Content uploaded by Tuzin Baycan
Author content
All content in this area was uploaded by Tuzin Baycan
Content may be subject to copyright.
Multidimensional Evaluation of Urban Green Spaces:
A Comparative Study on European Cities
Tüzin Baycan Levent
a
Ron Vreeker
b
Peter Nijkamp
b
a
Department of Urban and Regional Planning, Istanbul Technical University
b
Department of Spatial Economics, Free University Amsterdam
PN149TBLRV
Abstract
Urban green spaces play an important role in improving quality of life and sustainability in
cities and require a careful empirical assessment. Several factors such as social, economic,
ecological or planning aspects, and several functions such as utilization, production,
employment, education, regulation and preservation of urban green spaces form the basis for
the determination of the criteria and indicators relevant for the assessment of urban green
spaces. This multi-faceted ramification of urban green spaces needs therefore, a
multidimensional evaluation approach in an urban policy context. The aim of this paper is to
investigate the complex and heterogeneous structure of urban green spaces from a multi-
faceted assessment perspective. The paper examines urban green spaces from the viewpoint of
“quantity and availability of urban green spaces”, “changes in green spaces”, “planning of
urban green spaces”, “financing of urban green spaces” and “level of performance”, on the
basis of a comparison of 24 European cities by deploying a multi-criteria analysis for mixed
quantitative and qualitative information, coined Regime Analysis. It aims to highlight the
present situation and priorities in decision-making and to compare the “green performance”
of European cities in the process of urban green planning and management. A comparison of
urban green spaces in European cities by means of multi-criteria analysis brings to light the
critical elements in the present situation and sets out choice directions based on priorities in
decision-making and policy evaluation. This evaluation of several experiences in different
regions and countries provides a fascinating European picture in terms of urban green
planning and management.
1. Introduction
In the last decades a broad interest has arisen in planning and evaluation of environmental
goods, and ways of providing a “sustainable development”. The concept of sustainability has
become an important paradigm in urban planning, especially since a large share of the world’s
production, consumption and waste generation is concentrated in cities. Therefore, in recent
years a general concern for quality of life and sustainability, with a particular focus on the
city, has emerged. Given this increasing interest in sustainability, societies have become more
concerned about the built environment and with shaping nature in urban areas, and this has
led to the design of new landscape patterns in the countryside as well as to the creation of
parks and gardens in urban areas (Baycan-Levent et al, 2004; Dole, 1989; DTLR, 2001;
McCarthy et al, 2002; Naess, 2001; Priemus, 1999; Scottish Executive, 2001; Turner et al,
1999a).
The increasing focus on sustainability, the recognition of environmental degradation and the
incorporation of environmental preservation in the broad set of goals of sustainable
development policy has led to the development of proper methods to integrate environmental
explicitly in the decision-making process. Therefore, this new valuation problematic has been
investigated and discussed in several approaches, in particular, in the context of new emerging
disciplines, viz. ecological economics and environmental economics (see for an overview of
different valuation methodologies and techniques Bergh, 1999; Nunes et al, 2003; Turner et
al, 1999a). Such valuation studies aim to estimate the value of the flow of services from the
perspective of the sustainability of ecosystems and their interrelationships with the abiotic
environment. A review of evaluation methods shows that the more comprehensive the
techniques are, the more information is generated to assist in the appraisal of policy options
from a societal perspective (Turner et al, 1999b). Increasing complexity needs an evaluation
of multiple decision criteria and multiple effects. The overview of current evaluation practices
has drawn attention to the combination of methods, tools and data and the integration of
results from evaluations that use different strategies, carried out from different perspectives
and using different methods.
The present paper considers urban green spaces as an important constituent of sustainable
development of cities. Urban green spaces provide a range of benefits at both a national and
local level and offer many opportunities to people in different ways. They help to define and
support the identity of towns and cities, which can enhance their attractiveness for living,
working, investment and tourism. Therefore, they can contribute positively to the
competitiveness of cities. Not only do they provide opportunities for people, they also make
many contributions to social and economic life, and to the ecological and planning system,
and as a whole to the urban quality of life. Many studies refer to the contributions of urban
green spaces from several perspectives including social, economic, ecological or planning
dimensions (Baycan-Levent et al, 2004; Baycan-Levent and Nijkamp, 2004; Dole, 1989;
DTLR, 2001; Groot, 1994; Hart, 1997; Hough, 1984; Hueting, 1970; Jacobs, 1961; Priemus,
1999; Scottish Executive, 2001; Stanners and Bourdeau, 1995). This multidimensional
structure of urban green spaces and their various functions such as utilization, production,
employment, education, regulation and preservation need a multidimensional evaluation
approach.
2
The aim of this paper
1
is to investigate the complex and heterogeneous structure of urban
green spaces from a multidimensional evaluation approach. The paper examines them from
the perspectives of “quantity and availability of urban green spaces”, “changes in green
spaces”, “planning of urban green spaces”, “financing of urban green spaces” and “level of
performance” on the basis of a comparison of European cities by means of multicriteria
analysis. It aims to highlight the present situation and priorities in decision-making and policy
evaluation in the process of urban green planning and management. The next section
evaluates urban green spaces from the perspective of values and valuation methods and
describes the multi-faceted values of urban green and relevant evaluation models and
methodologies. Section 3 examines multidimensional evaluation approaches and application
fields. Section 4 explains the Regime Analysis, which is one of the decision-making
approaches of multi-criteria analysis that we deployed in this study on the evaluation of urban
green spaces. Section 5 compares the “green performance” of European cities in terms of the
present situation and priorities in decision-making and planning on the basis of the empirical
results of a multidimensional evaluation of urban green spaces. The final section offers some
concluding remarks, which focus on the critical aspects of green planning policies.
2. Evaluation of urban green spaces: values and valuation methods
The difficulties in developing a single unifying definition of value have induced several
precise definitions of value originating from different disciplines to meet different needs. A
transparent description of the values of urban green spaces is important in order to define and
measure their contributions to urban quality of life. It is also important to propose the
appropriate valuation methods for different values of urban green spaces. What are these
values? Several functions of urban green spaces such as utilization, production, employment,
education, regulation and preservation form the basis for the determination of the dimensions,
criteria and indicators that are relevant for their assessment. These functions also form types
or classes of values. A taxonomy of values for urban green spaces, which stems from its
functions and is based on several perspectives, viz. ecological, economic, social and planning,
has been described in a previous study by Baycan-Levent and Nijkamp (2004). In this
taxonomy the authors defined sixteen types of urban green space values classified according
to five distinctions: 1) ecological values: intrinsic natural value, genetic diversity value, life
support value; 2) economic values: market value; 3) social values: recreational value,
aesthetic value, cultural symbolization value, historical value, character building value,
therapeutic value, social interaction value, substitution value; 4) planning values:
instrumental/structural value, synergetic and competitive value; 5) multidimensional values:
scientific value, policy value (see Figure 1). A more detailed explanation of these values can
be found in the abovementioned study (see Baycan-Levent and Nijkamp, 2004). They
emphasized that the complex and multidimensional structure of urban green spaces make a
clear distinction among these dimensions difficult, while many values in the taxonomy refer
to more than one dimension. However, to avoid the duplication of the individual values, each
1
This paper is part of the EU project “Development of Urban Green Spaces to Improve the Quality of Life in
Cities and Urban Regions” (URGE) which is funded under Key-Action 4 “The City of Tomorrow and Cultural
Heritage” of the Programme “Energy, Environment and Sustainable Development” of the 5th Framework
Programme of the European Union (URGE, 2004). The URGE project aims to develop interdisciplinary tools for
scientists as well as for planners all over Europe for the planning and management of urban green spaces. The
main question addressed is how urban green spaces (both in a qualitative and a quantitative sense) can be
developed from ecological, economic, social and planning perspectives, and which tools and instruments are
helpful in this respect. The overall project goal is the elaboration and testing of an interdisciplinary, systematic
catalogue of methods and measures, based on a broad experience from various European cities.
3
value is placed only in one category/dimension according to its importance level among all
these dimensions. Arrows indicate the direct and indirect importance and impact of each value
for other dimensions. For example, the ecological values of urban green spaces provide
resources for economic production such as wood, fruits, compost and energy. The life support
value on the other hand, contributes positively to the urban micro-climate and as a result to
society by absorbing pollutants and releasing oxygen. Economic values are interrelated with
society in terms of production and employment. The preferences of society on the other hand,
affect economic values and impacts of urban green spaces in terms of hedonic values. Rather
complex relationships exist between society and planning, while social values constitute the
foundation stones of planning. Planning decisions may raise or reduce the satisfaction of
society at large; it generally plays an interconnected systemic role among the other values.
However, planning values in this taxonomy reflect a technical instrumental role. The common
values for all dimensions (scientific and policy values) on the other hand, are placed in a
separate dimension/category, which is based on four dimensions of urban green, described
here as multidimensional values. This taxonomy of urban green space values reflects a
comprehensive evaluation from several perspectives at a conceptual level. However, it is
obvious that the evaluation should focus on a representative number of well-chosen and
transparent criteria and indicators according to evaluation needs at the application level.
The taxonomy of urban green space values points out the complex and multidimensional
structure of urban green spaces. This complex and multidimensional structure makes the
description or design of a single “best” model difficult. In fact, it requires a set of relevant
models derived from several perspectives such as environmental, economic, social or policy
valuation to evaluate and integrate all dimensions of urban green spaces in order to provide
their input to each other. This multidimensional evaluation should comprise monetary and
non-monetary valuation methods for both quantitative and qualitative information.
Consequently, this evaluation should provide relevant policies and guidance for society and
planning authorities to improve the quality of life in cities. Therefore, the possible evaluation
methods for several dimensions of urban green spaces should be seen as an integrative part of
a comprehensive evaluation.
Baycan-Levent and Nijkamp (2004) also developed an operational taxonomy for the
evaluation of urban green spaces in parallel to their taxonomy of urban green space values.
This taxonomic framework offers a systematic assessment approach regarding the complex
and multidimensional structure of urban green spaces (see Table 1). The valuation methods
for several dimensions of urban green spaces in this taxonomy are based on “monetary
valuation” and “non-monetary valuation”. These valuation methods described in Table 1 can
be used to evaluate several dimensions/categories of urban green spaces. Clearly, from a
comprehensive methodological perspective an overall evaluation -in other words, an
“evaluation of evaluation”- may also be needed which is based on a broad “systems
approach”. For the “evaluation of evaluation” of several dimensions of urban green spaces
relevant methods could be “meta-analysis” and “meta-multi-criteria decision methods”.
While meta-analysis can be used to translate value estimates into other contexts, conditions,
locations or temporal settings that do not allow for direct valuation in “primary studies”,
meta-multi-criteria decision methods can be used to better understand the complex and
multidimensional structure of urban green spaces especially at the policy level.
Multidimensional evaluation approaches, application fields and its relevance for evaluating
urban green spaces are examined in the next section.
4
Intrinsic natural value
Genetic diversity value:
-preservation function
Life support value:
-regulation function
Figure 1 Taxonomy of urban green space (UGS) values
(Source: Baycan-Levent and Nijkamp, 2004)
5
Substitution value
Recreational value:
-utilization function
Aesthetic value
Cultural symbolization value
Historical value
Character building value
Social interaction value
Social values of UGS
Therapeutic value
Scientific value:
-education function
Policy value:
-financial function
-public function
Multidimensional values of UGS
Taxonomy of urban green space (UGS) values
Market value:
- production function
- employment function
-
hedonic function
Economic values of UGS
Synergetic and competitive value
Instrumental/structural value
Planning values of UGS Ecological values of UGS
Table 1 Types of urban green space (UGS) values and valuation methods (Baycan-Levent and
Nijkamp 2004)
Types of urban green space values Valuation methods
Ecological values of UGS:
• Intrinsic natural value (existence value)
• Genetic diversity value (bequest value)
• Life support value (indirect use value)
Monetary valuation: cost-benefit analysis,
travel cost method, replacement costs, tourism
revenues, production function, contingent
valuation
Non-monetary valuation: species and
ecosystem richness indices, genetic difference,
genetic distance, phenotypic trait analysis,
biodiversity index, keystone processes, health
index, ecosystem resilience and stability
analysis, hierarchical structure, population
viability analysis, eco-regions or eco-zones
Economic values of UGS:
• Market value (direct/indirect use value)
Monetary valuation: market analysis,
production functions, financial analysis,
economic cost-benefit analysis, travel cost
method, hedonic price method
Social values of UGS:
• Recreational value (direct use value)
• Aesthetic value (existence value)
• Cultural symbolization value (existence value)
• Historical value (bequest value)
• Character building value (indirect use value)
• Therapeutic value (indirect use value)
• Social interaction value (indirect use value)
• Substitution value (direct use value)
Monetary valuation: travel cost method,
tourism revenues
Non-monetary valuation: contingent valuation
Planning values of UGS:
• Instrumental/structural value (indirect use value)
• Synergetic and competitive value (existence value)
Monetary valuation: cost-benefit analysis,
contingent valuation, hedonic price method
Non-monetary valuation: geographical
information system (GIS) method, multi-criteria
decision method
Multidimensional values of UGS:
• Scientific value (indirect use value)
• Policy value (indirect use/existence value)
Monetary valuation: financial analysis, cost-
benefit analysis, cost-effectiveness analysis,
tourism revenues, taxes revenues
Non-monetary valuation: performance analysis,
multi-criteria decision method, meta-analysis,
value transfer, rough set analysis, fuzzy set
analysis, content analysis
3. Multidimensional evaluation approaches
In the past decades various decision support and evaluation methods have been developed. As
a response to the shortcomings of conventional evaluation studies, a great diversity of modern
assessment methods has been developed over the last decades in order to extend the range of
methods and to provide a complement to conventional cost-benefit analysis and to offer a
perspective for procedural types of decision-making in which various quality aspects are also
incorporated. Many of these methods simultaneously investigate the impact of policy
strategies (e.g. project, plan, or programme) on a multitude of relevant criteria, partly
monetary, partly non-monetary (including qualitative facets). They are often coined multi-
criteria methods and are also known as multi-assessment methods.
6
An important analytical tool in the framework of multidimensional evaluation analysis is
multi-criteria analysis. It can also been seen as a generalized and more flexible version of
cost-benefit analysis. Multi-criteria evaluation can be regarded as a more general method of
welfare analysis, and presents an empirical support mechanism for a traditional, neo-classical
economic analysis of social welfare and policy instruments (see Munda, 1997). Multi-criteria
analysis is part of decision theory, which aims to identify the best possible alternative out of a
set of rival choice options, where each option is characterized by multiple different judgement
criteria. These criteria may be mutually conflicting in nature and hence lead to complex trade-
off problems. Furthermore, the relative policy priority or weight attached to each individual
criterion impacts on the final choice. In a multi-criteria framework, different mutually
conflicting evaluation criteria are taken into consideration. Multi-criteria decision methods
constitute an important toolbox for helping a decision-maker to master actions involving
multiple criteria (Arrow and Raynaud, 1986; Roy, 1990). An evaluation method can support
the ranking of alternative choice options regarding management, policy, development
scenarios or projects. This can lead to a complete ranking, the best alternative, a set of
applicable alternatives, or an incomplete ranking. In general, the aim of these methods is to
combine assessment methods with judgement methods and to offer a solid analytical basis for
modern decision analysis (Nijkamp et al, 1990).
Multi-criteria methods can be distinguished into two classes depending on the level of
information: quantitative and qualitative methods. While quantitative methods require
numerical information about scores of each criterion, qualitative methods can be used if non-
numerical or categorical information on scores is available or if a mixture of quantitative and
qualitative scores is available (Janssen and Munda, 1999). Multi-criteria analysis comprises
various classes of decision-making approaches that are based on the characteristics of
information such as Regime Analysis, Flag Model, and Saaty’s Hierarchical Method (see for a
general overview Vreeker et al, 2002). Over the past decades, a wide range of multi-criteria
analysis methods has been developed. Multi-criteria analysis has become a useful tool in
evaluation and planning studies such as land-use and transportation, but also in relation to
urban sustainability analysis. It has also seen a wide range of applications in environmental
studies because of its potential to simultaneously consider different mutually irreducible or
incompatible judgement criteria, including non-monetary aspects.
4. Multidimensional evaluation of urban green spaces: Regime analysis
After a general overview on multidimensional evaluation approaches in the previous section,
here we evaluate the multidimensional nature of urban green spaces. First we explain the
Regime Analysis as a methodology which is one of the decision-making approaches of multi-
criteria analysis deployed in this study on the evaluation of urban green spaces and then, in
the next section, we evaluate the results of the Regime Analysis which enable us to compare
“green performance” of European cities.
Multi-criteria analysis comprises various classes of decision-making approaches. The multi-
assessment method used in our methodology is Regime Analysis. Regime Analysis is a
discrete multi-assessment method suitable to assess projects as well as policies. The strength
of Regime Analysis is that it is able to cope with binary, ordinal, categorical and cardinal
(ratio and interval scale) data, while the method is also able to use mixed data. This applies to
both the effects and the weights in the evaluation of alternatives.
7
The fundamental framework of the method is based upon two kinds of input data: an impact
matrix (structured information Table) and a set of (politically determined) weights (Nijkamp
et al, 1990, and Hinloopen et al, 1983). The impact matrix is composed of elements that
measure the effect of each considered alternative in relation to each policy-relevant criterion.
The set of weights incorporates information concerning the relative importance of the criteria
in the evaluation. In case there is no prioritization of criteria in the evaluation process, all
criteria will be assigned the same numerical weight value.
Regime Analysis is a discrete multicriteria method, and in particular, it is a generalized form
of concordance analysis, based in essence on a generalization of pair-wise comparison
methods. In order to gain a better understanding of Regime Analysis, let us reiterate the basic
principles of concordance analysis. Concordance analysis is an evaluation method in which
the basic idea is to rank a set of alternatives by means of their pair-wise comparisons in
relation to the chosen criteria. We consider a choice problem where we have a set of
alternatives i and a set of criteria k. We begin our analysis by comparing alternative i with
alternative k in relation to all criteria. After having done this, we select all criteria for which
alternative i performs better than, or is equal to, alternative k. This class of criteria we call a
"concordance set". Similarly, we define the class of criteria for which alternative i performs
worse than, or is equal to, alternative k. This set of criteria is called a "discordance set".
We now need to rank the alternatives. In order to do so, we introduce the concordance index.
The concordance index is the sum of the weights that are related to the criteria for which i is
better than k. We call this sum C
ik
. Then we calculate the concordance index for the same
alternatives, by considering the criteria for which k is better than i, i.e., C
ki
. After having
calculated these two sums, we subtract these two values in order to obtain the net concordance
index µ
ik
=C
ik
-C
ki
.
Because in most cases we have only ordinal information about the weights (and no trade-
offs), our interest is in the sign of the net concordance index µ
ik.
If the sign is positive, this
will indicate that alternative i is more attractive than alternative k; otherwise, the opposite
holds. We are now able to rank our alternatives. We note that due to the ordinal nature of the
information in the indicator µ
ik,
no information exists on the size of the difference between the
alternatives; it is only the sign of the indicator that matters.
We may also solve the complicating situation that it may not be able to determine an
unambiguous result, i.e. a complete ranking of alternatives, because of the problem of
ambiguity in the sign of the index µ. In order to solve this problem, we introduce a
performance indicator - as a semi-probability measure - p
ik
for the dominance of criteria i with
respect to criteria k as follows:
p prob
ij ij
=>()
µ
0
Next, we define an aggregate probability measure, which represents the success
(performance) score as follows:
p
I
p
ii
ji
=
−
≠
∑
1
1
j
where i is the number of chosen alternatives.
8
The problem here is to assess the value of p
ij
and of p
i
. The Regime Analysis then assumes a
specific probability distribution of the set of feasible weights. This assumption is based upon
the Laplace criterion in the case of decision-making under uncertainty. In the case of a
probability distribution of qualitative information, in principle, the use of stochastic analysis
will suffice, which is consistent with an originally ordinal data set. This procedure helps to
overcome the methodological problems we may encounter by applying a numerical operation
on qualitative data.
From the viewpoint of numerical analysis, the Regime method identifies the feasible domain
within which feasible values of the weights w
I
must fall in order to be compatible with the
condition imposed by their probability value. By means of a random generator, numerous
values of the weights can be calculated. This allows us at the end to calculate the probability
score (or success score) p
I
for each alternative i. We can then determine an unambiguous
solution and rank the alternatives.
Regime Analysis is able to examine both quantitative and cardinal data. In the case of choice
problems with qualitative data, we first need to transform the qualitative data into cardinal
data and then apply the Regime method. The Regime Software method is able to do this
consistently
2
and therefore is classified as an indirect method for qualitative data. This is an
important positive feature. When we apply the cardinalization of qualitative data through
indirect methods such as the Regime Analysis, we do not lose information as we do in direct
methods. This is due to the fact that in direct methods only the ordinal content of the available
quantitative information is used.
5. “Green performance” of European cities
Since this paper is related to the EU project “Development of Urban Green Spaces to Improve
the Quality of Life in Cities and Urban Regions” (URGE), the definition of urban green space
that is used here is almost similar to the one that is used in the URGE project, and it has been
formulated jointly by ecologists, economists, social scientists and planners. The following
definition was agreed upon:
By urban green spaces we understand public and private open spaces in urban areas,
primarily covered by vegetation, which are directly (e.g. active or passive recreation) or
indirectly (e.g. positive influence on the urban environment) available for the users.
In this study, we aim to compare the “green performance” of European cities in terms of the
present situation, priorities in decision-making and planning, and their success level from
proper evaluation perspectives. The sample in our study contains European cities, which aim
to share their experience in innovative green space policies and strategies. The data and
information used for comparison and evaluation are based on extensive questionnaires
obtained from relevant departments or experts of municipalities in European cities. We
examine urban green spaces from both quantitative and qualitative information on the basis of
multi-criteria analysis by means of the abovementioned Regime Analysis.
2
Regime analysis is included in the software package SAMIsoft, a deliverable of the EU project SAMI.
9
Description of data base
For the application of multi-criteria analysis, five thematic groups of criteria were
distinguished. The first criteria group, called “Quantity and availability of urban green
spaces”, focuses on the most important quantitatively well-definable physical features of
urban green spaces. In the second criteria group, “Changes in green spaces”, recent changes
in the total area of green spaces in the last 10 years is examined quantitatively. The third
criteria group, “Planning of green spaces”, contains qualitative criteria referring to the
planning system of a city. In the fourth criteria group, “Financing of urban green spaces”, the
changes in the budget are investigated quantitatively. The last criteria group, “Level of
performance”, reflects the success level of urban green space policy in light of the objectives
of a city from the representatives’ own evaluation perspectives. Table 2 summarizes these five
thematic groups of criteria and accompanying sub-criteria.
Table 2 Dimensions of urban green spaces policy and accompanying criteria
Criteria Sub-Criteria Data Type Expected value
Quantity and availability
of urban green spaces
Proportion of green spaces with respect
to total area (%)
Quantitative Higher is better
Proportion of green spaces per 1000
inhabitant (m2)
Quantitative Higher is better
Existence of a regional green space
system
Qualitative Existence is better
Changes in green spaces
Recent changes in the total area of
green spaces in the last 10 years
Quantitative Increase is better
Planning of green spaces
Importance of green spaces to the city
compared to other functions
Qualitative Higher is better
Existence of general goals and
strategies for the planning of urban
green
Qualitative Existence is better
Existence of special planning
instruments for urban green spaces
Qualitative Existence is better
Experience with citizens participation Qualitative Experience is better
Financing of urban green
spaces
Changes in the budget for greenery in
the last two years
Quantitative Increase is better
Level of performance
Success level of urban green space
policy in light of the objectives of a
city from the representatives’ own
evaluation perspectives
Qualitative Higher is better
The definition for each sub-criterion in Table 2 and its data sources are given below. These
definitions also explain the data set in Table 3, which we use in our Regime Analysis.
Proportion of green spaces with respect to total area (%): This is the proportion of total green
spaces in terms of land-use within the administrative area of the city. Total green spaces
consist of gardens, urban parks, quarter parks, historical gardens, green squares and plazas,
green playgrounds and other city-specific green spaces. This information is obtained by
presenting two different questions: the first one pertains to the land-use data, while the second
explores the types of urban green spaces.
Proportion of green spaces per 1000 inhabitant (m
2
): This data is obtained directly from the
representatives of municipalities by questionnaires.
10
Existence of a regional green space system: A regional green space system is defined by
green fingers (following natural lines, e.g. rivers), green corridors (following traffic routes),
greenbelts, disjointed green, urban forests and other city specific green spaces. This
information about existence of a regional green space system is obtained in terms of “yes” or
“no”.
Recent changes in the total area of green spaces in the last 10 years: This data is collected
directly from the representatives of municipalities via questionnaires. The changes are defined
as an increase, a decrease, or no change in the total area of green spaces in the last 10 years.
Importance of green spaces to the city compared to other functions: With this question it is
the aim to highlight the importance and the priority of urban green spaces in the city from the
perspective of planning authorities. The importance of green spaces is defined in five
categories; (1) very important, (2) important, (3) medium, (4) less important, (5) not
important.
Existence of general goals and strategies for the planning of urban green: This information
stems directly from the representatives of municipalities via questionnaires in terms of “yes”
or “no”.
Existence of special planning instruments for urban green: This information is gained directly
from the representatives of municipalities via questionnaires in terms of “yes” or “no”.
Experience with citizen participation: This information is derived directly from the
representatives of municipalities via questionnaires in terms of “yes” or “no”.
Changes in the budget for greenery in the last two years: This data is gleaned directly from
the representatives of municipalities via questionnaires. The changes are defined as an
increase, a decrease, or no change in the budget for greenery in the last two years.
Level of performance: With this question we aim to highlight the success level of urban green
space policies in the light of the objectives of a city from the representatives’ own evaluation
perspectives. The performance is defined in five categories: (1) very successful, (2)
moderately successful, (3) marginally successful, (4) low success, (5) no success at all.
Results of regime analysis
In our case study we evaluated urban green spaces in 24 European cities (see Table 3). In
order to do so we used 10 indicators defined in the previous section, emphasizing different
aspects of the quality of urban green. The measurement scales of these indicators differ and
range from ordinal scales (yes/no indicators) to ratio scales (percentages). This mixture of
data limits the application of quantitative data analysis techniques. Therefore, in this case
study we conducted a multicriteria evaluation tool, which is capable of encapsulating mixed
(qualitative and quantitative data) data sets. This was done on the basis of the previously
described multicriteria evaluation tool Regime Analysis.
In our analysis of urban green spaces we used Regime Analysis, the data set presented in
Table 3, and a uniform weight sector. A uniform weight vector implies that all indicators are
of equal importance and are assigned the same weight value. In a subsequent sensitivity
analysis, we will apply different weighting schemes, resulting in a robust emphasis on
11
specific types of indicators. This analysis will indicate whether the obtained results are robust
enough for changes in the values of weights.
12
Table 3 Impact Matrix
MULTICRITERIA EVALUATION OF URBAN GREEN SPACES
Quantity and availability of urban green
spaces
Changes in green
spaces
Planning of green spaces Financing of urban
green spaces
Level of
performance
Cities
1) Proportion of
green spaces
with respect to
total area (%)
2) Proportion of
green spaces per
1000 inhabitants
(m2)
3) Existence
of a regional
green space
system
4) Recent changes in
the total area of green
spaces in the last 10
years
5) Importance of
green spaces to
the city compared
to other functions
6) Existence of
general goals
and strategies
for the planning
of urban green
7) Existence of
special planning
instruments for
urban green
8)
Experience
with citizens
participation
9) Changes in the
budget for greenery
in the last 2 years
10) Level of
performance
Antwerp 11,3 51509 No + + Yes No Yes - 1
Berlin 14,3 37786 Yes + ++ Yes Yes Yes - 3
Bern 10,4 42519 Yes + + Yes Yes Yes - 2
Birmingham 14,0 20000 Yes + 0 No Yes Yes - 2
Budapest 21,3 61800 Yes - 0 Yes No No + 3
Cracovia 2,6 65455 Yes - 0 Yes Yes Yes + 3
Dublin 16,4 40000 Yes 0 + Yes Yes Yes + 2
Edinburgh 25,0 144592 Yes - + Yes Yes Yes - 3
Espoo 1,0 24465 Yes + + Yes Yes Yes + 2
Genoa 13,1 49394 No + + No No No 0 3
Helsinki 7,6 94154 Yes - + Yes Yes Yes - 2
Istanbul 0,5 2675 Yes + ++ Yes No Yes + 1
Leipzig 14,8 89617 Yes + 0 Yes Yes Yes - 2
Ljubljana 2,6 25966 Yes - 0 Yes No Yes + 4
Lodz 4,0 14947 Yes 0 0 Yes No No + 4
Malaga 59,3 4614815 Yes + + Yes No Yes + 2
Marseilles 39,3 118225 Yes + ++ Yes No No + 1
Montpellier 11,0 27729 Yes + + Yes No Yes + 1
Salzburg 11,4 51755 Yes - + Yes Yes Yes - 2
Sarajevo 1,2 11818 Yes - + Yes No Yes + 3
Turin 13,5 19444 Yes + 0 Yes No Yes 0 2
Vienna 14,4 36863 Yes - + Yes No Yes 0 2
Warsaw 22,3 68499 Yes - + Yes Yes Yes + 2
Zurich 17,4 44253 Yes - + Yes Yes Yes + 1
13
A first investigation of the data set points us towards various so-called inefficient alternatives
or green spaces. Inefficient alternatives are alternatives that are dominated by at least one
other alternative with respect to all indicators. The results of this first inspection are presented
in Table 4.
Table 4 Inefficient alternatives according to the regime results
City Dominated by
Birmingham Leipzig
Budapest Malaga, Marseilles, Warsaw
Cracovia Warsaw
Genoa Malaga, Marseilles
Ljubljana Cracovia, Dublin, Malaga, Montpellier,
Warsaw, Zurich.
Lodz Dublin, Malaga, Marseilles, Montpellier.
Salzburg Warsaw.
Sarajevo Dublin, Malaga, Montpellier, Warsaw, Zurich.
Turin
Malaga
Vienna Dublin, Malaga, Warsaw, Zurich
The ranking obtained by means of the Regime Analysis is depicted in Figure 2. The results of
the Regime Analysis, with the application of a uniform weight vector, indicate that the green
spaces in the cities of Marseilles and Malaga both obtain the highest position in our ranking.
Both cities dominate the other cities with respect to the availability of urban green spaces.
This is highlighted by the high scores on Percentage of green space and Proportion of
green spaces per 1000 inhabitants. The reason behind this result can be explained by
specific local circumstances of these cities. Marseilles, for example, has specific urban green
spaces, “espaces naturels” that cover 90 km
2
. With its forest-like parks of 426 km
2
, urban
parks of 1100 km
2
and a regional green system of “natural parks” of 49.957 km
2
, Malaga, on
the other hand, reflects a very special situation. However, when we take a look at the planning
related indicators (indicators 5, 6 and 7), we can see that both Marseilles and Malaga perform
on average and are dominated by the Northern-European Cities of Berlin, Bern, Dublin,
Edinburgh, Espoo, Leipzig, Salzburg, Warsaw and Zurich.
After performing a sensitivity analysis on the applied weighting vector, we can conclude the
following:
- No weight vector exists whereby Genoa will attain the first position in our ranking or will
be ranked above Malaga and Marseilles.
- Malaga will attain a second position in our ranking in case much emphasis is placed on
the special planning instruments.
- Istanbul will together with Malaga, gain the first position in our ranking in case much
emphasis is placed on changes in total area of green spaces.
- Genoa and Antwerp will change positions if more weight is placed on the indicator
changes in the budget.
14
Re g i m e A n a ly sis
Marseilles
Malaga
Genoa
A
ntwerp
Montpellie
r
Budapest
Warsaw
Zurich
Dublin
Vienna
Leipzig
Edinburgh
Turin
Berlin
Bern
Istanbul
Birmingham
Helsinki
Salzburg
Lodz
Espoo
Cracovia
Ljubljana
Sarajevo
Result
0,98
0,97
0,9
0,84
0,77
0,72
0,71
0,67
0,61
0,55
0,5 0,5
0,47
0,44
0,4
0,37
0,35
0,31
0,26
0,23
0,21
0,11
0,07
0,05
Figure 2 Ranking of European cities by means of “green performance”
15
To summarize, the Regime Analysis places the cities of Malaga and Marseilles first in the
overall ranking. However, the indicators related to the availability of green spaces mainly
determine this ranking. If we take a look at the planning performance indicators, we may
conclude that the northern European cities have higher scores. Many of these cities have a
below average score on the availability of green spaces. High scores on the planning might
indicate significant efforts to maintain the quality of the existing urban green spaces.
6. Concluding remarks
Urban green spaces have multi-faceted values comprising ecological, economic, social and
planning dimensions. This multidimensional nature of urban green spaces requires a
multidimensional set of characteristic attributes and a set of judgement criterion in a broader
context of relevant policy angles from a methodological perspective.
This paper focuses on the complex and multidimensional structure of urban green spaces
especially at a policy level. It compares the “green performance” of European cities in terms
of the present situation, priorities in decision-making and planning, and their success level
from proper evaluation perspectives on the basis of multi-criteria analysis through Regime
Analysis. The results of this Regime Analysis show that the indicators related to the
availability of urban green spaces determine the green performance and ranking of European
cities. However, when the planning performance indicators are taken into consideration, the
northern European cities have higher scores.
The results of this analysis should be interpreted with caution. When coming to a final verdict,
one has to take into account the local circumstances. For example, the Finnish cities of
Helsinki and Espoo both score poorly in our ranking. This is mainly caused by the below
average availability of urban green spaces in Espoo and by the decrease in both the total area
of green spaces, and the budget for greenery in Helsinki. This does not indicate that
inhabitants of those cities do not have access to green areas. Espoo and Helsinki are both
surrounded by enormous natural green areas such as green fingers and forests.
On the other hand, it should be kept in mind that the performance indicator used in this
analysis is not an objective indicator, as it reflects the perceived performance of the city from
the representatives’ evaluation perspectives. The representatives evaluate the performance in
greater detail from several viewpoints. For example, some of the representatives of the cities
in our sample indicated that they are successful in the conservation and management of urban
green spaces, but they have no success in creating new green spaces, while some others
emphasized that they are successful in central parts of the city, but that the availability of
green spaces decreases between the city centre and the suburbs. Clearly, this is caused by the
fact that we analysed the performance of a city by means of a performance score provided by
that city. In conclusion, the use of a multi-criteria analysis has allowed us to create a ranking
of the performance of urban green space policy in various cities, as seen through the eyes of
the cities’ representatives.
16
References
Arrow, K.J., Raynaud, H. (1986) Social Choice and Multicriterion Decision Making,
Cambridge, Massachusetts, London: MIT Press.
Baycan-Levent, T., Nijkamp, P. (2004) Evaluation of Urban Green Spaces, in D. Miller, D.
Patassini (Eds) Accounting for Non-Market Values in Planning Evaluation: Alternative
Methodologies and International Practices, Ashgate Publishing Limited (forthcoming).
Baycan-Levent, T., Leeuwen, E. van, Rodenburg, C., Nijkamp, P. (2004) Urban Sustainability
and Green Spaces, Indian Journal of Regional Science, (forthcoming).
Berg, J.C.J.M. van den (ed.) (1999) Handbook of Environmental and Resource
Economics, Cheltenham: Edward Elgar Publishers.
Dole, J. (1989) Greenscape 5: Green Cities, Architects’ Journal, 10 May 1989, pp. 61-69.
DTLR (2001) Green Spaces, Better Places: Interim Report of the Urban Green Spaces
Taskforce, Department for Transport, Local Government and the Regions, London.
Groot, R.S. de (1994) Environmental Functions and the Economic Value of Natural
Ecosystems, in A.M. Jansson, M. Hammer, C. Folke, R. Constanza (eds) Investing in
Natural Capital: The Ecological Economics Approach to Sustainability, Washington:
Island Press.
Hart, R. (1997) Children’s Participation: The Theory and Practice of Involving Young
Citizens in Community Development and Environmental Care, London: Earthscan
Publications
Hinloopen, E., Nijkamp, P., Rietveld, P. (1983) Qualitative Discrete Multiple Criteria Choice
Models, Regional Planning, Regional Science and Urban Economics, pp. 77-102.
Hough, M. (1984) City Form and Natural Processes. London: Croom Helm.
Hueting, R. (1970) Wat is de natuur ons waard? (What is nature worth to us?) Baarn:
Wereldvenster.
Jacobs, J. (1961) The Death and Life of Great American Cities, Harmondsworth: Penguin
Books.
Janssen, R., Munda, G. (1999) Multi-criteria methods for quantitative, qualitative and fuzzy
evaluation problems, in J.C.J.M. van den Berg (ed) Handbook of Environmental and
Resource Economics, Cheltenham: Edward Elgar Publishers.
McCarthy, J., Lloyd, G., Illsley, B. (2002) National Parks in Scotland: Balancing
Environment and Economy, European Planning Studies, 10(5), pp. 665-670.
Munda, G. (1997) Multicriteria Evaluation as a Multidimensional Approach to Welfare
Measurement, in J.C.J.M. van den Bergh, J. van der Straaten (eds) Economy and Ecosystems
in Change: Analytical and Historical Approaches, Cheltenham: Edward Elgar Publishers.
17
18
Naess, P. (2001) Urban Planning and Sustainable Development, European Planning Studies,
9(4), pp. 503-524.
Nijkamp, P., Rietveld, P., Voogd, H. (1990) Multicriteria Evaluation in Physical Planning,
Amsterdam: North Holland.
Nunes, P.A.L.D, Bergh, J.C.J.M van den, Nijkamp, P. (2003) The Ecological Economics of
Biodiversity: Methods, Values and Policy Applications, Cheltenham: Edward Elgar
Publishers.
Priemus, H. (1999) Sustainable Cities: How to Realize an Ecological Breakthrough: A Dutch
Approach, International Planning Studies, 4(2), pp. 213-236.
Roy, B. (1990) Decision Aid and Decision Making, in C.A. Bana e Costa (ed) Readings in
Multiple Criteria Decision Aid, Berlin: Springer Verlag.
Scottish Executive (2001) Rethinking Open Space, The Stationery Office, Kit Campbell
Associates, Edinburgh.
Stanners, D., Bourdeau, P. (1995) Europe’s Environment: The Dobris Assessment, Office
for Official Publications of the European Communities, Luxembourg.
Turner, R.K., Button, K., Nijkamp, P. (eds) (1999a) Ecosystems and Nature: Economics,
Science and Policy, Environmental Analysis and Economic Policy: 7, Cheltenham: Edward
Elgar Publishers.
Turner, R.K., Subak, S., Adger, W.N. (1999b) Pressures, Trends, and Impacts in Coastal
Zones: Interactions Between Socioeconomic and Natural Systems, in R.K. Turner, K. Button,
P. Nijkamp, Ecosystems and Nature: Economics, Science and Policy, Environmental
Analysis and Economic Policy: 7., Cheltenham: Edward Elgar Publishers.
URGE (2004) www.urge-project.org
Vreeker, R., Nijkamp, P., Welle, C.T. (2002) A multicriteria decision support methodology
for evaluating airport expansion plans, Transportation Research Part D, 7, pp. 27-47.
