Water Front
Water Front
Water Front
and Response IV
WITeLibrary
Home of the Transactions of the Wessex Institute.
Papers presented at FRIAR 2014 are archived in the WIT eLibrary
in volume 184 of WIT Transactions on Ecology and the Environment (ISSN: 1743-3541).
The WIT eLibrary provides the international scientific community with immediate and
permanent access to individual papers presented at WIT conferences.
Visit the WIT eLibrary at www.witpress.com.
Fourth International Conference on
Flood Recovery, Innovation and Response
FRIAR 2014
Conference Chairmen
D. Proverbs
University of the West of England, UK
C.A. Brebbia
Wessex Institute of Technology, UK
C. Booth
D. De Wrachien
H. Hashimoto
M. Holicky
S. Mambretti
D. Mioc
M. Mohssen
D. Molinari
Organised by
Wessex Institute of Technology, UK
University of the West of England, UK
Sponsored by
WIT Transactions on Ecology and the Environment
International Journal of Safety and Security Engineering
WIT Transactions
Transactions Editor
Carlos Brebbia
Wessex Institute of Technology
Ashurst Lodge, Ashurst
Southampton SO40 7AA, UK
Editorial Board
Editors
D. Proverbs
University of the West of England, UK
C.A. Brebbia
Wessex Institute of Technology, UK
Editors:
D. Proverbs
University of the West of England, UK
C.A. Brebbia
Wessex Institute of Technology, UK
Published by
WIT Press
Ashurst Lodge, Ashurst, Southampton, SO40 7AA, UK
Tel: 44 (0) 238 029 3223; Fax: 44 (0) 238 029 2853
E-Mail: witpress@witpress.com
http://www.witpress.com
ISBN: 978-1-84564-784-1
eISBN: 978-1-84564-785-8
ISSN: 1746-448X (print)
ISSN: 1743-3541 (on-line)
The texts of the papers in this volume were set individually by the authors or under their
supervision. Only minor corrections to the text may have been carried out by the publisher.
No responsibility is assumed by the Publisher, the Editors and Authors 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. The Publisher does not necessarily endorse the ideas held, or views expressed
by the Editors or Authors of the material contained in its publications.
FRIAR 2014 is the fourth Conference of this successful series. The conference
started at the Institute of Civil Engineers in London 2008 and was reconvened at
the Lombardy Region Headquarters in Milano in 2010 and in Dubrovnik in 2012.
The damage caused by the flooding over the winter period is estimated to be
£1.1bn in the UK alone; but of course this does not reflect the longer term impacts
to lives and communities and businesses, who will be affected for many months
beyond the flooding itself. For some home owners and businesses, insurers will
assist in the recovery process by providing the necessary funding and services
to restore properties back to a habitable state. For others including those without
insurance, the recovery process will be very challenging indeed and it is likely that
many businesses will simply collapse as a consequence.
Research has shown that in the aftermath of the summer 2007 floods in the UK,
the vast majority of flood affected properties were reinstated to their previous
condition, leaving them equally vulnerable to future flood events. This goes against
the principles of climate change adaptation and represents a missed opportunity
to build back better and improve the resilience of homes and businesses that were
affected. Hopefully, the financial support now being made available to businesses
and homes in the UK will help to ensure resilient measures are installed during the
recovery process.
We know that it is impossible to entirely eliminate the risk from flooding and
that there is considerable uncertainty about future extreme weather patterns .
Clearly, further research is needed to improve our understanding of the challenges
associated with making our rural and urban environments and the communities
that exist within them, more resilient to the effects of flooding. This includes the
development of new innovative solutions as part of an integrated approach to flood
risk management at the community level. The complexity of these challenges
means that we need to work across disciplines and draw on a range of expertise,
recognising the use of both structural and non-structural measures towards arriving
at novel solutions to suit local circumstances.
WIT Press, the publishing arm of the Wessex Institute has produced this volume
which is distributed around the world by its own offices in Europe and the USA
and an extensive distribution network. The book is produced in hard copy and
digital format to reach as many colleagues as possible. Furthermore, all conference
papers have been archived online in the Institute eLibrary (http://library.witpress.
com) where they are immediately and permanently available to the international
community.
The Editors are grateful to the authors for the quality of the papers published in
this book and particularly indebted to the members of the International Scientific
Advisory Committee and other colleagues who helped to select them, in this
manner ensuring their names the quality of this volume.
The Editors
Poznan
2014
Contents
Risk management and emergency response for a 300 km2 sub-sea level area
with a million citizens against extreme storm surge and flood due to the
“Super Ise-Bay Typhoon”
T. Tsujimoto, M. Igarashi & K. Kobayashi...................................................... 251
Abstract
Flood warning mainly depends on reliable flood forecast models. Literature is
rich in flood modelling techniques, but failures of these models, especially on the
very short scale such as hourly flows, do often cause devastating impacts on
the communities affected by these floods, and on many occasions result in loss of
lives. This paper presents a new approach for flood forecasting of river flows
based on the projection theorem in Hilbert space.
The new modelling process obtains the projection of hourly flow rates on
hourly rainfalls over the catchment at previous hours to the projected flow rate. A
total of 25 flow events observed for the Leith River in Dunedin, New Zealand,
along with their corresponding observed rainfalls at two sites in the catchment
have been identified and applied to calibrate and validate the derived model. The
proposed modelling technique was capable of simulating the flow process for
the Leith River, and is a promising tool for flood forecast when other models
fail. The proposed model is easy to apply, doesn’t imply a lot of assumptions or
parameters, as other models usually require, and can be used for long term forecast
based on forecasted hourly rain one day or more before the event, or real time
forecast during the event itself based on rainfall which has been already gauged.
However, for real time (short term) forecast, the forecast time can be a few hours
based on the catchment area and its topography which can lead to a fast flow to
the outlet.
Keywords: flood forecast, flood modelling, rainfall-runoff, projection in Hilbert
Space.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140011
4 Flood Recovery, Innovation and Response IV
1 Introduction
Natural disasters cause devastating damages to all types of lives on earth, and their
negative impacts can last for long periods with a huge cost to mitigate. Floods are
the most common natural disasters, and unlike other forms of natural disasters
which usually occur in specific regions such as earthquakes, volcanoes,
hurricanes, or tornadoes, floods occur almost everywhere, and no community is
immune from their devastating damages. Flood warning can be quite effective in
mitigating the impacts of a coming event, simply by getting prepared. Even with
the existence of flood protection schemes, there is usually the potential for a bigger
flooding event than what the scheme was designed for. The Environment Agency
of UK and the strategic plan for the US National Weather Service indicated the
urgent need for major investment to develop new forecast models for flood
warning [1, 2].
Flood forecasting is the corner stone for an efficient flood warning system.
New technology and the use of satellite and radar data have significantly improved
our capability of forecasting rainfall, even on an hourly basis, for short term
periods such as the next few hours or longer forecast such as the next 24/48 hours.
However, due to the complexity of this natural event, and the high spatial and
temporal variability of rain, the main driving force for flooding, in addition to the
complex hydrological aspects and characteristics of the catchment area, it is
usually hard to accurately forecast the coming flood event [1]. Many of the
available forecast models in the literature, especially those based on watershed
modelling and hydraulic/hydrologic routing, require a lot of data and include a lot
of assumptions for solving the concerned equations, which adds to their
complexity and applicability. There are many reports in the media and anger in
the communities over failure of their governments/authorities to provide proper
flood warning [3–7].
Time series analysis and modelling, such as ARIMA models, have been applied
in the literature for simulating streamflows. However, these models work more
for longer time periods where stationarity conditions can be assumed, or achieved
by removing apparent cycles or trends [8]. For hourly flows during a significant
event, the series is quite non-stationary, and flow rates react directly and are highly
related to the rainfall intensity during the period preceding this flow. ANN has
been recently applied for flood forecasting, and several techniques have been
suggested for their applications to hourly time steps [9, 10].
In New Zealand, floods are the most costly natural disaster. About 935
devastating floods occurred during the period 1920 to 1983 in New Zealand [11].
Dunedin is the second largest city in the South Island of New Zealand, with a
population of about 120,000 (Statistics New Zealand, 2013). The Leith River,
which drains about 45 km2 of mainly hilly areas around Dunedin, goes through the
city and passes by the prestigious University of Otago. Most of the northern part
of Dunedin lies within the flood plain for the Leith River. A big flooding event
for the Leith River can cause significant damage and loss to Dunedin in particular,
and the whole Otago Region in general. The Leith River has history of flooding,
and Dunedin experienced extensive damage and inundation during the 1877, 1923
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 5
and 1929 flood events. The Otago Regional Council has recently conducted
studies for flood protection schemes for the Leith River, and has applied for
consents to carry out the needed work.
The catchment area of the Leith River is not big, and this usually results in a
significant component of the runoff contributing to its high flow hydrograph,
compared to the base flow component which is usually very small (as shown in
Fig. 2). Thus, the straight line approach for separating the base flow has been
applied to estimate the runoff hydrograph due to the rainfall event over the
catchment [12]. This approach should result in good estimates of the runoff
hydrograph, as any error in estimating this very small base flow will not have
effect on the much bigger runoff component. The runoff hydrograph is obtained
by simply subtracting the estimated base flow from the flow hydrograph.
The basic concept of this model is based on the projection in Hilbert Space [8]
of the hourly river flows on the span of hourly rainfall data preceding these flows.
This model represents an extension of the models developed by [13] and [14]
for the univariate and multivariate flood forecast of lake levels.
Thus, the flow rate at time t, Qt, is projected on the span of rainfalls at
antecedent times: Rt-j, j = L1 to L2, where L1 and L2 represent lag-1 and Lag-2
hours before time t. Thus:
∑ (1)
Thus,
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
6 Flood Recovery, Innovation and Response IV
Equation (3) produces a system of (L2 – L1 +1) linear equations, which can be
solved simultaneously to obtain the parameters , j = L1 to L2. The projection
theorem guarantees that the produced solution is the unique mapping of Qt onto R.
The projection theorem guarantees that the model provided by (3) will produce
coefficients of (Rt-j) for the best forecasts of Qt. It is assumed in this research that
the relationship between Qt and Rt-j is linear, which might not be the best choice.
However, based on the model application which is shown later, this proved to be
satisfactory. More research is recommended to consider alternative relationships.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 7
60
50
Flow (m3/s)
40
30 Base Flow
20
10
0
11/02/05
12/02/05
13/02/05
14/02/05
15/02/05
16/02/05
17/02/05
Date
Figure 2: A high flow event for the Leith River showing the base flow.
A total of 25 high flow events have been selected from the available record during
the period March 2000 until November 2013. Twenty three events have been used
for model calibration and two events have been utilised to test the validation of the
developed model. Rainfall over the whole catchment was estimated by applying
Thiessen polygon method to calculate the weight for each rainfall site, and in turn
obtain the average rainfall over the whole catchment area. Thus, one time series
of average hourly rainfalls has been estimated and used in this case study. For the
calibration process, hourly lagged rainfalls for all the events were joined together
in one input file to the model so that the estimated parameters , j = L1 to L2 are
based on all the 23 events, and not only on one event. Figure 3 presents lagged
correlations between runoff flow rates and observed rainfalls at lags 0 to 10 hours
prior to the flow rate. The figure indicates that lags 3 and 4 are the highest, and it
is a must to include these rainfalls for the flood forecast of the Leith River. If L1
equals 3, then this will produce a 3 hours warning before this flow rate for a real
time forecast during the rainfall event. However, if this forecast is based on
rainfalls during the next day, the warning time would be much longer.
Figure 4 shows the observed versus the “forecasted” flows for the combined 23
events which were used in the calibration process.
In general, the model simulated “satisfactorily” the underlying hourly runoff
process, but underestimated some of the significantly high events, and also
overestimated others. However, it has to be stated that it is usually very hard for
any model on an hourly basis to simulate accurately the underlying hydrologic
process. Add to this, that this newly developed model does not “explicitly”
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
8 Flood Recovery, Innovation and Response IV
0.9
0.8
Correlation Coefficient
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 2 4 6 8 10 12
Lag (hrs)
Figure 3: Lagged cross correlations between runoff flow rates and rainfalls.
100
80
Flow (m3/s)
60
40 Observed
20 Forecasted
0
117
175
233
291
349
407
465
523
581
639
697
755
1
59
Time (hrs)
3 Model testing
Validation of the fitted model was carried out by applying the model to rainfall
events which were not included in its calibration process. Thus, these estimated
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 9
parameters are not “biased” toward these events. These two high flow events
occurred during the periods 30 July to 1 August 2008 and 15 to 20 June 2013.
Table 1 shows the forecasted peak flows versus the observed ones for the two
events, while figures 5 and 6 show the simulation of the fitted model to forecast
hourly flows for the two events. The table shows that the forecasted peak flows
were within 12% to 23% of the observed peaks, with determination coefficients
(R2) and Filliben correlation coefficients (FC) higher than 90%. It should be noted
that each event has two peaks, and the model was capable of capturing this
behaviour for the second event, but was not able to “properly” simulate the second
30
25
20
Flow (m3/s)
15
Observed
Forecasted
10
0
1/08/08
1/08/08
30/07/08
31/07/08
31/07/08
31/07/08
Date
Figure 5: Observed vs. forecasted runoff hydrographs for the rainfall event
July 2008.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
10 Flood Recovery, Innovation and Response IV
60
50
40
Flow (m3/s)
30
Observed
Forecasted
20
10
0
15/06/13
16/06/13
17/06/13
18/06/13
19/06/13
20/06/13
21/06/13
Date
Figure 6: Observed vs. forecasted runoff hydrograph for the rainfall event
June 2013.
peak of the first event. There is only one determination coefficient and one
Filliben correlation coefficient for each event, as shown in the table. The figures
confirm the conclusion that the model is capable of forecasting the Leith River
high flows, and responded well to the rising limb and the recession of the two
events.
4 Conclusions
A newly derived approach to forecast river flows based on the projection theorem
in Hilbert space has been presented and applied for the Leith River in Dunedin,
New Zealand. The model, once derived and calibrated, is easy to apply and can
be used for forecasting during a rainfall event with a lead time of 3 hours, or can
be used for a much longer time if forecasted rainfall is used. The model required
only hourly rainfall and flow data for its calibration, and only hourly rainfall data
for its application for flood forecast. Despite the fact that the model, in its current
form, does not “explicitly” has a function to account for hydrologic abstractions
from rainfall, still it produced satisfactorily results with its implicit inclusion of
rainfall losses during the projection process.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 11
References
[1] Bye, P. & M. Horner, Easter 1998 Floods Report by the Independent
Review Team to the Board of the Environmental Agency, vol 1,
Environmental Agency, Bristol, 1998.
[2] Demeritt D., H. Cloke, F. Pappenberger, J. Thielen, J. Bartholmes & Maria-
Helena Ramoset, Ensemble predictions and perceptions of risk, uncertainty,
and error in flood forecasting, Environmental Hazards, vol 7, pp. 115-127,
2007.
[3] McClure, M & T. Howell, Forecast failure: how flood warning came too
late for southern Albertans, Calgary Herald, December 31, 1991.
[4] Datta, S., CWC failed to forecast, alert about floods, DNA, New Delhi, June
26, 2013.
[5] Socialist Equality Party (Australia), Australia’s floods: a failure of
government and the profit system, WSWB World socialist web site, January
29, 2011.
[6] Daily Express newspaper, Anger over flood warning failure, Daily Express
paper, London, July 9, 2012.
[7] Handmer, J., Are Flood Warnings Futile? Risk communication in
emergencies, the Australian Journal of Disaster and Trauma Studies,
2000–2.
[8] Brockwell, P. J. & R. A. Davis., Time Series: Theory and Methods,
Springer-Verlag New York Inc., pp. 46-51, 1991.
[9] Tiwari, M. K., Chatterjee, C., Development of an accurate and reliable
hourly flood forecasting model using wavelet–bootstrap–ANN (WBANN)
hybrid approach, J. of Hydrology 394, pp. 458-470, 2010.
[10] Chen-ShenHsien; Lin-YongHuang; Chang-LiChiu; Chang-FiJohn, The
strategy of building a flood forecast model by neuro-fuzzy network. Journal
of Hydrological Processes, 20(7), pp. 1525-1540, 2006.
[11] McSaveney, E., Floods – New Zealand’s number one hazard, Te Ara – the
Encyclopedia of New Zealand, updated 2-Mar-09.
[12] Chow, V.T., D. R. Maidment & L. W. Mays, Applied Hydrology McGraw-
Hill, 1988.
[13] Mohssen, M. and Goldsmith, M., Flood Forecasting of Lake Levels: A New
Concept. Int. J. of Safety and Security Eng., 1(4), pp. 363-375, 2011.
[14] Mohssen, M., A Multivariate Model for Flood Forecasting of Lake Levels.
Int. J. of Safety and Security Eng., 3(2), pp. 141-152, 2013.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
This page intentionally left blank
Flood Recovery, Innovation and Reponse IV 13
Abstract
Flooding continues to cause significant disruption to individuals, organisations
and communities in many parts of the world. In terms of the impact on
businesses in the United Kingdom (UK), flooding is responsible for the loss of
millions of pounds to the economy. As part of a UK Engineering and Physical
Sciences Research Council funded project on flood risk management, SESAME,
research is being carried out with the aim of improving business response to and
preparedness for flood events. To achieve this aim, one strand of the research is
focused on establishing how agent-based modelling and simulation can be used
to evaluate and improve business continuity. This paper reports on the
development of the virtual geographic environment (VGE) component of an
agent-based model and how this has been combined with inundation prediction
to enable the identification of businesses affected by flooding in any urban area
of the UK. The VGE has been developed to use layers from Ordnance Survey’s
MasterMap, namely the Topography Layer, Integrated Transport Network
Layer and Address Layer 2. Coupling the VGE with inundation prediction
provides credibility in modelling flood events in any area of the UK. An initial
case study is presented focusing on the Lower Don Valley region of Sheffield
leading to the identification of businesses impacted by flooding based on a
predicted inundation. Further work will focus on the development of agents to
model and simulate businesses during and in the aftermath of flood events such
that changes in their behaviours can be investigated leading to improved
operational response and business continuity.
Keywords: floods, businesses, agent-based modelling and inundation prediction.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140021
14 Flood Recovery, Innovation and Response IV
1 Introduction
In recent years, many parts of the world have experienced and suffered from
severe flooding which continues to cause significant disruption to individuals,
organisations and communities. In terms of the impact on UK businesses,
research conducted by the Environment Agency (EA) has estimated the financial
cost of floods in 2012 as being nearly £600 million [1]. Further, the EA has
indicated flooding cost an average of £60,000 for every business affected. Such
significant economic loss has led to flood risk management becoming high on
the political agenda. Consequently, means of reducing the economic impact of
interruptions attributable to flooding at the business level, and thus more widely,
are receiving growing attention. Effective business continuity management
(BCM) is recognised as one means of reducing the effect of flooding on business
operations and enabling a more rapid return to normality. Indeed, BCM is
viewed as an important tool for business survival in the face of a range of
disruptive events [2, 3], including flooding, and a key part of any successful
flood response [4]. The International Organization for Standardization’s ISO
22301, which is related to BCM, is described as the requirements which will help
organisations to be better prepared and handle disruptions of any type [5].
Despite the existence of such standards, in the UK, organisational engagement
with business continuity remains low with less take-up by Small and Medium
Enterprises (SMEs) relative to larger businesses and public sector bodies [6, 7].
The Engineering and Physical Sciences Research Council funded SESAME
project is related to organisational operational response and strategic decision
making for long term flood preparedness in urban areas [8]. The project aims to
create a unified framework of academic knowledge that can be used to influence
the behaviours of businesses, particularly SMEs, faced with flooding and flood
risk. This framework will assist businesses in understanding how they might
reduce the disruption and economic loss associated with flood events thus
strengthening their resilience to flooding and that of the wider economy. To
realise this aim, four interdisciplinary research objectives are being pursued: (i)
achieve a better understanding of how businesses behave in the immediate and
longer term aftermath of flood events; (ii) establish how agent-based modelling
and simulation can be used to assess the behaviours of different types of
businesses at risk of flooding; (iii) assess the impacts of flooding on economic
systems both within and beyond the immediately affected urban area and explore
how changes in businesses’ behaviour could influence these impacts; (iv)
develop and evaluate approaches promoting organisational behaviour change and
adaptive learning throughout the flood cycle. In order to achieve these research
objectives, the SESAME project brings together the academic fields of business
continuity management, agent-based modelling and simulation, flood modelling,
economic modelling and the social/behavioural sciences. This paper focuses on
the development of the virtual geographic environment (VGE) component of an
agent-based model along with how this has been brought together with
inundation prediction, via flood modelling, to identify businesses affected by
flooding in any urban area of the UK.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 15
2 Related work
In the disaster management domain, research in the area of agent-based
modelling and simulation has focused on emergency response to major natural
and manmade events with agents representing emergency responders and/or
members of the public [9 –12]. In relation to flood risk management, agent-based
models have usefully been employed to model evacuation strategies involving
agents representing members of the public [13, 14]. However, despite the
concept of using agent-based modelling in business and organisation problems
[15–22], there is lack of research in the context of modelling businesses faced
with the challenges of ensuring business continuity when subjected to flooding.
This current dearth of research in agent-based modelling and simulation in the
context of business response to flooding offers scope for significant
contributions to knowledge to be made in relation to (i) identifying the specific
businesses affected by a flood event in a particular geographical area which can
then be modelled as agents, (ii) modelling these business agents’ actions and
interactions when responding to flood events based on field data gathered
through interviews with businesses at risk of flooding and/or which have
experience of flooding, and (iii) performing what-if analysis via agent-based
simulations of businesses’ responses to flood events in order to establish the
effect of changes in their behaviour and different approaches taken such as
adhering to flood plans.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
16 Flood Recovery, Innovation and Response IV
Agent-based modelling
1 3 4
identifies implemented in
Static
2
Dynamic 5
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 17
Figure 3: VGE for the Lower Don Valley region of Sheffield with flood extent.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
18 Flood Recovery, Innovation and Response IV
Using the database created for the Lower Don Valley region of Sheffield,
4037 businesses were identified as being within the bounded area considered
with 531 of those businesses affected directly by flooding based on the
inundation prediction. In this context, “affected directly by flooding” signifies
that both the building, and the road link associated with the building, of a
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 19
particular business were under a depth of water greater than or equal to 1 mm.
Based on the inundation prediction, the depth of flood water can be established
for every business. For the Lower Don Valley region of Sheffield, Table 1
presents a profile of the number of businesses’ buildings and road links affected
by flooding in relation to depth of flood water.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
20 Flood Recovery, Innovation and Response IV
(ISIC) of All Economic Activities, Rev.4, (21 high level activities and 99 sub-
activities) and the Cambridge Econometrics’ Multisectoral Dynamic Model
(MDM-E3) of the UK economy (46 industry types) respectively. Mapping
between these different classifications will be required once agent models are
developed in preparation for agent-based simulations to be performed.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 21
coastal), thus informing the attributes and behaviour of business agents. Also,
these interviews will be used to elicit types of potential behaviour changes a
business could make in terms of how it prepares for and/or responds to flood
events, which could subsequently be investigated using agent-based simulation.
Depending on the type of business being modelled, examples of such changes
could relate to: flood insurance; registering to receive EA flood warnings;
relocating key operations; training staff; installing flood barriers; making
premises more resilient to flooding using water resistant materials; developing a
flood plan; backing-up customer databases and electronic files; moving storage
areas out of reach of flood waters; being able to quickly move equipment,
computers, furniture, paper files, electrical items above ground level or to an
upper level of the building if possible; identifying alternative supply and
distribution routes; developing relationships with service and supply companies
in advance of flooding such that essential work can be undertaken rapidly to
quicken recovery and reduce business interruption.
Acknowledgement
The authors gratefully acknowledge the funding provided by the UK’s EPSRC
under grant EP/K012770/1.
References
[1] http://www.environment-agency.gov.uk
[2] Elliott, D., Herbane, B. & Swartz, E., Business Continuity Management,
Routledge: London, 2001.
[3] Herbane, B., The evolution of business continuity management: A
historical review of practices and drivers, Business History, 52(6), pp.
978–1002, 2010.
[4] Pitt, M., The Pitt Review: Lessons learned from the 2007 floods, Cabinet
Office, 2008.
[5] http://www.iso.org
[6] Musgrave, B. & Woodman, P., Weathering the Storm: The 2013 Business
Continuity Management Survey, Chartered Management Institute: London,
2013.
[7] Herbane, B., Small business research: Time for a crisis-based view,
International Small Business Journal, 28(1), pp. 43–64, 2010.
[8] Coates, G., Hawe, G.I., McGuinness, M., Wright, N.G., Guan, D., Harries,
T. & McEwen, L., A framework for organisational operational response
and strategic decision making for long term flood preparedness in urban
areas, Proceedings of the 3rd International Conference on Disaster
Management, 2013.
[9] Kitano, H. & Tadokoro, S., RoboCup Rescue: A Grand Challenge for
Multiagent and Intelligent Systems, Artificial Intelligence Magazine,
22(1), pp. 39–52, 2001.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
22 Flood Recovery, Innovation and Response IV
[10] Mysore, V., Narzisi, G. & Mishra, B., Agent Modeling of a Sarin Attack
in Manhattan, Proceedings of the 1st International Workshop on Agent
Technology for Disaster Management in the 5th International Conference
on Autonomous Agents and Multi-Agent Systems, pp. 108–115, 2006.
[11] Bellamine-Ben Saoud, N., Ben Mena, T., Dugdale, J., Pavard, B. & Ben
Ahmed, M., Assessing large scale emergency rescue plans: an agent based
approach, International Journal of Intelligent Control and Systems:
Special Issue on Emergency Management Systems, 11(4), pp. 260–271,
2006.
[12] Hawe, G.I., Wilson, D.T., Coates, G. & Crouch, R.S., Investigating the
Effect of Overtriage on Hospital Arrival Times of Critically Injured
Casualties during a Major Incident using Agent-Based Simulation,
Proceedings of the 6th International Conference on Soft Computing and
Intelligent Systems and the 13th International Symposium on Advanced
Intelligent Systems, 2012.
[13] Liu, Y., Okada, N., Shen, D. & Li, S., Agent based flood evacuation
simulation of life-threatening conditions using Vitae system model,
Journal of Natural Disaster Science, 31(2), pp. 33–41, 2009.
[14] Dawson, R., Peppe, R. & Wang, M., An agent based model for risk-based
flood incident management, Natural Hazards, 59(1), pp. 167–189, 2011.
[15] Nagendra Prasad, M.V. & Chartier, D.A., Modeling Organizations using
Agent-based Simulations, Proceedings of the Workshop on Agent
simulation: Application, Models & Tools, pp. 54–66, 1999.
[16] Gilbert, N. & Terna, P., How to Build and Use Agent-based Models in
Social Sciences, Mind & Society, 1, pp. 57–72, 2000.
[17] Bonabeau, E., Agent-based modeling: Methods and techniques for
simulating human systems, Proceedings of the National Academy of
Sciences of the United States of America, 99(3), pp. 7280–7287, 2002.
[18] North, M. J. & Macal, C.M., Managing business complexity: discovering
strategic solutions with agent-based modeling and simulation, Oxford
University Press: Oxford, 2007.
[19] Gilbert, N., Agent-based models, In: UNSPECIFIED Quantitative
Applications in the Social Sciences, Sage Publications Inc., 2007.
[20] Dignum, V. & Tick, C., Agent-based Analysis of Organizations:
Formalization and Simulation, Proceedings of the IEEE/WIC/ACM
International Conference on Intelligent Agent Technology, pp. 244–247,
2007.
[21] Wilkinson, I.F., Marks, R.E. & Young, L., Toward Agent-based Models of
the Development and Evolution of Business Relations and Networks, In
Unifying Themes in Complex Systems - Volume VI: Proceedings of the 6th
International Conference on Complex Systems, ed. Ali A. Minai, Dan
Braha & Yaneer Bar-Yam, Springer, New York, pp. 414–421, 2010.
[22] Fioretti, G., Agent-Based Simulation Models in Organization Science,
Organizational Research Methods, 16(2), pp. 227–242, 2013.
[23] http://www.ordnancesurvey.co.uk
[24] Environment Agency, Review of 2007 summer floods, December 2007.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 23
Abstract
The study of dam-break waves (DBW) is extremely important in providing the
information needed for risk assessment and management of coastal and riverine
areas. Adequate and acceptable preparedness for such an event to allow
mitigation of adverse impacts requires modelling of the flood as well as accurate
estimation of potential flood depths, flow velocities, and timing of the flood
arrival. This study investigated the effect of floodwater waves on various wall
surfaces and wall slopes in a 4.7m long wave tank by modelling a dam-break
phenomenon. The paper reports the novel simple methods (the Imaging System
(IS) and the sensor Signal Capture (SSC) technique) used for the estimation of
wave front propagation velocity which are the adaptations of the commonly used
Particle Image Velocimetry (PIV). The two techniques demonstrated good
agreement with the dam break wave theory as well as agreement between each
other. However, the SSC method with wave probes at a shorter separation
distance (0.41m apart) appears better and more in line with the results obtained
by previous investigators. The development represents a useful laboratory
scheme that is well suited for educational and initial research studies.
Keywords: dam-break, flow velocity, particle image velocimetry, flood waves.
1 Introduction
The concept of traditional flood protection is increasingly being replaced by
comprehensive risk management, which includes structural and non-structural
measures [1]. Hazard and risk maps are of particular importance for planning
purposes, risk awareness campaigns and the encouragement of private preventive
measures. Flood hazard risks are characterised by flood impact parameters such
as water depth and flow velocity. However, there has been a strong focus on
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140031
24 Flood Recovery, Innovation and Response IV
inundation depth as the main determinant for flood damage probably due to
limited information about other parameters characterising the flood, e.g. flow
velocity.
A systematic review of flood impacts on buildings and structures by Kelman
and Spence [2] revealed various damage mechanisms including hydrodynamic
actions related to waves and velocity as a result of turbulence. Dam-breaks have
been known for destroying buildings and infrastructures and also being
responsible for numerous losses of life in coastal and riverine areas. They
generally result in flash flood runoff in rivers and streams, debris flow surges
and tsunami run-up on dry coastal plains. In all these cases, the surge front is a
sudden discontinuity characterized by extremely rapid variations of flow depth
and velocity.
Flow velocity is generally presumed to influence flood damage. According to
Kreibich et al. [3] a significant influence of flow velocity on structural damage
could be shown in contrast to a minor influence on monetary losses and business
interruption. Forecasts of structural damage to road infrastructure is determined
to be based on flow velocity alone while the energy head is suggested as a
suitable flood impact parameter for reliable forecasting of structural damage to
residential buildings [3]. However, it is generally accepted that the higher the
flow velocity of the floodwater, the greater the probability (and extent) of
structural damage [4].
USACE [5] states that velocity is a major factor that could aggravate
structural and content damage during flooding events. High velocities limit the
time available for emergency measures and evacuation. The additional force of
high velocities creates greater danger of foundation collapse and forceful
destruction of contents [5]. For instance, Smith [6] states that a velocity of 3m/s
acting over a 1m depth will produce a force sufficient to exceed the design
capacity of a typical residential wall. The study shows further critical
combinations of water depth and flow velocities for building failure for three
different residential building types. These range from above 0.5m water depth
and 4m/s flow velocities to above 3m water depth with no flow velocity for
single storey weatherboard buildings [6].
The study of dam-break flow is important in providing vital information
needed for risk assessment and management of river valleys and coastal plains.
Such information may include useful data on dam-break flow variables such as
initial dam conditions, water depth downstream, flow velocity etc.
Moreover, physical modelling of dam-break waves is relatively limited. Most
predictions on dam-break waves are often based upon numerical predictions,
validated by limited data sets. According to Chanson [7] current knowledge of
dam-break waves in dry channels remains rudimentary despite a few available
studies.
In this paper, an experimental study of a dam-break flow is presented. Most
existing studies about dam-break flows are focused on variables such as
measurements of velocity profile and the water level using a Particle Tracking
Velocimetry (PTV) algorithm and/or Particle Image Velocimetry (PIV)
algorithm. However, in this study, an indirect way of measuring the flow
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 25
velocity was applied. The present study developed simple methods for estimating
instantaneous dam-break floodwater front velocity over the whole flow depth in
a dry channel using image acquisition techniques. The main feature of this
development is its simplicity that is well-suited to initial investigations.
2 2 (1)
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
26 Flood Recovery, Innovation and Response IV
3 Experimental work
This work was carried out in the Materials and Hydraulic Laboratory of the
School of the Built Environment, Liverpool John Moores University. A Low
Cost Wave Tank (LCWT) was primarily designed and constructed to simulate
dam failure in order to generate floodwater waves. The main aim was to
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 27
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
28 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 29
The results of the front velocity with dry-bed and wet-bed downstream
conditions are shown in Table 1. Initial downstream water depths of 0.05m,
0.10m and 0.15m were investigated with varying initial reservoir depths and
compared (Table 1). Figure 4 emphasizes that the velocity decreases as the
downstream initial water depth increases. A dry-bed downstream gave some
unexpected results in this case. Figure 4 as well as visual and video analysis also
indicated that higher values and complexity of flow characteristics were obtained
for the lower downstream water depth case than for the case with higher
downstream water depths. Figure 4 also shows that the initial slope of the
velocity variation decreases as the downstream initial water depth increases. For
all depth ratios, the velocity profiles eventually became quite stable after the bore
developed downstream which is considered to be satisfactory for the downstream
subcritical flow region.
Table 1: Computed front water velocity for dry-bed and wet-bed downstream
at varying reservoir depths.
Depth of
water in the Velocity, v Velocity, v Velocity, v Velocity, v
reservoir, do (dry-bed) (Hds=0.05m) (Hds=0.1m) (Hds=0.15m)
(m) (m/s) (m/s) (m/s) (m/s)
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
30 Flood Recovery, Innovation and Response IV
Figure 4: Comparison of the front water velocities for dry- bed and wet-bed
at various initial water depths downstream.
The accuracy of the flow velocity relies on several factors. In the present
experiment the flow velocity is mainly associated with the precision of the time
interval between image pairs and the exactness of the displacement
measurement. Thus, floodwater front velocity was again computed using the
SSC technique described earlier to validate the reliability and accuracy of the IS.
Figure 5 compares the results of the two methods. The results obtained using the
two techniques indicated a good agreement with the dam-break wave theory
however; it was observed that the velocities obtained using the SSC method with
wave probes at a shorter distance away from each other (0.41m apart) appeared
to be closer in agreement to that obtained using IS (Figure 5). This result follows
Chegini [17] concept that this distance needs to be small enough to maintain a
degree of correlation and accuracy in the measurements of floodwater
propagation velocity.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 31
Some previous investigators interchanged wave celerity with front water flow
velocity. This concept is verified in the present study. An approximation of wave
celerity was obtained from the shallow water relationship taken as:
(2)
The solitary wave theory gives celerity for the steep waves as:
1 (3)
where H = water depth in the channel and d = water surface elevation from Still
Water Level (SWL).
However, considering the dry-bed downstream condition, eqn (3) is
simplified to the form of shallow water relationship equivalent to eqn (2). This
allows the use of eqn (3) to compute floodwater wave celerity for the present
experiments. Also, using the empirical and analytical equations for the flow
velocities proposed by various previous investigators, the velocities obtained
from their models are compared with the celerity of the present study. It can be
seen from Figure 6 that the flow celerity of the present study is in close
agreement with the front water velocity of Lauber and Hager [8] while other
investigators appeared to overestimate the front water velocity in relation to
celerity of the flow.
Similarly, Figure 7 compares the front water flow velocity of previous
investigators with the floodwater front velocity obtained in the present study. It
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
32 Flood Recovery, Innovation and Response IV
Figure 7: Comparison of the computed front water velocities for this study
with various existing theories.
can again be seen from the figure that the front water velocity computed using
SSC method with wave probes 0.41m apart and that of IS are in close agreement
with Liem and Kongeter’s theory [13] as well as with Ritter’s predictions [7].
Hunt’s theory has fair agreement with front water velocity computed using SSC
method with wave probes of 2.2m apart. It should be noted that a comparison
with Hunt’s theory may be incorrect at the upstream end of the channel since
Hunt’s equation is said to be valid only once the wave front has travelled a
distance of more than 4 times the reservoir length [18].
5 Conclusions
The use of digital imaging for qualitative and quantitative characterisation of
fluid flows is not new. In recent years however, with the rapid development of
powerful digital cameras at affordable prices and the advances in robust and fast
image processing techniques, this tool has become very popular.
In the present study, propagation velocities of floodwater flow have been
computed in an idealized dam-break problem using various adaptations of the
commonly used PIV method. The IS and the SSC methods described in section 3
have been adopted. The results obtained using the two techniques demonstrated
good agreement with the dam-break wave theory. However, it was observed that
the velocities obtained using the SSC method with wave probes at shorter
distance away from each other (0.41m apart) appeared closer in agreement to the
IS than that of the SSC with 2.2m separation. In addition, it is also indicated that
the propagation velocity obtained using the SSC method with wave probes
0.41m apart and that of the IS appeared in close agreement with some previous
researchers, particularly Chegini’s concept [17].
It was also revealed that most previous investigators over-estimated front
water velocity by interchanging it for the wave celerity which implies that
caution should be taken when doing this as it is only applicable in certain
circumstances. Furthermore, the results of comparison of front velocity with
various downstream water levels (DSWL) revealed that higher DSWL reduces
the speed of the bore, which indicates that the water in front of the travelling
bore reduces the speed of the flood wave. More importantly, this development
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 33
References
[1] Sayers, P., Hall, J., Dawson, R., Rosu, C., Chatterton, J. and Deakin, R.,
Risk Assessment of Flood and coastal Defences for strategic Planning
(RASP) – A high level Methodology. DEFRA Conference of Coastal and
River Engineers, Keele University, HR Wallingford, 2002.
[2] Kelman, I. and Spence, R., An overview of flood actions on buildings.
Eng. Geol., 73, pp. 297–309, 2004.
[3] Kreibich, H., Piroth, K., Seifert, H., Maiwald, H., Kunert, U., Schwartz, J.,
Merz, B. and Thieken, A. H., Is flow velocity a significant parameter in
flood damage modelling? Natural Hazards Earth System Science, 9, pp.
1679 –1692, 2009.
[4] Soetanto, R. and Proverbs, D. G., Impact of flood characteristics on
damage caused to UK domestic properties: the perceptions of building
surveyors. Structural Survey, 22 (2), pp. 95–104, 2004.
[5] USACE, Design of Revetments, Seawalls and Bulkheads. EM 1110 – 2 -
1614, 1996.
[6] Smith, D. I., Flood damage estimation – A review of urban stage damage
curves and loss functions. Water SA, 20 (3), pp. 231–238, 1994.
[7] Chanson, H., Applications of the Saint-Venant Equations and method of
Characteristics to the Dam Break Wave Problem. Hydraulic Model
Reports of Department of Civil Engineering, University of Queensland,
Report No. CH55/05, ISBN 1864997966, 2005.
[8] Lauber, G. and Hager, W. H., Experiments to dam-break waves:
Horizontal channel. Journal of Hydraulic Research, 36 (3), pp. 291–307,
1998.
[9] Stansby, P. K., Chegini, A. H. N. and Barnes, T. C. D., The initial stages
of dam-break flow. Journal of Fluid Mechanics, 374, pp. 407–424, 1998.
[10] Adrian, R., Engineering Application of Particle Image Velocimeters. Proc.
of ICALOE, Laser Institute of America, pp. 56–71, 1989.
[11] Douxchamps, D., Spinewine, B., Capart, H., Zech, Y. and Macq, B.,
Particle-Based Imaging Methods for the Characterisation of Complex
Fluid Flows. Proc. of the IEEE Oceans, pp. 20–25, 2004.
[12] Chegini, A. H. N., Pender, G., Slaouti, A. and Tait, S. J., Velocity
measurement in dam-break flow using imaging system. Proc. Of the 2nd
Int. Conf. On Fluvial Hydraulics, IAHR/AIRH, 2 (June), pp. 858–867,
2004.
[13] Liem, R. and Kongeter, J., Application of High-Speed Digital Image
Processing to Experiments on Dam Break Waves. Proc. of Concerted
Action on Dam-Break Modelling (CADAM) Conference, European
Community Workgroup, pp. 399–411, 1999.
[14] Van Dyke, M., An Album of Fluid Motion, Publication of Parabolic Press,
1982.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
34 Flood Recovery, Innovation and Response IV
[15] Capart, H., Young, D. L. and Zech, Y., Voronoi imaging methods for the
measurements of regular flows. Experimental Fluids, 32, pp. 121–135,
2002.
[16] Raffel, M., Willert, C. E. and Kompenhaus, J., Particle Image Velocimetry
– A Practical Guide. Springer-Verlag: Berlin and New York, 1998.
[17] Chegini, A., Fundamental Investigations of Dam-break Flows. Ph.D.
Thesis, Department of Civil Engineering, University of Manchester,
Manchester, UK, 1997.
[18] Hunt, B., Dam-break solution. Journal of Hydraulic Engineering, ASCE,
110 (6), pp. 675– 686, 1984.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 35
Abstract
Typhoon Morakot struck central and southern Taiwan on August 8, 2009, and
the high rainfall intensity and accumulated rainfall-induced several floods,
landslides, and debris flows. In this study, the destructive debris flow caused by
Typhoon Morakot in the Sha-Xinkai gully of the Liouguei District in southern
Taiwan was selected as a case study for analysis. A two-dimensional model
(FLO-2D software) was used to simulate debris flow. First, hydrological and
geomorphological data were collected on the debris flow event and the
rheological properties of slurry collected from the field were analyzed. Next,
the relationship between debris flow discharge and water flow discharge was
obtained. The simulation results for the deposited area and depth were then
compared to aerial photos taken during a field investigation. Finally, the bulked
coefficient of discharge and the resistant parameters used in the model were
presented. The results showed that the maximum deposited depth in the debris
flow inundated area was over 6 m; the maximum velocity, 6.6 m/s; and the
deposited volume, almost 1,000,000 m3. The simulated deposition depth and
inundation area matched the results from the field investigation reasonably well.
In this study, the parameters and processes needed for the simulation of
landslide-induced debris flows were proposed to provide a reference for hazard
zone mapping and debris flow hazard mitigation.
Keywords: Typhoon Morakot, FLO-2D, rheological property, bulked coefficient.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140041
36 Flood Recovery, Innovation and Response IV
1 Introduction
Typhoon Morakot struck central and southern Taiwan on August 8, 2009. The
extreme rainfall (maximum hourly rainfall of 123 mm and 48-h rainfall of
2,361 mm measured at the Alishan rainfall station) associated with the typhoon-
induced several landslides, debris flows, and floods (Chen et al. [1]; Wang et al.
[2]). In the basin of the Raolung River in southern Taiwan, many landslide-
induced debris flow hazards originate from a gully that has a small watershed
area ( A ) (e.g., A smaller than 40 ha) and a high landslide ratio ( RL ) (e.g., RL >
30% where RL is the ratio of landslide area AL to watershed area A or RL =
AL / A ). Gullies with small watershed areas are generally unknown or overlooked
by people, and they are often the cause of serious disasters during extreme
rainfall events. The number of extreme rainfall events in Taiwan has increasing
trend in recent years, which has resulted in a greater number and magnitude of
debris flows during the last decade (Chen et al. [3]). Hence, the development of
techniques that can identify and possibly prevent debris flows in gullies is a very
important research topic for hazard mitigation efforts.
The FLO-2D [4] routing model is software designed for two-dimensional
mathematical modeling of water movement and fast flowing slope processes
including debris flows. The FLO-2D model has been used successfully for debris
flow simulations by many researchers in a variety of countries (Lin et al. [5];
Tecca et al. [6]; Sosio et al. [7]; Stolz et al. [8]; Jakob and Weatherly [9]; Hsu et
al. [10]; Sodnik and Mikos [11]). Data required for model simulations include a
digital terrain model, an inflow hydrograph, rheological properties of the
sediment water mixture, and the Manning roughness coefficient. The results
from debris flow simulations are especially sensitive to the inflow hydrograph
and rheological parameters associated with volumetric sediment concentrations.
The inflow hydrograph may be underestimated for small watershed areas with
high landside ratios. However, previous research has generally focused on debris
flows from large watershed areas. Furthermore, the rheological parameters used
in previous studies are usually determined by the back analysis method or by
comparisons between model simulations and field observations. In contrast, the
rheological parameters used in this study were determined via laboratory
experiments. Two volumetric concentrations were used to simulate landslide-
induced debris flow, and the empirical coefficient (i.e., the discharge bulked
coefficient) that described the relationship between debris flow discharge and
water flow discharge was determined in this study. Results of this study can
provide a basic framework Results of this study can provide a basic framework
for determining debris flow discharges and select rheological parameters in
simulations of landslide-induced debris flows, which is important for hazard
zone mapping and debris flow hazard mitigation.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 37
2 Study area
The Sha-Xinkai gully study area is located in the Shinfa Village of the Liouguei
District, Kaoshing city, in southern Taiwan (Figure 1). It has a catchment area of
29.7 ha, a main stream length of 542 m, and an average stream bed slope
of 22.5°.
Liouguei
Kaohsiung Sha-Xinkai gully
Watershed area
N Deposition area on land
Sha-Xinkai watershed
Kaohsiung city
Shinfa rain station
Taiwan
2.1.2 Rainfall
The hourly and cumulative rainfall data collected from the Shinfa rain gauge
station, which is located approximately 2 km away from the Sha-Xinkai gully,
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
38 Flood Recovery, Innovation and Response IV
1600
120 1400
1200
Disaster caused by
80 large debris flow 1000
(20:30-21:00)
800
600
40
400
200
0 0
15 18 21 0 3 6 9 12 15 18 21 0 3 6 9 12 15 18 21 0 3 6 9 12 15 18 21 0 3 6 9 12 15 18 21 0 (hr)
Aug. 7 Aug. 8 Aug. 9 Aug. 10 (Month day)
Date
Figure 2: Rainfall data collected from August 7, 2009 to August 10, 2009 at
the Shinfa rain gauge station and the time that a debris flow was
triggered.
Rheological properties are very important when modeling debris flows. In the
FLO-2D model, the rheological parameters, including the mixture yield stress
( y ) and the mixture viscosity ( ), are used to describe the rheological
characteristics of debris flows. The rheological parameters are dependent on the
volumetric concentration ( cV ), and they have a significant effect on debris flow
processes and the final deposition morphology (FLO-2D [4]). To determine the
rheological parameters of debris flow, soil samples with a particle diameter of
less than 1 mm were collected from the flow area of the Hong-Shui-Xian gully,
which is located next to the Sha-Xinkai debris flow. The soil samples were
analyzed in a laboratory experiment using a Brookfield viscometer (type DV-III).
The relationship between the shear stress and shear strain for the soil sample at
various cV values was analyzed. The results showed that the rheological
properties of the debris flow slurries could be described by the Bingham model.
The Bingham model contains two rheological parameters: yield stress ( y ) and
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 39
viscosity ( ). The y (in dynes/cm2 units) and (in poise units) both
exponentially increased with an increase in volumetric concentration ( C ) V
0.0485 e14.94 c V
(2)
The results from Eqs. (1) and (2) were consistent with the bounds reported in
previous studies (FLO-2D [4]; Dai et al. [13]; Fei [14]). Because the lithological
characteristics and grain sizes of deposits in the Sha-Xinkai gully were almost
identical to the Hong-Shui-Xian gully, the rheological relationships from
Eqs. (1) and (2) were used to determine the rheological parameters for debris
flow simulations in this study.
3 Methods
3.1 FLO-2D model
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
40 Flood Recovery, Innovation and Response IV
2. Rainfall data: Rainfall data were collected from the Shinfa rain gauge
station. The maximum hourly rainfall data from this station were used to
determine peak water flow discharges in the Sha-Xinkai gully during
Typhoon Morakot.
3. Parameters for simulation: The relationships for rheological parameters
(Eqs. (1) and (2)) were used to simulate debris flow. In addition to the
rheological parameters, other important parameters included the Manning
roughness coefficient ( n ) and the resistance parameter for laminar flow ( k ).
The n value depends on the land surface, and it can be determined by
referencing the FLO-2D user’s manual [4]. In the Sha-Xinkai gully, the n
value ranged from 0.10 to 0.20. Hence, a n = 0.15 was adopted for use in
this study. The k value can range from 24 to 50,000. For modeling debris
flow, a calibrated k value of 2285 (FLO-2D [4]) was used to simulate the
Sha-Xinkai debris flow event.
where cb is the discharge bulked coefficient. The value for Qwp is generally
determined from the rational formula, Qwp C I A / 360 , where C is the runoff
coefficient, I is the maximum hourly rainfall intensity (mm/h), and A is the
watershed area. In the Sha-Xinkai gully study area, C = 0.8 (SWCB [16]), I =
103 mm/h (i.e., the maximum hourly rainfall observed at the Shinfa rain gauge
station during Typhoon Morakot), and A = 29.7 ha. Hence, Qwp was 6.8 m3/s
according to the rational formula described above. The discharge bulked
coefficient ( cb ) depends on conditions of sediment supplementation. The cb
value can be high when a watershed has a high landslide ratio or when there is
high sediment supplementation. The debris flow discharge in this study was
determined by Eq. (3), and the cb value was calibrated by comparing the results
obtained from numerical simulations to those obtained in the field investigation.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 41
the FLO-2D user’s manual [4]. Stage one had a cV = 0.55–0.65 for landslides
and stage two had a cV = 0.48–0.55 for debris flows. The inflow hydrograph
used in this study is shown in Figure 3.
Qdp= cb Qwp
Discharge (m3/s)
1 2
PM
4 5 6 7 8 9 10 11 12
Time (h)
Figure 3: The inflow hydrograph used for this study. The hydrograph was
divided into stages 1 and 2 for simulations of debris flow.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
42 Flood Recovery, Innovation and Response IV
4 Results
4.1 Discharge bulked coefficient
If the water discharge contained in the debris flow discharge is solely contributed
by direct runoff ( Qwp ) (i.e., water flow discharge), the debris flow discharge
( Qdp ) is directly related to Qwp , and it is equivalent to the sum of Qwp and the
sediment discharge ( Qs ) (where Qs = cV Qdp ). The discharge bulked coefficient
( cb ) in Eq. (3) can be expressed as:
cb (1 cV ) 1 (4)
Similar to Eq. (4), Takahashi [17] derived cb (1 k c *cV ) 1 for debris flows
generated from gully bed erosion where k c * c* 1 and c* is the volumetric
concentration of the sediment layer on the gully bed. The maximum cV values
observed ranged up to 0 .9 c * (Takahashi [17]). Based on Takahashi’s research,
the maximum cb = 10 if cV = 0.9c* . This implies that the maximum debris flow
discharge is 10 times that of the water flow discharge. However, in the
relationships for cb (1 cV ) 1 or cb (1 kc *cV ) 1 , ground water or the water
contained in the sediment layer was not considered. Hence, the cb value
calculated by cb (1 cV )1 or cb (1 kc*cV ) 1 may underestimate the discharge for
debris flows induced by large landslides. The peak water flow discharge ( Qwp )
determined by the rational formula in this case study was 6.8 m3/s. Here, the cV
value was calculated by the relationship of the equilibrium concentration
(Takahashi [17]):
tan
cV (5)
(Gs 1)(tan tan )
where Gs is the specific gravity; , the friction angle; and , the angle of the
gully bed in flow section. Using a Gs = 2.65, = 35 o , and = 17 o , the cV
value determined from Eq. (5) was cV = 0.47. Also, the cb = 1.89 according to
Eq. (4). These data imply that the debris flow discharge was 1.89 times that of
the water flow discharge ( Q dp 1.89 Q wp ). The inundation area was modeled using
an inflow hydrograph of debris flow discharge of Qdp = 16.2 m3/s ( 1.89 Qwp ), a
duration of 2 h, and the rheological parameters (i.e., y and b ) computed with
Eqs. (1) and (2). Additionally, a cV = 0.47, n = 0.15, and k = 2285 were used
as inputs. Figure 4 shows the inundation area of debris flow from the FLO-2D
simulations. The inundation area and deposition depth from the simulations were
smaller than those determined from the field investigation due to an
underestimation of debris flow discharge. Besides direct runoff, the water flow
that initiated the debris flow likely came from ground water or water contained
in sediments that was brought in by the landslides. Furthermore, water flow
could have been blocked by the sediment brought in by landslides, which would
have rapidly increased water storage in the watershed. When the stored water
combined with sediments burst over a short period of time, this could have led to
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 43
a high debris flow discharge. The cb value calculated with Eq. (3) ranged from
15 to 20 when the inflow hydrograph followed the type shown in Figure 3 and
the debris flow volume was estimated at 1,000,000 m3. The cb value in the study
area was calibrated by comparisons of the numerical simulations to field
investigation data.
When n = 0.15, cb = 18, and the cV values for stages one and two were 0.64
and 0.50, respectively, in the inflow hydrograph (Figure 3), the depositional area
and deposition depth from the simulations were close to those observed during
the field investigation (Figure 5). The deposited depth in the debris flow
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
44 Flood Recovery, Innovation and Response IV
inundated area was over 6 m. The debris flow discharge was 18 times that of the
peak water flow discharge due to the high landslide ratio (40.7%) in the Sha-
Xinkai gully. The simulated results also showed that the debris flow rapidly
inundated the downstream area at 8:30–9:00 PM on August 8, 2009, with a
maximum velocity of 6.6 m/s.
5 Conclusions
The study area in the Sha-Xinkai debris flow gully had a small watershed area
( A ) and a high landslide ratio ( RL ) ( A 29.7 ha and RL = 40.7%). For this type
of the debris flow gully, the peak water discharge computed from the rational
formula was small and it likely underestimated the inflow hydrograph in debris
flow simulations. The depositional area and deposition depth in the simulations
were strongly affected by the inflow hydrograph that was associated with the
discharge bulked coefficient ( cb ) and the volumetric concentration ( cV ). A
method to reasonably determine cb and cV is important for debris flow
simulation research. In this study, the relationship between debris flow discharge
( Q dp ), peak water flow discharge ( Q wp ) (or cb value), and cV values in the
inflow hydrograph were calibrated by comparing the results obtained from the
numerical simulations to data from a field investigation. The debris flow
discharge in the Sha-Xinkai gully had a Q dp = 18 Q wp or cb = 18. Two cV
values for two different stages of the hydrograph were used to evaluate
rheological parameters (yield stress y and viscosity ), and these values were
cV = 0.64 and 0.50 for stages one and two, respectively. Calculation results also
indicated that the simulated sediment volume was approximately 1,000,000 m3,
the maximum flow velocity was about 6.6 m/s, and the maximum depth on the
flow was over 6 m. The simulated average depth was close to the depth observed
in the field investigation. These data may useful as a reference for future hazard
zone mapping and debris flow hazard mitigation.
References
[1] Chen, T. C., Wu, C. C., Weng, M. C., Hsieh, K. H. & Wang, C.C., Slope
Failure of Lawnon Basin Induced by Typhoon Morakot, Sino-Geotechnics,
122, pp. 13–20, 2009 (in Chinese).
[2] Wang, C.M., Lee, S.P., Li, C.C., Tsang, Y.C. & Shieh, C.L., Disasters
caused by Typhoon Morakot, Journal of the Taiwan Disaster Prevention
Society, 2(1), pp. 27–34, 2010 (in Chinese).
[3] Chen, J.C., Huang, W. S., Jan, C.D. & Yang, Y.H., Recent Changes in the
Number of Rainfall Events Related to Debris-Flow Occurrence in the
Chenyulan Stream Watershed, Taiwan, Nat. Hazards Earth Syst. Sci., 12,
pp. 1539–1549, 2012.
[4] FLO-2D, FLO-2D Users Manual, Ver. 2009. FLO-2D Software Inc,
Nutrioso, AZ, USA, 2009.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 45
[5] Lin, M.L.,Wang, K.L. & Huang, J.J., Debris flow run off simulation and
verification – case study of Chen-You-Lan Watershed, Taiwan., Nat.
Hazards Earth Syst. Sci., 5, pp. 439–445, 2005.
[6] Tecca, P.R., Genevois, R., Deganutti, A.M., & Armento, M.C., Numerical
modelling of two debris flows in the Dolomites (Northeastern Italian
Alps)., Debris-Flow Hazards Mitigation: Mechanics, Prediction, and
Assessment, Chen & Major, eds, Millpress, Netherlands, pp. 179–188,
2007.
[7] Sosio, R., Crosta, G.B. & Frattini, P., Field observations, rheological
testing and numerical modelling of a debris-flow event. Earth Surf.
Process. Landforms, 32, pp. 290–306, 2007.
[8] Stolz, A. & Huggel, C., Debris flows in the Swiss National Park: the
influence of different flow models and varying DEM grid size on
modeling results. Landslides, 5, pp. 311–319, 2008.
[9] Jakob, M. & Weatherly, H., Integrating uncertainty: Canyon Creek
hyperconcentrated flows of November 1989 and 1990., Landslides, 5,
pp. 83–95, 2008.
[10] Hsu, S. M., Chiou, L. B., Lin, G. F., Chao, C. H., Wen, H. Y. & Ku, C. Y.,
Applications of simulation technique on debris-flow hazard zone
delineation: a case study in Hualien County. Taiwan, Nat. Hazards Earth
Syst. Sci., 10, pp. 535–545, 2010.
[11] Sodnik, J. & Mikos, M., Estimation of magnitudes of debris flows in
selected torrential watersheds in Slovenia, Acta geographica Slovenica,
46(1), pp. 93–123, 2006.
[12] SWCB, Disasters caused by Typhoon Morakot in Taiwan, 1999. Soil and
Water Conservation Bureau (SWCB), Taiwan, 2009 (in Chinese)
[13] Dai, J., et al., An experimental study of slurry transport in pipes. Proc.,
Int. Symposium on River Sedimentation, pp. 195–204, 1980.
[14] Fei, X. J., Bingham yield stress of sediment water mixtures with
hyperconcentration, J. Sediment Res., 3, Beijing, China, pp. 19–28, 1981.
[15] Chen, J.C., Jan, C.D., and Lee, M.S., Reliability Analysis of Design
Discharge for Mountainous Gully Flow, Journal of Hydraulic Research,
46(6), pp. 835–838, 2008.
[16] SWCB, Technical Handbook of Soil and Water Conservation, Soil and
Water Conservation Bureau (SWCB), Taiwan, 2005. (in Chinese)
[17] Takahashi, T., Debris Flow. IAHR Monograph. Balkema, Rotterdam,
1999.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
This page intentionally left blank
Section 2
Risk assessment
This page intentionally left blank
Flood Recovery, Innovation and Reponse IV 49
Abstract
Catastrophe models often cover large geographic areas spanning multiple
countries or, in the case of flood models, entire watersheds. Models must be
sufficiently detailed to accurately account for hydrologic variation, which is
notably challenging when the modeled region is large. This is particularly true for
flood models, which require a highly detailed dataset, usually derived from a
digital terrain model (DTM), for reliable floodplain mapping. For one-dimensional
(1D) hydraulic models, the floodplain mapping approach tends to yield flat
surfaces often resulting in artefacts and inconsistencies near river confluences.
Because flood extent is limited by the length of cross-sectional lines along the
floodplain, these flat surfaces tend to drop sharply when the simulation reaches a
flat delta. The use of a two-dimensional (2D) model avoids these problems, but at
a high computational cost, and requires high quality terrain and bathymetry data.
This paper presents a new methodology for mapping floodplains using water
elevation points along a river network obtained from a 1D hydraulic model and a
DTM. The methodology applies kinematic and diffusion wave equations in
a simplified manner, using water elevation points as internal boundary conditions.
Several parameters control the expansion and smoothing algorithms that generate
realistic flood extent maps for different return periods. This methodology is
particularly suitable for modeling large domains since it produces accurate results
but requires much less computational time than a 2D model. In addition, because
the computation uses several source points per cross section, the flood extent is
not limited by the cross-sectional length, making this methodology appropriate for
levee breaches and in cases where river banks are not well defined and the cross-
sectional geometry is derived from a DTM.
Keywords: risk assessment, flood, flood mapping, large scale model.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140051
50 Flood Recovery, Innovation and Response IV
1 Introduction
Flood events in combination with human activity and land use changes threaten
both life and property in much of the world. The human and economic losses
inflicted by flood events have forced communities and governments to adopt new
direct and indirect measures to prevent, assess, and reduce the risk of flooding.
Over the last thirty years, a plethora of 1D, 2D and coupled flood models have
been developed, most of which are commercially available. Risk maps often
require the evaluation of risk for a given non-exceedance probability (often
referred to as return period maps), and thus, steady state models are often
preferred, again for their computational efficiency [1]. Flood hazard maps are
typically produced by governmental agencies, such as FEMA in United States,
ZÜRS in Germany, and many others across Europe. In this context, large scale
catastrophe risk models are catching the attention of more and more researchers in
academia and in industry. These models are used to assess the effect of
catastrophic flood on larger areas as opposed to local and detailed studies.
Despite the large number of such models, the need for nationwide medium-to-
high resolution inundation maps has led to the development of fast numerical
solutions with reduced computational effort.
In this study, we provide a tool for mapping hydraulic model results from a 1D
steady state model using a simplified quasi-physical approach that alternates
between the use of kinematic waves and diffusion waves to interpolate water
elevations between cross sections. This methodology eliminates the presence of
artefacts and drops in flood maps at confluence and reduces the computational
effort required in comparison to 2D models. The same kinematic and diffusion
waves approach is used to predict the flow level through a breach, using, as
boundary conditions, the solution obtained from 1D numerical model and a
volume hydrograph. Finally, the mapping algorithm provides an envelope of the
maximum depth for a given return period.
2 Methodology
One dimensional steady state models can rapidly assess the intensity and the extent
of flooding at specific return periods, provided that the effect of floodplain storage
is negligible [2, 3] and provided that the accuracy of such models is checked at
stream junctions, branches, and lateral inflow [1, 4]. One dimensional model
results, however, are only available locally at predefined model cross sections.
Therefore, 1D model solutions need to be interpolated between cross sections to
fill the gaps between them and can also be used in conjunction with 2D models to
simulate levee overtopping or flow through levee breaches.
Examples of interpolation of 1D hydraulic model results to create a flood extent
map include triangular irregular networks (TIN) generated by HEC-GeoRAS, as
well as chained interpolation between cross sections in MIKE-11.
The industry often requires large scale models (often national or continental) at
medium-to-high resolutions (specifically, 30 m to 90 m). DTMs and land use
datasets provide an essential source of input to generate such large scale models.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 51
Free surface wave propagation along rivers is generally approximated using the de
Saint-Venant equation [7, 8]:
/
0 (1)
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
52 Flood Recovery, Innovation and Response IV
and 2) a second smoothing phase, in which the water elevations from the
expansion phase are smoothed to reduce the number of artefacts from the elevation
maps (for example, an unrealistic flood elevation between two adjacent cells, i.e.,
wall of water).
For the first iteration, each source point will propagate to the next cell using
either a diffusion or kinematic wave to compute the elevation at the empty
neighbouring cells (subscript C). Figure 1 shows the neighbouring convention
used by this algorithm, in which z is the DTM elevation from the reference datum.
Each source point will loop through the 8 neighbouring cells.
zc , ζ c zc , ζ c zc, ζc zc , ζ c zc, ζc
zc, ζc zc, ζc
zc , ζ c zc, ζc zc , ζ c
(a) (b)
Figure 1: Neighbouring scheme and expansion steps for: a) 1st iteration and
b) second iteration.
For each successive step, every neighbour cell at the previous step will become
a source point and the original source point will be removed from the set of source
points (Figure 1(b)). The iteration will continue until the ground elevation will not
allow any further expansion from source cell; that is, when the list of source cells
is empty, or when the maximum number of iterations is reached. After each
expansion step, flooded cells’ elevations are added in a smoothing array, while the
source cell evaluated directly from the cross section will not be added, in order to
prevent any alteration of the 1D solution along the cross section.
Thus, for each expansion step, the water elevation of each cell will be
calculated according to eqn. (2):
∆ (2)
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 53
maximum water depth propagation computed from cell to cell in steady state
conditions will assume the form:
∙ (3)
Note that the source water depth propagates without dispersion in eqn. (3).
In contrast, in the case of a diffusion wave, the wave propagates to its maximum
depth while reducing its peak during wave propagation. The dissipation ∆ can be
calculated by rearranging the diffusion wave in eqn. (1) and evaluating the friction
slope through Manning equation:
∙ ∙
∆ ∆ ∙ ∙ ,∆ (4)
Herein, ∆L=∆X·θ is the distance from the cell centres, ∆X is the cell size, and θ is
a direction factor equal to θ=1 for horizontal and vertical neighbours (θ=1.41 for
diagonal cells), S0= |ζs-ζc| is the DTM slope, nL(T) is the longitudinal (transverse)
Manning coefficient, VL(T) is the longitudinal (transverse) flow velocity, and ψL(T)
is the downhill scale factor. Manning’s n coefficient and velocity are set constant
and do not depend upon the orography of the area. Therefore, ψ and S0 are used to
scale the flow velocity from flat areas to steep areas, and obtain different energy
dissipation values for catchments with different slopes.
The maximum drop ∆Emax is enforced to reduce the energy dissipation in
presence of artefact that may present high slope values.
The selection between kinematic and diffusion wave is based upon the
definition of longitudinal and transversal wave propagation. In case of longitudinal
wave propagation, the diffusion wave assumes velocity and Manning coefficients
that approximate the wave propagation along the river centre line. Conversely, the
transverse wave propagation assumes lower Manning’s n, velocity and thus energy
dissipation, to simulate propagation normal to the river centre line.
For transversal wave propagation, in fact, the free surface slope cannot be
approximated using the ground slope and, thus, the kinematic assumption would
lead to high error in predicting the water surface [7, 9] and cannot be used to
evaluate the water elevation.
In this study, the distinction between longitudinal and transverse wave
propagation is achieved through the location of the expansion cell C originating
from a source point lying on a cross section.
In order to differentiate between the two different wave propagation, it is
necessary to first introduce the Euclidean polygons. These polygons are defined
as the sets of point with a minimum Euclidean distance from the source cross
section. The DTM space is, therefore divided into different Euclidean polygons,
corresponding to each cross section.
Wave propagation of cells within the Euclidean polygon will be assumed to
have transverse wave propagation, while cells outside the Euclidean polygon will
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
54 Flood Recovery, Innovation and Response IV
be assumed to have longitudinal wave propagation, and thus, will have higher
energy dissipation to smoothly fill the gap between cross sections.
When the expansion cell lies inside the Euclidean polygon, the algorithm
selects the water elevation to be the maximum of the two elevations computed by
the diffusion or by the kinematic wave. During the longitudinal wave propagation
(thus, outside the Euclidean polygon), the water elevation is evaluated as the
minimum yielded by the kinematic and diffusion equations. Figure 2 shows
the results of the expansion algorithm from the 1st iteration to the final map extent
(Rhone River near Martigny). Figure 2(a) shows the first step of the expansion
algorithm. Here, the green line represents the cross section automatically
generated from the catchment and flow line characteristics (green lines), while the
green dots represent the source point location.
(a) (b)
(c)
Only the source points with an elevation above the DTM elevation can expand
from the 1st iteration, as shown by the flooded cells (blue) under the source point.
Herein, not all the points have been activated to show the potential of the algorithm
to flood area outside the cross section limit. As the algorithm executes, gaps
between cross sections are filled in both directions (Figure 2(a)). Finally, as the
algorithm proceeds, area outside the cross section extents will be flooded, in both
longitudinal and transversal direction. As shown in the figure (see the lower right
quadrant of Figure 2(c)), the flood extent map does not stop at the end of the cross
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 55
section; rather, the algorithm is capable of expanding the modeled flood extent
into those areas that would be left dry using traditional approaches.
The general framework presented here, i.e., kinematic and diffusion wave
propagation, can be used to evaluate the water depth from levee overtopping or
breach to provide a solution that adapts to the morphology of a floodplain. The
simplified 2D mapping algorithm has parameters selected to maximize the
correspondence between the 1D solution and the 2D solution. This is done to
prevent “wall of water” and map discontinuities at junctions observed using TIN-
based flood maps in which cross sections are not long enough to cover the extent
of the floodplain. However, in cases of levee breach or overtop, the longitudinal
dissipation may be either too low or too high; therefore, models of levee failure or
overtopping need to be driven by different physical assumptions, volume carried
by the hydrograph and wave duration.
Several reduced complexity models, or simplified models, have been
developed to assess the risk of levee failure, overtopping, or inundation of urban
areas. Typical approaches include models based on the discretisation of the
diffusive wave equation on Cartesian grids [16], regular and irregular storage cell
models [3, 5, 10–14] and raster based inertial models [15].
These reduced complexity methodologies are based on different assumptions,
but they all aim to decrease the computational cost of assessing levee failure
compared to fully physical 2d shallow water solutions. However, these reduced
complexity methodologies also analyze the evolution of the flood extent within a
certain event, and, thus may need data post processing, which requires additional
analysis and computational overhead, especially over a large domain.
In this paper, we use these types of risk maps to assess the maximum flood
elevation and extents associated with a certain return period in case of levee breach
and overtopping. Velocity within the floodplain will be neglected, due to the
relatively lower velocity in the floodplain compared to the main channel.
Starting from the source points at the end of a cross section, the average depth
can be evaluated using the information from the volume of water available within
a certain time or from a given hydrograph.
To account for the added water volume that may come from a breach, the
expansion parameters must be dynamically changed to adapt the solution to the
average available volume associated with a certain return period.
For each step, the available volume is redistributed in the expanded area such
that the average water depth for the next expansion multiplied by the total
expanded area matches the available volume.
Thus, the expansion volume at the step n, , is calculated as:
̅ (5)
where:
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
56 Flood Recovery, Innovation and Response IV
∆
∆
(6)
where, similarly to the method adopted by Liu and Pender (2010) [5], is the
total maximum volume available from the breach hydrograph at the expansion
step n from the beginning of the simulation, either obtained from a historical event
or from a simulated flood event. In this case, if the error has a positive bias, ψ
increases by a certain amount, in order to increase the dissipation for the next step
and adjust the volume. Since each expansion step depends on the model resolution
and not on a time step, the total volume available at each expansion step needs to
be evaluated from the hydrograph associating a certain time step to each expansion
step. In first approximation, this time step can be associated with typical flood
plain velocity, although more complex formulation can assume a dependency on
the average velocity evaluated from the flow surface gradient. Herein, the
floodplain velocity uexp will be assumed constant and needs to be calibrated to
produce reasonable extents, as discussed later in the validation.
To be more explicit, for a triangular flood hydrograph of which the volume
overtopping or flowing through the flood defence is known, the volume at
the expansion step n will be calculated as:
for n (8)
1 for n (9)
where nTP=TP / (Δx·uexp), nTP is the number of expansion iterations at the
hydrograph peak TP, nTE=TE / (Δx·uexp) is the number of expansion iterations at
the end of the hydrograph, uexp= is the average flood expansion velocity, and Δx is
the cell grid size.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 57
where the P is the generic cell in both model (M) and pilot domain (D) either
flooded (subscript 1) or dry (subscript 0). Modeled flood extents have been created
using the U.S. national elevation dataset (NED) with a resolution of 30 m.
Figure 3 compares the FEMA flood extent maps and the modeled flood extent
(AE zones in light blue and X zones protected by levee in yellow) for the Greater
Cincinnati region; the modeled flood extents were determined using the procedure
described in this paper, in which water elevations are propagated from cross
sections.
(a) (b)
The agreement between the FEMA extent map and the modeled map in
Cincinnati urban area considering both AE and X protected zones is F=0.76. The
levee breach model has been validated by comparing the 2011 breach of Cairo,
Illinois and the USACE simulation [6]. The Birds Point, Illinois, breach reached
its maximum volume after 3 days of operation with a maximum volume of around
1.3·109 m3 breach. For reference, USACE provides a model of the 2011 event
Cairo breach showing the floodway evolution after 120 hours (5 days) [6], which
shows an average propagation velocity of 0.2 m/s. The USGS provides a large
dataset of field measurements for this event (around 20 gage stations inside the
floodway, shown by the green dots in Figure 4). The maximum value of each gage
has been selected and used to calibrate the model. A range of different uexp in the
range 0.01 m/s< uexp <0.5 m/s has been selected. According to these data,
the minimum RMSE has been obtained using for uexp=0.2 m/s (RMSE=1.21 m;
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
58 Flood Recovery, Innovation and Response IV
F=0.7). Figure 4 compares the observed flood extent obtained from Landsat image
to the modeled flood extent. Water extent from Landsat images have been
extracted using the modified normalized difference water index (MNDWI [18,
19]), using a value of MNDWI>0.3 to detect water features on the image.
Two types of boundary conditions mainly affect the raster model, i.e.,
topography and Manning’s n coefficient correlated with the land use. However,
the results of both the general flood extent model and the levee failure model
presented in this study fit well with observed flood data. Further, the performance
of both models is comparable to that observed by Bates and De Roo [16] at similar
DTM resolutions.
In addition, we noted that a simplified flood model that employs a DTM with
coarser resolution produces modeled flood extent maps that fit less well to
observed flood extents. For example, Bates and De Roo [16], observed that
choosing a 25 meter resolution land use raster rather than 100 m resolution raster
inflicts a 10% loss in fit quality between the modeled and observed flood
extents.” In both Ohio River and New Madrid floodplain, fine resolution DTM
topography accounts for additional and localized levee protection, which cannot
be correctly assessed using the original resolution and need to be manually
surveyed and added on the DTM.
(a) (b)
Figure 4: Comparison between (a) simulated levee failure with uexp=0.2 m/s
(green dots: USGS gage station; pink line: USACE accredited
levees, red line: Mississippi river centre line) and (b) Landsat
extracted water bodies (MNDWI>0.3) for New Madrid floodway.
Other biases may derive from the use of a global land use in both modelling
approaches that may account for localized differences between predicted extents
and satellite images. Moreover, the use of a single uexp used for the levee breach
approach to estimate the amount of volume available for each expansion step may
lead to depth underestimates during the first development phase, where flow
expansion is generally faster due to a steeper free surface. In contrast, depth
overestimates can occur during the last ponding expansion phase of the breach,
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 59
where velocity may reach lowest expansion value. Finally, the model does not
account for ponding and backwater effects, which further reduce the
computational performance of this methodology.
4 Conclusion
A new framework to map 1D model results both along river flood plain and due
to overtopping or levee failure has been presented. The model is based on simple
development of kinematic and dynamic wave propagation. The modeled results
are comparable to those produced by other simplified models, while maintaining
a reduced computational cost. Boundary conditions, such as topography and land
use, account for the majority of discrepancy between the FEMA flood map (where
floodplains are manually surveyed) or historical inundation maps. Finally, further
analysis of the sensitivity of the model on the parameters used to reproduce flood
extent maps in both the general and levee framework should be conducted.
References
[1] Mapping the Zone: Improving Flood Map Accuracy; National Research
Council; Washington, DC: The National Academies Press, 2009.
[2] Horrit, M.S. & Bates, P.D., Evaluation of 1D and 2D numerical models for
predicting river flood inundation, Journal of Hydrology, 268, pp. 87–99,
2002.
[3] Krupka, M., Pender, G., Wallis, S., Sayers, P.B. & Mulet-Marti, J., A rapid
flood inundation model, In Proceedings of the 32nd Congress of the
International Association For Hydraulic Research, pp 1–28, 2007.
[4] Büchele, B., Kreibich, H., Kron, A., Thieken, A., Ihringer, J., Oberle, P.,
Merz, B. & Nestmann, F. Flood-risk mapping: contributions towards an
enhanced assessment of extreme events and associated risks, Nat. Hazards
Earth Syst. Sci., 6, pp. 485–503, 2006.
[5] Liu, Y. & Pender, G., A new rapid flood inundation model, In Proceedings
of the First IAHR European Congress, ed. S. Arthur, Edinburgh, UK, 2010.
[6] DeHaan, H., Stamper, J. & Walters, W., Mississippi River and Tributaries
System 2011 Post-Flood Report, USACE, Mississippi Valley Division,
2012.
[7] Ponce, V.M., Li, R.M. & Simons, D.B., Applicability of kinematic and
diffusion-models, Journal of the Hydraulics Division — ASCE, 104 (3),
pp. 353–360, 1978.
[8] Mujumdar, P.P., Flood wave propagation – The Saint Venant Equation,
Resonance, 6 (5), pp. 66–73, 2001.
[9] Singh, V.P. & Aravamuthan, V., Errors of kinematic-wave and diffusion-
wave approximations for steady-state overland flows, Catena, 27 (3–4),
pp. 209–227, 1996.
[10] Cunge, J.A.., Two-dimensional modeling of flood plains. Water Resources
Publications, 17, 705–762, 1975.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
60 Flood Recovery, Innovation and Response IV
[11] Moussa, R. & Bocquillon, C. On the use of the diffusive wave for modelling
extreme flood events with overbank flow in the floodplain, Journal of
Hydrology, 374, pp. 116–135, 2009.
[12] Castellarin, A., Domeneghetti, A. & Brath, A., Identifying robust large-scale
flood risk mitigation strategies: A quasi-2D hydraulic model as a tool for the
Po River, Physics and Chemistry of the Earth, 36, pp. 299–308, 2011.
[13] Gouldby, B., Sayers, P., Mulet-Marti, J., Hassan, M. & Benwell, D.,
A methodology for regional-scale flood risk assessment. Proceedings of the
Institution of Civil Engineers - Water Management, 161, pp. 169–182, 2008.
[14] Falter, D., Vorogushyn, S., Lhomme, J., Apel, H., Gouldby, B. & Merz, B.,
Hydraulic model evaluation for large-scale flood risk assessments, Hydrol.
Process., 27, pp. 1331–1340, 2013.
[15] Bates, P.D., Horritt, M.S. & Fewtrell, T.J., A simple inertial formulation of
the shallow water equations for efficient two-dimensional flood inundation
modelling, Journal of Hydrology, 387, pp. 33–45, 2010.
[16] Bates, P.D. & De Roo, A.P.J., A simple raster-based model for flood
inundation simulation, Journal of Hydrology, 236, pp. 54–77, 2000.
[17] Aronica, G., Bates, P. D. & Horritt, M. S., Assessing the uncertainty in
distributed model predictions using observed binary pattern information
within GLUE, Hydrol. Process., 16, pp. 2001–2016, 2002.
[18] Xu, H., Modification of normalised difference water index (NDWI) to
enhance open water features in remotely sensed imagery, International
Journal of Remote sensing, 27 (14), pp. 3025–3033, 2006.
[19] Ho L.T.K., Umitsu M. & Yamaguchi Y., Flood hazard mapping by satellite
images and SRTM DEM in the Vu Gia – Thu Bon alluvial plain, Central
Vietnam, International Archives of the Photogrammetry, Remote
Sensing and Spatial Information Science, 38 (8), pp. 275–280, 2010.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 61
Abstract
Today, flood risk in Japan occurs mainly in high density populated areas, as a
consequence of the rapid urban development of the deltaic plains of Japan during
the second half of the 20th century. At the end of the 20th century risk
management began to shift from mainly structural management to a more
“integrated” management. The evacuation process is one of the factors revealing
this shift. In Nagoya the evacuation process enhancement started with the Tokai
flood disaster (September 2000) and continues to this day. The most recent flood
events (urban flood of 2008 and typhoon No. 14 of 2011) highlight, however,
how the crisis management can still be vulnerable regarding evacuation. Our
research intends to assess the vulnerability factors of the crisis management
system, and especially of the evacuation process through interviews and a
questionnaire analysis method, in order to propose an integrated way of dealing
with evacuation in the case of a flood, imputing on GIS geographical as well as
social characteristics and evacuation patterns. Our research shows that the
evacuation process is effective despite low evacuation rate during past flood
event. In that regard improving the evacuation process cannot be separated from
the improvement of informational tools, but it can be seen that the possession of
hazard maps have few impact on evacuation decision. The efficiency of the
evacuation process in the case of a small to moderate flood event could therefore
be enhanced as the large-scale evacuation broadcast tends to target a population
in which more than half of the people do not need to evacuate. In the case of a
small flood event those repeated evacuation demands can increase a relatively
false sense of security and a loss of interest to flooding in general.
Keywords: vulnerability, adaptive capacity, floods, evacuation, GIS, Japan.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140061
62 Flood Recovery, Innovation and Response IV
1 Introduction
With its three biggest cities located in deltaic fluvial plains, Japan is at a high
flood risk. In Japanese megacities, although innovative and efficient measures
have been taken, the flood risk remains, and the vulnerability to flood risk is
expected to increase with the combination of natural factors characterizing the
hazards (expected and unexpected effects of climate change), territorial factors
(settlements of stakes in lowlands area, apparition of “new” urban-type floods),
and societal factors (knowledge and acceptance of flood risk, willingness to
evacuate).
As the flood risk is changing, so is its management, and on concentrating our
research on the evacuation process the purpose of this paper is to define small
scale vulnerability of dwellings and its relationship to the evacuation
management at city and prefectural scale. Through the changes in flood risk
management since the disaster of the 11 and 12th September 2000 in Nagoya-city
will be analysed the vulnerability and adaptive capacity concepts to flood risks in
flood risk management, from risk actors standpoint and GIS mapping.
The concept of vulnerability is one of the numerous tools that can be used in risk
analysis. It has been described as the flip-side of the resilience concept [1] and of
the robustness concept [1, 3], as it is described in broad terms as “the
susceptibility to be harmed” [4]. The resilience is described as “the ability of a
system to absorb shocks, to avoid crossing a threshold into an alternate and
possibly irreversible new state, and to regenerate after disturbance” [5], and the
robustness concept as “a systems ability to remain functioning under
disturbances” [3]. But if the vulnerability has not always the positive aspect of
the here above two concepts, it consists in a helpful analysis tool when
confronted to the evaluation of a system’s evolution, as it can be considered as
“the potential for a change or transformation of the system when confronted with
a perturbation, rather than the outcome of this confrontation” [2].
The use of the vulnerability concept in risks studies has been used in two
related and complementary approaches. The first, classic (end of 1970s)
approach consisted of measuring the potential exposure of the different stakes in
a system [6]. In this case, the evaluation of the vulnerability is made through the
exposure as an attribute of the relationship between the system and the
perturbation [2]. It corresponds to a technical and effective management of
disasters, which could be summed as the following non mathematical equation:
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 63
The second approach, enriched by research on natural hazards [9, 10], helped
to understand how the vulnerability concept cannot be taken outside of the
system it’s related to. The vulnerability being the propriety of a system, evolving
as the system evolves, and can be revealed during an event, or a disaster.
A key-component to the vulnerability concept is the “adaptive capacity” concept,
“the flexibility of ecosystems, and the ability of social systems to learn in
response to disturbances” [11, 12].
Because the flood risk management in Nagoya these last years knew
noticeable transformations, the vulnerability as propriety of a system increasing
this system to be harmed in case of a perturbation, as well as the concept of
adaptive capacity as formulated hereinabove by Turner et al. [12] will be used to
analyse the ability of the flood risk Japanese management system to evolve
during the past years and the influence of this evolution on the evacuation
process in Nagoya-city.
In order to build an efficient risk management, aside from the adaptive capacity
is also needed, and is paired with it, the ability to build efficient risk governance.
It has been made clear in natural hazards and in climate change research that
vulnerability is the propriety of a system. Megacities are a good example of
complexes, human-made, multi-level chain reactions, highly vulnerable type of
systems. Flood risk management in such systems needs the collaboration of very
different actors, with purposes and focuses that may differ. This is why it is
needed to take into account risk governance in the risk management system
vulnerability and adaptive capacity factors.
Building risk governance can be summarized in collecting, analysing and
communicating relevant risk information (through a complex web of actors,
rules, conventions, processes and mechanisms), taking risk management
decisions at the right time, and for those information and decisions to be
understood by the public concerned [13]. Although a difference has to be made
between risk management and risk governance. The definitions for
risk management are scarce, and can have different meanings, from risk
response – the risk management being the management of the crisis to which the
actors are confronted [14] – to the management of the risk at all times of the risk
(mitigation, preparedness, response and recovery [15]). Will be considered here
that risk governance is part of the risk management system, which entails the
different actors, the actions they decide to setup (hardware and software
measures), and the concrete results in the risk system and the consequences those
measures will have. Risk governance would be then the central part of the risk
management.
Building an efficient risk governance, to go further, includes the idea brought
up by the studies on climate change that risks in general have to be thought as
long-term duration processes. Adger [16] made a clear differentiation between
effectiveness and efficiency in adaptability capacity. The purpose of
effectiveness consists in responding to objectives that have been fixed, in
reducing the impacts of hazards and exposure, or to reduce the risk and avoid the
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
64 Flood Recovery, Innovation and Response IV
3 Methodology
3.1 Study area
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 65
Nagoya-city is the 4th largest Japanese city in Japan, with a total population of
2,272,075 [18] on the 1st of January 2014. Floods in Nagoya-city are not an
unknown event, the main water-related disasters that happened during the past
70 years were the Ise-bay typhoon of 1959, the flood of 2000 (called Tokai
flood) and the 2008 urban flood. The Tokai flood occurred on September 11 and
12, and damaged part of the Tokai region, due to heavy rainfall, amounting up to
a total of 567 mm (one third of the average annual rainfall). The Shin River and
Tenpaku River suffered levee breaches, the Shonai River and Yahagi River
flooded by levee overtopping, and the rainfall accumulating near the levees could
not be evacuated due to lack of drainage ability [19, 20]. The total loss for the
Tokai region reached the amount of 978.3 billions of yen, 155 injured people and
10 fatalities for the Tokai region, and 37% of Nagoya-city urban territory
flooded, 45 injured and 4 fatalities.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
66 Flood Recovery, Innovation and Response IV
INTERVIEWS
To official risk managers To population
Last emergency evacuation
Software measures setup
lived
Vulnerability Willingness to evacuate
factors Hardware measures setup
in the future
Risk knowledge
4 Results
It has to be pointed out that despite the amount of physical damages the human
damages during the Tokai flood in 2000 were low. The risk managing system
succeeded in an emergency evacuation of more than 5,500,000 people [19],
despite the absence of official evacuation process at that time. Evacuation was
difficult because it happened as an emergency measure after levees on Shin
River breached and most of the people had to evacuate while the water level was
high, but considered all in all successful. The Tokai flood marked for Nagoya the
starting point of the creation of evacuation process measure. The evacuation
process measure gathered old risk management actors: the national River Bureau
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 67
It has been difficult to find persons willing to talk about flood disaster and
evacuation processes. As a consequence, the data collected during the interviews
to the population were analyzed in comparison with the survey for evacuation in
case of flood realized by Aichi prefecture in December 2011 (3 months after the
2011 flood event).
When interrogated about the risk culture, most of the interviewees answer
knowing the major past flood disasters (Ise-bay typhoon and Tokai flood), but
also the small event of the precedent year. The same findings have been found by
the Aichi prefecture survey, although during the interviews it was obvious that
resident who experienced the Tokai flood were reluctant to refer to it when asked
about floods (flood risk is accepted, but with limits). The risk culture is also
enhanced by the preparation (survival kits, knowledge of safe areas and risky
areas) to more general type of disasters (earthquakes).
The Tokai flood seems to be remembered by the people who lived in flooded
areas in 2000. It is not described as a shock, although the experiences related
clearly showed that the experience was not pleasant: car stuck in the high water
or because of the traffic, and impossibility to move, doubt about what to do and
where to go when the water reached the house… The rare persons talking about
the flood without referring to unpleasant experience were two people who stayed
at home “because I knew it to be safe”. The levee breaches especially seem to
remain a shocking enough event to be considered in the survey [21] a risk more
important than other flooding types. The trust bestowed upon authorities was not
a subject broached during the interviews. Most of the interviewed people seemed
to have high confidence in structural measures realized after 2000, but somehow
did not seem to be interested in evacuation warnings (most of the time “my
house if safe” was the main reason called upon). Comparing these results with
the 2011 Aichi prefecture survey, it seems that the trust in authorities is high
(expectation for government and the local residents association to improve the
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
68 Flood Recovery, Innovation and Response IV
Comparing the empirical data gathered during on the field and the statistical data
available for housing and households from Nagoya office, an error of 268
apartments (18% of all the 1424 apartments evaluated) have been noticed and
will be taken into account when evaluating the number of households
and persons in need of an evacuation for the three reference model floods. The
number of family members living in one household will also be considered.
The Komoharachou district being more an industrial type of district, the average
family members for a household is 1.7 whereas in the two other districts, it
reaches 2.5.
The repartition between housing and non-housing buildings is close to equal
(46.5% non-housing building for 56.5% building housing). Repartition of
evaluated population by building class is however very differentiated (figures 2
and 3 and table 2).
Depending on the hazard type the rate of persons that need evacuation differs,
knowing that most of the people living in risky areas live in high condominiums
(63%) they are therefore not in grave danger in case of a flood disaster, and a
large amount of persons do not need to evacuate, even if the more vulnerable
private two-stories house type is the most common. For Shonai River model
flooding 41% of the population evaluated would need to evacuate, 39% for a
Shin River model flood, and 0.8% on case of urban flooding.
In order to represent population need of evacuation in case of flood, maps
have been realized, to help understand patterns in housing vulnerabilities. The
repartition of housing is however too homogenous in these districts to notice
patterns.
C C
3% 1%
A B
17% 36%
A
B 63%
80%
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 69
5 Discussion
Adaptive capacity is one of the strong points of the risk management in Nagoya-
city. Confronted to a new kind of disaster, the risk management system respond
quickly and effectively (has been seen in 2000, and 2008) to new kind of hazards,
and new kind of disasters. The lessons learned from the past, and different
disasters are also observable (1995 Kobe earthquake), making the flood
risk management system flexible. This adaptive capacity allows the flood risk
management to evolve and to integrate new actors, and new purposes, building
therefore better risk governance.
The interviews to the population revealed less clear adaptive capacity, and
more potential vulnerability to flood in the fact that it is difficult to evaluate the
number of persons willing to evacuate in case of a disaster. From the population
standpoint, it was also difficult to evaluate their perception of the risk
management and how they were integrated in it, as the answer both to the
interviews and the survey [21] were unclear. It can be said, though, that
the acceptance of the flood risk might be high, and the preparedness good,
demonstrating a good adaptive capacity. The acceptance of a flood disaster is
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
70 Flood Recovery, Innovation and Response IV
6 Conclusion
Japanese flood management system in Nagoya-city has changed during these
past years, demonstrating a high adaptive capacity to new challenges flood risk
poses in megacities today. New flood risks appeared; they are caused by more
intensive hazards or generated by urban shape. The risk management aims and
succeeds in improving structural measures, developing software measures
through the integration of new actors, developing a more integrated management
and therefore better risk governance for flood risks. In that regard, the flood risk
vulnerability can be considered low, and the system aiming for efficiency. The
low evacuation rates for the last flood events seem to better correspond to
effectiveness achievement goal. Efficiency in evacuation procedures would be
better achieved with a clear understanding for the population of the received
information, making it easier for them to make a choice when confronted to a
disaster. On that matter, risk governance still has progress to do, despite the high
adaptive capacity of both risk managers and residents in risky areas.
References
[1] Folke, C., Carpenter, S., Elmqvist, T., Gunderson, L., Holling, C.S.,
Walker, B., Bengtsson, J., Berkes, F., Colding, J., Danell, K., Falkenmark,
M., Gordon, L., Kaspersson, R., Kautsky, N., Kinzig, A., Levin, S.A.,
Maler, K.-G., Moberg, F., Ohlsson, L., Olsson, P., Ostrom, E., Reid, W.,
Rockstro¨ m, J., Savenije, H., Svedin, U., Resilience and sustainable
development: building adaptive capacity in a world of transformations.
Report for the Swedish Environmental Advisory Council 2002:1, ed.
Ministry of the Environment, Stokholm, pp. 437-440, 2002.
[2] Gallopin, C.G., Linkages between vulnerability, resilience, and adaptive
capacity. Global Environmental Change, 16, pp. 293-303, 2006.
[3] Mens, M.J.P., Klijin, F., de Brujin, K. M., van Beek, E., The meaning of
system robustness for flood risk management, Environmental Science &
Policy, 14, pp. 1121-1131, 2011.
[4] Adger, W. N., Vulnerability, Global Environmental Change, 16,
pp. 268-281, 2006.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 71
[5] Miller, F., Osbahr, H., Boyd, E., Thomalla, F., Sukaina, B., Ziervogel, G.,
Walker, B., Birkmann, J., van der Leeuw, S., Rockström, J., Hinkel, J.,
Downing, T., Floke, K., Nelson, D., Resilience and vulnerability:
complementary or conflicting concepts? Ecology and Society 15(3):11,
pp. 1-25, 2010.
[6] Thouret, J. C., D’Ercole, R., Vulnérabilité aux risques naturels en milieu
humains : effets, facteurs et réponses sociales, Cahiers de Sciences
Humaines, 32(2), pp. 407-422, 1996.
[7] Reghezza, M., Réflexions autour de la vulnérabilité métropolitaine : la
métropole parisienne face au risque de crue centennale, doctorate thesis,
Paris Nanterre-University, pp. 58-63, 2006.
[8] Reghezza, M., La vulnérabilité, un concept problématique (chapter 3), la
vulnérabilité des sociétés et des territoires face aux menaces naturelles,
ed. F. Leone, F. Vinet, PUM: Montpellier, pp. 35-41, 2005.
[9] Wisner B., Westgate, K., O’Keefe, P., Taking the Naturalness out of
Natural Disasters , Nature, 260, pp. 566-567, 1976.
[10] Blaikie I. et al. At Risk: Natural Hazards, People’s Vulnerability, eds.
Routledge, London, pp. 88-123, 2003.
[11] Gunderson, L., Holling, C. S., Panarchy, Island: Washington DC,
pp. 103-120.
[12] Turner, B.L., Kasperson, R. E., Matson P. A., McCarthy J. J., Corell R.
W., Christensen L., Eckley N., Kasperson, J. X., Luers, A., Martello, M.
L., Polsky, C., Pulsipher, A., Schiller, A., A framework for vulnerability
analysis in sustainability science, Proceedings of the National Academy
of Sciences of the United States of America, 100(14), pp. 8074-8079, 2003.
[13] Renn, O., Risk Governance. Coping with uncertainty in a complex world.
London, Earthscan, pp. 5-11.
[14] Heitzmann, K., Sudharshan, C., Siegel, P. B., Guidelines for assessing the
sources of risk and vulnerability, Social protection discussion paper series,
0218, pp. 1-60, 2002.
[15] Schelfault, K., Pannermans, B., van der Craats, I., Krywkow J., Mysiak, J.,
Cools J., Bringing flood resilience into practice: the FREEMAN project,
Environmental Science & policy, 14, pp. 825-833, 2005.
[16] Adger, W. N., Successful adaptation to climate change across scales,
Global Environmental Change, 15, pp. 77-86, 2005.
[17] Smit, B., Wandel, J., Adaptation, adaptive capacity, and vulnerability.
Global Environmental Change, 16, pp. 282-292, 2006.
[18] Nagoya-city Mayor office, http://www.city.nagoya.jp/
[19] Tominaga, A., Lessons learned from Tokai heavy rainfall, Journal of
Disaster Research, 2(1), pp. 50-51, 2007.
[20] Zhai, G., Public preference and willingness to pay for flood risk reduction
(chapter 4), Toward resilient society to emerging disaster risks in mega-
cities, eds Ikeda S., Fukuzono T., Sato T., TERRAPUB and NIED,
pp. 57-87, 2006.
[21] Aichi prefecture, Survey concerning evacuation in case of flood, pp. 1-63,
2011 (Japanese).
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
This page intentionally left blank
Section 3
Flood management
This page intentionally left blank
Flood Recovery, Innovation and Reponse IV 75
Abstract
Thailand mega floods in 2011 highlighted the need for an integrated approach to
a flood risk management approach, combining local level community-based
action and a national strategic policy in preparation and reduction of
vulnerability of a country as a whole. This paper provides fact about a flood
crisis in 2011 and a set of lessons learned of community-based flood risk
management from affected communities scattered around the great flood areas in
central Thailand. Data and insightful information were drawn from a field visit
and a three-day participatory workshop attended by over 50 participants who had
experience of the flood. Included in this were community people, representatives
of local administration organizations and centralized agencies responsible for
dealing with natural disaster and crises.
Lessons learned from the workshop are conceptualized into six knowledge
platforms (KPs), highlighting the community best practices in response to the
situation during and after the crisis. They include 1) structural measures;
2) nonstructural measures; 3) emergency responses; 4) how to cope with the
community financial risk; 5) risk information and decision making; 6) dealing
with floods crisis recovery planning.
Keywords: Thailand, community-based, flood response, flood risk management.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140071
76 Flood Recovery, Innovation and Response IV
1 Introduction
The concept of flood risk management (FRM) has been widely embraced over
the past decade. In many instances this conceptual acceptance has resulted in a
change in decision making and practices highlighting risk management as
potentially more complex, but more efficient and effective than a traditional
engineering standard-based approach (Sayers et al. [3]).
The 2011 flooding crisis in Thailand undermined public confidence in the
capacity of government to manage water resources, to guide responsible
development, and to tackle recurrent and unforeseen emergencies. Many agreed
that what was missing from the country’s emergency flood response.
In retrospect, the problem was that the government pursued these measures
with limited engagement of civil society, civil volunteers, the private sector, and
the non-profit sector. As people suffered tremendously, the experience has
propelled communities to take collective action in a manner consistent with
traditional values of self-reliance. In these circumstances, citizens and civil
society organizations began to help themselves by recruiting volunteers to gather
new information on households in affected communities, and confirming the
information and knowledge that they needed to secure proper assistance. Local
communities and authorities, therefore, played a lead role in FRM, both in the
short and long terms. It is also importance for local communities to review the
process of FRM through identifying lessons from their knowledge and
experiences that could be learned from past experiences and make improvements
for future practices.
Data and information were drawn from field visit and a three-day
participatory workshop. Over 50 participants who had had experience of the
flood from both urban and rural community attended. As well, representatives of
local administration organizations and centralize agencies responsible for dealing
with natural disaster and crises. The workshop was manage based on the
dialogue theory that can better motivate people to share experiences and
knowledge (Bohm [2]). Every session of the workshop was taped recorded and
subsequently transcribed into a text form of over 300 pages. Ethnographic
interpretation was undertaken in order to draw upon some measures and future
mechanisms that community could be integrated into future flood risk
management.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 77
billion was estimated, due to the flooding impact over a long period of time.
Major drivers causing the flood include the following.
2.1 The highest record of rainfall and tropical storms and the flow capacity
of rivers
Since 1901, the rainfalls in 2011 were the historical record in Thailand (World
Bank 2012). The heavy rainfalls in the latter period were the consequences of 5
tropical storms, (between the end of June to the beginning of October 2011 – see
also Figure 1). Together, a study of World Bank (2012) states that one of the
main causes of the flood crisis in 2011 was the low-flow capacity of the river
(Lower North and Central Plains of Chao Phraya River and tributaries – see also
Figure 2). Hence, water runoffs from major rivers had caused the overtopping of
river dykes and breaching in any river tributaries.
Rapid and unplanned urbanization and unsuitable land use in the flood plain
areas is probably one of the most important factors worsening the floods in 2011.
Ayutthaya province, where industrial and housing estates were located in the
areas, were supposed to be the flood plains many infrastructural facilities had
caused the blockage of the flood way. In Nonthaburi province, especially in
Bangyai district, as semi urban-rural area located at western side of the lower
Chao Praya river basin, flood plains and canals were also blocked by both the
public and private infrastructure and urban sprawls. Many public canals simply
disappeared because of illegal encroachments. Such changes in land use took
away the ability to drain water from the northern part of Bangkok into the canals
and drainage systems, and then to the drainage stations by the sea coast of the
city.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
78 Flood Recovery, Innovation and Response IV
2.3.1 The weakness of the flood master plan, action plan, and policy
responses from the central government
Despite the severity of flood and the government quick responses, evidence
showed there has been no concrete studies on the impact of the 2011 flood, by
drafting a flood management master plan and allocating about 330,000 million
baht (USD 11.3 billion) for the flood protection action plan as well as assistance
and compensation for the flood victims. The action plan budget consisted of
immediate flood compensation budget and budget for the flood action plans.
Although the master plan consisted of both the plan on infrastructural
investment, rehabilitation and maintenance, and the non-infrastructural plan, it
does not give much attention to the latter, particularly in term of local
agencies/communities involvement. No concrete policy and measures have been
proposed, specially, inadequate attention to the complex long-term issues of
fragmented water management and required institutional changes of integrated
water management to cope extreme weather conditions, the appropriate
combination of single command authority and decentralization.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 79
During the flood crisis, all participants pointed to two common information
problems: i) the lack of real-time information on conditions and on coordination
among parties (that is, on who is doing what); and ii) the loss of critical public
records vital to reconstruction.
With regards to the first point, during the floods crisis the national
government collected information from municipal governments, while additional
information was crowd-sourced and channeled through social media and the
Internet. Many post-disaster situations were made worse by the lack of
communications strategies that make use of appropriate media to deliver critical
messages. Good information enabled individuals and communities to not only
stay safe but also contribute more effectively to relief and recovery. It also
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
80 Flood Recovery, Innovation and Response IV
ensured that communities have a realistic set of expectations about relief and
reconstruction. If communication was to help people stay safe and minimize the
disruption to their lives, they must be able to trust the information and its
sources. Together, communication regarding evacuation, temporary shelters, and
emergency food distribution was mismanaged, creating confusion throughout the
crisis phase of the flood.
3.2 New crowd sourced information and the use of social media
Social media, “community” radio stations were extensively used for searches and
rescues. Social media included web-based applications that use the Internet to
connect people (prominent examples are Twitter and Facebook), web sites and
computer applications that enable users to collaborate and create contents, such
as YouTube. Emergency “community” local radio stations also played a crucial
role in the aftermath of the crisis. When the emergency communication systems
in many cities broke down due to power failures and lack of emergency backup
power, community radio stations were able to get useful information out to
residents.
With the relatively high levels of mobile phone penetration in Thailand, social
media could be very useful during disasters, to the extent that they are already
used in normal times. They could also serve to link up with communities outside
the flood-stricken areas to facilitate the acquisition and allocation of aid and
assistance. In many developing countries, lack of physical accessibility to
disaster-affected sites is a key issue. Mobile networks and social media can be
used to collect and share localized information to improve accesses to rescue and
relief efforts. Reliability and trustworthiness of information is an extremely
important factor in the use of social media.
At the beginning, the government predicted a low probability of the floods risk
occurring, and underestimated its size and the incoming monsoon risks. The
official risk depicts areas that were small than the area actually affected by the
floods.
In addition, accurate risk assessment and interactive communication systems
which could connect local communities, government agencies, and experts,
made people less vulnerable and more resilient. We have learned that under the
enormous crisis and mismanagement of communication, community members
should not be encouraged to stick to a single scenario. Community networking –
“flood information on land,” (e.g., rise of water levels and flooding areas) would
allow what was happening and what kind of preparation needed to the upstream
and downstream flooding indication. At the same time, people in these
communities all needed as frequent weather forecasts—“information from the
sky,” (weather monitor and forecast) as much as possible.
A better understanding of nature and limitations of risk planning among
communities, local authorities and the population at large would have to improve
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 81
At evacuation centers, the needs of women, kids, cross cultural people and the
disabled were not fully met. The overwhelming majority of the leaders of
community organizations were male. Relief goods delivered to the shelters were
biased in favor of male evacuees. New measures are needed to assure privacy
and security for all vulnerable groups and should be planned in advance. These
measures call for empowering marginalized groups for long-term recovery and
including a gender perspective in planning and managing shelters, which will
require women to be more deeply involved in shelter management. Vulnerable
groups must not only be protected but also engaged in decision making.
Understanding and meeting the challenges of the elderly, children, and women,
both during the emergency and in its aftermath, are priorities for effective post
disaster response. Local cultural knowledge sound solutions that take account of
special needs among segments of the population should be planned in advance in
order to enhance resilience and facilitate recovery and reconstruction.
During the crisis, it was reported that shelters provided for did not give
sufficient privacy for anyone, particularly for women, many of whom did not
have private spaces where they could change their clothes or breast-feed their
babies. At the peak of the relief effort, more than 2,000 people were housed in
one evacuation centre, while some left their communities and stayed with
relatives and friends who lives outside the flooded area. Most facilities, such as,
schools and community centers, were publicly owned and were urgently set up
as evacuation centers.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
82 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 83
Generally, check dams and dikes are both necessary and effective in preventing
ordinary floods, which are relatively frequent, but they are of limited use against
the extreme events that occur less frequently. As the case of Bang Chom Sri dam
showed, construction standards and stability performance under worst-case
scenarios should be further investigated. Structures should be able to withstand
floods that exceed their designed flow, reducing the force of the water before
they collapse and thereby mitigating damages.
The Central Government master plan generally put an emphasis on the
structural flood management and little attention was given to the issues of non-
structural aspects of flood management. Efforts within flood risk management
have to create solutions based on community ownership and consensus. By
preparing and increasing community awareness and capacity of local
governmental authorities to handle flood situations has been recognized as a
focal point for flood risk management. After mega flood, government launch a
mega project for flood protection, however it only focusing on construct dam and
flood way.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
84 Flood Recovery, Innovation and Response IV
During the 2011 flood crisis, full financial impact (including direct and indirect
impacts) form flood disaster will not be known for some time. The government
must play an important role in alleviating the disaster’s impact on households
and businesses through measures that ensure the stability of the financial system,
timely approvals of supplementary budgets, and provisions for rapid
disbursement disaster assistances, all of which helped citizens jumpstart their
recovery processes. The financial resources for recovery and reconstruction are
being funded by taxes to avoid leaving the cost to future generations. Flood
insurance helps people get back on their feet. Governments can play an
important role in fostering the growth of this kind of infrastructure, thereby
enabling the private insurance industry to offer cost-effective and affordable
insurance solutions.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 85
into account every conceivable eventuality and utilizing all the tools science has
to offer.
The sharing of information among governments, communities, and experts
left much to be desired. While science-based early-warning systems are
important during a disaster, it was best for the target population if it could
combine with information on the ground through regular sharing of pre-disaster
information at the local level. The sharing should be accompanied – over time
and with the community’s involvement – by disaster drills, community mapping
exercises, and other measures. In recent years, remote-sensing data have been
used around the world to rapidly map the damage resulting from natural
disasters.
4.6 Recovery planning
In the areas affected by the floods crisis, communities were organized on
recovery planning committees from the earliest stages. In general, authorities
should be prepared for disasters by designating temporary storage sites, traffic
routes for transporting waste; including hazardous and toxicity waste. The role of
the private sector in debris management, as well as cooperation with
organizations and government bodies outside the affected areas, should be
explored. The possibility of recycling should be considered. Finally, rice and
vegetable seeds as well as young fruit trees are of most needed by rural
communities, as they could save a great deal of money on food and begin a new
agricultural season without spending money on much of them.
5 Conclusion
Lesson learned from Thailand has pointed that, flood risk management
implemented by communities and local government is crucial. Communities and
local government have the opportunity to design solutions that are adaptable to
the needs of their local communities and are consistent with local policies and
priorities. The measures and possible future mechanisms in addressing
community’s flood risk-management. From the people’s view, the integrated
flood management mechanism does not have to rely entirely on the: “predict-
and-act’ approach, which is conventionally used for the designed structural
measures. The Thai experience showed that success of flood risk management
lay in community involvement. Effective flood risk management requires close
coordination among all affected areas, including all responsible municipalities as
well as their agencies and departments, in order to support all-inclusive and
country relevant solutions. Policy makers and urban/rural development experts –
both structural and non-structural aspects – should be well advised to listen to the
communities and empower them to be the focal part of the solution.
Acknowledgements
The authors wish to thank, Asian Development Bank, for providing the budget
for the workshop, community leader district and representatives from local
government agencies in Thawung and BangYai Sub-district, as well as
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
86 Flood Recovery, Innovation and Response IV
References
[1] Asian Development Bank, Community-based Flood Risk Management and
Disaster Responses the report, 2013.
[2] Bohm, David, On Dialogue, London Routledge, 1996.
[3] Paul Sayers et al., Flood Risk Management A Strategic Approach, Paris
UNESCO, 2013.
[4] World Bank, Thai Flood 2011. Rapid Assessment for Resilient Recovery
and Reconstruction Planning, Bangkok, 2012.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 87
Abstract
Although available water resources are limited, water demand is continuously
increasing due to population increases, economic development, and additional
uses, such as recreational and environmental uses. Constructing new reservoirs
has traditionally been the approach to develop new water resources. However,
such construction can be hampered by negative perceptions, adverse
environmental effects, and opposition from NGOs to dam construction. Although
Andong and Imha reservoirs are located close to each other, and they have
similar hydrological and meteorological characteristics, the storage capacity of
Imha reservoir is only about half that of Andong reservoir. This makes the
operation of both reservoirs inefficient. This paper evaluates the effect of a
diversion tunnel connecting Andong and Imha in the flood season. Water yield
and spillway release reduction capability with 95% reliability were analyzed
using historical daily inflows data for 30 years. By changing the reservoir
operation methods, the reservoir system performance was evaluated. The system
operation of the reservoirs with the diversion tunnel showed better results than
the individual operation.
Keywords: flood control, reservoir system operation, diversion tunnel.
1 Introduction
Demand for water resources has been rapidly increasing because of population
increases and economic development. Recently, demand for water necessary for
recreation and environmental improvements has been also continuously
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140081
88 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 89
Andong dam and Imha dam belong to the Nakdong river basin, which occupies
approximately 25% of the territory of Korea. Andong dam is located at the
uppermost stream of the Nakdong River, and Imha dam is located at
the Banbyeon stream, which is the first branch of the Nakdong River. The
Andong dam basin corresponds to approximately 6.8% (1,584 km2) of the entire
Nakdong river basin (23,384 km2), and the total length of flow paths in it is 31
km. The Imha dam basin occupies approximately 5.8% (1,361 km2) of the entire
Nakdong river basin area, and the total length of flow paths in it is 75 km.
Figure 1: Diagram of the Andong dam and Imha dam basins and the diversion
tunnel.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
90 Flood Recovery, Innovation and Response IV
The reservoirs’ planned monthly water supply volumes were divided into
irrigation water, domestic water, industrial water, and instream flows. Andong
reservoir’s annual water supply is 926 10 m and that of Imha reservoir
is 591.6 10 m .
The annual domestic and industrial water supplies of Andong reservoir and
Imha reservoir are 450 10 m and 363.6 10 m , respectively. The annual
irrigation water supplied by Andong reservoir is 300 10 m , and that supplied
by Imha reservoir is 13 10 m . With regard to irrigation water, irrigation water
consumption in the basin during the busy farming season from April to October
is reflected in the planned monthly water supply volumes. Variations in the
planned monthly water supply volumes are larger in Andong reservoir compared
to Imha reservoir. The annual instream flow of Andong reservoir is 176
10 m and that of Imha reservoir is 215 10 m when the instream flow of
Imha reservoir includes the volume supplied to Yeongcheon raceway (4.8 m3/s).
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 91
In this study, normal monthly water supply volumes were selected to analyze
the effects of the connected operation (via the diversion tunnel) of the reservoirs
on water supply volumes and additional discharges.
Storage capacities were calculated using the reservoir continuity equation based
on the specifications of the Andong reservoir and Imha reservoir to determine
discharges, additional supply volumes, power generation discharges, and
spillway releases. The storage capacity of day t was determined by the storage
capacity and inflow of day t-1. The discharge volume and the volume of the
diversion tunnel on day t and the reservoir discharge volume were obtained by
applying an additional supply rate for day t to the basic planned (normal) supply
volume. To determine the discharge volume that leads to reduced supply or
spillway release, the calculated storage capacity of day t was compared to the
storage capacities that correspond to the full water level and the low water level.
The power generation discharges of the reservoirs were discharged first, and any
discharge volumes that occurred in excess of the maximum power generation
capacity of the two reservoirs were counted as spillway release volumes
(equation (1)).
, (1)
where is the storage volume on day t, is the storage volume on day t-1, I
is the inflow volume on day t, is the discharge volume on day t, and is the
diversion tunnel diversion volume on day t.
The volume of the diversion tunnel was calculated considering the
entrance/exit head losses and the friction head loss. This was considered the
reservoir’s inflow volume. In this case, 0.2 was used as the entrance loss
coefficient, and 1.0 was used as the exit loss coefficient. The friction head loss
was calculated using the Darcy–Weisbach formula, and a roughness coefficient
of 0.015 was assumed. The sum of the diversion volumes of the two reservoirs
through the diversion tunnel was 0, and the diversion volume of each reservoir
increased or decreased according to water movements between the two reservoirs
(from Andong to Imha or from Imha to Andong). The formulas for calculating
the volumes of the diversion tunnel resulting from head loss differences are as
follows:
∆H W (2)
∆
V (3)
∆
D AV (4)
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
92 Flood Recovery, Innovation and Response IV
.
f – , (5)
where ∆H is the water level difference between the two reservoirs (m),
is the entrance loss coefficient,
W is the water level (m) of Andong reservoir,
is the exit loss coefficient,
W is the water level (m) of Imha reservoir,
is the friction loss coefficient,
is the entrance head loss (m),
is the length (m) of the diversion tunnel,
is the exit head loss (m),
is the diameter (m) of the diversion tunnel,
is the friction head loss (m),
n is the roughness coefficient,
D is the volume (m /s) of the diversion tunnel diversion,
is the flow velocity (m/s) of the diversion tunnel
A is the cross-sectional area (m ) of the diversion tunnel.
No diversion through the tunnel occurs when the water levels of the two
reservoirs are lower than the height of the diversion tunnel and water can be
moved between the two reservoirs only when the water levels of the two
reservoirs are higher than the diversion tunnel. Therefore, if the water level of
only one reservoir is higher than the diversion tunnel, diversion will occur until
the higher water level of the reservoir goes down to the height of the diversion
tunnel. If both the water levels of the two reservoirs are higher than the diversion
tunnel, diversion will occur from the reservoir with the higher water level to the
reservoir with the lower water level until the water levels of the two reservoirs
become the same.
As both reservoirs are installed with hydroelectric generation facilities and
generate hydroelectric power through the power discharge. Generation is
calculated through the discharge. The power generation, P, was calculated using
equation (6), and a generator efficiency value of 0.86 and a hydraulic turbine
efficiency value of 0.95 were applied to both reservoirs.
P 9.81 , (6)
where P is the power generation (GWh),
γ is the generator efficiency,
γ is the hydraulic turbine efficiency,
Q is the reservoir discharge volume (m3) for time T,
H is the Imha reservoir water level (m),
T is the friction loss coefficient (h).
The head loss difference, H (equation (7)), value was obtained by deducting
the tail water level from the head water level. The low water level, which is the
head water level, was determined by the average of the water levels of the two
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 93
reservoirs at time t and the next time t+1. The tail water level is the water level at
which the water was discharged. This is usually determined by the water level of
the regulating reservoir. However, in this paper, the value obtained by deducting
the average value of tail water levels from the average value of forebay water
levels was applied because the water level of the regulating reservoir could not
be considered and the simulation was conducted focusing on main reservoirs.
′ ′
H , (7)
where W is the low water level at time t, W is the low water level at time
t 1, HWL′ is the full water level of the regulating reservoir, and LWL′ is the
low water level of the regulating reservoir
The number of days of water shortage was calculated based on the low water
level. Days on which the water level of the reservoir was the same as the low
water level were counted as days of water shortage. That is, cases where the
water level of the reservoir dropped to the low water level and could not satisfy
the basic supply volume were defined as cases of water shortage, and the
numbers of days of water shortage determined in this way were counted.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
94 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 95
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
96 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 97
4 Conclusion
In this study, the effects of the connected reservoir operation of Andong
reservoir and Imha reservoir using a diversion tunnel were analyzed to prevent
floods and to ensure an uninterrupted water supply during drought periods by
efficiently managing the water resources that are spillway-released during flood
seasons. Structural methods using a connected reservoir operation and a
nonstructural a system operation method were applied. Using daily discharge
data for 30 years from 1979 to 2008, daily simulations were conducted using the
reservoir continuity equation, and water yields and effects of reducing spillway
release volumes were reviewed based on 95% reliability.
According to the results of the analyses based on the concept of deficit supply,
the number of times of spillway releases and spillway release volumes decreased
the most during the separate operation of the reservoirs, followed by the
reservoir system operation, the connected operation of the reservoirs, and the
connected reservoir system operation in order of precedence. In addition, the
possible annual additional supply volumes increased the most during the separate
operation of the reservoirs, followed by the connected operation of the reservoirs,
reservoir system operation, and the connected reservoirs system operation in
order of precedence.
Although system operation without the diversion tunnel enabled reducing
spillway release volumes and securing additional supply volumes during drought
periods, larger effects were obtained when the diversion tunnel and system
operation were used simultaneously. Given these results, using the diversion
tunnel and system operation together is considered to result in a structurally
stable connected reservoir operation.
In this study, the effects of using the diversion tunnel and the results of
system operation were examined in terms of their flood season spillway release
reducing effects and water supply during water utilization periods. With regard
to connected reservoir operations, a comprehensive analysis of hydroelectric
power generation, water quality improving effects, and economic and
sociological benefits is needed to address.
Acknowledgement
This work was supported by a National Research Foundation of Korea (NRF)
grant funded by the Korean government (MEST) (No 2013-065006).
References
[1] Kojiri, K., Ikebuchi, S. & Yamada, H., Basinwide flood control system by
combining prediction and reservoir operation. Stochastic Hydrology and
Hydraulics, 3, pp. 31–49, 1989.
[2] Cheng, C. & Chau, K.W., Flood control management system for reservoirs.
Environmental Modeling & Software, 19, pp. 1141–1150, 2004.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
98 Flood Recovery, Innovation and Response IV
[3] Xiang, L., Shenglian, G., Pan, L. & Guiya, C., Dynamic control of flood
limited water level for reservoir operation by considering inflow uncertainty.
Journal of Hydrology, 391, pp. 124–132, 2010.
[4] Kumar, N.D., Baliarsingh, F. & Raju, S.K., Optimal Reservoir Operation for
Flood Control Using Folded Dynamic Programming. Water Resour Manage,
24, pp. 1045–1064, 2010.
[5] Vonk, E., Xu, Y.P., Booji, M. J. & Augustijn, D.C., Adapting Multireservoir
Operation to Shifting Patterns of Water Supply and Demand. Water Resour
Manage, 28, pp. 625–643, 2014.
[6] Hirsh, R. M., Cohon, J. L. & ReVelle, C. S., Gains from joint operation of
multiple reservoir systems. Water Resources Research, 13, pp. 239–245,
1997.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Section 4
Considering ‘Blue-Green’
approaches to
Flood Risk Management
(Special session
organised by J. Lamond)
This page intentionally left blank
Flood Recovery, Innovation and Reponse IV 101
Abstract
This study develops a conceptual framework to inform thinking around the social
research approach adopted to consider the development of ‘Blue-Green’
approaches to Flood Risk Management (BG-FRM) in UK cities. The framework
informs the manner in which research is conducted and data analysed, to
understand current and possible future household and business behaviours as BG-
FRM becomes more established, and so possibly (or not) more ‘normalised’, as
well as the influences upon these behaviours that can potentially be played by key
stakeholders. A conceptual map is drawn up that outlines the key players, their
domains of agency and lines of influence concerning larger-scale (neighbourhood,
city-level) and smaller-scale (household, business) approaches. A conceptual
framework is then developed, thinking about the motivations and barriers that
could encourage or inhibit adoption of blue-green approaches and the behaviour
changes necessary for their sustainability, before surveying research already
conducted in this area. Social Practice Theory (SPT) is suggested as a new manner
of framing research to understand the ways in which behaviour may change, or
fail to change, and the opportunities and barriers to any such changes. SPT, it is
argued, could provide a means by which to consider present behaviours and
attitudes, so that we might more effectively look for opportunities to encourage
progressive behavioural developments that could increase the chances of BG-
FRM’s sustainability.
Keywords: Blue-Green, Flood-Risk Management, sustainable, behaviour, Social
Practice Theory (SPT).
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140091
102 Flood Recovery, Innovation and Response IV
1 Introduction
With the increased incidence of flood events in recent years and the hypothesis
that flooding may increase (or currently be increasing) as a result of development
pressures and climate change (King [1], Whitmarsh [2]), governments are taking
very seriously the need to deal with the economic and social threats from this
(DEFRA [3, 4], Environment Agency [5, 6], SEPA [7]). Thinking is shifting
away from simple notions of resisting outright inundation towards developing
resilience to flooding – living with water and making space for water gaining
prominence in academic literature and policy ([3], Pitt [8], Bowker [9], McBain
et al. [10]). Thinking has also moved away from erecting structural defences to
establishing softer and more sustainable FRM that retains, filters and makes use
of water-flows. The latter has begun happening for a number of reasons: the
environmental, aesthetic and socio-economic impacts of structural work; the
need to adapt urban areas to cope with a changing climate (using fewer resources
and emitting less waste), and an argued need to rethink our relationship with
water, reintegrating the natural water-cycle with the urban environment,
producing ‘water sensitive cities’ (Brown et al. [11]. Howe and Mitchell [12],
Kazmierczak and Carter [13], BGD [14]). BG-FRM approaches involve
improving green infrastructure, raising water-absorption capacity and promoting
natural channelling rather than containing and culverting (Abott et al. [15]).
A number of authors have begun publishing research around public attitudes
around Blue-Green approaches to Flood Risk Management (BG-FRM) (Bastien
et al. [16], Wright et al. [17], Kenyon [18], Apostolaki and Jefferies [19],
Werritty [20], Johnson and Priest [21]), some drawing conclusions as to how
people will behave around them. Results are apparently quite split between those
who found strong preferences for structural defences [20, 21] and others who
found preference for more sustainable solutions [16, 17]. In Apostolaki’s study,
awareness of SuDS’ flood functions was argued to be low, whilst others have
found it to be quite high (around 75% of respondents [19]).
What has not yet been done is to think about how behaviour might change
over time. Public attitudes may be cynicism and mistrust if people are not
involved in discussions from the outset. All parties will have a lot to contribute,
from scientific-technical assertions about BG-FRM in/efficacy to local
knowledge which could dismiss certain options or illustrate that others were
relevant and likely to work. Dialogic learning will be imperative to thinking
about viable BG-FRM options; for instance, significant differences have been
found to exist between actual and perceived SuDS’ safety levels (cf. McKissock
et al. [22]). Bastien et al. [16] and Apostolaki and Jefferies [19] found small
amounts of litter considered significant ‘pollution’, highlighting the need for
agreed maintenance systems. There could initially be hesitancy in uptake, with
safety fears around poorly lit green spaces (Bixler and Floyd [23]), water-butts
being seen as something for keen ‘productive’ gardeners (Chappells et al. [24]),
and green roofs and permeable paving possibly acceptable or workable only if no
behaviour-change were required (cf. Whitmarsh [25] on the asymmetry of
intentions and actions concerning climate change). There may be reluctance to
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 103
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
104 Flood Recovery, Innovation and Response IV
not yet considered how behaviour might alter with its establishment and
normalisation.
2 Stakeholder map
Understanding preferences and what affects these (and so behaviour) is a
necessary first step in negotiating, and overcoming, barriers and concerns. In
order to gain this understanding we need to draw up a stakeholder map before
outlining a conceptual framework of perceptions around BG-FRM, and thinking
through the motivators and barriers to behavioural changes that could facilitate
SuDS’ functioning. This work could then be used to monitor shifts in behaviour
and perceptions as BG-FRM grew, developed and became more normalised.
The system would need to be bounded to allow for proper analysis and
appraisal. An appropriate framing would be city boundaries, although some
agents will have wider operations and the watershed may stretch beyond, so
factors from outside would need consideration. Nonetheless, framing the
‘system’ at city level makes sense administratively and will be appropriate for
the majority of stakeholders. These would include communities (households,
governmental and non-governmental organisations, public service providers and
businesses and their representatives), the front end of dealing with inundation.
Environmental and wildlife groups will have a strong interest in BG-FRM, as
will landowning and advisory bodies such as farms and Natural England [27].
Water companies would be affected by BG-FRM insofar as water supply and
disposal of wastewaters would be altered by interventions. The Planning,
Development and Building industries would be affected, with new opportunities,
responsibilities and demands placed upon their work (RTPI [28], Scottish
Government [29]).
Key actors with responsibility for larger-scale BG-FRM would be Local
Authority, Environment Agency and national government bodies such as
DEFRA (see Figure 1). However the aggregate of household- and business-level
BG-FRM
could also make a significant contribution to reducing flooding. The Commission
for Architecture and the Built Environment (CABE [30]) found that whilst
increasing green space and tree cover in urban areas by 10% would reduce
surface water run-off by around 5%, adding green roofs to all buildings could
reduce it by 20%. So this, combined with water-butts and replacement of hard-
standing, could impact significantly on potential flooding.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 105
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
106 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 107
their neighbours will free-ride on the benefits. Activities will need to be scaled
up to community level, which in turn necessitates dialogue working towards
consensus on action. Local authority or government regulation may be needed to
ensure all parties act appropriately, but given that all would stand to benefit from
BG-FRM over time then this may be avoidable through dialogue and co-
construction of viable preferred solutions.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
108 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 109
5 Conclusion
This paper has outlined a conceptual framework to guide how social research
into practices (and changes in practices) regarding city-level adaptations to
increase BG-FRM could be undertaken. Principal stakeholders have been
provisionally identified and lines of influence between these for BG-FRM
approaches suggested. Introducing BG-FRM to a city will be a complex matter
affecting many different stakeholders, and this is but one reason why the process
will need to be as inclusive as possible from the very beginning, to ensure that all
relevant and concerned voices are being listened to.
The stabilising and destabilising factors and motivations affecting people’s
behaviour have also been outlined and considered. While some research has
looked at a time-slice of attitudes, none has as yet tried to think around why and
how behaviour patterns might change as infrastructure developed and became
more normalised. This is an essential next step in thinking about the viability of
BG-FRM over time, understanding more about people’s current thinking and
how this could develop, to develop a clearer picture of likely outcomes. The
Social Practice lens has identified new avenues for interrogating this behaviour.
Studying this further using an SPT approach will require close work with
communities and other stakeholders at the start and throughout the proposed
changes, to understand how multiple considerations might settle or shift
regarding FRM options under a variety of hypothetical situations over time.
Acknowledgements
This research was performed as part of an interdisciplinary project programme
undertaken by the Blue-Green Cities Research Consortium
(www.bluegreencities.ac.uk). The Consortium is funded by the UK Engineering
and Physical Sciences Research Council under grant EP/K013661/1, with
additional contributions from the Environment Agency, Rivers Agency
(Northern Ireland) and the National Science Foundation.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
110 Flood Recovery, Innovation and Response IV
References
[1] King, D.A., Climate change science: adapt, mitigate, or ignore? Science,
303(5655), pp. 176–177, 2004.
[2] Whitmarsh, L., Are flood victims more concerned about climate change
than other people? The role of direct experience in risk perception and
behavioural response. Journal of Risk Research, 11(3), pp. 351–374, 2008.
[3] DEFRA, Making Space for Water. Defra: London, 2005.
[4] DEFRA, Consultation on policy options for promoting property-level
flood protection and resilience, Defra: London, 2008a.
[5] Environment Agency, Flooding in England: A National Assessment of
Flood Risk, Environment Agency: Bristol, 2009.
[6] Environment Agency, Flooding in Wales: A National Assessment of Flood
Risk, Environment Agency: Bristol, 2009.
[7] Scottish Environment Protection Agency, Flooding in Scotland: A
Consultation on Potentially Vulnerable Areas and Local Plan Districts.
SEPA: Stirling, 2009.
[8] Pitt, M., Learning Lessons from the 2007 Floods: An Independent Review,
Cabinet Office: London, 2008.
[9] Bowker, P., Flood Resistance and Resilience Solutions: An R&D Scoping
Study, Defra: London, 2007.
[10] McBain, W., Wilkes, D. & Retter, M., Flood Resilience and Resistance for
Critical Infrastructure, CIRIA: London, 2010.
[11] Brown, R., Keath, N. & Wong, T., Transitioning to water sensitive cities:
historical, current and future transition states. 11th International
Conference on Urban Drainage, 10, 2008.
[12] Howe, C. & Mitchell, C., Water Sensitive Cities, IWA Publishing:
London, 2012.
[13] Kazmierczak, A. & Carter, J., Adaptation to climate change using green
and blue infrastructure. A database of case studies. University of
Manchester: Manchester, 2010.
[14] Blue-Green Dream website, http://bgd.org.uk, accessed on 10th September
2013, Imperial College London: London, 2013.
[15] Abbot, J., Davies, P., Simkins, P., Morgan, C., Levin, D. & Robinson, P.,
Creating Water Sensitive Places – scoping the potential for Water
Sensitive Urban Design in the UK, 2013.
[16] Bastien, N., Arthur, S. & McLoughlin, M.J., Valuing amenity: public
perceptions of sustainable drainage systems ponds. Water and
Environment Journal, 26(1), pp. 19–29, 2011.
[17] Wright, G.B. Arthur, S., Bowles, G., Bastien, N. & Unwin, D., Urban
creep in Scotland: stakeholder perceptions, quantification and cost
implications of permeable solutions. Water & Environment Journal, 25(4),
pp. 513–521, 2011.
[18] Kenyon, W., Evaluating flood risk management options in Scotland: A
participant-led multi-criteria approach. Ecological Economics, 64(1),
pp. 70–81, 2007.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 111
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
112 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 113
Abstract
A Blue-Green City aims to recreate a naturally-oriented water cycle while
contributing to the amenity of the city by bringing water management and green
infrastructure together. The Blue-Green approach is more than a stormwater
management strategy aimed at improving water quality and providing flood risk
benefits. It can also provide important ecosystem services and socio-cultural
benefits when the urban system is in a non-flood condition. However, quantitative
evaluation of benefits and the appraisal of the relative significance of each benefit
in a given location are not well understood. The Blue-Green Cities Research
Project aims to develop procedures for the robust evaluation of the multiple
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140101
114 Flood Recovery, Innovation and Response IV
1 Introduction
The combined impacts on social, economic and environmental systems make
flooding one of the World’s most serious hazards. Over 2.4 million properties in
England alone are at risk of fluvial or coastal flooding, with a further 2.8 million
properties susceptible to surface water flooding [1]. Increasing frequency and
magnitude of intense precipitation events in future decades are predicted to
increase flooding and damages incurred [2], particularly in cities where the
consequences of flooding are especially severe. Increasing urbanisation, economic
growth, and the concomitant increase in impermeable surfaces will further
exacerbate the urban flood risk. There is thus a demand for new and innovative
research that can help reduce the probability and/or consequences of urban
flooding while helping cities become more resilient and able to adapt to new flood
risks imposed by climate change [3] and economic development.
Non-traditional measures for flood risk management (FRM) aim to reduce the
amount of water entering man-made drainage systems and offer an alternative to
traditional grey infrastructure (e.g. piped drainage and waste water treatment
systems for pollution control). Natural measures are gaining increasing support as
efforts are made to better integrate the water cycle with urban design and
development needs, particularly in light of future climate change and the limited
adaptability of grey infrastructure to events that exceed the design standard. A
move towards urban water management that holistically considers the
environmental, social and economic consequences of different strategies is
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 115
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
116 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 117
flood states. BGI that perform to the design standard will fulfil the primary goal
of reducing the risk of surface water inundation during a flood event. In addition,
when in the flood state, BGI may reduce water pollution and improve water
quality, help control the water supply and prevent the cascade of negative socio-
economic impacts that generally occur in the aftermath of a flood, e.g. high repair
costs, displacement from homes, damage to health, decline in business and
reduced economic prosperity. Furthermore, construction and maintenance of BGI
is often cheaper than the grey alternative, as illustrated by Portland’s “Green
streets” project to reduce stormwater runoff and the risk of combined sewer
overflow. $250 million in hard infrastructure costs was saved through the design
and landscaping of soil and plants into the urban streetscape to aid infiltration and
reduce peak stormwater flow (at a cost of $8 million) [14].
Blue-Green Cities also offer numerous benefits when the system is in a non-
flood state. Environmental benefits include; reduction in the urban heat island
effect, improved air quality, noise reduction, carbon sequestration and a carbon
emission reduction potential through avoiding highly carbon intensive
alternatives, groundwater recharge, increased biodiversity, habitat enhancement
and related ecosystem services. Socio-cultural benefits include; traffic calming
and road safety, reduction in water demand and water recycling, improved health
and wellbeing, attractive landscape, improved quality of place, crime reduction
and education potential. BGI may also augment the ability of cities to mitigate and
adapt to climate change [14] and is frequently a key component of economic
regeneration projects to improve the liveability of urban environments [15].
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
118 Flood Recovery, Innovation and Response IV
research components (Fig 2) are denoted by Work Packages (WP), held together
by a strong communications package to promote interdisciplinarity and coherent,
integrated results, based on shared conceptual, methodological and theoretical
ideas [19]. A strong communications network, both internally and with external
stakeholders, is central to our goal to investigate the linkages between human
behaviours, physical processes and policy constraints regarding FRM. We aim to
progress from the multi-disciplinary approach where discrete disciplinary work
packages are completed and subsequently combined at the end of the project, with
little cross-discipline engagement during the research process. Rather, we aim for
data exchanges and common epistemological approaches to marry the
interdisciplinary appeal with the disciplinary mastery [20]. This will create
knowledge that is solution oriented and socially robust [21], and transferable to
both scientific and societal practice. Co-evolution of understanding and
knowledge, aided by tight integration within the team, will ensure that the sum of
the whole (in terms of deliverables) exceeds the sum of the parts.
The aim of the Consortium is to develop new urban FRM strategies as part of
wider, integrated planning intended to achieve urban renewal and environmental
enhancement in which multiple benefits of BGI are rigorously evaluated and
understood. Focussing on a common case study (Newcastle) in the third year of
the project (2015) is key to visualising the Consortium aim and converging on
common deliverables, with success relying on the co-production of knowledge and
multi-way exchange within the Research Consortium and wider stakeholders.
Communication is often ineffective and one-way between academia and end-
users, e.g. key stakeholders (including decision makers) and local communities
(those at risk of flooding and directly affected by decisions and hence should take
an active role in decision making regarding FRM [18, 22]). We aim to facilitate
discussion and include these groups from the outset.
Research will focus primarily on fluvial and pluvial flooding; the latter typically
caused by extreme local storms and insufficient capacity of subsurface drainage
networks. The Consortium is developing urban flood models that realistically
represent the urban environment (land use and terrain) in its complexity. Coupled
surface/sub-surface hydrodynamic models will produce inundation predictions
across a range of events of different frequencies and lengths, visualised in
probability maps for inundation across an urban area. Flood inundation modelling
is being developed to include the movement of water through Blue-Green features
such as blue and green roofs, retention ponds, permeable paving, green space and
bioswales, to enable a comparison of flow velocity, depth and inundation extent
before and after the adoption of BGI. BGI as a FRM strategy will be assessed by
a set of scenarios including ‘business and usual’ (no additional BGI) and a Blue-
Green future (BGI as preferred assets).
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 119
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
120 Flood Recovery, Innovation and Response IV
Successful simulation of the movement of water and sediment through the urban
environment will indicate design benefits of select infrastructure components and
generate recommendations to achieve multiple benefits. However, the physical
system cannot be assessed in isolation. Societal perceptions of the costs and
benefits of different FRM approaches play an important role in progressing
research into policy [22]. Interaction and involvement in the evolution of Blue-
Green design by the stakeholder community is essential to the concept of Blue-
Green Cities. Individual and institutional agents will be incorporated into the flood
inundation model to illustrate how behavioural changes impact on flooding and
vice versa. Such knowledge is crucial when making the case that agents need to
be part of the decision-making process for FRM. Fieldwork will be used to identify
and understand the behavioural responses of individuals and institutions to a range
of FRM strategies including Blue-Green. Evidence-based rules are being
developed using stated preference models to represent those behaviours and will
provide the data input to an agent-based model to investigate alternative scenarios
of future Blue-Green FRM strategies under different socio-economic conditions.
We are developing an understanding of how agents respond to stimulus and
change in the physical landscape, and how this may alter the probability of
flooding. We are also interested in how agents behave in a way to reduce the
consequences of flooding. Potential barriers to the implementation of FRM
strategies arise depending on where and to whom the benefits of BGI accrue during
times of no flood. This, and the potential for positive and negative interactions
with wider urban infrastructure, may act as an incentive/disincentive for the
adoption of innovative, non-traditional solutions.
Tools and methodologies are being developed to represent FRM and Blue-Green
networks in a single urban environment, as part of a wider complex ‘system of
systems’ that services urban communities. Series of interrelationships link energy,
transportation, water (supply and wastewater), emergency services, and
information and telecommunication sectors. Disrupting these dependencies can
have significant socio-cultural and economic consequences that may extend to
regional and national level, particularly during times of extreme flood. Research
will illustrate how changes in both the physical interfaces (flood pathways and
BGI) and institutional responsibilities (policy, planning and governance
structures) cascade across the wider urban system, and identify intervention points
to ensure rapid adoption, optimum functionality and reduced risk in other
infrastructure areas. The Three Points Approach (3PA) of Fratini et al. [23] will
be adopted and illustrates a more holistic process towards urban FRM that
simultaneously considers technical optimisation of urban drainage systems, spatial
planning to increase resiliency, and everyday performance under the green, non-
flood, condition as a foundation for social preparedness. Three system states have
been developed from the 3PA; non-flood (green condition), design standard, and
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 121
Methodologies are being developed to assess, quantify and value the multiple
benefits of adopting BGI in urban FRM strategies at both the local/regional and
global/international scales. Such methodologies will also robustly evaluate the
multiple functionalities of BGI components and address the inherent uncertainties
of cost/benefit analysis. By evaluating the relative significance of benefits in
context specific locations we aim to establish preference ratings linked to a multi
criteria analysis for component selection. This will provide sound science and
recommendations for design guidance to assist policy makers in the choice of
FRM strategy. Despite the 2007 SuDS (Sustainable urban Drainage Systems)
Manual (C697) [24] providing extensive guidance, the lack of recent UK
legislation is a key barrier to the limited uptake of BGI and SuDS.
We adopt a novel method of performance appraisal against a set of diverse
criteria that addresses environmental, socio-cultural and economic costs and
benefits that accrue beyond the realm of effective FRM. Surface water
management objectives, such as the minimization of runoff quantity, reduction of
peak stormwater flows, and improvement to runoff quality may be achieved by
grey or Blue-Green infrastructure. Both incur costs; capital materials, energy
inputs and maintenance, yet those for BGI are typically much lower [14]. Life
cycle assessment (LCA) and similar methods of economic costing are often used
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
122 Flood Recovery, Innovation and Response IV
for comparison and selection of asset design [25]. Whatmore et al. [22] contend
that choice of FRM solution based solely on economic viability (benefits > costs)
restricts the range of FRM solutions to be explored. The full net-benefit of BGI
development can only be realized by a comprehensive accounting of their multiple
benefits [14]. Quantitative evaluation of benefits and the appraisal of the relative
significance of each benefit in a given location are not well understood. BGI is
acknowledged as providing additional benefits that grey infrastructure cannot,
such as counteracting urban heat island effects, reducing energy costs, creating
community amenities and improving habitats [14], and multi-functional landuse
is paramount to optimise BGI benefit accrual.
4 Summary
The Blue-Green Cities Research Project adopts an interdisciplinary approach to
identify and rigorously evaluate the multiple benefits of natural flood risk
management strategies using Blue-Green infrastructure. This paradigm shift from
traditional grey infrastructure designed to remove water as quickly as possible
from the urban surface is in line with WSUD and urban water management that
holistically considers the environmental, social and economic consequences FRM
strategies. A Blue-Green City offers effective performance of the drainage
network to achieve high levels of flood protection and resilience to some future
climate change, while supporting multiple non-flood benefits, often maximised by
the integration of blue and green assays and creation of networks. Throughout
2014–15 the Blue-Green Cities Research Consortium will model how changes in
policy and associated agent behaviour/attitudes can impact on flooding and vice
versa. This linking of physical processes to human behavioural patterns for
different scenarios is highly innovative and will provide for an analysis of the
urban ‘system of systems’ and highlight where, when and to whom the multiple
benefits will accrue under different future scenarios. This will allow us to
rigorously, and where possible, quantitatively, evaluate the costs and benefits of
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 123
different strategies and appraise the relative significance of each benefit in a given
location. The attitudes and perceptions of people and society towards Blue-Green
and grey infrastructure is critical in demonstrating to policy makers how non-
traditional infrastructure may be utilised to achieve maximum benefit while
ensuing ‘agents’ become part of the decision-making process. The applicability of
the research methods will be tested in the demonstration case study (Newcastle,
UK) and will endeavour to incorporate the understanding and interest of key
stakeholders in urban FRM and connect this with the potential impact of adopting
the Blue-Green vision in a practical, real-life setting.
References
[1] Bennett, O., Flood defence spending in England, Standard Note: 14th
March 2013, House of Commons Library, London, 2013.
[2] Bates, B., Kundzewicz, Z.W., Wu, S., & Palutihof, J., (eds). Climate
Change and Water, Technical Paper, Intergovernmental Panel on Climate
Change (IPCC), 2008.
[3] Wilby, R.L. & Keenan, R., Adapting to flood risk under climate change.
Progress in Physical Geography, 36(3), pp. 348–378, 2012.
[4] Ashley, R., Lain, L., Ward, S., Shaffer, P., Walker, L., Morgan, C., Saul,
A., Wong, T., Moore, S., Water-sensitive urban design: opportunities for
the UK. Proceedings of the ICE-Municipal Engineer, 166(2), pp. 65–76,
2013.
[5] Wong, T., & Brown, R., The water sensitive city: principles for practice.
Water Science & Technology, 60(3), pp. 673–682, 2009.
[6] Potter, K., Ward, S., Shaw, D., Macdonald, N., White, I., Fisher, T., Butler,
D., & Kellagher, R., Engineers and planners: sustainable water management
alliances. Proceedings of the ICE-Engineering Sustainability, 164(4),
pp. 239–247, 2011.
[7] Brown, R.R., & Clarke, J. M., Transition to water sensitive urban design:
The story of Melbourne, Australia. Facility for Advancing Water
Biofiltration, Monash University Melbourne, Australia, 2007.
[8] Stahre, P., Blue-green fingerprints in the city of Malmö, Sweden: Malmö’s
way towards a sustainable urban drainage. VASYD, Malmö, Sweden, 2008.
[9] Portland “Grey to Green” initiative; The City of Portland Environmental
Services website, http://www.portlandoregon.gov/bes/47203
[10] Maksimović, S., Xi Liu, S., & M. Lalić, M., Blue Green Dream Project’s
Solutions for Urban Areas in the Future. Reporting for Sustainability,
pp. 49–54, 2013, available online at http://www.sciconfemc.rs
/PAPERS/BLUE%20GREEN%20.pdf
[11] Lowe, P. & Phillipson, J., Reflexive interdisciplinary research: the making
of a research programme on the rural economy and land use. Journal of
Agricultural Economics, 57(2), pp. 165–184, 2006.
[12] Hoyer, J., Dickhaut, W., Kronawitter, L., & Weber, B., Water sensitive
urban design: principles and inspiration for sustainable stormwater
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
124 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 125
Abstract
With the increasing densification in urban settlements the economic and social
disruption caused by pluvial flooding events globally is significant and growing.
Furthermore these problems are compounded where many cities are located in
areas where climate change predictions are for increased rainfall frequency
and/or intensity. One possible solution is the wide scale retrofit with green roof
technology as a means of mitigating stormwater runoff in urban settlements.
However, it is not known currently where the most effective location for and
siting of the retrofitted green roofs in a city or town would be. Moreover, the
number of and type of green roof required to reduce pluvial flooding is
unknown.
This paper describes a proof of concept framework for an assessment of the
potential to reduce pluvial flood hazard through the retrofit of green roofs
combining an evaluation of the retrofit potential of office buildings in the Central
Business District (CBD) with state-of-the-art urban rainfall inundation
modelling. Using retrofit scenarios for Melbourne CBD commercial buildings
built between 1998 and 2011 and the rainfall profile of the February 2011 event,
the modelled depths of flooding were compared. The results show that the
potential to mitigate extreme events via retrofit would be enhanced by
consideration of buildings within the wider catchment.
Keywords: flood, green roof, inundation modelling, retrofit, CBD.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140111
126 Flood Recovery, Innovation and Response IV
1 Introduction
Globally weather patterns are changing [1], whilst it is not possible to attribute
specific extreme events to changing climates there is consensus that the
frequency of intense rainfall events is rising and will continue over most land
masses, including those where average rainfall is decreasing [1]. Intense rainfall
events can cause flash floods, particularly in dense urban areas with low
permeability.
For Australia, major flooding occurred over the densely populated East Coast
area for two consecutive years from 2010. The State Emergency Service
responded to over 100 requests for flood-related damage when a storm caused 29
mm of rain in half an hour at Perth Airport [2]. In March 2012 the Bureau of
Meteorology issued Flood Warnings and broad-scale Severe Weather Warnings
for heavy rain and flash flooding over much of northern and eastern Queensland.
The estimated costs of remediation of flood damaged buildings is A$20 billion
[3, 4].
The incidence of pluvial flooding is, in part, attributable to changing weather
patterns, and climate predictions for Australia include increased intense rainfall
in south-western and south-eastern Australia [5]. Though there has been a
general trend of declining autumn and winter rainfall in south-western and south-
eastern Australia, Australian average annual rainfall has increased slightly,
largely due to increases in spring and summer rainfall, most markedly in north
western Australia [5]. Added to this are development pressures and increasing
urban density which add to the growth in damage caused by these events [6].
Green roofs have been shown to be highly suitable stormwater controls for
retrofitting in dense urban areas [7]. As roofing areas can account for 40–50% of
the impermeable surfaces in urban locations, such modification offers the
potential to mitigate pluvial flooding without additional land-take being required
[8]. There are two main types of green roof: extensive (incorporating shallow
rooted species in a relatively thin substrate) or intensive (deep rooted species
found in roof gardens); as the load-bearing capacity of extant structures is a key
constraint, the lighter extensive type is generally more appropriate for retrofitting
applications. Mitigation of stormwater impacts occur via two processes: water is
absorbed by the growing medium, thereby delaying the onset of runoff and
attenuating peak flows; the stored water is then released by a combination of
evaporation and transpiration, mediated by the foliage [9]. These processes can,
therefore, relieve pressure on existing piped drainage systems; in extreme storm
events, however, capacity will be exceeded and the design must take this into
account [10]. Rose and Lamond’s meta-analysis [11] notes that reported
performance ranged from 42–90% of annual rainfall, whilst average retention
during storm events varied from 30–100%.
Detailed specific study is, however, needed in order to make the case for
retrofitting green roofs: firstly, the surface area and location of candidate roofs
within the urban space will have a great influence on the quantity and pattern of
runoff attenuated; secondly, a number of physical factors need to be considered
in determining retrofit suitability; thirdly, the meteorological conditions
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 127
(including the typical rainfall patterns) and hydrological factors (such as runoff
characteristics) both need to be examined. The geographical location of green
roofs has, unsurprisingly, an impact on their performance, owing to regional
climatic variation: vegetated roofs in a sub-tropical Mediterranean climate [for
example, 12] will perform differently from those in a temperate maritime climate
such as the UK [13]. Further key factors influencing performance have been
found to include roof characteristics such as overshadowing (which can inhibit
vegetation growth) and the degree of pitch of the roof [14]; other variables
include substrate type; species mix; vegetation height and local weather
characteristics (intensity of rainfall; antecedent moisture conditions) [11]. It is
also necessary to engage the support of the owners and occupiers of relevant
properties, if a retrofitting programme is to be successful; an aid to uptake can be
provided via incentive schemes such as those employed in New York City [15]
and Portland [16]. Therefore, in this paper, a framework for evaluation of runoff
attenuation through retrofit of an urban area is described, with particular
reference to a case study in Melbourne Australia.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
128 Flood Recovery, Innovation and Response IV
Figure 2: Melbourne CBD showing typical rooftops with Yarra River to
right hand side (land generally slopes left to right in this image).
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 129
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
130 Flood Recovery, Innovation and Response IV
3 Results
3.1 Summary of retrofit potential
The map of Melbourne CBD (Figure 3) shows the historic centre area known as
the ‘Hoddle Grid’ on the north bank of the Yarra River, in which buildings were
assessed for suitability for retrofitting green roofs. Although the CBD is long
established the analysis revealed that the majority of the 526 commercial
buildings in the database (60%) were constructed after 1940 and over half were
constructed post 1960 representing a large amount of stock which is potentially
due for renewal and upgrade. However, the heights of the buildings are highly
variable, with the majority of low to medium rise buildings being mixed in with
the approximately 30% of high rise buildings and skyscrapers. Such an
arrangement of buildings could mean that existing buildings which have
adequate structural strength to accommodate retrofitting with green roofs may be
unsuitable because of overshadowing, which adversely affects planting.
However it is possible that consideration of other plant types with substrates
designed to be more absorbent than existing specifications might change this
finding. Furthermore, the orientation of many of the buildings in the CBD was
also seen to be unhelpful to successful plant growth.
Figure 3: The ‘Hoddle Grid’ area of Melbourne CBD with the modelled area
outlined in black (courtesy N. Bhattacharya-Mis).
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 131
Through visual inspection of the roof, using the Google Earth and Google
Map software, an evaluation of individual rooftops was carried out and each roof
classified as suitable, not suitable or indeterminate with respect to retrofit
potential. This evaluation was based on pitch, roof coverage by service plant and
roof construction. From this detailed study it appeared that only 15% of
Melbourne CBD buildings were considered suitable for retrofit; of the
remainder, 80% were seen not to be suitable and 5% were indeterminate. The
database covered only commercial office building stock, however, and many
other land use types exist in the CBD, including retail, residential and
educational. It is likely that some of the roofs of these buildings would also be
suited to retrofit, which would be expected to change the outcome. Finally, it
appears that the origin of the body of water which lead to the flooding in the
CBD was to the north of, and outside of, the Hoddle Grid and it is logical that
consideration of green roof retrofit potential in this stock is undertaken and
analysed.
The model outputs are colour-coded to show water depths (as illustrated in
Figure 4). The results of the inundation model clearly showed that major flows
within the Melbourne CBD during intense storms originate from outside the
CBD grid: for example, the route of the Yarra tributary culverted below
Elizabeth Street appears here as the white line running on a diagonal from north-
west to south-east, entering the main river to the south of the railway station.
A comparison of the modelled outputs reveal that the water depths could be
expected to decrease if green roofs were to be retrofitted in the area. In the
vicinity of Flinders Street Station, for example (Figure 5) the south-west corner
of the road grid has very dark shading (water over 1 metre depth) in scenario 1
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
132 Flood Recovery, Innovation and Response IV
(no green roofs), but this is replaced by a light grey (around 0.5 metre depth) in
scenario 3 (suitable roofs retrofitted). The granularity of the comparisons is, of
course, more apparent in the full colour images.
Figure 5: Detail of Flinders Street Station area without green roofs (left) and
with green roof retrofit (right).
3.4 Summary
The results suggest green roofs offer a potential for mitigation, in that flood
levels could be reduced in the affected areas, leading to concomitant reduction in
impacts; however, flooding of any depth presents issues for business continuity.
The key finding is that the historic flow pathways, normally hidden beneath the
urban district, will continue to re-establish their flow patterns during extreme
events, directing pluvial flows from the wider catchment towards the Yarra
River: in order to mitigate pluvial flooding in the CBD it will, therefore, be
necessary to take a wider catchment approach.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 133
the CBD area was greatly reduced, reflecting the fact that high density, prime
real estate property is not particularly sympathetic to green roof technology. The
implication of this finding for the city of Melbourne is that the potential to
mitigate flooding within the CBD solely by using CBD buildings is limited,
although other benefits deriving from the use of green roofs and other green
infrastructure are well recognised.
The results of the inundation model clearly show that major flows within the
Melbourne CBD during intense storms originate from outside the CBD grid. The
historic channels that are hidden beneath the urban district re-establish their flow
patterns during extreme events, directing pluvial flows from the wider catchment
towards the Yarra River: in order to mitigate pluvial flooding in the CBD it will
be necessary to take a wider catchment approach.
The potential of the framework to aid in decision making is clearly
demonstrated through the case study and therefore it is recommended to widen
the scope if the framework to the wider Melbourne area in order to evaluate
where the greatest retrofit opportunities exist.
Acknowledgements
This research was partially funded by the RICS research trust (Project no 464)
Retrofit of Sustainable Urban Drainage (SUDS) in CBD for improved flood
mitigation
Part of the research was also performed as part of an interdisciplinary project
programme undertaken by the Blue-Green Cities Research Consortium
(www.bluegreencities.ac.uk). The Consortium is funded by the UK Engineering
and Physical Sciences Research Council under grant EP/K013661/1, with
additional contributions from the Environment Agency, Rivers Agency
(Northern Ireland) and the National Science Foundation.
Thanks are due to the following for data supplied to the project:
Contour and shapefile data: David Hassett, GIS Team Leader, Property
Services, GPO Box 1603, Melbourne, Victoria 3001 Australia
Bureau of Meteorology data: Dr Blair Trewin, National Climate Centre, GPO
Box 1289, Melbourne, Vic 3001, Australia
References
[1] Solomon, S. and D. Qin, Climate Change 2007: The Physical Science
Basis. Contribution of working Group 1 to the Fourth Assessment Report
of the Intergovernmental Panel on Climate Change. 2007, Cambridge
University Press.: Cambridge, UK and New York, USA.
[2] Bureau of Meteorology. Heavy rain and Flooding. 2012 9th July 2012;
Available from: http://www.bom.gov.au/wa/sevwx/perth/floods.shtml.
[3] Bloomberg (2012) Insurers count cost in Queensland as floods peak.
[4] Companies and Markets (2011) Australian Flood Damage Reconstruction
Likely to Cost Billions.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
134 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 135
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
This page intentionally left blank
Section 5
Property-level flooding
and health consequences
(Special session
organised by C. A. Booth)
This page intentionally left blank
Flood Recovery, Innovation and Reponse IV 139
Abstract
In June 2013 the UK Government and the ABI announced plans for a new system
of insurance called ‘Flood Re’. This announcement was the first step towards
setting up a ‘not for profit’ scheme which aims to ensure the continuation of
affordable insurance for households with the highest flood risk. This research
investigates whether the widespread provision of flood insurance is a factor in the
low uptake of property level resilience measures. In the context of transition it
further examines whether there is growing impetus for the concept of
incorporating such measures. The literature establishes that the historic insurance
regime provided few incentives for installation of flood resilience measures and
that there are a number of factors beyond the provision of insurance which
influence the low uptake of measures. The impact of potential changes in the
insurance regime is explored in more detail through a number of semi-structured
interviews with key flood risk management professionals and academics. The
research finds that links between the provision of insurance and the installation of
resilience measures are significant. Flood resilience measures will continue to be
part of the wider strategy of community engagement with an integrated approach
to flood risk management. The newly proposed ‘Flood Re’ is intended to be a
transitory measure that will allow householders to adapt and take the necessary
measures to protect themselves. However, in the long term, the anticipated move
towards risk based pricing in whatever form may provide better incentives to
households to adapt and this could be reinforced by other measures to support
households in adaptation.
Keywords: adaptation, flood risk management, insurance, resilience.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140121
140 Flood Recovery, Innovation and Response IV
1 Introduction
Flooding is the biggest natural threat facing the UK and flood risk is predicted to
increase due climate change, development and the gradual deterioration of flood
defence assets (ABI [1]). The UK has benefited from a private insurance system
for flood risk that has existed for over half a century and this has formed the main
source of financial protection for households in flood risk areas. A series of
agreements were in place between the government and the insurance industry
which ensured that the majority of households had access to affordable insurance
for flooding. These started with what was referred to as the ‘Gentleman’s
Agreement’ and more recently the ‘Statement of Principles on the provision of
flood Insurance’ (DEFRA [2]). On the 27th of June 2013, the Government and the
insurance industry announced a new agreement which would guarantee
the availability of insurance for households in flood risk areas. The preferred
solution would be an industry-run, not-for-profit scheme called ‘Flood Re’. This
scheme will effectively cap the maximum amount paid by the 1–2 % of households
at highest risk of flooding. It would be funded by an industry backed levy set to
be £180 million per year for the first 5 years, an equivalent of £10.50 for every
UK household. It will take time for ‘Flood Re’ to become operational and therefore
the insurance industry has voluntarily agreed to abide by the Statement of
Principles until such a time that ‘Flood Re’ can be introduced (DEFRA [2]).
The considerable uncertainty surrounding the decision over the future of
household flood insurance, together with an increased frequency of flood events,
has resulted in increased awareness of the concept of incorporating flood resilience
at property level. However, uptake of these measures remains low. In 2008,
DEFRA announced less than 5000 homes have adopted flood resilient and
resistance measures (Bichard and Kazmierczak [3]). To encourage an increase in
uptake in resilience measures DEFRA launched its property level flood protection
scheme. The 2 year programme ran until March 2011 and it delivered £5.2 million
to 1,109 households, the average cost to households for these measures was £4,832
(Environment Agency [4]). Whilst the UK Government has sought to influence
householders to take up flood protection measures, the strategies employed have
not been as successful as they had hoped (Bichard and Kazmierczak [3]). One
reason for this could be that the wide availability of insurance to households has
distorted their perception of risk. This study therefore sought to investigate the
extent to which the widespread provision of flood insurance is contributing to the
low uptake of property level resilience measures.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 141
The Government and the ABI have agreed upon a Memorandum of Understanding
which sets out how ‘Flood Re’ is likely to operate in order to progress with the
development of Government policy. However, there are still many issues which
need to be resolved. The Government are to introduce new legislation in the Water
Bill to enable the introduction of ‘Flood Re’. The main powers will be to compel
all insurers offering household insurance to participate in ‘Flood Re’ and provide
for ‘Flood Re’ to be funded through an industry levy (DEFRA [2]). Householders
should be aware that ‘Flood Re’ will be a transitional measure, intended to be
phased out within 20 –25 years. A ‘Sunset Clause’ will be included in the primary
legislation to set an expiry date for ‘Flood Re’, as well as powers to ensure the
orderly winding down of the scheme (DEFRA [2]). The policy objective is that
there should be a gradual transition towards risk reflective pricing (a free market),
which is intended to increase incentives for flood risk to be managed properly. The
Government intends to seek powers in the Water Bill to allow them to stand ready
to regulate if ‘Flood Re’ can’t be made to work for consumers and insurers. This
‘Flood Insurance Obligation’ will require insurance companies to insure a
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
142 Flood Recovery, Innovation and Response IV
proportion of properties from a register of high risk households. This should create
a level playing field and overcome the competitive pressures on insurers to
withdraw from flood risk areas (DEFRA [2]).
In the current insurance market, there is very little to suggest that the installation
of FRe technology to protect property will result in reduced premiums and
excesses for householders. A number of surveys have concluded that for the
majority, there was no evidence to suggest the installation of FRe technologies
would result in cheaper insurance (Bell [9]; Cobbing and Miller [10]; Harries
[11]). The way in which resistance and resilience measures can help is by avoiding
the need to involve insurance companies, or reducing the size of the claim made.
This can help to maintain access to mainstream insurance but is little incentive for
the installation of such measures.
For individual properties, and for those properties in areas that flood frequently,
FRe technology can be a cost effective means of reducing damage and disruption
(Harries [11]). However, many people perceive that flood resilience measures may
adversely affect property value or make their properties harder to sell. This is
perhaps a genuine concern; research by Lamond et al. [12] found that many
property buyers and sellers are often unaware of the flood risk to their property. In
fact there is evidence to suggest that flood prone properties aren’t discounted in
price over the long term. This was illustrated by Lamond et al. [12] who point to
properties in Bewdley which showed dips in value following flood events in 2001
and 2002. However, over the long term their value recovered (cited in Lamond
[p. 332, 13]). These measures therefore could be seen as a deterrent for potential
buyers. This perceived barrier is something which needs to be overcome.
There is a longstanding and growing debate that the provision of insurance may
prevent some from taking the necessary steps to protect themselves (Priest et al.
[14]; Lamond and Proverbs [15]; Harries [16]; O’Neill and O’Neill [17]). The term
‘Moral Hazard’ is defined by O’Neill and O’Neill [17] as a situation in which
individuals or organisations do not bear the costs of a particular risk and hence
lack incentives to change behaviour to reduce that risk. The question of ‘Moral
Hazard’ has been raised with regards to the behaviour of homeowners in protecting
their own properties. It is used when there is a tendency towards less responsible
behaviour by those who believe they are insulated from financial risk by insurance
(Harries [16]). Understanding ‘Moral Hazard’ on behalf of the homeowner is
critical to ensuring that a system of insurance is put in place which encourages
homeowners to protect themselves. A balance needs to be found between
providing affordable cover for those who need it and encouraging some form of
self-protection which will reduce the impact of a future flood event. The difficulty
with this is that if insurance is the default position, and it is also widely available,
then policy holders are unlikely to consider other avoidance strategies (Lamond et
al. [12]).
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 143
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
144 Flood Recovery, Innovation and Response IV
Due to the large volumes of data it was important to adopt a method of analysis
which made sense of the information and presented the findings in a logical and
coherent way. The transcripts were analysed using methods adapted from the
hermeneutic analysis method. ‘Hermeneutics’ is characterised by Haigh [22] as
examining the inter-relationship of the response from the interviews and relating
this to the aims of the research at large. In this case, one of the main challenges
was ensuring that the analysis remained focused upon the aims and objectives of
this thesis.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 145
4 Analysis of interviews
The majority of Interviewees acknowledged that there was a link between the
availability of affordable insurance and households protecting their own
properties. The general feeling was that if people had always been paid out on
insurance then they would not be motivated to protect their property. However,
Interviewees C and F made the point that there is more to the issue than just
financial impact. The emotional stress that is suffered from flooding is also
significant. Interviewee C said that “they hoped that where it is beneficial for
someone to protect their property they would do this irrespective of the
affordability and the availability of insurance”. In the literature review however,
there was little evidence to support this statement. It was found that uptake of flood
protection measures is still very low. The complexity of dealing with flood risk
was cited by interviewee D as a possible reason for the low uptake of these
measures: “It’s not as clear cut as installing a safety lock to your front door…it is
less clear how these mechanisms are going to work”.
It was apparent there was no simple answer to the problem of persuading
householders to take up FRe technology. It was surprising that only one
Interviewee (B) thought reduced premiums could be used. This reflects the
perceived complexity of implementing such a scheme. Interviewee A thought that
outright refusal of cover, which would make the property un-mortgageable, would
be a possible driver. They also thought pilot grant schemes from Defra which have
now evolved into partnership funding schemes were motivational for people.
Interviewee G felt that people were more likely to take up flood resilient measures
if they were passive, such as flood doors that look like normal doors and kite
marked one way valves. “We are getting there but the PLP industry is still very
young and these products are still very expensive”.
Integration into building regulations and more robust planning were responses
that also stood out.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
146 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 147
From an insurer’s point of view, this kind of assurance is essential because if they
are ever going to offer discounts for flood protection measures they will need to
know the quality of the workmanship involved. Interviewee A pointed out that
there are precedents linking the benchmarking of quality with insurance. He
suggests standards for sprinkler systems and burglar alarms need to be adhered to
as a requirement for insurance against fire and theft. Respondent G says an in depth
report on what did and didn’t work in 2012 was required from the Environment
Agency and Defra in order to drive up standards.
The general feeling amongst interviewees was that building professionals
lacked the necessary experience and training to deal effectively with flood risk.
Even from those within the surveying profession. Respondent B highlighted the
complexities of the drying process was an area that needed better understanding.
There was acknowledgement that there are professionals who specialise, but that
for the moment they are in the minority. There are dedicated facilities in the UK,
such as the National Flood School, which specialise in training in flood restoration
for building professionals. There was concern with Interviewees E, D and F, that
Local Authorities, with their greater responsibility for FRM, may not yet have the
skills necessary to take on this role.
The response to whether we can learn from flood policies from other parts of
the world was that even though there are always lessons that can be learned from
other countries, there is not one specific model that will necessarily solve all the
issues. Other countries will have different climates, landscapes, populations and
social structures. There are many elements that forbid the applicability of a generic
solution to individual cases in the UK. Interviewee D rightly suggested that the
EU strategy which led to the ‘making space for water’ directive, has set out a
comprehensive new approach and triggered a rethink in the UK as to how flooding
is dealt with.
5 Conclusions
The research has established several key points that need to be achieved to promote
and incentivise the use of property level resilience. Amongst these, stronger
partnerships between insurance companies and those developing FRe technology
should be established. Passive measures which are deployed automatically could
be a key factor in encouraging insurance companies to reduce premiums as this
would remove the element of human error when it comes to deploying these
products. Insurers could be more influential in promoting and incentivising
property level resilience. For this to work there needs be assurances that the
products are going to reduce the amount the insurance companies pay out in the
event of a flood. There needs to be standardisation, both for the products and the
installation. Kite mark schemes are improving, and there are a lot more tested
products on the market. However, standards for installation need to come from
regulation, at the moment there is plenty of guidance and codes of practice, but in
an emergency these tend to be less adhered to. A relatively small reduction in the
cost of premiums will not be enough of a driver for change, because the initial
costs for the installation of FRe technology may be too high. Community schemes
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
148 Flood Recovery, Innovation and Response IV
need to allocate money to those households that are most vulnerable and need to
encourage innovative ideas for community resilience beyond the use of FRe
technology.
It has been established from the research that the provision of flood insurance
does impact upon householders decisions of whether or not to install property level
resilience measures. The term ‘Moral Hazard’ is used by many researchers to
describe a tendency towards less responsible behaviour by those who believe they
are insulated from financial risk by insurance. The difficulty is that if insurance is
the default position then it is unlikely that householders will consider other
avoidance strategies. It is important to emphasise that the measures proposed to
protect the availability of affordable flood insurance under ‘Flood Re’, are only
intended to be temporary. They will be phased out within 20-25 years, when the
market will move towards risk reflective pricing. There is a danger that some
householders that could benefit from FRe technology will be drawn into a false
sense of security by the availability of affordable insurance. It is therefore
important that during this transitional period of ‘Flood Re’, opportunities are taken
to develop and promote flood resilience for properties which will benefit most
from these measures. This assertion is backed up by the findings of the literature
and the interviews, which have highlighted that property level resilience measures
will need to play an increasingly important role in managing future flood risks.
References
[1] Written evidence to the Environment, Food and Rural Affairs Select
Committee inquiry into flood funding. Association of British Insurers, ABI.
Online. http://www.publications.parliament.uk/pa/cm201213/cmselect
/cmenvfru/writev/flood/m07.htm
[2] Securing the future availability and affordability of home insurance in areas
of flood risk. Department for Environment, Food and rural Affairs, DEFRA.
Online. https://consult.defra.gov.uk/flooding/floodinsurance
[3] Bichard, E. & Kazmierczak, A., Are homeowners willing to adapt to and
mitigate the effects of climate change? Climate Change (2012) pp. 112:
633–644, 2011.
[4] Guidance on surface water flood mapping for lead Local Authorities.
Environment Agency: Bristol, 2012.
[5] SMARTeST – Glossary. Online. tech.floodresilience.eu/attachments/article
/40/smartest-glossary.pdf
[6] White, I., Lawson, N., O’Hare, P., Garvin, S. & Connelly, A., Six Steps to
Property Level Flood Protection – Guidance for local authorities and
professionals: Manchester, 2012.
[7] Kazmierczak, A., and Connelly, A., Buildings and Flooding – a risk
response case study. EcoCities project, University of Manchester.
Manchester, 2011.
[8] Adaptation Sub-committee., Climate change – is the UK preparing for
flooding and water scarcity: Adaptation Sub – Committee Progress Report
2012, Committee on Climate Change: London, 2012.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 149
[9] Bell, A., Morpeth Flood Action Group Insurance Survey – Results and
Analysis, Online. http://www.morpethfloodaction.org.uk/survey.html
[10] Cobbing, P., and Miller, S., Property level protection and insurance: Main
report – 2012, National Flood Forum: Bewdley 2012.
[11] Harries, T., Review of the Pilot Flood Protection Grant Scheme in a
Recently Flooded Area. Department for Environment Food and Rural
Affairs: London, 2009.
[12] Lamond, J. E., Proverbs, D.G., and Hammond, F.N., Accessibility of flood
risk insurance in the UK: confusion, competition and complacency, Journal
of Risk Research, 12 (6), pp. 825–84, 2009.
[13] Lamond, J. E. (2012) Financial Implications of Flooding and the Risk of
Flooding on Households, in: Lamond, J., Booth, C., Hammond, F., and
Proverbs, D. (eds.) Flood Hazards: Impacts and Responses for the Built
Environment. CRC Press: Boca Raton pp. 317–326, 2012.
[14] Priest, S.J., Clark, M.J., Treby, E.J., Flood Insurance: The challenge of the
uninsured, Area 37.3 Royal Geographical Society, pp. 295–302, 2005.
[15] Lamond, J.E., Proverbs, D, G., Flood Insurance in the UK – a survey of the
experience of flood plain residents, in: Proverbs, D., Brebbia, C.A., and
Penning-Rowsell, E. (eds.) Flood Recovery, Innovation and Response, WIT
Press: Southampton, 2008.
[16] Harries, T., Why Most “At Risk” Homeowners Do Not Protect Their Homes
From Flooding, in: Lamond, J., Booth, C., Hammond, F., and Proverbs, D.
(eds.) Flood Hazards: Impacts and Responses for the Built Environment,
Taylor and Francis Group: Boca Raton, 2012.
[17] O’Neill, J., and O’Neill, M., Social Justice and the future of flood insurance,
Joseph Rowntree Foundation: York, 2012.
[18] Botzen, W.J., Van Den Bergh, J.C.J.M., Monetary valuation of insurance
against flood risk under climate change, International Economic Review,
53, pp. 1005–1026, 2012.
[19] White, I., O’Hare, P., Garvin, S., Connelly, A. (2012) Barriers to flood
resilience: Findings from the SMARTEST project, The University of
Manchester and BRe: Manchester, 2012.
[20] Boobier, T. (2012) The Development of Standards in Flood Damage Repair:
Lessons to be learned from the United Kingdom Example in: Lamond, J.,
Booth, C., Hammond, F., and Proverbs, D. (eds.) Flood Hazards: Impacts
and Responses for the Built Environment, CRC Press: Boca Raton, pp. 129–
139, 2012.
[21] Kidd, B., Tagg, A., Escarameia, M., von Christierson, B., Lamond, J.,
Proverbs, D. Guidance and standards for drying flood damaged buildings,
Signposting current guidance – BD2760, 2010.
[22] Haigh, R. Interviews: A negotiated partnership, in: Knight, A., and Ruddick,
L. (eds.) Advance Research Methods in the Built Environment. Oxford:
Blackwell: Oxford, pp. 111–120, 2008.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
This page intentionally left blank
Flood Recovery, Innovation and Reponse IV 151
Abstract
Installing perimeter floor drains is a waterproofing option used inside buildings,
which manages water ingress in basements. Newly designed products providing
a means to access the inverts of perimeter floor drains and facilitate inspection
and maintenance have been designed and are Patent Pending GB1117089.1,
GB1102662.2, and GB1102661.4. The new system incorporates pivotally
connected fittings with water deflector plates combined with straight lengths of
perimeter floor drain. The fittings ensure that secure joints with both axial and
invert alignment are maintained throughout, which are essential for movement of
water through the level perimeter floor drains and accessibility as recommended
by British Standard BS8102: 2009. The patents demonstrate several practical
advantages over those of existing designs and systems, which is evidenced by the
commercial uptake by Safeguard Europe Ltd. and product installations in
hundreds of flood-risk basement apartments in the UK.
Keywords: property adaptation, perimeter floor drainage, flood resilience,
patent product.
1 Introduction
Provision of a building basement can increase housing density without a
reduction in habitable space and so enable more homes to be built in a
development where building height or footprint is limited [1]. Basements have
long been used in commercial buildings for plant rooms, storage space and car
parking but in recent decades the arcane construction used in basements has been
transformed by modern technology to provide fully acceptable below ground
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140131
152 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 153
Standards BS8102: 2009 that all drainage systems and installations must now be
able to be tested, incorporate accessibility and be maintainable for the life of the
structure. In the past the ‘Achilles heel’ of Type-C protection has always been
accessibility into perimeter drainage systems that were buried within the
structure, which often resulted in the removal of whole floors to investigate
defects.
New products which provide means to access the inverts of perimeter floor
drains and facilitate inspection and maintenance are now commercially available
[9]. The designs featured are Patent Pending GB1117089.1, GB1102662.2, and
GB1102661.4 [10–12]. Perimeter floor drains are used inside buildings that are
subject to water ingress, typically basements and flood situations. In a typical
basement construction the perimeter floor drain collects water from Type-C
(drained cavity) installations as described in BS8102: 2009. A waterproofing
installation to BS8102: 2009 Type-C uses waterproof structures to form a cavity
between floor, adjacent wall, and/or ceiling which is then drained into the
perimeter floor drain. However, the Type-C system is incapable of accepting any
hydrostatic pressure and the collected water must be removed from the system or
leaks will occur. The perimeter floor drain is one of the most important parts of
the water collection and management system that must function properly to
prevent water from entering the basement.
The perimeter floor drain is laid directly onto the level floor of the building at
the internal wall to floor join, therefore the invert of the perimeter floor drain has
to be laid level on the basement floor. The perimeter floor drain is installed
around the perimeter of the basement and a T-piece is incorporated at some point
to transfer all the water collected by the perimeter floor drain into a link drain
which then leads to a sump/pump unit. The sump/pump unit is typically located
within the central floor area having a discharge pipe leading to an outside
domestic drain. All water ingress through walls, floor, ceilings, the vulnerable
wall to floor join and collected from the waterproofing cavities is transported
along the perimeter floor drain to the link drain and then into the sump/pump for
removal from the building. Most importantly these perimeter floor drains and
link drains have no gradient to move the water as they are laid directly onto a
level floor. It is the successful movement of the water along the perimeter floor
drain that presents the current problem because if the water is allowed to build
up then as the Type-C installation cannot withstand hydrostatic pressure we will
get leaks into the internal basement space. Furthermore, if water is allowed to
pond within the perimeter floor drain then the standing water leads to
recrystallization of salts which have entered as a solution within the water.
Blockages can then occur due to the deposition of recrystallized salts, which
cause more standing water and more blockages etc.
Water can only move along level perimeter floor drains by hydraulic head due
to the gravitational effects on the water. This driving force is very weak and it is
therefore essential that the invert of the perimeter floor drain is completely level
and has no obstructions. Any minor obstruction or misalignment of a perimeter
floor drain invert causes an increase in depth of water upstream of the
obstruction which may then exert hydrostatic pressure on the waterproofing
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
154 Flood Recovery, Innovation and Response IV
structure and cause a leak into the occupied basement. As a simple illustration
we can consider a perimeter floor drain installed in a basement that is 8 x 6m on
plan. The water collected by the perimeter floor drain at the furthest point from
the link drain will have to travel at least 14m and negotiate at least two elbow
bends and a T-piece connection into the link drain when the depth of the
perimeter floor drain and all that needs to be filled up in order to cause a leak is
only 40mm depth of standing water within the drain. This example does not take
into account the fact that matters are often much worse as there may be a back-
fall due to the floor itself not being exactly level across its surface. A floor that is
8m in length may well be 25mm lower at the point furthest away from the link
drain, combine this with a mere 5mm invert obstruction at each of the two
misaligned bends and the T-piece and we have the perimeter floor drain full of
water with a potential leak into the occupied building. In these circumstances,
which are unfortunately often found on site, the perimeter floor drain cannot be
accessed in order to be flushed out and cleaned and it is not possible to maintain
the system as per the requirements of BS8102: 2009.
The perimeter floor drain itself is usually made from plastic and has a smooth
invert. Usually it is the elbow bends and T-piece, where inverts become
misaligned, that causes problems with obstructions that lead to a leak. As a result
of problems with Type-C installations and subsequent leaks all new work
involving the use of perimeter floor drains is now subject to revised BS8102:
2009 code of practice for protection of below ground structures against water
from the ground. This applies to basement and flood situations where
accessibility and repairability must be allowed for in design. For example
BS8102: 2009 page 13: shows a cross-sectional view of Type-C (drained)
protection that details a perimeter floor drain as a maintainable drainage channel,
BS8102: 2009: 10.2.1.2. requirements: where the floor cavity incorporates
perimeter floor drain channels, which discharge into sump(s), both the channels
and the sumps should be cleaned before, during and after installation of the
membrane to allow uninterrupted drainage, BS8102:2009: 10.3.1. requirements:
access points that allow routine maintenance of channels and outlets should be
incorporated into the design of the waterproofing system, BS8102: 10.3.2.
requirements: immediately after the installation of a cavity drain system the
perimeter floor drainage channels and sumps should be cleaned out and tested.
The servicing requirements for the waterproofing system should be clearly set
out in the documentation supplied by the designer to the client, including the
need for regular planned maintenance of the drainage and/or pumping systems
not less than once a year.
In order to have perimeter floor drains that function correctly and also satisfy
the requirements of British Standards for flushing out and cleaning perimeter
floor drains, it is necessary to have perimeter floor drain joints, elbow bends and
T-pieces that are securely fitted together and accurately aligned both axially and
across inverts. The current methods of perimeter floor drain installation use a
straight butt joint that is often held together with adhesive tape. These butt joints
are easily disturbed both during their own installation and also during the
subsequent floor laying and framed wall building operations that are carried out
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 155
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
156 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 157
(Figure 1(b)) is being used on an installation the straight channel of the perimeter
floor drain is placed against the end of the extended elbow bend channel (S) and
then the top section of the perimeter floor drain is snapped into place, spanning
across the channel joint to create a secure staggered joint. There is no longer a
straight butt joint that passes directly through both top and channel sections of
the perimeter floor drain which is the major disadvantage associated with current
rigid bends. If a one part perimeter floor drain is being used for the straight
lengths of perimeter floor drain then the protruding channel section (S) will slide
inside any manufacturers’ one part perimeter floor drain channel currently
available to form a secure joint. The pivotally connected elbow bend is universal
and can therefore be used in installations of two part and one part perimeter floor
drains and in both cases will provide a secure joint to the straight lengths of
perimeter floor drain and the pivotal connection will ensure that soffits and
inverts are kept at the same level across the bend to prevent obstructions to water
flow [11].
Figure 3(b) shows the pivotally connected T-piece with internal deflector
plate. The top section (U) of a length of two part perimeter floor drain is
pivotally connected (Q) to the link drain top section (V). This pivotal connection
allows adjustment of the angle of intersection at the T-piece and ensures that on
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
158 Flood Recovery, Innovation and Response IV
a construction site installation of the link drain is axially aligned into the T-piece
in order to prevent obstructions to water flow and also achieve a secure joint.
The link drain top section (V) is set under the top section (U) so that the invert in
the link drain is lower than that of the perimeter floor drain to encourage water to
flow from the level perimeter floor drain invert into the link drain invert. A
length of two part lower channel section has the side wall cut away and is fixed
into the top section (U), similarly a lower channel section is cut and fixed into
the link drain top section (V). In both cases the channel sections are longer than
their respective top sections and protrude out as shown (S). As previously
described above for the elbow bend (see figure 3(a)) the protruding channels
provide secure joints to both one part and two part perimeter floor drain straight
lengths to ensure axial and invert alignment and hence no obstructions to flow.
The lower channel section at the point of intersection must have the sidewall
removed to allow water to pass into the link drain and removal of the sidewall
weakens the construction of the T-piece and reduces the capacity of the flat top
soffit section to handle floor loadings. A water deflector and support is internally
fitted that spans between invert and flat soffit top section at the point of
intersection. The support sits inside the T-piece in the channel section and serves
to both support the weak flat top section and due to its shape also deflects water
into the link drain passageway [12].
(a) (b)
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 159
Figure 4 is a plan view of perimeter floor drain, T-piece, link drain and sump
chamber to show the accessibility provided by the new pivotally connected T-
piece with water deflector plate and support (b) and an existing standard T-piece
in (a). The left hand side (a) shows perimeter floor drain A with a standard T-
piece W that joins the perimeter floor drain to the link drain G. Link drain G runs
to the sump/pump chamber H. The drain inspection camera or hose R can enter
through the sump/pump chamber lid and be pushed along the link drain G. When
R reaches the T-piece it cannot negotiate the corner into the perimeter floor drain
and hits against the channel wall, there is no way to direct the camera or hose
around the corner and along the perimeter floor drain. In Figure 4 the right hand
side (b) the new T-piece has the internal water deflector plate and support D
attached. The drain inspection camera or hose R can now pass around the corner
into the length of the perimeter floor drain. By means of pushing R along
different sides of the link drain both lengths of perimeter floor drain on either
side of the T-piece can now be accessed. The internal water deflector plate and
support D provides a constant radius for the bend in the inspection camera or
hose and prevents kinking as they are fed through the T-piece W along the
perimeter floor drain A.
Figure 4: Plan views of the perimeter floor drain, T-piece, link drain and
sump chamber to show the accessibility provided by the new
pivotally connected T-piece with water deflector plate.
Figure 5(a) is a plan view to show water and/or sediment flowing into the
sump/pump chamber. The flushing water has been introduced through jetting
points set into the perimeter floor drain. In this plan the perimeter floor drain A
is connected to the link drain G with a T-piece W. The T-piece W has an internal
support D attached. This plan shows the drainage system in use. The arrows C
show the flow of water through the system. The perimeter floor drain collects
water though pre-drilled holes in the channel sides, this water runs to the T-piece
where it is passed into the link drain G and hence on to sump/pump chamber H.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
160 Flood Recovery, Innovation and Response IV
The arrows C show flow of water and/or flow of sediment when the drainage
system is being cleaned by flushing out. The support D stops water and/or
sediment being washed back and forth across the end of the link drain and flows
into the link drain G. Subsequently, sediment can be removed by way of the
sump/pump chamber lid. As shown in Figure 4 one side of the new T-piece can
be used to introduce a hose into the perimeter floor drain and the flushing water
will then travel around the perimeter floor drain to carry sediment back to the
new T-piece where the deflector plate will direct it into the link drain and then
onto the sump for removal.
Figure 5(b) is a line drawing of the new T-piece to show the internal water
deflector plate that also acts as a support for the T-piece soffit. For illustration
purposes this drawing shows a rigid connection at the T-piece join, whereas in
practice the new T-piece also features an adjustable joint and an invert level
slightly lower in the link drain connection to aid water movement from perimeter
floor drain to link drain and sump [12].
(a) (b)
Figure 5: (a) Plan drawing to illustrate water and/or sediment flowing into
the sump/pump chamber and (b) a sectional drawing of the new T-
piece to show the internal water deflector plate that also acts as a
support for the T-piece soffit
5 Discussion
The products/system portrayed offers several advantages over existing
approaches to the option of installing a drained cavity construction in a basement
apartment. Pre-made elbow bends and T-piece make installation much easier and
quicker during construction. The new T-piece with internal water deflector
enables access into the perimeter floor drain system for inspection and effective
flushing via the sump chamber. The pre-made bends can be fitted to the exact
corner angle and then the secure joints hold the system together during assembly.
The T-piece link drain connection can be accurately aligned to meet the sump
location. The whole installation process needs less skill to complete, as the
operatives no longer have to try and mitre odd shaped plastic mouldings with
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 161
hand tools. With pre-made items the perimeter floor drain installation is simply
snapped together with invert levels and axial alignment guaranteed (Figure 6).
The deflector plates incorporated into corners and T-piece provide the
accessibility as recommended by the revised British Standard BS8102 2009 for
inspection and maintenance. The contractor can demonstrate on handover to the
client that the system works and is now able to build in the accessibility to
investigate any faults and also offer periodic maintenance contracts for the peace
of mind of the client.
Figure 6: Photos of the newly designed, easy to fit, perimeter floor drain
being installed by a semi-skilled professional in a basement
apartment.
6 Conclusions
Shortage of housing is encouraging the conversion of building basements into
habitable spaces. Mitigating the impact of below-ground living means there is a
need to adapt and protect accommodation against water-ingress through the
building envelope. Recommendations and guidance on the available approaches
for dealing with the entry of water from surrounding ground into a structure
below ground level include the use of a waterproofing barrier applied to the
structure, creation of a structurally integral watertight construction
or installation of a drained cavity construction. With the latter approach proving
popular, new flood-resilient products, which provide means to access the inverts
of perimeter floor drains and facilitate inspection and maintenance, are Patent
Pending GB1117089.1, GB1102662.2, and GB1102661.4. These can
demonstrate several practical advantages over those of existing designs and
systems. Recognition is demonstrated by the commercial uptake by Safeguard
Europe Ltd. and their installation in several hundred flood-risk basement
apartments in the UK
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
162 Flood Recovery, Innovation and Response IV
References
[1] Building Research Establishment, (2007) Good Building Guide 72:
Basement Construction and Waterproofing. Part 1: Site Investigation and
Preparation. Part 2 Construction, Safety, Insulation and Services.
Amersham: IHS BRE Press.
[2] Construction Industry Research Information Association (1995) Water-
Resisting Basements- Report 140. London: CIRIA.
[3] Tovey, A. and Keyworth, B. (1998) Basements: Land Use and Energy
Conservation – Evaluation with Market and Construction Survey.
Crowthorne: British Cement Association.
[4] Basement Information Centre (2004) Basement Information Centre
Approved Document – Basements for Dwellings. Camberley: BIC.
[5] Department of the Environment Transport and the Regions (1998) English
House Condition Survey 1996. London: The Stationary Office.
[6] Wolcox, S. and Perry, J. (2013) UK Housing Review: 2013 Briefing
Paper. Coventry: Chartered Institute of Housing.
[7] British Standards Institution (1990) BS8102:1990 Code of Practice for
Protection of Structures against Water from the Ground. London: BSI.
[8] British Standards Institution (2009) BS8102:2009 Code of Practice for
Protection of Below Ground Structures against Water from the Ground.
London: BSI.
[9] www.safeguardeurope.com/products/aquadrain.php
[10] Beddoes, D.W. (2011a) Cleanable Perimeter Drain System. Pat Pend.
117089.1.
[11] Beddoes, D.W. (2011b) Drain Deflector. Pat Pend. 1102662.2.
[12] Beddoes, D.W. (2011c) Perimeter Floor Drain T-Piece. Pat Pend.
1107397.0.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 163
Abstract
Despite the increasing impacts of recurrent flooding, there is dearth of research
involving businesses preparedness and recovery. This research therefore focused
on investigating the patterns of preparedness and trends in recovery among
business properties. A review of literature was performed primarily to recognize
the gaps requiring investigation followed by identification of two case studies
(Wakefield and Sheffield in the UK) for empirical data collection. The survey
enquired about the level of preparedness among a sample of the flood-affected
business community using a self-administered questionnaire. Questions addressed
the type of mitigation and preparedness activities and measures that they engaged
in and adopted for recovery along with factors like time cost of recovery and
sources of finances. Results from the survey suggest that business interruption was
highly influential in terms of differential cost and time of recovery. It was not the
direct impact of flooding rather the under-researched and lesser-perceived
business interruption through indirect factors that were more significant for cost
and time of recovery. Furthermore, evidence of businesses relying highly on self-
finance was also apparent from the survey. Knowledge gained from the survey for
preparedness measures indicated that out of flood-affected samples that flood
experience is an important indicator of preparedness and mitigation actions. The
outcome of the research has highlighted some of the least researched phenomena
in the flood-affected business property sector and can demonstrate the need for
more widespread efforts to improve disaster recovery among businesses and a
novel input for future research.
Keywords: businesses, flood risk, damage, disruption, preparedness, recovery.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140141
164 Flood Recovery, Innovation and Response IV
1 Introduction
Ensuring continuity of businesses in times of disaster is necessary for business
sector and it is necessary to synthesize prevention and protection measures in a
pre-disaster scenario in order to respond and recover faster during and after an
event and ensure continuous business operation [1] . Reduction of direct impact
among business enterprises require emergency relief services for cleaning up,
rebuilding and restoring properties. On the other hand, mitigation of indirect
effects demand financial assistance, employees’ return to job, suppliers and
consumer adjustment to the market, and essential service management. The
Committee on Disaster Research in Social Sciences has rightly suggested that
enterprises or businesses who are engaged in preparedness and mitigation
activities will be less vulnerable to natural disasters [2]. In theory insuring property
and businesses against flood damage can be treated as one of the effective tools of
mitigation; however, literature suggest that about 90% of the small and medium
enterprises (SME) in UK are under-insured [3]. Pitt’s report after the 2007 flood
event recommended the necessity of adoption of property level resistance and
resilience measures for all types of properties in the UK [4]. Research has
previously shown that business properties lack in such sources of protection
against impacts of flooding [5, 6] but such research has been limited in scale and
scope. Therefore, the main focus of this study is to further identify and investigate
patterns of preparedness and link this to trends of recovery using a case study
approach. The paper is structured in four sections. First, existing literature is
reviewed to gather impression of the flood risk and response situation in general
among flood plain population. Based on the rationale gained from literature review
methodology for specific case study areas were discussed and finalized. This is
followed by section on observed patterns of preparedness and recovery from the
selected case studies and finally, recommendations for future studies were
proposed before concluding remarks.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 165
reduction through appropriate flood response and recovery for flood plain
population involves factors such as being kept well informed through media; early
response to warnings; consideration of warning dissemination time and evacuation
time from the building [23, 24]. To respond to indirect effects of flooding, it is
essential to recover and restore vital records (insurance papers, tax return
documents, tracing orders etc.). This is greatly facilitated through appropriate
preparation and backup in advance of flooding [1, 28] and such activities may be
specified through a continuity plan. Financial constraint can make the recovery
process take longer [26] therefore adequate insurance is indicated. .Without
insurance, larger enterprises have greater financial capacity to respond to flood
effects and therefore tend to recover faster from floods while smaller enterprises
might suffer more as a result of their financial constraints [25, 26]. Factors such
as reluctance of finance companies to supply loans for repair of the affected
property and high premiums set by insurance companies for flood prone properties
can prove to be fatal for the recovery process [24]. Such actions can have
catastrophic impacts on many businesses; one report suggests that around 43% of
the properties closed down after a disaster and about 29% of those closed down
within two years [29]. Based on the insights gained from literature, the following
section will detail the methodology adopted in collection of empirical data from
two selected case study locations to analyse the situation of preparedness and
recovery in flood-affected areas for business properties of flood plain population.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
166 Flood Recovery, Innovation and Response IV
3 Methodology
It was necessary to identify areas for empirical data collection which have a
historical record of flooding and have a comparatively large population of
commercial properties at risk. Case study approach (although being
geographically limited) was appropriate for the purpose because of the scattered
nature of flood-affected properties and the lack of publicly available national data
sources with evidence of commercial properties affected by flooding in the past.
Therefore to increase the probability of tracing a comparatively large sample a
larger population at risk was selected through a systematic case study selection
approach.This is general consensus in literature that better prepared businesses
will fare well in case a disaster strikes [30–32]. The questionnaire survey enquired
about the level of preparedness among flood-affected sample population by asking
questions concerning type of mitigation and preparedness activities they are
engaged in. Number of preparedness and mitigation measures was provided in the
questionnaire with a range of activities to choose from. Enquiry was also done to
observe whether the preparedness measures were adopted before or after any flood
event.
The national assessment of flood risk in England states that the second area after
London at highest risk of flooding with largest number of people living at risk is
Yorkshire and Humber region [33]. Yorkshire and Humber region has a long
history of flooding and flooding in 2007 caused record breaking disruptions in the
area. It was mainly caused by heavy rainfall and river overflows. An Environment
Agency data report released in November 2007 showed that number of businesses
flooded in the region was 3718 which is the highest in the entire country [34].
Therefore, this area was selected as the area of interest for the research. The four
worst-affected locations were identified in the region: Sheffield, Hull, Doncaster
and Wakefield. In both Sheffield and Hull more than 1000 commercial properties
were affected as a result of 2007 flooding. Sheffield was chosen as one of the case
study areas because of the historical evidence of higher frequency of flooding in
the area than Hull which was one of the essential factors for sampling area
selection. In Doncaster not enough businesses were flooded and most of its
vulnerable areas were residential in nature as compared to Wakefield. Therefore,
Wakefield was chosen as the second case study location suitable for this study.
The primary unit of analysis for the research are commercial property occupiers.
There was no readily available data set of the members of this target population
that have been affected by direct or indirect sources of flooding from which a
sample population could be selected. Therefore a sample set was constructed from
a combination of different data sources. For example, available literature and flood
risk maps were relied upon to build a picture of the areas affected. It was therefore
difficult to determine the exact sample size relative to the target population since
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 167
Knowledge gained from the overall scenario indicated that out of the 69 flood-
affected responses 33 (48%) undertook some sort of preparatory measures and 36
(54%) did not engage in any of the given preparatory actions. Similarly, when the
type of preparedness measures implemented by prepared part of the sample
population were analysed it was apparent that they preferred easy to procure
temporary preparatory measures and fewer long term permanent solutions for risk
reduction. Other popular measures are Environment Agency flood warning and
property and business insurance (see Figure 1).
EA warning
15% 11% 13%
Other measures Property Insurance
Business disruption 9% 10% 10%
Business Insurance
plan
6% 5%
Emergency plan 8% Resilient fittings
0% 5%
8% 2% Temporary flood
Data backup 4% 15% installations
4% 4%
Alternative fuel Permanent flood
source installations
Alternative power Alternative location
source
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
168 Flood Recovery, Innovation and Response IV
Temporary flood installations were largely adopted which were not adequate
for higher magnitude of flooding. More than half (total 55%) of business occupiers
who adopted any sort of preparatory measures took up only one or two measures,
19% restricted themselves with 3 to 4 measures and 26% were prepared for
flooding with more than 4 different combination of measures. This shows that
although taking measures for flood risk reduction is not very prevalent among
occupiers there is certain group of business occupiers who have started preparing
for the inevitable. Flood experience can be seen as having significant impact on
the level of adoption of protection measures. An interesting pattern was observed
based on the responses from the population who were flooded once and more than
once in the study areas (Table 2).
One hundred percent (100%) of the people flooded more than twice have taken
up some preparatory measures, 88% of those flooded twice have at least one
measure, and 39% were prepared after only one event. The average number of
measures adopted by businesses did not show much variance based on the level of
experience. The range of number of adopted measures varied between 1.6 (flooded
once), 2 (flooded twice) and 1.8 (flooded more than twice). This is slightly
different outcome from the usual trend seen in the residential sector where it takes
more than two or three times for the flood-affected population to understand the
importance of mitigation [35]. In commercial sector it seems that those who decide
to undertake mitigation activities choose to do so in the light of fewer events.
Respondents were asked to rank between 1 and 5 (1-no cost incurred and 5 highly
expensive) the different factors that affect cost of recovery. Table 3 illustrates the
percentage of differential cost incurred by respondents based on their differential
ranking.
Disruption of sales was scored highest while employee compensation and legal
charges were among the lowest ranked factors. Other factors like clean up charges,
machinery and sales disruption, working hour loss and repairing ranked among the
next four most costly factors in terms of recovery. Out of 100% of total cost
incurred 62% of the total cost was incurred for indirect flood impacts. Therefore,
it is evident that the cost incurred on recovering from indirect sources of damage
was more dominating than its counterpart. Answering questions regarding
financing sources for recovery the responses were clearly dominated by two
sources of finances; self- finance and property insurance. Table 3 indicates how
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 169
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
170 Flood Recovery, Innovation and Response IV
Some businesses indicated that they can still feel the effects of flooding and never
recovered completely. This was around 10% of the flood-affected sample.
Therefore this might be possible that those businesses which could not recover
from the impacts of flooding were not represented in this data because they might
have closed or moved to another location. This is one of the drawbacks of self-
administered questionnaires to be fully explained, especially a questionnaire with
such great detail of information. It is interesting to notice that the preparedness
actions taken by businesses before flood event were mainly concentrated on
reducing direct damages, however data indicated that the impact of indirect effect
of flooding costs them more to recover. Therefore the insight gained from the
empirical analysis suggests that attention in reducing effects of indirect disruptions
and reducing impacts which originates offsite is also necessary.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 171
by losing more customers and work hour loss. Other factors such as repair inside
building and supply disruption, machinery repair, structural damage and recovery
services ranked 4th, 5th and 6th respectively.
References
[1] Y. Zhang, M. K. Lindell, and C. S. Prater, “Vulnerability of community
businesses to environmental disasters,” Disasters, vol. 33, no. 1, pp. 38–
57, 2009.
[2] Committee on Disaster Research and Social sciences, “Facing Hazards
and Disasters: Understanding Human Dimensions.” National Academy
Press., Washington D.C., 2006.
[3] K. Clemo, “Preparing for Climate Change: Insurance and Small
Business.,” Geneva Pap. Risk Insur. – Issues Pract., vol. 33, no. 1,
pp. 110–116, Jan. 2008.
[4] M. Pitt, “The Pitt Review – Learning Lessons from the 2007 floods,”
Cabinet office, London, 2008.
[5] G. Wedawatta, B. Ingirige and D. Proverbs, “Adaptation to flood risk: the
case of businesses in the UK,” in International conference on building
resilience, 2011.
[6] E. P. Evans, J. D. Simm, C. R. Thorne, N. W. Arnell, R. M. Ashley, T. M.
Hess, S. N. Lane, J. Morris, R. J. Nicholls, E. C. Penning-Rowsell, N. S.
Reynard, A. J. Saul, S. M. Tapsell, A. R. Watkinson, H. S. Wheater, “An
update of the foresight future flooding 2004 qualitative risk analysis,”
Cabinet Office, London, 2008.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
172 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 173
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
This page intentionally left blank
Flood Recovery, Innovation and Reponse IV 175
Abstract
In the face of increased flooding in the UK, it is becoming increasingly
important to understand the ways in which flood experience can affect
homeowners’ attitude towards taking precautionary measures to protect their
homes. This could include investing in flood adaptation measures to reduce
likely flood damage and hence exposure to flood risk. This research sought to
investigate, the level of awareness, implementation and the costs of resilience
measures, from those homeowners who had experienced flood damage to their
properties in 2007 summer flooding in England. A questionnaire survey was thus
employed to elicit the extent to which flood experience influenced the decision
to adopt flood resilient measures during reinstatement works. The findings
revealed that some 82% of houses inundated were returned to their pre-incident
condition i.e. with no improved resilience to future flooding. It was found that
the level of awareness of resilience measures among the respondents was high;
however, the level of implementation was quite low. Only 10% of those who
indicated that they invested resilience measure actually implemented a full
package of the measures. This shows that more needs to be done by flood risk
management stakeholders to encourage full uptake of resilience measures. Loss
adjusters and surveyors are better placed to advice homeowners of the potential
risk reduction measures, which can be implemented during reinstatement period.
Further, there is a need for policy development in the form of revising the current
Building Regulations for refurbishing or reinstating flood damaged buildings in
order to encourage the up-take of resilient reinstatement.
Keywords: flood adaptation, flood damage, flood experience, resilience
measures.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140151
176 Flood Recovery, Innovation and Response IV
1 Introduction
The cost of flood damage in the UK has risen significantly since 1998 [1].
Currently, in the UK, over 5.2 million properties and 2.4 million people are at
risk of flooding, and annual average damages are estimated to be more than
£1 billion [2]. However, climate change and the increasing urbanisation of our
societies are increasing flood risk [2, 3]. In particular, there now appears to be
clear evidence that climate change will lead to an increase in the frequency and
severity of extreme precipitation and other weather events [4]; for the UK, this
may well result in wetter and stormier winters [5]. As such, The “Foresight
Future Flooding” report raises the prospect of a 4–10-fold increase in coastal
flood risk by the 2080s as a result of sea level rise alone [2]. The UK
Government policy on flood management can be summed up by the strategy of
“Making space for Water” which combines the provision and maintenance of
engineered flood defences with the recognition that flooding can never be
prevented entirely [6].
The direct financial damages related to the flooding of residential properties
can be significant. Depending on flood depth, duration of flooding and property
types, it is estimated that the cost of flooding can range from £15,000 to over
£80,000 for a single residential property and its contents [7]. The impact of
flooding at an individual household level can also result in less direct,
insurance-related impacts [8, 9], with premiums and flood-related excesses
potentially increasing following a flood event and as a result of making
insurance claims [10, 11].
Whilst large scale flood defences can be effective in reducing widespread
flood risk, such developments are costly, both in terms of time and financial
resources. Consequently, cost benefit analysis does not always yield a favourable
result for large scale defence schemes, and the extensive flooding that has
recently occurred within the UK has strengthened calls for greater use of
adaptation measures [12, 13]. In the UK, such measures are generally classified
as resistance (measures to keep water out of properties) or resilience measures
(installed to reduce the damaged impact of flooding on the fabric of building)
[14]. Keeping water out is a natural desire of property owners but, it is not
always possible or cost effective to prevent flooding of property, especially when
the anticipated flood depth is up to 1000 mm. Resilience measures are often
preferred, allowing water into the property in the knowledge that preparations
have been taken to minimise the damage caused. These adaptation measures are
designed to achieve two important objectives: to limit the financial impact on the
flood victim or their insurer by reducing damage to contents and building fabric
and to reduce the time used to reinstate properties, thereby, allowing
communities to return to normality quickly in the aftermath of the flood event.
Research has shown that implementing adaptation measures during flood
recovery period can effectively reduce the cost of the measures [15].
Review of extant literature revealed that the uptake of resilience measures in
residential properties remains persistently low [15], with one study finding that
only 16% of households and 32% of small-medium enterprises (SMEs) in areas
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 177
of significant flood risk have taken practical steps to reduce their exposure to the
potential flood risk [16]. Common reasons for the low uptake of the measures
include underestimation of flood risk, a lack of understanding about flood
protection responsibilities and concerns over the costs and aesthetics of such
measures [17, 18].
This study explores, the level of awareness, implementation and the costs of
resilient reinstatement after 2007 summer flooding in England. Those
homeowners who had experienced flood damage to their properties were the
focus of the study. The 2007 summer flood event provides an interesting case
study as it was reported to be widespread, and the cost of reinstatement work
during the recovery process was the highest insurers had ever paid (prior
to 2007) in England on a single flood event. Adapting existing properties to
potential future flood risk can be achieved by investing either in resistance or
resilience measures, or the combination of the two measures. The focus of this
study is on resilience measures. The concept of resilient reinstatement and
specifications, which can be incorporated during the flood reinstatement process
and the costs of resilient reinstatement are discussed in this paper. The
concluding part of the paper outlines some of the lessons learned from the level
of awareness, implementation and the actual cost spent by those homeowners
who implemented one form of resilience measures as a result of reinstatement
work to their properties following the 2007 summer flood event.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
178 Flood Recovery, Innovation and Response IV
prevent water seeping into the fabric of a building. Other measures include re-
fitting electrical sockets and electricity meter boxes above the anticipated flood
levels. Despite the extra cost of these measures, it has been suggested that the
implementation of resilient measures will reduce the repair costs in the long-term
assuming repeat flooding [16]. Table 1, shows the most widely used and
recognised resilience measures/specifications.
Resilience measures
Replace timber floors with concrete and cover with tiles.
Replace carpet with ceramic tiles.
Replace chipboard/MDF kitchen and bathroom units with plastic equivalents or
stainless steel.
Replace gypsum plaster with more water-resistant material, such as lime plaster or
cement sand render.
Apply water resistant paint to walls.
Move service meters, boiler, and electrical points well above likely flood level.
Replace softwood timber skirting with plastic or hardwood and apply water resilience
paint.
Replace softwood door and window frames with water resilient alternative.
Replace mineral insulation with cell insulation.
Source: Joseph [7].
Previous research carried out on behalf of the ABI [15], revealed that, on
average, resilient reinstatement costs over 40% (£12,000) more than traditional
reinstatement. It was stressed that there are significant variations around this
40% average, both between house types (i.e. bungalow, block of flats, terraced,
semi-detached and detached houses) and within house types. Although, the
authors further reiterate that resilient reinstatement could costs as little as 15% or
as much as 70% more than traditional reinstatement [15]. The reasons for the
wide variation were; property owners’ individual preferences and different
approaches to reinstatement methods adopted by different surveyors, despite the
available guidance such as Proverbs and Soetanto [21]; Garvin et al. [22] and
PAS 64 [23]. Some resilient measures can be introduced on a cost neutral basis,
and therefore not all aspects of resilient reinstatement measures increase the cost
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 179
3 Research methodology
An extensive survey was undertaken among those homeowners, who
experienced flood damage to their properties in the summer 2007 flood event, in
order to gain a better understanding of their flood experiences; to investigate
their understanding of resilience reinstatement; and to examine their
responsiveness to resilient reinstatement while their properties were being
repaired. This contributed to the evidence base needed to inform the effective
promotion of resilient reinstatement during flood recovery period. The
investigation took the form of postal questionnaire surveys. The mix of
the targeted population, which comprises of young and elderly people, dictates
the postal approach instead of online method of questionnaire distribution. The
summer 2007 flood event in England was selected as the focus of the study. This
flood event was widespread and it affected much of the UK during June and July
2007 which followed the wettest-ever May since national records began in 1766
[12]. The survey was carried out in 2013 some 6 years after the event and was
designed to gather information in two key areas;
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
180 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 181
4 Research results
Detailed analysis of the dataset was carried out and is presented in this section.
Respondents were asked if they had experienced flood damage to their properties
before and after 2007. Most respondents (77%) had no previous flood experience
prior to the 2007 flood event; 16% reported that they had experienced one
previous damaging flood to their properties prior to the 2007 event; and
approximately 4% had been flooded twice and 3% had been flooded more than
twice. This information is important because it is anticipated that those
respondents who had been flooded more than once, are more likely to invest in
resilience measures. Research has shown that experience of flooding can be a
source of motivation to individuals to undertake precautionary measures against
future flooding [24]. Further, respondents were asked if they had experienced
further flood damage to their properties following the 2007 flood event. Some
91% of respondents did not experience a flood event after the 2007 summer
flood event. This means that only 9% of respondents had experienced further
flooding after 2007.
Despite the relatively high number of respondents being aware of (at least) one
form of resilience measure to protect their property, the results presented in
Figure 3, show that fewer people actually used the opportunity of the 2007 flood
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
182 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 183
Cell insulation 6%
Stainless steel kitchen units 5%
Plastic kitchen units 4%
Upvc doors 11% Series1
Plastic skirting 23%
Water resistant paint 25%
Tanking 2%
Gas meter above flood line 14%
Electrical socket above flood line 18%
Water resistant plaster 12%
Floor tile 20%
Concrete floor 25%
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
184 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 185
The last resort is that, if the level of uptake of resilience reinstatement after
flood event continues to be as low as what was revealed in this study, it is
recommended that, Government needs to revise the current Building Regulations
for refurbishing or reinstating flood damaged buildings in order to force the
up-take of resilient reinstatement after flood event.
References
[1] Association of British Insurers. Summer floods 2007: Learning the
lessons. London. Association of British Insurers. London, 2007.
[2] Evans, E. P., Ashley, R., Hall, J., Penning-Rowsell, E., Sayers, P., Thorne,
C. R. & Watkinson, A. Foresight. Future Flooding. 1 and 2 London, 2004.
[3] Office of Science and Technology Climate Change Science. Postnote No.
295. Parliamentary Office of Science and Technology, London, 2007.
[4] IPCC Managing the risks of extreme events and disasters to advance
climate change adaptation. A special report of working groups I and II of
the intergovernmental panel on climate change. Cambridge University
Press, Cambridge, 2012.
[5] UKCIP. UK Climate Projections. UK Climate Impacts Programme,
DEFRA. http://www.defra.gov.uk/publications/files/pb13274-uk-climate-
projections-090617.pdf, 2009.
[6] Rooke, D. The summer of storm. Water and Environment Magazine, (10),
pp. 8-9, 2007.
[7] Joseph, R. Development of a comprehensive quantification of the costs
and benefits of property level flood risk adaptation measures in England.
PhD Thesis. University of the West of England, Bristol, 2014.
[8] Ball T., Geddes A., Werritty A., Black A. & Easton A. Flood insurance
provision and affordability beyond the statement of principles: implication
for Scotland. CREW, University of Dundee, 2012.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
186 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 187
Abstract
A highly important but under researched impact of flood events is the long term
psychological effect of the distress and trauma caused by damage and losses
associated with repeated flooding of communities. As a part of the recovery
process responders need to consider flooded households and offer support to
mitigate against the stress of flooding. This research aims to consider how the
risk of repeat flooding and flood memory can affect the needs of communities
with respect to post disaster support. Previous research has identified a variety of
influencing factors that affect the prevalence of mental health disorders in the
aftermath of flooding. Using a structured literature review and novel conceptual
model this research examines the role of flood experience and memory in the
impact of flooding on mental health and the needs of flooded communities. It is
found that the memory of previous flooding can influence future outcomes in a
variety of ways, with some positive incentives towards actions that may result in
lower damages in future events. These actions, that affect future trauma, have the
potential to mitigate the impact of repeated flooding. Therefore appropriate post
disaster needs assessment should not only identify vulnerable individuals but
also take account of the risk of future flooding.
Keywords: flood memory, PTSD, Anxiety, frequent flooding, flood impact,
mental health, flood recovery.
1 Introduction
The impact of flooding on the physical and mental health and wellbeing of
communities can endure long after the loss and damage due to direct contact
with floodwater is repaired. Quite apart from the possible loss of life and
irreversible injury; studies have demonstrated that a variety of other physical
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140161
188 Flood Recovery, Innovation and Response IV
ailments and mental health issues can arise in the aftermath of a flood [1]. It is
clear that, for the UK, Europe and the majority of the developed world, the
mental health impacts of flooding are at least as important as the risks from
physical illness [1, 2]. However, the detailed level of understanding needed by
responders and agencies in order to provide appropriate support throughout the
disaster cycle is lacking [3–5]. Furthermore it is not entirely clear how the
resources available to offset the longer term effects of flooding on the
psychological resilience of individuals and communities should be directed.
Research has explored various factors that can influence the severity of
mental health impacts of flooding including flood characteristics [6, 7];
individual characteristics [8]; socio economic factors [6]; preparedness [9]; and
duration of reinstatement activities [10]. It also seems intuitive to suggest that
flood memory or experience of past flooding will have bearing on the severity of
mental health impacts. However research in this area is lacking and it has been
identified by the UK Health Protection Agency as an area in need of further
research [11].
Therefore this paper seeks to explore various factors, including flood memory
on the mental health and wellbeing of flood affected communities and
individuals. The eventual aim is to improve the understanding of mental health
consequences from repeated flood experience leading to improved provision of
support services and targeting of resources to those potentially most vulnerable
to future mental health problems as a result of flooding and flood risk [5].
2 Methodology
The research adopted an enquiry based qualitative approach through a structured
review of available literature on the basis of research questions designed to
answer the main research aim. Literature from the wider field of disaster
management was combined with flood specific research in order to address the
following research questions:
1. What are the main mental health problems caused by flooding?
2. What factors affect the prevalence and severity of mental health issues
in flooded communities?
3. How long does the impact of flooding on mental health endure and does
the memory of flooding affect mental health issues following flooding?
A keyword search of academic literature databases provided the majority of the
literature, recent publications were prioritised and the presence of several
overarching reviews was capitalised upon in order to optimise the coverage of
older and diverse literature. Over 80 studies were accessed directly but the pool
of background studies was far larger because of many wide ranging reviews.
A novel conceptual framework was then developed based on the available
evidence on the research questions. This illustrates the influencing factors, role
of memory and mitigating interventions in the context of communities at risk of
frequent and repeated flood events. The construct validity of the proposed
framework derives from the thorough nature of the qualitative enquiry.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 189
3 Research results
Studies investigating the impact of flooding on mental health span the disciplines
of Flood risk management, disaster management, public health, epidemiology,
environmental management, climate change and more. Findings from this
diverse knowledge base are structured below as they relate to the three research
questions. However the emphasis is on identifying lessons for appropriate needs
assessment rather than examining appropriate clinical diagnosis or treatment.
Models of health impacts in the literature include Few [16] and Tapsell et al. [5].
They suggest a list of factors that make a difference to the prevalence of mental
health issues post disaster. These can be grouped into pre-existing conditions,
impact of the stressor event and post event conditions and stresses [9]. However
the scale and direction of the influencing factors are not consistent across studies.
There are some confirmatory and strong results such as the conclusion that
low socio economic status relates to higher level of distress [17]. However, other
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
190 Flood Recovery, Innovation and Response IV
factors such as age display much more complex and conflicted relationships with
mental health problems. Furthermore the risk of development of severe mental
health issues, or PTSD, was found to be related to individuals with extreme
pre-existing conditions and the presence of extreme aspects in the stressor event.
So for examples survivors who may have a higher than typical risk for PTSD
include those with a history of trauma exposure; chronic illness; chronic social
problems; or other major life stressors such as single parenting [5]. Table 1
summarises the influencing factors identified in the literature review.
Vulnerability is related to gender. Tapsell et al. [5] observed that women and
girls exhibited stronger effects than men and boys in 42 out of 45 studies. The
effects were most marked in the study of PTSD and within traditional cultures.
The presence of strong spousal support in mitigating stress was also less helpful
for women than men with women apparently burdened by close social ties [5].
The effect of age as an influencing factor is more complex. While it is clear
that age contributes to physical challenges that result in increased physical health
impacts, injury and mortality [18], conflicting results are reported in mental
health studies in the disaster field [19]. For example, in Korea younger people
(under 45) were found to have most symptoms after a flood event [20] and other
authors have found similar patterns [21] possibly due to the older generation’s
coping strategies [22]. However the protective influence of older age was not
observed in Vietnam [23]. Some studies have demonstrated that middle aged
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 191
adults are the most prone to mental health problems after disasters [12]. Tapsell
et al. [5] suggest an explanation for this is that flooding adds to the greater
responsibilities they already face. Children as a distinct group have also been
studied, but no conclusive evidence demonstrates whether children are more or
less likely to display mental health impacts [5, 24]. Age as a factor is highly
confounded with other life stage, physical abilities and stressor variables to the
extent that any true impact of chronological age can rarely be established.
The available research on culture and ethnicity shows ethnic minorities are
more vulnerable to disasters in general possibly due to social deprivation and
marginalisation [12, 25]. The impact of cultural expectations may also have an
influence on the tendency of an individual to seek help [26, 27].
Evidence that implicates the flood severity in mental health disorders strongly
suggests this is related to direct, indirect, tangible and intangible losses [13].
These findings also seem to hold true for disasters generally [5, 12, 28].
Indicators include the number of casualties, deaths, losses and disease. For
example in Thailand severe flooding quadrupled the incidence of PTSD
symptoms whereas in Korea risk of PTSD and depression were influenced by
injury, death of a relative and damage [14]. However, the categorisation of
severity differs across studies and a full diagnostic would be difficult to establish
without considerable further research [9, 29].
On an individual level, pre-existing health conditions and personality factors
are good predictors of post-disaster mental health problems for a given disaster
severity [12]. This has also been demonstrated specifically for flood events [5].
After a flood the stress associated with lack of basic services, evacuation and
poor living conditions, can also damage mental health [4, 13, 30]. Events seen to
be accidents cause less distress than those seen as preventable [5].
Disaster studies that focus on aspects of the stressor have found that
psychological impacts are most likely when at least two severe event factors are
present: Extreme and widespread property damage; serious financial hardship;
human causes for the disaster; high levels of injury and deaths [12]. However
research has also demonstrated differences between different types of mental
health issues with PTSD related more to event stressors and depression
associated with both event stressors and life stressors [9].
It is clear from the above discussion that it is necessary to consider multiple
influencing factors relating to the flood and the population. However research
has tended to focus on factors in isolation, rather than investigate interactions
between multiple stressors and characteristics rendering the evidence indicative
rather than predictive. Recognising these factors may nevertheless help to
identify those most at risk of developing PTSD, anxiety and depression, however
a deeper understanding of the interactions may allow disaster managers and
health professionals to offer specifically targeted support.
Duration of impact and flood memory are linked because remembering a flood
will have an influence on the length of psychological symptoms. While treatment
pathways are outside the scope of the research the existing evidence does allow
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
192 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 193
Mental health consequences of flood events have not been fully addressed in the
past either in disaster or health fields. But the purpose of this present review is
not to predict the required mitigating actions in advance of flooding. Few [16]
and Tapsell et al. [5] have pointed out the difficulty and futility in making this
attempt. It is important to avoid pathologising a natural reaction to trauma and
offer appropriate post disaster support that can detect those individuals that may
be more vulnerable and likely to develop more severe and longer term symptoms
[17]. In this respect the memory and frequency of flooding could be instrumental
in setting up conditions that could trigger higher levels of emotional distress.
Therefore it is relevant to consider what actions can be taken in recently flooded
locations to mitigate against the impact of a second flood.
We can deduce that mitigation of mental health impacts for a given individual
with fixed personality and socio economic conditions for the next flood might be
achieved through one of the following approaches:
The first approach is to assess what steps could be useful in preventing the
development of psychological problems after an event. The model suggests
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
194 Flood Recovery, Innovation and Response IV
helpful interventions might be: support for faster reinstatement particularly for
those without insurance; sympathetic insurance and reinstatement professionals;
good advice on coping strategies that help individuals forget or assimilate the
trauma of their experiences; strengthening of community networks and other
social support; and provision of counselling and direct mental health support.
The second approach reduces future vulnerability to flooding through taking
steps to increase the capacity and resilience of people and the built environment
[39]. Helpful interventions might include: resilient reinstatement of buildings
[40], disaster preparedness training; and warning systems. Indeed a whole range
of flood risk management measures can be used to lower the risk of physical
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 195
damage and loss [41]. The resilience of the population at risk can also be
enhanced through many means including: the provision of peace of mind via
insurance; good advice on coping styles that enable rational actions while
reducing trauma; preparedness training; improving general wellbeing; and
boosting community cohesion [39].
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
196 Flood Recovery, Innovation and Response IV
for successive floods. Furthermore, action to limit the severity of flooding in the
successive flood will contribute to lower levels of trauma and therefore may lead
to improved mental health outcomes.
Coping strategies need to be explored. Those strategies appropriate to short
term recovery following a single flood may be different from those appropriate
for those at risk from frequent and repeated flooding. However an initially
detached coping style may enable a more rational consideration of the options
available to act.
Finally the results show that future research in this field may benefit from a
multi-dimensional approach to measurement of impacts and further
consideration of the complex relationship between concepts of flood memory
and mental health. The formulation of a conceptual model of the impacts of
flood memory on mental distress following repeated flooding will be helpful in
deriving appropriate multi-dimensional research designs that include the effect of
previous flooding.
Acknowledgement
This research was funded by the UK Engineering and Physical Sciences
Research Council under grant EP/K013513/1 Flood MEMORY: Multi-Event
Modelling Of Risk & recoverY.
References
[1] B. Carroll, R. Balogh, H. Morbey, and G. Araoz, “Health and social
impacts of a flood disaster: responding to needs and implications for
practice,” Disasters, vol. 34, pp. 1045–1063, 2010.
[2] L. Fewtrell and D. Kay, “An attempt to quantify the health impacts of
flooding in the UK using an urban case study,” Public Health, vol. 122,
pp. 446–451, 2008.
[3] S. Hajat, K. L. Ebi, S. Kovats, B. Menne, S. Edwards, and A. Haines, “The
human health consequences of flooding in Europe and the implications for
public health: a review of the evidence,” Applied Environmental Science
and Public Health, vol. 1, pp. 13–21, 2003.
[4] V. Mason, H. Andrews, and D. Upton, “The psychological impact of
exposure to floods,” Psychology, Health & Medicine, vol. 15, pp. 61–73,
2010/01/01 2010.
[5] S. M. Tapsell, S. M. Tunstall, and S. Priest, “Developing a conceptual
model of flood impacts upon human health,” FloodSite Report T10-09-02,
2009.
[6] T. W. Collins, A. M. Jimenez, and S. E. Grineski, “Hispanic Health
Disparities After a Flood Disaster: Results of a Population-Based Survey
of Individuals Experiencing Home Site Damage in El Paso (Texas, USA),”
Journal of Immigrant and Minority Health, vol. 15, pp. 415–426,
2013/04/01 2013.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 197
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
198 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 199
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
This page intentionally left blank
Flood Recovery, Innovation and Reponse IV 201
Abstract
During the past 30 years, floods have resulted in over 200,000 fatalities and
affected more than 2.8 billion others worldwide. Flood victims are vulnerable to
long-term physical and psychological health effects, which persist for an
undefined time period in the aftermath of a flood event. Following a flood event,
secondary stressors, which are indirectly related to the event, can potentially
prolong and intensify the health impacts on affected individuals and communities.
These secondary stressors consist of economic stressors, including loss of income,
but also social stressors such as isolation due to prolonged flooding. A significant
gap in the research to date is in relation to repeated flooding and its impact on the
extent to which individuals are affected by these secondary stressors. This review
examined studies focusing on repeated flooding, concentrating on the secondary
stressors resulting from repeated flood events. It also considered the awareness,
preparedness and resilience of the study populations in order to determine the
potential for these communities to be impacted by secondary stressors. This review
indicated that both rural and urban communities in developed and developing
countries are significantly affected by economic, social and psychological
secondary stressors. The majority of communities do have a basic awareness of
flood risk; however, many residents do not take flood risk seriously and thus take
little preventative action. Community resilience was higher in urban and rural
areas in developing countries, but also in rural areas in developed countries. Future
work should take into consideration the secondary stressors that affect different
communities and how to minimise their impact in order to increase resilience.
Keywords: flooding, repeated flooding, health, secondary stressors, resilience,
awareness, preparedness, urban, rural.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140171
202 Flood Recovery, Innovation and Response IV
1 Introduction
Flooding has become the most frequent type of major disaster globally within both
developing and developed countries [1]. The World Health Organisation
concluded that during the past 30 years, flooding resulted in over 200 000 fatalities
and affected more than 2.8 billion others worldwide [2]. The statistics illustrate
that flooding is a worldwide phenomenon and an unquestionable cross border
issue.
Flooding poses multiple risks to health and growing evidence worldwide
indicates that the health impacts of flooding penetrate a lot deeper than the
immediate physical impacts such as injuries and drowning [3]. An increasing
recognition is that following extreme events such as floods, secondary stressors,
which are indirectly related to the event, can potentially prolong and intensify the
health impacts on affected individuals and communities [4]. These secondary
stressors take in economic stressors including the impact on property values, but
also social stressors such as forced isolation due to a prolonged flood event.
Although numerous secondary stressors of extreme events have been identified,
there remains a need to establish whether repeated flooding has an impact on the
extent to which communities are affected by these stressors and also to investigate
if these stressors have similar impacts on different types of communities, such as
urban and rural areas.
Therefore the aims of this paper are:
1. To identify studies which have examined the long-term impact of
repeated flooding.
2. To summarise and critically review the published literature to date on the
secondary stressors impacting urban and rural flooded populations.
3. To establish the awareness, preparedness and level of resilience of
communities which have suffered repetitive flooding in order to
determine their vulnerability to secondary stressors.
4. To determine the knowledge gaps in the research relating to the long-
term health impact of repeated flooding on communities.
2 Methods
A literature search using Proquest, Science Direct, Medline and Web of Science
was conducted. The search was limited to peer-reviewed articles published in
English. Table 1 outlines the search strategy that was used to identify studies to be
included in the paper. It included a combination of key words relating to exposure,
health outcomes, susceptibility to flooding and the location of the flood event.
Studies were eliminated that did not focus on the health impacts of flood events or
secondary stressors of flood events. It was also decided to disregard articles which
focused solely on immediate impacts of flooding such as mortality, diarrhoeal
diseases etc. In addition, papers were excluded that addressed only the health
impacts of single flood events. After the search strategy was executed it was
decided to discount studies where the sample population did not permanently
reside in the area that had been repeatedly flooded i.e. studies on caravan sites.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 203
The full texts of the remaining articles which met the inclusion criteria were then
critically reviewed by the first author, the key findings of which are summarised
in this paper.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
204 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 205
3.1.1.3 Insurance and house prices Following a flood event, the key priority for
homeowners is to restore their property to its pre-flood state. However, with the
increasing prevalence of flooding, a rising problem is obtaining flood insurance
[17, 23, 24]. A UK study by Lamond et al. noted that 13% of respondents were
refused a quote for insurance due to flood risk and 3% were denied a renewal due
to flood risk [24].
In developed countries the increased prevalence of flooding has led to the need
for homeowners to invest in flood insurance. Flood insurance is often non-existent
in developing countries and flood victims have to use their savings to cover the
financial costs. Obtaining insurance for flood risk areas is now very difficult in
developed countries and although the majority of homeowners do eventually
obtain insurance, the strain on individuals already traumatised by other flooding
related stressors should not be underestimated.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
206 Flood Recovery, Innovation and Response IV
9]. This is particularly difficult for young children and the elderly who may not
understand the situation and feel anxious away from normal surroundings and
belongings.
Repeated flooding can eventually leave residents with only one option, that of
relocation. A study in New Orleans found that 23% of respondents were
considering, trying or actually selling their home [25]. Recurrent damage to
property or successive flooding resulting in damage to crops can encourage
migration [14]. Migration, although a means of removing exposure to repeated
flooding, can be a secondary stressor in itself as displacement from familiar
surroundings and severing of ties from family and neighbours can be extremely
distressful [5, 17, 24]. In many circumstances, flood victims wish to remain in
their homes despite the high flood risk, due to fear of the unknown. A study in the
USA found that while 190 households accepted the relocation offer, 47 rejected it
[22]. However, often the constant fear of reoccurring flooding can become too
stressful, leaving migration as the only alternative.
It is important to recognise that when the decision to migrate is eventually
made, the stress does not necessarily come to an end, due to financial issues such
as problems with selling homes or higher house prices in non-flood risk areas [5,
15]. A study in Alaska identified that the main obstacle to migration was
government funding, as the cost of relocating the village was too high [22]. In
addition, home owners are not always offered what they consider the true value of
their property, leading to added stress when deciding whether to agree to
government resettlement [15]. Property owners who perceived the condition of
their property as higher or were attached to their neighbourhood found it more
difficult to relocate, however, fear of future flooding made acceptance easier [16].
Dun identified that when government resettlement occurs, households are
provided with a five year loan to buy a housing plot; thus resettlement can actually
become an economic stressor, causing individuals to go into debt [14].
Temporary evacuation is common in both rural and urban areas. It is
unavoidable in both developed and developing countries in order to protect the
health of flood victims. However, this study has found that migration is generally
a rural issue, as individuals decide to move permanently to urban areas where they
feel that flood risk is a lesser concern. It is essential to recognise that while
migration and resettlement reduces physical exposure to the health impacts of
repeated flooding, it can increase economic and social vulnerabilities; therefore
migration can act as both a solution and a stressor.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 207
3.1.3.2 Stress, anxiety and depression Repeated flood events place individuals
and communities under severe stress and anxiety due to both social and economic
reasons [18, 19, 24, 27]. Stress and anxiety can sometimes have a significant
impact on human behaviour. Biswas et al. found that 70% of mothers and 40% of
fathers abused their children during flood events [18]. The possible long-term
health effects of abuse at an early age are widely recognised, including a tendency
towards eating disorders and depression [18]. It is important to recognise that the
extreme stress and anxiety placed on recurrent flood victims has the potential to
lead to mental health problems. In a 2013 study on repeated flood events, Wind et
al. found that mental health symptoms, such as anxiety and depression were
notably higher than other studies on natural disasters, suggesting that repeated
events have an even greater influence on mental health [27].
Stress, anxiety and depression due to flooding were identified in both urban
and rural studies in this review. Individuals from developing countries are
particularly susceptible as they have limited resources to recover from flooding.
Stress, anxiety and depression generally occur as a result of other secondary
stressors such as damage to property. It is therefore crucial to acquire a greater
knowledge of the impact of flood related secondary stressors to minimise the
mental health impacts of recurrent flooding.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
208 Flood Recovery, Innovation and Response IV
This review established that lack of preparedness for flooding was a significant
problem [6, 7, 10, 17, 20, 28, 29]. This problem was evident in both urban and
rural and developing and developed countries. An additional issue was that the
majority of preparation made was reactive rather than proactive [6, 10, 11]. The
key reasons were lack of awareness regarding mitigation measures, lack of
financial resources to invest in preventative measures, lack of ‘know-how’ and a
limited time frame between successive flood events [5–7, 11, 13, 15, 20, 29].
Several of the studies also recognised that often residents thought they were
prepared for flooding, but in reality they had not taken sufficient mitigation action
or their insurance did not cover flood events [17, 23, 24]. Furthermore, some
residents, despite being aware of future flood risk, chose to ignore it, believing that
it would not reoccur [16]. In developed countries, homeowners and business
owners are often dismissive of flood preparedness, suggesting it is entirely the
government’s responsibility [17, 23].
A key finding of this review was that the majority of flood victims who were
flood prepared resided in developing countries [5, 9, 11–13, 15, 19, 21]. However,
two of the studies in developed countries also illustrated flood preparedness [24,
25]. A potential explanation is that in developing countries flood victims see
mitigation and preventative action as their own responsibility rather than that of
government agencies, while flood victims in developed countries have a higher
expectation in relation to government assistance. Additionally Mishra et al. found
that place attachment significantly influenced flood preparedness, indicating that
residents who had lived in the same location for generations would be more likely
to take mitigation action in order to protect their heritage [28].
The review indicates that although a basic level of flood awareness did exist in
all the studies, there was a lack of awareness relating to mitigation measures and
the health implications of flooding. Lack of preparedness was a major issue of
concern in both developed and developing countries, however, flood victims in
developing countries, in both urban and rural areas were more likely to be flood
prepared due to their lack of dependence on agencies. The findings of the review
suggest that communities are very vulnerable to being affected by secondary
stressors due to their lack of flood preparedness. It is important to raise awareness
concerning the ramifications in communities if they do not prepare for flooding,
emphasising that lack of awareness and preparedness for flood events can escalate
the long-term impact of secondary stressors on health [10].
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 209
4 Conclusion
This review concluded that both rural and urban communities in developed and
developing countries are significantly affected by secondary stressors. We
identified that loss of income and damage to property had an almost universal
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
210 Flood Recovery, Innovation and Response IV
References
[1] Ahern, M., Kovats, R.S., Wilkinson, P., Few, R. and Matthies, F., Global
health impacts of floods: epidemiologic evidence. Epidemiologic reviews, 27
(1), pp. 36–46, 2005.
[2] World Health Organisation. Floods in the WHO European Region: Health
effects and their prevention. WHO: Copenhagen, 2013.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 211
[3] Few, R. and Matthies, F., Flood hazards and health: responding to present
and future risks, Earthscan: London, 2006.
[4] Lock, S., Rubin, G.J., Murray, V., Rogers, M.B., Amlôt, R. and Williams,
R., Secondary stressors and extreme events and disasters: a systematic
review of primary research from 2010–2011. PLoS currents, 4, 2012.
[5] Zahari, R.K. and Ariffin, R.N.R., Community-Based Disaster Management
in Kuala Lumpur. Procedia-Social and Behavioral Sciences, 85, pp. 493–
501, 2013.
[6] Hoggart, S., Hanley, M., Parker, D., Simmonds, D., Bilton, D., Filipova-
Marinova, M., Franklin, E., Kotsev, I., Penning-Rowsell, E. and Rundle, S.,
The consequences of doing nothing: The effects of seawater flooding on
coastal zones. Coastal Engineering, 87, pp. 169–182, 2014.
[7] Braun, B. and Aßheuer, T., Floods in megacity environments: vulnerability
and coping strategies of slum dwellers in Dhaka/Bangladesh. Natural
Hazards, 58 (2), pp. 771–787. 2011.
[8] Boon, H.J., Disaster resilience in a flood-impacted rural Australian
town. Natural Hazards, 71 (1), pp. 683–701, 2014.
[9] Paul, S.K. and Routray, J.K., Flood proneness and coping strategies: the
experiences of two villages in Bangladesh. Disasters, 34 (2), pp. 489–508,
2010.
[10] Tschakert, P., Sagoe, R., Ofori-Darko, G. and Codjoe, S.N., Floods in the
Sahel: an analysis of anomalies, memory, and anticipatory learning. Climatic
Change, 103 (3-4), pp. 471–502, 2010.
[11] Shimi, A.C., Parvin, G.A., Biswas, C. and Shaw, R., Impact and adaptation
to flood: A focus on water supply, sanitation and health problems of rural
community in Bangladesh. Disaster Prevention and Management, 19 (3), pp.
298–313, 2010.
[12] Mavhura, E., Manyena, S.B., Collins, A.E. and Manatsa, D., Indigenous
knowledge, coping strategies and resilience to floods in Muzarabani,
Zimbabwe. International Journal of Disaster Risk Reduction, 5 pp. 38–48,
2013.
[13] Nguyen, K.V. and James, H., Measuring Household Resilience to Floods: a
Case Study in the Vietnamese Mekong River Delta. Ecology & Society, 18
(3), 2013.
[14] Dun, O., Migration and displacement triggered by floods in the Mekong
Delta. International Migration, 49 (1), pp. 200–223, 2011.
[15] Chhotray, V. and Few, R., Post-disaster recovery and ongoing vulnerability:
ten years after the super-cyclone of 1999 in Orissa, India. Global
Environmental Change, 22 (3), pp. 695–702, 2012.
[16] Kick, E.L., Fraser, J.C., Fulkerson, G.M., McKinney, L.A. and De Vries,
D.H., Repetitive flood victims and acceptance of FEMA mitigation offers:
an analysis with community–system policy implications. Disasters, 35 (3),
pp. 510–539, 2011.
[17] Keogh, D.U., Apan, A., Mushtaq, S., King, D. and Thomas, M., Resilience,
vulnerability and adaptive capacity of an inland rural town prone to flooding:
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
212 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Section 6
State-of-the-art on flooding
damage survey and assessment
(Special session
organised by D. Molinari)
This page intentionally left blank
Flood Recovery, Innovation and Reponse IV 215
Abstract
The aim of this paper is to present a “procedure” to collect and store damage data
in the aftermath of flood events. The activity is performed within the
Poli_RISPOSTA project (stRumentI per la protezione civile a Supporto delle
POpolazioni nel poST Alluvione), an internal project of Politecnico di Milano
whose aim is to supply tools supporting Civil Protection Authorities in dealing
with flood emergency. Specifically, the aim of this paper is to discuss the present
implementation of the project, highlighting challenges for data collection,
storage, analysis and visualisation. Data can have different formats (e.g. paper
based vs. digital form, different digital files extensions), refer to different aspects
of the phenomenon (i.e. hazard, exposure, vulnerability and damage), refer to
different spatial and temporal scales (e.g. micro vs. meso scale, different phases
of the flood event) and come from different sources (e.g. local authorities, field
surveys, crowdsourcing). Therefore a multidisciplinary approach which includes
expertise from ICT, geomatics, engineering, urban planning, economy, etc. is
required. This paper first describes a conceptual map of the issue at stake, then it
discusses the state of the art of the implementation, taken as reference the
Umbria flood in November 2012. Impacts of the project are discussed with
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140181
216 Flood Recovery, Innovation and Response IV
1 Introduction
In recent years, awareness of a need for more effective disaster data collection,
storage, and sharing of analyses has developed in many parts of the world, also
in the wake of several policies that, at different levels of government, implicitly
or explicitly required to face the problem at stake (e.g. the Hyogo framework for
Action [1], the EU disaster prevention framework [2], the European Union
Solidarity Fund [3], the Green Paper on Insurance of Natural and Man-made
Disasters [4]).
Among natural disasters, this paper focuses on floods. Having more reliable
data on flood impacts is of paramount importance for improving pre and post
event risk reduction strategies. For instance De Groeve et al. [5] suggest three
application areas for (flood) loss data: loss accounting, disaster forensics and risk
modelling. In the aftermath of flood events, the principal motivation for
recording the impacts of floods is loss accounting. This information is crucial at
different levels of governance/risk management. At the local level, civil
protection and policy makers (i.e. mayors) need loss accounting in order to
identify priorities for the emergency and the recovery-reconstruction phases
while insurers use this information to compensate victims. At the sub-
national/national level, loss accounting is required by policy makers for fund
allocation, for addressing damage compensation and recovery. At the
international level the interest is on financial and humanitarian aid.
In peace time, flood loss data are required to improve knowledge of the
mechanisms leading to flood impacts; to analyse the causes of disasters through
measuring relative contribution of hazard, exposure, vulnerability and coping
capacity (i.e. the response to the flood). This is what is called disaster forensic.
The objective of disaster forensic is twofold: (i) to enhance disaster management
from lessons learnt, and (ii) to improve risk mitigation strategies by increasing
the capacity of modelling and forecasting flood damage.
Within this context, this paper presents the Poli-RISPOSTA project
(stRumentI per la protezione civile a Supporto delle POpolazioni nel poST
Alluvione), an internal project of Politecnico di Milano supporting
interdisciplinary research with a direct impact on the society. The main intention
of Poli-RISPOSTA is to build with and for the Civil Protection (CP) a model,
tools and advanced technical solutions for collecting, mapping and evaluating
post-flood damage data. In fact, as the consequence of the policies discussed
above, the need for enhanced methods and procedures for post-event damage
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 217
2 Problem conceptualisation
The general objective of Poli-RISPOSTA can be identified in the development of
a “complete” flood scenario describing both the physical features of the forcing
event (i.e. the flood) as well as its impacts and the capacity of societies to face
them. In order to accomplish with risk mitigation objectives (e.g. those imposed
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
218 Flood Recovery, Innovation and Response IV
by the EU Floods Directive) such a scenario must be developed both ex-ante and
ex-post. Before an event occurs “complete” scenarios provide a picture of the
most risky areas and allow identifying strategies to mitigate risk and to cope with
hazardous events. After the event occurrence, the objective is instead to figure
out real impacts and to identify priorities for the emergency and the recovery
phase (fig. 1). A comparison of the ex-ante and ex-post scenarios allows finally
to infer lessons towards an improvement of both the capacity of predicting the
event (and its consequences) and to cope with it.
To achieve these goals, tools and advanced technical solutions to collect,
store, analyse and represent a multitude of data must be developed within Poli-
RISPOSTA (fig. 1). After an event occurs, such data regard both physical effects
of the forcing event (as flooded areas, water depth and velocity inside it, the
occurrence and localisation of landslides, etc.) and observed damages on the
different sectors of the society (i.e. people, economic, and human activities),
the natural and built environment (i.e. residential and industrial buildings,
infrastructures, public and cultural heritage, ecosystems). Damages can be due to
the physical contact of the flooding water (i.e. direct damages) or induced by the
first (i.e. indirect damages); both (ii) tangible (i.e. monetary) and intangible data
must be taken into account. Moreover, data on mitigation actions implemented
by emergency services and lay people before and during the flood is of interest
as these actions influence both physical effects and damages.
Before the event, data regard instead results from hazard, exposure and
vulnerability modelling. Also in this case, information must be managed with
respect to the different variables characterising the physical scenario as well as
required to estimate risk on the different items potentially affected by the floods.
PEACE TIME
Exposure Vulnerability Hazard modeling
modeling modeling
EMERGENCY/RECOVERY
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 219
Flood
analysts
Flood
Flood analysts/
analysts ICT experts
The complexity of the problem at stake implies several challenges for Poli-
RISPOSTA, with respect to the current state of the art. In the following the most
relevant are discussed.
First, tools for systematic loss accounting are not very well developed. The
way in which flood damage data are presently collected and stored implies
several problems for an efficient, multipurpose use of data as wished in [5] and
[6]. The main problem of existing disaster databases concerns data comparison
and management. This is due to a lack of agreed standard to collect and store
damage data. Specifically, several differences can be found in existing databases
regarding:
- recorded losses. This depends on: (i) the intent of the reporting activity (i.e.
insurance companies, governmental agencies and NGOs collect data for
different purposes; for this reason, flood loss records are often not
representatives of the real impact of floods as they focus only on certain items
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
220 Flood Recovery, Innovation and Response IV
at risk and/or types of damage), (ii) the time of reporting, and (iii) present
capacity of estimating all types of damages (e.g. indirect or secondary damages
are not so evident in the aftermath of an event and are difficult to evaluate in
monetary terms).
- The scales of reporting. Flood loss data can be recorded at different spatial and
temporal scales, according to the intent of the report and to who is leading the
reporting activity (i.e. their role and responsibility). However,
aggregating/disaggregating damage over space and time is not straightforward.
- The economic rationale. There are different methods to evaluate monetary loss,
e.g. taking into account inflation (i.e. depreciated value), purchasing parity (i.e.
replacement value), insured losses, etc.
One of the main challenges Poli-RISPOSTA has to face is to develop tools
for the survey and collection of flood loss data that overcome the above limits,
guaranteeing high quality, consistent and reliable data, in the philosophy that
“the quality of disaster databases can only be as good as the reporting system”
[7]. Contrary to common practice, Poli-RISPOSTA wants to work at the local
level in order to meet two basic requirements of flood loss data: (i) going into
details of phenomena/aspects leading to damage and (ii) reporting all the events,
including small ones (like multi spot flash floods in mountain regions) which are
presently discounted by national/international databases [8][9]. Data at upper
levels, for strategic and policy making purpose, can be obtained in a second step
by proper aggregation rules. On the other hand, Poli-RISPOSTA wants to
provide a “complete picture” of a disaster, identifying damage to various sectors
of the economy and society. From this perspective, the PDNA - Post Disaster
Needs Assessment methodology resulting from the collaboration of a number of
institutions, including the EU Commission, United Nations, the World Bank and
others is a very important example (for an application see [10]).
Linked to the previous point is the development of technological solutions for
data acquisition. Indeed, while damage data at the meso or macro scale can be
inferred from indirect sources (e.g. public accounting, researches, newspapers,
and regulations), local data are often collected by means of field surveys. Tools
should then be developed in order to support data survey in digital format. Such
tools should provide real time data storage (in a database) and their visualisation
in terms of maps, supporting this way the field survey/emergency phase (e.g.
supporting the coordination of survey team). With respect to this, the DARMsys
developed by the Queensland Reconstruction Authority in Australia can be taken
as reference [11].
The need of managing collected data also in terms of visualisation and spatial
analysis represents another challenge for Poli-RISPOSTA. Since there is not a
standardized way to collect spatial data in the case of floods, data collected in the
aftermath of flood events are commonly in different formats that make data not
immediately usable for spatial analysis. This is not the case, e.g., in ex-ante risk
assessments where data of interest are directly produced to be handled by GIS
tools [12]. Creating flood databases is common practice, but not always in
GIS standard compatible formats. This point must be addressed by Poli-
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 221
RISPOSTA, providing tools for data storage which also satisfy requirements of
data visualisation and spatial analysis.
Last but not least, as any technological problem, an interdisciplinary approach
is required by the project that put together computer scientists and domain
experts.
With respect to this, the further challenge of Poli-RISPOSTA is that not only
domain experts come from different disciplines but also computer scientists are
heterogeneous. As regard domain experts, the need of analysing both the
physical feature of the event as well as its consequences (both in monetary terms
and with respect to intangible damages) implies expertise from engineering,
urban planning, sociology, economy, etc. With respect to computer sciences ICT
experts are required for the collection, storage and management of data of
interest. Moreover, the need of representing data in terms of maps as well as of
carrying out spatial analyses (see section 2) requires involving experts from
geomatics.
According to the cyclic process described in fig. 2, the first step in designing the
IS consists of an analysis concerning both data characteristics and types of
required elaboration to be performed. This analysis was carried out on the basis
of the flood event occurred in the Umbria Region – Central Italy in November
2012 [13]. On that occasion the regional CP asked Politecnico di Milano to
develop a report (under construction) describing the event and its consequences
at the regional level. Researchers activity focused on two aspects: (i) the
development of an ex-post scenario to help CP to figure out event impacts [14],
to identify priorities for recovery and reconstruction and to verify effectiveness
of emergency plans; (ii) the development of an ex-ante scenario to be compared
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
222 Flood Recovery, Innovation and Response IV
with the first to verify whether or not existing risk assessment and mitigation
strategies are suitable to deal with flood risk in Umbria. Indeed, results from this
experience are the starting point of Poli-RISPOSTA, allowing both to recognise
needs and requirements in terms of data analysis and elaboration for the scenario
development and to analyse the features of (available or required) data on which
such activity should be performed (see section 3.2).
With respect of data features, besides the fact that they refer to the different
domains discussed in section 2 (i.e. risk modelling, observed physical effects and
damage on the different sectors, mitigation actions), other important features
were recognised which imply requirements for the IS (see section 3.2);
availability in time is one of them. Data are available at different times. This is
due, on the one hand, to the nature of the data itself; for example, modelling data
are available before an event occurs, indirect damages (e.g. disruption of
economic activities, of basic services to the population, the loss of rental income)
are not evident in the aftermath of an event but some months later, etc. On the
other hand, norms regulating damage compensations count. The latter identify
which damages are refunded by law and which are the deadlines to ask for
compensation; for this reason, both public and private subjects give priority to
determine reimbursable damages while other types of losses are assessed in a
second step (e.g. in Italy damage to infrastructures must be declared by regional
authorities 20 days after the event while damage to residential buildings 90 days
after). However, generally speaking, data of interest are available before the
event or after the event, in a time window ranging from few days to 1 year.
Another important feature is the spatial scale; data can refer to individual
objects (e.g. damage to a building, a bridge, a fabric), the local scale (e.g.
number of evacuees in a municipality), the large scale (e.g. traffic disruption at
the province, flood zones in the river basin) or to the regional
/national/international scale (e.g. indirect damage to ecosystems). The Euclidian
dimension is also linked to this feature and to the need of representing data
(analyses) in terms of maps (see below). Data can be represented as points (e.g.
the damage to a building), lines (e.g. length of damaged roads) or areas
(e.g. flooded areas).
The source of data is another important aspect. Some data are acquired by
means of field surveys (e.g. direct damage to buildings, water depth inside the
flooded area) for which suitable tools should be designed (see sections 3 and 4).
Other data are directly produced by the CP (e.g. flood forecasts); finally several
data are recorded by other subjects (e.g. local authorities, service suppliers,
research centres) and must be “simply” collected by the CP.
The present data format is an additional characteristic to be taken into
account. Coming from different sources, data can have different formats: papery
based or digital. In the second case, recognised formats are heterogeneous:
features, texts, spreadsheets, images or multimedia.
Last but not least, not only quantitative (e.g. observed damage) but also
qualitative data (e.g. vulnerability features, emergency actions) are of interest.
In order to reproduce the complete event scenarios, the experience in the
Umbria region highlighted that data of interest not only are heterogeneous but
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 223
1. Temporal tracking and storage of data: some data are available before the
event, some are gathered during, other are collected after it; thus the system
should allow to store and manage data collected at different times. As some
information might deteriorate quickly after a flood event (e.g. the level of water,
the memories of people affected by the event, etc.), the system must recall users
on data to be recorded at each time. Moreover, for some data the interest is in
keeping their history over time (that is, the way they change over time), for
others only their current (or most updated) status needs to be known. Criteria
must be defined in this regard to be embedded in the IS.
2. Data aggregation/disaggregation: Data are gathered at different scales; an
approach is required to identify rules according to which data must be
aggregated/disaggregated to guarantee information coherence.
3. Data redundancy prevention: most of data come from different sources; this
reflects in several issues. In fact, data are gathered in many different formats
(spreadsheets, documents, audio, video, etc.) that provide information of
invaluable importance that could also be repeated. The system should define
criteria according to which data are stored or not (e.g. quality of data, time of
acquisition, source reliability, etc.).
4. Data pre-processing: Data can come in several formats that are not necessarily
compatible with their storage in a database or for spatial analysis. The IS can
support only pre-defined format(s). Accordingly, procedures for data pre-
processing (for users) must be defined. Likewise, it is important to support the
process of structuring (when possible) and organizing that information that is
semi-structured or unstructured (e.g. pictures, drawings, audio files and so on).
5. Data acquisition: Some data are collected by means of field surveys. A tool
should be developed to support data survey in digital format (e.g. tablet). Other
data must be simply collected from other sources; accordingly the IS must allow
data acquisition from different sources/users (see the next point).
6. Multi-owners environment: Different users will use the IS in different way
(e.g. to insert data, to analyse data, to visualise data elaboration). Possible users
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
224 Flood Recovery, Innovation and Response IV
must be identified as well as allowed actions for each users. This implies to
create different user permissions in the IS (also in remote).
7. Data management: The IS must support several data analysis:
aggregation/disaggregation, visualisation, filtering, querying, etc. both during the
collection/survey phase and at the end of this activity. Pre-defined tools for data
analysis must be developed within the IS to facilitate/make quicker the scenarios
development.
4 Next steps
Coherently with the cyclic process in fig. 2, next steps of Poli-RISPOSTA are
towards the implementation of the IS; in particular, the first milestone concerns
the development of a first IS prototype. The design and implementation of the
prototype will be performed in close collaboration with the CP by means of
participatory processes, exercises, etc. according to the project philosophy
which considers the involvement of stakeholders as key to get efficient and
feasible solutions.
Next efforts of Poli-RISPOSTA can be grouped specifically into three main
activities which are all required to develop the IS:
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 225
5 Conclusion
The objective of this paper is to present the Poli-RISPOSTA project, an
interdisciplinary project of Politecnico di Milano providing novel and enhanced
methods and procedures for post-flood damage assessment. The latter are a key
prerequisite for improving pre and post event risk reduction strategies as required
(among the others) also by the EU Floods Directive. Having more reliable flood
loss data is of paramount importance for loss accounting, disaster forensic and
risk modelling.
Efficient solutions imply the use of advanced technological tools; for this
reason an interdisciplinary approach is required that put together expert’s domain
(i.e. flood analysts) and computer scientists.
By describing the current level of implementation of the project, the paper wants
to highlight two peculiarities of the problem at stake. On the one hand, its
complexity both in terms of methodological gaps, data to be handled,
elaborations to be performed and the variety of expertise which is required. On
the other hand, the need to work with stakeholders (i.e. the users of developed
tools) to get feasible and effective solutions.
Acknowledgements
The authors acknowledge all the people involved in the Poli-RISPOSTA project
for their useful feedback on the paper. Authors also acknowledge with gratitude
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
226 Flood Recovery, Innovation and Response IV
the Umbria Region Civil Protection authority (and its staff), which strongly
encourages/actively takes part in this research.
References
[1] ISDR, Hyogo framework for Action 2005-2015: Building the resilience of
nations and communities to disasters, http:\\www.unisdr.org/wcdr, 2009.
[2] Council of the European Union, Council Conclusions on a Community
framework on disaster prevention within EU. 2979th Justice and Home
Affairs Council meeting. Brussels, 30 November 2009.
[3] Council Regulation (EC) No 2012/2002 of 11 November 2002
establishing the European Union Solidarity Fund.
[4] EC, 2013. Green Paper on the Insurance of Natural and Man-made
Disasters. COM/2013/0213 final.
[5] De Groeve, T., Poljansek, K. & Ehrlich, D., Recording Disaster Losses.
Recommendations for a European Approach, JRC Scientific and Policy
Report. Report EUR 26111 En, 2013.
[6] Wirtz, A., Kron, W., Low, P. & Steuer M., The need for data: natural
disasters and challenges of database management, Nat Hazards, 70,
pp. 135-157, 2014.
[7] Guha-Sapir D. & Below R., The quality and accuracy of disaster data. A
comparative analyses of three global data sets. ProVention Consortium
(World Bank), 2002.
[8] Llsat et al., Towards a database on societal impact of Mediterranean floods
within the framework of the HYMEX project, Nat. Hazards Earth Syst.
Sci, 13, pp. 1337-1350, 2013.
[9] Mysiak, J., Testella, F., Bonaiuto, M., Carrus, G., De Dominicis, S.,
Ganucci Cancellieri, U., Firus, K. & Grifoni, P., Flood risk management in
Italy: challenges and opportunities for the implementation of the EU
Floods Directive (2007/60/EC), Nat. Hazards Earth Syst. Sci., 13,
pp. 2883-2890, 2013.
[10] Wergerdt, J. & Mark, S.S., Post-Nargis Needs assessment and monitoring.
ASEAN’s Pioneering Response, Final report, Asean Secretariat, 2010.
[11] Queensland Reconstruction Authority, Australia,
http://qldreconstruction.org.au/about/darmsys
[12] Jonkman, S.N., Bočkarjova, M., Kok, M. & Bernardini P., Integrated
hydrodynamic and economic modelling of flood damage in the
Netherlands, Ecological Economics, 66, pp. 77-90, 2008.
[13] Servizio Protezione Civile – Regione Umbria. Evento alluvionale 11-14
Novembre 2012: Rapporto di evento. Available on line at:
www.cfumbria.it
[14] Molinari, D., Menoni, S. , Aronica, G.T., Ballio, F., Berni, N., Pandolfo,
C., Stelluti, M., Minucci, G., Ex-post damage assessment: an Italian
experience, Nat. Hazards Earth Syst. Sci (accepted for publication).
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 227
Abstract
Results from cooperation between the insurance business and the Norwegian
Water Resources and Energy Directorate on a flood damage survey after a major
flood in Norway 2013 is presented, as well as results from similar cooperation
after a flood in 1995. Benefits for flood risk management of including flood
parameters in future damage surveys are presented.
Keywords: damage survey, flood, natural hazard, insurance, risk management.
1 Introduction
Data from flood events are collected by different stakeholders for different
purposes. This paper presents results from cooperation on flood damage survey
after a major flood in Norway in May 2013, between the insurance business and
the Norwegian Water Resources and Energy Directorate (NVE). The idea is that
the survey made by the insurance business for their purpose potentially could
provide valuable information for other purposes within flood risk management. In
the first instance the idea was to improve the basis for damage functions relevant
for Norway. The paper expands on this and presents ideas for future collection of
data.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140191
228 Flood Recovery, Innovation and Response IV
dealing with floods and landslides. The Government states that it will continue its
efforts in preventing damage from floods and landslides according to a holistic
approach including mapping, land use planning, protection measures, monitoring,
early warning, contingency and crisis management. The Norwegian Water
Resources Directorate (NVE) is the agency at directorate level responsible for
coordinating the implementation of the national policy. NVE’s work in preventing
damage from floods and landslides is structured according to the holistic approach,
in the following tasks:
- Hazard and risk mapping
- Assistance and control of land use planning in the municipalities
- Planning and construction of structural protection measures
- Monitoring and early warning: Floods, Debris flows, Snow avalanche
- Assistance to the police and municipalities in emergency situations
- Research & Development, Communication
The White Paper [1] outlines how responsibilities for dealing with floods and
landslides are distributed among the main actors. The importance of cooperation
between the relevant actors is highlighted, and a national strategy for cooperation
and coordination will therefore be developed.
Every municipality is obligated according to the Civil Protection Act to
perform an overall Risk and Vulnerability (RAV) analysis for its territory as a
basis for preparedness to deal with harmful events and for land use planning. The
municipality is responsible for making sure that natural hazards are being
evaluated and taken properly into account in every new development scheme,
according to the Planning and Building Act.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 229
(a) (b)
(c)
Figure 1: (a) A typical residential house; (b) flooded houses during a flood in
1995; (c) the interior of a house during repair after flood damage. (All
photos: NVE.)
After a major flood in South Eastern Norway in 1995, cooperation between the
Norwegian Natural Perils Pool (NNPP) and NVE was established in order to
collect data on water level in buildings and the corresponding damage. The
surveyors for the insurance companies were asked to register maximum water
level in buildings relative to ground floor level. Wathne et al. [2] developed stage-
damage functions based on a limited part of the data set (as shown in figure 2).
Some years later more of the data from the survey in 1995 was systemized and
used by Gottschalk and Krasovskaia [3] in the Interreg III B project FLOWS. They
established damage functions for different categories of objects, as shown in
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
230 Flood Recovery, Innovation and Response IV
figure 3. Some statistical parameters related to the same data set is presented
in table 1.
1200
1000
800
D am age (kN O K )
Constr. yr 1850-1942
200
0
0 100 200 300 400 500 600
Water level (cm) (ground level = 200)
1000
damage cost [kNOK]
100
10
-1 0 1 2 3 4 5
water level from basement floor [m]
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 231
A major flood occurred in Norway late May 2013. The most severe flood and
subsequent damage occurred in River Gudbrandsdalslågen and its tributaries.
Figure 4 show photos of damaged buildings in the village Kvam. After the event
the insurance companies immediately starts the process of assessing damage as
basis for the compensation to the policy holders. For this the insurance companies
hire surveyors with relevant education and experience.
Shortly after the flood, contact was established between NVE, Finance Norway
and the NNPP. An agreement was made to include data on water levels in the
survey, similar to what was done in 1995. Ad hoc a form was developed for the
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
232 Flood Recovery, Innovation and Response IV
As of February 2014 a total of 243 cases have been reported by the surveyors. The
processing of these data is not yet completed as the compensation process is still
ongoing in a substantial part of cases. Compensation paid to the policy holders so
far span from more than 1 Mill NOK to 6000 NOK.
Some preliminary figures concerning the data from the forms could still be
presented. Concerning water level the following data appear:
Further investigation needs to be made into the cases with value “0” or no value.
A preliminary review indicates that supplementary information in the form in
some cases includes information on water level. In other cases it appears that
damage was only to the garden and the courtyard.
Concerning the other parameters in the form, the following results appear as
shown in table 3.
We see from Gottschalk and Krasovskaia [3] that there is a great variability in
the data set. The question is if more factors could be identified to create relations
with less variability. This was the idea behind including more parameters than
water level, such as erosion and mass deposition, in the form used in 2013.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 233
Another obvious parameter to test, is the total value of the building. This is a topic
for further research.
The stage-damage functions developed based on the data from the 1995 flood, is
among others useful for cost-benefit analyses related to flood protection schemes.
To NVE cost-benefit analyses is key input to the decision on governmental
financial support or not.
NVE’s cost-benefit analyses are based on a common concept of risk among
engineers: risk is a product of probability and consequences. The probability part
is usually well covered, for instance through flood mapping. NVE has since 1998
produced flood inundation maps presenting areas prone to flooding with high
precision based on analyses of flood frequency, hydraulic modeling and GIS-
analysis with a detailed digital elevation model. Limited access to data on damage
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
234 Flood Recovery, Innovation and Response IV
Based on the cooperation from the 2013 flood, NVE and Finance Norway has
started a discussion on the possibilities for including registration of flood
parameters as standard in damage surveys by the insurance business. Included in
this is an investigation of the advantages of more detailed positioning of damage
points in map coordinates, including relating water levels to the standard map
elevation basis.
NVE has highlighted that a better positioning will open for a much wider use
of the data, and hence increase the value substantially. All parts of flood risk
management benefit from information on events and improved mapping.
A good positioning of damage will pinpoint areas at risk and thus provide
important input to the municipalities in their overall RAV-analyses. Put together
such point observations could be the basis for flood event maps showing areas
exposed to one particular flood. If more sophisticated mapping has not been
performed, event maps are valuable for land use planning, flood protection and
emergency preparedness.
Observation of water level from actual flood events is important for validation
and calibration of flood models, such as the hydraulic models used in flood
inundation mapping. The access and quality of calibration data in the form of
observed flood levels significantly affects the quality of the maps.
Information from events in itself or via flood maps is the key for taking flood risk
properly into account in land use planning. Similarly the quality of flood maps is
important for the planning of protection measures. Better models could potentially
lead to reduced development cost as safety/uncertainty margins could be reduced.
The models developed in the mapping process are also used during flood
situations as a tool for the crisis management. Better models will accordingly
improve the basis for decisions on measures to be taken during crises.
7 Conclusions
FNO has started a pilot project aimed at clarifying if damage data from storm
water, backwater in sewer systems and natural hazards could be useful for the
municipalities in their work on identifying vulnerable areas and the performance
of RAV analyses. The project is due to deliver its results by the summer of 2014.
The cooperation referred in this paper is not part of the pilot project, but stand
as an example of how data from the insurance business could be used by NVE and
other authorities in the prevention of flood damage. Given that the data are being
used as suggested above, it could contribute to improved decisions regarding flood
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 235
risk. Ultimately the result would be reduced damage related to flood events and/
or reduced cost for measures taken.
Registrations of more parameters related to floods, should preferably not be
based on ad hoc initiatives such as in 1995 and 2013, but rather be part of standard
procedures of the survey after a flood event.
Before deciding on this in any direction, more investigation is necessary on a
number of issues:
- How to perform the registration in practice.
- What are the most cost-effective solutions providing sufficient quality
of data
- Format and organization of data.
- Sensitivity of data; what could/could not be published.
Potentially such investigations could be part of a follow-up project. In any case,
the processing and evaluation of the 2013-data will continue and hopefully
contribute to the further investigation of these issues.
References
[1] White paper no. 15. Meld St 15 (2011-2012) Melding til Stortinget. Hvordan
leve med farene – om flom og skred. In Norwegian. Ministry of Petroleum and
Energy, Oslo, 2012.
[2] Wathne, M., Skoglund, M. & Eggestad, H.O. Samfunnskostnader på grunn
av flom i vassdrag. HYDRA report no. R02. Norwegian with English
summary. Norwegian Water Resources and Energy Directorate, Oslo, 1999.
[3] Gottschalk, L. & Krasovskaia, I. Expected damage (risk) of flooding. Interreg
IIB FLOWS report, sub project 1b, Oslo, 2006.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
This page intentionally left blank
Section 7
Emergency preparedness
and response
This page intentionally left blank
Flood Recovery, Innovation and Reponse IV 239
Abstract
This paper gives an overview of the on-line flood warning implementation in the
province of New Brunswick, Canada. The on-line flood warning applications are
available via the “River Watch” website provided by the New Brunswick
Department of Environment. Advanced GIS technology combined with
hydrological modelling, provide a mapping and visualization tool that can be used
by emergency managers and the general public to predict possible flood zones.
The applications developed for “River Watch” support the processing of large
amounts of digital terrain and hydrological data, which are then, quantified and
displayed on digital maps allowing decision makers and the general population to
comprehend and visualize the possible area and impact of the flooding. The
WebGIS applications that are available from the “River Watch” web site provide
snow reports and maps, flood warnings and interactive maps. The searchable
historical database containing reports about the impact of past floods and
estimated damages provides a valuable insight into the past of the province of New
Brunswick and the motivation for development of the system for flood prediction
and management.
Keywords: flood maps, flood prediction, flood management.
1 Introduction
In the province of New Brunswick (Canada), river valleys and flood plains can
pose a risk because of ice jams, harsh weather and floods of annual spring thaw.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140201
240 Flood Recovery, Innovation and Response IV
Another danger comes from hurricanes, tropical storms, erosion, or other harsh
seasonal weather events, which may cause tidal and ice surges in coastal areas (see
Figure 1). All of these can cause a threat of flooding with material damage to
people and even the loss of human lives [1].
In Canada, the province of New Brunswick was the first province to join the
Flood Damage Reduction Program signing General, Mapping and Studies
Agreements in March 1976 [2]. The first outcome of this agreement was mapping
of the flood plain, where one in a 100-year flood was used to delineate and
designate flood plains in 13 areas [2].
Figure 1: Historic ice jams in the Stain John river basin (from:
http://www2.gnb.ca/content/dam/gnb/Departments/env/pdf/Water-
Eau/SaintJohnRiverBasin-BassinFleuveSaintJean.pdf).
Within this program, a sub-agreement on structural controls centered on
building sea dykes in the Petitcodiac area was made. This separate agreement for
flood control (where each party assumed one third of the costs in the Marsh Creek
area) was negotiated with the federal and provincial governments and the city of
Saint John. The flood management related works provided by this program
included channel improvements, improvements to outlet control structures, the
construction of a reservoir and the reconstruction of a bridge [2].
Additional studies agreement funded ice research on the Restigouche River and
the international section of the Saint John River [2].
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 241
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
242 Flood Recovery, Innovation and Response IV
flooding that usually happen in spring time [3], sometimes even with the
ice jams (see Figure 3).
Flash flooding – in which warning time is extremely limited – can be
caused by hurricanes, violent storms or dams breaking, that became more
frequent in recent years [3] what is attributed to the climate changes.
Many Canadian rivers experience frequent flooding. The potential for
flood damage is very high if residential or commercial development is
allowed on low-lying, flood-prone lands [3]. The regulation for building
permits should exclude construction on the flood plain.
Figure 3: Ice Jam Flooding: an ice jam in the St. John River caused major
flooding, impacting homes, businesses and public infrastructure in
the Perth-Andover area (from: http://www2.gnb.ca /content
/gnb/en/multimedia/mrenderer .2012.03.2012-03-25_1.jpg.html).
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 243
Figure 4: The Web map application showing the snow depth for the whole
province of New Brunswick (from: http://www2.gnb.ca/content
/gnb/en/news/ public_alerts/river_watch/survey_depth_cm.html).
form directly at the web site [4]. The interactive maps that show flooded areas in
near real time are updated on a daily basis (see Figure 6). In Figure 7, a more
detailed view of the Saint John River watershed is shown, and Figure 8 shows a
detailed daily flood map for the City of Fredericton, with interactive graphs
providing the readings for water gauges along the hydrographic network.
While the flood maps are produced and updated on a daily basis [5],
hydrological modelling [6] provides the possibility for forecasts for the next two
days of the water levels along the Saint John River (shown on the table in
Figure 9).
A service for reports and warnings about ice jams is developed as well (see
Figures 10 and 11). The valuable knowledge about the past floods and their
impacts to the people living in the province of New Brunswick is compiled and
available via the utilization of the historical database. The database provides
search utilities and reports to the users (see Figures 12 and 13).
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
244 Flood Recovery, Innovation and Response IV
Figure 6: Interactive flood warning map for the province of New Brunswick
[4].
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 245
Figure 7: Interactive flood warning map for the province of New Brunswick,
“zoom-in” for Saint John River watershed [4].
Figure 8: Interactive flood warning map for the province of New Brunswick,
“zoom-in” for the City of Fredericton [4].
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
246 Flood Recovery, Innovation and Response IV
Figure 9: Two days forecast for Saint John River (from: http://www2.gnb.ca
/content/gnb/en/news/public_alerts/river_watch/st_john_river_two
-dayforecast.html).
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 247
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
248 Flood Recovery, Innovation and Response IV
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 249
3 Conclusions
An overview of the flood prediction and mapping applications available on-line
from the “River Watch” web site has been presented. The online applications allow
access to flood forecast data and mapping services for ice jams, roads accessibility
or closure and near real time flood plain delineation. The historical database
provides the utilities for search and access to data about past flood events and
damages that were caused. The flood prediction and mapping applications and
other services provided by “River Watch” are accessible to the decision makers
and general public in order to assist them to comprehend the impacts and potential
damages of the flooding.
Acknowledgements
This project was financially supported, in part, by the N.B. Emergency Measures
Organization and by the Canadian Department of Natural Resources
Geoconnections program as well as by University of New Brunswick and New
Brunswick Innovation Foundation (NBIF). The IT Division of the City of
Fredericton and Geological Survey of New Brunswick provided datasets available
for this project. The New Brunswick Department of Environment has provided
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
250 Flood Recovery, Innovation and Response IV
References
[1] Hazards in New Brunswick, http://www.getprepared.gc.ca/cnt/hzd/rgnl/nb-
eng.aspx, accessed on-line, March 2014.
[2] Environment Canada, Flood damage reduction program,
https://www.ec.gc.ca/eau-water/default.asp?lang=En&n=B5349463-1,
archived information, accessed on-line, March 2014.
[3] Hazards in New Brunswick - Flood http://www.getprepared.gc.ca
/cnt/hzd/flds-eng.aspx#a1, accessed on-line, March 2014.
[4] River Watch, http://geonb.snb.ca/riverwatch/index.html#, accessed on-line,
March 2014.
[5] Mioc, D., Nickerson, B., Anton, F., Fraser, D., McGillivray, E., Morton, A.,
Tang, P., Arp, J.P. & Liang, G., Web-GIS application for flood prediction and
monitoring, International Conference on Flood Recovery Innovation and
Response, London, WIT Transactions on Ecology and the Environment
(ISBN: 978-1-84564-132-0), WIT Press, 2008, pp. 145-154.
[6] Mioc, D., Anton, F., Nickerson, B., Santos, M., Adda, P., Tienaah, T., Ahmad,
A., Mezouaghi, M., MacGillivray, E., Morton A. & Tang, P., Flood
Progression Modelling and Impact Analysis, Efficient Decision Support
Systems - Practice and Challenges in Multidisciplinary Domains, Chiang Jao
(Ed.), ISBN: 978-953-307-441-2, InTech, 2011, pp. 227-246.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 251
Abstract
There is a land of 300 km2 lower than sea level with a million citizens facing Ise-
bay in the central part of Japan, which is located on a possible route of typhoons
and is exposed to a risk of serious storm surge and flood. This area was attacked
by storm surge by “Ise-bay Typhoon” in 1959 and more than 5,000 people were
killed. In spite of a protection infrastructure constructed in this half decade after
the event, recent climate change may cause extreme typhoons exceeding the
level of protection and response, and resilience against such an enormous
disaster has not yet been prepared. We have made efforts to prepare an action
plan of risk management and emergency response since 2005. Once an extreme
storm surge breaks the protection infrastructure, a wide area will be inundated
with various risks and drainage from there will take a long time. Meanwhile,
with recent progress in weather forecasting of magnitude and course of big
typhoons, we may have a lead time of 36 hrs. We introduce 4 phases: Risk
management before typhoon arrival (Phase 0), emergency response within 0–
72 hrs (Phase I) and successive stages (Phases II and III). In particular, we study
how to make a wide preliminary evacuation possible with proper operation in
Phase 0. We have organized a working group to support the authority including
all the stakeholders related to disaster mitigation to make an action plan of risk
management and emergency response.
Keywords: typhoon, storm surge, catastrophe management, risk management,
emergency response.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140211
252 Flood Recovery, Innovation and Response IV
1 Introduction
There is a land of 300 km2 lower than sea level facing the Ise-bay in central
Japan. This area includes a part of Nagoya Metropolis whose population is more
than 2 million, and has developed as an industrial centre with high economic
activity. On the other hand, this area is located on a possible route of typhoons
and major rivers neighbouring this area cause huge flooding from mountain areas.
This area was attacked by a serious storm surge due to a typhoon (Ise-bay
typhoon, “Vera”) in 1959, and 500 km2 was flooded for a few months. More than
5,000 people were killed and the daily lives of a large number of citizens were
affected for a long time [1, 2]. Within around a half decade after this event, we
have completed protection infrastructures against that level of storm surge and
floods due to typhoons.
However, the level of typhoon may exceed the protection level because of
probabilistic phenomena, and recent climate change may cause such a super-
class typhoon. At the news of serious flooding of New Orleans by Hurricane
Katrina in 2005 [3], we learned of the emergency response there [4] and we
started to discuss risk and emergency management for the Ise-bay area. The
situation of this area such as the route of a typhoon and geographical
characteristics can cause high storm surge. There is a wide land below sea level
with a million citizens, and the large number of human activities is very similar
to New Orleans. Moreover, not only the area facing Ise-bay but areas facing
Tokyo-bay and Osaka-bay are in the same situation. In other words, 3 major
metropolises, Tokyo, Osaka and Nagoya, have similar risks of wide and long
term flooding due to storm surge to threaten a large number of lives and human
activities. Certainly the protection infrastructure has been accomplished during
this half decades, but catastrophic disaster may be estimated once a super
typhoon exceeding the level of our present protection level attacks those areas.
We have not prepared an appropriate risk management and emergency response
plan.
City Center
Ibi River Shonai R.
Sumida R. Edo R.
Mississippi River
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 253
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
254 Flood Recovery, Innovation and Response IV
record in Japan) and travels on a route that will cause the severest storm surge on
the Ise-bay. The time path of this typhoon is shown in Figure 2 [5].
Because risky points of coastal levee breach can be estimated based on the
numerical calculation of storm surge (rising of sea level due to astronomic tide,
lift up by pressure drop and waves), levee breaches are assumed there. In
addition, heavy rainfall with return period of 1,000 years is postulated. Then, we
assumed river levee breaches at several places along class A rivers. Such
assumptions are practically familiar in making a “flood hazard map” [5].
Figure 3 shows the flooded area [5] which is around 500 km2 (520 km2 with only
202 km2 caused by storm surge), and almost equal to the actual flooded area on
Nagara R.
Ibi R.
Kiso River
Maximum
inundation
Depth (m)
Shonai River
5.0 ~
4.0 ~ 5.0
3.0 ~ 4.0 Nagoya
2.0 ~ 3.0 City
1.0 ~ 2.0
0.9 ~ 1.0
0.8 ~ 0.9
0.7 ~ 0.8
0.6 ~ 0.7
0.5 ~ 0.6
0.4 ~ 0.5 Nagoya
0.3 ~ 0.4
0.2 ~ 0.3 Port
0.1 ~ 0.2
0.0 ~ 0.1 Ise Bay
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 255
the occasion of an Ise-bay Typhoon in 1959 (531 km2 with only 310 km2 caused
by storm surge). In Figure 3, the grey-scale presentation cannot show a detailed
spatial distribution of maximum flood depth but from the legend of the figure
one can recognize the range of maximum flood depths and the resolution of the
simulation.
In this scenario, overflow due to storm surge begins 18:00, and the tidal level
shows the maximum at 22:00, while the levee breaches along rivers due to
flooding happens after 01:00 of the next day. On the other hand, storm with
stronger wind than 20 m/s begins at 18:00 [5].
Number of days
required for 3~7
levee closure
and unwatering days
7 ~ 14 days
3 ~ 7 days 7~14days
0 ~ 3 days
3~7
days
Once the levees are broken, a wide area is flooded and immediately the levees
should be closed and drainage efforts are required in Phase I. Closure of levee
breaches and drainage have several technical problems (district division and
arranging pumping vehicles), but we roughly calculated the necessary terms of
drainage for individual districts, as shown in Figure 4 [8]. Some areas may
remain flooded for a few weeks and during this period people cannot live their
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
256 Flood Recovery, Innovation and Response IV
daily lives. Considering that the tasks to be done in Phases I and II are many,
there is less possibility to support life in such a flooded area. Thus, preliminary
evacuation from the area expected to be exposed to severe flooding is strongly
recommended and it should be achieved before the disaster happens. That is
“preliminary evacuation in a wide area” (over communities). If a disaster
happens, people staying in refuges inside the seriously flooded area must move
to the shelters in the dry areas because they cannot continue their daily lives
without lifeline services there (secondary evacuation), and they need special
transportation in the flooded areas.
The numbers of evacuees of respective cities or towns (or wards of Nagoya
city) who have no shelters within their communities were surveyed and the
results are shown in Figure 5 [5–7], where several neighbouring communities are
grouped as one block. The most important emergency support function of
Phase 0 is preliminary evacuation to a safe area (where flooding is not predicted)
and the evacuation destination must be different communities and such
evacuation necessitates a long journey.
balanced
Block 1
Block 2 balanced
Block 3
Block 4
5700
78400
Block 5
Block 6
25400 Block 7 Block 8
31700 50000
balanced
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 257
Usually, each warning level corresponds to each action for evacuation guidance
as follows:
Level 1: Recommendation of voluntary evacuation;
Level 2: Evacuation completed for handicapped persons;
Level 3: Issue of evacuation advisory by community head;
Level 4: Issue of evacuation order by community head.
However, in this case of a catastrophic typhoon, since evacuation requires long
travelling distances and time, the above guideline is not available or it may be
too late.
p g
名古屋
Nagoya
■18:00時点■
18:00
北緯33.4度,東経135.9度
Stage
■ステ ージ4 4 ((-6:00~0:00)
1 2 :0 0 -1 8 :0 0 )
避難指示を 発令する 。 間に合わな い地区や時
Inundation
伊勢湾岸にて高潮越流
による浸水開始 Evacuation
間によ っ て は緊急避難を command
指示する 。
Stage 3 (-3:00~-6:00)
■ステ ージ 3 ( 9 :0 0 -1 2 :0 0 )
Wide Range
避難勧告を 発令する 。Evacuation
広域避難を 行う 。
12:00
■12:00時点■
北緯31.0度,東経136.0度
Stage
■ステ ージ2
2 ((-12:00~-9:00)
6 :0 0 -9 :0 0 )
Evacuation準備・
要援護者の避難を of HC 開始・ 完了する 。
09:00
■09:00時点■
北緯29.9度,東経136.4度 Stage
■ステ ージ1
1 ((-24:00~-12:00)
1 日前1 8 :0 0 -当日の6 :0 0 )
自主避難を 呼びかける 。 精度の高い台風進路
Voluntary
予測、 高潮予測が発表さ evacuation
れる 。
Stage 0 (-36:00~-24:00)
■ステ ージ 0 ( 1 日半前~1 日前の1 8 :0 0 )
情報共有本部を 設立し 、 関係機関で 情報を 共
Set Pre-JFO
有する 。
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
258 Flood Recovery, Innovation and Response IV
completion for each community (from the center of the origin community to the
center of the destination communities) for several sets of combinations of cars,
buses and trains. The capacity and travelling time on the main routes such as
highways and national roads to connect origin and destination communities are
investigated for transportation by cars and buses. The statistics of railroad
companies are taken into account in the calculations for transportation by train.
Three cases of combinations are tested:
Case 1: cars (70%), buses (10%), trains (20%);
Case 2: cars (40%), buses (40%), trains (20%);
Case 3: cars (10%), buses (40%), trains (50%).
The results are summarized in Table 1, where the required time for completion of
evacuation is balanced with the number of evacuees from the origin block and
capacity of destination community. Table 1 suggests that there is an unbalance
between numbers of evacuees and capacities for respective communities (as a
total the capacity is less than the number of evacuees) and that the advantages in
the required time for evacuation change depend on the cases (combinations of
transportation means are different from one another).
Though no detailed names of communities are indicated in Table 1, Nagoya
city lacks the capacity of shelters and most citizens and the ward governors
consider that taller buildings may become refuges. However, once the area is
flooded, it might be quite difficult to support daily life there. People should know
that their daily lives depend on various lifelines, which are at risk of damage and
require a long time for their repairs. On the other hand, in the Nagoya city area,
if evacuees change their evacuation means from cars to trains, the required time
for evacuation can be efficiently reduced. Bus transportation is most efficient for
some other communities. Evacuation by individual cars requires a long time
though many inhabitants would use their own cars because buses and trains are
not a convenience in their daily lives.
Block O-community No.of evacuee Capacity of shelter Time for Evacuation (hrs)
(D-Community) Case 1 Case 2 Case 3
2 A 79,400 70,900 40 24 41
B 50 30 35
4a C 53,300 59,300 5 3 3
D 27 16 7
4b E 57,000 51,700 28 17 42
F 6 4 2
4c G 160,800 58,400 13 8 10
H 41 25 8
5 I 22 13 5
8 J 50 30 9
7a K 15 9 11
L 3 2 1.5
6 M 26,100 37,100 25 15 11
N 7 5 2
7c O 1.3 1.2 1.1
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 259
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
260 Flood Recovery, Innovation and Response IV
Local
Prefecture town
Other inhabitants Municipality
Organizations
Local
Informations: Inhabitants
・weather & river information
・Institution and action of organizations
・Information for evacuation:
Numbers of evacuees
Refuge condition, evacuation route
・Information on Handicapped persons Pre Joint Field Office
supporting system
・Traffic information:.
Traffic control, traffic jam, etc.
・Number of evacuee, staying peoples, etc.
Set‐up of Pre FEO :Flow of staffs
:Information Flow :Staff of FEO
Negotiation among
Tops of organizations
:Information System :TNT member
Belonging to TNT
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 261
7 Concluding remarks
The area facing the Ise-bay is a low land more than 300 km2 below sea level and
exposed to a risk of flooding by storm surge and floods due to super typhoons.
We postulated the “super Ise-bay typhoon” as a possible maximum one, and
discussed risk management and emergency response in this paper. From the view
point of emergency management against flooding of wide areas with failures of
lifelines, preliminary evacuation in a wide area was investigated as the key in a
risk management action plan. In this area, a few hundred thousand citizens are
considered as evacuees who must travel long distances within a day. Without a
plan and means for controlled evacuation, they cannot succeed. In this paper,
some model of the combination of transportation means was investigated by
using simulation. The numbers of evacuees and capacity of shelters are not
balanced, and depending on conditions of the respective communities the
appropriate combination of transportation means for evacuation are different
from one another. It is one of the difficulties in this problem but conversely this
point may give us a key to find an appropriate action plan. Furthermore, we have
proposed headquarters for information sharing and arrangements of resources
among different stakeholders, and it is expected to smoothly continue with
headquarters for disaster control after the disaster happens.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
262 Flood Recovery, Innovation and Response IV
References
[1] Chubu Association of Regional Design, 50 Years after Ise-bay Typhoon,
(supervised by T. Tsujimoto), 109p., 2009 (in Japanese).
[2] Japan Water Forum, Typhoon Isewan (Vera) and Its Lessons, 60p., 2005.
[3] White House, The Federal response to Hurricane Katrina: Lessons
Learned, 228p., US White House, 2006.
[4] Tsujimoto, T., Field survey on repair and restoration process and
improvement of emergency response in US after Hurricane Katrina and risk
and emergency management of Ise-bay area below the sea level against
storm surge and flood, Jour. Hydroscience & Hydraulic Eng., JSCE,
Vol. 54, pp. 889–894, 2010 (in Japanese).
[5] Tokai Nederland regional Authority against Storm Surge and Flood, Risk
Management Action Plan, 2nd Ver., 177p., 2009 (in Japanese).
[6] Tsujimoto, T., T. Kohno and S. Tanaka, Action plan for risk management
against large scale inundation due to “super Ise-bay typhoon” – Tokai
Nederland action plan against extreme storm surge and flood, Abstract, 5th
International Conf. on Flood Management, Tokyo, 2011.
[7] Kobayashi, K., On the Authority against Storm Surge and Flood in Tokai
Nederland –TNT Risk Management Action Plan, Lecture Note, 49th Summer
Seminar Series on Hydraul. Eng., JSCE, A-1, 2013 (in Japanese).
[8] Chubu Regional Bureau, Ministry of Land, Infrastructure, Transport and
Tourism, Unwatering Plan in the Nobi Plain, 132p., 2013 (in Japanese).
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 263
Abstract
CASSANDRA robotic system developed at LTR s.r.o. company and Brno
University of Technology is described. The system contains an operator’s station
controlled with one operator and a couple of robots – small and big ground robots,
flying robots (quadrocopters), and mapping robot. The robots are primarily
controlled by the operator with an advanced user interface with visual telepresence
and augmented reality. Nevertheless, the robots include the possibility of semi-
autonomous operation based on self-localisation. The user interface consists of a
computer, joypad, head-mounted display with inertial head-tracker,
communication device, and Cassandra software developed by our team in
Microsoft .NET. Orpheus class robots are described in the text. The robots are
made to be reliable and to be able to work in extreme conditions, they are tested
by a series of MIL-STD military tests for environmental parameters, EMC,
vibrations and shocks, contamination/decontamination, etc. Orpheus-X3 is a
general US&R robot with enhanced victim search capabilities, Orpheus-HOPE is
made for water contamination measurements, Orpheus-AC2 is a ruggedized
version for environmental parameter measurement. Two flying drones developed
completely by our team are described, as well as EnvMap mapping robot for real-
time construction of spatial digital maps with texture mapping. All the robots can
be controlled with the help of visual telepresence and augmented reality – that
makes robot control much more intuitive, and lets the rescuer concentrate on the
mission itself. The control station may be used as a self-containing wearable
system. The fusion system with multispectral measurement containing tricolor
cameras, thermal imagers and TOF camera is described.
Keywords: robot, user interface, telepresence, augmented reality, data fusion.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140221
264 Flood Recovery, Innovation and Response IV
1 Introduction
The reconnaissance of dangerous areas is one of the most challenging tasks for
today’s robotics. According to many indications, e.g. from the Robocup Rescue
League community where the DCI team is involved [1, 3], it seems that nowadays
the development of practical and usable reconnaissance robots is aimed at the
following tasks:
• A larger number of robots controlled by one operator, in such cases as when
the operator must concentrate on crucial tasks, such as victim identification,
while the robots perform basic tasks, like mapping, autonomously.
• Easy and intuitive human-to-robot interface should be optimized, since the real
operators will be rescuers rather than robotic specialists.
• For many kinds of reconnaissance missions it would be highly beneficial if the
user interface would somehow emphasize alive people – since they are often
the main objective (earthquake or floods victims, injured soldiers, criminals or
terrorists).
The remote robotic reconnaissance of dangerous areas is a very complex and
interdisciplinary task [7], and only well-tuned robotic systems, with good
software, hardware, communication and sensory subsystem, may succeed [17].
Mobility and the ability to work reliably in hard conditions are very important. It
also induces that mechanical construction and the hardware of the robots play a
very important role in this complex task [13].
The authors propose a possible solution of the abovementioned problems
through an advanced user interface program called CASSANDRA and show its
application on several reconnaissance robots developed by our team.
Although the technical features of individual robots are supposed to differ, the
robots can be divided into certain “classes” of robots that are capable of being
controlled with the control system. The classes are listed below with an emphasis
on their mapping and self-localization abilities.
• Bigger and more complex robots with sufficient mapping and self-localization
capabilities (e.g. Orpheus).
• Small robots with limited mapping and self-localization capabilities (e.g.
Perseus).
• Rotorcraft Unmanned Aerial Vehicles (UAVs) with self-localization only (e.g.
Uranus).
• Mapping robots with exceptional mapping and self-localization capabilities
(EnvMap).
At present, the reconnaissance robots and the operator’s telepresence control
system, the CASSANDRA, are completed; thus, each robot can be effectively
controlled by the system. Multispectral data-fusion for colour and thermal image
mixing with help of TOF camera, is finished as well. The current task consists in
enabling the automatic mapping and self-localization of the robots, both outdoors
and indoors, and implementation of enhanced reality mixing real telepresence data
with the data from the multispectral maps.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 265
2 Orpheus robots
Orpheus robots have been developed at Department of Control and
Instrumentation (DCI) and LTR s.r.o. (spin-off Brno University of Technology)
since 2003. The first version was called simply Orpheus, and our team was quite
successful in Robocup Rescue 2003 world competition in Padova, Italy – we won
the competition (see [1]). In 2003–2006 we improved/rebuilt the robot to the
version Orpheus-X2 (see [2]). In 2006 we were asked to make a military version
of the robot. The prototype was finished in 2007 and named Orpheus-AC (Army
and Chemical). In 2009 we started development of second generation, based
on Orpheus-A2 platform. We decided to make two basic modifications – Orpheus-
AC2 for chemical and nuclear contamination measurements and
Orpheus-Explorer for more general reconnaissance missions and victim search.
2.1 Orpheus-AC
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
266 Flood Recovery, Innovation and Response IV
2.2 Orpheus-HOPE
Its main difference to the other Orpheus robots is the sensory head on 1 DOF
motorised manipulator with heavy-metal analysis probe and water dive sensor –
both the sensor and probe were newly developed by our teams. The robot, in its
current status is only a proof-of-concept, we are currently working on practically
usable device with peristaltic pump-based system of water sampling.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 267
2.3 Orpheus-X3
The Orpheus-X3 (see Fig. 4) is an experimental reconnaissance robot based on the
Orpheus-AC2 model. It offers the same drive configuration as its predecessor,
namely the four extremely precise AC motors with harmonic gears directly
mechanically coupled to the wheels; this configuration makes the robot very
effective in hard terrain and enables it to achieve the maximum speed of 15 km/h.
The main difference consists in the chassis, which is not designed as completely
waterproof but consists of a series of aluminium plates mounted on a steel frame
of welded L-profiles. This modular structural concept makes the robot markedly
more versatile, which is a very important aspect in a robot made primarily for
research activities. Furthermore, the device is equipped with a 3DOF manipulator
for the sensory head. The manipulator, again, comprises very powerful AC motors
combined with extremely precise, low backlash harmonic drive gearboxes by the
Spinea Company. The presence of such precise gearboxes can be substantiated by
several reasons, mainly by the fact that the robot will be used not only for
telepresence but also for mobile mapping and SLAM [9, 10]. As currently planned,
the robot’s only proximity sensor will be the TOF camera.
1 3 1
2 2
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
268 Flood Recovery, Innovation and Response IV
• Two thermal Imagers (see 2 in Fig. 4). MicroEpsilon TIM 450 with a wide
lens, 382x288 pixels resolution, temperature resolution of 0.08K.
• One TOF camera (see 3 in Fig. 4). A Mesa Imaging SR4000 with the range of
10m, 176x144 pixels resolution. The field of view is 56°(h) x 69°(v).
The scheme of the presented system is indicated in Fig. 5, right.
THERMO L EMBEDDED PC
THERMO R EMBEDDED PC
TOF
ETHERNET
CCD L SWITCH
CCD R
OPERATOR STATION PC
ETHERNET
USB
Figure 5: Calibration pattern for TOF camera, CCD camera and thermal
imager (left), scheme of multispectral sensory head connections [4]
(right).
The technique was already studied by our team in the past (see [14]), but as the
sensory prices decreased rapidly and TOF cameras further developed, the method
may be improved to reach a significantly more advanced stage.
Image transformations are applied for data fusion. The range measurements of
the TOF camera can be displayed into images of CCD cameras and thermal imagers
using spatial coordinates. The procedure is outlined in the diagram on Fig. 6. The
input data include the range measurement, the image coordinates of all sensors, and
the results of the previous calibration.
The spatial coordinates X, Y, Z are computed from Eqs (1) and (2), where d is
the measured distance, xc, yc are the calibrated TOF image coordinates, and f is the
focal length of the TOF camera. The homogeneous transformation is determined
by Eq. (4), where R[3×3] is the rotational matrix, t[3×1] is the translation vector, and
X', Y', Z' are the spatial coordinates of the second sensor. The image coordinates of
the TOF camera in the next frame xc',yc' are computed according to perspective
projection (see Eq. (4)), where f' is the focal length of the second sensor.
Z d cos arctan
yc cos arctan xc
2 f
f 2 xc
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 269
Zxc Zy
X Y c
f f
Range measurement and image coordinates of TOF camera
Spatial coordinates
Homogeneous transformation
Perspective projection
Correction of principal point
Displaying overlapping images
Measured Intrinsic and
TOF, CCD FUSION
Range
spatial 3D extrinsic
image + +
points [X, Y, Z] parameters
Range image projected to CCD image
TOF, TERMO FUSION
Measured Intrinsic and
Range
spatial 3D extrinsic
image + +
points [X, Y, Z] parameters
Range image projected to thermal image
CCD, TERMO FUSION
ID point in CCD ID point in
image thermal image
Figure 7: Scheme of data fusion: up – TOF and CCD data fusion; centre –
TOF and thermal data fusion; down – CCD and thermal data fusion.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
270 Flood Recovery, Innovation and Response IV
X ' X
Y '
R t Y
Z ' 0 1 Z
1 1
According to the identical (ID) points of the TOF camera transformed into the
frames of the CCD camera and the thermal imager, the thermal image can be
displayed into the CCD image and vice versa.
Figure 8: (Left) complete fusion for one eye; (right) top right: the range
image; bottom: fusion of the range and thermal images by the
described algorithm.
This is performed for two stereo pairs of cameras, and thus the resulting image
may be presented to a head-mounted display with a stereovision support [16]; the
operator therefore receives a very good spatial representation of the environment
under any visibility conditions.
Figure 9: Data-fusion evidence grid with colour only (left) with thermal
imaging (right).
It has to be pointed out that the sensors on the sensory head are not used only
for this technique; simultaneously, we are also developing a SLAM technique and
similar texture-mapping algorithms [11] with robot evidence grids and octree [15].
Both of these maps contain color information and thermal information [12], so e.g.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 271
alive humans can be easily emphasized in the image – see Fig. 8. The octree map
has the advantage of great loseless data-compression (up to 1:512 for the scene on
Fig. 9 and resolution 1.28 cm), while the evidence grids are easy-to-modify.
Currently we are able to combine both of them in one image.
Figure 11: Spatial robot evidence grid (left) and height map (right) scanned by
EnvMap robot.
4 Other robots
A couple of other robots were developed by our team as a part of CASSANDRA
system. Uranus is our own multicopter [19] system. It currently contains two
quadrocopters – Uranus-ALU with 350 g payload capacity, and Uranus-CARB
with approx. 1500 g payload capacity (see Fig. 12).
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
272 Flood Recovery, Innovation and Response IV
Scorpio is an indoor robot based on Dr. Robot drive system similar to iRobot
Packbot. The robot is intended for indoor operation, it is able to climb up-stairs.
Our team only developed the electronics and camera manipulator. Perseus is one
of our small robots capable of operation in hard terrain (see Fig. 12).
5 CASSANDRA software
All the mentioned robots may be controlled by CASSANDRA software,
developed by our team. It is basically a universal user-interface program
developed in Microsoft .NET 4.5, WPF. It has many displaying capabilities (see
Fig. 13). The most important central part is filled with main camera image, while
the corners can be covered by configurable virtual head-up displays containing
video from other robot cameras or video from other active robots, as well as, other
data from robot sensors or depicting system status. The system can work with
variety of head mounted displays equipped with head movement sensors.
One of the main advantages of the whole CASSANDRA system is, that since
all of the parts (i.e. all the robots, software, communication protocols) are
developed by our team, it is possible to control all of the robots by one operator’s
station equipped with CASSANDRA software (see Fig. 14). So the operator has
the possibility easily switch among the robots and select the most appropriate one
for the task.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 273
6 Conclusion
The presented CASSANDRA system represents work-in-progress, rather than
completely finished system. The telepresence part of the system is considered
finished, currently the team works on integration of semi-autonomous and
autonomous functions, like self-localisation and autonomous real-time map
building. Several parts of the system are currently practically usable, e.g. Orpheus-
AC2 robot, that is in active military service in the Czech Army.
Acknowledgements
This work was supported by CEITEC – the Central European Institute of
Technology (CZ.1.05/1.1.00/02.0068) utilizing the European Regional
Development Fund.
This work was also supported by VG 2012 2015 096 grant named Cooperative
Robotic Exploration of Dangerous Areas by the Ministry of Interior, Czech
Republic, program BV II/2-VS.
References
[1] Zalud, L., (2004). Rescue Robot League – 1st Place Award Winner. In:
RoboCup 2003: Robot Soccer World Cup VII, Springer, Germany, ISBN 3-
540-22443-2.
[2] Zalud, L., (2001). Universal Autonomous and Telepresence Mobile Robot
Navigation. In: 32nd International Symposium on Robotics – ISR 2001,
pp. 1010-1015, Seoul, Korea.
[3] L. Zalud, “Integration of 3D Proximity Scanner to Orpheus Robotic
System”, in 16th IFAC World Congress. Prague, Czech Republic: 2005,
pp. 1209-1215.
[4] Kocmanova, P., Zalud, L., Spatial Calibration of TOF Camera, Thermal
Imager and CCD Camera. In Mendel 2013: 19th International Conference
on Soft Computing. Brno: Brno University of Technology, 2013, pp. 343-
348. ISBN 978-80-214-4755-4.
[5] Hartley, R., Zisserman, A., Multiple View Geometry in Computer Vision.
Cambridge, Cambridge University Press, 2003, ISBN 05-215-4051-8.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
274 Flood Recovery, Innovation and Response IV
[6] Zhang, Z., Flexible camera calibration by viewing a plane from unknown
orientations, In Computer vision, Vol. 1, 1999, pp. 666-673.
[7] Lundberg, C., and Christensen, H. I. Assessment of man-portable robots for
law enforcement agencies. In PerMis (Gaithersburg, MD, Aug 2007), R.
Madhavan and E. Messina, Eds., ACM/IEEE.
[8] Leonard, J.J., Durrant-Whyte, H.F. (1991). “Simultaneous map building and
localization for an autonomous mobile robot”. Intelligent Robots and
Systems’ 91. ‘Intelligence for Mechanical Systems, Proceedings IROS’91.
IEEE/RSJ International Workshop on: 1442–1447. doi:10.1109
/IROS.1991.174711. Retrieved 2008-04-08.
[9] Karlsson, N., Di Bernardo, E., Ostrowski, J, Goncalves, L., Pirjanian, P.,
Munich, M. (2005). “The vSLAM Algorithm for Robust Localization and
Mapping”. Int. Conf. on Robotics and Automation (ICRA).
[10] Ju, X., Nebel, J.-C., Siebert, J. P., 3D Thermography Imaging
Standardization Technique for Inflammation Diagnosis. In Proceedings of
the SPIE, Vol. 5640, 2005, pp. 266-273.
[11] Prakash, S., Pei Yean Lee, Caelli, T., 3D Mapping of Surface Temperature
Using Thermal Stereo, In Control, Automation, Robotics and Vision, 2006,
pp. 1-4.
[12] Tournas, E.-Tsakiri, M., Distance Error Estimation for Range Imaging
Sensors. In: Proceedings of the ISPRS Commission V Mid-Term
Symposium “Close Range Image Measurement Techniques”,
Vol. XXXVIII, 2010, Part 5, pp. 581-585. Newcastle upon Tyne, United
Kingdom, 21.-24. 6. 2010.
[13] Wise, E., (1999). Applied Robotics, Prompt Publications, USA, ISBN: 0-
7906-1184-8.
[14] Henning Eberhardt, Vesa Klumpp, Uwe D. Hanebeck, Density Trees for
Efficient Nonlinear State Estimation, Proceedings of the 13th International
Conference on Information Fusion, Edinburgh, United Kingdom, July,
2010.
[15] Posted on (19 December 2011). “Understanding Requirements for High-
Quality 3D Video: A Test in Stereo Perception”. 3droundabout.com.
Retrieved 29 March 2012.
[16] Robin Ritz, Markus Hehn, Sergei Lupashin, and Raffaello D’Andrea,
“Quadrocopter Performance Benchmarking Using Optimal Control”,
IEEE/RSJ International Conference on Intelligent Robots and Systems,
pp. 5179-5186, 2011.
[17] “CBRN Defence Market Forecast 2014-2024”.
http://www.visiongain.com/Report/1206/CBRN-Defence-Market- Forecast
-2014-2024 London, U.K. 14 February 2014. Retrieved 24 March 2014.
[18] Zalud, L., Burian, F., Kopecny, L., Kocmanova, P. (2013). Remote Robotic
Exploration of Contaminated and Dangerous Areas, International
Conference on Military Technologies, pp 525-532, Brno, Czech Republic,
ISBN 978-80-7231-917-6.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Section 8
Adaptation to flood risk
This page intentionally left blank
Flood Recovery, Innovation and Reponse IV 277
Abstract
The rising sea-level and the frequency of devastating floods have already
increased in a considerable way. At the same time, the population is
continuously rising, along with the demand for adequate housing and sufficient
space. In this context the so-called floating houses are a future-oriented solution
for settlements along coastlines and river districts or on little islands.
In highly industrialized countries which export products or the associated
licenses it is an opportunity for the development and construction of floating
houses. Up to now in most cases the floating objects are built on pontoons as
usual buildings on a fixed ground. The special boundary conditions caused by
water waves, water chemistry and climate components should be considered in
order to prevent damage. Concomitantly the floating objects bring chances for
mobility and use of alternative energies due to the water environment.
In Lusatia, a landscape in the eastern part of Germany southeast of Berlin, a
lot of former lignite open-cast mines were filled with water and the worldwide
largest artificial lake landscape was created among others by the assistance of the
International Building Exhibition “Fürst-Pückler-Land”. Different types of
floating houses have already been built. The department Building Physics of the
university BTU-CS has carried out a lot of investigations with regard to
materials, energy use and climate boundary conditions by means of
measurements and numerical simulations. Besides, water as a building ground
must be cost effective and exhibit an affordable floating architecture.
Keywords: floods, urban strategies, floating house, heat exchanger.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140231
278 Flood Recovery, Innovation and Response IV
Contrary to the German and European situation, the global population is growing
(fig. 1). Above all, the aspiration level increases unrestrained in our society also
with regard to the demand for living room. For instance in Germany 70 ha of
land per day are used additionally for building in spite of the decreasing
population in this country.
Independent of the knowledge of the reasons for climate change, the sea level is
rising and endangers the infrastructure of settlements in many regions (fig. 2).
Figure 2: Examples for the effect of sea level rising in the Netherlands and Asia.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 279
2 Floating houses
2.1 General situation
Floating houses have a long history [1]. The technique and architecture of these
buildings depend on climate boundary conditions, culture and raw materials
which were available in various places. Nowadays one can find exquisite
examples of floating buildings all over the world (fig. 3).
Figure 3: Seoul, floating amusement park opened 2011, Hangang river at night.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
280 Flood Recovery, Innovation and Response IV
3 Adaptation
3.1 Material and construction
3.1.1 Materials
Building materials used especially for pontoons are highly dependent on the
quality of the surrounding water in addition to economic reasons (e.g. the pH
value of mining lakes plays a major role in the corrosion of steel and concrete).
3.2 Construction
Floating houses are built on mobile ground. The buoy in fig. 9, installed on the
lake of a former opencast lignite mine, records the water waves data and other
sensors the effects to the construction of floating houses.
Another considerable load for a construction is the frost load of the pontoons
and piles (fig. 7a). By means of so-called heat pipes alternative energy is used to
avoid or at least to reduce the frost action upon the piles. Fig. 7b displays such a
manufactured heat pipe for a pile and figure 8 shows its installation with the help
of a crane.
a b
Figure 7: a: Piles subjected to freezing conditions. b: Prefabricated heat pipe.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 281
3.3 Self-sufficiency
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
282 Flood Recovery, Innovation and Response IV
Figure 11: Assembly of the heat exchanger into the pontoon by crane.
Figure 12: Scheme of the arrangement of six heat exchangers, valves and
heat pump.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 283
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
284 Flood Recovery, Innovation and Response IV
Figure 17: Measuring results of the thermal response method. Left: spiral heat
exchanger, right: compact heat exchanger – temperature of
surrounding water: 9.8 °C.
Heat storage
Solar energy can be stored in the pontoon’s space in connection with the classic
solar thermal energy. It is also possible to collect the solar energy directly
through a transparent, insulating cover in a floating storage box (fig. 18) [2]. In
the latter one the detection of temperature distribution is of interest (fig. 19).
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 285
Figure 18: Floating box with a cover of heat-insulating glass for experiments.
Figure 19: Figure 19: The influence of the asymmetric solar radiation on the
water surface (left) is eliminated by convection within the water
(right).
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
286 Flood Recovery, Innovation and Response IV
4 Outlook
Figure 21 depicts the potential of floating houses. Thanks to its mobility
principle, a floating object can be transported to another location after fulfilling
its purpose. For instance a floating stadium is used for another purpose after a
football championship. Nowadays amphibious buses already travel among
islands or between the canals of Amsterdam and the Schiphol airport (fig. 22).
Another possibility is buildings floating up in the case of floods if water
occurs. It could be an adaptation strategy for flood preparedness near river
districts in the future.
Figure 21: Design of a floating soccer stadium for the World Cup 2022 in
Qatar, Architectural Office Düsseldorf, Peter Knoebel.
References
[1] Stopp, H.; Strangfeld, P.: Schwimmende Wohnbauten, Beuth Verlag,
Berlin-Zürich-Wien 2012.
[2] Harnath, M., Heating and water supply of floating houses in compliance
with energy issues, Master thesis, University of Appl. Sciences, Hochschule
Lausitz (FH), 2011.
[3] www.floatingdutchman.nl, www.toursales.com/Floating-Bus.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 287
Abstract
In current measures taken in Europe to cope with growing flood risks, various
elements characterize the strategic and practical choices involving anticipation,
protection or mitigation. One crucial element in all flood-related projects is
space. In quantitative and qualitative aspects, most flood adaptation strategies
imply a morphological transformation of city and landscape, as well as the
redefinition of land use and status, which in its turn can lead to new deals among
territorial players. These multi-scale interplays can eventually put financial,
political and social status-quo under unknown pressure, and transform the role of
urban and landscape design, which gains in importance but also in complexity.
The nine contemporary flood-related projects reviewed reveal that the fluctuating
conditions and multiple interests in which they evolve require, in addition to
creative approaches, openness, perseverance and diplomatic skills. Landscape,
urban or architectural design becomes then an open and dynamic platform for
spatial renegotiation and adaptation, challenging design practices in flood-prone
areas as well as democratic structures.
Keywords: flood risk, urban and landscape design, adapted spatial design,
negotiation platform.
1 Introduction
Flood-related riverine projects involve two precious resources for urban
civilizations: water (the river) and space. In fact, they all imply physical and/or
cognitive redefinitions of space: local public space and civil works can become
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
doi:10.2495/FRIAR140241
288 Flood Recovery, Innovation and Response IV
regional defence infrastructures; land that seemed suitable for building can
become junk bond for investors if declared risk zone; purely agrarian areas
can turn into water storage. Along with these transformations, city and landscape
negotiate a new potential damage distribution, building up inter-linkages and
engaging into a reflexive redefinition of their respective roles. Besides, within
the city, riverine spaces generate simultaneously growing fears, waterfront
development ambitions and new functional combinations. These multi-scale
interplays, the economic and social pressures linked to them, and the diversity of
territorial players involved represent an extra layer of complexity in the remit
of spatial designers. Yet, some contemporary European flood-related projects do
recognize both space and water dynamics as crucial variables of flood adaptation
strategies. Our methodological approach was thus to review nine of these
projects, aiming to identify how spatial design has fulfilled its task of negotiation
platform. Case study was adopted as the research method, as it allows to gain a
comprehensive view of the targeted projects, thanks to its simultaneous attention
to “the complex relationships between context, product and process that govern
every design process” [1].
After introducing an overview of the multiple stakes involved in most flood
adaptation strategies (Section 2), the paper will pinpoint to how the notion of
‘river space’ has taken spatial design as a new dimension of flood management
(Section 3), implying for flood-prone territories a New Deal generated by design
(Section 4). We will then summarize some of the main roles played by spatial
design (acting as a dynamic negotiation platform) to shape these new deals in the
analysed cases (Section 5), before concluding with general implications and
prospects for future developments in flood-prone territories.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 289
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
290 Flood Recovery, Innovation and Response IV
Traditional European planning regulations long defined flood zones on the basis
of previous events and/or flood models, to then apply limitations in land use and
construction. Although still essential to most planning practices, this passive
method shows today its limits, especially in densely built flood-prone
environments. Numerous constructed obstacles have modified the contours and
behaviour of flooding. Contemporary flood zones are thus no longer determined
only by natural elements, but increasingly by the effects of man-made civil
works, earthworks, buildings or planted vegetation, as a result of past political
decisions – if not the sum of faits accomplis. Furthermore, when potential flood
space covers all or large parts of the living territory, the question cannot be
solved in simple terms of building limitations or natural floodplain restorations,
but also involves a crucial negotiating aspect to define what needs to be
floodable in order to accommodate higher discharges and to protect the most
valuable assets. Permanent river widening, as applied to the upper Rhône River,
or the creation of temporary flood spaces, such as the Dutch calamiteitenpolders
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 291
and the French champs d’inondation contrôlée (controlled flood fields) currently
implemented along the Isère River, are successful examples of diversion
strategies.
Contemporary flood adaptation programmes therefore suppose a notion of
acceptable loss of safe ground and often a notion of acceptable damage. These
two notions are both dynamic (as they cannot be exactly predefined) and
reflexive (as by accepting flooding in certain areas of a land or city, other parts
of the territory will be spared). The newly designated ‘river space’ thus
encompasses much more than the surface of the stream, but refers to the
necessary space of fluctuation, whose contours are not fixed but rather gradual
(from permanent stream, seasonal riverbed, foreshore, retention areas, flood
zones); each level of permeability allowing different activities to take place, as
long as primary hydraulic functions are guaranteed. In this context, the
redefinition, expansion or transformation of flood-prone areas exclude any
purely objective, unique and final configuration, but imply complex negotiations,
painful arbitrages and dynamic designs to reach optimal risk-safety distribution
and land valorisation. In the investigated projects (listed in Table 1), this
redefinition resulted in a new territorial deal among owners, users and
beneficiaries of the adaptation project, involving both material elements (land
and infrastructure) and immaterial ones (value of areas and degree of risk
allocated to them), all merged into a new territorial structure. The studied cases
have shown that this new deal has clear implications on, among others, the limits
of flood-prone areas, land statuses, related rules, as well as on practices of
riverine users, as presented below.
Flood adaptation
Country River Location Spatial type
intervention
BE Scheldt Antwerp Floodplain expansion Intra-urban park
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
292 Flood Recovery, Innovation and Response IV
While river space has long been defined as a negative space, progressively
reduced to maximize productive areas and expand building lands, the current
shift from flood defence to flood adaptation implies a reverse approach that first
defines the space needed to accommodate expected high waters, and
subsequently seeks to adapt the surrounding areas to provide the needed
capacity. Contrarily to the passive definition of flood zones, the definition of
adaptive measures, though elaborated with scientific tools, remains in essence a
political choice in its spatial translation, which implies a consensus on the
principle of the intervention and its perimeter. The interventions decided within
the Room for the River programme, located along river courses, were motivated
by the raise of national norms for river capacities that followed the 1993–1995
near flooding along the Rhine and Meuse rivers. For each measure, the type of
intervention, the financial means and expected effects on water level were set,
while the precise definition of river space and flood areas was left to regional and
local players, in collaboration with the national water authorities. The definition
of the new flood zones within the Noordwaard area and Overdiepse polder
eventually incorporated various elements: efficiency in hydrological terms, cost
targets, spatial quality, ecological value and sustainability of remaining farms
and dwellings. In the Swiss upper Rhône Valley, the decision of widening the
riverbed in order to increase its capacity was set by the Canton authorities at
the turn of the century, but various options can still be implemented locally,
including reinforcing the existing dikes or dredging the river. There again,
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 293
agreeing on the general means and objectives sets a discussion frame, but merely
opens the negotiations that will ultimately modify the area where water will be
allowed to fluctuate with more or less freedom.
Figure 1: Spatial adaptation in the Noordwaard area – Room for the River,
the Netherlands (Rossano).
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
294 Flood Recovery, Innovation and Response IV
the most obvious urban land use to absorb flood adaptation projects. However,
status change can also take place in the opposite direction, from flood area to
building land within new conditions (Bordeaux), or from restricted floodplain
to open public recreation space (Munich), which in its turn increases quality of
life and values up the immediate surroundings. However, as the Munich case
shows, new status generates new practices, which are not always foreseen, as the
new combinatory land uses overlap various regulations and mores. Local
authorities welcome the success of the new Isar River space, but simultaneously
struggle to control crowds’ behaviour in what is altogether a new kind a public
park, a nature area and still a floodplain.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 295
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
296 Flood Recovery, Innovation and Response IV
river” [12]. He also stressed that “the exhibition was a success, and people really
appropriated the term Sion-sur-Rhône”, title given to the design studio and to the
following publication [13]. A design competition for the reconfiguration of the
public spaces along the Rhône was eventually organized a year later, and is, right
now, in the process of further detailing by the winning design team. Although the
initial design envisioned by ETH Zurich’s students was not literally adopted, it
bridged the gap between City and Canton at a crucial moment in the planning
process (namely when urban ambition and flood mitigation could be connected
for the benefit of both), and it mobilized a population that hitherto had shown no
interest in what they saw as an abstract and purely technical issue.
The design competition held in Munich for the most central segment of the
Isarplan shows how spatial design can reveal latent expectations and oppositions.
Differently from the strictly internal process that supported the restoration of the
southern part of the river, the project commissioners decided in 2006 to organize
a landscape design competition on the Isar segment crossing the city centre. The
winning design envisioned a central linear sculptural element separating
the main stream and the new recreational open space, acknowledging the existing
technical constraints that would make impossible to set the river free. The second
prize was awarded to a completely different proposal, with an organic, informal
design. Yet, a public quarrel followed, showing that a significant part of the
population had expected a more spectacular ‘renaturation’ project and rejected
the urban aspect of the winning design. A period of intense and often emotional
discussions followed, involving city, local districts, water board and citizens,
giving the opportunity to express wishes and constraints, and eventually leading
to a consensual proposal that guaranteed the safety of urban infrastructures and
still suggested a certain natural freedom, most wanted inside the city. In the
words of the former Head of the city planning department, “it was important to
show, on the one hand, how little freedom there is when the river is so important,
but, on the other, to speak with people about this limited freedom we have,
because people perhaps expected something much more impressive. (...) These
competitions were more an education project. It was necessary to communicate”
[14]. Through an intense debate that could only have been ignited by concrete
proposals, the Isarplan left the secluded world of environmental and
technological expertise, on the one hand, and the realm of romantic dreams, on
the other. The design was thus not the result of a predefined image, but initiated
new perspectives leading to alternative trajectories.
The opening of the design process to a wider panel, in the programmatic phase,
offers the opportunity for participants to match more general expectations and
possibilities with specific options of spatial configuration. Shared scenarios can
then be created through the discussion about the distribution of land
and investments, and the elaboration of the envisioned spatial framework and
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 297
physical interventions. This matchmaker role was well illustrated by the process
of elaboration of the Meerstad project: during the planning workshops,
participants received basic programme elements (represented by on-scale pieces
of coloured paper, proportional to the land requested for water, wetlands, woods,
housing and industry), and by playing with them onto the area’s map, they could
quickly elaborate spatial distribution scenarios. Apart from allowing all present
players to envision their preferred options, this scenario-based participative
approach made them conscious of the difficulty of combining various elements
into a legible and attractive spatial framework. The community participation in
the design process that took place in the Overdiepse polder was even more
radical, as local inhabitants actually anticipated the planning process, right after
the area was identified by the government as a suitable floodplain extension.
They had to deal with a relatively simple equation (but a sensitive matter), as it
was clear that not all 16 existing farms could sustain their activity in the area.
They grasped the chance to develop their own plan, with support from the
national and regional authorities, choosing the most convenient project from
their point of view (the reconfiguration of the whole polder into a temporary
expansion space for high waters with nine heightened platforms for the future
farms). In this particular case, the design allowed the building of a consensus not
only between authorities and farmers, but first of all among local players
themselves, who could better deal with the economic and human aspects of the
project, and translate them into an agreed and shared framework.
6 Final considerations
Territorial design, by nature, is a complex task of organizing multiple collective
intentions, uses, desires, possibilities and constraints in a balanced, sensitive and
also inspiring spatial arrangement. Yet, when the existing constraint is linked to
riverine flood risk, the designers’ task is made even more challenging, as the
possibility to turn flood proneness into a great spatial opportunity is latently
offered. The analysed projects showed that despite all its complexities, the
design of flood-prone spaces can be performed as an open negotiation platform.
As a dynamic process, design is allowed to evolve: it can take into consideration
natural fluctuations as well as ever-changing sociocultural aspects, and can also
orchestrate the interdisciplinary approach needed to balance (apparently)
concurrent objectives with different time horizons. As an open platform,
designing river spaces comprises two complementary characteristics: by
fostering a wider participation, it promotes a valuable interchange of inputs
between stakeholders and designers, where unforeseen combinatory options can
actually emerge. On the other hand, by reintroducing the free will that
characterizes the practice of spatial design [15], it facilitates the emergence of
collective choices and consensual territorial visions beyond problem solving. Not
only this can eventually increase players’ sense of project ownership but surely
acts as a powerful sensitization tool, bringing flood risk closer to people’s daily
lives, and helping democratic societies build positive and shared answers to
flood risk challenges.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
298 Flood Recovery, Innovation and Response IV
Acknowledgements
This study was funded by the Swiss National Science Foundation and the
Portuguese Foundation for Science and Technology, through PhD grants. The
authors are thankful to all designers and territorial players who generously
provided information to both PhD researches.
References
[1] Foqué, R., Building knowledge in architecture. UPA: Brussels, p. 174,
2010.
[2] Harries, T., Feeling secure or being secure? Why it can seem better not to
protect yourself against a natural hazard. Health, Risk and Society, 10(5),
pp. 479-490, 2008.
[3] Klijn, F., Bruin, D., Hoog, M.C., Jansen, S. & Sijmons, D.F., Design
quality of room-for-the-river measures in the Netherlands: role and
assessment of the quality team (Q-team). International Journal of River
Basin Management, 11(3), pp. 287-299, 2013.
[4] Viganò, P., Extreme cities and bad places. International Journal of
Disaster Risk Science, 3(1), pp. 3-10, 2012.
[5] Barroca, B. & Hubert, G., Urbaniser les zones inondables, est-ce
concevable? Développement Durable et Territoires.
http://developpementdurable.revues.org/index7413.html
[6] De Meulder, B. Personal communication, 5 October 2013,
Architect/Urbanist, WIT Architecten, Louvain, Belgium.
[7] Ruimte voor de rivier. Samen werken aan een veilig en mooi
rivierengebied.
http://www.ruimtevoorderivier.nl/media/78773/corporate_brochuredefdoo
rlopend.pdf
[8] Ministerie van Verkeer en Waterstaat, Ministerie van Volkshuisvesting,
Ruimtelijke Ordening en Milieubeheer, Ministerie van Landbouw, Natuur
en Voedselkwaliteit. Nationale Waterplan 2009-2015, pp. 6, 71-73, 2009.
[9] Warner, J., Edelenbos, J. & van Buuren. A., Making space for the river:
governance challenges (Chapter 1). Making space for the river, eds. J.
Warner, A. van Buuren & J. Edelenbos. IWA: London, pp. 1-13, 2012.
[10] Schreuder, A. Wie woont er graag in de calamiteitenpolder?, Rotterdam,
02.12.2002, http://vorige.nrc.nl/binnenland/article1554360.ece
[11] Van Hall, A. Personal communication, 13 March 2013, Dike Master,
Water Board Hunze & Aa’s, Veendam, the Netherlands.
[12] Gross, D. Personal communication, 24 July 2013. Municipality Urbanist,
Sion, Switzerland.
[13] Girot, C., Rossano, F. & Duner, I., Sion-sur-Rhône, un nouveau paysage
pour la vallée du Rhône à Sion. GTA: Zurich, 2010.
[14] Tahlgott, C. Personal communication, 25 June 2013. Former Head of the
planning department, Munich, Germany.
[15] Sijmons, D., Mind the gap (Chapter 2). Rising waters, shifting lands, eds.
C. Girot & F. Rossano, GTA: Zurich, pp. 9-12, 2012.
WIT Transactions on Ecology and The Environment, Vol 184, © 2014 WIT Press
www.witpress.com, ISSN 1743-3541 (on-line)
Flood Recovery, Innovation and Reponse IV 299
Author index
Adegoke P. B. ............................ 23 Joseph P. .......................... 175, 201
Ahilan S. ............................ 13, 113 Jukrkorn N. ................................ 75
Al Khaddar R. M. ...................... 23
Allen D. ................................... 113 Kang J. ....................................... 87
Anton F. ................................... 239 Kilsby C. .................................. 113
Ardagna D................................ 215 Kim H. ....................................... 89
Arias C. .................................... 215 Kobayashi K. ........................... 251
Arthur S. .................................. 113 Kocmanova P........................... 263
Atherton W. ............................... 23 Kopecny L. .............................. 263
Atun F. ..................................... 215
Lamond J. ................101, 113, 125,
Beddoes D. W. ......................... 151 ................................. 163, 175, 187
Berg H...................................... 227 Lawson E. ................................ 113
Bhattacharya-Mis N. ................ 163
Booth C. A. .............................. 151 McGillivray E. ......................... 239
Burian F. .................................. 263 Mant J. ..................................... 113
Maskrey S. ............................... 113
Cameron D. .............................. 139 Mazuran M. ............................. 215
Cameron R. .............................. 201 Mezouaghi M........................... 239
Carnacina I. ................................ 49 Minucci G. ............................... 215
Chen J.-C. .................................. 35 Mioc D. .................................... 239
Choi C. ....................................... 87 Mofford L. ............................... 239
Chuang M.-R. ............................ 35 Mohssen M. ................................. 3
Coates G. ................................... 13 Molinari D. .............................. 215
Mount N................................... 113
Ebeltoft M. ............................... 227
Everett G. ......................... 101, 113 Nielsen J. ................................. 227
This book presents the results of an ambitious research activity designed to understand
why Early Warning Systems (EWSs) fail. However, from the beginning, the objective
of the research proved to be challenging for two reasons. First, as yet there is not
a shared understanding of what an EWS is among either research or practitioner
communities. Second, as a consequence, it is equally unclear when an EWS can be
considered successful or not. Because of this, the research needed first to define EWS
and identify its components, functions, peculiarities, and weak points. Only at that
point was a first attempt to evaluate EWSs performance possible.
Flood Early Warning Systems Performance is organised according to the conceptual
steps required by the research. In part I the “open questions” about the definition and
the role of EWSs are handled, the aim being the identification of how to evaluate
EWSs effectiveness/performance. Part II focuses on the real aim of the research,
providing concepts and tools to assess EWSs performance; suggested tools are also
implemented in a case study to describe how they can be applied in practice. The
sections are independent of each other to allow readers to focus only on the content
they are most interested in.
The book is designed for a wide audience. The book can serve as a sort of manual
for EWSs designers, managers, and users, but also has appeal for general readers
with an interest in the subject. While the focus of the book is flood risk in mountain
regions, most of the results can be applied to other hazards as well.
Traditionally early warning systems (EWSs) have been identified with monitoring and
forecasting systems and their assessment has therefore focused only on the accuracy of
predictions. The authors propose a shift in thinking towards the more comprehensive
concept of total warning systems, where monitoring and forecasting systems are
coupled with risk assessment, emergency management and communication aspects.
In line with this, a new approach to assess EWSs is proposed that is based on system’s
capacity of reducing expected damages, with the hope that improved EWSs will result.
ISBN: 978-1-84564-688-2 eISBN: 978-1-84564-689-9
Published 2013 / 196pp / £84.00
...for scientists by scientists
Tsunami
From Fundamentals to Damage Mitigation
Edited by: S. MAMBRETTI, Universidade Estadual de Campinas, Brasil
This volume is the first in a new series that covers various aspects of Safety and Security
Engineering with the aim of developing a comprehensive view on risk mitigation. This
volume is devoted to floods, since one-third of annual natural disasters and economic
losses, and more than half of the victims of natural disasters are flood-related.
The risk from flooding, and the demand for protection from it, has been growing
exponentially as a result of a burgeoning global population and growing wealth,
climate change and urban development. These factors make it imperative that we
change the way flood risk is managed.
Knowledge and scientific tools play a role of paramount importance in the strain of
coping with flooding problems, along with capacity building in the context of political
and administrative frameworks. Therefore, governments need to establish clear
institutional, financial and social mechanisms and processes for flood risk management
in order to ensure the safety of people and property and, thereby, contribute to both
flood defence and sustainable development.
The present volume contains selected papers presented at Conferences organised by
the Wessex Institute of Technology. The papers have been revised by the Authors to
bring them up to date and to integrate them into a coherent understanding of the topic.
It covers: Risk Assessment; Mathematical Models for Flood Propagation; Effect of
Topographic Data Resolution; Social and Psychological Aspects; Decision Making
and Management; Legislations and Directives; Alternatives in Flood Protection;
Response and Recovery; Damages and Economic-related Problems; Case Studies
The quality of the material makes the volume a most valuable and up-to-date tool
for professionals, scientists, and managers to appreciate the state of the art in this
important field of knowledge.
Series: Safety & Security Engineering
ISBN: 978-1-84564-646-2 eISBN: 978-1-84564-647-9
Published 2012 / 160pp / £65.00
This page intentionally left blank