03-DHI - Urban Climate Change Guidelines

URBAN
CLIMATE CHANGE
GUIDELINES
HOW TO ACHIEVE SUSTAINABLE ADAPTATION

IN URBAN AREAS.
JANUARY 2012
CONTENTS
1 Introduction ______________________________________________________________1
1.1 Background _________________________________________________________________ 1
1.2 Purpose ____________________________________________________________________ 2
1.3 Definition of the problem ______________________________________________________ 3
1.4 An Example – Legislation in Denmark_____________________________________________ 4
1.4.1 Sewerage plan ______________________________________________________________________ 4
1.4.2 Sewer renewal plan __________________________________________________________________ 5
1.4.3 Contingency plan ____________________________________________________________________ 5
1.4.4 Additional municipal planning, local plans, plans for water quality, etc. ________________________ 5
1.4.5 Risk assessment and analysis of extreme rain _____________________________________________ 6
1.5 The EU Flood Directive ________________________________________________________ 6
1.5.1 Preliminary flood risk assessment ______________________________________________________ 7
1.5.2 Flood hazard and flood risk maps _______________________________________________________ 7
1.5.3 Flood risk management plans __________________________________________________________ 8
1.5.4 Publication of maps, plans, etc. ________________________________________________________ 8
1.5.5 Summary __________________________________________________________________________ 8
2 Status of knowledge on climate change _______________________________________10

2.1 Scenarios for future climate change _____________________________________________10
2.2 Data base for the analysis of climate change impacts on drainage systems _____________13
2.3 Future rain over Denmark _____________________________________________________13
2.4 Future water levels in marine waters ____________________________________________15
2.5 Future water levels in lakes and rivers ___________________________________________15
2.6 Future water levels / pressures in groundwater ___________________________________16
2.7 Determination of input data and boundary conditions - Recommendations _____________16
2.7.1 Choice of precipitation ______________________________________________________________ 16
2.7.2 Coupled rainfall ____________________________________________________________________ 17
2.7.3 Choice of downstream boundary condition – Future water levels in recipients of storm water systems
18
2.7.4 Sea level rise and climate-related extreme water levels ____________________________________ 18
2.7.5 Rivers, lakes and ground water zone ___________________________________________________ 18
3 Compliance with the performance requirements under the influence of climate change 20
3.1 Description of desired standard of services _______________________________________20
3.1.1 Service level and compliance with operational requirements under the influence of climate change 23
4 Modelling of flooding from storm water system and rivers ________________________25

4.1 From surface depression maps to sophisticated hydraulic models_____________________25
4.1.1 Model tools _______________________________________________________________________ 25
4.1.2 Decision making process _____________________________________________________________ 26
4.1.3 The multidisciplinary angle ___________________________________________________________ 26
4.1.4 Economy _________________________________________________________________________ 26
4.2 Calculation Methods _________________________________________________________27
4.2.1 Terrain model - QA _________________________________________________________________ 28
4.2.2 Surface depression maps ____________________________________________________________ 29
4.2.3 Hydraulic surface calculations_________________________________________________________ 33
4.2.4 Hydrodynamic drainage model 1D _____________________________________________________ 35
The Climate Cookbook – 2nd Edition i Greve Forsyning, VandCenter Syd,

© DHI 2012 , PH-Consult og DHI
4.2.5 Combined hydrodynamic drainage model and surface model 1D-1D__________________________ 35
4.2.6 Combined hydrodynamic drainage model and surface model 1D-2D__________________________ 40
4.3 Comparison of methods ______________________________________________________42
5 Flooding from the sea ______________________________________________________45
5.1 Calculation Methods _________________________________________________________45
5.1.1 Water elevation map ________________________________________________________________ 46
5.1.2 Water elevation map in connection to the sea ___________________________________________ 48
5.1.3 1D-urban drainage systems __________________________________________________________ 50
5.1.4 1D-1D two-layered hydrodynamic model _______________________________________________ 50
5.1.5 Hydrodynamic calculation of flood _____________________________________________________ 50
5.1.6 1D-2D hydrodynamic modelling of the flow on terrain combined with a hydrodynamic model of the
drainage system __________________________________________________________________________ 53
6 Urban climate adaptation __________________________________________________55

6.1 Risk analysis ________________________________________________________________55
6.1.1 Damage assessment using a risk analysis ________________________________________________ 55
6.1.2 Risk analysis _______________________________________________________________________ 56
6.1.3 Risk of flooding from extreme rainfall __________________________________________________ 58
6.2 Priority adapting to floods under a changing climate _______________________________62
6.2.1 Urban prioritization in Greve municipality _______________________________________________ 63
6.2.2 Description of the method for prioritization in Greve ______________________________________ 63
6.2.3 Quality assurance of priority __________________________________________________________ 64
6.3 Options for adapting the urban drainage system __________________________________65
6.3.1 Physical measures on drainage system _________________________________________________ 65
6.3.2 Infiltration of storm water ___________________________________________________________ 66
6.3.3 Separation of combined systems ______________________________________________________ 66
6.3.4 Increasing pipe sizes ________________________________________________________________ 67
6.3.5 Overflow _________________________________________________________________________ 67
6.3.6 Basins ____________________________________________________________________________ 68
6.3.7 Local storage ______________________________________________________________________ 68
6.3.8 Control and regulation of drainage system ______________________________________________ 68
6.3.9 Use of the road system ______________________________________________________________ 69
6.3.10 Augmentation on private property __________________________________________________ 69
6.3.11 Drainage of road space ____________________________________________________________ 71
6.3.12 Examples of local handling of rainwater ______________________________________________ 71
6.4 Flood emergency preparedness ________________________________________________73
6.4.1 Before the rainfall event _____________________________________________________________ 74
6.4.2 During the rainfall event _____________________________________________________________ 78
6.4.3 After the rainfall event ______________________________________________________________ 78
7 Summary and conclusion ___________________________________________________79

8 References _______________________________________________________________81
APPENDIX
A Odense – Sports Arena

B Climate Analysis of Greve Midt and Karlslunde
The Climate Cookbook – 2nd Edition ii Greve Forsyning, VandCenter Syd,

Disclaimer
The information contained in this Guideline report is for general information purposes only. The information
is provided by DHI and while we endeavor to keep the information up to date and correct, we make no
representations or warranties of any kind, express or implied, about the completeness, accuracy, reliability,
suitability or availability with respect to the Guideline report or the information, products, services, or
related graphics contained in the Guideline report for any purpose. Any reliance you place on such
information is therefore strictly at your own risk. In no event will we be liable for any loss or damage
including without limitation, indirect or consequential loss or damage, or any loss or damage whatsoever
arising from loss of data or profits arising out of, or in connection with, the use of the Guideline report.
Through the Guideline report you are able to link to other websites which are not under the control of DHI.
We have no control over the nature, content and availability of those sites. The inclusion of any links does
not necessarily imply a recommendation or endorse the views expressed within them.
The Climate Cookbook – 2nd Edition iii Greve Forsyning, VandCenter Syd,
1 INTRODUCTION
1.1 Background
This document describes how cities around The World can respond to flooding
problems expected due to impact from climate change. These guidelines provide an
overview of information available about climate change, exemplified by using Denmark
as a case. Focus is on impacts on the urban drainage systems. The most significant
factors are the increase in extreme rainfall and rise in sea level. Examples of how floods
can be prevented and avoided describe available methods for analysing existing urban
drainage networks and evaluating the impact from various flood mitigation
augmentations.
During the recent years several cities have been affected by very long-term extreme
rainfall which showed limitations in the urban drainage in combination with the
remaining water cycle (groundwater and streams). These rainfall events emphasize the
need for municipalities to focus on flood management within urban drainage systems
and the entire water cycle. This edition of the climate cookbook is extended with
descriptions of response to flooding from the sea. The described methods are limited to
storm water systems, streams and seas.
The update contains guidelines for municipalities with respect to:
 Perform analysis and augmentation work of drainage systems in urbanized areas

 Undertake the principles described in the Climate Cookbook based on the latest
figures from the IPCC
 Incorporate updated climate-related maximum sea levels by including principles
for calculating wind induced increase in sea level as a supplement to the IPCC’s
mean sea levels
 Identify areas at risk of flooding due to climate change from rivers and seas
 Perform analysis and regulation of rivers in urbanized areas.
The report is financed by DANVA and the project partners. The report is prepared by
Greve Forsyning A/S, VandCenter Syd, PH-Consult and DHI. The members of the
working group were:
Birgit Paludan, Greve Forsyning

Nanna Høegh Nielsen, PH-Consult
Lina Nybo Jensen, PH-Consult
Annette Brink-Kjær, VandCenter Syd
Jens Jørgen Linde, PH-Consult
Ole Mark, DHI
The project was undertaken during the period November 2008 to June 2010.
The report provides detailed descriptions of methods for analysis of flooding from
simple "surface depression maps" or "blue spots" to the advanced hydraulic computer
The Climate Cookbook – 2nd Edition 1 Greve Forsyning, VandCenter Syd,

models. The methods can be used by both municipalities and wastewater utilities for
their detailed action against floods as well as in implementation of the flood directive.
This version of the guidelines (Climate Cookbook) represent a translated version of the
original Danish version under the title: "En kogebog for analyser af klimaændringers
effekter på oversvømmelser i byer".
1.2 Purpose
Design of urban drainage systems through time has been undertaken using multiple
methods such as hand calculated methods: Constant rain intensity, time-area method,
runoff diagrams, rain images or the rational method. In recent decades, these methods
have been supplemented with computer-based methods to support the dimensioning and
analysis of the urban drainage system. All over The World it is common practice to use
hydraulic models (such as MOUSE or MIKE URBAN) for analysing the function of the
urban drainage system and to calculate the effect of proposed augmentations.
Along with developments in the calculation methods, the requirements for the
dimensions of urban drainage system have evolved. For many years the common
practice was that in combined systems conduits were only allowed to reach full running
capacity every second year, and in storm water systems only once a year. National
rainfall series or some older gauged rainfall data have been used. Most of the Danish
urban drainage network is probably dimensioned from the above requirement. With the
release of Guide 27 (SVK Skrift 27) the dimension requirements were revised to a
maximum frequency of damaging flood events occurring from the systems. The urban
drainage system will now be dimensioned based on a selected return period of
damaging flood events. Additionally, it is recommended that the impact on the urban
drainage systems from even more extreme rainfall events than the urban drainage
system is designed for, are analyzed in order to minimize damage to infrastructure and
the environment.
Climate change leads to on-going changes in dimension standards because of increased

rainfall intensities and sea level rises. This implies that any dimension of infrastructure
must be assessed for a given time horizon, adapted life span and economy. The
infrastructure must at any time meet the desired standard of services for the system
listed in Guide 27.
When the rain fall exceeds the design rainfall of the systems, then there is often no
legislation or standards that describe which actions are required for the different entities
in charge of the urban drainage systems. Currently, there is no requirement for
assessments of when and where a city might be inundated caused by extreme rainfall.
Nevertheless, a real demand is present for these assessments in the light of recent years’
flood inundation events in many cities. In this report the technical procedures to
undertake the flood assessments are described. It is not considered “who you are”, but a
general urge is given to consider responsibilities so that these types of assessments are
undertaken in the future. In this way it is ensured that conscious choices are taken in
assessing flood inundation risks in cities.

1.3 Definition of the problem
This project is specifically focused towards the impacts of climate change on urban
drainage systems, in particular the impact from more frequent and extreme rainfall. It
also focuses on methods describing the impact of sea level rise and increased
groundwater table. The primary problem being considered is flooding of urbanized
areas.
Urban drainage systems are loaded directly by precipitation, and the hydraulic capacity
must be sufficient to convey water volumes to the receiving waters, otherwise when the
hydraulic capacity is met the network may discharge to the surface resulting in flooding.
The receiving waters may also be affected by climate change in terms of higher water
levels which can have significant back water effects and reduced outlet discharge
capacity in the urban drainage system. Rivers may also be affected by increased runoff
from other sources - including groundwater – which can lead to damages caused by
flooding because of the reduced hydraulic capacity. There are many contexts that should
be assessed and taken into account when looking at the function of urban drainage
systems influenced by climate change.
At this stage there is no solid knowledge about the correlation between the various key
inputs. Preliminary estimates of water levels in Køge Bay does not indicate any
correlation with extreme rainfall, but on the other hand it does not state that it is
meteorologically unlikely. In many coastal settings with low-lying built-up areas and
locations with low elevated overflow structures it is important to analyse these elements
in conjunction. If an overflow structure is not able to function as intended because of
the water level in the receiving waters, the risk of flooding increases significantly.
Similarly, if the seawater rises above quayside or dike crowns, larger areas may be
flooded with seawater and the urban drainage system may not be able to function. The
urban drainage system may in this situation have a negative effect by transporting water
to otherwise protected areas of the catchments.
Climate change is expected in some places and at certain times of the year to give rise to
increased groundwater levels which affects the urban drainage system since it can lead
to increased groundwater infiltration. The increased groundwater infiltration may lead
to reductions in the hydraulic capacity of drainage system and increasing the load to the
WWTP. Rising groundwater tables may also affect the local percolation and infiltration
systems which may lead to both local flooding and flooding due to reduce hydraulic
capacity in the urban drainage system. It is likely that extreme rainfall will occur in the
summer period, while maximum groundwater levels will occur during winter, but local
assessments should be undertaken to ensure that the effect is limited to local conditions.
It should always be assessed whether a local solution in one place may cause problems
elsewhere. The risk of this can be significant in situations where the drainage network is
operated at full capacity or is overloaded. In those cases it is recommended that the total
water cycle in the watershed is evaluated and combined with other climate-dependent
effects such as wind and seawater currents and levels. Any calculations of the urban
drainage systems should ideally be undertaken with boundary conditions that are
determined from the same expectations of climate change as rainfall. If possible, an
integrated model simulation analysis of all factors affecting runoff should be
undertaken. This integrated analysis is quite extensive and requires many details to

obtain a better result than what can be achieved by more simplified analysis
supplemented with sensitive analyses. Additionally, other analyses are required to
provide the likelihood of i.e. high water level in the receiving waters combined with
extreme rainfall runoff.
Flooding assessments can be undertaken at different levels depending on the available

information that exists on the watershed being evaluated and the type of problems to be
assessed. In the case of floods in well-defined areas, frequency and distribution of flood
may be calculated and assessed quite easily.
If the terrain and drainage system is more complex then modeling of drainage system
operation can be necessary to provide an overview of the situation. It may be an
advantage to include a runoff calculation on the surface of the watershed.
A comprehensive hydrological modeling approach may be required if the system gets

even more complicated where the water level in the receiving waters is included as a
boundary condition.
The choice of calculation level must always be adapted to the current problem and
available input and boundary conditions. It should be noted that this is the calculation of
future extreme situations based on forecasts of changes in load. There is considerable
uncertainty in all results and level of detail in the calculations, and the detail in the
model approach should be adapted to this uncertainty.
1.4 An Example – Legislation in Denmark

Any work on climate change impacts on urban drainage systems should be
interdisciplinary and will influence the dimensioning, operation and maintenance of
urban drainage networks and wastewater treatment plants. Work on climate effects
can/should influence the following official documents:
1.4.1 Sewerage plan

All municipalities are required by legislation to develop a sewerage plan. The sewerage
plan describes the status and approach for managing the sewerage in the municipality.
The infrastructure of the urban drainage system, connected industries etc. must be
described. In the sewerage plan, the municipality or wastewater utility must address
relevant land use planning objectives (including river basin management objectives) and
other local planning schemes. The sewerage plan must include measures to ensure that
wastewater discharges from wastewater treatment plants and storm water discharges do
not exceed water quantity and quality objectives for the receiving waters.
As a result of desired standards of services described in Wastewater Committee, Guide

No. 27, it is recommended that municipalities incorporate operational practices in their
respective sewerage plans. By doing this the municipalities produce a legal document in
terms of desired standards of services. When determining the desired standards of
services the practices should be examined for possible climate change effects which are
then incorporated into the sewerage plan.

1.4.2 Sewer renewal plan
In relation to the sewerage plan, the municipality is liable to develop a sewer renewal
plan (also called sewer maintenance plan, or sewer improvement plan). The sewer
renewal plan must include: An assessment of existing sewers’ condition, inspection
methods and frequency, target settings, a statement of sewer renewal needs and finally a
plan for renewal. Sewer renewal should be performed to meet the desired standards of
services specified in the sewerage plan in which the climate change effects should be
accounted for. There should be close coordination of sewer renewal plan and a possible
adaptation of urban drainage systems.
1.4.3 Contingency plan

All municipalities must, as part of the overall civilian preparedness, create an
emergency plan. There is no requirement for a specific contingency for the operation of
sewers and treatment plants. Some municipalities have, however, made such plans. A
contingency plan for sewers and treatment plants will usually take a number of factors
into account that are critical to the operation of the sewage plant, i.e. power failure,
flood damage in exposed locations, staff readiness for emergencies in order to maintain
a minimum service.
Following a risk assessment of flooding event it may be desirable to have:
 Actual physical measures to reduce the effects of a flood situation.
 Preparedness for emergency ad hoc response.
 Information / alerts both internally within the municipality's operations and

externally.
There is not necessarily a requirement for a full emergency plan, but only to assess the
significance of flood situations, as input to an emergency plan.
1.4.4 Additional municipal planning, local plans, plans for water quality, etc.
Climate change effects may affect other planning, particularly municipal, local plans,
plans for water quality, water levels, etc. It is important to incorporate the effects of
climate change and extreme rainfall in the work of sewerage plans. In the preparation of
sewerage and storm water plans, the municipalities are obliged to deal with the other
planning and ensure compliance. Thus, municipal plans and local plans could be
controlling the sanitation, particularly in terms of allowable imperviousness and
minimum allowable foundation elevations. Likewise, the municipality must deal with
the impact on water quality in the recipients by more extreme rainfall events.
The municipal plan represents the municipality’s plan for the area, and in this plan
guidelines should be incorporated for allocation of areas where flooded water can be
stored such as wet meadows, basins etc. Guidelines for foundation levels can be written
in the municipality plan for preventing flooding.
It should be noted that the horizon in the climate scenarios of 50 - 100 years is far
beyond time schedules in both sewerage plans and future water plans under the Water
Act directives, which operate with time horizons of 15 to 30 years.

In the draft Water Act directives there is no obligation to account for climate change.
The sewerage plan should be coordinated with the Water Act directives and policies in
order to reduce the risk of flooding due to climate change. The municipal plans can have
impact on the boundary conditions for the urban drainage system, such as options
diversion of water to the receiving waters. If the requirements of water plans become
very restrictive regarding the diversion of rainwater, it may have significant impact on
the handling of rainwater in the urban drainage system. In addition practices for
maintenance of watercourses, such as altered weed clearance, can have a major impact
on flood risk from rivers and can reduce the option of discharging water to the river.
Work on climate change will require a coordinated planning effort by the municipality.
Climate change takes place over a long time horizon. Despite this the magnitude of the
potential impacts and the consequent need for investment requires analysis of the
impacts on drainage systems now. This is required in order to prioritize, manage and
implement measures that reduce potential future damage to society as a result of climate
changes.
1.4.5 Risk assessment and analysis of extreme rain

Urban drainage networks are currently designed to overflow with a return period of
either 1 or 2 years. According to Guide 27 overflows once every 5 or every 10 years,
respectively for combined and separate sewer drainage systems. In addition, it is
recommended to establish contingency plans for incidents where sizing criteria is
exceeded. There is no formal legal requirement to assess extreme rainfall events beyond
return periods of 5 or 10 years, e.g. extreme rainfall events with frequencies of 50 years
or 100 years. Municipalities and their consultants have often in practice depending on
the area being sewered, undertaken more qualitative assessments of extreme rainfall in
model simulations, however, a formalized requirement for assessing the impact from
extreme rainfall in model simulations does not exist.
Since it is commonly expected that climate changes will lead to more extreme rainfall
and increased rise in sea level, it would be appropriate to assess the impact of extreme
rainfall for the drainage system - including assessment of any damage caused by
surcharged water. The methods described in Chapters 4 and 5 can be used for this work.
The Guide 27 is used by almost all Danish municipalities to design and maintain urban
drainage systems for conditions where there is no water on the ground. There are no
guidelines in Denmark for analysis and handling of the situation when there is water on
the terrain / flooding. This results in situations where important infrastructures are
flooded, e.g. as it happened in 2007 and 2010 with damages of more 1 billion dollars.
1.5 The EU Flood Directive

The Flood Directive is given by the European Parliaments and Councils Directive
2007/60/EC of 23rd October 2007 for assessment and management of flood risks.
Flooding originating from the sea is covered by "Assessment and risk of flooding from
the sea, bay or other parts of the territorial sea". Ministry of Environment and the
Coastal Directorate have identified (April 2011) nine risk areas based on a preliminary

assessment of flood risk from rivers, lakes, oceans and estuaries. This is currently in
public hearing.
The background for the Flood Directive is the major floods that occurred in Central
Europe. The objective of the Flood Directive is to establish a framework for measures
for reducing the risk of flood damages. The Flood Directive starts with article no. 1 by
defining what type of floods is covered by the Flood Directive:
1. Flooding:
A temporary coverage with water from rivers and lakes is applied to land areas
which are normally not covered by water.
2. Flood Risk:
The combination of the probability of a flood event and the potential adverse
consequences for human health, environment, heritage and economic activity
associated with floods.
1.5.1 Preliminary flood risk assessment

The first step is to undertake a preliminary flood risk assessments based on existing
studies and data. The assessment must include at least:
a) Topographic map of appropriate scale which includes land use coverage.
b) A description of the historic flooding that has caused extensive damage and are
likely to be repeated in the future. The description must include the extent of the
flood, flow routes and an assessment of damages.
European Member States shall complete the preliminary flood risk assessment by 22
December 2011.
1.5.2 Flood hazard and flood risk maps

European Member States shall compile flood hazard maps and flood risk maps. The
flood hazard maps shall cover geographical areas which may be flooded according to
the following scenarios:
a) low probability of flooding or extreme events.

b) medium probability of flooding (probably once every 100 years).
c) high probability of flooding, where appropriate.
For each scenario, the following information is shown:
a) flood level.
b) water depth or water level, whichever is applicable.
c) flow rate or the appropriate water quantities when appropriate.
Flood risk maps shall show:
a) the estimated number of inhabitants potentially affected.

b) the nature of economic activity in the area potentially affected.
c) infrastructure or installations which might cause accidental pollution in case of
flooding.
European Member States shall ensure that the maps of flood hazard and flood risk are
completed by 22 December 2013.
1.5.3 Flood risk management plans

Based on the maps of flood hazard and flood risk, flood risk management plans will be
developed. These plans will focus on: Prevention, protection and preparedness. Climate
change must be included in the plans when appropriate.
European Member States must set "suitable targets" for the management of flood risks,
with emphasis on reducing the potential negative impacts of floods on public health,
environment, heritage and economic activity. If this is relevant emphasis should be laid
on non-structural initiatives and /or reducing the likelihood of flooding.
Flood risk management plans must include relevant aspects such as costs and benefits,
flood extent, flow paths and areas which have the potential to retain flood water.
Flood risk management plans must include all aspects of risk management with special
emphasis on prevention, protection and preparedness, including flood forecasting and
warning systems. The flood risk management plans may also include the promotion of
sustainable land use practices and improve water retention as well as the controlled
flooding of certain areas in cases where flooding is avoidable.
European Member States shall ensure that flood risk management plans are completed
and published by 22 December 2015.
1.5.4 Publication of maps, plans, etc.

European Member States shall prepare preliminary assessment of flood risk available to
the public in terms of flood hazard maps, flood risk maps and flood risk management
plans. Review of maps, plans, etc. must be done every 6. year and as part of the
revisions include the climate change impact on the occurrence of floods.
1.5.5 Summary
Timeline for implementation of the directive is:
1. Preliminary flood risk assessment – deadline 2nd December 2011

2. The maps of flood hazard and flood risk – deadline 22nd December 2013
3. Flood risk management plans – deadline 22nd December 2015
Revisions and updates of the issues mentioned in the plans listed in section 1-3 above
start in 2018 and are then updated every six years. From this date the likely climate
change impacts on the occurrence of floods must be added to the plans etc.

The methods described in this report may be used in the work of implementing the
Flood Directive. It also provides an opportunity to consider the entire water cycle and
apply a holistic approach in the effort against floods.
Flood risks from urban drainage systems may not be included in the implementation of
the Floods Directive, but it is recommended that it will be included in the calculations
of the risk of flooding, because drainage system in many cases may convey water into
the critical areas of the city from both the sea and streams. In addition, flooding due to
the capacity of the urban drainage system should be known by the local authorities in
order to produce comprehensive solutions to flooding problems and prepare
contingency plans.

2 STATUS OF KNOWLEDGE ON CLIMATE CHANGE
The expected global climate changes will result in an increase in temperature and
average sea level. Locally around The World there are both predicted increase and
decrease in rainfall, and it is difficult to generalize the impacts from climate changes.
Hence, it is recommended e.g. to use statistical downscaling to find new design rainfall,
etc. Generally, it is expected in Denmark that the temperature, evaporation, rainfall and
sea level will increase. Additionally, it is expected that the rainfall events will have
higher intensities and wind.
2.1 Scenarios for future climate change

The climate change numbers presented in this report are based on Intergovernmental
Panel on Climate Changes (IPCC) reports from 2001 (IPCC, 2001). In April 2011,
IPCC published updated estimates of climate change. However, there are still only few
parts of The World where local estimates of rainfall for are available. In Denmark the
Guide 29 is used to determine the climate related extreme rainfall in Denmark. In
regards to the latest prognosis for climate related sea level rises much more data are
available, e.g. at places like the portal for climate adaption:
http://www.klimatilpasning.dk
It should be noted that the methods described in this report are generic, so that when
new figures for climate scenarios are available, only changes to the input values (rainfall
and sea level) are required to rerun the calculations. All methods and analysis will be
unchanged.
Denmark works currently with three climate scenarios (A2, B2, EU2C) calculated by
the Climate Center at DMI as well as climate scenario A1B which may be used for
analysis until year 2050. The reason is that the dispersion on (A2, B2, EU2C) is
insignificant until 2050 and A1B represents the average of the three scenarios. The three
scenarios (A2, B2, EU2C) is calculated based on IPCC’s reports from 2001 (IPCC,
2001) and are valid until year 2100. It has been decided by Danish officials that the
three scenarios be given equal priority, thus there is no preference for one of the
scenarios. It is therefore up to the individual water utilities or municipalities to choose
which climate scenario should be used.
The main features of these three climate scenarios are:
 A2 describes a heterogeneous world with local development, resulting in a

continued increase in the world population. The Economic development takes
place primarily at regional level, economic growth and technological changes
are more fragmented and slower than in other scenarios.
 B2 describes a world where the emphasis is laid on local solutions that are
economically, socially and environmentally sustainable. It is a world with a

steadily increasing global population at a lower rate than A2, and with
intermediate levels of economic development that are less rapid and more
diverse in technological changes than the other scenarios. Lastly, environmental
protection and social equity is part of this scenario.
 EU2C describes a world where the international community and the individual
countries enter agreements to develop technological solutions and changes in
behavior to decrease dangerous anthropogenic influences on Earth's climate
leading to global temperature increases exceeding 2°C above pre-industrial
levels. The scenario corresponds to the EU's objective as expressed by the
European Council in 2005.
The three scenarios are shown in Figure 2.1. It should be noted that the three climate
scenarios are fairly similar within the first 30-50 years, after which the differences
between the scenarios increase.
Figure 2.1: The evolution of global climate scenarios from the IPCC in 2001 (IPCC, 2001). Climate
scenarios A2, B2 and EU2C are prioritized equally in Denmark.

Figure 2.2: The evolution of global climate scenarios from the IPCC in 2007 (IPCC, 2007).
Figure 2.2 represents the last calculated information from the IPCC (IPCC, 2007) on the
temporal development of the globally accepted climate change scenarios. By comparing
the climate scenario A2 - calculated by the IPCC in 2001 and 2007 - no significant
differences were found at the global level. It should be noted here that local variations in
climate scenario A2 have not yet been calculated for Denmark. It is therefore not
possible to assess the importance of the IPCC 2007 estimates for existing Danish
climate scenarios.
Table 2.1: Changes in Denmark's climate up to 2071-2100 (relative to the mean of the period 1961-
1990) for the three Danish selected scenarios.
Climate Scenario A2 B2 EU2C

Country
Average Yearly Temperature +3,1°C +2,2°C +1,4°C
Average Yearly Rainfall +9% +8% +2%
Summer Rainfall (June, July, August) ÷15% ÷7% ÷3%
Maximum Daily Rainfall (June, July, August) +21% +20% +22%
Ocean
Average Wind (December, January, February) +4% +2% +1%
Maximum Wind Force (December, January, +10% +1% +1%
February)
Maximum Flood at West Coast +0,45-1,05 m Data N/A Data N/A
Source: Government, 2008. All numbers shows deviation from the period 1961-1990.

EU2C scenario is obtained by a simple scaling of A2 and B2 scenarios. DMI and the
Danish Environmental Protection Agency plans to implement a project that tries to
establish a robust climate scenario corresponding to the EU objective in which the
anthropogenic effect on global mean temperature should not exceed 2°C compared to
pre-industrial levels.
Climate scenarios, as they are considered in the IPCC assessments, are clean climate
models with a time resolution which is in the order of days, not hours or minutes. In
connection with the runoff of rainwater, especially in cities, it is crucial that there are
time series of precipitation which represent high-intensity rainfall and precipitation
linked to well-defined repetition periods now and in the future in high temporal
resolution. Projections of future precipitation made not available, but the Wastewater
Committee in Denmark has analysed and advocated which precipitation should be
dimensioned for the future when it comes to high-intensity rainfall, see Guide 29. For
coupled rain, where the frequency of precipitation is interesting, there are no
recommendations for climate-proofing of the frequency of precipitation. In chapter 2.3
the recommendations for use of future rainfall is described.
2.2 Data base for the analysis of climate change impacts on

drainage systems
The A2 scenario produced by DMI and Guide 29 has been used in this report to predict
future climate change and precipitation in Denmark. DMI's scenarios cover a 30-year
period, 2071-2100. For this reason it is only possible to comment on the temporal
development of rainfall intensities until 2071 in general, qualitative terms. It would be
highly relevant to know the expected development of rainfall intensities in the near
future (i.e. the next 30 years), but the information is not yet available. Data from the A2
scenario was used in the simulation of future effects on the aquatic environment in the
multidisciplinary research project, CONWOY (water and weather in 100 years, 2006).
DMI was a partner in this research study.
2.3 Future rain over Denmark

New measurements suggest that the change in rainfall patterns is already underway
(Guide 28, 2006) and has been for several years (Water and weather in 100 years,
2006). More intensive extreme rainfall is expected in the future, especially during
summer periods. The question now is which changes will occur and how quickly will
they come? Both issues are important to consider in planning the augmentations of
urban drainage systems. In some cases there is a need for estimates of future
precipitation in the rest of this century, while in other cases estimates of the future
precipitation is only required for a shorter time scale. It depends entirely on the actual
problem, not least on the potential to continuously adapt the system to increased load
and to judge the consequences of an error estimate in the prediction.
Based on DMI's modeling of the change in precipitation from 1960-1990 to 2070-2100

(Grum et al., 2005), it is estimated that there is likely an increase in the peak hour
rainfall of 20-50% over this period (Environmental Protection Agency report
Environmental Project No. 1123, 2006). Model calculations have provided hourly
values of precipitation for areas of approximately 12.5 by 12.5 km. The variation

between the areas is quite large. Some areas in Denmark have increases of up to 100%
while others have no increases. No explanation for these differences has been found.
They are evidently not dependent on either terrain or region. They appear to be random.
It is therefore chosen at the moment to recommend the same factor range for the entire
country. In Guide 29 a similar analysis of rainfall data has been undertaken and resulted
in climatic factors listed in Table 2.2 which is used for dimensioning Danish urban
drainage networks. The data analysis showed that there was a tendency for the factor to
increase with return period. The factors listed in the table are now applied in urban
drainage calculations.
Table 2.2Recommended climate change factors according to Guide 29.
Return Period 2 Year 10 Year 100 Year

Climate Factor 1.2 1.3 1.4
The data source for predicting the changes to rainfall in the rest of the century is still
limited. Given the uncertainty in the prediction it is considered reasonable to apply a
linear increase. It should be noted that this increase may not be applied when using a
factor 1.2 to 1.4 on the depression rain-series sampled until now. It is only a factor of
extreme rainfall.
For the climate scenarios (A2, B2, EU2C) it is expected that annual precipitation in
Denmark will not change much compared to present, but that precipitation patterns will
change. It is therefore uncertain whether the current description of CDS-rain (Guide 28,
2006) can be applied for the future by simply multiplying with a factor on present CDS
rain to obtain a future extreme rainfall event. The error in this must be accounted for by
including the uncertainty in the climate factor.
Over the next 100 years it is expected that a 10 year storm event will increase by 30%.
It means that parts of the urban drainage systems, which currently meet the desired
standards of services according to Guide 27, at some point in the future will no longer
be able to meet the functional requirement because of the increased size of the 10 year
storm event.
In order to get an overview of the future performance of the urban drainage network
model simulations can be undertaken loaded with the current 10 year and 5 year storm
events adjusted for climate change. In case the predicted climate change at a given
location results in a 30% increase in the 10 year storm event, then the urban drainage
network can be simulated with the following scenarios in order to analyze the urban
drainage network’s resilience to climate change.
Table 2.3: Expected increase in rainfall
Time Horizon Expected increase in rainfall

Present 10 year rain event 0%
In 10 year 3%
In 25 year 7.5%
In 50 year 15%
In 100 year 30%

Since there is currently no information available about the temporal development of the
climate change scenarios, the projection is linear.
2.4 Future water levels in marine waters

In the future, the mean sea level along the coasts will increase due to climate-related sea
level rise. In addition to this, new climate-related extreme wind fields will lead to
backwater or increases in near shore sea level. Changing sea levels will affect the
hydraulic conditions at the outlets in drainage systems which discharge by gravity to the
sea.
More specifically the mean sea level is expected to rise by 28-81 cm in Danish waters
during this century. Additionally, a wind contribution by storm surges of up to 30 cm in
maximum sea water levels is expected on the west coast of Jutland.
(http://www.klimatilpasning.dk/DA-DK/KYST/PÅVIRKNING/Sider/Forside.aspx,
December 2009).
The impact of the new extreme water levels for coastal municipalities will be calculated
as: Sea level rise plus backwater from new climate-related extreme wind fields. Output
from these oceanographic calculations will be extreme water levels for relevant return
periods including a measure of duration (i.e. day maximum or similar). This information
can be used to estimate the required pumping in a given urban drainage system and
whether inflow from the sea through rivers may occur.
DMI and Danish Coastal Authority recommend that the sensitivity of a given scenario is
examined against the projected future seawater condition and the possibility of even
higher sea water level rises associated with risk (http://www.kyst.dk/sw17746.asp).
2.5 Future water levels in lakes and rivers

Rivers have large differences in water flow patterns. Some creeks have a relatively
constant water flow with small differences between winter and summer flow. They can
have a little sensitivity to extreme rainfall events, while other streams and rivers have
large differences between winter and summer flow, and they can have a strong response
to extreme rainfall events. This type of stream may have a tendency to dry out in
summer.
In the future it is expected that rivers will also be affected by climatic changes. Sea level
rise will affect river and reduce the drainage capacity, and in situations of extreme water
levels the impact is increased. Since the rivers often act as boundary condition to the
urban drainage system, it is recommended that the river hydraulics are described
together with the urban drainage network.
Generally, the rivers are increasingly affected by intruding saltwater wedges, which
may alter the biological conditions in the river. Impacts from climate change on rivers
other than the hydraulic conditions are outside the scope of this study.

2.6 Future water levels / pressures in groundwater
Any changes in the groundwater conditions are interesting from an urban drainage
technical point of view. If the groundwater table changes it might impact the infiltration
to the urban drainage system, the secondary groundwater zone which has impact on the
runoff on terrain to storm water and combined systems as well as the rivers. This effect
has for the past five years had a large impact where coupled extreme rainfall events
resulted in increased groundwater table. Hereby the surface runoff from permeable areas
to the urban drainage systems increased.
Forecasts of the climate change calculated by DMI with resolution of 25 km x 25 km or

possibly finer (Sonnenborg et al., 2006) can be used for analyzing changes in
groundwater conditions.
The expected increase in precipitation events with high intensity may have local impact
on the groundwater conditions. Especially in areas with coarse, sandy sediments a rapid
rise in groundwater table may occur under very intense rainfall with potential
infiltration into sewers, basements and other underground structures. At the same time
there may be disturbances in the dewatering plant pumping from the upper aquifers.
The projected sea level rise in coastal areas would cause a rise in groundwater
potentials. The risk of intrusion of saline water in coastal groundwater wells will
increase and the drainage of coastal catchment areas will be disturbed or altered. In
many cases the sea level rise will only have very local impact on groundwater
conditions. However, it may have significant impact on runoff conditions in coastal
rivers, where gradients are small, especially combined with the anticipated increased
intensity of rainfall. This is particularly true in rivers located in West Jutland where a
fine balance between rainfall intensity (and long periods of high rainfall) and the risk of
flooding exists.
In the winter of 2006/2007 Storåen went over its banks in the area of Holstebro because
of the high precipitation in autumn and winter months which lead to increased surface
runoff to the receiving water. Flooding from the river reoccurred in 2011 because of a
combination of frozen soil and fast snowmelt in connection with rainfall.
2.7 Determination of input data and boundary conditions -

Recommendations
A calculation of future hydraulic conditions in a given area requires determination of all
relevant input data and boundary conditions. The most important is usually the choice
of climate change rainfall scenario, but also water level in the receiving waters can be
critical and in some cases where it is impacted by groundwater conditions. Additionally,
the physical planning of the area and forecasts for the development must be taken into
account as well as other changes in the society. The following summarizes how these
assumptions can be determined.
2.7.1 Choice of precipitation

Time series of precipitation over the past approximately 30 years are available for most
areas in Denmark. In most cases all rain events in such time series are used in

calculations and the simulation results processed statistically. Hereby the effects of all
rainfall events having different return periods are obtained.
Another method is to undertake a calculation using a simplified version of the model

representing the considered watershed. The simplified model simulations are used to
identify the rain events which represent the required return periods to be analysed in
detail.
Finally, it is possible on the basis of local rainfall time series or the regional model to
create CDS rain representing the required return periods. The use of CDS rain is also
the only option when rain events with long return periods are required. Because of the
relatively short period of the rainfall time series, rainfall events having return periods
higher than 10 years are encumbered with high uncertainty. In this case it is
recommended to use CDS rain.
The above options can all be used when using historical data. When taking into account
the increase in precipitation due to climate change, only the latter option is directly
applicable. As described in section 2.3 it is easy to apply a climate factor to the CDS
rain and thus load the urban drainage systems with the expected increase in
precipitation.
It is recommended to make a projection of the likely rainfall based on knowledge of

past rainfall at the given location in the country. Until improved knowledge is available
the projection should be done by applying a climate factor, refer Table 2.3 (Guide 29)
As mentioned earlier the climate factor can only be applied to extreme rainfall.
Within a few years it will probably be possible to generate all local rainfall series
corresponding to a chosen climate projection. It will then be possible also for the future
conditions to calculate e.g. discharges to recipients and other issues that require all the
rain events to be applied. Currently, we have no experience in this country in
developing and applying such artificial rain time series in the model simulations.
2.7.2 Coupled rainfall

In recent years the floods have occurred due to high intensity rainfall fallen in a very
short time, but in several cases the flooding has also occurred because the rainfall has
fallen at times when the systems: channels, streams, lakes, reservoirs and groundwater
zone have not been emptied after previous rainfall. This is called coupled rainfall.
Coupled rainfall is taken into account in the calculations by using long rainfall time
series measured in the local area. This is obviously only possible for the existing
conditions and no future relationship!
If a climate factor is simply multiplied to an existing rainfall series in order to account

for rain coupled by climate adaptation, then it is not taken into account that in the future
there might be shorter time between rainfall events and thus more likely to achieve
coupled rain.
There are currently no recommendations for how to take coupled rainfall into account
when applying climate adaptation.

2.7.3 Choice of downstream boundary condition – Future water levels in
recipients of storm water systems
The downstream boundary condition for an urban drainage system is usually a lake, a
river or the ocean. Usually, the mean water level in the receiving waters is used as
downstream boundary condition. If there is a coincidence of high tide in the receiving
waters and the design rain then it is necessary to undertake an analysis of the correlation
between rainfall and high tides in the receiving waters in terms of an analysis of the
"joint probability" between the return period for the water level in the recipient and
return period of rainfall event.
2.7.4 Sea level rise and climate-related extreme water levels

The extreme water levels along the coastline change because of a climate related sea
level rise and a backwater effect as a result of new climate-related extreme wind fields.
These new extreme sea water levels will affect the urban drainage systems in the coastal
municipalities. The new extreme sea water levels along coastal municipalities are
calculated by sea level rise plus backwater based on new climate-related extreme wind
fields. The Danish Coastal Authority has calculated the new maximum water levels for
selected coastal areas in Denmark and this information can be found on their website
and in the report: "Climate Change effect on the coast, Danish Coastal Authority, 2008".
In order to estimate local sea level rise as a result of climate-related wind fields a 2D
oceanographic model can be used. Output from these oceanographic calculations will be
extreme water levels for relevant return periods including a measure of duration (i.e.
day maximum or similar). In many cases simply using current daily sea levels (typically
an average level) or extreme water level with a specific return period (i.e. T = 10 years
estimated from a measured time series in the sea) will provide reasonable estimates. An
example of a calculation of new extreme water levels in Køge Bay is given in Appendix
D.
2.7.5 Rivers, lakes and ground water zone

The water level in lakes and streams are boundary condition for runoff from storm water
system. In each area to be analyzed, it should therefore initially be assessed whether it is
likely that the boundary conditions will change significantly compared to the urban
drainage system.
The water level in the stream depends on runoff intensity and volumes from urban and
rural hinterlands. When rainfall intensity increases in the future, it is expected that the
maximum flow of the rivers increases as well. The change of base flow in rivers mainly
depends on groundwater level including changes in the soil moisture capacity in the
future.
If the boundary condition of urban drainage system is a stream, it will in most cases be
necessary to include the effect of the watercourse in the calculations of the impact
because of the interdependency between streams and urban drainage systems.
In those cases where the water level in the receiving waters is currently affecting the
retention and discharge of urban drainage systems, it would also be appropriate to
include the water level of the receiving waters in future scenarios. If the overflow/outlet
is discharging to the sea, it is recommended to use future water levels in the ocean as the

new downstream boundary condition. If the overflow/outlet to the recipient is located
away from the sea (above the present sea level + expected change in sea level due to
climate change) an analysis of future water levels in the receiving waters is required.
The future water level needs to be estimated before the new climate-related sea level
(boundary condition) can be applied to the drainage system.
The water level in the lakes will depend on the future flow in the rivers,
infiltration/recharge of groundwater and the amount of evaporation which according to
DMI will increase as a result of increased temperatures.
The water level in the groundwater zone is inversely correlated with rainfall intensity as
increased intensity will increase the surface runoff. The rain intensity during summer
increases in future rainfall scenarios which tend to reduce the groundwater potential.
The sea is the groundwater potential boundary condition. Since the sea rises in the
future a potential increase in groundwater level in coastal zones is expected. In addition,
the groundwater level in the upper soil layers will depend on temperature, evaporation
and on farmers’ anticipated changes in drainage conditions due to the expected drier
summers.
To include all of the above effects of climate change, which potentially could influence
the future drainage system, it is necessary to establish a complete hydrological model
which includes ocean, groundwater, lakes, streams and storm water system. This is
currently beyond the scope of most municipalities and it is considered sufficient to use
existing boundary condition data corrected by the IPCC recommendations (at the sea
water level).

3 COMPLIANCE WITH THE PERFORMANCE REQUIREMENTS
UNDER THE INFLUENCE OF CLIMATE CHANGE
3.1 Description of desired standard of services

Construction of urban drainage systems must comply with national practice (In
Denmark: Guides from Wastewater Committee) specifying how much rainfall the urban
drainage system is required to convey. This is the dimension criteria. If an extreme
rainfall event occurs, then the urban drainage system is not required to have capacity to
convey the water. Figure 3.1 shows an illustration of the principles of dimensioning of
urban drainage systems. The figure shows the overall existing conditions in a
municipality. Note that service levels can be described similarly for increased
groundwater potential or seawater level.
Figure 3.1: Existing conditions in the drainage systems
The amount of precipitation (the further to the right the greater precipitation) is given on
the horizontal axis, and the probability of flooding is displayed on the vertical axis. The
entry on line 1) represents a normal rainfall, i.e. between 10 and 20 mm. In this situation

flooding due to rainfall should never happen. Additionally, it is appropriate that
artificial lakes, canals, streams etc. fit into the city and appear as realistic recreational
elements (i.e. there is a minimum and clean water flow in rivers, relatively clean water
in the lakes and channels are designed so that they appear "nice" in the city / landscape,
etc.).
Along the line to the right the flood risk limit is reached for the current dimensioned
urban drainage systems. For existing urban drainage systems this situation corresponds
to a rainfall with an intensity of 8-9 mm over 10 minutes, which could previously be
expected to occur every second year (averaged over many years). In this situation the
capacity of the storm water systems will be reached, but there will be no flooding. This
is under the condition that urban drainage systems are sufficiently maintained.
If there is an increase in precipitation (higher up on the line), the risk of flooding (as
shown on the vertical axis) will increase. Within the existing network flooding can be
reduced by putting a contingency plan in operation as illustrated by 2) -> 3) in the
figure: Before the rain falls, the lakes and channels can be drained in order to have
available storage capacity, pumping water with mobile pumps, dig out additional
appropriate storage locations, cleaning grates, dam rivers, etc. In addition, if a flood
incident is forecasted in a specific area, fences can be put up to avoid accidents due to
flooding.
Point 4 in the figure illustrates the situation when flooding will occur. The rainfall is so
extreme in this situation that it will not be possible from an economic point of view to
avoid flooding.
In Figure 3.2a vision for the development of urban drainage system and implementation
of climate adaptation is given.
Based on the increasing rainfall, there is a need to revise the current dimensioning
standard so that the urban drainage systems can handle the increased amount of rainfall.
Point 3 in Figure 3.2 illustrates ....?.
The effect of Guide 27 is that the guaranteed minimum standard of services is now at
point 3 when new construction is planned. The rainfall needs to be more extreme in
order to cause flooding – which effectively means that minimum standard of service
level is raised for current rainfall but that the population will experience the same
occurrence of flooding as today when the full climate effect has been reached (approx.
year 2100).
A rainfall event corresponding to point 3) represents now the guaranteed limit against
flooding. Further along the horizontal line towards the right (increased precipitation)
there will be a risk of flooding. Contingency plans can reduce the damages illustrated by
moving from 3) to 4) in Figure 3.2.

Figure 3.2: Vision for the conditions in drainage systems in Greve Kommune (color explanation similar
to that in Figure 3.1).
The last item on line 4) indicates the conditions of the extreme rainfall, as seen in recent
years in many municipalities such as Greve in August 2002 with 100 mm in three hours
and in July 2007 with 250 mm within a period of three weeks (and approx. 60 mm in
one day). In these situations (4) there are no guarantees against flooding. Contingency
plans might reduce flooding and damage, but it is likely that flooding will occur.
Making a city resilient against such precipitation conditions the guaranteed minimum
service levels must be moved towards Point 4) with extensive investments as a result.
It is a political question if the minimum guaranteed service level should be raised

through expansion of existing system. It is a question of what is acceptable and
willingness to pay.
Water Utilities are responsible for fulfilling the national requirements as described by
Land Commission Guides for old systems and in the Wastewater Committee Guides in

recent years, (point 2) in Figure 3.2. These requirements may be considered as Danish
practice in this area for constructions undertaken before 2005. Constructions established
after 2005 must comply with the new requirements described in Guide 27,
corresponding to point 3 in Figure 3.2.
If a municipality wants to establish a higher level of service than point 2 for

constructions undertaken before 2005, or possibly also for point 3 constructions after
2005, it is a local political issue.
As it is today, municipalities are not responsible for flooding in the cases where
precipitation is above point 2, and there is no legislation requiring authorities to take
into account these precipitation events.
3.1.1 Service level and compliance with operational requirements under the
influence of climate change
The changes in rainfall over Denmark as described in section 2.3 result in an altered
response in urban drainage systems in the future.
Work to reduce the negative impact from climate changes on drainage systems will
have the overall aim of reducing potential damage to community values - either by
reducing the flood level, or by reducing damage through efficient management of
potential flooding.
Danish urban drainage systems constructed after 2005, must comply with functional
requirements formulated by the Wastewater Committee in Guide 27:
 In combined sewered areas surcharges to terrain is only allowed once every 10

years
 In separate sewered areas surcharges to terrain is only allowed once every 5

years.
These are minimum requirements, and municipalities are allowed to set higher
standards to meet additional and / or more stringent requirements. In Guide 27 it is
stated that it is the actual situation in the catchments that is decisive based on e.g. the
actual frequency of surcharges. It is pointed out that when sizing and analysing urban
drainage systems, future changes in climate must be taken into account, particularly
changes in rainfall conditions and water level in receiving waters in order to ensure that
the system meets the required functional requirements throughout the life expectancy.
In order to check if service levels are maintained, calculations (corresponding to

precipitation described in Section 2.3) can be undertaken to analyze the response in
urban drainage system to changes in climate with return periods of 10, 25, 50 and 100
years. This analysis will reveal locations and timing in the urban drainage system where
function requirement is exceeded. This information can be included in plans for
maintenance and expansion. It is recommended by the Environmental Protection
Agency report, Environmental Project No. 1123 (2006) that a problem identification
and prioritization of the effects of climate change on three levels must be undertaken:

Level 1. Analysis of whether climate change will result in a violation of current
operational requirements.
Level 2. Mapping of the level of flooding as a result of the new design rainfall based on
forecasts of future climate changes. The water depth and spatial distribution of the
flooding is assessed as well as the potential damage.
Level 3. If significant flood water is surcharged to terrain, an assessment of any damage

and management of water on the ground must be undertaken. This assessment is part of
the planning when a drainage system should be adapted so it can handle the extra
rainfall due to climate change.
When the problems are identification they are prioritized using the risk analysis
procedure described in Chapter 6.
An example of such a projection of a 10-year design rainfall including calculations of

functional specifications to the urban drainage network is provided in Appendix A, B, C
and D. Municipalities can use this method to prioritize measures against floods. These
calculations will show the trend in the exceeding of the functional requirement. It
should be noted that any planned urban development should be included in the
calculations.

4 MODELLING OF FLOODING FROM STORM WATER SYSTEM
AND RIVERS
4.1 From surface depression maps to sophisticated hydraulic

models
4.1.1 Model tools

In connection with the choice of modeling tool for use in the analysis of climate
adaptation, it is important to focus on the following questions:
 What is the purpose of calculations?

 What is the terrain in the area like?
 What data is available?
In many cases only knowledge about whether a flood event may arise or in which areas
it may happen is required. In other cases there is a need for detailed calculations of
return periods and extent of flooding.
The types of modeling tools that are useful for different problems depend largely on the
urban drainage system dynamics and the shape of the terrain. In a simple urban drainage
system where the dynamics are less important, it is often possible to implement a
qualified calculation of flood extent based on the calculation of water balance. In urban
drainage systems that are more complicated, it is necessary to use a dynamic drainage
model.
Similarly, the terrain types are divided into simple cases, which are characterized by
gully pots / basins without dynamics and more complicated cases where the dynamics
on the surface are important.
There is often great variation in the level of detail and quality of data. For instance if
there is any calibrated dynamic model of drainage area available and if it is possible to
acquire a terrain model and GIS data of the houses and roads.
In GIS the digital terrain model, DTM is used to locate indentations in the surface and
nominate these depressions as known risk areas. An increasing number of people are
using DTM’s to assess the risk of storm water flooding in cities. In many cases, these
depressions are not necessarily risky, as there may be drainage channels draining these
gully pots, such as rivers, ditches, storm drains, etc. DTM’s are also used to assess flood
risks from the sea in the same way as for storm water flooding, however; in this case the
analysis is not necessarily complete: The depressions that appear as gully pots are only
filled if there is a system directing water to the depression.
The analytical methods can be improved by modifying the DTM so that the assessments
take into account any drainage systems directing water away from or towards the
indentation. Finally, it takes into account that the water flow has a certain duration,
which is then included in the application of numerical hydraulic models.

As shown, the DTM can be used at different levels of analysis of climate change impact
on flood risk in the city: They can be used right from the very simple to the most
complex methods. From surface depression maps to sophisticated computer models.
The technical analysis method, using the terrain model at one of the levels can be called
"technical path" see Figure 4.1.
4.1.2 Decision making process

When using terrain models for analysis of climate change, it is very important to
consider what the analysis will be used for: Is it for the initial flood risk assessments? Is
it for determining the desired standards of services of storm water and river systems? Is
it for assessing various climate adaptation options? Looking at the method from surface
depression maps to sophisticated computer model as an odometer from left to right (see
Figure 4.1: Illustration of the modeling tool (technical path) and what the different
levels can be used to) is like looking at a “decision-climate-meter” that ranges from "we
must investigate climate adaptation’, to "we need climate adaptation", to "the city is
climate adapted and we will monitor the city". The application can be called "the
political path”. The political path follows the technical path since the policy-making
processes require different degrees of technical knowledge.
4.1.3 The multidisciplinary angle

When implementing climate adaptation within municipalities terrain models and
hydraulic assessments may be used for very different things: they can provide the city
planner with an idea of the areas to be exempted from other applications and used as
storage areas for storm water; they can be used by the officer to assess whether a
planning permission can be given or not for a specific area or whether further analysis is
required to authorize a permission; they can be used by environmental planner / river
authority for advice on the establishment of wetlands for retention of nutrients and the
landscape architect to assess how various natural elements may be placed. Again, the
different disciplines require different applications of technical knowledge and follow
"the climate-meter" for the technical path: City planners can use a simple "surface
depression map" to determine land use, but in the local plan the hydraulic conditions in
combination with terrain model must be known in more details before a definite land
use may be established in a locality. The River Authority must know the specific water
levels in rivers and thus hydraulic models of rivers and terrain levels are required in
order to implement a stream-regulation which also has a positive impact on the climate
adaptation.
The Landscape Architect may in some cases use the simple surface depression maps for
example to place a lake, but in other cases known water levels are required to ensure
that the lake will actually contain water! At the “climate-meter” a "multidisciplinary"
path can be added, where applications of different disciplines in technical departments
can be anchored.
4.1.4 Economy
The price for achieving the various levels of the “climate-meter” ranges from relatively
cheap to relatively expensive, from left to right: One can relatively easily work out a
"depression map" on un-quantified data, but very detailed and hard work is required to
establish an advanced hydraulic model which includes the entire water cycle.

Figure 4.1: Illustration of the modeling tool (technical path) and what the different levels can be used to.
Similarly, many other aspects of climate adaptation fit into the “climate-meter” to
support the decision where to start and where to end, e.g. who may use the different
methods: State, regions, municipalities or utilities, solution methods, environmental
conditions and economy.
The illustration in Figure 4.1 can be used to communicate decisions. In subsequent

chapters, the technical path will be reviewed in detail, examples of climate adaptation
will be shown, and suggestions for carrying out risk assessments and prioritization of
adaptation to climate change will be given.
4.2 Calculation Methods

This section describes the hydraulic models that can be used for assessing and
prioritizing climate adaptation. The section is structured according to Figure 4.1, by
starting with the terrain model (DTM) ending with the advanced hydraulic models.
1. DTM - Depression Map: GIS analysis, using the DTM is calculated depressions
in the surface. The method takes only surface runoff into account and thus not
the storage volume and hydraulics of the pipes.
2. Quality assured DTM Depression Maps: GIS analysis, using a quality assured
elevation model (DTM) to calculate depressions in the surface. The method
includes only surface runoff and does not include the hydraulics in the urban
drainage network.
3. Depression Maps combined with simple runoff models. Simple volume account
of precipitation and runoff capacities on terrain is incorporated in the
"Depression Maps" calculation. This method provides a first estimate of risk
zones.
4. Hydrodynamic model in 1D (MOUSE). The hydraulics in the pipes and

channels are simulated. The method does not include overland terrain flow and
does not simulate floods of terrain. The method does provide information on
critical points in the urban drainage network.

5. Combined hydrodynamic drainage model and surface model (1D-1D). The
surface is implemented in the 1D model (e.g. MOUSE) as basins connected to
each other by overflow weirs. The method shows the risk zones, the amount to
ponded water as well as how far the water reaches spatially. The method does
not describe the velocity and flow paths on the surface.
6. Combined hydrodynamic drainage model and 2D surface model (1D-2D) MIKE

Flood (MOUSE/MIKE11 and MIKE 21). The urban drainage model and the
hydrodynamic 2D surface model are combined. The method includes the flow
velocities and flow paths on the surface. It is relatively easy to set up, but is very
computation intensive.
7. Simulation of the total water cycle. In the future it will be possible to combine
models of runoff from the groundwater zone, streams, urban drainage system
and the sea. This will provide a comprehensive overview of runoff in cities.
Currently, the urban drainage system and streams can be combined with the
groundwater hydrology in the zone, but it is very computation intensive and
difficult to set up models in a detail that makes it worth executing the great
work.
The following sections describe these methods in detail:
4.2.1 Terrain model - QA

A terrain model is a digital survey of the topography of an area, i.e. heights on the
surface. A terrain model in itself provides some information as to where the extreme
rainfall is likely to cause damaging floods. It is possible to identify depressions where
water can gather.
In areas where there is not an existing drainage model and where the terrain mainly
consists of gully pots, the terrain model (without a hydrodynamic model) is suitable for
estimating smaller impact and for feasibility studies. The advantage of this method is
that it can provide a quick overview of the problem. On the other hand, there may be
considerable uncertainty in the method, and one should be very aware of the errors
arising from ignoring the hydrodynamics of the system.
Processing of elevation data

Elevation data should be quality assured and then used to generate a terrain model for
further calculations. Since most bridges will act as a barrier in the raw elevation data,
there should be a review of these areas and where necessary culverts built. Similarly, all
river sections should be examined because crossing pipe lines also act as obstacles. All
areas around houses should be quality assured. An example of processing of terrain data
is given in the following example.
Usually, there is a GIS registry of the largest bridges in a municipality. This can either
be directly incorporated into the terrain model or used as a template for a manual
revision of the areas. Figure 4.2 shows the changes in water levels that can be produced
by proper validation of the digital terrain model.

Figure 4.2: Calculated depressions before culverts are opened (left) and after having been opened up
(right). (Vejdirektoratet, 2009).
The analysis can be used to make an assessment of flow capacity under bridges, since
the analysis before opening the bridges gives a picture of a potential flood event where
the culverts and bridges are blocked.
4.2.2 Surface depression maps
Calculation of depressions
In calculating the potential flood areas, it is assumed that all surfaces are completely
impermeable. There is no option for infiltration or runoff via drains and drainage
system. The maps show potential areas of risk - namely depressions in the terrain where
there is a possibility that water can gather and cause flooding. Maps can show the
distribution of missing information about the spatial distribution and volume, and partly
a list of the depths where the depressions fill up with water. When all depressions in a
map are shown it may result in a noisy picture of the terrain model. Filtering the data
can reduce the coverage of low risk flooding areas, e.g. by showing all depressions or
depressions just above e.g. 100 or 500 m3, as shown in Figure 4.3.
Figure 4.3: Reduction in the number of depressions based on volume. The first part shows all
depressions, the second part shows depressions of 100 m3 and the third part shows
depressions over 500 m3.
When depressions have been identified depths can be calculated, see Figure 4.4.
Information about depressions can fairly easily be linked with data for the groundwater
table.

This method provides no information on how much rain is required to fill the
depressions in the terrain. Filling of the depressions may in the majority of cases be
described as "worst case", i.e. an event which has a very high return period.
Depth
Figure 4.4: Identifying depressions and calculations of depths.
The following section describes a very simple method of how rainfall and drainage
capacity may be taken into account based on only volume considerations.
Waterways on the surface and contributing areas

Waterways on the surface can be calculated with and without houses in the terrain
model. The difference between the two is shown in Figure 4.5.
Figure 4.5: Waterways on the surface calculated without houses (left) and including houses (right).
Choice of method depends on the level of detail since a terrain model including houses
may cause some extra localized flooding, see Figure 4.6.

Figure 4.6: Depressions calculated without houses (left) and including houses (right).
Catchment boundaries are defined for all depressions over 100 m3, see Figure 4.7. If
larger basin boundaries are required, then larger volume or maximum depth of
depressions can be selected.
Figure 4.7: Contributing areas.
The area and imperviousness of the catchment can be used to calculate how many
millimeters of rain are required to fill the depressions, see Figure 4.7. This may be
misleading since flooding due to a half filled large depression may be critical.

Road
Stream
Catchment
Depression
mm rainfall
Figure 4.8: Millimeter rain per depression (Vejdirektoratet 2009).
A simple calculation of flood extent can be implemented by comparing the geometry of

the depressions with the knowledge of the available capacity in the drainage system.
Level of floods in small areas can be calculated based on knowledge of the capacity of
the drainage system and out of the drainage system and catchment area, see Figure 4.9.
Precipitation
Catchment
Inlet Capacity / Upstream Catchment

Outlet Capacity / Backwater
Figure 4.9: Sketch of smaller system.
As a hypothetical example Figure 4.10 shows a basin of 4 ha, where the capacity of the
outlet pipe is 80 l/s and the volume of drainage system is 43 m3. Based on observations
of water depth on the road in various heavy rain events, the imperviousness is calibrated
to 45%. Figure 4.10 shows the effect of a 76 mm rain falling in 120 minutes (equivalent
to a 100 year rainfall event). It is assumed that there is no backwater from the
downstream system. The flooded area is 0.4 hectares and it will take two to three hours
before capacity is regained in the drainage system.

Figure 4.10: Flooded area is shown in blue.
Data from the terrain analysis can also be used for calculation of the return periods of
maximum water level in depressions from the surface area, imperviousness, volume
calculation and upstream contributions, see section 4.2.3 for more details. This means
that the dynamics of the drainage system and any backwater effects are disregarded.
4.2.3 Hydraulic surface calculations

The depression map method calculates only depths in the depressions, the spatial
distribution of depressions and flow on the surface. By using a photo analysis (Nielsen,
2010) and / or GIS analysis of terrain data, the surface can be supplemented with
imperviousness. A proper hydraulic surface runoff calculation of flooding to current
rainfall events can then be undertaken. The calculations can be implemented as 1D or
2D surface runoff calculations (see section 4.2.5 and 4.2.6).
The calculations provide a good estimate of the relationship between the contribution
from the catchment and depression volume, but they do not account for contributions
and capacity of the urban drainage system. In general, this leads to over-estimation of
flooding upstream, while flooding downstream tends to be under estimated.
An example of a 1D surface calculation is shown in Figure 4.11. Flows at the surface

are described using channels. Figure 4.11 shows the calculated maximum water depths
for a 500 year event. There is a flooded area under the bridge deck which has a greater
extent than the width of the road above. This is due to the bridge deck being removed in
the original terrain model. Elevations under the bridge deck are uncertain, and the
calculated volume and water depths will be affected by this. There is less volume
available in reality which will lead to a wider flooding extent.

T500, Climate Factor 1.4
Meter
Figure 4.11: Example of a pure surface runoff model made for the Road Directorate in 2009.
An example of comparison between calculated flood using Depression Method and a

clean surface calculation is shown in Figure 4.12.

Figure 4.12: Example of flood results when hydraulic surface charge of 100 years of rain (red) compared
with Depression Method (gray).
4.2.4 Hydrodynamic drainage model 1D

A hydrodynamic drainage model (1D) can be used to provide an initial overview of
where the water first surcharges to the terrain. This can be combined with knowledge of
how water flows on the surface, and where water can accumulate in depressions. If the
water surcharges at several locations near a hazard zone classified by the Depression
Map Method further calculations in this area should be undertaken.
Figure 4.13: Example of use of Depression Map Method together with backwater effect calculations using
hydraulic drainage model. [DEFRA 2006].
The hydrodynamic models should be validated and calibrated to measurements to

ensure that the models represent real flows to a sufficiently high degree.
VandCenter Syd and Greve Forsyning have used data from meters to calibrate and
validate models since 2004. The use of meters within Greve Forsyning is described in
Appendix F.
4.2.5 Combined hydrodynamic drainage model and surface model 1D-1D

A hydraulic drainage model (MOUSE, MIKE URBAN, Info Works CS, SWMM5 or
similar) provides an approximate determination of where water will surcharge to the
terrain, but it does not describe the flow on the surface and hence no interaction between
surface and drainage system. The consequence is that a traditional urban drainage model
is flawed as soon as water surcharges to terrain requiring that the ground surface is
implemented in the urban drainage model.
Digital analysis of a terrain model shows a complex system of flow patterns on the
surface. Further analysis may reveal that the description of the surface can be simplified
to be comprised of a series of depressions, channels and overflows similar to the
calculation methods implemented in dynamic 1D drainage models. Many of the

damaging urban floods in Denmark are caused by water collected in depressions, from
which the water only flows when capacity is regained in the drainage system or by
infiltration, or routed as surface flow when the depressions are filled up. Generally, the
surface can be described as hydraulic reservoirs and overflow. In some cases the surface
description can be supplemented with channels.
On the basis of the digital analysis a one-dimensional surface model can be generated
consisting of ponds, weirs and channels after which the surface is implemented directly
into the hydraulic drain model without coupling between different types of hydraulic
models. Figure 4.12 shows the principle diagram of the 1D-1D system.
Figure 4.14: Principle sketch of 1D-1D coupling of drainage system and surface.
Analyses of DTM data will reveal that the depressions and indentations on the surface
are rarely neat and regular closed depressions. A depression calculated by the
Depression Method as a single depression usually consists of many smaller depressions.
Figure 4.15 illustrates a hypothetical example of two depressions. Often a depression
found by the Depression Map Method consists of several hundred depressions
depending on the grid size in the DTM.

Figure 4.15: Depressions and inundations - depending on whether the differences between the threshold
level and minimum elevations in depression 1 and depression 2 are combined in one
depression or kept as two depressions.
A method to generate a workable model of the surface is to make it as simple as

possible without excessive loss of information. As an example, a model can be made up
by ponds, canals and overflow when the number of depressions is substantially reduced.
Reduction criteria such as minimum threshold depth in a depression, minimum area and
/ or minimum volume at threshold can be applied.
If the threshold for a given depression is less than the chosen threshold, then the
depression can be merged with the adjacent depression. The consequence of merging of
depressions versus separation is illustrated in Figure 4.16 and Figure 4.17. The selected
values for threshold depths are mainly important in cases where water is primarily
supplied to one depression and where both depressions have a relatively large volume
under the common threshold.

Figure 4.16: Illustration of the filling of depressions.
Figure 4.17: Quantitative illustration of the filling of depressions.

When the detail method and level of detail is provided, surface depressions with
associated level-area curve and overflow edges can be generated through GIS
applications (Nielsen et al, 2009 and Jensen et al, 2010), after which they can be
imported into a hydraulic 1D drainage model as basins and overflow.
The most accurate and detailed model is achieved by low threshold values of
aggregation of depressions, but at a very high level of detail the complexity of 1D-1D
model and the calculation time increase significantly. If a very high level of detail on
the surface is required then it may be advantageous to use a 2D model for surface
calculations (see Section 4.2.6).
A number of sensitivity analyses of detail level and edge effects have been undertaken.
The main conclusions are that the level of detail of the result is higher or the same as the
chosen degree of detail in the model. The error caused by the level of detail is greatest
in sub-catchments situated on slopes with large storage capacity on the slope. In these
cases the capacity to retain water on the slope is ignored. In addition the velocity with
which water flows on the surface is overestimated (from nodes to reservoirs and
between reservoirs) unless overland flow canals are incorporated. By contrast, the
volume is conserved in the system corresponding to the selected drainage model and the
selected time step.
Based on GIS-generated surface depressions and results from the 1D model the flood
inundation and flood levels can be plotted in GIS, see Figure 4.18.
Depth
Figure 4.18: Example of representation of flood results.
With 1D-1D model technique it is possible to implement flood calculations for model
areas of considerable size and with a reasonable calculation time - without using sub-
models. The model simulations of Odense and Greve have been undertaken by applying
the 1D-1D model concept. (See Annex B and E).
In Greve, the method has been used for quality assurance of the priority of climate
adaptation (as described in Chapter 6) and to generate risk maps for documentation to
EPA authority.

4.2.6 Combined hydrodynamic drainage model and surface model 1D-2D
The hydrodynamic drainage model can be combined with a 2D hydrodynamic
description of the runoff on the surface, i.e. MIKE FLOOD, (Mark et al., 2006). With
the 2D surface model is it possible to give a detailed description of an extreme rainfall
situation where both the dynamics of drainage system and the ground surface are
included. Terrain data of good quality is required to take full advantage of the level of
detail, and the calculation time increases significantly. At this stage it is not practicable
with long time series of rain events due to calculation time.
The advantage of a combined urban drainage model and surface model is that there is
potential for great precision of the flood extent during extreme rainfall events.
A calibrated MOUSE / MIKE URBAN model can be coupled with a 2D description of

the surface water flows in MIKE FLOOD. A smaller sub-region can be selected where
the two models are coupled. In the couplings water is exchanged dynamically between
pipe flow model and the surface flow model. Because of the high calculation time, it is
important to choose the area carefully. If there is a need for a high degree of detail in
terrain, i.e. a grid size down to 1 x 1 m, it is recommended to select a smaller area. A
typical grid size will be 2-4 m.
Input to MIKE FLOOD consists of elevation models, which either originate from Laser
Scans from airplanes or helicopters (also known as LIDAR - Light Detection and
Ranging) or the xyz-coordinates for a terrain model and GIS-based polygons for houses
and roads. A terrain model including roads, houses, etc. may be purchased from a local
distributor of such data, or it can be generated by putting themes together, e.g. by
raising houses four meters above the terrain level and lower roads by 20 cm. This
process is outlined in Figure 4.19.
Terrain model +
Buildings and
roads =>
Runoff Model
Figure 4.19: Surface model can be generated from terrain data, GIS theme of houses and roads.
Selected manholes from MOUSE / MIKE URBAN are coupled to the 2D terrain model.
It is important to consider which manholes should be coupled to the surface, where and
how by considering the density of pipelines and manholes and their terrain levels. In
some cases it may be advantageous to move the coupling point in the surface model to a

model grid point with a lower level or create an artificial weir to compensate for the
volume located in all the pipelines and manholes not described in the urban drainage
model.
Physical System
Figure 4.20: Sketch of the physical system versus model.
There will be some uncertainty about how water is exchanged between the two models.
How quickly the water can surcharge through grates and manholes in the selected area
and how quickly it can return to the pipe network. Obviously, this can be a calibration
parameter, but as a rule of thumb it must be assumed that there is a limit to how much
pressure may be in the system before the water surcharges through all the grates and
manholes. Similarly, it is believed that the water on the surface reenters the urban
drainage system as soon as the hydraulic capacity is regained.
Calibration data for surface model may be based on observations of the extent of flood
inundation in certain areas during one or more historical rain events, but often there is
very little data available. The main tasks in a calibration are:
 Recalibration impervious parameters for different land uses in a drainage model

 Possible overland flow paths. Houses and other land uses may be aggregated
during the conversion of the surface grid
 Are manholes appropriately coupled, and is the right amount of water exchanged
between the two models?
 Roughness (Manning Number) on the surface is in most cases of minor
importance, and it only affects the shape of the runoff hydrograph.
It is relatively simple to develop a coupled model, but calculation time is long and
requires high quality inputs. Figure 4.21 shows results from a calculation with coupled
models. An aerial photo is used as background.

Figure 4.21: Example of calculation results from a coupled model.
4.3 Comparison of methods

The terrain model and Depression Map Method provides a good description of areas
where water can potentially be stored in depressions in the terrain. Analysis in GIS also
enables the calculation of overland flow paths on the surface and the extent of
contributing areas (catchment areas of depressions and gaps). The terrain model and
Depression Map Method are both powerful tools in the initial surveys of catchment
areas and initial ranking of priorities. The methods are especially suitable in areas where
the database for pipe registration is poor or nonexistent.
Analysis of terrain data provides no information on the influence from drainage system,
so it is important to emphasize that the results must be interpreted with caution. Based
on overland flow paths and the extent of contributing areas, the analysis may
characterize an area as low-risk zone without regard to contributions from the pipes in
the urban drainage network. A GIS analysis would not have given rise to increased
preparedness at Greve Hall or Ejersmindevej in Odense since the contributing areas
have limited geographical extent.

Figure 4.22: Terrain Models and Depression Method gives a good overview of potential flooding and flow
on the surface roads.
A hydraulic surface calculation of flooding due to current rainfall events may be

undertaken on the surface, but when it does not take into account the contributions from
and capacity in the urban drainage system the results are erroneous.
Figure 4.23: Correlation between results of Depression Method (gray), hydraulic surface calculation (red)
and combined hydrodynamic drainage model and surface model 1D-1D (yellow) - Blue
coincide hydraulic surface calculation and combined hydrodynamic drainage model.
The traditional hydrodynamic drainage model provides a good indication of where the
water will initially surcharge to terrain, but does not describe overland flow on the
surface and the interaction between surface and drainage systems. It is therefore only
possible to reliably estimate the intensity and the volume which the urban drainage
system has the capacity to capture. When water is present on the terrain, results from the
model are unreliable.
By combining data, methods and models of the different approaches, e.g. terrain model,
Depression Method and drainage model, the required amount of information is gathered
for undertaking high-level calculations to describe the interaction between the drainage

system and surface. Depending on the purpose of calculations, 1D or 2D method can be
chosen for calculating the flow on the surface.
The one-dimensional surface description provides an operational model with low

calculation time and high stability. By contrast, there is a risk of overestimating the
flood velocity, especially for areas with long flow paths and minor slopes. The two-
dimensional surface description increases processing time significantly, but calculates
the velocity on the surface which gives a more accurate description of the flood
dynamics.
For both model types increased detail of the surface reduces the error in the surface flow
description, but increases the calculation time. For 1D calculations, the number of
depressions in surface controls the surface flow, while 2D surface is controlled by grid
size. The results for the two surface models are comparable (Nielsen et al., 2009), but in
areas where flow time on the surface is large (long flow paths with relatively low slope),
a 2D surface description should be considered. Alternatively flow paths should be
included in the 1D description for describing the flow on the surface.
Figure 4.24: Results from models with, 1D and 2D surface descriptions respectively.

5 FLOODING FROM THE SEA
Flooding from the sea is characterized by high water levels in the ocean resulting in
inundation of lands which are usually not under water. This type of flood can occur
either by water running off terrain or by sea water inundation through rivers or drainage
systems.
As with the analysis of flooding from storm water system, there are several methods
available to calculate flooding from the sea. These methods range from simple GIS
analysis for the use of hydraulic computer models to .......?. The methods are described
in this chapter and illustrated in Figure 5.1.
Figure 5.1: Illustration of the modeling tool and what the different modeling levels can be used for.
5.1 Calculation Methods

Methods for analysis of the risk of flooding from the sea include:
1. "Water Elevation Maps": GIS analysis using only a digital terrain model to
identify flooded areas. The method uses sea water levels and terrain elevations.
This method shows only where there could potentially be water on terrain if
there was a connection to the coast.
2. “Water Elevation Maps of areas in connection to the sea” (in GIS terminology
called “Cost Distance”: GIS analysis where the influence from dykes and
barriers on the flooding is taken into account. For this analysis a quality assured
DTM is required.
3. “1D-MOUSE/MIKE URBAN calculation: where only the pipes and possibly

streams are simulated under various increasing sea-level scenarios. The
calculation shows the potential extent of flooding from the sea through the storm
water system and possibly rivers, but only at nodes.

4. 1D-1D depressions in the terrain model are implemented in the MOUSE/MIKE
URBAN model including the complete drainage system (rivers and storm
water/combined system). The water from the sea routes through drainage system
and terrain via the MOUSE/MIKE URBAN basins and pipe network. The model
does not model flow velocities at terrain.
5. Hydrodynamic modeling of flow on the terrain: A 2D hydrodynamic model is

used to simulate water flow from the ocean and onto the land. The routing of
water follows the terrain elevation.
6. Hydrodynamic modeling of flow in the terrain, combined with a hydrodynamic

model of the drainage system: The method uses a 2D hydrodynamic model to
simulate water flow from the ocean and onto the land. The routing of water
follows the terrain elevation. At the same time the water flow is calculated by
the model of the drainage system and low-lying areas which are linked to
openings in the drainage system that can become flooded.
Terrain Models
A terrain model is a digital topographic representation of the area. This model can either
simply describe the topography in the form of contour lines (called a digital terrain
model) or it can also include structures, in the form of dykes, houses, walls, etc. known
as a height model.
When choosing which type of terrain model is the most appropriate, factors in the
model area that may affect water currents need to be evaluated. Physical structures
(dykes, houses, etc.) which significantly affect the water flow, must be included in the
terrain model. Information on houses might be generated based on GIS layers
containing information on the location of buildings. This information is often available
in the municipality. A terrain model with descriptions of houses, roads and other vital
structures can typically be purchased from a data supplier.
5.1.1 Water elevation map

A GIS analysis could easily provide information about where a given water level may
flow over land and fill up to the given water elevation.
The flood map can be developed by calculating the local water depth as the difference
between sea level and the local terrain elevation. All areas in the terrain model, which
lie below sea level will be flooded, even those not connected to the sea. This is done in
practice by setting the signature for water level to blue until the required elevation is
visualized (e.g. 2m level). This method is obviously subject to considerable uncertainty
because the method assumes that areas which have no direct contact to the sea are
inundated. If a method in which only areas with direct contact to the sea (through the
terrain) are required, the tool "Cost-Distance" in ESRIs software can be applied.
The described method is most suitable for coastal areas with large slopes, meaning only
areas near the coast are inundated. Limitation of the method is the fact that natural
physical processes are ignored, e.g. the time it takes the water to flow on terrain. The
method yields the maximum flood image and is especially good for rapid screening of
the areas that are at risk of flooding.

Water level Coastal area terrain
Inland depression
Figure 5.2: Illustration of elevation-map method using GIS analysis.
In the following example a flood map for a maximum sea level of 2.24 m is provided.
The digital terrain model of the area has a grid size of 1.6 x 1.6 m. The calculated flood
is illustrated in Figure 5.2. The map shows flooding along the coast and in the areas
behind the coast line.
Figure 5.3: Flood maps based on elevation-map method - a pure GIS analysis.
If it is assumed that only those areas which are in direct contact with the sea are flooded
by the rising sea level, then the flooding is reduced significantly as shown in Figure 5.4.

Figure 5.4: Flood maps based on Cost-Distance method (see below) - a GIS analysis. Only land which
is directly linked to the sea is likely to be flooded.
5.1.2 Water elevation map in connection to the sea

Obviously, it is more appropriate to analyze the areas that are directly in contact with
the sea and therefore inundated. The Cost-Distance function considers which areas will
be affected directly from a defined source (the ocean) using the information in the
height model (DTM). At this stage special attention is required to areas where a gap is
defined between sea and land in the DTM. Limitations in the existing definition of the
DTM are listed below:
1. Rivers are not wide enough to be described in the DTM (if e.g. the stream is 1 m
wide and the DTM resolution is 1.6x1.6m), which potentially may reduce the
flooding modeled by the model compared to reality.
2. Rivers passing through culverts under bridges may not have been opened in the
DTM (e.g. the roads and small bridges), which potentially reduces the flooding
in the model compared to reality.
3. Dikes are not represented because the DTM is prepared before a dike is
established (which leads to artificial flooding in the city by the model).
4. Bridges have been removed in the DTM, thus giving an artificial "gap" into the
city (e.g. if there is a structure under a bridge that closes at high tide which is not
represented in the DTM and results in artificial flooding of the city by the
model).
5. Storm water systems and combined sewer systems with overflow to the sea are
not described in the DTM. This is an overall situation because the DTM does not
have any information about the pipe network under the surface. This restriction
gives potentially less flooding in the model than in reality, unless non-return
valves are installed on all outlet pipes and constructions.

Finally, it should be noted that the DTM only shows surface height and no information
on whether the dikes which are described in the model can withstand the pressure
provided by the water. Around Denmark there are many beach dunes, which from above
look like dikes, but which may be washed away by storm surges.
The dynamics of the system (energy losses in streams) and duration of high water level
are not included in this analysis and should be assessed by other criteria.
If the DTM is modified to address the limitations of the items 1-5 listed above, the Cost
Distance method is a powerful tool to show the worst case flooding of cities combined
with high water levels in the ocean.
The dynamics of the system (including energy losses in streams) and duration of high
water level is still not part of the approach, but can be assessed by analysing the
durations of storm surge and estimate the importance of this by using an average water
velocity (e.g. 0.25 m/s). If the distance from the sea to a low-lying area is short, the
duration has less significance compared to a long distance.
To ensure that "worst case" is assessed it is obvious that the DTM is quality assured. In
Figure 5.5 an example of a cost-distance calculation is shown with and without correct
description of the stream. As Figure 5.5 shows, it is necessary to ensure that all terrain
features such as rivers, dikes, bridges and culverts are represented in the DTM.
This method does not account for storm water and the ability of sanitary pipes to
transport water into the city. A rapid screening where this ability is included can be
undertaken by opening the terrain model where pipes are located in the ground. If this
analysis is carried out, then it is very important to be critical of the outcome, since the
water will only enter the network at manholes.

Figure 5.5: Illustration of a flood calculation of the cost-distance method using a raw data DTM (top) and
a quality assured DTM (bottom), where rivers, dikes, bridges and culverts are represented
(analysis and calculations are only carried out to illustrate the method).
5.1.3 1D-urban drainage systems

Calculations of flooding in the city resulting from inflow of water due to high sea levels
can be undertaken with 1D hydrodynamic model. The sea water level is used as
boundary condition at the outlets of the model. The result from the 1D simulations
provides information at which manholes flooding may potentially occur in the city.
5.1.4 1D-1D two-layered hydrodynamic model

By using the method as described in 4.2.5 simulations of the risk of flooding can be
undertaken by setting the boundary conditions at outlets to the sea level.
5.1.5 Hydrodynamic calculation of flood

This method uses a 2D hydrodynamic model to model the water flow on the terrain, and
describes the flow between land and sea by applying the sea water level as boundary to
the model. Compared to Cost Method, this method provides additional information in

terms of flow velocities on the surface during flooding. This physically based method
describes in detail the flow and time delay of water flowing through low-lying areas
with complex geometry.
Input to the hydrodynamic calculation of flood

The description of the flow on terrain is based on input in the form of a digital terrain
model describing the topography in model domain. The sea level represents the
boundary conditions for the model domain, which define the variation of water level
and the following potential flooding.
When the model domain is defined it must be ensured that the elevation of the inland
area is greater than the maximum expected water levels in the ocean. In addition, the
rivers, dikes, etc. must be described in the terrain model to ensure accurate descriptions
of the flow around / over / through them.
dtm_5_06feb2009.dfs2
Elevation (m)
-1.98 - -1.74
-1.73 - -1.54
-1.53 - 0.32
0.33 - 1.09
1.10 - 1.83
1.84 - 2.58
2.59 - 3.54
3.55 - 4.58
4.59 - 5.65
5.66 - 7.94
Figure 5.6: Model area when the expected maximum water level is 3.0 m.
It is possible to calculate the effect in terms of floods from time-varying water levels in
the ocean.

Climate Scenarie
Future A2
Scenario, T=100
Water 3.0
Level
Water Level (m) 2.5
2.0
1.5
1.0
0.5
0.0
0 20 40
Duration (h)
Time (hours)
Figure 5.7: Illustration of a time-varying water levels in the ocean - as input to a flood calculation.

Results
Model simulation results are temporal variations of water depths on the terrain. In
addition, information on flow rates can be plotted, e.g. for an analysis of the influences
from the forces of the water. The methods do not include the conveyance of water in
streams and pipe networks to and from the city.
MzResultView1
6166600
6166400
6166200
6166000
6165800
6165600
6165400
711000 712000
07/02/07 16:00:00, Time step 20 of 40
Figure 5.8: Example of model results showing water depths and velocities.
5.1.6 1D-2D hydrodynamic modelling of the flow on terrain combined with a

hydrodynamic model of the drainage system
A model describing the flooding from the sea can be combined with a model of the
drainage system. This combination makes it possible to describe the water flow on land
and through the drainage system. This is done by coupling a drainage model (e.g.
MOUSE/MIKE URBAN) and a 2D hydrodynamic model (e.g. MIKE 21). It is
important to describe the pipe system, if it is located in an area where the inland terrain
is lower than the coast. The advantage of this model application is the mapping of areas
with flood risk which would otherwise erroneously be assumed flood safe.
Water level
Water level
Drainage pipes
Drainage pipes
Figure 5.9: Illustration of the components in a calculation with a drainage model in combination with a
model describing the flooding from sea.

A 2D model describing water flow on the terrain is built as described above and can be
linked dynamically to a calibrated model. The coupling of the two models can be
reduced to the locations where the drainage system can potentially contribute to
flooding on the surface. In areas which are significantly higher than the expected
maximum water levels in the ocean, there is no need for linkage between the two
models. In addition, the level of detail in the drainage model can be reduced at those
locations far away from potential flooding areas. Simplified drainage models with fewer
couplings to the 2D flow model will reduce the simulation time.
Figure 5.10: Only selected areas of the drainage model (shown in red) are coupled to 2D flow model.
This optimizes the calculation time.
A time series with the variation of sea level is used as boundary condition for
calculation. In addition, the urban drainage model can be loaded with rainfall. Then the
combined effect of high tide in the ocean coinciding with a rain event can be analysed.
Results
Model results are temporal variations of water depths on terrain along with flooded
areas. In addition to this information on flow velocities can be plotted, e.g. for analysing
the impact of forces from the water as well as information on water flow and water
levels in drainage system.

6 URBAN CLIMATE ADAPTATION
Climate change adaptation in urban areas is described in this chapter by applying risk
assessments in practical solutions in order to achieve desired standards of services in
preparation and layout of contingency plans.
6.1 Risk analysis
6.1.1 Damage assessment using a risk analysis

Assessing the risk of damages in an urban drainage catchment can be undertaken at
different levels, from broad qualitative analysis to quantitative analysis. Other impacts
may be taken into account in the analysis. Besides the influence of extreme rainfall
there are also risks in the general operation of drainage systems.
A complete risk analysis of the system can be undertaken by systematically examining

how the drainage system operates under different conditions during both extreme rain
events and periods with disruptions in the service by weighting the various disruptions
by their importance. A simpler risk analysis focused on extreme flooding is also
possible. Analyses at both levels are very useful tools that can be used in prioritizing the
maintenance, and operational actions should be continuously made to upgrade the
drainage system.
Traditionally, damage caused by surcharged water on terrain is divided into three

categories:
 Direct damage - typically damage caused by standing or flowing water.

 Indirect damages – e.g. traffic accidents because of aquaplaning, traffic
disruptions, administrative costs, labor costs, loss of production, etc.
 Social costs - negative long term effects of a more economical nature, such as
reducing the value of property in areas subject to flooding and slower economic
growth.
A big advantage of a risk analysis is that all causes of flooding are assessed and
weighted. Hence optimizing the time and avoiding disproportionate spending of time on
some measures, while others, perhaps more important, are overlooked. As an example a
pump failure of a pumping station due to obstruction or power failure during a moderate
rain event could result in flooding comparable to the flood caused by an extreme rainfall
event. One method to find the cost related to flooding in urban areas is to collect
information on documented flood incidents by the insurance companies, as e.g. made in
Norway (König et al., 2002), Denmark (DANVA, 2005) or Brazil (Nascimento et al.,
2005). An internationally recognized technique to quantify the damage is the use of
"Flood Damage Curves", describing the extent of the damage as a function of land use
and water level, refer (Speight, 2006) and (Nascimento et al., 2005). Currently, such
"Flood Damage Curves" do not exist for any areas in Denmark.
The following issues should be included in an assessment of damage related to flooding:

 To prevent that the population is brought into contact with a mixture of sewage
and rainwater due to overloading of the drainage systems
 That vital community functions, such as electricity supply, water supply, heat
supply, communication points and access to hospitals are not out of operation
due to flooding
 That the number of affected basements and buildings are minimized
 That the number of flooded electrical power cabinets and other equipment is
minimized
 The impact from flooding on traffic is minimized.
6.1.2 Risk analysis
Definition of risk concept
Risk is the combination of the probability of an adverse event (e.g. failure of wastewater treatment plant /
pump station, basement flooding, releases of hazardous substances, errors in management / SCADA) and
the magnitude of the consequences (e.g. damage to facilities, personal injury, odor, traffic delays, fish kills)
and severity (is the release of 1 litre or 100 litres, is it the hospital that gets flooded, how many are injured).
Mathematically expressed as: risk = probability times consequence.
A plan for managing risk includes the following seven steps:
1. Identify the risks (e.g. what can go wrong?)

2. Assess the likelihood and consequences of these risks.
3. Determine the risk mitigation options
4. Assess the economic, environmental, public relations and operational costs and
benefits of the options
5. Prioritize the mitigation option
6. Identify the decision makers
7. Develop the implementation plan.
The steps in a risk analysis are illustrated in Figure 6.1.
Figure 6.1: The process of risk analysis.

The first step is data collection, where knowledge of the drainage system is obtained.
This is followed by a coarse risk analysis during which a screening of infrastructure is
undertaken by experts and special risk tools. After the coarse risk analysis, there are two
options; either to prepare a detailed risk analysis with focus on selected areas from the
coarse risk analysis or to go directly on to identify mitigation measures. If it is decided
to proceed with the detailed risk analysis then it is possible to quantify different priority
risk mitigation measures.
In order to prioritize the selected sites, it is necessary to establish three matrices:

 A frequency matrix
 A consequence matrix
 A risk matrix
The frequency matrix consists of seven intervals named F1 to F7. F1 is an event that
statistically occurs less frequently than once every 10,000 years. F7 is an event that
statistically occurs 10-100 times a year. The frequency ranges are constructed according
to a logarithmic scale. Because of the logarithmic scale it is not important to know the
frequencies of adverse events accurately. It is important to know the magnitude of a
given event to be used. The frequency matrix is shown in Figure 6.1.
Frequency Interval Classification Frequency per year

daily to monthly F7 10 – 100
Monthly to year F6 1 -10
1 – 10 year F5 0.1 – 1
10 – 100 year F4 0.01 – 0.1
100 – 1000 year F3 0.001 – 0.01
1000 – 10000 year F2 0.0001 – 0.001
< 10000 year F1 0.00001 – 0.0001
Figure 6.2: The frequency matrix.
A logarithmic scale is used between the individual impact categories in the matrix to
make it possible to compare the impact groups. "Negligible" for instance indicates an
economic value of 10,000 to 100,000 DKK, while "Marginal" indicates a value between
100,000 and 1 million DKK.
The economic scale used in the consequence matrix is not arbitrary. Each figure is
estimated from available sources and practical guidance numbers.
The consequence matrix can describe the different impact categories ranging from
no/negligible impact to the disastrous impact described in both qualitative and
quantitative terms.
The accumulated risk matrix is shown in Figure 6.3.

Risk Matrix
Consequences
Classification of risks
None Insignificant Marginal Serious Critical Disastrous
Frequency classes
0 1 2 3 4 5
Number per year
10 – 100 7 7 8 9 10 11 12
1 – 10 6 6 7 8 9 10 11
13
10
0.1 – 1 5 5 6 7 8 12 9 10
8 4 8
0.01 – 0.1 4 4 5 6 7 7 9
5 6
2
0.001 – 0.01 3 3 4 5 3 6 1 7 8
9 11
0.0001 – 0.001 2 2 3 4 5 6 7
0.00001 – 0.001 1 1 2 3 4 5 6
Larger than 7 Not tolerable

6 or 7 Not desirable
5 Tolerable
Less than 5 Negligible
Figure 6.3: Risk matrix. In the matrix are examples of selected sites in sewers placed in relation to the
assessed frequencies and consequences.
Four colors are used in the risk matrix to indicate whether the calculated risk level for a
given event is tolerable or not. A risk level above six or seven shall lead to
implementation of defined actions to reduce risk levels. According to Figure 6.3
identified mitigation measures must be implemented for the two events classified in the
non-tolerable region (indicated by circle No. 4 and No. 13 in Figure 6.3).
All items located in the yellow area should be evaluated based on a cost-benefit-analysis
that can determine what and how much is required to reduce the level of risk and
whether an investment should be made here and now, or only when the impact occurs.
An analysis provides the basis for assessing the risk level for the entire drainage system
and to assess this level relative to the acceptance threshold defined in the risk matrix.
For incidents above the acceptance threshold risk mitigation measures must be
identified and implemented. For incidents which lie in the acceptance area, an
identification of optional mitigation measures must be undertaken and assessed through
a cost-benefit-analysis.
6.1.3 Risk of flooding from extreme rainfall

A risk analysis of flooding from extreme rainfall alone can be based on flood maps, see
Chapter 4. The simulation results of rain with high return periods may be plotted using
GIS themes or aerial photos in order to identify problematic areas. Each area must be
assessed as to whether flooding is a problem and whether there may be damages.
The assessment shall be based on the following considerations:
 If a park or football field is flooded for a given return period, is it acceptable? Is

the inundation from a separate or combined system? How long does it take
before the area can be used again and is clean up required?
 What flood levels will affect basements, first floor, electrical cabinets or parked
cars, etc.?

 How much does the number of different damages increase caused by climate
change? Is there a risk of more damages related to urban development and what
is the flood impact from planned upgrades of drainage system? Can damages
caused by flooding be exported to other locations?
 What is the level of uncertainty in the model results? How well is the model
calibrated, and has a safety factor been included? Is it reasonable to interpret the
results directly, or should a safety factor be applied to the results?
 Compilation of damages.
The cost of flood damage varies depending on what is damaged, if the damaged items
have been completely or partly written off, replacement cost, etc. Moreover, the cost
depends on whether the flooding was caused by rain water and sewage, and where the
flooding occurred. It is therefore very difficult to generalize the damage costs. A general
list that accurately describes the cost of flooding of electrical cabinets, basements,
houses etc. cannot be developed. It is therefore recommended to first determine the
number of damages by type and then to cost the damage.
To quantify the loss by flooding it is desirable to have a geographical overview of what

values might be flooded. Typically, municipalities have records of where the buildings
are located, and housing registration (BBR in Denmark) contains information about
location of basements. The basis for the comparison is established by combining the
building theme and BBR data. Public buildings will often have a higher value than a
single dwelling, so it may be appropriate to categorize the public institutions in terms of
use, i.e. as kindergarten or a nursing home.
Streets convey rainfall water into the drains. However, when the capacity of the
drainage system is exceeded the water may surcharge to the roads. The roads are then
used to convey the excess water during the rainfall event. In these situations it is
important to know estimates of water depths, water velocities and where the water
flows. Roads are usually designed to drain storm water quickly and efficiently.
However, when there are significant amounts of water on roads, it might conflict with
the original design of the road. If an analysis shows that a road under future climate
conditions will be flooded more frequently, it should be discussed and resolved with the
road authorities. The road construction may be adapted. In connection with damage
assessments of roads it is pertinent to examine the criteria for the operation: How much
water on the road is allowed by the Authority before the road must be closed? And to
investigate the road quality, so it can be determined how much water it takes to destroy
the foundation of the road and how long the road can be flooded before damage occurs.
Valuation
The following parameters can be used for valuation of flooding:
 Housing
 Crèches
 Kindergartens
 Nursing homes and sheltered housing

 Water distribution. Flooding of the building of the water treatment plant causing
possible contamination of clean water
 Water wells, flooding may lead to contamination of the bore field
 Petrol stations where there may be a risk of water flowing into the tanks, so the
petrol runs out (Service Stations with newsstand sales, Auto Service etc.)
 Areas of storage of oil and hazardous waste near recipients
 Companies with oil and petrol separators connected to the sewage system and
storm water system. Oil/gasoline may either surcharge inside a building or
outside. (It will run approx. 50-100 l from each separator.)
 Especially for wastewater systems:
o Avoid overflows from sewage pumping stations.
o Avoid swimming pools becoming contaminated with sewage.
It is important that people with the greatest knowledge of the area being examined are
consulted to determine appropriate values for the various categories above. In some
cases, the GIS staff has a good overview of the information available. The boundaries of
what can be illustrated and calculated from GIS primarily depends on what information
is available. Below are a few examples for inspiration.
Figure 6.4 shows a theme with houses inundated by various return periods. The GIS
layer of simulated floods is linked to the GIS layer of houses taking into account the
foundation level.
In Figure 6.5 the electricity cabinets are illustrated with floods exceeding 40 cm, by
which flooded electricity cabinets can be identified and counted.
Figure 6.6 and Figure 6.7 show the specific buildings plotted together with extent of the
flooding and flood depth. In this example schools, kindergartens and service stations are
shown. This type of GIS illustration shows how and where health or environmental
issues may arise.
T = 20 yr
T = 50 yr
T = 100 yr
Figure 6.4: Example of GIS theme of the houses flooded at different return periods.

Figure 6.5: Example of GIS theme of electrical cabinets damaged. Placement of electrical cabinets is
pictured together with water levels above 40 cm.
Figure 6.6: Example of GIS theme showing specific buildings. Flood propagation is pictured together
with the location of schools and kindergartens and gas stations.
Figure 6.7: Example of GIS theme showing specific buildings. Flood levels are pictured together with
the location of schools, kindergartens and gas stations.

For valuation of a new road foundation the following should be noted:
 it varies according to thickness, etc..

 most expensive is asphalt layered roads. In this situation it will be necessary to
remove and deposit the asphalt before construction of new foundation, followed
by a new asphalt pavement.
 in parts of the foundation there will be cables, and costs associated with coping
and any repairs due to damage caused by replacement of the foundation are
impossible to estimate. Worst Case = much more expensive than road
foundation and asphalt replacement.
6.2 Priority adapting to floods under a changing climate

As it appears the risk analysis can be used as a basis for prioritizing actions to prevent
floods and to adapt to climate change, but in many cases it will not be necessary to
implement the full risk analysis to get started.
Analyses of climate adaptation can be achieved at many different levels (as described in
the introduction to Section 4). These methods can be used for different degrees of
priority: Establishing a basis for getting started, where models shall be established,
prioritizing measurement programs, prioritization of specific climate adaptation in the
form of installations, priority for emergency action, etc.
In relation to the specific climate adaptation it will be necessary to implement priorities

of both the economy as technical measures, which are highly political decisions, but the
political decisions must obviously be made on a sound technical basis.
Priorities can be carried out based on assessments of risks of flooding, but can also be
implemented based on economic assessments: where do you get the greatest reduction
in flood risk or most climate change adaptation for the money?
Prioritization of climate change adaptation can be based on the climate-meter. Among

others it will be possible to prioritize where to undertake registration of the pipe
network if it is not available in a digital form. The digitization can be undertaken based
on the relatively simple Depression Map Method combined with a simple hydraulic
model, refer 4.2.2.
In Greve Municipality it was decided that the entire urban drainage network must be
upgraded so that the water level in these systems only surcharge to terrain once every 10
year. Based on the experiences from the flooding in July 2007 and a flood risk map
prepared using the so-called quality-assured Depression Method, the entire municipality
is divided into 42 urban catchments, and priorities of the climate change adaptation is
planned for the next 12-15 years.
The hierarchy is based on the idea that those who have been hit hardest must be climate
change adjusted first. This plan is politically decided.

6.2.1 Urban prioritization in Greve municipality
Model domains in general follow the storm watersheds in order to ensure that hydraulic
analysis of climate change adaptation can be adequately undertaken for the urban
drainage network.
The following information was used to help guide the prioritization of upgrade works:
 Experiences from the floods in July 2007, as reported directly from citizens or
landowner associations.
 The digitization of the storm water system.
 The digital terrain model for Greve Municipality which is used to calculate the
depth of the depressions in the surface, the "Depression Map Method".
 GIS theme of buildings in the municipality and the theme of business and public
buildings.
6.2.2 Description of the method for prioritization in Greve

Urban areas ranked by the experience of flooding in 2007.
Prioritized urban watersheds
Figure 6.8: Prioritization of urban water sheds. The numbering reflects the order of implementation
Figure 6.8 shows the prioritization of urban areas. Urban area no. 1 is the first area to
become adapted to the climate change, followed by urban area no. 2, etc.
In order to give an idea of the flood risk to commercial and public buildings, the number
of these located in the depressions is shown for each urban area. However, these
commercial and public buildings were not included in political decision on the
prioritization of climate change adaptation.

6.2.3 Quality assurance of priority
The hydraulic models will be improved and updated through monitoring programs and
calibration. Improvements in the models due to additional data may lead to future
changes in prioritization. Similarly, economic aspects may lead to reprioritization. It
may for instance appear that a relatively simple and inexpensive augmentation will have
a major positive impact on climate change adaptation (such as is done with the
establishment of an outlet pump at a main outlet or the provision of terrain planning).
A similar ranking, where the number of affected properties is investigated, is

implemented by using 1D-1D surface calculations as described in 4.2.5. After the
hydraulic calculation the number of houses within the flooded areas is counted, and
priorities are implemented and compared with the first priorities.

6.3 Options for adapting the urban drainage system
The expected higher rainfall in the cities should either be discharged or stored in order
to avoid flooding. Possibly, part of the water can infiltrate locally before it enters the
urban drainage system. A wide range of technical options exist to solve this. The
drainage system can be built out with additional or larger pipes, and ponds for storing
can be constructed. In the following examples various augmentation options are given.
Main groups of options are:

 Active reduction of inflow of rainwater to the drainage system, i.e. through
increased infiltration of rainwater
 Temporary controlled storage of rainwater, i.e. using wetlands
 Augmentations in the drainage system that increases capacity, i.e. larger pipes,
basins, etc.
6.3.1 Physical measures on drainage system

Addressing the increased rainfall from our urban areas in order to meet the standard of
services and reduce the flooding can be done in a variety of ways.
There are three types of solutions: To avoid the increased volume of water discharged to
the drainage system, increasing the discharge or the storage capacity of the drainage
system or possibly a combination of these. Reduction of inflow to the drainage system
can usually only be achieved by establishing local infiltration of water. Drainage of
storm water can be done through open channels or closed pipes to the recipient, to larger
infiltration units, or with any wastewater to treatment plants.
Storage systems can be either traditional basins like concrete boxes or pipe-basins, or it
can be lakes and ponds. Beyond the physical conditions in the catchment, treatment
plant capacity and conditions in the receiving waters, it is crucial whether the drainage
system is a combined system or a separate system. In a separate storm water system, it
is usually much easier to find diversion options for a local recipient, than it is for
overflow originating from a combined system. Figure 6.9 shows an overview of
possible ways of regulating storm water into a drainage system.
Figure 6.9: Schematic overview of possible ways of regulating runoff. Source: Parkinson & Mark 2005.

The following describes some of the most common approaches expected to be used for
augmenting existing drainage systems, so that they can meet performance requirements
during the future increased load. The municipality should be aware that actions often
lead to a need for revision of both wastewater and discharge or infiltration permits, e.g.
if overflow volumes or local infiltration is increased/established.
6.3.2 Infiltration of storm water

Where it is geologically and hydrologically possible, infiltration of storm water can be
established at each property, or complete infiltration solutions can be made for small
urban areas. Storm water from roads, sites, car parks etc. may in some cases be
infiltrated, but in those cases in must be determined whether the infiltration may pose a
threat to groundwater quality.
Infiltration is basically the best environmental approach for discharging unpolluted

rainwater, as it largely corresponds to the natural way and results in only limited
interventions in the natural water cycle. Drainage structures should be designed so that
there is emergency overflow from infiltration facilities to the public storm water system.
This reduces the risk of flooding and the size of the infiltration facilities required is
limited. However, this has the unfortunate consequence that during heavy rain the
fascines can be filled up, resulting in a quite instantaneous and uneven flow back to the
drainage system, which must therefore be designed to cope with these peaks in the flow.
The capacity of fascines may typically be of a size equivalent to 20-30 mm of rain, but
there is no assurance that fascines are empty at the beginning of rainfall events. For this
reason it is not certain during an extreme rainfall event that flooding is reduced
significantly. However, fascines can reduce the yearly runoff volume considerably and
increase groundwater recharge.
Combination of infiltration and storing

As mentioned infiltration systems for storm water usually have a limited capacity,
requiring water to be stored during periods of major inflows. The optimal combination
of storage size and infiltration capacity can be calculated or estimated based on
knowledge of soil infiltration capacity, flow conditions, etc., refer Guide 25.
6.3.3 Separation of combined systems

Many of the most appropriate measures to address the increased rainfall are ill-suited
for combined systems. The mixture of sewage and storm water is so polluted that the
water must be treated with caution. Human contact with the water poses a risk of
disease, and there are aesthetic problems at outlets. Functional requirements are
therefore much more stringent to the combined systems than to storm water systems. It
is natural to consider changing the old combined systems to separate systems, especially
if the spare capacity of the combined system is limited requiring major built-outs. In
practice, this involves so many problems that only a few places exist where it is
implemented. It is very expensive and very difficult to build a completely new drainage
system, which also requires that the pipes located at each parcel are converted to a
separate system.
Separating the combined system is carried out in smaller communities and new built-
outs, but rarely in the old city centers where the need is often the greatest. Therefore,

there is a need for other solutions to address at these sites. There is currently a
“standard” solution for these city problems.
6.3.4 Increasing pipe sizes

If the conceptual layout of a drainage system cannot be changed, it can be chosen to
simply increase the dimension of all the pipes in the system, so the capacity will meet
the requirements in the standard of services.
Alternatively, an additional pipeline can be added along the existing pipeline. Prior to
this augmentation the drainage system should be carefully analysed in order to only
make the necessary substitutions, and it should be considered to increase existing
dimension on some stretches, and whether there are alternative pipeline options, which
can reduce costs for expansion.
Trunk mains
The increased runoff flow from a catchment can be conveyed through larger pipes or
stored in basins. Due to lack of space it may be difficult to expand the sewage system,
and an option could be to build tunnels conveying the water from strategically well-
placed nodes in a catchment area to the recipient or main trunk line. The tunnels can
also act as extra storage capacity. The technical and economic feasibility of using such
solutions have been considerably improved in recent years. It should be noted that the
increased water flow can be critical to the rehabilitation method that can be used, and
thus the expense.
6.3.5 Overflow
In combined systems overflow or spillways are often installed to prevent the water level
in the drainage system from exceeding a certain level which protects areas from
flooding as well as ensures that only the designed volumes of water are conveyed
through the system. Overflow discharges across a weir to basin, outlet pipe or recipient.
To ensure the same hydraulic functionality at the overflow structure during higher
inflow and constant outflow, it will be required to increase the width of the weir crest or
lower the weir crest level. The latter will, however, have the unfortunate consequence
that the number of overflow increases.
To ensure the best hydraulic function of an overflow structure, i.e. ensuring that most
water flows through the structure without an increase in the backwater, the weir
structure can be equipped with movable weir, dynamically controlled crest level or a
moveable flap. This can also maximize the basin effect in the upstream drainage system.
In addition to the hydraulically justified augmentations, treatment measures are also

installed at overflow structures. Normally, automatically cleaned strainers or grids but
in some cases more extensive treatment measures, i.e. removal of nutrients and
sanitation. The development will certainly lead to such cleaning being more and more
prevalent, providing increased and better cleaning methods for use by local treatment. If
the discharged water is sufficiently cleaned, the cleaning can compensate for the
increased overflow volumes, so that the impact on the recipient is reduced despite the
increased overflow.

6.3.6 Basins
At many locations where it is chosen to reduce the hydraulic capacity in an outgoing
pipe, basins are built which can act as buffer in the drainage system. Basins are often
constructed with an overflow ensuring only overflows to the recipient at a chosen
frequency. The basins can both be designed to store the extreme peaks of the runoff, so
flooding is avoided or reduced, and they may also reduce the overflow to the recipient.
In combined systems the stored water is conveyed to the water treatment plants in the
normal way after a rainfall event. In separate systems, basins are usually used to smooth
runoff flow, but also add some treatment of the water before it is discharged into the
recipient.
Sizing of basins in combined systems can be made from the discharge capacity of the
basin and selected return period of overflow. There are formulas in Guide 26 to
determine the required volume of the basins, but it is recommended that updated rain
series are used and in addition the effect of climate change incorporated. A subsequent
calculation is undertaken with historical rain to verify the function of the basin.
Basins in separate storm water systems can often be engaged in recreational areas and
therefore have other functions than just smoothing the runoff. In this case the size of the
basins may be determined by the permissible water level variation, i.e. of requirements
to retention time limits. The retention time may not be too short because it gives too
little withdrawal of substance, and it must not be too long as it can cause excessive
algae growth in the basin/pond. This type of basins may also be recommended because
they are often very flexible to increased inflows partly because overflows due to the
location do not cause major damage.
6.3.7 Local storage

Wherever there is a possibility it will be a good idea to store rain water in extreme
situations. It should therefore be considered to place basins in as many small watersheds
as possible, i.e. in drainage of minor roads, parking lots, etc. Perhaps at some locations
the storm water inlets can be made so large that they can act as smaller basins during
extreme rainfall by reducing the outgoing pipe capacity. Developments such as these
can be performed when there is still rehabilitation undertaken, and can thus assist in
compensating for extreme rainfall beyond the level of service.
It could be considered at the planning stage that newly paved areas can serve multiple
purposes, so that planned activities in this area are not harmed by water depths of
approx. 5-10 cm in the area during extreme rainfall in a short period.
6.3.8 Control and regulation of drainage system

Drainage systems are dimensioned to handle a design rainfall and thereby meet
performance requirements. Since rainfall often falls unevenly across a catchment basin
and the capacity of the pipeline system is often varied, there may be a good opportunity
to improve the use of a drainage system by introducing dynamic control of certain
elements in the system, e.g. the outflow from the basins. This can contribute to both
reduced flooding and in combined systems reduced overflow to recipients. For drainage
systems with multiple basins, pumping stations, etc. it is strongly recommended to
investigate the potential for dynamic control. As part of augmentations in the system it
may be appropriate to examine whether control can allow for more appropriate solutions

to problems such as the storage basins or large pipe basins can be better placed in the
system.
6.3.9 Use of the road system

Normally, runoff from the roads is conveyed to the drainage system to avoid water or
aqua planning on the road. In some cases it may be considered to exploit the road
profile to convey water away during extreme rainfall. If the terrain is suitable and a
model can be assessed in terms of how the system will operate, then it may be an
excellent way to get the water transported from critical areas to suitable recipients or
storage options. The method can be recommended only to be used to separate storm
water systems and in situations where the design rainfall has been exceeded (i.e. in
emergency situations).
6.3.10 Augmentation on private property

Physical measures
It may be useful to encourage owners to collect and divert rain water on their own land
in order to avoid that the water is collected and hence requiring large drainage capacity.
Additionally, less effect on the water circulation in the area is achieved by infiltrating
the water locally. However, it requires that groundwater, soil and terrain conditions
fulfill certain conditions so that it is possible to divert the water locally, without
introducing local problems and damages.
If a parcel with a relatively small impervious area of 150 m2 is considered, this

corresponds to the fact that landowner must be able to store and dispose 7.5 m3 of
rainwater in a 5-year rainfall event (rainfall equivalent 50 mm), if there is no connection
of storm water to sewer. This amount equates to 30 standard rain barrels or a pond in
the grounds of 5 by 5 m and is 30 cm deep. If more water falls the owner needs to have
a management plan in place for handling this extra amount of water volume locally to
prevent flooding on his own or other people’s parcels.
Managing storm water on their own land without drainage to the combined drainage
system may therefore be recommended primarily for environmental reasons and in
order to recharge groundwater. When looking at the hydraulic balance, these
constructions are not the solution, but a complement to climate adaptation.
It is recommended not to base an adaptation solely on efforts by private landowners for

many reasons. The fact alone that it is not possible to control when landowners are
ready to disconnect their storm water system is reason enough not to rely on this method
from a hydraulic standpoint.
There are several methods that can be used by landowners if they want to reduce storm
water discharge to the public system:
Infiltration of rainwater
This refers to the diversion of rainwater into fascines on the site. Infiltration requires
adequate soil conditions. Fascines are often designed in a size equal to 20-30 mm of
rainfall, but no certainty exists that the complete capacity is available when the rainfall
starts.

If e.g. grass armor stone or similar surfaces are used in parking spaces, etc., a large part
of rainfall is infiltrated on site depending on the soil type. But in case of intense rainfall
water will run on the surface and be discharged to the drainage system.
Rainwater barrels
By collecting rainwater in rain barrels a reduction in the discharge to the drainage
system is obtained and water consumption is reduced in cases where the water replaces
the standard drinking water supply used e.g. for garden watering. The volume that can
be collected is often very limited, 200-500 l is often seen, and this is only a modest
proportion of the volume of extreme rain on a roof. Rainwater barrels can be full at the
start of the rainfall and therefore not reduce runoff at all.
Reuse of rainwater
Use of rainwater in homes as a substitute for water supply has only been implemented
in a few places, but has the same advantages as rain barrels and the further advantage
that consumption - as opposed to irrigation - is more evenly distributed over time. A
major drawback is also here that the storage capacity is limited and therefore there is no
guarantee that systems can store water during critical situations.
In Rørcentrets guidance "Use of rainwater" the use of a tank of 3 m3 for such plants is
recommended, and it is estimated that a tank like this will be able to store a large part of
the annual precipitation to be used in the dwelling for toilet flushing and washing
machines. In connection with extreme rainfall an overflow is required to divert water,
since a 5 year storm event alone requires at least 7.5 m3 for a single-family house. Re-
use of rainwater should be in regulatory guidelines (see EPA guidelines and Rørcenteret
guidelines).
Green roofs
Techniques have been developed for using the so-called green roofs, where a grid of
growth layers in which plants can grow, are laid out on the roofs of buildings. The aim
is to store the water fallen on the roof in the growing layer where it is absorbed by
plants. However, the storage ability of the growth layer is limited, only 6-10 mm, so the
storage effect during extreme rainfall is limited. However, on an annual basis a quite
good effect can be achieved in terms of reduced inflow to drainage systems.
Green roofs can therefore only be recommended for reasons of aesthetics and have very
limited effect on the hydraulic system in the event of rain.
As shown in the mentioned examples of actions on private land, it is hard to find

solutions for the public which are as safe and easy as discharging to the public drainage
system, and it is hard to find solutions that can handle the very critical periods of
extreme rainfall.
Private prevention of basement flooding

If a landowner needs to guard the basement against flooding from sewers non-return or
check valves can be installed so that water cannot flow backwards into the basement. If
the basement drainage system is connected to a pump wet well pumping to drainage
system, then even higher security is gained towards basement flooding. This also
ensures that the private installations can be used regardless of the water level in the
public drainage system.

Administrative actions against private property
There is an opportunity in the law that the municipality allows a private property which
diverts rainwater to the municipal drainage, to be disconnected to the drainage system
for all or part of runoff. It requires a voluntary agreement between the municipality and
the property owner. The municipality may reimburse connection charge for this storm
water drainage.
6.3.11 Drainage of road space

Drainage of road space works in most places very effectively. This is also the target
from the road authority's side, since water on the roadway constitutes a danger to traffic
and water in the road paving and road base layer may damage the road. Drainage of
especially smaller roads and streets, however, could perhaps be made so that the water
in a lesser degree were led directly to the drainage system, but first had to pass through
some retention devices such as infiltration devices. There is also scope for increased use
of semi-permeable pavements, where part of water on the road infiltrates through.
6.3.12 Examples of local handling of rainwater

The following are examples and illustrations of local handling of rainwater.
It is estimated that water in the urban environment will help to increase the recreational
value. Water storage in the urban environment will often require a landscape design that
involves green areas and thereby achieve positive side effects such as cooler urban areas
in summer and greater recreational value.
Projects in Odense and Copenhagen have shown that a solution based solely on local
infiltration is probably not sufficient, both because of increases in the groundwater table
and required space for creation. It will need additional measures such as storage of
rainwater in the terrain for which modeling is required. In Odense a contingency plan
has been established for handling rainwater at Sports Park.
Figure 6.10: Relief channel to reduce flooding at Sports Park in Odense, established in 2008.

The dimensioning of the channel was based on simulations with the surface model so
that both the size and location led to the desired effect. The channel has subsequently
been in operation for a number of rain events and has functioned as intended. Since it
was established there have been no damages to buildings and other infrastructure. The
intention is to extend the urban drainage system and establish alternative storm water
management in the area, so that the channel will increasingly become a part of the
solution which only comes into operation during rain event with a return period of
approx. 10 years and higher.

6.4 Flood emergency preparedness
Municipalities shall determine the desired standards of services which they will offer
citizens. This has to be done under conditions that are more extreme than the defined
desired standards of service which cannot prevent floods. However, it is possible to
minimize damage and inconvenience by increasing emergency preparedness. The level
of flood emergency preparedness needs to be balanced by the financial effort (see
Section 6.1 Risk).
Preparedness involves a wide range of activities and assessments that can protect assets
and people from damage caused by water. The contingency plans should of course
contain important phone numbers and other important administrative information, but in
this report only the hydraulic aspects of preparedness will be discussed.
Preparedness can be divided into before, during and after because the state of
emergency must be investigated and planned before it occurs, actions may be required
during the emergency situation and there will be an evaluation after the event where the
experience will be evaluated and possibly incorporated into new updated contingency
plans.
In preparedness. The possible components of flood preparedness is shown.
Figure 6.11: Illustration of a flood preparedness.
In the following sections the flood preparedness components are elaborated.

6.4.1 Before the rainfall event
Establishment of contingency plans if necessary for climate change adaptation

All municipalities must, as a part of the overall civilian preparedness, have a
contingency plan in place. There is currently no requirement for the municipalities to
develop a specific plan for operation of urban drainage systems and wastewater
treatment plants. Some municipalities have, however, made such plans which
accommodate a number of issues that are critical to the operation of the urban drainage
system, e.g. power failures, flood damages in exposed locations, staff/contractor
preparedness for emergencies that could maintain a minimum service level.
Contingency plans are plans which are used by municipalities to respond to overloads to
urban drainage system and water surcharges to terrain, and they include:
 Actual physical measures to reduce the effects of an extreme rainfall situation

and resulting floods such as earth embankments and walls designed to hold
water back in pre-defined depressions.
 Preparedness for emergency ad hoc efforts, i.e. placement of sandbags and use
of mobile pumps
 Information/alerts both internally within the municipality's operations and
externally
 Preparatory work must be undertaken where all details related to physical
measures, acute ad hoc efforts, information and alerts are reviewed.
Priority for contingency plans

Contingency plans should be available to all urban and possibly rural areas where it is
estimated that flooding may cause significant either human health related or costly
damages.
Once contingency plans have been established for a large area, e.g. a municipality or a
region, priorities for all catchment areas in the region should be set. This must be done
before a critical situation occurs, because there might not be sufficient personnel and
equipment available to implement the effort in all catchments at once. A prioritized
contingency plan would be a good decision support tool for incident management team.
The hierarchy of plans can be implemented using the same principles as the
prioritization of climate change adaptation.
Structural responses to flooding

In relation to obtaining a climate change service level for a catchment area, analysis and
detailed projects will uncover the critical issues within the area. The solutions that exist
could to some degree be expanded without significant additional costs. It may also be
required to be able to protect vulnerable housing areas using banks of earth or terrain
regulated through embankments or excavation so that water can be conveyed to less
critical areas.

Examples of permanent measures include embankment at Godsparken in Greve to
prevent a river from flowing into an urban area, and a gutter near the Sports Park in
Odense, which convey the water onto the running path to avoid damage to floors in
buildings, see Appendix C.
The embankment at Godsparken is not expensive in construction and ensures not only
against extreme long term rainfall, but also against extreme water levels in the ocean.
Figure 6.12: Establishment of embankment at Godsparken in Greve.
Mobile preparedness actions

Besides the stationary emergency response there is a wide range of possibilities for
mobile emergency measures, e.g. mobile pumps, sandbags and shutters. Partly through
terrain analysis, calculations and experiences a strategy can be created in advance for
how surface water is conveyed in an emergency situation and the necessary dimensions
for pumps and mobile walls can be assessed. It is essential that the number and precise
location of such sandbags is known and that everything is available in stock and ready
before the situation arises.

Gate Sandbags Online Measurement
Water discharged away

from residential areas
Figure 6.13: Sample proposals for mobile emergency response in combination with online measurement.
Figure 6.14: Mobile Pump used at Greve Gymnasium during extreme rainfall in Greve on 5th July 2010.
Warning
It is essential that the municipality and wastewater utility is warned about possible
adverse events that should be acted upon. Meanwhile, it is also appropriate that citizens
are warned that flooding is expected and adviced to secure personal belongings. .
A number of meteorological institutes forecast heavy rainfall events today, but the risk
of subsequent floods in cities are often based on experience. This is inadequate because
local conditions in the urban drainage systems determine if flooding occurs or not.
Alerts are currently used in selected locations abroad to reduce costs associated with
flooding. Can the urban drainage system for instance be partially emptied until rain

arrives, or traffic radio can be used to warn people to stay away from urban areas at risk
of flooding, cf. Chumchean et al. 2005, Rene 2011.
Some floods are acceptable if people are informed in a timely and appropriate manner
about how to behave. However, this requires that the municipality is in possession of an
appropriate action and contingency plan that can be executed when an extreme rainfall
is warned. If an analysis shows that there will be flooding in an area under future
climate change conditions which are not acceptable, then it will take some time from the
analysis is performed until new infrastructure is built. In this period a warning is useful.
When the warning comes into force, it is important that the wastewater utility has a
communication channel set up through which information to the citizens about the
measures affecting them can be communicated. Before the emergency situation occurs,
citizens must be aware of how to seek information: Website, radio or similar.
It is appropriate to have a contingency plan based on a warning of heavy rainfall for

viaducts or similar flood prone sites. Using the warning, such sites can be isolated
before the flood is so high that people are at risk if attempting to walk or drive through
the water. Whether the warning will be appropriate and economically viable must be
assessed on a case-by-case basis.
Responses to the flooding may depend on:
 Existing storage basins, canals, streams, rivers and lakes that can be drained
before the emergency situation arises, ensuring an optimum volume available in
the systems.
 How soon operational staff can be warned so they are ready to implement
contingency plans
 Existing grates, outlets, non-return valves, etc. are reviewed to ensure that they
are fully operational before the rainfall occurs.
Control and supervision

Control can be implemented in urban drainage systems if there is additional storage
available or long transport times present.
A control strategy could be developed when a sufficient understanding of how the urban
drainage system functions are available. The coupled models described in Chapter 4
may be useful to provide an overview of where it is appropriate to control the water, e.g.
by implementing gates or pumps. The analysis may have identified a critical area where
it may be useful to store water upstream by flooding less critical areas. By applying on-
line meters at strategic locations in the urban drainage system, gates or pumps can be
controlled by set points and be in operation at the right time. Additional storage can be
created by emptying existing ponds, canals, rivers and lakes before the situation arises.
Establishment of supervision of critical locations in the systems is highly relevant for

decision making during a flood situation. Thus it will be possible to implement
management of pumps and valves in a critical situation and to prioritize efforts.

A combination of warning and control can be used to ensure the required volume in the
system where it is expected to be most needed, see Appendix F.
6.4.2 During the rainfall event

During the extreme rainfall and the time just after (depending on rainfall character) the
urban drainage systems are monitored and the contingency plans are initiated when
required.
Capture of evidence and experience

To ensure that the entire organization will be wiser from the experience gained during
the incident, it is very important to conduct a detailed documentation of the incident.
The documentation should include at least the log of adjustments and operation actions
in urban drainage system (who has done what and when), and preferably include notes
of why and on what basis the action was implemented. Observations in the field are
very valuable (preferably with pictures) when experience should be used in further
analysis and possible in updating the contingency plans.
After a flood event very detailed knowledge of what exactly happened during the
incident is required.
Control
The control is implemented before the incident as outlined in the plans.
6.4.3 After the rainfall event

After the floods a clean-up of both the urban drainage systems and the terrain is
required. It must be ensured that facilities have not have damaged and that the function
and capacity have not been reduced by trapped items.
Updating contingency plans

Thorough documentation and experience of the flooding incident may be used to
evaluate whether the contingency plans need to be updated. This includes evaluation of
the prioritization of the plans, whether the hydraulic model requires recalibration after
the incident as well as finding solutions to the challenges or ensuring that service levels
are met.
Operating experience
It is valuable to compare experiences with expectations and conclude if flooding is
caused by operational problems or lack of capacity in the urban drainage systems.

7 SUMMARY AND CONCLUSION
The Climate Cookbook is an update of the previous edition from 2007 and contains the
status of knowledge on climate change that has been reported from the Danish state and
recommended for use when calculating the risk of flooding from the sea in the cities. In
regard to calculations of the effects of changes in precipitation, it is recommended to
use the recommendations in the Guidelines published by the Wastewater Committee of
the Danish Engineering Association.
Methods for analysis of flooding and flood risks in cities from both the rainfall as well
as the sea is described in the report. The methods are differentiated from the very simple
GIS method to the most advanced hydraulic model that includes several elements in the
hydraulics of the drainage system.
The calculation methods are used in many contexts to assess interventions against
flooding due to climate change. It includes examples from Greve and Odense.
The methods described in this report are generic and can thus be used when the existing
climate scenarios is updated with new estimates of precipitation and variation in sea
level.
Providing the level of desired standard of service on runoff from cities is very important
when deciding whether to analyse climate change and especially when making very
costly decisions about specific adaptation of drainage systems. The Climate Cookbook
contains a description of service on rainwater and water conditions and a proposal for
how this can be disseminated.
A challenge in management and prioritization of climate change is estimating damages

to society due to flooding. The principles behind the calculation of damages due to
floods is described along with methods to reduce damages. There are still some
unresolved issues, including pricing of damage, before it can be more than just an
enumeration of the various types of damage.
The methods described in the Climate Cookbook can be used as the basis for the
preparation of risk assessments and maps, providing decision support on many different
levels and to prioritize climate adaptation efforts. It includes a description of how to
implement a detailed risk analysis of floods, how to prioritize, and finally it provides
examples of what type of solutions can be deployed at the specific climate adaptation.
In recent years, floods in cities all over The World show that there is a very real need of
having a good contingency plan of rainwater when rainfalls exceed what the systems
can handle. The Climate Cookbook is a comprehensive approach to precipitation
emergency described and given examples of contingency operations "before, during and
after" extreme precipitation events.
Thus, the Climate Cookbook is a guide for both technicians to conduct the tests and the
basic hydraulic work that planners must establish to inform decision makers on climate
adaptation and work with the effort against flooding in cities.

The descriptions of the methods are supported by examples in Odense and Greve. The
Climate Cookbook shows that there is no justification for postponing the analysis of the
risk of flooding due to rainfall and sea level rise - on the contrary methods are available
for analysis of the challenges, so it is just getting started, so that conscious choices on
solutions to known challenges can be taken.

8 REFERENCES
Akademiet for de Tekniske Videnskaber (ATV), (2003)
Effekter af klimaændringer - tilpasninger i Danmark
ISBN: 87-7836-031-5.
Chumchean, S., Einfalt, T., Vibulsirikul, P., Mark, O. (2005)

To prevent floods in Bangkok: An operational radar and RTC application - Rainfall
forecasting
10th International Conference on Urban Drainage, Copenhagen, Denmark, August 2005
DANVA (2005)
Funktionspraksis for afløbssystemer under regn – Baggrundsrapport for Skrift 27
Domingo, N.D.S., Paludan, B., Madsen, H., Hansen, F., Sunyer, M., Mark, O. 2010.
Climate Change and Storm Surges: Assessing Impacts on Your Coastal City Though
MIKE FLOOD Modeling. 2010 MIKE by DHI Conference, Copenhagen, Denmark, 6-8
September 2010. DHI ref. 41/10
Domingo, N.D.S., Paludan, B., Hansen, F., Madsen, H., Sunyer, M., Mark,
O.”MODELING OF SEA LEVEL RISE AND SUBSEQUENT URBAN FLOODING
DUE TO CLIMATE CHANGES”. SimHydro 2010:Hydraulic modeling and
uncertainty, 2-4 June 2010, Sophia Antipolis
Domingo, N.D.S., Refsgaard, A., Mark, O., Paludan, B. “Flood analysis in mixed-urban
areas reflecting interactions with the complete water cycle through coupled hydrologic-
hydraulic modeling. IWA Publishing 2010, Water Science & Technology—WST | 62.6
| 2010.
Grum, M., Jørgensen A.T., Johansen, R.M. and Linde, J.J. (2005)
The Effect of Climate Change on Urban Drainage: An Evaluation Based on Regional
Climate Model Simulations
Intergovernmental Panel on Climate Change (IPCC) (2001)

Climate Change 2001: The Scientific Basis
Intergovernmental Panel on Climate Change (IPCC) (2007)

Climate Change 2007: The Physical Science Basis
Kystdirektoratet, (2008)
”Klimaændringers effekt på kysten”
Mark, O., Weesakul, S., Apirumanekul, C., Boonya-Aroonnet, S., Djordjević. S. (2004)
Potential and limitations of 1-D modelling of urban flooding
Journal of Hydrology
Mark, O., Djordjević, S., (2006)

While waiting for the next flood in your city…
7th International Conference on Hydroinformatics – Nice, France.
Miljøstyrelsen – Miljø- og Energiministeriet (2006)

Afløbssystemer under påvirkning af klimaændringer – Hovedrapport
Miljøprojekt nr. 1123, 2006
http://www.mst.dk/udgiv/publikationer/2006/87-7052-253-7/Html
Miljøstyrelsen – Miljø- og Energiministeriet 2006)

Katalog over tiltag til reduktion af effekten fra klimaændringer på afløbssystemer –
Tillægsrapport
Miljøprojekt nr. 1124, (2006)
http://www.mst.dk/udgiv/publikationer/2006/87-7052-255-3/html
Nascimento, N., Baptista, M., Silva, A., Machado, M. L., (2005)

Flood-damage curves: Methodological development for the Brazilian context
Paludan, B., Brink-Kjær, A., Nielsen, N.H., Linde, J. J., Jensen, L.N., Mark, O.
“Climate change management in drainage systems – A “Climate Cookbook” for
adapting to climate changes. Novatech, France 2010
Parkinson, J., Mark, O. (2005)

Urban Stormwater Management in Developing Countries
A book - 225 pages published by “The International Water Association” (IWA)
ISBN: 1843390574
PH-Consult (2003)
Hydrauliske undersøgelser af opland F7 i Odense
PH-Consult for Odense Vandselskab
Regeringen (2008)
Strategi for tilpasning til klimaændringer i Danmark
ISBN: 978-87-7844-720-3
Sonnenborg, T.O., Christensen, B.S.B., van Roosmalen, L., H.J. Henriksen (2006).
Klimaændringers betydning for vandkredsløbet i Danmark.
GEUS rapport.
Speight, L., (2006)

Analysis of the causes of flood risk in urban areas
The University of Newcastle upon Tyne
Spildevandskomiteen (1995)
Skrift 25 - Nedsivning af regnvand - dimensionering
Skrift 26 - Regional variation af ekstremregn i Danmark

Skrift 27 - Funktionspraksis for afløbssystemer under regn
Skrift 28 – Regional variation af ekstremregn i Danmark – ny bearbejdning (1979-
2005)
Skrift 29 – Forventede ændringer i ekstremregn som følge af klimaændringer
Sørensen, S., Adeler, O. F., Bentsen, L., (2005)

Risikoanalyse på vej ind i spildevandsbranchen
(NO-DOG-info 4/2005)
Teknologirådets rapport (2005)

Nyt klima – nyt liv
http://www.tekno.dk/subpage.php3?article=870&toppic=kategori7&language=dk
Thodsen, Hans (2007)

The effect of climate change on the transport of water, sediment and nutrients in Danish
rivers
PhD Thesis at Institute of Geography, University of Copenhagen
Vand og vejr om 100 år (2006)

Klimaforandringer og det danske vandmiljø
Redigeret af Morten Søndergård, Brian Kronvang, Morten Pejrup og Kaj Sand-Jensen.
Udgivet på Forlaget Hovedland 2006. ISBN 87-7739-889-0.

APPENDIX A
Odense – Sports Stadium

(Project implementation: 2007-2009)
Climate Cookbook – Appendix A C.1
© DHI 2012
C.1 Appendix C. Odense – Sports Stadium
C.1.1 Terrain regulation at Odense Stadium
The new sports centre of approx. 1,400m2 at Odense Stadium has been flooded three
times since its completion in 2005, most recently on 30-06-2007. Photos from one of
the flooding events are shown in Figure C.1.
Figure C.1 Map of the Odense Stadium area.
In the following a 2-D model is applied for simulation of surface run-off resulting from
water held back in the sewage system. Calculations are made on the flooding events
on 13th and 18th August 2006 together with a 100 years’ CDS rainfall, and calculations
with suggested solutions based on terrain adjustment are made.
If the rainfall event on 13th August is compared to the historical rainfall series (standard
national rain series) it is found that the rainfall event is T = 5-7 years for high
intensities, up to a variation of approx. 15 minutes. In this connection it should be
mentioned that precipitation in August 2006 was record-breaking. From 1st August to
13th August precipitation was 78.2 mm, of which 42 mm fell on 11th August. On 13th
August the drainage system in the Sports Stadium has obviously still been draining off
water from the event on 11th August. The water level in the Bolbro trench has probably
also been higher than normal, also due to a high groundwater level.
On 17th August 17.8 mm rain fell in 28 minutes. Highest intensity was 30 µm/s. The
rain was a 4-year rainfall event for 5-minute intensities, but a 2-year rainfall event for
10-minute intensities, i.e. a less intense rainfall than the one on 13th August, however, it
did result in damages in the form of flooding.
Drawing material of the drainage of the Sports Stadium has been received from
Odense Vandcenter and it shows the following, see figure c.2:
© DHI 2012
Drainage to the rain water system:
 The whole area around the sports hall drains surface water to the rain water pipe.
The new athletic facility probably also drains to the ø1000 rain water pipe.
 Ice stadium and skating hall discharge surface water to the ø1000 rain water pipe.
 The cabin house in the biking stadium also discharges to the ø1000 rain water pipe
via a ø300 pipe under Møllemarksvej
 According to municipal information four drainage pipes are connected to the rain
water pipe in the Sports Stadium. There are probably also a number of connections
upstream. Presumed drainage volume to the rain water pipe is 20-30 ha.
Drainage to combined sewers:

 A ø400 concrete combined sewer passes through the eastern part of the soccer
stadium. However, it is placed too high for drainage from the athletic and soccer
fields
 The bowling hall drains off surface water to the combined sewer in Møllemarksvej.
 The tunnel under the biking stadium can be emptied by means of a pump
discharging to the combined sewer in Møllemarksvej.
Figure C.2 Map of the Sports Stadium
© DHI 2012
Figure C.3 shows the sewer system around the Sports Stadium. The blue line is a
ø1000 rainwater pipeline discharging to the Bolbro trench. The pipe starts in Ole
Worms Gade and receives overflow water from the combined system in the wells
marked with a red circle (wells J50O025 and J51F16X).
Figure C.3 Sewer system in the Sports Stadium
Figure C.4 shows a MOUSE model and surface (elevation conditions) for the area
around the Sports Stadium.
Odense
Idrætspark
Figure C.4 MOUSE model and surface at Odense Sports Stadium
© DHI 2012
C.1.2 Design assumptions
The calculation tool MIKE FLOOD is used as a link module between the programs
MOUSE (sewer system), MIKE21 (surface 2D) and MIKE11 (streams etc.). The last-
mentioned can be left out and is not applied in the following. Configuration of these
programs is made individually. MOUSE and MIKE21 are executed simultaneously from
the link module MIKE FLOOD. A time interval of 2 seconds is applied, resulting in a
running velocity of the models, cf. surface extract figure C.4, of approx. 5-6 hours per
real hour (2.6 GHz Pentium).
Inlet hydrographs and the sewer system is calculated in MOUSE. The model applied
has been supplemented with drain pipes and other pipes in the area around the Sports
Stadium, among others drainage systems have been established under the football
stadium (Fionia Park) and the athletic field. This information stem from sewage plans
received.
For calculation of existing conditions precipitation data from SVK rain gauge 28181,
Bolbro Waterworks, have been applied due to its geographical location relative to the
Sports Stadium.
The terrain model is supplied in DVR90, and therefore all levels in the MOUSE model
has been converted from DNN to DVR90, cf. section 1.2.2.
Data basis for calculating surface flows is a so-called digital terrain model (DTM).
Contrary to DTH (digital height model) DTM does not include roads, houses, etc. The
DTM applied is based on a 5 m grid where the height uncertainty is 35 cm. The
reference system is System34/DVR90. Appendix A is an extract from the DTM model
in the Sports Stadium area. The figure shows the density of ground levels which form
the basis for the MIKE21 calculations.
The interpolation between height levels results in the terrain being levelled out, e.g. on
the soccer and athletic fields. Therefore, the terrain levels have been manually
adjusted to match the actual conditions.
The MIKE21 model is set up in a j,k-system with a chosen grid size of 4 metres rotated
50.650 relative to the MOUSE model, in order that among others relevant roads
(Møllemarksvej and Højstrupvej) are placed approximately parallel between the axis of
the j,k-system. This is the most appropriate method since the flux between the grid
points can only take place parallel to the j or the k axis. The surface covers an area of
1.448 m x 1.140 metres corresponding to 362 j-points and 285 k-points. The grid size
of 4 metres is chosen partly due to the solution of the digital terrain model, partly in
consideration of the calculation velocity.
Roads, houses and other sealed areas appear as shape files (ArcGIS). The shape files
are converted and exported to the terrain model in MIKE21. Roads and sealed areas
are digitised separately in ArcMap prior to export, since the items do not appear as
closed polygons (areas). Roads and other sealed areas are imported into the terrain
model deducting 20 cm, and houses are raised by 3 metres compared to the terrain
model.
© DHI 2012
Figure C.5 Surface of the area surrounding Odense Sports Stadium. The map is turned as the extract in
figure C.4
C.1.3 Status calculations

Initial calculations are made for the events on 13th and 17th August 2006 respectively
in order to verify the model compared to the observations and to show how the two
events resulted in damages in the Sports Stadium due to surface run-off. Furthermore,
a calculation of a scenario with 100 years’ rain (CDS) has been made.
On 13th August 40.6 mm precipitation within 4 hours and 50 minutes was registered by
rain gauge 28181 at Bolbro waterworks. 30-40 cm water was observed in the new
gymnasium, 10 cm water in the club rooms of the ice skating stadium, and 30-40 cm
water in the locker and storage room of the athletic field.
Figure C.6 shows a snapshot at 16:16 (GMT), where the water depth in front of the
entrance to the gymnasium is calculated to 35-40 cm (yellow colour). This is the max.
water depth in front of the gymnasium, and it corresponds more or less to the
observation mentioned above. The rainfall started at 11.11 (GMT), i.e. the max. water
level is found approx. 5 hours later.
It should be noted that the new sports hall (gymnasium) established in 2005 is not
included in the aerial photo, however, it has been added to the surface.
© DHI 2012
Figure C.6 13/8 2006. Status calculation with data from rain gauge 28181 Bolbro waterworks. Water
depths at 16:16 GMT are shown.
On 17th August 17.8 mm precipitation within 28 minutes was registered by rain gauge
28181 at Bolbro waterworks. The rainfall started at 12:23 (GMT). The result was
basically the same as on 13th August, however, the max. water depths were a little bit
smaller. Figure C.7 shows the situation at 13:03.
Figure C.7 17/8 2006. Status calculation with data from rain gauge 28181 Bolbro waterworks. Water
depths on 17th August at 13:03 GMT are shown.
© DHI 2012
A calculation with a 100 years’ CDS (Chicago Design Storm) rainfall with a duration of
2 hours has been made. This results in a max. intensity of 61.4 m/sec (3.7 mm/min)
and a rainfall depth of 48.9 mm, cf. figure c.8.
RAINFALL [my-m/s] CDS100år
60.0
55.0
50.0
45.0
40.0
35.0
30.0
25.0
20.0
15.0
10.0
5.0
0.0
00:00:00 00:16:00 00:32:00 00:44:00 01:00:00 01:16:00 01:28:00 01:44:00

1-1-1900
Figure C.8 100-years’ CDS rain with a storm duration of 2 hours
The rainfall results in water depths in the area in front of the sports stadium of up to 70
cm and water depths at the athletic field of approx. 50 cm.
Figure C.9 100 years’ CDS rainfall. Water depths 1 hour and 35 minutes after the rainfall started are
shown.
© DHI 2012
C.1.4 Suggested solution
The height difference between the floor level and the athletic field grass is approx. 2
metres. This could be used to establish a trench or a canal with an average gradient of
6.4 per mille and with outlet in a yet to be determined place in the area surrounding the
athletic field which is the lowest point in the area (level 6.85 DVR90). The distance from
the front side of the sports stadium to the athletic field grass is approx. 300 metres. In
this way the athletic field is used as storage capacity. After this the area is trained to
the Bolbro trench. Dikes must be established in some places to keep water away from
the buildings, among others the ramps at the ice skating stadium, the facilities at the
athletic field and other facilities where water may enter.
The property office of Odense municipality has prepared a sketch of above-mentioned

proposal for terrain regulation at Odense Sports Stadium, cf. figure C.10. The plan with
terrain regulation is to direct water to the lowest point in the area, i.e. the athletic
stadium. Flooding of the athletic stadium is not critical compared to the wooden floor in
the sports hall being destroyed by water.
Figure C.10 Odense municipality’s proposal for terrain regulation.
The proposal of the municipality increases the recurrence interval for flooding of the
sports hall floor from approx. 1-2 years to less frequent than every 5 years. The
proposal can be optimised by choosing a surface with less resistance (concrete
slapping), a more steep gradient at the start before the profile narrows at the corner
around the sports hall, and a wider profile along the wall of the sports hall towards
north-west. The final canal is shown in figure C.11. The canal has been in use twice
and has prevented flooding of the sports hall.
© DHI 2012
Figure C.11 Photos from the area after terrain regulation.
© DHI 2012
APPENDIX B
Climate Analysis for Greve Midt and Karlslunde
Climate Cookbook for DANVA – Appendix B D.1 Greve Forsyning, VandCenter Syd, PH-Consult and DHI
© DHI 2012
D.1 Appendix D. Climate Analysis for Greve Midt and Karlslunde
D.1.1 Introduction
On 1st August 2002 Greve centre received a rainfall event of 100 mm in three hours.
The incident caused damaging flooding at the city hall and high school. The rainfall
event is estimated to be an event that statistically occurs less than every 500 years.
On 3rd August 2002 the same area received another rainfall of 30 mm, which became
detrimental because all storage capacities and volumes within the municipality were
filled from the previous rainfall.
Since this incident Greve Municipality has focused on optimizing the operation of the
storm water systems and ensuring that the desired standard of services is met and
staying on top of problems before they occur despite the fact that it is beyond what is
required through the desired standard of services.
Greve Municipality has chosen to use many resources to improve the operation of the
storm water system when it became clear that it is not a question of whether there will
be extreme rainfall events, but rather when they will occur. It is indifferent to Greve Mu-
nicipality whether the more frequent extreme rainfall events are due to climate change
or not.
In August 2002, the centre of Greve was hit hard as the centre of the rainfall event was
just above Greve centre where the city hall and gymnasium are located.
At that time it was not clearly stated who was responsible for doing what, i.e. who
would be responsible for pumping when and where. In the hindsight it would have been
helpful to have identified who should do what in such an extreme rainfall event. The
centre of Greve is therefore used in this case, as it illustrates how the overloading of
drainage systems can be analysed.
The drainage system in the centre of Greve has a separate sewer system. All storm
water drains to a main storm water channel called Streget, which drains into the Bay of
Køge via a gravity outlet with invert level in Streget in elevation 0m DNN. Additionally to
this there is a combined sewer outlet and a pump which in 2005 was upgraded to pro-
vide 1 m3/s and is controlled by the water level in the Streget. In this case the focus is
solely on the separate storm water system.
The Greve Utility works currently on upgrading the storm water system in the centre of
Greve, so it fully meets the new desired standards of services for newly built drainage
systems. This case describes the solutions being applied to achieve functional re-
quirement and focuses on describing the damage that can be expected if no action is
taken on the system and what actions can alleviate some of the problems occurring
during extreme rainfall events.
During the period 2003 to 2007 Greve Municipality worked on getting a good under-
standing of the physical conditions in the drainage systems through the use of continu-
ous measuring systems, where water level, velocity and rainfall data were measured
simultaneously in different catchments. The measurements were used for calibration of
hydraulic models, analysis of the operational situation and to some extent to alert risks
of flooding. In this case, possible warning systems for Greve will be described.
© DHI 2012
The model used in this case is covering the storm water system upstream the high
school and has been calibrated using the water level, velocity and rainfall gauges. In
the case study urban climate cook book concept is applied.
D.1.2 Catchment description

The city area around the centre of Greve has a separate sewer system with a total area
of 520 ha divided into 11 catchments. The total impervious area of 260 ha drains to the
main storm water channel, Streget. Streget drains by gravity through an outfall pipe
and by pumping at the outlet to Køge Bay. The area is very densely populated. In the
area there are eight basins, 50 km storm water pipes and approx. 1,250 rainwater
manholes.
D.1.3 Future rainfall - Greve

According to DMI's climate forecasts we can expect during this century both a modified
annual distribution of precipitation and changes in rainfall intensities of the individual
rainfall events. A significant increase in extreme rainfall and rainfall intensities especial-
ly during summer is expected. The DMI (pt. "worst") climate scenario A2 (period: 2071
to 2100) predicts the annual precipitation to be slightly increased, and the extreme rain-
fall events during summer to be more frequent and more powerful. Summer rainfall is
expected in total to be slightly smaller.
At DTU's Department of Environment & Resources they have in some projects calcu-
lated the increase which climate models predict for extreme rain events (repetition pe-
riods of 5 and 10 years). For climate scenario A2 it is found that for these repetition pe-
riods increases in rainfall depth of approx. 20-50% can be expected. However, the
modelling does not show that a different intensity distribution of precipitation is fore-
seen. It is therefore believed that the increase in precipitation is best recognized by a
factor (climate factor) of rain as the ones used today. Due to the uncertainty about cli-
mate change it has been decided in Greve to use a climate factor of 1.2. In addition,
the MOUSE model for Greve is partially calibrated, and this is why a model uncertainty
of 20% is chose, i.e. the total factor on future CDS rain is 1.44. A current and a climate
projected 10-year CDS rain (including model uncertainty) for Greve seen in Figure D.1.
© DHI 2012
60
50
Regnintensitet (my meter/s)
40
I dags 10 års regn

30
10 års regn for klimascenarium A2
20
10
0
0 50 100 150 200 250 300 350 400
Tid (minutter)
Figure D.1 Present design and climate projected 10-year CDS rainfall for Greve. Time step = 1 minute.
Climate factor = 1.2. The climate projected rainfall is based on scenario A2.
© DHI 2012
D.1.4 Development in meeting functional requirements in Greve
Greve municipality is interested in analysing the development of how the sewage sys-
tem meets the present functional requirements under the influence of climate changes.
In order to prioritise initiatives made on the sewage system, the time-related develop-
ment in climate changes is simulated by linear extrapolation of a 10-year CDS rain and
the water level in Køge Bay for climate scenario A2. This results in the following simu-
lations:
Time horizon Expected sea level rise Rainfall

Today 0 cm 10-year CDS rainfall
In 10 years 3 cm 10-year CDS rainfall x 1.22
In 25 years 7.5 cm 10-year CDS rainfall x 1.26
In 100 years 50 cm 10-year CDS rainfall x 1.44
This means that problem identification is made first by running ordinary MOUSE pipe
calculations and identifying flooded places. Since the terrain is not described in the
most common MOUSE models, the indicated flooding depth is only indicative. The re-
sult of the four simulations is shown in Figures D.2 to D.5.
[m] FLOOD Meter - Maximum 10y6h_0%.PRF

130700.0
130600.0
130500.0
130400.0
130300.0
130200.0
130100.0
130000.0
129900.0
129800.0
129700.0
129600.0
129500.0
129400.0
129300.0
129200.0
129100.0
129000.0
128900.0
128800.0
128700.0
128600.0
128500.0
128400.0
128300.0
128200.0
128100.0
128000.0
127900.0
127800.0
127700.0
127600.0
-92500.0 -92000.0 -91500.0 -91000.0 -90500.0 -90000.0 -89500.0 -89000.0 -88500.0 -88000.0
[m]
Figure D.2 Indication of flooding depths i.e. places where the drainage system will run into problems to-
day for a 10-year rain. Input 10-year CDS rainfall.
© DHI 2012
130700.0
130600.0
130500.0
130400.0
130300.0
130200.0
130100.0
130000.0
129900.0
129800.0
129700.0
129600.0
129500.0
129400.0
129300.0
129200.0
129100.0
129000.0
128900.0
128800.0
128700.0
128600.0
128500.0
128400.0
128300.0
128200.0
128100.0
128000.0
127900.0
127800.0
127700.0
127600.0
-92500.0 -92000.0 -91500.0 -91000.0 -90500.0 -90000.0 -89500.0 -89000.0 -88500.0 -88000.0
[m]
Figure D.3 Indication of flooding depths, i.e. places where the sewage system in 10 years will have
problems for an extrapolated 10-year rainfall. Input: 10-year CDS rainfall and water level in
Køge Bay extrapolated for climate scenario A2.
130700.0
130600.0
130500.0
130400.0
130300.0
130200.0
130100.0
130000.0
129900.0
129800.0
129700.0
129600.0
129500.0
129400.0
129300.0
129200.0
129100.0
129000.0
128900.0
128800.0
128700.0
128600.0
128500.0
128400.0
128300.0
128200.0
128100.0
128000.0
127900.0
127800.0
127700.0
127600.0
-92500.0 -92000.0 -91500.0 -91000.0 -90500.0 -90000.0 -89500.0 -89000.0 -88500.0 -88000.0
[m]
© DHI 2012
130700.0
130600.0
130500.0
130400.0
130300.0
130200.0
130100.0
130000.0
129900.0
129800.0
129700.0
129600.0
129500.0
129400.0
129300.0
129200.0
129100.0
129000.0
128900.0
128800.0
128700.0
128600.0
128500.0
128400.0
128300.0
128200.0
128100.0
128000.0
127900.0
127800.0
127700.0
127600.0
-92500.0 -92000.0 -91500.0 -91000.0 -90500.0 -90000.0 -89500.0 -89000.0 -88500.0 -88000.0
[m]
After having made a problem identification, the extent of the flooding must be quantified
and damages assessed. For quantification of flooding a digital terrain model is set up to
be used as surface model for MOUSE/MIKE URBAN. To illustrate the method it has
been chosen to focus on the area at Greve High School and Greve Town Hall, i.e. the
two areas which were struck by flooding in 2002. As it appears from Figure D.2 a great
deal of flooding is expected in this area for a 10-year CDS rainfall today. The combined
MOUSE and hydrodynamic surface model (MIKE FLOOD) are used to calculate the
flooding in detail. Figure D.6 shows that flooding (of up to 33 cm at the high school) can
be expected in 10 years, if climate changes follow climate scenario A2. Please note the
uncertainty factor of 20%. Figure D.7 shows the calculated flooding depths for climate
scenario A2 in 100 years. It is seen that climate changes are now expected to cause
flooding of up to 83 cm at the high school. A comparison of the calculated distribution
of flooding for a 10-year rainfall at Greve High School in 10 years and 100 years re-
spectively is shown in Figures D.8 and D.9. It is seen that the development in a 10-year
rainfall and in sea water levels gives rise to more extensive flooding, which is expected
to further increase over time.
In the present situation (today’s 10-year rainfall) no safety factor is applied in the area,
since the model is calibrated for the present situation. For the future situation with cli-
mate changes and e.g. extension, a safety factor of 1.2 is applied.
© DHI 2012
Thus, the above calculation reflects the function of the sewage system in two fields:
1) Development in fulfilment of the functional requirement, and 2) Conditions under ex-
treme events where the design precipitation and water level in Køge Bay has been ex-
ceeded. It will be 2) which will form the basis of a decision-making as to how to react in
case of very extreme events where the sewage system cannot and is not expected to
be able to handle the rain water. A solution could be warning systems utilizing the ex-
isting systems more efficiently in order to minimize damages and lead the water to
places where it is less damaging, e.g. green areas.
Figure D.6 Max. flooding depths calculated for a 10-year CDS rainfall extrapolated by 10 years. Input:
10-year CDS rainfall and water level in Køge Bay extrapolated for climate scenario A2.
Today, Greve municipality has quite a lot of digitalised information on the infrastructure.
This information is among others found in the GIS system of the municipality, making it
possible to easily estimate the extent of damages under the designed conditions. Im-
portant information on infrastructure in relation to flooding could be: Supply boxes
(boxes placed at all cadastres); fiber point (junction box with optical fibres); stations
connecting supply boxes; buildings with basements; public buildings and environmen-
tally heavy industrial companies. Examples of comparisons of the municipal GIS data
with designed flooding are shown in Figure D.8. Among other things it is seen that the
flooding in the area around the town hall is expected to develop from a scenario with
quite a lot of water in the streets to a scenario with flooding of a child care centre. At
the same time it is seen that the service station will be flooded already in 10 years, and
therefore the consequences of a flooding like this should be taken into account.
© DHI 2012
F
Figure D.7 Max. Flooding depths calculated for a 10-year CDS rainfall, extrapolated by 100 years. In-
put: 10-year CDS rainfall and water level in Køge Bay extrapolated for climate scenario A2.
Figure D.8 Yellow/brown colour indicates flooding calculated for a 10-year CDS rainfall, extrapolated by
10 years. Rosy/pink colours show flooding calculated for a 10-year CDS rainfall, extrapolat-
ed by 100 years. Input: 10-year CDS rainfall and water level in Køge Bay extrapolated for
climate scenario A2.
© DHI 2012
D.1.5 Alternative solutions to reduce damages from flooding in Greve
Immediately behind the town hall and the high school there is a football field belonging
to the high school. During the flooding in 2002 this area was not flooded, because it is
at a higher level than the high school and town hall. Figure D.9 shows the football field.
Figure D.9 Football field behind Greve High School which could be used for storage of rain water during
extreme rainfall.
Since a mound has already been established around the football field, it was investi-
gated whether it is possible to close the mound completely and then use the football
field as rain water basin during extreme rainfall. If so, water from the sewage system
will be pumped to the football field. Two pumps were added to the sewage system
model, each with a capacity of 0.5 m3/s. Furthermore, in the terrain model 1.2 metres
have been digged from the whole surface of the football field in order to obtain extra
basin volume during extreme rainfall.
The result of the calculation shows that the extent and depths of the flooding is heavily
reduced. Figure D.10 shows that the max. water depth in the terrain outside the high
school has been reduced from approx. 83 cm to 17 cm. Furthermore, the extension of
flooding in the area has been heavily reduced, cf. Figure D.9. In this figure it is also
seen that the utilisation of the football field for storage of water has the additional effect
that the tank station and child care centre are no longer flooded in the climate scenario.
It should be noted, however, that the quality of the football field after a possible excava-
tion of 1.2 metres of soil has not been tested in the project. In order to secure the high
school in a situation as mentioned above, the curb in front of the school entrance must
be increased by approx. 20 cm.
© DHI 2012
3
Figure D.10 Blue/purple colour indicates flooding after installation of two 0.5 m /s pumps and utilisation
of the football field as rain water basin. The red colour shows the extent of flooding before
installation of the two pumps. Input: 10-year CDS rainfall and water level in Køge Bay ex-
trapolated for climate scenario A2.
© DHI 2012
3
Figure D.11 Max. water levels after installation of two 0.5 m /s pumps and utilisation of the football field
as rain water basin. Reference calculation is shown in Figure D.6. Input: 10-year CDS rain-
fall and water level in Køge Bay extrapolated for climate scenario A2.
D.1.6 Effects of climate changes on the water level in Køge Bay

The expected climate changes will result in increased water level in Køge Bay. As de-
scribed in section 7.2 the expected time-related development in the mean water level
will be as follows:
Time horizon Expected sea level rise

Today 0 cm
In 50 years 15 cm
In 100 years 50 cm
A relatively quick analysis of the ”direct effect” of increased sea water level on a coastal
municipality like e.g. Greve can be made with advantage. “Direct effect” means the
possibility of backwater from the sea to urban areas in the event of extreme high wa-
ters.
In Greve municipality a GIS analysis has been performed, where the terrain level at all
cadastres in the city have been compared to the existing extreme water level and the
future level (estimated to be 0.5 metres higher, which is probably a little lower than
what can be expected, since changed wind patterns will lead to increased storm surg-
es). Figure D.12 shows the recurrence intervals for the water level at Drogden Light-
house, which has turned out to be representative for Mosede Habour in Greve. It is
seen that a 10-year event corresponds to a water level of approx. 1.3 metres.
© DHI 2012
Figure D.12 Recurrence interval for max. water level at Drogden lighthouse, data from 1992-2002.
In the existing situation the GIS analysis shows that 165 cadastres are situated below
this water level and thus potentially threatened in the event of extreme high waters. If
the level is increased to 1.8 metres (1.3 + 0.5 metres) the number of potentially threat-
ened cadastres is increased to 1,179, cf. figure D.9. This analysis can be extended to
include also cadastres including houses with basements, levels to ground floor, etc.
© DHI 2012
Figure D.13 Number of cadastres lying below sea level today and under future climate changes due to
increase in extreme water level in Køge Bay.
Whether in the future there will be a convergence between extreme water level and ex-
treme rainfall cannot be predicted at present. However, in order to clarify conditions in
the sewage system it is a very relevant analysis which should be performed. Under the
existing conditions there is no convergence between extreme rainfall and extreme wa-
ter level in Køge Bay.
D.1.7 Summary
This example shows how you can extrapolate climate rainfall and analyse the sewage
system as to where problems may arise, and how you can quantify the extent of the
flooding and potential damages.
© DHI 2012

03-DHI - Urban Climate Change Guidelines

Uploaded by

Copyright:

Available Formats

03-DHI - Urban Climate Change Guidelines

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

03-DHI - Urban Climate Change Guidelines

Uploaded by

Copyright:

Available Formats

URBAN

HOW TO ACHIEVE SUSTAINABLE ADAPTATION

2 Status of knowledge on climate change _______________________________________10

4 Modelling of flooding from storm water system and rivers ________________________25

The Climate Cookbook – 2nd Edition i Greve Forsyning, VandCenter Syd,

6 Urban climate adaptation __________________________________________________55

7 Summary and conclusion ___________________________________________________79

A Odense – Sports Arena

The Climate Cookbook – 2nd Edition ii Greve Forsyning, VandCenter Syd,

The update contains guidelines for municipalities with respect to:

 Perform analysis and augmentation work of drainage systems in urbanized areas

Birgit Paludan, Greve Forsyning

The Climate Cookbook – 2nd Edition 1 Greve Forsyning, VandCenter Syd,

Climate change leads to on-going changes in dimension standards because of increased

The Climate Cookbook – 2nd Edition 2 Greve Forsyning, VandCenter Syd,

The Climate Cookbook – 2nd Edition 3 Greve Forsyning, VandCenter Syd,

Flooding assessments can be undertaken at different levels depending on the available

A comprehensive hydrological modeling approach may be required if the system gets

1.4 An Example – Legislation in Denmark

1.4.1 Sewerage plan

As a result of desired standards of services described in Wastewater Committee, Guide

The Climate Cookbook – 2nd Edition 4 Greve Forsyning, VandCenter Syd,

1.4.3 Contingency plan

Following a risk assessment of flooding event it may be desirable to have:

 Actual physical measures to reduce the effects of a flood situation.

 Preparedness for emergency ad hoc response.

 Information / alerts both internally within the municipality's operations and

The Climate Cookbook – 2nd Edition 5 Greve Forsyning, VandCenter Syd,

1.4.5 Risk assessment and analysis of extreme rain

1.5 The EU Flood Directive

The Climate Cookbook – 2nd Edition 6 Greve Forsyning, VandCenter Syd,

1.5.1 Preliminary flood risk assessment

a) Topographic map of appropriate scale which includes land use coverage.

1.5.2 Flood hazard and flood risk maps

a) low probability of flooding or extreme events.

For each scenario, the following information is shown:

Flood risk maps shall show:

a) the estimated number of inhabitants potentially affected.

The Climate Cookbook – 2nd Edition 7 Greve Forsyning, VandCenter Syd,

1.5.3 Flood risk management plans

1.5.4 Publication of maps, plans, etc.

1. Preliminary flood risk assessment – deadline 2nd December 2011

The Climate Cookbook – 2nd Edition 8 Greve Forsyning, VandCenter Syd,

The Climate Cookbook – 2nd Edition 9 Greve Forsyning, VandCenter Syd,

2.1 Scenarios for future climate change

The main features of these three climate scenarios are:

 A2 describes a heterogeneous world with local development, resulting in a

The Climate Cookbook – 2nd Edition 10 Greve Forsyning, VandCenter Syd,

The Climate Cookbook – 2nd Edition 11 Greve Forsyning, VandCenter Syd,

Climate Scenario A2 B2 EU2C

The Climate Cookbook – 2nd Edition 12 Greve Forsyning, VandCenter Syd,

2.2 Data base for the analysis of climate change impacts on

2.3 Future rain over Denmark

Based on DMI's modeling of the change in precipitation from 1960-1990 to 2070-2100

The Climate Cookbook – 2nd Edition 13 Greve Forsyning, VandCenter Syd,

Table 2.2Recommended climate change factors according to Guide 29.

Return Period 2 Year 10 Year 100 Year

Table 2.3: Expected increase in rainfall

Time Horizon Expected increase in rainfall

The Climate Cookbook – 2nd Edition 14 Greve Forsyning, VandCenter Syd,