Vous êtes sur la page 1sur 134

A multi-criteria optimisation of scenarios

for the protection of water resources in


Europe
Support to the EU Blueprint
to Safeguard Europes
Waters
Ad de Roo, Peter Burek, Alessandro Gentile,
Angel Udias, Faycal Bouraoui, Alberto Aloe,
Alessandra Bianchi, Alessandra La Notte, Onno
Kuik, Javier Elorza Tenreiro, Ine Vandecasteele,
Sarah Mubareka, Claudia Baranzelli, Marcel Van
Der Perk, Carlo Lavalle, Giovanni Bidoglio
2012

Report EUR 25552 EN

European Commission
Joint Research Centre
Institute for Environment and Sustainability
Contact information
Prof.Dr. Ad de Roo
Address: Joint Research Centre, Via Enrico Fermi 2749, TP 261, 21027 Ispra (VA), Italy
E-mail: ad.de-roo@ec.europa.eu
Tel.: +39 0332 78 6240
Fax: +39 0332 78 6653
http://ies.jrc.ec.europa.eu/the-institute/units/water-resources.html
http://www.jrc.ec.europa.eu/
This publication is a Reference Report by the Joint Research Centre of the European Commission.
Legal Notice
Neither the European Commission nor any person acting on behalf of the Commission
is responsible for the use which might be made of this publication.
Europe Direct is a service to help you find answers to your questions about the European Union
Freephone number (*): 00 800 6 7 8 9 10 11
(*) Certain mobile telephone operators do not allow access to 00 800 numbers or these calls may be billed.

A great deal of additional information on the European Union is available on the Internet.
It can be accessed through the Europa server http://europa.eu/.
JRC25552
EUR 25552 EN
ISBN 978-92-79-27025-3
ISSN 1831-9424
doi:10.2788/55540
Luxembourg: Publications Office of the European Union, 2012
European Union, 2012
Reproduction is authorised provided the source is acknowledged.
Printed in Italy

Contents
Summary............................................................................................................................................. 5
Introduction......................................................................................................................................... 7
Setup of the study................................................................................................................................ 8
Scenarios......................................................................................................................................... 9
Studies from ENV used in this work.............................................................................................. 11
Basic data used.................................................................................................................................. 12
Meteorological data....................................................................................................................... 12
Meteorological data used in the scenario modelling ....................................................................... 15
Hydrological data for calibration and validation ............................................................................ 15
Irrigation water requirement data 2006 and 2030 ....................................................................... 15
Water abstraction for Livestock ..................................................................................................... 18
Public Water Withdrawals ............................................................................................................. 18
Industrial water withdrawals.......................................................................................................... 20
Energy water withdrawals EPRTR.............................................................................................. 21
Total water abstraction .................................................................................................................. 22
Point sources data.......................................................................................................................... 22
Cost of water retention scenarios ................................................................................................... 23
Cost of other scenarios .................................................................................................................. 24
Benefits (reduced economic loss for sectors, reduced flood damage) ............................................. 25
Modelling methods............................................................................................................................ 28
CAPRI .......................................................................................................................................... 28
LUMP ........................................................................................................................................... 29
LISFLOOD ................................................................................................................................... 30
EPIC for N & P fluxes................................................................................................................... 34
LISQUAL ..................................................................................................................................... 35
The Optimisation routine............................................................................................................... 45
The use of water-regions and macro-regions............................................................................. 49
Sensitivity analysis and uncertainty ............................................................................................... 51
The Baseline conditions .................................................................................................................... 54
Costs of the scenarios per region ....................................................................................................... 56
Results of the Individual Scenarios.................................................................................................... 58
The Baseline Scenarios.................................................................................................................. 58
The 11-Riparian zone scenario ...................................................................................................... 68
The 12-Afforestation scenario ....................................................................................................... 70
The 21-Urban25 scenario .............................................................................................................. 72
The 22-Urban50 scenario .............................................................................................................. 74
The 31-Grassland scenario............................................................................................................. 76
The 32-Buffer strip scenario .......................................................................................................... 78
The 33-Grassed waterways scenario .............................................................................................. 80
The 34-Crop practices scenario...................................................................................................... 82
The 43-Meander scenario .............................................................................................................. 84
The 51-N-fixing scenario............................................................................................................... 86
The 52-Optimum Fertilization scenario ......................................................................................... 88
The 71-Desalination scenario......................................................................................................... 90
The 91-Irrigation Efficiency scenario ............................................................................................ 92
The 93-Water Re-Use scenario ...................................................................................................... 94
The 94-Water Saving scenario....................................................................................................... 96
3

The 95-Leakage scenario............................................................................................................... 98


The 96-Wastewater re-use scenario ............................................................................................. 100
Results of the multi-criteria optimisation ......................................................................................... 102
Example of the procedure of the multi-criteria optimization for the Danube ................................ 103
Region 11 Danube: Crop optimisation ...................................................................................... 111
Region 11 Danube: Flooding optimisation ............................................................................... 113
Region 13 Iberia Mediterranean: Water saving optimisation ..................................................... 115
Region 13 Iberia Mediterranean: Crop optimisation.................................................................. 116
Region 13 Iberia Mediterranean: Flooding optimisation............................................................ 117
Region 10 France Atlantic: Water saving optimisation.............................................................. 118
Region 10 France Atlantic: Crop optimisation .......................................................................... 120
Region 10 France Atlantic : Flooding optimisation ................................................................... 122
Limitations of the current study, and further work needed ............................................................... 124
Conclusions..................................................................................................................................... 125
References....................................................................................................................................... 128

Summary
A modelling environment has been developed to assess optimum combinations of water retention
measures, water savings measures, and nutrient reduction measures for continental Europe. This
modelling environment consists of linking the agricultural CAPRI model, the LUMP land use model,
the LISFLOOD water quantity model, the EPIC water quality model, the LISQUAL combined water
quantity, quality and hydro-economic model, and a multi-criteria optimisation routine.
Simulations have been carried out to assess the effects of water retention measures, water savings
measures, and nutrient reduction measures on several hydro-chemical indicators, such as the Water
Exploitation Index, Environmental Flow indicators, N and P concentrations in rivers, the 50-year
return period river discharge as an indicator for flooding, and economic losses due to water scarcity for
the agricultural sector, the manufacturing-industry sector, the energy-production sector and the
domestic sector. Also, potential flood damage of a 100-year return period flood has been used as an
indicator.
The study shows that technically this modelling software environment can deliver optimum scenario
combinations of packages of measures that improve various water quantity and water quality
indicators, but that additional work is needed before final conclusions can be made using the tool.
Further work is necessary, especially in the economic loss estimations, the water prices and priceelasticity, as well as the implementation and maintenance costs of individual scenarios.
The most promising individual scenarios are the following:
The Water saving in households scenario, which aims to achieve 25% water savings in the domestic
sector by simplistic measures (replacing showerheads, etc.), improves the Water Exploitation Index,
and reduces the amounts of abstracted and consumed water. Positive changes are observed in several
regions in the following order of importance: Great Britain (GB), Po (Milan area), Mediterranean
Iberia, southern Italy and Odra/Vistula (Warsaw area), with even higher local effects around populated
areas. Reductions in water consumption are also simulated in many other areas.
The Irrigation efficiency scenario aims to improve irrigation efficiency levels from the current
average of 74% (Eastern Europe) - 77% (Western Europe) to 93% by applying drip irrigation in all
areas. This scenario improves the Water Exploitation Indices and the Environmental Flow Indices,
especially in the Danube, Iberia/Mediterranean, southern Italy, Sicily, Sardinia, Greece/Evros, and the
France/Atlantic macro-region. Improvements are also simulated in other regions. An additional benefit
is also that the use of deep (geological) groundwater is reduced by around 20% in all areas. Due to the
larger amount of water available when less irrigation water is consumed, economic losses are also
reduced for industry, the public sector, and agriculture.
The scenario Water re-use in industry, which assumes that 50% of the water abstracted for industry
is re-used, leads to improvements in the Water Exploitation Index of around 10% in several regions,
with most effects being simulated in the industrial Elbe/Ems, GB, the Rhine/Meuse/Scheldt region,
southern Italy, Sardinia, Sicily, the Po and the Odra/Vistula region.
The Leakage reduction scenario, which assumes that 50% of the current leaking in the public water
supply is repaired, improves the Water Exploitation Index in all regions, most dramatically in GB
(24%), Ireland (38%), Po (6.2%), Adige/Balkan (5.8%), and Greece/Evros (4.1%), with local effects
even higher. It also improves the Environmental Flow Indicators by several days per year, especially in
the region of GB (8.6%), Ireland (5.8%), Sardinia (3.7%) and Sicily (2.2%).
5

The Urban-greening 25% scenario - establishing urban greening measures - reduces flood peaks
(Q50) slightly, for example by 0.7% in the region of GB. This scenario consequently also reduces the
potential flood damage by 27% in the region of GB in general, and by even more locally in England.
Further positive effects are simulated in the regions of the Rhine/Meuse/Scheldt (0.2% Q50 decrease,
12% flood damage decrease), Elbe/Ems (0.3% Q50 decrease, 6.5% flood damage decrease), Po (0.2%
Q50 decrease, 4.5% flood damage decrease) & Mediterranean Iberia (0.2% Q50 decrease, 6.0% flood
damage decrease). On the other hand, reduced runoff from cities results in less availability of water for
extraction, and thus leads to a slight deterioration of the WEI in those areas. However, environmental
flow (10th percentile) conditions improve in GB by 0.4%, likely because of increased baseflow as a
consequence of increased infiltration. Because of reduced total river flow amounts, N concentrations in
rivers are simulated to slightly increase under the urban25 scenario in the regions of GB (0.2%) and
the Elbe/Ems (0.3%).
The N-fixing Scenario and the Optimum Fertilization Scenario both reduce Nitrate and Phosphate
concentrations in all regions that have significant agriculture, most dramatically in the Elbe/Ems
region (68% and 26%), France/Atlantic (61% and 25%), Denmark/northern Germany (61% and 45%),
the Rhine/Meuse/Scheldt (63% and 39%), and GB (75% and 26%).
The Re-Meandering Scenario, which increases the meandering of the current rivers by increasing the
length and storage capacity of the river bed reduces flood peaks in all European regions, and is
estimated to significantly reduce the flood damage potential especially in the Elbe/Ems (11%), Danube
(10%), Odra/Vistula (9.8%), Po (6.8%), Rhine/Meuse/Scheldt (5.3%) and France/Atlantic regions
(5.8%). At the same time, environmental flow conditions improve in some areas, for example in GB
(0.1%) and Ireland (0.3%), while environmental flow conditions are simulated to deteriorate by 0.5 to
0.9% in Mediterranean regions. Nitrate concentrations decrease in all regions, with maximum changes
of 20-25% in GB and Ireland.
The Crop Practices Scenario, which simulates the effects of the implementation of combined
methods of improved crop practices (reversed/reduced organic matter decline and increased mulching
and tillage), results in reductions of potential flood damages in all regions, including areas with high
absolute flood damages in regions such as the Odra/Vistula (6.2% reduction), the Rhine/Meuse (7.8%
reduction), Great Britain (15.9% reduction) and the Danube (8.2% reduction).
Installing desalination plants along the coastlines would improve the Water Exploitation Index in
several European macro-regions, and decrease the number of days during which Environmental Flow
cannot be respected, especially in Spain and Italy.

Introduction
In the context of the impact assessment for the forthcoming "Blueprint to Safeguard Europe's Water
Resources", the European Commission is developing a baseline scenario that will bring together
climate, land-use and socio-economic scenarios and look at the implication for water resources
availability and use under different policy scenarios. This work is carried out by the Joint Research
Centre of the European Commission. The objectives of the assessment are:
1. The development of an optimisation model linked with dynamic, spatially explicit water
quality and quantity models that allows for the selection of measures affecting water
availability and water demand based on environmental and economic considerations,
2. The application of this model to all European River Basins for a baseline scenario and a
number of alternative policy and socio-economic scenarios.
The aim of the modelling exercise is to help maximise the net social benefits gained from the use of
water by economic sectors including a range of components, such as welfare impacts for water users,
valuation of key ecosystem services provision, and valuation of the external costs of degrading
ecological and chemical status and energy consumption triggered by water abstraction and return.

Setup of the study


To achieve this, the JRC is building a hydro-economic model platform based on a combination of the
following elements:
- A simplified, pan-European, biophysical modelling tool, which integrates and/or mimics the
results of full runs of the Integrated Water Modelling Platform (IWMP). The IWMP is
currently developed by the JRC and already partially employed in support of the Blueprint. The
IWMP already allows for the definition and implementation of alternative scenarios of water
management in connection with socio-economic drivers. The following modelling tools,
maintained and used at the IES, will be primarily used: the JRC LISFLOOD model for water
quantity, the SWAT/GREEN/EPIC models for water quality, and LUMP (the JRC Land Use
Modelling Platform) for land use change scenarios. Agri-economic scenarios are provided by
the CAPRI model following on the work performed for the CAP assessment.
-

An optimisation tool that builds on an existing optimisation model based on multi-criteria


analysis, and which has already been applied earlier in a pilot study on the river basins in
Catalonia, Spain1. This multi-objective optimisation, based on a genetic algorithm, has proven
powerful for addressing issues of conflicting management goals and objectives.

A set of economic functions representing water demand and supply which allows the impacts
of strategic measures to be assessed, defined and discussed with stakeholders.

For this study, the biophysical modules described above must be combined with economic components
that account for the economic profits or losses at each demand site and the benefits to, for example,
ecological and environmental uses. The defined scenarios are analysed with this model.
The suite of models allows the variability of the quantity and quality of water resources to be
accounted for. All hydrologic elements will be linked to a network of sources and demand sites for
urban and industrial water uses, for energy production, for ecological and environmental flows, and for
agricultural needs. The latter will be based on specific modules accounting for river-basin scale
agricultural practices and their impacts on water. Withdrawals and return flows are developed for each
of these demand nodes, or determined using empirical relations between water and type of productive
uses. Taking economic and environmental constraints into account, the optimisation module is
envisaged to allocate available water to all end users while ensuring the best trade-offs for economic
and environmental sustainability.
The optimisation tool as described above can only work if it is fully and dynamically linked to a
biophysical model. Most of the existing biophysical models that are coupled with optimisation tools
only address sectoral aspects of the water management cycle, e.g. water distribution networks or
selected water quality components only.
In the short term, the existing JRC biophysical models (LISFLOOD, SWAT, etc.) are too complex to
achieve this dynamic coupling within the available time schedule of the Blueprint. Therefore, to meet
the Blueprint time schedule, we are using a pragmatic approach.
In the context of the present contract, the JRC is developing a - somewhat simplified - biophysical
modelling tool with simulations at pan-European scale on a 5x5 km regular grid, with a daily
simulation timestep. The modelling tool working title LISQUAL - integrates the results of full

http://www.analisisderiesgos.org/docs/TR/2010/Technical_Report_2010_23.pdf

simulation runs of the detailed water quantity model LISFLOOD and the water quality model output
produced by the EPIC model.
This approach offers two advantages: 1) continental coverage and dynamic coupling with the
optimisation model, 2) the taking into account of specific measures which are modelled at higher
resolution with models such as LUMP, the Land Use Modelling Platform.

Figure 1 Flowchart of the modelling exercise carried out in this study.

Scenarios
The identification of the classes of measures to be included in the scenarios is essential for the
definition of the objectives to be evaluated in the multi-criteria decision, e.g.

Water supply from other basins, the sea or fossil resources;

Increase water retention (artificial/natural);

Improve the quality of available water;

Improve efficiency in the use of water.


The table below excludes measures that address chemicals (other than N and P) and ecological status
which, while they are very relevant for the transformation agenda under the Blueprint, cannot be
integrated into the optimisation procedure with hydro-economic modelling in the context of this
assessment due to their higher complexity.
Measures such as water pricing or targets are not modelled per se. They are parameters, constraints or
outputs of the optimisation.

The measures listed below are included in the analysis:


Category

Scenario

Description

BASELINE2030

0.0 Baseline 2030

LUMP 2030, 2010 fertilisation application, 2010 point sources

BASELINE2006

0.1 Baseline 2006


1.1 Riparian Afforestation, CAP
consistent
1.2 Afforestation in mountainous
areas

As Baseline 2030, but with Landuse 2006

1-FOREST

2-URBAN

Afforest areas from LUMP-CAP scenarios


Afforest areas in mountainous areas (LUMP)
Green infrastucture, Green roofs, Rain Gardens, Park Depressions; For
all urban areas: Direct Runoff Fraction << 50%, Evapotanspiration >>
50%
Green infrastucture, Green roofs, Rain Gardens, Park Depressions; For
all urban areas: Direct Runoff Fraction << 25%, Evapotanspiration >>
25%

2.1 50% Green

2.2 25% Green


3-AGRICULTURE

4-NATURAL RETENTION

5-NUTRIENTS

6-POINT SOURCES

7. WATER SUPPLY

8. TECHNICAL
RETENTION

3.1 Grassland

3.2 Buffer strips

Convert areas from LUMP-CAP scenarios to grassland


5m wide grass buffer strips within arable fields, on slopes < 10%, every
200m; 2.5% of arable land converted to grassland, only on slopes <
10%

3.3 Grassed waterways

10m wide grass-covered areas in valley-bottom; 1% of arable land


converted to grassland, in valley-bottoms > 5%

3.4 Crop practicies

Reverse OM decline and increase mulching; increased infiltration,


porosity, modified hydraulic parameters

4.1 Wetlands

Riparian wetlands along rivers; Change cross section

4.2 Polders

Introduce flood retention polders along rivers

4.3 Re-meandering
4.4 Buffer ponds in headwater areas
1
4.5 Buffer ponds in headwater areas
2

natural retention ponds in headwater areas with 5000 m3 storage per


25km2
natural retention ponds in headwater areas with 10000 m3 storage per
25km2

5.1 N-fixing winter crops

updated N & P fluxes

5.2 optimum fertilisation application

updated N & P fluxes

5.3 N-fixing winter crops & optimum


fertilisation application
6.1 New wastewater treatment plants
(WWTP)

updated point information

6.2 Changing type of WWTP

updated point information

7.1 groundwater extraction

updated point water availability

7.2 desalination
7.3 large-scale water-transfer
infrastructures

updated point water availability

8.1 constructing dams and reservoirs

new dams/resoirvoir to temporarily store water

updated N & P fluxes

transfer of water between river basins

8.2 hard infrastructure for flood risk


9. EFFICIENCY

9.1 Irrigation management


9.2 Water efficiency in power
generation
9.3 Water efficiency in industrial
processes
9.4 Water efficiency in
Buildings/households

optimizing crop water requirements


Save water in power generation, as compared to current use
Save water in industry, as compared to current use
Save water in households, as compared to current use

9.5 Leakage reduction

Fix all leakages 90% or 100% (reduce water abstraction)


Reduce deep groundwater use for irrigation and replace by treated
wastewater

9.6 Wastewater reuse for irrigation

Figure 2 Overview of the scenario measures included in this study.

10

Some assumptions were made in these scenarios:


Scenario 7.1 Desalination
o Plants are assumed to treat 60 Megalitres per day
o Water is used at a maximum distance of 150km inland, inside the same Major Region,
but can contain more Water Regions
o Costs are calculated including distance to plant and height difference
Scenario 9.1 Irrigation efficiency
Change efficiency from the current average of 74% (Eastern Europe) - 77% (Western
Europe) to 93% by applying drip irrigation. Costs 153 Euro/ha.
o Current efficiency stored in maps
o ScenIrgf.map defines additional irrigation efficiency (16-19%)
o Applied only in the non-drip irrigation areas (currently 3% in Eastern Europe, 18%
in Western Europe
Actual area of irrigation is used (hectare per 5x5km model pixel)
Costs: Adjusted for national price levels
Scenario 9.3 Water re-use in industry
Assumed that 50% of water is re-used
Costs are 0.013 Euro per re-used m3
Adjusted for national price levels
Water re-use for irrigation is treated in scenario 9.6
Scenario 9.4 Water savings in domestic a sector
25% savings achieved
since payback times are less than 30 years, 0 costs assumed
Scenario 9.5 Leakage reduction
50% reduction compared to current leakage
Scenario 9.6 Wastewater re-use for irrigation
50% reduction in the use of deep groundwater for irrigation; instead, treated urban
wastewater is used for irrigation
Studies from ENV used in this work
The following studies of DG ENV have been used in this analysis:
The STELLA 2012 study on natural water retention measures including their costs,
complemented by JRC hydrological and land use modelling
The COWI 2011 study on the investment needed to rehabilitate water distribution networks
(ENV.G.1/FRA/2006/0073, September 2011)
Studies contributing to Water Scarcity and Droughts Gap Analysis
Discussion Paper Environmental flows in the EU in the context of the Expert Group on
water scarcity and droughts, subcontracted to TYPSA (Jan 2012)
Water Scarcity & Drought indicator set development document (April 2012), on the Water
Exploitation Index

11

Basic data used

Meteorological data
The meteorological variables of precipitation, minimum/maximum/average/dewpoint temperature,
wind speed, potential evapotranspiration, and evaporation rates for open water and bare soil surfaces
for driving LISFLOOD are derived from various data sources for the period 1/1/1990 until 31/12/2010.
These data sources include the JRC MARS database (http://mars.jrc.ec.europa.eu/mars/), data obtained
from MeteoConsult, SYNOP data, as well as data from the European Climate Assessment & Dataset
(ECA&D, http://eca.knmi.nl/). An overview of the definitions of the different meteorological variables
and the data sources used for those variables is given in Table 1.

Variable
ws
pr
tn
tx
td

ta

e0
et
es

Definition
Mean daily wind speed at 10 metres/second (m/s)
from 0-24 UTC
Precipitation (mm) between 6 UTC on the day
specified and 6 UTC on the following day
Minimum temperature (C) between 18 UTC and
6 UTC (i.e. during the preceding night)
Maximum temperature (C) between 6 UTC and
18 UTC (i.e. during daytime)
Average of all available dewpoint temperature
measurements between 6 UTC on the day
specified and 6 UTC on the next day.
If the daily maximum temperature (tx) and the
daily minimum temperature (tn) is known, mean
daily
temperature
is
calculated
as
ta = (tx + tn) / 2.
Penman potential evaporation from a free water
surface (mm/day)
Penman potential transpiration from a crop
canopy (mm/day)
Penman potential evaporation from a moist bare
soil surface (mm/day)

Data Sources
JRC MARS, SYNOP,
ECA&D
JRC MARS, SYNOP,
ECA&D, MeteoConsult
JRC MARS, SYNOP,
ECA&D
JRC MARS, SYNOP,
ECA&D
SYNOP

JRC MARS, SYNOP,


ECA&D

JRC MARS
JRC MARS
JRC MARS

Table 1 An overview of the observed meteorological variables as used in LISFLOOD and LISVAP

All meteorological variables are interpolated on a 5 x 5 km grid using inverse distance weighting with
a weight of d-2 and a maximum number of 5 points for the interpolation. Temperature variables are
first corrected using the elevation obtained from a DEM with a resolution of 1 x 1 km and using a
constant lapse rate of 0.006 (0.002 for dewpoint temperature) and are then interpolated onto the 5 x 5
km grid. An example of the spatial distribution of precipitation observations is given in Figure 4.

12

Figure 3 Spatial distribution of available precipitation observations on 15 Dec 2010.

Figure 5 shows the temporal distributions of the different meteorological variables from 01.01.1990
until 31.12.2010. On average more than 2000 5 x 5 km pixels have at least one precipitation
observation and more than 1,200 have a temperature or wind speed observation for the time period
between 1990 and 2003. Since about 2004 on average more than 2,500 5 x 5 km pixels for
precipitation and more than 2,000 for temperature and wind speed are observed at least once.

13

3500

ws

td

ta

tx

tn

pr

3000
2500
2000
1500
1000
500
0
1/
1/
90
1/
1/
91
1/
1/
92
1/
1/
93
1/
1/
94
1/
1/
95
1/
1/
96
1/
1/
97
1/
1/
98
1/
1/
99
1/
1/
00
1/
1/
01
1/
1/
02
1/
1/
03
1/
1/
04
1/
1/
05
1/
1/
06
1/
1/
07
1/
1/
08
1/
1/
09
1/
1/
10

Number of pixels with at least one observation

4000

Figure 4 Temporal distribution of the number of observations available for the interpolation of precipitation,
minimum/maximum/average/dew point temperature, and wind speed.

Potential evapotranspiration, and evaporation rates for open water and bare soil surfaces, are calculated
in two different ways during the 21-year time span. For periods before 23/01/2003 e0, et and es are
taken from the JRC MARS database directly and interpolated as described above. However, from
23/01/2003 onwards, LISVAP (van der Knijff, 2008), an evaporation pre-processor for LISFLOOD, is
used to derive the maps using the observed variables minimum daily temperature (tn), maximum daily
temperature (tx), dewpoint temperature (td) and windspeed (ws). Figure 6 shows the annual average
precipitation for Europe, computed from the data we have available for this study.

Figure 5 Annual average precipitation (mm) (1990-2010), based on spatially interpolated ground station
measurements, using the JRC CGMS/Mars database and the JRC EUFloodGIS database. Source: JRC / Salamon,
Burek (2011).

.
14

Meteorological data used in the scenario modelling


Weather data to carry out the scenario calculations are derived from baseline 1981-2010 climate
simulations from the data portal of the ENSEMBLES project. This is done to in a later stage carry out
climate change scenarios using the same setup and compare the results with the baseline scenario.
Within the ENSEMBLES project a large number of climate simulations using different combinations
of state-of-the-art General Circulation Models (GCMs) and Regional Climate Models (RCMs) have
been run for Europe. All climate simulations are forced by the SRES-A1B scenario defined by the
IPCC. For this work, we employ three climate simulations obtained from a combination of two GCMs
(HadCM3Q0 and ECHAM5) and three RCMs (HIRHAM5, RACMO2 and HadRM3) (see Table 1 for
details). These climate simulations have been selected in order to cover as much as possible of the
climate uncertainty.
Climate simulations from the 3 models have been bias corrected by JRC, based on the algorithms
developed by Piani et al. (2010) and recently applied to correct climate simulations from
ENSEMBLES by Dosio and Paruolo (2011) and Rojas et al. (2011).
Model

1
2
3

Driving GCM

RCM

Institute

ECHAM5-r3 HIRHAM5 Danish Meteorological Institute


DMI-HIRHAM
ECHAM5-r3 RACMO2 The Royal Netherlands Meteorological Institute
KNMI-RACMO
HadCM3Q0 HadRM3Q0 UK Met Office, Hadley Centre for Climate Prediction and Research METO-HadRM3Q
Table 1: Climate simulations used to force LISFLOOD in the period 1981-2010
Hydrological data for calibration and validation
Observed river flow data at gauging stations from Europe were used from the Global Runoff Data
Centre (GRDC). These daily observed data have been used to calibrate and validate the LISFLOOD
model setups for Europe. For Europe, historical river flow data from 435 stations (Figure 2) have been
used for model calibration and validation.

Irrigation water requirement data 2006 and 2030


A detailed description of the methods used to derive irrigation requirements for Europe can be found in
Wriedt et al. (2008, 2009). Irrigation requirements were estimated for 1984 until 2008. For the purpose
of the scenario study here, monthly averages were used derived from these 25-year period, and used
for the baseline 2006 scenario. Since an irrigation water requirement data set consistent with the 2030
land use scenario is not yet available, the 1984-2008 average irrigation data are also used for the
baseline2030 scenario and all measures scenarios.
The generation of the irrigation map followed a two-step procedure. First, irrigated area was
distributed to crop categories at sub-regional level (NUTS3) based on statistical information and
distribution rules. Next, the regional information was disaggregated to a high resolution dataset based
on the crop distribution and a global irrigation dataset (Siebert et al., 2005). EPIC uses a base-map to
identify irrigated areas, which originates from FAO (Siebert et al., 2005). Some boundaries between
the countries can clearly be seen on this map (e.g. FR-DE) and this causes high irrigation
requirements, different across boundaries and across rivers (Rhine for instance).

15

Figure 6 Irrigated area (Source: FAO, Siebert et al., 2005)

Based on crop growth, soil water and the EPIC nutrient model, we estimated irrigation water
requirements on a daily basis at a 10 x 10 km grid scale using automatic irrigation under different
strategies, including unlimited irrigation, various water stress thresholds, and a no irrigation strategy.
We then chose the irrigation strategy that yields at least 80% of the yield obtained under the unlimited
irrigation strategy, while also minimising water use. Selecting different irrigation strategies allowed us
to adapt irrigation more specifically to soil conditions and crop growth. For rice we took a fixed
irrigation strategy applying an amount of 3,500 mm per year. The calculated average irrigation
requirements for irrigated crops range from a minimum of 13 mm/yr in Switzerland to 900 mm/yr and
more for Spain, Portugal, Greece and Italy.
EPIC runs for the five dominant crops of each grid cell. This resulted sometimes in the loss of certain
irrigated crops in Germany and UK in particular. To solve this issue we forced EPIC to run for the
irrigated crops even though these were not in the 5 dominant crops. So for Germany and the UK we
forced EPIC to run for irrigated potatoes and sugar beet. In France, there initially was an
overestimation of irrigation in Brittany, which has been resolved with the final data used. A map of
irrigation requirements is shown in Figure 7. Clearly irrigation requirements for both Germany and UK
are low compared to those of Southern countries.

16

Figure 7 Average annual irrigation water requirements for Europe (mm per 25 km2 grid). Source:
JRC/Bouraoui/Aloe (2012), updated from Wriedt (2008).

A graph showing the estimated versus reported abstraction using EPIC for year 2000 is shown in
Figure 8. No estimation was done on uncertainty as the raw data reported by Member States is already
subject to high unquantifiable uncertainty. However the estimations of EPIC are in the range of
reported abstraction values.

Figure 8 Estimated versus reported irrigation abstraction using EPIC for year 2000 (Aloe & Bouraoui, 2012)

17

Water abstraction for Livestock


Daily maps of livestock water withdrawal were calculated using the Food and Agriculture
Organization (FAO) livestock density maps and output from the CAPRI agricultural model. The water
use was calculated based on the specific water requirement (taken from literature) and spatial
distribution of each type of livestock (cattle, pigs, poultry, sheep and goats). A temperature function
was derived and applied to describe daily variations in water use for each livestock group.

Figure 9 Average annual livestock water withdrawal (2005-07) (mm/yr per 25 km2 grid).
Source: JRC/Mubareka, Vandecasteele, Bianchi (2012).

Public Water Withdrawals


The water withdrawal maps for the public, industry and energy sectors were calculated based on a
disaggregation of sectorial water use statistics. The OECD/EUROSTAT Joint Questionnaire on Inland
Water provides country-level statistics on annual freshwater abstraction by source and sector and water
use by supply category and user (Nagy et al.). The questionnaire covers the EU27 countries plus some
data on Iceland, Norway, Switzerland, Croatia, the Former Yugoslav Republic of Macedonia, Turkey,
Bosnia and Herzegovina, and Serbia. For the reference year 2006, we used the average sectorial water
withdrawals for the period 2005-7 from EUROSTAT, supplemented by the 2003-2007 average from
FAO - AQUASTAT. Where there were still missing values the respective sectorial European average
per unit was used. Since AQUASTAT provides only a total value for the industrial sector, we used the
EU27 average to split the totals into the respective proportion of water used for industry
(manufacturing), 40% and energy production, 60% of total.
18

Monthly public water withdrawal maps were computed, while industrial and energy withdrawals were
assumed to remain constant throughout the year. The final maps were aggregated to fit the 5k mask,
and values converted from hm3 to mm per year.

Figure 10 Average annual public water withdrawl for Europe (2006) (mm/yr per 25 km2 grid).
Source: JRC/Mubareka, Vandecasteele, Bianchi (2012).

2006
Public water withdrawals were assumed to be those made by residents and tourists in urban areas, so
that the spatial distribution of the withdrawals was assumed to be directly related to the combined
population and tourist density. Since tourists are known to have a higher water use than residents, the
tourist density maps were given a greater weight when assigning the water withdrawals we used a
ratio of 300/160 (Gssling et al., 2012).
Population density maps were available for 2006 at 100m resolution (Batista e Silva et al., submitted).
Tourist density maps were created using the number of nights spent by non-residents at NUTS2 level
(EUROSTAT) further disaggregated to NUTS3 level using the number of bedplaces (EUROSTAT).
The monthly distribution of tourism was calculated using the country-level percentage of nights spent
per month (EUROSTAT). For both cases, where data was missing national statistics or regional
averages were used, always taking the closest available year to 2006.
The total number of tourists per month at NUTS3 level for each country was disaggregated to the
refined Corine classes 111 and 112 (urban fabric), 141 (green urban areas), and 142 (sport and leisure
facilities). The number of nights spent abroad by residents per quarter year was also calculated and
subtracted from the population density maps. The final map, to which the country-level public water
withdrawal statistics were disaggregated, was then calculated as:

19

Weighted number of users per pixel = Population density map 2006 nights abroad map
(quarterly) + 300/160*tourist density map (monthly)

2030
A population density map for 2030 was computed, using population projections from EUROSTAT,
and modelled land use for 2030 from the EUClueScanner100 model. Both the tourist density maps and
the nights spent abroad maps were multiplied by a specific predicted annual growth factor taken from
the Tourism vision 2020 projections (WTO, 2000). The temporal (monthly) and spatial distribution of
tourism was kept constant.
The public water withdrawal per capita was also kept constant, so that the total public water
withdrawals for 2030 directly reflect the projected population and tourism densities.

Industrial water withdrawals


2006
All water used for manufacturing purposes was assumed to be withdrawn within industrial areas.
Where refined Corine land use maps were available (Batista e Silva et al., 2012), the water
withdrawals were disaggregated to the Industry classes 121 (industry and commercial units), 123 (port
areas), 124 (airports), 131 (mineral extraction sites), and 133 (construction sites).

Figure 11 Average annual Industrial water withdrawal (2005-2007) (mm/year per 25km2 grid). Source:
JRC/Mubareka, Vandecasteele, Bianchi (2012).

20

Where there was no coverage by the refined Corine land use map (for MD, UA, BY, RU), water
withdrawals were assumed to occur within the globcorine land use class 0 (Urban and associated
areas). For countries where no industrial water withdrawal totals were available (AD, FO, IM, LI,
MC, SM, VA) we used the EU27 average water use per pixel of industrial land.
2030
In order to calculate water withdrawals for 2030, we first computed a change factor per country. We
assumed the driving force for water withdrawals in time to be the GVA for industry.
Looking at historical data (1990-2010), there is a trend towards more efficient water use in industry
due to technological improvements, and therefore, on average, a decreasing water use per unit in
Europe. The average (decreasing) trend in total industrial water withdrawals for 2000-2006 for DE,
ES, FR & UK, was taken as an efficiency factor (-1.33 %/year) to correct for this observation. This
factor was subtracted from the (increasing) trends given by the GVA (%increase/year, for EU25, from
GEM-E3) for each country. The EU average trend in GVA was used to fill in for any missing values.
Country change factor (%/yr) = GVA for industry (%/yr) efficiency factor (%/yr)
For countries where the land use was kept constant (non-EU27), the 2006 industrial water withdrawal
map was multiplied directly by this change factor (x24 years) to compute the 2030 water withdrawals.
For the EU27 countries the total water withdrawal for 2006 was first multiplied by the country-specific
change factor (x24 years) to give the 2030 values, and then disaggregated to the industrial area for
2030 (modelled using EUClueScanner100, Lavalle et al., 2011).

Energy water withdrawals EPRTR


2006
Energy water withdrawals are those made for cooling during electricity production; we therefore
assumed this withdrawal to be attributed almost exclusively to thermal power stations.
For the EU27 countries, the country-level water withdrawals for use in the energy sector were
disaggregated using the number of thermal power stations per 5km pixel extracted from the European
Pollutant Release and Transfer Register data base, E-PRTR (any records having the same location
were first removed to avoid double counting).
For non-EU27 countries, for which this data was not available, the water withdrawals were
disaggregated to the Corine refined industry class 121, or to the globcorine class 0, depending on
availability. Where energy water withdrawal statistics were not available, the EU27 average
withdrawal per pixel was used.
2030
We assumed the driving force for water withdrawals in time to be energy consumption. The average
trend in water use 2000-2006 for DE, ES, FR, UK & PL was taken as an efficiency factor (-1.69
%/year), which was subtracted from the trends given by energy consumption (POLES output,
%increase/year) for each country. The EU average trend in energy consumption was used to fill in for
any missing values.
Country change factor (%/yr) = energy consumption (%/yr) efficiency factor (%/yr)
The 2006 energy withdrawals map was multiplied directly by this change factor (x24 years) to
compute the 2030 energy water withdrawals.

21

2030
The country change factors were computed as described in the EPRTR methodology. For the EU27
countries, the land use was updated to the modeled industry class 121. For non-EU27 countries, the
2006 energy withdrawals map was directly multiplied by the country change factors.

Total water abstraction


Figure 10 shows the sum of the water abstractions for irrigation, industry, livestock and domestic
purposes.

Figure 12 Total average annual water demand in Europe for irrigation, livestock, industry (manufacturing and
energy production) and the domestic sector. Source: JRC/Bouraoui, Wriedt, Mubareka, Vandecasteele, Bianchi
(2012).

Point sources data


We have used to UWWTP database distributed publically by the EEA. It covers 28 countries. We
cleaned as much as possible and then converted the data into PE (person equivalent) discharging to a
specific stream according to the following flow chart:

22

Figure 13 Flowchart to estimate point sources.

We have added the level of treatment to each treatment plants in order to generate the load of waste
water in terms of water and BOD, nitrogen and phosphorus. The discharged loads and their location
are available as input to LISQUAL and for further use in the optimisation.

Cost of water retention scenarios


For the natural water retention measures, the contractor Stella has collected the costs of measures,
specified per country. Five main categories of costs, including sub-categories, have been identified by
STELLA:
Land requirement: Acquisition and compensation;
Construction and rehabilitation: Investment, design and contingency;
Construction and rehabilitation: Operation and maintenance;
Administrative costs: Enforcement costs, monitoring, extension of networks;
Other costs.
For several scenarios, a unit investment cost and operation and maintenance cost is estimated per ha:

Figure 14 Unit investment and operation costs (source: Stella, 2012)

23

Cost of other scenarios


For the other scenarios, the sub-contractor IVM/VU (Free University of Amsterdam) has collected
information on unit cost across all EU member states of the measures:

Optimum fertiliser application

New wastewater treatment plants

Change type of wastewater treatment plant

Groundwater extraction

Desalination

Large-scale water transfer infrastructures

Constructing dams and reservoirs

Hard infrastructure for flood risk

Irrigation management

Water efficiency in power generation

Water efficiency in industrial processes

Water efficiency in building/households

Figure 15 Comparitive Price Levels (2010)

24

Benefits (reduced economic loss for sectors, reduced flood damage)


Not only costs of the measures are taken into account, but also some benefits, specifically:
reduced expected annual flood damage as a consequence of the measures in the scenario
reduced economic loss for the domestic sector
reduced economic loss for the agricultural sector
reduced economic loss for the industrial sector
reduced economic loss for the energy sector
The Expected Annual Flood Damage is calculated by combining the waterlevels of several return
periods of flooding, with 100m resolution potential damages calculated by combining local elevation,
water level, land use, and economic loss data which are a function of waterdepth and land use.
Economic losses are defined as the benefits forgone to water users as they receive less water than in a
normal year (i.e., less than their baseline water demand). The benefits of a unit of water at current
levels of provision are (under competitive conditions) equal to its market price. At lower levels of
provision, the forgone benefits per unit of water will typically rise for water users. The relationship
between the level of provision and the forgone benefits per unit of water can be depicted in an
(inverse) demand curve. The slope of the (inverse) demand curve expresses the sensitivity of the
forgone benefits (i.e. the economic losses) to the change in provision. If the slope is very steep, losses
increase sharply when the provision of water falls. If, in contrast, the slope is shallow, losses increase
slowly as the provision falls. Figure 15 presents an example of an (inverse) demand curve that is used
to calculate economic losses in this study. In this Figure, the economic loss due to a reduction in the
provision of water is measured by the area under the demand curve that has the difference between the
baseline water availability (3.70E+08 m3) and the restricted availability (for example 3.00E+08 m3) as
its base.
The slope of the demand curve at a specific level of provision is measured by the so-called price
elasticity of demand. This elasticity is negative, indicating that the price (i.e. benefit per unit)
increases as supply decreases. If the elasticity is close to zero, the slope is very steep and the demand
for water is said to be inelastic. If the elasticity is less than -1, the demand is said to be elastic.
This study constructed demand curves for the domestic, industrial, energy and agricultural sectors.
Note that we consider the economic welfare loss of water scarcity to private households to be just as
real as economic losses to industrial enterprises (including agriculture), although the former losses
may not become recorded in financial accounts. This approach of treating economic welfare losses to
households as real is in line with standard economic practice in Social Cost-Benefit Analysis and is
for example also used in the Californian CALVIN (California Value Integrated Network) economicengineering optimization model of Californias water supply system (Jenkins et al. 2004).
To construct demand curves, information is needed on market prices, the slope of the curves, and
baseline water demands for the sectors across Europe. Baseline water demand (m3) is in the base data
of LISQUAL (see above). There is considerable heterogeneity in water prices across uses and regions
in Europe. Figure 14, shows average residential water prices (including charges for sanitation) across
Europe (Kuik 2012). There is very little empirical information on the specific form of the demand
curves for water of industry and energy. Based on the review of literature of Kuik (2012), a price
elasticity of -0.71 has been used for all industrial and energy sectors.
There is more empirical evidence on the relationship between price and demand for water for the
residential sector. Based on the review of European studies by Kuik (2012), this study uses countryspecific price elasticities that range from -0.187 to -0.402.
25

There is also more empirical evidence on the relationship between the price of irrigation water and
demand in agriculture. The evidence, however, shows extreme variation across locations and crops.
This study uses an extremely simplified linear demand curves that are cut-off from above by a socalled choke price, i.e. a price of irrigation water at which irrigation is stopped, resulting in
maximum damage for agriculture. Depending on the ratio of irrigation water availability and the total
irrigation water demand, the economic loss is calculated. Choke prices of 0.35 Euro/m3 for low-value
crops and 1.25 Euro/m3 for high-value crops are used, according to data provided by Kuik (2012).

Figure 16 Water prices used in this study (source: Kuik 2012)

The calculation of the economic losses across scenarios is a very rough first approximation, due to the
lack of data on location- and use-specific water prices, the dearth of empirical estimates of the
sensitivity of demand for water of industrial and energy firms to different water prices, and the huge
variety and context-specificity of agricultural demand for irrigation water. It is even impossible to say
at this moment whether our assumptions lead to an over- or underestimate of the real economic costs
of water scarcity. There is a huge scope for improvement in this area.

26

Figure 17 Example of the Economic Loss Model which estimates the loss (in Euro's) if water availability (m3) is less
than the water demand (the value at zero loss). For cost of water 0.6 Euro per m3 is used, and for price elasticity 0.71.

27

Modelling methods
CAPRI
CAPRI stands for Common Agricultural Policy Regionalised Impact analysis and is both the
acronym for an EU-wide quantitative agricultural sector modelling system and of the first project
centred around it. CAPRI was developed with European Commission research funds, the first phase
being 1996-1999. Operational for more than a decade (1999), the CAPRI model supports decisionmaking related to the Common Agricultural Policy and Environmental policy related to agriculture
based on sound scientific quantitative analysis.
CAPRI is a global, comparative static partial equilibrium model for primary and secondary agricultural
commodities, designed for ex-ante impact assessment. The model has a European focus but trade
policies from seventy-seven countries, divided into 40 trade blocks, are incorporated through the
global, multi-commodity market module. The agricultural sector for the EU27 plus Turkey, Norway
and the Western Balkans is taken into account within the supply module through 280 regional models
or 1900 farm-regional models. Together, the market and supply module are iteratively linked. The
outputs include farm supply details, such as land requirements for commodities on a NUTS 2 level.
Certain variables, namely crop shares, yields, stocking densities, and fertilizer application rates, are
downscaled to 1x1 km grid cells.
CAPRI builds upon an analysis of observed historical trends, on expert information for particular
issues, and on standard economic modelling.
The CAPRI baseline used for this study was generated in the frame of the EC4MACS project, funded
by the EU-LIFE program.
The baseline for agricultural policies includes the following assumptions:

The Health Check of the Common Agricultural Policy (CAP)

Agricultural premiums are largely decoupled from production levels

Biofuels as projected by PRIMES (see details later)

Impacts of the Nitrates Directive

The economic and sectorial outlook for the EU economy until 2030 is based on the latest short- term
forecasts published by DG-ECFIN in May 2009 and the DG ECFIN Ageing 2009 report of April 2009,
which include projections of long-term GDP, population and labour force developments. The multisector and multi-country general equilibrium model GEM-E3 has been used in EC4MAC to ensure
internal consistency among the relevant aspects while reproducing the short-term GDP forecasts as
well as the long-term GDP and demographic projections of DG-ECFIN. The baseline scenario
assumes for the EU-27 that economic recovery will compensate for some of the loss in GDP during the
recent recession, but in the longer term will not entirely compensate these GDP losses.
An extensive assessment of uncertainties related to the CAPRI model has been performed in the frame
of EC4MAC and is available at:
http://www.ec4macs.eu/content/report/EC4MACS_Publications/UNC_Rep_final%20in%20PDF/CAP
RI_Uncertainties_Final.pdf
The outputs from CAPRI are used to drive the agricultural land use allocations performed by LUMP in
the various simulations described in this report. This ensures consistency between the economic and
market assumptions (CAP compliant) and the physical results of the allocation. During the simulation,
the socio-economic assumptions set in the CAPRI baseline are not modified.
28

Furthermore, the CAPRI output is used for the EPIC N & P modelling used in this project.
LUMP
The JRC Land Use Modelling Platform (LUMP) is used to simulate CAP-consistent land use scenarios
for the year 2030 at a 100x100m spatial resolution at a pan-European scale. This scenario is used as a
baseline for calculations in this study.
For several specific natural water retention measures, specific runs are done with LUMP to simulate
alternative land use patterns for 2030. This is done for e.g. afforestation scenarios in riparian areas,
afforestation in mountainous areas, and a CAP-consistent run converting arable land into grassland.
Upscaling to the 5km spatial resolution LISFLOOD hydrological model is done with a series of 80
tables defining sub-grid variability within a 5km grid, aimed at maintaining the detailed 100m LUMP
simulations.
Land use/cover scenario baseline 2030
Future land claims for arable land and pasture were derived from the CAPRI-PE 2030 scenario with the assumption of nationalflat rates.
Future land claims for urban land were derived from Eurostat data (EUROPOP2008), based on a single convergence scenario,
whereby demographic structural differences between countries are assumed to fade out by 2150.
Land use change from forest or semi-natural vegetation to agricultural land is only allowed outside protected areas (i.e. Natura
2000).
Land use change from agricultural land to urban or industrial land is only allowed outside protected areas (i.e. Natura 2000).
Abandoned land is driven by economic factors, i.e. emerges as a result of the decline in agricultural claims, and thus its
definition does not take directly into consideration any other variable related with economic or demographic conditions (e.g.
holdings with low income or proportion of farmers close to retiring age).
Land use change to arable land and permanent crops is encouraged in Less Favoured Areas (art.18 and 20) and discouraged in
environmental sensitive areas: a 50m strip width along water courses in currently designated Nitrate Vulnerable Zones; and in
erosion sensitive areas (where erosion is between 20 and 50 ton/ha/year and higher than 50 ton/ha/year).

29

LISFLOOD
LISFLOOD is a GIS-based spatially-distributed hydrological rainfall-runoff model developed at the
JRC. It includes a one-dimensional hydrodynamic channel routing model (De Roo et al., 2000; Van
der Knijff et al., 2010). LISFLOOD is currently used at the JRC for simulating water resources in
Europe and Africa. Driven by meteorological forcing data (precipitation, temperature, potential
evapotranspiration, and evaporation rates for open water and bare soil surfaces), LISFLOOD calculates
a complete water balance at a daily time step and every grid-cell. Processes simulated for each grid
cell include snowmelt, soil freezing, surface runoff, infiltration into the soil, preferential flow,
redistribution of soil moisture within the soil profile, drainage of water to the groundwater system,
groundwater storage, and groundwater base flow. Runoff produced for every grid cell is routed
through the river network using a kinematic wave approach. Although this model has been developed
with the aim of carrying out operational flood forecasting at the pan-European scale, recent
applications demonstrate that it is well suited for assessing the effects of land-use change and climate
change on hydrology (Feyen et al., 2007; Dankers and Feyen, 2009).
To account properly for land-use dynamics, some conceptual changes have been made to render
LISFLOOD more land-use sensitive. Combining land-use classes and modelling aggregated classes
separately is known as the concept of hydrological response units (HRU). This concept is used in
models such as SWAT (Arnold and Fohrer, 2005) and PREVAH (Viviroli et al., 2009) and is now
implemented in LISFLOOD on the sub-grid level. A forest fraction map, water fraction and direct
runoff fraction have been derived from the 100m resolution land use Land Use Modelling Platform
(LUMP) maps. The spatial distribution and frequency of each class is defined as a percentage of the
entire 5 x 5 km grid. To address the sub-grid variability in land use, we model the within-grid
variability by running the soil modules separately for fractions of land use.

Figure 18 Input data used for the European LISFLOOD model setup

30

The model has also options to simulate lakes, reservoirs, and retention polders, which are relevant for
low-flow analysis (as they tend to increase low flows) as well as for simulating flood protection during
high flows. In the current setting for Europe, 173 lakes and reservoirs are included. Several large lakes
and reservoirs are also included for Africa.
The current pan-European setup of LISFLOOD uses a 5-km grid and spatially variable input
parameters and variables obtained from European databases (Figure 1). Elevation data are obtained
from the Shuttle Radar Topography Mission (SRTM) (Farr et al., 2007) and river properties were
obtained from the Catchment Information System (Hiederer and de Roo, 2003). Soil properties were
obtained from the European Soil Geographical Database (King et al., 1994) whereas porosity,
saturated hydraulic conductivity and moisture retention properties for different texture classes were
obtained from the HYPRES database (Wsten et al., 1999). Vegetative properties and land use cover
were obtained from the JRC Land Use Modelling Platform (LUMP). The meteorological and
hydrological data used are described in the next chapter.
For Africa, a 0.1-degree resolution is established.
The LISFLOOD model output can be any internal variable calculated by the model, either as time
series, summary maps or stacked maps over the complete time period. Examples of output are
discharge hydrographs, summary maps of evapotranspiration, soil moisture or groundwater recharge.
For modelling water supply and water demand, the model output is the daily accumulated amount of
surface and groundwater in millimetres for each grid cell (daily local runoff).
With a grid size of 5 x 5 km (Europe) and 0.1 x 0.1 degree (Africa), LISFLOOD is developed for
simulating medium and large river basins. Satisfactory results can be obtained in basins of a few
hundred km2 up to the size of the entire Danube basin. A limiting factor is the availability of good,
accurate and homogenous input data for the entire pan-European or pan-African scale, for example soil
data or measured discharge data. Human influences (e.g. dams, reservoirs, polders, irrigation) also are
difficult to quantify. This is an especially important factor for low-flow simulations.
With a 5-km grid resolution, the current European-wide model setup uses spatially variable parameters
on soil, vegetation and land use derived from European datasets. The Shuffled Complex Evolution University of Arizona (SCE-UA) algorithm (Duan et al., 1992) was selected for carrying out the
calibration of the LISFLOOD model at the European scale. A set of 9 parameters that control
infiltration, snowmelt, overland and river flow, as well as residence times in the soil and subsurface
reservoirs, have been estimated for 479 catchments by calibrating the model against historical records
of river discharge. The results of 435 of these catchments have been used to set up the model, while
44 were rejected because of unreliable data input. The calibration period varied between the different
catchments depending on the availability of discharge measurements, but all spanned between 1 and 9
years between 1996 and 2009. A more detailed description of the calibration of LISFLOOD for
different European catchments is given by Feyen et al. (2007, 2008).

31

Figure 19 Location of the 435 discharge gauging stations used in the calibration of the hydrological model

The location of the stations used in the calibration and validation are represented by the black dots in
Figure 2. This overview shows that the coverage is sufficient in most parts of northern and central
Europe. For the Balkan area, Greece, southern Italy and parts the Iberian Peninsula, no discharge series
were available at the time of the model calibration. For catchments where discharge measurements
were not available, simple regionalisation techniques (regional averages) were applied to obtain the
parameters.

Figure 20 Observed versus simulated averages, 95% and 99% discharge for each of the 435 calibration stations for
a 3-year validation period

Figure 3 shows observed versus simulated averages, and 95% and 99% discharge for each of the 435
calibration stations shown in Figure 2, for a 3-year validation period. This period varyies between the
different catchments depending on the availability of discharge measurements but includes the latest
available data. Visual inspection and the values for the coefficient of determination (r2) show that the
observed flow statistics are reasonably well reproduced by the LISFLOOD simulation with a general
tendency towards better performance for average flows and with increasing catchment size.
32

Notwithstanding the overall good agreement between the observed and simulated flow statistics, large
discrepancies do occur at a small number of stations. Deviations from the observation-based statistics
may be attributed to errors in meteorological forcings, the spatial interpolation of meteorological data,
as well as to errors in the hydrological model, its static input and the calibration of its parameters.
Some of the differences may also be due to manmade modifications of flow regimes in many
catchments, which are not accounted for in the hydrological model.

33

EPIC for N & P fluxes


Nitrogen and phosphorous fluxes from agriculture are calculated with the pan-European set-up of the
EPIC model, ran at 10x10km grid resolution.
A land use map (LUM) was developed combining the land use distribution of CORINE Land Cover
2000 with the FSS statistics on crop areas for the year 2000. 43 crop categories were distinguished
according to FSS crop categories. The crop areas were distributed to arable land, grassland and
permanent cropland of the CORINE data set. FSS crop areas were directly assigned to corresponding
land use classes where possible. Field crops were distributed to irrigated and non-irrigated agricultural
land. Where FSS crop areas do not match CORINE land cover data, a set of enlargement and shrinking
rules was applied to ensure consistency with FSS data. Altogether, the LUM contains 98 land use
classes in 1ha resolution for EU25. The land use raster grid with 100m resolution was then tabulated to
obtain crop areas per 10km grid cells.
For each of the crop available in the land use map, fertilization was taken from CAPRI in the form of
manure (consistent at NUTS2 level with the number of animal unit present reported by EUROSTAT)
and mineral fertilizer consistent at NUT0 with the national data reported to EUROSTAT. Baseline
irrigation was calibrated for year 2000 and then auto-irrigation was used for all scenarios for the
irrigated crops. For the baseline EPIC was run with the observed weather taken from the JRC MARS
database.
For the optimized fertilization, EPIC was run then using an auto-fertilization option where an
application of fertilization is performed each time a crop is under nitrogen stress. For catch crop we
use non fertilized alfalfa
Four scenarios were run:
Baseline scenario with crop distribution and fertilizer application per crop from CAPRI
Scenario identical to the previous one but with a winter/catch crop cover
Scenario with optimum fertilization with crop distribution identical to scenario 1
Scenario with optimum fertilization with crop management identical to scenario 2 (crop from
CAPRI with winter/ catch crop)
Using EPIC, the following fluxes are calculated for each scenario with a daily timestep, but are used
with a monthly average values in LISQUAL

Nleach (nitrogen from leaching)

Pleach (phosphorous from leaching)

Nrunoff (nitrogen in surface runoff)

Prunoff (phosporous in surface runoff)

34

LISQUAL
The aim of the LISQUAL model is to be able to carry out integrated simulations of water quantity and
quality on continental and sub-continental scale, in order to assess a variety of measures and policy
options. The model is taking into account water abstraction, demand and consumption. Furthermore,
the model includes principles such as environmental flow and the water exploitation index. Also, the
model contains an economic loss estimation routine which estimates the economic loss if water
availability is not sufficient to meet the demand.
Using these individual daily fluxes allows later on in the LISQUAL model routing of water after
additional abstractions or return flows.
Spatial and Temporal Inputs
All inputs specified below are spatial inputs, i.e. values can be assigned to every single 5*5km grid,
the spatial resolution of the model. The temporal resolution is 1 day, with where appropriate daily
input data, monthly average input data, or annual average input data.
Water quantity (from LISFLOOD runs)
o Total runoff produced in cell, contributing to river
o Individual water fluxes (required for chemical load estimates)
 Surface runoff
 Preferential flow
 Subsurface flow
Water quality (from EPIC runs):
o N concentration in runoff from agricultural land
o N concentration in leaching from agricultural land
o P concentration in runoff from agricultural land
o P concentration in leaching from agricultural land
o Equilibrium Phosphate concentration (Error! Reference source not found.)
o Average atmospheric Nitrogen input
Point sources
o N & P point sources from waste water treatment
Additional meteorological data:
o Daily 2m surface temperature (to estimate water temperature)
o Potential Evapotranspiration data (to estimate lake evaporation)
Topographic data
o River flow network;
o River dimension (cross section) parameters;
o MacroRegions: 21 large European regions (no water transfers between those regions);
these consist of one or several entire river basins;
o WaterRegions: 1246 smaller European sub-regions; these regions are used as units for
computing e.g the water exploitation index; water transfers between these regions - but
within a certain MacroRegion are possible;
Land use data
o Fraction of arable land within 5*5km pixel
o Fraction of sealed (urban) area within 5*5km pixel
Water abstraction, demand and consumption data
o Average monthly irrigation water need (from EPIC)
o Average daily abstraction of water for households
o Average monthly livestock water need
o Average annual industrial water abstraction
o Average annual water abstraction for energy production
35

o Current irrigation efficiency


o Current leakage loss of water
o Current desalination installations and their capacity
o Current water re-use in industry
o Current percentage water savings in households
Economic data
o Water price
o Price elasticity for domestic water use
o Comparative price levels
For scenarios
o Any change in one of the above parameters
o Additional irrigation efficiency
o Leakage reduction
o Percentage water re-use in industry
o Percentage water savings in households
o New desalination plants with their capacity and area of water use

Figure 21 Equilibrium Phosphorous Concentration. Values are a function of lithology

36

Model equations

Kinematic wave routing of river discharge (from LISFLOOD), using the LISFLOOD
calibration parameterization, which is also used in the operational EFAS flood warning system;
Routines to simulate the effects of lakes and reservoirs;
Daily routing of surface and river N and P, as a function of flow velocity;
Exponential first order - removal of Nitrogen, as a function of water temperature, water depth
and flow velocity;
First order Phosphorous removal, taking into account an equilibrium phosphorous
concentration depending on sediment characteristics, derived from geological maps
Irrigation water use, taking into account irrigation demand, and irrigation efficiency
Industrial water use, taking into account abstractions, consumptive use and potential re-use of
water
Domestic water use, taking into account abstraction, leakage, and water savings
Livestock water use, taking into account abstraction and consumptive use

Environmental Flow
The WFD does not specify the term environmental flows but requires the flow regime should provide
conditions consistent with the achievement of the values specified for the Biological Quality
Elements. In this context, environmental flows to achieve the Good Ecological Status can be defined
as the hydrological regime necessary to achieve the values specified for the biological quality elements
in order to be classified as Good Status. The concept of environmental flow is taken into account by
calculating the 10th and the 25th percentile of daily river discharge at each location on a monthly basis,
based on model simulations of a period of 30 years (1981-2010), while not including the current water
abstraction and consumption. This method is based on work by Sanchez & Schmidt (2012) and
bilateral discussions with JRC staff (W. Van De Bund, D. Van Ham). The method here corresponds to
a simplified application of the Range of Variability Approach (RVA), as part of the Level1Preliminary Eflow Assessment.
An example for the 10th percentile map for June is given below. Next, the LISQUAL routine flags
every day that the environmental flow threshold is not respected -i.e. discharge is lower than the
environmental flow thresholds.
At the end of a scenario simulation run, the total number of days at which discharge is lower than the
10th percentile is calculated. If this values exceeds 10 percent of the days so for this report larger than
37 days per year the scenario does not fulfil the environmental flow requirements. If for a scenario
the number of days that the environmental flow threshold is exceeded is larger than the baseline, this
indicates that the scenario has negative consequences in view of the environmental flow principle.

37

Figure 22 Example Environmental Flow map: 10th percentile for July.

Water Exploitation Index


A Water Exploitation Index is calculated along the lines of the Water Scarcity and Drought Expert
Group (WSDEG) 2012 document on indicators.
Within LISQUAL, a Water Exploitation Index + (here called WEIcns) is calculated as:
WEIcns = (abstraction return flow) / (external inflow + internal flow)
With:
Internal flow = net generated water (rainfall evapotranspiration + snowmelt)
External inflow = inflow from upstream areas
Abstraction = water abstraction in the region
Return flow = water abstraction minus water consumption in the region
(Water consumption = abstraction minus return flow)
The index is calculated for single entire years (from 1st October until 1st October) for the entire
simulation period, in this case 30 years, and then averaged. A monthly calculation would be
38

technically possible but requires detailed information and simulation of seasonal water storages, which
is a considerable effort in data acquisition, as is recognized by the WSDEG.
In addition, a WEI(abs) is calculated as follows:
WEIabs = abstraction / (external inflow + internal flow)
which is essentially the WEI+ (WEIcns) without the return flow

Figure 23 Example of the WEI+(cns) indicator, here fore the Baseline2030 scenario.

Instead of calculating the 30-year average of the individual annual WEIabs indicators, we also
investigated the use instead of a median (50th percentile, WEI50) WEI, the maximum WEI (100th
percentile, WEI100), or the intermediate 80th and 90th percentile WEI (WEI80 and WEI90). The 80th,
39

WEIabs100 (-)

WEIabs90 (-)

WEIabs80 (-)

WEIabs50 (-)

WEIabs (-)

WEIcns (-)

Region

90th and 100th percentile give the WEI for one or more extreme years within the 30-year simulation
period

01 N. Scandinavia
0.001 0.004 0.004 0.005 0.005 0.007
02 S. Scandinavia
0.004 0.019 0.018 0.022 0.024 0.031
03 Baltic
0.018 0.061 0.059 0.075 0.085 0.115
04 Denmark/N.Germany 0.025 0.095 0.090 0.114 0.120 0.153
05 Odra/Vistula
0.082 0.282 0.270 0.362 0.416 0.568
06 Elbe to Ems
0.073 0.281 0.266 0.346 0.391 0.533
07 Rhine/Meuse/Scheldt 0.043 0.160 0.155 0.185 0.201 0.265
08 GB
0.021 0.073 0.074 0.088 0.095 0.105
09 Irland/N.Ireland
0.006 0.017 0.017 0.019 0.021 0.026
10 France Atlantic
0.050 0.100 0.095 0.119 0.138 0.209
11 Danube
0.143 0.238 0.224 0.308 0.355 0.475
12 Iberia Atlantic
0.029 0.061 0.058 0.075 0.084 0.131
13 Iberia Mediterranean
0.198 0.322 0.301 0.398 0.466 0.784
14 France Mediterranean 0.027 0.069 0.067 0.084 0.094 0.116
15 Po
0.075 0.154 0.146 0.181 0.211 0.272
16 Corsica
0.008 0.061 0.056 0.075 0.084 0.118
17 Sardinia
0.164 0.232 0.201 0.308 0.345 0.502
18 Sicily
0.358 0.475 0.448 0.593 0.694 0.921
19 South Italy
0.197 0.297 0.282 0.372 0.407 0.596
20 Adige/Balkan
0.020 0.045 0.042 0.055 0.064 0.073
21 Greece/Evros
0.097 0.159 0.150 0.202 0.232 0.288
Figure 24. WEI indicators for the 21 European macro-regions for the Baseline 2030 scenario.

Later on in the report, well keep on using the average WEIabs and WEIcns. As a rule of thumb, figure
24 indicates that WEI90 is around 50% larger than the average WEI, and WEI100 the maximum
WEI in a single year is approximately double the average WEI. However, scenarios tend to react in a
non-linear way to changes, so this rule of thumb should be taken with care.

40

Figure 25 WEIAbs 50th, 80th, 90th and 100th percentile indicators for the Baseline 2030 scenario. The 50th
percentile is the medium Water Exploitation Index, whereas the 100th percentile represents the maximum Water
Exploitation Index in a single year within the 30years simulated.

41

Model output
The LISQUAL model produces the following output. In addition, any internal model variable could be
output as well:
Spatial output (maps) used in the Blueprint optimization process:
Discharge
o Daily discharge maps
o Maps with maximum annual discharge
 Used to calculate flood return periods

Figure 26 50year return period river discharge

Average NO3 concentration in rivers


Average PO4 concentration in rivers
Water Exploitation Index
o WEI-abs (not including return flow, using abstraction only)
o WEI+ (WEI-cns, including return flow, using net consumption)
Environmental Flow indicator
o Number of days with discharge lower than 10th percentile environmental flow
o Number of days with discharge lower than 25th percentile environmental flow
Costs
o Costs of the scenario
42

 Calculated before the start of a run


Economic Losses (further explained on page 25)
o Expected Annual Flood damage as a consequence of the scenario
 Calculated in post-processing, after a run
o Estimated economic loss for the domestic sector
o Estimated economic loss for the industrial sector
o Estimated economic loss for the energy sector
o Estimated economic loss for the agricultural sector

Model assumptions
Environmental flow: thresholds 10 & 25 percentile of daily discharge are used, per month, so
the January 10-percentile is used to evaluate January daily discharges in the scenarios, etc.
o Env10flw monthly maps
o Env25flw monthly maps

So called water regions are used to calculate the Water Exploitation Index Plus indicator.
These regions are typically single river basins or subbasins, but transboundary catchments are
split in separate regions following country borders.

Irrigation water use:


o Effective crop water use from irrigation is estimated using the EPIC model by JRC;
actual irrigation water abstraction is larger due to an efficiency lower than 100%.
o An initial CurrentIrrigationEfficiency is used, which aims to mimic irrigation
efficiency as in 2010. According to Kuik (2012), this efficiency is around 77% in
Western Europe, and 74% in Central in Eastern Europe; The Efficiency is expressed as
a fraction (0.77 and 0.74).
o An AdditionalIrrigationEfficieny is used for scenario calculations; it is used as a
fraction, so can at maximum be 0.25, totaling 100%
Map: scenirgf.map
Only applied in irrigation areas
o It is assumed that 5% of the not-efficiently used irrigation water (the difference
between abstraction and crop use) is lost (evaporated), while 95% of that part is return
flow.

Manufacturing Industry water use:


o Industrial water abstractions are estimated by JRC using Eurostat and similar data
sources, using the 100m landuse data to downscale;
o Consumptive water use in industry IndustryConsumptiveUseFraction is assumed at
0.15; values between 0 and 1; this is water which is not returned (evaporates during
cooling etc);
o A WaterReUseFraction (for baseline set at 0.0) can be used for scenario simulations:
e.g. 0.5 = 50% of the water used in industry is re-used) (baseline=0, maximum=1, all
water used twice)
Map: scenruse.map
Only applied in industry areas

Energy sector water use:


o Water abstractions for energy production are estimated using the methodology
described on page 21.
o As for consumptive use, a distinction is made between tower cooled / recirculating
energy production facilities, and once through cooled energy production facilities
(RWE Npower (2012)).
43

o Once through facilities are assumed at locations along large rivers and along the coast,
and are assumed to have a net water consumption of 2.5%.
o Tower cooled / recirculating facilities are assumed at locations elsewhere (along small
rivers), and are assumed to have a net water consumption of 33%.

Household water use and leaking:


o Household water abstractions are estimated by JRC using Eurostat and similar data
sources, using the 100m landuse data to downscale;
o Consumptive water use in households, in the model called the
DomesticConsumptiveUseFraction is assumed at 0.20; (source EEA 2007) values
between 0 and 1; this is water which is not returned (gardens etc);
o As for domestic water supply, actual reported leakage is taken into account with the
LeakageFraction. This is the fraction of leakage of public water supply (0=no
leakage, 1=100% leakage)
o An additional LeakageReductionFraction (for baseline set at 0.0) can be used for
scenario simulations: e.g. 0.5 = 50% leaking as compared to current leakage)
(baseline=0, maximum=1, no leaking anymore)
Map: scenleak.map
Assumed for scenario: 0.5 everywhere; baseline 0.0 everywhere
o An additional WaterSavingsFraction (for baseline set at 0.0) can be used for scenario
simulations: e.g. 10% = 0.1 as compared to current use (baseline=0, maximum=1)
Map: scenwsav.map
Assumed for scenario: 0.25 everywhere; baseline 0.0 everywhere
o It is assumed that 5% of the leaked water is lost (evaporated), while 95% of that part
is return flow.

Livestock water use:


o Livestock water abstractions are estimated by JRC using Eurostat and similar data
sources, using the 100m landuse data and actual temperature data to downscale;
o Consumptive water use is estimated using the LivestockConsumptiveUseFraction,
estimated at 0.15

Desalination:
o Desalination plants can be located anywhere at the coast, and the area needs to be
defined where the generated water is actually exported to, to estimate the costs
o Plants are here assumed to be 60 Megalitres per day
o Water is used max 150km inland, inside the same Major Region, but can contain more
WaterRegions
o Costs are calculated including distance to plant and height difference

To match water supply and water demand, the following assumptions are made:
Water can be used from the defined river (sub)-basin - no water transfers between river basins
are taken into account, because of a lack of continental data on this;
Water can be used only from within the country, so no major water transfers take place
between countries above the normal cross-border river flow;
Groundwater recharge volumes are included in the analysis, but fossil (non-renewable)
groundwater is not included;

44

The Optimisation routine

Investment ()

Optimization problems with multiple conflicting objectives lead to sets of trade-off solutions each of
which is no better or worse than the other. In general, natural resource development, use, and
management decisions involve multiple conflicting objectives and criteria, as well as incommensurable
units for measuring goods and services. In these cases, several optimization criteria need to be satisfied
simultaneously. Moreover, it is often not advisable to combine them into a single objective. While it
may sometimes happen that a single solution optimizes all of the criteria, the more likely scenario is
when one solution is optimal with respect to a single criterion while other solutions are best with
respect to the other criteria. The increase of the goodness of the solution with respect to one
objective will produce a decrease of its goodness with respect to the others. While there are no
problems in understanding the notion of optimality in single objective problems, multi-objective
optimization requires the concept of Pareto-optimality or Pareto efficiency.

Environmental impact
Figure 27 Example of Pareto optimal efficient (red) and non-efficient (blue) strategies according to environmental
impact and restoration investment criteria.

MOEA is an iterative search algorithm that is based on the principles of evolution (Goldberg 1989). A
solution is represented as a genome. The optimization starts with an initial population of
genomes (see Figure 27). In each iterative step the genomes of the population are evaluated with
a defined objective function and the fittest genomes are chosen to be recombined. The newly
generated solutions or offspring genomes also are evaluated and the least fit genomes are excluded
from the population to maintain the original population size. The MOEA algorithm applies the usual
45

procedures of selection (Tournament), crossover (Multi-point) and mutation (uniform) to generate the
new population. The Pareto fitness function and the tournament selection approach can be relatively
efficient methods of incorporating into fitness a points non-dominated tendencies. It also introduces
elitism by searching for and storing Pareto optimal points (separate from the general population) that
surface with each one.
This process is continually repeated for a given number of iterations know as generations: a big
population and a higher value of iterations is usually a guaranty of a better GA performance, but it
requires longer computation time to reach optimal solutions. The output of the optimization process is
a range of non-dominated solutions know as Pareto-optimal solutions that can be visualized in a two or
more dimensional plot to see the optimal solution tradeoff between the different objective functions

Interface for Model Optimisation (IMO)


The aim of the IMO interface is to be able to integrate the biophysical model LISQUAL with a multi
objective evolutionary algorithm (MOEA). The IMO interface consists of a Python/R/PCRaster script
interfacing the biophysical model LISQUAL (Processor) and the Optimization tool (Optimizer),
composed of 5 main parts (scripts):
1.
PreProcessor;
2.
Processor;
3.
PostProcessor;
4.
CostFunction;
5.
Optimizer.
InitProcessor

Optimizer
optimal combination
of percentage
scenarios

new percentages of
scenarios

OptInterface

PostProcessor
CostFunction

Statistical comparison
of scenarios and
baseline

PreProcessor

(scenarios-baseline) *r
OUTPUT
N
P
Cost
EnvFlow
Wei_abs

Simplified
Biophysical
Model

Figure 28 The Optimisation Cycle

46

INPUT
ttoc
srun
pflow
sgw
nleach
pleach
nrunoff

The IMO interface starts generating initially random percentages (fractions) of combinations between
the different scenarios (numbers [0,1] to be used for creating new combinations of scenarios), by an
initial processor (InitProcessor), e.g.:

Scenario Name
Combination
Name
Percentages of
combination (1st
iteration)
2nd iteration
3rd iteration

Afforest

Urban 25

Crop

Irrigation

Watersaving

Leakeage

Combination1

Combination2

Combination3

Combination4

Combination5

Combination6

0.425

0.017

0.865

0.378

0.277

0.518

0.941
0.926

0.899
0.418

0.945
0.832

0.128
0.997

0.951
0.512

0.541
0.782

The percentages of a scenario (fraction between 0 and 1) are assigned according to a so-called priority
map which is established for each individual measure/scenario. Highest priority (1) is given to areas
where water shortage is highest, and the measure is possible (Figure 30).
IMO does a large number of iterations to allow the optimization tool to select the best combination of
scenarios. At the end of each iteration, a new set of percentage numbers are generated, initially in
random way. After a block of 10 (or more) iterations the Optimizer starts a loop (generation) to select
the best solutions for each parameter.
Each iteration cycle runs, in sequence, the PreProcessor, the Processor, the PostProcessor, the
CostFunction and the Optimizer for each new combination. In each iteration, starting from the initial
random percentages, the PreProcessor generates a new combination of scenarios for each parameter
(ttoc, sgw, srun, etc.).
parameter(combination) = parameter(baseline2030) + (parameter(scenario1)-parameter(baseline2030))*random1 +
+ (parameter(scenario2)-parameter(baseline2030))* random2+
+ (parameter(scenario3)-parameter(baseline2030))* random3 +
Processor runs the LisQual model for the new combination. The inputs for LisQual are the following
parameters: ttoc, pflow, srun, sgw, nleach, nrunoff, pleach, prunoff, envflow, etc. The LisQual outputs
are: N, P Cost, EnvFlow and Wei_abs

PostProcessor compares the LisQual outputs of the new combination with their corresponding values
in the baseline, e.g.:

Result = (N(combination) N(baseline2030))/ (N(baseline2030))


If the output (e.g., N, P, EnvFlow, etc.) is an improvement over the baseline the Result will be a
negative number.

CostFunction calculates the cost for each new combination, since the sum of the differences for the
lossindu (reduction of cost of water use for the industry-manufacturing wateruse), lossdoms
(reduction of cost of water use for the domestic wateruse), lossagri (reduction of cost of water use for
the agricultural sector), lossener (reduction of the cost of water use for the energy production sector)
47

and loss_flood (reduction of cost to be paid in case of a 100 year flood) parameters between the new
combination vs baseline plus the cost to implement the measure (cost_implement_measure):

Cost(combination) = cost_implement_measure + lossdoms(diff) + lossindu(diff) + lossagri(diff) + lossener(diff) +


cost_flood(diff)
Where:
lossdoms(diff)
lossindu(diff)
lossagri(diff)
lossener(diff)
cost_flood(diff)

= lossdoms(combination) - lossdoms(baseline2030)
= lossindu(combination) - lossindu(baseline2030)
= lossagri(combination) - lossagri(baseline2030)
= lossener(combination) - lossener(baseline2030)
= cost_flood(combination) - cost_flood(baseline2030)

Figure 29 Priority map for the Afforestation Scenario, as an example input needed for the optimisation routine.

48

The use of water-regions and macro-regions

Figure 30 21 regions of Europe defined by river basins, climate and socio-economics

The scenarios are evaluated in 21 European macro regions (Figure 30). These regions consist of entire
river basins, e.g. region 13 consists of the entire Danube basin. Smaller river basins are merged with
one single region. These 21 regions are defined to avoid a single pan-European optimisation, but also
avoiding detailed optimisation which can be better done at regional/MS level. The regions are also
identified to separate major islands (Sicily, Sardinia, Corsica), for which it is unrealistic to assume that
water transfers take place with the main land. Since we wanted to avoid defining hundreds of regions,
smaller islands are included in continental macro regions (e.g. Baleares, attached to Iberia
Mediterranean). This may be unrealistic, but is done for practical reasons.
The comparison of water availability against water demand is carried out in so called water-regions,
which are much smaller areas than the macro-regions discussed above. The water-regions consist of
sub-riverbasins, within a single country. This is done to prevent that water scarcity in one macroregion is compensated and averaged out by a water surplus in another part of the macro-region.
The water-regions (Figure 31) are used for the computation of the Water Exploitation Index and the
Economic Loss for the various sectors. For all those water-regions the total available water (m3) is
compared to the needs for abstraction and consumption from various sectors.

49

Figure 31 The 'Water-Regions' used in the computation of the Water Exploitation Index and the Economic Loss due
to water scarcity

50

Sensitivity analysis and uncertainty


The coupling of the deterministic models used in this study - CAPRI for agro-economics, LUMP for
land use, EPIC for irrigation and nutrient fluxes, LISFLOOD for water quantity, and LISQUAL for
routing and hydro-economic estimations results in a vast number of spatial variables used, parameter
values, and several assumptions.
We have been collecting the required spatial information to the best of our ability, but it is clear that
improvements are possible for many variables. While detailed data maybe available in some Member
States, pan-European data however are in many cases not (yet) available.
Slight variations of these variables may lead to small or large variations of the model results. A
complete thorough sensitivity analysis could not be carried out within the timeframe of this study.
However, several sensitivity runs have been executed.
An analysis was done to investigate what would be the result if water abstraction would be reduced by
a factor 1.5 or increased by a factor 1.5 (Figure 32).

Figure 32 Sensitivity analysis of water abstraction for the Danube. Calculations are done for the baseline2030
abstraction (abs100), and a reduced abstraction (abs066), and an increased abstraction (abs150).

Figure 32 shows that especially the economic loss values calculated are very sensitive to the changes
in water abstraction. Changes in WEI and abstracted and consumed water are linearly related to
changes in abstraction values, but for example a 50% increase in abstraction leads to a 450% increase
of the economic loss in the domestic sector, and 250-300% increases in the industrial and agricultural
sectors.

51

Figure 33 Sensitivity analysis of water consumption for the Danube. Calculations are done for the baseline2030
abstraction (con100), and a reduced consumption (con050), and an increased consumption (abs200).

Another analysis was done to assess the effect of changes in the water consumption percentages used
(Figure 33). Changing water consumption percentages seems to be far less sensitive than changes in
absolute water abstractions.
The economic loss estimations are linearly related to the water price used (Figure 34). So any variation
or uncertainty in the water price results in a linear change of the economic loss.
0.35

0.3

Economic Loss (Euro)

0.25

0.2

0.15

0.1

0.05

0
0

0.5

1.5

2.5

Water Price (Euro)

Figure 34 The influence of water price on economic loss

52

3.5

4.5

Furthermore, the calculated economic loss is not very sensitive to changes in price elasticity used if
price elasticity is lower than -0.2. As soon as price elasticity approaches zero, the resulting economic
loss increases exponentially. Therefore, care should be taken be low values of price elasticity.

14

12

economic loss (Euro)

10

0
-2

-1.8

-1.6

-1.4

-1.2

-1

-0.8

-0.6

price elasticity (-)

Figure 35 The influence of varying price elasticity on the calculated economic loss

53

-0.4

-0.2

The Baseline conditions


In this study, we use two baseline scenarios: baseline-2006 and baseline-2030. The reference to
compare all scenario-runs to is called Baseline-2030. This baseline runs use the following data and
assumptions:
Both 2006 and 2030 Baseline scenarios:
Climate data 1981-2010 (source: KNMI, Ensembles project). These data are derived from the
GCM ECHAM5 coupled with the RCM RACMO2. The data set has been bias-corrected by
JRC using the observed meteorological data described in the data chapter.
Water abstraction data for the reference year 2006 are used for Livestock, both for 2006 and
2030 (an updated analysis for 2030 could not be finalised)
Economic data are mostly used for the year 2010, costs are corrected for individual country
price levels.
Consumptive use fractions are estimated based on literature values, and reflect current (2011)
consumption
o LivestockConsumptiveUseFraction is estimated at 0.15 (thus 85% is returned);
o IndustryConsumptiveUseFraction is estimated at 0.15 (thus 85% is returned);
o EnergyConsumptiveUseFraction for once through cooling facilities is estimated at
0.025 (thus 97.5% is returned) along large rivers and along the coast; for other
locations along small rivers where recirculating cooling towers are used a
consumption of 33% is assumed;
 Source: RWE Npower (2012), Torcellini et al. (2003)
o DomesticConsumptiveUseFraction is estimated at 0.20;
 Source: EEA (2005) State of Environment
o IrrigationWaterLoss is estimated at 0.75 (the water that is lost from ineffective
irrigation)
o LeakingWaterLoss is estimated at 0.75 (the water that is lost from leakage)
 Note: all these fractions remain constant for all scenarios
The Baseline2006 scenario:
Landuse 2006 is using the CORINE Land Cover 2006 refined map (Batista e Silva, F., Lavalle,
C., Koomen, E. (in press) A procedure to obtain a refined European land use/cover map, Journal of Land Use
Science).

Water abstraction data for the reference year 2006 are used for Industrial Manufacturing,
Energy Production, and Public Use (Domestic);
Water requirements for irrigation: average monthly water requirements are used based on
estimations using the EPIC model from 1981-2010 (a 30year period);
Nitrogen and Phosphorous fluxes from agriculture are estimated using EPIC, and based on
CAPRI baseline simulation; Since fluxes are calculated per km2 arable area, the fluxes in the
Baseline2006 reflect the fluxes related to the land use 2006 scenario.

The Baseline2030 scenario:


Landuse 2030 derived with the Land Use Modelling Platform (LUMP), starting from the 2006
Corine reference land use. This baseline scenario represents the current socio-economic and
environmental trends with current policy provision maintained (business-as-usual scenario).
No further specific policy options are implemented. This scenario is consistent with the Status
Quo scenario developed for the IA of CAP post-2013. The policy provisions taken into
consideration in the implementation of Baseline 2030 scenario are detailed as follows:
54

o Natura 2000: Council Directive 92/43/EEC of 21 May 1992 on the conservation of


natural habitats and of wild fauna and flora and Directive 2009/147/EC of the
European Parliament and of the Council of 30 November 2009 on the conservation of
wild birds;
o Nitrate Vulnerable Zones (NVZ): Council Directive of 12 December 1991 concerning
the protection of waters against pollution caused by nitrates from agricultural sources
(91/676/EEC);
o Erosion sensitive areas: the current GAEC framework (Council Regulation (EC) No.
73/2009, Annex III);
o Less Favoured Areas (LFA): this payment scheme promotes agriculture production in
areas with natural handicaps (Articles 18 and 20 of Council Regulation (EC)
1257/1999).
The main assumptions used in the Baseline 2030 scenario were:
o Future land claims for arable land and pasture were derived from the Extended version
of the CAPRI Baseline. Future land claims for urban land were derived from Eurostat
data (EUROPOP2008).
o Land use change from forest or semi-natural vegetation to agricultural land, and from
agricultural land to urban or industrial land is only allowed outside protected areas (i.e.
Natura 2000).
o Abandoned land is driven by economic factors, i.e. emerges as a result of the decline in
agricultural claims, and thus its definition does not take directly into consideration any
other variable related with economic or demographic conditions (e.g. holdings with
low income or proportion of farmers close to retiring age).
o Land use change to arable land and permanent crops is encouraged in Less favoured
Areas and discouraged in environmental sensitive areas: potential riparian areas in
currently designated Nitrate Vulnerable Zones; and in erosion sensitive areas (where
erosion is between 20 and 50 ton/ha/year and higher than 50 ton/ha/year).
o The CAPRI Baseline includes the biofuel information from the PRIMES reference run
which assumes that national targets under the Renewables directive 2009/28/EC and
the GHG Effort sharing decision 2009/406/EC are achieved in 2020.;
Modified water abstraction data for the reference year 2030 are used for Industrial
Manufacturing, Energy Production, and Public Use (Domestic);
Water requirements for irrigation: based on estimations using the EPIC model with 1980-2010
climate data, but with 2030 crops coming from CAPRI and the LUMP platform
Nitrogen and Phosphorous fluxes from agriculture are estimated using EPIC, and based on
CAPRI baseline simulation; Since fluxes are calculated per km2 arable area, the fluxes in the
Baseline2030 reflect the fluxes related to the land use 2030 scenario.

55

Costs of the scenarios per region


For each of the 21 European macro regions, the total costs of the individual measures are shown in the
tables below.

Figure 36 Total costs of individual scenarios per macro region (costs in Million Euro) (1st part)

These costs are calculated by multiplying the area of the applied scenario per 5*5km model grid with
the costs per hectare of the scenario, corrected with the comparative price level per country.
These costs are valid if the entire scenario is implemented in all the areas indicated by the land use
model LUMP or as assumed for the scenario.
Figure 38 shows and example of the investment and maintenance costs of the scenario increasing
irrigation efficiency

56

Figure 37 Total costs of individual scenarios per macro region (costs in Million Euro) (2nd part)

Figure 38 Cost of the irrigation efficiency scenario

57

Results of the Individual Scenarios


The Baseline Scenarios
In this section, the results of the Baseline2006 and the Baseline2030 scenarios are presented, and the
differences between them.
Indicator
Costs (MillionEuro)
TotalLoss (MillionEuro)
FloodDamage (MillionEuro)
Loss Agri (MillionEuro)
Loss Industry (MillionEuro)
Loss Domestic (MillionEuro)
Q50 (m3/s)
WEIcns (-)
WEIabs (-)
Env10 (days)
Env25 (days)
NO3 (mg/l)
PO4 (mg/l)
Abstracted (Mm3/yr)
Consumed (Mm3/yr)
Groundwater (Mm3/yr)

Description
The total costs of the scenario (implementation costs plus 30 years maintenance)
Total estimated economic losses of the scenario for the 30 year period
Total estimated flood damage (here: the cost of the 100yr return period flood)
Total estimated economic loss for agriculture as a consequence of water shortage (*)
Total estimated economic loss for industry as a consequence of water shortage (*)
Total estimated economic loss for the public sector as a consequence of water shortage (*)
River flow (m3/s) for a 50year return period flood (region average)
Water Exploitation Index, including return flow (region average)
Water Exploitation Index, excluding return flow (region average)
Environmental Flow: number of days per year with discharge less than 10-percentile (36.5 is threshold)
Environmental Flow: number of days per year with discharge less than 25-percentile (91 is threshold)
NO3 average annual concentration in the main rivers (region average)
PO4 average annual concentration in the main rivers (region average)
Total amount of water abstracted for use by various sectors (region total per year)
Total amount of water consumed by various sectors (region total per year)
Total amount of deep (geological) groundwater abstracted by various sectors (region total per year)
(*) average 2006 water availability is taken as the current water requirement at which loss is 0

Figure 39 Explanation of the indicators used in the scenario analyses below.

The indicators used for comparing the regions and the different scenarios are explained in Figure 34.
As explained before, for WEI the average annual WEI are included in the tables. We experimented
with assessing the 50th (median), 80th, 90th and 100th (max) percentile WEIs as well. As a rule of
thumb, the WEI90 is 50% larger than the average WEI, while the WEI100 is twice the WEI-average
value.
The differences between the baseline2006 and baseline2030 are due to changes in land-use, causing
e.g. different evaporation and infiltration response, and due to changes in water demand for the public
sector, irrigation, energy production and industrial manufacturing. Note: water use for livestock is held
constant due to time constraints of the analysis.

58

Groundwater (Mm3/yr)

Consumed (Mm3/yr)

Abstracted (Mm3/yr)

PO4 (mg/l)

NO3 (mg/l)

Env25 (days/yr)

Env10 (days/yr)

WEIabs (-)

WEIcns (-)

Q50 (m3/s)

Loss Domestic (MillionEuro/yr)

Loss Industry (MillionEuro/yr)


Loss EnergyProduction
(MillionEuro/yr)

Loss Agri (MillionEuro/yr)

FloodDamage (MillionEuro/yr)

TotalLoss (MillionEuro/yr)

Region

01 N. Scandinavia
259 259 0
0
0
0 313 0.001 0.003 37 92 0.18 0.04 2332
02 S. Scandinavia
178 178 0
0
0
0 283 0.003 0.013 41 94 0.68 0.06 3341
03 Baltic
58
58 0
0
0
0 317 0.012 0.035 73 122 0.53 0.05 4410
04 Denmark/N.Germany
24
24 0
0
0
0 41 0.022 0.082 101 145 3.67 0.21 2110
05 Odra/Vistula
368 368 0
0
0
0 333 0.044 0.162 135 174 1.63 0.09 12186
06 Elbe to Ems
2233 271 0 240 1243 479 282 0.047 0.196 124 164 3.46 0.12 14427
07 Rhine/Meuse/Scheldt
635 635 0
0
0
0 744 0.039 0.134 114 155 3.34 0.15 39651
08 GB
993 993 0
0
0
0 110 0.017 0.060 58 107 2.53 0.08 7932
09 Irland/N.Ireland
70
70 0
0
0
0 106 0.004 0.014 44 98 0.63 0.11 1107
10 France Atlantic
684 683 0
0
0
1 288 0.077 0.124 132 173 3.74 0.16 30320
11 Danube
1990 1142 57 25 535 232 896 0.138 0.202 141 180 1.08 0.11 56783
12 Iberia Atlantic
54
54 0
0
0
0 339 0.039 0.070 123 162 0.72 0.09 7597
13 Iberia Mediterranean
599 125 58 16 301 98 226 0.197 0.311 177 208 0.48 0.14 27117
14 France Mediterranean 157 157 0
0
0
0 565 0.031 0.068 110 148 0.71 0.12 11741
15 Po
440 431 0
0
7
1 707 0.076 0.146 145 176 0.83 0.10 16046
16 Corsica
0
0 0
0
0
0 90 0.007 0.057 102 147 0.33 0.01
336
17 Sardinia
3
3 0
0
0
0 108 0.171 0.229 175 200 0.63 0.08 2067
18 Sicily
1
1 0
0
0
0 75 0.365 0.475 211 228 1.40 0.13 3671
19 South Italy
151 121 4
7
3 16 198 0.194 0.285 175 199 1.38 0.12 20032
20 Adige/Balkan
28
28 0
0
0
0 234 0.020 0.041 78 123 0.49 0.12 3785
21 Greece/Evros
32
32 0
0
0
0 210 0.091 0.136 162 195 0.45 0.09 12406

484
124
863
382
1160
57
540
268
3051
61
3313
3
10425 1197
2258
0
353
0
17388 11115
34488 12906
4284 2608
17300 11280
4898 2408
8064 4235
40
21
1537 1062
2812 1953
13200 8270
1871
848
8607 5395

Figure 40 Summary indicators for the Baseline2006 scenario for the macro regions. Most values are region
averages, but the cost/loss/damage numbers are region total values, averaged for a year.

As explained before, the environmental flow indicator is based on thresholds calculated for current
rivers but without the current water abstraction and consumption. Figure 40 shows that almost
everywhere in Europe, except the Scandinavian regions and Ireland, the environmental flow indicator
is already much higher than the basic threshold. Very high values (>> 100 days per year, where 37
would be the number when environmental flow conditions are always met) are simulated in Sicily,
Iberia Mediterranean, South Italy, Sardinia and Greece/Evros. The Danube, Po and Oder/Vistula are
other areas with large values for the environmental flow indicator. The same regions are also the
regions with the highest average Water Exploitation.
Economic loss for various sectors due to water scarcity is simulated for the Elbe/Ems region, the
Danube region, Iberia/Mediterranean and Southern Italy.

59

N. Scandinavia
S. Scandinavia
Baltic
Denmark/N.Germany
Odra/Vistula
Elbe to Ems
Rhine/Meuse/Scheldt
GB
Irland/N.Ireland
France Atlantic
Danube
Iberia Atlantic
Iberia Mediterranean
France Mediterranean
Po
Corsica
Sardinia
Sicily
South Italy
Adige/Balkan
Greece/Evros

Groundwater (Mm3/yr)

Consumed (Mm3/yr)

Abstracted (Mm3/yr)

PO4 (mg/l)

NO3 (mg/l)

Env25 (days/yr)

Env10 (days/yr)

WEIabs (-)

WEIcns (-)

Q50 (m3/s)

Loss Domestic (MillionEuro/yr)

Loss EnergyProduction
(MillionEuro/yr)

Loss Industry (MillionEuro/yr)

Loss Agri (MillionEuro/yr)

FloodDamage (MillionEuro/yr)

TotalLoss (MillionEuro/yr)

Region
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21

250 250
0
0
0
0 313 0.001 0.004 37 92 0.170 0.037 3252
600
124
171 171
0
0
0
0 282 0.004 0.019 41 94 0.661 0.051 4533 1093
432
58
58
0
0
0
0 314 0.018 0.061 82 129 1.049 0.058 8144 1881
58
25
25
0
0
0
0 40 0.025 0.095 108 152 3.907 0.230 2437
621
281
567 370
0
12 160
26 326 0.082 0.282 172 205 2.591 0.095 20789 5567
140
9550 298
0 1135 5563 2554 277 0.073 0.281 143 181 5.231 0.123 20628 5171
92
1125 945
0
14 149
16 742 0.043 0.160 122 161 3.454 0.156 47962 11629 1105
5258 5258
0
0
0
0 113 0.021 0.073 65 111 2.586 0.081 9914 2818
80
185 185
0
0
0
0 107 0.006 0.017 45 98 1.081 0.135 1399
464
1
759 758
0
0
0
0 289 0.050 0.100 124 166 4.777 0.155 25718 11698 6352
4160 1203 141
80 2160 576 875 0.143 0.238 151 189 1.640 0.115 67242 35033 11311
57
57
0
0
0
0 340 0.029 0.061 119 159 0.769 0.093 7404 3950 2310
656 149 54
26 287 140 227 0.198 0.322 180 210 0.544 0.137 28833 18014 11629
158 158
0
0
0
0 563 0.027 0.069 106 145 0.650 0.113 11664 4018 1542
446 441
0
0
4
0 707 0.075 0.154 141 172 0.891 0.094 16919 7990 4065
0
0
0
0
0
0 89 0.008 0.061 102 148 0.339 0.015
360
49
26
2
2
0
0
0
0 107 0.164 0.232 175 200 0.709 0.083 2080 1469 1001
1
1
0
0
0
0 75 0.358 0.475 211 228 1.417 0.123 3662 2755 1902
183 132
7
16
4
24 199 0.197 0.297 174 197 1.346 0.118 20990 13373 8319
30
30
0
0
0
0 234 0.020 0.045 79 124 0.504 0.116 4119 1886
813
61
31
0
2
27
2 209 0.097 0.159 167 199 0.499 0.089 14061 9036 5395

Figure 41 Summary indicators for the Baseline2030 scenario for the macro regions.

Figure 41 shows for 2030 economic loss for various sectors due to water scarcity in the Odra/Vistula
region, the Rhine/Meuse region and the Greece/Evros region, as compared to no economic loss in the
2006 baseline.
Figure 42 shows the differences between the 2006 and the 2030 baseline runs. The results indicate
significant environmental flow deterioration in the Odra/Vistula, Elbe/Ems and the Danube, with more
than 10% more days that environmental flow conditions cannot be met, in areas that already have nonfavourable environmental flow conditions in the baseline2006 scenario: > 100 days/year where
discharge is less than the 10th percentile environmental flow threshold.
Large increases in water abstraction and consumption are observed in the Baltic region (85%),
Odra/Vistula (71%) and Elbe/Ems (43%). As a consequence, in those macro-regions the 50-year return
period discharge (Q50) is reduced, which leads to a slight reduction of flood damage potential in the
Baltic, whereby through land use changes the flood damage potential in the Elbe/Ems still is simulated
to increase with 9.8%.
The Water Exploitation indicators are deteriorating almost everywhere in Europe, with large changes
in the Odra/Vistula (from 0.162 to 0.282; 74% change), the Elbe/Ems (from 0.196 to 0.281, 43%
change) and the Danube (from 0.202 to 0.238; 18% change). For some regions such as France/Atlantic
and Iberia Atlantic, a small improvement in WEI is simulated.

60

Groundwater (%change)

Consumed (%change)

Abstracted (%change)

PO4 (change mg/l)

NO3 (change mg/l)

Env25 (change days/yr)

WEIcns (change)
0.000
-0.001
-0.006
-0.003
-0.038
-0.026
-0.004
-0.004
-0.001
0.027
-0.006
0.010
-0.001
0.004
0.001
-0.001
0.007
0.007
-0.003
0.000
-0.006

Env10 (change days/yr)

Q50 (m3/s change)


0.26
0.31
2.38
0.60
7.51
4.71
1.48
-2.99
-1.61
-1.05
21.22
-0.72
-1.37
2.45
0.63
0.25
1.47
0.49
-1.27
-0.16
1.32

WEIabs (change)

Loss Sectors (%change)

FloodDamage (%change)

0
3.49
0.0
0
4.09 -113.6
0
0.23
0.0
0
-3.26
-16.8
0
-0.35 -1000.0
0
-9.85 -371.5
0 -48.82 -1000.0
0 -429.74
0.0
0 -163.32
0.0
0 -10.99
42.6
0
-5.41 -248.3
0
-5.23
25.4
0 -19.28
-7.0
0
-0.75
-13.2
0
-2.33
44.9
0
4.01
31.9
0
8.77
0.0
0
2.96
0.0
0
-9.11
-66.4
0
-5.05
-36.5
0
3.94 -1000.0

Costs (MillionEuro)

Region
01 N. Scandinavia
02 S. Scandinavia
03 Baltic
04 Denmark/N.Germany
05 Odra/Vistula
06 Elbe to Ems
07 Rhine/Meuse/Scheldt
08 GB
09 Irland/N.Ireland
10 France Atlantic
11 Danube
12 Iberia Atlantic
13 Iberia Mediterranean
14 France Mediterranean
15 Po
16 Corsica
17 Sardinia
18 Sicily
19 South Italy
20 Adige/Balkan
21 Greece/Evros

-0.001 0.21 0.44 0.01 0.00 -39.5 -24.1


0.0
-0.005 -0.48 -0.14 0.01 0.00 -35.7 -26.6
-13.1
-0.027 -9.52 -7.53 -0.52 0.00 -84.7 -62.2
-0.7
-0.012 -6.46 -6.54 -0.23 -0.02 -15.5 -15.2
-4.7
-0.120 -36.78 -31.33 -0.96 0.00 -70.6 -82.5 -127.3
-0.084 -18.93 -16.61 -1.77 0.00 -43.0 -56.1 -1000.0
-0.026 -7.75 -6.58 -0.11 0.00 -21.0 -11.6
7.7
-0.013 -6.59 -4.25 -0.06 0.00 -25.0 -24.8
0.0
-0.003 -0.96 -0.07 -0.45 -0.02 -26.3 -31.6
0.0
0.023 8.67 7.02 -1.04 0.00 15.2 32.7
42.9
-0.036 -10.27 -9.04 -0.56 0.00 -18.4 -1.6
12.4
0.009 3.53 2.83 -0.04 0.00 2.5 7.8
11.4
-0.010 -2.21 -1.53 -0.06 0.00 -6.3 -4.1
-3.1
-0.001 4.17 2.99 0.06 0.00 0.7 18.0
36.0
-0.008 4.00 3.47 -0.06 0.00 -5.4 0.9
4.0
-0.004 -0.46 -0.50 -0.01 0.00 -7.4 -21.7
-24.7
-0.002 -0.34 -0.12 -0.08 0.00 -0.7 4.4
5.8
0.001 0.38 0.38 -0.02 0.00 0.2 2.0
2.6
-0.012 1.38 1.60 0.04 0.00 -4.8 -1.3
-0.6
-0.004 -0.81 -0.50 -0.01 0.00 -8.8 -0.8
4.2
-0.022 -4.82 -4.03 -0.05 0.00 -13.3 -5.0
0.0

Figure 42 Differences between Baseline2030 and Baseline2006 scenario. Most indicators are presented as %change
relative to 2006. Positive numbers indicate an improvement from 2006 to 2030. Negative numbers indicate a
deterioration from 2006 to 2030.

Figure 43 and further show the various Water Exploitation Indicators for 2006 and 2030 for the
waterregions in Europe. Figure 47 and Figure 48 show the differences between 2006 and 2030.
Areas with Water Exploitation + (WEI-consumption) currently (Figure 43) already above 0.400
(considered to have extreme water stress) are in Spain, Hungary, Rumania, Italy and Greece. Towards
2030 (Figure 44), WEI+ is strongly deteriorating in Northern Germany, Poland and Hungary.
The total surface area with Water Exploitation (WEI-abstraction) currently (Figure 45) already above
0.400 are obviously more areas than when only consumption is considered (Figure 43). Especially
areas with large numbers of population and industrial activity are also appearing then.
Figure 48 shows large increases in abstractions in Northern Germany, Poland, Hungary, Bulgaria,
Rumania and Lithuania for the 2030 period as compared to the 2006 baseline period. These changes
are not due to a climate signal (the climate has been kept constant here), but due to land use &
agricultural changes and water demand changes due to industrial and population changes.

61

Figure 43 Water Exploitation Index (consumption) for 2006

62

Figure 44 Water Exploitation Index (consumption) for 2030

63

Figure 45 Water Exploitation Index (abstraction) for 2006

64

Figure 46 Water Exploitation Index (abstraction) for 2030.

65

Figure 47 Change in WEI-consumption between 2006 and 2030

66

Figure 48 Change in WEI-abstraction between 2006 and 2030

67

01 N. Scandinavia
02 S. Scandinavia
03 Baltic
04 Denmark/N.Germany
05 Odra/Vistula
06 Elbe to Ems
07 Rhine/Meuse/Scheldt
08 GB
09 Irland/N.Ireland
10 France Atlantic
11 Danube
12 Iberia Atlantic
13 Iberia Mediterranean
14 France Mediterranean
15 Po
16 Corsica
17 Sardinia
18 Sicily
19 South Italy
20 Adige/Balkan
21 Greece/Evros

7945
7587
7587
0
0
0
0 313 0.001 0.004 1086 2703 0.17 0.04 3252
4842
4902
4902
0
0
0
0 282 0.004 0.019 1213 2814 0.65 0.05 4533
4062
1704
1704
0
0
0
0 314 0.018 0.061 2471 3865 1.05 0.06 8144
5342
676
676
0
0
0
0 40 0.025 0.095 3212 4531 3.76 0.22 2437
2378 16021 10278
3
339
4657
744 325 0.082 0.282 5151 6154 2.57 0.09 20789
6751 369955 60778
9 36770 180324 92073 273 0.074 0.283 4316 5449 5.07 0.12 20628
4367 33738 28530
0
415
4326
467 742 0.043 0.160 3647 4842 3.40 0.15 47962
2027 188473 188473
0
0
0
0 114 0.021 0.073 1912 3303 2.57 0.08 9914
1009
5785
5785
0
0
0
0 108 0.006 0.017 1340 2905 1.09 0.14 1399
4677 23268 23236
0
2
20
9 288 0.050 0.101 3712 4974 4.76 0.15 25718
9739 124636 35909 4236 2380 64790 17322 873 0.144 0.238 4549 5669 1.62 0.11 67242
1623
1789
1789
0
0
0
0 341 0.029 0.061 3582 4769 0.75 0.09 7404
1431 19578
4598 1575
760
8511 4134 228 0.197 0.320 5374 6279 0.54 0.14 28833
1218
4718
4718
0
0
0
0 562 0.027 0.069 3166 4361 0.65 0.11 11664
840 13806 13660
8
15
109
14 709 0.075 0.154 4223 5156 0.89 0.09 16919
15
5
5
0
0
0
0 89 0.008 0.061 3046 4403 0.33 0.02
360
108
78
78
0
0
0
0 108 0.163 0.231 5237 6004 0.70 0.08 2080
323
21
21
0
0
0
0 75 0.357 0.473 6319 6825 1.40 0.12 3662
1151
5563
4042 208
484
116
713 200 0.197 0.297 5211 5917 1.33 0.12 20990
227
911
911
0
0
0
0 234 0.020 0.045 2361 3710 0.50 0.12 4119
656
1858
947
0
62
801
48 209 0.097 0.159 5012 5960 0.49 0.09 14061

Groundwater (Mm3/yr)

Consumed (Mm3/yr)

Abstracted (Mm3/yr)

PO4 (mg/l)

NO3 (mg/l)

Env25 (days)

Env10 (days)

WEIabs (-)

WEIcns (-)

Q50 (m3/s)

Loss Domestic (MillionEuro)

Loss EnergyProduction (MillionEuro)

Loss Industry (MillionEuro)

Loss Agri (MillionEuro)

FloodDamage (MillionEuro)

TotalLoss (MillionEuro)

Scenario Costs (MillionEuro)

Region

The 11-Riparian zone scenario

600
124
1093
432
1881
58
621
281
5567
140
5171
92
11629 1105
2818
80
464
1
11698 6352
35033 11311
3950 2310
18014 11629
4018 1542
7990 4065
49
26
1469 1001
2755 1902
13373 8319
1886
813
9036 5395

Figure 49 Results of the 11-riparian scenario.

This scenario involves the afforestation of all potential riparian areas alongside rivers, consistent with
CAP targets. A LUMP run was executed. Taking as a reference the riparian zones identified by Clerici
et al. (2012) , the suitability for forest was be enhanced within potential riparian areas. Since this
dataset describes several degrees of probability of having riparian areas, the enhancement of forest
might reflect these differences.

68

Groundwater (%change)

Consumed (%change)

Abstracted (%change)

PO4 (change mg/l)

NO3 (change mg/l)

Env25 (change days/yr)

Env10 (change days/yr)

WEIabs (change)

WEIcns (change)

Q50 (m3/s change)

Loss Sectors (%change)

FloodDamage (%change)

Costs (MillionEuro)

Region

01 N. Scandinavia
7945
-1.31 0.0 -0.23 0.000 0.000 1.05 1.57 0.00 0.00 0.0 0.0 0.0
02 S. Scandinavia
4842
4.43 0.0 0.66 0.000 0.000 0.60 0.65 0.01 0.00 0.0 0.0 0.0
03 Baltic
4062
2.59 0.0 0.46 0.000 0.000 0.09 0.32 0.00 0.00 0.0 0.0 0.0
04 Denmark/N.Germany 5342 10.69 0.0 0.16 0.000 0.000 0.59 0.71 0.15 0.01 0.0 0.0 0.0
05 Odra/Vistula
2378
7.33 3.1 0.28 0.000 0.000 0.09 0.08 0.02 0.00 0.0 0.0 0.0
06 Elbe to Ems
6751 -580.33 -11.4 4.36 -0.001 -0.003 -0.74 -0.73 0.16 0.00 0.0 0.0 0.0
07 Rhine/Meuse/Scheldt 4367
-0.59 3.2 0.90 0.000 0.000 0.12 0.07 0.05 0.00 0.0 0.0 0.0
08 GB
2027 -19.49 0.0 -0.64 0.000 0.000 0.78 0.78 0.01 0.00 0.0 0.0 0.0
09 Irland/N.Ireland
1009
-4.11 0.0 -0.91 0.000 0.000 0.74 1.58 -0.01 0.00 0.0 0.0 0.0
10 France Atlantic
4677
-2.18 -47.0 0.20 0.000 0.000 -0.01 0.03 0.01 0.00 0.0 0.0 0.0
11 Danube
9739
0.53 0.0 2.36 0.000 -0.001 -0.20 -0.21 0.02 0.00 0.0 0.0 0.0
12 Iberia Atlantic
1623
-4.85 0.0 -0.51 0.000 0.000 0.09 0.08 0.02 0.00 0.0 0.0 0.0
13 Iberia Mediterranean
1431
-2.75 1.5 -0.61 0.001 0.002 0.38 0.38 0.00 0.00 0.0 0.0 0.0
14 France Mediterranean 1218
0.30 0.0 0.65 0.000 0.000 0.02 0.01 0.00 0.00 0.0 0.0 0.0
15 Po
840
-3.16 -1.4 -1.89 0.000 0.000 0.36 0.36 0.00 0.00 0.0 0.0 0.0
16 Corsica
15
-1.78 0.0 -0.17 0.000 0.000 0.90 1.08 0.01 0.00 0.0 0.0 0.0
17 Sardinia
108
-4.61 0.0 -0.65 0.001 0.001 0.34 0.35 0.01 0.00 0.0 0.0 0.0
18 Sicily
323
-2.93 0.0 -0.17 0.001 0.001 0.47 0.42 0.02 0.00 0.0 0.0 0.0
19 South Italy
1151
-2.30 0.4 -0.48 0.000 0.000 -0.09 -0.11 0.01 0.00 0.0 0.0 0.0
20 Adige/Balkan
227
-1.97 0.0 -0.20 0.000 0.000 -0.01 -0.02 0.00 0.00 0.0 0.0 0.0
21 Greece/Evros
656
-2.41 0.4 -0.19 0.000 0.000 -0.09 -0.04 0.01 0.00 0.0 0.0 0.0
Figure 50 Differences between the 11-riparian scenario and the baseline2030 scenario. Positive changes indicate that
the scenario is an improvement as compared to the baseline2030 conditions.

In these simulations, establishing riparian zones reduces extreme river flow (Q50) in some areas, e.g.
Elbe/Ems (1.6% improvement) & Danube (0.3% improvement), and reduces potential flood damage in
some areas as well, e.g. in the Danube region with 0.5%. However, there are other areas where the
opposite effect is simulated. Changes are subtle.
It should be noted that the riparian zone land use allocation has been carried out within the LUMP land
use model, while respecting the boundary conditions as set by CAPRI. Thus afforestation in one area
may be compensated by converting forest to other land use in other areas. This sometimes leads to
increased discharges while decreased discharges would be expected from the implementation of
riparian zones.

69

01 N. Scandinavia
02 S. Scandinavia
03 Baltic
04 Denmark/N.Germany
05 Odra/Vistula
06 Elbe to Ems
07 Rhine/Meuse/Scheldt
08 GB
09 Irland/N.Ireland
10 France Atlantic
11 Danube
12 Iberia Atlantic
13 Iberia Mediterranean
14 France Mediterranean
15 Po
16 Corsica
17 Sardinia
18 Sicily
19 South Italy
20 Adige/Balkan
21 Greece/Evros

7945
7644
7644
0
0
0
0 313 0.001 0.004 1081 2697 0.17 0.04 3252
4842
4919
4919
0
0
0
0 282 0.004 0.019 1215 2816 0.65 0.05 4533
4062
1711
1711
0
0
0
0 314 0.018 0.061 2472 3869 1.05 0.06 8144
5342
675
675
0
0
0
0 40 0.025 0.095 3213 4532 3.76 0.22 2437
2378 15809 10078
3
338
4647
742 324 0.082 0.282 5145 6149 2.56 0.09 20789
6751 369310 60516
9 36734 180172 91878 272 0.074 0.284 4317 5450 5.07 0.12 20628
4367 34464 29270
0
414
4315
465 742 0.043 0.160 3642 4838 3.40 0.15 47962
2027 188613 188613
0
0
0
0 114 0.021 0.073 1908 3299 2.57 0.08 9914
1009
5945
5945
0
0
0
0 108 0.006 0.017 1342 2907 1.09 0.14 1399
4677 23207 23186
0
1
13
6 287 0.050 0.101 3708 4971 4.77 0.15 25718
9739 139222 50378 4230 2378 64908 17327 871 0.144 0.238 4550 5668 1.61 0.11 67242
1623
1768
1768
0
0
0
0 342 0.029 0.061 3588 4779 0.74 0.09 7404
1431 19497
4411 1591
769
8532 4194 227 0.198 0.322 5400 6306 0.53 0.14 28833
1218
4609
4609
0
0
0
0 561 0.027 0.069 3170 4377 0.64 0.11 11664
840 14101 13956
8
15
109
14 710 0.075 0.154 4227 5161 0.89 0.09 16919
15
6
6
0
0
0
0 90 0.008 0.061 3048 4408 0.33 0.02
360
108
86
86
0
0
0
0 108 0.163 0.230 5200 5976 0.69 0.08 2080
323
21
21
0
0
0
0 75 0.358 0.474 6321 6829 1.39 0.12 3662
1151
5889
4374 207
482
115
711 201 0.198 0.298 5226 5935 1.33 0.12 20990
227
963
963
0
0
0
0 235 0.020 0.045 2362 3713 0.50 0.12 4119
656
1822
929
0
61
785
47 209 0.097 0.159 5031 5979 0.49 0.09 14061

Groundwater (Mm3/yr)

Consumed (Mm3/yr)

Abstracted (Mm3/yr)

PO4 (mg/l)

NO3 (mg/l)

Env25 (days)

Env10 (days)

WEIabs (-)

WEIcns (-)

Q50 (m3/s)

Loss Domestic (MillionEuro)

Loss EnergyProduction (MillionEuro)

Loss Industry (MillionEuro)

Loss Agri (MillionEuro)

FloodDamage (MillionEuro)

TotalLoss (MillionEuro)

Scenario Costs (MillionEuro)

Region

The 12-Afforestation scenario

600
124
1093
432
1881
58
621
281
5567
140
5171
92
11629 1105
2818
80
464
1
11698 6352
35033 11311
3950 2310
18014 11629
4018 1542
7990 4065
49
26
1469 1001
2755 1902
13373 8319
1886
813
9036 5395

Figure 51 Results of the 12-afforestation scenario.

Areas above 500m altitude and with slope gradients > 10% were afforested or reforested including
filling gaps within existing forested areas, non-CAP consistent. A LUMP run was carried out with a
strong enhancement of the conversion to forest from pastures, arable land, permanent crops and seminatural vegetation in areas above 500m and with slopes greater than 10%. It should be noted that this
scenario is based on the configuration of the Baseline 2030. However, the encouragement of the land
use change to arable land and permanent crops in Less Favoured Areas was not implemented, due to
the conflict between art.18 (mountains) of Council Regulation (EC) 1257/1999 and the aim of this
scenario.
Assumptions for this scenario are:

LUMP model is used with respect to the demand given by CAPRI

areas above 500 m altitude

slope gradients larger than 10%

70

Groundwater (%change)

Consumed (%change)

Abstracted (%change)

PO4 (change mg/l)

NO3 (change mg/l)

Env25 (change days/yr)

Env10 (change days/yr)

WEIabs (change)

WEIcns (change)

Q50 (m3/s change)

Loss Sectors (%change)

FloodDamage (%change)

Costs (MillionEuro)

Region

01 N. Scandinavia
7945
-2.07 0.0 -0.18 0.000 0.000 1.22 1.78 0.00 0.00 0.0 0.0 0.0
02 S. Scandinavia
4842
4.10 0.0 0.60 0.000 0.000 0.55 0.58 0.01 0.00 0.0 0.0 0.0
03 Baltic
4062
2.21 0.0 0.40 0.000 0.000 0.02 0.19 0.00 0.00 0.0 0.0 0.0
04 Denmark/N.Germany 5342 10.77 0.0 0.16 0.000 0.000 0.56 0.68 0.15 0.01 0.0 0.0 0.0
05 Odra/Vistula
2378
9.13 3.3 1.34 0.000 0.000 0.31 0.26 0.03 0.00 0.0 0.0 0.0
06 Elbe to Ems
6751 -577.40 -11.2 4.70 -0.001 -0.003 -0.78 -0.79 0.16 0.00 0.0 0.0 0.0
07 Rhine/Meuse/Scheldt 4367
-3.20 3.4 -0.04 0.000 0.000 0.28 0.19 0.06 0.00 0.0 0.0 0.0
08 GB
2027 -19.58 0.0 -0.71 0.000 0.000 0.92 0.91 0.01 0.00 0.0 0.0 0.0
09 Irland/N.Ireland
1009
-6.99 0.0 -0.96 0.000 0.000 0.67 1.51 -0.01 0.00 0.0 0.0 0.0
10 France Atlantic
4677
-1.96 4.1 1.49 0.000 0.000 0.10 0.10 0.00 0.00 0.0 0.0 0.0
11 Danube
9739 -39.56 -0.2 4.60 0.000 -0.001 -0.21 -0.17 0.03 0.00 0.0 0.0 0.0
12 Iberia Atlantic
1623
-3.61 0.0 -2.36 0.000 0.000 -0.12 -0.26 0.03 0.00 0.0 0.0 0.0
13 Iberia Mediterranean
1431
1.41 0.8 0.74 0.000 0.000 -0.50 -0.52 0.02 0.00 0.0 0.0 0.0
14 France Mediterranean 1218
2.62 0.0 1.45 0.000 0.000 -0.12 -0.55 0.01 0.00 0.0 0.0 0.0
15 Po
840
-5.39 -0.8 -2.91 0.000 0.000 0.22 0.19 0.00 0.00 0.0 0.0 0.0
16 Corsica
15
-2.57 0.0 -0.45 0.000 0.000 0.84 0.92 0.01 0.00 0.0 0.0 0.0
17 Sardinia
108 -14.15 0.0 -1.30 0.001 0.001 1.57 1.27 0.02 0.00 0.0 0.0 0.0
18 Sicily
323
-3.50 0.0 -0.05 0.000 0.000 0.42 0.29 0.03 0.00 0.0 0.0 0.0
19 South Italy
1151 -10.70 0.8 -2.03 -0.001 -0.001 -0.60 -0.72 0.01 0.00 0.0 0.0 0.0
20 Adige/Balkan
227
-7.76 0.0 -1.22 0.000 0.000 -0.06 -0.12 0.00 0.00 0.0 0.0 0.0
21 Greece/Evros
656
-0.43 2.4 0.09 0.000 0.000 -0.72 -0.67 0.00 0.00 0.0 0.0 0.0
Figure 52 Differences between the 12-afforestation scenario and the baseline2030 scenario. Positive changes indicate
that the scenario is an improvement as compared to the baseline2030 conditions.

In these simulations, establishing afforestation reduces extreme river flow (Q50) in some areas, e.g.
Elbe/Ems (1.7%), Danube (0.5%), Odra/Vistula (0.4%), France/Atlantic (0.5%), France/Mediterranean
(0.3%).
The scenario reduces potential flood damage in some areas as well, e.g. in the Odra/Vistula area with
9%, However, increases in flood damage are simulated in the Danube region with 40%, in the
Elbe/Ems area with more than 100%, and in the GB-region with 20%. The latter effect is probably to
be explained by compensating effects in the land use model used. It should be noted that the
afforestation in mountainous areas land use allocation has been carried out within the LUMP land use
model, while respecting the boundary conditions as set by CAPRI. Thus afforestation in one area may
be compensated by converting forest to other land use in other areas. This sometimes leads to
increased discharges while decreased discharges would be expected from the implementation of
afforestation.

71

01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21

N. Scandinavia
8499
7470
7470
0
0
0
0 313 0.001 0.004 1118 2750 0.17 0.04 3252
S. Scandinavia
17447
5114
5114
0
0
0
0 282 0.004 0.019 1231 2835 0.66 0.05 4533
Baltic
7663
1746
1746
0
0
0
0 314 0.018 0.062 2490 3891 1.05 0.06 8144
Denmark/N.Germany
10850
737
737
0
0
0
0 40 0.025 0.095 3238 4563 3.91 0.23 2437
Odra/Vistula
18389 17069 10955
3
364
4948
799 325 0.082 0.284 5176 6177 2.60 0.09 20789
Elbe to Ems
56965 304521
8352
9 35691 174994 85475 276 0.074 0.283 4313 5446 5.24 0.12 20628
Rhine/Meuse/Scheldt 111481 30771 25033
0
455
4746
536 741 0.043 0.161 3658 4853 3.46 0.16 47962
GB
141710 114919 114919
0
0
0
0 113 0.021 0.073 1928 3329 2.59 0.08 9914
Irland/N.Ireland
9753
5001
5001
0
0
0
0 107 0.006 0.017 1362 2954 1.08 0.13 1399
France Atlantic
51229 22260 22080
0
11
115
54 288 0.050 0.101 3716 4980 4.78 0.15 25718
Danube
73456 126783 35199 4406 2452 66400 18326 874 0.144 0.239 4553 5673 1.64 0.11 67242
Iberia Atlantic
7799
1698
1698
0
0
0
0 340 0.029 0.061 3585 4772 0.77 0.09 7404
Iberia Mediterranean
40779 19894
4205 1688
796
8838 4367 227 0.199 0.324 5392 6298 0.54 0.14 28833
France Mediterranean 22745
4699
4699
0
0
0
0 563 0.027 0.069 3168 4363 0.65 0.11 11664
Po
16199 12791 12646
8
15
109
14 706 0.075 0.154 4238 5170 0.89 0.09 16919
Corsica
69
5
5
0
0
0
0 89 0.008 0.061 3072 4436 0.34 0.02
360
Sardinia
761
75
75
0
0
0
0 107 0.164 0.232 5247 6015 0.71 0.08 2080
Sicily
2071
20
20
0
0
0
0 75 0.358 0.475 6336 6840 1.42 0.12 3662
South Italy
14047
5432
3889 212
490
117
724 199 0.197 0.297 5210 5916 1.35 0.12 20990
Adige/Balkan
4071
865
865
0
0
0
0 234 0.020 0.045 2364 3712 0.50 0.12 4119
Greece/Evros
16376
1836
910
0
64
814
49 208 0.097 0.159 5014 5963 0.50 0.09 14061

Figure 53 Results of the 21-urban25 scenario.

Figure 54 Reduction of the damage of a 100yr return period flood under the 21-urban25 scenario

72

Groundwater (Mm3/yr)

Consumed (Mm3/yr)

Abstracted (Mm3/yr)

PO4 (mg/l)

NO3 (mg/l)

Env25 (days)

Env10 (days)

WEIabs (-)

WEIcns (-)

Q50 (m3/s)

Loss Domestic (MillionEuro)

Loss EnergyProduction (MillionEuro)

Loss Industry (MillionEuro)

Loss Agri (MillionEuro)

FloodDamage (MillionEuro)

TotalLoss (MillionEuro)

Scenario Costs (MillionEuro)

Region

The 21-Urban25 scenario

600
124
1093
432
1881
58
621
281
5567
140
5171
92
11629 1105
2818
80
464
1
11698 6352
35033 11311
3950 2310
18014 11629
4018 1542
7990 4065
49
26
1469 1001
2755 1902
13373 8319
1886
813
9036 5395

Groundwater (%change)

Consumed (%change)

Abstracted (%change)

PO4 (change mg/l)

NO3 (change mg/l)

Env25 (change days/yr)

Env10 (change days/yr)

WEIabs (change)

WEIcns (change)

Q50 (m3/s change)

Loss Sectors (%change)

FloodDamage (%change)

Costs (MillionEuro)

Region

01 N. Scandinavia
8499 0.25
0.0 0.01 0.000 0.000 -0.01 -0.01 0.00 0.00 0.0 0.0 0.0
02 S. Scandinavia
17447 0.30
0.0 0.02 0.000 0.000 -0.01 -0.03 0.00 0.00 0.0 0.0 0.0
03 Baltic
7663 0.19
0.0 0.11 0.000 0.000 -0.58 -0.53 0.00 0.00 0.0 0.0 0.0
04 Denmark/N.Germany
10850 2.67
0.0 0.09 0.000 0.000 -0.28 -0.37 0.00 0.00 0.0 0.0 0.0
05 Odra/Vistula
18389 1.23
-3.1 0.50 0.000 -0.001 -0.73 -0.70 0.00 0.00 0.0 0.0 0.0
06 Elbe to Ems
56965 6.51
-6.7 0.95 -0.001 -0.003 -0.63 -0.65 -0.01 0.00 0.0 0.0 0.0
07 Rhine/Meuse/Scheldt 111481 11.74
-6.7 1.48 0.000 -0.001 -0.25 -0.32 0.00 0.00 0.0 0.0 0.0
08 GB
141710 27.14
0.0 0.76 0.000 -0.001 0.23 -0.11 -0.01 0.00 0.0 0.0 0.0
09 Irland/N.Ireland
9753 9.98
0.0 0.10 0.000 0.000 0.00 -0.06 0.00 0.00 0.0 0.0 0.0
10 France Atlantic
51229 2.91 -745.5 0.26 0.000 0.000 -0.17 -0.19 0.00 0.00 0.0 0.0 0.0
11 Danube
73456 2.49
-3.2 1.23 -0.001 -0.002 -0.33 -0.34 0.00 0.00 0.0 0.0 0.0
12 Iberia Atlantic
7799 0.53
0.0 0.09 0.000 0.000 -0.02 -0.03 0.00 0.00 0.0 0.0 0.0
13 Iberia Mediterranean
40779 6.04
-3.1 0.44 -0.001 -0.002 -0.24 -0.25 0.00 0.00 0.0 0.0 0.0
14 France Mediterranean 22745 0.71
0.0 0.19 0.000 0.000 -0.02 -0.06 0.00 0.00 0.0 0.0 0.0
15 Po
16199 4.50
-0.8 1.10 0.000 0.000 -0.13 -0.13 0.00 0.00 0.0 0.0 0.0
16 Corsica
69 0.02
0.0 0.01 0.000 0.000 0.01 0.00 0.00 0.00 0.0 0.0 0.0
17 Sardinia
761 0.40
0.0 0.03 0.000 0.000 -0.01 -0.02 0.00 0.00 0.0 0.0 0.0
18 Sicily
2071 1.12
0.0 0.07 0.000 0.000 -0.09 -0.10 0.00 0.00 0.0 0.0 0.0
19 South Italy
14047 1.58
-1.1 0.23 0.000 0.000 -0.06 -0.08 0.00 0.00 0.0 0.0 0.0
20 Adige/Balkan
4071 3.16
0.0 0.16 0.000 0.000 -0.10 -0.09 0.00 0.00 0.0 0.0 0.0
21 Greece/Evros
16376 1.63
-1.3 0.34 0.000 0.000 -0.16 -0.15 0.00 0.00 0.0 0.0 0.0
Figure 55 Differences between the 21-urban25 scenario and the baseline2030 scenario. Positive changes indicate that
the scenario is an improvement as compared to the baseline2030 conditions.

This scenario simulates the implementation of a combination of measures in urban areas like green
infrastructure, green roofs, rain gardens, park depressions, infiltration devices. The percentage of
impermeable area within CORINE land use class 1 (continuous urban fabric) was reduced with 25%
sealed area.
In these simulations, establishing urban greening measures reduces flood peaks (Q50) slightly, for
example in the GB-region (0.7%). This scenario consequently also reduces the potential flood damage
with 27% in the GB-region, and even higher in England (Figure 54). Further positive effects are
simulated in the Rhine/Meuse/Scheldt (0.2% Q50 decrease, 12% flood damage decrease), Elbe/Ems
(0.3% Q50 decrease, 6.5% flood damage decrease), Po (0.2% Q50 decrease, 4.5% flood damage
decrease) & Iberia-Mediterranean (0.2% Q50 decrease, 6.0% flood damage decrease). In Scandinavia
however, there are no effects, likely caused by the fact that cities are more open there already. On the
other side, reduced runoff from cities result in less water available for extraction, and thus slightly
deteriorating the WEI in those areas. Environmental flow (10th percentile) conditions improve however
in the GB with 0.4%, likely because of increasing baseflow as a consequence of increased infiltration.
Because of reduced total river flow amounts, N concentrations in rivers are simulated to slightly
increase under the urban25 scenario in the GB region (0.2%) and the Elbe/Ems region (0.3%).

73

01 N. Scandinavia
02 S. Scandinavia
03 Baltic
04 Denmark/N.Germany
05 Odra/Vistula
06 Elbe to Ems
07 Rhine/Meuse/Scheldt
08 GB
09 Irland/N.Ireland
10 France Atlantic
11 Danube
12 Iberia Atlantic
13 Iberia Mediterranean
14 France Mediterranean
15 Po
16 Corsica
17 Sardinia
18 Sicily
19 South Italy
20 Adige/Balkan
21 Greece/Evros

17270
7453 7453
0
0
0
0 313 0.001 0.004 1118 2750 0.17 0.04 3252
34778
5098 5098
0
0
0
0 282 0.004 0.019 1231 2835 0.66 0.05 4533
15374
1742 1742
0
0
0
0 314 0.018 0.062 2507 3906 1.05 0.06 8144
21991
718
718
0
0
0
0 40 0.025 0.095 3246 4573 3.91 0.23 2437
37821 17141 10816
3
381
5112
830 324 0.083 0.285 5199 6200 2.60 0.09 20789
117101 324939 7913
10 37459 183633 95925 275 0.075 0.286 4332 5466 5.26 0.12 20628
226878 28694 22588
0
483
5037
585 739 0.043 0.161 3665 4863 3.46 0.16 47962
283421 86312 86312
0
0
0
0 112 0.021 0.074 1920 3332 2.60 0.08 9914
19506
4546 4546
0
0
0
0 107 0.006 0.017 1362 2956 1.08 0.13 1399
105660 22017 21633
0
24
239
121 288 0.050 0.101 3722 4986 4.79 0.16 25718
146245 128901 34454 4584 2516 67955 19392 873 0.145 0.241 4563 5683 1.64 0.11 67242
15614
1689 1689
0
0
0
0 340 0.029 0.061 3586 4773 0.77 0.09 7404
81558 20194 4003 1768
822
9063 4539 226 0.200 0.326 5399 6305 0.54 0.14 28833
46605
4666 4666
0
0
0
0 562 0.027 0.069 3168 4365 0.65 0.11 11664
33348 12272 12126
8
15
109
14 704 0.075 0.154 4242 5174 0.89 0.09 16919
143
5
5
0
0
0
0 89 0.008 0.061 3072 4435 0.34 0.02
360
1569
74
74
0
0
0
0 107 0.164 0.232 5247 6015 0.71 0.08 2080
4270
20
20
0
0
0
0 75 0.358 0.475 6339 6843 1.42 0.12 3662
28972
5386 3827 215
495
119
731 199 0.197 0.297 5212 5918 1.35 0.12 20990
8272
839
839
0
0
0
0 234 0.020 0.045 2367 3714 0.50 0.12 4119
32834
1834
895
0
64
825
50 208 0.097 0.159 5019 5968 0.50 0.09 14061

600
124
1093
432
1881
58
621
281
5567
140
5171
92
11629 1105
2818
80
464
1
11698 6352
35033 11311
3950 2310
18014 11629
4018 1542
7990 4065
49
26
1469 1001
2755 1902
13373 8319
1886
813
9036 5395

Figure 56 Results of the 22-urban50 scenario.

Figure 57 Reduction of the damage of a 100yr return period flood under the 22-urban50 scenario

74

Groundwater (Mm3/yr)

Consumed (Mm3/yr)

Abstracted (Mm3/yr)

PO4 (mg/l)

NO3 (mg/l)

Env25 (days)

Env10 (days)

WEIabs (-)

WEIcns (-)

Q50 (m3/s)

Loss Domestic (MillionEuro)

Loss EnergyProduction (MillionEuro)

Loss Industry (MillionEuro)

Loss Agri (MillionEuro)

FloodDamage (MillionEuro)

TotalLoss (MillionEuro)

Scenario Costs (MillionEuro)

Region

The 22-Urban50 scenario

Groundwater (%change)

Consumed (%change)

Abstracted (%change)

PO4 (change mg/l)

NO3 (change mg/l)

Env25 (change days/yr)

Env10 (change days/yr)

WEIabs (change)

WEIcns (change)

Q50 (m3/s change)

Loss Sectors (%change)

FloodDamage (%change)

Costs (MillionEuro)

Region

01 N. Scandinavia
17270 0.48
0.0 0.01 0.000 0.000 0.00 0.00 0.00 0.00 0.0 0.0 0.0
02 S. Scandinavia
34778 0.62
0.0 0.06 0.000 0.000 0.01 -0.03 0.00 0.00 0.0 0.0 0.0
03 Baltic
15374 0.42
0.0 0.22 0.000 -0.001 -1.13 -1.03 0.00 0.00 0.0 0.0 0.0
04 Denmark/N.Germany
21991 5.11
0.0 0.19 0.000 0.000 -0.52 -0.70 -0.01 0.00 0.0 0.0 0.0
05 Odra/Vistula
37821 2.49
-6.7 1.06 -0.001 -0.003 -1.51 -1.45 -0.01 0.00 0.0 0.0 0.0
06 Elbe to Ems
117101 11.43
-14.2 1.93 -0.001 -0.006 -1.25 -1.32 -0.03 0.00 0.0 0.0 0.0
07 Rhine/Meuse/Scheldt 226878 20.36
-13.5 2.96 0.000 -0.001 -0.47 -0.63 -0.01 0.00 0.0 0.0 0.0
08 GB
283421 45.28
0.0 1.50 0.000 -0.001 0.49 -0.19 -0.01 0.00 0.0 0.0 0.0
09 Irland/N.Ireland
19506 18.17
0.0 0.20 0.000 0.000 0.01 -0.11 0.00 0.00 0.0 0.0 0.0
10 France Atlantic
105660 4.87 -1000.0 0.54 0.000 0.000 -0.34 -0.38 -0.01 0.00 0.0 0.0 0.0
11 Danube
146245 4.56
-6.5 2.42 -0.002 -0.003 -0.64 -0.67 0.00 0.00 0.0 0.0 0.0
12 Iberia Atlantic
15614 1.06
0.0 0.18 0.000 0.000 -0.05 -0.06 0.00 0.00 0.0 0.0 0.0
13 Iberia Mediterranean
81558 10.55
-6.4 0.90 -0.003 -0.004 -0.48 -0.51 0.00 0.00 0.0 0.0 0.0
14 France Mediterranean 46605 1.41
0.0 0.38 0.000 0.000 -0.04 -0.12 0.00 0.00 0.0 0.0 0.0
15 Po
33348 8.42
-1.6 2.23 0.000 0.000 -0.25 -0.26 0.00 0.00 0.0 0.0 0.0
16 Corsica
143 0.07
0.0 0.01 0.000 0.000 0.03 0.01 0.00 0.00 0.0 0.0 0.0
17 Sardinia
1569 0.83
0.0 0.07 0.000 0.000 0.00 -0.04 0.00 0.00 0.0 0.0 0.0
18 Sicily
4270 2.22
0.0 0.14 0.000 -0.001 -0.20 -0.20 0.00 0.00 0.0 0.0 0.0
19 South Italy
28972 3.15
-2.1 0.47 0.000 0.000 -0.13 -0.15 0.00 0.00 0.0 0.0 0.0
20 Adige/Balkan
8272 6.16
0.0 0.32 0.000 0.000 -0.21 -0.18 0.00 0.00 0.0 0.0 0.0
21 Greece/Evros
32834 3.22
-2.7 0.69 0.000 0.000 -0.32 -0.31 0.00 0.00 0.0 0.0 0.0
Figure 58 Differences between the 22-urban50 scenario and the baseline2030 scenario. Positive changes indicate that
the scenario is an improvement as compared to the baseline2030 conditions.

This scenario simulates the implementation of a combination of measures in urban areas like green
infrastructure, green roofs, rain gardens, park depressions, infiltration devices. The percentage of
impermeable area within CORINE land use class 1 (continuous urban fabric) was reduced with 50%
sealed area.
In these simulations, establishing urban greening measures reduces flood peaks (Q50) slightly, for
example in the GB-region (1.3%). Potential flood damage is simulated to decrease with 45% in the
GB-region, with local effects in England even larger (Figure 57). Further positive effects are simulated
in the Elbe/Ems (11% decrease), Rhine/Meuse/Scheldt (20% decrease), Ireland (18% decrease), Po
(8.4% decrease) and Iberia-Mediterranean (10.5% decrease). In Scandinavia however, there are no
effects, likely caused by the fact that cities are more open there already.
On the other side, reduced runoff from cities result in less water available for extraction, and thus
deteriorating the WEI in those areas. Environmental flow conditions (10th percentile) slightly improve
however in the GB (0.8%), likely because of increasing baseflow as a consequence of increased
infiltration.
Because of reduced river flow amounts, N concentrations in rivers are simulated to increase under the
urban50 scenario, with 0.4% in the GB region and with 0.5% in the Elbe/Ems region.

75

01 N. Scandinavia
1
7697
7697
0
0
0
0 313 0.001 0.004 1084 2697 0.18 0.04 3252
02 S. Scandinavia
3
4936
4936
0
0
0
0 282 0.004 0.019 1212 2812 0.66 0.05 4533
03 Baltic
8
1709
1709
0
0
0
0 314 0.018 0.061 2476 3873 1.04 0.06 8144
04 Denmark/N.Germany
2
675
675
0
0
0
0 40 0.025 0.095 3213 4532 3.76 0.22 2437
05 Odra/Vistula
19 16172 10400
3
341
4681
748 325 0.082 0.282 5155 6157 2.57 0.09 20789
06 Elbe to Ems
26 370200 60641
9 36800 180474 92276 272 0.074 0.284 4316 5448 5.06 0.12 20628
07 Rhine/Meuse/Scheldt
51 28138 22916
0
416
4337
469 741 0.043 0.160 3649 4845 3.40 0.15 47962
08 GB
48 198579 198579
0
0
0
0 114 0.021 0.073 1913 3304 2.57 0.08 9914
09 Irland/N.Ireland
281
5949
5949
0
0
0
0 107 0.006 0.017 1365 2944 0.72 0.11 1399
10 France Atlantic
108 23019 22998
0
1
14
6 288 0.050 0.101 3717 4979 4.73 0.15 25718
11 Danube
121 125237 36454 4235 2382 64873 17293 872 0.144 0.239 4557 5676 1.61 0.11 67242
12 Iberia Atlantic
27
1746
1746
0
0
0
0 340 0.029 0.060 3574 4759 0.75 0.09 7404
13 Iberia Mediterranean
25 19622
4642 1576
761
8507 4136 228 0.197 0.320 5373 6278 0.53 0.14 28833
14 France Mediterranean 30
4720
4720
0
0
0
0 562 0.027 0.069 3165 4359 0.65 0.11 11664
15 Po
18 13654 13509
8
15
109
14 708 0.075 0.154 4230 5161 0.89 0.09 16919
16 Corsica
1
6
6
0
0
0
0 89 0.008 0.061 3043 4401 0.33 0.02
360
17 Sardinia
0
78
78
0
0
0
0 108 0.163 0.231 5238 6004 0.70 0.08 2080
18 Sicily
3
21
21
0
0
0
0 75 0.358 0.475 6337 6841 1.40 0.12 3662
19 South Italy
27
5535
4014 208
484
116
713 199 0.197 0.297 5210 5916 1.33 0.12 20990
20 Adige/Balkan
3
909
909
0
0
0
0 234 0.020 0.045 2361 3709 0.50 0.12 4119
21 Greece/Evros
7
1855
944
0
62
801
48 209 0.097 0.159 5013 5960 0.49 0.09 14061

Figure 59 Results of the 31-grassland scenario.

76

Groundwater (Mm3/yr)

Consumed (Mm3/yr)

Abstracted (Mm3/yr)

PO4 (mg/l)

NO3 (mg/l)

Env25 (days)

Env10 (days)

WEIabs (-)

WEIcns (-)

Q50 (m3/s)

Loss Domestic (MillionEuro)

Loss EnergyProduction (MillionEuro)

Loss Industry (MillionEuro)

Loss Agri (MillionEuro)

FloodDamage (MillionEuro)

TotalLoss (MillionEuro)

Scenario Costs (MillionEuro)

Region

The 31-Grassland scenario

600
124
1093
432
1881
58
621
281
5567
140
5171
92
11629 1105
2818
80
464
1
11698 6352
35033 11311
3950 2310
18014 11629
4018 1542
7990 4065
49
26
1469 1001
2755 1902
13373 8319
1886
813
9036 5395

Groundwater (%change)

Consumed (%change)

Abstracted (%change)

PO4 (change mg/l)

NO3 (change mg/l)

Env25 (change days/yr)

Env10 (change days/yr)

WEIabs (change)

WEIcns (change)

Q50 (m3/s change)

Loss Sectors (%change)

FloodDamage (%change)

Costs (MillionEuro)

Region

01 N. Scandinavia
1
-2.78 0.0 -0.25 0.000 0.000 1.13 1.78 -0.01 0.00 0.0 0.0 0.0
02 S. Scandinavia
3
3.79 0.0 0.61 0.000 0.000 0.65 0.72 0.00 0.00 0.0 0.0 0.0
03 Baltic
8
2.34 0.0 0.38 0.000 0.000 -0.09 0.05 0.01 0.00 0.0 0.0 0.0
04 Denmark/N.Germany
2 10.86 0.0 0.16 0.000 0.000 0.57 0.66 0.15 0.01 0.0 0.0 0.0
05 Odra/Vistula
19
6.23 2.6 0.45 0.000 0.000 -0.02 -0.04 0.03 0.00 0.0 0.0 0.0
06 Elbe to Ems
26 -578.80 -11.5 4.50 -0.001 -0.003 -0.73 -0.71 0.17 0.00 0.0 0.0 0.0
07 Rhine/Meuse/Scheldt
51 19.20 2.9 1.16 0.000 0.000 0.04 -0.06 0.06 0.00 0.0 0.0 0.0
08 GB
48 -25.90 0.0 -0.73 0.000 0.000 0.74 0.74 0.02 0.00 0.0 0.0 0.0
09 Irland/N.Ireland
281
-7.07 0.0 0.07 0.000 0.000 -0.08 0.29 0.36 0.02 0.0 0.0 0.0
10 France Atlantic
108
-1.13 1.1 0.26 0.000 0.000 -0.19 -0.15 0.05 0.00 0.0 0.0 0.0
11 Danube
121
-0.98 -0.1 2.84 0.000 -0.001 -0.45 -0.43 0.03 0.00 0.0 0.0 0.0
12 Iberia Atlantic
27
-2.33 0.0 -0.33 0.000 0.000 0.36 0.38 0.02 0.00 0.0 0.0 0.0
13 Iberia Mediterranean
25
-3.75 1.5 -0.62 0.001 0.002 0.39 0.39 0.01 0.00 0.0 0.0 0.0
14 France Mediterranean 30
0.27 0.0 0.41 0.000 0.000 0.06 0.08 0.00 0.00 0.0 0.0 0.0
15 Po
18
-2.01 -1.0 -0.92 0.000 0.000 0.13 0.16 0.01 0.00 0.0 0.0 0.0
16 Corsica
1
-2.26 0.0 -0.17 0.000 0.000 0.99 1.15 0.01 0.00 0.0 0.0 0.0
17 Sardinia
0
-4.45 0.0 -0.63 0.001 0.001 0.31 0.33 0.01 0.00 0.0 0.0 0.0
18 Sicily
3
-2.18 0.0 -0.13 0.000 0.000 -0.12 -0.13 0.02 0.00 0.0 0.0 0.0
19 South Italy
27
-1.60 0.4 -0.33 0.000 0.000 -0.06 -0.09 0.02 0.00 0.0 0.0 0.0
20 Adige/Balkan
3
-1.76 0.0 -0.06 0.000 0.000 -0.01 -0.01 0.00 0.00 0.0 0.0 0.0
21 Greece/Evros
7
-2.11 0.4 -0.21 0.000 0.000 -0.10 -0.05 0.01 0.00 0.0 0.0 0.0
Figure 60 Differences between the 31-grassland scenario and the baseline2030 scenario. Positive changes indicate
that the scenario is an improvement as compared to the baseline2030 conditions.

The CAP-demand for grassland/pasture was increased by 10% on slopes larger than 10%. This
required a LUMP run in which each NUTS2 region received an increase of 10% for grassland/pasture
with respect to the demand given by CAPRI. Additionally, the presence of grassland/pastures was
enhanced on land with slopes greater than 10%. If the current land use was already grassland/pasture
its maintenance was enhanced. The conversion of forest/natural vegetation to grassland was prohibited.
Reductions in extreme discharge (Q50) are simulated in the Danube region (0.3%), the Elbe/Ems
region (1.6%), and the Rhine/Meuse/Scheldt region (0.2%). While grassland improves the infiltration
capacity of the soil as compared to arable land, the spatial reconfiguration of land use resulting from
the LUMP run is not largely beneficial to reduce potential flood damages in many regions.

77

01 N. Scandinavia
02 S. Scandinavia
03 Baltic
04 Denmark/N.Germany
05 Odra/Vistula
06 Elbe to Ems
07 Rhine/Meuse/Scheldt
08 GB
09 Irland/N.Ireland
10 France Atlantic
11 Danube
12 Iberia Atlantic
13 Iberia Mediterranean
14 France Mediterranean
15 Po
16 Corsica
17 Sardinia
18 Sicily
19 South Italy
20 Adige/Balkan
21 Greece/Evros

426
7491
7491
0
0
0
0 313 0.001 0.004 1118 2750 0.17 0.04 3252
1320
5131
5131
0
0
0
0 282 0.004 0.019 1232 2834 0.65 0.05 4533
1344
1752
1752
0
0
0
0 314 0.018 0.061 2473 3876 1.04 0.06 8144
861
764
764
0
0
0
0 40 0.025 0.095 3230 4552 3.87 0.23 2437
2174 17031 11109
3
350
4800
770 326 0.082 0.282 5153 6156 2.56 0.09 20789
2358 286351
8949
9 34020 166811 76563 277 0.073 0.280 4295 5428 5.15 0.12 20628
2337 33790 28392
0
430
4477
492 743 0.043 0.160 3652 4845 3.40 0.16 47962
1788 157596 157596
0
0
0
0 113 0.021 0.073 1937 3328 2.54 0.08 9914
259
5550
5550
0
0
0
0 107 0.006 0.017 1364 2954 1.07 0.13 1399
5526 22829 22797
0
2
21
9 289 0.050 0.100 3715 4978 4.70 0.15 25718
4379 124755 36184 4217 2383 64704 17268 876 0.143 0.238 4540 5660 1.62 0.11 67242
1668
1707
1707
0
0
0
0 340 0.029 0.061 3585 4772 0.76 0.09 7404
5345 19686
4472 1613
772
8619 4210 227 0.198 0.322 5385 6290 0.54 0.14 28833
993
4733
4733
0
0
0
0 563 0.027 0.069 3168 4362 0.64 0.11 11664
218 13388 13244
8
15
108
14 707 0.075 0.154 4234 5167 0.89 0.09 16919
15
5
5
0
0
0
0 89 0.008 0.061 3073 4436 0.34 0.02
360
196
75
75
0
0
0
0 107 0.164 0.232 5245 6012 0.70 0.08 2080
243
21
21
0
0
0
0 75 0.358 0.474 6330 6834 1.41 0.12 3662
1073
5472
3945 209
485
116
716 199 0.197 0.297 5210 5915 1.34 0.12 20990
164
893
893
0
0
0
0 234 0.020 0.045 2360 3709 0.50 0.12 4119
1394
1840
930
0
62
800
48 208 0.097 0.159 5008 5958 0.50 0.09 14061

600
124
1093
432
1881
58
621
281
5567
140
5171
92
11629 1105
2818
80
464
1
11698 6352
35033 11311
3950 2310
18014 11629
4018 1542
7990 4065
49
26
1469 1001
2755 1902
13373 8319
1886
813
9036 5395

Figure 61 Results of the 32-bufferstrip scenario.

5m wide grass buffer strips were simulated within arable fields, on slopes between 2% and 10%,
every 200m slope length (ideally constructed along the contours). Thus, it was assumed that 2.5% of
the arable land on these slopes is converted to grassland.

78

Groundwater (Mm3/yr)

Consumed (Mm3/yr)

Abstracted (Mm3/yr)

PO4 (mg/l)

NO3 (mg/l)

Env25 (days)

Env10 (days)

WEIabs (-)

WEIcns (-)

Q50 (m3/s)

Loss Domestic (MillionEuro)

Loss EnergyProduction (MillionEuro)

Loss Industry (MillionEuro)

Loss Agri (MillionEuro)

FloodDamage (MillionEuro)

TotalLoss (MillionEuro)

Scenario Costs (MillionEuro)

Region

The 32-Buffer strip scenario

Groundwater (%change)

Consumed (%change)

Abstracted (%change)

PO4 (change mg/l)

NO3 (change mg/l)

Env25 (change days/yr)

Env10 (change days/yr)

WEIabs (change)

WEIcns (change)

Q50 (m3/s change)

Loss Sectors (%change)

Costs (MillionEuro)
426
1320
1344
861
2174
2358
2337
1788
259
5526
4379
1668
5345
993
218
15
196
243
1073
164
1394

FloodDamage (%change)

Region
01 N. Scandinavia
02 S. Scandinavia
03 Baltic
04 Denmark/N.Germany
05 Odra/Vistula
06 Elbe to Ems
07 Rhine/Meuse/Scheldt
08 GB
09 Irland/N.Ireland
10 France Atlantic
11 Danube
12 Iberia Atlantic
13 Iberia Mediterranean
14 France Mediterranean
15 Po
16 Corsica
17 Sardinia
18 Sicily
19 South Italy
20 Adige/Balkan
21 Greece/Evros

-0.03 0.0 -0.02 0.000 0.000 0.00 0.00 0.00 0.00 0.0 0.0 0.0
-0.02 0.0 0.00 0.000 0.000 -0.01 -0.01 0.01 0.00 0.0 0.0 0.0
-0.15 0.0 -0.13 0.000 0.000 -0.01 -0.03 0.01 0.00 0.0 0.0 0.0
-0.87 0.0 -0.04 0.000 0.000 0.01 0.00 0.04 0.00 0.0 0.0 0.0
-0.15 0.1 -0.27 0.000 0.000 0.02 0.02 0.03 0.00 0.0 0.0 0.0
-0.17 0.1 -0.19 0.000 0.000 -0.03 -0.03 0.08 0.00 0.0 0.0 0.0
-0.10 -0.4 -0.20 0.000 0.000 -0.06 -0.05 0.05 0.00 0.0 0.0 0.0
0.08 0.0 -0.01 0.000 0.000 -0.08 -0.08 0.05 0.00 0.0 0.0 0.0
0.12 0.0 -0.01 0.000 0.000 -0.03 -0.05 0.01 0.00 0.0 0.0 0.0
-0.24 -51.7 -0.10 0.000 0.000 -0.11 -0.12 0.08 0.00 0.0 0.0 0.0
-0.24 0.2 -0.46 0.000 0.000 0.10 0.10 0.02 0.00 0.0 0.0 0.0
-0.03 0.0 0.06 0.000 0.000 -0.03 -0.03 0.01 0.00 0.0 0.0 0.0
0.07 0.0 -0.14 0.000 0.000 0.01 0.01 0.01 0.00 0.0 0.0 0.0
-0.02 0.0 -0.08 0.000 0.000 -0.02 -0.03 0.01 0.00 0.0 0.0 0.0
-0.02 -0.4 -0.07 0.000 0.000 -0.01 -0.01 0.00 0.00 0.0 0.0 0.0
0.06 0.0 -0.01 0.000 0.000 -0.01 -0.02 0.00 0.00 0.0 0.0 0.0
0.49 0.0 0.14 0.000 0.000 0.08 0.06 0.01 0.00 0.0 0.0 0.0
-1.06 0.0 -0.07 0.000 0.000 0.11 0.10 0.01 0.00 0.0 0.0 0.0
0.15 0.0 0.05 0.000 0.000 -0.04 -0.04 0.01 0.00 0.0 0.0 0.0
0.07 0.0 -0.01 0.000 0.000 0.01 0.01 0.00 0.00 0.0 0.0 0.0
-0.58 0.5 0.61 0.000 0.000 0.04 0.01 0.00 0.00 0.0 0.0 0.0

Figure 62 Differences between the 32-bufferstrip scenario and the baseline2030 scenario. Positive changes indicate
that the scenario is an improvement as compared to the baseline2030 conditions.

The outputs of the scenario are only marginally different as compared to the Baseline2030 scenario.
Simulations indicate that runoff amounts are hardly changing under this scenario. Earlier studies
available in literature also indicate that bufferstrips may help to reduce soil loss and capturing of
nutrients, but that runoff effects are less clear.

79

01 N. Scandinavia
02 S. Scandinavia
03 Baltic
04 Denmark/N.Germany
05 Odra/Vistula
06 Elbe to Ems
07 Rhine/Meuse/Scheldt
08 GB
09 Irland/N.Ireland
10 France Atlantic
11 Danube
12 Iberia Atlantic
13 Iberia Mediterranean
14 France Mediterranean
15 Po
16 Corsica
17 Sardinia
18 Sicily
19 South Italy
20 Adige/Balkan
21 Greece/Evros

1206
7491
7491
0
0
0
0 313 0.001 0.004 1118 2750 0.17 0.04 3252
2383
5130
5130
0
0
0
0 282 0.004 0.019 1232 2834 0.65 0.05 4533
3043
1744
1744
0
0
0
0 314 0.018 0.061 2472 3874 1.03 0.06 8144
2780
773
773
0
0
0
0 40 0.025 0.095 3229 4552 3.82 0.23 2437
4983 16994 11079
3
349
4794
769 326 0.082 0.282 5152 6155 2.54 0.09 20789
3981 286238
8918
9 34010 166761 76540 277 0.073 0.280 4295 5428 5.14 0.12 20628
4438 33379 27986
0
429
4473
491 743 0.043 0.160 3653 4846 3.39 0.15 47962
2613 156032 156032
0
0
0
0 113 0.021 0.073 1938 3329 2.53 0.08 9914
827
5521
5521
0
0
0
0 107 0.006 0.017 1364 2955 1.07 0.13 1399
9158 22875 22841
0
2
22
10 289 0.050 0.100 3715 4978 4.68 0.15 25718
9578 124385 36128 4203 2376 64488 17191 876 0.143 0.237 4537 5657 1.62 0.11 67242
2728
1696
1696
0
0
0
0 340 0.029 0.061 3585 4772 0.76 0.09 7404
5421 19664
4447 1613
773
8618 4212 227 0.198 0.322 5385 6290 0.54 0.14 28833
1298
4728
4728
0
0
0
0 563 0.027 0.069 3168 4362 0.64 0.11 11664
1595 13366 13224
8
14
106
13 707 0.075 0.154 4234 5166 0.88 0.09 16919
23
5
5
0
0
0
0 89 0.008 0.061 3073 4436 0.34 0.02
360
283
74
74
0
0
0
0 107 0.164 0.232 5244 6012 0.70 0.08 2080
516
21
21
0
0
0
0 75 0.357 0.474 6330 6834 1.39 0.12 3662
2320
5437
3902 211
488
117
720 199 0.197 0.297 5209 5914 1.33 0.12 20990
518
884
884
0
0
0
0 234 0.020 0.045 2361 3709 0.50 0.12 4119
2394
1831
925
0
62
796
48 208 0.097 0.159 5007 5958 0.50 0.09 14061

Groundwater (Mm3/yr)

Consumed (Mm3/yr)

Abstracted (Mm3/yr)

PO4 (mg/l)

NO3 (mg/l)

Env25 (days)

Env10 (days)

WEIabs (-)

WEIcns (-)

Q50 (m3/s)

Loss Domestic (MillionEuro)

Loss EnergyProduction (MillionEuro)

Loss Industry (MillionEuro)

Loss Agri (MillionEuro)

FloodDamage (MillionEuro)

TotalLoss (MillionEuro)

Scenario Costs (MillionEuro)

Region

The 33-Grassed waterways scenario

600
124
1093
432
1881
58
621
281
5567
140
5171
92
11629 1105
2818
80
464
1
11698 6352
35033 11311
3950 2310
18014 11629
4018 1542
7990 4065
49
26
1469 1001
2755 1902
13373 8319
1886
813
9036 5395

Figure 63 Results of the 33-grassed waterway scenario.

10m wide grass-covered areas were simulated in headwater valley bottoms. GIS pre-processing was
used to select all areas equalling 25km2 upstream area, representing the headwater areas in the
modelling system. These areas were then overlain on the arable land defined by the land use maps, and
the vegetation and soil parameter tables were recalculated assuming 5 % of arable land converted to
grassland.
Assumptions for this scenario are:
used for all arable land, but only in headwater areas with a catchment size of 25km2
5% of arable land converted to grassland (50m grass strip every 1000m)

80

Groundwater (%change)

Consumed (%change)

Abstracted (%change)

PO4 (change mg/l)

NO3 (change mg/l)

Env25 (change days/yr)

Env10 (change days/yr)

WEIabs (change)

WEIcns (change)

Q50 (m3/s change)

Loss Sectors (%change)

Costs (MillionEuro)
1206
2383
3043
2780
4983
3981
4438
2613
827
9158
9578
2728
5421
1298
1595
23
283
516
2320
518
2394

FloodDamage (%change)

Region
01 N. Scandinavia
02 S. Scandinavia
03 Baltic
04 Denmark/N.Germany
05 Odra/Vistula
06 Elbe to Ems
07 Rhine/Meuse/Scheldt
08 GB
09 Irland/N.Ireland
10 France Atlantic
11 Danube
12 Iberia Atlantic
13 Iberia Mediterranean
14 France Mediterranean
15 Po
16 Corsica
17 Sardinia
18 Sicily
19 South Italy
20 Adige/Balkan
21 Greece/Evros

-0.03 0.0 -0.02 0.000 0.000 0.00 0.00 0.00 0.00 0.0 0.0 0.0
-0.01 0.0 0.02 0.000 0.000 -0.01 -0.01 0.01 0.00 0.0 0.0 0.0
0.31 0.0 -0.18 0.000 0.000 0.05 0.03 0.02 0.00 0.0 0.0 0.0
-2.08 0.0 -0.07 0.000 0.000 0.01 -0.01 0.08 0.00 0.0 0.0 0.0
0.11 0.2 -0.57 0.000 0.000 0.06 0.06 0.05 0.00 0.0 0.0 0.0
0.17 0.1 -0.26 0.000 0.000 -0.03 -0.03 0.09 0.00 0.0 0.0 0.0
1.33 -0.3 -0.13 0.000 0.000 -0.08 -0.08 0.07 0.00 0.0 0.0 0.0
1.08 0.0 0.04 0.000 0.000 -0.09 -0.09 0.05 0.00 0.0 0.0 0.0
0.63 0.0 0.02 0.000 0.000 -0.06 -0.08 0.02 0.00 0.0 0.0 0.0
-0.44 -61.5 -0.15 0.000 0.000 -0.12 -0.12 0.09 0.00 0.0 0.0 0.0
-0.08 0.5 -0.86 0.000 0.000 0.20 0.18 0.02 0.00 0.0 0.0 0.0
0.61 0.0 0.23 0.000 0.000 -0.03 -0.02 0.01 0.00 0.0 0.0 0.0
0.61 0.0 0.00 0.000 0.000 0.00 0.01 0.00 0.00 0.0 0.0 0.0
0.09 0.0 -0.06 0.000 0.000 -0.03 -0.03 0.01 0.00 0.0 0.0 0.0
0.14 1.3 -0.30 0.000 0.000 0.01 0.02 0.01 0.00 0.0 0.0 0.0
-0.08 0.0 0.00 0.000 0.000 0.00 -0.01 0.00 0.00 0.0 0.0 0.0
1.31 0.0 0.25 0.000 0.000 0.09 0.08 0.01 0.00 0.0 0.0 0.0
-0.99 0.0 -0.05 0.000 0.000 0.12 0.11 0.02 0.00 0.0 0.0 0.0
1.24 -0.5 0.43 0.000 0.000 -0.03 -0.03 0.02 0.00 0.0 0.0 0.0
1.08 0.0 0.11 0.000 0.000 0.01 0.01 0.00 0.00 0.0 0.0 0.0
-0.04 1.0 0.83 0.000 0.000 0.07 0.04 0.00 0.00 0.0 0.0 0.0

Figure 64 Differences between the 33-grassed waterway scenario and the baseline2030 scenario. Positive changes
indicate that the scenario is an improvement as compared to the baseline2030 conditions

Some reduction on flood damages are simulated in the Rhine/Meuse/Scheldt area (1.3%), Great Britain
(1.1%), Iberia/Atlantic (0.6%), Sardinia (1.3%) and Southern Italy (1.2%).

81

01 N. Scandinavia
02 S. Scandinavia
03 Baltic
04 Denmark/N.Germany
05 Odra/Vistula
06 Elbe to Ems
07 Rhine/Meuse/Scheldt
08 GB
09 Irland/N.Ireland
10 France Atlantic
11 Danube
12 Iberia Atlantic
13 Iberia Mediterranean
14 France Mediterranean
15 Po
16 Corsica
17 Sardinia
18 Sicily
19 South Italy
20 Adige/Balkan
21 Greece/Evros

7404
7357
7357
0
0
0
0 312 0.001 0.004 1118 2750 0.17 0.04 3252
13437
4932
4932
0
0
0
0 282 0.004 0.019 1230 2833 0.66 0.05 4533
15725
1560
1560
0
0
0
0 311 0.018 0.062 2492 3891 1.04 0.06 8144
11942
571
571
0
0
0
0 39 0.025 0.096 3267 4602 3.85 0.23 2437
25534 16755 10404
3
383
5140
825 321 0.083 0.287 5211 6213 2.56 0.09 20789
21771 290822
8140
9 34570 169496 78607 272 0.074 0.283 4332 5466 5.18 0.12 20628
23695 32027 26154
0
466
4857
551 738 0.043 0.161 3665 4861 3.44 0.16 47962
13079 132670 132670
0
0
0
0 113 0.021 0.074 1951 3344 2.56 0.08 9914
4850
5399
5399
0
0
0
0 107 0.006 0.017 1351 2942 1.08 0.13 1399
47397 19385 19172
0
13
135
65 284 0.051 0.103 3774 5041 4.76 0.16 25718
55209 129582 33152 4694 2612 69739 19385 864 0.148 0.244 4639 5751 1.63 0.11 67242
15147
1551
1551
0
0
0
0 337 0.029 0.061 3612 4801 0.77 0.09 7404
31948 19467
3909 1656
789
8810 4303 224 0.200 0.325 5409 6315 0.54 0.14 28833
8139
4558
4558
0
0
0
0 561 0.027 0.069 3167 4362 0.65 0.11 11664
8968 12465 12294
9
17
128
16 701 0.075 0.155 4249 5181 0.89 0.09 16919
149
5
5
0
0
0
0 89 0.008 0.061 3068 4432 0.34 0.02
360
1793
63
63
0
0
0
0 106 0.165 0.233 5253 6020 0.70 0.08 2080
2773
14
14
0
0
0
0 71 0.365 0.484 6383 6891 1.40 0.12 3662
14198
5217
3614 220
509
122
752 196 0.199 0.300 5219 5927 1.34 0.12 20990
3101
831
831
0
0
0
0 233 0.020 0.045 2358 3707 0.50 0.12 4119
14686
1913
771
0
78
1002
62 205 0.098 0.161 5053 5999 0.49 0.09 14061

600
124
1093
432
1881
58
621
281
5567
140
5171
92
11629 1105
2818
80
464
1
11698 6352
35033 11311
3950 2310
18014 11629
4018 1542
7990 4065
49
26
1469 1001
2755 1902
13373 8319
1886
813
9036 5395

Figure 65 Results of the 34-crop practices scenario.

Figure 66 Reduction of the potential damages of a 100-yr return period flood when the 34-crop scenario is
implemented.

82

Groundwater (Mm3/yr)

Consumed (Mm3/yr)

Abstracted (Mm3/yr)

PO4 (mg/l)

NO3 (mg/l)

Env25 (days)

Env10 (days)

WEIabs (-)

WEIcns (-)

Q50 (m3/s)

Loss Domestic (MillionEuro)

Loss EnergyProduction (MillionEuro)

Loss Industry (MillionEuro)

Loss Agri (MillionEuro)

FloodDamage (MillionEuro)

TotalLoss (MillionEuro)

Scenario Costs (MillionEuro)

Region

The 34-Crop practices scenario

Groundwater (%change)

Consumed (%change)

Abstracted (%change)

PO4 (change mg/l)

NO3 (change mg/l)

Env25 (change days/yr)

Env10 (change days/yr)

WEIabs (change)

WEIcns (change)

Q50 (m3/s change)

Loss Sectors (%change)

FloodDamage (%change)

Costs (MillionEuro)

Region

01 N. Scandinavia
7404 1.76
0.0 0.26 0.000 0.000 0.01 0.02 0.00 0.00 0.0 0.0 0.0
02 S. Scandinavia
13437 3.85
0.0 0.34 0.000 0.000 0.04 0.04 0.00 0.00 0.0 0.0 0.0
03 Baltic
15725 10.86
0.0 3.26 0.000 -0.001 -0.62 -0.56 0.01 0.00 0.0 0.0 0.0
04 Denmark/N.Germany 11942 24.55
0.0 1.00 0.000 -0.002 -1.25 -1.65 0.06 0.00 0.0 0.0 0.0
05 Odra/Vistula
25534 6.19
-7.1 4.85 -0.001 -0.005 -1.91 -1.90 0.04 0.00 0.0 0.0 0.0
06 Elbe to Ems
21771 8.89
-1.8 4.77 -0.001 -0.003 -1.26 -1.32 0.05 0.00 0.0 0.0 0.0
07 Rhine/Meuse/Scheldt 23695 7.79
-9.2 4.58 0.000 -0.001 -0.48 -0.58 0.01 0.00 0.0 0.0 0.0
08 GB
13079 15.89
0.0 0.75 0.000 -0.001 -0.54 -0.61 0.02 0.00 0.0 0.0 0.0
09 Irland/N.Ireland
4850 2.83
0.0 0.30 0.000 0.000 0.39 0.35 0.00 0.00 0.0 0.0 0.0
10 France Atlantic
47397 15.69 -898.3 4.84 -0.001 -0.003 -2.09 -2.21 0.02 0.00 0.0 0.0 0.0
11 Danube
55209 8.16
-8.7 11.35 -0.004 -0.007 -3.18 -2.96 0.01 0.00 0.0 0.0 0.0
12 Iberia Atlantic
15147 9.13
0.0 2.95 0.000 -0.001 -0.94 -1.01 0.00 0.00 0.0 0.0 0.0
13 Iberia Mediterranean
31948 12.65
-2.3 3.47 -0.002 -0.003 -0.81 -0.82 0.00 0.00 0.0 0.0 0.0
14 France Mediterranean 8139 3.70
0.0 2.16 0.000 0.000 0.01 -0.02 0.00 0.00 0.0 0.0 0.0
15 Po
8968 7.16 -18.8 5.41 -0.001 -0.001 -0.50 -0.48 0.00 0.00 0.0 0.0 0.0
16 Corsica
149 0.53
0.0 0.12 0.000 0.000 0.17 0.13 0.00 0.00 0.0 0.0 0.0
17 Sardinia
1793 15.69
0.0 1.30 -0.001 -0.002 -0.19 -0.19 0.01 0.00 0.0 0.0 0.0
18 Sicily
2773 29.45
0.0 3.83 -0.007 -0.009 -1.64 -1.78 0.02 0.00 0.0 0.0 0.0
19 South Italy
14198 8.52
-5.0 2.92 -0.002 -0.003 -0.36 -0.45 0.01 0.00 0.0 0.0 0.0
20 Adige/Balkan
3101 6.99
0.0 1.39 0.000 0.000 0.10 0.07 0.00 0.00 0.0 0.0 0.0
21 Greece/Evros
14686 16.64 -24.9 4.19 -0.001 -0.002 -1.46 -1.34 0.00 0.00 0.0 0.0 0.0
Figure 67 Differences between the 34-crop practices scenario and the baseline2030 scenario. Positive changes
indicate that the scenario is an improvement as compared to the baseline2030 conditions.

This scenario simulates the effects of the implementation of combined methods of improved crop
practices on arable land. Modelled effects are reversed/reduced organic matter decline and increased
mulching and tillage. Infiltration was increased by changing soil hydraulic parameters (e.g. bulk
density/porosity). According to literature, improved crop practices should decrease bulk density and
increase organic matter. The scenario Crop practice is applied to all arable land.
Results indicate reductions to potential flood damages in all regions, and also in areas with high
absolute flood damages such as Odra/Vistula (6.2% reduction), Rhine/Meuse (7.8% reduction), Great
Britain (15.9% reduction), Danube (8.2% reduction).
Figure 69 shows the areas where the models simulated the largest reductions in potential flood
damages for the crop practices scenario.

83

01 N. Scandinavia
02 S. Scandinavia
03 Baltic
04 Denmark/N.Germany
05 Odra/Vistula
06 Elbe to Ems
07 Rhine/Meuse/Scheldt
08 GB
09 Irland/N.Ireland
10 France Atlantic
11 Danube
12 Iberia Atlantic
13 Iberia Mediterranean
14 France Mediterranean
15 Po
16 Corsica
17 Sardinia
18 Sicily
19 South Italy
20 Adige/Balkan
21 Greece/Evros

23
6906
6906
0
0
0
0 311 0.001 0.004 1107 2733 0.16 0.04 3252
8
4599
4599
0
0
0
0 280 0.004 0.019 1220 2815 0.60 0.05 4533
3
1601
1601
0
0
0
0 313 0.018 0.061 2493 3887 0.93 0.06 8144
1
730
730
0
0
0
0 40 0.025 0.095 3273 4583 3.65 0.23 2437
3 15945 10004
3
351
4815
773 322 0.082 0.283 5210 6198 2.35 0.09 20789
2 285510
7936
9 34038 166900 76627 274 0.073 0.281 4322 5447 4.98 0.12 20628
3 32238 26860
0
428
4461
489 740 0.043 0.160 3683 4869 3.31 0.16 47962
3 151341 151341
0
0
0
0 112 0.021 0.073 1933 3322 2.05 0.07 9914
1
5365
5365
0
0
0
0 107 0.006 0.017 1359 2949 0.80 0.10 1399
8 21523 21426
0
6
62
29 286 0.050 0.101 3751 5002 4.41 0.16 25718
9 121320 32475 4227 2388 64926 17304 869 0.143 0.238 4575 5687 1.54 0.12 67242
5
1594
1594
0
0
0
0 337 0.029 0.061 3612 4790 0.72 0.09 7404
6 19405
4182 1614
773
8620 4216 226 0.198 0.322 5414 6309 0.49 0.14 28833
3
4572
4572
0
0
0
0 561 0.027 0.069 3194 4381 0.62 0.12 11664
3 12489 12345
8
15
108
14 702 0.075 0.154 4252 5179 0.84 0.10 16919
0
5
5
0
0
0
0 89 0.008 0.061 3073 4432 0.34 0.02
360
0
73
73
0
0
0
0 106 0.164 0.232 5252 6016 0.69 0.09 2080
0
19
19
0
0
0
0 74 0.358 0.475 6339 6842 1.36 0.13 3662
3
5300
3773 209
486
116
716 198 0.197 0.297 5233 5934 1.30 0.12 20990
1
804
804
0
0
0
0 232 0.020 0.045 2366 3713 0.49 0.12 4119
2
1800
885
0
63
804
48 208 0.097 0.159 5040 5981 0.48 0.09 14061

600
124
1093
432
1881
58
621
281
5567
140
5171
92
11629 1105
2818
80
464
1
11698 6352
35033 11311
3950 2310
18014 11629
4018 1542
7990 4065
49
26
1469 1001
2755 1902
13373 8319
1886
813
9036 5395

Figure 68 Results of the 43-meander scenario.

Figure 69 Reduction of the damage of a 100yr return period flood under the 43-meander scenario

84

Groundwater (Mm3/yr)

Consumed (Mm3/yr)

Abstracted (Mm3/yr)

PO4 (mg/l)

NO3 (mg/l)

Env25 (days)

Env10 (days)

WEIabs (-)

WEIcns (-)

Q50 (m3/s)

Loss Domestic (MillionEuro)

Loss EnergyProduction (MillionEuro)

Loss Industry (MillionEuro)

Loss Agri (MillionEuro)

FloodDamage (MillionEuro)

TotalLoss (MillionEuro)

Scenario Costs (MillionEuro)

Region

The 43-Meander scenario

Groundwater (%change)

Consumed (%change)

Abstracted (%change)

PO4 (change mg/l)

NO3 (change mg/l)

Env25 (change days/yr)

Env10 (change days/yr)

WEIabs (change)

WEIcns (change)

Q50 (m3/s change)

Loss Sectors (%change)

FloodDamage (%change)

Costs (MillionEuro)

Region
01 N. Scandinavia
02 S. Scandinavia
03 Baltic
04 Denmark/N.Germany
05 Odra/Vistula
06 Elbe to Ems
07 Rhine/Meuse/Scheldt
08 GB
09 Irland/N.Ireland
10 France Atlantic
11 Danube
12 Iberia Atlantic
13 Iberia Mediterranean
14 France Mediterranean
15 Po
16 Corsica
17 Sardinia
18 Sicily
19 South Italy
20 Adige/Balkan
21 Greece/Evros

23 7.78
0.0 1.27 0.000 0.000 0.37 0.57 0.01 0.00 0.0 0.0 0.0
8 10.34
0.0 2.09 0.000 0.000 0.38 0.64 0.07 0.00 0.0 0.0 0.0
3 8.47
0.0 0.74 0.000 0.000 -0.66 -0.40 0.12 0.00 0.0 0.0 0.0
1 3.61
0.0 0.55 0.000 0.000 -1.42 -1.05 0.26 0.00 0.0 0.0 0.0
3 9.80
-0.2 3.43 0.000 0.000 -1.86 -1.40 0.24 0.00 0.0 0.0 0.0
2 11.16
0.0 2.73 0.000 0.000 -0.92 -0.69 0.25 0.00 0.0 0.0 0.0
3 5.30
0.0 1.98 0.000 0.000 -1.07 -0.85 0.14 0.00 0.0 0.0 0.0
3 4.05
0.0 0.89 0.000 0.000 0.06 0.14 0.54 0.01 0.0 0.0 0.0
1 3.45
0.0 0.28 0.000 0.000 0.12 0.12 0.28 0.03 0.0 0.0 0.0
8 5.78 -355.4 2.61 0.000 0.000 -1.31 -0.91 0.36 0.00 0.0 0.0 0.0
9 10.04
-0.2 5.77 0.000 0.000 -1.04 -0.81 0.10 0.00 0.0 0.0 0.0
5 6.60
0.0 2.95 0.000 0.000 -0.92 -0.63 0.05 0.00 0.0 0.0 0.0
6 6.54
-0.1 1.77 0.000 0.000 -0.97 -0.62 0.05 0.00 0.0 0.0 0.0
3 3.39
0.0 2.18 0.000 0.000 -0.91 -0.65 0.03 -0.01 0.0 0.0 0.0
3 6.77
0.0 4.77 0.000 0.000 -0.59 -0.42 0.05 0.00 0.0 0.0 0.0
0 0.89
0.0 0.29 0.000 0.000 -0.01 0.13 0.00 0.00 0.0 0.0 0.0
0 2.62
0.0 0.52 0.000 0.000 -0.17 -0.07 0.02 0.00 0.0 0.0 0.0
0 8.84
0.0 0.47 0.000 0.000 -0.20 -0.17 0.05 0.00 0.0 0.0 0.0
3 4.50
0.0 1.40 0.000 0.000 -0.84 -0.67 0.05 0.00 0.0 0.0 0.0
1 9.99
0.0 1.61 0.000 0.000 -0.17 -0.13 0.01 0.00 0.0 0.0 0.0
2 4.27
0.0 1.04 0.000 0.000 -1.01 -0.73 0.01 0.00 0.0 0.0 0.0

Figure 70 Differences between the 43-meander scenario and the baseline2030 scenario. Positive changes indicate
that the scenario is an improvement as compared to the baseline2030 conditions.

Increasing the meandering of the current rivers - i.e. increasing the length and storage capacity of the
river bed reduces flood peaks in all European regions, and is estimated to significantly reduce the
flood damage potential especially in the Elbe/Ems (11%), Danube (10%), Odra/Vistula (9.8%), Po
(6.8%), Rhine/Meuse/Scheldt (5.3%) and France/Atlantic regions (5.8%) (Figure 70). Figure 69 shows
large regional flood damage potential reductions in Southern England.
At the same time, environmental flow conditions improve in some areas, for example GB (0.1%) and
Ireland (0.3%), while especially in Mediterranean regions, environmental flow conditions are
simulated to deteriorate with 0.5 to 0.9%.
Nitrate concentrations are benefitting in all regions, with maximum changes of 20-25% in GB and
Ireland.

85

01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21

N. Scandinavia
S. Scandinavia
Baltic
Denmark/N.Germany
Odra/Vistula
Elbe to Ems
Rhine/Meuse/Scheldt
GB
Irland/N.Ireland
France Atlantic
Danube
Iberia Atlantic
Iberia Mediterranean
France Mediterranean
Po
Corsica
Sardinia
Sicily
South Italy
Adige/Balkan
Greece/Evros

32
7489
7489
0
0
0
0 313 0.001 0.004 1118 2750 0.17 0.04 3252
5
5130
5130
0
0
0
0 282 0.004 0.019 1231 2834 0.66 0.05 4533
-7
1750
1750
0
0
0
0 314 0.018 0.061 2473 3875 1.05 0.06 8144
-17
757
757
0
0
0
0 40 0.025 0.095 3230 4552 3.80 0.22 2437
-16 17322 11065
3
368
5078
809 326 0.087 0.287 5182 6179 2.51 0.09 21271
-89 286503
8908
9 34041 166910 76636 277 0.081 0.288 4374 5490 5.04 0.12 21270
-238 34365 28365
1
476
4958
566 741 0.045 0.163 3670 4859 3.30 0.15 48675
-128 157730 157730
0
0
0
0 113 0.021 0.073 1935 3326 2.57 0.08 9914
-83
5556
5556
0
0
0
0 107 0.006 0.017 1363 2953 1.08 0.13 1399
-221 22705 22705
0
0
0
0 289 0.052 0.103 3707 4974 4.56 0.15 26258
-95 134744 35775 5526 2688 69858 20898 876 0.168 0.265 4793 5862 1.54 0.12 75788
-81
1707
1707
0
0
0
0 340 0.029 0.061 3592 4774 0.73 0.09 7438
-146 19691
4477 1613
772
8621 4208 227 0.198 0.322 5392 6296 0.54 0.14 28833
-55
4726
4726
0
0
0
0 562 0.027 0.069 3168 4360 0.63 0.11 11723
-25 13357 13212
8
15
109
14 705 0.076 0.156 4294 5210 0.83 0.09 17136
-1
5
5
0
0
0
0 89 0.008 0.061 3102 4459 0.34 0.02
361
-7
75
75
0
0
0
0 107 0.164 0.232 5247 6014 0.71 0.08 2080
0
20
20
0
0
0
0 75 0.358 0.475 6333 6837 1.42 0.12 3662
-20
5481
3954 209
486
116
716 199 0.197 0.297 5222 5926 1.32 0.12 20992
-3
898
898
0
0
0
0 234 0.021 0.046 2429 3765 0.48 0.12 4212
-7
1835
920
0
63
804
48 209 0.097 0.159 5056 5997 0.47 0.09 14089

600
124
1093
432
1881
58
621
281
6018
214
5773
190
12301 1344
2818
80
464
1
12207 6754
43033 14989
3982 2334
18015 11629
4073 1582
8195 4221
49
26
1469 1001
2755 1902
13375 8321
1972
866
9063 5411

Figure 71 Results of the 51-Nfixing scenario.

Figure 72 Average NO3 concentration reduction in rivers as a consequence of the 51-Nfixing scenario

86

Groundwater (Mm3/yr)

Consumed (Mm3/yr)

Abstracted (Mm3/yr)

PO4 (mg/l)

NO3 (mg/l)

Env25 (days)

Env10 (days)

WEIabs (-)

WEIcns (-)

Q50 (m3/s)

Loss Domestic (MillionEuro)

Loss EnergyProduction (MillionEuro)

Loss Industry (MillionEuro)

Loss Agri (MillionEuro)

FloodDamage (MillionEuro)

TotalLoss (MillionEuro)

Scenario Costs (MillionEuro)

Region

The 51-N-fixing scenario

Groundwater (%change)

Consumed (%change)

Abstracted (%change)

PO4 (change mg/l)

NO3 (change mg/l)

Env25 (change days/yr)

Env10 (change days/yr)

WEIabs (change)

WEIcns (change)

Q50 (m3/s change)

Loss Sectors (%change)

FloodDamage (%change)

Costs (MillionEuro)

Region

01 N. Scandinavia
32 0.00
0.0 0.00 0.000 0.000 0.00 0.00 0.00 0.00 0.0 0.0
0.0
02 S. Scandinavia
5 0.00
0.0 0.00 0.000 0.000 0.00 0.00 0.00 0.00 0.0 0.0
0.0
03 Baltic
-7 0.00
0.0 0.00 0.000 0.000 0.00 0.00 0.00 0.00 0.0 0.0
0.0
04 Denmark/N.Germany
-17 0.00
0.0 0.00 0.000 0.000 0.00 0.00 0.11 0.01 0.0 0.0
0.0
05 Odra/Vistula
-16 0.24 -5.6 -0.01 -0.005 -0.005 -0.95 -0.75 0.09 0.00 -2.3 -8.1 -52.9
06 Elbe to Ems
-89 0.29
0.0 -0.20 -0.007 -0.008 -2.66 -2.12 0.19 0.00 -3.1 -11.7 -107.6
07 Rhine/Meuse/Scheldt -238 -0.01 -11.6 1.54 -0.002 -0.002 -0.64 -0.52 0.16 0.00 -1.5 -5.8 -21.7
08 GB
-128 0.00
0.0 0.00 0.000 0.000 0.00 0.00 0.01 0.00 0.0 0.0
0.0
09 Irland/N.Ireland
-83 0.00
0.0 0.00 0.000 0.000 0.00 0.00 0.00 0.00 0.0 0.0
0.0
10 France Atlantic
-221 0.16 100.0 -0.20 -0.002 -0.002 0.14 0.02 0.21 0.00 -2.1 -4.3
-6.3
11 Danube
-95 0.90 -11.6 -0.88 -0.025 -0.027 -8.31 -6.66 0.10 0.00 -12.7 -22.8 -32.5
12 Iberia Atlantic
-81 -0.01
0.0 0.01 0.000 0.000 -0.25 -0.11 0.04 0.00 -0.5 -0.8
-1.1
13 Iberia Mediterranean
-146 -0.05
0.0 0.01 0.000 0.000 -0.23 -0.19 0.00 0.00 0.0 0.0
0.0
14 France Mediterranean -55 0.14
0.0 0.35 0.000 0.000 -0.05 0.02 0.02 0.00 -0.5 -1.4
-2.6
15 Po
-25 0.23 -0.8 1.71 -0.002 -0.002 -1.99 -1.46 0.06 0.00 -1.3 -2.6
-3.8
16 Corsica
-1 -0.02
0.0 0.00 0.000 0.000 -0.98 -0.76 0.00 0.00 0.0 -0.1
-0.2
17 Sardinia
-7 0.00
0.0 0.00 0.000 0.000 0.00 0.00 0.00 0.00 0.0 0.0
0.0
18 Sicily
0 0.00
0.0 0.00 0.000 0.000 0.00 0.00 0.00 0.00 0.0 0.0
0.0
19 South Italy
-20 -0.07
0.0 -0.02 0.000 0.000 -0.47 -0.40 0.02 0.00 0.0 0.0
0.0
20 Adige/Balkan
-3 -0.46
0.0 0.02 -0.001 -0.001 -2.28 -1.87 0.02 0.00 -2.2 -4.6
-6.5
21 Greece/Evros
-7 0.50
0.0 0.01 -0.001 -0.001 -1.56 -1.26 0.03 0.00 -0.2 -0.3
-0.3
Figure 73 Differences between the 51-Nfixing scenario and the baseline2030 scenario. Positive changes indicate that
the scenario is an improvement as compared to the baseline2030 conditions.

This scenario reduces Nitrate concentrations in rivers in all regions with significant agriculture, while
most dramatically in the France/Atlantic (4.5%), Elbe/Ems (3.7%) and Rhine/Meuse/Scheldt (4.6%)
regions, but also in the Danube (6.0%) and Po region (6.7%).
The costs of the scenario are defined as the reduction of fertilizers, thus resulting in negative numbers
in the cost column.

87

01 N. Scandinavia
02 S. Scandinavia
03 Baltic
04 Denmark/N.Germany
05 Odra/Vistula
06 Elbe to Ems
07 Rhine/Meuse/Scheldt
08 GB
09 Irland/N.Ireland
10 France Atlantic
11 Danube
12 Iberia Atlantic
13 Iberia Mediterranean
14 France Mediterranean
15 Po
16 Corsica
17 Sardinia
18 Sicily
19 South Italy
20 Adige/Balkan
21 Greece/Evros

-1054
7489
7489
0
0
0
0 313 0.001 0.004 1118 2750 0.15 0.04 3252
-1134
5130
5130
0
0
0
0 282 0.004 0.019 1231 2834 0.38 0.04 4533
101
1750
1750
0
0
0
0 314 0.018 0.061 2473 3875 0.59 0.05 8144
-663
757
757
0
0
0
0 40 0.024 0.094 3222 4546 1.54 0.13 2427
-254 17018 11092
3
351
4802
771 325 0.082 0.282 5154 6156 1.10 0.09 20788
-1600 286503
8933
9 34038 166897 76626 277 0.073 0.281 4294 5427 1.66 0.09 20627
-3538 33741 28363
0
428
4461
489 742 0.043 0.160 3650 4843 1.28 0.10 47938
-2100 157730 157730
0
0
0
0 113 0.021 0.073 1935 3326 0.64 0.06 9914
-1183
5556
5556
0
0
0
0 107 0.006 0.017 1363 2953 0.43 0.09 1399
-1916 22760 22738
0
1
14
6 289 0.050 0.101 3713 4976 1.87 0.12 25806
-110 124889 36076 4228 2398 64871 17316 875 0.144 0.238 4545 5664 1.10 0.11 67316
-342
1707
1707
0
0
0
0 340 0.030 0.061 3597 4780 0.62 0.07 7683
-299 19648
4479 1601
769
8613 4186 227 0.197 0.321 5384 6290 0.37 0.13 28823
-237
4735
4735
0
0
0
0 563 0.027 0.069 3180 4372 0.66 0.10 11741
-301 13401 13257
8
15
108
14 707 0.075 0.154 4237 5168 0.83 0.07 16968
12
5
5
0
0
0
0 89 0.008 0.061 3076 4437 0.33 0.01
361
19
75
75
0
0
0
0 107 0.167 0.234 5280 6037 1.50 0.07 2104
20
20
20
0
0
0
0 75 0.358 0.475 6343 6847 2.70 0.12 3667
270
5478
3948 210
486
116
717 199 0.199 0.299 5220 5923 1.90 0.10 21212
-33
893
893
0
0
0
0 234 0.021 0.045 2359 3708 0.54 0.11 4147
68
1839
924
0
63
804
48 209 0.097 0.159 5010 5958 0.43 0.08 14063

600
124
1093
432
1881
58
612
269
5566
140
5169
91
11607 1097
2817
80
464
1
11781 6417
35103 11357
4212 2509
18005 11622
4090 1599
8036 4101
49
26
1491 1017
2759 1905
13583 8477
1912
829
9039 5396

Figure 74 Results of the 52-Optimum Fertilization scenario.

Figure 75 Average NO3 concentration reduction in rivers as a consequence of the 52OptFertilization scenario

88

Groundwater (Mm3/yr)

Consumed (Mm3/yr)

Abstracted (Mm3/yr)

PO4 (mg/l)

NO3 (mg/l)

Env25 (days)

Env10 (days)

WEIabs (-)

WEIcns (-)

Q50 (m3/s)

Loss Domestic (MillionEuro)

Loss EnergyProduction (MillionEuro)

Loss Industry (MillionEuro)

Loss Agri (MillionEuro)

FloodDamage (MillionEuro)

TotalLoss (MillionEuro)

Scenario Costs (MillionEuro)

Region

The 52-Optimum Fertilization scenario

Groundwater (%change)

Consumed (%change)

Abstracted (%change)

PO4 (change mg/l)

NO3 (change mg/l)

Env25 (change days/yr)

Env10 (change days/yr)

WEIabs (change)

WEIcns (change)

Q50 (m3/s change)

Loss Sectors (%change)

FloodDamage (%change)

Costs (MillionEuro)

Region

01 N. Scandinavia
-1054 0.00 0.0 0.00 0.000 0.000 0.00 0.00 0.02 0.00 0.0 0.0 0.0
02 S. Scandinavia
-1134 0.00 0.0 0.00 0.000 0.000 0.00 0.00 0.28 0.01 0.0 0.0 0.0
03 Baltic
101 0.00 0.0 0.00 0.000 0.000 0.00 0.00 0.46 0.01 0.0 0.0 0.0
04 Denmark/N.Germany
-663 0.01 0.0 0.00 0.000 0.000 0.26 0.20 2.37 0.10 0.4 1.5 4.3
05 Odra/Vistula
-254 0.00 0.0 0.01 0.000 0.000 0.00 0.00 1.50 0.01 0.0 0.0 0.1
06 Elbe to Ems
-1600 0.00 0.0 0.00 0.000 0.000 0.00 0.00 3.57 0.03 0.0 0.0 0.2
07 Rhine/Meuse/Scheldt -3538 0.00 0.0 -0.01 0.000 0.000 0.02 0.01 2.17 0.06 0.0 0.2 0.7
08 GB
-2100 0.00 0.0 0.00 0.000 0.000 0.00 0.00 1.95 0.02 0.0 0.0 0.5
09 Irland/N.Ireland
-1183 0.00 0.0 0.00 0.000 0.000 0.00 0.00 0.65 0.05 0.0 0.0 1.3
10 France Atlantic
-1916 0.01 -3.2 0.03 0.000 0.000 -0.06 -0.05 2.91 0.04 -0.3 -0.7 -1.0
11 Danube
-110 0.06 -0.1 -0.03 0.000 0.000 -0.05 -0.05 0.54 0.01 -0.1 -0.2 -0.4
12 Iberia Atlantic
-342 -0.01 0.0 0.00 -0.001 -0.001 -0.41 -0.29 0.15 0.02 -3.8 -6.6 -8.6
13 Iberia Mediterranean
-299 -0.09 0.3 -0.07 0.000 0.000 0.03 0.00 0.17 0.00 0.0 0.1 0.1
14 France Mediterranean -237 -0.04 0.0 -0.06 0.000 0.000 -0.44 -0.35 0.00 0.01 -0.7 -1.8 -3.7
15 Po
-301 -0.12 -0.2 -0.40 0.000 0.000 -0.09 -0.06 0.07 0.02 -0.3 -0.6 -0.9
16 Corsica
12 0.00 0.0 0.00 0.000 0.000 -0.12 -0.05 0.01 0.00 0.0 -0.2 -0.3
17 Sardinia
19 -0.03 0.0 0.00 -0.002 -0.003 -1.11 -0.77 -0.79 0.01 -1.1 -1.5 -1.6
18 Sicily
20 0.04 0.0 0.01 -0.001 -0.001 -0.33 -0.31 -1.28 0.01 -0.1 -0.2 -0.2
19 South Italy
270 0.06 -0.2 0.01 -0.002 -0.003 -0.40 -0.33 -0.55 0.02 -1.1 -1.6 -1.9
20 Adige/Balkan
-33 0.02 0.0 0.00 0.000 0.000 0.06 0.03 -0.04 0.01 -0.7 -1.4 -2.0
21 Greece/Evros
68 0.06 0.0 0.04 0.000 0.000 -0.01 0.02 0.07 0.00 0.0 0.0 0.0
Figure 76 Differences between the 52-optfertilization scenario and the baseline2030 scenario. Positive changes
indicate that the scenario is an improvement as compared to the baseline2030 conditions.

This scenario reduces Nitrate and Phosphate concentrations in all regions with significant agriculture,
while most dramatically in the Elbe/Ems region (68% and 26%), France/Atlantic (61% and 25%),
Denmark/N-Germany (61% and 45%), Rhine/Meuse/Scheldt (63% and 39%), and GB (75% and
26%). Figure 76 shows that the GB changes are mainly occurring in England.
For southern Italy, Sicily and Sardinia, results are that the simulated Nitrate concentrations are
increasing as compared to the Baseline2030. What is simulated for those regions is largely controlled
by the simulation of wheat, Within the EPIC model, wheat has been set to be irrigated. The basic
fertilization provided by the CAPRI model is rather low and the yield of the wheat is also rather low.
When moving to auto-fertilization, the EPIC model applies higher quantities of nitrogen because there
is enough water due to irrigation. So, the application of N is doubled, and the yield also has almost
doubled. This leads to higher leaching losses of N.
The costs of the scenario are defined as the reduction of fertilizers, thus resulting in negative numbers
in the cost column. The costs do not take into the account the effort - and thus also costs - of finding
out the optimum fertilization.

89

01 N. Scandinavia
28271
7520
7520
0
0
0
0 313 0.001 0.004 1030 2625 0.17 0.04 3252
02 S. Scandinavia
16436
5169
5169
0
0
0
0 282 0.004 0.019 1116 2710 0.66 0.05 4533
03 Baltic
7889
1758
1758
0
0
0
0 314 0.018 0.061 2425 3819 1.05 0.06 8144
04 Denmark/N.Germany
2630
785
785
0
0
0
0 40 0.025 0.094 2993 4353 3.87 0.23 2437
05 Odra/Vistula
3287 17024 11099
3
350
4802
771 326 0.082 0.282 5131 6132 2.59 0.09 20789
06 Elbe to Ems
2630 286452
8939
9 34032 166858 76615 277 0.073 0.280 4281 5416 5.23 0.12 20628
07 Rhine/Meuse/Scheldt
1972 33773 28395
0
428
4461
489 742 0.043 0.160 3646 4840 3.45 0.16 47962
08 GB
8547 158655 158655
0
0
0
0 113 0.021 0.073 1855 3262 2.09 0.07 9914
09 Irland/N.Ireland
7232
5610
5610
0
0
0
0 107 0.006 0.017 1189 2787 0.82 0.10 1399
10 France Atlantic
5260 22787 22768
0
1
12
5 289 0.050 0.100 3689 4956 4.77 0.15 25718
11 Danube
657 124803 36099 4221 2387 64805 17291 875 0.143 0.238 4541 5660 1.64 0.11 67242
12 Iberia Atlantic
3945
1708
1708
0
0
0
0 340 0.029 0.060 3550 4744 0.77 0.09 7404
13 Iberia Mediterranean
9862 19226
4485 1542
733
8607 3859 227 0.196 0.318 5276 6209 0.52 0.13 28833
14 France Mediterranean 4602
4735
4735
0
0
0
0 563 0.027 0.069 3087 4302 0.65 0.11 11664
15 Po
3287 13355 13257
5
10
74
9 707 0.074 0.153 4213 5149 0.89 0.09 16919
16 Corsica
657
5
5
0
0
0
0 89 0.008 0.061 2459 3890 0.34 0.02
360
17 Sardinia
1315
75
75
0
0
0
0 107 0.163 0.230 5067 5878 0.70 0.08 2080
18 Sicily
1315
21
21
0
0
0
0 75 0.354 0.469 6240 6770 1.41 0.12 3662
19 South Italy
9204
5193
3960 157
401
96
580 199 0.194 0.293 5132 5856 1.34 0.12 20990
20 Adige/Balkan
4602
897
897
0
0
0
0 234 0.020 0.045 2288 3646 0.50 0.12 4119
21 Greece/Evros
8547
1840
927
0
63
802
48 209 0.096 0.157 4901 5865 0.50 0.09 14061

Figure 77 Results of the 71-desalination scenario.

90

Groundwater (Mm3/yr)

Consumed (Mm3/yr)

Abstracted (Mm3/yr)

PO4 (mg/l)

NO3 (mg/l)

Env25 (days)

Env10 (days)

WEIabs (-)

WEIcns (-)

Q50 (m3/s)

Loss Domestic (MillionEuro)

Loss EnergyProduction (MillionEuro)

Loss Industry (MillionEuro)

Loss Agri (MillionEuro)

FloodDamage (MillionEuro)

TotalLoss (MillionEuro)

Scenario Costs (MillionEuro)

Region

The 71-Desalination scenario

600
124
1093
432
1881
58
621
281
5567
140
5171
92
11629 1105
2818
80
464
1
11698 6352
35033 11311
3950 2310
18014 11629
4018 1542
7990 4065
49
26
1469 1001
2755 1902
13373 8319
1886
813
9036 5395

Groundwater (%change)

Consumed (%change)

Abstracted (%change)

PO4 (change mg/l)

NO3 (change mg/l)

Env25 (change days/yr)

Env10 (change days/yr)

WEIabs (change)

WEIcns (change)

Q50 (m3/s change)

Loss Sectors (%change)

FloodDamage (%change)

Costs (MillionEuro)

Region

01 N. Scandinavia
28271 -0.42 0.0 -0.06 0.000 0.000 2.93 4.18 0.00 0.00 0.0 0.0 0.0
02 S. Scandinavia
16436 -0.76 0.0 -0.09 0.000 0.000 3.84 4.12 0.01 0.00 0.0 0.0 0.0
03 Baltic
7889 -0.46 0.0 -0.03 0.000 0.000 1.61 1.86 0.00 0.00 0.0 0.0 0.0
04 Denmark/N.Germany
2630 -3.77 0.0 -0.04 0.000 0.001 7.89 6.64 0.04 0.00 0.0 0.0 0.0
05 Odra/Vistula
3287 -0.06 0.0 -0.02 0.000 0.000 0.77 0.80 0.00 0.00 0.0 0.0 0.0
06 Elbe to Ems
2630 -0.06 0.0 -0.01 0.000 0.000 0.44 0.37 0.00 0.00 0.0 0.0 0.0
07 Rhine/Meuse/Scheldt
1972 -0.11 0.0 -0.01 0.000 0.000 0.14 0.12 0.00 0.00 0.0 0.0 0.0
08 GB
8547 -0.59 0.0 -0.05 0.000 0.000 2.66 2.14 0.49 0.02 0.0 0.0 0.0
09 Irland/N.Ireland
7232 -0.97 0.0 -0.18 0.000 0.000 5.78 5.53 0.26 0.03 0.0 0.0 0.0
10 France Atlantic
5260 -0.12 11.7 -0.03 0.000 0.000 0.74 0.61 0.01 0.00 0.0 0.0 0.0
11 Danube
657 0.00 0.0 0.00 0.000 0.000 0.09 0.08 0.00 0.00 0.0 0.0 0.0
12 Iberia Atlantic
3945 -0.07 0.0 -0.05 0.000 0.000 1.13 0.90 0.00 0.00 0.0 0.0 0.0
13 Iberia Mediterranean
9862 -0.24 3.1 -0.04 0.002 0.004 3.63 2.71 0.02 0.00 0.0 0.0 0.0
14 France Mediterranean 4602 -0.05 0.0 -0.06 0.000 0.000 2.66 1.96 0.00 0.00 0.0 0.0 0.0
15 Po
3287 -0.11 31.8 -0.07 0.000 0.001 0.72 0.57 0.00 0.00 0.0 0.0 0.0
16 Corsica
657 -0.21 0.0 -0.03 0.000 0.000 20.47 18.20 0.00 0.00 0.0 0.0 0.0
17 Sardinia
1315 -0.40 0.0 -0.04 0.002 0.002 5.99 4.54 0.01 0.00 0.0 0.0 0.0
18 Sicily
1315 -0.20 0.0 -0.03 0.004 0.005 3.10 2.24 0.01 0.00 0.0 0.0 0.0
19 South Italy
9204 -0.22 19.2 -0.08 0.003 0.004 2.55 1.92 0.01 0.00 0.0 0.0 0.0
20 Adige/Balkan
4602 -0.40 0.0 -0.07 0.000 0.000 2.44 2.11 0.00 0.00 0.0 0.0 0.0
21 Greece/Evros
8547 -0.28 0.2 -0.07 0.001 0.001 3.61 3.13 0.00 0.00 0.0 0.0 0.0
Figure 78 Differences between the 71-desalination scenario and the baseline2030 scenario. Positive changes indicate
that the scenario is an improvement as compared to the baseline2030 conditions.

This scenario estimates the effects of installing a number of desalination plants along the coastlines.
The number of plants is indicated in the figure below. In the Iberia/Mediterranean region for example,
15 additional desalination plants are added in the simulations. The assumed capacity of these plants is
60 Megalitres per day each. It is further assumed that water is used max 150km inland, inside the same
macro region (e.g. Mediterranean Iberia), but a single plant can supply one or more Water-Regions.
Costs are calculated including distance to plant and height difference. Based on the data reported by
Kuik (2012), we assumed 0.04 Euro/m3/100km distance as well as 0.04 Euro/m3/100m vertical lift. It
is assumed that desalination plants are conventional, costing 1 Euro/m3 produced. The costs for
desalination plants using renewable energy would be up to 9 times more expensive.
Installing desalination plants improves the Water Exploitation Index in several macro-regions (Figure
78), for example in France/Mediterranean (0.7%), Southern Italy (1.3%) and Greece/Evros (1.1%).
Implementing desalination decreases the number of days that Environmental Flow cannot be
respected, especially in Spain (2% improvement) and Italy (1.5-3.4% improvement). Obviously, in
Northern Scandinavia installing these plants has hardly an effect, since abstraction is low compared to
high natural availability of water. Since more water is simulated to remain in rivers if desalinated
water is used, potential flood damages are simulated to increase slightly (less than 1%).
91

PO4 (mg/l)

NO3 (mg/l)

Env25 (days)

Env10 (days)

100
356
46
232
112
75
911
66
1
5240
9040
1906
9593
1272
3354
21
826
1569
6863
669
4308

WEIabs (-)

573
1048
1805
589
5472
5092
11318
2803
464
10573
30784
3530
15864
3708
7275
44
1285
2405
11840
1720
7886

WEIcns (-)

7
7361
7489
-128
0
0
0 313 0.001 0.004 1117 2750 0.17 0.04 3216
35
4846
5130
-284
0
0
0 282 0.004 0.019 1228 2831 0.66 0.05 4474
7
1515
1750
-235
0
0
0 314 0.017 0.060 2463 3865 1.05 0.06 8042
35
-556
757 -1313
0
0
0 40 0.023 0.093 3206 4534 3.90 0.23 2393
6 16498 11095
-316
330
4646
743 326 0.080 0.280 5144 6149 2.59 0.09 20662
18 285431
8930
-180 33942 166337 76402 277 0.072 0.279 4283 5418 5.23 0.12 20523
28 31916 28371 -1820
427
4450
488 741 0.042 0.159 3637 4833 3.45 0.16 47547
16 157746 157753
-7
0
0
0 113 0.021 0.073 1933 3324 2.59 0.08 9895
0
5556
5556
0
0
0
0 107 0.006 0.017 1363 2953 1.08 0.13 1398
165 17107 22697 -5590
0
0
0 289 0.045 0.094 3649 4925 4.76 0.15 24218
210 91765 36397 -12680 1914 53785 12348 874 0.125 0.214 4460 5595 1.63 0.11 61576
75
-518
1708 -2226
0
0
0 340 0.026 0.057 3526 4726 0.77 0.09 6845
396
4660
4478 -11169
625
7444 3283 227 0.174 0.290 5294 6219 0.54 0.14 25966
52
3258
4727 -1470
0
0
0 562 0.025 0.066 3124 4329 0.65 0.11 11251
237 11187 13246 -2127
7
54
7 706 0.068 0.145 4181 5121 0.89 0.09 15966
2
-17
5
-23
0
0
0 89 0.008 0.060 3137 4482 0.34 0.02
354
7
-466
75
-541
0
0
0 107 0.144 0.204 5176 5965 0.70 0.08 1835
19 -1008
20 -1028
0
0
0 75 0.312 0.414 6279 6789 1.40 0.12 3195
88
329
3952 -4413
298
71
421 199 0.174 0.266 5158 5869 1.34 0.12 18945
25
445
895
-450
0
0
0 234 0.019 0.043 2289 3652 0.50 0.12 3898
244
-318
930 -2162
63
804
48 209 0.085 0.143 4915 5884 0.49 0.09 12527

Q50 (m3/s)

Groundwater (Mm3/yr)

Loss Domestic (MillionEuro)

Loss EnergyProduction (MillionEuro)

Loss Industry (MillionEuro)

Loss Agri (MillionEuro)

FloodDamage (MillionEuro)

TotalLoss (MillionEuro)

Consumed (Mm3/yr)

N. Scandinavia
S. Scandinavia
Baltic
Denmark/N.Germany
Odra/Vistula
Elbe to Ems
Rhine/Meuse/Scheldt
GB
Irland/N.Ireland
France Atlantic
Danube
Iberia Atlantic
Iberia Mediterranean
France Mediterranean
Po
Corsica
Sardinia
Sicily
South Italy
Adige/Balkan
Greece/Evros

Abstracted (Mm3/yr)

01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21

Scenario Costs (MillionEuro)

Region

The 91-Irrigation Efficiency scenario

Figure 79 Results of the 91-Irrigation Efficiency Scenario

Figure 80 Areas where the Water Exploitation Index is improved when the Irrigation Efficiency Scenario is
implemented.

92

Groundwater (%change)

Consumed (%change)

Abstracted (%change)

PO4 (change mg/l)

NO3 (change mg/l)

Env25 (change days/yr)

Env10 (change days/yr)

WEIabs (change)

WEIcns (change)

Q50 (m3/s change)

Loss Sectors (%change)

FloodDamage (%change)

Costs (MillionEuro)

Region

01 N. Scandinavia
7 0.00
0.0 0.00 0.000 0.000 0.02 0.02 0.00 0.00 1.1 4.6 19.2
02 S. Scandinavia
35 0.00 1000.0 0.00 0.000 0.000 0.12 0.09 0.00 0.00 1.3 4.1 17.5
03 Baltic
7 -0.01
0.0 -0.02 0.001 0.001 0.32 0.33 0.00 0.00 1.3 4.1 20.1
04 Denmark/N.Germany
35 0.03 1000.0 0.00 0.001 0.002 0.78 0.60 0.00 0.00 1.8 5.3 17.5
05 Odra/Vistula
6 -0.03
8.9 -0.01 0.002 0.002 0.33 0.26 0.00 0.00 0.6 1.7 20.1
06 Elbe to Ems
18 0.04
0.4 0.08 0.001 0.001 0.36 0.28 0.00 0.00 0.5 1.5 17.6
07 Rhine/Meuse/Scheldt
28 -0.03
34.1 1.00 0.001 0.001 0.45 0.35 0.00 0.00 0.9 2.7 17.5
08 GB
16 -0.01
0.0 0.00 0.000 0.000 0.07 0.05 0.00 0.00 0.2 0.5 17.5
09 Irland/N.Ireland
0 0.00
0.0 0.00 0.000 0.000 0.00 0.00 0.00 0.00 0.0 0.0 17.5
10 France Atlantic
165 0.20 1000.0 0.09 0.005 0.007 2.06 1.64 0.01 0.00 5.8 9.6 17.5
11 Danube
210 -0.83
37.6 1.66 0.018 0.024 2.76 2.27 0.01 0.00 8.4 12.1 20.1
12 Iberia Atlantic
75 -0.06 1000.0 0.05 0.003 0.004 1.94 1.49 0.00 0.00 7.6 10.6 17.5
13 Iberia Mediterranean
396 -0.07
98.8 -0.10 0.023 0.031 3.04 2.38 0.00 0.00 9.9 11.9 17.5
14 France Mediterranean 52 0.11 1000.0 0.33 0.002 0.003 1.42 1.06 0.00 0.00 3.5 7.7 17.5
15 Po
237 -0.03 1000.0 1.07 0.007 0.009 1.76 1.53 0.01 0.00 5.6 8.9 17.5
16 Corsica
2 -0.04 1000.0 0.01 0.001 0.001 -2.15 -1.54 0.00 0.00 1.7 9.4 17.5
17 Sardinia
7 -0.15
0.0 -0.02 0.021 0.028 2.38 1.64 0.00 0.00 11.8 12.6 17.5
18 Sicily
19 0.15
0.0 0.00 0.046 0.061 1.81 1.60 0.01 0.00 12.8 12.7 17.5
19 South Italy
88 -0.02 337.2 -0.05 0.023 0.031 1.66 1.48 0.01 0.00 9.7 11.5 17.5
20 Adige/Balkan
25 -0.14 1000.0 -0.01 0.002 0.002 2.40 1.90 0.00 0.00 5.4 8.8 17.7
21 Greece/Evros
244 -0.58 236.4 -0.18 0.012 0.016 3.14 2.48 0.00 0.00 10.9 12.7 20.1
Figure 81 Differences between the 91-irrigation_efficiency scenario and the baseline2030 scenario. Positive changes
indicate that the scenario is an improvement as compared to the baseline2030 conditions.

The scenario Irrigation efficiency assumes that the current irrigation efficiency is improved from
current average 74% (Eastern Europe) - 77% (Western Europe) to 93% by applying drip irrigation
everywhere where it is not yet applied. Costs are estimated at 153 Euro/ha where drip irrigation is not
applied yet. According to literature, current drip-irrigation areas currently cover 3% in Eastern Europe,
and 18% in Western Europe. Costs are again adjusted for national price levels.
The scenario improves the Water Exploitation Indices (Figure 81), especially in the Danube (12.6%),
Iberia/Mediterranean (11.8%), Sicily (12.7%), South Italy (11.8%), Sardinia (12.6%), Greece/Evros
(12.1%) and France Atlantic (9.8%). Figure 80 shows the specific areas where WEI is improved under
this scenario. At the same time, the Environmental Flow Indices in those areas are improved as well
with maximum 3% (Greece/Evros). In the other regions improvements are simulated as well. An
additional benefit is that the use of deep (geological) groundwater is reduced by around 20%
everywhere (last column).
Due to the larger available amount of water when less irrigation water is consumed, economic losses
for industry, the public sector, and agriculture, are reduced also.

93

01 N. Scandinavia
698
6319
7491
0 -1172
0
0 313 0.001 0.003 1109 2742 0.17 0.04 2464
02 S. Scandinavia
878
3301
5139
0 -1838
0
0 282 0.003 0.013 1194 2806 0.66 0.05 3298
03 Baltic
987
1396
1750
0
-354
0
0 314 0.017 0.058 2436 3844 1.05 0.06 7907
04 Denmark/N.Germany
411
387
757
0
-370
0
0 40 0.024 0.089 3224 4547 3.90 0.23 2276
05 Odra/Vistula
2086 14838 11100
2
-920
4022
635 326 0.080 0.271 5133 6139 2.59 0.09 20064
06 Elbe to Ems
4178 207078
8957
6 11736 134524 51855 277 0.069 0.255 4267 5405 5.22 0.12 18840
07 Rhine/Meuse/Scheldt 12074 18014 28457
0 -14338
3532
363 743 0.040 0.143 3629 4827 3.45 0.16 42681
08 GB
627 158137 158293
0
-156
0
0 113 0.020 0.065 1915 3312 2.58 0.08 8904
09 Irland/N.Ireland
1197
5539
5568
0
-29
0
0 107 0.006 0.016 1362 2952 1.08 0.13 1351
10 France Atlantic
6065 20567 22749
0 -2181
0
0 289 0.049 0.095 3702 4967 4.77 0.15 24252
11 Danube
6993 117104 36140 4139 -2785 63016 16595 874 0.142 0.232 4526 5650 1.64 0.11 64391
12 Iberia Atlantic
766
1316
1708
0
-393
0
0 340 0.029 0.059 3579 4767 0.77 0.09 7163
13 Iberia Mediterranean
2208 17841
4480 1532
-530
8441 3918 227 0.197 0.316 5377 6284 0.54 0.14 28312
14 France Mediterranean 1999
3641
4741
0 -1100
0
0 563 0.026 0.064 3152 4350 0.65 0.11 10896
15 Po
4376 11507 13334
7 -1935
90
11 707 0.072 0.135 4174 5121 0.89 0.09 14821
16 Corsica
4
-2
5
0
-8
0
0 89 0.008 0.061 3068 4433 0.34 0.02
355
17 Sardinia
45
-75
75
0
-150
0
0 107 0.161 0.213 5075 5912 0.70 0.08 1913
18 Sicily
56
-122
20
0
-143
0
0 75 0.355 0.454 6307 6819 1.41 0.12 3502
19 South Italy
1134
3264
3954 134 -1416
84
508 199 0.194 0.274 5179 5890 1.34 0.12 19205
20 Adige/Balkan
724
493
896
0
-403
0
0 234 0.020 0.040 2350 3701 0.50 0.12 3682
21 Greece/Evros
618
1042
925
0
-569
648
38 209 0.096 0.153 4945 5912 0.50 0.09 13484

Groundwater (Mm3/yr)

Consumed (Mm3/yr)

Abstracted (Mm3/yr)

PO4 (mg/l)

NO3 (mg/l)

Env25 (days)

Env10 (days)

WEIabs (-)

WEIcns (-)

Q50 (m3/s)

Loss Domestic (MillionEuro)

Loss EnergyProduction (MillionEuro)

Loss Industry (MillionEuro)

Loss Agri (MillionEuro)

FloodDamage (MillionEuro)

TotalLoss (MillionEuro)

Scenario Costs (MillionEuro)

Region

The 93-Water Re-Use scenario

482
124
907
432
1846
58
597
281
5458
140
4902
92
10837 1105
2666
80
457
1
11478 6352
34606 11311
3913 2310
17936 11629
3902 1542
7675 4065
48
26
1444 1001
2731 1902
13105 8319
1820
813
8950 5395

Figure 82 Results of the 93-water-re-use scenario.

Figure 83 Areas where the Water Exploitation Index is improved upon implementing the 93-water re-use scenario

94

Groundwater (%change)

Consumed (%change)

Abstracted (%change)

PO4 (change mg/l)

NO3 (change mg/l)

Env25 (change days/yr)

Env10 (change days/yr)

WEIabs (change)

WEIcns (change)

Q50 (m3/s change)

Loss Sectors (%change)

FloodDamage (%change)

Costs (MillionEuro)

Region

01 N. Scandinavia
698 -0.03
0.0 -0.01 0.000 0.001 0.29 0.28 0.00 0.00 24.2 19.7 0.0
02 S. Scandinavia
878 -0.18 1000.0 -0.02 0.001 0.006 1.25 0.92 0.00 0.00 27.3 17.0 0.0
03 Baltic
987 -0.02
0.0 0.02 0.001 0.003 1.25 1.02 0.00 0.00 2.9 1.9 0.0
04 Denmark/N.Germany
411 -0.01 1000.0 -0.01 0.001 0.006 0.18 0.17 0.00 0.00 6.6 3.9 0.0
05 Odra/Vistula
2086 -0.08
36.9 -0.09 0.002 0.011 0.71 0.59 0.00 0.00 3.5 2.0 0.0
06 Elbe to Ems
4178 -0.26
28.6 -0.15 0.004 0.026 0.91 0.71 0.01 0.00 8.7 5.2 0.0
07 Rhine/Meuse/Scheldt 12074 -0.34 294.2 -0.77 0.003 0.018 0.71 0.56 0.01 0.00 11.0 6.8 0.0
08 GB
627 -0.36
0.0 -0.01 0.001 0.007 0.67 0.47 0.00 0.00 10.2 5.4 0.0
09 Irland/N.Ireland
1197 -0.20
0.0 0.00 0.000 0.001 0.03 0.03 0.00 0.00 3.4 1.5 0.0
10 France Atlantic
6065 -0.03 1000.0 -0.03 0.001 0.006 0.30 0.25 0.00 0.00 5.7 1.9 0.0
11 Danube
6993 -0.11
8.7 0.96 0.001 0.006 0.58 0.43 0.00 0.00 4.2 1.2 0.0
12 Iberia Atlantic
766 -0.09 1000.0 -0.01 0.000 0.002 0.19 0.14 0.00 0.00 3.3 0.9 0.0
13 Iberia Mediterranean
2208 -0.11
12.2 -0.02 0.001 0.006 0.27 0.20 0.00 0.00 1.8 0.4 0.0
14 France Mediterranean 1999 -0.18 1000.0 -0.11 0.001 0.005 0.49 0.35 0.00 0.00 6.6 2.9 0.0
15 Po
4376 -0.69 1000.0 -0.67 0.003 0.019 2.01 1.52 0.00 0.00 12.4 3.9 0.0
16 Corsica
4 -0.01 1000.0 0.00 0.000 0.001 0.15 0.08 0.00 0.00 1.5 1.6 0.0
17 Sardinia
45 0.01
0.0 -0.02 0.003 0.019 5.74 3.42 0.00 0.00 8.1 1.7 0.0
18 Sicily
56 0.00
0.0 0.00 0.003 0.021 0.89 0.62 0.00 0.00 4.4 0.9 0.0
19 South Italy
1134 -0.07 145.2 -0.01 0.003 0.023 0.98 0.77 0.00 0.00 8.5 2.0 0.0
20 Adige/Balkan
724 -0.30 1000.0 0.00 0.001 0.005 0.35 0.27 0.00 0.00 10.6 3.5 0.0
21 Greece/Evros
618 -0.01
87.2 0.15 0.001 0.006 2.16 1.57 0.00 0.00 4.1 1.0 0.0
Figure 84 Differences between the 93-water-re-use scenario and the baseline2030 scenario. Positive changes indicate
that the scenario is an improvement as compared to the baseline2030 conditions.

The scenario Water re-use in industry assumes that 50% of the water abstracted for industry is reused, which leads to a reduction in freshwater abstraction. Costs are assumed at 0.013 Euro per m3 reused, and adjusted for national price levels.
Slight improvements in the Water Exploitation Index are simulated almost everywhere (Figure 84), but
most effects are simulated in the industrial Elbe/Ems (9.2%), the GB (10.9%) and the
Rhine/Meuse/Scheldt (10.9%) region, as well as South-Italy (7.7%), the Po (12.4%), Sardinia (8.1%),
Sicily (4.3%), and the Odra/Vistula region (4.0%).

95

01 N. Scandinavia
02 S. Scandinavia
03 Baltic
04 Denmark/N.Germany
05 Odra/Vistula
06 Elbe to Ems
07 Rhine/Meuse/Scheldt
08 GB
09 Irland/N.Ireland
10 France Atlantic
11 Danube
12 Iberia Atlantic
13 Iberia Mediterranean
14 France Mediterranean
15 Po
16 Corsica
17 Sardinia
18 Sicily
19 South Italy
20 Adige/Balkan
21 Greece/Evros

Groundwater (Mm3/yr)

Consumed (Mm3/yr)

Abstracted (Mm3/yr)

PO4 (mg/l)

NO3 (mg/l)

Env25 (days)

Env10 (days)

WEIabs (-)

WEIcns (-)

Q50 (m3/s)

Loss Domestic (MillionEuro)

Loss EnergyProduction (MillionEuro)

Loss Industry (MillionEuro)

Loss Agri (MillionEuro)

FloodDamage (MillionEuro)

TotalLoss (MillionEuro)

Scenario Costs (MillionEuro)

Region

The 94-Water Saving scenario

0
7217
7490
0
0
0 -273 313 0.001 0.004 1117 2749 0.17 0.04 3160
575
124
0
4223
5133
0
0
0 -910 282 0.004 0.018 1210 2818 0.66 0.05 4228 1009
432
0
1534
1750
0
0
0 -217 314 0.018 0.061 2470 3872 1.05 0.06 8072 1858
58
0
-821
748
0
0
0 -1569 40 0.024 0.090 3245 4563 3.91 0.23 2323
594
281
0 14450 11100
2
305
4148 -1105 326 0.080 0.275 5139 6144 2.59 0.09 20251 5414
140
0 251793
8946
8 32450 159156 51233 277 0.071 0.274 4278 5414 5.23 0.12 20095 5030
92
0 26936 28409
0
400
4172 -6044 743 0.041 0.155 3636 4833 3.45 0.16 46430 11230 1105
0 157957 158492
0
0
0 -535 113 0.017 0.060 1795 3222 2.57 0.08 8193 2255
80
0
5020
5594
0
0
0 -574 107 0.004 0.014 1316 2910 1.08 0.13 1159
366
1
0 19852 22681
0
0
0 -2830 289 0.049 0.097 3697 4963 4.77 0.15 24767 11377 6352
0 113583 36133 4112 2248 62728 8361 875 0.142 0.233 4521 5644 1.64 0.11 65762 34523 11311
0
563
1706
0
0
0 -1143 340 0.028 0.058 3564 4756 0.77 0.09 7061 3840 2310
0 13616
4477 1476
686
8322 -1345 227 0.194 0.308 5345 6260 0.54 0.14 27691 17715 11629
0
3116
4743
0
0
0 -1627 563 0.026 0.065 3139 4340 0.65 0.11 11122 3846 1542
0 11866 13368
7
14
100 -1624 707 0.072 0.146 4162 5111 0.89 0.09 16040 7661 4065
0
-7
5
0
0
0
-13 89 0.008 0.061 3038 4411 0.34 0.02
356
47
26
0
-10
75
0
0
0
-85 107 0.162 0.227 5121 5938 0.71 0.08 2033 1452 1001
0
-185
20
0
0
0 -205 75 0.352 0.460 6250 6785 1.41 0.12 3547 2712 1902
0
3542
3949 164
400
96 -1066 199 0.193 0.286 5176 5888 1.34 0.12 20147 13056 8319
0
324
894
0
0
0 -570 234 0.019 0.042 2325 3683 0.50 0.12 3877 1803
813
0
1215
925
0
59
754 -523 209 0.095 0.155 4961 5922 0.50 0.09 13675 8869 5395

Figure 85 Results of the 94-watersaving scenario

Figure 86 Areas where the Water Exploitation Index is reduced due to scenario 94-WaterSaving in the public sector

96

Q50 (m3/s change)

WEIcns (change)

WEIabs (change)

Env10 (change days/yr)

Env25 (change days/yr)

NO3 (change mg/l)

0.00
-0.01
-0.01
0.00
-0.08
0.05
-0.56
-0.04
-0.07
0.01
0.04
0.03
-0.05
-0.06
-0.76
0.00
-0.01
0.00
0.01
0.00
0.00

0.000
0.000
0.000
0.001
0.002
0.002
0.001
0.004
0.001
0.001
0.002
0.001
0.004
0.001
0.003
0.000
0.002
0.006
0.004
0.001
0.002

0.000
0.001
0.000
0.005
0.007
0.007
0.005
0.013
0.003
0.003
0.005
0.003
0.013
0.004
0.008
0.001
0.005
0.015
0.011
0.003
0.004

0.04
0.71
0.10
-0.50
0.50
0.54
0.48
4.66
1.55
0.46
0.75
0.68
1.32
0.94
2.39
1.17
4.19
2.76
1.07
1.20
1.62

0.04
0.51
0.09
-0.38
0.40
0.43
0.36
3.45
1.44
0.37
0.61
0.49
1.01
0.70
1.85
0.83
2.53
1.76
0.86
0.86
1.23

0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.02
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.01
0.00
0.00
0.00

Groundwater (%change)

Loss Sectors (%change)


0.0
1000.0
0.0
1000.0
43.5
12.5
127.4
0.0
0.0
1000.0
12.7
1000.0
39.9
1000.0
1000.0
1000.0
0.0
0.0
126.7
1000.0
68.3

Consumed (%change)

FloodDamage (%change)
-0.02
-0.06
-0.03
1.19
-0.08
-0.13
-0.16
-0.48
-0.68
0.26
-0.10
0.03
-0.06
-0.21
-0.96
0.00
0.03
0.17
0.05
-0.03
-0.09

Abstracted (%change)

Costs (MillionEuro)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0

PO4 (change mg/l)

Region
01 N. Scandinavia
02 S. Scandinavia
03 Baltic
04 Denmark/N.Germany
05 Odra/Vistula
06 Elbe to Ems
07 Rhine/Meuse/Scheldt
08 GB
09 Irland/N.Ireland
10 France Atlantic
11 Danube
12 Iberia Atlantic
13 Iberia Mediterranean
14 France Mediterranean
15 Po
16 Corsica
17 Sardinia
18 Sicily
19 South Italy
20 Adige/Balkan
21 Greece/Evros

0.00 2.8 4.1 0.0


0.00 6.7 7.7 0.0
0.00 0.9 1.3 0.0
0.00 4.7 4.5 0.0
0.00 2.6 2.7 0.0
0.00 2.6 2.7 0.0
0.00 3.2 3.4 0.0
0.00 17.4 20.0 0.0
0.00 17.1 21.3 0.0
0.00 3.7 2.7 0.0
0.00 2.2 1.5 0.0
0.00 4.6 2.8 0.0
0.00 4.0 1.7 0.0
0.00 4.6 4.3 0.0
0.00 5.2 4.1 0.0
0.00 1.2 3.0 0.0
0.00 2.3 1.2 0.0
0.00 3.1 1.6 0.0
0.00 4.0 2.4 0.0
0.00 5.9 4.4 0.0
0.00 2.7 1.8 0.0

Figure 87 Differences between the 94-water saving scenario and the baseline2030 scenario. Positive changes indicate
that the scenario is an improvement as compared to the baseline2030 conditions.

This scenario assumes water savings in the domestic sector, achieving 25% savings in water demand
by simplistic measures (replacing showerheads etc).
Improvements in the Water Exploitation Index are simulated (Figure 87), as well as reductions in
abstracted and consumed water. Changes are observed in several regions in the following order of
magnitude: Sicily (3.1%), Iberia/Mediterranean (4.2%), Great Britain (17.4%), Southern-Italy (3.6%),
Po (Milan area) (5.3%), and Odra/Vistula (Warsaw area) (2.5%), with local effects around populated
areas even higher (Figure 86). Reductions in water consumption are also simulated in many other
areas.
Zero costs are assumed here for the implementation of the scenario since no better figure could be
found. However, some costs exist for implementing these measures.

97

01 N. Scandinavia
2368
7416
7490
0
0
0
-74 313 0.001 0.004 1117 2749 0.17 0.04 3227
02 S. Scandinavia
2798
4865
5133
0
0
0 -267 282 0.004 0.018 1212 2820 0.66 0.05 4444
03 Baltic
1920
1640
1751
0
0
0 -110 314 0.018 0.061 2470 3872 1.05 0.06 8107
04 Denmark/N.Germany
1579
521
753
0
0
0 -231 40 0.024 0.094 3238 4558 3.91 0.23 2419
05 Odra/Vistula
7736 16146 11099
3
335
4572
138 326 0.080 0.280 5141 6146 2.59 0.09 20610
06 Elbe to Ems
5785 279488
8940
9 33736 165423 71381 277 0.072 0.279 4280 5415 5.23 0.12 20493
07 Rhine/Meuse/Scheldt 15407 32492 28398
0
423
4407 -736 743 0.042 0.159 3641 4836 3.45 0.16 47597
08 GB
10352 158128 158406
0
0
0 -278 113 0.016 0.066 1768 3202 2.56 0.08 9019
09 Irland/N.Ireland
440
4992
5605
0
0
0 -613 107 0.004 0.014 1284 2878 1.07 0.13 1167
10 France Atlantic
14393 21091 22707
0
0
0 -1616 289 0.049 0.098 3694 4961 4.77 0.15 25175
11 Danube
9714 116726 36159 4149 2241 62752 11425 875 0.140 0.234 4510 5636 1.64 0.11 66292
12 Iberia Atlantic
2003
1149
1707
0
0
0 -558 340 0.028 0.060 3567 4759 0.77 0.09 7225
13 Iberia Mediterranean
3075 18311
4478 1586
756
8565 2925 227 0.196 0.319 5359 6271 0.54 0.14 28548
14 France Mediterranean 5833
3991
4746
0
0
0 -756 563 0.026 0.067 3137 4340 0.65 0.11 11403
15 Po
3510 12321 13416
8
14
102 -1219 708 0.070 0.148 4128 5084 0.88 0.09 16257
16 Corsica
65
-2
5
0
0
0
-7 89 0.008 0.061 3029 4404 0.34 0.02
358
17 Sardinia
194
11
75
0
0
0
-64 107 0.161 0.228 5053 5899 0.70 0.08 2044
18 Sicily
572
-136
20
0
0
0 -156 75 0.349 0.463 6196 6754 1.41 0.12 3575
19 South Italy
3659
3996
3951 174
420
100 -648 199 0.191 0.289 5159 5874 1.34 0.12 20348
20 Adige/Balkan
880
583
894
0
0
0 -310 234 0.019 0.044 2320 3680 0.50 0.12 3976
21 Greece/Evros
3211
1137
926
0
56
722 -567 209 0.093 0.154 4928 5896 0.50 0.09 13629

581
124
1025
432
1854
58
608
281
5433
140
5070
92
11356 1105
2146
80
290
1
11291 6352
34321 11311
3815 2310
17801 11629
3822 1542
7494 4065
47
26
1442 1001
2690 1902
12892 8319
1779
813
8713 5395

Figure 88 Results of the 95-leakage reduction scenario.

Figure 89 Areas where the Water Exploitation Index improves due to the 95-leakage reduction scenario

98

Groundwater (Mm3/yr)

Consumed (Mm3/yr)

Abstracted (Mm3/yr)

PO4 (mg/l)

NO3 (mg/l)

Env25 (days)

Env10 (days)

WEIabs (-)

WEIcns (-)

Q50 (m3/s)

Loss Domestic (MillionEuro)

Loss EnergyProduction (MillionEuro)

Loss Industry (MillionEuro)

Loss Agri (MillionEuro)

FloodDamage (MillionEuro)

TotalLoss (MillionEuro)

Scenario Costs (MillionEuro)

Region

The 95-Leakage scenario

Groundwater (%change)

Consumed (%change)

Abstracted (%change)

PO4 (change mg/l)

NO3 (change mg/l)

Env25 (change days/yr)

Env10 (change days/yr)

WEIabs (change)

WEIcns (change)

Q50 (m3/s change)

Loss Sectors (%change)

FloodDamage (%change)

Costs (MillionEuro)

Region

01 N. Scandinavia
2368 -0.02
0.0 0.00 0.000 0.000 0.03 0.03 0.00 0.00 0.8 3.1 0.0
02 S. Scandinavia
2798 -0.06 1000.0 -0.01 0.000 0.000 0.64 0.46 0.00 0.00 2.0 6.2 0.0
03 Baltic
1920 -0.05
0.0 -0.01 0.000 0.000 0.11 0.11 0.00 0.00 0.5 1.5 0.0
04 Denmark/N.Germany
1579 0.58 1000.0 0.00 0.001 0.001 -0.26 -0.19 0.00 0.00 0.8 2.2 0.0
05 Odra/Vistula
7736 -0.06
14.9 -0.02 0.002 0.002 0.44 0.35 0.00 0.00 0.9 2.4 0.0
06 Elbe to Ems
5785 -0.07
2.5 0.09 0.001 0.002 0.48 0.39 0.00 0.00 0.7 2.0 0.0
07 Rhine/Meuse/Scheldt 15407 -0.13
23.9 -0.48 0.001 0.001 0.32 0.25 0.00 0.00 0.8 2.4 0.0
08 GB
10352 -0.43
0.0 -0.04 0.005 0.007 5.58 4.11 0.02 0.00 9.0 23.8 0.0
09 Irland/N.Ireland
440 -0.89
0.0 -0.11 0.002 0.003 2.62 2.47 0.01 0.00 16.6 37.5 0.0
10 France Atlantic
14393 0.15 1000.0 -0.02 0.001 0.002 0.58 0.46 0.01 0.00 2.1 3.5 0.0
11 Danube
9714 -0.17
9.2 -0.24 0.003 0.004 1.10 0.89 0.00 0.00 1.4 2.0 0.0
12 Iberia Atlantic
2003 -0.03 1000.0 0.01 0.001 0.001 0.59 0.40 0.00 0.00 2.4 3.4 0.0
13 Iberia Mediterranean
3075 -0.09
9.1 -0.05 0.002 0.003 0.85 0.63 0.00 0.00 1.0 1.2 0.0
14 France Mediterranean 5833 -0.30 1000.0 -0.04 0.001 0.002 0.99 0.70 0.00 0.00 2.2 4.9 0.0
15 Po
3510 -1.31 861.4 -1.07 0.005 0.006 3.55 2.74 0.01 0.00 3.9 6.2 0.0
16 Corsica
65 0.00 1000.0 0.00 0.000 0.000 1.47 1.05 0.00 0.00 0.7 3.8 0.0
17 Sardinia
194 -0.03
0.0 -0.02 0.003 0.004 6.46 3.83 0.00 0.00 1.7 1.8 0.0
18 Sicily
572 0.23
0.0 0.00 0.008 0.011 4.59 2.79 0.01 0.00 2.4 2.4 0.0
19 South Italy
3659 0.00
97.0 0.01 0.006 0.008 1.63 1.30 0.01 0.00 3.1 3.6 0.0
20 Adige/Balkan
880 0.00 1000.0 0.00 0.001 0.002 1.34 0.96 0.00 0.00 3.5 5.7 0.0
21 Greece/Evros
3211 -0.20
76.9 -0.03 0.004 0.005 2.73 2.08 0.00 0.00 3.1 3.6 0.0
Figure 90 Differences between the 95-leakage reduction scenario and the baseline2030 scenario. Positive changes
indicate that the scenario is an improvement as compared to the baseline2030 conditions.

The leakage scenario assumes that 50% of the current leaking in the public water supply is repaired,
which is similar to the scenario applied in the COWI (2011) report, where also the costs for 50%
leakage reduction were deducted from.
The scenario improves the Water Exploitation Index (abstraction) in all regions (Figure 90), most
dramatically in the GB (24%), Ireland (38%), Po (6.2%), Adige/Balkan (5.8%), and Greece/Evros
(4.1%), with local effects even higher (Figure 89). It also improves the Environmental Flow Indicators
with several days per year, especially in the GB region (8.6%), Ireland (5.8%), Sardinia (3.7%) and
Sicily (2.2%).

99

Loss Domestic (MillionEuro)

Loss EnergyProduction (MillionEuro)

Loss Industry (MillionEuro)

Loss Agri (MillionEuro)

FloodDamage (MillionEuro)

TotalLoss (MillionEuro)

Abstracted (Mm3/yr)

PO4 (mg/l)

NO3 (mg/l)

Env25 (days)

Env10 (days)

WEIabs (-)

62
216
29
141
70
46
552
40
1
3176
5655
1155
5814
771
2033
13
500
951
4159
406
2697

WEIcns (-)

7489
7489
0
0
0
0 313 0.001 0.005 1119 2752 0.17 0.04 3319
662
5130
5130
0
0
0
0 282 0.005 0.019 1251 2848 0.66 0.05 4745 1292
1750
1750
0
0
0
0 314 0.018 0.061 2476 3877 1.05 0.06 8173 1909
771
771
0
0
0
0 40 0.030 0.100 3276 4587 3.95 0.23 2577
754
17130 11089
4
371
4884
782 325 0.084 0.284 5163 6164 2.59 0.09 20858 5632
286904
8932
9 34082 167154 76727 277 0.074 0.281 4300 5431 5.24 0.12 20674 5214
33693 28309
0
428
4465
490 742 0.045 0.162 3677 4866 3.47 0.16 48509 12144
157658 157658
0
0
0
0 113 0.021 0.073 1941 3330 2.11 0.07 9954 2855
5559
5559
0
0
0
0 107 0.006 0.017 1363 2953 0.83 0.10 1399
465
22856 22726
0
16
72
42 289 0.063 0.114 3856 5091 4.82 0.16 28898 14693
139239 35938 5142 2943 73345 21870 876 0.166 0.262 4636 5739 1.69 0.12 72896 40325
1707
1707
0
0
0
0 340 0.036 0.068 3714 4872 0.78 0.09 8542 5021
25973
4450 2785 1112 11095 6531 227 0.257 0.385 5581 6446 0.59 0.14 34666 23507
4745
4745
0
0
0
0 564 0.033 0.075 3268 4438 0.68 0.12 12528 4832
13791 13312
29
68
322
59 709 0.093 0.173 4350 5266 0.94 0.10 18858 9816
5
5
0
0
0
0 89 0.010 0.064 3117 4468 0.34 0.02
373
61
75
75
0
0
0
0 107 0.217 0.288 5406 6129 0.88 0.09 2581 1941
265
21
17
72
41
115 75 0.474 0.598 6451 6940 1.69 0.16 4613 3651
7647
3935 572 1122
294 1725 199 0.256 0.360 5311 6007 1.46 0.13 25149 17290
898
898
0
0
0
0 234 0.024 0.049 2449 3780 0.51 0.12 4527 2270
1892
914
4
73
806
95 208 0.123 0.187 5167 6085 0.51 0.09 16757 11559

Q50 (m3/s)

Groundwater (Mm3/yr)

?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?

Consumed (Mm3/yr)

01 N. Scandinavia
02 S. Scandinavia
03 Baltic
04 Denmark/N.Germany
05 Odra/Vistula
06 Elbe to Ems
07 Rhine/Meuse/Scheldt
08 GB
09 Irland/N.Ireland
10 France Atlantic
11 Danube
12 Iberia Atlantic
13 Iberia Mediterranean
14 France Mediterranean
15 Po
16 Corsica
17 Sardinia
18 Sicily
19 South Italy
20 Adige/Balkan
21 Greece/Evros

Scenario Costs (MillionEuro)

Region

The 96-Wastewater re-use scenario

Figure 91 Results of the 96-wastewater re-use scenario.

This scenario assumes that the use of deep geological groundwater for irrigation is reduced by 50%
and instead treated wastewater is used for irrigation. The costs of this scenario are unknown to the
authors, and there are also unknown effects of wastewater re-use on water quality.

100

Groundwater (%change)

Consumed (%change)

Abstracted (%change)

PO4 (change mg/l)

NO3 (change mg/l)

Env25 (change days/yr)

Env10 (change days/yr)

WEIabs (change)

WEIcns (change)

Q50 (m3/s change)

Loss Sectors (%change)

FloodDamage (%change)

Costs (MillionEuro)

Region

01 N. Scandinavia
?
0.00
0.0 0.00 0.000 0.000 -0.04 -0.04 0.00 0.00 -2.0 -10.4 50.0
02 S. Scandinavia
?
0.00
0.0 0.00 -0.001 -0.001 -0.65 -0.49 0.00 0.00 -4.7 -18.2 50.0
03 Baltic
?
0.00
0.0 0.00 0.000 0.000 -0.08 -0.09 0.00 0.00 -0.4 -1.4 50.0
04 Denmark/N.Germany ? -1.81
0.0 0.00 -0.005 -0.006 -1.54 -1.18 -0.04 0.00 -5.8 -21.3 50.0
05 Odra/Vistula
?
0.02
-1.9 0.02 -0.002 -0.002 -0.31 -0.24 0.00 0.00 -0.3 -1.2 50.0
06 Elbe to Ems
?
0.02
-0.1 -0.06 -0.001 -0.001 -0.19 -0.15 0.00 0.00 -0.2 -0.8 50.0
07 Rhine/Meuse/Scheldt ?
0.19
-0.1 0.04 -0.002 -0.002 -0.89 -0.74 -0.02 0.00 -1.1 -4.4 50.0
08 GB
?
0.05
0.0 0.00 0.000 0.000 -0.19 -0.14 0.48 0.01 -0.4 -1.3 50.0
09 Irland/N.Ireland
? -0.05
0.0 0.00 0.000 0.000 0.00 0.00 0.25 0.03 0.0 -0.1 50.0
10 France Atlantic
?
0.07 -512.6 0.01 -0.013 -0.014 -4.83 -3.89 -0.05 0.00 -12.4 -25.6 50.0
11 Danube
?
0.44 -16.4 -0.94 -0.022 -0.024 -3.09 -2.56 -0.04 -0.01 -8.4 -15.1 50.0
12 Iberia Atlantic
?
0.00
0.0 -0.18 -0.007 -0.008 -4.34 -3.36 -0.02 0.00 -15.4 -27.1 50.0
13 Iberia Mediterranean ?
0.56 -41.5 0.21 -0.060 -0.064 -6.54 -5.19 -0.04 0.00 -20.2 -30.5 50.0
14 France Mediterranean ? -0.26
0.0 -0.88 -0.006 -0.006 -3.38 -2.55 -0.03 0.00 -7.4 -20.3 50.0
15 Po
? -0.53 -233.0 -2.28 -0.018 -0.019 -3.87 -3.32 -0.05 0.00 -11.5 -22.9 50.0
16 Corsica
?
0.12
0.0 0.01 -0.002 -0.002 -1.48 -1.07 0.00 0.00 -3.6 -24.9 50.0
17 Sardinia
?
0.44
0.0 0.06 -0.053 -0.056 -5.31 -3.81 -0.17 -0.01 -24.1 -32.1 50.0
18 Sicily
? -0.44
0.0 -0.01 -0.116 -0.124 -3.91 -3.42 -0.27 -0.03 -26.0 -32.5 50.0
19 South Italy
?
0.40 -143.1 0.16 -0.059 -0.063 -3.43 -3.12 -0.11 -0.01 -19.8 -29.3 50.0
20 Adige/Balkan
? -0.53
0.0 -0.10 -0.004 -0.004 -2.96 -2.36 -0.01 0.00 -9.9 -20.4 50.0
21 Greece/Evros
?
1.12
-6.9 0.39 -0.026 -0.028 -5.24 -4.22 -0.01 0.00 -19.2 -27.9 50.0
Figure 92 Differences between the 96-wastewater-reuse scenario and the baseline2030 scenario. Positive changes
indicate that the scenario is an improvement as compared to the baseline2030 conditions.

The outcome (Figure 92) shows that when the use of deep groundwater is reduced with 50% (last
column), more surface water is used for irrigation which is partially evapotranspirating - , thus
deteriorating the Water Exploitation Indicators and the Environmental Flow Indicators. Since less deep
groundwater is used, this scenario is more sustainable.

101

Results of the multi-criteria optimisation


Using the optimisation routine linked to the LISQUAL model, both described previously, three groups
of scenarios are evaluated. The number of scenarios per group is limited here to 5 or 6 to limit the
number of iterations needed to establish the Pareto front.
The following scenarios are evaluated (see Table 2):
A group called flooding, containing 5 scenarios that are likely increasing water retention and
reduce flood risk, consisting of afforestation (12), urban green measures (21), crop tillage
measures (34), re-meandering of rivers (43), and grassland (31).
A group called crop, containing 6 scenarios that are reduce N and P loads in rivers and reduce
water use in agriculture and industry, consisting of winter cover crops (51), optimum
fertilisation (52), winter cover crops and optimum fertilisation (53), irrigation water use
efficiency (91), crop tillage measures (34), and water re-use in industry (93).
A group called water saving, containing 6 scenarios that are aimed at water saving and should
reduce water abstraction, consisting of desalination (71), urban green measures (21), irrigation
water use efficiency (91), water re-use in industry (93), water saving in households (94), and
reduction of the current leakage in public water supply (95).

Table 2 Overview of scenarios used for optimisation

FLOOD

CROP

WATER
SAVING
12afforestation 51Nfixing
71Desalination
21urban25
52OptFertilization 91Irrigation
34crop
53Combined
93Reuse
43meander
91Irrigation
94WaterSaving
31grassland
34crop
95Leakage
93Reuse
21urban25

As explained before, the scenarios are evaluated and optimised in 21 European macro regions (). These
regions consist of entire river basins, e.g. region 13 consists of the entire Danube basin. Smaller river
basins are merged with one single region. These 21 regions are defined to avoid a single pan-European
optimisation, but also avoiding detailed optimisation which can be better done at regional/MS level.
The regions are also identified to separate major islands (Sicily, Sardinia, Corsica), for which it is
unrealistic to assume that water transfers take place with the main land. Since we wanted to avoid
defining hundreds of regions, smaller islands are included in continental macro regions (e.g. Baleares,
attached to Iberia Mediterranean). This may be unrealistic, but is done for practical reasons.
An example of the optimisation process is given below for the Danube region for the group of Water
saving measures, consisting of six scenarios (see table 2, third column).

102

Example of the procedure of the multi-criteria optimization for the Danube


Region 11 Danube: watersaving optimization
Scenario weighting - decision space
For each scenario a priority map is defined with values from 0-100% in order to rank the priority of
occurrence of this scenario measure. For example in Figure 93 (left) the priority of the irrigation
efficiency scenario is bigger east of the Carpathian Arc (darker green colours).
For each run of the optimisation process a scenario weighting is calculated for each scenario. Priority
and scenario weighting defines at which pixels the scenario is active and where it is not active. A
pixel with a higher priority has a bigger chance to be selected i.e. a pixel with priority: 80% is selected
if the scenario weighting is between 20% and 100% while a pixel with priority 10% is only selected if
the scenario weighting is between 90-100%. Figure 93 (right/top) shows the area which is selected if
the scenario weighting is 50%, (right/bottom) shows a bigger area if the scenario weighting is 75%. A
value of 0% indicates that this scenario is not used at all while a value of 100% indicates that this
scenario is used on every pixel where irrigation is possible (non-grey coloured areas in Figure 93).

Figure 93: Scenario weighting Example: Irrigation Efficiency for Region 11 - Danube

103

Objective functions solution space


Goal of the optimization procedure is to find the optimum combinations of scenarios with fit best the
objective functions below:
- Minimize the net costs of the scenarios (Cost)
- Reducing the number of days below a certain minimum flow (Env10Flow)
- reducing the Water Exploitation Index (WEI)
Costs:
Costs are the difference between the values calculated from a scenario combination and the Baseline
2030 run. It includes the cost to implement the measure, the estimated economic loss for the domestic
sector, estimated economic loss for the agricultural sector, estimated economic loss for the industrial
sector, estimated economic loss for the energy sector and the expected flood damage of one 100 year
flood. Costs are summed up for the considered area (i.e. Danube catchment) and the period of time (30
years) and displayed as cost in 1000 per km2.
It should be noted, that scenario combination may end up with negative costs, meaning that the
benefits of the scenario - i.e. reduced flood damage, reduced economic losses are larger than
investment and maintenance of the measures.

Environmental flow:
The concept of environmental flow is taken into account by calculating the 10th percentile of daily
river discharge at each location on a monthly basis, based on model simulations of the baseline 2006
scenario of a period of 30 years with no water abstraction (see also chapter on modeling methods,
LISQUAL). For the Baseline 2030 and for each scenario combination the days below the 10th
percentile are counted. A lower number of days means less days with discharge below the 10th
percentile. For the optimization, Env10Flowscenario. is calculated for each single river pixel for a period
of 30 years and spatially averaged for the catchment area. Here a lower Env10flow indicates that this
scenario combination has less days of low flow per year.

WEI (abs):
A Water Exploitation Index is calculated along the lines of the Water Scarcity and Drought Expert
Group (WSDEG). Here the absolute WEI is use, which is essentially the WEI without the return flow.
For the optimization the 90% quantile of annual WEI of a period of 30 years is calculated and only
WEI greater equal 0.5 are taken into account for calculating the spatial average. Here the aim is to
optimize for a lower WEI90%.
As there are three objective functions to optimize, it is in most cases impossible to find one optimum
scenario combination but instead you get several combinations, each one being an optimum on a
certain aspect, creating a range between a scenario combination with the lowest cost, with the lowest
EnvFlow and with the lowest WEI. These optimum scenario combinations are called the Pareto
front.

104

First optimization runs


First run of any optimization consists of the Baseline 2030 scenario, which is used to calculate the
reference costs, - consisting of the estimated economic loss under baseline conditions -, Env10Flow
and WEI. Next, six runs, each for a single scenario are executed. For example run 2 is using 100% of
the 21-urban25 scenario with all other single scenarios at 0%.
Table 3 shows the first 7 run with the weighting of each scenario (blue part) and the resulting objective
functions. The first run includes no scenario (all weighting are set to 0). The next six runs give 100%
weighting to only a single scenario.

Table 3: Details of the first scenario combinations in the Water Saving group with respect to Costs, WEI and
EnvFlow first 7 runs
Region 11
"Water saving"
Scenario
combination
Baseline 2030

Decision space
Scenario combination [weighting in %]
21_UG 71_DS 91_IE
93_WRI 94_WSH

95_LR

Solution space
Objective function
Costs
Env10Flow
[T per [days]
km2]

WEI90
[]

0.0

153.8

1.20

100

98.7

154.1

1.21

71 Desalination

100

0.0

153.8

1.20

91 Irrigation

100

-49.3

151.2

1.06

93 Re-use

100

-2.3

153.2

1.18

94 WaterSaving

100

-17.3

153.1

1.18

95 Leakage

100

-0.8

152.7

1.18

21 Urban25

shows the solution space for the first 7 optimization runs for each objective function against
the other two. The red point indicates the results of the Baseline 2030 run as starting point. Better
solutions are moving towards the utopia point in the lower left corner.

Figure 94

Baseline 2030
scenario

Run which is
Pareto

Utopia point

Figure 94 Region 11 (Danube) Water Saving: First 7 optimisation runs

105

After 30 runs
For the next optimization runs the weighting of the scenarios is calculated using a multi-criteria
genetic algorithm (more about this in chapter Modelling methods, The Optimization routine). Each
new generation (every 10 runs) is improving towards are a better Pareto front. Table 4 shows the
weighting of combinations together with the objective functions resulting from running these
combinations with LISQUAL. A value of 34 (Table 4, Comb. 26 and 21_UG) indicates that this
scenario is used on every pixel where the priority for this scenario is higher than 66% (100 - 34). A
value of 100 (Table 4, Comb. 28 and 91_IE) indicates that this scenario is used on every pixel where
the priority for this scenario is higher than 0%. If a scenario weighting is 20% only pixels of the
priority map above 80% are used i.e. only pixel in Figure 94 in dark red are used for Irrigation
Efficiency
Table 4: Details of some scenario combinations in the Water Saving group with respect to Costs, WEI and
EnvFlow
Region 11
"Water saving"
Scenario
combination

Decision space
Scenario combination [weighting in %]
21_UG 71_DS 91_IE
93_WRI 94_WSH

95_LR

Solution space
Objective functions
Costs
Env10Flow
[T per [days]
km2]

WEI90
[]

Comb. 26

34

100

71

100

57

-59.4

150.8

1.05

Comb. 27

30

96

100

71

100

100

-54.5

150.0

1.04

Comb. 28

49

89

14

22

46

-0.1

153.7

1.19

Comb. 29

24

94

100

72

99

98

-50.5

150.8

1.05

Comb. 30

77

12

46

37

14

89

0.3

153.1

1.16

Figure 95 shows the objective function plots and the Pareto front for each of them.
94 the points (and the Pareto front) are moving towards the lower left corner.

Pareto front

Figure 95 Region 11 Danube - Water Saving: First 30 optimization runs

106

Compared toFigure

After 244 runs


Figure 96 tries to visualize the three dimension problem the optimization of three goals (Costs, WEI
and Env10Flow) at the same time. The red points are the Pareto points in a 3D space. Points which are
high on the y-axe (high costs) can be as well Pareto (because WEI or/and Env10Flow are low) than
points which are low on the y-axe (low costs), but with less negative values for WEI and Env10Flow.

Figure 96 Region11 Danube - Water saving: 3D perspective from different angels

Because 3D plots are difficult to interpret Figure 97 switches back to three 2D plots, each one
explaining x-y axe, x-z axe and y-z axe separately.
Figure 97 shows an evolution of points towards the lower left corner compared to Figure 96. The left
(Cost vs. WEI) and the middle (Cost vs. Env10Flow) picture show a clear Pareto front. The right
picture shows a single optimal point, due to a high correlation between WEI and Env10Flow. All
points on the red line are on the Pareto front and therefore the optimum solutions for each pair of
objective functions. The blue points in each picture are the Pareto points of the other two figures, i.e.
on the right picture (WEI vs. Env10Flow) you will find all the Pareto points from the left and middle
picture.

107

Figure 97 Region 11 - Danube - Water Saving: After 244 optimisation runs

From optimized solution space to spatial distributed decision space


The optimization procedure produces optimum points of objective function. Each of the points in the
decision space Figure 97 stands for a combination of scenarios with the points marked with a red
border being Pareto. Table 5 shows the final Pareto points. Well take points/combinations C17 and
C66 as examples.
Table 5: Pareto points for Region 11 - Danube - Water Savings Group Optimisation
Region 11
"Water saving"
Scenario
combination

C17
C150
C60
C33
C225
C12
C181
C27
C66

Decision space
Scenario combination
21_UG 71_DS 91_IE

0
24
9
28
27
28
27
30
27

43
49
81
94
94
92
94
96
13

100
100
100
100
100
100
100
100
100

93_WRI

94_WSH

95_LR

0
22
81
70
88
73
70
71
96

100
100
100
100
100
100
100
100
100

10
52
61
66
83
96
99
100
100

Solution space
Objective functions
Costs
Env10Flow
[T per [days]
2
km ]

-60.4
-59.9
-59.8
-59.5
-59.0
-58.1
-56.2
-54.5
-53.7

150.8
150.8
150.8
150.8
150.6
150.5
150.3
150.0
149.9

WEI90
[]

1.05
1.05
1.05
1.05
1.05
1.05
1.05
1.04
1.04

The two points/combinations C17 and C66 highlighted inTable 5 and Figure 98 (orange and blue
dot), represent spatial scenario combinations as shown in Figure 99.
Figure 99 shows the spatial characteristic of all six scenarios for the two sets of combination C17 and
C66.

108

Figure 98: Region 11 Danube Water saving: two selected points along the Pareto front

Figure 99 Region 11 Danube Water saving: comparison between the two selected Pareto points (C17 and C66)

The main objectives for optimizing the water saving scenarios-group are:
costs
water exploitation index (abstraction only, without return flow), and
environmental flow (here the 10th percentile of daily river discharge on a monthly basis).
Within Figure 98, the left and middle picture show a small Pareto front with points close to eachother
but shifted to the left lower corner compared to the red starting point indicating the baseline scenario.
The right picture shows a single best point due to the strong correlation between WEI and Env10flow.
Figure 99 shows the comparison of the spatial configuration of the single scenario weightings for two
points on the Pareto front: scenario combination C17 and scenario combination C66
109

21 Urban25:
All Pareto points have low values for this scenario with a maximum up to 30% which indicates that
including only a few locations with urban measures leads to better results. The high costs associated
with this scenario probably also explains that only few locations are present in points near the Pareto
front.
71 Desalination, 93 Re-use and 95 Leakage:
The Pareto points C47 and C71 have different values for these single scenarios, but together they lead
to similar optimised output indicators. So, the same water savings objectives can be reached by
different configurations of scenario combinations.
91 Irrigation and 94 Watersaving:
Irrigation efficiency (scenario 91) and watersavings in households (scenario 94) are very important to
improve the results of the optimization. These are the two controlling scenarios to improve water
efficiency.

110

Region 11 Danube: Crop optimisation

Figure 100 Region 11 Danube Crop: Optimization after 244 runs with two selected points along the Pareto front

The main objectives for Crop optimization are:


Costs (including initial and beneficial costs),
concentration of Nitrate (N in [mg/l]), and
environmental flow (Env10flow).
Run No C05 (orange point in Figure 100) with 100% Irrigation Efficiency and all other measures at 0%
leads to an optimum results for cost and env10flow but not optimum for nitrate reduction.
Run No. C90 (blue point) includes almost 100% of scenarios 51-Nitrate-fixing, 52OptimisedFertilization, 91-irrigation-efficiency and 93-Water Re-use and therefore the entire
scenario combination is a little less cost effective, results in a few more days with flow conditions
below the threshold for environmental flow, but shows the best nitrate reduction.

111

Figure 101 Region 11 Danube Crop - comparison between the two selected Pareto points (C90 and C05)

34-Crop tillage measures:


Due to high initial costs and no gain for Env10flow this scenario is not included in the points at the
Pareto front (weighting 0%).
51-Nfixing and 52-OptimisedFertilization:
To optimise Nitrate concentration it is best to have these scenarios 100% in, but it has almost no effect
on WEI or Env10flow, and inclusion leads to slightly higher net costs.
91-IrrigationEfficiency and 93-water-re-use
Increasing irrigation efficiency is the dominant factor to decrease net costs and Env10flow. Water reuse will also reduce cost and env10flow but to a lesser extent and solutions with less than 100% re-use
can be still be on the Pareto front.

112

Region 11 Danube: Flooding optimisation

Figure 102 Region 11 Danube Flooding: Optimization after 244 runs with two selected points along the Pareto
front

(middle and right picture) shows that reducing the flood damage potential will deteriorate
the environmental flow objective function (Env10flow).

Figure 102

Figure 103 Region 11 Danube Flooding - comparison between the two selected Pareto points (C06 and C23)

The blue point is indicating a Pareto solution with 100% scenario 43 meander in. This is most cost
beneficial solution which includes also some benefits for flood reduction. If we include the scenarios
113

12-Afforestation, 21-Urban25 and 31-Grassland it will decrease the flood damage but increases the
overall costs and deteriorates Env10flow.

114

Region 13 Iberia Mediterranean: Water saving optimisation

Figure 104 Region 13 Iberia Mediterranean Water saving: Optimization after 140 runs

Figure 105 Region 13 Iberia Mediterranean Water saving - comparison between the two selected Pareto points

Similar to the Danube region, her also the 91-IrrigationEfficiency and the 94-Watersaving
scenarios are the only two significantly improving the objective functions (costs, WEI, Env10flow).
71-Desalination, 93-Re-use and 95-LeakageReduction do not make significant differences.
Scenario 21_Urban does not improve any of the objective functions therefore it is at 0%.

115

Region 13 Iberia Mediterranean: Crop optimisation


Run No C4 (blue point) with only 100% Irrigation leads to optimum results for net cost and env10flow
but not for nitrate reduction. Here we find the same result as for the Danube catchment. Run No. C5
(orange point) includes only 100% of scenarios 52OptimisedFertilization. It has the same value for
Nitrate concentration but is the most cost effective and env10flow effective run. For the Danube
catchment also 51-NitrateFixing reduces N significantly and 91-IrrigationEfficiency, 93-WaterRe-use to a lesser extent. For Iberia, the 52-Fertilization is the most important.

Figure 106 Region 13 Iberia Mediterranean Crop: Optimization after 139 runs

Figure 107 Region 13 Iberia Mediterranean Crop - comparison between the two selected Pareto points

116

Region 13 Iberia Mediterranean: Flooding optimisation

Figure 108 Region 13 Iberia Mediterranean Flooding - Optimization after 109 runs

Scenario 43 meander (run no. C6 - orange point) will decrease flood damage with optimum total net
costs. Including scenarios 12-Afforestion and 34-crop will further reduce flood damage but
increase total net costs. Different to the Danube, the scenarios 21-Urban25 and 31-Grassland are
not part of the Pareto front points, while scenario 34-CropTillage does play a role.

Figure 109 Region 13 Iberia Mediterranean Flooding: - comparison between the two selected Pareto points

117

Region 10 France Atlantic: Water saving optimisation

Figure 110 Region 10 France Atlantic Water saving: Optimization after 81 runs

Figure 111 Region 10 France Atlantic Water saving: - comparison between the two selected Pareto points (C21
and C60)

118

In the optimization group water saving the scenarios 91-IrrigationEfficiency and 94-WaterSaving
are the most effective in the France/Atlantic region with respect to net costs, env10flow and WEI.
Including further scenarios 71-desalination, 93-re-use and 95-LeakageReduction will improve
WEI and Env10flow further (orange point) but with higher net costs. Here we find the same results as
for region 11 (Danube) and 13 (Spain).

119

Region 10 France Atlantic: Crop optimisation

Figure 112 Region 10 France Atlantic Crop: Optimization after 220 runs

For region 10 France/Atlantic similar results are obtained as compared to region 13 (Iberia /
Mediterranean).
Scenario 91-IrrigationEfficiency does not reach optimum Nitrate concentration (right graph) but
reduces Env10flow and net costs (see blue point).
Scenario 52-OptimizedFertilization reaches optimum Nitrate concentration (left graph) but has a
worse Env10flow indicator (middle graph) and slightly less optimum net costs (left and middle graph).
Adding scenario 51-Nitrate-fixing improves reducing Nitrate concentration but deteriorates
environmental flow conditions (Env10flow) on the other side.

120

Figure 113 Region 10 France Atlantic - Crop: - comparison between the two selected Pareto points

121

Region 10 France Atlantic : Flooding optimisation

Figure 114 Region 10 France Atlantic Flooding: Optimization after 130 runs

As for the flooding group of scenarios in the France/Atlantic Region, the orange point (run No. C6 100% meander) shows the optimization run with the lowest net costs but also with the lowest flood
damage compared to the Baseline scenario (red point).
Including scenarios 31-grassland and 34-CropTillage (blue point - Run 67) will further reduce the
flood damage but on the other side increase overall net costs and deteriorate environmental flow
conditions (Env10flow).

122

Figure 115 Region 10 France Atlantic - Flooding: - comparison between the two selected Pareto points

123

Limitations of the current study, and further work needed


The study shows that this modelling software environment can technically deliver optimum scenario
combinations of packages of measures that improve various water quantity and water quality
indicators, but that additional work is needed before final conclusions can be made about using the
tool, especially in the areas of economic loss estimations, water prices and price-elasticity, and the
implementation and maintenance costs of individual scenarios.
The current study has a few limitations:
Water abstraction data for the reference year 2006 are used for Livestock, both for 2006 and
2030; an updated analysis for 2030 could not be finalised in time for this study, but will be
carried out at a later stage.
The price of water used in this study was that of the public sector, due to a lack of complete
data for the other sectors. As this price is typically different for industry and agriculture, in
future updated studies the different prices for different sectors will need to be taken into
account.
The definition of the implementation and maintenance costs of the water retention measures
needs further work.
The definition of the smaller water regions, which is made here based on sub-basin
boundaries and country borders, needs to be streamlined to more closely reflect the reality in
regions; since water scarcity is estimated within those regions, the size and location of these
regions matter significantly to the outcome of the study.
Large scale inter river basin water transfers (e.g. Ebro to south of Spain, Lake Constance to
Stuttgart, the Rhine to Amsterdam, etc.) are not included at the moment, but technically could
be included quite easily; a database on large-scale water transfers would need to be established.
Although the first results of the optimisation tool are quite promising and already indicate which
scenarios are the most beneficial for different objectives, it is also found to have some shortcomings
which were identified by the individual runs. To get meaningful results, more effort has to be invested
into describing objectives such as environmental flow, the Water Exploitation Index, and flood risk
and especially into the economic cost calculations (see conclusions on the individual runs).
Another problem arises from averaging values in time and space. For the optimisation tool, single
values are needed to characterise a 30-year period over a whole region. This aggregation leads to some
averaging effects such as, for example, extreme values appearing in small areas and for a limited time
being smoothed over. To overcome this problem, flood risk is evaluated for extreme values (a 100year return period) and the 90% percentile is calculated for the Water Exploitation Index.
The following is needed to update the current studies:
Redefine the costs of scenarios, e.g. the 94-Watersaving scenario
Update the economic losses resulting from water shortages in energy production, industry and
irrigation
Incorporate and assess large-scale water transfers
Run climate scenarios for the periods 2011-2040 and 2041-2070
The following data gaps are identified:
Large-scale water transfers
Local water storages

Discharge data for southern European countries, which is necessary to validate models
124

Conclusions
A modelling environment has been developed to assess optimum combinations of water retention
measures, water savings measures, and nutrient reduction measures for continental Europe. This
modelling environment consists of linking the agricultural CAPRI model, the LUMP land use model,
the LISFLOOD water quantity model, the EPIC water quality model, the LISQUAL combined water
quantity, quality and hydro-economic model, and a multi-criteria optimisation routine.
Simulations have been carried out to assess the effects of water-retention measures, water-saving
measures, and nutrient-reduction measures on several hydro-chemical indicators, such as the Water
Exploitation Index (WEI), Environmental Flow indicators, N and P concentrations in rivers, the 50year return period river discharge as an indicator for flooding, and economic losses due to water
scarcity for the agricultural sector, the manufacturing-industry sector, the energy-production sector and
the domestic sector. The potential flood damage of a 100-year return period flood has been used as an
indicator.
The study shows that technically this modelling software environment can deliver optimum scenario
combinations of packages of measures that improve various water quantity and water quality
indicators, but that additional work is needed before final conclusions can be made using the tool. This
additional work is especially needed for economic loss estimations, water prices and price-elasticity,
and the implementation and maintenance costs of individual scenarios.
Sensitivity analysis demonstrated that estimated economic loss values are very sensitive to changes in
water abstraction amounts. Where changes in the WEI and abstracted and consumed water are linearly
related to changes in abstraction values, economic losses increase up to 9-fold as a result of changes in
water abstraction. The water price has a linear influence on economic loss, but price elasticity values
between 0 and -0.2 influence the estimated loss to a much greater extent. Therefore, additional focus
needs to be given to the collection of economic data.
As for water quantity and quality, the following are the most promising individual scenarios:
The Water saving in households scenario, which aims to achieve 25% water savings in the
domestic sector through simplistic measures (replacing showerheads, etc.), improves the Water
Exploitation Index, and reduces the amounts of abstracted and consumed water. Greater water savings
are observed in several regions in the following order of importance: Great Britain (GB), Po (Milan
area), Iberia Mediterranean, southern Italy and Odra/Vistula (Warsaw area), with even greater effects
locally around populated areas. Reductions in water consumption are also simulated in many other
areas.
The scenario Irrigation efficiency which assumes that the current irrigation efficiency is improved
from the current average of 74% (Eastern Europe) - 77% (Western Europe) to 93% by applying drip
irrigation in all regions, improves the Water Exploitation Indices and the Environmental Flow Indices,
especially in the Danube, the Mediterranean Iberia, southern Italy, Sicily, Sardinia, Greece/Evros, and
the France/Atlantic macro-region. Improvements are also simulated in the other regions. An additional
benefit is the fact that the use of deep (geological) groundwater is reduced by around 20% in all
regions. Due to the larger amount of water available when less water is consumed for irrigation,
economic loss reductions are also experienced by industry, the public sector and agriculture.
The scenario Water re-use in industry, which assumes that 50% of the water abstracted for industry
is re-used, leads to improvements in the Water Exploitation Index of around 10% in several regions,
with most effects being simulated in the industrial Elbe/Ems, the regions of GB and the
Rhine/Meuse/Scheldt, as well as in southern Italy, Sardinia, Sicily, the Po and the Odra/Vistula region.
125

The Leakage reduction scenario, which assumes that 50% of the current leaks in the public water
supply are repaired, improves the Water Exploitation Index in all regions, most dramatically in GB
(24%), Ireland (38%), the Po region (6.2%), the Adige/Balkan region (5.8%), and Greece/Evros
(4.1%), with local effects even higher. It also improves the Environmental Flow Indicators by several
days per year, especially in the region of GB (8.6%), Ireland (5.8%), Sardinia (3.7%) and Sicily
(2.2%).
The Urban-greening 25% scenario - establishing urban greening measures - reduces flood peaks
(Q50) slightly, for example by 0.7% in the region of GB. This scenario consequently also reduces the
potential flood damage by 27% in the region of GB, with even greater reductions locally in England.
Further positive effects are simulated in the regions of the Rhine/Meuse/Scheldt (0.2% Q50 decrease,
12% flood damage decrease), the Elbe/Ems (0.3% Q50 decrease, 6.5% flood damage decrease), the Po
(0.2% Q50 decrease, 4.5% flood damage decrease) and Mediterranean Iberia (0.2% Q50 decrease,
6.0% flood damage decrease). On the other hand, reduced runoff from cities result in less water being
available for extraction, and thus the WEI deteriorates slightly in those areas. However, environmental
flow (10th percentile) conditions improve in the region of GB with 0.4%, likely because of increasing
baseflows as a consequence of increased infiltration. Because of reduced total river flow amounts, N
concentrations in rivers are simulated to slightly increase under the urban25 scenario in the regions of
GB (0.2%) and the Elbe/Ems (0.3%).
The N-fixing Scenario and the Optimum Fertilization Scenario both reduce Nitrate and
Phosphate concentrations in all regions that significantly engage in agriculture, most dramatically in
the Elbe/Ems region (68% and 26%), the France/Atlantic region (61% and 25%), Denmark and
northern Germany (61% and 45%), the Rhine/Meuse/Scheldt (63% and 39%), and GB (75% and
26%).
The Re-Meandering Scenario, which increases the meandering of the current rivers by increasing
the length and storage capacity of the river bed, reduces flood peaks in all European regions, and is
estimated to significantly reduce the flood damage potential especially in the Elbe/Ems (11%), Danube
(10%), Odra/Vistula (9.8%), Po (6.8%), Rhine/Meuse/Scheldt (5.3%) and the France/Atlantic region
(5.8%). At the same time, environmental flow conditions improve in some areas, for example in GB
(0.1%) and Ireland (0.3%), while environmental flow conditions are simulated to deteriorate by 0.5 to
0.9% in Mediterranean regions. Nitrate concentrations decrease in all regions, with maximum changes
of 20-25% in GB and Ireland.
The Crop Practices Scenario, which simulates the effects of the implementation of combined
methods of improved crop practices (reversed/reduced organic matter decline and increased mulching
and tillage), results in reductions to potential flood damages in all regions, including areas with high
absolute flood damages such as the Odra/Vistula (6.2% reduction), the Rhine/Meuse (7.8% reduction),
GB (15.9% reduction) and the Danube (8.2% reduction).
Installing desalination plants along the coastlines would improve the Water Exploitation Index in
several European macro-regions, and decrease the number of days during which the Environmental
Flow cannot be respected, especially in Spain and Italy.
The effects of the individual water retention scenarios are explained in detail in the NWRM report
(Burek et al., 2012).
As for multi-criteria optimisation, this report demonstrates that the optimisation tool works and
shows first experimental results. However, further work has to be carried out, especially on the cost
estimation in the modelling tools.
126

The current study shows that multi-criteria optimisation in general can be used to evaluate scenarios in
order to arrive at trade-off solutions for multiple objectives, which can be in conflict with each other.
The multi-objective evolutionary algorithm (MOEA) is an accepted approach to address issues of
opposing management goals such as best flood reduction, improved environmental flow situation
combined with lowest possible costs. A working tool was created which includes multi-objective
optimisation methods with different biophysical and economic models on a pan-European scale.
In order to reduce the number of iterations needed to reach an optimal solution, three separate groups
of measures for optimisation were selected, each including 5 to 6 single scenarios to optimise:

A group called water saving, containing 6 scenarios that are aimed at water saving and
reducing water abstraction
A group called crop, containing 5 scenarios that relate to agricultural measures and which
should reduce N and P loads in rivers and reduce water use in agriculture
A group called flooding, containing 5 scenarios that are expected to increase water retention,
reduce flood risk, and possibly improve low flow conditions.

As a first demonstration of the optimisation tool, three of the 21 European macro regions were selected
to optimise each of these three scenario groups. These are:

region 11 Danube
region 13 - Mediterranean Iberia
region 10 France/Atlantic.

In the group of water-saving scenarios, improving water use efficiency in irrigation (scenario 91) and
water savings in households (scenario 94) seem to be most effective for all three regions with respect
to costs, while also reducing the Water Exploitation Index (WEI) and Environmental Flow
(Env10flow). Including desalination (71), water re-use in industry (93) and reducing leakage from
public water supply (95) will only give a small decrease in WEI and Env10flow and with slightly
higher costs.
In the group of crop scenarios, the Optimum Fertilization Scenario (scenario 52) was the most
effective in decreasing N and P loads. The N-fixing scenario (winter cover crops, scenario 51),
increasing irrigation water use efficiency (91) and to a lesser extent the water re-use in industry
scenario (93) only helped to reduce N loads for region 11 (Danube).
In the group of flooding scenarios, all three regions show that the re-meandering of rivers scenario
(43) will decrease net costs, including a reduction of flood damage potential. The three regions
perform differently in minimising flood damage. For the Danube region (region 11) the scenarios
afforestation (12), urban green measures (21) and to a lesser extent the grassland scenario (31) will
decrease flood risk further. For region 13 (Mediterranean Iberia), afforestation (12) and crop tillage
measures (34) decrease flood damage most. For region 10 (France/Atlantic), the grassland (31) and
crop tillage measures (34) are the most cost effective in reducing flood damage.

127

References
Alcamo, J., P. Dll, F., Kaspar, and S. Siebert (1997): Global change and global scenarios of water use
and availability: An application of WaterGAP 1.0, Report A9701, Center for Environmental Systems
Research, University of Kassel, Germany.
Arnold, J.G and Fohrer, N. (2005): SWAT2000: current capabilities and research opportunities in
applied watershed modelling. Hydrological Processes. Special Issue: SWAT 2000 Development and
Application, 19 (3) 563572. doi:10.1002/hyp.5611.
Balsamo, G., S. Boussetta, P. Lopez, and L. Ferranti (2010): Evaluation of ERA-Interim and ERAInterim-GPCP-rescaled precipitation over the U.S.A., ECMWF ERA Report Series, 5(1-25 (available
online: http://www.ecmwf.int/publications/library/do/references/list/782009 ).
Batista e Silva, F., Gallego, J. and Lavalle, C. (submitted): A high resolution population grid map for
Europe, Journal of Maps.
Batista e Silva, F., Lavalle, C., Koomen, E. (in press) A procedure to obtain a refined European land
use/cover map, Journal of Land Use Science.
Bdis, K. (2009): Development of a data set for continental hydrologic modeling, European
Commission, Joint Research Centre, Institute for Environment and Sustainability. doi: 10.2788/46925.
Burek, P., Sarah Mubareka, Rodrigo Rojas, Ad de Roo, Alessandra Bianchi, Claudia Baranzelli, Carlo
Lavalle, Ine Vandecasteele (2012): Evaluation of effectiveness of Natural Water Retention Measures.
JRC report.
Dankers, R. and Feyen, L. (2009): Flood hazard in Europe in an ensemble of regional climate
scenarios, J. Geophys. Res., 114, D16108, doi:10.1029/2008JD011523.
De Roo, A.P.J., Wesseling, C.G., Van Deursen, W.P.A. (2000): Physically-based river basin modelling
within a GIS: The LISFLOOD model. Hydrological Processes, 14, 1981-1992.
De Roo, A., Odijk, M., Schmuck, G., Koster, E. and Lucieer, A (2001): Assessing the effects of land
use changes on floods in the Meuse and Oder catchments, Physics and Chemistry of the Earth, Part B,
26, 593599, doi:10.1016/S1464-1909(01)00054-5.
Duan, Q., Sorooshian, S. and Gupta, H. (1992): Effective and efficient global optimization for
conceptual rainfall runoff models, Water Resources Research (28) 10151031,
doi:10.1029/91WR02985.
FAO (2007): Gridded livestock of the world 2007, by G.R.W. Wint and T.P. Robinson, FAO, Rome.
Farr, T. G., Rosen, P. A., Caro, E., Crippen, R., Duren, R., Hensley, S., Kobrick, M., Paller,
M.,Rodriguez, E., Roth, L., Seal, D., Shaffer, S., Shimada, J., Umland, J., Werner, M., Oskin,
M.,Burbank, D. and Alsdorf, D. (2007): The Shuttle Radar Topography Mission, Reviews of
Geophysics, (45). RG2004, doi:10.1029/2005RG000183.
Feyen, L., Vrugt, J., ONuallin, B., van der Knijff, J. and De Roo, A. (2007): Parameter optimization
arnd uncertainty assessment for largescale streamflow simulation with the LISFLOOD model, Journal
of Hydrology (332) 276289, doi:10.1016/j.jhydrol.2006.07.004.
128

Feyen, J., Kalas, M. and Vrugt, J. (2008): The value of semidistributed parameters for large-scale
streamflow simulation using global optimization, Hydrological Sciences Journal (53) 293308,
doi:10.1623/hysj.53.2.293.
Gssling, S.; Peeters, P.; Hall, C. M.; Ceron, J. P.; Dubois, G.; Lehmann, L. V.; Scott, D., (2012),
Tourism and water use: supply, demand, and security. An international review., Tourism Management,
Vol. 33 No. 1 pp. 1-15
Joint Research Centre (2003): Global Land Cover 2000 database, European Commission,
Documentation and data. URL: http://bioval.jrc.ec.europa.eu/products/glc2000/glc2000.php .
Haddeland, Ingjerd, Douglas B. Clark, Wietse Franssen, Fulco Ludwig, Frank Vo, Nigel W. Arnell,
Nathalie Bertrand, Martin Best, Sonja Folwell, Dieter Gerten, Sandra Gomes, Simon N. Gosling,
Stefan Hagemann, Naota Hanasaki, Richard Harding, Jens Heinke, Pavel Kabat, Sujan Koirala, Taikan
Oki, Jan Polcher, Tobias Stacke, Pedro Viterbo, Graham P. Weedon and Pat Yehm (2011):
Multimodel Estimate of the Global Terrestrial Water Balance: Setup and First Results. Journal of
Hydrometeorology (12) 869-884. doi:10.1175/2011JHM1324.1.
Hiederer, R. and De Roo, A. (2003): A European flow network and catchment data set, Joint Research
Centre, European Commission.
Hoekstra, A.Y. and M.M.Mekonnen (2011): Global water scarcity: the monthly blue water footprint
compared to blue water availability for the worlds major river basins. Value of Water Research
Report Series No. 53. UNESCO-IHE.
King, D., Daroussin, J., and Tavernier, R.(1994): Development of a soil geographic database from the
Soil Map of the European Communities, Catena, 21, 3756, doi:10.1016/0341-8162(94)90030- 2.
Klein Goldewijk, K. and G. van Drecht (2006): HYDE 3: Current and historical population and land
cover in Integrated modelling of global environmental change: An overview of IMAGE 2.4 by A.F.
Bouwman et al. (Eds), pp. 93-112, MNP (Netherlands Environmental Assessment Agency), Bilthoven,
the Netherlands.
Lavalle, C., C. Baranzelli, et al. (2011). A High Resolution Land Use/Cover Modelling Framework for
Europe: Introducing the EU-ClueScanner100 Model Computational Science and Its Applications ICCSA 2011. B. Murgante, O. Gervasi, A. Iglesias, D. Taniar and B. Apduhan, Springer Berlin /
Heidelberg. 6782: 60-75.
Nagy M., Lenz K., Windhofer G., Frst J., Fribourg-Blanc B., (2007), Data Collection Manual for the
OECD/Eurostat Joint Questionnaire on Inland Waters Tables 1 7. Concepts, definitions, current
practices, evaluations and recommendations, Version 2.21.
RWE Npower (2012), Water use at thermal power plant quantification, metrics, and social benefit.
Report ENV/485/2012 prepared for DG Environment Water Blueprint ad hoc Modelling Group, pp.
39.
Snchez, Rafael and G. Schmidt. 2012. Environmental flows in the EU - discussion paper - draft 3.0.
European Commission, Intecsa-Inarsa s.a. and Typsa.
Shiklomanov, I.A. (ed.) (1997): Assessment of water resources and water availability in the world.
Comprehensive assessment of the freshwater resources of the world. WMO and SEI.
129

Simmons, A., S. Uppala, D. Dee and S. Kobayashi (2007): ERA-Interim: New ECMWF reanalysis
products from 1989 onwards, in Newsletter 110 - Winter 2006/07, ECMWF, pp. 11.
Torcellini, P., N. Long, R. Judkoff (2003), Consumptive Water Use for U.S. Power Production.
National Renewable Energy Laboratory, Golden, Colorado. NREL/TP-550-33905.
Van Beek, L. P. H., Y. Wada, and M. F. P. Bierkens (2011): Global monthly water stress: 1. Water
balance and water availability, Water Resour. Res., 47, W07517, doi:10.1029/2010WR009791.
Van der Knijff, J., J. Younis, and A. de Roo (2010): LISFLOOD: a GIS-based distributed model for
river basin scale water balance and flood simulation, Int. J. Geogr. Inf. Sci., 24, 189212,
doi:10.1080/13658810802549154, 2010.
Viviroli, D., Zappa, M., Gurtz, J., & Weingartner, R. (2009): An introduction to the hydrological
modelling system PREVAH and its pre- and post-processing-tools. Environmental Modelling &
Software, 24(10), 12091222.
Vrsmarty, C. J., C. Leveque and C. Revenga (2005): Chapter 7: Freshwater ecosystems, in
Millennium Ecosystem Assessment Volume 1: Conditions and Trends by C. Caudill et al. (eds.),
Island Press, Washington DC.
Wada, Y., van Beek, L. P. H. and M. F. P. Bierkens (2011a): Modelling global water stress of the
recent past: on the relative importance of trends in water demand and climate variability, Hydrol. Earth
Syst. Sci., 15, 3785-3808, doi:10.5194/hess-15-3785-2011.
Wada, Y., van Beek, L. P. H., Viviroli, D., Drr, H. H.,Weingartner, R. and M. F. P. Bierkens
(2011b): Global monthly water stress: II.Water demand and severity of water, Water Resources
Research (47) W07518, doi:10.1029/2010WR009792.
World Tourism Organization (2000), Tourism 2020 Vision, Volume 4. Europe
Wsten, J., Lilly, A. and Le Bas, C. (1999): Development and use of a database of hydraulic properties
of European soils, Geoderma, 90, 169185, doi:10.1016/s0016-7061(98)00132-3.
Wriedt G and Bouraoui F. (2011): A spatially distributed assessment of water allocation in EU27 for
Year 2000. EUR 24913 EN. Luxembourg (Luxembourg): Publications Office of the European Union.
Wriedt G, Van Der Velde M, Aloe A and Bouraoui F. (2008): Water Requirements for Irrigation in the
European Union. EUR 23453 EN. Luxembourg (Luxembourg): OPOCE.
Wriedt G, Van Der Velde M, Aloe A and Bouraoui F. (2009): A European Irrigation Map for Spatially
Distributed Modelling. Agricultural Water Management 96 (5). p. 771-789.
WRI (World Resources Institute) (1998): World Resources: A Guide to the Global Environment 199899. Washington DC.
Data sources:
EUROSTAT Annual water abstraction by source and by sector
http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=env_watq2&lang=en
130

FAO AQUASTAT Water withdrawal by sector


http://www.fao.org/nr/water/aquastat/data/query/index.html?lang=en
POLES Prospective Outlook on Long-term Energy Systems Model - administered by the JRC,
IPTS, Seville
http://ipts.jrc.ec.europa.eu/activities/energy-and-transport/documents/POLESdescription.pdf
GEM-E3 - General Equilibrium Model for Economy Energy Environment, administered by the
E3M Lab, National Technical University of Athens, Greece
http://ipts.jrc.ec.europa.eu/activities/energy-and-transport/gem-e3/
European Pollutant Release and Transfer Register (E-PRTR) data base
http://prtr.ec.europa.eu/

131

European Commission
EUR 25552 EN Joint Research Centre Institute for Environment and Sustainability
Title: A multi-criteria optimisation of scenarios for the protection of water resources in Europe
Author(s): Ad de Roo, Peter Burek, Alessandro Gentile, Angel Udias, Faycal Bouraoui,
Alberto Aloe, Alessandra Bianchi, Alessandra La Notte, Onno Kuik, Javier Elorza Tenreiro, Ine Vandecasteele, Sarah
Mubareka, Claudia Baranzelli, Marcel Van Der Perk, Carlo Lavalle, Giovanni Bidoglio
Luxembourg: Publications Office of the European Union
2012 134 pp. 21.0 x 29.7 cm
EUR Scientific and Technical Research series ISSN 1831-9424
ISBN 978-92-79-27025-3
doi: 10.2788/55540

Abstract
A modelling environment has been developed to assess optimum combinations of water retention measures, water
savings measures, and nutrient reduction measures for continental Europe. This modelling environment consists of
linking the agricultural CAPRI model, the LUMP land use model, the LISFLOOD water quantity model, the EPIC water
quality model, the LISQUAL combined water quantity, quality and hydro-economic model, and a multi-criteria
optimisation routine.
Simulations have been carried out to assess the effects of water retention measures, water savings measures, and
nutrient reduction measures on several hydro-chemical indicators, such as the Water Exploitation Index, Environmental
Flow indicators, N and P concentrations in rivers, the 50-year return period river discharge as an indicator for flooding,
and economic losses due to water scarcity for the agricultural sector, the industrial sector, and the public sector. Also,
potential flood damage of a 100-year return period flood has been used as an indicator.

z
LB-NA-25552-EN-N

As the Commissions in-house science service, the Joint Research Centres mission is to provide
EU policies with independent, evidence-based scientific and technical support throughout the
whole policy cycle.
Working in close cooperation with policy Directorates-General, the JRC addresses key societal
challenges while stimulating innovation through developing new standards, methods and tools,
and sharing and transferring its know-how to the Member States and international community.
Key policy areas include: environment and climate change; energy and transport; agriculture
and food security; health and consumer protection; information society and digital agenda;
safety and security including nuclear; all supported through a cross-cutting and multidisciplinary approach.

Vous aimerez peut-être aussi