Stockholm Resilience Centre
Research for Governance of Social-Ecological Systems
Master’s Thesis, 60 ECTS
Ecosystems, Resilience and Governance Master’s programme 2010/12, 120 ECTS
Challenges and opportunities for governing ecosystem services
in an urban world
A systematic review and synthesis
Marte Sendstad
1
MSc thesis Ecosystems, Resilience and Governance
Stockholm Resilience Centre Marte Sendstad
Challenges and opportunities for governing ecosystem services in an urban world: a systematic review and
synthesis
Supervisor: Thomas Elmqvist Co-Supervisor: Cathy Wilkinson
2
Abstract
Governance of ecosystem services for an urban population is an issue that is of global concern as people all over the world are moving into cities, and cities are also drawing on ecosystem services from all over the world. Efforts in science and policy have however been lagging behind in understanding and responding to the complex relationship between urbanisation and ecosystem services. To assess the state of the art in the scientific literature regarding
governance of urban ecosystem services, a methodology for systematic review and qualitative synthesis was designed and applied drawing on the experiences from health science and grounded theory. The approach was found to be useful in handling a large and diverse body of information and categorizing the main challenges and opportunities in governance of urban ecosystem services found in the literature. The results suggest that challenges can broadly be related to lack of knowledge, equity, lack of institutional capacity, conflicting views among stakeholders, and finally, decision makers often prioritizing economic growth over ecosystem services (ES) provision. Two over-arching challenges found are related to scale mismatch and trade-offs, suggesting that governance of ES should be considered in terms of its effects on different temporal and spatial scales, associated trade-offs, also in terms of beneficiaries. The main opportunities identified relate to planning and management – principles, approaches and tools, civil society as a source of knowledge and management capacity, participatory learning processes, leadership and ecological citizenship. Research and governance need to recognize how urban societies influence and depend upon ecosystem processes and to strengthen the evidence base for urban ES governance. This is the first time a systematic review and
synthesis has been performed addressing this topic. It aims to support the emerging discipline of urban ecology, and policy by informing the Cities and Biodiversity Outlook.
3
Table of content
Abstract ... 2
Introduction ... 5
An emerging environmental agenda ... 6
Problem statement ... 7
Research questions ... 7
Structure of the thesis and limitation ... 8
Theoretical context ... 9
What is an urban area? ... 10
Ecosystem services ... 10
Governance ... 13
Method ... 15
Introduction to method ... 15
The traditional review and the systematic review ... 15
Grounded theory ... 19
The method designed ... 20
Results and synthesis ... 26
Search report from systematic review ... 26
The main challenges found for governance of urban ES ... 29
The main opportunities found for governance of urban ES ... 44
Discussion ... 59
Synthesis of challenges and opportunities ... 59
Discussion related to method ... 67
Conclusion ... 72
References ... 73
Appendix A ... 86
Appendix B ... 95
4
Acknowledgements:
I would first like to thank my supervisors, Cathy and Thomas, for their advice and for involving me in the work on the Cities and Biodiversity Outlook.
I would also like to thank Oliver, Andre, Chantal and Fabiana for welcoming me to the CBD secretariat as a part of the cities and local authorities’ team and taking the time to comment on my preliminary results.
I would like to thank Sue, Lars K, and Karianne H for sharing insight and literature. I would also thank my fellow students for the time spent at SRC and in particular my thesis group for their feedback on my work.
Finally, I would like to thank Sverre who has been an important support throughout the time spent on this study.
5
Introduction
The world is becoming increasingly more urban; close to 70% of the global human population has been estimated to be living in cities by 2050 (OECD 2012). Even though rates of
urbanisation are slowing, the size and the numbers of cities are still growing (UN POP 2011, Potts 2012). The urbanizing trends are most drastic in the developing countries (UN POP 2011, OECD 2012); where cities are estimated to absorb more than 95% of the total expected population growth (Grimm et al. 2008). Nature is the life support system of our society, and people in cities, as elsewhere, depend on ecosystems for their life and welfare (Odum 1989, Daily 1997). Ecosystem services are however largely being degraded and unsustainably used (MA 2005, CBD 2010a). The unambiguous trend towards a more urban world increasingly gives cities an important role in global environmental governance as centres of demand of ecosystem services (ES) and accordingly, sources of environmental impact, representing a risk of eroding the capacity of ecosystems to support human welfare within and beyond cities (Kroll et al. 2012). It has been shown that urban populations depend on ES provided by land often covering several hundred times the area covered by the city (Rees & Wackernagel 1996, Folke et al. 1997, Warren-Rhodes & Koenig 2001). Governance of such ES for an urban population is therefore an issue that is of global concern as people all over the world are moving into cities, and cities are also drawing on ecosystem services from all over the world.
Urbanisation is however not inherently a threat to ES. Cities will have a large impact on demand of ES merely due to hosting a large population, but it is arguably an effective spatial organisation using less land per capita (for inhabitancy) and holding potential economies of scale that is of particular relevance to a growing global population (Glaeser 2011, McDonald
& Marcotullio 2011, Puppim de Oliveira et al. 2011). The relationship between the urban populations and ES are complex, and a growing number of studies have focussed on understanding the interlinked social and ecological dynamics of cities within the emerging field of urban ecology (Young & Wolf 2006). Knowledge of the diversity of insights from different disciplines and approaches to this topic is however dispersed, and there has not been any comprehensive scientific review of the experiences of the challenges and opportunities in
6 governing ES for an urban population. Doing a review and developing methodology for dealing with this kind of complex and interdisciplinary knowledge to inform policy is therefore timely.
An emerging urban environmental agenda
Cities started to grow rapidly in Europe and North-America following the industrial
revolution. Soon, pollution became a serious issue affecting human health and the integrity of ecosystems, through e.g. disruption of the biochemical cycles (Haughton & Hunter 1994).
After World War II and a following liberalization of global trade, cities, being engines of economic growth, went from having mainly local to regional impacts, to influence the environment on a global level (Marcotullio & McGranaham 2007, Lieberherr-Gardiol 2008).
The contemporary environmental agenda focusing on global environmental change emerged in the early 1970s, with an increased awareness in civil society and among decision makers concerning environmental degradation (Haughton & Hunter 1994) and the planet as a system with limits to growth (cf. Meadows et al. 1972).
The report Our Common Future (WCED 1987) recognizes a rapid urbanization on a global scale, the central role of cities in the global economy, and that the prospects of any city
depend on and influence the hinterland. It also argues that national authorities are not enabling local authorities to deal with challenges related to rapid urbanisation. This focus on local authorities in environmental governance gained further attention following the Earth Summit in Rio de Janeiro in 1992. However, the relationship between cities and ecosystems has not been a focus in global policy making (Puppim de Oliveira et al. 2011). The landmark report on ES, the Millennium Ecosystem Assessment (MA 2005) only had a small section on cities, and The World Development Report (World Bank 2009), a central publication on
urbanisation trends, does not include ecosystems or the services they provide.
At the Conference of the Parties (COP10) of the Convention on Biological Diversity (CBD) parties recognized the policy gap regarding urbanisation and the environmental agenda requesting an “assessment of the links and opportunities between urbanization and
biodiversity” (CBD 2010b: X/22). This study will inform the response to that request, The
7 Cities and Biodiversity Outlook (CBO-1), by providing knowledge support for the synthesis report and the CBO-1 scientific foundation report to be presented at CBD COP11 and at Rio+20.
Problem statement
Despite the clearly critical nexus between cities and natural resources, the relationship between urbanization and ecosystem services has not been prioritized in research (Grimm et al. 2008, Niemelä et al. 2010), in environmental governance or in urban intergovernmental policy processes (see above, Puppim de Oliveira et al. 2011). There has not been any
systematic review of the scientific evidence regarding governance of urban ES, resulting in a knowledge gap that is pertinent to address to inform both the broader urban environmental agenda and the Scientific Foundation for the Cities and Biodiversity Outlook, as well as further scientific endeavours. Furthermore, it is important to develop appropriate
methodology to examine and synthesise large bodies of information regarding complex interlinked social and ecological issues with knowledge stemming from multiple disciplines like the topic at hand.
Research questions
This thesis aims to fill the above mentioned knowledge gap by developing a global scientific systematic review and qualitative synthesis of urban ES governance to support further research efforts and inform the Cities and Biodiversity Outlook. It will explore important governance opportunities and challenges directly related to ES represented in scientific literature by answering the following research questions:
1) What are the main challenges presented in the literature for governing urban ecosystem services?
2) What are the main opportunities presented in the literature for governing urban ecosystem services?
3) How can systematic review and qualitative synthesis contribute to generating insights into governance of ecosystem services for an urban population?
8 Structure of the thesis and limitations
Due to this thesis being a literature review and synthesis based on an inductive approach, the structure of a traditional thesis, often presenting a case study, has been amended to fit the purpose of this project. The next section aims to present key concepts for the study and place the study in a theoretical tradition. The study then proceeds with a method section presenting both the theoretical background of the method as well as elaborating on how it was designed and applied. The results section will provide a search report, as well as a narrative of the challenges and opportunities found. The thesis will be finalised with a discussion of the synthesis and the contribution of the method.
9
Theoretical context
Urbanization and governance of urban ES is a complex topic with social, economic, technical, and ecological components. Most of the articles included in this review can be related to the emerging field of urban ecology. Urban ecology is defined, by McDonnell (2011:9) as merging “...both basic (i.e. fundamental) and applied (i.e. problem oriented), natural and social science research to explore and elucidate the multiple dimensions of urban
ecosystems”. Urban areas have not been a research priority among ecologists until late 20th century (Grimm et al. 2008), and early studies were conducted with a more narrow focus, in particular on urban biotopes and introduced species (Sukopp 2002). These local studies within cities are often referred to as studies of the ecology in cities, in contrast to the ecology of cities. The latter perspective, with a focus on the city as a whole, emerged after 1970,
investigating energy flow and nutrient cycling (Sukopp 2002). The more recent development within urban ecology has been focusing on cities as “...heterogeneous, dynamic landscapes and as complex, adaptive, socioecological systems, in which the delivery of ecosystem services links society and ecosystems at multiple scales” (Grimm et al. 2008: 756). Social science has however a longer tradition of studying human-environmental interaction in cities (Grimm et al. 2008). Over time studies of social and ecological, as well as economic and technical aspects of the city, have become more integrated in urban ecology, including sociology, economics, planning, and psychology (Young & Wolf 2006). This study builds on this diverse tradition and focus on two key concepts in addition to urbanisation, namely governance and ecosystem services. Together these direct attention to the social and
ecological dynamics in cities and also its relations to the wider landscape, the ecology in and of cities, and are of relevance to further knowledge development in urban ecology. The three main concepts of the study are presented in the following sub-section; urban areas, ecosystem services and governance.
10
What is an urban area?
Definitions of urban areas or cities, which are often not explicit in the literature, can
emphasize different parameters like population, political status, and density (see Hardoy et al.
2001 and references therein). A general definition considered to match all studies included in this review is that cities/urban areas are centres of human activity dominated by built
environment that are often characterised by a gradient of density from a core, or clusters of such, surrounded by suburbs and peri-urban areas connecting the urban to the rural. Defining a city or urban area can be a particular challenge when studying them as social-ecological phenomena since the area included in the more generic definition above includes merely a fraction of the spatial and ecological effects of urbanisation (cf. Rees 2010).
When discussing whether an issue is local it is generally meant within the urban jurisdiction, in contrast to regional, which is meant to include the city and its surrounding hinterland including e.g. watersheds, forests and agricultural areas, involving higher level jurisdiction authorities. Issues described as having a global nature is however thought to cross national borders and commonly across major continents. This is done for simplicity although it is recognized that there are grey zones; e.g. megacities may be more correctly characterised as city-regions, and some cities may draw upon resources across national borders for the same processes as others draw on “regional” resources, like shared watersheds.
Ecosystem services
ES has only recently become a widely used term within science (Dick et al. 2011), being defined as “conditions and processes through which natural ecosystems, and the species that makes them up, sustain and fulfil human life” (Daily 1997:3). It is thus an anthropocentric and utilitarian concept (Borgström 2006) that enhances a focus on how ecosystem functioning provides a broad range of services underpinning and enabling our economy and society to operate (Folke et al. 2007, Merson et al. 2010). An ecosystem can be defined as consisting of all living organisms within a biological community, their interactions with each other, the abiotic environment, and with adjacent ecosystems (cf. Odum 1989). Bolund & Hunhammar (1999:294) argue that the city as a whole can be viewed as one large ecosystem, and apply the
11 term urban ecosystems “for all natural green and blue areas in the city, including in this
definition street trees and ponds“.
Using ES to understand and address ecosystems has resonated in research and policy circles due to an awareness of how efforts to protect ecosystems, including the large variety of benefits people obtain from these, have not been successful (Dick et al. 2011). The
Millennium Ecosystem Assessment (MA 2005) highlighted this concept presenting a set of categories of services related to different constituents of human well-being, like basic materials for a good life, health and security (MA 2005:VI).
As defined in this report, ES include (MA 2005:5):
• Provisioning services, which are material goods like food, water, timber, and fiber
• Cultural services which are non-material benefits supported by ecosystems like recreational, aesthetic, and spiritual experiences.
• Regulating services, regulating ecosystem processes like climate, floods, disease, waste, and water quality
• Supporting services, which underpin the categories above by e.g. soil formation, photosynthesis, and nutrient cycling.
Urban areas draw on a number of different ES on different spatial scales from all the MA categories (Appendix A, table S1). ES varies in terms of whether they can be transported over longer distances, by natural or human means. Global climate regulation rely on ecosystems all over the globe, water quality depends on the watershed on the regional level, recreational services from green areas in peoples’ everyday life should be possible to reach within a few kilometres, and shade from trees or noise reduction are provided only locally (Bolund &
Hunhammar 1999, McDonald & Marcotullio 2011).
The categories overlap and support each other, indicating the importance of having a system perspective when thinking of ES and managing the number of different ES across the landscape. There are thus synergies between the categories, like regulating services
supporting a number of other services (Raudsepp-Hearne et al. 2010). Trade-offs are however
12 also common, in decision making or as unconscious consequences of actions, prioritizing some ES at the cost of reducing the provision of others (Rodríguez et al. 2006)
There may be different kinds of trade-offs; between ES, between ES and other policy
objectives - across spatial and temporal scales, as well as degree of reversibility (Rodríguez et al. 2006), and trade-offs between beneficiaries (Rounsevell et al. 2010).
Biodiversity and drivers of change
ES rely on the structure and function of ecosystems, which depend on biodiversity (Hooper et al. 2006); “the diversity within species, between species and the diversity of ecosystems”
(CBD 1992). How ES are affected by loss of biodiversity is often not known (Kremen &
Ostenfeld 2005, Borgström 2006). It may depend on which species or systems are studied and on what time scale (Walker & Salt 2006), or by resource availability, prevailing patterns of disturbance, and climate (Hooper et al. 2006). In general, an ecosystem experiencing species loss is likely to become less able to deal with disturbance while retaining structure and function (Walker & Salt 2006). Furthermore, it is more likely that an ES will be sustained if there are more species within a functional group (Walker & Salt 2006). Loss of certain species, so called keystone species, may have a larger effect on the whole system and thus its function (Hooper et al. 2006).
There are a number of direct and indirect drivers of loss of biodiversity, and thus ES, at work from the local to the global scale (Table 1). The most important direct driver of biodiversity loss presented is land cover change. According to the MA (2005), there is a need for
substantial changes in institutions, policies and practises to mitigate these, and the report states that these changes do not seem to be underway. Furthermore, the Global Biodiversity Outlook (GBO-3) emphasized that an important reason for the failure by the global
community to stop the loss of biodiversity was that measures were not implemented on a sufficient scale and were not addressing the underlying drivers (CBD 2010a).
13 Table 1: The most important direct and indirect drivers of biodiversity loss pointed out in MA (2005) and GBO-3 (CBD 2010a). Economic drivers include e.g. policy on trade and market regulation, while socio-political drivers are more general policy, including legal tools and institutional structure. Cultural and religious drivers include motivation for behaviour like consumption choices. The urban characteristics related to the direct drivers are drawn from Puppim de Oliveira et al. (2011) and references therein. The columns “Global indirect drivers of loss” and “Global direct drivers of loss” are not directly linked horizontally within the table.
Global indirect drivers of loss
Global direct drivers of
loss Urban characteristics related to direct drivers of loss
Demographic
Land use change
Urban sprawl/habitat fragmentation in urban to peri-urban areas.
Consumption activities by urban citizens can impact remote ecosystems, e.g. by import of larger quantities of timber and agricultural products causing land use change.
Economic
Invasive alien species
(IAS) Tendency of higher number of IAS closer to urban centres.
Socio-political Over exploitation
Urban areas are centres of human consumption activities and urban residents can support trade of a variety of goods from distant ecosystems.
Science and
technology Pollution
Air and water borne pollutants stemming from urban activities or production/transport activities for urban consumption.
Cultural and religious
Climate change
Urban areas are sources of green house gases and cause further emissions through direct/indirect land use change.
Urban heat island effect impact the local climate within cities
Governance
Governance is often contrasted to government, representing a shift in policy and focus of academia (Healey et al. 2002). Government is associated with the state and its formal authority to enforce its decisions (Rosenau 1992) legitimized by elected political
representatives (Healey 2007). Governance entails a change in government and implies the
“creation of a structure or an order which cannot be externally imposed but is the result of the interaction of a multiplicity of governing and each other influencing actors’ (Kooiman & Van Vliet 1993:64). Thus, governance does not exclude government, but does neither rely on its formal authority. The governance literature is mainly focusing on societies commonly
14 referred to as democratic, although this is often implicit and they don’t exclude its validity to other contexts (Lee 2003).
Governance can be viewed as “all ‘collective action’ promoted as for public purposes, wider than the purposes of individual agents” (Healey 2007:17), as well as a network of partnerships (Stoker 1998), facilitating social coordination (Lee 2003). This can include semi-autonomous relationships between the authorities on various levels, the civil society and private sector and its dynamics over time, with partly conflicting and overlapping agendas. This study uses governance as a heuristic term to be open to discovery of different perspectives on the capacities, relations of actors and the dynamics between them (cf. Healey et al. 2002).
The focus on governance may be due to an increased awareness of governments operating in a world with a range of other actors and factors influencing outcomes (Stoker 1998). This includes recognizing the capacity of civil society (Lee 2003), how some actors have more influence than others (Healey 2007), how governments are influenced by actors and dominant agendas at other scales (Marcotullio & McGranahan 2007), how governance outcomes are shaped outside the arenas of public control, and the limits of capacity of the present public institutions (Healey et al. 2002). Which factors influence governance and shape outcomes thus depends on the local context.
In governance of ES the emphasis and normative aim will be conservation, maintenance and enhancement of such, and social equity in regards to this. In practice, governance of urban ES in this study manifests itself in the form of environmental management and planning
activities. While “managers” here may have a mandate to work on conservation, maintenance and enhancement of ES in the urban environment, “planners” may have a wider mandate related to guiding urban development in general (cf. Friedmann 1987).
15
Method
Introduction to method
Considering the scope of the study, a systematic approach followed by a qualitative synthesis, was considered to be a relevant method. Furthermore, since systematic review is a quite recent methodology, which has not been used much in reviews of topics pertinent to this study, this section also provides the background for research question 3.
This section will start by presenting and contrasting this approach to the more widely used traditional review. It will be followed by a presentation of the main principles of grounded theory; the theoretical background for the method developed to synthesize the data found through the systematic review. Finally, the methodology applied in this study will be described and illustrated. The structure and nature of the content of the result section will together with this method section provide the basis for a discussion about research question 3 under discussion related to method.
Literature review and synthesis
A literature review is herein understood as the “whole process of bringing together a body of evidence which can be drawn from research and other sources..” (Mays et al. 2005:S1), while synthesis on the other hand “refer to the stage of a review when the evidence extracted from the individual sources is brought together in some way. This may entail simply juxtaposing findings from different sources or identifying and extracting common themes across sources or integrating data from several sources to produce new insights or theories” (Mays et al.
2005:2).
The traditional review and the systematic review
The traditional narrative review is the most common and often implicit method, and is usually based on a set of articles gathered by the author in a subjective manner, which are critically evaluated and presented in a systematic way to support a narrative, insight or argument (Pai et
16 al. 2004, Jesson et al. 2011). A strength of the traditional narrative review is its flexibility; it facilitates an exploratory, curiosity based study where the door is always open to explore new branches of knowledge (Jesson et al. 2011). A common criticism of this approach is therefore that it is hard for others to know how literature was found, excluded and analyzed before potentially being described in the article or emphasized in the discussion. It is thus difficult for the reader to judge how rigorous and comprehensive the scientific foundation for the review is (Pai et al. 2004, Jesson et al. 2011). It has therefore been argued that such reviews are of lesser value for knowledge support for policy making (Jesson et al. 2011), and with a growing body of knowledge regarding complex research topics, it will be difficult for individuals to stay updated. Such arguments, which I believe are relevant for the study at hand, have been used to promote systematic review strategies based on comprehensive keyword searches in e.g. health science (Pai et al. 2004, Littel et al. 2008).
Jesson et al. (2011) describe how a main distinction between the systematic review and the traditional narrative review is the use of predefined criteria when identifying and examining studies. A systematic review is an approach that “aims to comprehensively locate and synthesize research that bears on a particular question, using organized, transparent, and replicable procedures at each step in the process” (Little et al. 2008:1). It is “A method of making sense of large bodies of information, and a means to contributing to the answers to questions about what works and what does not” (Petticrew & Roberts 2006:2). There are several guidelines available for systematic reviews following the principles presented above.
The Cochrane Collaboration and The Campbell Collaboration are the two most known international, interdisciplinary collaborations that have developed guidelines for systematic reviews and that are performing and gathering such studies in online databases (Littel et al.
2008). A Cochrane review focuses mainly on effects of healthcare interventions, whereas the Campbell review methodology builds on the Cochrane tradition and focuses on effects of policies/practices, but particularly “effects of interventions in social, educational and behavioural areas..” (Campbell & Schryer-Roy 2008:7).
17 A general step by step approach to a systematic review is presented below (based on figure 7.1, Jesson et al. 2011)
1, Scope and map 2, Plan and protocol 3, Document
4, Inclusion and exclusion criteria 5, Search and screen
6, Quality appraisal 7, Data extraction 8, Synthesis
Discussion with policy makers asking for advice and dissemination of results is also
sometimes included since the purpose of most systematic reviews is to aim for evidence based advice to inform policy (Campbell & Schryer-Roy 2008). It should be mentioned that there are also efforts within environmental science to apply the systematic review methodology to strengthen the evidence base to be used by practitioners, focusing on quantitative data and assessing the effectiveness of interventions in environmental conservation (Pullin & Knight 2001, Sutherland et al. 2004).
The distinction between the systematic and the traditional review is a somewhat artificial one;
there are many more variations within the research genre of literature review. The systematic review and the traditional narrative review can, as pointed out by Jesson et al. (2011), be viewed as two end points of a gradient, between the exploratory, not explicitly defined and creative approach and the rigorous, replicable and transparent. Authors use a range of different terms for reviews along this gradient; e.g. the narrative review, content analysis, realist synthesis, the narrative synthesis and bibliometric overview. There is no clear
agreement in the literature on what all these labels tell the reader about the method. The labels and approaches to literature reviews is a part of an ongoing discussion (Jesson et al. 2011).
18 Most systematic reviews are based on quantitative data, though qualitative systematic reviews have become more commonplace within health science the last years (Atkins et al. 2008). It has attained an increased appreciation for its benefits to the progress and evaluation of complex topics (Ring et al. 2010, Karianne Hammerstrøm pers. comm.; Trials search
coordinator - The Campbell Collaboration). There is however no frameworks agreed upon for synthesizing a diverse range of qualitative data and adapt the systematic review principles to the different nature of such studies (Mays et al. 2005).
Synthesis of qualitative data
Systematic reviews are usually followed by a synthesis (cf. the step by step approach shown above), which is the part of this methodology where studies based on qualitative and
quantitative data differs the most. While quantitative data can be merged from different studies for statistical analysis, this is not always possible or beneficial for qualitative data drawn from studies that are not fully comparable. A qualitative synthesis in contrast, is
defined by (Sandelowski et al. 1997:366) as “the theories, grand narratives, generalizations, or interpretive translations produced from the integration or comparison of findings...”. Ring et al. (2010:2) provide a guide for how to follow up a systematic review with synthesis of qualitative research in medical science, and point out that it is possible that “..the relevant qualitative studies on the same topic can identify a range of common themes as well as any divergent views”. Qualitative data can contribute to a richer understanding in text analysis, like stated in the guide referred to above; “Such information can be particularly useful in understanding the barriers and facilitators to the delivery or implementation of new
technologies in practice” and understanding why some measures work better for some groups (Ring et al. 2010:2).
Qualitative synthesis in systematic reviews draws on the tradition of qualitative studies, like grounded theory (Ring et al. 2010), one of several approaches to text analysis and synthesis (Miles & Huberman 1994, Ring et al. 2010). The core of the synthesis is exploring
relationships and finding connections between studies, adding knowledge beyond each individual study (Hart 1998). Furthermore, qualitative synthesis does also differ from quantitative in systematic reviews in that text analyzed often are influenced by a set of
19 concepts/theories or a philosophical position (Ring et al. 2010). This can be explicit or
implicit in the particular study. Grounded theory is one approach to analyzing such diverse body of research evidence when re-assembling selected data (Ring et al. 2010), that forms the methodological backdrop for the synthesis in this study.
Grounded theory
Grounded theory is in essence an inductive process. Texts are carefully investigated (through a process called microanalysis) to see “what is going on” (Strauss & Corbin 1998:114) and selected text of interest are labelled as more general phenomena/central analytic ideas separated into more abstract categories of suitable concepts with suggested relationships (open coding). Concepts often flourish in the beginning of a study and the researcher is encouraged to make sense of the diversity trough a process Strauss & Corbin (1998) refer to as conceptual ordering, where concepts can be elaborated and delineated from other
categories in terms of their properties and dimensions, representing how phenomena varies within them. Here the aim is to move beyond description to “abstracting, reducing and relating” where the researcher has a more conceptual lens trough which she/he views the data (Strauss & Corbin 1998:66). Phenomena can later be coded into subcategories related to a particular concept code, through a process called axial coding (Strauss & Corbin 1998).
Grounded theory is an iterative process of constant comparison of the selected texts, where categories emerge rather than being based on a pre-designed set based on
expectations/theories (Glaser & Strauss 1967). Grounded theory, as presented in Glaser &
Strauss (1967) and Strauss & Corbin (1998), is however developed mainly focusing on primary studies or cross-case comparison of such.
I considered the principles and main steps recommended in systematic review approaches as pertinent for building up a knowledge base about challenges and opportunities of urban ES found across the literature to aim for systematic, broad coverage of literature across scientific traditions. That being said, it will be impossible and not always fit for purpose to rigorously try to follow all the requirements of any of the distinct approaches to answer the research questions of interest. The study at hand will thus not be a systematic review or grounded theory synthesis that follows any premade fixed review criteria. It will (cf. research question
20 3) be an exploration of how research on a different complex problem, governance of ES, can draw on the existing experience from the above mentioned approaches using the approach described below.
The method designed
A step by step approach was designed in three parts as presented below and elaborated further down in this section. Before the following process research questions and scope of the study was decided.
Systematic search
1, Two main searches were performed. First with combinations of two key word categories (category 1 and 2 table 2) in addition to “governance”. Secondly with two key word
categories in combination with “ecosystem service” (Category 2 and 3 Table 2).
2, All searches were performed in two databases; SCOPUS and ISI Web of Science (ISI WS).
Critical appraisal
3, All abstracts found were read to consider relevance to the research questions.
4, The list of included articles were presented to leading researchers and policy makers within the field to make sure that there were no “landmark-articles” missed.
Synthesis
5, The articles that were considered to be relevant were read.
6, Notes were taken from each of the articles and quotes were sorted as either “governance challenges” or “governance opportunities”.
7, The quotes were interpreted into phenomena
8, The phenomena were coded into concepts and subcategories through an iterative process.
9, These findings created a foundation for discussions with policy makers in the field of local authorities, urbanisation and biodiversity involved in the preparation process of the CBO based at the secretariat of the CBD.
10, A presentation of the results of the coding process was written according to the content in each of the delineated concepts and subcategories.
21 11, Dissemination of results. The findings were included in the report Cities and Biodiversity Outlook.
Systematic search
The categories of words used in the search aim to represent the three main focus areas of the study; governance – of ecosystem services – in an urban setting (Table 2).
Table 2: The table presents the words chosen for each of the search word categories generating the literature reviewed in the study.
Category 1 Category 2 Category 3
Ecosystem services Urban Governance
Ecosystem Cities Planning
Biodiversity Metropolitan Management
Green Institutions
Resilient
Category 1 included several complementary words due to “ecosystem services” being a quite recent term used in science (Dick et al. 2011). Words used in different theoretical traditions were also considered and led to the inclusion of “resilient” and “green”, the latter to include planning literature. All combinations of category 1 and 2 were used the chosen databases together with “governance” from category 3 in the first search. In the second search all words combinations of category 2 and 3 were used in addition to “ecosystem services” from
category 1. The same search was done in ISI WS and SCOPUS to avoid database bias. These databases were chosen as they were considered to include a broad range of journals, thus all searches were done in all journals available to the respective databases. It was aimed for high sensitivity by not excluding any of the journal categories in the search settings to find studies from different theoretical traditions. The rationale behind this choice was the belief that when addressing complex issues like the one at hand, insights may evolve through the interaction of disciplines and moving beyond the natural-social science division (Bayá Laffite 2009). The selected words were searched for using the Boolean logic term “AND” and “OR”. In ISI WS the search was done under “topic” while in SCOPUS under “Article title, abstract, keywords”.
Both the databases searched for the selected keywords in abstract, title and keywords
22 described in the articles. Both checked singular, plural, and possessive forms for all keywords entered (ISI Web of Knowledge 2012, SCOPUS 2012). The search was performed without time limitations. In SCOPUS both “article and review” was included. This choice is not included in the ISI WS. Only peer reviewed articles was included in both databases, which is not in line with the Campbell review, which argues that this strategy can lead to a publication bias (Hammerstrøm et al. 2010). This was however a norm for the CBO scientific foundation report. Searches were documented throughout the process; dates, databases used, keyword combination, number of articles found and titles of articles for later reference.
To summarize, after this step I had a number titles and abstracts of articles based on the chosen combination of key words used in the two databases. The next step was to evaluate these.
Critical appraisal
Definitions of key concepts were clarified before evaluating articles (c.f Introduction). When evaluating abstracts, the criteria for inclusion was whether the study focused on governance of ES in cities; by urban actors, or actors cooperating with urban actors, for conservation,
maintenance or enhancement of concrete ecosystem services. The article did (c.f. the use of different keywords) not need to use the term “ecosystem service”, but a direct link to at least one such service, including challenges and opportunities related to governance of it, should be addressed in the text. It was not considered sufficient if an article described one or more ES or urban environmental characteristics in detail, but ended with a general statement about the need for action without describing concrete measures (opportunities) or challenge.
An example would be that an article documenting pollution in an urban river would not be sampled if it does not also investigate possible governance opportunities/challenges related to this. If in doubt after considering abstracts by the criteria, then the main text was searched through for more information, if it was still not clear then the text was read. This study did not rank or exclude any studies due to an assessment of quality since there is a large debate with no consensus in the literature on how to use such criteria in qualitative studies (Sandelowski 1997, Mays et al.2005, Atkins et al. 2008, Jenkins et al. 2011, Nagata et al. 2011). The final list of articles was sent to my supervisors at Stockholm Resilience Centre (Thomas Elmqvist
23 and Cathy Wilkinson) and my supervisor at the secretariat for the CBD where I was based (Oliver Hillel) for possible inclusion of missed landmark studies.
To summarize, after this step I had a number of articles considered to be relevant to the topic of study. The next step was to read and synthesise the relevant information from these.
Synthesis
The first step of the analysis was to sample text representing challenges or opportunities from the included articles for further coding. Challenges and opportunities, both stated explicitly in the literature by the authors and after careful evaluation of problems/barriers presented, were sampled.
A challenge is here interpreted as a barrier that hinders protection, enhancement or
maintenance of ES. Challenges often manifest themselves related to particular measures while others are more general. I did not consider whether I believed that the concrete measures planned were effective or not in relation to the ES targeted. A challenge could include an opportunity and vice versa. When this was the case, the text was coded as both, and different aspects of the text were included in codes. A remark was made when particular opportunities and challenges were related in a text, e.g. a concrete strategy had particular challenges related to it.
Opportunities represent outcomes of the type aimed for in systematic reviews in medicine;
“what works” (cf. definition by Petticrew & Roberts 2006). However, in the literature this study is based upon the opportunities are not that clear-cut. There was hence a need to specify different categories of opportunities regarding the evidence base. One opportunity category was measures or tools that have been implemented for governance of ES, which do not need to be seen as ideal as long as it is considered to have positive effects. The second category was governance recommendations based on a case study or review of experiences from case studies, which have not been implemented. Finally, there are opportunities that are based on theoretical explorations based on challenges noted from case studies or arguments of
24 challenges that might emerge in the future. Opportunities sampled were thus categorized both according to content related concept codes, and to evidence base.
Table 3: The asterisk system used to separate the three levels of evidence for sampled opportunities.
Sign marking references in the result section
Description of chosen categories of evidence base
*
Opportunity measures or tools that have been implemented for governance of ES with a degree of success. This includes also direct references to concrete implementation, but was not investigated in detail in that given study (including both case studies and reviews).
**
Governance recommendations based on experiences from a case study or review of such, but that has not been implemented.
***
Opportunities that are based on theoretical explorations of tools or approaches based on challenges from case studies or arguments of challenges that might emerge in the future, but that is not related to a specific case.
The selected text were mined for meaning to be understood in more abstract terms, to see what kind of general phenomena they represent, seeing over time how the different pieces of selected text could be related in concepts codes and subcategories (see examples of quotes, phenomena, concept categories and subcategories in table S2, Appendix A). The synthesis thus becomes a result of constant comparison; between articles and the data selected from them, and within and across concept codes. These comparisons, which were recorded continuously, also lead to awareness about similarities, variation and gaps. As an increasing number of articles were read, patterns of the phenomena in codes and subcategories started to emerge, and was further refined. It was considered important to not “force” summaries into categories, but to see if patterns emerged over time. If it was not clear what phenomenon a text was an instance of, or what concept or subcategory a described phenomenon represented,
25 a decision was made to wait and return to it at a later point. Some pieces of text were coded into several categories as they represented more than one concept (see example in table S2, appendix A). In such instances links were noted in the coding table. Over time it became possible to see how different concepts were linked according to my interpretation of the literature.
Half way through the synthesis I presented preliminary results to colleagues at the secretariat for the CBD (SCBD) to get their comments. I had one discussion with Oliver Hillel
(Programme Officer, Sustainable Use, Tourism and Island Biodiversity), and one with Andre Mader (Programme Officer, Local Authorities and Biodiversity) and Chantal Robichaud (Programme Assistant). The intention was to see how the results matched their experience, i.e.
how well the method worked and if potential gaps in the results could be found (cf. research question 3).
Additional information
In addition to challenges and opportunities, some additional information was collected that was considered to be potentially relevant for further consideration. This is referring to;
geographical focus, scale addressed, actor/s in focus and finally if any particular ES was in focus. It was also noted if the study was a case related to a particular location or a comparison of such or if it was a more theoretical exploration or review. Reflections on the designed method and evidence base of the study can be found in the discussion.
26
Results and synthesis
This section will present the results of the systematic review by presenting a search report, before presenting the results from the synthesis corresponding to research question 1 and 2;
the main challenges and opportunities.
Search report from systematic review
The first round of searches generated 219 articles in SCOPUS and 192 in ISI Web of Science, of which in total 158 and 106 articles, respectively, were considered potentially relevant based on abstracts. After removing overlaps, excluding articles not available in English or in full text version, and excluding articles that were not relevant based on inclusion criteria when evaluating the full text (cf. method), 55 unique titles were included for analysis. The second round of article searches generated 275 articles in SCOPUS and 35 in ISI WS of which 184 and 27 articles were considered to be potentially relevant based on abstracts. After the same exclusion process as above, in addition to removing overlap with the first round of key word search, 83 new articles were included, resulting in a total of 138 articles being reviewed from 76 journals (Appendix B). The number of studies included from SCOPUS and ISI WS, number of unique studies form each, and percentage overlap can be found in table S3
(Appendix A). Geographical representativeness was approximately the same before and after applying exclusion criteria. The searches performed (round 1 and 2 ) resulted mostly in articles written in English, articles in other languages, like Mandarin and Spanish, were excluded. Included studies were grouped based on geographical area investigated; a summary of countries and cities being covered in these articles is shown in table S4 appendix A. Totally 88 cities or urban regions from 23 countries were represented in studies. There was a clear bias towards Europe (32 studies from 27 cities/urban regions from 9 countries), North- America (28 studies from 26 cities/urban regions in USA and Canada) and China (22 studies of 11 cities/urban regions). In addition to these studies there were also some studies looking at a large number of cities within a given country, e.g. studying land use change response to policy. Most of the studies retrieved were case studies. The oldest study included is from
27 1999, but the majority of studies are from 2005 or later, after which there has been rapid growth in studies on the topic at hand (figure 1). In total, only seven of the studies were from before 2005. Africa, South America and parts of Asia are to a large extent missing in the included literature. No articles were added after sending the final list of articles to my supervisors. Table 4 present the titles of the concept codes and subcategories of challenges and opportunities found. The order of challenges and opportunities do not reflect importance.
Figure 1: Number of publications sorted by year for studies included in the literature review.
0 5 10 15 20 25 30 35
28 Table 4: Concept codes and subcategories of challenges and opportunities.
CHALLENGES: concept codes subcategories
Lack of knowledge Lack of scientific knowledge and tools
Weak link between existing scientific knowledge and policy
Lack of awareness of nature and ecosystem complexity
Equity Access
Disparities in who cause and pay for environmental degradation
Lack of institutional capacity Lack of appropriate strategies and regulations Lack of implementation of strategies and regulations
Ability to acquire and handle information The response does not match the urgency or seriousness of problems
Cooperation between people in different departments or levels of organization
Constrained public budges and the involvement of private actors
Growth first No subcategories
Conflicting views: balancing management objectives No subcategories Challenges related to scale mismatch No subcategories Challenges related to trade-offs No subcategories
OPPORTUNITIES: concept codes subcategories
Planning and management - principles and
approaches Green networks and protection of larger green areas
Restoration, complementation and reconciliation Green innovation
Planning and management tools Regulation of land use Planning tools
Economic instruments and valuation tools Civil society – a source of knowledge and management
capacity
No subcategories
Learning Learning in civil society
Leadership Learning among decision makers
Ecological citizenship No subcategories
29
The main challenges found for governance of urban ES
This section presents the synthesised challenges according to concept codes, which will be described for clarity where necessary, followed by results in subsequent subcategories as it was coded. The coding process with examples of quotes from articles, coded phenomena, subcategory, and concept code is exemplified in table S2, appendix A. Concept codes should not be viewed as isolated entities of phenomena, and some phenomena are coded into more than one concept code. Two of the concept codes, challenges related to scale mismatch and trade-offs, are overarching challenges meaning that they are common in the literature, but only coded in addition to one of the other concept codes. The category “lack of institutional capacity” showed the greatest amount of overlap with other categories, most notably “growth first” and “lack of knowledge”. Examples of texts that were double coded can be found in table S2 (appendix A).
Lack of knowledge
This concept entails both scientific and non-scientific knowledge, its interaction and its influence on policy. Scientific knowledge includes peer reviewed research and research tools relevant to inform policy about governance of urban ES. Non-scientific knowledge includes knowledge held by practitioners and people interacting with nature on a regular basis.
Lack of scientific knowledge and tools
This subcategory is focused on a need for research for knowledge support for protection, maintenance and enhancement of ES to be used by other researchers, planners and all other relevant stakeholders. The underlying concern is that lack of scientific knowledge may lead to a lack of knowledge among decision makers and the public as to how and why governance actions should address ES and ecosystem integrity. As has been mentioned previously, there is a lack of knowledge about ecosystem structure and function (see also Boyer & Polasky 2005). This shortage found in the literature concerns in particular knowledge about urban ecosystems (Niemelä et al. 2010), like the functioning of supporting services e.g. soil processes and nutrient cycling (Lal & Lorenz 2009) and the challenge for planners and
30 decision makers of dealing with scientific uncertainty (Fang et al. 2006, Niemelä et al. 2010, Su & Fath 2012). There is a lack of local ecological data needed to support regional planning and policy development (Peterson et al. 2007), like calibrated approaches to methods for urban planning focusing on biotopes/habitats (Mendiondo 2008), ecosystem valuation (Boyer and Polasky 2005), and for considering options and uncertainty in light of climate change and varying scenarios (Niemelä et al. 2010). A broadly covered challenge is the lack of
appropriate scientific tools and methods to capture the complexity, and limits of ecosystems, and influencing drivers of change (Merson et al 2010, Puppim de Oliveira et al. 2011); e.g.
methods to quantify ecosystem integrity and regulating services at the landscape scale (Kroll et al. 2012), valuation techniques (Li et al. 2010a), indicators that go beyond quantity to measure the quality and variety of ES provided (Xu et al. 2011) also across scale (Mendiondo 2008), and indicators for assessing urban ecosystem state (Su and Fath 2012), like water environment criteria (Meng 2009). There is a need for a variety of tools adapted to the urban context (Borgström et al. 2006b), as was shown by Tratalos et al. (2007) in their study of change in ecosystem performance related to urban densification. Lundy and Wade (2011) point out that transdisciplinary research is promising for addressing urban ES challenges in a holistic manner, but that it remains a challenge that this field is not as respected and well funded as disciplinary research.
Weak link between existing scientific knowledge and policy
It is challenge to make sure that relevant decisions are informed by existing knowledge about ES (Alonso & Heninen 2011). Despite having access to relevant knowledge, it may take time before this has an effect on policy, public awareness and political action (Lieberherr-Gardiol 2008). Niemelä et al. (2010: 3238) found in interviews with land-use planning professionals in Finland that the ES concept was perceived as “abstract and too complex” for concrete application.
The outcome of planning processes depend on the existing knowledge and the planners’
ability to make use of it, but also the perspectives of local decision-makers and the views of local stakeholders (Alfsen-Norodom et al. 2004, Yli-Pelkonen et al. 2006, Bayá Laffite 2009).
In a study from New York the connection between science and policy was found to be weak
31 due to the scientific view being just one of many stakeholders involved in decisions (Alfsen- Norodom et al. 2004). According to Bayá Laffite (2009) science and the views of stakeholders should not be viewed as in conflict, but as a potential for knowledge co-production. It may however be a challenge for academia to interact with non-scientific knowledge (Bayá Laffite 2009).
Lack of awareness of nature and ecosystem complexity
There is a lack of awareness among people in general, but also decision makers and planners, about the diversity of nature, its complexity, as well as human dependence on ecosystem functions across scale. It is therefore a challenge to mainstream awareness about the ecology of and in cities to achieve political support and changed habits (Wolch 2007).A personal experience of nature is important for caring about its protection, and through this, contribute to governance measures for protection, maintenance and enhancement of ES (Dearborn &
Kark 2009). This challenge relates specifically to urban dwellers who spend less time outside or do not have access to green areas (Dearborn & Kark 2009), but goes beyond that as people more generally lack a connection to the environment they draw their resources from
(Borgström et al. 2006b). People do in many cases not rely directly on local ecosystem functioning (Borgström et al. 2006b) and are often not aware of how their behaviour affect ecosystem services provided outside cities (Puppim de Oliveira et al. 2011). This may lead to a limited understanding of ecological processes and complexity (Borgström et al. 2006b) and, at the local scale, limited understanding of both existing and potential ES from urban
ecosystems (Jim & Chen 2006), like the value of conserving peri-urban agriculture (Merson et al. 2010). In a study byJim & Chen (2006:342) in Guangzhou (China), residents that were asked to rank different ES from urban green space, placed high values on services like air quality and aesthetic enhancement in contrast to facilitation of water treatment and flood abatement. The authors suggested that this could be related to the fact that wastewater in that area so far has been treated through a technical water facility and flood has not been a
problem. In addition, they also found that people ranked contact with nature quite high, while placing low values on the service “wildlife habitat and species conservation”, displaying a potential lack of system understanding. Environmental education activities are rarely focusing on the urban landscape (Dearborn & Kark 2009).
32 Several studies highlight the lack of awareness and narrow understanding of ecosystem
functioning among decision makers. On the local scale, this has been linked to the failure to recognize multiple values of trees (Moll 2005, Li et al. 2005b), whereas on the regional scale, it has been related to peri-urban land use change and the value of conserving agricultural land (Merson et al. 2010). Additionally, decision makers may not recognise the global ecological impact of cities, limiting ES governance to local green spaces (Puppim de Oliveira et al.
2011). Planning practices based on a view of nature as predictable and generally stable, with a fixed number of species to be managed in designated areas, is a challenge as it may fail to respond to change (Asikainen & Jokinen 2009), and understand species movement across relevant scales (Mendiondo 2008). This finalizes the results on lack of knowledge.
Equity
The concepts relates to the balance of benefits compared to burdens (Greenberg & Cidon 1997) and the “distribution of both material and nonmaterial benefits derived from public policy that does not favour those who are already better off at the beginning” (Fainstein 2010:36).
Access
Access to ES is often not equally distributed within the city (Li et al. 2005a), low income/minority groups tend to have lower access than their more affluent counterparts (Wolch 2007, Perkins 2010). In Leeds, While et al. (2004) found that projects leading to environmental degradation tend to be located where people are less vocal in political debates.
Dependence on land trusts, private donations (Warren et al. 2011) and volunteers (Perkins 2010) for access and quality of green areas may intensify such inequities as some
communities do not have the capacity to compensate for lack of public support. Efforts to increase urban centre attractiveness to urban dwellers and investors, and to reduce sprawl, can lead to increased access to ES in traditional deprived areas, but also to increased housing prices (c.f. table S1, Appendix A) and a risk of gentrification/displacement. This has been documented in Vancouver (Quastel 2009, Hutton 2011) and Portland (Hagerman 2007), which often are perceived as front runners in progressive urban development. Policies for reducing peri-urban land use change through densification may be ineffective due to a relative
33 small group of urban citizens that prefer low density housing and is responsible for a large impact on loss of natural habitat, as was shown in a study of 274 metropolitan areas in USA (McDonald et al. 2010).
Disparities in who cause and pay for environmental degradation
People who have a higher per capita responsibility for degradation of ES are often not the ones experiencing the cost. Costs related to environmental degradation leading to quantitative or qualitative loss of ES may be displaced across temporal and spatial scales. Most articles focus on distribution within the urban area, where a small group of urban citizens may have a higher per-capita responsibility for loss or degradation of ES, while the costs of this loss is carried by the public (McDonald et al. 2010). Powerful actors often dominate urban planning, especially in implementation (Hagerman 2007, Peterson et al. 2007, Asikainen & Jokinen 2009). Conflicts between housing requirements and concerns of land use change are often paralleled by issues of equity. Cape Town is an example of a city where rapid urbanisation has led to increasing social inequity where settlements of poor communities are constructed in areas vulnerable to natural disturbance further undermining the natural protection of these areas and access to other ES, while wealthy and spacious suburbs are growing elsewhere (Erntson et al. 2010a). In the Rochester region in USA, where the city core is characterized by a predominantly African-American population while the suburbs are dominated by white residents, Wekerle & Abbruzzese (2010) found that attempts to raise awareness and coordinate efforts to reduce sprawl (promote smart growth policies) had little effect.
Furthermore, they also pointed out that it might be particularly challenging to implement progressive planning solutions securing ecologically valuable land in urban regions with weak economies, as actors may be less willing to give up benefits for the common good. A similar case can be seen in several Indian cities, where low income groups are being sidelined in processes related to access to and conservation of ES, in regards to e.g. hindering traditional practices of sustainable use (D`Souza & Nagendra 2011) or displacement (Zérah 2006). The perception that poor people as responsible for environmental degradation in spite of their situation and relatively low impact is a challenge (Zérah 2006, D`Souza & Nagendra 2011).
As pointed out above, inequity also occurs between urban and rural regions, e.g. where urban residents are suffering from lowered water quality due to externalities from upstream land
34 management (Sarker et al. 2008), or the opposite, when urban residents draw on provisional services and nature conservation from rural hinterlands, leaving a marginalized rural
population and degraded environment due to inappropriate payment for these services (Gutman 2007). Following globalization, equity is not merely a local or regional issue; the social and ecological costs of improved urban living conditions, in particular in what is often referred to as developed countries, but also for wealthy urban dwellers elsewhere, are
transferred through global trade flows (Hagerman 2007, Meng 2009). There is an increasing body of literature from Chinese cities documenting severe degradation of ecosystems and ES by a variety of production activities for global consumption (e.g. Meng 2009, Wang et al.
2009).
Lack of institutional capacity
Lack of institutionalcapacity is a comprehensive category. Institutional capacity is here a matter of challenges related to formal authority and structures ability to plan and regulate governance of ES including the ability of such structures to acquire and handle relevant information and cooperate across levels of decision making.
Lack of appropriate strategies and regulations
National orinternational levels of governance have often not focused on cities in policies on governance of ecosystem (Puppim de Oliveira et al. 2011). The literature reviewed does however focus more on challenges related to lack of appropriate strategies and regulations at the local to regional level. Biodiversity conservation is not mainstreamed across urban authorities (Puppim de Oliveira et al. 2011). In China, an important challenge has been the lack of legal tools and plans for governance of ES since the central planning system was developed before decision makers had any significant awareness of the value of integrating environmental concerns into urban planning (Fang et al. 2006, Xu et al. 2011). The value of ES may also be underestimated in the economic system (Gaodi et al. 2010), and compensation mechanisms may fail to protect green areas from real estate development if the fee developers must pay to build on green areas is significantly lower than the income prospects (Li et al.
2005a). Strategies and regulation may not take the full range of ES or its relevant scale into account in urban development and land use planning (Lucero & Tarlock 2003, Bayá Laffite 2009, Huang et al. 2011). Provisional services can be prioritized and their status regulated,
