Innovative Solutions for the Delta

Innovative Solutions for the Delta
The DeltaCompetition: Creative ideas from students all over the world
for the sustainable development of densely populated delta areas
deltacompetition
2006
Innovative Solutions for the Delta
The DeltaCompetition: Creative ideas from students all over the world
for the sustainable development of densely populated delta areas
deltacompetition
2006
Innovative
Solutions
for the Delta
contents
foreword
6
preface
7–9
panel of judges CVs
11–14
Paper 1. Floating City IJmeer
Accelerator for Delta Technology
Author: R. de Graaf, M. Fremouw, B. van Bueren, K. Czapiewska, M. Kuijper
15–34
Paper 2. Sustainable Growth in Urbanised Delta Areas
The opportunities of a geographical approach to the Pearl River Delta
Author: G. van Rens, A.L. Nelissen, C. Schamhart, N. Lugt
35–56
Paper 3. All You Need is Space
Long-term and large-scale sustainable development in deltaic areas
Author: L.M. van der Burgh, M.B. Brommer
57–76
Paper 4. The Flood House Concept
A new approach in Reducing Flood Vulnerability
Author: H. Vreugdenhil, L. Meijer, L. Hartnack, T. Rijcken
77–94
Paper 5. A Hydrologic Flood Forecasting System for Mesoamerica
Author: J.E. Villalobos, P.E. Rodriguez, P. Saksa
95-–110
Innovative
Solutions
for the Delta
foreword
deltas in the world
Dear Reader,
Half of the 6.5 billion people on our planet live in densely populated delta areas. These areas are the most
dynamic and fragile areas we have. Many deltas contain fast growing and economically flourishing cities. These
conglomerations desperately need space for work, living and recreation. But the delta also provides agricultural
land, which we urgently need to feed all the hungry mouths. At the same time, the delta is threatened. Deltas are
enormously sensitive to flooding and coastal erosion. The rising sea and river levels increase the risks to people
and nature. During every hurricane season we see heartrending pictures of the full effects of flooding. The disaster
in New Orleans also made the wealthy nations face the facts. No one can remain shielded from the effects of
climate change and extremes of weather. I can safely say without exaggeration that the utmost will be demanded
of mankind if we are to protect our fragile delta areas in the years to come. Existing plans for structuring and
protecting deltas are no longer enough. New solutions are required.
‘Innovative Solutions for the Delta’ is a summary of the five best entries in the Royal Haskoning
DeltaCompetition. Right from the outset, the sustainable development of delta areas has been a key theme in
Royal Haskoning’s work. Therefore, as part of our anniversary, we have initiated the DeltaCompetition in which we
have challenged students from all over the world to come up with new, innovative and sustainable solutions for
the threatened deltas.
The DeltaCompetition has encouraged talented students to devise new ideas. In all papers that were
submitted, the devotion to find a solution for the growing problems in the Deltas of the world is heart warming.
This book brings together the five most promising ideas. From floating cities in the Netherlands to protection
strategies for the Pearl River in China and hydrologic flood forecasting systems in Mesoamerica, but also long-term
and largescale development in deltaic areas and a new approach in reducing flood vulnerability.
The ideas stem from collaboration between different fields of expertise. From water managers and civil
engineers to architects and aerospace engineers. The solutions presented in this book are all of a multidisciplinary
nature, which we have in particularly encouraged. After all, the problems that are important in the delta are
pre-eminently problems that transcend discipline boundaries and require solutions with contributions from
several areas of expertise.
I hope you enjoy reading this book and that the solutions outlined inspire you to work with us in the further
development of creative and sustainable solutions to the threats facing our delta areas.
Jan Bout,
Chairman of the Board of Management
Royal Haskoning
Innovative
Solutions
for the Delta
preface
introduction
The DeltaCompetition aimed to stimulate PhD, master’s and bachelor’s programme students from all over the
world to develop innovative, sustainable solutions to the threats and issues facing densely populated delta
areas because of climate change. In particular we were looking for new, inspiring and bold ideas and unexpected
solutions to these issues and problems, trying to find a combination of ideas from different disciplines with
surprising results that could contribute to the sustainable development of delta areas.
The call for papers in January 2006 resulted in the registration of 33 teams from the UK, the Netherlands,
Germany and the USA to Zimbabwe, Bangladesh, Vietnam, Indonesia and China. All over the world bachelor’s,
master’s and PhD students formed multidisciplinary teams to participate in the DeltaCompetition. In July 2006
13 teams submitted papers that met the requirements of the DeltaCompetition. Of these, seven came from teams
studying at universities in the Netherlands. The others came from such universities as the Asian Institute of
Technology (Thailand), the University of Newcastle upon Tyne (United Kingdom), Louisiana State University (USA)
and Birzeit University (Palestinian Territories). Their solutions ranged from sustainable infrastructure in flood-prone
areas, flood forecasting systems and methods to prevent coastal erosion to integrated planning systems, spatial
development strategies, floating cities and solutions for increasing risk awareness and preparedness. The panel of
judges got to work on these 13 papers.
We paid particular attention to the presence of a sound scientific basis, original or innovative elements in the
solution and the quality of the presentation of the paper. Of the 13 papers submitted, five clearly stood out, scoring
very well on these criteria. These papers are included in this book.
In their paper “A hydrological flood forecasting system for Mesoamerica”, Jose Villalobos, Patricia Rodriguez and
Philip Saksa from Lousiana State University have come up with the idea of developing an internet-accessible, userfriendly hydrological system that performs real-time flood forecasting for the Mesoamerican region. The system
uses publicly available satellite remote sensing and geographical information to obtain the data necessary to
process hydrologic models with the ability to predict flooding in coastal areas. Its uniqueness lies primarily in the
incorporation of satellite rainfall estimates. Another positive feature of the idea is its applicability not only in the
Mesoamerican region but also in other flood-prone areas in the world with limited access to relevant data.
The Flood House Concept constitutes a new approach to reducing vulnerability to flooding. In their paper,
Heleen Vreugdenhil, Leo Meijer, Lars Hartnack and Ties Rijcken from Delft University of Technology present their
ideas of combining a memorial to Hurricane Katrina with a way to learn lessons and come up with measures
to prevent future disasters. To achieve this, the team has come up with the flood house, a flood-proof water
management centre in which people, science, government and business come together to learn, discuss and create
new ideas about water management and flood related issues.
Innovative
Solutions
for the Delta
Both papers draw on a sound scientific basis and have sought solutions to improve awareness of and readiness to
face the threats facing delta areas because of climate change. Both papers show ingenuity and are well-written
and presented. However, the three other papers proved to be even better.
Lisette van der Burgh (University of Twente) and Marit Brommer (Delft University of Technology) joined forces
for the DeltaCompetition to come up with an integrated approach to incorporating long-term morphological
processes as well as the socio-economic development of delta areas in order to constitute a more sustainable basis
for policy and decision making. In “All you need is space, long-term and large-scale sustainable development in
delta areas”, they argue that although sea level rise constitutes a major threat to delta areas, much more attention
should also be paid to long-term, gradual morphological processes that shape coastal configurations. Moreover,
while current policy focuses much more on the relatively short-term development of the coastline, we should
adopt a more strategic approach to the long term, including the effects of human interference on coastal evolution
models and scenarios. It was primarily the scientific basis and originality of this paper that made us decide to
award the third prize of the DeltaCompetition to Lisette van den Burgh and Marit Brommer.
Another team from Delft University of Technology came up with a geographical design approach to spatial
and urban development in the Pearl River Delta, focusing on the city of Guangzhou. Geert van Rens, Anne Loes
Nilissen, Carianne Schamhart and Nadia Lugt effectively show in their paper that urban expansion in this rapidly
developing delta could be better structured by an ecological zone that doubles as a green river. Not only will this
reduce flood risks, but it will also make the city much more attractive and probably increase the value of adjacent
houses and offices. Their geographical design approach is applicable anywhere and makes it possible to identify
opportunities for interaction between functions and design solutions that fit well into their environment.
Like the three other papers from Delft University of Technology, which were written in a combined and
coordinated effort1, the presentation of the paper is excellent. Their analysis effectively integrates two relevant
scales of analysis – that of the basin and of the city – and applied a layer approach in their analysis and
recommendations for future planning and development, which the panel of judges considered to be very
innovative. Moreover, the geographical design approach is also relatively easily applicable and can be adopted
in all other urbanized delta areas in the world. Therefore, the panel of judges awarded the second prize in the
DeltaCompetition to this team.
The most original, innovative as well as attractive paper was submitted to the DeltaCompetition by Rutger
de Graaf, Michiel Fremouw, Bart van Bueren, Karina Czapiewska and Maarten Kuijper from Delft University of
Technology. Their paper “Floating city IJmeer, accelerator for delta technology” is not only highly visionary, but also
provides us with a pragmatic management approach to realize their long-term, sustainable solution to typical
urban pressure in a typical delta - the IJmeer in the Netherlands. Moreover, their innovative floating city concept
addresses a very urgent need to meet urbanization needs in the Amsterdam-Almere region. We believe that this
paper is fine reading material for decision makers and stakeholders currently involved in this process. It combines
technical solutions with a management approach based on transition theory, which is currently very popular
in (public) management sciences. Finally, the paper is very accessible and the team has worked out inspiring
graphical displays of their floating city. All in all, this paper is a highly innovative solution to an urgent problem in
a typical delta area without losing sight of practical applicability by integrating technical, environmental, social,
economic and public administration knowledge. It is not surprising that the panel of judges came to a unanimous
decision to award the first prize of the DeltaCompetition to “Floating city IJmeer”.
Innovative
Solutions
for the Delta
The panel of judges wishes to thank and congratulate all participants in the DeltaCompetition. Not only has the
DeltaCompetition inspired young students all over the world, but the solutions they have come up with are very
promising indeed. We especially recommend reading these five papers to all experts, decision makers and other
stakeholders involved. We believe that these contributions constitute a source of inspiration for urgent problems
facing many densely populated delta areas throughout the world. And we encourage the students who have
participated in the DeltaCompetition, and especially the prize winners, to continue their work and maintain their
inspiration and enthusiasm when they graduate and become professionals and decision makers themselves.
On behalf of the Panel of Judges,
Prof. Ir. L. de Quelerij, Chairman
1The
DeltaSync team consisted of four teams from Delft University of Technology that each submitted a paper. DeltaSync was
founded and coordinated by Ties Rijcken. Gert Jan de Werk from the Faculty of Technology, Policy and Management was their
reference teacher.
Innovative
Solutions
for the Delta
Innovative
Solutions
for the Delta
10
panel of judges CVs
Prof. Ir. L. de Quelerij
Louis de Quelerij was born on 29 July 1952 in Vlissingen, the Netherlands. He studied civil engineering at the Delft
Technical University of Technology and obtained his Masters degree in 1976.
From 1976 to 1986, he was employed by the Ministry of Transport, Public Works and Water Management at the
Rijkswaterstaat Department. From 1986 to 2002 he worked at Fugro Ingenieursbureau B.V., one of the operating
companies of Fugro NV. As managing director, he was responsible for the geotechnical services in the Netherlands.
He has been working for Fugro as director for one day/week from October 2002 to date. From that same date he
has been professor of Civil Engineering and dean of the faculty of Civil Engineering and Geosciences at the Delft
University of Technology. He has more than 25 years experience as geotechnical consultant in many infrastructure
projects.
He published over 30 international papers related to hydraulic engineering, sheetpiles, dynamic compaction,
waste materials, maintenance and risk analysis.
He is the final editor of the Dutch Guideline for Foundation Engineering (SBR).
He is a member of the Dutch Royal Institute of Engineers (KIVI), Netherlands Society for Soil Mechanics and
Geotechncial Engineering (NSSMGE), registered engineer of the Professional Association of Dutch Consulting
Engineers (ONRI).
Currently he is chairman of the KIVI/NIRIA department Geotechnics, the BSIK research programme Delft
Cluster, the TU Delft Research Centres Water, Earth and Mobility.
w w w. c i t g . t u d e l f t . n l
Prof. Dr. R. J. Nicholls
Robert Nicholls was born in 1958 in the United Kingdom. He obtained his PhD on “The Stability of Shingle Beaches
in the Eastern Half of Christchurch Bay” in 1985 from the University of Southampton. He has worked in the
University of Plymouth, the University of Maryland, Flood Hazard Research Centre at Middlesex University, and
since 2004 has been Professor of Coastal Engineering in the University of Southampton.
His main research interests are long-term coastal engineering and management, especially the issues
of coastal impacts and adaptation to climate change, with an emphasis on sea-level rise. This includes lead
authorship of chapters in four reports of the Intergovernmental Panel for Climate Change (IPCC): Second
Assessment Report (1996); the Regional Assessment (1998); the Special Report on Technology Transfer (2000); and
the Third Assessment Report (2001). Currently, he is convening lead author for the “Coastal Systems and Low-Lying
Areas” chapter in the IPCC 4th assessment, which is due to be published in 2007.
He participated in the UK-funded “Fast Track” assessments of climate change as the coastal expert, led the
SURVAS Project which reviewed vulnerability of coastal zones around the world from 1999 to 2001 and is leading
the coastal research theme in Phase II of the Tyndall Centre for Climate Change Research.
He was a lead investigator on the DINAS-COAST Project which lead to the development of the DIVA tool for
sub-national to global vulnerability assessment.
w w w. c i v i l . s o t o n . a c . u k /m a i n _ c i v i l . h t m
Innovative
Solutions
for the Delta
11
Prof. Dr. P. Hooimeijer
Pieter Hooimeijer (born in 1955) obtained his PhD on ‘Demographic Change, Household Evolution and the Housing
Market’ from the University of Utrecht with Marianne Linde in 1988.
From 1983 to 1984 Dr. Hooimeijer worked as Lecturer of Statistics at the Department of Geography Free
University Amsterdam. He was Research Assistant at the Faculty of Geographical Sciences, University of Utrecht in
1985. From 1986 to 1988 he worked as NWO Research Fellow at the Faculty of Geographical Sciences University of
Utrecht.
Dr. Hooimeijer was Director of Education of the Netherlands Graduate School of Housing and Urban Research
from 1988 to 1990.
From 1991 to 1995 he worked as Associate Professor of Demography and Population Geography at the Faculty
of Geographical Sciences Utrecht University and as Professor of Demography at the Institute of Planning and
Demography at the Faculty of Environmental Sciences University of Amsterdam.
From 1995 to 1998 he was Professor of Demography at the Faculty of Geographical Sciences Utrecht University.
From 1999 to 2000 he worked as Scientific Director of the “Netherlands Graduate School of Housing and Urban
Research” and Professor of Demography at Utrecht University.
From 2001 to 2005 he was Dean at the Faculty of Geographical Sciences and Professor of Demography at
Utrecht University.
At this moment Dr. Hooimeijer is Professor of Human Geography, Head of Department Human Geography &
Planning, Scientific Director “Netherlands Graduate School of Urban and Regional Research”.
w w w. g e o . u u . n l
Prof. Dr. R. A. Meganck
Prof. Richard A. Meganck is the director of the UNESCO-IHE Institute for Water Education, the largest water
education facility in the world. Prof. Meganck is a US citizen whose career spans twenty-five years of experience
in international development and management of natural resources. He was Assistant Secretary General and
Director for Sustainable Development and the Environment at the Organization of American States. In this
capacity he managed over twenty major projects in trans-boundary water resources, renewable energy, natural
hazard vulnerability reduction, climate change/sea level rise, public participation in decision-making and
biodiversity management.
Prof. Meganck was also the founding Director of the United Nations Environmental Program International
Environmental Technology Center in Osaka, Japan. He served as the Director and Regional Representative of the
United Nations Environmental Program for Asia and the Pacific in Bangkok.
Prof. Meganck received his BS and MS from Michigan State University and received his PhD from Oregon State
University. He is the author of more than 70 journal articles and published reports and has published more than
100 scientific papers.
Prof. Meganck serves as an international advisor for several organizations including the Corporate Board of
Advisors of the United Nations University and the Board of Governors of the World Water Council.
w w w. u n e s c o - i h e . o r g
Innovative
Solutions
for the Delta
12
Dr. J.M. Visser
After visiting the Netherlands as part of the New Orleans delegation and hearing about the DeltaCompetition,
Dr. Jenneke Visser, associate professor at the Coastal Ecology Institute of the Louisiana State University also joined
the team of jury members.
Dr. Visser is particularly involved in coastal restoration and wetland protection research and projects. Among
others, she is modeling coastal processes and landscape dynamics in the Louisiana Delta Plain and providing
university assistance in the development of the Coastal Wetland Planning, Protection and Restoration Act.
Dr. Visser was born in the Netherlands, where she did her bachelors and masters in Biology, at the University of
Amsterdam. After that she moved to the USA and focused on marine sciences during her PhD at the Louisiana
State University. In 2004 she became an associate professor research at this University.
C.Th. Smit, LLM, MSc
Tom Smit is Member of the Management Council of Royal Haskoning and the Director of the Spatial Development
Division of Royal Haskoning. Besides his management position in the firm, he is management consultant and
senior legal advisor in the field of water, spatial development and the environment. As a legal advisor he has been
responsible for drafting a large number of water and environmental laws, among others in:
>> Greece, advisory services for the Ministry of Environment concerning the new Environmental Law (1983-1986),
>> Egypt, advisory services for the Ministry of Water Resource Management concerning waterboard policies and
rules (2001-2002),
>> Romania, advisory services for the Ministry of Environment concerning environmental and safety policies and
rules (1998),
>> Bosnia- Herzegovina, advisory services for three Entity’s, an Organization of Independent Supervisors and the
EU concerning the River Basin Authority policies and their implementation (2001 - 2002),
>> Netherlands, responsible for the extension and transition of the general terms of the Environmental Hygiene
Law to the Environmental Management Law (1980-1986).
w w w. r o y a l h a s k o n i n g . c o m
Innovative
Solutions
for the Delta
13
Dr. Ir. M. J. van der Vlist
Maarten van der Vlist, who was born on 25 October 1954, studied town and country planning at Wageningen
University and until 1999 he worked there in the Spatial Planning Development Department. In 1998 he obtained
his doctorate with a thesis on ‘Duurzaamheid als planningsopgave’ (sustainability as a planning challenge) in
which he investigated the area-focused coordination of spatial, environmental and water policy on a regional scale.
Maarten van der Vlist has been a vision and development consultant on the staff of the Director-General for
Public Works and Water Management of the Netherlands since April 2005. Prior to this Mr van der Vlist, worked for
six years at the Institute for Inland Water Management and Waste Water Treatment (RWS-RIZA) in Lelystad
(1999-2005). He was head of the Space and Water Department.
This relationship between water and space was also at the core of the scenario project ‘Extreme toekomst:
waterlast of waterlust’ (extreme future – the water we want and the water we don’t). He also contributed to the
trend report ‘Koersen op tijdgeest’ (setting a course in tune with the times) produced by the Ministry of Transport,
Public Works and Water Management. This knowledge formed part of the basis on which he advised policy
directorates in The Hague on water policy.
w w w.v e r k e e r e n w a t e r s t a a t . n l
Dr. A. Datta
Dr. Anjan Datta, born in 1952, studied Economics and Development Studies. After receiving a Masters in Economics
with Honours, he started his career as a development worker and later moved into research leading to a PhD in
Development Studies.
In 2002 he joined the United Nations Environment Programme as Programme Officer and currently acts as the
Head of Non-Point Source Pollution Unit in UNEP Coordination Office of the Global Programme of Action for the
Protection of the Marine Environment from Land-based Activities.
Prior to joining UNEP, during 1976-1999 he worked with various research institutes, UN agencies, the World
Bank, NGOs and bi-lateral donor agencies. During this time he also worked with the Institute of Social Studies, The
Hague, the Netherlands and the MacArthur Foundation in the USA as a Fellow. In mid 1999 he joined the University
of Leeds, United Kingdom as a staff member. During his tenure in the University of Leeds he acted as a Research
Coordinator for the South Asia Sustainable Livelihood Policy Research Programme, and also worked as a Senior
Fellow at the Centre for Water Policy and Development housed within the School of Geography.
Over the years he published extensively on land and water resources management. He is the author of
three books, published numerous articles for academic journals and edited books, and written many research
monographs.
w w w. u n e p . o r g
Innovative
Solutions
for the Delta
14
FLOATING CITY IJMEER
ACCELERATOR FOR DELTA TECHNOLOGY
Floating City IJmeer
Accelerator for Delta Technology
PrizewinnerofTheDeltaCompetition2006
RuTGER DE GRAAF (TEAM LEADER)
PhD Student Water Management
MICHIEL FREMOuW
Msc Student Building Technology
BART VAN BuEREN
Msc Student Architecture & Building Technology
KARINA CZAPIEWSKA
Msc Student Real Estate & Housing
MAARTEN KuIJPER
Msc Student Civil Engineering
Innovative
Solutions
for the Delta
1
Abstract
Climate change, sea level rise, population growth and ongoing urbanization result in higher vulnerability of
the Rhine Delta because it will result in increased flooding frequency, increasing investments and increased
use of water, energy and other resources. The Rhine Delta also faces strong competition from thriving
economies such as China and India. After agriculture and industry, services are moving away to low cost
countries as well. Conventional urbanization increases flood risk. This study presents a more self-supporting,
non risk increasing concept of urbanization in the Almere-Amsterdam region, that will contribute to further
economic and technological development of the Rhine Delta. This transition experiment aims to learn and
build experience with non risk increasing modes of urbanization in order to contribute to the societal change
to a sustainable Rhine Delta. It uses local water and energy resources instead of external resources only.
Moreover, its addition to housing capacity does not result in an equal addition to economic damage in case
of flooding. The floating city is an accelerator for delta technology; it offers a testing ground for pilots of new
water, energy and floating technology. It will enhance knowledge based activities in the Amsterdam-Almere
region, a unique area with high economic potential. The testing ground in this region will attract knowledge
intensive hi-tech companies as well as highly educated knowledge workers. The development of the floating
city is combined with large scale wetland development in the IJmeer to create an ecological improvement.
Moreover, Floating City IJmeer contributes to reducing regional mobility and housing problems and anticipates
on an expected increased demand for water recreation. The transition management governance model is used
to place the floating city in a broader perspective and offers guidance on turning this concept into reality.
1 Introduction
The Rhine Delta is of high economic and geographic
importance; Europe’s largest harbour Rotterdam and an
economic engine of high importance, the Amsterdam-Almere
area, are located in this delta. The IJsselmeer-Waddenzee
area is a wetland of European importance for ecology. The
Amsterdam-Almere area is selected as a specific case study
because in this region, balancing economy, ecology, recreation,
housing and mobility requires development of innovative delta
concepts.
Water engineering made economic development of this
once marshy delta possible, but also resulted in an area that
is largely located below sea level. High investments and
population density have made this area highly vulnerable to
extreme flooding events with low probability (Graaf and Ven,
submitted). Estimated damage in case of sea defense failure
is 400 billion euros, an amount larger than the Dutch yearly
available national budget (Ale, 2006).
Climate change, sea level rise, population growth and
ongoing urbanization result in higher vulnerability of the Rhine
Innovative
Solutions
for the Delta
16
Fig. 1 – Dutch areas below sea level (AHN)
Delta because it will result in increased flooding frequency, increased investments and increased use of water,
energy and other resources. These resources are extracted from increasingly remote areas. As a result dependence
of external areas, including politically unstable areas, will increase. The Rhine Delta also faces strong competition
from thriving economies such as China and India. After agriculture and industry, services are moving away to low
cost countries as well. Making the switch to knowledge-intensive high value added economic activities is therefore
of vital importance for the Rhine Delta.
Conventional urbanization increases flood risk. The aim of this study is to develop a more self supporting, non
risk increasing concept of urbanization in the Almere Amsterdam region that will contribute to further economic
and technological development in the Rhine Delta. In this paper we analyze national trends, developments in
the Amsterdam region and technical innovations. We present a transition experiment of a more self-supporting
floating city in the IJmeer. This transition experiment aims to learn and build experience with non risk increasing
modes of urbanization in order to contribute to the societal change to a sustainable Rhine Delta. The floating city
is an accelerator for delta technology; it offers a testing ground for pilots of new technology.
2 Scientific background
2.1 Sustainability
Since the publication of the UN report ‘Our common future’ of the Brundtland commission (WCED, 1987),
sustainability is considered primarily in terms of continuing to improve human well being, whilst not undermining
the natural resource base on which future generations will have to depend. There is no common agreement on
the exact definition of sustainability. However, most approaches on sustainability mention balancing of interests
between social, ecological and economical aspects. (Elkington, 2001; Rotmans, 2003; Butler, 2006).
2.2 Transitions to more sustainable urban areas
More sustainable urban areas means balancing economic developments (mobility), social developments (flood
security, housing demand) and ecological developments. Urbanization that also uses local resources of water
and energy and which results in no further increase of expected damage (risk times impact) in case of flooding,
contributes to sustainability. However, changing to other modes of urbanization by system innovation is a long
term process, impeded by long expected lifetime of urban infrastructure (Butler, 1997; Hiessl et al., 2001), high levels
of invested capital, high levels of uncertainty and institutional problems (Czemiel and Hyvonen, 2002). One of the
main institutional problems is the number of institutions that are involved in urban development.
Future developments in the Rhine Delta, such as climate change and economic, technological and
demographical developments, are characterized by wide margins of uncertainty, creating fundamental problems
for policy makers and water managers. Decisions are made without knowing the exact consequences or knowing
future societal demands. A management approach to deal with complex problems under conditions of high
uncertainty is transition management (Rotmans, 2003), which is aimed at realization of societal transformations
to increase sustainability. A transition is a structural change in the way a societal system operates, and as such a
long-term process (25-50 years) (Rotmans, 2003).
Transition management thus is a governance model for long-term policies, since transitions typically take a
generation or more to unfold. It starts from the recognition that for these highly complex and uncertain problems,
blueprint solutions are impossible to develop on beforehand. Therefore, experimenting and learning-by-doing are
crucial. Executing transition experiments is about learning new modes of urbanization (system innovation) rather
than optimizing existing infrastructure. The information and experience gained by these experiments are used to
improve other experiments, knowledge and skills until there is enough know-how and preparedness to transform
current urban structure.
Innovative
Solutions
for the Delta
17
3 Analysis
For a successful transition, the condition is that developments on macro, meso and micro scale have an enhancing
effect on each other (Rotmans, 2003). Therefore, developments on three levels will be evaluated. National
developments on macro scale, regional developments on meso scale and technical innovations on micro scale.
Based on a workshop meeting with all Deltasync teams, we selected trends that are particularly important for the
Rhine Delta.
3.1 Macro level
Demographic developments
Population growth will probably continue until 2035 and stabilize just below 18 million inhabitants (CBS, 2006).
Slow population decrease will start around 2040 according to the average scenario. Moreover, the number of
inhabitants per house will also decrease from 2.31 now to 2.17 in 2020. (CBS, 2006) This will cause continued
growing of housing demand in the Netherlands for the coming decades. In the Dutch urban delta, the expected
rise in households will be much higher than the average Dutch level. As a result, housing demand in this area will
be very high.
Another important demographic development is aging of the Dutch population. The percentage of people of
65 years and older will rise from 14.4% to around 20% in 2025 (CBS, 2006). The worker-pensioner ratio currently is 4
to 1. However, in 2040 this will have increased to 2.5 to 1. As a result the financial burden for the working part of the
population will increase. Innovation and development of the knowledge economy will be important to sustain the
social system by increasing competitive capacity of the economy and production efficiency.
x mln
22
20
18
16
14
2002
2005
2010
2015
2020
2025
2030
2035
global economy
strong europe
transatlantic market
regional communities
2040
2045
2050
Fig. 2 – Long term population growth
according to various scenarios (Jong, 2004)
prognosis
Climate change
Innovative
Solutions
for the Delta
18
Climate projections in Table 1 (KNMI, 2006) indicate that in the Netherlands more rainfall can be expected in
winter whereas the summers become drier. Droughts will occur more frequently and the variation of water
resources increases. The expected summer discharge of the Rhine will decrease with 10% in an average climate
scenario and even 60% in a dry scenario (NMP, 2005). Another effect of climate change is sea level rise, this effect
combined with land subsidence and lower river discharge in summer can result in problems with salinity and
increases flooding frequency. Consequently, water intake from rivers in delta areas becomes more difficult, the
chance of water shortage increases. The frequency, at which years with very high salinity in the delta (like 1976)
occur, increases with 80% in an average climate scenario (RIZA, 2005). Overall, creating space for water storage
will be increasingly important as well as developing building technology that contributes to reduced flooding
vulnerability.
2050
G
G+
W
W+
Global temperature rise
+1°C
+1°C
+2°C
+2°C
Change air circulation pattern in Western Europe
no
yes
no
yes
Winter
Average temperature
+0,9°C
+1,1°C
+1,8°C
+2,3°C
Average precipitation amount
4%
7%
7%
14%
10 days precipitation sum with expected
exceeding frequency of once in 10 years
4%
6%
8%
12%
Highest daily average wind velocity per year 0%
2%
-1%
4%
Average temperature
+0,9°C
+1,4°C
+1,7°C
+2,8°C
Average precipitation amount
3%
-10%
6%
-19%
10 days precipitation sum with expected
exceeding frequency of once in 10 years
13%
5%
27%
10%
Potential evaporation
3%
8%
7%
15%
Absolute increase
15-25 cm
15-25 cm
20-35 cm
20-35 cm
Summer
Sea level
Table 1 - Climate scenarios for the Netherlands (KNMI, 2006)
Knowledge based economy
After agriculture and manufacturing, services are also being ‘outsourced’ to low cost countries such as India and
China. In the IT sector, for instance, outsourcing is expected to grow 44% this year (Ernst&Young, 2006). As set out
in the Lisbon strategy in 2000, the European Union should focus its efforts in the realms of science, technology and
sustainability to create growth and jobs. Europe seems to be experiencing a brain drain however, as there is a net
migration of people with a tertiary education to the US (EC, 2003).
A renewed effort is required to reach the Lisbon-strategy goals (EC, 2006), either by means of funding or by
projects. The Amsterdam-Almere region includes two universities, several research institutes and numerous hightech companies, making it an area of interest for Dutch efforts in reaching these goals. At this moment, however,
the Amsterdam area is rated at place 11 in Europe for public and private R&D investments as percentage of Gross
Regional Product (Regio Randstad, 2005). In addition, to make the step from science and knowledge towards
economic impacts, applying science in practice is necessary. Starting small scale transition experiments where
all kinds of innovative technologies can be applied, tested and further developed will enhance knowledge based
activities in this region. This testing ground will attract knowledge intensive hi-tech companies as well as highly
educated knowledge workers.
3.2 Meso level
Development of recreation and tourism
Both Amsterdam and Almere are expanding towards the IJmeer. Population increase in the region will result
in additional pressure on existing recreational locations and a demand for new ones. This is not necessarily a
negative influence on local ecology, according to a collective of nine interest groups (both governmental and
non-governmental) combined (Stuurgroep Verkenning IJmeer, 2005), because it can be combined with ecological
enrichment. Additionally, recreational activities with a low environmental impact (like sailing) can be stimulated by
building additional facilities in the region.
Innovative
Solutions
for the Delta
19
Development of the Amsterdam-Almere area
The IJmeer region is part of the Dutch ecological main structure, and plays an important role in bird migration
in Europe. ‘Stuurgroep Verkenning IJmeer’, has ascertained that the ecosystem in the region has been rapidly
declining since the early nineties, and that not intervening will result in a continuation of this process.
On the other hand, the city of Almere is continuing to grow, and next to the added pressure of housing projects,
the increase in population requires additional recreational functions in the area.
However, these two demands are not necessarily mutually exclusive. A balance needs to be found between
ecological preservation demands and housing needs. Urban and recreational developments in the IJmeer
combined with large scale nature development can strengthen both economic and ecological values in this region.
(Stuurgroep Verkenning IJmeer, 2005).
The presence of Schiphol – one of Europe’s main air hubs, the Netherlands’ second largest seaport and a
prestigious business district on the Zuidas makes the so-called Randstad ‘Noordvleugel’ an economic centre of
national importance. At this moment, the Amsterdam region is listed in the top-5 favourite places to locate a
business in Europe. (Regio Randstad, 2005). To continue growth and attract more expat knowledge workers, the
Zuidas business district would also benefit from additional high quality housing. Its unique location and innovative
nature will make the floating city a growth catalyst in the region.
The VROM council and the V&W council argue that for continued growth of Amsterdam and Almere, these
main cities in the Noordvleugel should be developed not separately but as a conjoined urban area, a twin city. The
opportunities for housing growth in Amsterdam are limited, whereas Almere can still expand to the west.
(VROM/V&W, 2006).
Demand for housing
The estimated shortage of houses in the Amsterdam-Almere area was 65.700 in 2002 (Vermazen, 2002). Building
locations in the direct vicinity of Amsterdam are necessary to reach the ambitious targets set by the municipal
board of Amsterdam. In the period to 2010, 20.000 new houses will have to be built in the Amsterdam area.
According to the head of the Housing Corporations in Amsterdam, new building sites are essential to reach those
targets. (Peijs, 2006).
Because of its location between the two large cities, the IJmeer is an area of interest in this matter. With IJburg,
Amsterdam has already created a new housing project in the area, and Almere is soon to follow with Almere
Pampus and Almere Poort along the IJmeer’s rim (Fig. 3).
Demand for infrastructure
Innovative
Solutions
for the Delta
20
The economic damage caused by traffic congestion in the Netherlands is
estimated to be a650 million in 2006, most of which originates from urban
beltways and main traffic axes (RWS-AVV, 2004).
According to Rijkswaterstaat, the province of Flevoland has the longest
living-working distances in the Netherlands and that performance of the
A6 connection between Almere and Amsterdam is among the lowest
Fig. 3 – Urban development of
(Rijkswaterstaat, 2005).
Amsterdam and Almere (Peijs,
The ministry of Economic Affairs considers motor vehicle accessibility in
structuurdocument, 2006)
the Randstad’s North Wing to be of national priority (EZ, 2006). A preliminary
study published by the ROA (Regional Body of Amsterdam) in April of 2006
suggests an improved public transportation connection between Almere and Amsterdam is a necessity for
continued development of the region. It also suggests an additional road connection over the IJmeer, while not
strictly necessary, will improve current traffic conditions in the region (ROA, 2006).
3.3 Micro level
Small scale transition experiments can contribute in achieving sustainable development and can stimulate the
development of a knowledge based economy. In this paragraph some innovative technologies with particular
opportunities for Dutch delta technology are further elaborated.
Floating structure technology
Current floating homes are constructed using concrete pontoons as a basis, due to inherent stability, low cost, no
required maintenance (Hendriks, 1999) and the ability to use the pontoon itself as a living space (Aqua Struenda,
2004).
With a few adaptations this technology should be able to provide a basis for an entire city. To this end either
large quantities of pontoons need to be prefabricated, or fewer, larger elements in a dry-dock near the site
(specially constructed for this purpose). It’s possible to create elements up to 150 meters long and wide in this
fashion (Kuijper, 2006).
Joining elements brings the added benefit of a fail-safe mechanism to the structure, because even if an
accident causes one element to lose buoyancy, its neighbours can still distribute its load. Larger elements can
benefit from this fail-safe mechanism by internal compartmentalization. The resulting compartments can be
utilized, for example as storage rooms, parking garages or building engineering spaces.
A recent alternative to this technology is the combined use of EP-foam and concrete. EP-foam can be moulded
into complex shapes, and by shaping these in such a way that concrete can be poured into the openings between
the blocks, the result is an unsinkable float with a concrete framework (Rijcken, 2003). The flexibility of this system
allows for creation of complex, organically shaped floats, making it possible to create more natural settings within
the floating city.
Fig. 4 – Floating element (Rijcken, 2003)
Innovative
Solutions
for the Delta
21
Water technology
Cities currently still rely on a centralized and energy intensive water supply infrastructure. Rainfall in urban areas is
a relatively clean source, which could be used as resource for local drinking water production. Instead, storm water
is transported as wastewater in combined sewer systems.
For new urban development, research in the Netherlands shows that local rainfall can fulfil residential water
requirements in an average year (Graaf and Ven, 2005). In the IJmeer, storm water shouldn’t be drained off by sewer
systems, but instead be reverted back to the IJmeer.
Useful application of wastewater effluent will become more feasible as wastewater treatment technology
continues to improve. Separation of storm water and sewage also results in more efficient water treatment plants.
Like storm water, treated waste water can be reverted back to the IJmeer. Effluent is no longer waste that needs
disposing of, but can instead be considered a valuable resource, because local wastewater treatment facilities can
be used as a biogas energy source for the new urban development (Jong, 2006).
For the floating city in the IJmeer, applying water technology means that it won’t be connected to a large
centralized water infrastructure. Instead, local drinking water production and decentralized wastewater treatment
will be applied and tested in this city. Experiments with many types of decentralized water technology can
take place, adding to the development of these systems. The focus on local solutions means that no expensive
infrastructure and energy intensive transportation of water and wastewater are necessary.
Energy technology
Even though dependence on fossil fuels is still dominant, energy generation technologies are now available that
are capable of significantly reducing this. The first step towards a sustainable energy policy was set by ratification
of the Kyoto protocol in 2004, which focuses on reducing greenhouse gas output.
The importance of a sustainable energy policy is twofold. First of all fossil fuels are in relatively limited supply
and will at some point cease to be available – crude oil around 40 years, natural gas around 65 years (Stofberg et
al, 2000). Secondly, using fossil fuels for power generation has had a negative environmental impact over the past
years, both in mining/processing and in expenditure.
Reducing dependence on fossil fuels is required, and the Trias Ecologica (Duijvestein, 2002) is a three step
strategy to achieve this. The first step is to prevent unnecessary energy usage, for example by clustering buildings,
improving thermal insulation properties and application of Low Temperature heating systems. The second step
is to apply sustainable energy where possible. This is not limited to power generation; using sunlight for space
and water heating, applying shading during summer and more effective use of daylight are also considered to be
‘applying of sustainable energy’. The last step is to use the remaining non-sustainable energy sources as efficiently
as possible.
Because the floating city is located in the Dutch Ecological Main Structure, there are some restrictions to
deployment of power generation technologies like wind turbines, although this can be overcome by displacing
part of the production, for example to Flevoland or the North Sea.
A development worth mentioning in this respect is that of the Osmaal, which uses the difference in potential
between the fresh water of the IJsselmeer and the salt water of the North Sea to generate a significant amount of
power (Huisman, 2006).
Innovative
Solutions
for the Delta
22
4 Concept of the floating city
4.1 Strategy
Based on the analysis, a strategy towards a design can be devised with a number of key elements – those that
will have the most influence in the design stage itself. The goal of this project being both a relatively short term
solution for certain regional problems as well as a breeding ground for (long-term) future solutions, means that
the initial size will be moderate – approximately 10.000 inhabitants. The city will be modular both in planning and
construction, so it can be expanded or its elements upgraded.
Urban and ecological development
Current Amsterdam and Almere housing projects and plans towards the IJmeer mainly turn water into land,
reducing capacity of the main Dutch fresh water storage, as well as decreasing wet nature area. Using floating
structures does not reduce that capacity and creates housing which anticipates on climate change projections.
Using floating, constructed wetlands as breakwaters improves the role of the IJmeer as part of the European
Ecological Main Structure, by increasing appeal to migratory birds.
Technology
Another short term goal of the project is to put together new and existing technologies to show that the concept
of a floating city is a viable one, and a solution to many problems. The long term goal however is to create a testing
ground for future technologies, thus accelerating floating technology development and contributing toward a
knowledge based economy.
Sustainability
The floating city uses as much local resources as possible where power and water supply are concerned. The
structures themselves are designed to make efficient use of daylight and the heat of the sun. Where local
resources are not an option (for example during the construction phase), an effort should be made to apply either
sustainable, or durable non-local resources. Modularity of the project will increase adaptability, itself a feature of
sustainability.
Transportation
The IJmeer bridge will not only improve the connection between the two main urban areas in the region, but also
act as an on-ramp for the floating city. Providing a road connection (both vehicles and public transportation) next
to ferry services and water taxis will improve short-term acceptance of this project, both for its inhabitants and
visitors.
Tourism and economy
The project itself should prove to be a regional landmark, not just because it’s a floating city but also because of
the surrounding wetlands and facilities with tourist appeal, like a marina, a floating hotel and a floating theater.
Not only will increased tourist revenue be beneficial to the regional economy, public interest from abroad also
helps the floating city become an export product in its own regard, just like the Delta Works helped export Dutch
delta technology.
Innovative
Solutions
for the Delta
23
Characteristics
The combination of these elements results in a project with the following characteristics:
>> a living space for thousands, reducing the need to claim water area
>> a positive influence on regional ecology
>> a landmark, further increasing tourist appeal of the IJmeer
>> a pilot project for larger floating cities
>> a testing ground or accelerator for floating technology, resulting in knowledge and products that can be exported
The next paragraphs will go into more detail about implementation of this concept.
4.2 Concept
Combining the elements of the strategy is not intended to have a single, definitive result. The purpose of the
project of course is to be a living, breathing testing ground in which new delta technologies can be applied, so after
ten years the floating city will look different from when it started. This is a vision of its first incarnation.
A highway bridge (1, numbers refer to the next two pages) connects the cities of Amsterdam and Almere. The
structure is a hollow vessel with a lowered road surface, where the sides act as noise barriers, and an open top. The
road surface has enough buoyancy to support its dead load, which means that the road will not sink even after
sustaining damage. The low super-surface profile of the bridge allows vehicles to stay out of sight reducing noise
pollution as well. Where the floating bridge crosses the shipping lane, a high overpass bridge allows for vessels to
pass.
Along the highway bridge, a floating city is moored, surrounded by floating vegetation. This outer ring serves
both as a breakwater and a constructed wetland, reducing wave height in the city and cleaning its affluent at
the same time. Part of the ring is meant to become valuable ‘wetland’ nature, compensating for use of the water
surface. Openings in the wetlands allow boats and yachts to enter and leave.
The transport hub on the floating bridge provides two highway exits, parking facilities (5) and a metro station
(3), as well as a dock (6) for numerous single-vehicle, automated ferries capable of carrying inhabitants’ cars to
their homes – catering for those who prefer that over using the transit hub’s parking facilities.
As opposed to large ferries this system is not a batch process, but rather a continuous mechanism of ferries
departing and arriving, so there is no waiting time. During transit, driver and passengers can stay in their vehicle
resulting in a streamlined means to get from shore to floating home with maximum use of the potentials of water
– both as an infrastructure and as open space.
Within the city, residents can move about using floating pathways (7) (8), either on foot or by bike. Car access is
only allowed in case of emergencies.
There are two groups of pathways perpendicular to one another on different levels, effectively creating long
waterways which allow for water based traffic. These pathways intersect inside the larger living clusters (9) with a
3D bridge connection (10).
Within the city, the means of transport are small (6) and big ferries (4), private boats (4), walking and biking.
Innovative
Solutions
for the Delta
24
Spacious courtyards (12) within the clusters provide access to people’s homes. These courtyards fill the need for
public space, something that requires careful attention in floating cities. An ETFE canopy similar to the iris dome
(Hoberman, 1994) covers the courtyard, acting as a rain screen and helping maintain a desirable temperature. A
natural airflow will ventilate the atrium. The rolling doughnut-stream effect (14) will increase the Venturi (15) and
Bernouilly (16) effects on top of the dome combined with thermal effects (Baldwin, 1977). Additional cooling within
the structure can be provided with a closed loop water heat exchanger (16).
Spatial dimensions and flexibility of the structure’s interior are designed with the disabled and elderly in mind (17).
Innovative
Solutions
for the Delta
25
Innovative
Solutions
for the Delta
26
Its circular nature allows for all homes to be provided with the delightful water diorama (18) surrounding the
structure.
A spherical shape is the optimal ratio for surface and volume. This saves on materials and energy loss. By
flattening the volume, internal spaces become more practical in use. The large living clusters consist of 60 houses
with corresponding public and parking space (9). Smaller clusters contain up to 12 houses (19), and are basically set
up in the same way. This approach of an architecture with strong references to water invites more daring designs
(Tang, 2006) (20). A single ’foot’ on the water floor fixes the structure’s horizontal position while allowing for
vertical movement, a method of mooring currently subject of further development (Attika, 1994) (21). Because of
a minimum of connections to both water floor and city infrastructure, all units and clusters can be prefabricated,
and moved to and from locations for a dynamic urban planning. City densities and capacities are easy to alter.
4.3 Design of the regional connections
Connecting the floating city with its surroundings is the floating bridge, which provides for a motorway and metro
line between Amsterdam and Almere. The motorway will be an extension of the secondary Amsterdam beltway
A9, originating from the A1-A9 junction and connecting to the S101 in Almere.
The metro line will connect to the Amsterdam metro network in Diemen Zuid station (8 minutes from the
floating city) and stretches from the Zuidas to Almere (26 minutes for a single trip).
Recent research shows that around 60% of commuters use cars for work-related travel (RWS-AVV, 2004), so a
road connection will improve short-term social acceptance of the floating city by potential residents.
The bridge will be a good impulse for further development of the Almere region, given its relationship with
Amsterdam (Heijden, 2006) and is an alternative to the proposed, controversial Naardermeer traverse.
The third means of regional transportation will be a (fast) ferry connecting the floating city to Amsterdam
Central Station (20 minutes).
Fig. 5 – Regional connections
Innovative
Solutions
for the Delta
27
5 The transition: from vision to reality
In this paper a governance model of transition management is used to contribute to making the step from vision
to reality (Kemp and Loorbach, 2005). Transition management consists of the phases illustrated in Fig. 6.
Fig. 6 – Phases in transition management
(Loorbach and Kemp, 2005)
1.
2.
3.
4.
Establishing a transition arena and vision development
Developing coalitions and transition agendas
Mobilizing actors and executing transition experiments
Evaluating, monitoring and learning
5.1 Establishing a transition arena and vision development
The first step in transition management is the development of a transition arena. This is a group of people, able
to think creatively across domains and towards the future, and together developing a vision on sustainable
urbanization. Preferably, the group consists of people from companies, NGOs, governments and research
institutions. Many disciplines, for instance spatial planning, engineering, ecology, architecture and recreation
expertise should be involved to develop an ‘out of the box’, ‘multi-domain’ vision.
5.2 Developing coalitions and transition agendas
The problems described in this paper are not limited to one domain. On the contrary; problems of mobility,
demography, ecology, economy and flood control interlock in the densely populated area of the AmsterdamAlmere region. The strategy of the floating city aims to contribute to a solution for all these delta problems. This
means multiple domains will have to be integrated and many actors will have to be involved in the plan. Transition
management aims to establish a broad societal coalition of governments, research institutes, universities,
companies, developers and consultants, to bring knowledge together in an innovative transition experiment. The
vision frame of the first phase will be used to develop coalitions and a transition agenda. The actors in could be
involved in the transition coalition. Participation can be consultation, cooperation and informing.
Innovative
Solutions
for the Delta
28
actor
European Government
main problem/objective in
amsterdam-almere area
means for developing a
floating city in ijmeer
Environmental quality
Issue of permission for
developments based on Bird and
Habitat Directive
Funding
National Government
Ministry of transport, public works
and water management
Congestion of traffic
Funding
Flood control
Technical knowledge
Decision power
Ministry of housing, spatial
planning and the environment
Ministry of economic affairs
Housing demand
Funding
Sustainable development
Spatial knowledge
Shortage of space
Decision power
Economic growth
Funding
Competitive knowledge based
economy
Economic knowledge
Decision power
Provincial Government
Provinces of Flevoland and
Noord-Holland.
Economic position
Funding
Traffic problems
Regional knowledge
Availability of housing locations
Funding
Accessibility, connection AlmereAmsterdam-Schiphol
Local knowledge
Decrease of ecological quality
Ecological knowledge
Insufficient possibilities for
recreation
Recreation and mobility knowledge
Development of international
business location
Corporate knowledge
Municipal Government
Municipalities of Amsterdam and
Almere
Environmental Organizations
Natuurmonumenten,
Staatsbosbeheer, Milieudefensie e.a.
Recreational and Mobility
Organizations
ANWB, Watersportverbond e.a.
Economic Organizations
Projectbureau Zuidas
Housing location for expat
knowledge workers
Regional connections
Airport Schiphol
Expansion
Investment capacity for connections
Regional connections
Energy company
Reliable, cost effective energy
supply
Investments, knowledge
Water company
Reliable, cost effective water supply
Investments, knowledge
Innovative
Solutions
for the Delta
29
actor
main problem/objective in
amsterdam-almere area
means for developing a
floating city in ijmeer
Public transport company
Reliable cost effective public
transport
Investments, knowledge
Availability of new housing and
company locations
Investments, knowledge of projects,
technical knowledge
Project Developers
NEPROM
Inhabitants
Availability of houses
Quality living environment
Jobs
Knowledge on demands of target
group
Research Institutes
Universities
Investments in R&D
Innovations, creative ideas, designs
for floating cities
Consultants
New projects
Knowledge of projects area
developments, innovation and
technology
Architects
New projects
Designs
New projects
Technical knowledge of floating
structures
Knowledge institutes
Constructors
Table 2 – Possible participating actors, objectives and means in the floating city project
5.3 Executing the transition experiment
By cooperation of public and private parties a balance between economic, ecological and social development
can be accomplished. This means that both experimenting with new technologies and learning by doing are
important.
The development of the Amsterdam-Almere region is a matter of national importance; firstly the national
government should issue a decree to initiate this process. Government efforts should be focused on improving
accessibility, ecology, economy and knowledge development by constructing a floating city in the IJmeer.
Important in the execution of the pilot project is the fact that this project should start with creating ecological
surplus by constructing new nature areas. This is necessary to make other developments possible within the
limits of European legislation (Bird and Habitat Directive). Because the project is a multi-domain one, funds for
infrastructure, economic development, ecologic development and research funding can be combined in this project
increasing the feasibility of reaching the required budget.
Innovative
Solutions
for the Delta
30
5.4 Evaluating, monitoring and learning
The transition experiment should be continuously monitored and evaluated. A collective memory about combining
ecology, mobility, housing and recreation will need to be built. One floating city is not yet a sustainable delta.
Therefore experiences from the floating city should be used to start and improve other transition experiments in
the Rhine Delta and worldwide to obtain knowledge and skills until there is enough know-how and preparedness
to transform the urban delta system to a lower risk, more self supporting area.
6 Discussion
Building in the IJmeer has recently attracted a lot of public attention. On 30 June the cabinet made a decision on
building in the IJmeer. These plans partly have the same background as our vision. However, our focus is more on
developing new building technologies such as floating structures, rather than applying the proven conventional
concept of raising terrain by hydraulic fill. Developing a floating city contributes to the national objective of
strengthening the knowledge based economy while reducing its vulnerability to flooding.
For a maximized learning experience in the floating city, it would be useful to do experiments with radical new
technologies. However, a city is built for people. Therefore, in the floating city on the IJmeer, a balance should be
found between testing for technological development and making an urban area where it is good for people to
live. In practice it will be partly state-of-the-art technology and new technology that is applied in the floating city.
Moreover, the floating city will attract people that like the idea of innovation and technology development. This in
turn will create room for the transition experiments.
Many delta areas all over the world have problems similar to the Rhine Delta. Technology developed in the
IJmeer floating city may prove to become a valuable export product.
7 Conclusions
The aim of this study as introduced in the beginning of this paper is to develop a more self supporting, non
risk increasing concept of urbanization in the Almere Amsterdam region that will contribute to further
economic and technological development of the Rhine Delta. For this purpose a combination of analysis,
strategy and concept development has been applied. National trends, developments in the Amsterdam region
and technical innovations have been analyzed. These developments are characterized by wide margins of
uncertainty. A management approach to deal with complex problems under conditions of high uncertainty
is transition management. A transition experiment of a more self-supporting floating city in the IJmeer has
been presented. It uses local water and energy resources instead of external resources only. Moreover, it does
not result in increased expected damage in case of flooding. Next to being more self-supporting and non risk
increasing, this floating city is an accelerator for delta technology; it offers a testing ground for pilots with
new water, energy and floating technology. It will enhance knowledge based activities in the AmsterdamAlmere region, a unique area with high economic potential. The floating city is a high quality living area that
will attract highly educated knowledge workers. The development of the floating city is combined with large
scale wetland development in the IJmeer to create an ecological improvement. Moreover, it contributes to
reducing the mobility and housing problems in the region, stimulates the regional economy and anticipates
on an expected increased demand for water recreation. Overall, the floating city IJmeer anticipates on multiple
trends and contributes to development of a knowledge based economy and new technolgies while not further
increasing flood risks.
Innovative
Solutions
for the Delta
31
References
AHN, Actueel Hoogtebestand Nederland (http://www.ahn.nl/). Elevation model of the Netherlands.
Ale, B. (2006). Hoogleraren over rampenbestrijding. Webarticle on http://www.vpro.nl/programma/tegenlicht/
Aqua Struenda (2004). Waterwonen in de 21ste eeuw, Civieltechnische oplossingen voor drijvende wijken. Heerlen,
the Netherlands.
Baldwin, J. (1996). Bucky Works: Buckinster Fuller’s Ideas for Today. New York, United States of America.
Biesboer, F. (2006). Het Osmaal. De Ingenieur 07-4-2006 p34-40.
Butler, D. and J. Parkinson (1997). Towards sustainable urban drainage. Wat. Sci. Tech. Vol. 35 No 9. pp. 53-63.
Butler, D. (2006). Water sustainability, a journey into the unknown?. Presentation for the Sustainable Urban Water
Management symposium of the Students Society of Water Management, Delft, The Netherlands.
CBS (2006), Demografische Informatie. http://statline.cbs.nl/StatWeb/
Commissie Waterbeheer 21e eeuw, CW21. (2001). Anders omgaan met water, waterbeleid voor de 21e eeuw.
Czemiel, J. and I. Hyvönen (2002). Are there sustainable alternatives to water-based sanitation system? Practical
illustrations and policy issues. Water Policy 4 (2002) 515-530.
Duijvestein, C.A.J. (2002). Het nieuwe ecologisch bouwen. Delft, The Netherlands.
Elkington, J. (2001). The Chrysalis Economy. How Citizen CEO’s and corporations can fuse value and value creation.
Capstone, Oxford, UK.
European Commission (2006). Time to move up a gear – The new partnership for growth and jobs. Luxembourg,
Luxembourg.
European Commission, Directorate-General for Research (2006). Third European report on science & technology
indicators. Brussels, Belgium.
Gennip, C.E.G. van (2006). Pieken in de Delta – Gebiedsgerichte economische perspectieven. Policy document issued
by the state secretary of economic affairs. The Hague, The Netherlands.
Graaf, R.E. de and F.H.M. van de Ven, (2005). Transitions to more sustainable concepts of urban water management
and water supply. Conference proceedings 10th International Conference on Urban Drainage, Copenhagen,
Denmark.
Graaf, R.E. de, F.H.M. van de Ven and N.C. van de Giesen. (submitted). The Closed City as a strategy to reduce
vulnerability of urban areas for climate change. Conference on Innovations in coping with water and climate
related risks. International Water Association. Amsterdam.
Heijden, B. et al (2006). Verkenning Regionale IJmeerverbinding. Issued by the Regional Body of Amsterdam (ROA).
Amsterdam, The Netherlands.
Hendriks, Ch. F. (1999), Duurzame bouwmaterialen. Best, the Netherlands.
Hiessl, H, R. Wals, D. Toussaint (2001). Design and Sustainability Assessment of Scenarios of Urban Water
Infrastructure Systems. Conference proceedings 5th international conference on Technology and innovation.
Innovative
Solutions
for the Delta
32
Huisman, J. (2005). Vier alternatieve-energiekaarten van Nederland. De Ingenieur 04-2-2005 p34-35.
Jong, A.H. de et al (2004). Lange-termijn bevolkingscenario’s voor Nederland. Report about long-term population
growth, jointly issued by CBS and RIVM-MNP.
Jong, K. de (2006). Lage EPC voor woningen in Polderwijk door biobrandstof. Technisch weekblad, 13 mei 2006.
KNMI (2006). Klimaat in de 21e eeuw. Vier scenario’s voor Nederland. De Bilt, The Netherlands.
Kuijper (2006). De drijvende fundering, een stabiele basis voor waterwonen in de 21ste eeuw, hst. 7, 10 en 18. Delft,
the Netherlands.
Loorbach, D. and R. Kemp (2005). Dutch Policies to Manage the Transition to Sustainable Energy. Jahrbuch
Okologische Okonomik. J. Meyerhoff. Marburg, Metropolis Verlag. 4: 123-151, 2005.
Nieuwenhout, F. et al (2005). Flexible electricity grids. Petten, The Netherlands.
NMP (2005). Effecten van klimaatverandering in Nederland. Bilthoven, The Netherlands.
Peijs, K. (2006). Bouwlocaties dringend gewenst. Metro, 10 mei.
Peijs, K. (2006). Samenhang in ontwikkeling – Structuurdocument Noordvleugel. Den Haag, Netherlands.
Regio Randstad (2005). Randstad Holland in Europe. Statistical booklet with information on population, economy,
connections and business locations. Utrecht, the Netherlands.
RIZA (2005). Aard, ernst en omvang van watertekorten in Nederland. RIZA report 2005.016, http://www.
droogtestudie.nl/
Rotmans, J. (2003). Transitiemanagement, sleutel voor een duurzame ontwikkeling. Koninklijke Van Gorcum. Assen,
the Netherlands.
Rijcken, T. (2005). De drijvende bouwsteen. Article in Cement no. 5, Nov 2005.
Rijkswaterstaat, Adviesdienst Verkeer en Vervoer (2004). Fileverkenning – de ontwikkeling van vertragingen op het
Nederlandse snelwegnet. Rotterdam, The Netherlands.
Rijkswaterstaat, Directie IJsselmeergebied (2004). Actueel Dieptebestand Markermeer 2004. Based upon
measurements from the AHN database.
Rijkswaterstaat, Directoraat Generaal personenvervoer (2005). Kerncijfers regionale bereikbaarheid 2005. The
Hague, The Netherlands.
Stofberg, F. et al (2000). Vademecum energiebewust ontwerpen van nieuwbouwwoningen. Issued by Novem,
BOOM and the national centre for sustainable design (Nationaal DuBo Centrum). Sittard, The Netherlands.
Stuurgroep Verkenning IJmeer (2005). Toekomstvisie IJmeer – Naar een Waterpark IJmeer binnen het Wetland
IJsselmeer. Amsterdam, the Netherlands.
Swanenvleugel, B. and Zwanikken, T.H.C. (2006). Briefadvies Noordvleugel. Advice from the VROM and V&W
councils about development of the northern region of the Randstad.
Vermazen, I. (2002). Kernpublicatie WBO 2002, Dienst Wonen Amsterdam en Regionaal orgaan Amsterdam.
VROM, LNV, VenW en EZ, Ministeries van (2005). Nota Ruimte. http://www.vrom.nl/
Walraven & Markovich (2004). Development and applications of high performance hybrid-fibre concrete. Delft, the
Netherlands.
WCED (1987). The Brundtland Commission Report. Our Common Future. Oxford University Press. New York. USA.
Innovative
Solutions
for the Delta
33
Innovative
Solutions
for the Delta
3
SUSTAINABLE GROWTH IN URBANISED DELTA AREAS
THE OPPORTUNITIES OF A GEOGRAPHICAL APPROACH TO THE PEARL RIVER DELTA
Sustainable Growth in Urbanised Delta Areas
The opportunities of a geographical approach to the Pearl River Delta
SecondprizeofTheDeltaCompetition2006
GEERT VAN RENS
PhD Student Mechanical Engineering
ANNE LOES NILLESEN
Graduate Architecture & urbanism
CARIANNE SCHAMHART
Graduate Aerospace Engineering
NADIA LuGT
BSc Student Civil Engineering
Innovative
Solutions
for the Delta
3
Abstract
The attractions of delta areas have boomed economies and founded major cities, but the threats of the
adjacent water have persisted and natural resources have declined. The objective to facilitate sustainable
urban growth in delta areas can only be met by a simultaneous approach of all the stakeholders concerned.
This paper will demonstrate the result of a geographical design approach for a fast growing, often flooded and
environmentally endangered area: the Pearl River Delta in China. It will be shown that with a careful analysis
of the local situation (geography) and local requirements translated to geographical elements, solutions can
be found that blend in with their surroundings and meet the needs of present and future.
1 Introduction
River deltas have always provided unique opportunities for settlement with their fertile plains and natural
infrastructure, but at the same time the water remained a threat. Flooding by the river or the sea is a permanent
risk for many deltas worldwide and global climate changes aggravate local weather extremes. As part of the
Royal Haskoning DeltaCompetition this paper will investigate the relation between urbanization and flooding
aimed at a sustainable coexistence of people and water in the future.
The scope of the investigation favoured a multidisciplinary team and the use of an integral design approach.
The team consists of an architect, mechanical, civil and aerospace engineer and the design approach used
combines different aspects in a geographical manner. China’s Pearl River Delta (PRD) is chosen as an example to
investigate the relation between urbanization and flooding. This delta develops at extreme speeds and faces many
of the problems associated with modern delta areas. A solution is sought that matches the local priorities with the
need for space for water and growth.
2 Design approach
Sustainable coexistence of people and water in urbanized delta areas means that development meets current needs and
does not compromise the ability to meet future needs4. This requires the evaluation of all needs of a delta and integration
of the different problems to yield 1 set of requirements. To be able to cope with such different aspects a geographical
design approach is used based on the philosophy of Ciro Najle5. This design approach combines an integral perspective
with the use of inspiration from other fields to serve as a starting point for the solution. This means that many different
aspects of the site are summarized in maps. The combination of these maps demonstrates the main problems of a
site but also integrates problems of a different nature to a specific location. This offers the opportunity of identifying
synergy and combined solutions. The philosophy also stresses the importance of inspiration from other fields like natural
phenomena, history or culture. These can all provide ideas for systems that fit well within the local circumstances. This
yields an integration of spatial planning with technical solutions to the problems encountered at a specific location.
This paper will use this method on two levels. First the Pearl River Delta (PRD) is considered as one problem area. An
analysis illustrates the general solution procedure and yields the framework and the main requirements for sustainable
growth in the PRD. But, within the scope of this research it is not possible to achieve the required detail for the entire
region. Therefore a specific location is chosen as an example to demonstrate the full capabilities of the design approach.
The background information of this location is combined with the framework and requirements found in the analysis of
the entire PRD. The combination yields one set of requirements for sustainable growth.
Innovative
Solutions
for the Delta
36
Most requirements are related to water management. Therefore numerical model is developed to calculate the
effects of water management and check the applicability of proposed solutions. Inspiration from other fields
is used to find possible water management systems that can be integrated within the area according to the
calculations of the numerical model. This yields a specific set of water management systems that can be applied
within a framework of spatial planning aimed at sustainable growth.
3 The Pearl River Delta
3.1 Why Pearl River Delta?
Urbanized river deltas demonstrate different
characteristics depending on their local climate,
geographical position, economical and political
circumstances. Therefore not all deltas can easily
be compared to other deltas. But many deltas
have developed along the same patterns and
have experienced the same problems as other
deltas did at the same phase in their development,
although not always to the same extent. A general
pattern for transition can be derived from the main
benefits that attracted people to the deltas: fertile
plains and fishing allowed agriculture and high
population densities; transport over river and sea
enabled trade; sea harbours, sufficient fresh water
and the presence of a local market and distribution
network favoured industry and recently water has
been recognized for its recreational potential and
has become a major asset for housing projects.
China’s Pearl River Delta (PRD) was chosen
Fig. A – The Pearl River Delta is situated in the Guangdong
because it develops at a different pace compared
province in the South of China
to European deltas. It can be reviewed as a delta in
its industrial period, but developing much faster
than the European deltas. Therefore the lessons learned in Europe can be used to understand the developments in
the PRD but the pace of development is so high that the effect of change can be seen much faster and therefore
considered as a test case for European deltas.
The PRD is also an interesting site to investigate due to the variety of problems associated with the rapid
growth, the political urgency of water quality problems and flooding and the financial opportunities for change.
Innovative
Solutions
for the Delta
37
3.2 Background information on the Pearl River Delta
Historically the PRD is created by man6. Early farmers channelled the rivers to use the fresh water for their crops
and this network of small dikes and channelled streams resulted in a land reclamation process that gradually
changed the PRD to a delta with broad flat plains intersected by streams and rivers. The highly fertile soil and
sub tropical climate allow three harvests a year and the PRD has long been important as an agricultural power
within China7. Villages were small and scattered over the entire PRD with Guangzhou as only regional urban
centre. Guangzhou was part of the sea bound trading route for silk and was involved in small scale international
trade. These opportunities for trade also attracted the Portuguese and British in 1557 AD and 1843 AD respectively.
Their settlement on the coast of the PRD led to the foundation of Macau and Hong Kong. The British Hong Kong
superseded Macau’s position as trading port and became the largest city in the region.
Recent developments are mostly related to political changes in the last 30 years that have boomed the
economy. The first important change was the reform of the economy in 1978 from a centralized plan-economy to
a system that allowed local authorities to decide what they wanted to do with a large part of the tax revenues
(up to 70%)8. In addition two areas in the PRD (Shenzhen and Zhuhai) were appointed as ‘Special Economic Zones’
with entirely open market economies and Guangzhou was appointed as an ‘open city’ in 1984, followed by the
entire PRD in 1985. An ‘open city’ was open to foreign investments, but less open than the Special Economic Zones9.
The second important change was a reform of the land policy. For a long time land use had been virtually free of
charge, but from 1987 the Chinese government started experimenting with the lease of state-owned land. The
lease of land increased the income of local governments that used this money to improve the infrastructure,
thereby increasing the value of the rest of the land10 and facilitating further economic growth.
An important consequence of the local control of a large part of tax income is the large competition between
the cities and villages within the Pearl River Delta. Cities try to develop better circumstances to attract industry
like harbors and airports and are reluctant to enforce environmental and social legislation. Another aspect of the
competition between the cities is a strong branding of their own identity to distinguish themselves from the
others. In this respect the PRD has become one of the few regions in China that values its own culture and history.
At present the PRD is one of the most important economical forces in China. Since the first political changes
in 1978 the economy grew with almost 20% per year11 and changed from a predominantly agricultural economy
to industry and services. The current division in the origin of the Gross Domestic Product (GDP) illustrates this
change with only 5.8 percent of GDP originating from the primary sector, while secondary and tertiary sector
account for 49.6 and 44.6 percent of the GDP12. This has also given rise to concerns for the reduction of agricultural
land and the associated reduction of food security. The connections of Hong Kong to the Western markets and to
the PRD led to many joint ventures of Hong Kong industrialists and the labor force of local communities. As the
economy grew the villages grew as well, which yielded a scattered urbanization. At the same time the economic
circumstances have resulted in a strong migration from the rest of China to the PRD. The labor force in the PRD is
thus constantly growing.
The subtropical climate in the PRD is dominated by seasonal rainfall from May to September, differing from
earlier in the north (Guangzhou) to later in the south (Shenzen). More than half of the total precipitation falls
between June and August13, while the main season for typhoons is between July and September. The smaller
tropical storms occur approximately 27 times in a year, while severe tropical storms occur nine times a year14.
Tropical storms are usually accompanied by heavy rains and storm surge. The heavy rainfall can cause flooding in
urban areas, transport problems or flooding of rivers and the storm surge can cause serious flooding by the sea
around the estuary.
Innovative
Solutions
for the Delta
38
Water pollution is an important problem for all of the PRD. The water consumption divided between 70%
municipal and 30% industrial water usage. Only 9% of domestic wastewater receives treatment before discharge
and although 85% of industrial wastewater is treated heavy metals and chemicals are still part of the non-treated
15%. Combined with agricultural runoff of fertilizer and pesticides15 the water quality has severely deteriorated. In
times of drought the low water quality is further affected by salt intrusion. A dry season of the catchment area of
the main tributaries of the Pearl River causes a salt tide to move upstream and affects the water supply of more
than 15 million people in the Delta16.
3.3 Geographical analysis of the main cities in the PRD
The geographical analysis of the PRD is used to form an impression of the developments and main problems in
the region. The location is foreign to all group members and therefore the initial analysis was aimed at a general
insight to the region. Next to this impression the general analysis provides the information required to choose a
specific location for the next phase of design.
The available information of the PRD is often not sufficiently detailed to make continuous maps and the time
required to make such maps is simply not available. Therefore the analysis of the PRD is limited to Hong Kong,
Shenzhen, Dongguan, Guangzhou, Foshan, Zhongshan, Zhuhai and Macau combined with some overall data.
The population size and growth are visualized in figure B to identify the cities with a recent history of growth
and compare the growth to their current size. The economical size and growth of cities in figure C are used to find
the opportunities for investment and future financial perspectives.
The impact of flooding on the different cities is summarized with the number of mortalities per year due to
flooding, rain storms or typhoons as visualized in figure D. This is used as an indication of the severity of water
related problems for the cities. To find possible reasons for the water problems figure E and F present the land
elevation and major land reclamations for the entire PRD respectively. The land reclamation also illustrates the
growth potential of the land itself. Figure G gives the land use in the PRD as an indication of the distribution of the
economical, functional and social land value.
The PRD has a scattered urbanization with high
growth for the individual cities
Figure B demonstrates that Guangzhou and Hong Kong
are the main urban centers in the region. Although the
population of more established cities grows slower only
Macau has actually stabilized in size; all other cities
are growing. Most of the growth of Hong Kong has
moved to Shenzhen in recent years; Guangzhou and the
surrounding cities all demonstrate a high growth. The
cities on the west side of the PRD are of a smaller scale
and relatively scattered.
Fig.B – City size and growth
Innovative
Solutions
for the Delta
39
Fig. C – Economical size and growth17
Hong Kong is the main economical power of the PRD, but the smaller centers demonstrate the highest
growth
The economical size of Hong Kong (GDP 166 USD in 2004) is almost four time that of Guangzhou (GDP 46 USD)
and Shenzhen (GDP 41 USD). Foshan (GDP 20 USD) is next in line while Dongguan (GDP 14 USD), Macau (GDP 10
USD), Zhongshan (GDP 7 USD) and Zhuhai (GDP 7 USD) are truly of a different size. The economical growth of Hong
Kong is partly situated in Shenzhen which is comparable to the situation described with the population growth.
The economical growth of the established cities is considerably lower than the growth in the younger cities. From
the younger cities only Zhuhai has figures below 20% growth in GDP between 2003 and 2004 although it still
grows much faster than the adjacent Macau (11 versus 3 % respectively). The overall picture demonstrates a strong
economic activity with high growth rates spread over multiple urban centers in the entire PRD.
Innovative
Solutions
for the Delta
40
Flooding is mostly threatening the largest cities
The largest water problems manifest themselves in the
most populated cities. This indicates a strong relation
between population density and the mortality due to
flooding, but also confirms the assumption that water
problems increase if the space for water is decreased.
Fig. D – Mortalities due to flooding per year
between 1997 and 2005
The PRD consists of large flat plains with a few hills
and mountains
The difference in land elevation for most of the PRD is
limited to 1 or 2 meter and most of the area consists of
flat plains. But the presence of a few hills and mountains
change the local circumstances considerably.
Hong Kong has a much higher precipitation due
to the mountains between the Hong Kong peninsula
and the main land and has a limited area suitable for
urbanization. Macau, Zhuhai, Guangzhou and Shenzhen
are all in the vicinity of higher ground (above 200m NAP),
while Foshan, Dongguan and Zhongshan are situated in
the middle of flat plains with Zhongshan on the lowest
ground. Water management systems will have to be
adapted to these conditions.
Fig. E – Land elevation18
Innovative
Solutions
for the Delta
41
Land reclamation in the PRD continues to add cultivated land
to the delta
There are no major land reclamations near large cities. Hong Kong
has used land reclamation to increase its urban area directly, but only
on a scale that can not be visualized in figure F. Most of the reclaimed
land is used as cultivated land situated in the PRD itself or on the
coast line. They fit the historical land reclamation pattern of the PRD
although on a modern scale.
Fig. F – Land reclamation
The PRD is dominated by rice fields combined with some fish
ponds, forests and scattered urbanization
Figure G illustrates the scattered position of urban areas within an
overall green environment. Next to some large forests most land
is in use for agriculture. More specifically the agricultural land is
reserved for rice. The floodplains are used as flat rice fields (paddy)
while the surrounding areas are transformed to terraced rice fields
(terraced paddy). The diked ponds are a combination of fish farms
with commercial agriculture on the surrounding dikes (bananas, sugar
cane, silk, eg). This kind of agriculture is typical for this climate and has
a high protein yield.
If these maps are considered simultaneously, it can be noted
that the problems with flooding seem more related to the number
of citizens and urbanized area than to the local land elevation or
presence of mountains. Therefore spatial planning within cities and
Fig. G – Land use
their vicinity could make a significant change to the impact of floods.
The space for water should be an integral part of spatial planning
and design. Land use is also not directly related to the differences in land elevation. Although forests correspond
to hills and mountains the differences in elevation of the flat plains can not be matched with the land usage. This
suggests that land use is relatively flexible and different types of land use can be chosen for a specific location. This
increases the flexibility of spatial planning.
This leads to the conclusion that sustainable urbanization in the PRD should anticipate growth in both
economical sense and population size and should include the space for water in its design as one of the main
requirements. The economical and political independence of the local governments makes it possible to realize
designs that require major spatial and financial commitments as long as they also match the cities perceived
branding or give it an competitive edge.
Innovative
Solutions
for the Delta
42
4 Example: Guangzhou
To demonstrate the effect of a geographical problem definition in finding local solutions Guangzhou is used as an
example. This city combines most of the evaluated problems and is relatively well documented.
4.1 Background
Guangzhou (also referred to as Canton) is said to be founded in 214 B.C. as Panyu. It was the only city in an
agricultural area of small villages and after the foundation of Macau and Hong Kong it remained the regional
capital as only Chinese city. Therefore the tertiary sector has always been relatively important, combined with small
scale international trade (silk, luxury items). At present the city is inhabited by approximately 8 million people and
grows quickly both economically and in population.
Guangzhou has a strong local government that wants to develop Guangzhou as a green city within the
PRD with an ecological zone that connects all major parks and nature reserves. A strong historical and cultural
awareness is also used to distinguish the city within the PRD and the local government is involved in several
programs to develop the city in accordance with these objectives. Therefore any solution to flooding and rapid
growth related problems should also match these city objectives.
Most water related problems in Guangzhou are either directly related to the heavy rains or indirectly by the
rise of the river due to heavy rainfall upstream. The high surface runoff leads to a surface flow to the river and
inundation of the lower areas, while the flow of the river can barely be contained within the diked space. Another
important water problem is drinking water. Pollution is a big threat to drinking water facilities. In 1998 only two
of Guangzhou’s nine water treatment plants could supply drinking water quality, the other seven could not meet
this standard due to contamination. 19Urban water supply, sewage discharge and waste water treatment are
managed by different organizations. This complicates the construction and operation of non-industrial waste
water treatment plants20.
4.2 Requirements for sustainable growth
Aim is to develop a sustainable growth strategy that integrates the concern for the decreasing agricultural space,
limits the effects of heavy rains, matches historical values, incorporates green into the city and protects the
urbanized areas from flooding by the river. The different water aspects are evaluated to find specific targets and
the other requirements are also further specified.
To be able to grow in a sustainable manner the water management should keep the citizens safe and provide
sufficient drinking water in a way that does not deteriorate in the future. Since the city is constantly growing
the solutions should be an integral part of the urban region, because otherwise the load on the (shrinking)
surroundings will constantly increase. Therefore water retention and space for the river will have to be an integral
part of the cities spatial planning. Only drinking water resources are on a scale that can be met outside the city
boundaries.
The threat of flooding by the river can be illustrated with the mass flow of the Pearl River. Officially
the maximum mass flow that can be contained within the current waterways is 33.000 m3/s. This occurs
approximately once in five years, while 40.000 m3/s statistically happens only every 20 years and 47.000 m3/s only
every 100 years. In reality Guangzhou floods every 10 to 20 years21.
For the growing city these statistics should be improved because the consequences of flooding increase for
higher population densities and higher economical land value. The amount of water concerned is too large for
storage; therefore the threat of the river can only be reduced if there is more space for the water to flow. This can
be accomplished with a green river. This is a dry river that can be used as extra river channel in case of high water.
Innovative
Solutions
for the Delta
43
It should be placed as bypass for the most vulnerable parts of the river side and be wide and deep enough to
accommodate the extra water.
To reduce uncontrolled flooding in official numbers to once every 20 years:
maximum flow velocity u=2 m/s22
additional mass flow Q=7.000 m3/s
maximum water depth of green river d=4 m23
with width w = Q
24
u.d
Therefore the width of the combination of green rivers should be at least 875m. This space should be provided
along the river either at the river sides or in dry streams that empty directly into the estuary or on a part of the
river that can contain that amount of water.
Excessive rain and drought can be moderated by retention areas. These areas are filled in time of rain with
direct rainfall and with the runoff of their surroundings and the water can be used in times of drought. For
Guangzhou these retention areas would have to be able to cope with the average yearly maximum precipitation of
one month. Evaporation and a connection to the river should make it possible to manage the water level in a way
that maintains the buffer function of the retention areas. Appendix A gives the numerical model used to test these
assumptions.
A retention area for Guangzhou should be:
average maximum precipitation p = 280 mm/month25
runoff in urbanized area r = 50%26
urban area Aurban=1 m2
required retention volume V = r . p . Aurban = 0.5.0.28.1=0.14 m3
Therefore every 1 m2 of urbanized area requires 0.14m3 retention volume.
The third water aspect is the availability of drinking water. Water pollution is a major concern for Guangzhou
since most of the households and still 15% of the industry dispose their sewage directly to the Pearl River. This
will have to be changed to achieve sustainable growth, but it is not within the scope of this evaluation. The
required measures are comparable to those that were implemented in Europe to limit the same problems. The
requirements for drinking water are considered from a perspective that the water quality in the Pearl River is only
to low for drinking water intake in the driest season, because of salt intrusion from the estuary.
10% of household water consumption should be of drinking quality27. At present Guangzhou uses 785*106 m3
of water for household consumption per year28. Therefore in the driest 3 months of the year 19.6 *106 m3 of water
should be available of drinking quality. This is comparable to a basin of 2.6 kilometres squared and 3m deep. To
provide clean water a basin should be upstream from Guangzhou, expandable for an increasing urban population
and free from direct pollution.
Next to these three water requirements the growing city should include as much agriculture as feasible to
alleviate the concerns of decreasing food production. Spatial planning should be inline with the local history and
continue the ecological zone of Guangzhou.
Innovative
Solutions
for the Delta
44
4.3 Proposed solution
The first element of Ciro Najles design approach was to use geography to derive one set of requirements that
includes the needs of a specific location. The second element is to actively search for inspiration. Therefore a broad
research was conducted for analogies between a water management system and natural phenomena (eg. water
management in cells, salt management of sharks), but also for historical water management systems. This yielded
the inspiration for the second part of the proposed solution, since the first part of the proposed solution seems to
follow directly from the requirements: use the ecological zone as green river to structure new development. The
second part was inspired by a historical water management system that is still in use near Foshan: the fish pond.
The requirement to continue the ecological zone of Guangzhou coincides with the need for green rivers and
also makes it possible to keep some agriculture within the city boundaries. A combination of parks, nature reserves
and agricultural fields can be connected to act as green rivers in case of emergency and as ecological structure for
normal use. American experience with green rivers has demonstrated that they can become a major asset of a city
by upgrading the value of the adjacent houses and offices and they can have a strong impact on the cities image.
The extensive fish pond region near Foshan is barely affected by rain or drought, because the fish ponds buffer
the water. Fish ponds are a combination of fish farms with agriculture on the surrounding dikes. Especially the
combination of fish with mulberry trees for silk production on the dikes has a very high cultural historic value and
would strengthen the green image of the city, since silk worms are sensitive to air pollution and can therefore
only exist in clean (‘green’) areas. The interaction between the dikes and the ponds allows for a very high yield.
The waste of the silk cocoons is extra food for the fish, while the accumulated waste of the fish in the mud on the
bottom of the fish pond is used as fertilizer for the trees on the dikes. A comparable relation exists between cash
crops like banana’s or sugar cane, small cattle or poultry and fish. This kind of agriculture could therefore reduce
the concern for the decreasing space for agriculture by intensifying the use of the space.
Next to the cultural historic value and the high agricultural yields the fish ponds can fulfil another important
part of the requirements. They can limit the extremes of rainfall and drought. If the average water depth in the fish
pond is 2.5m with a minimum of 1.5m and maximum of 3.5m each 1m2 fish pond could retain the water of 14m2
urbanized area. Appendix A gives the numerical model that evaluates the opportunities of a fish pond to meet
the water needs. To be able to use the fish ponds as retention space for urbanized area the fish ponds should be
connected to the runoff water of their surroundings. But the water quality of the runoff water must be quite high
not to harm the ecosystem. Therefore a direct connection to the roof of buildings is proposed. This also makes it
possible to use the difference in height to power most of the transportation.
Fig. H – Fish ponds and agriculture in the PRD29.
Fish farming combined with agriculture on the
surrounding dikes (mulberry trees)
Innovative
Solutions
for the Delta
45
4.4 Geographical analysis of Guangzhou
The analysis is based on the spatial expansion program of Guangzhou until 2010. This southbound expansion (see
figure I) still has a lot of open space, but is close enough to Guangzhou centre to expect further urbanization in the
future.
To identify opportunities for an ecological zone that can act as green river the present green, agriculture and
water are considered.
The ecological zone connects the existing green and the river to
create green rivers
Figure I shows the parks and nature reserves as expected in 2010 and the
agricultural area adjacent to water. The parks and nature reserves need
to be connected to create an ecological structure, while the agricultural
fields provide opportunities to connect a green river to the flow of the
Pearl River. The proposed green structure is indicated with the dotted
lines. The northern line proposes the main ecological zone that connects
all major green areas. The southern line is a proposition for a green
river over mostly agricultural fields alongside existing water. Both are
oriented ‘down stream’ parallel to the river. The connections (branches)
between the two zones create a more robust structure.
Next these zones are combined with the present infrastructure to
minimize the intersections of infrastructure and ecological zone as can
be seen in figure J where the additional red lines indicate major roads
and the dashed lines represent the rail infrastructure.
Fig. I – Ecological zone
Refining location ecological zone with the present infrastructure
Fig. J – Present infrastructure
Innovative
Solutions
for the Delta
46
It can be seen that some small relocations of the two branches aligns
them with the major infrastructure rather than crossing them. The
combination also reveals an additional advantage of the spatial
orientation of the southern green river: an overlap with the rail track
for most of its route. The presence of the rail already defines the
route, requires part of the surrounding infrastructure to use split level
intersections, and makes the area less suitable for habitation. A green
river also requires some bridges or tunnels to connect the two sides
in case the river is flooded and most off all needs space. Therefore this
combination is very favourable. Many other aspects could be included
to refine the exact location of the green rivers and to improve the
actual functionality and fit within its surroundings.
The fish ponds are identified by combining urbanized area with
agricultural area. The fish ponds should preferably be located directly
adjacent to urbanized area to profit from the runoff of the buildings.
Therefore figure K indicates all areas where agricultural fields are
situated next to urbanization.
Opportunities to place fish ponds directly adjacent
to urbanization
Fig. K – Preferred location fish ponds
A detailed planning should ensure that every 14 m2
of urbanized area is combined with 1 m2 of fish pond
and define the infrastructure required to transport the
runoff water to the ponds and over flow from the ponds
to the river.
The fish ponds can respond to urban growth in three
ways depending on the specific type of growth. Limited
growth could be met by decreasing the space for the
dikes, thus increasing the space for water. This reduces
the agricultural yield, but maintains all other functions.
Stronger growth could be facilitated by moving the fish
ponds and aligning them with the new city boundaries.
The most likely development in the Pearl River Delta
is scattered growth. Therefore the initial change from
field agriculture to fish ponds could be repeated with
the same advantages of increased agricultural yield and
water storage for new areas of urbanization.
The geographical analysis of Guangzhou
demonstrates that both green rivers and fish ponds can
be well integrated to their environment and match the
cities spatial planning objectives.
Innovative
Solutions
for the Delta
47
5 Conclusions and recommendations
A sustainable coexistence of densely populated areas and flood prone regions can only be achieved by
integrating water management within the urbanized area. Especially growing cities will need to incorporate
space for water within their boundaries to avoid an ever increasing load on their surroundings.
The combination of the analysis of the entire Pearl River Delta with Guangzhou showed that for a
sustainable future the city would need to be able to grow both economical and in population; it would have to
reduce the threat of flooding by the river and the consequences of excessive rain; it should improve the water
quality for drinking water intake; it should maintain food production; the solution should match the culture
and history of Guangzhou and strengthen the green image of the city.
This has resulted in a spatial development objective for the expansion of Guangzhou: expansions should
be structured by the ecological zone that doubles as green river. For the projected south bound expansion two
ecological zones are defined that are connected to the present ecological zone and parallel to the river. They
consist of nature reserves, parks and agricultural fields and are connected to the river at several places. The
width is sufficient to reduce flooding to once every 20 years (official statistics) instead of once every five years
at present.
The long stretch of green will strengthen the green image of the city and will most likely have a positive
effect on the prices of the adjacent houses and offices. The reduced flood risk will also help to promote long
term investments and can give Guangzhou a competitive edge in the Pearl River Delta.
The second spatial measurement is the introduction of fish ponds to manage the rainfall. A fish pond
combines fish farming with agriculture on the surrounding dikes and has a high agricultural yield. Connected
to the runoff water of their surroundings each m2 of fishpond can store the water of 14m2 of urbanized area.
Therefore they should preferably be situated adjacent to urbanization to limit the transportation effort.
The fish ponds can reduce the consequences of excessive rain or drought, intensify food production and
they have a strong historical value for the region. The combination of mulberry trees with silk worms on the
dikes with fish could become a land mark for the city, since silk worms are sensitive to air pollution their very
existence would prove the ‘greenness’ of the city.
The geographical design approach made it possible to identify opportunities for interaction between
functions and design solutions that fit well in their environment. It would yield very different results for
another location. Therefore the design method is suitable for other deltas, but the solutions that were found
are specific to Guangzhou and will not have the same effect in other circumstances.
Innovative
Solutions
for the Delta
48
Bibliography
Prof. van den Akker and H. Savenije, Hydrology I, TU Delft Press.
Robert Ash 2003, ’The emergence of regional economies in China and its implications, with special reference to Hong
Kong and Guangdong’, Asia Europe Journal 1: pp 281-289, Springer Verlag, ISSN:
1610-2932 (Paper) 1612-1031 (Online).
Citizens’ party 1999, ‘Improving water quality in the Pearl River Delta, opportunities & challenges, Role for Hong Kong’.
Anthony Gar-on-yeh and Xia Li 1999, ‘Economic development and agricultural land loss in the Pearl River Delta,
China’, Habitat international vol. 23 no 3 pp373-390, Elsevier Science, ISSN 0197-3975, 1999.
Zhong Gongfu 2002, ‘The types, structure and results of the dike-pond system in South China’, Guangzhou institute
of Geography.
Ciro Najle 2004, ’Hydrotypes, water processing devices, Research Studio 2004-2005’, Berlage Institute Rotterdam.
W. Seabrooke, C.W. Yeung, M.F. Ma, Y. Li 2004, ‘Implementing sustainable urban development at the operational level
(with special reference to Hong Kong and Guangzhou)’ Habitat international vol 28 pp 443-466, Elsevier Science,
ISSN 0197-3975, 2004.
J. Shen, Z. Feng, KY Wong 2003, ‘Dual track urbanisation in a transitional economy: The case of pearl river delta in
South China’, Habitat international, Elsevier Science, ISSN 0197-3975, Article in press.
C. Tuan and Linda F.Y. Ng 2001, ‘Regional division of labor from agglomeration economies’ perspective: some evidence’,
Journal of Asian Economics 12 pp 65–85, Elsevier Science Inc., ISSN: 1049-0078.
US Consulate General Guangzhou 2000 ‘Guangdong environment: Some progress, but many problems remain’ .
Qihao Weng, ‘Modelling urban growth effects on surface runoff with the integration of remote sensing and GIS’,
Department of geography and geology, Indiana state university.
Innovative
Solutions
for the Delta
49
Appendix A: Computer model of the fish pond
Starting points
>> Control of the water level in the fish pond in a simple way.
>> A constant flow of water is taken from the pond during the whole year. This can be used for agricultural
purposes, a grey water circuit or (most likely after purification) drinking water.
Basic formula
The water level is calculated by:
with:
>> h(x) the water level at day x,
>> h(x-1) the water level at day x-1,
>>
the input flow per squared meter pond (units [m/day]) at day x,
>>
is the flow per squared meter pond that evaporates (units [m/day]),
>>
is the flow per squared meter pond that is discharged by the controller,
>>
is the flow per squared meter pond that is constantly removed from the pond (units [m/day]).
Note that it is assumed that no water exits the pond through the sides and bottom of the pond. This assumption
can only be justified if the ponds are constructed from, or lined with a material that is barely water permeable or
even water impermeable, like clay and concrete.
Boundary conditions
For the maximum allowable level in the pond a typical value of 3.5m is used30. This value is used for the simplified
static calculation in the main text of the paper as well.
The lower level is chosen as a level that is deemed sufficient to sustain fish life in the pond. The value of 1.5m is
the same as used in the main text of the paper.
Input by precipitation
Three different scenarios for precipitation (see table A.1) are used to calculate whether the control is sufficient to
not exceed the boundaries or not. The input by precipitation in days is the average precipitation per day based
on the precipitation per month. This is due to the limited availability of precipitation on daily basis. That this
assumption is not entirely valid can be seen in the last column of table A.1. The input to the pond is calculated
using a run-off coefficient of 0.5 for the city31. The rain that falls on the dikes around the pond is conservatively
assumed to end up in the pond, which means that a coefficient of 1 is used for the dikes. The level in the pond is
however not very sensitive on the coefficient used for the dikes. Obviously all rain directly into the pond is stored in
the pond.
The ratio of build-area-to-water and the ratio of dike and water is chosen in accordance with the static
calculation from the main text, i.e. build-area to water 14:1 and dike to water ratio=1:1. the input into the pond can
be calculated as:
With rocity the run-off coefficient for the city and Cdike the coefficient for the dike. hrain is the rain per m2 terrain
(unit [m/day]) A is for area. The areas themselves are not specified, just the ratio of the areas.
Innovative
Solutions
for the Delta
50
Table A.1: Delta precipitation
Month
Precipitation 2003
[mm/month]32
Precipitation 1997
[mm/month]33
Average
precipitation [mm/
month] 34
Mean number of
precipitation days
[mm]31
January
33.1
65.9
43
8
February
7.6
105.6
65
11
March
65.1
59.6
85
15
April
59.7
197.1
182
16
May
189.4
165.6
284
18
June
329.8
468.6
258
19
July
61
248.4
228
16
August
189
282.4
221
16
September
367.7
203.6
172
13
October
3.2
142.6
79
7
November
30.9
9.7
42
6
December
1.2
48.8
24
5
Total
1338.2
1997.9
1683
150
Evaporation
In the Guangzhou area approximately 2/3 of the rainfall evaporates. The evaporation from the pond is assumed
to be approximately 2/3 of the average yearly rainfall directly into the pond, which equals 1100mm. It is assumed
that evaporation is constant throughout the year. These are rather bold assumptions. For use of this model for the
actual construction of a pond this assumption needs to be verified with local data.
The amount of water evaporating can also be estimated using an energy balance for the pond per m2:
with Esolar the energy from the sun, Ereflection the reflected energy from the sun, Eradiation the amount of heat radiated by
the pond, Econvection the energy released into the air due to convection, Econduction the energy away from the pond due to
conduction and finally Eevaporation is the energy necessary to evaporate the evaporating water.
Econduction is approximately zero, because air is a very good isolator.
Esolar is for the Guangzhou area on average 3.5 kWh per m2 per day35.
Ereflection is reflectivity times Esolar. The reflectivity is 0.0636.
Eradiation=esTpond4. With e the emissivity of the medium which is by approximation 1-reflectivity, s is the StefanBoltzmann constant, which is 5.67.10-8 W/(m2K4)
Econvection is linked to Eevaporation by:
Source of this formula is Wikipedia, but it is rewritten to SI-units, with T temperature in Kelvin (or Celsius) and
P total pressure in Pascal, and pv,s,pond and pv,air the vapour pressure at the surface of the pond and actual vapour
pressure in the air far away from the pond.
With Tpond 30°C and Tair 25°C and 90% humidity (estimated values) it can be calculated that 0.8 meter of water
evaporates per year. This means that the assumption of 1.1 m evaporation seems fair.
Innovative
Solutions
for the Delta
51
Control of the water level in the fish pond
A drain with certain dimensions at a certain height is the simplest way of a control of the water level. The flow per
area at moment t is related to the water height by:
for h(t)>href
Formula III
K is the control parameter. The larger K the larger the response on the exceedance of the reference height and the
quicker the discrepancy between the actual level and the reference level is reduced.
Note that in the actual discharge over a day is dependant of the average water level over the period under
consideration. At this moment the water level of one time step earlier is taken as the average water level, this
induces a small error.
The reference height and control parameter K can be chosen freely (within the physical possibilities to fit the
water level between the lower and upper boundary. For further use of the reference height see constant flow of
water from the pond.
Formula III and the factor K, that were boldly introduced above have got a physical background, which is
explained below.
For a tank (or in this case a pond) with a drain an energy balance over a streamline can be made:
wv is the viscous work, ws the shaft work, q heat transport. a is a loss factor due to friction, V is for velocity, u
internal energy, g the gravitational constant, r the density of the medium and z the height, which may be chosen
(if chosen consistently) as absolute or relative.
This formula can be simplified considerably assuming:
1. Steady flow
2 Incompressible flow
3. Frictionless flow
4. Flow along a single streamline
5. No shaft work between pond and drain
6. No heat transfer between pond and drain
7. ppond=pdrain, i.e. no backpressure on the drain
Assumptions 1 to 6 reduce the formula to the Bernoulli equation. With assumption 7 this leads to the rather
simple formula:
This formula can be translated to the height relative to the bottom of the pond:
Formula I
If we use a mass-balance and assume no heat transfer and no compression (already assumed) the mass balance
equals a volume balance.
Formula II
With A the surface and V velocity through the surface. Note that Vpond equals the decrease in height in the pond
per second, or in other words, if we define jcontrol in m/day, jcontrol=Vpond*3600*24.
Combination of formula I and II leads to formula IV.
Innovative
Solutions
for the Delta
52
Formula IV
Note that mostly
<<1 and formula IV is often simplified as:
Formula V
In reality assumption 3 and 4 do not hold. It is common engineering practice to correct formula IV (or formula V)
for these assumptions by introducing a so-called discharge coefficient cd. The value of this discharge coefficient
varies between 0.6 en 1.
Hence the control parameter K is defined as:
with cft the conversion factor to get from seconds to the time unit used in the simulation.
Constant flow of water from the pond
The water level in the pond cannot exceed the lower boundary of 1.5m. Therefore the constant flow from the pond is
automatically calculated assuming that the flow can be ensured during six totally dry months (see formula below).
Note that this formula does not incorporate the removal of water by evaporation, which is incorporated in the
model. The assumption of 6 totally dry months is a very conservative assumption however, and the formula is valid
up to a minimal reference level of 1.6 m.
Processing
As the starting level for the calculation the minimum level was taken. Two consecutive years have been calculated,
because due to the initial level the first year is not representative.
Results
The conservative assumption for the constant drain resulted in a water level that was never below the lower limit
for the three scenarios and the chosen reference water levels. The exceedance of the upper level is strongly related
to the precipitation and the value of control parameter K. A higher value of K results in no exceedance of the upper
limit, but results in a stronger response on precipitation (or in other words a higher run-off coefficient from the pond).
Five different sets of parameters are shown in table A.2 that fulfil the boundary conditions. Set 1 gives the
minimum reference level. It is not equal to the minimum level due to evaporation of water.
Innovative
Solutions
for the Delta
53
Table A.2
Set 1
Set 2
Set 3
Set 4
Set 5
href [m]
1.57
1.7
2.1
2.5
3
K [m/(day2)]
0.025
0.03
0.04
0.06
0.12
f constdrain
[m/day]
0*
0.011
0.033
0.0055
0.0082
ratio constant drain
to drinking water
consumption# [%]
0
134
403
671
1006
* Manually set to zero
# To put the constant drain into perspective. Based on value drinking water quality consumption in Guangzhou, discussed in the main body of the paper and assuming whole Guangzhou was Area of Guangzhou of 7400 km2. Source: www.blikopnieuws.nl and www.cnhomestay.com/city/cityguids/guangzhou.
The most desirable dataset can be chosen, depending on whether a slow response or a large constant flow is
preferred.
Output
Input Rainfall
Examples of output of the program with dataset 2 and the
scenario with precipitation in 2003.
12
Rainfall per day (mm/day)
10
8
6
4
2
0
100
200
300
400
Time [Day]
500
600
700
The black striped line represents the minimum level in the
pond. The upper level is indicated by a same line, but is not
visible due to the relative small amount of rain in this scenario.
Variation waterlevel in pond starting first of January
3
2.5
Level in fishpond (m)
2
1.5
1
0.5
0
Innovative
Solutions
for the Delta
54
400
450
500
550
Time [Day]
600
650
700
Flow from pond by "controller"
0.18
1.8
0.16
1.6
0.14
1.4
0.12
1.2
0.1
0.08
1
0.8
0.06
0.6
0.04
0.4
0.02
0.2
0
350
400
450
500
550
Time [Day]
600
run-off coefficient from pond
2
run-off coefficient (-)
Flow from pond per m2 water in pond (m)
0.2
650
700
750
0
400
450
500
550
days
600
650
700
Run-off coefficient is in this case defined as drain from the pond divided by the rainfall. In dry moments the drain
exceeds the rainfall, which results in values larger than 1.
Conclusion and recommendations appendix A
It is shown that a fish pond can indeed mitigate run-off peaks due to excessive rainfall at the calculated ratio
in the paper of 14 m2 city to 1 m2 fish pond. This ratio is too small for the fish pond to act as a storage of rainfall
during long periods, but this was not the aim of this design. It is also shown that a fish pond can be used as a
source for constant water withdrawal. It is possible to modify the behaviour of the fish pond to favour a higher
constant water withdrawal or to favour more smoothening of the run-off peaks by modifying vertical position and
dimensions of the drain. The water level can be controlled with more advanced controls, but a drain at a certain
height fulfils all the requirements.
For actual dimensioning of the drain it is strongly recommended to
use rainfall data on a daily but prefarably hourly basis. This means
that the calculation needs to be repeated with smaller timesteps.
It is recommended to use data on evaporation in Guangzhou with
a higher time resolution.
Innovative
Solutions
for the Delta
55
1Photo: Muztagh
2Photos
3Photo
4Based
5Ciro
travel: www.muztagh.com
by Elsrud family: www.elsrud.net
by Mr. Heywood: www.uwm.edu/~heywood
on the definition for sustainable development of the Brundtlandt Commission (1987).
Najle 2004; Hydrotypes, water processing devices, Research Studio 2004-2005; Berlage Institute.
6http://web.mit.edu/11.952/www/en/region/regional_water.html
7A.W.M Wong
River Delta.
8Robert
and M.H. Wong, 2003, Recent socio-economic changes in relation to environmental quality of the Pearl
Ash 2003, The emergence of regional economies in China and its implications.
9http://www.china.org.cn/e-china/openingup/sez.htm
10Anthony Gar-on-yeh and Xia Li 1999, Economic development
11Robert
Ash 2003, The emergence of regional economies in China and its implications.
12J.Shen, Z. Feng, KY Wong
South China.
13Travel
14Exact
and agricultural land loss in the pearl river delta.
2003, Dual track urbanisation in a transitional economy: The case of pearl river delta in
China Guide.
figures: Tropical Storms > 34 knots: 26.7/year, Tropical Cyclones > 63 knots: 16.9/year, Category 3+ Tropical
Cyclones > 95 knots: 8.5 times a year.
15‘Improving
16China
water quality in the Pearl River Delta’ Citizens Party.
Daily.
17http://www.gddoftec.gov.cn/en/Statistical/200412/4.htm
18De
grote Bosatlas 50th edition, 1988, Wolters Noordthoff BV Groningen.
20US
Consulate General Guangzhou 2000 ‘Guangdong environment: Some progress, but many problems remain’.
19Asia
Environmental Trading Ltd., China Environmental Review, August 1998.
21Zhenguo
Huang et al. 2004 ‘Coastal Inundation due to sea level rise in the Pearl River Delta, China, Guangzhou
institute of Geograpgy’.
22prof
van den Akker, hydroloog, CT TU Delft.
23prof. van
24prof. van
den Akker, hydroloog, CT TU Delft.
den Akker and H. Savenije, Hydrology I.
25http://iri.columbia.edu/climate/cid/Dec2004/download.pdf
26Prof. Savenije
‘Modelling urban growth effects on surface runoff with the integration of remote sensing and
GIS’,Qihao Weng, Department of Geograph and Geology, Indiana State University’.
27http://www.vewin.nl/bestanden/internet/Publicaties/Rapportage_watergebruik_thuis_2004.def.II.pdf
28Improving
water quality in the Pearl River Delta 1999, opportunities and Challenges.
29Photo: Food
30The
31Prof
and agriculture organization of the united nations by Mr. Zhang (www.fao.org).
types, structure and results of the dike-pond system in South China. Zhong Gongfu, Guangzhou institute of Geography.
Savenije (personal communication) and ‘Modelling urban growth effects on surface runoff with the
integration of remote sensing and GIS’, Qihao Weng, Department of geography and geology, Indiana state
university.
32http://www.stats.gov.cn/english/statisticaldata/yearlydata/yb2004-e/html/A0115ae.htm
33http://www.stats.gov.cn/english/statisticaldata/yearlydata/YB1998e/A1-13E.htm
34http://www.worldweather.org/001/c00241.htm
35www.solar4power.com/solar-power-global-maps
Innovative
Solutions
for the Delta
56
36Hydrologie
1 Prof. van den Akker and H. Savenije.
All You Need is Space
Long-term and large-scale sustainable development in deltaic areas
Third prize of The DeltaCompetition 2006
Lisette M. van der Burgh
Water Engineering & Management, University of Twente, Enschede
Marit B. Brommer
Civil Engineering and Geotechnology, Delft University of Technology, Delft
Innovative
Solutions
for the Delta
57
Abstract
Knowledge on the long-term (100 years and longer) morphological evolution of deltaic areas is indispensable
for sustainable coastal zone management. For the implementation of coastal policy measures, long-term
forecasts together with insight into the main governing processes in the delta, both natural as well as
socio-economic, are needed.
However, present-day models aimed at forecasting coastal change do not provide such an integrated
approach. This paper addresses the problems in forecasting long-term deltaic evolution and presents a
conceptual model in which all the factors and processes in the delta are taken into account. In addition, we
provide an approach to improve forecasting of long-term deltaic evolution.
1 Introduction
Deltaic areas are amongst the most heavily populated areas around the world and are the place of intensive
economic development (Ministry of Transport, Public Works and Water Management, 1993). Most deltaic areas are
threatened by floods from both the land side (rivers) and the sea side (storm surges and sea-level rise). It is widely
recognized that climate change will pose an ever bigger threat on deltas due to an accelerated sea-level rise, higher
river discharges and more extreme storm conditions. Socio-economic activities together with natural processes
will continue to exert pressure on the available amount of space in the delta; space will become more and more a
scarce good.
In the Netherlands, policy makers recognize the threats in the Dutch delta. From 1990 onwards, different policy
documents have appeared which address the aspects of safety in the coastal zone (e.g. Eerste Kustnota, 1990)
together with the present and future use of space (Nota Ruimte, 2004).
Until approximately the year 2000, coastal policy documents mainly focused on maintaining the predefined
safety standards of the coastal zone and on the design of strategies to withstand a so-called ‘super’ storm (a storm
with a frequency of occurrence of e.g. once per 10.000 years). Strategies include ‘hard’ measures (dikes and dams)
and ‘soft’ measures (beach nourishments) to maintain the coastline position.
However, these policy measures are predominantly concerned with maintaining safety on rather short time
periods (a few decades), in stead of considering safety on time periods of a century and longer. In addition, only rare
storm events were considered as being relevant in shaping the coastal morphology, and hence, only these events
were believed to determine the amount of coastal erosion, whereas the century-scale processes and evolution of
the coastal zone were believed to be of minor importance as erosional or depositional agents.
With the appearance of the Third Coastal Policy Document (Derde Kustnota) in 2000, emphasis shifted from
this small-scale approach towards a larger-scale approach. Since population growth and economic pressure
are likely to increase in the deltaic zone not only in the near future, but also centuries from now, it is of great
importance to gain insight in how different socio-economic activities can adapt to the long-term (centuries) and
large-scale (10s km2) natural coastal dynamics.
Innovative
Solutions
for the Delta
58
Research aim and approach
In this paper, we argue that the long-term, rather gradual processes actually shape and have shaped the coastal
configuration we witness today (Summerfield (1991), Battjes (2006), Dronkers (2005)) and, therefore, that
forecasting long-term and large-scale coastal evolution is a prerequisite to determine which deltaic areas are
and will be suitable for sustainable development. We introduce the following key issues which we believe are
underexposed and not well developed in current coastal forecasting exercises and long-term scenario building:
1. Integrate long-term knowledge of the past (geological case studies) into physical models aimed at forecasting
coastal evolution
2. Include the effects of human interferences (socio-economic developments) in coastal evolution models and
scenarios
3. Focus on hinterland characteristics (assessment of sediment budgets)
As a result, we present a conceptual model, in which we address the above mentioned issues and we discuss
possibilities on how to include these issues in long-term coastal evolution modelling and scenario building.
2 Present-day coastal research: an overview
Dutch coastal zone policy
Knowledge on the long-term evolution of deltaic areas is indispensable for coastal management purposes, in
which safety and sustainable use of the limited amount of available space are aspects of paramount importance.
The Third Dutch Coastal Policy Document (3de Kustnota, 2000) emphasized the aspect of safety in accordance
with the use of space and contended the consideration of larger temporal scales in coastal zone policy and
spatial planning measures. Furthermore, the Fourth Policy Document on Water Management (Vierde Nota
Waterhuishouding) advocated more coherence between policies with regard to water and spatial planning,
by taking into account different interests, such as safety, nature, agriculture, and drinking-water facilities. An
important aspect of this document was creating enough space for all these purposes: an integrated approach was
needed to reconcile purposes and possible effects of measures on a national level and on a more local level.
More research was needed to evaluate the impacts of human interferences on the natural system. However,
reconciling different disciplines, such as the natural and social sciences is a difficult task, which we consider further
in section 3.
Innovative
Solutions
for the Delta
59
Present coastal research activities
In coastal research, use is made of different modelling approaches to simulate coastal evolution, since it is
impossible to design one model which is able to cover the whole range of processes operating at different spatial
and temporal scales in the coastal zone.
However, these models are not appropriate to forecast evolution on geological timescales (100 years and
longer) due to the large amount of assumptions and schematisations made, nor do they take into account the
effects of socio-economic developments in the coastal zone.
In addition, until recently, only exceptional storm surge conditions – the so-called high magnitude, low
frequency events – were considered as being relevant in shaping the coastal zone, whereas the century-scale
processes and evolution of the deltaic area were ignored as being important in shaping the coastal environment.
This conviction has changed throughout the last years and it is now recognized that large-scale processes are of
great importance in shaping the overall coastal configuration.
During the Holocene period, the position of the shoreline has shifted position continuously and coastal
morphology has changed as well, mainly due to the effects of the rising sea level (Summerfield, 1991). Extreme
events, such as storm surges, also can have a significant effect on coastal morphology, but these effects are usually
temporarily, i.e. in the order of a few years; their impact usually diminishes after a couple of years and the system
adapts to the gradual, but continuous interaction of processes and forms again (this is called the morphologic
equilibrium, (e.g. De Boer, 1992 and Phillips, 1992)).
Summarized, although large scale coastal behaviour and the modelling of large-scale processes and behaviour
are recognized as being important for Integrated Coastal Zone Management (ICZM), present modelling practices
still adopt ‘civil engineering’ scales, i.e. coastal processes and system evolution on time spans of decades.
Incorporating geological knowledge and scales in model practices is still underexposed.
Furthermore, much attention is given to the effects of sea-level rise, while we believe that changes in
hinterland characteristics (e.g. river input) can be of equal or even more importance, especially in fluvial dominated
deltas such as the Rhine-Meuse Delta (the Netherlands) and the Po Delta (Italy).
River sediments are the source of 80 to 90% of beach sand and fluvial dominated deltas are able to expand
seawards (and thereby creating ‘new space’!), provided that sediment supply is sufficient and that the delta is able
to keep pace with rising sea-levels (Summerfield, 1991).
Furthermore, Coleman and Wright (1971) stress the point that changes in the hinterland will influence the
sedimentation and the hydrology of the deltaic plain and the stability of the shoreline.
Syvitski et al. (2005) give estimates on the global reduction in sediment flux due to sediment retention
in (man-made) reservoirs of 10 to 20 %. The reduction in sediment flux to the coastal zone can have a strong
impact on coastal zone erosion in addition to an accelerated sea level rise. This justifies considering hinterland
characteristics an important aspect in ICZM.
Innovative
Solutions
for the Delta
60
3 Natural and socio-economic processes in the coastal zone: the issue of scale
One of the main problems in forecasting large scale and long-term deltaic evolution is that it is difficult to assess
the (combined) effects of natural processes and socio-economic processes in time and space. This problem is
mainly due to the fact that natural processes and socio-economic processes have a different notion of scale.
Scales are considered an important part of scientific research. Scales can be referred to as the “spatial, temporal,
quantitative, or analytical dimensions used by scientists to measure and study objects and processes” (Gibson et
al., 2000).
In the natural sciences, choosing a scale on which to examine phenomena refers to time and space. Scale in
socio-economic research however, has a more abstract meaning and can often not be expressed in terms of time
and space.
This section discusses processes and changes in the natural and socio-economic system and will address the
difficulties in forecasting changes in these systems.
Natural processes and evolution
Coastal processes operate at a wide range of temporal and spatial scales, ranging from small scale turbulent
processes to the Holocene evolution of the Holland coast.
One of the problems in predicting long-term and large-scale coastal evolution is that there is insufficient
knowledge on which processes are important in the long run. An EU-funded project called PACE (Predicting
Aggregated-Scale Coastal Evolution) was launched to provide approaches towards forecasting large scale coastal
behaviour.
One of the results of this project is discussed by Cowell et al. (2003). They present a hierarchy approach which
provides a framework for the aggregation of processes in long-term and large-scale coastal simulations. Cowell et
al. (2003) call this hierarchy approach ‘The Coastal Tract Cascade’ (see Fig. 1).
Fig. 1 – Coastal Tract Cascade (After: Cowell et al., 2003)
Innovative
Solutions
for the Delta
61
The cascade provides the means of separating out large-scale coastal change from processes on smaller spatial
and temporal scales. Different modelling approaches are used for forecasting processes and evolution on different
levels in the Coastal Tract Cascade.
Roughly, a division can be made between process-based models and behaviour-oriented models. Process-based
models are useful for simulating the smaller-scale hydrodynamic processes, whereas behaviour-oriented models
are designed to cope with difficulties regarding the prediction of large-scale phenomena. These latter models
do not take into account complete process knowledge and, as a result, many assumptions are made with regard
to the governing processes. Another disadvantage of behaviour-oriented models is the rough spatial coastal
schematization, which makes it difficult to evaluate the effects of smaller-scale coastal policy measures.
Process-based models have, in general, a rather sound scientific basis. This is mainly due to the fact that
small scale processes are easy to measure and can often be simulated by means of laboratory experiments.
A disadvantage of these models is the restricted range of temporal and spatial scales over which they can be
implemented (Bras et al. (2003)).
Forecasting processes on larger temporal and spatial scales (the First and Second order of the Coastal Tract
Cascade) is complicated by the fact that processes that may be ignored at the small scale may have effects on the
large scale. Moreover, the larger the scale of interest, the greater the amount of assumptions made with regard to
the constituting processes, hence the greater the uncertainty in forecasting coastal evolution. According to Haff
(1996), predicting coastal evolution on the First and Second order of the Coastal Tract Cascade remains a difficult
task because of the lack of knowledge of initial conditions and the presence of system heterogeneity, which all
become more important with increasing system size. Furthermore, large-scale geomorphic systems are often
unique and control and/or repeatability, aspects which are present in the laboratory, are usually absent.
We argue that in order to forecast evolution on the First and Second order of the Coastal Tract Cascade, we first
need to gain insight in past deltaic behaviour, since insight into the driving forces of past deltaic evolution on large
time scales will assist in the development of scenarios for future deltaic evolution, given the present boundary
conditions (both natural and human-induced), on large time-scales (Van der Spek, 2005).
Knowledge on important governing processes over time scales which cover more than a century can be gained
by examining the stratigraphic record of a specific site. By examining sediment types and changes in sediment type
throughout such a record, changes in environmental conditions can be determined. Hence, these records provide
useful information in past coastal settings, but can also help us to understand present and future changes in
the environment: the past is the key to the present (Kroonenberg, 2005). Unfortunately, in present day modelling
practices, little use is made of this geological knowledge.
Furthermore, insight in past deltaic behaviour provides datasets to calibrate our large-scale physical models
upon. We will discuss past deltaic behaviour in greater detail in the next section.
Innovative
Solutions
for the Delta
62
Socio-economic processes
Since long-term deltaic change is not only governed by natural processes, but also by the effects of human
interferences, we need to find an approach to take into account socio-economic processes in the scope of ICZM.
However, to link the socio-economic scales in the coastal zone (see Fig. 2) to the scales of natural processes is
not a straightforward task.
Natural processes are expressed in terms of changes in space and time. Since socio-economic research is
concerned with changes in human behaviour and with how micro-level, individual behaviour evolves to macrolevel, entire system behaviour, it cannot easily be expressed in terms of time and space (Van der Veen & Otter,
2002).
Therefore, assessing the impacts of socio-economic processes remains a difficult and ambiguous task, since
this requires a thorough understanding on human behaviour. In addition, motives of people change rapidly, so
forecasting human behaviour and associated socio-economic changes on long timescales are surrounded by great
uncertainty.
At present, models to simulate the effects of human interferences are poorly developed. Land-use models,
which simulate human changes reflected by changes in land-use patterns, are promising tools in assessing socioeconomic impacts in time and space. We will discuss these land-use models in greater detail when introducing our
conceptual model in section 5.
Fig. 2 – Different scales of the socio-economic system in a coastal area
(After: KUST2005 research program)
Innovative
Solutions
for the Delta
63
4 Geological case studies
Fig. 3 – Location case studies:
1. Rhine-Meuse delta (The Netherlands). 2. Po Delta (Italy)
Insight in natural deltaic development will assist in decision-making programmes with regard to spatial planning
measures. In addition, most deltas worldwide experience human interferences from the medieval period (12001600 AD) onwards, which impacted deltaic behaviour worldwide. It would be an asset to unravel deltaic “natural’
behaviour and deltaic “human-induced’ behaviour. Geological case studies provide records that go beyond the scale
of human perception. They provide insight in understanding the holistic delta system: natural and human-altered.
Innovative
Solutions
for the Delta
64
Case Study 1: Rhine-Meuse delta evolution
Historical growth of Rhine-Meuse delta
Approximately 2600 yr BC the Dutch coast closed. 600 years later the northernmost distributary of the Rhine
(Vecht) silted up and the Old Rhine became the northern border of the Roman Empire (50 BC-400 AD). During
the Roman Empire human interferences were limited to the building of artificial mounds on the shores of (tidal)
creeks. From 800 AD onwards, the construction of embankments and dikes started. After damming several
distributaries (Hollandse IJssel 1285 AD; Linge 1307 AD) the number of Rhine distributaries was reduced to three.
Between 1250 and 1550 AD, land reclamation started. When the windmill was introduced polders were developed.
The Golden Age period (1550-1675) was characterized by large-scale land reclamations and the intensive use of
windmills for drainage purposes (Ministry of Transport, Public Works and Water Management, 1993). The digging
of the Pannendens Canal (PK in Fig. 5) in 1707 AD significantly altered the discharge distribution over the Rhine
distributaries and hence sediment supply to the delta. The Haringvliet (Fig. 5) soon became the main outlet. The
Haringvliet was closed recently, which has forced the main flow through the Rotterdam Waterway (dug in 1872 AD
(Berendsen, 2005)).
Modern Rhine-Meuse Delta
The River Rhine splits into two major distributaries, the Lek (which becomes the Nieuwe Maas) and the Waal.
The Waal merges with the Maas and forms the Merwede. The Merwede and the Bergsche Maas convolute in the
Hollandschdiep (Fig. 5). The third distributary, known as the Kromme Rhine, flows to Utrecht and is renamed the
Oude Rijn. The average discharge of the Rhine is ~2260 m3/s, and peak discharges have been measured as high as
~13 000 m3/s (Berendsen, 2005). The Rhine-Meuse delta has an area of 80.000 km2, is heavily populated and has a
large economic value. The delta is completely engineered (canals, dikes, dams, stop banks etc.), and has little room
to adapt to climate change scenarios. However, there is recent support for the “Ruimte voor de Rivier” project that
aims to enlarge the dynamic space for heavily controlled rivers such as the Rhine.
Fig. 4 – Palaeo-geographic snapshots of the Rhine-Meuse Delta.
A) Closure of the coast and onset of “modern” Rhine-Meuse Delta.
B) The Rhine-Meuse Delta is still in progress (Modified from: Berendsen, 2005)
Innovative
Solutions
for the Delta
65
Fig. 5 – Overview of present-day Rhine-Meuse delta: PK – Pannendens Canal; RW – Rotterdam Waterway
(From: Berendsen, 2005)
Case Study 2: Po Delta evolution
Historical growth of Po Delta
Innovative
Solutions
for the Delta
66
At the end of the Bronze Age (1000 B.C.) until approximately 1000 AD the Po consisted of two separate river
networks (Po di Adria and Po di Spina) and several distributary channels that evolved through natural avulsions
(Ciabatti, 1966, Bondesan, 1990). Po River discharge was dispersed though the distributary channels and cuspate
deltas formed at their outlets. The cuspate morphology suggests that these deltas were wave-dominated
(Correggiari et al., 2005). The Adige and Reno belonged to the Po River System. This combined Po river system
prograded seaward with a rate of ~4 m/yr (Syvitski, 2005). From 1000 AD onwards discharge and sediment flux is
highest through two major channels (Po di Primaro and Po di Volano). The Adige and Reno are disconnected from
the Po system. Because of some natural breakthroughs the first major diversion took place in 1150 AD in Ficarolo
(Ciabatti, 1966) upon which the Po di Primaro and Po di Volano lost their influence on the system.
Fig. 6 – Progradation of the Po Delta in
subsequent time intervals (Modified
from: Stefani and Vincenzi, 2005)
Between 1200 and 1600 AD, inhabitants attempted to revive the branches of the Po di Primaro and Po di Volano.
These attempts were not successful and as a result the northern coast started to prograde at a rate of 25 m/yr.
The northern progradation posed a threat to the Venice lagoon and the Venice Republic endorsed the construction
of the Pila canal (Porto Viro diversion: 1604 AD) to prevent silting of the lagoon. The Porto Viro diversion led to the
onset of the modern Po Delta (Visentini and Borghi, 1938; Correggiari, 2005). The confinement of the Po Delta
system to a relatively small area at a time of increased soil erosion due to human activity had a huge impact on
the progradation rate of the Po Delta system. The progradation rate of the Po Delta system increased to 86 m/yr
with some branches prograding at even higher rates (120-180 m/yr). In the nineteenth century the progradation
rate started to slow down to between 60-100 m/yr, possibly due to a reduction in precipitation and thus discharge
(Syvitski et al., 2005). Since 1886 AD the distributary channel Po di Pila became dominant and advanced at a rate of
47 m/yr (Visentini and Borghi, 1938; Correggiari et al., 2005).
Modern Po Delta
The modern Po Delta is a fluvial dominated delta and is heavily engineered; all distributary channels have stop
banks (flood protection levees) and water gates to confine and control the flow through the channels (Syvitski
et al., 2005a). Average discharge of the Po River is ~1500 m3/s, with peaks (spring and autumn) of ~10.000 m3/s
(Correggiari, 2005). The modern Po Delta progradation rate is estimated to be between 7 and 13 m/yr (Nelson, 1970).
The Po watershed (75.000km2) is heavily populated (16 million people) and is as well one of the most agriculturally
productive areas in Europe.
Innovative
Solutions
for the Delta
67
Synthesis
Important conclusions from the two case studies can be drawn: Even though both deltas are supply dominated
and are heavily engineered, they exhibit different present-day characteristics. The Rhine-Meuse delta is much more
confined in its space than the Po Delta, and as a result the Rhine-Meuse delta hardly has any room left to respond
dynamically to changes in hinterland (rivers) or basin (sea level change) regimes. The Po Delta on the other hand
is still prograding at a reasonable rate (~10 m/yr) even though sea level is rising. This is due to the fact that canals
and dikes which surround the different distributaries confine the river flow which results into the extension of the
delta into the sea.
The case studies demonstrate both natural deltaic responses and human-induced deltaic responses to climate
change. Although these natural and human-induced responses act on different (temporal and spatial) scales, they
are closely linked, since they belong both to the deltaic system (see section 3).
Summarized, deltaic behaviour is sensitive to sediment supply (hinterland processes), sea level rise (processes
in the basin) and human interferences.
If we want to obtain insight in the future amount of available space in the delta (hence, delta progradation), we
need to know the quantity of sediment that will retain in the delta. We can quantify the amount of sediment in the
delta by means of calculating sediment budgets, which is discussed in the next section.
Sediment Budgets
The concept of sediment budgets in a deltaic area refers to the balance between sediment added to and removed
from the delta (Morton, 2003). When sediment input to the delta is higher than sediment removal, there is a
surplus of sediment and the delta may prograde seaward. On the other hand, when more sediment is removed
than is added, there is a deficit in sediment input and the delta retreats landward. We argue that the sediment
budget concept is an extremely important yet underexposed research topic in ICZM.
However, establishing a sediment budget for any deltaic area is a difficult task. To calculate a sediment budget
all sediment sources to the delta must be identified. Subsequently, the sinks - areas where sediment is temporarily
or permanently stored – must be determined. Finally, estimates on net sediment deposition versus net sediment
erosion should be made.
The sediment budget concept is a powerful tool in erosion control and sustainable delta management. The
concept applies to multiple scales (from a local scale to an entire region), and offers quantitative output to coastal
managers and decision-makers.
There are several tools that can be of aid in determining sediment budgets: e.g. historical maps and documents
(which, for the Dutch delta, covers a period of approximately 200 years due to the regular morphological
monitoring from 1843 onwards), analyses of land use changes and numerical models for simulating hinterland
characteristics and/or delta development.
In the next section we present our integrated approach for enhanced coastal development scenario-building in
which the importance of our three key issues, long-term knowledge, socio-economic developments and sediment
budgets, are emphasized.
Innovative
Solutions
for the Delta
68
5 Results: The Integrated Delta Model
Our Integrated Delta Model (IDM) forms a multi-disciplinary integrated approach to enhance long-term and largescale future delta development scenarios. The previous chapters have demonstrated what we believe is important,
but still underexposed in present-day coastal research, namely:
1. The integration of long-term knowledge of the past (geological case studies) into physical models aimed at
forecasting coastal evolution.
2. Include the effects of human interferences (socio-economic developments) in coastal evolution models and
scenarios.
3. Focus on hinterland characteristics (assessment of sediment budgets).
We will first discuss our approach towards considering these factors in ICZM. The results of these considerations
ultimately lead to our integrated view on future coastal research: the Integrated Delta Model.
A geological approach in forecasting coastal evolution
Forecasting coastal evolution on large temporal and spatial scales is surrounded by uncertainties and incomplete
knowledge of the processes involved (Cowell et al., 1995). Important in forecasting long-term evolution however, is
to discover a trend (see Fig. 7).
From Fig. 7 it becomes clear that the overall trend determines
the direction a system will evolve to. Smaller scale fluctuations
are bounded by the larger scale trend and the effects of smaller
scale processes can usually be treated as noise on longer time
spans. Examples of smaller scale fluctuations are for instance the
seasonal differences, reflected by an increase of storm events (and
hence, erosion) during winter time and quiet conditions during
summertime (with as a result deposition of material). Although
these smaller scale processes might hinder recreation, they are of
minor importance on longer time scales.
Fig. 7 – Scales of variations in coastline position
(After: Terwindt and Kroon, 1993)
Discovering a trend in the evolution of a system can be reached by means of stratigraphic analysis (see section 3).
Incorporating stratigraphic knowledge into (numerical) models to predict future coastal evolution can be
considered valid since the stratigraphy records the results of sediment transport processes over time periods which
are long enough to distinguish mean depositional trends from fluctuations associated with storms and other
unpredictable occurrences (Stolper et al., 2005).
Stolper et al. (2005) introduce the GEOMBEST model as a tool to aid understanding of coastal evolution
at geological time scales. This model quantifies all assumptions involved in reconstructing geological history.
It provides a link between geological data gained from the stratigraphic record and our knowledge of coastal
processes. At present, GEOMBEST can be used as a tool for ‘inverse simulations’, meaning that the present
morphology is used to reconstruct geological history, but the model can also be used to forecast future shoreline
positions on geological time scales which are of importance for long-term coastal management.
Innovative
Solutions
for the Delta
69
Human impacts
We also need to evaluate the effects of socio-economic and demographic developments at the delta plain. One of
the methods to simulate the effects of people on nature and landscape is discussed by De Nijs et al. (2004). They
constructed spatially detailed land-use maps of the Netherlands for the year 2030, in which the effects of different
national economic and demographic scenarios were taken into account. This model is called the Environment
Explorer. It is important to note that the land-use maps do not show what the Netherlands will exactly look like in
2030, but rather show the possible changes in land-use if Dutch society were to develop according to one of the
scenarios.
At present, we consider this approach a promising tool in assessing how future land-use might develop
according to current spatial developments and also how the socio-economic system functions. However, we should
strive for a more complete picture, namely the linking of the socio-economic/ human system to the natural
(long-term) system.
Fig. 8 demonstrates that natural long-term coastal processes and
evolution interact with the human system and that the human
system exerts pressure on the natural system. Present land-use
models do not take into account the mutual interactions between
the natural and socio-economic system. These models assume that
the natural system merely provides the boundary conditions to
possible human alternations of the landscape, but they do not take
Fig. 8 – Coupling the natural long-term
into account that due to a combination of human interferences and
coastal evolution to a land-use model
natural evolution, these boundary conditions may shift in time. We
suggest establishing an interactive link between land-use models
and long-term and large-scale future delta development scenarios.
Hydrological Models
A reduction of river-supplied sediment will directly influence coastal retreat (Liquete et al., 2004). The importance
of monitoring terrestrial sediment supply to the coast and human perturbations on sediment supply was
recognized and put on the agenda by the International Geosphere Biosphere Programme (IGBP) through its core
project Land Ocean Interaction in the Coastal Zone (LOICZ; Syvitski, 2003).
Unfortunately, so far, not even 10 % of the global rivers have been monitored on their respective sediment
loads (Syvitski, 2003). Because of this scarcity of data with respect to sediment supply to the coast, scientists have
put a lot of energy in developing hydrological transport models which enable the prediction of sediment flux and
discharges of global rivers (Syvitski and Alcott, 1995; Morehead et al., 2003; Syvitski et al., 2005b).
In this paper we advocate using climate-driven hydrological models for generating synthetic water discharges
and sediment load records at the river outlet. An example of such a model is HydroTrend (Kettner et al., in
press). The model HydroTrend serves several purposes: (1) to simulate the discharge and sediment loads where
observational data are limited, (2) to assess human impacts on natural sediment loads (Syvitski et al., 2005b) and
(3) to gain insight in palaeo-discharges and sediment loads of the geological past in specific areas (Eastern coast of
the United States, Overeem et al., 2005; Adriatic Coast, Italy, Brommer and Weltje, 2006; Syvitski and Kettner., subm.)
Innovative
Solutions
for the Delta
70
As mentioned in section 4, we stress the importance of establishing a sediment budget for the delta in terms of
sediment sources and sinks. For our integrated approach, we believe it is important to use hydrological models
which generate sediment loads and discharges for past and future time intervals. Simulated sediment loads of the
past can be matched to the stratigraphic record of the delta to obtain sediment budgets with an estimate of the
residual uncertainty. As a result, we will obtain better knowledge with regard to the building blocks of the delta.
Calculating sediment budgets for any given deltaic area worldwide and integrating these budgets in forecasting
the long-term delta evolution provides a quantitative framework to estimate the future amount of space in the
delta.
Integrated Delta Model
Fig. 9 shows the Integrated Delta Model (IDM). Fluvial processes (hinterland characteristics), changes in sediment
budget (sediment transport processes) and human activities are at present omitted in forecasting long-term
large-scale deltaic changes and are therefore indicated with a cross. The mega scale (tectonic movements and
the geologically inherited substrate) provides the boundary conditions for the processes which governs changes
in the delta. These processes operate on different scales, making it hard to understand which process is of most
importance or produces the biggest uncertainty when dealing with long-term and large-scale deltaic morphology.
The present approach in coastal research is indicated on line three. Process-based and behaviour-oriented
models in conjunction with global climate change models and sea-level rise scenarios are used to simulate future
deltaic development.
At the bottom of Fig. 9, we indicate the previously discussed issues that should be incorporated in future delta
scenario building.
We argue that deltaic space is controlled by fluvial sediment supply and sea-level rise. We should, therefore,
not only focus on sea-level rise scenarios but also on changing sediment supply scenarios. Hydrological models
like HydroTrend can be used to determine past and future sediment loads, upon which these numbers can be
integrated in forecasting delta development.
The stratigraphic record provides important
lessons concerning past deltaic development,
and helps to determine the present-day
and future trend of delta development. The
GEOMBEST model offers possibilities to use
stratigraphic knowledge to forecast the
future trend.
To budget the delta and treat the delta as
a ‘bank account’ presents a quantitative
framework for forecasting scenarios. Finally,
socio-economic developments at the delta
plain are of utmost importance to take into
account. Land-use models are a potential
powerful tool to assess responses of the
delta to human alternations.
Fig. 9 – The Integrated Delta Model (IDM):
A multi-disciplinary integrated approach to enhance future delta
development scenario-building
Innovative
Solutions
for the Delta
71
6 Discussion and recommendations for further research
The IDM provides useful insight into all the processes operating in the delta on different temporal and spatial
scales and it shows the ‘white spots’ in present research activities with regard to forecasting future delta changes.
For a sustainable coastal zone policy, we argue that more research effort has to be spent in addressing the impact
of these white spots on delta development and also on examining how the different systems (natural and maninduced) influence one another.
Natural versus social system and model linking
The IDM, and in fact any other integrated model, exhibits a large range of complexity. This complexity is mostly
due to the fact that models simulating natural evolution are based on physical laws and equations, which govern
quantitative answers based on temporal and spatial changes, whereas models simulating social changes are
of a stochastic nature (e.g. models based on game theory principles), which normally only provides qualitative
information.
To examine how these different systems influence one another is therefore not a straightforward task.
However, in order to have an operational IDM rather than a merely conceptual model, we should strive for finding
ways on how to link the different models which constitute the conceptual model.
The problem of ‘linking’ natural and social systems is widely acknowledged (e.g. the KUST2005 research
program). So far, most social research in the delta has focussed on socio-economic changes in isolation from
the natural system. Ideally, we should have a means to quantify the human impacts in the coastal zone (e.g.
quantify changes in sediment budget due to human interferences, apart from the natural changes in sediment
budget). Syvitski et al. (2005b) strive for quantifying changes in sediment load due to human interferences with
a hydrological model. This model is a promising tool in simulating future changes in sediment load due to both
natural and human processes.
Fortunately, some links of the conceptual model may be easier to establish. For instance, some models which
are at present used to simulate coastal processes on engineering time scales (e.g. the behaviour-oriented PonTos
model (Steetzel & Wang, 2003)), offer possibilities to include time dependent sources and sinks and their impact
on coastal evolution, which can be provided by the hydrological model.
Calibration of the integrated delta model
In our research statement we stated that studying past deltaic behaviour provides datasets upon which physical
models can be calibrated. It is of importance however, to make a clear distinction in past deltaic behaviour in
response to climate change (both in hinterland and sea level) and in response to human interferences in the delta
plain (demonstrated in the case studies). Once this distinction is made, we can establish a “natural” dataset (deltaic
response to climate change, reflected in changes in the stratigraphic record) and a “human induced” dataset
(deltaic response to human interferences, by making use of historical data together with socio-economic scenarios).
With these two different types of datasets we can calibrate our conceptual model.
Innovative
Solutions
for the Delta
72
Validation of the integrated delta model
Validating models is a difficult task in any discipline. With regard to our case, the main difficulty is that we
demonstrate a conceptual model that should forecast coastal evolution at a large temporal and spatial scale. But
validating models which are used to forecast evolution that still needs to take place almost sounds irrational and
impossible.
However, even though this problem seems to have no adequate solution, we argue, again, that “data from the
past” may be used to validate (components of) our conceptual model.
Careful selection of several deltas that can be ranked in distinct classes (type of delta, sediment input etc.) can
be analysed on the key issues we proposed before.
Our workflow could be as follows: first we have to assess delta adaptation in response to natural evolution and
to human-induced changes. Then we determine a well constrained time interval, for example the past 150 years.
Subsequently, we examine how the model ‘predicts’ (this is called hind casting) coastal evolution in this time
interval. If simulations do not match, we need to calibrate our model again. This is an iterative procedure until we
have reduced our uncertainties to an acceptable level.
Sediment budget analysis also offers a dataset to validate our model upon. It is often possible to reconstruct
delta development by means of historical maps and documents. When compared to present-day digital elevation
or terrain models, volumetric changes of the deltaic space over a certain time interval can be calculated. This
‘historic’ result can then be compared to volumetric changes determined by means of numerical modelling. If both
results are in reasonable agreement, we may use the numerical model to estimate a future budget as a result of
changes in hinterland, sea-level or human interferences.
Uncertainties in forecasting long-term changes
In addition to the validation topic mentioned above, we should be aware of the fact that forecasting social and
natural changes (and their mutual influence) are surrounded by great uncertainties. We propose, in accordance
with a research initiative of LOICZ, long-term scenario constructing as one of the means to deal with these future
uncertainties. Scenarios provide a range of possible outcomes, rather than one answer (surrounded with great
uncertainty). This range of possible outcomes could then be communicated to the coastal manager. The scenarios
can provide answers on how future social and natural changes affect the level of risk in an area. Insight into risk
provides the coastal manager a tool to design appropriate (safety) measures.
Concluding remarks
In this paper, we have demonstrated that for a sustainable coastal zone policy, we need insight into the past and
future long-term large-scale deltaic evolution. In present day modelling exercises, important information with
regard to long-term natural processes, hinterland characteristics and human interferences are omitted. We have
provided an integrated approach towards dealing with all the relevant actors in the coastal zone. We recommend
using this integrated approach along with sea-level rise scenarios to forecast future delta space, accommodating
people, ecology and economics.
Innovative
Solutions
for the Delta
73
References
Battjes, J.A, 2006. Developments in coastal engineering results, Coastal Engineering 53, 121-132.
Berendsen, H.J.A., 2005. The Rhine-Meuse delta at a glance. 8th International Conference on Fluvial Sedimentology,
Delft, The Netherlands. Mid-conference excursion guide, 10 August, 2005.
Bondesan, M., 1990. L’area deltizia padana: caratteri geografici e geomorfologici. In: Libri, Spazio (Ed.), Il Parco dei
delta del Po: studi ed immagini, 2, 10– 48. Bras, R.L, Tucker, G.E, Teles, V., Six Myths About Mathematical Modeling in
Geomorphology. In: Wilcock, P. and Iverson, R (eds). 2003, Prediction in Geomorphology, AGU.
Brommer, M.B., Weltje, G.J., 2006. Provenance and sediment budget analysis of the late-Holocene prodelta in the
Adriatic Basin, Italy. In: Proceedings of Sediment 2006, 7-9 June 2006, Göttingen.
Ciabatti, M., 1966. Richerche sull’ evoluzione del delta padano. Giornale di Geologia 34, 381-410.
Coleman, J.M., Wright, L.D., 1971, Analysis of major river systems and their deltas, Technical Report No.95, Coastal
Studies Institute, Louisiana State University, Baton Rouge.
Correggiari, A.M., Cattaneo, A., Trincardi, F., 2005. The modern Po Delta system: Lobe switching and asymmetric
prodelta growth. Marine Geology 222-223, 49-74.
Cowell, P.J., Roy, P.S., Jones, R.A., 1995, Simulation of large-scale coastal change using a morphological behaviour
model, Marine Geology 126, 45-61.
Cowell, P.J., Stive, M.J.F., Niedoroda, A.W., de Vriend, H.J., Swift, D.J.P., Kaminsky, G.M., Capobianco,M., 2003. The
Coastal-Tract (Part 1): A conceptual approach to aggregated modelling of low-order coastal change, Journal of
Coastal Research 19(4), 812-827.
De Boer, D. H., 1992, Hierarchies and spatial scale in process geomorphology: a review, Geomorphology 4, 303-318.
Dronkers, J., 2005, Dynamics of coastal systems, Advanced Series on Ocean Engineering – Volume 25, World
Scientific Publishing Co. Pte. Ltd., pp. 519.
Gibson, C.C., Ostrom, E., Ahn, T.K., 2000. The concept of scale and the human dimensions of global change: a survey,
Ecological Economics 32, 217-239.
Haff, P.K., Limitations on Predictive Modeling in Geomorphology, in: Rhoads, B.L., Thorn, C.E. (eds.), 1996, The
Scientific Nature of Geomorphology: Proceedings of the 27th Binghamton Symposium in Geomorphology, 337-358.
Kettner, A.J., Syvitski, J.P.M., in press. HydroTrend version 3.0: a climate-driven hydrological transport model that
simulates discharge and sediment load leaving a river system.
Kroonenberg, S.B., 2006. De menselijke maat, De aarde over tienduizend jaar, Uitgeverij Atlas Amsterdam/
Antwerpen, pp. 334.
Kustverdediging na 1990, Eerste kustnota, 1990, Ministerie van Verkeer en Waterstaat, pp. 65. Hydrology and Water
management of Deltaic Areas, 1993, Ministry of Transport, Public Works and Water Management, report 93-5,
Centre for Civil Engineering Research and Codes, pp. 312.
Liquete, C., Canals, M., Arnau, P., Urgeles, R., Durrieu de Madron, X., 2004. The Impact of Humans on Strata Formation
Along Mediterranean Margins. Oceanography 17, 70-79.
Ministerie van Verkeer en Waterstaat, 1998, Water kader. 4e Nota waterhuishouding, Regeringsbeslissing, Ando,
Den Haag.
Innovative
Solutions
for the Delta
74
Morehead, M.D., Syvitski, J.P.M., Hutton, E.W.H., Peckham, S.D., 2003. Modeling the temporal variability in the flux of
sediment from ungauged river basins. Global and Planetary Change 39, 95-110.
Morton, R.A., 2003. An overview of coastal land loss: with emphasis on the Southeastern United States. USGS open
file report 03-337.
National Spatial Strategy; creating space for development (Nota Ruimte), 2004, Interdepartementale Projectgroep
Nota Ruimte, Ministeries van VROM, LNV, VenW en EZ, The Hague, The Netherlands.
Nelson, B. W., 1970. Hydrography, sediment dispersal, and recent historical development of the Po river delta, Italy.
SEPM Special Publication 15, 152-184.
Overeem, I., Syvitski, J., Hutton, E., Kettner, A., 2005. Stratigraphic variability due to uncertainty in model boundary
conditions: A case-study of the New Jersey Shelf over the last 40,000 years. Marine Geology 224, 23-41.
Phillips, J.D., 1992, Qualitative Chaos in Geomorphic Systems, With an Example from Wetland Response to Sea Level
Rise, Journal of Geology 100, pp. 365-374.
Steetzel, H. and Wang, Z. B., 2003, Development and application of a large-scale morphological model of the Dutch
coast, technical report prepared for National Institute for Coastal and Marine Management/ Rijkswaterstaat-RIKZ, pp. 79.
Stefani, M., Vincenzi, S., 2005. The interplay of eustasy, climate and human activity in the late Quaternary
depositional evolution and sedimentary architecture of the Po Delta system. Marine Geology 222-223, 19-48.
Stolper, D., List, J.H., Thieler, E.R., 2005, Simulating the evolution of coastal morphology and stratigraphy with a new
morphological-behaviour model (GEOMBEST), Marine Geology 218, 17-36.
Summerfield, M. A., 1991, Global Geomorphology: An introduction to the study of landforms, Longman Singapore
Publishers (Pte) Ltd., pp. 537.
Syvitski, J.P.M., Alcott, H.M., 1995. RIVER3: Simulation of water and sediment river discharge from climate and
drainage basin variables. Computers and Geosciences 21, 89-101.
Syvitski, J.P.M., Peckham, S.D., Hilberman, R., Mulder, T., 2003. Predicting the terrestrial flux of sediment to the global
ocean: a planetary perspective. Sedimentary Geology 162, 5-24.
Syvitski, J.P.M., Kettner, A.J., subm. 2004. Predicting discharge and sediment flux of the Po River, Italy since the Last
Glacial Maximum. Continental Shelf Research.
Syvitski, J.P.M., Kettner, A.J., Correggiari, A.M., Nelson, B.W., 2005a. Distributary channels and their impact on
sediment dispersal. Marine Geology 222-223, 75-94.
Syvitski, J.P.M., Vörösmarty, C.J., Kettner, A.J., Green, P., 2005b. Impacts of humans on the flux of terrestrial sediment
to the global coastal ocean. Science 308, 376-380.
Terwindt, J.H.J and Kroon, A., 1993, Theoretical concepts of parametrization of coastal behaviour, Large Scale Coastal
Behavior ‘93, U.S. Geological Survey Open File Report 93-381, edited by J.H. List.
Traditie, Trends en Toekomst, Derde kustnota, 2000, Ministerie van Verkeer en Waterstaat, pp. 122.
Van der Spek, 2005, Long-term Morphodynamics, lecture notes NCK Summerschool 2005.
Van der Veen, A., Otter, H., 2002. Scales in space, Integrated Assessment, Volume 3(2-3), 160-166.Visentini, M., Borghi,
G., 1938. Le spiagge padane. Ricerche sulle variazioni delle spiagge italiane. CNR Report, Roma, 7, pp. 137.
Wright, L D and Thom, B G, 1977, Coastal depositional landforms: a morphodynamic approach, Progress in Physical
Geography 1(3), 412-459.
Innovative
Solutions
for the Delta
75
Innovative
Solutions
for the Delta
THE FLOOD HOUSE CONCEPT
A NEW APPROACH IN REDUCING FLOOD VULNERABILITY
The Flood House Concept
A new approach in reducing flood vulnerability
Fourth nominee of The DeltaCompetition 200
HELEEN VREuGDENHIL (TEAM LEADER)
PhD Student Policy Analysis
LEO MEIJER
MSc Student Water Management
LARS HARTNACK
MSc Student Architecture
TIES RIJCKEN
PhD Student Strategic Product Design & Building Technology
Innovative
Solutions
for the Delta
Abstract
Deltas throughout the world are vulnerable to natural hazards. New Orleans provides a recent and obvious
example. We analyzed the situation in New Orleans and the Mississippi Delta after Hurricane Katrina has
passed, from a vulnerability perspective. Vulnerability can be subdivided into four components: threshold
capacity, coping capacity, recovery capacity and adaptive capacity. The vulnerability of New Orleans could be
decreased by increasing these capacities.
We propose Flood House as a concept to improve the four vulnerability capacities simultaneously. The
Flood House is a flood-proof water management centre, in which people, science, government and business
come together, first to remember victims of Katrina, but second to learn, discuss and create new – joint – ideas
on all kinds of water management and flood related issues. This results in innovative strategies, policies and
products to increase individual and collective safety against further future floodings.
Pre-conditions for success are adaptation to local circumstances and a differentiated approach to, and
platforms for, the different societal stakeholders. The Flood House facilitates an innovative approach to
governance on disaster management and could therefore even have a function during hurricanes. Once it has
been tested and proven itself, the concept could be transplanted to other delta regions throughout the world.
1 Introduction
Innovative
Solutions
for the Delta
78
The destructive force of water is high on the agendas of many scientist, politicians and people living in sea and
river shore areas. Global climate change research predicts rising sea levels, more storms and higher peaks in
rainfall. Next to this, recent dramatic events, such as the Tsunami in Asia and Hurricane Katrina in New Orleans,
have raised our awareness on the vulnerability of deltas.
In New Orleans, the floods caused by Katrina killed over 1500 people and sent 400.000 people in exile, which
is about 85% of the population. Until now, not even half of the population has returned to the city. New Orleans
is empty and devastated. People grieve. Federal aid is limited and the general moral is low. Comparable disasters
could happen in other deltas throughout the world.
In a way, New Orleans has an advantage over other deltas. New Orleans can be rebuilt according to new ideas.
It has real-life experience with disasters, momentum, and (literally) space for alternative protective measures. For
example, the Delta Works of the Netherlands could only be built after the floods of 1953. Maybe New Orleans can
also benefit from the disaster. But should it be by similar gigantic structural interventions such as the Dutch dams
and barriers? Maybe the 21st century demands for less prestige-oriented solutions. Subtle interventions could have
a positive effect on a shorter term, involve more people and can flexibly adapt to increasing insight on the longer
term.
Questions like ‘what exactly is safety’, ‘who is responsible for which threat’ and ‘how do you protect yourself
or the city against a flood’ are important to be considered by all stakeholders in New Orleans. Not only to recover
from Katrina, but also to prepare for future threats. This can be derived by increasing the threshold and coping
capacity, but moreover by being adaptive now and in the future.
The disaster in New Orleans shows the world the vulnerability of deltas in general and the Mississippi delta
specifically. At the same time it offers us an opportunity to propose a concept, the Flood House, which could
contribute to reduce this flood vulnerability. This is not a new storm surge barrier design, not a policy to get more
funding, no plea for restoring the whole area to its natural state, but a new attitude towards flood protection, in
which all members of society are heard and will be offered to do what they do best.
2 Concept of vulnerability
Vulnerability is a concept used to indicate ‘the sensitivity of a system for exposure to shocks, stresses and
disturbances, or the degree to which a system is susceptible to adverse effects’ (Turner, 2003; IPCC, 2001).
Disturbances can be both exogenous or endogenous (e.g. a drought can be caused by low river flow, bad water
management or both). Graaf and Ven (2006) identify a combination of four components of vulnerability in deltas.
These four components are the ‘Threshold capacity’, ‘Coping capacity’, ‘Recovery capacity’ and ‘Adaptive capacity’.
2.1 Threshold capacity
In all deltas there is a certain threshold capacity. The threshold capacity consists of the volume and flux of water
that the water system can handle without causing any damage to everyday life. For example, the threshold
capacity could be determined by long term discharge forecasts resulting in a system of dikes and a centralized
spatial planning for the water system. The spatial planning could include the use of the water system as a leading
principle, resulting in restrictions on building, the widening of rivers and floodplains and the application of
waterproof building concepts and technologies such as floating constructions in hazardous regions (Commissie
Waterbeheer 21e eeuw, 2000; Smits, 2000).
2.2 Coping capacity
If the volume of water is larger than the actual discharge capacity of the water system, the excess of water will
cause flooding. The ability of a society to deal with floods or excessive rainfall is described by the term coping
capacity. There are two main aspects determining this coping capacity. The first is the management of the excess
of water. If there is an agreement or an organisation determining where the excess water should flow, supported
by a physical structure, the flood could be (re)directed and controlled in order to minimise damage. Emergency
inundation polders as applied in the Netherlands are an example of attempts to control exceptionally high river
discharges. The second aspect of coping capacity is disaster management. This includes the organisation of
the evacuation of inhabitants before and after an area is flooded, but also preventing looting, evacuation of pet
animals and reducing damage to loose objects such as cars.
2.3 Recovery capacity
The third component of vulnerability is the recovery capacity. This component deals with the effort, costs and time
for the system to recover to the original (or desired) state. The system comprises many elements, including ecology,
economy, livability and water management. For instance, in many flooded areas huge unexpected problems arose
with household and industrial chemicals, which caused severe environmental pollution and put a threat on the
livability of the area. The costs and time to undo these negative effects and to make the area livable again are main
indicators of the recovery capacity. Furthermore, the ability to manage the period in between the disaster and the
fully recovered status is also part of the recovery capacity.
Innovative
Solutions
for the Delta
79
2.4 Adaptive capacity
In contrast to the other three components of vulnerability, the adaptive capacity is a rather holistic notion. In
general, adaptive management is an approach based on collaboration among agencies, researchers and local
stewards. It sees resource management as a continuous learning-by-doing process that recognizes public
participation and collaborative learning (Stockholm International Water Institute, 2005). Adaptive management
recognizes structural uncertainties. Resilience in the ecosystem and flexibility in the management institutions
are essential factors for adaptive management (Gunderson, 1999) in order to deal with these deep uncertainties.
Pragmatic solutions for water management would be to seek for restoration of resilience and flexibility, which
can occur through novel assessments, small-scale experiments or when an unforeseen policy crisis allows for
reformation or restructuring of power relationships among actors (Gunderson, 1999). To increase the adaptive
capacity, ecosystem processes would be the leading factor to which other policies are adapted instead of
increasing control over the resources. Furthermore, the institutions should have certain flexibility and can change
easily if the situation asks for it. As such, a net could be developed in case an aspect fails.
These four components of vulnerability are interrelated. For instance, in case of an increased adaptive capacity,
the pressure on the other components decreases. A resilient ecosystem is often able to store large water quantities
and could in that case reduce a flood peak. A reduced flood peak could imply less stress on the threshold capacity
and since the chance for a flood and subsequent damage decreases, coping and recovery capacity are less needed.
3 Katrina and New Orleans
The history of New Orleans showed that Katrina was not a unique event. The most recent hurricanes with large
social impacts include ‘Camille’ and ‘Betsy’ in the 1960’s. In fact, a disaster like Katrina has been forecasted by
Fischetti in Scientific American in 2001 and by Brouwer in the Civil Engineering Magazine in 2003. Fischetti warned
for the threat from the sea: with the disappearing wetlands the protective nature of the river delta at the coast
around New Orleans is disappearing. Brouwer focused on the threat of Lake Pontchartrain that in case of high
speed winds from a certain direction could overtop and breach the levees. This, in combination with the heavy
rains, is exactly what happened during the Katrina disaster.
3.1 Threshold capacity: wetlands and levees
The threshold capacity of New Orleans is derived from a natural system of wetlands and water storage bodies
and from an artificial system of levees, pumps and canals. New Orleans is located along the largest wetlands of
the USA. These wetlands function as hurricane and flood protection (Restore America’s Estuaries, 2006). However,
several threats, both natural and human, are diminishing this natural buffer. The area is subsiding with high speed
(Dixon, 2006; Fontenot, 2006) and economic developments and urbanisation reduce the size of the wetlands.
With respect to the civil structures, it turned out that the quality of the levees was marginal (Rogers, 2006;
Independent Levee Investigation Team, 2006) as a result of limited interest, subsequent marginal funding (Seed,
2006) and poor cooperation between the many levee boards (Sharky, 2006). Due to their limited quality of they
could not withstand the heavy winds, flood and rains, resulting in several breaches of levees that should have
protected the city. Since half of the city lies below sea level, a breach means serious inundation for large areas of
the city. Besides the levees, there were limited measures as compartimentation, flood proof architecture or excess
water storage basins.
Innovative
Solutions
for the Delta
80
Box 1: Report on Hurricane Katrina.
[Rijkswaterstaat, 2005; NOS, 2005]
Katrina
Friday, 25 August, 2005, Hurricane Katrina moves over the southern tip of Florida into the Gulf of Mexico. As
predicted, the hurricane strengthens to a category 3 on Saturday, 27 August. Voluntary evacuation starts in the
threatened section of the coast. The highway lanes around New Orleans are used to transport evacuees and
the Louisiana Superdome is announced as a shelter of last resort. On Sunday, 28 August Katrina increases in
strength to Category 5 on the Saffir-Simpson scale. Despite a mandatory evacuation order for New Orleans, an
estimated 100,000 people remain in the city.
On Monday morning Katrina lands and causes massive damage. Initial reports speak of a total of 55 death
and $60 billion in damage. As the winds above New Orleans subside, the city seems to have been spared.
However, within a few hours, water from the Gulf of Mexico and from Lake Pontchartrain begins to inundate
the city through breaches in the levees and the worst case scenario becomes a reality. The main disaster thus
did not directly come from Katrina, but rather from the winds raising the water levels of Lake Pontchartrain to
immense heights and the heavy rains. When about 80% of the city flooded, over 1500 died in the devastating
forces of the water and entire communities became homeless.
Besides the personal disaster, Katrina caused huge damage to the economy and the environment. Part
of the people remaining in the town did not want to leave, but another part could not leave because of the
deficiencies in the evacuation plan that did not fully account the large number of people without private
transport. Only days after the storm, all people were evacuated, but for many it was too late.
3.2 Coping capacity: disaster management
The coping capacity of New Orleans was compiled in a disaster management plan including evacuation of citizens
and water disposal by pumping. Technical failures (e.g. electricity breakdown), limited experience and familiarity
with the plans, and unexpected human behavior caused unforeseen developments. For instance, the City of New
Orleans waited quite a long time before issuing a mandatory evacuation order possibly in fear of being held
liable for a loss of income had the hurricane not affected the city (Rijkswaterstaat, 2005). Local officials were
overwhelmed by the magnitude and consequences of the disaster including break-downs in vital communication,
thereby causing wide-scale confusion and delays, such as
not putting all the available buses into action for evacuation.
Other examples causing unexpected developments were
the breakdowns of the pumps and the love of people for
their pets. Eventually, about 100,000 people remained in the
city during the hurricane. These were mainly poorer people
without possibilities of transportation and they were mainly
living in the most affected areas. Many evacuees found
shelter in the Superdome or other cities such as Baton Rouge
and Houston.
Fig. 1 – The New Orleans ‘Diaspora’
(Source BBC News)
Innovative
Solutions
for the Delta
81
3.3 Recovery capacity: a devastated and abandoned city
After the hurricane, recovery faced –and still faces- many practical problems. Pumping the water out of the city,
cleaning the city from pollution and rebuilding the infrastructure, houses and public buildings are a few of the
challenges New Orleans is facing. Due to limited pumping capacity, it took several months before all the excess
water had been disposed of. Excess water causes infrastructural problems and mud deposition, but also spreads
and severe pollution from households and industrial chemicals. Recovery of the system from this pollution takes a
long time, thereby affecting both environment and public health. The first priority of the United States Army Corps
of Engineers (USACE) was to restore the levees to the pre-Katrina level of safety before the new hurricane season
(Setliff, 2006), in which they succeeded (USACE, 2006).
To rebuild the city, a committee (Bring New Orleans Back) has been installed by Mayor Ray Nagin. Their main
task is ‘to keep the city culturally intact while at the same time improving the areas that were marginal’ (Legarde,
2006). Difficulties include differences in perspectives on urban and spatial planning and limited resources. The
federal government limits its help to New Orleans ‘because it would be inconsistent with our responsibility to US
taxpayers to distinguish between deltaic regions’ (Duncan, 2005).
An important aspect in the recovering the city is the return of the people of New Orleans. In January 2006,
only an estimated 125.000 people resided in New Orleans, 300.000 have not yet returned (Sharky, 2006). The
expectation is that in a few years only half of the city will return. The people most affected by the hurricane were
those living in the low-lying parts of the city and they were generally poorer people. They showed ability to move
elsewhere and started a new life. They often do not want to or cannot return. This diaspora has major implications
for New Orleans, including the limited availability of workers to rebuild infrastructure, a drop in tax income and
abandoned neighbourhoods with possibilities for pauperization and criminality.
Less obvious are organisational and planning problems for neighbourhoods where the majority of the houses
are empty and demolished, even after a year. For instance, schools with less than 20% of the students present are
hard to run and loosing the black community with the specific jazz culture is a loss for the cultural value of New
Orleans as a whole. On the other hand, more space could offer opportunities in reconstructing the city in terms
of quality of neighbourhoods and water storage. Not only for New Orleans the diaspora has major implications,
also for the cities where the refugees went, this sudden increase in number of inhabitants causes problems. For
instance in Houston, a direct extra police force was lacking that could deal with the increase in criminality (City
Journal, 2006).
3.4 Adaptive capacity: reborn wetlands and innovation opportunities
The resilience of the water system of New Orleans has been limited in the past by, among others, canalisation,
restricted sedimentary processes, reduced wetlands and ground subsidence. This made the area vulnerable for
floods. The limited flexibility in institutions appeared in institutions that did not overtake responsibilities and the
absence of scenarios and funds for failing mechanisms.
The adaptive capacity, as hard as it is to measure, will reveal itself the coming years and decades. Ecosystembased management and participatory approaches offer possibilities to increase the adaptive capacity (Gunderson,
1999; Pahl-Wostl, 2006). For instance, using the available space in the city for water storage if needed or restoring
natural processes in the wetlands thereby slowing down speed of subsidence (Saeijs, 2004), could increase the
resilience of the ecosystem. The business-spirit and persistence of the Americans could also contribute to the
general adaptivity. Facing the damage of the flood so directly should provide means for innovation by developers,
insurance companies and entrepreneurs. The coming years, the integration of the physical and societal system is a
challenge for New Orleans.
Innovative
Solutions
for the Delta
82
4 Water management culture
Every country has its own water management culture developed over time. As such, different countries could learn
from each other or at least better understand their own culture if compared with other cultures. In understanding
some aspects of American attitude towards water management, a brief comparison with Dutch attitude towards
water management is depicted. This understanding gives input for fundamental discussions at the Flood House.
The differences are derived from cultural, socio-economic and institutional characteristics. A first difference is
the role of the government to protect residents and the individual responsibilities of residents. In the Netherlands,
the role of the government in water management is very large, which results in large-scale spatial plans and
water management plans all directed by governmental organisations. Private partnership and public participation
are limited, although the Dutch government acknowledges the importance of this. The trust in the government
arranging protection against floods is generally large and people are willing to spend tax money on this. As a
result, the sense of collective safety is large, but this approach also implies that individual choices are usually
subordinated to general policies. In some cases people even have to move out if a water management measure has
been planned on where they live. In the USA the focus is much more on the individual safety. This could result in a
larger adequacy in dealing with disasters of individuals, but the management of the water system as a coherent
system is less adequate.
Another difference is the attitude towards risks. In the Netherlands, risks are diminished as much as possible.
Dealing with risks in general enlarges the opportunities for innovations on a business level, but limiting risks for
floods could increase safety. However, reducing flood risks creates a paradox of a diminished level of risk-awareness
and a subsequent reduced level of ability to deal with risks and investing in originally flood-prone areas. In case
of flooding, the damage will be even larger. A society should ask itself to what extent it is willing to pay to reduce
risks. According to Link (2006) the limited involvement of both government and people in public safety can be
explained by the limited available resources for flood protection. The importance of return on investments is
another explanation why the investments needed to reduce risks are lower in the USA than in the Netherlands.
Fig. 2 – Three pillars or ‘floaters’ under the concept of The Flood House, creating a ‘trimiran’ (compare: catamaran)
Innovative
Solutions
for the Delta
83
5 Trimaran: A new water management approach
New Orleans is currently trying to
recover from Hurricane Katrina. Large
numbers of inhabitants have not
returned yet. With the inhabitants,
a large part of New Orleans’ specific
culture got lost. Creating job
opportunities, rebuilding the city and
building trust will contribute to the
return of the inhabitants. However,
focus should not be single-sided
on recovery. It is important to look
ahead and consider opportunities
to decrease the vulnerability as a
whole for future threats. Therefore,
serious reconsiderations on the water
management system are necessary.
Questions should be raised and old
viewpoints reconsidered. What level
of collective safety do we strive for?
To what extent are we willing to pay
for this level of safety now and in the
future, and who is responsible for
what? What is the best combination between large structural interventions such as storm surge barriers and other
measures like smart spatial planning and restoring the natural water system and wetlands?
Open public debates could increase understanding and awareness and could also lead to creativity and
supported management actions, which is a precondition for success. Scientific and creative ideas could be
transferred into entrepreneurial activities. Integration of societal actors, collaboration between stakeholders and
participatory planning are part of a global democratic trend, relevant in the 21st century (Rischard 2005). However,
offering a platform for integration is only sensible and feasible if it is adapted to the local characteristics of New
Orleans and the US water management approach. These three components form the basic structure of the Flood
House, represented by the ‘trimaran’ in figure 2.
Innovative
Solutions
for the Delta
84
6 The Flood House
From an historical perspective, the main focus of water management in New Orleans and many other deltas has
been on increasing the threshold capacity. Nowadays, with the new knowledge on interrelated physical and social
elements of the water system, a more holistic approach towards flood vulnerability would be more appropriate.
Governments are expected to offer a certain level protection against floods, but government responsibilities
and effectiveness are not always sufficient to achieve this. If more information about the many aspects of flood
vulnerability had been provided and different stakeholders had been acting integrated, responsibilities would have
been clearer, people would have been better prepared and flood related business opportunities would have arisen.
Therefore, governments, businesses, people and science should find a way to effectively collaborate. They should
grasp and discuss the concept of flood vulnerability scientifically resulting in action and entrepreneurship. This can
happen in an environment that is both neutral and stimulating.
The Flood House is a multi-purpose, flood-proof, information centre and platform in New Orleans, in which one
can find all kinds of flood management related issues and products. Through multiple activities awareness for the
shared problems will rise. The Flood House should be self supporting and will have four major functions:
>> Educating and informing people about water management and related issues;
>> Enabling people to buy products that will increase their collective and individual sense of safety;
>> A conference and knowledge centre that will bring together several actors involved in water management and
will combine their (research) activities;
>> Memorial for the people who lost their lives during the Katrina disaster.
The Flood House concept could be regarded as an attempt to increase the adaptive capacity in New Orleans, since
it encourages and stimulates collaboration among different type of actors and stakeholders. Furthermore, by
the nature of the education and knowledge forums, increases in resilience of the ecosystem and flexibility of the
institutions could be reached. Because of the current crisis in the New Orleans water management, a ‘window of
opportunity’ (Kingdon, 1984) has opened that could be used to structurally increase the adaptive capacity and as
such decrease the vulnerability of the delta as a whole.
The design of the Flood House should allow for flexibility and it functions may have to change in time. The
actual decision over the appearance and location of the Flood House should be left to the people of New Orleans.
In a design competition everybody could be allowed to participate and present their design. The people of New
Orleans should in the end decide which Flood House should be built.
Innovative
Solutions
for the Delta
6.1 Functions of the Flood House
Survival Shop
In the Flood House products related to floods will be advertised and sold. Experts give advise to minimise flood
damage and on life saving products, tax structures, risks, insurances, and so on. A number of ideas are listed below
and some of them are illustrated with exemplary posters.
Products to increase the threshold capacity:
>> Adopt-a-dike: this program connects civilians to a certain piece of dike, either financially or with practical
activities such as inspection, mowing or maintenance. In return they will be honoured similarly to the existing
American “adopt-a-highway” programmes.
Innovative
Solutions
for the Delta
>> Adopt-a-wetland: the principle of this program is equal to that of the previous one, but here pieces of natural
wetlands can be adopted by (groups of) civilians or schools.
>> Garden water storage: farmers or people with big gardens are encouraged to store large quantities of rainwater
in their garden during excessive rain showers.
>> Waterproof buildings: project developers can advertise flood-resistant (highrise) buildings in amphibious zones,
such as the proposed Greenway Buffers (Sharky, 2006).
Products to increase the coping capacity:
>> Life jackets: Everybody in New Orleans should have a life jacket and should know how they work. In the shop life
jackets will be offered for only $2, subsidised by governmental grants and donations to the ‘Flood House Fund’.
>> Boats: The shop will offer several kinds of small boats. From inflatable rowing boats to small motorized boats.
>> Floating cars and attributes: amphibious cars and external air-bags and lifelines prevent vehicles submerging
or floating away and damage during a flood.
>> Home flooding kit: a waterproof box for storage of passports, insurance policies, chemicals, (non-) prescription
drugs, etc.
>> Life-in-a-bubble: protection from water and pollution by a plastic sleeve that automatically covers your house
during a flood.
>> Floating technology: floating houses, amphibious or on existing water, do not get damaged during a flood.
>> Floating dog kennels: when a flood rises, a dog or cat might retreat to his kennel, which remains floating.
Possibly, it automatically locks when the dog is inside and the kennel has reached a certain height.
Products to increase the recovery capacity:
>> Instant house: A house that will be self supporting and can be dropped shortly after a flood and will provide
temporary housing.
>> Temporary utilities: generators, water pyramid (Nitsche, 2006), vaccinations against diseases that will be there
after a flood, temporary hospitals.
>> First-aid-after floods: health courses specially focussed on flood related problems.
>> Visual shields: plants or fences hide ugly damaged and abandoned houses.
Innovative
Solutions
for the Delta
87
6.2 The Flood House Information Centre
The Flood House provides space for research and development of new flood protection measures. It does not
facilitate full-time researchers, but brings together the people of New Orleans, governmental, business and
residential organisations that focus on water management, disaster management and the reconstruction of the
city. Conferences can be held and the acquired knowledge can be exported to other delta areas. This knowledge
centre will be easily accessible for people, business and government. A website will be maintained and it will hold a
large library on flood related literature.
Many people in New Orleans did not realise the importance of a sound water system. The possible effects
of hurricanes in every day life are not very clear. In the Flood House permanent education will be given on water
management and the water system. Special programs for specific groups should be offered to create a better,
general understanding on the water management system of the city. This could lead to in-depth discussions on
water management. The main focus in this part of the Flood House could shift in time: the coming years there
is plenty to tell and learn from the Katrina disaster, but in the future other topics could be interesting as well.
The measures and changes in the water system can be explained to civilians, business and government. Disaster
management strategies can be discussed and adjusted with everyone involved.
Examples for education and workshops include:
>> Flood protection and disaster management
>> Innovative technologies on water management
>> Spatial planning (e.g. the role of wetlands in water management and the role spatial planning could serve in
preserving them)
>> Architecture: Advice on flood prone building and furnishing of your house
>> Institutional discourses (e.g. responsibilities, implications of changes)
>> Field trips for schools or kids parties (e.g. biology classes)
>> Workshops in the Louisiana wetlands: ranging from ecology to the importance of wetlands for economy (e.g.
nation’s oil and gas supply, seafood)
Innovative
Solutions
for the Delta
88
6.3 Cultural activities
After the Katrina disaster a permanent memorial is needed for the people who lost their lives to the hurricane. This
memorial should have a central place in the Flood house. In the meeting area several general features as a bar and
stage area, lavatories and management offices can be installed. Besides of serious issues, the Flood House could
also function as a community place, where people can have fun enjoying jazz concerts and eating in the ‘Water
Restaurant’. This restaurant serves all kind of local dishes with food derived from the wetlands.
6.4 Financing
The Flood House concept taps into the money that individuals are willing to spend on flood protection, businesses
on exposure and knowledge development and governments on water management in general. Considering the
financial and emotional damage to restore and prevent, there should be enough money available. It can be stated
that the Flood House has no competitors, since it aims at embracing and facilitating anyone who works in the field.
The Flood House will be a not-for-profit organisation that is financially self supporting. The financial structure is
divided in ongoing (daily) budgeting and project-based budgets. Each budget gets income from profit, subsidies or
donations, depending on the nature of the project, participants and responsibilities. The profit made in the shop
and knowledge centre will be used for the general maintenance of the Flood House and to attract new artists and
inventors to develop new products or services. Excess money is stored in the ‘Flood House Fund’ to finance special
projects.
The products and programs should allow all people to participate, independent of their social or financial
status. Several basic survival products, such as the life jacket, will be offered very cheap. The poorer people can also
participate in the adopt-a-dike program by doing maintenance. If enough investors can be found for this program,
there might even be financial compensation for the people who offer to do the maintenance. The wealthier people
can invest in dikes, increase their safety and as such decrease their insurance premium. To get the Flood House
started, a number of organisations should collaborate and invest. Ideally it should be a balanced mix between
governmental, corporate and individual capital. The City of New Orleans, some federal money, insurance companies
and developers should unite for their benefit and the benefit of New Orleans.
Innovative
Solutions
for the Delta
89
6.5 Marketing
The architecture competition and the idea in general will probably get enough media attention to get known
throughout New Orleans. The first feature of the marketing strategy is on getting people to the Flood House. Jazz
festivals, carnival activities and the water restaurant could be used to attract people. Since the Katrina disaster is
still very present in the memory of many people, there will be a general interest in the city’s water management
strategy. When people enter the Flood House they will first see the memorial for the people who lost their lives
during the disaster. After this they will be led to the informative and educative section. Changing expositions and
general information will coincide to give people – from all walks of life and all ages – the things they need to know
on the (future) water management of New Orleans. People can be informed about the activities in the Flood House
by flyers. Information is of course also available on the website, which could serve as an easily accessible forum for
several topics.
When people have gained the information they are being given the possibility to participate. In the survival
shop there will be both practical and simple products that can be used instantaneously and more sophisticated
products and programs. People can commit themselves to these programs by either financing them or
participating, such that every income group has access to knowledge and products. These programs will have
to be actively supported by the local government and it should – in time – become a social naturalness for every
inhabitant of New Orleans. In order to get people used to these products some will be given away for free during
competitions or lotteries. Also the participation should become something important or fashionable. People
should feel they mean something to society.
Besides of individual inhabitants, other groups include schools, charity, NGO’s, social clubs, businesses and
government. Schools educate the next generation and the Flood House could contribute to the educational
program of the school. Social clubs could decide to collectively participate in larger projects such as adopt-a-dike
and as such contribute to society. Businesses could be attracted by offering an extended network of organisations
and contacts nearby, public relations, and practical issues such as conference availabilities. Governmental
organisations could be attracted to the idea of the Flood House for ethical reasons, but moreover actively gain
profits from increasing public participation.
Innovative
Solutions
for the Delta
90
Box 2: A new hurricane scenario (fictional)
[CNN, 2037]
Melissa
Thursday, October 15th 2037, Hurricane Melissa moves into the Gulf of Mexico and strengthens to a category 5
on Friday. The hurricane is aiming for New Orleans. It is the end of the hurricane season and due to the global
climate change the hurricane season lasts longer and category 5 hurricanes are no longer an exception. The
weather services have calculated that it will pass New Orleans approximately 50 kilometres to the west. So
the storm surges are closed and everybody has collected some extra food and water. Mandatory evacuation
orders are not given anymore because of legal conflicts of the past. Inhabitants are expected to decide for
themselves if they prefer to stay or leave. Sunday morning, Melissa hits land. The hurricane did not follow its
predicted pathway, but hit New Orleans right in the middle. All authorities are puzzled, how could the weather
services be so wrong? After all, there were billions of US dollars invested in a better warning system? With
wind speeds over 400km per hour and heavy rain showers this is the biggest storm ever to hit the US, and it
hits New Orleans right in the middle.
The city is flooding piece by piece, but in a slow and predicted way. The restoration of the wetlands had a
balancing effect on the storm and the levees were in a good state of maintenance. The compartimentation
of the city has resulted in a predictable pattern of flooding. The new built area, with floating homes was
one of the first areas to flood, but with very little damage. Other areas followed. The city alarm did work and
people were able to take their last minute preparations on there homes, such as storing valuable things and
chemicals in the special designed boxes. Other people took shelter in there boats and the precautionary
measured seemed to work.
Around 8 o’clock the next morning the Flood House starts to undertake action. Together with the police,
in boats, they start to hand out blankets and a hot breakfast. Some of the injured people were taken in the
Flood House for first aid. The neighbourhood coordinators knew where aid was needed most and where
people could assist other citizens. The first help was very efficient. The city had more or less anticipated on an
excess of water and emergency power and pumps were installed. After two weeks the water was pumped out
of the city and people started to rebuilt their houses. The Flood House has found its place again and the first
exhibition is planned. The rebuilding plans are made and will be presented, and there will be a debate on flood
management. Ray Nagin will be a giving a lecture about the differences between Melissa and Katrina.
Innovative
Solutions
for the Delta
91
7 The future of the Flood House
7.1 Melissa: Katrina’s bigger sister
In box 2 a fictional future scenario is presented. The suggested measures and products will not prevent the
next disaster happening. However, it does show what role the Flood House could play in New Orleans. The
products for sale and implemented programs will result in an increase of the threshold capacity. Due to the fact
that inhabitants and authorities are better informed about the water management and are familiar with the
emergency plans, less surprise and panic occurs. As such, the coping capacity has been increased. Since authorities
and citizens are prepared for a disaster, the process of cleaning up and restoring the city is also a lot easier. The
amount of damage is limited and major environmental pollution has been prevented. The fact that people can deal
with a disaster is a result of an increased adaptive capacity. By informing and educating people, through shared
knowledge and collective planning the Flood House has become a stage of incredible value.
7.2 Transplantation of the Flood House concept
The Flood House is designed for New Orleans, but the concept could offer benefits in many other deltas
throughout the world. The term institutional transplantability is used to describe the borrowings of institutional
experiences from elsewhere (Mamadouh e.a., 2002). De Jong (2002) indicate several factors and conditions for
success. The first includes the careful consideration of the relation between the goals of the transplantation and
the actual transplanted institution. In other words, the concept should actually have proven itself in New Orleans
and should potentially contribute to the goals of the specific area to which it is transplanted. Only if the conditions
of the host country are clear, transplantation should be considered. Adaptation to local circumstances is crucial.
A second factor mentioned is that an idea in general is more likely to be transplantable than specific legal
frameworks or procedures. This implies for the Flood House that the concept itself should be considered as
transplantable, and not too much the results that emerge from it. Another condition for success is that the
transplant should be considered as a loosely defined model rather than a strict model to be followed. In fact,
this factor stresses again the need to adapt to local circumstances and use those parts of the concepts which
actually contribute something and leave others out. A fourth factor for success is the timing of transplantation.
Periods with a sense of emergency and urgency, facilitate the transplantation process. This also holds for the
establishment of the concept, since only after a disaster like Hurricane Katrina a concept like the Flood House could
emerge. In periods of stability, transplantation is less easy to accomplish.
In case of the Flood House, transplantation seems possible and reasonable since it is a concept that is subject
of transplantation. It could be attractive not only for the host country, but also for the donor country since it
increases knowledge and availability of expositions that could be exchanged. However, it is crucial to translate
the conceptual ideas to the local institutional, physical and socio-economic circumstances as was done for New
Orleans. Before considering transplantation, the concept first needs to prove its value in New Orleans and it should
be very well explored where to transplant it. The European Union is searching for the establishment of practical
solutions to enhance active involvement of all interested parties in implementation of water management
practices (HarmoniCOP, 2005). The Flood House concept could be such a solution and as such, the European Union
or the individual member states should keep an eye on the development of the Flood House in New Orleans.
Innovative
Solutions
for the Delta
92
Acknowledgements
We would like to thank the following people for their contribution to this paper and the development of the Flood
House concept:
1. Tineke Ruijgh-van de Ploeg
2. Jill Slinger
3. Stef Janssen
4. Darby Grande
References
Literature
Brouwer, G., (2003). The creeping storm. Civil engineering magazine.
Commissie Waterbeheer 21e eeuw (2000). Waterbeleid voor de 21e eeuw.
Rijkswaterstaat (2005). Coastal hazard and disaster management in the United States, Lessons for the Netherlands.
De Jong, M., Lalenis, K., Mamadouh, V., (2002). The theory and practice of institutional transplantation, experiences
with the transfer of policy institutions, Dordrecht: Kluwer academic publishers.
Dixon, T., Amelung, F., Ferreti, A., Novali, F., Rocca, F., Dokka, R., Sella, G., Kim, S., Wdowinski, S., Whitman, D. (2006).
Space geodesy, subsidence and flooding in New Orleans. Nature, 441, 587-588.
Fischetti, M. (2001). Drowning New Orleans. Scientific American.
Gunderson, L., (1999). Resilience, flexibility and adaptive management – Antidotes for spurious certitude?
Conservation ecology 3(1):7.
Harmonising Collaborative Planning (HarmoniCOP) (2005). Learning together to manage together- improving
participation in water management.
Independent Levee Investigation Team, (2006). National Science Foundation.
IPCC (2001). Climate change 2001, impacts, adapation and vulnerability, Section, Hydrology and Water Resources.
Report of Working Group II of the Intergovernmental Panel on Climate Change.
Kingdon, J. (1984). Agendas, Alternatives, and Public Policies. New York: HarperCollins.
Mamadouh, V., de Jong, M., Lalenis, K., (2002) ‘An introduction to institutional transplantation’. In: The Theory and
Practice of Institutional Transplantation. Experiences with the Transfer of Policy Institutions, de Jong, M., Lalenis, K.,
Mamadouh, V. (eds). Kluwer Academic Publishers: Dordrecht.
Pahl-Wostl,C. (2006) The Importance of Social Learning in Restoring the Multifunctionality of Rivers and Floodplains.
Ecology and Society 11 (1): 10.
Rischard, 2005 ‘Vijf voor twaalf’. Uitgeverij Lemniscaat, Rotterdam.
Saeijs, H., Smits, T., Overmars,W., Willems, D., (2004) Changing Estuaries, Changing Views, Rotterdam.
Smits, A.J.M., Nienhuis, P.H., Leuven, R.S.E.W. (2000) New approaches to river management, Leiden.
Stockholm International Water Institute, (2005) Resilience: Going from Conventional to Adaptive Freshwater
Management for Human and Ecosystem Compatibility.
Innovative
Solutions
for the Delta
93
Turner, B.L., II, 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., (2003) A framework for vulnerability analysis in sustainability
science. Proceedings of the National Academy of Sciences of the United States of America 100 (14) (8 July),
8074–8079.
Graaf, R.E. de and F.H.M. van de Ven (2006, submitted). The Closed City as a strategy to reduce vulnerability of urban
areas for climate change. Conference on Innovations in coping with water and climate related risks. International
Water Association. Amsterdam, 2006.
Newspapers, interviews and presentations
City Journal, (4 January 2006). Katrina Refugees Shoot up Houston. FEMA should help the Texas city to control its
crime spike. By: Nicole Gelinas.
Duncan, “Expert Views On Hurricane & Flood Protection & Water Resources Planning For A Rebuilt Gulf Coast”
(20 October, 2005) (website: http://www.house.gov/transportation/press/press2005/release116.html) visited:
May 2006.
Fontenot, (2006) New Orleans Interventions in the Urban Landscape, conference lecture March 2006 Nirov, the
Netherlands.
Legarde, (date unknown) BNOB committee in: BBC News, New Orleans rebuilding unveiled.
Link, E. University of Maryland/USACE, in: Diane Rehm Show, 11 March 2006.
NOS journal (8 September, 2005). New Orleans in dodelijke kom.
Nitzsche, M. (2006). Waterpyramid. Internet: www.waterpyramid.nl, visited June 2006.
Restore America’s Estuaries Newsletter (2006), 3rd National Conference and EXPO on Coastal and Estuarine Habitat
Restoration (2): 4.
Rogers, D, (1 June 2006) University of Missouri. Drilling for truth in New Orleans, a geologist’s story. In: Q&A, Nature,
441, 556.
Sharky, B. (2006) Living on the Edge, rebuilding a city in uncertain terrain. Presentation at Slim.nu conference, The
Hague.
Seed, R., University of California, in: Diane Rehm Show, 11 March 2006.
Setliff, C.L., USACE. In: Diane Rehm Show, 11 March 2006.
USACE, Weekly Focus (31 May 2006) Repairing New Orleans Hurricane Protection System Before 1 June.
NRC next (2 June 2006) Niet klaar voor een nieuwe Katrina p.6-7.
Innovative
Solutions
for the Delta
94
A Hydrologic Flood Forecasting System for Mesoamerica
Mesoamerica Sane & Safe
Fifth nominee of The DeltaCompetition 2006
JOSE E. VILLALOBOS
PATRICIA E. RODRIGUEZ
PHILIP SAKSA
Innovative
Solutions
for the Delta
95
Overview
Frequent flooding from heavy rains is a common occurrence in the region of Mesoamerica (Fig. 1), causing both
death and destruction – often among these nations’ poorest populations. With a flood forecasting system in
place, the safety and security of both people and property in this region can be greatly improved. This project
will create an internet accessible, user-friendly hydrologic system that performs real-time flood forecasting
for the region of Mesoamerica. Our system will be able to employ recent and forecasted precipitation data,
which are then entered into hydrologic models used to predict areas of potential flooding. Due to the poor
availability of observed data in the region, the precipitation data will be derived from satellite images, and
combined with additional Geographic Information System (GIS) data to determine watershed characteristics
such as land use distribution. Much of the data needed for this project is already publicly available from
several independent databases. The next step, which we are proposing, is to combine all of this information
into one central hydrologic system with the ability to effectively retrieve, analyze, and output the model results
in both numeric data and geographic formats. The Hydrologic Flood Forecasting System (HFFS) will also be
available via the internet by using a web mapping application to develop an interactive system accessible to
scientists, educational institutions, and government officials, as well as to the general public.
Fig. 1 – A map of Mesoamerica and the Caribbean obtained from the
SERVIR website – one of the data sources for the proposed Hydrologic
Flood Forecasting System
1. Introduction
Innovative
Solutions
for the Delta
96
Mesoamerica is a region that not only links North and South America geographically, but also ecologically and
culturally. The Mesoamerican territory covers nearly one million square kilometers, and has approximately 60
million inhabitants. The region includes the southern part of Mexico and the seven countries of Central America
– Belize, Guatemala, El Salvador, Honduras, Costa Rica, Nicaragua, and Panama. Dominated by mountainous terrain,
Mesoamerica has more than 100 large volcanoes, some more than 4,000m high. Along the western front, the land
surface slopes down rather abruptly from the mountain crests to a narrow coastal plain along the Pacific Ocean.
On the opposite, eastern coast, the land descends more gradually from the mountains to a broad plain along the
Caribbean Sea.
Many small streams drain from the steep western slopes into the Pacific Ocean while the longest rivers in
Mesoamerica flow through the broad eastern plains to the Gulf of Mexico and Caribbean Sea. Temperatures
in Mesoamerica, which is situated between the Tropic of Cancer and the Equator, vary principally according to
altitude rather than latitude. The Caribbean Coast and eastern mountain slopes generally receive twice as much
annual precipitation as the western mountain slopes along the Pacific coast. Precipitation in the entire region is
highest along the Mosquito Coast of eastern Nicaragua, where the town of San Juan del Norte (Fig. 1) receives
about 6,350mm of rain per year. Soconusco, an area spanning the Pacific coast border of Mexico and Guatemala
(Fig. 2), receives the most rain in the western region at 4000mm annually.
Although Mesoamerica is rich in terms of culture and biodiversity, the region has poor social and economic
conditions that are exacerbated by a vulnerability to natural disasters. Flooding is a major problem along the
coasts, particularly for the people who living in low lying areas. Just within the past decade, there have been
numerous events where regional flooding has resulted in extensive death and damage to property such as houses,
buildings, plantations, and livestock. One of these events occurred in October 1998, when Hurricane Mitch savaged
Mesoamerica killing at least 11,000 people, leaving thousands more missing, and displacing more than two million
others. Nicaragua and Honduras absorbed the brunt of the damage, but El Salvador, Guatemala, Belize, and other
countries in the region were also heavily impacted. Some observers have called Mitch the worst natural disaster
ever to strike Mesoamerica.
Furthermore, in September and October 1999, a tropical storm system sparked by Hurricane Floyd intensified
the already heavy rains, typical at that time of the year in Mesoamerica, and the waterlogged region was drenched
continuously for many days. There were widespread casualties and damage in southern Mexico, Guatemala,
El Salvador, Honduras, Nicaragua, and Costa Rica. In some of these countries, several regions were inundated
with overflowing rivers and dams that destroyed highways and bridges and swept away entire communities in
floodwaters and mudslides. In December 2002 and October/November 2004, the rainy season was more severe
than usual in Costa Rica and Panama, while heavy storms lashed the Caribbean coastal lowlands. The rains caused
floods and landslides, displacing many families living in vulnerable and high-risk areas. Most recently, Hurricanes
Stan and Beta caused similarly devastating effects throughout the region in October 2005, causing many deaths
and displacing hundreds of thousands of people.
Due to the frequency and extent of flooding described above, a system to predict the location of flooding using
real-time forecasting is very desirable for the region.
Such a system would require a real-time prediction of storm runoff. This can be achieved with the use of recent and
forecasted precipitation maps along with river basin characteristics, to process a combination of meteorological
and hydrologic models capable of identifying areas susceptible to flooding. The key issue in real-time flood
prediction is the availability of accurate precipitation information. There are few operational rain gauges, so
dependable rainfall estimation is a major challenge. However, satellite remote sensing measurements providing
a detailed display and information of precipitation currently exist, reliably estimating precipitation in the
Mesoamerican region. In addition, the system would use watershed characteristics, such as land use and elevation,
that can be obtained from remote sensing methods and managed in a Geographic Information System (GIS). With
the satellite derived precipitation and watershed characteristics determined, hydrologic models will then be able
to process the information to predict areas of flooding. The system
could also be used to process hypothetical storm events to identify
the most vulnerable areas of flooding in order to complement
land use planning with flood protection and prevention measures.
Finally, a web mapping application of the GIS will be used to make
the system accessible to users through the internet.
The aim of this project is to develop an efficient Hydrologic
Flood Forecasting System (HFFS) for the Mesoamerican region.
The system would disseminate the information using an internet
mapping portal that is easy to operate, such that any user would be
capable of running the system with minimal requirements.
Fig. 2 – Map showing flooding within the
past decade along the Pacific and Northern
coasts of Mexico and Guatemala
Innovative
Solutions
for the Delta
97
2. Objectives
This project proposes to incorporate precipitation information derived from satellite observations with watershed
characteristics, also retrieved from satellite observations and GIS databases, into hydrologic models to perform
real-time flood forecasting in Mesoamerica. The system will be accessible via the internet through a web mapping
application. Moreover, the system could also be utilized for the implementation of integrated water resources
management for this region.
The primary objective of this project is to develop an innovative, modular, user-friendly and internet-accessible
watershed modeling system, which will combine hydrological models with new technologies in data collection and
handling, such as GIS, remote sensing, and web mapping. The specific objectives of this project are to:
1. Use remote sensing and GIS methods to obtain precipitation and watershed characteristics for the
Mesoamerica region.
2. Identify the most commonly employed public and proprietary hydrologic models (e.g. HSPF, MIKE SHE, HECHMS, TOPMODEL, SAC-SMA) to be used in the flood forecasting system.
3. Develop program interfaces to input the data to the selected models.
4. Use a web mapping interface to make the system available to users through the internet.
5. Test the proposed hydrologic system over at least three different coastal river basins in Mesoamerica for
hydrological operational predictions and flash flood forecasting. This stage will naturally include the calibration
of the model parameters and validation of the model results.
3. Data processes and models
Here we will present our sources of data, demonstrate how that data is processed, and discuss the modeling
techniques that will be used to predict flooding. This section describes the core components and structure of the
Hydrologic Flood Forecasting System.
3.1 Remote Sensing and GIS Data
Clearly, in the last decade, the great increase in availability of geospatial technologies has been proven to aid
human efforts to prevent natural hazards. According to Waser (2002), the power of geospatial technologies comes
from their ability to enable the acquisition of massive quantities of data, linked by geo-referencing to specific
physical locations on the planet’s surface, with the ability to retrieve, analyze, and distribute this data in a variety of
combinations and permutations that can be tailored to meet a diverse array of end-user needs.
Bennett (1997) affirms that in fully integrated systems GIS users should have access to simulation models
through software ‘hooks’ and/or built in macro-languages. This integration strategy can also provide access
to a consistent user interface and data structure, but software available at that time did not support model
development, or user interaction, during simulated events. He continues with the assertion that such integrated
system software must support the construction, execution, and manipulation of geographical simulation models
in a seamless, user-friendly environment. Finally, he emphasizes that users should be able to visualize ongoing
simulations and suspend the simulation process to query intermediate results, investigate spatial/temporal
relations, and even modify the underlying models used to simulate geographical processes. One can conclude that
current remote sensing and geographic information system technologies provide sources for rapid collection of
field data and prompt data processing. These sources of data are proposed to be employed in this project to obtain
the input rainfall and watershed characteristics necessary for hydrologic modeling.
Innovative
Solutions
for the Delta
98
To make some of the powerful geospatial technologies for Mesoamerica accessible, an internet based Regional
Monitoring and Visualization System (SERVIR initials in Spanish) was recently launched. This system is funded
mainly by NASA and the United States Agency for International Development (USAID). Utilizing satellite imagery
and other data sources, this system is designed to aid in environmental management and disaster prevention.
SERVIR (Fig. 3) provides data at no cost to scientists, educators, and policymakers to monitor and forecast
environmental changes and disasters response. This project proposes to obtain GIS data and satellite images from
SERVIR to determine watershed characteristics (elevation, land use, stream network) for the HFFS.
Fig. 3 – SERVIR Website and data portal
Innovative
Solutions
for the Delta
99
3.2 Meteorological Data and Modeling
Hong and others (2004) assert that rainfall measurements derived from meteorological satellites have become
an attractive option because of their high spatial and temporal sampling frequencies. In this context, it is worth
noting that the Global Precipitation Measurement (GPM) Mission, which is a partnership between the United
States National Aeronautics and Space Administration (NASA), the Japanese Aerospace Exploration Agency (JAXA)
and other international agencies, is developing a constellation of satellites exclusively for measuring precipitation.
This system is being designed to provide data with more frequency, and greater accuracy than any system or
satellite currently in operation (Berger, 2005).
Hou (2006) states that the GPM Mission uses advanced precipitation radar, with a constellation of passive
microwave radiometers, to improve the accuracy, sampling, and coverage of global precipitation measurements.
It is a scientific mission with integrated application goals focusing on (1) advancing the knowledge of the global
water/energy cycle variability and freshwater availability and (2) improving weather, climate, and hydrological
prediction capabilities through more frequent measurements of global precipitation with increased accuracy. He
also anticipates that the GPM Core satellite, which carries a JAXA-provided dual-frequency precipitation radar and
NASA-provided microwave radiometers with high-frequency capabilities for light rain and frozen precipitation
measurements, is expected to be launched in the 2010 timeframe. The GPM Core will serve as a precipitation
physics laboratory and a calibration system for improved precipitation measurements by creating a heterogeneous
constellation of dedicated and operational microwave radiometers. NASA also plans to provide an additional “wild
card” constellation member with a copy of the radiometer carried on the GPM Core to be placed in an orbit to
maximize the coverage and sampling of the constellation.
Fig. 4 – The Hydrologic Data and Information System (HyDIS) Interactive Website displaying a six hour precipitation
map encompassing Mesoamerica
Innovative
Solutions
for the Delta
100
Anticipating the operational status of the GPM Mission, the proposed HFFS intends to employ the precipitation
measurement algorithm, called Precipitation Estimation from Remotely Sensed Information using Artificial Neural
Networks (PERSIANN). This is a satellite-based algorithm used to estimate rainfall with a nearly global coverage.
The PERSIANN system uses neural network function classification and approximation procedures to estimate
the rainfall rate from the infrared brightness temperature image provided by geostationary satellites at a scale
of 0.25° x 0.25°. Sorooshian and others (2005) explain and discuss this algorithm in detail, and they point out
that its precipitation predictions have been successfully used over the years in a number of hydrologic research
and application studies. This data is available to the public through the Hydrologic Data and Information System
(HyDIS). HyDIS (Fig. 4) is a river basin and country-based internet GIS system that includes a global precipitationmapping server (Sorooshian et al 2005). Until the GPM mission is in operation, the HFFS will use hourly satellite
precipitation measurements from the HyDIS database.
Meteorological models are used to predict weather patterns and movement. They use universal laws of
atmospheric physics and empirical relationships to estimate or simulate precipitation, wind flow, temperature,
humidity, and vertical air mixing in time and space throughout the modeled area. In these models, the main links
between the land surface and the atmosphere are formulated on the basis of the energy balance, heat balance,
and moisture balance. In the early 1970’s, researchers started developing an atmospheric prediction model derived
from remote sensing. At that time, Richard Anthes started developing a predictive, hydrostatic meteorological
model (Anthes and Warner, 1978), which has since evolved into the Fifth-Generation Pennsylvania State University/
National Center for Atmospheric Research (PSU/NCAR) Mesoscale Model (MM5). This version of the model has
been operational since 1999.
A reliable, real time quantitative precipitation forecast has long been recognized as a prerequisite to hydrologic
forecasting for flash-floods (Georgakakos, 2002a). The MM5 is a limited-area, non-hydrostatic, terrain-following
sigma-p coordinate model designed to simulate and predict mesoscale atmospheric circulation. The model is
supported by several pre- and post-processing programs, which are referred to collectively as the MM5 modeling
system. The MM5 modeling system software is mostly written in Fortran, and has been developed at Penn State
and NCAR as a community mesoscale model, with additional contributions from users worldwide. The MM5
modeling system software is freely provided and supported by the Mesoscale Prediction Group in the Mesoscale
and Microscale Meteorology Division, NCAR (http://www.mmm.ucar.edu/mm5/).
Innovative
Solutions
for the Delta
101
3.3 Hydrologic modeling
Flooding is a critical subject of study in hydrology, and such events can be predicted with the use of hydrological
models. According to Singh & Frevert (2006), watershed models simulate natural processes such as the flow of
water, sediment, chemicals, nutrients, and microbial organisms within watersheds, as well as quantify the impact
of human activities on these processes. Generally, watershed models are used to predict the output (streamflow)
of a system (watershed or river basin) in response to some amount of input (precipitation). The temporal and
spatial variations of both system and input have recently been the driving factors in the development of more
physically based, spatially explicit approaches to hydrologic modeling.
Watershed models can be classified based on: process description, timescale, techniques of solution, land use,
and model use. In 1996, The American Society of Civil Engineers also separated flood analysis models into four
categories: Event-based precipitation-runoff models, Continuous precipitation-runoff-models, Reservoir regulation
models, and Flood frequency analysis models (Singh and Frevert, 2002). In this project, we plan on considering both
lumped and distributed models for determining flood forecasting. This is a classification based on the description
of the hydrologic process that contributes to the system output in conjunction with the system characteristics
(watershed), also established by Singh (1995).
A lumped parameter model does not clearly account for spatial relationships between model parameters
and inputs or outputs. Distributed parameter models explicitly account for spatial relationships among model
variables and parameters. The number of variables and parameters necessary to run a distributed model are far
greater than for a lumped model of the same basin. This data demanding process generates difficulties in the
parameterization, calibration and validation of the distributed model. Since comprehensive spatial data for the
entire Mesoamerica region is not currently available, this project proposes to employ both lumped and distributed
hydrologic models for flood forecasting. The intended approach is a top-down system (Klemes, 1983) that starts
with the simplest conceptual model configuration that can gradually build process complexity.
In the end, the type of model to be used is often dictated by the availability of data. When the necessary
data either does not exist or are not fully available, regionalization and synthetic techniques are useful (Singh
and Frevert, 2002). The simplest regionalization approach is to fix watershed model parameters to average
values for a region (Vogel 2006). Shmagin and Kanivetsky (2006) have proposed a new integrative approach
in regional hydrology following the conceptual systemic or system analyses. This approach uses a multilevel
system for landscapes in which the watershed is a subsystem, taking the place of regional, dual scale (regional
and basin), hydrological analysis for temporal and spatial variability of river stream runoff. They state that the
conceptual systemic model of a watershed as a multidimensional area unit, reflecting quantitative and qualitative
characteristics of all landscape properties, was developed based on the cybernetic model of a geographic sphere
reported by Krcho (1978).
Dominguez and others (1996) presented an interesting regional analysis for the Grijalva River Basin in Chiapas,
Mexico (part of Mesoamerica). In that piece, they used available information about maximum rainfall and
runoff events, as well as a transformation function, to homogenize the sample and achieve regional equations
to predict peak flows using the size of the river basin area. According to Vogel (2006), the transfer of hydrologic
characteristics of watersheds from data-rich to data-poor environments is one of the most fundamental
challenges in the field of hydrology. Since the Mesoamerican region has poor data availability, this project proposes
to employ a top-down approach (Klemes, 1983) that starts with a similar regional analysis configuration and
gradually, as more data becomes available, descends to an individual basin scale.
Innovative
Solutions
for the Delta
102
Most processes that occur in nature are not completely understood, and therefore mathematical depictions
of these processes contain different levels of uncertainty. Stochastic models explicitly account for uncertainty
in model parameters. Deterministic models, on the other hand, characterize processes with specific values.
Uncertainty is not considered in these processes, thus the same set of input values will always give the same set of
output values. Since there exists scattered rainfall and runoff information in the Mesoamerican region, this project
intends to employ stochastic models as complements of deterministic models in order to make flood forecasting
a downward approach (Klemes, 1983). This process starts with a minimal data demanding model, and gradually,
as more information is obtained, proceeds to increasingly data demanding models in search of the balance cited
before.
3.4 Coupling meteorological and hydrological models
In the meteorological sciences, the coupling of atmospheric and hydrologic models is seen as a tool to validate
precipitation in atmospheric models through the integration of basin effects as an alternative to point-bypoint verification of a highly variable field, with the aim of improving atmospheric simulation through a better
representation of the surface water budget. In this context, and in regards to flood prediction, Tomassetti and
others (2005) affirm that a strategic goal of applied meteorology is to try to predict with high spatial resolution
the segments of a drainage network where floods may occur. One possible way to reach this goal is to couple the
meteorological mesoscale model with high resolution hydrological models.
Bae and others (1995) were the first to report that coupled meteorological-hydrological models for
simultaneous rainfall and flow prediction were in operational use at that time. Georgakakos (2002b) also
confirmed that the formulation of the rainfall prediction component in integrated hydrometeorological models
was expanded to incorporate the use of data from weather radars, generate spatially distributed rainfall
predictions over large mesoscale domains with high resolution, use forecast fields from large-scale numerical
weather prediction models, and to include mountain terrain effects on rainfall. He proposed a catchmentaggregate process-based model (rainfall prediction, soil moisture, and channel flow), complemented by a state
estimator to update the model and make it suitable for real time flash flood prediction. The state estimator also
creates the capability of probabilistic forecasting that quantifies uncertainties due to errors from observation
sensors, model structure, parameters, or inputs (Georgakakos, 1987).
From the perspective of flood forecasting in real time, the HFFS (Hydrologic Flood Forecasting System) will
require the use of interactive meteorological and hydrological models. These models should be coupled in such a
manner that permits a prediction of the time and space distribution of both rainfall and the resultant flooding.
To do so, these tools must take advantage of existing and new developments in data acquisition, processing and
management.
3.5 Flood modeling
Flood models, such as WetSpa or DBSIM, comprise a combination of hydrologic and hydraulic models. The
hydrologic model determines the runoff that occurs following an individual precipitation event. The primary
outputs from the hydrologic model are hydrographs at varying locations along the stream network describing the
quantity, rate and timing of streamflow resulting from precipitation events. These hydrographs then become a
key input into the hydraulic model. The hydraulic model simulates the movement of flood waters through stream
reaches, storage elements, and hydraulic structures. The hydraulic model calculates flood levels and flow patterns
while modeling the complex effects of backwater, overtopping of embankments, waterway confluences, bridge
constrictions and other hydraulic structure behaviors.
Innovative
Solutions
for the Delta
103
Creutin and Borga (2003) indicate that flash floods develop at space and time scales that conventional observation
systems of rain and discharge in rivers are unable to monitor. They proceed to describe the three main problems
that hamper the progress of flash-flood research as: 1) downscaling due to the incoherent space and time scales
between atmospheric models and the flash-flood triggering processes, 2) small basins prone to flash-floods are
seldom gauged and must be modeled without calibration, and 3) knowing the soil retention limits of runoff
under the range of accumulated precipitation considered. Finally they considered the development of integrated
hydrometeorological approaches for flash-flood real-time hazard assessment as one of two key developments in
the future of flash flood research. This project aims to integrate satellite rainfall estimation and regional hydrology
parameters into the hydrologic/hydraulic (flood) models to confront the flash flood monitoring problem.
3.6 Web Mapping
Loucks and others, cited in Garrote and Bras (1995), state that a real-time flood forecasting system must combine
a data acquisition system and state of the art hydrologic modeling to provide the decision maker with the best
information possible. Since the decision making process must be carried out in real-time, the availability of an
adequate software environment is very important in facilitating the task of model users, especially if the operation
involves the use of complex, data-intensive distributed models.
This project proposes to use a web mapping application aimed towards decision-makers with some GIS and/or
hydrological training and experience. Scientists, educational institutions, and government officials will be the
target audience of the HFFS, although it will also be available to the public at large. This assumes that the users
will range anywhere from beginners in computerized mapping, to having some GIS exposure, to experts in the
field. This range in user experience will require several levels of complexity in the HFFS in order to face one of the
ubiquitous deficiencies pointed out by Singh and Frevert (2006), which is the lack of user-friendliness of watershed
models.
4. Calibration of models
Calibration is a key component in producing accurate, reliable model results. However, performing this process in
ungauged or data poor areas, such as in much of Mesoamerica, is extremely difficult. This section explains and
discusses several statistical techniques that can be used for calibration in areas with insufficient information. It is
important to note that these methods are still being developed and need to be explored more fully to determine
the procedure that produces the most accurate results and predictions for this system. There is no definitive
method for determining this procedure, and it will take many trials and user/model feedback to develop a working
tool for this type of evaluation. Therefore, this section of the HFFS will be created after the initial setup of the
system to improve model outputs in watersheds with few observational resources.
4.1 Regional Parameter Estimation
As we have previously stated, the region of Mesoamerica is characterized by poor hydrometeorological data
availability. However, since the main goal of this project is to predict floods in this location, it is necessary to
research techniques for determining model parameters that can be employed with limited amounts of data. In this
context, Kundzewicz (2002) argues that if there is no data observed in a catchment, one has to use methods which
do not require the availability of a lengthy time series of hydrological records. Even if there are only a few gauged
sites among many similar and adjacent catchments, one can try to establish regionally valid laws. Models can then
be developed for gauged catchments and used to link their parameters to physical characteristics (e.g., by not-very-
Innovative
Solutions
for the Delta
104
illuminating linear regression). Once this is completed, the regional approach can be applied to ungauged basins,
and the necessary physical characteristics can be determined.
Additionally, Vogel (2006) states that as watershed models become increasingly sophisticated and useful, there
is a need to extend their applicability to locations where they cannot calibrated or validated. He also emphasizes
that without streamflow data a watershed model cannot be calibrated or validated, hence regional methods
are needed which relate easily measured watershed characteristics to watershed model parameters. Inside his
literature review about methods for the regional calibration of watershed models, he highlights several promising
approaches:
1. Methods for grouping catchments on the basis of their hydrologic homogeneity,
2. The hybrid method, where first cluster analysis and principal components analyses are employed to break the
region into homogeneous sub-regions, in which similar drainage basins that are close to ungauged basins are
used to develop regional flow duration curve models, and subsequently used to calibrate watershed models at
ungauged locations, and
3. The regional calibration methodology, which attempts to calibrate the model at all sites in a region
simultaneously, while concurrently attempting to achieve the best possible regional relationships among
watershed model parameters and watershed characteristics.
Finally, he underlines a recommendation that a hybrid regionalization method should combine recent advances
in regional hydrologic statistics and the determination of hydrologically homogeneous regions with the regional
calibration methodology. Included is a need to enable proper accounting of the impacts of both serial and spatial
covariance structures of watershed models residuals.
4.2 Determination of Watershed Similarity
Statistical multivariate techniques provide a rational framework for analyzing the similarity of units, which
vary with respect to numerous characteristics. Regarding the determination of hydrologically homogeneous
regions, this project proposes to use the properties of Mesoamerica’s catchments (area, mean altitude, average
precipitation, drainage density, land use) through both ordination and classification (classifying units into discrete
subsets) techniques described by Omi and others (1979) as follows:
1. The dimensionality of the problem is reduced using factor analysis. This ordination technique reduces the
original number of variables to a smaller number of “factors” which are linear combinations of the original
data set. The factor loadings, which are correlations between the original variables and the new factors, are
then used to calculate the “scores” for each unit along each factor. These factor scores are used in place of the
original variables in the subsequent analysis.
2. Initially, groups of similar units are identified by a cluster analysis of the factor scores. This classification process
involves the determination of the units that are most alike, based on systematic comparisons of the factor
scores for all units. The cluster analysis could be performed on the raw data, thus eliminating the need for the
factor analysis. However, the factor analysis clarifies relationships between measured characteristics, mitigates
the effect of redundant or correlated variables, and helps infer the structure and importance of phenomena
that underlie the raw data.
3. The initial groups are tested, and units are reclassified using discriminant function analysis. Linear
combinations of the factor scores, which maximize the distinction between groups, enable a unit to be placed
into the group to which it has the highest probability of membership.
Innovative
Solutions
for the Delta
105
To investigate the possibility of identifying homogeneous precipitation areas, this project also plans to use the
simple model (S-model) of the Principal Components Analysis (PCA) in order to have a better understanding of the
complex atmospheric phenomena resulting in the seasonal rainfall distribution over Mesoamerica - as was used
in Austria by Ehrendorfer (1987). According to this author, by using PCA – which can be considered a simple datadescriptive method – it is possible to describe a given data set completely by means of new variables (principal
components or PCs), which have two fundamental properties: (a) two different uncorrelated components and (b)
each component is derived from an empirical orthogonal variable, accounting for a maximum in residual total
variance of the original data set. As stated by Johnson and Wichern (2002), a PCA is concerned with explaining the
variance-covariance structure of a set of variables through a few linear combinations of these variables, with the
general objectives of data reduction and interpretation. They also affirm that an analysis of principal components
often reveals relationships that were not previously suspected, and thereby allows interpretations that would not
have been shown otherwise.
One method that accounts for both the spatial covariance among flow, climate and watershed characteristics,
as well as the temporal covariance associated with flow, climate and model residuals, is the generalized least
squares method (Vogel, 2006). As indicated by Hofrichter and others (2006), annual maxima of the discharge
of a river are commonly used to predict flood waters. Annual maxima from several rivers can then be modeled
to predict flood levels for rivers where little or no discharge data is available. In order to explore the capability of
creating a state estimator for the Mesoamerica region used for updating the models that will be utilized in the
HFFS, this project plans to use an approach similar to the one successfully applied in Austria by Hofrichter and
others (2006).
Due to the non-normal distribution of the annual maxima, Hofrichter and others (2006) did not consider
linear regression models. Instead, they ran the modeling within the context of quasi-likelihood estimation in the
generalized linear model (GLM) framework. Additionally, to take the correlation among observations of different
rivers at one year periods into account, the model was augmented with a random effect, which lead to the broad
class of generalized linear mixed models (GLMMs). The first analysis of the annual maxima indicates that it was
reasonable to assume temporal independence over the years. Therefore, only spatial dependency of the data
was considered. The fitted values were plugged into the method of moment estimator of the parameters in the
extreme value distribution (Gumbel Distribution) to obtain estimates of certain quantiles. These quantiles are
commonly used to predict flood levels of rivers. The authors concluded that this approach provides satisfactory
estimates of flood levels, and can be applied for rivers where the only data available are the catchment properties.
Alternatively, for the establishment of the state estimator inside of the HSSF, the method used in Australia
by Post (2004) will also be considered. Using this method, flow duration curves (FDC) are represented using a
logarithmic transformation. The FDCs have been defined using two parameters: (1) the ‘cease to flow’ point, and (2)
the FDC slope. These two parameters defining the FDC have been related to the area, mean annual precipitation,
drainage density and total stream length of the catchments under consideration. Finally, a regionalization
procedure has been developed whereby the FDC for an ungauged catchment can be predicted based upon the
attributes of that catchment. The author concludes that this simple model is sufficient to adequately define the
flow duration curve in a variety of catchments. It is therefore compatible with the ‘top-down’ approach, where
additional levels of complexity are added to a model only when the necessary data becomes available. Lastly, he
states that the mirror-image nature of the FDC may imply that it is possible to predict high flows upon analysis of
low flows in the same catchment.
Innovative
Solutions
for the Delta
106
5. Proposed methodology
This project aims to build a modular hydrologic flood forecasting system (HFFS) for the Mesoamerican region. The
modular design of the HFFS will enable the final user to select modules in custom combinations. The application
of the HFFS is based upon satellite-based algorithm rainfall data to create flood inundation maps, which will then
be internet accessible. The system will be geospatially integrated over Mesoamerica and contain: 1) an interface
with the SERVIR and HyDIS databases, 2) a central GIS database to extract watershed characteristics (watershed
divisions, shapes, elevation, slopes, soils, land use, drainage networks, and surface runoff directions), 3) the public
simulation models available for flood analysis or the preparation of data to be input into proprietary models, and
4) the proper interfaces to relay the input data to the models, and the resulting outputs to the final users and
decision-makers through both data tables and maps (Fig. 5).
To accomplish the HFFS, it is proposed to perform the following tasks:
1. Construct Module-I (Interface-I) to retrieve watershed characteristics and parameters from satellite images
and the geo-database of Mesoamerica obtained from SERVIR. These data could be input manually or via GIS
software. This interface will include menus to load a general map of the basin, map layers containing the
watershed characteristics, and a drainage network layer defining the flow direction. Additionally, the interface
will permit the user to edit the drainage network and direction of overland or channel flow.
2. Construct Module-II (Interface-II), which will receive the rainfall maps from the HyDIS, and transfer these data
into the geo-data base for Mesoamerica. Eventually, this module will be used to receive precipitation data from
the Global Precipitation Measurement (GPM) Mission.
3. Selection of the suitable lumped and distributed models.
4. Construct Module-III (Interface-III), which will convert the geo-referenced rainfall and watershed data to the
formats required by the hydrologic models.
5. Construct Module-IV (Interface-IV) that will call the execution of a publicly available simulation model or
output the data in a model-ready format.
6. Construct Module-V (Interface-V) which will be the graphical user interface that will convert the simulation
model outputs into a web mapping application to allow the user to edit and visualize the results through the
internet.
7. Develop/obtain tools to calibrate and validate the hydrologic models.
Innovative
Solutions
for the Delta
107
6. Requirements
1.
2.
3.
4.
A server capable of hosting the HFFS
Access to the SERVIR database
Access to the HyDIS server
Acquisition of remote sensing data, computers, and software (GIS, Hydrologic Models, Web Mapping, and
software development)
7. Expected results
This project would provide a final product consisting of:
1. A modular Hydrologic Flood Forecasting System (HFFS) for the Mesoamerica region
2. Insights into the use of remote sensing rainfall estimates for hydrological forecasting
3. A tool which will allow the exploration of the competitiveness of different hydrologic models for flood
forecasting in the region of Mesoamerica
8. Conclusions
Mesoamerica has been shown to be an area susceptible to heavy rains and coastal flooding that has resulted
in many deaths and much destruction over the past decade. To help mitigate the losses resulting from these
floods, we propose a real-time Hydrologic Flood Forecasting System. This system will use publicly available
satellite remote sensing and geographic information to obtain the data necessary to process hydrologic
models with the ability to predict flooding in coastal areas.
The development of the HFFS for Mesoamerica would prove to be a great asset in the planning
and prevention of the frequent flood related disasters that inundate this region. It would not only be
a technological achievement, as one of the first modeling applications to incorporate satellite rainfall
estimations, but it also has the potential to positively affect the lives of the many people that live along
coastal areas. The results of this system could then be used to implement a similar setup in the many other
areas of the world that have limited observed data.
Innovative
Solutions
for the Delta
108
9. References
Anthes, R.A., and T.T. Warner. 1978. Development of Hydrodynamic Models Suitable for Air Pollution and other
Mesometeorological Studies. Monthly Weather Review, vol. 106, 1045-1078.
Bae, D.H., K.P. Georgakakos, and S. K. Nanda. 1995. Operational Forecasting with Real-Time Databases. ASCE Journal
of the Hydraulics Division, vol. 121(1), 49-60.
Bennet, D.A. 1997. A framework for the integration of geographical information systems and modelbase
management. International Journal of Geographical Information Science, vol. 11(4), 337-357.
Berger, B. 2005. NASA May Accelerate Global Precipitation Measurement Mission. Space News, Business Report. 07
Jul 2005.
Creutin, J.D. and M. Borga. 2003. Radar hydrology modifies the monitoring of flash-flood hazard. Hydrological
Processes, vol. 17, 1453-1456.
Domínguez, R. J.E. Villalobos, and D. Guichard. 1996. Contribución al análisis regional de lluvias y escurrimientos
máximos en la cuenca del río Grijalva (Contribution to the regional analysis of rainfall and peak flows in the
Grijalva River Basin) Proceedings of XVII Congreso Latinoamericano de Hidráulica. Ecuador.
Ehrendorfer, M. 1987. A Regionalization of Austria’s Precipitation Climate Using Principal Component Analysis.
Journal of Climatology, vol. 7, 71-89.
Garrote, L. and R. L. Bras. 1995. An Integrated Software environment for real-time use of a distributed hydrologic
model. Journal of Hydrology, vol. 167, 307-326.
Georgakakos, K. P. 1987. Real-Time Flash Flood Prediction. Journal of Geophysical Research, vol. 92 (D8), 9615-9629.
Georgakakos, K. P. 2002a. Hydrologic Short-Term Forecasting with QPF Input. White Paper for USWRP Warm Season
Precipitation Workshop.
Georgakakos, K. P. 2002b. Chap. 18: Hydrometereological Models for Real time Rainfall and Flow Forecasting. In
Mathematical Models of Small Watershed Hydrology and Applications. Singh, V.P. and D.K. Frevert (Eds.). Water
Resources Publications, USA.
Hofrichter, J., T. Harum, and H. Fried. 2006. Statistical Modelling of Annual Maxima in Hydrology. Austrian Journal Of
Statistics, vol. 35(1), 21–30.
Hong, Y., K.L. Hsu, S. Sorooshian, and X.G. Gao. 2004. Precipitation Estimation from Remotely Sensed Imagery using
an Artificial Neural Network Cloud Classification System. Journal of Applied Meteorology. 43 (12), 1834-1852.
Hou, A. 2006. The Global Precipitation Measurement (GPM) Mission: An overview. Geophysical Research Abstracts,
vol. 8.
Johnson, R.A. and D.W. Wichern. 2002. Applied Multivariate Statistical Analysis. 5th ed. Prentice Hall, USA.
Klemes, V. 1983. Conceptualization and Scale in Hydrology. Journal of Hydrology, vol. 65, 1-23.
Krcho, J. 1978. The Spatial Organization of the Physical-Geographic Sphere as a Cybernetic System Expressed by
means of Measure of Entropy. Acta Facultatis Rerum Naturaium Universitatis Commenianae, Ceographica 16, 57-147.
Innovative
Solutions
for the Delta
109
Kundzewicz, Z.W. 2002. Presentation at the kick-off meeting on “Prediction in ungauged basins” held in Brazil, 20-22
November 2002.
Omi, P. N., L.C. Wensel, and J. L. Murphy. 1979. An Application of Multivariate Statistics to Land-Use Planning:
Classifying Land Units into Homogeneous Zones. Forest Science, vol. 25(3) 399-414.
Post, D. 2004. A New Method for Estimating Flow Duration Curves: an Application to the Burdekin River Catchment,
North Queensland, Australia. In Pahl-Wostl, C., S. Schmidt, A.E. Rizzoli, and A.J. Jakeman (eds). Complexity and
Integrated Resources Management, Transactions of the 2nd Biennial Meeting of the International Environmental
Modelling and Software Society, iEMSs: Manno, Switzerland, 2004. vol. 3, pp 1195-1200.
Shmagin B., and R. Kanivetsky. 2006. Regional Hydrology: Tools vs. Ideas. Published in Coastal Hydrology and
Processes (ed. By V. P. Singh & Y. J. Xu), 183-196. Water Resources Publications, LLC, USA.
Singh, V.P. (Ed). 1995. Computer Models of Watershed Hydrology. Water Resources Publications, USA.
Singh, V.P. and D.K. Frevert.. 2002. Chapter 1. Mathematical Modeling of Watershed Hydrology. In Mathematical
Models of Large Watershed Hydrology. Singh, V.P. and D.K. Frevert (eds). Water Resources Publications, USA.
Singh, V. P. and D.K. Frevert. 2006. Chapter 1: Introduction. In Watershed Models, Singh, V.P. and D.K. Frevert (eds).
Taylor & Francis Group, USA.
Sorooshian, S., K. Hsu, B. Imam, and Y. Hong. 2005. Global Precipitation Estimation from Satellite Image Using
Artificial Neural Networks. Journal of Applied Meteorology, Vol. 36, 1176-1190.
Tomassetti, B., E. Coppola, M. Verdecchia, and G. Visconti. 2005. Coupling a distributed grid based hydrological model
and MM5 meteorological model for flooding alert mapping. Advances in Geosciences, vol. 2, pp. 59–63.
Vogel, R.M. 2006. Chapter 3: Regional Calibration of Watershed models. In Watershed Models, Singh, V. P. and D.K.
Frevert (eds.). Taylor & Francis Group, USA.
Wasser, K. 2002. Geospatial technologies: Tools for understanding complex systems. Editor’s note. Arid Lands
Newsletter, No. 51.
Innovative
Solutions
for the Delta
110
Innovative
Solutions
for the Delta
111
Royal Haskoning
DeltaCompetition
Postbus 151
6500 AD Nijmegen
The Netherlands
telephone +31 (0) 24 328 4725
fax +31 (0) 24 360 9566
e-mail: [email protected]
We are always interested in any comments or questions
you may have about anything you have read regarding the
DeltaCompetition.
Editing:
Royal Haskoning
PR & Communications
Layout:
Charles Whalley Advertising Limited
United Kingdom
Printing:
Verenigde Offset-bedrijven bv
The Netherlands
You can find more information about the DeltaCompetition
on our website:
www.deltacompetition.nl
www.royalhaskoning.com
Innovative
Solutions
for the Delta
112
Innovative
Solutions
for the Delta
113
Innovative
Solutions
for the Delta
114
All rights reserved. No part of this book may
be copied, reproduced or transmitted in any
form, by any electrical or mechanical means
without prior permission in writing from the
publisher, except by a reviewer who may quote
brief passages.
© Royal Haskoning 2006
Published by Royal Haskoning,
P.O. Box 151, 6500 AD Nijmegen,
Barbarossastraat 35, 6522 DK Nijmegen,
The Netherlands