Ecological Indicators 73 (2017) 597–611

Contents lists available at ScienceDirect

Ecological Indicators
journal homepage: www.elsevier.com/locate/ecolind

Original Articles

Modeling and measuring urban sustainability in multi-criteria based

systems — A challenging issue
Fazia Ali-Toudert ∗ , Limei Ji
TU Dortmund University, Faculty of Architecture and Civil Engineering, Chair for Energy Efficient Buildings EEB, Wilhelm-Dilthey-Str. 1, 44227 Dortmund,
Germany

a r t i c l e i n f o a b s t r a c t

Article history: This paper addresses how urban sustainability is modeled and the ways criteria-based systems deal with
Received 15 August 2016 its measurability for an effective and reliable assessment. Twelve sustainability models are reviewed and a
Accepted 28 September 2016 subset is briefly presented. More importantly, this research work investigates five national rating systems
of sustainable urban development compared with the newly developed CAMSUD system. The comparison
Keywords: focuses on the systems’ structure, categorization, technical content and measurability. The main findings
Sustainability model
about the selected national rating systems thoroughly discussed in the paper are: (i) They all have a tree-
Indicator quantification
like structure, (ii) their conceptualization and categorization follow three or four sustainability pillars
Criteria interaction
Weighting
models, sustainability topics or spatial scale; (iii) they use either planning-oriented or performance-
Rating system oriented weighting approaches; (iv) the criteria are defined as sustainability goals, action measures or
CAMSUD assignments to be fulfilled; (v) the sustainability items can hardly be juxtaposed since they are differently
handled, (vi) overlapping criteria might occur, (vii) similar criteria can be categorized under different
categories and this affects the emphasis put on these categories, (viii) all criteria are independently rated
with no consideration of mutual interrelationships. In an attempt to solve some of these weaknesses, the
newly developed CAMSUD system is introduced as alternative and relies on the following: (i) the system
structure is considered as a network, (ii) the conceptualization and categorization is based on spatial
scaling as well as on sustainability topics and pillars, (iii) many criteria are directly planning-relevant
(23 of 40), (iv) the criteria are defined as sustainability goals rather than action measures and (v) the
quantification of criteria is planned as to account for mutual interactions.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction the theoretical modeling as well as the practical measurability of

The concept of sustainability is differently perceived depending
on the view perspective given by the background discipline of the
observer. Disciplines with a focus on social issues (e.g. Fücks, 2013;
Rückert-John, 2013; Lorenz, 2014), on technology (e.g. Cerone et al.,
2014), on energy (e.g. El Bassam et al., 2013; Sayigh 2014) or on
policy (e.g. Brannstrom and Vadjunec, 2013), just to name a few of
them, address the theme by highlighting different contents.
Indeed, the sustainability goals at urban scale are newly
addressed, in comparison to the building scale for which relatively
mature rating systems developed over decades exist. First sus-
tainability rating systems for urban areas are available but their
endorsement is still underway. This study reports on an analysis of
some of them and reveals that methodological aspects related to
∗ Corresponding author.
E-mail address: fazia.alitoudert@tu-dortmund.de (F. Ali-Toudert).
urban sustainability are still unsolved. It points out the deficiencies,
discusses the matter in detail and seeks to find ways in addressing
the problem. The paper includes: (1) a report on the theoretical
discourse prevailing in the literature on the comprehension of sus-
tainability with an effort to highlight the most operational concepts,
and (2) a transverse analysis and discussion of the measurability of
urban sustainability criteria of selected rating systems as well as in a
new assessment system developed by the authors (see Ali-Toudert
et al., 2016).
1.1. Models of sustainability
Many attempts, aimed at wholly comprehending the concept of
sustainability, have yielded various theoretical models. Fig. 1 gives
a schematic summary of some recurrent models from the litera-
ture. It refers to: [1] Spindler (2011); [2] Costanza and Wainger
(1991), Neumayer (1999); [3] Kleine (2009), Spindler (2011); [4]
Bott and Grassl (2013); [5] Munasinghe (1993); [6] Augenbroe and

http://dx.doi.org/10.1016/j.ecolind.2016.09.046
1470-160X/© 2016 Elsevier Ltd. All rights reserved.
598 F. Ali-Toudert, L. Ji / Ecological Indicators 73 (2017) 597–611

Fig. 1. Review of some existing sustainability models from the literature.

Pearce (2010); [7] Spindler, 2011; [8] Stoke (2008), Thwink, (2016); Gillette (2007), Higgins (2015) and [12] Curwell et al., 2007. Fig.
[9] Spangenberg (1997), Valentin and Spangenberg (2000), Lozano 1 shows at first glance some consensus but, when looked at in
(2008); [10] NZMCH (2006); [11] Hawkes (2001), Duxbury and detail, also exhibits clear differences. Basically, sustainable devel-
 F. Ali-Toudert, L. Ji / Ecological Indicators 73 (2017) 597–611
opment is represented as a 3-pole or 3-pillar model (see [1]–[6]
in Fig. 1), either explicitly or tacitly bringing together the envi-
ronment, economy and society dimensions. However, there are
different views on the relationship exhibited between these three
topics in question. In Model [1] ecology, economy and social mat-
ters are considered relatively independent to each other. The basis
for sustainability in this model is the consideration of the sub-
themes as stand-alone with little or no interactions whatsoever
between the elements. This viewpoint is widely disputed because
of the lack of consideration of interdependences between the ele-
ments (see Spindler, 2011). Indeed, the value of the whole model is
more than the summation of individual elements. Model [2] high-
lights a hierarchical difference where economy is embedded in
social activities, both of which are enclosed in the environmen-
tal element. The hierarchy transits from economy set at the lower
end to the all-encompassing environment (Costanza and Wainger,
1991; Neumayer, 1999). The most widespread scheme represents
the elements in an interacting circuit diagram as shown in model
[3]. Here, the elements are considered to interact with each other
and the relationship between them leads to shared sustainability
goals expressed by viability, bearableness and equity (see Kleine,
2009; Spindler, 2011). Although similar in form to [3], model [4]
is primarily different by considering efficiency, sufficiency and
consistency as performance goals in lieu of the aforementioned
dimensions in [3]; meanwhile the 3 basic dimensions of ecology,
society and economy remain tacit. Model [5] also focuses on and
depicts interactions between the three main topics of sustainability
(Munasinghe, 1993). Model [6] argues a shift from an old paradigm,
based on a balanced consideration of time, cost and quality to a
new paradigm focusing on human satisfaction, reduced consump-
tion and minimal environmental impact. More models introduce
other dimensions as additional pillars, mostly 4-pole models as
shown in Fig. 1 ([7]–[12]). Typically, culture and governance are
the additional dimensions disputed as affecting sustainability in
some manner. When culture is added, it is seen as an extended
understanding of the human dimension represented so far by the
society. This is the case in model [7], [10] and [11]. Governance, in
the other hand, adds the managerial and decision making compo-
nent as well as introducing policy and institutional responsibility
as a factor required in successfully implementing sustainability as
in model [8], [9] and [12]. Worthy of note is the diversity of terms
used throughout the literature in depicting the concept of sustain-
ability. Whereas some can be considered as interchangeable such as
environment and ecology, or society and community, others show
slight deviations in meaning and thus emerge as “neighbor terms”
used to highlight specific facets of one subject such as well-being
or justice within the social dimension.
1.2. Urban sustainability models
The city is a system which strongly interacts with its surround-
ing environment; hence implementing urban sustainability would
involve minimizing the burden on ecosystems while improving the
quality of life (Alberti, 1996). Urban sustainability is a challenge
given the complexity and diversity of this task, which additionally
to ecology involves the combination of economic and socio-cultural
considerations in enhancing environmental quality, economic effi-
ciency and human well-being via the proposed action measures.
This complexity arises from the high concentration of human
activities of diverse origin, alongside their operation and mutual
interactions which have to be considered. The spatial delimitation
of the urban area and the rest field provides the ecological load or
footprint in material consumption as an approach for urban sus-
tainability modeling (e.g. Rees, 1992; Wackernagel et al., 1999;
Moran et al., 2008). Another view focuses on the self-organizing
complexity of the city with people acting as “system engineers”
599
in the city system (Wu, 2008). “Policy and management” were
sometimes understood as a macro dimension of sustainability (e.g.
Drakakis-Smith, 1995). From the aspect of organizational process
of urban development, the DPSIR analysis (DPSIR: driving, force,
pressure, state, impact, response) was developed by the Organisa-
tion for Economic Cooperation and Development (Dizdaroglu and
Yigitcanlar, 2014). With this model, the solutions for sustainabil-
ity issues are addressed with the help of identification of problems
following a DPSIR-procedure. Another opinion about urban sus-
tainability is enhancing the city’s resilience which reduces the
vulnerability to natural and human hazards (Collier et al., 2013).
One further concept is the criteria-based approach which has the
advantage of being operational as it focuses on the factual expres-
sion of sustainability ”on-site” (see e.g. Mori and Christodoulou,
2012; Hiremath et al., 2013; Sharifi and Murayama, 2013; Reith
and Orova, 2015). Systems relying on this approach are the object
of the investigation hereafter.
2. National rating systems
The basis of this research is an extensive literature review and
extensive comparison of five well-known national rating systems
for sustainable urban development. (CASBEE UD, LEED ND, BREEAM
COM, DGNB NSQ and Green Star COM). The systems were chosen
for their representiveness, maturity and wide use, even interna-
tionally. The comparison analysis addresses the system intention,
applicability, concept, technical content, measurability and certi-
fication procedure (see Fig. 2 “analytical part”). The convergent
and divergent features of the investigated rating systems are also
expounded. Based on the findings, a new theoretical multi-criteria
system for urban sustainability named CAMSUD is developed as
a planning-oriented decision-support instrument (Fig. 2 “design |
implementation”) and is extensively described in Ali-Toudert et al.
(2016). Some key features of CAMSUD e.g. the definition of criteria
interactions, the link with the planning legislation and the database
of action measures are highlighted there.
Moreover, some findings from the present analysis e.g. the
construction of system categories, formulation of criteria, consider-
ation of criteria interactions etc. are heuristically used for the design
of the CAMSUD system. The highlights of CAMSUD are hence pre-
sented as outcome in Section 3. CAMSUD will be further constructed
as a computational tool at a later stage in order to cope with the
intrinsic complexity of the matter as discussed in this paper.
In this research, the current versions of the five selected national
rating systems for urban sustainability are studied:
- CASBEE for Urban Development version 2014 (CASBEE UD),
- LEED for Neighborhood Development version v4 2014 (LEED ND),
- BREEAM for Communities version 2012 (BREEAM COM),
- DGNB – New Urban Districts version 2012 (DGNB NSQ),
- Green Star – Communities version 0.2 2015 (Green Star COM).
2.1. System structure and categorization
2.1.1. System structure
The five investigated systems are characterized by a tree-like
structure as shown in Fig. 3: the main levels (e.g. categories) are
divided into sub levels (e.g. the criteria). Correspondingly, the sub
levels have clear membership to the main levels.
There are totally 6 levels found in the five rating systems. Level
2 and Level 5 are of crucial importance in all rating systems, hence
devoted extra attention in our analysis in this article. The reason is
that level 2 lays bare the rudiments and conceptualization process
of urban sustainability as applied in each rating system and level 5
addresses the scope and completeness of criteria in handling urban
600 F. Ali-Toudert, L. Ji / Ecological Indicators 73 (2017) 597–611

Fig. 2. Methodological structure of this study including a comparative analysis and the development of CAMSUD.

sustainability as well as highlights significant details on their con- terion” or “credit”. Level 6, which deals with the measurability i.e.
cretion. The levels are differently named but relative consensus is the quantification of the degree of fulfillment of a criterion, is less
observable for level 2 as “category” or “topic” and for level 5 as “cri- consensual (i.e. scoring criterion, assessment requirement, mea-
 F. Ali-Toudert, L. Ji / Ecological Indicators 73 (2017) 597–611
Fig. 3. The tree-like structure of the investigated systems including their successive levels.
surable indicator, action measure). The designations of each level
and an example for each are given in Fig. 3 to show the variety of
a system concept given by its structure and also the definition of
each level.
CASBEE UD exhibits the highest level of disparity in compari-
son with other systems. Level 1 is only available in CASBEE UD and
consists of two groups “quality” and “load”, which assess internal
quality and external environmental impact, respectively. While the
“quality” items are further segmented into different levels, “load”
item stops at level 2. Moreover, the chosen designations in CASBEE
UD are also quite different from the other systems. For example, the
“categories” or “topics” in level 2 are named “major items” whereas
“criteria” or “credits” on level 5 are named “minor items”. The out-
come in Fig. 3, expressed in points or level per criterion, is explained
in Section 2.3.2.
2.1.2. System categorization
The juxtaposition of the categories of these systems is hardly
possible due to the different logic governing the classification of
the various items. In lieu thereof, the logic in categorizing the sus-
tainability issues in each of the five rating systems is schematized
in Fig. 4 and discussed hereafter.
As aforementioned, CASBEE UD is the only system with the first
level. At level 1, a ratio of “environmental quality” and “environ-
mental load” is built based on the sum of scoring on level 2. The ratio
is named BEE – “built environment efficiency”. For the “environ-
mental quality”, CASBEE UD takes the three common sustainability
pillars as its three major items (= categories). The environmen-
tal impact of the three sectors (traffic, building and green) are set
as three major items and grouped into the “environmental load”.
Thereby, this concept puts more in the foreground the “load” on
the environment than the other systems.
LEED ND deviates from the structure usually based on the
three sustainability pillars. It rather adopts another viewpoint,
with the first three categories being planning-oriented and catego-
rized according to the spatial area of influence: total urban context
601
(smart location & linkage SLL), urban design (neighborhood pat-
tern & design NPD) and green buildings including infrastructure
(green infrastructure & buildings GIB). The category “innovation &
design process IN” promotes creative design, while “regional pri-
ority credit RP” is formulated for accessing geographically specific
environmental issues, in the USA (v2009) and worldwide (v4, 2014)
(USGBC, 2014).
The master planning process employed in BREEAM COM is dis-
tinctive in comparison to other systems in that all categories are
divided and executed in a chronological sequence as follows: step
1: establishing the principle of development; step 2: determining
the layout of the development and step 3: designing the details. This
accounts for why BREEAM COM recommends an early registration
of the projects in quest of certification.
DGNB NSQ defines its categories with the term “quality”. The
three common pillars of sustainability (ecology, society, economy)
are taken on as three main categories (ecological quality, economic
quality, socio-cultural and functional quality). The “technical qual-
ity” and “process quality” are two further categories, with distinct
operating terms valid through the whole system (DGNB, 2012).
DGNB NSQ is the only system highlighting “technical quality” in
a separate category, an implicit dimension in other systems. The
process quality deals with management and governance issues.
Besides the three pillars of sustainability interpreted as “envi-
ronment”, “economic prosperity” and “liveability” categories,
Green Star COM defines 3 further categories designated as “design”;
“governance” and “innovation”. The first five categories are strongly
oriented to the five principles “(1) enhancing liveability, (2) creating
opportunities for economic prosperity, (3) fostering environmental
responsibility, (4) embracing design excellence and (5) demon-
strating visionary leadership and strong governance”, which were
discussed at the beginning of the system’s development during
a consultation process with more than 180 organizations (GBCA,
2010). The last category “innovation” is set to award innovative
solutions (GBCA, 2015).
602 F. Ali-Toudert, L. Ji / Ecological Indicators 73 (2017) 597–611

Fig. 4. Structuring of the main categories within the investigated rating systems.

Hence, an apparent missing category in a system does not nec- 2.2. System technical content—criteria list
essarily imply that the significant issues of sustainability are not
accessed; the problem is the disparity in the structuring as well as Level 5 in Fig. 3 is the level of criteria for sustainability assess-
in the formal terms employed by the different systems in describing ment. A criterion can be defined as a rule or standard of judgment,
sustainability requirements. For more insight on the extensiveness therefore acting as a means of evaluation and subsequently aiding
of a system, it is more useful to consider level 5 dedicated to the in the decision process. A criterion operates like an index and is
single criteria as explained in the next section. measurable by a combination of several indicators acting as level 6
. in Fig. 3. This is worthy of note because the concepts ”criterion” and
 F. Ali-Toudert, L. Ji / Ecological Indicators 73 (2017) 597–611 603

Table 1
Comparison of the compliance criteria between the five selected rating systems.
604 F. Ali-Toudert, L. Ji / Ecological Indicators 73 (2017) 597–611

Table 1 (Continued)

”indicator” are often mixed-up in the literature creating thereby a 2.2.1. Principles of criteria formulation
certain confusion. At level 5, the thoroughness and extensiveness of The investigated rating systems use extensively diversified
the rating systems are comparable and this is discussed hereafter. terms for describing the criteria, which amplifies the perception of
 F. Ali-Toudert, L. Ji / Ecological Indicators 73 (2017) 597–611 605

Fig. 5. Weighting of the rating systems according to the categories set for the analysis.

Fig. 6. Weighting scale of the individual criteria in 4 of the investigated rating systems (in percentage %).

the complexity of the subject.Whereas some criteria formulations
are quite self-explanatory, others are ambiguous without reference
to the detailed description. Moreover, the denominations of the cri-
teria within each system are unequally detailed with some of them
kept minimal like “view” or “history and culture” and others at
full length like “possibility to make (energy) demand/supply sys-
tem smart” in CASEBEE UD. This is observable across the systems
making them inhomogeneous in form and content.
As Ali-Toudert et al. (2016) already mentioned, the designation
of criteria generally combines two main aspects: goals i.e.“target
to be attained” and the “action domain”. An example from LEED
ND: In “SLL credit: brownfield remediation”, the “remediation” is
the target and “brownfield” is the action domain. If with the defini-
tion of the action domain, the target is assumed to be clear enough,
e.g. “GIB credit: solar orientation” in LEED ND, the target could be
omitted. In some cases, however, this can be unclear at first and
hides different meanings, e.g. “LE 02 – land use” in BREEAM COM,
where only the action domain is known (land use planning) but the
target not easily understood as being about the investigation and
remediation of land contamination. In comparison with the same
606 F. Ali-Toudert, L. Ji / Ecological Indicators 73 (2017) 597–611
Table 2
Levels of final score in the investigated rating systems.
CASBEE UD LEED ND
0–0.5: poor C 40–49: certified
0.5–1.0: fairy poor B- 50–59: silver
1.0–1.5: good B+ 60–79: gold
1.5–3.0: very good A >80: platinum
>3.0 & “quality” ≥50%:
excellent S
DGNB NSQ Green Star COM
50% ≤ x < 65% & 10–19 points: 1 star:
y ≥ 35%: bronze minimum practice
65% ≤ x < 80% & 20–29 points: 2 star:
y ≥ 50%: silver average practice
x ≥ 80% & y ≥ 65%: gold 30–44 points: 3 star:
x: final score; good practice
y: minimum fulfillment 45–59 points: 4 star:
of each category best practice
60–74 points: 5 star:
Australian excellence
≥75 points: 6 star:
world leadership
term in DGNB NSQ “ENV 2.1: land use”, land use has another focus,
i.e. brownfield re-use and reduction of unsealed space consump-
tion. Likewise, the same criterion in CASBEE UD “3.1.2.2 land use”
focusses on land use efficiency via vertical extension and handling
soil contamination.
The advantage of defining the criteria as a target within a spe-
cific action domain is to enhance the operability of the criteria. The
“action domain” might be too constricted thereby restraining possi-
ble innovation in planning and design thereby putting the squeeze
on flexibility. For example “1.2.1.2 building top greening” in CASBEE
UD is less a target than a concretion measure or design strategy.
As aforementioned, the assessment criteria are principally
defined as “targets” but indeed some criteria are defined as plan-
ning measures or assignments to be fulfilled. The possible mix of
targets and measures is one of the manifest difficulties when for-
mulating a criteria list. For some of the criteria, a check list of
relevant assignments is provided. The criteria fulfillment is deter-
mined by counting the number of assignments in that list, if they
are undertaken during the urban project development or not. For
example, for the assessment issue RE 07 – Transport carbon emis-
sions in BREEAM COM, if “a feasibility study/desk-based assessment
is undertaken using the information from the transport assessment
or transport statement to establish appropriate alternative trans-
port options for the development” is defined as one criterion to
allocate credits (BRE Global, 2012) . It sets no matrix for evalu-
ating the design, so the quality of the activity or measure is not
directly assessed. In this case, the rating systems serve as a guide
to ensure that various targets of design are achieved by referring
to relevant specialist fields, rather than as a holistic medium for
quality assessment of the design.
Furthermore, the target may be too complex when influenced by
or applied to several action domains. The enlarged scope becomes
inappropriate for the evaluation under one single criterion. An
example is “Liv-3: healthy and active living” in Green Star COM.
2.2.2. Classification of criteria into categories
As previously mentioned, CASBEE UD, DGNB NSQ and Green Star
COM all adopt the three pillars of sustainability as the basis for
category formulation and these pillars intersect at shared themes,
hence, comparable criteria may be grouped into different cate-
gories. For example, in CASBEE UD, the criterion “efficiency of
energy system” is a subcategory of “economy”, but in DGNB NSQ
the criterion “primary energy demand and ratio of renewable pri-
mary energy” is assigned to the category “ecological quality”. In this
BREEAM COM
<30%: unclassified
≥30%: pass
≥45%: good
≥55%: very good
≥70%: excellent
≥85%: outstanding
two highlighted scenarios, where the emphasis is placed on within
the systems for the same criterion is different, the argument pred-
icates on the fact that the economic aspect of the energy issues in
CASBEE UD is more stressed than the ecological impact, while the
opposite is valid in the case of DGNB NSQ. Which criterion falls into
which category is a key issue since this influences the scoring of the
individual categories and calls for deliberation.
In order to make the criteria of the various 5 systems juxta-
posable, eight self-defined dimensions of urban sustainability are
used as framework of comparison (and later adopted as CAMSUD
categories) and shown in Fig. 5 and Table 1
Fig. 5 shows the weighting percentages and reveals important
differences. The categories (2) and (5) for land use and resources,
respectively, are most consensual, whereas the other categories
have larger variances, e.g. LEED ND seems to highlight the infras-
tructure and undervalues economy and management, whereas
DGNB NSQ privileges economic related topics to society related
ones.
2.2.3. Weighting and thoroughness of criteria
The basis for scoring the criteria in a rating system is sub-
ject to some weighting according to certain assumed importance.
The concept behind the various weighting perceptions has been
often deliberated as this governs the final result of the scoring,
consequently influencing the delivered certificate. Haapio (2012)
contends with the risk posed by the various yardstick upon which
criteria are examined, in that some criteria attracts more atten-
tion while other criteria with the same relevance are unwittingly
neglected. The number of criteria in the five systems vary from 30
(CASBEE UD) to 56 (LEED ND). The several voids left unfilled, like in
CASBEE UD, point out the relative imprecision of a system in com-
parison to another in describing an issue. The Weights of the eight
topics are allocated differently. The first three topics ([1] site loca-
tion | site ecology, [2] land use | urban fabric | building arrangements
and [3] infrastructure | transport) are mainly planning oriented, and
the rest are rather performance oriented. In this sense, LEED ND is
the most planning-oriented system of the five.
The thoroughness, simplification and emphasis employed in
each system are also visible in Table 1. Some aspects are in cer-
tain systems handled in detail while in other systems relatively
simplified, e.g. the criteria for the topic “site location | site ecology”
in LEED ND gain much more detail and importance than in other
systems. The blanks in Table 1 state for non-available criteria in the
corresponding systems. But this does not necessarily mean that the
 F. Ali-Toudert, L. Ji / Ecological Indicators 73 (2017) 597–611 607

relevant issues are not taken into consideration, in fact they might Table 3
Properties of CAMSUD in comparison with the five investigated national rating
have been incidentally or passively considered within other criteria
systems.
due to the interwoven nature of many of them.

2.2.4. Overlapping of criteria

Due to the complexity and the interdisciplinary nature of the
sustainability issues, the criteria in the systems are not stand-alone.
They always have inevitable interrelationships with each other.
This makes the scoring more difficult. In BREEAM COM, for example,
some overlapping could result from the chronological nature of the
rating system’s process. Indeed, BREEAM COM is the only system
which classifies the criteria into distinctive chronological order of
assessment. In step 1, it is more targeted towards conceptualization
and the criteria in step 3 are more construction-oriented. In other
words, some criteria involved in the latter stage are designated to
depict the detailed implementation of the initially derived solu-
tions from the former steps. Therefore, overlapping is imminent in
the criteria across the steps. As a demonstration, the criteria “rain-
water harvesting” in step 3 is part of the implementation of the
criterion “water strategy” in step 1.
Criteria overlapping can also occur as a result of different
spatial scales of observance employed as well as the variegated
nature of the topics involved. For example, “tree-lined and shaded
streetscapes” and “heat island reduction” are two criteria in LEED
ND which illustrate the possible redundancy due to different spa-
tial scales. “Urban design” and “efficient use of area” are two criteria
in DGNB NSQ within “socio-cultural and functional quality” and
“economic quality”, respectively highlighting the ease with which
redundancy could occur within a system. The overlapping is not
necessarily conceived as a disadvantage in this paper as this is
inherent to the complexity and multi-faceted character of urban
sustainability. However, ways of managing possible double scoring
must be provided. In this paper, criteria overlapping are one form
of criteria interactions (see also Ali-Toudert et al., 2016).

2.2.5. Mandatory requirements, optional criteria and bonus

criteria
The criteria list in the current rating systems generally consist
of a few mandatory criteria, prevalent optional criteria and excep-
tional bonus criteria that could be possibly fulfilled. The optional
criteria generally offer a certain degree of flexibility from low to
best compliance. LEED ND is the only system with prerequisite
criteria. The prerequisite credit in LEED ND usually has the same
name as its equivalent optional credit and mostly the same con-
tent as well, only the limit values are more constraining in the
latter case. BREEAM COM, DGNB NSQ and Green Star COM also have
such threshold requirements but implemented differently from
LEED ND. These are embedded into the main criteria and act as
the “minimum limit value” to attain. If this mandatory minimum
is not fulfilled, the rating cannot be achieved. The approach with
which bonus criteria is graded was not intended for measuring the
degree of fulfillment, but rather in encouraging additional achieve-
ments as well as acknowledging extra effort put in place in ensuring
better sustainability beyond the level governed by prevailing legal
standards and norms.

2.3. System measurability

2.3.1. Measurability of criteria
The criteria on level 5 are assessable by employing the indica-
tors on level 6 (Fig. 3). An indicator is a benchmark, a metric or a
statistic that indicates the state or condition of something, here the
measure of a criterion. The indicators on level 6 are measurable,
either quantitatively by setting clear-cut target compliance values
or qualitatively e.g. by proposing possible concretion measures. In
DGNB NSQ, some indicators are readily measurable, while others
are expressed in form of a cluster of parameters. For parameters
and indicators excluded from this aforementioned classification,
they are subjected to rules for allocating points with a degree of
flexibility governed by the prevailing conditions. These include
1) calculation: if an indicator or parameter is quantifiable, then
for attainable values, equivalent points are awarded, (2) check-
list for qualitative issues: for each achieved activity in the list,
certain points are awarded and (3) determination of accessibil-
ity: maximum number of attainable points are given for rating the
608 F. Ali-Toudert, L. Ji / Ecological Indicators 73 (2017) 597–611
SIT [4] URB [15] INF [4]
CLM [4]
RES [5]
SOC [2] | ECN [2] | MNG [4]
Fig. 7. Structure of the main categories of the CAMSUD system.
(1) analysis items (2) main findings
structure tree-like structure
according to three and four pillars models
categories according to sustainability topics
according to spaal scale
weighting: planning-oriented or performance-oriented
formulation: "target" or "action domain" or "target" +
"action domain" or "measures / acvies"
thoroughness: sometimes assessment criteria indirectly
considered
criteria
measurability
similar criteria differently categorized in different systems
criteria overlapping
point allocaon: criteria are scored as stand-alone issues
Fig. 8. The CAMSUD design principles against the main findings of the analysis of five existing rating systems.
accessibility to certain infrastructures and services. For example,
maximum time distance and/or maximum straight line distance to
some designated locations are set and the percentage of buildings
situated within this maximum distance operate as a multiplier fac-
tor. In this rule, the points finally awarded are the product of the
maximum possible points and that multiplier factor (DGNB, 2012).
These rules are similarly applicable in the other four systems.
BREEAM COM warns from setting planning and design targets
using individual issues and credit levels (level 5 in Fig. 3), as these
would limit design flexibility and have an impact on the project
costs (BRE Global, 2012). In order to encourage the innovative solu-
tions, DGNB NSQ has announced in its handbook that only target
values are defined but the measures are given a degree of flexibility
(DGNB, 2012). In fact, the target values are regulated by two rules
(1) and (3) aforementioned. It is worth mentioning that measures
are occasionally listed in lieu of qualitative targets. In general terms,
all the systems discussed utilize both the fulfillment of compliance
targets and the number of implemented measures provided by the
rating systems in the assessment of urban sustainability with the
exception of Green Star COM (no information available on level 6).
SIT: site locaon | site ecology
URB: land use |urban fabric | building arrangements
INF: infrastructure | transport
CLM: urban climate | climate change
RES: resources [energy, water, materials]
SOC: society | culture
ECN: economy
MNG: management | quality of service
[ ]: number of criteria in the corresponding category
(3) CAMSUD design
system structure:
network
concepon :
spatial scale + sustainability topics + pillars model
planning-oriented: SIT + URB + INF = 23 from 40 criteria
criteria formulation:
"targe o be aained" + "action domain"
measures and solutions:
collected in a database, supporting quantification of
interactions
definion and quantification ofcriteria interactions
2.3.2. Weighting and scoring by point allocation
The method with which points are allocated has a decisive effect
on the rating result. The methods include: (1) weight assignments
for different criteria, (2) degree of fulfillment of each criterion and
(3) levels of final score. Fig. 6 illustrates graphically the weighting
scale and logic behind the scoring in each system (except green
star).
CASBEE UD implements an equal level of importance through-
out the criteria list in assigning scores, thereby implying that this
system places the three dimensions of sustainability on same level
of significance; the adjoining subtopics are also on same scale of
influence within each dimension, as well as the criteria within each
subcategory. This simply connotes that the weight of each criterion
(minor item in CASBEE UD) is a cumulation of the number of minor
items underlying the small and middle items, hence the small items
with the most criteria (22 of 30) get the weight of 2.8%, while the
remaining 8 criteria are allocated different weights between 1.9%
and 11.1%. DGNB NSQ and Green Star COM treat each criterion dif-
ferently with each criterion been handled with varying levels of
importance. In the case of DGNB NSQ, the weight assigned for each
main topic is fixed. The first four main topics are allocated 22.5%
 F. Ali-Toudert, L. Ji / Ecological Indicators 73 (2017) 597–611
each, whereas the process quality is assigned 10%. The weights
are further allocated onto the criteria level unequally but typically
influenced by their level of significance named “significance factor”.
Each criterion gets a weight from 0.9% to 6.8%. Weight allocation
in BREEAM COM is typically regulated by the “impact of each cate-
gory against each of the three aspects of sustainability” (BRE Global,
2012). The weight of each criterion is estimated by the same princi-
ple. As a result, the criteria in BREEAM COM are weighted between
0.9% and 8.9%. In LEED ND, the weight of each criterion corresponds
to its number of credit directly, e.g. a criterion with 1 credit has
the weight of 1% since the total equals 110 credits. LEED ND is
the only system with a possible credit score of 0 corresponding
to prerequisites (Fig. 6).
Each criterion has a scale of fulfillment. For CASBEE UD, the
scale has five levels. For DGNB NSQ, it is 0–100 points, where 0
point corresponds to level 1 in CASBEE UD being the lowest value
or minimum requirement. Fifty points in DGNB NSQ and level 3
in CASBEE UD correspond to the norm and it refers to the current
state of technology or standard according to local mandatory reg-
ulation. One hundred points in DGNB NSQ and level 5 in CASBEE
UD are the highest attainable target values corresponding to best
practice (DGNB, 2012; IBEC, 2014). For LEED ND and BREEAM COM,
the scale of fulfillment is not static as in DGNB NSQ and CASBEE
UD and a criterion could have one to ten and one to eleven credits,
respectively.
BREEAM COM and DGNB NSQ utilize the scale 0–100 (%) in com-
puting the final score. LEED ND and Green Star COM count up to
110% because they offer extra bonus points to a maximum of 10%.
The range of classification for final aggregates are listed in Table 2.
The final scores resulting from the different systems are not easily
comparable due to the disparity in the different levels of strictness
prevailing in the criteria assessment, although all use similar scale
(from 0 to 100 or 110) for result presentation.
CASBEE UD is the only system that sets no minimum require-
ment to a certificate in its appraisal of the neighborhood. More
than just acting as a certification system, it is characterized by
much more flexibility by taking into account all possible quality
levels of urban districts. The overall scoring result of CASBEE UD is
depicted by a generated quotient BEE “quality/load” at a scale with
five possible levels (see Fig. 4).
As previously mentioned in Section 2.2.4, the intrinsic com-
plexity of urban sustainability leads to the difficulty of defining
“stand-alone” criteria i.e. without any form of interference between
them. However, the scoring procedure employed in the five systems
only evaluates the criteria by measuring the degree of fulfillment
of the specified target values without special consideration for the
resulting interaction with other subtle targets. Whether the con-
cretion measures of a criterion indirectly decrease the fulfillment
of another criterion is currently not addressed by the systems. How-
ever, the interactions between these criteria cannot be neglected
because considering possible interactions subsequently improves
the prediction of lateral changes which must be coordinated by the
planners in a compromise solution.
3. CAMSUD system
The name CAMSUD stands for Comprehensive Assessment
Method for Sustainable Urban Development. It was initiated by
Ali-Toudert (2008) and then progressively developed at the chair
for energy efficient buildings EEB at the TU Dortmund University
(Fährmann, 2012; Czempik, 2013; Ali-Toudert et al., 2016). The first
published version, named CAMSUD 1.0, is described in Ali-Toudert
et al. (2016).
Table 3 recalls from Ali-Toudert et al. (2016) dominant prop-
erties of CAMSUD, alongside brief description on similarity or
609
difference compared against the five national rating systems. This
section focuses then on the process and principles for the develop-
ment of the CAMSUD concept, technical content and measurability
relevant issues.
3.1. CAMSUD concept: structure and categories
In a bid to ensure clarity in the system’s structure, CAMSUD spec-
ifies only 3 hierarchical levels, which are (1) category (main topic),
(2) criterion (specific goal) and (3) indicator (measurable key met-
ric). This is comparable to most of the investigated rating systems
but the novel idea behind is a network structure. This means that all
indicators are interconnected for the definition and quantification
of their interactions in view of the final assessment of the criteria
(see Ali-Toudert et al., 2016). The advantage of a network struc-
ture instead of a tree-like structure will be explained together with
the way of identification and definition of criteria interactions in
CAMSUD system (see Section 3.2).
The system’s structure of CAMSUD is shown in Fig. 7. The
thematic categories refer to Ali-Toudert (2008) and are in good
agreement with the analyzed rating systems (see Section 2.1) as
well as with the literature review to urban sustainability. CAMSUD
highlights the 3 spatial dimensions of planning and design as sys-
tem’s core: site location | site ecology (SIT), land use | urban fabric
| building arrangements (URB) and infrastructure | transport (INF).
Then, about the impact on the environment, CAMSUD focuses on
urban climate and climate change related issues (CLM) and natu-
ral resource efficient use (RES) throughout the 3 planning poles.
Ultimately, economic, social and governance matters are handled
transversally across the system in explicit categories (SOC, ECN and
MNG).
Because the CAMSUD system is still under development, the
weighting of the categories has not yet been determined. The
weighting of CAMSUD categories and criteria shown in Table 1 and
Fig. 5 are given at the moment solely according to the number of
criteria. With this rough weighting, the three categories SIT, URB
and INF have a sum of 57.5%, which corresponds to the original
concept of making CAMSUD a planning-oriented system.
3.2. Technical content and measurability of CAMSUD system
The investigated national rating systems structure the content
of the criteria differently, sometimes with disputable consistency,
and this posed a challenge in conducting the analysis as reported in
Section 2.2.1. To solve this issue, CAMSUD strives to follow clear sys-
tematics in the definition of its criteria. CAMSUD criteria are given
in Table 1 and the principles guiding their designation are further
discussed herein. The criteria are concisely named and emphati-
cally denoted as “goals” in form of a “target to be attained” for a
specific “action domain”. For example, in criterion [15] “housing
typology” is the action domain and “diverse” is the target to be
reached. This aims to simplify every criterion giving it an explicit
meaning and clear-cut boundaries with respect to other criteria.
In some cases, where the target to be attained or the action to be
taken is obvious, it has been omitted. A typical case is criterion [10]
where the target of ensuring outdoor spaces and recreational qual-
ity are considered as tacitly understood. Other scenarios where the
designated goals are related, these are grouped in one criterion, e.g.
in the criteria [14] building layout | building typology, [20] urban
network | accessibility or [33] social diversity | equity, etc. In cri-
terion [12], expressing safety and security, the goal to be reached
suffices because the action domain applies to the whole project.
Furthermore, it was ensured that criteria which express tangi-
ble action measures are avoided in order not to confine the user’s
innovative skills. Recommendations about the best practices are
rather foreseen to be collected in an external database to act as sup-
610 F. Ali-Toudert, L. Ji / Ecological Indicators 73 (2017) 597–611
port in the planners’ decision-making and subsequently serve the
quantification at indicator level. This catalogue of measures is flex-
ible and expandable in order to accommodate the planners’ private
experience.
CAMSUD assessment approach contrasts with the following
findings from the analysis of existing rating systems (see Fig. 8):
(1) criteria, which are not available in the investigated systems, are
sometimes considered indirectly (see Section 2.2.3 and Table 1); (2)
because of the absolute membership of criteria to the categories, a
certain aspect of this criteria is emphasized whereas other possible
aspects are neglected (see Section 2.2.2); and (3) the overlapping of
criteria. So, CAMSUD adopts a new approach to handle the assess-
ment based on the identification and quantification of their mutual
interactions (see Fig. 8). This will be processed in form of weight-
ing assignments via shared action measures. A computational tool
(ECAMSUD) is envisaged as a management tool for the theoretical
CAMSUD system which will assist in a dynamic appraisal process
of the criteria in order to guarantee an appropriate weighting of
their respective influence. More details and also some examples of
the logic chain of criteria interactions are given in Ali-Toudert et al.
(2016).
4. Conclusion
An assessment method for urban sustainability is only viable if
it is properly modeled and appropriately measured. This issue has
been investigated on five national multi-criteria rating systems and
on the newly developed CAMSUD system. Urban sustainability is
indeed found to be a consensual objective, which concept is, how-
ever, still in a dynamic process of clarification in terms of content as
well as ways of implementation. In this respect, multi-criteria based
systems reveal to be useful operational approaches, yet requiring
further maturation in order to cope with the intrinsic complexity
of the subject.
The analysis showed that the criteria are variably formulated,
either as targets (i.e. goals), action measures or assignments to be
fulfilled. This makes a rating system unclear whether the fulfill-
ment of its sustainability goals is evaluated or the compliance to
predefined solutions or whether the use of planning instruments is
assessed. To make the criteria target-oriented, clear but still widely
operable, CAMSUD defines assessment criteria with the general for-
mulation of a goal as “target to be attained” applied to a specific
“action domain”.
A proper measuring of the criteria often faces the prob-
lem of completeness of sustainability items, overlapping criteria,
weighting assumptions and stand-alone assessment. Especially, the
possible interactions between the criteria are generally ignored in
the investigated national rating systems, which may lead to the
overestimation or underestimation of sustainability compliance.
In CAMSUD, a focus is put on the necessity of accounting for
criteria interactions which might be topic-related, scale-related
and/or time-related. To achieve this goal, a computational tool
based on mathematical methods helping decision making is one
privileged way to answer this complexity. This tool development is
set as the next task for CAMSUD implementation for the practice.
References
Alberti, M., 1996. Measuring urban sustainability. Environ. Impact Assess. Rev. 16
(4–6), 381–424.
Ali-Toudert, F., Ji, L., Fährmann, L. and Czempik, S., 2016. Comprehensive
Assessment Method for Sustainable Urban Development CAMSUD — A New
Multi-Criteria System for Planning, Evaluation and Decision-Making,
submitted to Ecological Indicators (in revision).
Ali-Toudert, F., 2008. From green building to sustainable urban settlements: a new
assessment method. PLEA, 6 pp.
Augenbroe, G., Pearce, A.R., 2010. Sustainable Construction in the United States of
America: A Perspective to the Year 2010. Georgia Institute of Technology
[Online] Available at http://www.irbnet.de/daten/iconda/CIB13602.pdf.
BRE Global, 2012. BREEAM Communities: Technical Manual SD202-0.2:2012
[Online], Watford Hertfordshire, Available at http://www.breeam.com/bre
PrintOutput/BREEAM Communities 0 2.pdf.
Bott, H., Grassl, G., 2013. Nachhaltige Stadtplanung, München. Institut f. intern
Architektur-Dok.
Brannstrom, C., Vadjunec, J.M. (Eds.), 2013. Routledge, New York.
Cerone, A., Persico, D., Fernandes, S., Garcia-Perez, A., Katsaros, P., Shaikh, S.A.,
Stamelos, L., 2014. Information technology and open source. In: Applications
for Education, Innovation, and Sustainability: SEFM 2012 Satellite Events,
InSuEdu, MoKMaDS, and OpenCert, Thessaloniki, Greece, October 1–2, 2012:
revised selected papers, Berlin, Springer.
Collier, M.J., Nedović-Budić, Z., Aerts, J., Connop, S., Foley, D., Foley, K., Newport, D.,
McQuaid, S., Slaev, A. and Verburg, P., 2013. ‘Transitioning to resilience and
sustainability in urban communities’, Current Research on Cities, 32,
Supplement 1, no. 0, pp. S21-S28 [Online]. 10.1016/j.cities.2013.03.010.
Costanza, R., Wainger, L., 1991. Ecological Economics: The Science and
Management of Sustainability. University Press, New York, Columbia.
Curwell, S.R., Deakin, M., Symes, M., 2007. Sustainable Urban Development: The
Environmental Assessment Methods. Taylor & Francis.
Czempik, S., 2013. Praxisorientierte Anwendung und Weiterentwicklung eines
neuen theoretischen planungsorientierten Indikatorensystems zur
Nachhaltigkeitsbewertung von Stadtquartieren. In: DegreeThesis. TU
Dortmund University, Germany.
DGNB, 2012. Neubau Stadtquartiere: DGNB Handbuch für Nachhaltiges Bauen;
Version 2012. Kohlhammer, Stuttgart.
Dizdaroglu, D., Yigitcanlar, T., 2014. A parcel-scale assessment tool to measure
sustainability through urban ecosystem components: the MUSIX model. Ecol.
Indic. 41 (0), 115–130, http://dx.doi.org/10.1016/j.ecolind.2014.01.037
[Online].
Drakakis-Smith, D., 1995. Third world cities: sustainable urban development 1.
Urban Stud.s 32 (4–5), 659–677.
Duxbury, N. and Gillette, E., 2007. Culture as a Key Dimension of Sustainability:
Exploring Concepts, Themes, and Models [Online], Centre of Expertise on
Culture and Communities, Creative City Network of Canada/Simon Fraser
University, Vancouver. Available at http://cercles.iii.com/arxius/pdf/E130054.
pdf.
El Bassam, N., Maegaard, P., Schlichting, M.L., 2013. Distributed Renewable
Energies for Off-Grid Communities: Strategies and Technologies Toward
Achieving Sustainability in Energy Generation and Supply. Elsevier, Oxford,
UK, Waltham, MA.
Fährmann, L., 2012. Entwicklung eines planungsorientierten Indikatorensystems
zur Nachhaltigkeitsbewertung von Stadtquartieren. In: Degree Thesis. TU
Dortmund University, Germany.
Fücks, R., 2013. Intelligent wachsen: Die grüne Revolution. Hanser, München.
GBCA (ed), 2010. Green Star Communities: Stage 1 Consultation Report [Online].
Available at http://www.gbca.org.au/uploads/152/2712/
GS%20Communities%20-%20Stage%201%20Consultation%20Report
FINAL%20Issue.pdf.
GBCA (ed), (2015) Green Star Communities: Scorecard [Online]. Available at http://
www.gbca.org.au/green-star/rating-tools/green-star-communities/rating-
tool/green-star-communities-pilot-v02-rating-tool/.
Haapio, A., 2012. Towards sustainable urban communities. Environ. Impact Assess.
Rev. 32 (1), 165–169.
Hawkes, J., 2001. The fourth pillar of sustainability: Culture’s essential role in public
planning, Melbourne,V ic, Cultural Development Network; Common Ground.
Higgins, K.L., 2015. From bud to blossom: nurturing sustainable stewardship. In:
Higgins, K.L. (Ed.), Economic Growth and Sustainability. Academic Press, San
Diego, pp. 167–180 (Chapter 13).
Hiremath, R.B., Balachandra, P., Kumar, B., Bansode, S.S., Murali, J., 2013.
Indicator-based urban sustainability–A review. Energy Sustain. Dev. 17 (6),
555–563, http://dx.doi.org/10.1016/j.esd.2013.08.004.
IBEC, 2014. CASBEE for Urban Development: Technical Manual (2014 Edition)
[Online]. Available at http://www.ibec.or.jp/CASBEE/english/download/
CASBEE-UDe 2014manual.pdf.
Kleine, A., 2009. Operationalisierung einer Nachhaltigkeitsstrategie: Ökologie,
Ökonomie und Soziales integrieren [Online]. Gabler Verlag, Available at http://
books.google.de/books?id=QLTazBpsHbAC.
Lorenz, S., 2014. Mehr oder weniger?: Zur Soziologie ökologischer
Wachstumskritik und nachhaltiger Entwicklung, Bielefeld, transcript.
Lozano, R., 2008. Envisioning sustainability three-dimensionally. J. Clean. Prod. 16
(17), 1838–1846.
Moran, D.D., Wackernagel, M., Kitzes, J.A., Goldfinger, S.H., Boutaud, A., 2008.
Measuring sustainable development—nation by nation. Ecol. Econ. 64 (3),
470–474.
Mori, K., Christodoulou, A., 2012. Review of sustainability indices and indicators:
Towards a new City Sustainability Index (CSI). Environ. Impact Assess. Rev. 32
(1), 94–106, http://dx.doi.org/10.1016/j.eiar.2011.06.001.
Munasinghe, M., 1993. Environmental Economics and Sustainable Development.
World Bank, Washington, D.C.
NZMCH—New Zealand Ministry for Culture and Heritage, 2006. Cultural
Well-being and Local Government: Report 1 (Online]. Available at http://www.
mch.govt.nz/files/report1.pdf.
 F. Ali-Toudert, L. Ji / Ecological Indicators 73 (2017) 597–611
Neumayer, E., 1999. Weak Versus Strong Sustainability: Exploring the Limits of
Two Opposing Paradigms. Elgar, Cheltenham, UK, Northampton, MA, USA.
Reith, A., Orova, M., 2015. Do green neighbourhood ratings cover sustainability?
Integr. Ecol. Indic. Sustain. Urban Ecosyst. Eval. Manag. 48 (0), 660–672, http://
dx.doi.org/10.1016/j.ecolind.2014.09.005.
Rückert-John, J. (Ed.), 2013. Springer VS, Wiesbaden.
Rees, W.E., 1992. Ecological footprints and appropriated carrying capacity: what
urban economics leaves out. Environ. Urban. 4 (2), 121–130.
Sayigh, A., 2014. Sustainability, Energy and Architecture: Case Studies in Realizing
Green Buildings. Academic Press, Oxford, UK.
Sharifi, A., Murayama, A., 2013. A critical review of seven selected neighborhood
sustainability assessment tools. Environ. Impact Assess. Rev. 38, 73–87, http://
dx.doi.org/10.1016/j.eiar.2012.06.006.
Spangenberg, J.H., 1997. Environmental space-based proactive linkage indicators:
a compass on the road towards sustainability. In: Moldan, B. (Ed.),
Sustainability Indicators: a Report on the Project on Indictors of Sustainable
Development. Wiley, Chichester, pp. 119–124.
611
Spindler, E.A., 2011. Geschichte der Nachhaltigkeit: Vom Werden und Wirken
eines beliebten Begriffes [Online]. Available at https://www.nachhaltigkeit.
info/media/1326279587phpeJPyvC.pdf.
Stoke, K., 2008. 3 Spheres for Sustainability [Online]. Available at http://kauaian.
net/blog/?p=1162.
Thwink.org, 2016. The Three Pillars of Sustainability [Online]. Available at http://
www.thwink.org/sustain/glossary/ThreePillarsOfSustainability.html.
USGBC, 2014. LEED v4 for Neighborhood Development [Online]. Available at http://
www.usgbc.org/sites/default/files/LEED%20v4%20ND 10.01.14 current 0.pdf.
Valentin, A., Spangenberg, J.H., 2000. A guide to community sustainability
indicators. Assess. Methodologies Urban Infrastruct. 20 (3), 381–392, http://dx.
doi.org/10.1016/S0195-9255(00)00049-4 [Online].
Wackernagel, M., Onisto, L., Bello, P., Callejas Linares, A., Susana López Falfán, Ina,
Méndez Garcıı́a, J., Isabel Suárez Guerrero, Ana and Guadalupe Suárez
Guerrero, Ma, 1999. National natural capital accounting with the ecological
footprint concept, Ecol. Econ., vol. 29, no. 3, pp. 375–390.
Wu, J., 2008. Making the case for landscape ecology: an effective approach to urban
sustainability. Landsc. J. 27 (1), 41–50.