Project no.: 022793 FORESCENE Development of a

Project no.: 022793

FORESCENE

Development of a Forecasting Framework and Scenarios to Support the EU Sustainable

Development Strategy

Instrument: STREP

Thematic Priority 8.1: Policy-oriented research, scientific support to policies, integrating and

strengthening the European Research Area

D.4.1-2 – Technical report

Baseline scenario assumptions and model projections for selected target

indicators

Submission date: February 2009

Start date of project: 1/12/2005 Duration: 36 months

Organisation name of lead contractor for this deliverable:

Wuppertal Institute for Climate, Environment and Energy

Revision: final

Project co-funded by the European Commission within the Sixth Framework Programme (2002-2006)

Dissemination Level

PU Public X

PP Restricted to other programme participants (including the Commission Services)

RE Restricted to a group specified by the consortium (including the Commission Services)

CO Confidential, only for members of the consortium (including the Commission Services)

2

FORESCENE D.4.1-2 – Technical report

Baseline scenario assumptions and model projections for selected target parameters

3

Baseline scenario assumptions and model

projections for selected target indicators

Technical Report of Work Package 4

Stefan Bringezu, Mathieu Saurat

Roy Haines-Young, Alison Rollett

Mats Svensson

Wuppertal Institute for Climate, Environment and Energy

University of Nottingham, Centre for Environmental Management

Lund University, Centre for Sustainability Studies

4



5

Table of Content

1 INTRODUCTION 11

2 METHODOLOGY AND MODEL STRUCTURE 15

2.1 Bayesian modelling 15

2.2 Dealing with time 15

2.3 Dealing with uncertainties 17

2.4 General structure and system boundaries 18

2.5 Integration of the model components, linkages between the environmental topics 21

2.6 Data Base 21

2.7 Methodology for forecasting 22

3 BASELINE ASSUMPTIONS FOR THE INPUT NODES 25

3.1 Resource use and waste generation submodel 25

3.1.1 Economy module 25

3.1.2 Mineral materials module 25

3.1.3 Fossil fuels module 27

3.1.4 Transport module 29

3.1.5 Land use module 30

3.1.6 Greenhouse gases module 31

3.2 Water use submodel 33

3.3 Biodiversity, soils and landscape submodel 34

3.3.1 Biodiversity and landscape module 34

3.3.2 Soils module 45

4 BASELINE FORECASTING RESULTS FOR SELECTED INTERMEDIARY AND

TARGET NODES 49

4.1 Resource use and waste generation submodel 49

4.1.1 Economy module 49

4.1.2 Mineral materials module 49

4.1.3 Fossil fuels module 52

4.1.4 Transport module 54

4.1.5 Greenhouse gases module 56

4.2 Water use submodel 58

4.3 Biodiversity, soils and landscape submodel 59

6

5 CONCLUSIONS 65

6 REFERENCES 67

7 APPENDIX 74

7.1 Conditional Probability Tables (CPTs) 74

7.2 Values for model input nodes of the biodiversity, soils and landscape submodel for

baseline scenarios 2000-2050 76

7.3 Definitions of forest land from Eurostat Forestry Statistics 2007 84



7

List of Tables Table 1: Data base of the FORESCENE model prototype 22

Table 2: Mean material intensity values in EU-25 in 2000 26

Table 3: Mean values for hidden and indirect flow coefficients for EU-25 in 2000 and baseline

assumptions for the development until 2020 (mineral materials) 27

Table 4: Baseline assumption for gross inland energy consumption (GIC) (EC 2008) 28

Table 5: Mean values for hidden and indirect flow coefficients for EU-25 in 2000 and baseline

assumptions for the development until 2020 (fossil fuels and nuclear) 29

Table 6: Baseline assumptions for the transport module 29

Table 7: Baseline assumptions regarding biofuel consumption for the land use module 31

Table 8: Baseline assumptions regarding biofuel yields for the land use module (outside EU)

31

Table 9: GHG emissions improvements due to the use of biofuels compared to conventional

fuel (without land use change) 32

Table 10: Carbon debt 32

Table 11: Baseline assumptions for the water submodel 34

Table 12: Probability of percentage spend on AEP 35

Table 13: Probability of non-compliance with cross compliance regulations 35

Table 14: Certified forest area by region, 2005-2006 36

Table 15: Conservation status for mammal and plant species 2000-2005 and projections for

2010 40

Table 16: Projected probability of favourable and unfavourable conservation status for EU25

bird species (input node) 42

Table 17: Assumed increase in percentage of protected land area in the EU25 (input node)

45

Table 18: Projected share of gross domestic expenditure in R & D (% of GDP) (input node) 46

Table 19: Values for model input nodes of the biodiversity, soils and landscape submodel for

baseline scenarios 2000-2050 76


baseline scenarios 2000-2050 (Continued 1) 77







8



Table 25: List of nodes values with value used in place of normal distribution in baseline

model test 82



9

List of Figures Figure 1: Socio-industrial metabolism and Driving Forces-Pressures-State-Impacts-

Response (DPSIR) framework in FORESCENE 13

Figure 2: The six questions of the FORESCENE framework 14

Figure 3: First option for dealing with time in Bayesian networks 16

Figure 4: Second option for dealing with time in Bayesian networks. 17

Figure 5: General overview of the structure of the FORESCENE model prototype. 18

Figure 6: Graphical representation of the Bayesian network for the FORESCENE meta

model 20

Figure 7: Forecasting method in the FORESCENE Bayesian network model prototype 24

Figure 8: Expected changes in habitat status from 2000-2050 under baseline projections

(input nodes) 38


(continued) 39

Figure 10: Expected changes in species status from 2000-2050 under baseline projections

(input node) 41

Figure 11: Methodology for determination of EU25 conservation status for bird species 42

Figure 12: Current trends in EU mammal and bird populations 43

Figure 13: Projected probability of declining, stable, increasing or unknown trend in mammal

and bird populations (input nodes) 44

Figure 14: Projected probability that alien species are decreasing, stable or increasing over

the time period 2000-2050 (input node) 45

Figure 15: Projected greenhouse gas emissions for the period 2000-2050 (input node) 47

Figure 16: Assumed nutrient transfer efficiency for grassland and arable from 2000-2050

(input node) 47

Figure 17: Projected change in fertiliser use intensity for the baseline scenario 2000-2050

(input node) 48

Figure 18: Baseline modelling results for GDP 50

Figure 19: Baseline modelling results for TMR minerals 51

Figure 20: Baseline modelling results for the ratio foreign vs. domestic TMR minerals 51

Figure 21: Baseline modelling results for TMR fossil fuels 53

Figure 22: Baseline modelling results for the ratio foreign vs. domestic TMR fossil fuels 53

Figure 23: Baseline modelling results for biofuel consumption 54

Figure 24: Baseline modelling results for transport 55

Figure 25: Baseline modelling results for GHG emissions from fossil fuels 56

10

Figure 26: Baseline modelling results for GHG emissions savings from biofuel without land

use change 57

Figure 27: Baseline modelling results for GHG emissions savings from biofuel including land

use change 57

Figure 28: Baseline modelling results for total water withdrawals 58

Figure 29: Baseline modelling results for water balance 59

Figure 30: Modelled probability of aquatic, terrestrial and overall biodiversity status 60

Figure 31: Predicted model probability for soil erosion and soil carbon and the output node

soil quality 62


soil quality 63

Figure 33: Predicted model probability of species conservation status and trend in species

number 64



11

1 Introduction

Since the Gothenburg summit in 2001, the implementation of the concept of sustainable

development has been a core challenge for policy making in the European Union. To

improve the basis for policy design, and also comply with the specific needs of ex-ante

impact assessments, there is a need for a forecasting framework to develop harmonised

middle and long-term baseline and alternative policy scenarios. In the context of the

Sustainable Development Strategy (European Commission 2001; Council of European

Union, 2006) the forecasting framework should allow to develop scenarios that can be used

for strategic policy preparation to better specify and disentangle the mutual relationships

between environmental, economic and social trends.

To be effective, policy development and appraisal needs to understand the key driving forces

and their cross-cutting linkages, which lead to increased pressure on different aspects of the

environment. Measures which are designed to solve single problems often risk shifting the

burden to other sectors, and they may be ineffective due to the complex interaction of

environmental effects. The need for policies to be based on a cross-cutting approach has

been highlighted in “The 2005 Review of the EU Sustainable Development Strategy: Initial

Stocktaking and future orientations” (COM(2005)37 final), as well as in the Commission’s

Communication on its Strategic Objectives 2005-2009 (COM (2005) 12). However, cross-

cutting driving forces which are relevant for various environmental and sustainability related

problems have not yet been analyzed in a systematic policy-oriented manner.

The FORESCENE project is an attempt to identify cross-cutting measures that are effective

and efficient, and which have the potential to mitigate several environmental problems at the

same time. As shown in Figure 1, three broad environmental themes are covered: ‘resource

use and waste generation’, ‘water’, and ‘biodiversity, soils and landscape’. The aim of the

project has been to capture the anthropogenic cross-cutting drivers that generate

environmental pressures and the inter-linkages between them, so that an integrated

forecasting framework could be constructed. It should, in turn, offer the possibility to test

alternative policy strategies against some baseline. To do so, the framework used in

FORESCENE was inspired and adapted from the backcasting methodology. As shown in

Figure 2, the project’s framework consists of six questions that were addressed during a

series of workshops. By applying the framework the project aimed to:

• determine cross-cutting (cross-thematic and cross-sectoral) driving forces for

environmental problems in the three targeted fields (relates to Question 1);

• define essential elements of sustainable development in the different topic areas,

particularly in the form of desired sustainability goals (relates to Question 2);

• describe cross-sectoral measures expected to exert a multi-beneficial impact over the

environmental fields considered, so that the sustainability goals identified can be

achieved (relates to Question 3); and,

• combine the scenario elements (driving forces, strategies, goals etc) into baseline

and alternative scenarios (relates to Questions 4 to 6).

12

The present report describes the approach taken to address Question 5 “What would happen

under business-as-usual conditions?” shown in Figure 2. As a result of Step 3 a model

prototype based on the Bayesian network methodology was developed. It consists of

interlinked sub-models featuring the environmental issues and cross-cutting drivers identified

in Step 1. The model prototype can then be used for forecasting exercises. The structure of

the model prototype, the methodological approach to Bayesian modelling, forecasting and

backcasting, and related issues are described in chapter 2.

Chapter 3 presents baseline assumptions made for the input nodes (driving and control

parameters) of the different modules of the model prototype. This lays the grounds for

modelling the baseline (or business-as-usual) scenario.

Using the forecasting methodology described in chapter 2, time series of key output variables

are modelled based on the assumptions made in chapter 3. The baseline projections for

selected intermediary and target parameters are presented in chapter 4. These baseline

modelling results will serve as a reference for comparison when alternative scenarios are

quantified in WP5.

Chapter 5 finally discusses the possibilities and perspectives to further develop the

framework and model. The ‘prototype’ baseline projections obtained in this workpackage also

offer perspectives for modelling alternative scenarios in WP5.



13

Figure 1: Socio-industrial metabolism and Driving Forces-Pressures-State-Impacts-

Response (DPSIR) framework in FORESCENE

14

Figure 2: The six questions of the FORESCENE framework



15

2 Methodology and model structure

2.1 Bayesian modelling

Bayesian networks (BNs) are an increasingly popular method applied to uncertain and

complex domains such as environmental modelling and management. They can be used as

or in combination with decision tools. BN models offer the possibility to incorporate

knowledge of different accuracies (e.g. absence/presence of an observation vs. quantitative

data from measurements) and from different sources (Marcot et al. 2001). Carefully elicited,

expert knowledge can be combined with empirical data, in a mathematically coherent

manner. Parameter values that come with uncertainties can be expressed as probability

distributions rather than average values. The higher the uncertainty, the wider is the

probability distribution. The end-point modelling results are also represented as probability

distributions which prevents overconfidence in single estimated values and allows for an

estimation of risks and uncertainties (Uusitalo 2007).

The Bayesian network approach to modelling presents, however, three main shortcomings

(Borsuk et al. 2004, Uusitalo 2007). First, BNs are acyclic graphs and therefore do not

support feedback loops (Jensen 2001). Second, and closely related shortcoming, is the

difficulty of modelling temporal dynamics in BNs. A possible workaround consists in using a

separate network for each time slice (Jensen 2001, Uusitalo 2007). This is however often

very tedious. The third difficulty associated with BN models is their limited ability to deal with

continuous data when used in compiled form. In such a form, the continuous variables and

parametric equations between variables need to be discretized over discrete domains

chosen by the user. This implies a trade-off as the discretization can only account for rough

characteristics of the original continuous distributions and relationships (Friedman and

Goldszmidt 1996). The problems inherent to the discretization of continuous variables can be

avoided by using the other major approach in Bayesian modelling beside Bayesian networks,

namely hierarchical simulation-based modelling (Uusitalo 2007, Gelman et al. 1995).

In the FORESCENE model prototype both branches of Bayesian modelling are implemented.

The network approach is used, for example, with the 'biodiversity, soils and landscape'

submodel that mainly consists of discrete variables. Once the submodel is compiled the BN

can provide instant responses to queries such as the modification of an input parameter. The

result is also directly visible in the graphical representation of the network.

The simulation-based approach is used for other modules of the model prototype that contain

mainly or exclusively continuous variables. The probability distributions of target parameters

are estimated by generating samples from these distributions by simulation (Gelman et al.

1995, Borsuk et al. 2003). The results, however, require some processing before they can be

shown. But, on the other hand, the least possible information is lost.

2.2 Dealing with time

It is important to note that the time dimension is not explicitly present in a Bayesian network

model. Using a collection of networks is an option to represent an evolution over time. Figure

16

3 shows how time dynamics is dealt with in FORESCENE, for some parts of the model

prototype (e.g. biodiversity and soils submodel). The networks of the time series differ from

one another by the input values of their marginal nodes, which correspond to initial input

values for given years. These inputs are calculated separately and for different scenarios.

The target nodes from each network then deliver output values for each year considered,

which can, in turn, be worked out into time series outside the networks.

Another option consists in having the different years in one overall network (see Figure 4).

This option is adapted for networks used with simulation based modelling (see previous

section). It also facilitates the updating of input node values, for example for different

scenarios, as there is only one file to update. However, this option is to be reserved to

networks that are not too large and ramified.

Figure 3: First option for dealing with time in Bayesian networks



17

Figure 4: Second option for dealing with time in Bayesian networks.

Note: Node types are as follows: input and control nodes (turquoise and orange), intermediary nodes

(pink), output/target nodes (yellow)

2.3 Dealing with uncertainties

Building a model and operating it is confronted to two main types of uncertainties: uncertainty

in the causal structure and parameter uncertainty. In the present model prototype, as in most

models, the first type of uncertainty is not quantified. Therefore, the real uncertainty in model

predictions will be greater than that suggested by the model. In the case of Bayesian

networks, the following options have been suggested for addressing uncertainty in model

structure: Bayesian model averaging, learning from additional data and model testing

(Borsuk et al. 2004). Considering the third option, we have compared the modelling results

from the model prototype with existing data. This comparison, however, does not provide a

true validation, as much of the model building was based on the same or related research

that generated the data.

However, in contrast to most modelling approaches, parameter uncertainty can be accounted

for at a very early stage in the modelling process, rather than at the end using sensitivity

analysis. When setting up input and control nodes, parameter uncertainty can be captured by

probability distributions. The modelling results from the output nodes can therefore also be

expressed as probability distributions that reflect the combined uncertainties from all input

and control parameters, and hence uncertainty in model predictions. This probabilistic

approach implies that even with a fixed criterion regarding the desired value of a given output

parameter, the choice of action on the control parameters depends on the degree of

confidence required by decision makers. That degree of confidence can be translated directly

in terms of percentiles of the probability distributions of target parameters.

In most modelling approaches, such a margin of safety in obtaining the desired outcome

would be sought through conservative model assumptions. Though such assumptions can

be justified, the modeller implicitly chooses a particular level of confidence in making them.

Such a task, however, is a risk management decision that should better be made by decision

makers. Therefore a model approach that accounts for prediction uncertainties provides an

18

explicit basis for choosing a decision criterion that includes a margin of safety. The size of

the margin, decided by decision makers, might be reduced by the decision makers

themselves if they settle for a lower degree of confidence, or by reducing prediction

uncertainty. The latter requires further data collection to reduce the uncertainty of input

parameters and of relationships between parameters. This, in turn, requires further

communication with stakeholders, experts and, eventually, decision makers. In that respect,

the explicit dealing with uncertainty of the Bayesian approach fits well in a heuristic

approach.

2.4 General structure and system boundaries

The general structure of the model prototype developed in the FORESCENE project is

shown in Figure 5. The geographical system boundaries are that of the EU-25 with regard to

the driving forces but the impacts on most of the environmental issues are considered at the

global level.

Figure 5: General overview of the structure of the FORESCENE model prototype.

Note: The components of the model considered beyond the EU-25 system borders are represented

with shadows.

More detailed, Figure 6 outlines the general structure of the model prototype, showing

variables and dependencies relevant for the selected environmental fields: ‘resource use and

waste generation’, ‘water’ and ‘biodiversity, landscape and soils’. The central part of the

figure represents the organising structure of the model, consisting of separately developed

modules. In the graphical representation of a Bayesian network, variables are depicted by

nodes, and a dependence between one variable and another is represented by an arrow.

The absence of an arrow between any two nodes implies the conditional independence.

Nodes can be either discrete (i.e. with a defined finite set of possible values called states) or

continuous (i.e. can take in a value between any other two values). The ‘sub-models’ are the

component parts of the model prototype, covering one of the three environmental topics, and

‘modules’ are component parts within a sub-model.

The following sub-models and modules are shown in Figure 6:



19

• Economy

• Mineral materials

• Fossil fuels

• Biofuels

• GHG emissions

• Agricultural land use

• Water

• Biodiversity and soils

Some modules have been or could be developed in more complex versions, in order, for

example, to further differentiate and parametrise mineral, fossil fuel, greenhouse gas and

crop types. This potential for refinement illustrates the strength of the Bayesian approach

which allows for gradual elaboration of individual elements. The biofuel module actually also

includes a simplified transport module. The biomass and agricultural land use is not fully

functional at this development stage of the model prototype. Solely land use for biofuel crop

production within and outside the EU is calculated from within the BN module at that stage.

Results from existing land use models are used in an aggregated form outside the BNs to

ensure consistency of the input data sets and of the results stemming from the functioning

part of the module.

Any Bayesian network can probably be best explained by starting with their end-points and

proceeding in the “up-arrow” direction. When building the diagram, intermediate variables

and relationships are only included if they contribute to the model’s ability to predict values of

the end-point indicators, that is if they are controllable (e.g. human dependent variables),

predictable or observable at the scale of the modelling exercise. If a variable does not have

one of these characteristics, then it is not explicitly included, and the resulting variability

becomes part of the uncertainty associated with the model (Borsuk 2004).

For some modules (e.g. economy) the end-point or intermediary nodes deliver interim results

that are used in the other modules. The end-point indicators of most modules (e.g. TMR

minerals for the minerals module) are proxies for the assessment of the progress made

towards the sustainability goals. In developing the model structure, one logically starts with

the end-point nodes, which are then related to their immediate causal variables, which are

then related back to their own causes, and so on, back to the drivers. The parentless nodes

at the top of the diagram can either be considered marginal variables representing natural

variability, or those that will be influenced by sustainability strategies.

Causal relationships between the nodes are characterized either with functional equations or

conditional probability tables (CPTs, see Appendix). The latter are the contingency tables of

conditional probabilities stored at each node, containing the probabilities of the node given

each configuration of parent values. The first option is, when possible, to be preferred

because the specification of functional equations between variables reduces the variability of

the results. Probabilities are introduced into the functional equations by assigning probability

distributions to the equation parameters, which represent knowledge uncertainty about the

parameter value from a Bayesian perspective (see section ‘Dealing with uncertainty’).

20

Figure 6: Graphical representation of the Bayesian network for the FORESCENE meta model

Central figure: submodels are indicated with rectangular blocks, model parameters with round nodes, and causal relationships are indicated with arrows.

Surrounding insets: networks representing submodels of the main network. (Note: network images used in this figure do not correspond to their latest

versions).


Baseline scenario assumptions and model projections for selected target indicators

21

2.5 Integration of the model components, linkages between the environmental topics

FORESCENE aimed to focus on the interdependencies between the three environmental topics

‘resource use and waste’, ‘water’, and ‘biodiversity and soils’. At the present stage of development

not all interdependencies considered are explicitly modelled in the Bayesian networks. Some

linkages, such as those between agricultural land use and biodiversity, are accounted for but are

partly exogenously determined. For example, because the biomass and land use module is not

fully functional at the present point (it is limited to biofuels), the inputs ‘agricultural land use in EU’

or ‘intensity of agriculture in EU’ are determined exogenously based on results from existing

models and studies and injected into the model, after ensuring consistency with results from the

other BN modules.

The in-built modular structure of the model prototype implies that attention needs to be paid to the

linkages between the modules, also within a sub-model (corresponding to one environmental

topic). The input nodes of one given module consist of its own marginal nodes and of output or

intermediary nodes from another module. Often, the purpose of a marginal node in one module is

to operationalize a linkage between this module and a ‘parent’ module that influences it. For

example, water intensity coefficients (marginal/input nodes of the water module) operationalize the

linkages with the mineral materials and fossil energy modules by associating water abstraction to

material use. Therefore, input data sets generated exogenously (such as water coefficients) and

fed into the marginal nodes need to be consistent across all modules as well as with regard to

modelling results coming as input data from other modules.

2.6 Data Base

Data mining and literature assessment, conducted in order to ‘populate’ the model prototype,

served the following purposes:

! provide a basis of empirical data and expert statements from which relationships between

nodes could be characterised and quantified (whether parameters in functional equations,

or probabilities in CPTs);

! account for parameter uncertainty with probability distributions when different studies

publish diverging data;

! compile time series of input data to be fed into the marginal nodes of the model prototype.

Assumptions and modelling results from existing business-as-usual forecasts were considered in

FORESCENE to build input data sets for the baseline scenario. Knowledge from different sources

could be integrated into one data set while acknowledging their diversity through the use of

probability distributions rather than point average values.

To each module in the model corresponds a spreadsheet containing time series (in five year time

slices) for the input nodes. Spreadsheets are linked in a way to ensure consistency of input data

across the modules (see also section 3.7 ‘Methodology for forecasting’ below). An overview of the

data base of the FORESCENE model prototype is shown in Table 1.

22

Table 1: Data base of the FORESCENE model prototype

Module Data sources

Economy Eurostat, GINFORS model, EEA

Mineral materials WI data base, ETC/RWM NAMEA, MOSUS

Fossil fuels WI data base, Eurostat, EEA, DG TREN

Biofuels WI data base, Eurostat, EEA, DG-ENER

GHG emissions ETC/AAC, IPCC, UNEP/IPSRM, DG TREN

Agricultural land use FAO, FAPRI, EEA, UNEP/IPSRM, Eurostat

Water Aquastat, Eurostat

Biodiversity and soils EEA, Eurostat, FAO, DG AGRI, and diverse

specialised literature sources

2.7 Methodology for forecasting

This section explains the forecasting technique used in FORESCENE. Figure 7 gives an overview

of the different steps involved. The upper, middle and lower parts of the figure summarize the

process into three phases:

! Construction of the network, parameterization and characterization of the linkages

! Construction of input data sets for each module

! Model runs, representation and analysis of the results.

The upper and lower parts of the figure are further divided into two boxes. On the left-hand side the

technical description concerns network modelling, while on the right-hand side simulation-based

modelling is described. The former option is reserved to BNs with discrete variables (e.g.

‘Agricultural management’ is ‘green’ or ‘brown’) or continuous variables that can be discretized

without increasing the variability of the results (e.g. ‘pollutant load’ with known thresholds). The

latter method regards BNs with continuous variables linked by functional equations, which would

be much less accurate if discretized.

Both approaches start with a graphical representation of the model structure (upper part of Figure

7). In FORESCENE, a commercial software called Netica1 was used. Parameters deemed

necessary to be considered can stem from participatory processes as in the Steps 1 and 2 of the

FORESCENE framework. The quantitative relationships characterizing the conditional distributions

between the model components need then to be established. The degree of belief and uncertainty

underlying the functional relationships between the variables is acknowledged through probabilities

in CPTs or equations with parameters expressed as probability distributions.

1 www.norsys.com



23

Once the linkages are characterized, the marginal nodes (without parent nodes) are updated with

their probability distributions (discrete or continuous). Necessary information are read for each time

slice, for each marginal node of each module by a vba-macro from a common spreadsheet and

written into the text-file (i.e. non graphic) version of the BN.

When the model is “populated”, one can re-open its graphical version and run the forecast. With

discrete nodes and a compiled network (lower part, left-hand side of Figure 7) the modelled

probability distributions of target nodes over their discrete states (e.g. ‘Biodiversity status’ is

declining, stable or improving) can be read directly in Netica. Each time slice is modelled in a

separate BN. If an input node is modified, the influence of the change immediately appears.

For simulation-based modelling, sample cases should be generated (Netica has a Monte Carlo-like

function for it) in a sufficient number to cover the extent of the probability distributions of the target

nodes. Then, using the open source softwares for numeric calculation Scilab2 or for statistics R3,

these distributions can be visualised and their characteristics calculated. With the simulation-based

Bayesian modelling, results are more than single point values: the predictive precision, given

model uncertainties, of policy relevant variables is quantified.

2 www.scilab.org

3 www.r-project.org

24

Figure 7: Forecasting method in the FORESCENE Bayesian network model prototype



25

3 Baseline assumptions for the input nodes

3.1 Resource use and waste generation submodel

3.1.1 Economy module

In its current stage of development, the model prototype does not feature an endogenous

representation of economic development. The economy module is not an econometric model.

Driving parameters such GDP per capita and population level are determined exogenously from

existing studies. The module then calculates GDP of EU-25 which level serves as a driver for

material and energy use, modelled in subsequent modules.

Projections of population growth are taken from Eurostat. From 2000 to 2015, population in EU-25

increases at a 1% rate per 5-year interval. The growth rate then slows down and becomes negative

from 2030 onwards. In 2050, the EU-25 hosts 2.3 million inhabitants less than in 2000.

Empirical data for GDP per capita (also referred to as economic activity) are taken from Eurostat

for the period 2000-2005 (7.5% increase over the period). For the period 2005-2020, an annual

average growth rate of 2% is taken for GDP per capita from the baseline of the GINFORS

economic model (MOSUS project). Other sources are slightly more optimistic, for example EEA

(2005) assumes an average annual growth rate of 2.5% for GDP per capita in the EU-25. To take

this observation into account and build the FORESCENE baseline scenario as a “corridor”

including both pessimistic and optimistic views, the growth of GDP per capita is modelled with a

normal distribution (mean = 2%, standard deviation = 0.3%). The distribution chosen is centred on

2% growth and 90% of its values occur in the interval 1.5%-2.5%. EC (2008), for example, uses a

+1.8% annual average growth rate in its baseline scenario. For the period 2020-2050, economic

activity is assumed to continue growing at a slower rate and is modelled with a normal distribution

centred on 1.5% (standard deviation = 0.3%).

As a result, real GDP increases on average by 2% per year until 2050. But, due to the uncertainty

about the growth rate of GDP per capita (represented by normal distribution), the confidence

interval around the mean value of GDP widens at each time step.

Domestic demand and exports are determined in comparison to GDP. In 2000, domestic demand

is as high as GDP4 while exports represent 36% of that value. The development trends given by

Eurostat until 2009 are prolonged until 2020 (-0.1% and +1.3% per year for the ratios domestic

demand and export to GDP, respectively). After that date, domestic demand and exports are

assumed to remain in the same proportions in comparison with GDP (98% and 51%, respectively).

3.1.2 Mineral materials module

Given the level of domestic demand5 and exports (see section on economy module), the domestic

material input of minerals (DMI minerals) of the European economy is driven by the respective

shares of services and goods in domestic demand and exports and their respective material

intensities. The corresponding input data are estimated from NAMEA6-based input-output analyses

4 At the EU-25 level, foreign trade is almost balanced (the difference between exports and imports represents

only 1% of GDP) which explains why domestic demand is as high as GDP.

26

conducted by the EEA ETC/RWM7 for 8 EU countries8 with monetary and physical IO tables from

1995 (2 countries) and 2000 (6 countries).

Averaging over the complete set of input-output results, the share of services9 in domestic demand

and export is estimated at 64% and 19%, respectively. These input values are assumed constant

over the whole timeframe in the baseline scenario.

Uncertainties arise from the sometimes questionable accuracy of the input-output analyses and

from scaling up the results for application at the EU-25 level. For example, MFA results for Europe

(Eurostat 2002, Moll et al. 2005, Schütz 2007) have been used to re-calibrate material intensities

which come out too high from the input-output analyses (because indirect foreign used extraction

associated with imports is included in the material inputs activated by final use of a product,

whereas DMI contains only imports and domestic used extraction). The proportionality relation

between material intensities of services vs. that of goods is preserved in the process. The point

values found for different material intensities in 2000 are presented in Table 2. Furthermore, it is

usually considered that DMI bears with itself an error of ±5% (we consider it as the 90% confidence

interval). To account for it, material intensities are modelled as normal distributions parameterised

with means as shown in Table 2 and standard deviations of 3% of the mean.

Table 2: Mean material intensity values in EU-25 in 2000

DMI activated

by final use

t / Mill. euros

Metal ores Industrial

minerals

Construction

minerals Total minerals

Goods 52 47 440 539

Services 5 7 69 81

The material intensity of goods is expected to decrease by 1.7% per year in the baseline scenario,

i.e. without the implementation of any particular policy instrument. It reflects the observed inherent

5 Includes the items “Final consumption expenditure by households”, “Final consumption expenditure by non-

profit organisations serving households (NPISH)”, “Final consumption expenditure by government”, “Gross

fixed capital formation” and “Changes in inventories and valuables”.

6 NAMEA: National Accounting Matrix including Environmental Accounts.

7 European Topic Centre on Resource and Waste Management (ETC/RWM, Topic Centre of the European

Environment Agency): Implementation Plan 2006 - task 7.1.2.1 "NAMEA-based Input-Output Analyses".

8 Germany, Denmark, Spain, Hungary, Italy, The Netherlands, Sweden and the United Kingdom.

9 The term ‘services’ includes the aggregated product groups “Wholesale and retail trade, repair of motor

vehicles, motorcycles and personal and household goods; hotels and restaurants; transport, storage and

communication”, “Financial intermediation; real estate, renting and business activities” and “Public

administration and defence, compulsory social security; education; health and social work; other community,

social and personal service activities; private households with employed persons; extra-territorial

organizations and bodies”. The term ‘goods’ corresponds to the aggregated product groups “Agriculture,

hunting, forestry and fishing”, “Total industry (excluding construction)” and “Construction”.



27

increase of material productivity. Further improvement is not expected to occur in the baseline.

Material intensities of services are assumed constant.

The self-sufficiency of EU regarding minerals (i.e. the imported share of DMI minerals) is taken

from EU-15 MFA studies (Schütz 2007). According to the same source, the share of imports in DMI

metal ores and industrials minerals increased on average by 1.5% and 1.8 % per year during the

decade 1990-2000 (from 62% and 24% up to 72% and 29%, respectively). This trend is prolonged

until 2020 and the share of imports is assumed to remain constant afterwards. Imports of

construction minerals are neglected in view of the massive domestic extraction (mainly sand,

gravels, stones etc).

Moving from direct material input (DMI) to total material requirement (TMR), the indirect flows (IF)

associated with imports and the unused domestic extraction associated with mining in Europe (or

hidden flows, HF) need to be calculated. Hidden flows and indirect flows coefficients come from

EU-15 MFA studies (see Table 3). For the baseline, the trends observed in the decade before 2000

are prolonged until 2020 and flattened afterwards. Depending on the type of mineral materials,

domestic hidden flow and foreign indirect flow intensities are expected to further decrease, and

respectively increase, until 2020 at annual average rates presented in Table 3. TMR is usually said

to come with an error of ±15% (we consider it as the 90% confidence interval). To account for it

(including the error already coming with DMI), HF and IF coefficients are modelled as normal

distributions parameterised with means as shown in Table 3 and standard deviations of 9% of the

mean.

Table 3: Mean values for hidden and indirect flow coefficients for EU-25 in 2000 and baseline assumptions for the development until 2020 (mineral materials)

t / t Metal ores Industrial

minerals

Construction

minerals Total minerals

Hidden flows of

domestic extraction 1.07 0.43 0.22 0.56

Annual change until

2000-2020 -2.9% -3.1% -0.3% -0.9%

Indirect flows of

imports 13 4.01 : 8.09

Annual change until

2000-2020 +1.9% +3.8% : +1.1%

3.1.3 Fossil fuels module

The use of fossil fuels in the EU has in common with the use of mineral materials modelled in the

previous module the issues of extraction of non renewable resources, and the associated indirect

flows and unused extraction. The problem of greenhouse gases emissions is, however, specific to

this category of materials, mainly burnt for energy generation.

28

Table 4 presents the initial conditions (year 2000) and the assumptions for the dynamics of the

energy intensity and energy mix parameters in the baseline scenario. The data come from EC

(2008) until 2030. For the baseline, it is assumed that the parameters remain constant afterwards.

The basic reasoning behind is that any assumption according to which past rates of change, which

had been observed over about a decade, will continue for a period longer than 20 years in the

future would be highly uncertain. Increasing uncertainty could have been reflected by widening the

percentile corridor; this might have lead to unacceptable vagueness of results. One may also argue

that growth or decrease rates of input parameters will level off with time in a baseline scenario (i.e.

without specific strategies implemented to sustain and foster growth or decrease of key driving

factors). Following this rationale, rates of changes are either reduced past 2030 (e.g. in the

economy module) or parameters are simply kept constant. The resulting non-linearity of the

baseline forecasts provide also evidence of the impact of ongoing rates of change. The fossil fuel

mix is implemented partly exogenously to calculate the direct material input of fossil fuels (DMI

fossil fuels) of EU-25, starting from primary energy use. Conversion factors of primary energy into

mass of fuel are taken from EU tables reported to UNFCCC.

The degree of self-sufficiency of Europe regarding fossil fuels, hidden flow (HF) and indirect flow

(IF) coefficients for the domestic and foreign parts of DMI fossil fuels are adapted from the material

flow data base of the WI. The energy mix is taken into account exogenously in the overall

spreadsheet of input values (see section 2.7) and aggregated values are then read into the

Bayesian network model. This way, the complexity and number of ramifications of the network can

be kept at a manageable level. Table 5 shows HF and IF factors, and their development over time

assumed in the baseline. Trends observed over the period 1990-2000 for EU-15 (WI MFA data

base) show little to no change. These flat trends are prolonged until 2020 and parameters are kept

constant afterwards. Only hidden flow coefficients associated with domestic extraction of solid fuels

and natural gas show a slight increase. With the deposits reaching near depletion levels, unused

extraction increases for the same output of usable material. The availability and quality of data is

quite disparate, hence HF and IF coefficients are modelled as normal distributions parameterised

with means as shown in Table 5 and standard deviations of 5% of the mean.

Table 4: Baseline assumption for gross inland energy consumption (GIC) (EC 2008)

Initial conditions (year 2000) 2000-10 2010-20 2020-30 2030-50

toe GIC / Mill. Euro GDP Annual average % change

Energy intensity 170.4 -1.3 -1.7 -1.6 0

% of gross inland consumption Annual average % point change

Solid fuels 14.8 -0.1 +0.07 -0.03 0

Oil 40.4 -0.23 -0.08 -0.02 0

Natural gas 23.3 +0.2 +0.08 -0.01 0

Nuclear 15.3 -0.1 -0.25 -0.15 0

Renewable energy forms 6.0 +0.22 +0.19 +0.19 0



29

Table 5: Mean values for hidden and indirect flow coefficients for EU-25 in 2000 and

baseline assumptions for the development until 2020 (fossil fuels and nuclear)

t / t Solid

fuels Oil Natural gas Nuclear Total fuels

Hidden flows of

domestic

extraction

7.15 0.01 0.1 : 4.4

Annual change

until 2000-2020 1.52% 0% 1.94% : 1.6%

Indirect flows of

imports 5.95 0.32 0.28 8500 1.8

Annual change

until 2000-2020 0.02% 0% 0% 0% 0%

3.1.4 Transport module

The transport module consists of a separate simplified transport model featuring road transport for

passengers and freight, and air transport for passengers. At the present stage of development of

the model prototype, the focus is to eventually determine the demand for biofuels in road transport

in order to link to the land use and greenhouse gas emissions modules.

The initial (year 2000) activity in freight (road) and passenger (air and road) transport are taken

from Eurostat (EU-25). The evolution of these parameters over time is modelled through elasticity

coefficients relating to the level of GDP growth (provided by the economy module). The baseline

scenario uses elasticities as shown in Table 6. Data up to 2030 come from the baseline scenario of

EC (2008) and are prolonged until 2050. Fuel efficiencies and improvement over time for each type

of transport are also taken from the baseline scenario of EC (2008).

The branch of the module dealing with road transport is then further expanded to model use for

biofuels. Seventy-five percent of the demand for biofuels is assumed to be covered by biodiesel

(own assumption, based on discussions with experts), the rest being bioethanol. The share of

biofuel actually blended with conventional fuel in 2000 and 2005 is taken from Eurostat. From then

until 2030, the baseline scenario values of EC (2008) are used (7.4% in 2020; 9.5% in 2030). After

2030, the highest objective of the EU Biofuel Directive (10%) is used until 2050.

Table 6: Baseline assumptions for the transport module

Unit Initial conditions (year

2005)

Evolution over time (2000-2050)

EU freight transport

(road)

Mt-km 1 793 390

GDP elasticity: 1.07 (2005-2010); 0.86

(2010-2015); 0.78 (2015-2020); 0.62

after 2030

EU passenger

transport (road)

MP-km 5 247 640 GDP elasticity: 0.58 (2005-2010); 0.54

(2010-2015); 0.53 (2015-2020); 0.51

30

after 2030

EU passenger

transport (air)

MP-km 507 071 GDP elasticity: 1.53 (2005-2010); 1.54

(2010-2015); 1.48 (2015-2020); 1.3

after 2030

EU fuel intensity

(freight road)

toe /

Mt-km

65.3 Annual energy intensity change:

-0.41% (2005-2030); constant after

EU fuel intensity

(passenger road)

toe /

MP-km


-0.8% (2005-2030); constant after

EU fuel intensity

(passenger air)

toe /

MP-km


-0.84% (2005-2030); constant after

3.1.5 Land use module

The current version of the land use module focuses on non food biomass, more precisely on

biofuel use in Europe and the consequences on land use in other regions of the world and on

greenhouse gas emissions. Therefore, the module receives input from the transport module

regarding demand for biofuels, and provides the greenhouse emission module with estimates of

land use change outside Europe.

Given the level of biodiesel and bioethanol consumption, and the degree of self-sufficiency of

Europe regarding these two types of biofuels (Schütz 2007), the module calculates the amount of

land needed outside EU for its supply (Table 7). For this purpose, production yields for biodiesel

from soybean and palmoil, and bioethanol from sugar cane in different exporting regions of the

world are considered exogenously. The land use module is fed with aggregated yields (one for

biodiesel, one for bioethanol; to keep the complexity of the network down) modelled as normal

probability distributions accounting for the variability of yields between crops and producing

regions. Assumptions are presented in Table 8.

The total amount of arable land needed outside European borders to produce biofuel for

consumption in Europe is compared between every two time slice. The increase in land use is then

assumed to happen at the expense of natural land which is converted into crop land. This result is

then an input for the greenhouse gas emission module.

Regarding further development of the module with food biomass production and consumption,

most data sources for baseline assumptions have been chosen. FAO statistics and FAPRI (2007)

shall provide data regarding the development of overall agricultural land use, yields, consumption

and production of biomass, and biomass trade. EEA (2006a, 2007a) has modelled, within a precise

framework of assumptions, the area of agricultural land available for bioenergy production. These

data are already considered exogenously in the spreadsheet of input data in order to ensure

consistency of the assumptions across the modules.



31

Table 7: Baseline assumptions regarding biofuel consumption for the land use module

Unit Initial conditions (year

2005)

Evolution over time (2000-2050)

EU share biodiesel %

(energy)

0.19 5.6% in 2020; 7.1% in 2030; 7.5%

afterwards

EU share

bioethanol

%

(energy)

0.06 1.9% in 2020; 2.4% in 2030; 2.5%

afterwards

EU self-sufficiency

in biodiesel

% 60 constant

EU self-sufficiency

in bioethanol

% 50 constant

Table 8: Baseline assumptions regarding biofuel yields for the land use module (outside EU)

Crops considered Yields (toe / ha)

Biodiesel produced

outside EU

Soybean and palm oil Normal distributions:

! = 1.9; ! = 0.4 in 2005

! = 2.8; ! = 0.6 in 2030

Bioethanol produced

outside EU

Sugar cane Normal distributions:

! = 2.5; ! = 0.3 in 2005

! = 3.7; ! = 0.4 in 2030

3.1.6 Greenhouse gases module

In the actual version of the model prototype, the output of the greenhouse gases module consists

of three main variables: greenhouse gas emissions from fossil fuels, greenhouse gas emission

savings from using biofuels instead of fossil fuels in road transport (excluding emissions from land

use changes due to biofuel production), and greenhouse gas emissions resulting from land

conversion for the production of biofuels. The GHG module is linked to the fossil fuel, transport and

land use modules that deliver the necessary inputs.

Regarding emissions from fossil fuels, the baseline assumptions presented in section 3.1.3 for the

energy mix, and especially the fossil fuel mix, translate as follows in terms of carbon intensity (EC

2008): in 2000 the economy of the EU emitted 2.23 tonnes of CO2 per tonne of oil equivalent (toe)

of gross inland energy consumption (all inclusive, i.e. with renewable sources); between 2000 and

2010 the carbon intensity is assumed to decrease on average by 0.3% per year; between 2020

and 2030 the carbon intensity is expected to decrease on average by 0.2% per year; and until

2050 the status quo is assumed as baseline.

With regard to emissions from the use of biofuels, Menichetti and Otto (2008) summarize activities

and findings of a comprehensive review of the most recent Life Cycle Assessment (LCA) studies

and energy and environmental balances of biofuels publicly available. For the model, the GHG

32

emissions saved life-cycle wide through the use of biofuels in the transport sector need to be

assessed for biodiesel and bioethanol produced in EU and outside (i.e. four estimates of GHG

savings needed). The geographical scope of the studies reviewed by Menichetti and Otto (2008)

usually does not match that of the EU-25. The overall GHG saving ranges given by the different

LCAs considered together are very wide for each of the four types of biofuels. To reflect these

results, normal distributions are built from the data in Menichetti and Otto (2008), assuming that the

ranges given in their review represent 99.8% confidence intervals (i.e. 3.09 standard deviations) for

the GHG emissions improvement compared to conventional fuel. The baseline assumptions finally

retained are presented in Table 9 and are assumed constant for the whole timeframe.

However, the emission savings calculated using LCAs data from the literature do not tell the whole

story because they overlook emissions induced by the conversion of natural land into arable land in

order to grow biofuel crops. This land use change is modelled in the previous module for biofuel

production outside Europe dedicated to the supply of Europe. The methodological framework and

baseline data uses Fargione et al.’s (2008) study. Table 10 shows the ‘carbon debt’ caused by the

conversion of 1 ha of natural land. Large variations of the carbon debt can be observed depending

on the region of production, and hence the type of land converted. These variations are captured in

the baseline assumptions shown in Table 10 in the form of normal probability distrbutions.

Table 9: GHG emissions improvements due to the use of biofuels compared to conventional

fuel (without land use change)

Crops considered

GHG emissions improvement

compared to conventional fuel

Biodiesel produced in

EU

Rapeseed Range: 20% - 80%

Normal distribution: ! = 50; ! = 9.7

Biodiesel produced

outside EU

Soybean and palm oil Range: -17% - 110%

Normal distribution: ! = 46.5; ! = 20.5

Bioethanol produced

in EU

Wheat Range: 2% - 90%


Bioethanol produced

outside EU

Sugar cane Range: 70% - 100%


Table 10: Carbon debt

Crops considered Carbon debt (t CO2 / ha)

Land converted for

biodiesel production

Soybean and palm oil Range: 33.2 - 1125.8 t CO2 / ha

Normal distribution: ! = 580; ! = 177

Land converted for

bioethanol production

Sugar cane Range: 57 - 165 t CO2 / ha

Normal distribution: ! = 111; ! = 17



33

3.2 Water use submodel

The baseline assumed certain specific values for driving variables, updated each fifth year. These

key variables were given a certain value, assuming either a stable value for the period 2000-2050,

linear increase of exponential increase for the period. The baseline scenario was run with the

following set of input values below.

Cooling water efficiency (LA1) – The efficiency was assumed to remain the same throughout the

period. The EPRI report (2002) states that “it is unclear whether total U.S. freshwater consumption

by the power generation sector will increase or decrease over the next twenty years”. It is assumed

that the same will be valid for the European energy sector.

Industrial water efficiency (LB1) – The efficiency is assumed to increase linearly during the time

period (Cosgrove & Rijsberman, 2000). The efficiency is increasing from a level of 59% to 68%

during the time period.

Household water efficiency (LC1) – The household water efficiency is assumed to increase in a

linear fashion during the time period to 2050, from a level of 69% to 78%, as a result of the

introduction of water pricing and improved technology (Hoekstra & Chapagain, 2007).

Forested area (LF) – The forested area of Europe is expected to continue to increase, with a net

formation rate of new forested area of 0.5%/yr (EEA 2005; Eurostat 2007b).

Evapotranspiration (LG) – The rate of evapotranspiration is mainly determined by climatic factors,

and as the climate change is highly likely to continue, with higher temperatures over vast areas of

Europe (IPCC 2007), this will also lead to increased evapotranspiration. The evapotranspiration is

calculated to increase from 52% to 54% during the time period.

Irrigated area (LJ) – The climate change will also lead to an increase in the amount of irrigated

areas, especially in the southern parts of Europe. It is expected here that the amount will increase,

from a level of 25 million ha to 43 million ha in 2050 (EEA, 2007b).

Desalinized water (LN) – The climate change will also drive the amount of desalinisation of sea

water made. The current level of 1.2 billion m3 per year is expected to increase exponentially to a

level of 80 billion m3 per year (Aquastat 2007).

Recycled wastewater (LR) – Recycled wastewater is assumed to increase exponentially during the

period, from a very low level of 0.025 billion m3 per year to 6.4 billion m3 per year (EEA 2007b).

Precipitation (LT) – Most climate models project increasing precipitation rates for central and

northern Europe and decreasing precipitation rates for southern Europe and more intense rainfall

events (EEA 2007b). It is there assumed a modest increase of 5% during the 2000-2050 period.

Water quality (LWQ) – Water quality is assumed to remain the same during the period, affecting

around 5% of the total amount of available water, by making it unusable (EEA 2007b).

34

Table 11: Baseline assumptions for the water submodel

Node Label Unit 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

LA1 Cooling

water

efficiency

%

10

10

10

10

10

10

10

10

10

10

LB1 Industrial

water

efficiency

%

59 60 61 62 63 64 65 66 67 68

LC1 Household

water

efficiency

%

69 70 71 72 73 74 75 76 77 78

LF Forested

Area

106 ha

158 160 165 170 175 180 185 190 195 200

LG Evapotrans

piration

% 52 52 53 53 54 54 55 55 56 56

LJ Irrigated

area

106 ha

25 27 29 31 33 35 37 39 41 43

LN Desalinized

water

109 m

3

/year 0,31 0,40 0,60 1,20 2,40 4,80 10,0 20,0 40,0 80,0

LR Recycled

wastewater

109 m

3

/year 0,025 0,025 0,05 0,10 0,20 0,40 0,80 1,60 3,20 6,40

LT Precipitation 109 m

3

/year 1 355 1 360 1 365 1 370 1 375 1 380 1 385 1 390 1 395 1 400

LWQ Water

quality

% 5 5 5 5 5 5 5 5 5 5

3.3 Biodiversity, soils and landscape submodel

The following sections specify the baseline assumptions for the sub-model biodiversity, landscape

and soils. Recent and projected trends for the input parameters are described together with the

data sources from which the trend information was either derived or based. Table 11 at the end of

Section 3.2 provides an overview summary of the modelled trends for the sub-model input nodes.

3.3.1 Biodiversity and landscape module

The agri-environment support node represents the proportion of Common Agricultural Policy Pillar

II (Rural Development) spending on agri-environment programmes. The baseline data (2000) is

derived from mid-term Rural Development reports by Member States/regions, giving an overview

on agri-environmental measures applied in the 2000-2006 Rural Development programming period

(European Commission, 2003; European Commission Directorate General for Agriculture and

Rural Development, 2005). EU Member States report a wide range of proportionate spending on

agri-environment measures and a mean value was used to establish the baseline. In the

subsequent rural development period, 2007-2013, expenditure on agri-environment measures has

been reported at 22% of the EAFRD budget (European Commission, 2007). As a result the model

includes a growth in the expenditure on agri-environment programmes at the start of the period

2000-2050. Due to a lack of data on spending beyond 2013 it is assumed that the probability that

spending on AEP will be greater or less than 50% of the total spent on Pillar II of the CAP will



35

follow a similar trend to that reported previously, i.e. it will show a small increase over each five-

year time step (Table 12).

Table 12: Probability of percentage spend on AEP

Year 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Probability % % % % % % % % % % %

AEP < 50% of Pillar II 85 85 78 78 69 69 58 58 47 47 36

AEP > 50% of Pillar II 15 15 22 22 31 31 42 42 53 53 64

Alliance Environnement, 2007 reports that non-adherence to cross compliance is approximately

12% (prior to 2005 there was no requirement for cross compliance so the model is set to 100%

non-compliance). Due to the short time cross compliance regulations have been in operation data

is fairly limited and predictions of future trends are, therefore, not available. Despite the lack of

predictive data to inform the model it has been assumed that the level of non-compliance will

decrease over the modelling period and a one-percent reduction in non-compliance has been

assumed for each model run (Table 13).

Table 13: Probability of non-compliance with cross compliance regulations

Year 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Probability % % % % % % % % % % %

>10% non-

compliance 100 90 89 88 87 86 85 84 83 82 81

<10% non-

compliance 0 10 11 12 13 14 15 16 17 18 19

The quality of forest management is difficult to assess; in the model forest management is defined

as either ‘green’ or ‘brown’. One indicator of forest management quality that can be used to

determine ‘green’ or ‘brown’ status is the area of forest that has been certified by the two main

certifying bodies operating in Europe (the Forest Stewardship Council and the Pan European

Forest Certification Council). Thus in the model ‘green management’ is taken to be the proportion

of forest area under certification. The ‘UNECE FAO Forest Products Annual Market Review (2006)’

reports that the area of certified forest increased by 12% between 2005 and 2006. However, the

figures for Europe show that over the same period the growth in certified forest area was much

smaller from 78.5 million hectares in 2005 to 78.9 million hectares in 2006, representing an annual

increase of about 0.5% (Table 14). An increase of 0.5% per annum is, therefore, assumed in this

model from 2005 onwards.

36

Table 14: Certified forest area by region, 2005-2006

Region Total forest

area

(million ha)

Total certified forest

area 2005 2006

(million ha)

Area certified

2005 2006

(% of total forest)

North America 470.6 140.2 157.7 29.8 33.5

EU/EFTA *155.5 78.5 78.9 50.5 50.7

EECCA 907.4 8.8 13.0 1.0 1.4

Oceania 197.6 3.4 6.4 1.7 3.3

Africa 649.9 6.2 2.1 1.0 0.3

Latin America 964.4 2.3 11.1 0.2 1.1

Asia 524.1 0.8 1.1 0.2 0.2

World total 3869.5 240.2 270.3 6.2 7.0

* Total forest area in this table refers to forest area only. Nodes LF, LFr have larger values as they include

areas classified as both forest and other wooded land (see Appendix 1).

(Source:UNECE/FAO Forest Products Annual Market Review, 2005-2006)

The conservation status of habitats data is derived from Article 17 reporting requirements under the

Habitats Directive (European Commission, 1992) for the period 2001-2006 for the EU2510. A major

component of the report is an assessment of the conservation of all habitats listed on Annexes I

and II of the Directive. The assessment is based around the definition of ‘favourable conservation

status’ set out in the directive, and combines assessments of range, area, structure and functions

and future prospects. Each of the parameters is reported as one of four classes, favourable,

unfavourable-inadequate, unfavourable and unknown. When the results from all four assessments

(range, area and so on) are unequal the results are weighted according to the following rules:

! If more than 25% is unfavourable bad then the result is unfavourable bad

! If more than 75% is favourable then the result is favourable

! If more than 25% is unknown then the result is unknown

! All other combinations the result is unfavourable-inadequate.

For the purpose of this model assessments returning the parameter value unknown have been

equally distributed between the other three categories. Combining the results of all four categories

makes an overall assessment; the habitat status nodes for 2000-2005 are populated with this data,

which was used to build the conditional probability tables. As noted above the assessment includes

a prediction of future prospects for conservation status for the subsequent reporting period. The

future prospects data has been used to establish the trends in future conservation status from 2010

(Figure 8). From 2015 onwards it is assumed that the trends in habitat status will continue as

10 http://biodiversity.eionet.europa.eu/article17/habitatsprogresswebsite



37

suggested by the predictions made under Article 17 reporting. Taking the example of grassland this

would mean that the unfavourable – bad category will decrease, the unfavourable – inadequate

category will increase and finally that the favourable category would grow over the same time

frame. From 2005 to 2050, grassland habitat classified as bad would reduce from 46% to 19% of

total grassland area, that categorised as inadequate from 27 to 50% and grassland with favourable

conservation status would increase from 19% to 31%.

38


(input nodes)

0%

20%

40%

60%

80%

100%

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Forest

Unfavourable – bad Unfavourable – inadequate Favourable

0%

20%

40%

60%

80%

100%

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Sclerophilus scrub




39


(continued)

0%

20%

40%

60%

80%

100%

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Bogs and mires


0%

20%

40%

60%

80%

100%

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Heath and Scrub

Unfavourable - bad Unfavourable - inadequate Favourable

0%

20%

40%

60%

80%

100%

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Aquatic


40

Data on the conservation status of mammal and plant species is also derived from Article 17

reporting requirements under the Habitats Directive (European Commission, 1992) for the period

2001-2006 for the EU2511. The assessment is based around the definition of ‘favourable

conservation status’ set out in the directive, and combines assessments of range, population,

suitable habitat and future prospects. Each of the parameters is reported as one of four classes,

favourable, unfavourable-inadequate, unfavourable and unknown. For the purpose of this model

assessments returning the parameter value unknown have been equally distributed between the

other three categories (Table 15). As for habitat data when the results from all four assessments

(range, population and so on) are unequal the results are weighted according to the rules detailed

above. The future prospects data has been used to establish the trends in conservation status from

2010. From 2015 onwards it is assumed that the trends in habitat status will continue as suggested

by the predictions made under Article 17 reporting.

Table 15: Conservation status for mammal and plant species 2000-2005 and projections for 2010

Species Group Conservation Status 2000 2005 2010

Unfavourable – bad

35.0% 35.0% 29%

Unfavourable – inadequate

36.0% 36.0% 33% Mammal species status

Favourable

29.0% 29.0% 38%

Unfavourable – bad

28.0% 28.0% 20%

Unfavourable – inadequate

43.0% 43.0% 42% Plant species status

Favourable

29.0% 29.0% 38%

The following figures illustrate the probability that plant and mammal species will be assessed as

unfavourable – bad, unfavourable – inadequate and favourable for the period 2000-2050. These

figures suggest that that the conservation status of these two species groups will improve and that

by 2020 for plant species and 2025 for mammal species conservation status is more likely to be

favourable than either of the two unfavourable categories.

11 http://biodiversity.eionet.europa.eu/article17/habitatsprogresswebsite



41

Figure 10: Expected changes in species status from 2000-2050 under baseline projections

(input node)

0%

20%

40%

60%

80%

100%

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Mammals


0%

20%

40%

60%

80%

100%

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Plants


The BirdLife (2004b) found that, 216 (48%) species out of 448 species had unfavourable

conservation status at the EU25 level. This showed that there is a higher proportion of species with

unfavourable status in the EU25 than at the Pan-European level (43% unfavourable). To assess

the conservation status of all wild bird species occurring naturally and regularly in the European

Union Birdlife collected the following data from each country:

! Breeding population size (in or around the year 2000).

! Breeding population trend (over the period 1990–2000).

Wherever possible, national coordinators supplied population trend data as actual percentage

change figures over the period 1990–2000. Where this was not possible data was supplied on

trend direction and magnitude using a fixed set of categories and codes. Trend categories ranged

from -5 to 5, with the sign indicating the direction of the change and with stable populations being

represented by a value of zero. The conservation status of birds in the EU25 was assessed using

the three-step process illustrated in Figure 11.

42

Figure 11: Methodology for determination of EU25 conservation status for bird species

Looking at the situation in 1990, the percentage of species with unfavourable conservation status

at (today) EU25 level was slightly higher (51%) than in 2000. This suggests the overall situation of

birds has slightly improved in the EU and in the new joining countries. In line with this trend a 3%

decrease in the probability of unfavourable conservation status was taken forward for each five-

year time step (Table 16).

Table 16: Projected probability of favourable and unfavourable conservation status for EU25

bird species (input node)

Year 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Probability % % % % % % % % % % %

Unfavourable 48.0 46.6 45.0 43.5 42.1 40.6 39.2 37.8 36.4 35.0 33.6

Favourable 52.0 53.6 55.0 56.5 57.9 59.4 60.8 62.2 63.6 65.0 66.4



43

Figure 12: Current trends in EU mammal and bird populations

Birds population trends

Decreasing

23%

Stable

51%

Increasing

9%

Unknown

17%

Mammal population trends

Decreasing

27%

Stable

32%

Increasing

8%

Unknow n

33%

The current status of mammal species, according to the 2007 publication ‘The Status and

Distribution of European Mammals’ (Temple and Terry, 2007), is that 27% of mammal species are

declining, 32% are stable, 8% are increasing and the final 33% have an unknown population trend

(Figure 12). These figures are used as the starting point for the model. ‘Global Biodiversity Outlook

2’ (Secretariat of the Convention on Biological Diversity, 2006), reports that vertebrate species

have declined by approximately 40% over the period 1970-2000. In line with this data an annual

increase in the probability of species declining of 1.3% per annum has been used to suggest

species trends for 2005-2050.

44

Figure 13: Projected probability of declining, stable, increasing or unknown trend in

mammal and bird populations (input nodes)

Mammals

0%

20%

40%

60%

80%

100%

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Declining Stable Increasing Unknown

Birds

0%

20%

40%

60%

80%

100%

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Declining Stable Increasing Unknown

Birdlife 2004 report that in 2000 51% of bird species populations were stable, 23% were

decreasing and 9% were increasing (Figure 13). In comparison with the figures for the 1990

assessment species were more likely to have moved from the decreasing to stable category than

vice versa. For that reason the projected probability of stable population increases from 51% to

65% over the period 2000-2050 (Figure 13).

Invasive alien species are non-native species that have been introduced by human action (either

intentionally or unintentionally) outside their natural habitats where they have established and bred

causing damage to biological diversity. Invasive alien species present a significant threat to

European biodiversity, acting as vectors for disease, altering the structure and functioning of the

ecosystem and competing for food and resources with native species. Globally invasive alien

species are considered as a second major threat to biodiversity and at the EU level both the EU 6th

Environmental Action Programme and the 2006 communication from the Commission of Halting

the Loss of Biodiversity by 2010 and beyond have identified invasive alien species as a priority

area for action.



45

Not all invasive species are equally harmful to native biodiversity. The 163 species identified as the

‘worst invasive species’ have been recognised as having the most impact on native European

biodiversity. Data from the EEA shows that every year from 1950 at least one species that will

significantly affect European biodiversity has been established. This trend shows a continual rate of

growth of approximately one species per year and this growth rate will be taken forward in this

model.

Figure 14: Projected probability that alien species are decreasing, stable or increasing over

the time period 2000-2050 (input node)

0%

20%

40%

60%

80%

100%

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Alien species

Decreasing Stable Increasing

The data on the percentage of protected area in the EU was taken from the World Database on

Protected Areas12 and has been used to establish the baseline level at 12%. Given the prediction

by the Global Biodiversity Outlook (Secretariat of the Convention on Biological Diversity, 2006) that

the coverage of protected areas is expected to increase the model includes a small rise in this area

over the period 2010 to 2050 (1% per five-year time step), so that the protected area reaches over

13% by 2050.

Table 17: Assumed increase in percentage of protected land area in the EU25 (input node)

Year 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Protected area (%) 12 12 12.12 12.24 12.36 12.49 12.61 12.74 12.87 12.99 13.12

3.3.2 Soils module

Research and development data was derived from Eurostat statistics on gross domestic

expenditure on R & D as a percentage of GDP (Eurostat, 2007); the 2000 baseline was established

12 http://sea.unep-wcmc.org/wdbpa/

46

as 1.9%. Despite the target set by the Lisbon summit for research expenditure to reach 3% of GDP

by 2010 there has been little change in the proportionate spending on R & D over the last decade

(Figure 10). In line with these findings the expenditure on GDP was assumed at a constant 1.9%.

Table 18: Projected share of gross domestic expenditure in R & D (% of GDP) (input node)

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

% of

GDP

1

.

9

1.8 1.8 1.8 1.9 1.9 1.9 1.9 1.9 1.9

Source: Eurostat Yearbook 2006-2007

Amongst the aims of the proposed Soil Framework Directive are the establishment of a common

framework to “protect soil on the basis of the principles of preservation of soil functions, prevention

of soil degradation, mitigation of its effects, restoration of degraded soils and integration in other

sectoral policies” (European Commission 2006, p.5). However, the Soil Framework Directive has

not yet been implemented. Despite this it was felt that the future introduction of such legislation

could have important implications for a number of soil-based parameters, which made this an

interesting node to include within the sub model. The baseline state for the Soil Framework

Directive has been set to ‘none’. Due to uncertainty about when the Directive will be implemented,

and in what format, the node is set to this state throughout the baseline scenario.

Estimates of greenhouse gas emissions at the global level serve as proxy for representing how

pressing the issue of climate is and will become. Greenhouse gas emission data is obtained from

IPCC data, which reports emissions of 44.7 Gt CO2-equivalent in the year 2000 (IPCC, 2007). Due

to the uncertainty associated with this estimate this is modelled as a normal distribution with a

standard deviation of about 5 Gt CO2- equivalent. The SRES13 Scenarios project an increase of

baseline global gas emissions of between 25 to 90% between 2000 and 2030. Beyond 2030, the

situation is less clear-cut with SRES scenarios predicting both increases and decreases in

greenhouse gas emission. However, data from the CAIT databases14 suggests that on average

emissions of CO2 (Mt CO2 from energy) will increase by an average of 1.5% between 2002 and

2050. Although not accounting for all sectors or all types of emissions this value is used as an

indication of future global greenhouse gas emissions and a 1.5% per annum increase is taken

forward in this model.

13 SRES scenarios are those described in the IPCC Special Report on Emissions Scenarios (2000). The

scenarios are grouped into four families, A1, A2, B1 and B2, which explore a wide range of driving forces

(economic, demographic and technological) and their resulting greenhouse gas emissions.

14 http://cait.wri.org/



47

Figure 15: Projected greenhouse gas emissions for the period 2000-2050 (input node)

0.0

20.0

40.0

60.0

80.0

100.0

120.0

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

GT

CO

2 e

quiv

ale

nt

Greenhouse gas emissions

Nutrient transfer efficiency data is taken from a number of sources (Domburg et al., 2000, Raun

and Johnson, 1999; IGER) and so the baseline value has been established using a normal

distribution with mean values of 28% (standard deviation 14%) for grassland and 52% (standard

deviation 26%) for arable crops. Improvements in nutrient use efficiency might be expected in the

future, however, the level of increase in efficiency that might be expected is difficult to quantify. For

that reason the level of nutrient transfer efficiency is increased by one percent for each five year

time period.

Figure 16: Assumed nutrient transfer efficiency for grassland and arable from 2000-2050

(input node)

Nutrient transfer efficiency

0.0

20.0

40.0

60.0

80.0

100.0

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Grassland Arable

Fertiliser use intensity for the EU is taken from the CAIT database, 2002 data for EU 27 (excluding

Slovakia, Luxembourg and Latvia). Using this resource a mean and standard deviation were

calculated so that fertiliser use for the baseline year could be modelled as a normal distribution,

reflecting the variability of this statistic. The European Fertiliser Association (EFMA) (2007) predicts

an increase in the use of nitrogen in the EU27 of 3.6% between 2007 and 2017. In contrast, over

48

the same period EFMA suggest that the use of phosphorus will decline by 4.4% and potassium by

2.6%. The FAO report, World Agriculture: 2015/2030 suggests that overall fertiliser use will

continue to grow in the developed world until 2030 by about 1% per annum. This latter figure

represents all nutrient sources so was used to represent the growth in fertiliser use intensity until

2030. Beyond 2030, the direction of future fertiliser use intensity is uncertain so between 2030 and

2050 the value in the baseline scenario will be held constant.

Figure 17: Projected change in fertiliser use intensity for the baseline scenario 2000-2050 (input node)

0

100

200

300

400

500

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

kg/ha

Fertiliser use intensity



49

4 Baseline forecasting results for selected intermediary and target

nodes

4.1 Resource use and waste generation submodel

This section presents baseline forecasts for key output nodes of the resource use and waste

generation submodel. As described in chapter 3, the use of probability distributions instead of point

values makes it possible to account for a degree of parameter uncertainty. As a consequence,

forecasts of target indicators are shown as “corridors” that represent the 90% confidence intervals

of the corresponding modelling results, given the input parameter uncertainties.

4.1.1 Economy module

In the economy module, future growth of GDP per capita is modelled using a normal distribution

centred in 2% until 2030 and in 1.5% afterwards (0.3% standard deviation in each case). Figure 18

(upper part) shows the baseline assumptions for population and GDP per capita in the EU-25. The

range defined by the normal distribution for GDP per capita was chosen to reflect differences in

assumptions observed in existing European scenario studies. The lower part of Figure 18 shows

the forecast for European GDP until 2050. The width of the confidence interval increases the

further we look into the future. Given the assumptions for the input parameters, GDP roughly

doubles between 2005 and 2050, though it is expected to increase at a slightly slower rate after

2030. Starting from the point value reported in European statistical data (about 10*106 million

euros), GDP’s mean value reaches about 21*106 million euros in 2050 (90% CI: 18.5 - 25.2 *106

million euros).

4.1.2 Mineral materials module

As a consequence of a growing economy, the 90% confidence interval for TMR minerals increases

from 7.8 – 9.2Gt in 2005 to 12.4 – 18Gt in 2050 (Figure 19). Though the direct material input levels

off over time, due to increasing material productivity and the assumption of a slower GDP growth

after 2030, TMR keeps increasing. This can be explained by the fact that the indirect flows,

especially unused extraction such as mining waste associated with direct material input, keep

increasing as a result of resource depletion. Furthermore, increasingly this waste production occurs

in other parts of the world that export to Europe. This worsening of the problem shifting is illustrated

by the ratio foreign TMR over domestic TMR in Figure 20, which keeps increasing. The forecasted

trend shows that in 2050 between 2.3 and 3.1 (90% CI) times more materials will be extracted, and

for a large part turned into waste, outside the EU than within in order to satisfy European demand

for mineral materials.

50

Figure 18: Baseline modelling results for GDP



51

Figure 19: Baseline modelling results for TMR minerals

Figure 20: Baseline modelling results for the ratio foreign vs. domestic TMR minerals

52

4.1.3 Fossil fuels module

For the period 2005-2030, the assumptions for the baseline improvement in energy efficiency and

the development of renewable energy sources in EU-25 were derived from EC 2008, as described

in section 3.1.3. The development of these parameters compensate the joint effects of a growing

economy and growing indirect flows per unit of extracted fossil fuel. Therefore the mean value of

TMR fossil fuels remains rather constant around 6.5 billion tonnes (Figure 21). The 90% confidence

interval, however, gets wider from 6.3-6.9 to 5.7-7.5 billion tonnes reflecting the uncertainties built

in model parameters such as hidden and indirect flow coefficients in the present module, and

economic growth in the “parent” economy module. Baseline assumption that Europe will

increasingly rely on imported fossil fuels for its energy supply also explains that the ratio foreign vs.

domestic TMR increases until 2030 (Figure 22).

Regarding the 2030-2050 period, input parameters such as energy efficiency and development of

renewable energy sources are conservatively kept constant at the baseline level they are expected

to reach in 2030. At the same time, economy is assumed to keep growing (though at a slower rate)

which explains that mean TMR fossil fuels increases from 6.5 to 8.4 billion tonnes between 2030

and 2050 (Figure 21). It is also assumed that, after 2030, the import structure of Europe regarding

fossil fuels has stabilised and the share of imports remain constant at 78%. Therefore, the ratio

foreign vs. domestic TMR remains constant (Figure 22). It means that, in relative terms, the

situation will worsen at the same pace outside and inside EU. However, the absolute amount of

fossil fuels extracted and waste generated outside the EU keeps increasing and remains higher

than in the EU.



53

Figure 21: Baseline modelling results for TMR fossil fuels

Figure 22: Baseline modelling results for the ratio foreign vs. domestic TMR fossil fuels

54

4.1.4 Transport module

Transport activities in the EU-25 are expected to be tight to economic growth, with elasticities

around 0.5 for passenger road transport, decreasing from 1 to 0.6 for freight road transport, and

remaining higher than 1 at 1.3-1.5 for passenger air transport. Expected activities in million tonne

kilometres and million passenger kilometres are shown in Figure 24. Following growth in road

transport and mandatory biofuel blending with conventional fuel (up to 9.5% in 2030 and constant

afterwards), biofuel use is expected to increase ten fold, from 3.3 to 34 million toe, between 2005

and 2030 (Figure 23).

Figure 23: Baseline modelling results for biofuel consumption



55

Figure 24: Baseline modelling results for transport

56

4.1.5 Greenhouse gases module

In this module, a forecast of greenhouse gas emissions from fossil fuels is generated and shown in

Figure 25. Unsurprisingly, it follows a similar development pattern as TMR fossil fuels shown in

section 4.1.3, as both are calculated from the level of fossil fuel use in EU. As biofuels start

replacing a fraction of the fossil fuels used in road transport, corresponding fossil greenhouse gas

emissions are avoided. Emission savings through biofuels as shown in Figure 26 are less than 1:1

savings when compared to the equivalent fossil energy use because they account for lifecycle-wide

emissions. However, emissions from conversion of land for biofuel production is not included in that

picture.

When including greenhouse gas emissions from land use change (LUC), savings associated with

the use of biofuel look much different, and are even negative. The wide probability distribution

shown in Figure 26 reflects the uncertainty on parameters such as biofuel crop yields or the actual

amount of greenhouse gases emitted per unit area of converted land. Due to the limited time

dynamic capabilities of the Bayesian Network approach, the “carbon debt” (Fargione et al. 2008)

associated with land use change outside EU is accounted for all at once, in the year when land

conversion occurs. It explains the large “negative savings” from biofuel use in EU in 2010 (-1.1Gt)

when demand is growing, supported by political decisions. Towards the end of the modelled time

period, when demand for biofuels has stabilised, and production yields and fuel efficiency have

increased, the median of “GHG savings from biofuels with land use change” is close to the x-axis,

meaning that the land use change induced GHG emissions are about the same order of magnitude

as the savings by biofuels without LUC. The savings are, however, still negative.

Figure 25: Baseline modelling results for GHG emissions from fossil fuels



57

Figure 26: Baseline modelling results for GHG emissions savings from biofuel without land

use change

Figure 27: Baseline modelling results for GHG emissions savings from biofuel including land use change

58

4.2 Water use submodel

The water usage efficiencies within the industrial, power generation and household sectors are

assumed to increase linearly but modestly in a business-as-usual scenario, while industrial

recycling and desalinization are assumed to increase exponentially, starting from a low level. The

overall usage of water is expected to increase during the period, driven by increasing material and

energy use, which is outside the water sub-model boundaries. It is also assumed that the number

of households will continue to increase, even if the EU-25 population level remains stable, which

will lead to increase in overall household water consumption. The water balance (difference

between withdrawals and supply) is not expected to change significantly, as increasing

precipitations (as a consequence of climate change, especially in the north of Europe), recycling

and desalinization (especially in the south of Europe) tend to offset increased consumption.

Regional variations, however, can be expected to be more contrasted. But they are, to date, not

captured by the model prototype.

Figure 28: Baseline modelling results for total water withdrawals



59

Figure 29: Baseline modelling results for water balance

4.3 Biodiversity, soils and landscape submodel

The biodiversity, soils and landscape sub-model consists of discrete and discretized nodes. In

order to reduce the problem of discretizing continuous variables (much information is often lost in

the process and the result can be misleading), for those nodes where the input parameter is

specified as a normal distribution with associated mean and standard deviation (Table 20) initial

model runs were undertaken using the mean value as a numeric input to the model. In addition,

due to lack of data from the water model at the time of the model runs, the value for water

abstraction was held constant (i.e. node LM1 change in water abstraction was set to 100%

probability that abstraction was stable). These considerations need to be borne in mind when

considering the initial model output presented below.

Figure 30 shows the modelled output for the nodes terrestrial, aquatic and overall biodiversity. The

model predicts that aquatic biodiversity status will improve over the period 2000-2050, i.e. there will

be a greater probability of favourable conservation status in 2050 than in 2000. However, it should

be remembered that the water abstraction node was held constant in this model run and allowing

this node to vary would produce more realistic results.

60

Figure 30: Modelled probability of aquatic, terrestrial and overall biodiversity status

0%

20%

40%

60%

80%

100%

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Aquatic biodiversity status


0%

20%

40%

60%

80%

100%

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Terrestrial biodiversity


0%

20%

40%

60%

80%

100%

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Overall biodiversity

Declining Stable Improving



61

In contrast, the model predicts that terrestrial habitat condition will decline over the same period

with the probability of unfavourable habitat status increasing from 25% in 2000 to 48% by 2050.

At the start of the modelling period the model predicts a 48% probability that overall biodiversity will

be declining and an almost equal chance that it will be either stable or increasing. By the end of the

modelling period the model prediction there is a similar probability of declining biodiversity (45%) to

that predicted in 2000, a 14% chance that biodiversity will be stable and a 41% chance that it will

be increasing.

Figure 31 shows the predicted model probability for fragmentation, wider ecological potential and

protected habitat condition, three of the four parent nodes for terrestrial biodiversity. The fourth

parent node for terrestrial biodiversity is area of protected land, which is derived from exogenous

data. In line with the node for terrestrial biodiversity the model predicts that fragmentation will

increase and that the wider ecological potential will be reduced and that overall habitat condition is

most likely to be unfavourable. Looking at the charts in Figure 31 the effect of abrupt changes in

state can be seen. For example, the parents of the fragmentation node (effective mesh size) are

the discrete nodes artificial, forest and agricultural land use change. For the first four model runs

(2000-2015) these nodes are set so that land use is stable (probability 100%) and therefore, the

model predicts that the status of fragmentation will also remain stable. However, by 2020 assumed

changes in land use (set as input) mean that the model now works with a 100% probability that

agricultural area is decreasing, a 100% probability that artificial area is increasing and 100%

probability that forest area remains stable. This change in parent node states can be seen in the

fragmentation chart where the probability of high fragmentation (low MEFF) is now abruptly

increased.

Figure 32 shows the node for soil organic carbon, soil erosion and soil quality. The predicted

probability for these nodes shows little change under modelled conditions as the Soil Framework

Directive, which is a parent node of a number of soil related parameters has been kept constant at

‘none’ (100% probability) for the entire modelling period.

The model predicts that there will be an increase in favourable conservation status over the fifty-

year period 2000-2050. However, the model also suggests that there will be an increase in the

number of species that are experiencing a downward trend in species number.

62


soil quality

0%

20%

40%

60%

80%

100%

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Fragmentation

MEFF low MEFF medium MEFF high

0%

20%

40%

60%

80%

100%

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Ecological potential - wider landscape

Green index low Green index medium Green index high

0%

20%

40%

60%

80%

100%

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Protected habitat condition




63


soil quality

0%

20%

40%

60%

80%

100%

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Soil erosion

Soil Erosion 0-0.5 t/ha/yr Soil Erosion 0.5-1 t/ha/yr Soil Erosion 1-2 t/ha/yr

Soil Erosion 2-5 t/ha/yr Soil Erosion 5-10 t/ha/yr

0%

20%

40%

60%

80%

100%

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Soil carbon

0 to 1% SOC 1 to 2% SOC 2 to 6% SOC

0%

20%

40%

60%

80%

100%

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Soil quality

Low Medium High

64

Figure 33: Predicted model probability of species conservation status and trend in species

number

0%

20%

40%

60%

80%

100%

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Conservation status - species

Unfavourable Favourable

0%

20%

40%

60%

80%

100%

2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Trend in species number

Decreasing Stable Increasing



65

5 Conclusions

The forecasting method using network and simulation-based Bayesian modelling as presented in

this report provided a set of modelling results constituting the baseline scenario of FORESCENE.

The baseline will serve as reference against which alternative scenarios will be elaborated in

workpackage 5. A basic finding from the business-as-usual scenario is that current trends in

material and energy productivity improvements are insufficient to offset the growing demand for

resources that result from continued economic growth. It also shows that a large and increasing

part of the environmental burden associated with material and energy use in Europe actually hits

the environment outside our boundaries.

The modelling approach chosen allows to represent results for quantitative indicators as “corridors”

that reflect the 90% confidence interval. This represents the uncertainties attached to input and

control parameters. As a consequence, the width of the results’ confidence intervals (hence of the

corridor of probability) tend to increase the further we look into the future.

Besides the continuous full probability distributions of quantitative parameters, some more

qualitative indicators can also be represented. In such a case, modelling results take the form of

discrete probability distributions. Nodes are linked probabilistically through the result of expert

judgement and literature data elicitation. Thus the modelling framework allows the inclusion of both

physically measurable and calculated data and more qualitative assessments.

The baseline forecasts provide some first plausible as well as interesting outcomes. However,

more testing is recommended to check the robustness of the various facets of the model prototype.

Endogenous and exogenous linkages between the modules need to be further tested and

improved. The behaviour of the ‘biodiversity and soils’ sub-network, whose long ramifications of

discrete and discretized nodes tend to buffer the effect of influencing parameters situated

upstream, need to be studied closer.

Finally, there are three major options for further development.

Firstly, biodiversity outside the EU has not yet been included in the model explicitly. Nevertheless,

the model shows that the increased land use change for biofuel production leads to enhanced

GHG emissions, and therefore one may conclude that biodiversity in other regions is also

negatively affected. With the modelling of appropriate expert knowledge, this relation, as well as

others, could be modelled using a module made of discrete nodes, similarly to the modelling of

biodiversity within Europe.

Secondly, the modelling of agricultural land use could be extended to food biomass in order to

account for the intra- and extra-EU land use associated with the total consumption of agricultural

goods (including non-food biomass for energetic and material use). This could be used to model

the effectiveness of strategies to enhance efficient use of resources (e.g. through car fleet

consumption caps; reductions of food waste, and longer-term changes in people’s diet). Data exist

but need to be adapted for use in a network model aggregated at the EU-25 geographical scale.

Relationships with the driving forces also need to be worked out and operationalized for Bayesian

modelling. It should be clarified whether this modelling is possible at the level of aggregation

considered or if a detailed, bottom-up network model is required.

Thirdly, the modelling of the external dimension of EU activites could be enhanced by extending

the FORESCENE model with a forestry module. Various member states foster the use of forestry

products for energetic and material purposes; as a consequence, Scandinavian countries have

66

become net importers of forestry products; the impacts on world forests and their biodiversity status

could be assessed in a first stage by expert assessment which may then be refined by more

analytical modelling.



67

6 References

Alliance Environnement (2007) Evaluation of the application of cross compliance as foreseen

under Regulation 1782/2003.

Alston, J.M., Chan-Kang, C., Marra, M.C., Pardey, P.G. and Wyatt, T.J. (2000) A meta-analysis of

rates of return to agricultural R&D: Ex Pede Herculem?, International Food Policy Research

Institute.

Borsuk, M. E., C. A. Stow and K. H. Reckhow (2003) Integrated approach to total maximum daily

load development for Neuse river estuary using Bayesian probability network model (Neu-BERN).

Journal of Water Resources Planning and Management 129(4): 271-282.

Borsuk, M. E. (2004) Predictive assessment of fish health and fish kills in the Neuse river estuary

using elicited expert judgment. Human and Ecological Risk Assessment 10: 415-434.

Borsuk, M. E., C. A. Stow and K. H. Reckhow (2004) A Bayesian network of eutrophication models

for synthesis, prediction, and uncertainty analysis. Ecological Modelling 173: 219-239.

Birdlife International (2004) Birds in Europe: population estimates, trends and conservation status.

BirdLife International . Cambridge, UK.

BirdLife International. (2004b). Birds in the European Union: a status assessment. BirdLife

International, Wageningen, The Netherlands.

Cain, J. (2001) Planning improvements in natural resources management – Guidelines for using

Bayesian networks to support the planning and management of development programmes in the

water sector and beyond. Centre for Ecology & Hydrology: UK.

Climate Analysis Indicators Tool (CAIT V&A) Version 2.0. (Washington, DC: World Resources

Institute, 2008 (http://cait.wri.org/).

Commission of the European Communities (2001) A Sustainable Europe for a Better World: A

European Union Strategy for Sustainable Development. Brussels, 15.5.2001, COM(2001)264 final.

Commission of the European Communities (2005a) Commission’s Communication on its Strategic

Objectives 2005-2009 (COM (2005) 12).

Commission of the European Communities (2005b) The 2005 Review of the EU Sustainable

Development Strategy (COM(2005)37 final).

Consultative Group on International Agricultural Research Science Council (CGIAR) (2003)

Benefit-Cost Meta-analysis of investment in the International Agricultural Research Centres of the

CGIAR, FAO organisation of the United Nations.

Council of the European Union (2006) Review of the EU Sustainable Development Strategy (EU

SDS) " Renewed Strategy. 10907/06.

Domburg, P. Edwards, A.C., Sinclair, A.H. and Chalmers, N.A. (2000) Assessing nitrogen and

phosphorus efficiency at farm and catchment scale using nutrient budgets. Journal of the Science

of Food and Agriculture 80: 1946-1952.

EEA (European Environment Agency) (2005) European environment outlook. European

Environment Agency. Copenhagen: EEA Report No 4/2005.

68

EEA (2006a) How much bioenergy can Europe produce without harming the environment?

Copenhagen: EEA Report No 7/2006.

EEA (2006b) Towards integrated land and ecosystem accounting, EEA Report No 11/2006.

EEA (2007a) Estimating the environmentally compatible bioenergy potential from agriculture.

Copenhagen: EEA Technical Report No 12/2007.

EEA (2007b) Europe's environment - The fourth assessment. Luxembourg: Office for Official

Publications of the European Communities.

EEA Data Concentration of Nitrates mg NO3/litre,

http://dataservice.eea.europa.eu/atlas/viewdata/viewpub.asp?id=2030

European Commission (1992) Council Directive 92/43/EEC of 21 May 1992 on the conservation of

natural habitats and of wild fauna and flora.

European Commission (2003) Fact Sheet: Rural Development in the EU.

European Commission (2006) Proposal for a Directive of the European Parliament and of the

Council establishing a framework for the protection of soil and amending Directive 2004/35/EC.

European Commission (2007) Report from the Commission to the Council and the European

Parliament On implementation of Council Directive 91/676/EEC concerning the protection of waters

against pollution caused by nitrates from agricultural sources for the period 2000-2003.

European Commission Directorate General for Agriculture and Rural Development (2005). Agri-

environment Measures. Overview on General Principles, Types of Measures and Applications.

European Communities (2008) European energy and transport: Trends to 2030 – Update 2007.

Directorate General for Energy and Transport. Luxembourg: Office for Official Publications of the

European Communities.

European Fertilizer Manufacturers Association (2007) Forecast of food, farming and fertilizer use in

the European Union 2007-2017.

European Union Directorate General for Agriculture and Rural Development (2007). Rural

Development in the European Union. Statistical and Economic Information.

Eurostat (2002) Resource use in the European Union 1980-2000: Indicators and analysis.

Luxembourg: Office for Official Publications of the European Communities.

Eurostat (2007) Science, technology and innovation in Europe: Luxembourg: Office for Official

Publications of the European Communities.

Eurostat (2007b). Forestry Statistics 2007 Edition. Luxembourg: Office for Official Publications of

the European Communities.

Fargione,J., J. Hill, D. Tilman, S. Polasky and P. Hawthorne (2008) Land clearing and the biofuel

carbon debt. Science 319(5867): 1235-1238.

FAO http://faostat.fao.org/site/377/default.aspx#ancor

FAO (2003) World agriculture: towards 2015/2030. An FAO perspective, Earthscan.

http://www.fao.org/docrep/005/y4252e/y4252e00.htm

FAPRI (2007) U.S. and world agricultural outlook 2007. Food and Agricultural Policy Research

Institute. Iowa U.S.A.: Iowa State University, University of Missouri-Columbia.



69

Freibauer, A., Rounsevell, M. D. A., Smith, P. and Verhagen, J. (2004). "Carbon sequestration in

the agricultural soils of Europe." Geoderma 122(1): 1-23.

Friedman, N. and M. Goldszmidt (1996) Discretizing continuous attributes while learning Bayesian

networks. In: Proceedings of teh 13th International Conference on Machine Learning (ICML).

Morgan Kaufmann Publishers, San Francisco, CA, pp. 157-165.

Fuglie, K.O. and Heisey, P.W. (2007) Economic returns to public agricultural reserach. Economic

Brief Number 10. United States Department of Agriculture Economic Research Service.

Gelman, A., J. B. Carlin, H. S. Stern, and D. B. Rubin (1995) Bayesian data analysis. In: Texts in

Statistical Science. Chapman & Hall.

IGER, Nutrient Use Efficiency by the Dairy Cow

http://www.iger.bbsrc.ac.uk/Practice/Publications_&_Leaflets/NitrogenDairy.htm

IPCC (2007) Fourth Assessment Report of the Intergovernmental Panel on Climate Change:

Synthesis Report.

Jensen, F. V. (2001) Bayesian networks and decision graphs. Springer-Verlag: New-York.

JRC (PESERA Model data) IN: European Union Directorate-General for Agriculture and Rural

Development (2007). Rural Development in the European Union. Statistical and Economic

Information.

Marcot, B. G., R. S. Holthausen, M. G. Raphael, M. Rowland and M. Wisdom (2001) Using

Bayesian Belief Networks to evaluate fish and wildlife population viability under land management

alternatives from an environmental impact assessment. Forest Ecol. Manage. 153, 29-42.

Menichetti, E. and M. Otto (2008) Existing knowledge and limits of scientific assessment of the

sustainability impacts due to biofuels by LCA methodology. International Panel for Sustainable

Resource Management, UNEP-DTIE (unpublished).

Moll, S., S. Bringezu and H. Schütz (2005) Resource use in European countries: An estimate of

materials and waste streams of the Community, including imports and exports using the instrument

of material flow analysis. Wuppertal Institute for Climate, Environment and Energy. Wuppertal:

Wuppertal Report.

Moser, B. Jaeger, A.G.., Tappeiner, U., Tasser, E. And Eiselt, B. (2007) Modification of the

effective mesh size for measuring landscape fragmentation to solve the boundary problem.

Landscape Ecology 22 (3) 447-459.

Raun, W.R. and Johnson, G.V. (1999) Improving nitrogen use efficiency for cereal production.

Agronomy Journal 91(3): 357-363.

Roseboom, J., Dierderen, P. and Kuyvenhoven, A. (2003) Optimising the allocation of agricultural

R&D funding: is win-win targeting possible? Proceedings of the 25th International Conference of

Agricultural Economists (IAAE) 16-23 August, 2003.

Schütz, H. (2007) Personal communication.

Smith, P., Andrén, O., Karlsson, T., Perälä, P., Regina, K., Rounsevell, M., and Van-Wesemaels,

B. (2005) Carbon sequestration potential in European croplands has been overestimated. Global

Change Biology 11: 2153-2163.

70

Secretariat of the Convention on Biological Diversity (2006) Global Biodiversity Outlook 2.

Montreal.

Temple, H.J. and Terry, A. (Compilers). 2007. The Status and Distribution of European Mammals.

Luxembourg: Office for Official Publications of the European Communities.

UN, European Forest Sector Outlook Study 1960-2000-2020: Main Report (2005) UNECE, FAO.

UNECE FAO (2006) Forest Products Annual Market Review: Geneva Timber and Forest Study

Paper 21. United Nations, New York and Geneva.

Uusitalo, L. 2007. Advantages and challenges of Bayesian networks in environmental modelling.

Ecological Modelling 203: 312-318.

Van-Camp, L., Bujarrabal, B., Gentile, A.R., Jones, R.J.A., Montanarella, L., Olazabal, C., and

Selvaradjou, S-K. (2004) Reports of the technical working groups established under the Thematic

Strategy for Soil Protection. Volume III: Organic Matter. EUR 21319 EN/3, 872 pp. Office for Official

Publications of the European Communities, Luxembourg.

Weber, J.L., Spyropoulou, R., Soukup, T. and Páramo, F. (2008) Net Landscape Ecological

Potential of Europe and change 1990-2000. EEA.

World Database on Protected Areas, http://sea.unep-wcmc.org/wdbpa/

74

7 Appendix

7.1 Conditional Probability Tables (CPTs)

Cain (2001) defines a Bayesian Network as a ‘graphical tool for building decision support systems

to help make decisions under uncertain conditions’. The key phrase to focus on in this definition is

‘uncertain conditions’. As Cain points out, BNs were originally developed to allow the impact of

uncertainty in management to be accounted for. Using the tools decision makers could balance the

desirability of an outcome against the chance that the management option selected might fail. The

representation of a system in terms of a set of relationships that have probabilities associated with

them is at the heart of the Bayesian approach. An example of a simple BN is shown in Figure 34.

Figure 34: A simple BN

Source: after Cain 2001

If we think about the sorts of things that might influence agricultural yield, for example, then these

might include water supply and fertilizer applications. The amount of water applied to the crop

might, in turn, be influenced by such factors as soil type and the level of irrigation. Figure 34 shows

this diagrammatically. The key variables (that is the things that can change, such as yield, fertilizer

applications and soil type etc.) are shown as a set of boxes or nodes. The relationships between

the variables are shown as a set of arrows. These arrows simply set out the linkages between the

variables; they show what influences what. The arrows describe what we think the causal

relationships are within the system; notice that the arrows have a direction to express this idea of

causality.

In Figure 34 each of the nodes are shown as being able to take various states. Thus yield can be

‘good’ or ‘poor’, or soil type can be ‘sandy’ or ‘clay’. In the diagram, the nodes are represented by a

special type of box, called a ‘belief bar’, which we can use to express the probability that the

variable (node) is in a particular state, and how this might influence the other nodes to which it is

linked. At the moment no real probabilities have been assigned, and all the nodes show a 50:50

chance of being in a particular state. What the BN allows us to do is to assign probabilities to the

states for the different nodes to describe how we think the world actually works. The way we do this

is to build Conditional Probability Tables (CPT) for each node.



75

Table 19: CPT for the Crop water Application Node

Table 19 shows the CPT for the node ‘Crop Water Application’ that is part of the network described

in Figure 34. Technically CPT is the contingency table of conditional probabilities15 stored at each

node, containing the probabilities of the node given each configuration of parent values. Clearly

when building a BN we do not have to confine ourselves to variables that have only two states. The

software can handle a number of discrete states or even continuous values.

15 The conditional probability of an event is the probability of the event occurring under certain given

conditions.

76

7.2 Values for model input nodes of the biodiversity, soils and landscape submodel for baseline scenarios 2000-2050

Table 20: Values for model input nodes of the biodiversity, soils and landscape submodel for baseline scenarios 2000-2050

Node Node Name Measurement 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

85% 85% 78% 78% 69% 69% 58% 58% 47% 47% 36% NAES Agri-

environment

support

Probability for

CPT 15% 15% 22% 22% 31% 31% 42% 42% 53% 53% 64%

100% 90% 89% 88% 87% 86% 85% 84% 83% 82% 81% NCC Cross

compliance

Probability for

CPT 0% 10% 11% 12% 13% 14% 15% 16% 17% 18% 19%

% Efficiency

(mean)

28.0 28.3 28.6 28.8 29.1 29.4 29.7 30.0 30.3 30.6 30.9 NGTE Nutrient transfer

efficiency

(grassland)

% Efficiency

(stdev)

14.0 14.1 14.3 14.4 14.6 14.7 14.9 15.0 15.2 15.3 15.5

% Efficiency

(mean)

52.0 52.5 53.0 53.6 54.1 54.7 55.2 55.8 56.3 56.9 57.4 NATE Nutrient transfer

efficiency

(arable) % Efficiency

(stdev)

26.0 26.3 26.5 26.8 27.1 27.3 27.6 27.9 28.2 28.4 28.7

50.8% 49.5% 48.2% 46.9% 45.6% 44.3% 42.9% 41.4% 40.0% 38.5% 36.9% NFM Forest

management

Probability for

CPT 49.2% 50.5% 51.8% 53.1% 54.4% 55.7% 57.1% 58.6% 60.0% 61.5% 63.1%



77

Table 21: Values for model input nodes of the biodiversity, soils and landscape submodel for baseline scenarios 2000-2050 (Continued 1)


Yield t ha-1

yr-1

(mean)

2.46 2.69 2.85 3.00 3.15 3.31 3.48 3.66 3.85 4.04 4.25 WBP9Y

Oilseed yield

Yield t ha-1

yr-1

(stdev)

0.16 0.17 0.18 0.19 0.20 0.21 0.22 0.24 0.25 0.26 0.27

Yield t ha-1

yr-1

(mean)

5.25 5.40 5.62 5.79 5.98 6.16 6.35 6.55 6.76 6.97 7.19 WBP10Y

Wheat yield

Yield t ha-1

yr-1

(stdev)

0.33 0.34 0.35 0.36 0.38 0.39 0.40 0.41 0.42 0.44 0.45

Yield t ha-1

yr-1

(mean)

2.43 2.45 2.55 2.60 2.66 2.71 2.77 2.82 2.88 2.94 3.00 WBP11Y

Grains yield

Yield t ha-1

yr-1

(stdev)

0.23 0.23 0.24 0.25 0.25 0.26 0.26 0.27 0.27 0.28 0.28

Yield t ha-1

yr-1

(mean)

54.24 57.32 60.98 61.60 62.22 62.85 63.49 64.13 64.78 65.43 66.10 WBP12Y

Sugar beet yield

Yield t ha-1

yr-1

(stdev)

3.51 3.71 3.95 3.99 4.03 4.07 4.11 4.15 4.19 4.23 4.28

Yield t ha-1

yr-1

(mean)

4.17 4.26 4.37 4.45 4.54 4.63 4.71 4.80 4.90 4.99 5.09 WBP13Y

Rice yield

Yield t ha-1

yr-1

(stdev)

0.18 0.18 0.18 0.19 0.19 0.19 0.20 0.20 0.21 0.21 0.21

WBP14Y Pasture yield for

beef, mutton

and milk

Yield t ha-1

yr-1

(mean)

0.32 0.30 0.30 0.30 0.30 0.30 0.30 0.30 0.30 0.30 0.30

78



Hectares 17,059,

563

17,490,

338

17,931,

991

18,384,

797

18,849,

036

19,324,

997

19,812,

978

20,313,

280

20,826,

216

21,352,

104

21,891,

272

NARA EU artificial

area

Hectares

(stdev)

852,97

8

874,51

7

896,60

0

919,24

0

942,45

2

966,25

0

990,64

9

1,015,6

64

1,041,3

11

1,067,6

05

1,094,5

64

Hectares 178,51

7,000

172,37

6,000

167,20

4,720

162,18

8,578

157,32

2,921

152,60

3,233

148,02

5,136

143,58

4,382

139,27

6,851

135,09

8,545

131,04

5,589

WBP1 EU agricultural

area

Hectares

(stdev)

8,925,8

50

8,618,8

00

8,360,2

36

8,109,4

29

7,866,1

46

7,630,1

62

7,401,2

57

7,179,2

19

6,963,8

43

6,754,9

27

6,552,2

79

Hectares 164,17

0,000

166,44

1,000

169,79

6,557

173,21

9,765

176,71

1,986

180,27

4,613

183,90

9,065

187,61

6,790

191,39

9,265

195,25

7,996

199,19

4,523

LF EU forest area

Hectares

(stdev)

8,208,5

00

8,322,0

50

8,489,8

28

8,660,9

88

8,835,5

99

9,013,7

31

9,195,4

53

9,380,8

39

9,569,9

63

9,762,9

00

9,959,7

26

54% 54% 46% 42% 38% 34% 31% 27% 24% 21% 19%

27% 27% 32% 35% 37% 40% 42% 44% 46% 49% 50%

NGHS Grassland

habitat status

Probability for

CPT

19% 19% 22% 23% 25% 26% 27% 28% 29% 30% 31%

41% 41% 27% 22% 17% 13% 10% 7% 5% 4% 3%

33% 33% 37% 38% 37% 37% 35% 34% 32% 30% 28%

NFHS Forest habitat

status

Probability for

CPT

26% 26% 36% 41% 46% 50% 55% 59% 62% 66% 69%

56% 56% 42% 35% 28% 22% 17% 12% 9% 6% 4%

37% 37% 43% 44% 44% 42% 39% 35% 31% 26% 21%

NBHS Bogs and mires

habitat status

Probability for

CPT

8% 8% 15% 21% 28% 35% 44% 53% 61% 68% 75%



79



37.0% 37% 28% 23% 19% 15% 12% 9% 6% 5% 3%

40.0% 40% 36% 32% 29% 24% 20% 16% 13% 10% 8%

NHHS Heath and

scrub status

Probability for

CPT

23.0% 23% 36% 44% 52% 60% 68% 75% 80% 85% 89%

10.3% 10% 16% 20% 24% 29% 33% 38% 43% 47% 52%

47.0% 47% 32% 25% 20% 15% 11% 8% 6% 4% 3%

NSHS Sclerophyllus

scrub status

Probability for

CPT

43.3% 43% 52% 55% 56% 56% 55% 54% 51% 48% 45%

34.3% 34% 20% 15% 11% 8% 6% 4% 3% 2% 1%

45.3% 45% 53% 55% 56% 56% 56% 55% 54% 52% 51%

NAHS Aquatic habitat

status

Probability for

CPT

20.3% 20% 27% 30% 33% 36% 39% 41% 43% 45% 48%

35.0% 35.0% 29% 26% 23% 20% 18% 15% 13% 11% 9%

36.0% 36.0% 33% 31% 29% 27% 24% 22% 19% 17% 15%

NMSS Mammal

species status

Probability for

CPT

29.0% 29.0% 38% 43% 48% 53% 58% 63% 68% 72% 76%

28.0% 28.0% 20% 16% 12% 9% 7% 5% 3% 2% 2%

43.0% 43.0% 42% 38% 34% 29% 24% 20% 16% 12% 10%

NPSS Plant species

status

Probability for

CPT

29.0% 29.0% 38% 46% 54% 62% 69% 76% 81% 85% 89%

48.0% 46.6% 45.0% 43.5% 42.1% 40.6% 39.2% 37.8% 36.4% 35.0% 33.6% NBSS Bird species

status

Probability for

CPT 52.0% 53.6% 55.0% 56.5% 57.9% 59.4% 60.8% 62.2% 63.6% 65.0% 66.4%

80

Table 24: Values for model input nodes of the biodiversity, soils and landscape submodel for baseline scenarios 2000-2050 (Continued

4)


27.0% 33% 39% 45% 51% 57% 63% 69% 75% 75% 75%

32.0% 28% 24% 20% 16% 12% 8% 4% 0% 0% 0%

8.0% 7% 6% 5% 4% 3% 2% 1% 0% 0% 0%

NMPOP Mammal

population

Probability for

CPT

33.0% 32% 31% 30% 29% 28% 27% 26% 25% 25% 25%

23.0% 22% 20% 19% 18% 17% 16% 15% 14% 13% 13%

51.0% 52% 54% 55% 56% 58% 59% 61% 62% 64% 65%

9.0% 9% 9% 9% 9% 9% 9% 10% 10% 10% 10%

NBPOP Bird population Probability for

CPT

17.0% 17% 17% 17% 16% 16% 15% 15% 14% 13% 12%

10.0% 8% 6% 4% 2% 0% 0% 0% 0% 0% 0%

15.0% 14% 13% 12% 11% 10% 7% 4% 1% 0% 0

NALSPP Trend in worst

alien species

Probability for

CPT

75.0% 78% 81% 84% 87% 90% 93% 96% 99% 100% 1

NPHA Protected

habitat area

% of EU land

area

12% 12% 12.12

%

12.24

%

12.36

%

12.49

%

12.61

%

12.74

%

12.87

%

12.99

%

13.12

%

NRD R & D Intensity % GDP 1.9 1.9 1.9 1.9 1.9 1.9 1.9 1.9 1.9 1.9 1.9

100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100%

0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0%

NSFD Soil Framework

Directive

Probability for

CPT

0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0%

Gt CO2eq

(mean)

44.7 48.2 51.9 55.9 60.2 64.9 69.9 75.3 81.1 87.4 94.1 NGHGE Global

greenhouse gas

emissions Gt CO2eq

(stdev)

4.5 4.8 5.2 5.6 6.0 6.5 7.0 7.5 8.1 8.7 9.4



81



Kg ha-1

yr-1

(mean)

172 181 190 200 210 221 232 232 232 232 232 NFERT Fertiliser use

intensity kg/ha

Kg ha-1 yr-1 (st

dev)

126 132 139 146 153 161 169 169 169 169 169

LMr Water

abstraction

2000

109 m

3 yr-1

246

LMr Water

abstraction

109 m

3 yr-1

82

Table 26: List of nodes values with value used in place of normal distribution in baseline model test


NGTE Nutrient transfer

efficiency

(grassland)

% Efficiency

(mean)

28.0 28.3 28.6 28.8 29.1 29.4 29.7 30.0 30.3 30.6 30.9

NATE Nutrient transfer

efficiency

(arable)

% Efficiency

(mean)

52.0 52.5 53.0 53.6 54.1 54.7 55.2 55.8 56.3 56.9 57.4

WBP9Y

Oilseed yield

Yield t ha-1

yr-1

(mean)

2.46 2.69 2.85 3.00 3.15 3.31 3.48 3.66 3.85 4.04 4.25

WBP10Y

Wheat yield

Yield t ha-1

yr-1

(mean)

5.25 5.40 5.62 5.79 5.98 6.16 6.35 6.55 6.76 6.97 7.19

WBP11Y

Grains yield

Yield t ha-1

yr-1

(mean)

2.43 2.45 2.55 2.60 2.66 2.71 2.77 2.82 2.88 2.94 3.00

WBP12Y

Sugar beet yield

Yield t ha-1

yr-1

(mean)

54.24 57.32 60.98 61.60 62.22 62.85 63.49 64.13 64.78 65.43 66.10

WBP13Y

Rice yield

Yield t ha-1

yr-1

(mean)

4.17 4.26 4.37 4.45 4.54 4.63 4.71 4.80 4.90 4.99 5.09



83

Node Node name Measurement 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

NARA EU artificial

area

Hectares 17,059,563 17,490,338 17,931,991 18,384,797 18,849,036 19,324,997 19,812,978 20,313,280 20,826,216 21,352,104 21,891,272

WBP1 EU

agricultural

area

Hectares 178,517,000 172,376,000 167,204,720 162,188,578 157,322,921 152,603,233 148,025,136 143,584,382 139,276,851 135,098,545 131,045,589

LF EU forest

area

Hectares 164,170,000 166,441,000 169,796,557 173,219,765 176,711,986 180,274,613 183,909,065 187,616,790 191,399,265 195,257,996 199,194,523

NGHGE Global

greenhouse

gas

emissions

Gt CO2

equivalent

(mean)

44.7 48.2 51.9 55.9 60.2 64.9 69.9 75.3 81.1 87.4 94.1

NFERT Fertiliser use

intensity

kg/ha

Kg ha-1

yr-1

(mean)

172 181 190 200 210 221 232 232 232 232 232

84

7.3 Definitions of forest land from Eurostat Forestry Statistics 2007

Forest is defined as land with tree crown cover (or equivalent stocking level) of more than 10%

and area of more than 0.5 ha. The trees should be able to reach a minimum height of 5 m at

maturity in situ. It may consist of either closed forest formations, where trees of various storeys

and undergrowth cover a high proportion of the ground, or of open forest formations, with a

continuous vegetation cover, in which tree crown cover exceeds 10% of the area. Young natural

stands and all plantations established for forestry purposes that have yet to reach a crown

density of 10%, or a tree height of 5 m, are included under forest. This is because they are

areas that normally form part of the forest area, that are temporarily unstocked as a result of

human intervention or natural causes, but that are expected to revert to forest.

Other Wooded Land is land either with a tree crown cover (or equivalent stocking level) of 5-

10% of trees able to reach a height of 5 m at maturity in situ, or a crown cover (or equivalent

stocking level) of more than 10% of trees not able to reach a height of 5 m at maturity in situ

(e.g. dwarf or stunted trees) and shrub or bush cover.

FOWL is the total of Forest and Other Wooded Land.

Documents

Project no.: 022793 FORESCENE Development of a