. Of this sentiment on the
part of his fellow-member, Vine, it must be supposed, had no inkling,
for one evening, bubbling over with hospitality and brotherly
kindness, he approached Baker Green in the club.
“I wish, my dear fellow, you would come down and spend a week at
my place at Chislehurst,” he said.

“Delighted,” replied the other.
“I live at Vine Court,” explained the knight.
Baker Green took out his pocket-book as if to make a note.
“What Court did you say?” he asked innocently.
“Vine Court,” replied the pleased Sir Somers.
“Yes—er—and what number?” inquired the remorseless Green.
It is perhaps needless to add that the proposed visit was never paid.
Sir W. S. Gilbert was an occasional visitor at the supper-rooms
beneath the club. The incident I am about to relate is scarcely
relevant to the subject with which the present chapter deals, but as
it happened on the premises, so to speak, I may be pardoned for
introducing it. At Evans’s it was the custom to pay for your supper
to a waiter who stood at the door—a lightning calculator who, by the
means of a legerdemain which was all his own, was able to add
about 25 per cent. to every bill without the victim being able to see
exactly how it was done. Gilbert rather resented the arithmetical
methods of “John,” and at last came to the determination to pay
“John” off by tipping him a penny instead of the sixpence which had
hitherto been his pourboire. On the night on which his resolution
was to be carried into effect his bill amounted to exactly hall a
crown. He handed that coin to the magic calculator, and then
handed his tip of one penny. “John” looked at the coin, smiled a
deprecating smile, and, handing it back to the donor, said in a tone
of subdued solicitude: “Perhaps you may be going over a bridge, sir.”
There was a toll levied on those crossing Waterloo Bridge in those
days. The retort hit in two ways. The first suggestion was that the
gentleman lived at the other side of the water; and the second, that
he had been reduced to his last copper. The comment was, in fact,
quite Gilbertian—as “John” himself was perfectly well aware.

The doyen of the club was W. B. Tegetmier. He seemed a survival
almost of another age. For he was the same W. B. Tegetmier to
whom Darwin, in his “Descent of Man,” makes so many
acknowledgments of assistance in connection with experiments in
the breeding of pigeons. He was one of the first men to use the
bicycle as a means of getting to and from his office at the Field,
which was then in the Strand. He must have been well over sixty at
the time, and he continued to use the machine till he was well over
seventy. A wonderful, wiry, active, peppery-tempered little man with
a kindly expression indicating a heart more kindly still. Not that he
could not say a hard thing when he thought it absolutely necessary.
By his intimates he was always called “Teg.” But should any man
who was not an intimate presume thus to address him, he would
quickly resent the familiarity. Thus, on one occasion Mr. Bowles, a
barrister and brother-Savage, finding the little naturalist there,
addressed him by his sobriquet.
“Hallo, how are you, Teg?” said the devoted man, bent on geniality.
“Quite well, thank you—Po!” answered the other icily.
I had the honour of attending two of the Saturday dinners of the
Savage Club. There was nothing quite like those dinners then; there
has been nothing quite like them since. No after-dinner speeches
were permitted, but when the meal—a very simple one—was at an
end, the members set about entertaining their guests and
themselves by song, anecdote, recitation, imitation, and playing
upon instruments—for some of the finest instrumentalists in England
were Savages. Old George Grossmith—father of George Grossmith,
the well-known illustrator of Gilbert and Sullivan opera and platform
entertainer, and grandfather of George Grossmith junior of the
Gaiety Theatre—gave us a reading from the first chapter of “Bleak
House”; Signor Foli sang “Simon the Cellarer”; Oscar Barrett and
John Radcliffe fluted to us; Hamilton Clarke presided at the piano;
Charles Collette pattered; George Honey gave some side-splitting
stories, ably seconded in this department by dear old “Lal” Brough.
The whole thing went with a “zip.” There was no hesitation on the

part of performers; the neophyte who “broke down” in his
performance was as heartily cheered as the veteran who rendered a
passage reserved for such a gathering. Indeed, the feeling that one
was listening to an entertainment which the public could not have
for love or money added not a little, I imagine, to the sense of
pleasure in those who took part in the post-prandial entertainment.
The Arundel and the Wigwam were conducted much on Savage
lines, and the Junior Garrick, to which I have made reference in an
earlier chapter, was decidedly a Bohemian institution. It had two
periods. It originally existed as a members’ club; but a large
number of influential members quarrelled with the committee and
withdrew. The financial position of those who remained was not
sufficiently strong to justify them in continuing it. And it seemed a
pity to close the doors; for the club occupied a fine house at the
corner of Adam Street and Adelphi Terrace. It remains an excellent
example of Adam architecture, and contains some magnificent Adam
ceilings and cornices. The drawing-room on the first-floor, with its
unrivalled view of the Thames, is a spacious and well-proportioned
apartment. The room beneath it was our dining-room, and the
billiard-room was at the top of the house.
Now, whereas the Savage never suffered from any schism, the
Junior Garrick was the victim of no less than two. The first while it
was a members’ club; the second, when it had become a proprietary
club. The first offshoot organized itself into the Green-Room Club,
which flourishes to this day, and is at present housed in Leicester
Square, nearly facing the Alhambra. This is now the principal club,
entirely composed of stage professionals. The second offshoot of
the old “J.G.C.,” as we liked to call it, was the Yorick. I know the
Yorick still exists, for I recently saw in the daily Press a letter dated
from that address.
In these days the Bohemian thinks it no longer good form to roam
around the town attired in the negligent seediness of the
impecunious student of the Quartier Latin. Unkempt locks, extreme
squalor, and dirty finger-nails, are no longer regarded as essential

characteristics of the social Bohemian. In the process of evolution
we have now arrived at the evening-dress Bohemian. The Eccentric
Club at Piccadilly Circus is his chosen resort. The phenomenal
success of this club is attributable to the fact that the principal
members of the original committee were business men; that it has
been enabled to develop on a very small capital—some £700, I
think; and that it was so fortunate as to acquire the premises,
furniture, and fixtures, of an expiring institution at a ridiculously
small figure.
This flourishing society grew out of the ashes of the old Coventry, a
proprietary club which existed for some years in Coventry Street.
When that rather cosy resort went the way of all proprietary clubs, a
few of us met at Rule’s, in Maiden Lane, with a view of seeing
whether a sufficient number of old Coventry members could not be
induced to found another social centre in which men who had for
some years come to regard the Coventry as their ordinary place of
meeting. The idea caught on. The title “Eccentric” was decided on
at our very first meeting. The old premises of the Pelican were to be
had on reasonable terms. And we commenced, with a good list of
members, in those sacred precincts. Among the actors who joined
were “Lal” Brough and Arthur Roberts, and among the artists were
Phil May, Julian Price, and Paleologue. The last-named gentleman
adorned the walls of the club-house with some very spirited mural
decorations. So spirited, indeed, was the fresco from the atelier of
Paleologue, that when the club gave what were called “ladies’ days”
Paleologue’s canvas had to be removed for the occasion. Knowing
who some of the ladies were, and understanding something also of
the characteristics of the committee-men who succeeded in carrying
this proposal, the arrangement always struck me as being
particularly quaint and insular.
One of the paintings of Julian Price was an inimitably clever likeness
of Drummond, our head-waiter. No man was ever half so
respectable as Drummond looked; and Price has caught his mild,
inquiring, deprecatory expression to a nicety. His trim black

whispers increase the pallor of his face, and, to mark the members’
appreciation of his high reputation, the artist has endowed him with
a halo. We had taken Drummond on from the Raleigh Club. In
carrying out his duties, Drummond was unaffected by the
circumstances passing around him. The most mirth-provoking joke
might be let off in his presence, but Drummond never turned a hair.
When joking took a practical turn, and when he became the subject
of the joke, affairs took on another complexion. And Drummond’s
reason for resigning at the Raleigh was—or was said to be—that
Lord Marcus Beresford, in an access of boyish irresponsibility, had
put Drummond into the ice-chest, shut the lid on him, and had then
forgotten all about him. Fortunately, another waiter had occasion to
go to the refrigerator before a fatality occurred, or poor Drummond
would have become just so many pounds of frozen meat.
This extraordinary man, notwithstanding his serious mood, was the
most painstaking, obliging, and solicitous club waiter I have ever
met. He understood the gastronomic tastes of every member, and
was infinitely desirous of giving satisfaction. He had one or two
curious methods of pronunciation; I believe they had been imposed
on him by facetious members of the Raleigh. Thus, he always said
“sooty” instead of “sauté.” It became quite a habit to ask
Drummond what potatoes were ready, for the sake of hearing his
quaint version: “What potatoes to-day, Drummond?” “Potatoes, sir?
There’s biled, mashed, and sooty.”
Drummond’s reason for accepting service at clubs which remained
open all night long, and frequently until four and five in the morning,
was a singular one. It seems that he was a proper religious man,
and held the office of deacon in connection with some conventicle in
the suburbs. In accepting a position in a club where all-night
sittings were the rule, he was free for every Sunday. I have seldom
heard of a man sacrificing more for his religion—have you? If
Drummond be still alive, he must be an old man by now, and may
his declining years be peaceful! If he be dead, may the turf lie light
on him!

The safeguard of a strong committee will never stand between a
proprietary club and eventual extinction. One of the strongest
committees I have known was got together by Mr. Earn Murray when
he founded the United Arts Club. The promoter was enterprising,
sanguine, and ambitious. But the only two private members of the
club who ever succeeded in achieving notoriety were “Old Solomon,”
the racing tipster, and Percy Lefroy, the murderer of Mr. Gold.
Our legislature, which always does things in a grandmotherly sort of
way, thought to purify the West End and suppress the Cyprian by
closing the night-houses in the Haymarket and in the streets
impinging thereon. The abolishing of those squalid dens did not,
indeed, result in her disestablishment, but in the betterment of the
conditions under which she carried on her sad but—if the unco’ guid
will permit the use of the word in this relation—necessary calling.
Phryne, like the poor, we shall always have with us. The obvious
duty of society, therefore, is not to take measures for her
suppression, but measures for her amelioration and regulation.
School Board education and an acquired knowledge of the laws of
hygiene have done much for her. When one compares the toilet, the
costume, and the manners, of the demi-mondaines who nightly
frequent the back of the dress-circle of certain houses of
entertainment with the tawdry, over-painted, giggling, solicitous
creature of thirty years ago, then, and only then, can one
understand the gratifying change that has taken place in the
habitude of this inalienable excrescence on the body politic.
When the night-houses were closed, and the police instructed to
keep the West End streets clear at midnight, there opened, here and
there, clubs for the accommodation of Phryne and her friends. So
that the closing of the frowsy saloons in which she had been wont to
congregate was a blessing in disguise, and, indeed, fixes the date of
the gratifying amelioration in her manners. For in the clubs a certain
decorum was observed even in the ballroom, which afforded the
raison d’être of social rialtos of the small-hours. The proprietors saw
to that; for the recurrence of disturbance or the report of sinister

incidents might occasion a raid. Election to these clubs was not, as
may well be supposed, a very difficult matter. One was proposed on
the doorstep, seconded on the hall mat, and unanimously elected a
member in the cloak-room. But the men “on the door” knew
perfectly well whom to admit and whom to dismiss. The bully, the
exploiter of frailty, the souteneur, were kept ruthlessly outside. Thus
the proprietor protected at once himself and his customers. He ran
a sort of bon marché in fact, where no middleman operated between
the goods and the patrons of the exchange.
The children of Israel—whose mission in these later years is to be
both our paymasters and our panders—were particularly zealous in
the promotion of this kind of réunion bohémiene. Belasco opened
the Supper Club in Percy Street, Tottenham Court Road. Sam Cohen
provided the “Spooferies” in Maiden Lane. He had previously run
the concern as a baccarat club, its useful career in that direction
having ended in a raid, and a prosecution of the greatest number of
persons ever called up at Bow Street to answer a single charge.
Sam must have been a bit of a cynic in his way, for the house in
which the “Spooferies” met was next door to the Jewish synagogue.
A Hebrew named Foster established a similar place in Long Acre,
and a coreligionist of his called Moore—a euphuism, I apprehend, for
Moses—opened the Waterloo Club in Waterloo Place, Pall Mall.
There were others. But those I have named are the only ones of
which I had a personal knowledge. This admission may, I fear,
horrify those readers who are of the dawn of the century. I can
assure my prudish friends, however, that were I mischievously
inclined I could give them a list of names of persons who were at
one time young men about town, but who now occupy prominent
positions in the Senate, at the Bar, and, generally speaking, in the
public life of the country, who were to be seen, in the jocund years,
thoroughly enjoying themselves in such Bohemian society as was to
be found at the “Spooferies” or the Supper Club.
I can see—in my mind’s eye, Horatio—some adipose, sleek, and
eminently respectable householder, some Member of Parliament,

London County Councillor, West End physician, fashionable painter,
or what not, who has taken up these reminiscences to while away an
hour. I can see this staid citizen, this respectable family man, this
stickler for morality, this Justice of the Peace, and all the rest of it,
squirming as he reads the above passage. With a blush he lays
down the book, and, looking suspiciously around, murmurs: “Damn
the fellow, he means me!” Yes, I undoubtedly mean you. But you
may read on without apprehension, my excellent friend, for I am the
soul of discretion. Your early trespasses are safe. In return I would
only ask this: that, remembering that you and I have sown some
wild-oats in the same fallows, you should exercise a little more
common-sense and charity in dealing with the peccadilloes of your
juniors, and that, generally speaking, you would carry yourself with
a less pompous air of conscious rectitude.

CHAPTER XI
THE JOKER
There are jokers and jokers. Professors of the art of practical joking
are disappearing before an advancing civilization like the Red Indian
of the Far West. The evanishment of the verbal joker is due to a
deplorable shrinkage in the national sense of humour. There will
soon be left to us the joker which is the fifty-third card in the pack,
and is incapable of any sense or emotion whatever.
But in the days of my vanity grown men carried with them into a
tun-bellied middle-age the fine flow of animal spirits and inordinate
capacity for fun which nowadays would be deprecated by the well-
regulated schoolboy. In Fleet Street one would have thought that
there would have been no time for any joking beyond an occasional
interchange of verbal pleasantries. But even in that busy
thoroughfare the practical joker found—or made—occasions for the
exercise of his fearsome talents.
It is something of a truism to say that the real man is very seldom
the man as he is observed in his public appearances. Who, for
instance, who only knew Edmund O’Donovan as the learned writer
of travel articles in the Quarterly Review, the accomplished special
correspondent of a one-time influential daily, the honoured guest of
savants, the respected lecturer before Royal Societies—who, I say,
who saw O’Donovan with his Society war-paint on could have
imagined the wild, undisciplined, half-mad, but wholly delightful
creature that was exhibited at intervals to Society in conventional
garb. He was the maddest and the most modest Irishman I ever

met. When he returned from his extraordinary adventures in Merv,
he did not put up at some swagger hotel in London, where he would
be easily accessible to Society intent on making him the lion of a
season. He lodged at a public-house in Holborn kept by a fellow-
countryman of his, named Peter Cowell. This house was at the time
known to the police in connection with the visits of Irish patriots of
the physical force party in national politics. It was the resort of the
scattered remnants of a disintegrated Fenianism.
Cowell revered his strange guest, and when customers heard the
sounds of revolver practice in the upper part of the house, you may
be sure that he did not give his patrons the true explanation of the
noise. The fact was that O’Donovan, in bed at midday, had grown
greatly annoyed at the crude art evinced in the engravings that
Cowell had hung upon his walls, and that he was engaged in
shooting those masterpieces into smithereens. This revolver practice
in his bedroom only ceased when there was nothing breakable left to
fire at. “Glory be to God!” said Peter Cowell, in relating the
circumstance to a correspondent, “there’s not a pictur’ nor a frame
nor a utinshill of anny kyoind that Misther O’Donovan hasn’t bruk an’
ped for!”
Two foreign gentlemen who refused to give their names, but who
had some important intelligence to convey, called at my office. I
signalled down that I would see them. I expected men in European
garb. But the two weird creatures who shuffled into my sanctum
were clothed in undressed animal skins reaching almost to their
feet. They were shod in the same material. And their head-dress
was also a fur so fashioned that only the eyes and nose of the
individuals were visible. The curious part of the equipment was that
the visitors carried pistols in their skin belts. I think that it was this
little circumstance that “gave the show away.” I looked very hard at
the taller of the two men, and then, feeling sure in my surmise, I
said cheerily:
“My dear O’Donovan, how are you? I’m delighted to see you.”

“Faith, I knew you’d know me!” he declared, in a tone that entirely
disguised his disappointment. “Come out and have a drink.”
Now, this hospitable invitation placed me in something of a
dilemma. For in the first place I did not wish to offend O’Donovan
by refusing, and in the second I had no desire to walk up Fleet
Street in the company of companions so strangely clad. I suggested
that, if O’Donovan and his friend would go on to the “Cheese,” I
would follow when I had finished writing the letter on which I was
then busy.
“That’s a beastly picture of Dizzy,” said O’Donovan quietly. He had
taken his revolver from his belt, and was pointing with it to “Ape’s”
cartoon of Beaconsfield which hung opposite my desk.
I understood the hint. I rose and accompanied my remorseless
friend. My worst anticipations were realized when I reached the
office door. Quite a large crowd of Fleet Street loafers—and I think
that in the Street of Adventure we could have boasted of as many
loafers to the square yard as any thoroughfare in London—pressed
round the door. The Fleet Street loafer is often exhilarated by the
sight of strange visitors; but he had never yet seen visitors quite so
strange as these. The crowd did not make any demonstration. But
Cockney criticisms of the general appearance of my companions
were freely bandied about. We had to cross the street and
encounter the jibes of cab-drivers and omnibus cads. The crowd
followed us right up to the doors of the tavern to which I had been
invited. Here was another assembly. For O’Donovan had already
visited the Cheshire Cheese, and had announced his intention of
returning to lunch. I believe that old Moore had during that
afternoon the most anxious time of his life. The fun waxed fast and
furious. But there is safety in a multitude of any kind, and the
intrepid traveller had so many friends and admirers in this gathering
that I was soon able to slip away unnoticed.
The man who accompanied O’Donovan on this occasion was Frank
Power—one of the most accomplished humbugs that ever made a

way in life by means of a glib tongue, a vivid imagination, and an
entire absence of scruple of any kind. O’Donovan subsequently
engaged him as secretary, and he was to have accompanied his
employer during the march with Hicks Pasha. It was characteristic
of Power that when the march was made Power remained behind in
Khartoum. He was once mentioned in the House of Commons. A
question was asked by an Irish Member as to the qualifications of
Mr. Frank Power, who had contrived to get himself made British
Consul at Khartoum. Mr. Gladstone, whose imagination was at times
as vivid as that of Power himself, replied promptly that the
gentleman in question was an “esteemed merchant” of that city.
In letters home, O’Donovan freely expressed his belief that the
chances of his ever returning to England alive were extremely small.
It is inconceivable that he should not have communicated this
opinion to Power. That young gentleman, holding that discretion is
the better part of valour, had an attack of dysentery at the very
moment when his services should have—under ordinary
circumstances—become of any value to his chief. He did not
accompany the intrepid column that marched across the sands to
inevitable and complete annihilation. As to O’Donovan, I know that
he died as he would have wished to die. No survivor of that ill-fated
expedition was allowed to escape with the story of the fight. But I
can picture O’Donovan in the midst of the mêlée, his eyes bright
with the fury of battle, his wild Irish “Whirroo!” appalling even his
frantic assailants, his desperate play with revolver, his final collapse
on the hot bosom of Mother Earth, his warm Irish blood reddening
the sands of the African desert.
John Augustus O’Shea, of the Standard, was another war-
correspondent who was very much given to practical joking, and
disguise generally played a prominent part in his plans. On one
occasion he was commissioned by his editor to describe a certain
Lord Mayor’s Show. Elephants were to play a part in this particular
pageant; and it occurred to the accomplished correspondent that
from the back of an elephant he might obtain an unrivalled view of

the rivals of the route. George Sanger was providing the elephants,
and O’Shea experienced no difficulty in obtaining permission to ride
in a howdah and illustrate the fidelity of Indian Princes to the
Empire. Sanger was also able to provide the Oriental costume
essential to the part, together with the stage diamonds without
which no self-respecting Prince ever goes out elephant-riding. His
face was made up to the proper tint; his turban was a triumph of
millinery; and as O’Shea passed through Fleet Street in the character
of an Eastern potentate, and in the train of a London Lord Mayor,
not a soul recognized him.
Indeed, the completeness of the disguise led to some
inconvenience. For when the show was at an end, and O’Shea went
on his elephant to Sanger’s stables in the Westminster Bridge Road,
he found himself pressed for time, and unable, therefore, to
abandon his disguise. He got into a hansom just as he was, and
drove off to Shoe Lane to write his descriptive article for the Evening
Standard. He was about to pass the commissionaire who stood
sentry at the office door. But that old soldier did not recognize a
member of the staff in the garb of a pious Hindu, and O’Shea,
unable to curb his love of practical joking, soundly rated the old
soldier in an improvised gibberish which the warrior, no doubt,
thought he recognized as something he had been acquainted with in
the East. O’Shea endeavoured to push past. The man “on the door”
barred his progress. The war of strange words between them grew
loud and furious. The commissionaire called to a member of the
crowd that was gathering round the door to go for the police, and
upstairs the sub-editor was anxiously waiting for O’Shea’s copy.
Before the police could arrive Gilbert Venables came on the scene,
recognized the correspondent under the disguise of the dusky
Indian, and explained matters to the faithful doorkeeper. The
anxiety of the sub-editor was soon appeased, and O’Shea sat down
to reel off a column of humorous descriptive copy such as he alone
on that staff could produce. “The Giniral”—as O’Shea was called in
Fleet Street—was one of those strange men who think that it is

Welcome to our website – the perfect destination for book lovers and
knowledge seekers. We believe that every book holds a new world,
offering opportunities for learning, discovery, and personal growth.
That’s why we are dedicated to bringing you a diverse collection of
books, ranging from classic literature and specialized publications to
self-development guides and children's books.
More than just a book-buying platform, we strive to be a bridge
connecting you with timeless cultural and intellectual values. With an
elegant, user-friendly interface and a smart search system, you can
quickly find the books that best suit your interests. Additionally,
our special promotions and home delivery services help you save time
and fully enjoy the joy of reading.
Join us on a journey of knowledge exploration, passion nurturing, and
personal growth every day!
ebookbell.com

Flood Risk Change A Complexity Perspective Andreas Paul Zischg

More Related Content

Similar to Flood Risk Change A Complexity Perspective Andreas Paul Zischg (20)

Recently uploaded (20)

Flood Risk Change A Complexity Perspective Andreas Paul Zischg