Please cite this article in press as: Ioj a, C.I., et al.

, Using multi-criteria analysis for the identication of spatial land-use conicts in the
Bucharest Metropolitan Area. Ecol. Indicat. (2013), http://dx.doi.org/10.1016/j.ecolind.2013.09.029
ARTICLE IN PRESS
G Model
ECOIND-1686; No. of Pages 10
Ecological Indicators xxx (2013) xxxxxx
Contents lists available at ScienceDirect
Ecological Indicators
j our nal home page: www. el sevi er . com/ l ocat e/ ecol i nd
Using multi-criteria analysis for the identication of spatial land-use
conicts in the Bucharest Metropolitan Area
Cristian Ioan Ioj a, Mihai R azvan Nit a, Gabriel Ovidiu Vn au, Diana Andreea Onose

,
Athanasios Alexandru Gavrilidis
Centre for Environmental Research and Impact Studies, University of Bucharest, N. Balcescu Boulevard, No. 1, Sector 1, 010041 Bucharest, Romania
a r t i c l e i n f o
Article history:
Received 22 April 2013
Received in revised form25 July 2013
Accepted 10 September 2013
Keywords:
Multi-criteria analysis
Land-use conicts
Spatial indicators
Urban development indicators
Bucharest Metropolitan Area
a b s t r a c t
The appearance of land-use conicts represents one of the expressions of the increasing human pressure
on the environment, especially in complex metropolitan areas. We used a multi-criteria analysis applied
in the Bucharest Metropolitan Area to create a tool for integrating land-use conicts into the strategies
for territory planning at the metropolitan level. We selected ten main criteria for the analysis, divided
them into two categories, i.e., spatial indicators and urban development indicators, and standardized
their values. Using the method of pair-wise comparison with an expert-opinion system, we determined
the relative importance of each criterion in the form of a criteria weight. The spatial indicators reveal
high probabilities for land-use conicts in the proximity of Bucharest (range: 7.1952.48), whereas urban
development indicators show a scattered spatial distribution of land-use conicts (range: 10.0030.01).
The variability of the total score for the spatial indicators is greater than that of the urban development
indicators (standard error of 0.64 > 0.30, respectively). The total scores reveal local administrative units
characterized by a critical or high incidence of spatial-land use conicts as well as their location in the
Bucharest Metropolitan Area. Further research should concentrate on improving the expert-opinion input
for assigning weights and revealing the potential for replication in other areas or on different subjects.
2013 Elsevier Ltd. All rights reserved.
1. Introduction
Because indicators represent a formof conscious connection to
the environment (Golusin and Ivanovic, 2009) and the resilience
of nature (Passeri et al., 2013), society uses indicators to simplify
reality or for comparison with a normal or reference state (EEA,
2010). Indicators measure the characteristic values of perceived
realities (Kurtz et al., 2001) to produce information and describe
phenomena relative to other parameters (AntropandVanEetvelde,
2000; Caeiroet al., 2012; de Leeuw, 2002; Hasse andLathrop, 2003).
In the context of the increasing human pressure on the envi-
ronment, especially in urban and metropolitan areas (P atroescu
et al., 2009), indicators represent instruments that can be used to
reduce information gaps (Uuemaa et al., 2013), analyze environ-
mental impacts (Ioj a et al., 2007), compare scenarios (Petrov et al.,
2011), communicate results to the public (Li et al., 2009) and aid in
the decision-making process (Jaeger et al., 2010).

Corresponding author. Tel.: +40 213103872.

E-mail addresses: iojacristian@yahoo.com(C.I. Ioj a),
mihairazvan.nita@g.unibuc.ro (M.R. Nit a), gabi vanau@yahoo.com
(G.O. Vn au), dianaandreea.onose@g.unibuc.ro, diana.onose@gmail.com
(D.A. Onose), athanasios gavrilidis22@yahoo.com(A.A. Gavrilidis).
Indicators of land-use changes in metropolitan areas have var-
ied over time, from informal methods to formal methods (i.e.,
cost-benet analysis or multi-criteria analysis) (Kamruzzaman and
Baker, 2013). The chosenindicators shouldallowintuitive interpre-
tation based on mathematical simplicity and modest data (Jaeger
et al., 2010) and should be quantiable, sensitive to changes in land
cover, temporallyandspatiallyexplicit andscalable(Larondelleand
Haase, 2013); they should be able to reect not only the analyzed
phenomena but also the particular needs and goals represented
by the diversity in a chosen study area (Shen et al., 2011). Interna-
tional institutions andorganizations have developedtheir ownsets
of indicators for monitoring of environmental aspects, e.g., the core
set indicators of the European Environment Agency (EEA, 2011)
or the sustainability indicators of the Organization for Economic
Co-operation and Development.
Urbanizationpromotes rapidsocial andeconomic development,
but at the same time, it leads to many environmental problems
(Li et al., 2009) that are becoming increasingly contentious. There-
fore, the need exists for instruments aimed at better understanding
and management of these issues (Asah et al., 2012). Frequently,
these environmental problems lead to the emergence of environ-
mental conicts (Kamruzzaman and Baker, 2013) between various
resource owners and users.
One of the most important environmental conict sources is
land-use changes (Sleeter et al., 2012) due to their effect on the
1470-160X/$ see front matter 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.ecolind.2013.09.029
Please cite this article in press as: Ioj a, C.I., et al., Using multi-criteria analysis for the identication of spatial land-use conicts in the
Bucharest Metropolitan Area. Ecol. Indicat. (2013), http://dx.doi.org/10.1016/j.ecolind.2013.09.029
ARTICLE IN PRESS
G Model
ECOIND-1686; No. of Pages 10
2 C.I. Ioj a et al. / Ecological Indicators xxx (2013) xxxxxx
natural energyandmaterial cycles of ecosystems (Tonget al., 2012),
the local climate conditions, biodiversity, and water resources (Liu
et al., 2011). Conicts over land uses represent a characteristic of
urban development, especially in the context of metropolitan areas
(Pacione, 2013), which are confronted with a high degree of pres-
sure fromvarious developers (P atroescu et al., 2009).
As a result, conicts emerge either between various land uses
(residential vs. industrial or agricultural) or between different eco-
nomic or social groups (residents, farmers, developers, etc.) (Darly
and Torre, 2013). Land-use conicts involve multiple parties who
choose terms that favor their respective positions (Shmueli, 2008)
and frequently exceed the community scale (Sze and Sovacool,
2013) to involve a range of societal actors, governmental bodies,
non-governmental organizations, and private interests (Saarikoski
et al., 2013).
The spatial component of land-use conicts has shown a high
manifestation in the last two decades in the countries of Eastern
Europe, where the shift from a centralized planning system to an
uncontrolled urban development (P atroescu et al., 2009) has facili-
tated the emergence of these types of conicts in various forms and
places.
Land-use conicts are increased by the fact that decisions on
projects and developments for different land uses are still made
based on incomplete information (Uuemaa et al., 2013); ideally,
these situations would require accurate and high-quality decision-
making to resolve these complex issues (Kamruzzaman and Baker,
2013), therefore illustrating the need for adequate indicators to
measure the relevant driving forces and impact factors (Mubareka
and Ehrlich, 2010).
Authorities and planners should nd a compromise to accom-
modate all of the land uses needed for a specic region (Helbron
et al., 2011) by controlling urbansprawl, preventing environmental
degradation and addressing the difculties posed by transport and
the provision of public services (Darly and Torre, 2013; Pacione,
2013; Koschke et al., 2012). At the same time, these authorities and
planners must continue to provide the minimumamount of build-
able land in response to demands for housing and services in rural
areas (Darly and Torre, 2013; P atroescu et al., 2011) and maintain
the balance between current and future needs (Sze and Sovacool,
2013; Koschke et al., 2012). The large volume of spatial data that
lends geographical expression to the economic, social, cultural and
ecological aspects of society (Jeong et al., 2013) requires analysis
methods that will integrate these aspects according to their impact
on the nal outcome (Saaty, 1990).
Multi-criteria analysis (MCA) is one method that can be used
in spatial planning to aid decision-makers in exploring and solving
multiple and complex problems (Jeong et al., 2013). The MCA rep-
resents decision-making analysis based on decision-science theory
that is able to quantitatively evaluate alternatives by taking into
account different perspectives and priorities to produce a common
output (Barfod et al., 2011; Convertino et al., 2013). Choosing the
MCA algorithm that best ts the research problem is challenging,
however, because the subjective scaling of the clear quantitative
numbers needed for pair-wise comparison of elements in the same
hierarchy may lead to losses in accuracy (Kowalski et al., 2009).
Various studies (Jeong et al., 2013; Koschke et al., 2012;
Convertino et al., 2013) have used MCA to assess the spatial distri-
bution of environmental problems, whereas others have combined
theMCA withcost-benet analysis (Barfodet al., 2011; Ghnemann
et al., 2012). All MCA analyses have a common pattern: dene the
alternatives to be ranked, identify the criteria that will inuence
the outcome, assign weights to the criteria and normalize them,
and determine the nal values (Convertino et al., 2013). Uncer-
tainties exist with respect to the impact levels and weights, but
in most cases, neither the time nor resources exist to followthe full
construction of the model (Ghnemann et al., 2012).
The assessment criteria are chosen to a greater or lesser extent
by how well they relate to the functional and livable dimensions
of land uses (Sze and Sovacool, 2013; Ghnemann et al., 2012).
The MCA is employed in land-use conicts analysis to classify the
risk levels of signicant impacts on the affected area (Helbron et al.,
2011) intheestablishment of public policycontexts (Kowalski et al.,
2009) or as applied to conict resolution(Kamruzzamanand Baker,
2013).
Spatial conicts have a high incidence in the Bucharest
Metropolitan Area due to this regions administrative hetero-
geneity (98 local administrative units from ve counties with a
reduced degree of law enforcement by local administrations) and
its dynamic characteristics (increasing proportion of constructed
surfaces, appearance of environmental degradation sources, modi-
cationof functional areas, reductionof greenareas, fragmentation
of properties) specic to the Eastern European countries (Zolin,
2007; Ioj a and Tudor, 2012).
The current study develops a methodology for identifying areas
prone tospatial land-use conicts byfocusingona post-communist
area in which phenomena such as the abandonment of agricul-
tural activities, increasing attractiveness for residential areas and
changes in the prole of industrial activities have a large incidence.
The aimof our study is to provide a tool for integrating land-use
conicts into strategies of territory planning at the metropolitan
level.
The main objectives of our paper are the following: (a) to
derive a set of indicators with spatial distributions that are use-
ful for analyzing the spatial land-use conicts in the Bucharest
MetropolitanArea and(b) tointegrate these indicators intoa multi-
criteria assessment that will allow (c) the spatial identication of
areas characterized by a high incidence of land-use conicts at the
metropolitan level.
2. Methodology
2.1. Study area
The Bucharest MetropolitanArea (Fig. 1) is situatedinthe south-
eastern region of Romania and contains 98 local administrative
units (cities and communes corresponding to EU NUTS 5) included
in ve counties (Nit a, 2012) with a total surface of 5080km
2
. We
removed Bucharest, the capital city of Romania from our study
area, because its characteristics and range of issues are of a com-
pletely different type (urban phenomena) than the remainder of
the metropolitan area (rural side mixed with urban functions). The
land use is mainly agricultural (76.8%), with the built-up environ-
ment representing only 4.65% of the metropolitan areas total land
(P atroescu et al., 2011).
The dominant relief is characterized by low elevation plains
and river oodplains (P atroescu et al., 2009). The main rivers are
tributaries of the Danube (which represents the southern limit of
the metropolitan area and the border to Bulgaria). Forests once
covered most of the study area, which is fragmented by lakes
and rivers (4.9% is covered by aquatic surfaces), but centuries of
human activities (mainly agricultural) have reduced the forests
to a small proportion (10.5% covered by forests) (P atroescu et al.,
2011). The study area has a total population of 571,315 inhabi-
tants, but numerous inhabitants of Bucharest have second homes
in this area that are inhabited for a smaller or larger period of time
(P atroescu et al., 2009). The average population density is 111.5
inhabitants/km
2
(Reyet al., 2007) withnotableterritorial variations
(range: 211010).
The changes in land properties after 1989 (shift from public
to private owned lands) has increased the pressure of land-use
changes according to individual purpose (Golusin and Ivanovic,
Please cite this article in press as: Ioj a, C.I., et al., Using multi-criteria analysis for the identication of spatial land-use conicts in the
Bucharest Metropolitan Area. Ecol. Indicat. (2013), http://dx.doi.org/10.1016/j.ecolind.2013.09.029
ARTICLE IN PRESS
G Model
ECOIND-1686; No. of Pages 10
C.I. Ioj a et al. / Ecological Indicators xxx (2013) xxxxxx 3
Fig. 1. Bucharest Metropolitan Area.
2009), regardless of the public interests, impacts on productivity
and the efciency of the surrounding areas (P atroescu et al., 2009).
These newareas have lost viability and suffered processes of aban-
donment or losses of productivity (Ioj a et al., 2011). Many land
uses have appeared in areas that were not able support them, and
external functions (by denition) have entered the built-up areas
of human settlements (Nit a, 2012).
We obtained statistical data for several indicators (accessibility
of the inhabitants to public infrastructure) fromthe National Insti-
tute of Statistics and City Halls of the Local Administrative Units
of the metropolitan area and extracted demographic data fromthe
1992 and 2011 national census. Spatial data for the dynamics of
residential areas were obtained from the analysis of topograph-
ical maps (at a scale of 1:25,000) published in 1977 and aerial
images from 2008 using the ArcGIS 9.3 software from ESRI. We
obtained data on the surface areas of agricultural land, built-up
areas and semi-natural surfaces that provide ecological services
from a combination of sources, including aerial images and the
localitys statistical charts.
2.2. Selection of indicators
We based our selection of indicators on their relevance to the
objectives of the study and the availability of data. Based on the
statistical criteria, we selected ten indicators (Table 1) in two cat-
egories: spatial indicators and urban development indicators. Our
indicators are intended to assess the potential for the appearance
of land-use conict and not to assess the current distribution of
land-use conicts.
Spatial indicators (X1X4) express the direct relationship
between the presence and absence of certain land uses (agri-
cultural, forestry, aquatic, built-up) at the metropolitan level.
Agricultural surfaces account for thehighest percent (76.8%) among
the land uses in the Bucharest Metropolitan Area and represent a
main direction for the development of built-up areas. Residential
surfaces represent the function with the highest vulnerability to
land-use conicts and have signicant importance for quality of
life. Semi-natural surfaces, in addition to providing much-needed
ecological services (air cleaning, climate regulation, water manage-
ment), also represent buffer areas between potential conicting
land-uses.
Urban development indicators (Y1Y6) are an indirect expres-
sion of the pressure on or needs of the population fromeach local
administrative unit for certain elements of public infrastructure.
Population tendencies and density are indicators of the magnitude
of the conicts and the potentially affected inhabitants, whereas
agingexpresses the dimensionof a sensible age groupthat is among
the rst affected by the negative externalities of land-use conicts.
Aging also expresses the resilience of the inhabitants of certain
localities to changes. By their presence, the indicators of public
infrastructure (water, gas, sewage) display a high potential for
urban development because areas that benet from these infras-
tructures are preferred by developers for their planned projects.
Scores of indicators withdifferent measurement scales (percent,
inhabitants/surface, rate of growth) cannot be compared directly
and must be standardized to a dimensionless value (Sudhakaran
et al., 2013). We standardized the data using Mathematical Pro-
gramming (Munier, 2004) via transformation to percent in a
process that considers the worst value equal to 1 (the minimum
data value for indicators X1, X3, Y4, Y5, Y6 and the maximumdata
value for indicators X2, X4, Y1, Y2, Y3) and the best value equal to
100. The standardized values were determined based on formulas
(1) and (2):
f (Z) =
1 100
max min
Z +
100 max 1 min
max min
(1)
if maximumrepresents the worst value, and
f (Z) =
100 1
max min
Z +
1 max 100 min
max min
(2)
Please cite this article in press as: Ioj a, C.I., et al., Using multi-criteria analysis for the identication of spatial land-use conicts in the
Bucharest Metropolitan Area. Ecol. Indicat. (2013), http://dx.doi.org/10.1016/j.ecolind.2013.09.029
ARTICLE IN PRESS
G Model
ECOIND-1686; No. of Pages 10
4 C.I. Ioj a et al. / Ecological Indicators xxx (2013) xxxxxx
Table 1
Selected indicators used in the analysis.
Categories of indicators Indicator code Indicator name Worst situation
expressed by
Explanation
Spatial indicators
X1 Agricultural land Minimumvalue Percent of agricultural areas fromthe total
surface in 2008
X2 Built-up areas Maximumvalue Percent of built-up areas fromthe total surface
in 2008
X3 Semi-natural surfaces
providing ecological
services
Minimumvalue Percent of forests and water bodies fromthe
total surface in 2008
X4 Residential area
tendency
Maximumvalue Rate of growth for the surface of residential
areas between 1977 and 2008
Indicators of urban
development
Y1 Population tendency Maximumvalue Rate of growth for the number of inhabitants
between 1992 and 2011
Y2 Population density Maximumvalue Density of population for each locality in 2011
Y3 Aging Maximumvalue Percent of elderly population (age above 60) in
2011
Y4 Sewage Minimumvalue Percent of inhabitants with access to public
sewage systems in 2008
Y5 Water Minimumvalue Percent of inhabitants with access to the public
water network in 2008
Y6 Gas Minimumvalue Percent of inhabitants with access to the public
gas distribution network in 2008
if minimum represents the worst value. In these equations,
max=the maximumvalue of a data set, min =the minimumvalue
of a data set, and Z =the value of indicators Xi and Yi for each local
administrative unit.
2.3. Multi-criteria analysis
The identication of areas with spatial land-use conicts was
conducted based on a multi-criteria analysis (MCA) (Beinat and
Nijkamp, 1998; Munier, 2004). The general outline of the method-
ology can be described by the following stages (Fig. 2), similar to
those of other studies (Jeong et al., 2013):
Fig. 2. Flowchart of the methodology for mapping the spatial distribution of land-
use conicts.
- Constructing a GIS database containing all of the spatial informa-
tion on the Bucharest Metropolitan Area and connecting it to the
data available for each local administrative unit.
- Selecting the representative indicators and transforming them
into elements of the multi-criteria analysis.
- Using a method of pair-wise comparisons to determine the rela-
tive importance of each criteria in the formof a criteria weight.
- Aggregating the data with the criteria weights and obtaining the
scores of each area.
- Spatially representing the distribution of the scores obtained for
each local administrative unit.
For theanalysis, we considered10maincriteria as denedabove
and 98 cases corresponding to the local administrative units from
the Bucharest Metropolitan Area. We selected the criteria for con-
sistency, transparency and ease of understanding as required in a
MCA analysis (Ghnemann et al., 2012).
Using the Analytical Hierarchy Process (Saaty, 1990), we estab-
lished the weights of each of the 10 criteria (g
i
) by considering the
sumof weights equal to1for the purpose of aggregatingthemintoa
singleoverall valueandproducingaranking. Theweights reect the
importance of each criteria and were assigned based on a pair-wise
comparisons (Table 2), although the weight value dened for each
one canstill be considered arbitrary (Kong et al., 2012; Ghnemann
et al., 2012). We established the weights of each criterion based on
an expert opinion applied by the authors of the present study.
By multiplying the weights of each criterion with the standard-
ized values, we obtained ten partial scores. We obtained the nal
score for each local administrative unit by aggregating the partial
scores with the aid of formula (3). The overall score is represented
by the weighted average of its scores on all of the criteria.
S =
10

i=1
(g
i
f (Z
i
)) (3)
As a result, the partial and total scores can vary between 1 and
100. In the nal stage, we represented the score for each criteria
and the total scores on maps of the Bucharest Metropolitan Areas
using a division of classes (Adler et al., 2010; Dixon et al., 2009)
to express differences between the magnitudes of spatial land-use
conicts.
Please cite this article in press as: Ioj a, C.I., et al., Using multi-criteria analysis for the identication of spatial land-use conicts in the
Bucharest Metropolitan Area. Ecol. Indicat. (2013), http://dx.doi.org/10.1016/j.ecolind.2013.09.029
ARTICLE IN PRESS
G Model
ECOIND-1686; No. of Pages 10
C.I. Ioj a et al. / Ecological Indicators xxx (2013) xxxxxx 5
Table 2
Relative importance of pair-wise comparisons and numerical rates after Jeong et al. (2013).
More important intensity Denition Less important intensity
1 Equal importance or preference 1
2 More or less equal to moderate importance or preference 1/2
3 More or less moderate importance or preference 1/3
4 More or less moderate to strong importance or preference 1/4
5 More or less strong importance or preference 1/5
6 More or less strong to very strong importance or preference 1/6
7 More or less very strong importance or preference 1/7
8 More or less very to extremely strong importance or preference 1/8
9 More or less extreme importance or preference 1/9
Table 3
Statistical parameters of the rawdata and intermediate standardized values.
Indicator Rawdata Intermediate
standardized values
Average Max Min St. dev. Average St. dev.
X1 76.42 93.60 34.71 13.23 29.88 22.24
X2 6.10 42.39 1.28 6.31 12.62 15.19
X3 0.11 0.43 0.00 0.10 75.05 22.14
X4 55.72 433.50 50.45 50.65 22.72 10.36
Y1 10.11 151.47 64.09 24.53 25.79 11.27
Y2 150.59 1009.94 21.38 155.33 13.94 15.56
Y3 24.73 41.65 13.22 6.79 41.08 23.65
Y6 16.83 74.50 0.08 17.04 77.72 22.67
Y7 7.03 105.00 0.00 13.61 93.37 12.83
Y8 13.81 129.90 0.00 24.78 89.47 18.89
3. Results
The differences betweenthe values of the rawdata for eachindi-
cator were signicant (Table 3), with certain of them showing a
spread of values (Y2 range of 21.381009.94, 155.33) while oth-
ers presented a more concentrated pattern (X3 range of 00.43,
0.10). Using the standardization method, we normalized the val-
ues to a common level (all standard deviations of the intermediate
values ranged in the 10.3623.65 interval).
We obtained the weights for each criterion from the pair-
wise comparisons conducted by each author of the present study
(Table 4). It is notable that although the number of spatial indi-
cators is smaller than that of urban development indicators, their
total weight is larger (0.53). Indicators with the highest weights
tend to represent residential surfaces (X4 0.19), the tendency of
population (Y1 0.16) and the total surface of built up areas (X2
0.15). The lowest weights were assigned to indicators representing
the infrastructure development (Y4Y6). The sensitivity analysis
in the expert weighting process has been assessed by constructing
scatter plots of the output variables against a randomly selected
input variable. The scatter plots revealed reasonable results across
the analysis, and no signicant differences were recorded.
We aggregated the weights of each indicator with the inter-
mediated standardized values, thus obtaining the scores for each
Table 4
Weights of the selected indicators.
Indicator Weight Indicator Weight
X1 0.13 Y1 0.16
X2 0.15 Y2 0.14
X3 0.06 Y3 0.05
X4 0.19 Y4 0.04
Total weight of
spatial indicators
0.53 Y5 0.04
Y6 0.04
Total weight of urban
development
indicators
0.47
criterion and for each local administrative unit fromthe Bucharest
Metropolitan Area.
The spatial indicators (Fig. 3) showa relatively similar distribu-
tion in the metropolitan area, with certain areas of concentrations
according to the characteristics of each indicator. Indicator X1
agricultural land displays heterogeneous values in the metropoli-
tan area, with the lowest values found in the eastern part. Indicator
X2 built-up areas shows the highest values concentrated in
localities situated in close proximity to Bucharest, and interme-
diate values in certain regional poles of development. Indicator
X3 semi-natural surfaces providing ecological services shows a
reversed homogenous distribution, with the majority of values dis-
tributed to the worst-case scenarios due to the lack of both forests
and water bodies and with low values corresponding to locali-
ties from the north and western regions characterized by large
forest areas and water bodies. Indicator X4 residential areas ten-
dency presents a dominance of two mediumclasses of values, both
characterizing increases in the surface of residential areas in the
respective localities.
The total aggregation of spatial indicators (Fig. 4) shows locali-
ties with a high probability of land-use conicts incidence situated
particularly in proximity to Bucharest and in local administrative
units that have a legal urban status. The scores range between 7.19
(lowest values in the eastern portion of the metropolitan area) and
52.48(Voluntari, whichneighbors Bucharest inthe northeast), with
four other local administrative units recording scores between 25
and 50 (all small cities situated in proximity to Bucharest).
Theurbandevelopment indicators (Fig. 5) varycorrespondingly:
Indicator Y1 population tendency and Indicator Y2 popula-
tion density showa small number of localities with high values for
the worst situation (characterizing localities that suffered impor-
tant developments of residential areas in the past 20 years) while
the others recorded medium values. Indicator Y3 aging shows
a concentration of high values in the eastern portion of the
metropolitan area, and smaller values are recorded in the cen-
ter. All of the infrastructure accessibility indicators, i.e., Indicator
Y4 sewage, Indicator Y5 water and Indicator Y6 gas, show
a reversed homogenous distribution, with the majority of values
distributed to the worst situation case, and only a small number of
local administrative units present the best scenario.
The total aggregationof urbandevelopment indicators (Fig. 6) in
the Bucharest Metropolitan Area reveals a scattered distribution of
a small number of local administrative units characterized by the
existence of spatial land-use conicts. The scores range between
10.00 (Joita) and30.01 (Chitila), withthe small values locatedinthe
periphery of the metropolitan area, and high values characterizing
localities that represent poles of development.
The variability of the total score for the spatial indicators is
greater than that of the urban development indicators (standard
error 0.64>0.30). The total aggregation of indicators characterizing
land-useconicts intheBucharest MetropolitanArea arepresented
in Fig. 7 with scores ranging between 17.53 (Gurbanesti in the
eastern portion of the metropolitan area) and 76.01 (Voluntari in
Please cite this article in press as: Ioj a, C.I., et al., Using multi-criteria analysis for the identication of spatial land-use conicts in the
Bucharest Metropolitan Area. Ecol. Indicat. (2013), http://dx.doi.org/10.1016/j.ecolind.2013.09.029
ARTICLE IN PRESS
G Model
ECOIND-1686; No. of Pages 10
6 C.I. Ioj a et al. / Ecological Indicators xxx (2013) xxxxxx
Fig. 3. Scores of the spatial indicators in the Bucharest Metropolitan Area.
Fig. 4. Aggregated scores of spatial indicators in the Bucharest Metropolitan Area.
Please cite this article in press as: Ioj a, C.I., et al., Using multi-criteria analysis for the identication of spatial land-use conicts in the
Bucharest Metropolitan Area. Ecol. Indicat. (2013), http://dx.doi.org/10.1016/j.ecolind.2013.09.029
ARTICLE IN PRESS
G Model
ECOIND-1686; No. of Pages 10
C.I. Ioj a et al. / Ecological Indicators xxx (2013) xxxxxx 7
Fig. 5. Scores of the urban development indicators in the Bucharest Metropolitan Area.
the north-eastern area of Bucharest), and small values character-
izing areas with a high rural degree. One local administrative unit
exhibits a value greater 75 (Voluntari) and three present a value
above 50(Bragadiru, Chiajna andChitila); all are small cities located
in proximity to Bucharest.
4. Discussion
The distribution of scores for the spatial indicators in
the Bucharest Metropolitan Area can be explained using the
characteristics of the environment and socio-economic aspects.
Agricultural lands (X1) show heterogeneous scores that directly
related to the total surface of the local administrative areas and
their position in the metropolitan area, with peripheral and large
localities showing an availability of agricultural lands for further
development (Golusin and Ivanovic, 2009) and thus mitigating the
spatial conict role of other land uses. Built-up areas (X2) and the
tendency toward residential areas (X4) have a direct connection
to the position of the local administrative unit in the metropolitan
area. High values characterize areas in close proximity to Bucharest
Please cite this article in press as: Ioj a, C.I., et al., Using multi-criteria analysis for the identication of spatial land-use conicts in the
Bucharest Metropolitan Area. Ecol. Indicat. (2013), http://dx.doi.org/10.1016/j.ecolind.2013.09.029
ARTICLE IN PRESS
G Model
ECOIND-1686; No. of Pages 10
8 C.I. Ioj a et al. / Ecological Indicators xxx (2013) xxxxxx
Fig. 6. Multi-criteria assessment of urban development indicators in the Bucharest Metropolitan Area.
and the northern area, which is a pole of growth for residential
development (Nit a, 2012) due to the large proportion of semi-
natural surfaces providing ecological services (as expressed by
indicator X3).
Population tendency (Y1) shows an evolution similar to that of
the residential areas (X4), whereas population density (Y2) is rel-
evant for local administrative units with small surfaces (generally
urban) that are situated in proximity to Bucharest. Aging (Y3) and
indicators of the accessibility of population to public infrastructure
(i.e., sewage systems (Y4), drinking water (Y5) andnatural gas (Y6))
express the rural degree of the metropolitan area, especially in its
eastern and peripheral areas. The lack of such basic infrastructure
is a situation that characterizes a large proportion of rural areas in
Romania (Rey et al., 2007).
The total aggregation of spatial indicators allows the iden-
tication of spatial land-use conicts in two categories of
local administrative units: urban areas situated in proximity to
Bucharest (Otopeni, Buftea, Chiajna, Voluntari) and those charac-
terized by elements of attractiveness for developers (Snagov). The
total aggregationof urbandevelopment indicators follows thesame
Fig. 7. Multi-criteria assessment of total indicators of spatial land-use conicts in the Bucharest Metropolitan Area.
Please cite this article in press as: Ioj a, C.I., et al., Using multi-criteria analysis for the identication of spatial land-use conicts in the
Bucharest Metropolitan Area. Ecol. Indicat. (2013), http://dx.doi.org/10.1016/j.ecolind.2013.09.029
ARTICLE IN PRESS
G Model
ECOIND-1686; No. of Pages 10
C.I. Ioj a et al. / Ecological Indicators xxx (2013) xxxxxx 9
patternas thespatial indicators but at asmaller scale. Thedifference
between the standard error of the spatial indicators (0.64) and the
urbandevelopment indicators (0.30) couldindicate the uncertainty
of the comparison and might require a renement of the indicators
included in the analysis. The total scores reveal local administra-
tive units characterized by a critical (Voluntari) or high (Bragadiru,
Chitila, Chiajna) incidence of spatial-land use conicts.
The identication of spatial land-use conicts allows managers
and planners to promote sustainable use of lands at both the local
and the regional levels (Li et al., 2009). When analyzing the location
of newlarge-scale projects (public or private,) the decision-makers
must be aware of where unbalanced situations already exist due to
conicting land-uses and where there are reserves of space with-
out such problems. In addition, because conicting land uses often
diminish the quality of life for inhabitants, any reduction in their
distribution can help correct the uneven distribution of environ-
mental conditions between poor and afuent residents (Romero
et al., 2012) and between the resident population and real-estate
developers duetodifferences inpurchasingpower betweenthetwo
categories. However, this process does not represent an absolute
solution because traditional planning should shift to an appropri-
atemanagement of landthat canintegratebothpublic participation
and private interest (Cerreta et al., 2012).
The use of MCA in analyzing spatial land-use conicts in the
Bucharest Metropolitan Area has shown that even if every land use
conict has its own particularities (Sze and Sovacool, 2013), our
criteria can provide instruments (Jeong et al., 2013) to land plan-
ners and administration for decision-making processes aimed at
coordinated urban and rural development between metropolitan
urban and rural areas (Shen et al., 2012). The simplicity and speed
of calculation of the MCA indicates that the method is effective in
response to rapid land-use changes (Uuemaa et al., 2013).
The method identies spatial land-use conicts at the
metropolitan level and the specic LAU involved or at the statis-
tical level for which the administration can provide the necessary
data. Additional in-depth analysis could be performed, provided
that each LAU can offer more spatially detailed data.
It remains to be proved whether this method is convenient for
the comparisonof different regions andphenomena because it gen-
erally involves area-proportionate additive measures (Jaeger et al.,
2010). Further attention should be focused on the structure of the
weight sets (Ghnemann et al., 2012) because they could have a
major impact on the nal outcome of the study. Additionally, the
method does not allowa clear differentiation of the nature of con-
icts (Darly and Torre, 2013); in most cases, a variation of land-use
conicts would be expected for such a large area.
5. Conclusions
The presented sets of indicators of spatial land-use conicts are
useful instruments for improving planning policies and strategies,
increasing the social awareness of the population and providing
direct economic projections for developers. Their integration into
a multi-criteria analysis allowed us to map the areas character-
ized by a high incidence of land-use conicts, which can be further
used for sustainable development of the metropolitan area. The
use of multi-criteria analysis in identifying the areas of land-use
conicts can be of real help to policy makers and authorities in the
decision-making process, especially in Eastern European countries,
because this methodology can represent a starting point for these
assessments.
The main limitations of the methods lie in the availability of
data required for the construction of each criterion. In addition,
no principles or guidelines are currently available to select the
most representative criteria for the characterization of land-use
conicts; therefore, the authors chose themaccording to their own
expertise and the availability of data. Improvements should be pur-
sued via organizing expert-opinion sessions with specialists from
different elds of study. Additionally, further research is necessary
to analyze whether such a multi-criteria assessment can be applied
based on other criteria or on other study areas.
Acknowledgements
This work was supported by a grant funded by the Min-
istry of Education and Research through the Young Research
Teams project PN-II-RU-TE-2011-3-0285. Modeling of environ-
mental impact induced by typologies of incompatible land uses in
human settlements.
References
Adler, S., Hubener, T., Drebler, M., Lotter, A., Andreson, N.J., 2010. Acomparisonof rel-
ative abundance versus class data indiatom-basedquantitative reconstructions.
Journal of Environmental Management 91, 13801388.
Antrop, M., Van Eetvelde, V., 2000. Holistic aspects of suburban landscapes: visual
image interpretation and landscape metrics. Landscape and Urban Planning 50,
4358.
Asah, S.T., Bengston, D.N., Wendt, K., Nelson, K.C., 2012. Diagnostic reframing of
intractable environmental problems: case of a contestedmultiparty public land-
use conict. Journal of Environment Management 108, 108119.
Barfod, M.B., Salling, K.B., Leleur, S., 2011. Composite decision support by combining
cost-benet and multi-criteria decision analysis. Decision Support Systems 51,
167175.
Beinat, E., Nijkamp, P., 1998. Multicriteria Analysis for Land-Use Management. In:
Environment & Management. Kluver Academic Publishers, Dordrecht.
Caeiro, S., Ramos, T.B., Huisingh, D., 2012. Procedures and criteria to develop and
evaluate household sustainable consumption indicators. Journal of Cleaner Pro-
duction 27, 7291.
Cerreta, M., Panaro, S., Cannatella, D., 2012. Multidimensional spatial decision-
makingprocess: local sharedvalues inaction. In: Murgante, B., Gervasi, O., Misra,
S., Nedjah, N., Rocha, A.C., Taniar, D., Apduhan, B. (Eds.), Computational Science
and Its Applications ICCSA 2012. Springer, Berlin, Heidelberg, pp. 5470.
Convertino, M., Baker, K.M., Vogel, J.T., Lu, C., Suedel, B., Linkov, I., 2013. Multi-criteria
decision analysis to select metrics for design and monitoring of sustainable
ecosystemrestorations. Ecological Indicators 26, 7686.
Darly, S., Torre, A., 2013. Conicts over farmland uses and the dynamics of agri-
urban localities in the Greater Paris Region: an empirical analysis based on
daily regional press and eld interviews. Land Use Policy 33, 9099.
de Leeuw, F., 2002. A set of emission indicators for long-range transboundary air
pollution. Environmental Science and Policy 5, 135145.
Dixon, S.J., Heinrich, N., Holmboe, M., Schaefer, L., Reed, R.R., Trevejo, J., Brereton,
R.G., 2009. Application of classication methods when group sizes are unequal
by incorporationof prior probabilities tothree commonapproaches: application
to simulations and mouse urinary chemosignals. Chemometrics and Intelligent
Laboratory Systems 99, 111120.
EEA, 2010. The European Environment State and Outlook 2010. European Envi-
ronmental Agency, Copenhaga.
EEA, 2011. Annual Report 2010 and Environmental Statement. Environmental Pro-
tection Agency, Bruxelles.
Golusin, M., Ivanovic, O.M., 2009. Denition, characteristics and state of indica-
tors of sustainable development incoutries of SoutheasternEurope. Agriculture,
Ecosystems and Environment 130, 6774.
Ghnemann, A., Laird, J.J., Pearman, A.D., 2012. Combining cost-benet and
multi-criteria analysis to prioritise a national road infrastructure programme.
Transport Policy 23, 1524.
Hasse, J.E., Lathrop, R.G., 2003. Land resource impact indicators of urban sprawl.
Applied Geography 23, 159175.
Helbron, H., Schmidt, M., Glasson, J., Downes, N., 2011. Indicators for strategic envi-
ronmental assessment in regional land use planning to assess conicts with
adaptation to global climate change. Ecological Indicators 11, 9095.
Ioj a, C.I., Tudor, C.A., 2012. Temporal analysis of incompatible land-use and land-
cover: the proximity between residential areas and gas stations in Bucharest
Suburban Area. Procedia Environmental Sciences 14, 4958.
Ioj a, I.C., Onose, D.A., Nit a, M.R., Vn au, G.O., P atroescu, M., Gavrilidis, A.A., Saghin, I.,
Zarea, R., 2011. The conversion of agricultural lands into built surfaces in Roma-
nia. In: Niola, V., Kala, T., Popescu, C. (Eds.), 2nd International conference on
Urban Sustainability, Cultural Sustainability, Green Development, Green Struc-
tures and Clean Cars (USCUDAR11),. WSEAS, Praga.
Ioj a, I.C., P atroescu, M., Matache, M.L., Pavelescu, G., Damian, R., 2007. Environmental
impact assessment of the vegetable cultivations using the Pimentel-Euleistein
model Case study Arges Lower Watershed. In: Ples u, V., S erban, P. (Eds.), Com-
puter Aided Process Engineering. Elsevier Press, pp. 12471252.
Jaeger, J.A.G., Bertiller, R., Schwick, C., Cavens, D., Kienast, F., 2010. Urbanpermeation
of landscapes and sprawl per capita: newmeasures of urban sprawl. Ecological
Indicators 10, 427441.
Please cite this article in press as: Ioj a, C.I., et al., Using multi-criteria analysis for the identication of spatial land-use conicts in the
Bucharest Metropolitan Area. Ecol. Indicat. (2013), http://dx.doi.org/10.1016/j.ecolind.2013.09.029
ARTICLE IN PRESS
G Model
ECOIND-1686; No. of Pages 10
10 C.I. Ioj a et al. / Ecological Indicators xxx (2013) xxxxxx
Jeong, J.S., Garca-Moruno, L., Hernndez-Blanco, J., 2013. A site planning approach
for rural buildings into a landscape using a spatial multi-criteria decision anal-
ysis methodology. Land Use Policy 32, 108118.
Kamruzzaman, M., Baker, D., 2013. Will the application of spatial multi criteria eval-
uation technique enhance the quality of decision-making to resolve boundary
conicts in the Philippines? Land Use Policy 34, 1126.
Kong, F., Yin, H., Nakagoshi, N., James, P., 2012. Simulating urban growth processes
incorporating a potential model with spatial metrics. Ecological Indicators 20,
8291.
Koschke, L., Frst, C., Frank, S., Makeschin, F., 2012. A multi-criteria approach for
an integrated land-cover-based assessment of ecosystem services provision to
support landscape planning. Ecological Indicators 21, 5466.
Kowalski, K., Stagl, S., Madlener, R., Omann, I., 2009. Sustainable energy futures:
methodological challenges in combining scenarios and participatory multi-
criteria analysis. European Journal of Operational Research 197, 10631074.
Kurtz, J.C., Jackson, L.E., Fisher, W.S., 2001. Strategies for evaluating indicators based
on guidelines from Environmental Protection Agencys Ofce of Research and
Development. Ecological Indicators 1, 4960.
Larondelle, N., Haase, D., 2013. Urban ecosystem services assessment along a
ruralurban gradient: a cross-analysis of European cities. Ecological Indicators
29, 179190.
Li, F., Liu, X., Hu, D., Wang, R., Yang, W., Li, D., Zhao, D., 2009. Measurement indicators
and anevaluationapproachfor assessing urbansustainable development: a case
study for Chinas Jining City. Landscape and Urban Planning 90, 134142.
Liu, Y., Yao, C., Wang, G., Bao, S., 2011. An integrated sustainable development
approach to modeling the eco-environmental effects from urbanization. Eco-
logical Indicators 11, 15991608.
Mubareka, S., Ehrlich, D., 2010. Identifying and modelling environmental indicators
for assessing populationvulnerability to conict using ground and satellite data.
Ecological Indicators 10, 493503.
Munier, N., 2004. Multicriteria Environmental Assessment. APractical Guide. Kluver
Academic Publishers, Dordrecht.
Nit a, M.R., 2012. Mapping favorability for residential development, case study:
Bucharest Metropolitan Area. Procedia Environmental Sciences 14, 5970.
Pacione, M., 2013. Private prot, public interest and land use planninga conict
interpretation of residential development pressure in Glasgows ruralurban
fringe. Land Use Policy 32, 6177.
Passeri, N., Borucke, M., Blasi, E., Franco, S., Lazarus, E., 2013. The inuence of farming
technique on cropland: a new approach for the ecological footprint. Ecological
Indicators 29, 15.
P atroescu, M., Nit a, M.R., Ioj a, I.C., Vn au, G.O., 2009. New residential areas in
Bucharest Metropolitan Area location, type and characteristics. In: In 14th
International Conference on Urban Planning, Regional Development and Infor-
mation Society Cities 3.0 Strategies, Concepts and Technologies for Planning
the Urban Future REAL CORP, Sitges, pp. 767772.
P atroescu, M., Vn au, G.O., Nit a, M.R., Ioj a, I.C., Ioj a, A.D., 2011. Landuse change inthe
Bucharest Metropolitan Area and its impacts on the quality of the environment
in residential developments, forumgeograc. Studii s i cercet ari de geograe s i
protect ia mediului 10, 5264.
Petrov, L.O., Shahumyan, H., Williams, B., Convery, S., 2011. Scenarios and indicators
supporting urban regional planning. Procedia Social and Behavioral Sciences
21, 243252.
Rey, V., Groza, O., Ianos , I., P atroescu, M., 2007. Atlas de la Roumanie. Reclus, Mont-
pelier, Paris.
Romero, H., Vsquez, A., Fuentes, C., Salgado, M., Schmidt, A., Banzhaf, E., 2012.
Assessing urban environmental segregation (UES). The case of Santiago de Chile.
Ecological Indicators 23, 7687.
Saarikoski, H., Raitio, K., Barry, J., 2013. Understanding successful conict reso-
lution: policy regime changes and new interactive arenas in the Great Bear
Rainforest. Land Use Policy 32, 271280.
Saaty, T.L., 1990. Multicriteria decision making the analytic hierarchy process.
Mc-GrawHill, NewYork.
Shen, L.-Y., Jorge Ochoa, J., Shah, M.N., Zhang, X., 2011. The application of urban
sustainability indicators a comparison between various practices. Habitat
International 35, 1729.
Shen, L., Jiang, S., Yuan, H., 2012. Critical indicators for assessing the contribution
of infrastructure projects to coordinated urbanrural development in China.
Habitat International 36, 237246.
Shmueli, D.F., 2008. Framing in geographical analysis of environmental
conicts: theory, methodology and three case studies. Geoforum 39,
20482061.
Sleeter, B.M., Sohl, T.L., Bouchard, M.A., Reker, R.R., Soulard, C.E., Acevedo, W., Grif-
th, G.E., Sleeter, R.R., Auch, R.F., Sayler, K.L., Prisley, S., Zhu, Z., 2012. Scenarios of
land use and land cover change in the conterminous United States: utilizing the
special report onemissionscenarios at ecoregional scales. Global Environmental
Change 22, 896914.
Sudhakaran, S., Lattemann, S., Amy, G.L., 2013. Appropriate drinking water treat-
ment processes for organic micropollutants removal based on experimental and
model studies a multi-criteria analysis study. Science of the Total Environment
442, 478488.
Sze, M.N.M., Sovacool, B.K., 2013. Of fast lanes, ora, and foreign workers: managing
land use conicts in Singapore. Land Use Policy 30, 167176.
Tong, S.T.Y., Sun, Y., Ranatunga, T., He, J., Yang, Y.J., 2012. Predictingplausible impacts
of sets of climate and land use change scenarios on water resources. Applied
Geography 32, 477489.
Uuemaa, E., Mander, ., Marja, R., 2013. Trends intheuseof landscapespatial metrics
as landscape indicators: a review. Ecological Indicators 28, 100106.
Zolin, M.B., 2007. The Extended Metropolitan Area in a new EU member state:
implications for a rural development approach. Transition Studies Review 14,
565573.