TOWARDS THE DEVELOPMENT AND EVALUATION OF ADAPTIVE
FLOOD RISK MANAGEMENT STRATEGIES FOR THE RHINE
ESTUARY - DRECHTSTEDEN
J. Kind1, R.Vos2, A. Tijssen1,3, A. Jeuken1 and N. Slootjes1
1. Deltares, Utrecht/Delft, the Netherlands
2. Ministry of Infrastructure and the Environment, Rijkswaterstaat Agency, Lelystad, the Netherlands
3. The World Bank, Washington DC, USA (since February 2014)
ABSTRACT: Flood risk in the western part of the Netherlands will increase due to climate change,
socioeconomic development and subsidence if no measures are taken. The Delta Programme is looking
100 years ahead to develop a flood risk management strategy which, in light of future uncertainties, is
both flexible and robust. Different possible strategies have already been studied in previous phases of the
program and some choices have already been made. The preferential strategy will be a strategy which
builds on the present system of strong dikes and flexible barriers. Within this strategy, opportunities exist
to replace part of the necessary dike reinforcements with room for the river measures. In this paper we
describe and apply a dynamic framework for the development and subsequent economic evaluation of
Key Words: Flood Risk, Adaptive Management, Cost-benefit Analysis, Decision Support, Tools
The western part of the Netherlands is a densely populated delta with high economic value and is home
to one of the world’s largest international harbours, the port of Rotterdam. The area is low lying and
vulnerable to flooding from both rivers and the sea. It is expected that the vulnerability will increase due to
the (combined) effects of sea level rise, higher river discharges, soil subsidence and socioeconomic
The problem of increasing flood risk is addressed in the National Delta Programme (2009-2015). This
entails multi-level collaboration between the national government, provinces, water boards and
municipalities. Five national ‘Delta Decisions’ are being prepared, including decisions on flood protection
standards (Van der Most et al., 2014; van Alphen, 2014; Eijgenraam et al., 2014; Kind, 2013), fresh water
supply, and spatial planning and adaptation. In six regional sub-programmes, integrated flood risk
management strategies are further developed involving regional and local stakeholders. Those strategies
contain the proposed specific measures that apply for implementation in the short, medium or long term.
This paper is focussed on one of the regional sub-programmes:the Delta Programme Rhine Estuary–
Drechtsteden (DPRD). This sub-programme is closely related to the Delta sub-programme for the Rivers
(Schielen and van der Aarsen, 2014).
1.1 Adaptive Planning in the Delta Programme
Within the Delta Programme one of the largest challenges is dealing with uncertainties about the future
climate, population, economy and society. The Delta Programme tries to handle uncertainty by an
adaptive way of planning, i.e. maximizing flexibility, keep options open and avoid “lock in”. Even so,
scenarios are used to assess whether the measures sort the same effect under different conditions.
1.2 Work done so far
The Delta Programme was initiated in 2009 after the advice of the (second) Delta Committee
(Deltacommisie 2008) and distinguishes several phases. The first phase (2009-2010) contained the initial
problem analysis. The purposes was to assess the plausible future changes in flood risk and fresh water
supply and demand, driven by factors such as climate change and socioeconomic developments. The
problem analysis indicated that water level could rise in the area with up to one meter by the end of the
century and potential flood damages would increase significantly if nothing would be done
(Deltaprogramma Rijnmond-Drechtsteden, 2012). The problem analysis formed the basis to develop flood
risk management strategies for the Rhine Estuary – Drechtsteden area.
In the second phase (2010-2011), ‘first generation’ strategies were investigated. Those strategies
indicated the ‘edges of the playing field’ and included large infrastructural works which changed the areas’
entire flood protection and fresh water supply system. First generation strategies ranged from a fully
closed Rotterdam area, with dams and sluices which ships had to pass and river discharge which had to
be diverted to the southern branches, to a fully open river-sea strategy relying solely on the protection by
strong dikes. In Jeuken et al. (2013b) it was concluded that those strategies were either economically
inefficient or had large negative environmental consequences. The reference strategy, which in essence
is a continuation of the existing system where flood protection is maintained with dikes, room for the river
projects and flexible storm surge barriers (amongst which the Maeslantbarrier), showed to be the
preferred ‘first generation’ strategy.
More recently (2011-2013), the idea of changing the river Rhine discharge distribution over its main
branches has been studied (van Rhee, 2013). One of those Rhine branches flows into the northern Lake
IJssel instead of the south-western delta. Changing the distribution could therefore improve the flood
protection in the Rhine-Meuse estuary. The studies, however, gave ambiguous results and the national
net benefit of changing the distribution discharge remained far from clear. In combination with the
unknown morphological risk of altering the discharge distribution, this led to the conclusion that a change
is at present not wanted.
In the last phase (2013-2014), the preferential flood risk management strategy was developed, taking the
most promising measures studied in all previous phases as potential “building blocks”. In the preferential
strategy, the existing system with flexible storm surge barriers and strong dikes is maintained. This
strategy is further assembled using the following main elements:
dike improvements necessary to comply with existing flood protection standards;
dike improvements needed to comply with updated flood protection standards;
dike improvements to compensate for rising water levels and subsiding land;
measures to lower the water levels (water retention and room for the river); and
measures which reduce the potential consequences of floods.
Cost-benefit analysis was used to derive an initial economically efficient flood risk management strategy
taking the cost of the measures and the benefits of flood risk reduction into account. By also considering
the environmental benefits and benefits due to an increase in spatial quality, as well as the preferences
from regional stakeholders, the final preferential strategy is now being developed by the program.
1.3 This paper
In this paper we illustrate the derivation of the economically efficient flood risk management strategy
(short: efficient strategy). The challenge we face is in finding a robust optimal combination and sequence
of flood risk measures, interlinking the short, medium and long term, while complying with updated flood
protection standards (that are still under discussion), and taking into account uncertain climate change
and socioeconomic development. We present (yet unfinished) work on the development of an evaluation
tool (the “Planning Kit DPRD”) and discuss some of the first results we obtained from analysing several
strategies. The approach we took suggests further (maybe academic, probably labour intensive) work to
extend this type of analyses with no-regret or real option analysis. Since we fully deployed our modelling
framework only in the last phase of the sub-programme, we expect that applying it to other delta regions,
which are in an earlier phase of policy formulation and where more strategic options are still being
considered, will yield additional valuable results to refine the approach.
2. DEVELOPING ADAPTIVE STRATEGIES
The challenge of finding a robust and optimal combination and sequence of interventions linking short
and long term agendas is also part of the challenge of adaptive planning and adaptive management. We
provide a brief overview of the current literature on the subject and then shift to the application in the
Rhine Estuary - Drechtsteden area.
2.1 Adaptive planning and management
According to Sayers et al. (2012), central elements of adaptive planning are: “responses to changes that
are effective under the widest set of all plausible future scenarios; responses do not foreclose future
options or unnecessarily constrain future choice; relevant changes are foreseen through targeted
monitoring and scenarios of the future are continuously being reassessed; and policies, strategies and
structure plans are appropriately redefined”. A general procedure to make an adaptive plan and comply
with the aims as described by Sayers et al. (2012) is given by the ‘Dynamic Adaptive Policy Pathways’
(DAPP) framework (Haasnoot et al., 2013) that combines the approach of adaptive policy making (APM)
(Kwakkel and Walker 2010) and adaptation pathways (AP) (Haasnoot et al., 2012a).
Adaptive delta management (ADM) is the adaptive planning approach chosen by the Delta Programme
for dealing with uncertainties to support the decision making process. It tries to link decision making now
to options for adaptation in the future (van Rhee, 2012). Central elements in ADM are the use of so called
Delta scenarios (Bruggeman et al., 2011), the assessment of key vulnerabilities by looking for adaptation
tipping points (Kwadijk et al., 2010), the use of adaptation pathways (Haasnoot et al., 2012) to envision
the portfolio of options, its effectiveness and its flexibility, and a systematic regional multi-stakeholder
approach to combine challenges across different agendas (water management, water policy and spatial
The adaptation tipping points (ATP’s) in combination with the Delta scenarios define the time range within
which the current water management strategy will not meet its objectives anymore, which is a trigger to
adjust the strategy or switch to a new strategy. In addition, opportunities may arise from other agendas to
shift necessary investments to an earlier timeframe (Koukoui et al., 2013). ATP’s or opportunities not
necessarily define the economic optimal moments for investments.
The idea of ADM, which has a lot of similarities with the Thames Estuary project in the UK (Reeder and
Ranger, 2012; Jeuken and Reeder, 2011), has been previously applied within the Delta Programme
Rhine Estuary-Drechtsteden (Van der Brugge et al., 2012). For this area the most dominant ATP’s are the
end of life time of the major storm surge barrier the ‘Maeslantbarrier’ and the limitation of freshwater
supply due to increasing salinization of the fresh water inlet near the city of Gouda (Kwadijk et al., 2010).
In addition it is clear that increasing sea level rise, local subsidence and higher river discharges, together
with socioeconomic growth, will increase the flood risks in the region if the protection level is not adjusted
or other measures are not taken (Jeuken et al., 2013b). The latter is difficult to express as an ATP related
to current management practice, since the existing policy already consists of a regular (6 yearly)
assessment and improvement of the dike system based on clear targets that should be met according to
the Water Act. Whenever a dike does not meet these targets, it is programmed for improvement in the
national Flood Risk Management Programme.
2.2 Adaptive planning and the preferential strategy
Considering the choices that have already been made in the previous phase in the Delta Programme, one
of the remaining challenges for the regional sub-programmes is to find out how to minimize the costs of
flood protection over a longer time period of 50 to 100 years ahead. and by doing that also widening the
scope by considering more options than dike improvement alone. Instead of waiting for local thresholds to
be exceeded, the Delta Programme therefore is looking forward in its anticipating strategies for
optimization of investment paths in time and in combination of different kinds of measures. Figure 1
shows the different measures that can be used to assemble the potential strategies in more detail.
Figure 1: Different elements of the potential strategies with indicative time window according to two
of the Delta Scenarios (Steam and Rest). Each colour represents a different category of measures. From
top to bottom: measures to reduce the consequences of floods, measures to protect unembanked areas
(purple), measures related to the storm surge barrier Maeslantbarrier (pink), measures related to ‘making
room for rivers’ (dark green). Measures for dike-improvement (blue) are divided into three categories.
Improvements are needed because dikes (i) are not meeting existing standards, (ii) are not meeting new
standards or (iii) have to be improved to cope with future water levels due to climate change or soil
subsidence. Measures indicated with *) are not assessed within our framework, those have either already
been rejected or have little interference with the other elements in the strategy.
To cope with increasing climate change, more and larger measures can be applied. This can be viewed
as the final portfolio of options from ADM. However, it is not a portfolio of options one can simply choose
from. This has to do with:
-the area is so large that different measures in different areas are needed to reach the targets set by
the programme. For instance, applying room for the river solutions is only reaching part of the targets
set by the programme because the size and impact area is limited. However, the impact areas of
different measures, do overlap;
-the effects of the different measures is different. Some measures increase the strength of a dike,
some the height, some decrease the water levels and some reduce the consequences of floods.
Especially the strength of the dikes (piping) is a problem in parts of the area, which can be repaired
by dike improvements only;
-there are good opportunities to combine the three challenges for dike improvement (mainstreaming
agendas). It is in general much cheaper to add a meter once (to comply both with existing and new
standards and to compensate climate change) than come back three times in a couple of decennia.
The only drawback could be that since the climate signal is uncertain, there is a risk of overinvesting,
since improvements in dikes which do not meet current standards, have to start in early years.
The bottom-line of the preferential strategy is thus a strategy of dike improvements that may or may not
be diversified with other measures in particular parts of the area. The earlier mentioned ATP for the storm
surge barrier is determining a decision point for replacing the barrier, or deciding differently. However,
after the phase of the first generation strategies, it has been decided to develop strategies for the
following phases in which the Maeslantbarrier is replaced.
Table 1 presents a short overview of the water level reducing measures which have been used as
building blocks and as such as partial alternatives for dike reinforcements within the preferential strategy.
Table 1 Summary of the final set of potential options other than dike improvements
Map Code Description
1MLK Maeslantbarrier Replace Maeslantbarrier
2 Grev Water storage lake
Grevelingen during high river
In 2015 the Lake Volkerak-Zoommeer is prepared for temporary water storage
when exceptional conditions result in the combination of a closed storm surge
barrier and high river discharge. In view of the expected climate change, it is
being considered to expand the storage capacity of Lake Volkerak with Lake
3 Chan Channel deepening
Boven and Nieuwe Merwede Channel deepening is also a measure to lower the water levels near the bottle
neck in the River Merwede. The deepening of the channel is located in the
Boven Merwede and Nieuwe Merwede.
4RftR1 Room for the River package 1 In the eastern part of the area is a bottleneck in the river Merwede during high
river discharges. This causes backwater effects. The three Room for the River
building blocks each consists of one or multiple measures with the aim to
widen the riverbed to lower the water levels. Three different packages were
developed. RftR1 and RftR2 can be deployed alone or in combination. RftR3
is a subset of RftR2 and can be deployed alone or in combination with RftR1,
but not in combination with RftR2
RftR2 Room for the River package 2
RftR3 Room for the River package 3
3. THE PLANNING KIT, A TOOL FOR EVALUATION
The Planning Kit DPRD was developed as a decision support tool to evaluate a wide variety of flood risk
management strategies, wider than the set of strategies finally evaluated since also measures which
reduce the flood consequences (damages and/or casualties) are incorporated. In the Dutch situation,
however, due to the existing high flood protection standards, measures which reduce flood consequences
are in general not cost efficient; they are therefore no longer considered as important building block.
The tool is implemented in the C# programming language and consists of a database, a calculation
module, a user interface and a post-processing module that transforms the results in the desired output
(Kind et al., in prep). The tool determines nominal and present values of all costs of measures and the
present value of the cost of expected flood damages.
The user can choose for the calculation the (new) flood protection standard, climate change and
socioeconomic development according to one of the four Delta scenarios, and the selection of other
measures which lower the water levels or reduce the consequences of a flood. Those other measures
can be assumed for implementation at any year in the period 2017-2100.
The flood protection standard is expressed as a flood probability per year (Van der Most et al. 2014).
Those can also be expressed in terms of design water levels - the water level the dike has to minimal
withstand in order to provide the required level of flood protection. In the Rhine Estuary – Drechtsteden
area, we have distinguished more than 200 dike stretches, with an average length of 3 kilometer. For
each dike, hydraulic information is provided in the database concerning the actual height, the water levels
for different return periods, water level rises due to climate change, soil subsidence, and so on. Also
information on the cost of improving the dikes, in steps of 10 centimeters up to 2 meters, is included in the
Dike reinforcements are considered when its height is lower than the design water level, causing water to
flow over the dike after which the dike eventually breaches and flooding occurs. This is valid under the
assumption that the dikes comply with existing design rules, which state that dikes should be strong
enough to withstand water up to the crest, in which case the dominant failure mechanism of the dike is
“overflow and overtopping”. However, we should note that recent research has indicated that a
substantial number of dikes in the Netherlands are not strong enough to withstand water up to the crest
and have (severe) ‘piping’ problems (Jongejan et al., 2011). For the dike sections where the piping
problem is known, the Planning Kit addresses the piping problem by strengthening the dikes, for which
also costs are included in the database. After addressing the piping problem, the dikes comply with
existing design rules and it is safe to assume that “overflow and overtopping” is the dominant failure
mechanism. Now we can reduce the probability of flooding by heightening the dikes or lowering the
(design) water levels.
Due to climate change, the water levels will rise over time. At the same time, subsidence is responsible
for a gradual decline in dike height. If no action is taken, in some future year the dikes will be too low and
will have to be heightened. Figure 2 shows as an example for one dike section the development in time
of the design water level (blue line) and the dike height (continuous green line).
Figure 2: Dike reinforcement scheme, taking into account water level rise due to climate change and
decline of dike height due to soil subsidence
The Planning Kit computes on an annual basis the design water level and dike height, taking into account
the effects of climate change and soil subsidence. For each year, the dike has to withstand the dike
height test. When a dike fails this test, the dike is heightened after a lead time of 15 years. We have
incorporated this lead time to accommodate for the time laps between rejection and reinforcement of a
dike, due to policy decisions, design of the reinforcements, procurements, etc. A period of around 15
years is also in line with todays’ practices.
The amount of dike heightening is determined following the current design practices in the Netherlands
(Ministerie van Verkeer en Waterstaat & Expertise Netwerk Waterkeren, 2007). To determine the amount
of necessary dike heightening, a design horizon of 50 years is used, starting from the year of rejection.
This means that after heightening, the dike has to withstand the dike height test for 50 years after the
rejection, on basis of projections for climate change, soil subsidence and - if agreed upon - planned future
measures which will lower the design water levels. An additional allowance for robustness of 30
centimeter in the design of the dike is added. The dashed green line in Figure 2 shows the amount of dike
heightening based on the design horizon. The continuous green line shows the added dike height due to
the allowance for robustness. Due to this allowance, the dike is expected to fail the test again only after
the year 2100.
To lower the design water levels in the area, the design water level reducing measures described in
section 2 can be selected in the Planning Kit, in any year between 2017 and 2100 and in (almost) any
combination. Those measures will lower the design water level, causing the dike to pass the dike height
test longer, and hence investments for dike reinforcement can be postponed. (In fact, the two small drops
of the design water level in 2030 and 2050 in Figure 2 are caused by two assumed measures located far
away from dike, causing it to have a minimal but visible effect). The postponement of dike investments
ultimately translates in a lower present value of the investment cost; this makes up an important part of
the benefits of the measures which reduce the water levels.
The Planning Kit also computes the expected flood damages in the area on an annual basis. This does
not only include property damages, but also damages due to business interruption, indirect damages, and
the damages due to casualties, following the approach of Kind (2013), Eijgenraam et al. (2014) and
Bokarjova et al. (2012). The Planning Kit contains information on damages and casualties based on in
total about 110 inundation scenarios (Kind, 2013 and Eijgenraam et al., 2014). The damages increase on
an annual basis depends on the Delta scenario chosen for the calculation (e.g., Steam or Rest).
The potential damages are combined with the flood probabilities of the dike. The flood probabilities are
computed on basis of the projected surplus or deficit in dike height in reference to the design water level.
Without measures, the probability of flooding, and therefore also the expected annual damage, will
increase over time due to climate change and subsidence. Investments in heightening the dikes or
lowering the design water levels will decrease the expected annual damage. Those are the second part of
The Planning Kit calculates the present values of all costs of measures and of expected damages. For
discounting costs and benefits, we use the 5.5% real discount rate per year, as prescribed by the Dutch
government. Although we are only interested in developing and evaluating strategies for the period till the
year 2100, we use a time horizon of 200 years (2015-2215). By using such long time horizon, there is no
need to worry about salvage values of measures for the remaining useful lifetimes after the year 2100,
which may affect the outcome of the analysis.
4. EVALUATING THE STRATEGIES
Given the constraints imposed by the policy decisions already taken in the previous phases of the
regional sub-programme, notably that (i) the ‘backbone’ of the preferential strategy for the Rhine Estuary
– Drechtsteden area is the continuation of the existing strategy with a flexible storm surge barrier, and (ii)
no change in the discharge distribution over the river Rhine branches is warranted, the preferential
strategy will in essence be a strategy based on further dike reinforcements. A number of measures which
lower the water levels are however available as (partial) alternatives for dike heightening, either alone or
in combination. Economic evaluation can help decision making by answering the following questions:
1. which water level reducing measures are cost efficient?;
2. when should those measures be considered for implementation?; and
3. what is an optimal combination of measures, sequence and timing?
Determining the optimal combination, sequence and time frame for implementing the measures is both
relevant and challenging, because (considering only one Delta scenario):
a) some of dikes have to be reinforced in the short term because of the existing problems with piping;
opportunities to combine the investments with dike investments needed for to climate change or
higher flood protection standards exist leading to substantial synergies and cost savings (hence:
choose dikes and postpone water level reducing measures);
b) other dikes are strong enough and have excessive height (hence: postpone water level reducing
c) still other do not suffer from piping problems and have no excessive height, those will soon have to
be reinforced due to climate change and/or increased protection standards, unless the water level is
lowered by other means (hence: speed up water level reducing measures);
d) the water level reducing measures will affect multiple dike segments characterized under (a), (b) or
(c) (hence: it is complicated to see what an optimal timing for alternative measure is); and
e) part of the effect area of the measures overlap (hence: system approach is needed).
If we take the five water level reducing measures from Table 1 and if we consider for each of them four
time frames for implementation – say short term (2030), medium term (2050), long term (2070), or not at
all - this would already give 1024 possible combinations of measures in time (‘strategies’). Such
evaluation calls for batch or optimization functionality in the Planning Kit, which we did not develop. The
actual relevant number of combinations is, however, much smaller, for example since the Room for the
River-3 package cannot be combined with the Room for the River -2 package, or due to the limited
overlap between the impact area of the retention measure in the Grevelingen and the other water level
reducing measures. Within our project, we have chosen to evaluate only 16 strategies, which are
schematically presented in the Figure 3.
Figure 3: Evaluated strategies, Steam scenario
We evaluated the strategies using two of the four Delta scenarios, Steam and Rest. In the Steam
scenario, high climate change is combined with high economic growth; the rise in future flood risk is
highest. The Rest scenario is at the other end of the spectrum – low climate change and low economic
growth. Strategies which perform well under all scenarios are often called “no-regret”, while strategies
which perform well under only few (normally high climate change and high socioeconomic growth)
scenarios do imply potential regret and may be considered for implementation in a more temporized
manner. We will illustrate the results for the Steam scenario.
Table 2 gives the present values of the costs of the measures and the cost of expected damages for the
different strategies. The strategy with the lowest total cost is the most efficient strategy.
Table 2: Summary of strategies and present values of costs and expected damages
No. Measures of the strategy with assumed year
of implementation Costs, in Million Euro Present Value, 2013 prices
Grev RftR1 RftR2 RftR3 Chan Cost of
Dikes Cost of
S1 1771 58 1763 3592
S2 2030 1730 184 1709 3623
S3 2050 1732 101 1752 3585
S4 2070 1767 73 1752 3591
S5 2030 1737 314 1749 3800
S6 2050 1742 146 1767 3655
S7 2070 1771 88 1762 3620
S8 2030 2030 1710 552 1740 4003
S9 2030 2050 1726 396 1750 3873
S10 2030 2070 1737 342 1748 3826
S11 2030 2030 1686 334 1749 3768
S12 2030 2050 1723 321 1750 3794
S13 2030 2070 1736 317 1747 3800
S14 2030 1734 125 1742 3601
S15 2050 1740 81 1765 3586
S16 2070 1771 66 1759 3596
Total costs range between 3.6 and 4.0 billion euro present value (nominal between 6.3 and 7.3 billion
euro). The relative small difference is because in each strategy the majority of the dike reinforcements are
needed to meet existing and future flood protection standards and to compensate for climate change and
soil subsidence. Strategies S1 (reference strategy with no water level reducing measures), S3
(Grevelingen in 2050), S4 (Grevelingen in 2070) and S15 (Channel deepening in 2050) are the strategies
with the lowest total cost.
On basis of the total costs of the strategies we are able to analyze the costs and benefits of individual
water level reducing measures. We distinguish two types of benefits of the measures, which we derive
from a comparison with a reference strategy. First is the reduction in the present value of the cost of dike
reinforcements. This benefit arises since investments in dikes are postponed as a consequence of water
level reducing measures. The second benefit is the reduction in the present value of the expected
damages. This benefit arises only when measures are implemented relatively early, before a dike which
has a positive water level reducing effect from the measure has failed the height test. This will lead to
temporarily excesses (above the legal standard) in flood protection for certain dikes, and hence to a
further reduction of the expected damages. Having defined those two types of benefits, we calculate
benefit-cost ratios by dividing the total of the two types of benefits by the cost of the measures, see Table
Table 3: Differences in present values of costs and damages compared to a reference situation
(in million euro), and benefit-cost ratios. Steam scenario.
1 2 3 4 5 6 7
No. Reference Cost differences between Strategy and Reference B/C=
S2 S1 -41 126 -54 31 0.75
S3 S1 -39 43 -11 -7 1.17
S4 S1 -5 15 -11 -1 1.10
S5 S1 -35 257 -14 208 0.19
S6 S1 -29 88 4 63 0.29
S7 S1 -1 30 -1 28 0.07
S8 S5 -26 238 -9 202 0.15
S9 S5 -10 82 1 72 0.11
S10 S5 0 28 -2 26 0.06
S11 S5 -51 20 -1 -32 2.63
S12 S5 -13 7 1 -6 1.85
S13 S5 0 2 -3 -1 1.24
S14 S1 -37 67 -21 9 0.87
S15 S1 -31 23 2 -6 1.26
S16 S1 -1 8 -4 4 0.54
Note that the Room for the River 2 and 3 packages (S8-S13) have been analyzed against a reference
including the Room for the River 1 package (S5), and all other strategies have been analyzed against a
“dikes only” strategy (S1).
Table 3 gives a good indication which measures are economically efficient and also in which period.
Retention in the Grevelingen basin is profitable at the margin, with a benefit-cost ratio of 0.75 in 2030
(S2), 1.17 in 2050 (S3) and 1.10 in 2070 (S4). Most dikes which benefit from retention in the Grevelingen
do not suffer from piping problems and have initially surplus height, hence waiting for climate change to
occur is a good strategy.1 Room for the River 1 and 2 are not efficient in the short, medium or long run
(S5-S10). Room for the River 3, however, which has only been analyzed in combination with 1, shows to
be very efficient (S11-S13). Channel deepening of the Merwede is a profitable measure in the medium
term (2050, S15). If we wait too long (2070, S16), the measure will not lead to a reduction in the cost of
the initial round of dike increases but will come too early to reduce the cost of the second round of dike
increases (after the year 2100, due to the 30 cm allowance for robustness).
The results suggest an optimal strategy for the Steam scenario, with retention in the Grevelingen and
channel deepening in 2050 (S3 and S15), and a small Room for the River 3 package in 2030 (on basis of
S11). Analyzing this strategy with the Planning Kit results in total costs of 3569 million euro, which is
lower than the total costs of all other strategies in Table 2.
5. DISCUSSION AND CONCLUSION
In this paper, we have described our approach on the economic evaluation of adaptive flood risk
management strategies for the Delta Programme Rhine Estuary - Drechtsteden. We developed a hydro-
economic model to evaluate the costs and benefits of single strategies (the Planning Kit DPRD). We have
shown that even if we consider a few measures and a few years for implementation, the number of
potential strategies to be evaluated may become already very large. Combining all results for different
strategies under different climate change and socioeconomic scenarios in order to derive a robust optimal
1 In the Rest scenario, this measure is not beneficial at any of the three moments we examine.
strategy will certainly require assessment models comparable to the Planning Kit DPRD with batch
functionality or optimization procedures.
An important simplification in our model is the summation of the effects of single measures on the design
water levels. The water levels in the area are influenced by both tidal and fluvial effects. This makes it a
complex system, with non-linear effects in the determination of water levels for high return periods. The
summation of effects on design water levels might introduce an error. The overall effect of a combination
of measures on the design water levels might be larger or smaller than the summation of the effects of
single measures. At this moment, we cannot deal with this limitation in our approach.
The adaptive strategies we study in this paper do not differ a great deal from each other. They are
basically strategies with dikes with a few options for other efficient measures. Comparable computations
for the Rest scenario result in comparable optimal strategies, with measures that can be implemented in
later years than in the Steam scenario. In this sense our approach provides robust guidance for the
development of the strategy. We expect the application of the methodology to be much more interesting
when the differences between strategies to be assessed are larger. A big challenge lies ahead to find
optimal adaptation pathways which take also uncertainties into account. The framework described in this
paper serves as a first step and can easily be modified to include no-regret analysis or real option
analysis, but much remains to be done. Extending the framework with non-monetized cost and benefits
could also be considered. We leave this all for future development.
Alphen, J. van, 2014: The Delta Programme and updated flood risk management policies in the
Netherlands. 6th International Conference on Flood Management (ICFM6), September 2014, Sao
Paulo, Brazil (in prep.)
Bokarjova M, P. Rietveld P and E.T. Verhoef, 2012: Composite valuation of immaterial damage in
flooding: Value of statistical life, value of statistical evacuation and value of statistical injury.
Brugge, R. van der, R. Roosjen, T. Morselt and A. Jeuken, 2012: “Adaptive Delta Management”. Water
Bruggeman W., M. Haasnoot, S. Hommes, A. te Linde and R. van der Brugge, 2011: Delta scenarios.
Exploration of physical and socio-economic developments in the 21st century, based on the KNMI
and WLO scenarios, for use in the Delta Program 2011-2012 (In Dutch). Report 1204151.002,
Deltacommissie, 2008: Working together with water, A living land builds for its future, Findings of the
Deltacommissie 2008. The Hague, The Netherlands.
Deltaprogramma Rijnmond-Drechtsteden, 2012. Probleemanalyse.
Eijgenraam, C., J. Kind. C. Bak, R. Brekelmans, D. den Hertog, M. Duits, K. Roos, P. Vermeer and W.
Kuijken, 2014: “Economically Efficient Standards to Protect the Netherlands Against Flooding”.
Interfaces 44(1);7-21. http://pubsonline.informs.org/doi/abs/10.1287/inte.2013.0721
Haasnoot, M., J.H. Kwakkel, and W.E. Walker, 2013: “Dynamic Adaptive Policy Pathways: A New Method
for Crafting Robust Decisions for a Deeply Uncertain World”. Global Environmental Change, 23 (2),
Haasnoot, M., Middelkoop, H., Offermans, A., van Beek, E., and W.P.A. van Deursen, 2012: “Exploring
pathways for sustainable water management in river deltas in a changing environment”. Climatic
Change, doi: 10.1007/s10584-012-0444-2.
Jeuken, A. and T.Reeder, 2011: “Short-term decision making and long-term strategies: how to adapt to
uncertain climate change”. Water Governance 1/2011, 29–35.
Jeuken, A., J. Kind and J. Gauderis, 2013b: “Cost-benefit analysis of flood protection strategies for the
Rhine-Meuse delta” Comprehensive flood risk management. Research for policy and practice.
Proceedings of the 2nd European Conference on Flood Risk Management (Klijn, F., Schweckendiek,
T. Eds.). CRC Press, Taylor & Francis Group, London, UK, p. 436.
Jeuken, A., M. Haasnoot, T. Reeder and P. Ward, 2013a: “Lessons learnt from adaptation planning in
four deltas and coastal cities”. Int. Journal of Water and Climate (under review)
Jongejan, R.B., H. Stefess, N. Roode, W. ter Horst andB. Maaskant. The VNK2-project: a large scale
quantitative flood risk analysis for The Netherlands. 5th International Conference on Flood
Management (ICFM5), 27-29 September 2011, Tokyo-Japan
Katsman, C.A., A. Sterl, J.J. Beersma, H.W. van den Brink, W. Hazeleger and 15 co-authors, 2011:
“Exploring high-end scenarios for local sea level rise to develop flood protection strategies for a low-
lying delta - the Netherlands as an example”. Climatic Change, 2011.
Kind, J., N. Slootjes, D. Wagenaar, A. Tijssen, E. van Zantvoort. B. Faassen and N. Kramer, 2014.
Achtergronddocumentatie Blokkendoos DPRD 1.0 en 2.0. Deltares, Utrecht (in prep.)
Kind, J.M, 2013: Economically efficient flood protection standards for the Netherlands, Journal of Flood
Risk Management, 2013. doi: 10.1111/jfr3.12026
Kwadijk, J.C.J., Haasnoot, M., Mulder, J.P.M., Hoogvliet, M.M.C., Jeuken, A.B.M., van der Krogt, R.A.A.,
van Oostrom, N.G.C., Schelfhout, H.A., van Velzen, E.H., van Waveren, H. & de Wit, M.J.M, 2010:
“Using adaptation tipping points to prepare for climate change and sea level rise: a case study in the
Netherlands”. Wiley Interdisciplinary Reviews: Climate Change, 1, 729-740.10.1002/wcc.64
Ministerie van Verkeer en Waterstaat and Expertise Netwerk Waterkeren, 2007. Leidraad Rivieren.
Most, H. van der, I. Tánczos, K.M. de Bruijn and D. Wagenaar, 2014: New, Risk-Based Standards for
Flood Protection in the Netherlands. 6th International Conference on Flood Management (ICFM6),
September 2014, Sao Paulo, Brazil (in prep.)
Reeder, T. and Ranger, N. 2012: How do you adapt in an uncertain world? Lessons from the Thames
Estuary 2100 project, Washington, DC.
Rhee, G. van, 2012. Handreiking adaptief deltamanagement.
Rhee, G. van, 2013. Kosteneffectiviteitsanalyse Rijn, Quick Scan.
Sayers PB, Galloway GE, Hall JW, 2012 Robust decision-making under uncertainty – towards adaptive
and resilient flood risk management infrastructure. In: Flood Risk (Sayers B ed) . pp 281-302
Schielen, R.M.J. and L.F.M. van den Aarsen, 2014. A river strategy to handle climate change and new
safety standards in the Netherlands. 6th International Conference on Flood Management (ICFM6),
September 2014, Sao Paulo, Brazil (in prep.)
Slomp, R.M., J.P. de Waal, E.F.W. Ruijgh, T. Kroon1 and J.S.L.J. van Alphen, 2014. The Dutch Delta
Model for Policy Analysis on Flood Risk Management In the Netherlands. 6th International
Conference on Flood Management (ICFM6), September 2014, Sao Paulo, Brazil (in prep.)