Space for mainstreaming? Learning from the implementation of urban forest strategies in metropolitan Melbourne

ABSTRACT Australia is experiencing an accelerated rate of climate-related extreme weather events, and many of the solutions to reduce the exposure to climate-risk are nature-based, governing urban forests, waterways, and stormwater. However, the governance of nature-based solutions in Australian cities is still fragmented and piecemeal, generally lacking a coherent narrative and widespread support. What is needed are institutional spaces that mainstream such solutions. In this paper, we draw on a case study of urban forestry implementation across metropolitan Melbourne, as a lens to examine the creation and evolution of such institutional spaces. We explain the functions and design characteristics of institutional spaces from the perspective of the requirements for establishing and maintaining institutional spaces and what is produced or the outcomes from institutional spaces. The mobilisation and evolution of institutional spaces are important to understand for the impact on the planning and governance of individual cities as well as the metropolitan region. Our key findings frame institutional spaces as relational, learning-oriented, collaborative, and empowering spaces that facilitate transformative agendas and actions for the mainstreaming of nature-based solutions in cities. From these findings, we identify seven recommendations for how practitioners can make the most of institutional spaces. Practitioner pointers Creating space to bridge silos, foster experimentation, and develop evidence-based policy is critical to mainstream nature-based solutions. Collaborative approaches are essential for effective institutional spaces, to participate in networking and knowledge co-production opportunities. Actors in institutional spaces facilitate mainstreaming by learning from and building on policy and practice legacies.


Introduction
Climate and ecological crises expose the vulnerabilities of human and natural systems, and highlight in a bleak manner, the consequences of inertia and inaction in adapting cities and their governance.The need for transformative action in cities to respond to climate change and biodiversity loss has been communicated clearly in recent international scientific reports (IPBES 2019;IPCC 2022).This requires an update to business-as-usual institutions so that they instead can facilitate, navigate, and promote the change that is needed.In this paper, we introduce an important piece of the puzzle for enabling this change: the opening up of 'institutional spaces' that embed the social dynamics for change.
The concept of 'institutional space' is still emerging, but has been used both explicitly and implicitly within the academic fields of policy studies, in relation to policy mixes (Edmondson, Kern, and Rogge 2019;Howlett and del Rio 2015), policy formulation (Salo et al. 2022), organisational studies (Droege and Marvel 2010), as well as sustainability transitions (Frantzeskaki et al. 2020;Frantzeskaki and Bush 2021;Frantzeskaki, Avelino, and Loorbach 2013;Lockwood 2022;Novalia et al. 2018;Pesch 2015;Rogers et al. 2015;Savini and Bertolini 2019).It is important to study institutional spaces because they enable how actors pursue, activate, and accelerate change agendas and pathways by carving out the space to deliver transformative action.
In this paper we outline recent research results on local government urban forest strategies in metropolitan Melbourne, as an example of what we refer to as mainstreaming of nature-based solutions (NBS).Within these efforts we focus on the design characteristics and functions of institutional spaces.This is done by drawing on and re-framing a socio-technical transitions perspective, and qualitative case study research relating to urban forest strategies in metropolitan Melbourne.Specifically, this paper provides a deeper understanding of how institutional spaces that support climate-adaptive action can evolve and their impact on urban planning.Therefore, in the discussion we also identify seven planning recommendations for how best to create and utilise the benefits of institutional spaces.

Background to (re)conceptualise institutional spaces
As is usually the case in the social sciences, it is important to develop a precise and exact language to describe the complex processes of human interactions.Without this language, it is impossible to adequately describe something as nuanced and dynamic as institutional space.Therefore, we here set the scene of our study by delving into key theoretical concepts and notions that are required in order to explain and discuss our study results, to explain how the mainstreaming of NBS in cities can be facilitated by this (re)conceptualisation of institutional spaces.

Governance of and with urban nature-based solutions
We focus on the concept of NBS, specifically from a governance perspective and how institutional change can happen.
The International Union for Conservation of Nature (IUCN) defines NBS as 'actions to protect, sustainably manage and restore natural or modified ecosystems, which address societal challenges (e.g., climate change, food and water security or natural disasters) effectively and adaptively, while simultaneously providing human well-being and biodiversity benefits' (Cohen-Shacham et al. 2016, xii).The implementation and governance of urban NBS is therefore increasingly important as a key complement, to challenge the unsustainable grey infrastructure that dominates the urban fabric, and to improve liveability outcomes (Frantzeskaki and McPhearson 2022).
Urban NBS are solutions that support climate change adaptation, for example by mitigating urban heat island effects, while simultaneously providing multiple co-benefits for human health and wellbeing (Cohen-Shacham et al. 2016).Therefore, urban NBS can play an important role in transforming how urban infrastructure services are provided (Dorst et al. 2021;2022).
Urban forestry strategies, their introduction and implementation, are examples of urban NBS projects and programmes.This recent recognition of the contribution of urban forestry for urban climate resilience, as cities increasingly adopt such projects and strategies, can therefore build from the extensive research on its characteristics, design, and benefits it can bring to cities (Davison and Kirkpatrick 2014;French 1975;1983;Johnston 1989;Moore 1993).Whilst urban forestry is not new, especially in relation to arboriculture and tree management (Davison and Kirkpatrick 2014;Moore 1993), we consider its systematic use for achieving climate resilience and human well-being a recent addition to urban climate resilience governance and conceptualise it as a niche innovation.Moreover, governance processes and dynamics have been identified as a key gap in the urban forestry literature (Lawrence et al. 2013;Ordóñez et al. 2019), and we therefore draw on literature that examines the governance of and with NBS in cities. Implementation of urban forestry strategies, as examples of NBS, thus require institutional spaces to allow for experimentation, and for its future adoption through configuring new relations and policy co-ordination with considerable potential for a broader planning transformation.
The mainstreaming of NBS challenges status quo institutions in places like Melbourne.This is because mainstreaming processes require governance to incorporate social-ecological values and benefits (Frantzeskaki et al. 2022) as well as participatory and inclusive approaches (Gulsrud, Hertzog, and Shears 2018).This means that approaches for the governance of NBS need to be innovative to plan, govern, and engage stakeholders to design and deliver urban NBS (Gulsrud, Hertzog, and Shears 2018;van der Jagt et al. 2020).As Kabisch, Frantzeskaki, and Hansen (2022) discuss, urban NBS need to be understood as integrated solutions that are delivered through systemic approaches in governance, which have novel designs to provide benefits for humans and non-humans, and that elevate justice in planning discourses, which are place-based, and are a way to facilitate social learning.
In this way, urban forest strategies across metropolitan Melbourne, as a case study, provide a general overarching and consistent setting to analyse how such nature-based transformation may be envisioned, planned, and achievedthrough the lens of institutional spacesfor improving urban climate resilience outcomes.

Framing sustainability transitions
The multi-level perspective is a core concept within the now widely adopted socio-technical transitions theory.According to this theory, socio-technical transitions occur across three nested levels: niches, regimes, and landscapes.The landscape refers to the external context of a system (slowest to change); the regime refers to the rules of the game (incremental change); and the niche refers to small innovations that are protected from market forces (radical innovation) (Geels 2002).We draw on this theoretical background to describe elements of transition processes, to develop how institutional change can be conceptualised beyond the traditional socio-technical framing of much transitions literature.
In this paper, we focus specifically on urban forestry as an innovation happening in the niche-level, as described in socio-technical literature (Raven 2006;Smith and Raven 2012).Therefore, urban forestry can be considered a niche intervention, in other words, 'a deliberately created niche to alter and improve the system' (Doyon 2018, 68) that is emerging and operated mostly by local governments.Furthermore, we consider niches as protected spaces 'for path-breaking innovations' (Smith andRaven 2012, 1025), i.e., that they are purposively created for trialling and implementing innovative policies and practices.Within such niche-level actions, what we refer to as institutional spaces are emerging and benefit from social-political shields from outside interference and risks, and thus provides enabling environments for innovation: (1) to be established and nourished, (2) the capacity to evolve, and (3) to be empowered by adapting institutions so it can be widely adopted (Smith and Raven 2012).In this way, niche spaces help to facilitate transformative change in institutions and systems (Smith and Raven 2012).This means that such 'niche spaces' can be enabled to succeed within the existing institutional context (i.e., actors exploit opportunities within the regime; to fit and conform) or because existing institutional context evolves or is forced to change (i.e., actors deliberately and strategically mobilise innovation to create or re-shape institutional space to succeed within; to stretch and transform) (Smith and Raven 2012).
Drawing from this theoretical background, in this paper, we frame urban forestry as an urban governance experimentation, occupying a niche space.This is important to clarify, as although there is momentum of urban forest strategies across metropolitan Melbourne, it still occupies a relatively restricted space within urban planning and development priorities, or pockets of change, and not a fundamental shift in urban infrastructure regimes (Dorst et al. 2021;2022;Tozer et al. 2022;Xie, Bulkeley, and Tozer 2022).

What is mainstreaming?
We define the process of mainstreaming (a key term for our research) as a transformative agenda in the planning practices and policies, decision-making, and politics of cities. Mainstreaming is underpinned by the purposeful and cumulative actions of multiple actors to enable, design, connect, and implement innovative solutions to enact climate and ecological adaptation strategies for the resilience of cities (systems) and the sustainability of (re)creating social-political dynamics for institutional change (Adams, Frantzeskaki, and Moglia 2023).

What are institutions and institutional context?
To understand institutional spaces, we need to first understand the meaning of institutions.Here, institutions are framed as 'clusters of rights, rules, and decision-making procedures' (Young 2008, 7) and impose the rules of governance to 'structure and coordinate social interactions across a number of settings' such as laws, regulations, and policies (Matthews 2012(Matthews , 1100)).In the literature on metropolitan institutional change, this has primarily been framed from the perspective of formal government restructuring to establish metropolitan authorities which can formally change the political organisation of a city (region) (Kübler 2012).However, in the context of Australian cities, this top-down mode of governance for the metropolitan region is not how metropolitan-wide governance is emerging.Instead, metropolitan-wide (informal) platforms are emerging, such as Resilient Melbourne (Frantzeskaki and Bush 2021), Living Melbourne (Bush, Coffey, and Fastenrath 2020;Coenen et al. 2020;Fastenrath, Bush, and Coenen 2020;Hartigan et al. 2021), and Regen Melbourne (https://www.regen.melbourne/).We refer to these metropolitanwide platforms, and how they interact with city-level governments, as examples of institutional spaces.
This emergence of metropolitan-wide institutional spaces that promote urban resilience, specifically urban NBS through urban forestry agendas and actions emphasises the necessity of new governance arrangements for the management of transformative stressors in metropolitan planning, such as climate change (Matthews 2012).Matthews (2012Matthews ( , 1100) ) further conceptualises these transformative stressors 'by their capacity to compel institutional change even in situations where resistance from institutional actors is present, or where institutional capacity may be limited'.The institutional spaces at play, mobilised by multi-actor collaborations, in cities and metropolitan regions are the innovative, niche actors that create institutional spaces to respond to these stressors and create conditions or proposals for institutional change.We, therefore, position our study in the shift away from technologically centric solutions towards NBS as innovations in sustainability transitions discourses (van der Jagt et al. 2020).

What are the functions of institutional spaces?
We identify two key themes from the literature for understanding the functions and characteristics of institutional spaces.These themes highlight the social-political factors that enable multi-actor collaboration that produces transformative impact and outcomes for nature-based climate adaptation in cities.
Firstly, institutional spaces are agency-based, which is sometimes called institutional work.Institutional work focuses on how agency influences institutional change by understanding the micro-dynamics of institutional change and how institutions are maintained over time (Patterson and Beunen 2019).Institutional work therefore elevates the importance of deliberative or discursive agency to create, maintain, and disrupt (in)formal institutions (Patterson and Beunen 2019;Rogers et al. 2015).This is important for understanding what power or influence niche actors have to create and/or exploit windows of opportunity (Edmondson, Kern, and Rogge 2019;Novalia et al. 2018).For example, agency can shift institutional boundaries (Droege and Marvel 2010), as institutions are underpinned by ongoing relationships and interactions among multiple actors or interplay to produce institutional change (Patterson and Beunen 2019).This is important for understanding how actors mobilise and attain resourcing (protect or secure) for institutional spaces (Edmondson, Kern, and Rogge 2019).These actors are often considered as niche actors; however, it is also important to consider regime actors for understanding how innovations can be embedded and diffused (Frantzeskaki, Thissen, and Grin 2016;Geels 2021, 47;Loorbach, Frantzeskaki, and Avelino 2017).
Secondly, institutional spaces are created around niches and experimentation.Niches can generate discursive spaces for change where the regime cannot (i.e., this allows for important protection of the niche) (Pesch 2015).We adopt the following definition of experimentation: a process, where emerging and fluid ideas, practices, expectations, technologies, and new social relations can develop and align into a new, potentially more sustainable socio-technical configuration that, if diffused more broadly, will radically alter the existing system (Fuenfschilling, Frantzeskaki, and Coenen 2019, 220).
Niches and experiments are essential concepts for understanding the characteristics of institutional spaces and they need to be understood in their wider institutional context.This niche-regime relationality is important for understanding and analysing the interactions, tensions, and co-evolution of institutional spaces.This includes the political processes that underpin these interactions (Savini and Bertolini 2019), actor capacities and capabilities to deal with enabling and constraining factors within institutional context (Novalia et al. 2018), and the impact of institutional context on policy design and mixes (Howlett and del Rio 2015;Lockwood 2022).These considerations are important for understanding the impact of niches and niche actors and their actions on what happens outside the niche (i.e., impact of niche on regime and regime actors) (Pesch 2015).For example, do new institutions co-exist with or undermine the regime?(Rogers et al. 2015).Therefore, niches and regimes are co-evolutionary, interdependent, and cumulatively shift over time and it is within these tensions that institutional spaces are created and evolve.
Therefore, to attain sustainability transformations, the contribution of incremental, concrete, small steps towards the broader goal, i.e., to build transformative pathways, through multiple small wins are critical to understand (Salo et al. 2022).This is important for breaking down knowledge and expertise silos as well as mobilising actors across boundaries (Willems and Giezen 2022).Furthermore, it is integral for understanding the evolution and expansion (up-scaling) of niches, such as niche actors gaining power and influence through coalition and network building (Edmondson, Kern, and Rogge 2019).
This actor-centred perspective is very useful for understanding how (local) urban actors shape (new) institutional spaces to adopt policies and practices that enable transformative change in the assumptions and decision-making agendas that underpin urban planning.
As a coherent definition of institutional spaces does not exist in the literature, we build on three definitions of institutional space from the literature.These definitions all highlight the importance of agency to create institutional spaces as well as the impact of institutional context on institutional spaces.Oteman, Wiering, and Helderman (2014, 4) define institutional space 'as the degree of discretionary freedom of community initiatives to decide autonomously about the design of a project (in terms of procedures and planning) and its contents (in terms of its goals and means)'.This is important for understanding whether institutional spaces are supported and steered from top-down and therefore the extent to which they are either able to embrace diversity or are subsumed into business-as-usual actions and diluted (Oteman, Wiering, and Helderman 2014).
Droege and Marvel (2010) conceptualise institutional space from the perspective of praxis-driven institutional change in the context of private enterprise in China's transitional economy.They position entrepreneurs as change agents who take advantage of 'the ambiguity of multilevel institutional inconsistencies to simply find a way to fix an immediate problem rather than change entire systems' (Droege and Marvel 2010, 207).Therefore, institutional spaces are negotiated spaces 'between "what is" and "what should be"' (Droege and Marvel 2010, 206).Savini and Bertolini (2019, 833) frame institutional spaces from a social practices' perspective, as niche construction and define: '[s]ocial practices can both reinforce the status quo or alter it … Niches can be defined as institutional spaces or episodes wherein social practices generate opportunities for disruption in their regulatory and physical context'.
In this paper, drawing on the above literature, we conceptualise institutional spaces as multi-actor settings in which new relationships, ways of cooperation, collaboration, and co-ordination are mobilised and configured with the aim to create opportunities for experimentation and the social-institutional conditions that allow, enable, design, and apply novel solutions.Therefore, institutional spaces offer new ways of organising actors and their actions and as a result, propose or make space for new institutions responding to a specified challenge (real-life problem) or opportunity.Institutional spaces, as we conceptualise in this paper, have the core characteristics of agencybased, place-based, time-specific, and action-oriented interventions that can (re)shape institutional spaces, to help answer practical questions such as: . WHO: which actors develop and maintain institutional space? .PLACE: where do institutional spaces unfold?
. TIMING: when are windows opened and exploited?
. ACTIONS: what experimental or innovative policy, governance, and practices are adopted to transform planning agendas?
Against this background, there is a need to better understand the institutional spaces in which actors operate to implement transformative policies and plans, as a way of attaining more climate-resilient cities.In this paper, we address the following question: What are institutional spaces and how can they contribute to the mainstreaming of nature-based solutions in cities?

Case study context
Urban forestry is an important approach to governance for nature-based climate adaptation in cities.In the context of metropolitan Melbourne, strategies for urban forestry are essential to change how we think about, value, and manage urban landscapes, to improve liveability, health and well-being, and resilience to address challenges including population growth and urban heat (City of Melbourne 2012; The Nature Conservancy and Resilient Melbourne 2019).This case study focuses on local government, as this is the level in which urban forestry governance is being actively pursued and advanced.This is important to note because the planning of Australian cities is a complex, and often fragmented, across Local, State, and rarely Federal Government authority and responsibilities (Davidson and Gleeson 2018;Phelan, Hurley, and Bush 2019).
Two key events in the urban forestry narrative for metropolitan Melbourne are important to highlight to understand the functions and characteristics of institutional spaces.Firstly, the pioneering, leadership, and international visibility of City of Melbourne's urban forest strategy (Gulsrud, Hertzog, and Shears 2018), especially its design, operation, and impact on planning approaches.Secondly, the collaborative, metropolitan-wide approach for urban forestry developed by Living Melbourne is crucial for resourcing, developing tools, and partnering opportunities to up-scale urban forestry actions in public and private realms (Bush, Coffey, and Fastenrath 2020;Fastenrath, Bush, and Coenen 2020; The Nature Conservancy and Resilient Melbourne 2019).

Case study design
We use an exploratory case study methodology (Yin 2003) to develop a deeper understanding of the existing and emerging institutional spaces across metropolitan Melbourne, in order to understand how the implementation of an urban forestry strategy can help transform the governance of NBS in cities.Within the case study, interviews were conducted with Council officers responsible for the urban forestry programmes in eighteen (18) of the thirty-two (32) local government areas (LGAs or Councils).Table 1 provides more detail on the participants in this study.The case study had an embedded design to capture understanding and insight into institutional settings of different Councils as well accounting for the broader context of the metropolitan region.This broader geographical scope is important because it helps us understand two scales at which the institutional spaces operate and change in terms of the formal context as well as the collaborative, emerging informal metropolitan platforms that support urban forestry actions (Coenen et al. 2020;Fastenrath, Bush, and Coenen 2020).

Data collection
We conducted thirty-two (32) interviews (online) during March-May 2022 with experts and practitioners who are involved in urban forestry across metropolitan Melbourne.Participants were selected for their involvement in Council urban forest strategies (n = 25), Living Melbourne (n = 1), or as academics (n = 6) undertaking research in this field (Table 1).Interviews were semi-structured to encourage and allow participants to reflect more flexibly on their practices and knowledges of urban forestry governance.The project received ethics approval from Swinburne University of Technology Human Research Ethics Committee (Ref: 20225938-9184/ 20225938-9531).Interviews were conducted online via Microsoft Teams and participants provided informed verbal consent to be interviewed, have the interview recorded, and that the resulting research may be published in journal articles or other formats.
The thirty-two (32) participants are anonymised, and Table 1 shows the participant type and which Council the local government officers work at.The interviewees will only be referenced by their participant code.Participants were purposively selected, starting with the City of Melbourne (n = 8), as an acknowledged pioneer in Australia (Gulsrud, Hertzog, and Shears 2018).We further used a snowballing approach to identify other potential participants (in local government and academia) and to expand the case study to other metropolitan Melbourne Councils.

Data analysis
In this study we deepen the conceptualisation of 'institutional spaces', building on themes in well-established socio-technical transitions theory, specifically niches within the multi-level perspective (Doyon 2018;Geels 2002;Smith and Raven 2012).While we link our study to established theory, this is not used as an analytical framework for data analysis.Interview data was analysed drawing on a grounded theory approach (Charmaz and Bryant 2008) to inductively develop, rather than presuppose, theoretical insights on the institutional settings emerging from the strategic development and implementation of urban forestry governance in metropolitan Melbourne.This analysis process examined the requirements for effective institutional spaces and the outcomes that institutional spaces produce.

Results: mapping the institutional spaces that support NBS mainstreaming
In this section we draw on insights from interviews to map and conceptualise the activities that take place in and constitute institutional spaces that support urban forestry in metropolitan Melbourne to understand what makes them more adaptive and supportive of NBS in general.From our research we identify and explain the activities of institutional spaces to be transformative in a way that supports NBS mainstreaming in cities.Specifically, we identify that institutional spaces have the following functions: . Relationship-building to promote transformation, including bridging silos in organisations; .Activities that generate momentum for transformative agendas to gain traction, i.e., positive pathdependency, defined as positive inertia, which is 'an emergent property at urban system level, where positive approaches are embedded into the city's identity and thus not influenced by shortterm political cycles' (Irvine and Bai 2019, 16); .Transformative tensions, or stressors (Matthews 2012), through activities such as critical dialogue and social learning for reviewing and reforming planning and existing policies; .Opportunities for co-producing knowledge and skills that underpin capacities to mainstream NBS.
Bridging silos: Many interviewees emphasised how the efforts of creating and implementing their urban forestry strategy had the effect of bridging silos, and that whole-of-Council approaches were a key success factor in this activity.Therefore, this is a key factor to opening up an institutional space, by breaking down organisational silos of operation and implementation.This relationship-building within Council is an essential component of institutional spaces, as one interviewee described relationships: they're a really intangible thing, you can't put that in a strategy, but that's actually how you get stuff done.That's sometimes the only way of navigating through these difficult things.[R1-9] Most interviewees were positive when speaking about the progress of urban forestry in terms of awareness and commitment of their Councils.Interviewees described their Council's internal relationships for how climate and ecological issues are viewed and understood by actors across Council departments: the engineering team we collaborated with because we needed to reframe their thinking around how they viewed trees as part of their work.So we did quite a lot of internal engagement with them … there was a lot of work done across Council to make sure that teams and departments were aware of the work that was going on, what the priorities were, but particularly around this idea of embedding innovation within existing decision making.[R1-6] we've got a bit of a reputation now, like our climate emergency response plan has been really impactful.And I think it has reframed how people think about our role as a Council and their individual roles within Council.And I think the expectation is that the urban forest strategy will do the same because we're linking it up to urban heat and the health and well-being of our community and we're linking it up to health and well-being in terms of sustainable transport.[R2-5] Relationships were also found to be important to create or extend the institutional space across the metropolitan region.This means that institutional spaces are not only about collaborative approaches internally within each Council, but relate to their impact and influence to facilitate collaboration across Councils.The configuration and facilitation of institutional spaces across the metropolitan region is important for understanding the impact of leadership, innovation, and learning on the evolution of urban forestry as a practice and discipline.Interviewees highlighted the impact of the City of Melbourne as a city-leader in urban forestry for the rest of the metropolitan Councils: And because so many of them have followed because of the willingness to share and to act as an ambassador for nature-based solutions, the City of Melbourne has been able to lift everybody up.And then there's enough of them to make Living Melbourne work.And that's a real coup, because most cities that are fractured into 32 council areas would find it very hard to speak as one.But they've managed it in Living Melbourne and that's through a general acceptance that this is a good thing.That having more vegetated systems for biodiversity, for cooling, for play, whatever it is, is a good thing.[R1-5] Ensuring opportunities for building relationships among key actors: Building relationships among multiple actors is required for opening and supporting the expansion of institutional spaces in terms of the outcomes they can produce.One example of this is the conscious leadership role that the City of Melbourne has undertaken to support the diffusion of urban forestry to other Councils and how their expertise influences the State Government, as one interviewee described: whether we like it or not, we're in a leadership role and I think we try to take that responsibility seriously … if we've got the funding to do innovative things or different projects then I think there's a greater benefit for everyone if we are sharing those learnings, participating in those networks.[R1-9] We don't have direct control [at State Government level], so it's an avid advocacy piece … We influence where we can, we advocate where we can.We need to get a bit creative about how we support them with our technical expertise to deliver better outcomes on their projects as well.[R1-9] Such leadership is an influential feature in this case study; however, our analysis indicates that institutional spaces for urban forestry governance across metropolitan Melbourne are evolving through different Councils adapting and contextualising the concept for their needs.This collaborative creation, through cumulative knowledge and policy processes happening in the communities of practices that contribute to urban forestry governance (such as practitioners, policymakers, academics, and consultants), reinforces relationships.It also produces new opportunities to continue the work to bridge silos and leveraging new or different priorities through networks to build capacity for, and to resource, urban forestry governance.This can be achieved, at least in part, through networking that shares (best) planning policies and practices from Council-to-Council.Such relationships have now moved beyond being mere pioneering activities, and shaping of the City of Melbourne, to now include a range of different voices, perspectives, and knowledge to help define the goals and pursue the implementation of urban forestry.As one interviewee stated: I'm really pleased to say that there's now the emergence of a whole range of other Councils who now also sit under that leadership banner who are doing their own things and who have great things to share and learn from.[R1-6] It was also noted that networking opportunities can build relationships and capacity through emerging metropolitan platforms and governance, such as Living Melbourne.This insight is also evident in the urban greening partnership of the western metropolitan Melbourne Councils called Greening the West, and the visibility of City of Melbourne's urban forestry programme in international city networks, for example C40 Cities and 100 Resilient Cities.Leadership and advocacy through (sub)regional networking (cross-boundary projects) are also important in order to produce new networks for learning.As one interviewee stated: Greening the North is probably in its inception stage, following on from the success of Greening the West.That's something that we're involved in the emergence of that group, to advocate and hopefully gain some sort of resources to green.[R2-9] This means that the increased adoption of urban forestry activities across metropolitan Melbourne is enabled and reinforced by different actors building on the legacies of prior urban forestry activities, through actively participating and learning to develop best practices.For example, in the (place-based) knowledge about urban forestry, what to include and how to formulate policies and plans, and investing in partnerships and collaboration to benefit from the cumulative improvement of knowledge and practice for urban forests.As one interviewee described, these characteristics mean that institutional spaces help facilitate opportunities for social and policy learning: working with others and they build on it too, City of Melbourne already had an urban forest strategy with targets in it, but then Living Melbourne was able to provide a richness to that.So, although our public realm canopy cover targets in our strategy were more ambitious, we didn't have targets for the private realm, that plan was able to develop more nuanced targets.And so that still adds value to us as well.So, by being involved in those partnerships and collaborative pieces of work, we're obviously able to help shape the direction that they go in and make sure that it adds value for us as well.[R1-9] This capacity is also built through networks that are shaped around urban forestry, greening, and conservation (i.e., depending on the ecological context) activities.Interviewees noted that Council-to-Council networks for the implementation of urban forestry are important for NBS mainstreaming because they directly foster relationships across Council jurisdictions for knowledge sharing, funding, advocacy, and delivering outcomes.As one interviewee stated: if you're looking at connecting … forest corridors from one area to the other, then it's that connection between Council and Council … I think collaboratively it works in terms of a voice as well.[R2-17] Building the momentum for transformative agendas Institutional spaces also need to be considered from the perspective of the potential and momentum that they can produce for transformative governance.This may lead to positive path-dependency; for example, by gaining political support as well as reinforcing commitment to mainstreaming agendas.As one interviewee noted of the City of Melbourne's urban forest strategy: If you had done it a couple of years earlier, would it have happened in the same way?No.It was just everyone was aligned at the same time.And I think it was just very, very fortunate that all of the right people were together in one space to push that.[R1-10] This opening of institutional space at the right time is critical in allowing it to flourish.In terms of the timing for prioritising urban nature and ecological systems in city planning, interviewees mentioned drivers such as the stability of political support for urban forestry, the statutory planning rules that provide the legal and policy justification to protect urban trees and canopy, and the capabilities and capacities of the right actors (e.g., champions) to be able to capitalise on the opportunities to propagate the adoption, implementation, and evolution of urban forestry when they arise.
Within urban Councils there is a common tension between development pressures and growing the urban forest.Within this context, this timing of opening an institutional space for change is critical for generating momentum to support and reinforce opportunities to prioritise urban forestry actions.Urban forestry, and urban NBS more generally, build on concepts for conservation planning, ecosystem services, and ecosystem-based adaptation and are therefore interconnected with place-based decision, policy, and action legacies.This means that for institutional spaces to be productive and relevant, they need to be contextualised for place-based needs and flexible and responsive to changing social-ecological needs.Therefore, these legacies can provide a positive, enabling context to push transformative agendas.For example, as one interviewee described: As a Council we've been a bit lucky.There was the sites of biological significance work that was done between 2018 and 2020.That was largely a contemporary version of a similar piece of work that was done way back in 1997, and the outcome of that piece of work was all the sites of biological significance or pretty much all of them had a vegetation protection overlay put over them.And there was also recognition at the time that a significant landscape overlay was put over the ridge lines, which have quite a bit of tree canopy cover.So the legacy of those decisions way back then is that we still have a relatively high urban canopy cover for a metropolitan Council, an urbanised council, so we've been fortunate in bold decisions made a long time ago have left us with a reasonable legacy.[R2-1] It is important to emphasise that these institutional spaces are shaped and underpinned by governance settings, such as the priority of participatory planning approaches and the extent of future-oriented planning, and whether there is a mandate for expanding the scope of urban forestry to include initiatives on private land as well as public land.The momentum in institutional spaces can trigger domino effects, i.e., what happens in individual Councils, further shapes how those Councils come together to collaborate, which in turn shapes the influence that urban forestry has at a larger geographical scale, here across metropolitan Melbourne as well as State Government priorities.As one interviewee stated: I think there is a groundswell of action both at Council level and at State Government agency levels … in essence, I think it's a real wave now of action in this space and very hard for someone to put the genie back in the bottle.[R1-7] The momentum of change that an institutional space generates is therefore produced by multi-actor interactions that consciously build positive pathways and therefore forming positive path-dependencies for implementing urban forestry actions and mainstreaming urban NBS.This leverages existing relationships and networks to produce these networking opportunities to expand their influence.For example, one interviewee described this process regarding the creation of Living Melbourne: the network of people, executives, staff that Resilient Melbourne made across metropolitan Melbourne was critical, not just in Councils but State Government organisations, water authorities, not for profit organisations.They'd created a network of people, which was absolutely critical to the development of Living Melbourne.Without that network it would have been very hard to build it from scratch, because The Nature Conservancy, whilst very capable, didn't have that network.[R1-7] Creating transformative 'tensions' for mainstreaming urban forestry Institutional spaces enable discursive space for urban forestry actors to advocate for, formulate and review planning policies, and implement projects for growing the urban forest.Urban forestry policies, plans, and projects are important examples of what institutional spaces can produce and deliver to build the evidence-base, create evidence-based policies, and the justification for (continued) political and/or community support.These can produce and leverage the transformative tensions or stressors in the knowledge, policies, and decisions that drive institutional change.This includes the strategic positioning and implementation of urban forestry, as one interviewee explained the methodology behind their Council's urban forest strategy: We went essentially with an arboricultural-led strategy rather than an urban design or landscape architecture one because I wanted something that would really give our arborists and our tree planting crew something practical that would guide them.[R2-12] The legacy, continuity, and learning in urban forestry are important indicators that tell us about the transformative tensions that underpin the reflexivity of institutional spaces, in other words: how they learn.As one interviewee said: now that it's here, it's here to stay, we get to build on it.We're very fortunate that it exists and that it laid the work for all the next policies that are going to come.[R1-12] This is also important for understanding and defining the strategic planning priorities for urban forestry.Specifically, the engagement, inclusion, and empowerment of local communities, their aspirations and concerns, in planning processes is important for the success of mainstreaming.As one interviewee noted: engaging the community, empowering and engaging and educating them … we're supposed to be here for the community.Otherwise, we just get caught up in all the policies and plans and we forget who they are for.[R2-13] (Co-)producing knowledge and skills to mainstream nature-based solutions Knowledge is an important requirement for ensuring institutional spaces are effective in enabling change, such as through the actions of key players, including practitioners, academics, and consultants, to build evidence-bases, understanding what change is needed, and practitioner and policy skills.More importantly, institutional spaces are dynamically changing in an ongoing and incremental manner, and this is underpinned by the knowledge and skills that they hold and nurture.
We frame interaction between (scientific) knowledge, policy formulation and (practitioner) skills as the science-policy interface.Based on this, we argue that the City of Melbourne has developed a productive science-policy interface for urban forestry over time.As noted by many interviewees (specifically R1-1 to R1-12), the City of Melbourne's science-policy interface is built on long-term and trustful relationships between local government officers and academics and the financial capacity to invest in commissioning (specific) research.This means that there is a porous boundary between knowledge (co)production and knowledge use in policy and planning practice.We found that the collaboration between research and practice actors has been essential for the institutional space to develop the capacity for urban forestry (as an example of NBS mainstreaming).This is based on respectful (co)production of knowledges and cultivating diverse and sophisticated expertise and knowledge for how urban NBS are thought of and implemented.As one interviewee noted: you come to realise that it's a very useful relationship for both sides.It's a matter of finding ways of having this relationship so understanding that both experiences and both voices are valuable.[R1-11] This type of constructive science-policy interface is, however, not a universally accessible characteristic of institutional space across metropolitan Melbourne.Many interviewees mentioned, specifically participants not from the City of Melbourne, that the absence of constructive collaboration across the science-policy interface is often a barrier to the mainstreaming of urban NBS.This shows frequently missing connections to foster urban experimentation in many metropolitan Melbourne Councils.In light of this, we argue that the mainstreaming of NBS can be undermined by disconnections in science-policy-practice interfaces.As one interviewee noted: there's a real disconnect.It's really quite a challenging conduit to make happen.[R2-5] This can be seen in research-implementation gaps: There's so much potential to gain science-based experimentation and trials in local government, but it doesn't seem to be happening very quickly.[R2-9] Another way to conceptualise these barriers is (in)access to research outputs, as one interviewee noted: Keeping up with what might be best practice or cutting edge is tricky.[R2-7] The same interviewee continued, emphasising that important sources of knowledge for local government include open access resources that are easily engaged with: Greener Spaces Better Places is a good source of information … these are public access and have free, easy to locate information.[R2-7] Discussion: transformative elements of institutional spaces for urban forestry in metropolitan Melbourne In this section we synthesise our findings about what makes institutional spaces more constructive and transformative.This allows us to conceptualise institutional spaces by mapping their functions of bridging silos, relationship-building, generating momentum, transforming tensions, and co-producing knowledge.
We argue that these functions and features of institutional spaces have the potential to catalyse transformative action and accelerate climate adaptation actions, and should lead to greater momentum and sophistication over time.This has insights and implications for understanding the actors who develop and maintain institutional spaces, where they unfold, when windows of opportunity can be exploited, and the innovative actions that contribute to the transformation of urban planning.

Design characteristics of institutional spaces
We define effective institutional spaces, broadly, as being collaborative (in terms of how they operate), shared (in terms of who is involved or has access to its products), and (re)shaped by multiple actors over time.Specifically, four key characteristics of institutional space is that they are relational, learningoriented, collaborative, and empowering spaces.

Relational institutional spaces
Institutional spaces are relational spaces that produce transformative agendas and actions through the connections and interactions among diverse actors.The literature identifies some of these elements, such as how citizen science programmes contribute to knowledge (Roger and Motion 2022), the awareness of the benefits of urban forests and the perceptions of urban forest managers (Kirkpatrick, Davison, and Harwood 2013;Ordóñez et al. 2020a;2020b), and the importance of private sector and civil society actors in urban forest management and implementation (Park and Youn 2013), such as environmental non-governmental organisations (e.g., The Nature Conservancy) (Foo 2018;Frantzeskaki and Bush 2021;van der Jagt and Lawrence 2019).The relationality of institutional spaces is a characteristic that helps explain how they evolve and mature.This means that within their context, institutional spaces eventually ought to become self-sustaining and selfreinforcing in their dynamics.This is driven by actors maintaining the institutional space, new actors re-purposing institutional space for their own place-based needs, or institutional space growing (up-scaling) to be inclusive and responsive to diverse implementation needs through networks, and outcomes that are produced can leverage continued political support and resourcing (e.g., ongoing funding sources and active community engagement (van der Jagt and Lawrence 2019)).

Learning-oriented institutional spaces
Institutional spaces represent opportunities for learning that are flexible and responsive to emerging needs, thereby allowing actors to participate in deliberative actions to transform planning (for mainstreaming of urban forestry in our case study).This framing of institutional spaces means that we can analyse the institutional capacity that can be produced by the creation and enactment of institutional spaces, as well as the institutional capital that can reinforce action and mobilise in different contexts, such as the role of practitioners, academics, and consultants in diffusing knowledge and policy transfer across metropolitan Councils.This highlights the constructive role of institutional spaces to enable learning and reflexivity as well as scaling up/down to manage different issuesi.e., at local or metropolitan scale.Institutional spaces need to be allowed to build on what has worked, as well as have the responsiveness and flexibility to change course when necessary.This can happen, for example, in legislative and regulatory changes (Atmiş 2016;Park and Youn 2013), co-ordination and communication skills to bridge silos, as well as resourcing to implement programmes (Ordóñez et al. 2020a;2020b).

Collaborative institutional spaces
Institutional spaces need to enable collaboration in a way that includes all relevant actors, knowledges, practices, and sectors.This is essential in order to legitimise and build the acceptance for co-produced and co-created transformative agendas and actions.This means that different institutional spaces will often co-exist and overlap across multiple governance levels (e.g., local, metropolitan) and can therefore influence each other.For example, pioneering institutional spaces often have a powerful champion to open and create this space, such as the innovation and leadership of the City of Melbourne (Bush et al. 2023b;Gulsrud, Hertzog, and Shears 2018) and the influence of Living Melbourne (Bush, Coffey, and Fastenrath 2020;Coenen et al. 2020;Fastenrath, Bush, and Coenen 2020;Hartigan et al. 2021), which can be used to leverage collaboration to build on these legacies in other contexts.Institutional spaces therefore have intermediating characteristics in which the capabilities and capacities of actors to create and participate in them can either be individual or collective.Conceptualising institutional space as porous is therefore important to include the single-level (for example within a Council) and multi-level characteristics of institutional spaces (for example across a metropolitan region).This can be instrumental for creating a constructive science-policy interface for long-term and ongoing engagement among city practitioners and academics relating to urban biodiversity and urban ecology (Bush et al. 2023b).The science-policy interface is useful for understanding the dynamics of communities of practice, and has been found to be an important element in the creation and adoption of urban climate and biodiversity related strategies in Australian (Bush et al. 2023a;2023b) and European cities (Frantzeskaki et al. 2019;Kabisch et al. 2016).Therefore, the collaborative characteristics of institutional spaces are underpinned by its supporting multi-actor networks that enable knowledge and practice diffusion across local and metropolitan scales.

Empowering institutional spaces
Institutional spaces need to be empowering spaces that are created by place-based actors (champions), and maintained by the active, purposeful, and inclusive participation of multiple actors over time.The evolution in the purpose and reasons for participation within institutional spaces drives their flexibility and maturity and help them to be responsive to diverse and changing needs.For example, successful institutional spaces may be exploited to leverage an acceleration of complex and sophisticated policy approaches to not only urban forestry, but urban biodiversity and urban ecology (Bush et al. 2023b), i.e., that institutional spaces can expand based on theme.Furthermore, institutional spaces can expand based on spatiality, such as the impact of Living Melbourne on the definition and capacity to implement urban forestry across metropolitan Melbourne Councils (Bush, Coffey, and Fastenrath 2020;Coenen et al. 2020;The Nature Conservancy and Resilient Melbourne 2019).This feature is important because it means they are not limited to well-resourced pioneers designing experimental institutional space.Lastly, such empowering spaces connect to the emerging literature on mosaic governance, which focuses on the relationships among citizens and local governments to implement and maintain urban greening projects (Buijs et al. 2017;2019;Gentin et al. 2022).
The multi-actor web that keeps institutional spaces working also need to be invested in for their continuation and persist with efforts to maintain (in)formal platforms for place-based outcomes.Through the multi-actor webs, actors are empowered to mainstream NBS over time, to learn and continue to improve and innovate for urban forestry, transform the way nature-based city infrastructure is understood and prioritised, i.e., thought of as essential and critical not a luxury.In this way, institutional space actors fight for the right to be disruptiveto challenge the status quo (regime).

Implications of institutional spaces for planning practice
Based on our case study findings, we argue strongly that institutional spaces are critical for NBS mainstreaming in cities because they create the social dynamics to support transformative governance agendas and actions for nature-based climate adaptation.It follows that the critical underpinning requirement for institutional spaces is the active engagement and empowerment of diverse actors.In our case study these actors are particularly in local government, but also include (or have the potential to include) representatives from water authorities, universities, consultants, communities, State Government departments, and the private sector, to participate in, (re)shape, and transform institutions through the evolution, maturity, and sophistication of urban forestry experimentation.Actors can leverage their empowerment and engagement for networking opportunities to diffuse NBS mainstreaming agendas and actions for local and/or metropolitan-wide outcomes.This is supported by recent research that focuses on the role of intermediary organisations such as environmental non-governmental organisations (Frantzeskaki and Bush 2021).
We emphasise the interdependence of the pioneers of urban forestry and the community of practice as well as the opportunity for emerging leaders to not only create but manage and maintain institutional spaces that are relevant and flexible for place-based adoption and implementation.
We have found that institutional spaces have been transformative in relation to urban forestry, and have provided spaces that embrace the urban forestry narrative, and through their operation embed urban forestry narratives in (local) urban planning agendas and actions.We think that this has implications more generally in how NBS for climate adaptation and urban resilience can be embraced and embedded in urban planning practices, by leveraging the opportunities created by the institutional spaces for urban forestry, which is also explored in recent research on City of Melbourne's urban nature programmes (Bush et al. 2023b).
In summary, institutional spaces can amplify and accelerate the adoption of innovative, systemic solutions, to capture the strategic groundwork of pioneers and leaders of experimentation, and ultimately support and reinforce the momentum for transformative change.We suggest that future research should focus on how our conceptualisation of institutional spaces can leverage planning reform across multiple levels of governance.

Planning recommendations for creating or navigating institutional spaces
The characteristics of institutional spaces are important to elevate within planning practice, especially in the process(es) of mainstreaming NBS in cities.This includes the social dynamics that create institutional spaces, how actors navigate the complicated settings of institutional spaces, as well as the readiness and capacity to capitalise on windows of opportunity when they arise.
Building relationships, by collaborating within the institutional space created for urban forestry, is important in order to facilitate the evolution of institutional space through best practices, collective action, and cumulative learning opportunities for experimentation.This can be achieved by multiple leaders contributing to the richness of urban forestry knowledge, practices, and implementation, as well as participation in networking and collective action projects or programmes for funding and implementation opportunities.
The momentum in transformative governance is produced by multi-actor collaborations that develop urban forestry policies, plans, and practice.This requires alignment of key actors to create and progress the agenda of urban forestry, to apply urban forestry at local and metropolitan scales, and its potential to influence urban planning policies at the State-level.
Momentum is often generated by champions who push for transformative agendas and build on legacies of urban environmental policies and plans.Therefore, positive path-dependencies are fostered by building on existing relationships, networks, and capacity to leverage the continuation of activities that are supported and reinforced through institutional spaces.
Transformative action requires that institutional spaces are relational and have the capacity to enable or leverage institutional change.This means that actors need to navigate transformative tensions to produce the knowledge, evidence-based policies, political buy-in and as a result, support for the implementation of urban forestry actions.This can be achieved through the configuration of institutional spaces that prioritise policy learning, reviewing policies, and community engagement.
The evolution of institutional spaces, specifically in relation to the implementation of urban forestry, is underpinned by the knowledge and skills to develop them constructively.This means that it is important to understand both science and policy inputs to the practice of urban forestry.Therefore, institutional spaces need to be empowering spaces for actors to learn, collaborate, and implement place-based solutions for nature-based climate adaptation.
We identify seven urban planning recommendations based on these characteristics to suggest how to best create and utilise institutional spaces in urban planning policies, practices, and politics to mainstream NBS in cities.
. Recommendation 1: Build relationships that bridge silos of operation (within Council) to encourage collective and collaborative actions to leverage ongoing political support and resourcing to reinforce experimentation (up-scaling, mainstreaming); . Recommendation 2: Encourage (local, place-based) experimentation and innovation in urban forestry and urban NBS, such as facilitating greening on private land; .Recommendation 3: Learn from champions and innovators in urban forestry governance and practices, such as by building on policy legacies of urban forestry, urban NBS, or environmental planning and biological conservation; .Recommendation 4: Participate in urban forestry and urban NBS networking opportunities at metropolitan, State, national, or international scales to improve knowledge, collective bargaining capacities for funding, and collaborative programmes (e.g., Greening the West, Living Melbourne); . Recommendation 5: Develop and maintain strong ties with local communities to gain acceptance and justification for urban forestry, including participatory planning and education programmes; .Recommendation 6: Enable policy processes that formulate and review state-of-the-art evidencebases and innovations in urban forestry; .Recommendation 7: Invest time and effort in developing relationships in the science-policy interface (with universities, consultants, etc.) either as a Council or collectively through networks to draw on the disciplinary knowledge, practices, and skills in urban forestry.

Conclusion
Social processes underpin the policies, knowledge, and decisions that are needed for climate adaptation, including the adoption of nature-based solutions.This paper reports on the study of such social processes, in relation to urban forestry governance in metropolitan Melbourne.Deeper understanding of these social processes, especially in relation to how institutions can transform over time is important for realising climate and ecologically resilient cities and shifting towards sustainability norms in urban planning.
The social processes that support transformative change occurs in what we label institutional spaces.We argue that planners need to recognise and use these institutional spaces to build readiness and capacities in the knowledge, policy, practices, and networks of urban forestry, and more broadly urban nature-based solutions.There is therefore an opportunity for actors involved in urban planning policies and practices to capitalise on institutional spaces that are emerging and mobilising for urban forestry to develop and produce capabilities and capacities in their own projects and programmes.This is important, not only for ensuring that planners are ready to adopt, implement, and adapt these innovative, systemic solutions in their own local government areas, but also to learn from and feed into the collective, cumulative disciplinary and practice knowledges for urban forestry.
We conclude that institutional spaces are driven by multi-actor co-ordination through a series of functions.These functions help create and maintain relational, learning-oriented, collaborative, and empowering spaces for designing and implementing innovative solutions in cities.By conceptually developing the understanding of urban institutional spaces for explaining how NBS mainstreaming can unfold over time is important for (re)framing how sustainability transitions processes are enacted and pursued in cities.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Table 1 .
Case study recruitment and participants.