Cultivating Governance: An Exploratory Analysis of Urban Garden Management Models for Long-Term Sustainability

Abstract

The transformation of food systems has driven the growth of community gardens as urban green infrastructure with potential social, health and environmental benefits; however, less attention has been paid to how their governance is configured and sustained in practice. This study analyses and compares the governance of seven urban allotments in the Valencia Functional Urban Area (Spain) and explores its social dimensions: participation, community cohesion, inclusion and learning. A qualitative cross-sectional design combined 10 semi-structured interviews and five focus groups with 37 institutional, associative, user and expert participants, selected through purposive sampling until thematic saturation and analysed thematically. The allotments form a continuum between community self-management and institutional management: three are formally institutional, one is hybrid and three are self-managed with support, although hybrid arrangements predominate in practice. Participants described institutional contacts, funding, tenure arrangements and organisational capacity as relevant to continuity. Reported difficulties included administrative constraints, concentrated workloads and volunteer succession. Social interaction, inclusion and learning were valued alongside cultivation. The findings identify differences in reported governance practices without establishing the superiority of any model.

1. Introduction

In recent decades, the transformation of the food system has gained global significance [1,2]. Research on food systems examines the implications of production, consumption, supply chains and access to food for health and the environment [3]. Although widely used, the concept of sustainable food systems remains loosely defined and is applied in different ways across disciplines, and its economic, social and environmental dimensions involve trade-offs that need to be addressed explicitly [4]. In this article, a sustainable food system is therefore understood as one that "delivers food security and nutrition for all in such a way that the economic, social and environmental bases to generate food security and nutrition for future generations are not compromised" [5]. The transition towards sustainable food systems is key to several Sustainable Development Goals (SDGs), such as zero hunger (SDG 2), good health and well-being (SDG 3), responsible consumption and production (SDG 12), climate action (SDG 13), life below water and life on land (SDGs 14 and 15) and peace, justice and strong institutions (SDG 16) [6,7]. Among the strategies aimed at healthy urban food systems, urban agriculture stands out as one that seeks to increase the availability of fresh, healthy food [1,8], strengthen local production and urban resilience [9,10,11] and promote social cohesion and learning [12,13].

Urban agriculture thus strives to deliver environmental, health and well-being benefits [1,14], although these outcomes cannot be taken for granted, and some authors question its contribution to food security due to a lack of space and limited productive capacity [15]. Community gardens form part of this and usually consist of small shared plots where food is grown seasonally [16]; they integrate environmental, social and urban dimensions and constitute urban green infrastructure that can generate ecosystem services [17]. Despite the sustained growth in research on their benefits [18], less attention has been paid to how governance, understood as the structures, relationships and participatory processes through which stakeholders share decision-making, is configured and sustained in practice, and to how the actors involved perceive its relationship with these outcomes [19,20]. The literature identifies different models situated on a continuum between “top-down” hierarchical structures and “bottom-up” self-managed approaches, with hybrid configurations and even bottom-up models supported by political or administrative bodies [21]. In this vein, a case study in Austin (Texas) found that higher levels of social and ecological performance were associated with bottom-up models, in which the community directly takes charge of decision-making and management [22].

Building on this literature [21,23], five models emerge most clearly: 1. community self-management or bottom-up management; 2. external support without a loss of local leadership; 3. hybrid arrangements; 4. institutional management mediated by NGOs or associations; and 5. municipal or top-down management [22,24,25,26].

The bottom-up, or grassroots, self-managed model places the gardeners at the centre of decision-making. Drake [25] identifies it as one of the three main ways of managing community gardens, alongside those organised externally and those actively managed by NGOs. Here, the community not only participates but also defines priorities, rules of use, the division of labour and conflict resolution mechanisms. In Austin, Ponstingel [22] associated bottom-up structures with the best social and ecological outcomes, noting that they tend to maximise ownership, a sense of belonging and local adaptive capacity, but depend heavily on available time, social capital and group stability. Without clear rules or mediation mechanisms, this autonomy can become a source of fragility, particularly where participant turnover is high or access to resources is contested.

The bottom-up model with external support retains community leadership but incorporates assistance from a local council, a university, an NGO or a public agency. In Toronto, Jacob and Rocha [27] associate this support with project longevity, particularly in low-income neighbourhoods, as it helps to secure land, funding, staff or training. External support may provide resources and operational assistance, while responsibilities and decision-making powers need to be specified between the actors involved.

The hybrid model seeks to sustain both institutional leadership and local autonomy. In Cape Town, Kanosvamhira [23] formulates it as “hybrid governance (neither top-down nor bottom-up)”, which avoids the false dichotomy between fully community-run and fully administered gardens, and warns that support from public agencies and nonprofit organisations can be important for sustainability but, if poorly managed, may erode local ownership. The model combines contributions from institutional and local actors, with different distributions of resources, responsibilities and decision-making powers.

The institutional, or third-party-mediated, model emerges when an external entity (a cultural institution, an NGO or a development association) becomes the central actor. In Warsaw and Poznań, Maćkiewicz and Jeziorek [26] warn that partnerships with cultural institutions may shift power away from the community towards more hierarchical arrangements, while in Turin, Tecco et al. [28] show that management by the local government and promoting associations, within its regulatory frameworks, faces challenges of adaptability. Intermediary organisations may provide access to permits, resources and operational support, while users’ involvement in decisions varies with the arrangements established.

In the municipal, or top-down, model, regulatory and strategic power lies with the public administration. Maćkiewicz and Jeziorek [26] describe configurations in which community members no longer influence strategic decisions, and Willman [29] shows that in Tampere the “operational space” of gardens is defined by administrative policies within a cautious, ambiguous and sometimes contradictory institutional culture. This model provides order, regulatory continuity and land protection, but may limit participation to instrumental tasks. In Geneva, Ernwein [30] observes that heavily regulated municipal allotments seek to curb the privatisation of space and the commodification of nature, yet may reinforce more individualised subjectivities.

Considered together, these typologies shift the question from “which model is best?” to “what combination of power, resources and autonomy does each garden need in its context?”. The literature identifies different combinations of institutional support, internal decision-making and coordination, whose implications depend on the context and objectives of each garden.

In this context, the focus of this study is the governance of urban allotments in the Valencia Functional Urban Area (Spain), and the continuum of governance models analysed here serves as an analytical framework (see Section 2.3). The study addresses three research questions. RQ1: How are responsibilities distributed amongst public authorities, associations, intermediary organisations and gardeners across seven urban allotments, and where on the continuum does each allotment fall, both formally and in practice, in terms of autonomy, participation and institutional support? RQ2: Which institutional, regulatory and organisational factors, according to the actors involved, sustain or erode each governance arrangement over time? RQ3: How do participation, community cohesion, social inclusion, and learning and knowledge transfer manifest themselves within each governance model? By addressing these questions, the study seeks to contribute to the literature on urban garden governance by moving beyond the classification of models towards an understanding of the conditions under which their position on the continuum is maintained or altered. The initial hypothesis is that these gardens do not follow a single model, but are distributed along this continuum, and that, as perceived by the actors involved, their governance structure is closely linked to their social, environmental and urban outcomes.

2. Materials and Methods

2.1. Spatial Context

The study was carried out in the Functional Urban Area (FUA) of Valencia. It is the third-largest FUA in Spain, with a population of 1.78 million in 2022 [31]. The area maintains close historical links with l’Horta de València, a peri-urban irrigated agricultural system of high environmental, cultural, landscape and productive value, whose legacy remains visible in the network of irrigation channels, traditional practices and the land protected by the Territorial Action Plan for the Management and Revitalisation of the Huerta de València [32]. At the same time, urban expansion, population growth and changes in land use are creating tensions regarding the preservation of agricultural spaces and the integration of green infrastructure.

Urban gardens have proliferated in recent decades in the Valencia metropolitan area as spaces for urban agriculture and green infrastructure, the benefits of which (social cohesion, community learning, physical and mental health, urban resilience and the provision of ecosystem services [10,12,17,18]) depend to a large extent on the specific forms of organisation, management and regulation adopted in each territorial context.

The empirical research focused on a selection of urban allotments in the Valencia metropolitan area that represent different positions along the continuum between bottom-up, hybrid and top-down models of urban governance [21,22,23,33], and incorporated the perspectives of institutional, political, technical and community stakeholders.

The seven allotments studied (Espai Verd Benicalap, Mislata, Malilla, Sociópolis-La Torre, Godella, Llíria and Alaquàs) were selected in two stages (see Figure 1). First, candidate gardens were identified on the basis of the research team’s previous fieldwork on urban gardens in the Valencia FUA [34,35], which had documented the diversity of implementation and management models in the area, including bottom-up, hybrid and top-down initiatives [34], and through contacts with managers and key informants facilitated by partners of the U-GARDEN project, such as the Universitat Politècnica de València. Access to each garden was negotiated with the actors able to grant it, such as municipal councillors and technical staff or the boards of the managing associations, who were contacted by telephone and email, either through the local councils or directly through the associations. Second, cases were selected according to two criteria: (a) being active during the fieldwork period; and (b) the willingness of the managing bodies to grant access and take part in the study. Several gardens initially considered could not be included because they had been affected by the floods of October 2024 and their managers could not be reached or did not respond; the most significant of these cases is described in the limitations section.

2.2. Participants and Procedure

A qualitative design was adopted because the research questions focus on how governance arrangements are configured, sustained and experienced by the actors involved, processes that require in-depth accounts rather than quantitative measurement [36,37]. Case selection followed a purposive sampling approach, aimed at capturing the maximum variability of governance models within a single territorial context and at selecting cases that were rich in information and analytically relevant, rather than seeking statistical representativeness [36,37]. The final sample comprised 37 participants across two complementary methods: 10 semi-structured individual interviews, coded E1 to E10, and 5 focus groups, coded GF1 to GF5, involving a total of 27 people. This combination made it possible to contrast the individual perspectives of institutional, political, technical, academic and associative informants with the group dynamics of community actors and service users.

For recruitment, the research team contacted, by email and telephone, the various stakeholders involved in the field of urban and community gardens: policy-makers and technical staff from local and supra-municipal authorities, staff from public innovation bodies, university lecturers specialising in urban planning and urban gardens, and representatives of the associations managing the gardens under study and of third-sector organisations.

The inclusion criteria were: (a) to hold an institutional, political, technical, academic, associative or user role with direct responsibility or documented experience in the management, regulation, use or study of the urban gardens within the FUA in Valencia; (b) to demonstrate at least six months of continuous involvement in that role or activity; and (c) to be able to express opinions and experiences verbally without restrictions. Individuals with only a one-off or circumstantial connection were excluded, as were those unable to respond independently to the interview guide.

All sessions, both individual and group, were conducted between November 2025 and June 2026 by the principal investigators following the thematic script detailed in Section 2.3; they lasted approximately between 60 and 90 minutes and were audio-recorded, with prior authorisation, for subsequent verbatim transcription and analysis. Basic data were also collected to characterise the sample (position or role, type of institution or organisation, field of activity, mode of participation and gender).

Individual interviews

The individual interviews were conducted in person, either at the institutional headquarters or at each informant’s place of work. The sample encompasses the various levels of governance involved: the supra-municipal level, the municipal level (in its technical and political aspects), public innovation, the academic sphere and the third sector (see

Table 1. Profile of interviewees (n = 10).

Code	Position/Role	Institution/Sector	Profile
E1	Environmental technician in agricultural land and urban allotments	Supra-municipal administration	Institutional technician
E2	Technician in sustainable food and food governance	Local government	Food governance specialist
E3	Agriculture and environment technician in municipal allotments	Local government	Allotment management technician
E4	Political office responsible for urban allotments	Local government	Policy-maker
E5	Political office for urban services, environment and agriculture	Local government	Policy-maker
E6	Gardening technician involved in allotment regulation	Municipal administration	Municipal technician
E7	Lecturer and researcher in urban and peri-urban gardens	Public university	Academic expert
E8	Technician in agricultural governance and allotment management	Local government	Agricultural governance specialist
E9	Urban innovation technician in green infrastructure projects	Public innovation body	Innovation technician
E10	Head of a third-sector organisation coordinating an inclusive municipal allotment	Nonprofit organisation, in collaboration with the local authority	Third sector

). Of the ten interviewees, six were men and four were women.

Focus groups

The sample includes the views of university lecturers specialising in urban planning and urban gardens, associations and groups managing community gardens, garden users and management committees, executive committees of managing associations, and users of a garden provided by an association. The composition of each focus group is detailed in

Table 2. Composition of the focus groups (n = 5; total participants = 27).

Code	Organisation/Venue/Profile	Participants	Relationship with Urban Gardens
FG1	University lecturers in urban planning	3	Academic, teaching and research experience in urban planning, green spaces and urban gardens.
FG2	Representatives of associations and groups managing a community urban allotment	7	Direct involvement in the community management, maintenance, social use and revitalisation of the allotment; includes members of the management committee, neighbourhood association and social groups.
FG3	Users, members of the management committee and local authority staff at an urban allotment	5	Direct involvement in the cultivation, maintenance and monitoring of the plots; providing guidance to other users; informal monitoring of compliance with rules; collaboration with the local authority.
FG4	Members of the executive committee of an urban allotment management association	5	Direct involvement in the management of the allotments by an association; formalisation of the association; advocacy for neighbourhood-led management carried out by the users themselves.
FG5	Participants in a socially inclusive allotment workshop	7	Direct participation in the day-to-day activities of the allotment as part of a workshop run by a socio-inclusive association, with support from the coordinating organisation and volunteers; linking gardening activities to processes of social inclusion and well-being.

.

Table 3. Distribution of participants by garden and profile (N = 37).

Garden/Group	Institutional Actors	Associations	Gardeners/Users	External Experts	Total
Sociópolis	1	–	–	–	1
Mislata	3	–	3	–	6
Alaquàs	1	–	–	–	1
Godella	1	–	–	–	1
Espai Verd Benicalap	1	7	–	–	8
Malilla	–	5	5	–	5
Pentagrama (Llíria)	–	1	7	–	8
Cross-cutting actors	3	–	–	–	3
Academic experts	–	–	–	4	4
Total	10	13	15	4	37

Participants with multiple roles are counted in each relevant category; the Total column counts each person once.

shows the distribution of participants by garden and profile. Overall, the sample included 10 institutional actors (municipal and supra-municipal technical and political staff), 13 representatives of associations, 15 gardeners or users and 4 external academic experts. Several participants held multiple roles: the five board members of the managing association in Malilla and the members of the management committee in Mislata are also plot holders, as is the participant in Mislata who provides informal training to new users, while one municipal officer in Mislata is a former plot holder. Representation across gardens is uneven: three cases (the institutional allotments of Sociópolis and Alaquàs and the hybrid allotment of Godella) are represented only by institutional actors, whereas the self-managed cases are represented mainly by associations and users. In addition, three institutional informants (E1, E2 and E6) provided a cross-cutting perspective on the municipal and supra-municipal framework rather than on a single garden.

Data collection continued until thematic saturation was reached, defined as the point at which the themes and categories derived from the interview guide began to recur in successive sessions without providing substantially new information on the governance models analysed [38,39].

2.3. Guide to Interview Topics

The aim was to analyse and compare the governance models of urban allotments in Valencia and to explore the social dimensions associated with participation, inclusion and learning. To this end, a flexible thematic interview guide was drawn up, based on the specialist literature, containing the minimum necessary content to ensure these dimensions were covered without influencing the respondents’ answers. The guide was structured into ten thematic blocks, each opening with a broad, non-leading question, the order and depth of which could be adapted according to the interviewee’s profile and the flow of the conversation. A preliminary coding framework informed by the interview guide and the specialist literature was refined inductively through iterative review of the transcripts, resulting in a final system of 10 categories and 39 subcategories.

Following the continuum of governance models described in the specialist literature, the first sections seek to situate each allotment between grassroots community self-management (bottom-up, in which the community decides and manages [22]) and top-down municipal management (in which the administration defines and controls [29]), including hybrid configurations in which both poles collaborate [23], models of external support in which actors from outside the community assume leadership or funding roles [40], systems for allocating plots within a public framework for individual use [41], and models mediated by third parties, such as cultural or community organisations that act as a link between the administration and the users [26].

Table 4. Governance models of community urban allotments along the bottom-up/top-down continuum.

Governance Model	Position on the Continuum	Who Decides and Controls	Community Autonomy	Role of the Administration and NGOs	Distinctive Feature	Main Advantage and Risk
1. Community self-management (grassroots)	Extremely bottom-up	The gardeners; the community sets priorities, rules, the division of labour and conflict resolution	Maximum	Minimal or non-existent; the organisations involved are grassroots, integrated within the community itself	The community not only participates but manages the allotment in its entirety	Advantage: maximum ownership, sense of belonging and local adaptation. Risk: vulnerability if rules or mediation are lacking.
2. External support without loss of local leadership	Bottom-up with support (close to the hybrid model)	The community takes the lead; a local council, university, NGO or public agency provides support	High	Support infrastructure (land, funding, staff, training), not a substitute for the community	External support provides backing without taking the lead; this is usually linked to the project’s longevity	Advantage: resilience and continuity, particularly in low-income neighbourhoods. Risk: dependence on external resources
3. Hybrid arrangements (co-governance)	Midway on the continuum (neither top-down nor bottom-up)	Shared leadership between the community and the organisation	Medium; division of roles	The institution provides land, funding, mediation, legal certainty and a link to urban policies; the community retains control over the day-to-day running of the allotment	Avoids the community/administered dichotomy; requires a finely tuned governance structure	Advantage: combines resources, public legitimacy and organisational capacity. Risk: unclear responsibilities and decision-making powers
4. Institutional management mediated by NGOs or associations	Cross-cutting; somewhere between hybrid and top-down depending on where real power lies	An external entity (cultural institution, NGO or promoting association) becomes a central actor	Medium–low; power may shift towards hierarchical arrangements	The third party manages the project, obtains permits, ensures continuity and provides resources	This arises when the allotment needs an intermediary to survive; it ranges from support to substitution	Advantage: ability to sustain and extend the life of the community garden. Risk: limited user involvement in decisions
5. Municipal or top-down management	Extreme top-down approach	The public administration centralises regulatory and strategic power	Low; the community loses influence over strategic decisions	The administration plans, regulates and protects the land; NGOs play a secondary or implementing role	It defines the “operational space” of the allotment; it can be both protective and disciplinary	Advantage: order, regulatory continuity and land protection. Risk: participation reduced to instrumental tasks and reduced sense of ownership

summarises these configurations, which informed both the interview guide and the preliminary coding framework.

The remaining sections address day-to-day operations, participation and social inclusion, conflicts and their resolution mechanisms, knowledge and learning, environmental and urban impact, and the long-term sustainability of the model. Each of these ten sections is broken down into subsections that first present a comparative analysis of the allotments studied and then the direct quotations that support it.

2.4. Data Analysis and Ethical Considerations

The empirical material was analysed using qualitative thematic analysis. All interviews and focus groups were transcribed verbatim, and the set of transcripts constituted the corpus for analysis. Based on the system of categories and subcategories described in Section 2.3, each relevant discourse fragment was assigned to the corresponding subcategory and labelled with the informant’s code (E1 to E10) or the focus group’s code (FG1 to FG5), enabling a cross-cutting analysis that compares cases within each subcategory. The combination of individual interviews and focus groups enabled intra-method triangulation, by contrasting the individual perspectives of institutional, technical and academic informants with the collective dynamics of community actors and service users. The process continued until thematic saturation was reached, and the complete set of coded quotations is presented, organised by category, subcategory and coder, in Table S1 (Supplementary Materials). Some quotations illustrate more than one dimension and were coded into two subcategories; they are identified by sharing the same code and, in Table S1, are transcribed only once under their main subcategory, with a cross-reference from the second.

Each allotment was assigned to a governance model on the basis of three analytical criteria derived from the dimensions used to characterise the models in Table 4: (a) who establishes the regulatory framework, including by-laws or internal rules, the allocation of plots and sanctions; (b) who takes day-to-day and internal decisions; and (c) whether institutional support consists only of providing resources or also involves control over decisions. A case was classified as institutional when the administration, or a body acting on its behalf, sets the framework and retains the final decision on allocation and sanctions, with user bodies playing an advisory role or coordinating activities among users. It was classified as hybrid when decision-making is formally shared: the administration sets the framework and retains key decisions such as the allocation and withdrawal of plots, while an association of users with its own governing bodies decides on day-to-day operation, internal rules and common resources. It was classified as self-managed with support when the community or managing association decides on allocation, internal rules, sanctions and day-to-day operation, while the administration provides the land or other resources and sets the conditions for their use, for instance through a tender or the approval of a management project, without deciding on internal matters. The provision of land alone was therefore not considered sufficient to classify a case as hybrid; the decisive criterion was the distribution of decision-making power. In borderline cases, in which an association with its own governing bodies operates within a municipal framework, the case was classified as hybrid when the council retains the allocation and withdrawal of plots, and as self-managed with support when these decisions lie with the association. As most cases combine elements of more than one model in practice, this classification refers to the formal position of each allotment on the continuum, whereas its actual operation is examined in the Results and Discussion Sections.

From an ethical standpoint, all participants were informed of the study’s objectives, the voluntary nature of their participation, the academic use of the data and the option to withdraw at any time; prior to each session, their informed consent and authorisation for recording were obtained. The data were treated confidentially and anonymised using alphanumeric codes (E1–E10 for individual interviews and GF1–GF5 for focus groups), and information that could directly link each code to a specific allotment was minimised, in accordance with the European Union’s General Data Protection Regulation (EU) 2016/679. Furthermore, the study was conducted in accordance with the applicable European and national ethical guidelines for research involving human subjects and received a statement of support following a prior ethical review by the Human Research Ethics Committee of the University of Valencia (Application Reference 2024-SOCTUR-3769066, 13 October 2025).

3. Results: Comparative Analysis of Focus Groups and Interviews

The presentation of the following results proceeds from the general to the specific, from the characterisation of the governance model to the conditions for sustainability and future prospects. The complete set of verbatim quotations underpinning the analysis, organised by category, subcategory and coder, is set out in Table S1 (Supplementary Materials).

3.1. Governance and the Institution–Community Relationship

3.1.1. Type of Governance Model

The community gardens studied fall into three governance models, viewed more as points on a continuum than as distinct types: (1) top-down (E8_1), with the administration setting the framework and associations providing limited, secondary support (E8_2), sometimes through associations created within the allotment itself (E8_3); (2) hybrid arrangements, in which an association of users, established within the terms of use set by the local council, manages day-to-day operation under municipal oversight (E3_4); and (3) bottom-up with support, where a community or association leads the management and the local authority provides land, funding or guidance without taking on day-to-day operations (GF2_E1_5, E10_6, GF4_E4_7). These models are hereafter referred to as institutional, hybrid and self-managed with support, respectively. Of the five configurations identified in the literature (Table 4), pure community self-management was not observed, since all cases depend on public land or institutional support. The model mediated by a third party is represented by the therapeutic allotment, managed by a third-sector association of people with mental health conditions and their families. Unlike the cases described in the literature, in which the intermediary tends to shift power towards more hierarchical arrangements [26], decision-making here lies with the association, so the case is classified among the self-managed models with support. The criteria used for this classification are set out in Section 2.4. Fundamentally, the allotments are conceived as urban facilities for social interaction, not merely as spaces for cultivation (GF1_E1_8).

Three cases are institutional. In the first, a municipal agricultural body manages the garden on behalf of the local council (E8_1); its manager prefers this format to self-management, which the manager associates with problems when those who cultivate also make decisions (E8_9), and relies on the separation between the manager and user (E8_10), involving associations only through agreements (E8_2), including one set up for that purpose within the allotment (E8_3). In a second municipal allotment, public management is deemed essential and self-management unfeasible with more than one hundred users (E4_11): the users’ committee advises and issues warnings, but neither makes decisions nor imposes sanctions (GF3_E4_12, GF3_E4_13), and the project originated as a political initiative to protect agricultural land (E4_14). In a third, despite its citizen-led origins (E5_15), the role of the administration is emphasised (E5_16), supported by an advisory committee (E5_17). The fourth case, by contrast, is hybrid: municipal operating regulations frame an ad hoc users’ association (E3_4) with compulsory membership (E3_18), which grew out of a university-led initiative (E3_19) and runs day-to-day operation through its own board and assembly, while the council allocates and withdraws plots and ensures that internal decisions remain within the regulations.

Three allotments are self-managed with support. In one, management was handed over to a federation of neighbourhood associations (GF2_E1_5), with compulsory membership for access (GF2_E2_20) and unified representation before the administration (GF2_E1_21); it emerged from a public innovation project intended to catalyse citizen participation (E9_22), from which the local authority later withdrew entirely (E9_23). The second, a therapeutic centre run by a third-sector organisation, operates on municipal land and in a municipal building (E10_6), combining high operational autonomy with dependence on public assets (E10_24) and an informal internal organisation centred on key figures (GF5_E6_25).

In the third, a neighbourhood association was set up so that the allotment holders themselves, rather than an external body, could manage it (GF4_E4_7, GF4_E5_26), in the belief that this fosters greater involvement (GF4_E3_27), although the land is municipal and management is granted through a periodic contract awarded by tender, for which the association must submit a management project to the council, whose framework also acts as a backstop for internal rules (GF4_E4_28). This case lies close to the hybrid model, as its municipal framework is set through a tender-based contract, but it is classified as self-managed because the association itself decides on admissions, internal rules and sanctions, including the withdrawal of plots, whereas in the hybrid case these decisions remain with the council. No case therefore occupies the extremes: top-down models rely on day-to-day support from associations, whilst bottom-up models depend on land and institutional support. Hybridisation between the institutional framework and community self-organisation is the cross-cutting feature of the sample. Figure 2 summarises the formal position of each allotment on the continuum and the shifts observed in its actual operation, and

Table 5. Comparative analysis of seven urban allotments: governance model, benefits and risks of the model, and social, urban and environmental benefits.

Allotment	Governance Model	Benefits of the Model	Risks of the Model	Social Benefits	Urban Benefits	Environmental Benefits
Malilla Allotment	Self-managed model with support (bottom-up): Management by the Geshuma neighbourhood association through a periodic municipal lease subject to a tender process, for which the association submits a management project to the Town Council. It has an extensive board of directors (18 members), a general assembly and its own internal regulations with a graduated system of disciplinary measures. It operates alongside five to seven partner associations.	High and sustained involvement (participation in joint activities exceeding 25 per cent), operational autonomy and a board with a rotation system. The association acts as the “eyes of the council” and ensures compliance with internal regulations. A model consulted by other local authorities interested in replicating it.	Periodic contract renewal (precedent of closure in San Isidro), lack of a stable municipal contact following the retirement of the key civil servant, and dependence on the electoral cycle. No involvement in the by-law and only reactive institutional support.	An intergenerational community (aged 20 to 85) and the inclusion of vulnerable groups (mental health, children, women, multiculturalism) through partner organisations; 220 people on the waiting list. Socio-professional diversity and educational activities with schoolchildren.	Integrated into Malilla Park as a green space planned under the PAI; partially compensates for the loss of the traditional vegetable garden. Densely built-up location, accessible on foot to around 50,000 residents without a metro link.	Sector-specific composting, biodiversity (insect hotels), agroecology and reused water. Organic farming is mandatory by by-law and makes use of shredded municipal prunings; a philosophy of a complete life cycle is demonstrated during school visits.
Sociópolis Allotment (La Torre)	Institutional model (top-down), with centralised management: The Municipal Agricultural Council (Consell Agrari Municipal) has managed Sociópolis on behalf of the City Council since 2008 and oversees all processes. No member of the Council has a plot, which guarantees impartiality. It operates alongside six associations under an agreement and in conjunction with the Water Tribunal (Favara irrigation channel).	Consolidated institutional stability and impartial decision-making through the separation of manager and user. Extensive support (training in organic farming, pest control and composting; 3+1-year leases) and de facto flexibility, with extensions granted for well-maintained plots.	Friction between two concurrent regulatory frameworks (restrictive guidelines from the council and a municipal by-law geared towards self-management). Bureaucratic overload and a digital divide incompatible with an older user demographic; rigid timetables and health-related issues caused by the heat.	For vulnerable groups (people with disabilities, those with mental health issues, marginalised women), it functions as a “healing project”; for older people, it functions as a structuring routine. Dynamics of reciprocity (cross-watering, time bank) and a strong sense of belonging.	Part of the Sociópolis urban development project (La Torre neighbourhood): a bridge between the city and the countryside and an element of spatial planning. Some 280 plots ranging from 65 to 90 m2 on public land set aside for community use, within a framework of traditional Valencian farming practices.	Organic farming (organic fertiliser, biological pest control, composting). Heritage value of traditional gravity-fed irrigation via the Favara irrigation channel; water scarcity remains an unresolved issue.
Mislata Allotment	Institutional model (top-down): A municipal initiative since 2017, created to protect agricultural land and revitalise the Pou del Quint. Municipal ownership and a regulatory by-law; 105 plots, a minimum of 5 years’ registration on the municipal register and a public lottery. A voluntary Management Committee that advises and inspects whilst the Town Council makes decisions.	A formal procedure with safeguards (hearing process, regulated revocation), low cost and a partnership with the Water Tribunal (free allocation of water). A high degree of operational autonomy within the formal framework and well-established waiting lists.	An ordinance in need of comprehensive review and the exhaustion of the volunteers who sustain the Committee. Administrative overload (around 200 cases) without a specific budget allocation; underhand tactics to accumulate plots and latent pressure on land.	High demand with a reserve list, predominantly comprising older people and pensioners. Daily interaction, exchange of produce, and physical, psychological and social benefits; inclusion through plots reserved for schools and an occupational centre.	A space for relaxation in one of Europe’s most densely populated municipalities. A central, pedestrianised location on land zoned as non-building land; the local council removed a roundabout from the master plan to preserve it.	Minimisation of chemical use, prioritisation of organic farming in the lottery selection criteria, and preservation of traditional irrigation via irrigation channels. Use of shredded prunings and future installation of a composting facility; the century-old well serves as a strategic reserve.
Godella Allotment	Hybrid model: The Town Council leases land from four private landowners and makes it available to users through the La Coscollosa association, which has compulsory membership, with an assembly, a board and self-funding (membership fee of €40). Regulations dating from 2013 and compulsory agroecological farming.	High operational autonomy and very smooth communication with the Town Council. Multidimensional support (land lease, annual grant of €3000, training, tools and legal support); well-established collaboration after twelve years.	The 2013 regulations are obsolete (the planned biennial lottery was never implemented) and there is a lack of flexibility in amending them. Dependence on private, rather than municipal, land, and pressure from “villa development” in a Grade 1 protected market garden area.	High demand and a very diverse membership (various social classes, ages and nationalities). A strong sense of belonging, mutual support and communal work days; benefits for older people and vulnerable groups through agreements.	Peri-urban location next to the Moncada irrigation channel, on Grade 1 protected agricultural land, acting as a transition between the town centre and traditional farmland. Advocates for the integration of allotments into urban design as a specific amenity.	Mandatory agroecological cultivation and preservation of traditional plant material (home-grown seedlings, seeds contributed by migrants). Conversion of a former rubbish tip into a reservoir of peri-urban biodiversity.
Espai Verd Benicalap Allotment	Self-managed model with support (bottom-up): Federative self-management on a municipally leased plot: the public foundation Las Naves provides support (initial funding from the European GrowGreen project, guidance) and then withdraws, with management falling to a federation of 15 to 20 neighbourhood groups with a governing committee. Distinguishing feature: The allotments must be collective, not individual, with statutes approved by the users.	Collective cultivation allows organisations to share tasks, while the federation coordinates external relations. Participants described operational autonomy alongside limited continuing support and concentrated workloads.	Reliance on the driving group and burnout amongst volunteers (low actual participation, 4–6 active members), recurrent vandalism and theft, lack of ongoing funding, and territorial uncertainty due to the reversible designation as an educational facility.	Intercultural and intergenerational cohesion in a low-income neighbourhood with a high mix of migrant communities; effective inclusion of vulnerable groups (an African woman has chaired the Espai). Access via open membership, with no income requirements.	Regenerative function: Transforms a derelict site used as an informal car park into a place for meeting and production. It addresses the shortage of public facilities in the Ciudad Fallera and forms part of the urban greening policy; a public catalogue of sites is proposed.	Organic farming required by the statutes, a microhabitat with pollinators in fallow areas, and an edible syntropic forest with native reforestation. It forms part of nature-based solutions (GrowGreen) to tackle climate change.
Alaquàs Allotment	Institutional model (top-down): A project stemming from a citizens’ initiative (ten years): a proposal included in the White Paper and submitted to the full council. Municipal management involving environmental and education specialists, and a de facto binding advisory committee. Two by-laws (on allocation and penalties) with provisions for different user categories.	Extensive institutional support (land, well water, fencing, car park, drip irrigation, technical expertise and training). Amendments to the by-law are proposed, for the most part, by the users; priority is given to social groups and the garden is open to the public on Saturdays.	Ban on direct sales (barter only) and immediate expulsion without notice for the use of chemicals. Private land held under a nominal lease until its transfer to the local authority; risks relating to hydrological and climatic conditions and future environmental disinvestment.	200 active users (more than half on a daily basis) and a waiting list of 60–70 people; seven or eight nationalities. An emotional refuge (“better than any medicine”), financial savings and social interaction (Saturday lunches).	Structural function in a very densely populated municipality with stunted growth: an intentional green corridor (historic market garden, urban woodland, allotments and tree-lined avenue). Public land repurposed to preserve the agricultural area, with a green space ratio well above the WHO standard.	A biodiversity reservoir providing biological pest control; 100 per cent of water sourced from municipal wells and permeable concrete that returns water to the water table. Reduction in CO2 emissions and the urban heat island effect; natural preparations, reported restrictions on cultivation inputs.
Pentagrama Association’s Allotment (Llíria)	Self-managed model with support (bottom-up), managed by a third-sector association within a light institutional framework: Municipal land and buildings provided to the Pentagrama Association, which manages the allotment, the greenhouse and the nursery. Llíria Town Council owns and funds the initial infrastructure; retired farmer volunteers pass on their technical knowledge. Distinguishing feature: There are no plots or formal regulations, and the allotment operates as a therapeutic tool within a mental health programme.	High operational autonomy with tasks tailored to each person’s abilities and regulatory flexibility allowing for incremental development. Blurring of roles between users, volunteers and professionals; 19 years of experience and proven replicability across the region.	Consumables are funded entirely by the association (structural but not recurring operational support from the local council), and administrative procedures are perceived as complicated. Dependence on older volunteers with no guaranteed succession and on a municipal land lease with an agreed time frame.	Core focus: People with chronic mental illness; physical, attentional and emotional therapeutic benefits, self-esteem and a sense of belonging (“Pentagrama experienced as a family”). Collaboration with disability, older people and children, with a normalising effect.	Regeneration of an underused municipal facility and integration into the municipal mental health policy, in which Llíria was a pioneer. Land set aside for community use at the headquarters, with a second school garden opposite; no land-use conflicts.	Environmental dimension subordinate to the therapeutic focus: Low-impact practices (drip irrigation, composting, manure and natural substrate, self-defined as “organic”), but with no data on biodiversity or ecosystem services.

Source: Own analysis based on interviews (E1–E10) and focus groups (GF1–GF5).

summarises, for each of the seven allotments, the governance model, its benefits and risks, and its social, urban and environmental benefits.

3.1.2. Distribution of Responsibilities and the Role of Key Actors

In top-down and hybrid models, the administration sets the framework (by-laws, land allocation and supervision) and delegates day-to-day management to an operational arm, such as a committee, a board or a contracting body. This mixed governance is valued as a more mature form of shared responsibility, in which a designated manager coordinates and reports to the local council, relieving the administration of direct oversight (GF1_E1_29).

The operational arm takes various forms. In the centrally managed allotment, a municipal agricultural body handles applications and allocates plots through agreements (E8_30), leaving associations a limited role with ad hoc material support (E8_31). In the municipal allotment with a volunteer management committee, the separation of functions is explicit: the committee advises, reports and inspects, while decisions rest with civil servants (GF3_E4_32); it was set up to manage a large number of plots (E4_33), on the premise that the administration’s role is to ensure the system functions (E4_34), with volunteers carrying out periodic inspections (GF3_E2_35) and a municipal technician drafting status reports (GF3_E4_36). In a community-run allotment, technical support from the local authority underpins its operation (E5_37) and is considered a prerequisite for viability (E5_38). In a garden governed by regulations, the local authority allocates and withdraws plots (E3_39) and intervenes when the association fails to resolve issues (E3_40), whilst the association ensures day-to-day compliance (E3_41).

In supported bottom-up models, the focus shifts to the associative structure. A committee maintains the infrastructure, formalises documentation and brings in new associations (GF2_E1_42), although responsibility tends to be concentrated in one or two people, which reveals its fragility (GF2_E7_43); a management committee with representation from each group structures decision-making (E9_44), whilst the public promoting body, operating under a “four-helix” model (E9_45), provides only discreet support (E9_46). In the therapeutic garden, the association manages resources on municipal land provided on loan, with occasional and administratively slow institutional support (E10_47), and distributes tasks among volunteers with agricultural expertise (GF5_E3_48, GF5_E5_49).

In the self-managed allotment of neighbourhood origin, the community took over management by setting up a board of directors after the handover from the previous management body (GF4_E1_50), assuming tasks previously performed by municipal services (GF4_E1_51), although communication with the council is channelled through an independent body (GF4_E3_52). Around these core actors, a network of supporting actors emerges: social groups with plots reserved for vulnerable populations (E5_53), educational centres benefiting from land transfers (E4_54) and traditional irrigation bodies (E4_55). The result is a tiered structure: the administration retains the framework and the final say, a community-based operational arm runs day-to-day management, and a wider network of stakeholders extends the allotment’s social function.

3.1.3. Degree of Formalisation

Formalisation is generally high, although the regulatory instrument varies. In institutional allotments, a municipal by-law regulates access, plot sizes, irrigation and penalties (E4_56, E4_57; E5_58), and in one case even the layout and allocation of plots (E5_59). Where management is based on an agreement with the local community, there is no by-law but a set of rules governing allocation and use (E3_60).

Formalisation is supported by monitoring tools such as the minutes of the management committee (GF3_E4_61) and a photographic record of neglected plots (GF3_E2_62). In the centrally managed allotment, however, it is contested: the manager applies their own rules and criticises that the by-law was drafted without consulting the managing body (E8_63) and presupposes a form of self-management he considers unworkable (E8_64).

In the self-managed allotments, formalisation is internal but robust. One has its own internal rules with a graduated system of offences that may lead to the loss of the plot (GF4_E1_65), and registration in the register of associations was a requirement for the allocation (GF4_E5_66). The other has internal regulations (GF2_E2_67) and statutes providing for plots to be reclaimed in the event of inactivity (GF2_E1_68).

In the allotment regulated by rules, formalisation combines rules of use with technical requirements, such as mandatory agroecological cultivation (E3_69), and decisions such as abolishing the public price to strengthen the association’s economic self-organisation (E3_70). The therapeutic garden is the exception, with low formalisation: it began as an informal proposal by the association for a municipal space with no defined use (E10_71). Overall, formalisation is high where the framework is institutional and shifts towards self-regulation where management is community-based, but no allotment operates without rules.

3.1.4. User Autonomy and Institutional Support

User autonomy varies inversely with the degree of centralisation. In the most centralised model, the managing body retains overall control but allows associations autonomy in their internal affairs (E8_72). In the hybrid model, users gain real influence when dialogue with the administration is fluid and their requests are addressed (E3_73).

Institutional support is crucial to the viability of the allotments. Technical support from the council underpins day-to-day operations (E5_37), and without a body to guide the process, a community initiative may become a victim of its own success (E9_74). In municipalities with fewer resources, higher levels of government compensate for this shortfall (E1_75), and the local authority responds promptly to material needs when these are raised (GF3_E1_76).

The flip side of autonomy is fragility when support fails. In a self-managed allotment, the community takes on burdensome tasks precisely because of the lack of municipal support (GF4_E4_77), and the loss of a key contact person leaves it without a link to the administration (GF4_E4_78). Autonomy thus appears sustainable only when accompanied by stable institutional support.

3.1.5. Dependence on the Local Council and the Quality of Collaboration

All the allotments ultimately depend on the local council, which provides the land and sets the framework, but the nature of this dependence varies. In self-managed gardens, it is existential: the project’s continuity was called into question when a change in management led the council to threaten closure (GF4_E4_79), and the provision of land is conditional upon a minimum level of use, reflecting the council’s interest in preventing the site’s deterioration rather than in promoting the garden (GF2_E6_80). In the therapeutic allotment, the dependence is financial: the association’s investment only makes sense if the land lease is secured in the medium term (E10_24).

The quality of collaboration also varies. At the positive end, the users’ committee describes itself as the “eyes” of the local council (GF3_E2_81) and dialogue with the administration is fluid and responsive (E3_73). At the other end, collaboration is perceived as non-existent or merely formal: one group states that it has received no financial support (GF2_E6_82), and another describes an administration that only reacts to crises, not to routine requests (GF4_E1_83). Even in the most established institutional case, the regulation was drafted without consulting the managing body (E8_63).

Collaboration tends to be reactive and poorly planned. Experts highlight the absence of a systematic municipal strategy, with support linked to political expediency rather than to a sustained public policy (E7_84). Thus, the allotments depend on the council for their very existence, but the quality of collaboration rests more on individual will and circumstances than on a stable framework of shared responsibility.

3.2. Day-to-Day Operations and Practices

3.2.1. Organisation of Day-to-Day Work

Day-to-day work combines individual effort with collective tasks, and its coordination depends on the irrigation system: canal irrigation concentrates work on irrigation days, whereas drip irrigation distributes it over specific days, as illustrated by the weekly shifts in Mislata (105 plots) (GF3_E3_85), the “pozo de secà” in Godella (E3_86), gravity irrigation in Sociópolis (E8_87) and the wells using natural slurry in Alaquàs (E5_88). In Malilla (GF4_E2_89) and the Espai Verd in Benicalap (E7_90; GF2_E2_91), collective participation is reinforced through the registration of activities or a mandatory community-based approach, in contrast to the allotments open to the public in Valencia (E6_92).

3.2.2. Decision-Making

Decisions operate at an operational (day-to-day) level and a strategic level (by-laws, irrigation, conflicts), distributed across a tiered structure of assemblies, meetings, committees, technical officers and councillors (GF1_E2_93). Mislata raises issues through its management committee and agrees on the by-laws with the technical officer and the councillor (GF3_E2_125); Malilla combines a monthly meeting with an assembly (GF4_E3_94); Godella distinguishes between a daily meeting and an annual assembly under non-intervening municipal oversight (E3_95); and in Alaquàs, the non-binding advisory committee effectively influences the Town Council’s decisions (E5_96). In Sociópolis, decision-making is centralised within the Consell Agrari (E8_31), whilst in the Valencia model, matters are resolved by the Parks and Gardens Department (E6_97).

3.2.3. Resource Management

The funding of basic resources (plot, water, tools, fencing, infrastructure and maintenance) ranges from almost full municipal funding to users covering most of the costs. Mislata receives land, weekly water supplies, fencing and the shredding of prunings in exchange for a deposit (GF3_E2_98); Alaquàs has its own municipal wells and a sustainable drainage system (E5_99); Godella leases private land and combines an annual grant of 3000 euros with user fees (E3_100); Malilla supplements municipal contributions with fees for additional costs (GF4_E2_101); the Espai Verd in Benicalap, after an initial grant of 12,000 euros for fencing, is sustained through recycled materials (GF2_E1_102) and the support of Las Naves, the public foundation of the Valencia City Council that led the European GrowGreen project; and Sociópolis relies on water bottles, as watering is limited to alternate Tuesdays (E8_103). In the Valencia model, the City Council provides land, mains water, lockers and toilets (E6_104), whilst the Provincial Council highlights the routine use of drinking water (E1_105) and the UPV points to the ban on selling surpluses as a structural handicap (GF1_E1_106).

3.2.4. Discrepancies Between Regulations and Practice

Discrepancies between regulations and practice fall into five categories: individual opportunism, tacit personal exceptions, difficulty in imposing sanctions without a waiting list, structural problems with the regulations, and clashes between by-laws and reality. In Mislata, plots are accumulated through family connections, and a merit-based system was rejected because it introduced subjectivity compared with a lottery (GF3_E2_107); in Malilla, there are instances of symbolic appropriation and covert subletting, penalised by the loss of the plot (GF4_E5_108); the Espai Verd replaces revocation for inactivity, which is inapplicable without a waiting list, with a temporary transfer (GF2_E1_109); in Godella, the biannual lottery has never been held and plots have been occupied since 2013 (E3_110); and Alaquàs tolerates abandonment due to illness but expels users immediately if chemicals are used (E5_111). In Sociópolis, the Valencia model and the surroundings of the Espai Verd Benicalap, further tensions arise from the impracticability of self-management (E8_112), uses not covered by the regulations, such as school gardens (E6_113), and perimeter issues such as informal parking (E9_114).

3.3. Regulatory Framework and Tensions

3.3.1. Legislation and Regulation

The regulatory framework combines municipal by-laws, tender specifications and internal regulations, forming a functional but friction-prone structure. In institutional allotments, the by-law is conceived as a facilitating instrument that ensures essential requirements are met without hindering use (GF3_E3_115), based on a deliberately simple set of rules for coexistence that nonetheless require periodic reminders (GF3_E2_116). In the hybrid allotment, the instrument is not a municipal ordinance but a set of rules governing allocation and use, which requires membership of an association and exclusively agroecological cultivation, and which municipal technical staff enforce in constant consultation with the association and the users (E3_117).

Experts stress that these spaces cannot develop organically without guidelines or a certain degree of control, lest they degrade the urban landscape (GF1_E1_118). One regularity stands out: the deeper the institutional roots, the denser the formal regulation; the greater the self-management, the heavier the internal self-regulation. No case, however, operates outside a body of rules.

3.3.2. Access and Allocation of Plots

Access is where the models differ most, and in almost all of them demand exceeds supply. In the municipal allotment with a management committee, plots are allocated by lottery amongst residents registered for at least five years, as a guarantee of equality in the use of a public asset (GF3_E4_119), and membership of the committee confers no privilege in the lottery (GF3_E2_120); the by-law also provides for direct allocations in the public interest to educational establishments and community groups (GF3_E4_121).

In the self-managed neighbourhood allotment, a public tender for management, which requires a project meeting all formal requirements (GF4_E3_122), coexists with an internal waiting list that far exceeds the number of plots; indeed, the association had to be formally established beforehand in order to apply (GF4_E3_123). In the allotment governed by regulations, standard access is also by lottery, with an annual fee and a deposit, whilst social plots are managed through social services and are exempt from payment. Access thus functions as the main mechanism for equity, but also as the point where unmet demand exerts the greatest pressure on the model.

3.3.3. Regulatory Rigidity Versus Flexibility

A good framework combines rigidity in essential matters with flexibility in operational ones, avoiding both a disconnect from practice and from arbitrariness. In the municipal allotment with a management committee, the rules are rigid regarding the lottery but flexible regarding justified absences, and appeals are addressed without turning management into punitive surveillance (GF3_E4_124); the by-law itself is continuously updated to add what is necessary and remove what is obsolete (GF3_E2_125). An attempt to replace the lottery with a merit-based system was rejected because it would introduce subjectivity instead of the objectivity of chance (GF3_E2_126).

Self-managed models share a flexible mechanism for revocation due to inactivity, with periods of two to four months set out in each set of rules (E7_127), which the association-based self-managed allotment does not apply to because it has no waiting list for replacements, making a permissive approach necessary (GF2_E1_128). Flexibility is thus the means by which everyday practice adapts rules that would prove dysfunctional if applied literally.

3.3.4. Tensions

The analysis identifies five types of recurring regulatory tension: property-related, political, multi-level, economic and procedural. Property-related tensions stem from precarious tenure: in the self-managed community-based allotment, the land is zoned for future public facilities and can be reclaimed by the local authority, a risk mitigated only by the symbolic backing of European funding for the project (GF2_E2_129). Political tensions arise from dependence on the electoral cycle and the stance of the local government, whose dialogue with associations may cool after a change of administration (GF4_E1_130).

Multi-level tensions stem from the division of powers and the difficulty of controlling land use on land owned by others, particularly private land (E7_131). Economic tensions centre on the ban on selling surplus produce, which confines the allotment to self-consumption and limits the associations’ financial autonomy (GF1_E1_132). Procedural tensions take the form of administrative burdens and rotation rules that, by forcing a change of users after a few years, reopen the waiting list and jeopardise continuity of participation (E6_133). These accounts identify legal and administrative difficulties alongside organisational vulnerabilities, including uneven participation, concentrated workloads and uncertain volunteer succession.

3.4. Participation and Community

3.4.1. Level of Active Participation

Most allotments have a waiting list, a sign of unmet social demand, although participation varies once people have a plot. In the self-managed allotment of neighbourhood origin, more than 25 per cent of registered members take part in each communal activity, a high level for a group of around eighty users (GF4_E4_134), driven by a group that took over management and chose to continue the project on its own initiative (GF4_E2_135). In the municipal allotment initiated by local residents, demand predated the project itself, sustained by neighbours with an agricultural background and by people from other regions accustomed to growing their own food (E5_136). The common pattern is high participation centred on a stable core group, mostly made up of retired people.

3.4.2. Sense of Belonging

The sense of belonging is underpinned by inherited rural traditions, by the routine and sense of purpose that the allotment offers older people, and by intergenerational transmission. In the municipal allotment with a management committee, losing one’s plot would mean losing a daily source of joy (GF3_E1_137), and the bond is described as a devotion to the countryside rather than as a productive activity (GF3_E2_138). In the therapeutic allotment, belonging to the group is considered the foundation of the project and a prerequisite for its effects to last beyond the site (E10_139). In the centrally managed allotment, belonging verges on a sense of quasi-ownership, which the manager corrects by reminding users that the plot is an administrative concession. The supra-municipal administration highlights the garden’s role in breaking isolation, even amongst neighbours who have their own gardens (E1_140).

3.4.3. Community Dynamics

Community dynamics include daily greetings, mutual assistance, the exchange of produce and shared celebrations, which often constitute the true value of the allotment, and are reinforced by sociocultural diversity. In the municipal allotment with a management committee, neighbours organise watering through a messaging group when a plot holder cannot attend (GF3_E3_141), and the allotment fosters bonds that extend beyond its boundaries (GF3_E1_142).

In the allotment governed by regulations, communal work days extend into shared lunches that foster integration (E3_143), and people of various nationalities coexist with mutual language support (E3_144). In the citizen-initiated municipal allotment, intercultural coexistence is reflected in users introducing crops from their countries of origin (E5_145), whilst in the self-managed neighbourhood allotment, the community network ensures that everyone knows one another and participates, even if not everyone works equally (GF4_E4_146). In the centrally managed allotment, reciprocity operates as a time bank and is linked to a neighbouring school through a third-sector organisation, creating an informal institutional framework that supports day-to-day operations.

3.4.4. Inequality in Involvement

Even in the most cohesive allotments, involvement is uneven: an active core sustains day-to-day operations, an intermediate group fulfils its duties without engaging in collective tasks, and a minority takes no part. In the municipal allotment with a management committee, some users systematically ask others to water their plots (GF3_E2_147), and the management core itself is small and ageing, which reveals its fragility (GF3_E2_148). In the therapeutic allotment, involvement is adapted to each person’s physical capacity, with tasks divided between the greenhouse and the vegetable patch (E10_149). The supra-municipal administration links this divide to differing life expectations across generations and the gradual abandonment of inherited land (E1_150).

3.5. Conflicts and Regulatory Mechanisms

3.5.1. Types of Conflicts

Conflicts are internal, external or structural, with one type predominating depending on the model. In the allotment with a management committee, the main issues are the abandonment of plots and hoarding and petty theft amongst users (GF3_E1_151). In the self-managed allotment run by an association, external conflicts such as theft and damage to equipment and infrastructure prevail (GF2_E1_152). Experts observe that, in squatted allotments, coexistence is sustained by informal balances amongst users (E7_153). In institutional models, conflict tends to be channelled administratively, whilst in self-managed gardens it is resolved through collective deliberation.

3.5.2. Causes of Conflict

The causes recur across models and fall into five categories. The first is an overestimation of commitment: many people sign up without realising the dedication that cultivation requires (GF3_E1_154). The second is the symbolic appropriation of plots by long-standing users, who perceive temporary tenancy as de facto ownership (GF4_E1_155).

The third is the overburdening of volunteers, with a small core taking on unsustainable tasks (GF2_E1_156), exacerbated by poor communication between the project’s driving force and the associations it represents (E7_157). The fourth is the appointment of external management bodies not aligned with the allotment project, which leads to secondary management and, ultimately, to the abandonment of plots (GF4_E4_158). The fifth is regulatory non-compliance with effects on neighbouring plots (chemicals, pests, weeds). Conflicts are therefore rarely purely relational: they arise from the friction between the individual logic of the plot holder and the collective logic of the allotment.

3.5.3. Resolution Mechanisms

Resolution mechanisms follow a tiered approach, more formalised in institutional models and more deliberative in self-managed ones, but with the same basic sequence: a verbal warning, a formal notification, a deadline to rectify the situation and, as a last resort, a penalty. In the municipal allotment with a management committee, due process is guaranteed through a formal notice, a ten-day deadline and a hearing before revocation (GF3_E4_159). In the citizen-initiated municipal allotment, the procedure is entirely institutional: a preliminary report, the opportunity to submit representations and, if non-compliance persists, publication in the regional official gazette for expulsion (E5_160).

In the self-managed allotment of neighbourhood origin, the assembly acts as an internal tribunal, in the format of a residents’ association without an external administrator (GF4_E3_161), and cases not covered by the regulations are resolved by a vote at a general meeting (GF4_E1_162). In the delegated models, administrative intervention is reserved for serious situations.

3.5.4. Legitimacy of Rules and Authority

The legitimacy of the rules rests on shared pillars: democratic ownership, proportionality of sanctions, separation of roles and respect for users. In the municipal allotment with a management committee, the objectivity of the lottery legitimises allocation because it treats everyone equally (GF3_E4_163), and sanctioning serious misconduct such as theft is accepted as necessary to protect the rights of others (GF3_E2_164). In the self-managed allotment of neighbourhood origin, legitimacy is also supported by a minimal external framework from the local council, which places everyone on an equal footing (GF4_E3_165). Technical and institutional support reinforces legitimacy in delegated models, whereas in self-managed models it rests on the collective authorship of the rules.

3.6. Knowledge, Learning and Education

3.6.1. Learning Amongst Users and the Transfer of Agricultural Knowledge

Learning amongst users is horizontal and informal, supported by the most experienced and shaped by a willingness to learn (GF3_E2_166). In the allotment with a management committee, the committee itself advises less experienced users on what to plant or how to tend the soil (GF3_E2_167). Some participants linked this knowledge to cultivation practices learned in childhood (GF3_E1_168).

3.6.2. Educational Activities

Educational activities are the most institutionalised form of learning. In the municipal allotment with a management committee, school visits predominate (GF3_E3_169), with a hands-on approach in which pupils work the land directly (GF3_E2_170). In the citizen-led municipal allotment, schools visit regularly for practical lessons on site (E5_171).

An academic participant proposed wider community involvement in school gardens to reduce dependence on individual teachers (GF1_E1_172), while another expert identified educational collaboration between gardens as an underused opportunity to share expertise in agroecology, pest control and composting (E7_173). In therapeutic gardens, however, links with schools are weak due to school timetable constraints (E10_174).

3.6.3. Integration of New Users

New users are integrated mainly through immersion, with varying degrees of support. In the community-run municipal garden, the process is informal but effective: newcomers acquire basic knowledge within a few days with the help of their peers (E5_175). In the therapeutic garden, integration occurs through imitation, which blurs roles to the point that participants, volunteers and professionals are not always distinguished (E10_176). In community-based models, integration also requires formal membership of the association, which involves learning how its assembly-based decision-making works.

3.7. Social, Environmental and Urban Impact

3.7.1. Social Benefits

Social benefits form the cross-cutting core of the garden’s value, encompassing physical, psychological and relational dimensions (GF3_E3_177). The garden also supports the integration of migrant and low-income populations (E1_178) and helps families save on food costs in a context of rising prices (E5_179).

The therapeutic garden’s manager perceived that gardening helped some users focus on current tasks and tentatively mentioned fewer intrusive thoughts (E10_180). These impressions were not assessed through clinical measures.

3.7.2. Environmental Benefits

Participants reported soil improvement in Mislata (GF3_E1_181), drip irrigation intended to limit water use in Alaquàs (E5_182), and plant-based preparations for fertilisation or pest management (E5_183). The Alaquàs informant also attributed CO₂ reduction and heat mitigation to the site (E5_184). These environmental effects were not measured.

3.7.3. Urban Function

The urban function of the garden varies from case to case but converges on its role as green infrastructure. In the community-led municipal allotment, it forms part of a green corridor that links the historic market gardens with the residential fabric and buffers the impact of major roads (E5_185), and small collective allotments acting as climate refuges are proposed to increase green infrastructure where it is scarce (E5_186). Experts also value the allotment’s connection to the neighbourhood, which integrates it into the surrounding urban fabric (E7_187).

3.7.4. Perception of the Value of the Allotment

The value of the allotment is perceived positively in all cases, albeit with nuances. In the municipal allotment with a management committee, it is expressed as an almost euphoric daily satisfaction at harvesting one’s own produce (GF3_E2_188). In the self-managed allotment run by an association, there is frustration at not realising its full potential due to a lack of volunteers (GF2_E2_189). In the therapeutic garden, its value lies in the self-esteem and sense of purpose generated by bringing something to life from nothing (E10_190). Experts conclude that the value of the garden is considerable, but that public policy does not make sufficient use of it.

3.8. Social Inclusion and Public Service

3.8.1. Accessibility

Accessibility depends on admission criteria and physical proximity. In the municipal allotment with a management committee, access is by lottery among registered residents, whereas in community-based models it is conditional upon membership of the association. In the citizen-initiated municipal allotment, most users walk there because it is close to the town centre (E5_191). Experts note the reverse case of private allotments, whose long waiting lists and discretionary access restrict entry in practice (E7_192). Peripheral locations or physical barriers disadvantage groups with reduced mobility.

3.8.2. Inclusion of Vulnerable Groups

The inclusion of vulnerable groups is supported by provisions in local by-laws, statutory preferences and agreements with the third sector. In the municipal allotment with a management committee, a plot is reserved for an occupational centre on grounds of public interest (GF3_E4_193). In the community-led municipal allotment, one group donates its produce to cookery workshops for migrants (E5_194). In the garden governed by regulations, social plots are allocated to organisations working with people facing social exclusion, in need of reintegration or with mental health conditions (E3_195).

In the centrally managed allotment, partnerships involve mental health organisations, organisations representing people with disabilities and other groups (E8_196). The therapeutic allotment is the extreme case, organised entirely around the inclusion of people with mental health issues. Inclusion is thus a structural feature of the public allotment model, not a secondary addition.

3.8.3. Social Function and Equity of Access

Participants described family and intergenerational activities (GF3_E2_197) and links with local cultivation practices (E5_198). Access procedures included lotteries, waiting lists, association membership and direct allocations, with different eligibility conditions. An academic participant questioned restrictions affecting collective use of produce (GF1_E1_199).

3.9. Territorial and Urban Planning Dimension

3.9.1. Location and Accessibility

Proximity to urban areas is key to the success of allotments. The supra-municipal administration identifies water and proximity as the two decisive factors (E1_200), adding that a good allotment needs basic facilities such as rest and shaded areas, which are essential in summer (E1_201). Experts also highlight functional access (on foot, by bicycle or by car) as a prerequisite for good management (GF1_E1_202). In the citizen-initiated municipal allotment, proximity to the town centre allows most people to walk there, although a car park was provided to ease congestion (E5_203).

3.9.2. Land Tenure

The type of land determines the legal stability of each allotment and its vulnerability to administrative decisions. Four configurations are identified. The citizen-initiated municipal allotment is located on land formerly used for sport, prioritising the conservation of scarce agricultural land (E5_204). The self-managed allotment run by an association occupies land zoned for educational and cultural purposes and reserved for a future facility, a temporary status that its users accept (GF2_E1_205).

The municipal administration of the large city recognises this logic and sees scope in the ageing school network to consolidate uses such as market gardening (E6_206). The other two configurations are fully municipal public land and private agricultural land under territorial protection, leased by the administration and subject to specific restrictions. Ownership and land-use classification do not always match the actual use, which introduces structural fragility into several models.

3.9.3. Land-Use Conflicts

Land-use conflicts are either active or latent. An underlying pattern is cultural: in municipalities with a tradition of vegetable gardens, residents find it difficult to understand the concept of urban allotments, whose value is better understood in larger cities (GF1_E2_207). The origin of some allotments also reveals a conflict between horticultural use and public space: reclassification as a privatised public space made opening hours necessary as a safeguard (E6_208).

In self-managed community-based allotments, the conflict is latent, as the planning classification allows the land to be reclaimed at any time; in the hybrid allotment, the tension shifts to unauthorised uses on protected land, which are handled by the regional authority in the absence of a dedicated penalty system. The supra-municipal administration also notes that inherited land tends to fall into disuse when new generations do not cultivate it (E1_209).

3.9.4. Integration into Urban Planning

Integration into urban planning follows three patterns: consolidated, stable but vulnerable, and contingent. Despite this variability, shared proposals emerge. The citizen-led municipal allotment and the experts advocate a mandatory land reserve for allotments in general plans, specifying the type of green space and without reducing existing green areas (E5_210, GF1_E1_211).

The municipal administration of the large city frames this integration within a strategy for citizens to take ownership of public space (E6_212), whilst experts advocate networking amongst allotments, although previous attempts have not yet taken root (E7_213). Other proposals point to existing tools, such as a public catalogue of available plots identifying suitable locations for community gardens.

3.10. Sustainability and the Future of the Model

3.10.1. Continuity of the Project

Continuity varies according to the governance model and the stability of the institutional framework. The municipal allotment with a management committee looks to the future with confidence, drawing on the consolidation of its by-law (GF3_E3_214) and the internal harmony of the group (GF3_E1_215). The self-managed neighbourhood allotment maintains cohesion and a broad committee with monthly meetings aimed at continuous improvement, but views the pending contract renewal with concern.

At the opposite end, the self-managed garden run by an association shows signs of exhaustion amongst its active core, which acknowledges that it is barely surviving after years of overwork (GF2_E1_216), and the therapeutic garden’s continuity is conditioned by its dependence on a time-limited municipal land grant (E10_217). Continuity thus favours institutional or mixed projects backed by a stable regulatory framework, whereas self-managed projects depend on the succession of their volunteers.

3.10.2. Risks

Five categories of risk combine in varying ways in each case. The first is political: dependence on the electoral cycle, illustrated by the stalling of a municipal urban agriculture plan after a change of government. The second is legal and administrative: the rigidity of tender specifications, which in the self-managed community garden raises concern about contract renewal, given that a neighbouring garden was closed due to a simple administrative delay (GF4_E1_218).

The third is related to land: the risk of land being reclaimed because of its planning classification. The fourth concerns volunteers: the exhaustion of the active core and the lack of successors, exacerbated in the therapeutic garden by reliance on elderly volunteers with agricultural expertise (E10_219). The fifth is operational: health incidents linked to users’ age and weather conditions (E8_220), water shortages under restrictions, and the takeover of management by organisations without horticultural expertise. Institutional neglect in neighbouring municipalities illustrates what happens without public support (GF3_E3_221).

3.10.3. Scalability and Replicability

Scalability depends on three structural factors identified across cases: proximity to the town centre, prior organised demand and a stable water supply. When all three are present, the model can be easily replicated. Those running the municipal allotment with a management committee recommend studying previous experiences before designing a new allotment (GF3_E3_222) and highlight water as an essential prerequisite (GF3_E2_223).

The supra-municipal administration notes a documented ripple effect across the province linked to funding schemes (E1_224), and the self-managed, association-based allotment has already inspired a replica in another neighbourhood (GF2_E1_225). The therapeutic allotment shows a different path to institutional scaling, having been replicated by a supra-municipal body (E10_226). Inter-municipal knowledge transfer thus emerges as the main driver of the model’s dissemination.

3.10.4. Vision for the Future and Proposals

Proposals for the future converge in five areas: land reserves in general development plans, infrastructure improvements, legal protection through local by-laws, controlled marketing of surplus produce, and networking between projects and local authorities. In the community-run municipal allotment, priorities are operational, such as drip irrigation and shared services, subject to the available budget (GF3_E3_227).

In the citizen-led municipal allotment, the vision for the future involves strengthening water self-sufficiency by interconnecting well irrigation networks (E5_228), whilst the supra-municipal administration observes a growing interest amongst urban residents in these neo-rural practices (E1_229). The shared vision is a model based on regulatory stability, ongoing institutional support and adequate infrastructure, capable of sustaining the community cohesion that is its main asset.

Note on coding. The format of Table S1 is followed: interviews E{n}_{order}; focus groups GF{group}_E{participant}_{order}. The participant number indicates the order in which the person spoke in the focus group, and the final number is the order in which the quotation appears in the text (a single, consecutive series from 1 to 229 throughout the document). The data have been anonymised: the text in square brackets replaces identifiable names or provides context without altering the meaning of the quotation; figures are retained for their analytical value.

4. Discussion

The aim of this study was to analyse and compare the governance models of urban allotments in the Valencia Functional Urban Area and to explore the associated social dimensions, in particular participation, community cohesion, social inclusion and learning processes. The results support the initial hypothesis: the allotments studied do not follow a single model, but are distributed along a continuum ranging from community self-management to municipal or institutional management, with hybrid models and those receiving external support occupying intermediate positions [21,33]. However, the analysis qualifies this hypothesis on one key point, given that rather than discrete types, what is observed are unstable balances between participation, regulation and sustainability, and the position of each allotment garden on the continuum depends less on a pre-existing design than on the day-to-day interaction between the community and the administration.

Regarding RQ1, according to the criteria set out in Section 2.4, three allotments are formally institutional, one is hybrid and three are self-managed with support, one of which lies close to the hybrid boundary. However, as Figure 2 shows, the dominant configuration in actual operation is hybrid: the local authority sets the framework and provides the land, whilst an association or committee, together with horizontal relationships amongst users, maintains day-to-day management. This de facto hybridisation is evident even in cases with institutional roots, such as the centrally managed allotment, where the municipal agricultural body coexists with an internal system of reciprocity based on a “time bank” that handles irrigation, planting and maintenance, or the municipal allotment with a management committee, whose operation relies on a voluntary management committee of users. In the Valencian context, this distribution mirrors the triad of rental, associative and municipal models described by Palau-Salvador et al. [42] for the same metropolitan area, and aligns with the continuum of top-down, bottom-up and mixed approaches proposed by Fox-Kämper et al. [21]. The predominance of the hybrid model is consistent with the European trend, in which collaborative arrangements between civil society and the public sector are in the majority [33], and reinforces the idea that the dichotomy between the fully community-run allotment and the fully administered one has little analytical value [23,43].

Participants described institutional resources and internal coordination as relevant to operation and continuity. In Mislata, municipal responsibilities coexist with a volunteer users’ committee, while in Sociópolis centralised management coexists with mutual assistance among users. These arrangements also involve difficulties, including dependence on key volunteers. Related studies examine administrative support, autonomy and collaboration in different contexts [23,27,33]; the present findings do not establish a ranking of governance models.

Regarding RQ2, and consistent with this idea, governance models for urban allotments are neither fixed nor permanent configurations, but living structures that evolve over time. The international literature has clearly documented this capacity for change. Governance structures tend to take on different forms at each stage of the project—planning, implementation and day-to-day management—and the allocation of responsibilities amongst public authorities, associations and users is frequently readjusted as the allotment matures [21]. Zhang et al. [44] illustrate this sequence in the “We Garden” programme in Shenzhen, where a government-led approach in the planning and construction stages is followed by grassroots autonomy in the management stage, with nonprofit organisations acting as a bridge between the administration and residents; they also warn that this autonomy deserves particular attention when public funding is reduced or withdrawn. Furthermore, a common shift towards more top-down configurations has been observed, in which the administration regains a leading role and the gardening community loses strategic influence over the decisions that affect it [26,28]. Conversely, the opposite trajectory has also been documented. An initially institutional model, with the community playing a supporting role, evolves over time towards a grassroots model, sustained by the community but politically backed by the local council [43]. Both trajectories—bottom-up and top-down—demonstrate that the position of each model within the institutional–self-managed continuum is not a structural feature, but a situational one subject to review. The Valencian cases illustrate this dynamism and provide concrete examples of how such shifts occur. The recent deterioration in the relationship between the municipal administration of the large city and the association managing the neighbourhood-based self-managed allotment illustrates how a stable arrangement between a self-managed allotment and the administration can be weakened by seemingly minor causes, such as the retirement of the key municipal officer without a replacement being appointed. The trajectory of the association-based, self-managed allotment shows a different shift; having started as a project supported institutionally by the public innovation body and through European funding, it evolved towards a form of self-management that was not deliberately chosen, but rather assumed due to the sudden absence of ongoing public support, leading to an excessive workload and burnout amongst the active core group. This trajectory can be compared with the sequence described in Shenzhen, from a government-led launch to grassroots management [44]. In this allotment, participants associated reduced continuing public support with increased workloads for the active group. The intermediary organisations that Zhang et al. [44] consider key to this process are also present in the Valencian cases, in the facilitating role of the public foundation that promoted this allotment and in the third-sector association that manages the therapeutic allotment. And the case of the major city in the area—examined through the institutional perspective of its City Council rather than as one of the seven allotments analysed—reveals the model’s sensitivity to the electoral cycle; the suspension of the municipal urban agriculture plan and the uncertainty surrounding the future of the municipal food council following the change of government in 2023 demonstrate that public policies on urban allotments, when not safeguarded by regulatory instruments of sufficient legal weight, are vulnerable to discontinuity. These three examples show that a model’s position on the continuum is renegotiated continually. Its trajectory hinges on very specific factors: whether there is a stable technical lead, whether funding continues, how sensitive institutional support is to changes in municipal government, and how robust the regulatory framework behind the project proves to be.

Participants also described difficulties concerning tenure, administrative procedures, institutional support and internal organisation. These difficulties varied across cases, and the study does not establish their relative importance. These difficulties tie in with the administrative barriers described by Willman [29] in Tampere, with the ambivalence—both protective and disciplinary—surrounding Geneva’s municipal allotments [30], and with the short-term social policy nature, rather than long-term sustainable planning, that Anthopoulou et al. [45] attribute to the Greek community gardens that emerged during the crisis. Palau-Salvador et al. [42] emphasise, in the same vein, that public policies are the decisive factor in consolidating these processes and that public support is particularly necessary where land tenure or water supply is insecure. The territorial protection of the Valencian horta through the Territorial Action Plan for the Horta [32], described by stakeholders as an umbrella regulation lacking its own system of penalties, illustrates the regulatory limitations described by participants.

Participants described disagreements over the suitability of some managing organisations. Maćkiewicz and Jeziorek [26] likewise document changes in community involvement and educational activities associated with institutional arrangements in two Polish gardens. Drake [25] offers a complementary interpretative framework in which the promoters’ intentions regarding the community do not automatically translate into the expected outcomes, but rather depend on how users’ expectations are recognised and integrated. This mismatch between design and practice runs through several of the Valencian cases, ranging from the rejection of the merit-based allocation system in the municipal allotment with a management committee to the failure to implement the biannual lottery in the allotment governed by regulations, and points to the need to address the power dynamics that Zerbian et al. [46] identify as key determinants in the city of Valencia’s food governance instruments.

Regarding RQ3, participants described social interaction, intergenerational activities, informal learning and participation by organisations serving particular groups. Related sociocultural functions [12,13,18] and perceived health benefits [16] have been examined elsewhere. Hasson’s comparison of London allotments and community gardens [47] provides context for differences in objectives and participation, without establishing equivalent differences in the present sample. The accounts from Valencia indicate that cultivation and social activities coexist in different organisational arrangements.

Viewed within a broader context, these findings reinforce the argument that urban agriculture should not be assessed solely on the basis of its quantitative contribution to food security—which is limited by the scarcity of urban land [15]—but rather on the basis of its multifunctional nature. Participants’ accounts can be considered alongside research on health [1,8,14], sociocultural functions [12,18], urban resilience [9,48] and climate adaptation [49,50]. These studies provide context for potential contributions; the present study did not measure health changes, environmental impacts or urban resilience. The Mediterranean specificity of this case, characterised by the tradition of peri-urban vegetable gardens and the expansion of community gardens following the financial crisis [42], helps to explain the relative importance of municipal and plot-level configurations, in which land-use planning and heritage protection act as structuring factors.

From a public policy perspective, the results suggest that urban food strategies in the Valencia Functional Urban Area, and in comparable Mediterranean contexts, would benefit from a portfolio approach that recognises the complementarity of municipal, community-led and hybrid models, rather than favouring a single template [33,42]. Such an approach could address tenure conditions, access procedures and operational support across different governance models, according to local needs and organisational capacity. The proposals emerging from the stakeholders themselves point in this direction: setting aside specific land in general development plans without taking it away from existing green spaces; providing legal safeguards for local by-laws against changes in government; opening up short supply chains currently blocked by land-use classifications; and creating a network of urban farming initiatives to share knowledge and exchange resources between municipalities. In this regard, the role of the “gardeniser” described by Pino Cabeza [43]—a professional figure straddling the roles of agricultural technician and community facilitator, combining horticultural knowledge with skills in mediation, organisational management and liaison with the administration—and the role of municipal food councils analysed by Zerbian et al. [46] offer concrete tools for organising this collaboration. The former would provide operational continuity and relieve volunteers of tasks that currently jeopardise their sustainability, whilst the latter would provide a stable framework for cross-sectoral deliberation between the administration, civil society and the productive sector. These instruments also highlight the importance of communication strategies. The results show that communication between the administration and the gardens is often reactive and dependent on individual contact persons, so that the loss of a key municipal officer can leave an association without a link to the administration, and that poor communication between the driving force of a project and the associations it represents is a source of conflict. Sereenonchai and Arunrat [51] argue that effective communication in urban agriculture relies on credible sources, community-led initiatives, feedback loops and formal collaboration with local governments and schools. In the Valencian cases, these elements are partly present, through the horizontal transfer of knowledge among gardeners, the management committees that relay information between users and the administration, and the school visits organised in several allotments, but they rarely form part of a planned strategy. Stable communication channels between municipal technicians and managing bodies, together with the inter-municipal network proposed by the stakeholders, would help to reduce this dependence on individuals.

Taken together, these findings make three contributions to the literature on urban garden governance. First, at the conceptual level, they show that the position of a garden on the continuum between self-management and institutional management is situational rather than structural, and they identify four factors on which it depends: the presence of a stable technical point of contact, the continuity of funding, political sensitivity and the robustness of the regulatory framework. This builds on work showing that governance structures change across the stages of a project [21] and on the notion of hybrid governance [23], by specifying the conditions under which these arrangements are sustained or eroded. Second, at the empirical level, the study documents institutional and organisational difficulties in a Mediterranean metropolitan context where urban allotments coexist with a peri-urban agricultural tradition [42]. Third, at the applied level, they support a portfolio approach to public policy that recognises the complementarity of municipal, community-led and hybrid models rather than promoting a single template.

This study has several limitations. Its qualitative and contextual nature limits the statistical generalisation of the findings, although it allows for an in-depth interpretative understanding of governance practices [36,52]. The focus on the Valencia Functional Urban Area, whilst useful for comparative analysis, calls for caution when extrapolating the results to other urban contexts with different institutional and cultural trajectories. A second specific limitation stemmed from the practical impossibility of incorporating into the analysis certain urban allotments initially envisaged in the research design. The most significant case is that of a community allotment in the metropolitan area, managed by an association and dedicated to organic farming, agro-environmental education and the revival of traditional market gardening; the study of this site could not be completed because the facilities were destroyed during the severe floods (DANA) that affected Valencia in October 2024. In addition, three cases (Sociópolis, Alaquàs and Godella) are represented only by institutional actors, so their accounts may reflect an institutional-response bias; this was partly mitigated by triangulation with the perspectives of associations, users and external experts in the other cases, but future research should incorporate the voices of users in these gardens. Furthermore, the research primarily reflects the perspective of the actors directly involved; therefore, future studies should incorporate the voices of non-participating residents and private landowners. It would also be valuable to adopt longitudinal designs capable of tracking how governance arrangements evolve over time and across the different phases of the garden’s development [21,26,46], as well as combining the qualitative approach with quantitative indicators of ecological, health and equity outcomes, including the measurement of ecosystem services such as carbon sequestration [12,50]. These approaches would help to consolidate the empirical basis needed to guide fairer, more resilient and sustainable urban allotment policies.

5. Conclusions

This study set out to analyse and compare the governance of urban allotments in the Valencia Functional Urban Area and to explore its social dimensions. Regarding RQ1 and RQ2, the results support, with some nuances, the initial hypothesis. Urban allotments in the Valencia Functional Urban Area do not follow a single model, but are distributed along a continuum between community self-management and institutional management. However, rather than discrete and stable types, what is observed are unstable balances between participation, regulation and sustainability. The position of each allotment on this continuum depends less on a pre-determined design than on the day-to-day interaction between the community and the administration, and is continually renegotiated according to factors such as the presence of a technical point of contact, the continuity of funding, political sensitivity and the robustness of the regulatory framework. The hybrid configuration, in which the administration sets the framework (by-laws, land allocation and supervision) whilst an association or committee manages day-to-day operations, is the most widespread in practice, even though only one case is formally hybrid; this undermines the analytical value of the dichotomy between the fully community-run allotment and the fully administration-run one.

Participants identified institutional resources, tenure conditions and administrative procedures as relevant to continuity. They also described organisational difficulties, including concentrated workloads and uncertain volunteer succession. These findings indicate that both institutional arrangements and organisational capacity require consideration across governance models.

Regarding RQ3, participants described social interaction, informal learning and perceived well-being benefits alongside cultivation. The findings suggest considering different governance arrangements according to local objectives, resources and access conditions. They do not establish that one arrangement produces better social outcomes than another.