Upscaling strategies for food forestry and the place-based conditions
shaping their success: insights from the Netherlands
W.B.A. (Wytze) Walstra University of Amsterdam Student nr. 12309079
Supervised by J. Willems, K.M. Wijsman & C.F. Rammelt
28 May 2021 Haarlem
Upscaling strategies for food forestry and the place-based conditions
shaping their success: insights from the Netherlands
W.B.A. Walstra a ✉
a BSc student Future Planet Studies, University of Amsterdam
Abstract
In response to multiple sustainability crises unfolding in the Dutch agricultural sector, food forestry has emerged as an alternative agricultural system. Early research results indicate that food forests, productive ecosystems designed by humans in imitation of a natural forest, are capable of restoring biodiversity, mitigating soil degradation and capturing significant amounts of carbon. However, the intended food forest area of 1,000 hectares in 2030, which has been adopted by the Dutch national government in 2017, represents less than 0.1 percent of the total agricultural land area in the Netherlands. For such an alternative agricultural practice to play a significant role in the transition towards sustainable agriculture, further upscaling is necessary. The present study identifies strategies for upscaling food forestry in the Netherlands beyond 1,000 ha and the place-based conditions shaping their success. It revolves around two central themes stemming from studies on ‘the geography of sustainability transitions’, place and scale, which were found to affect transition dynamics. Across four domains (ecological, socio-cultural, policy, socio-demographic) emerging from eighteen identified upscaling strategies, several place-specificities were found relevant to food forestry. Seven tiers of governance involved in upscaling food forestry were identified and it is demonstrated how these tiers interact in three distinctive ways: complementation, obstruction and ignorance. The study shows how stakeholders can embed upscaling strategies in spatial contexts in order to accelerate upscaling. In doing so, the study contributes to 1) theory on the geography of sustainability transitions and 2) the emerging research agenda on food forestry.
Contents
1. Introduction 2
2. Theoretical & conceptual framework 4
2.1 The geography of sustainability transitions 4
2.2 Strategies for upscaling 5
2.3 Food forestry 5 3. Methodology 6 4. Results 8 4.1 Ecological domain 8 4.2 Socio-cultural domain 9 4.3 Policy domain 10 4.4 Socio-demographic domain 10
5. Analysis & discussion 11
5.1 Dynamics of place 11
Highlights
• Food forestry was studied as a radical innovation in the Dutch agri-food sector to explore the spatiality of upscaling strategies in agriculture.
• Across four domains, place specificities were found affecting transition dynamics.
• Results nuance findings of earlier research showing fixed actor constellation in the agricultural sector, as governments may guide land use change by involving in land acquisition.
• Which upscaling strategies fit a place best is context-dependent.
• Seven tiers of governance involved in upscaling in food forestry and their interactions are identified.
5.2 Dynamics of scale 12 5.3 Limitations & recommendations for further research 13
6. Conclusion 14
Acknowledgements 14
References 15
1. Introduction
Around the world, the Netherlands is known as the “tiny country feed[ing] the world” (Viviano, 2017, p. 1). According to QS World University Rankings, the Netherlands is home to the No. 1 agricultural & forestry university on the globe (QS Quacquarelli Symonds, n.d.). It is also the second agricultural exporter worldwide (Statistics Netherlands, 2019). The Dutch agricultural sector could be seen as the ultimate plea for agriculture intensification. A lesser-known perspective is that the Netherlands is in the midst of an accelerating, multifaceted agricultural crisis (MAC) that first became apparent in the 1970s (Grin, 2010). Figure 1 provides an overview of these facets, including alarming biodiversity loss, an ongoing crisis over nitrogen deposition that puts at risk €14 billion worth of construction projects (Stokstad, 2019) and a rapidly decreasing farming population. Until recently, the incumbent agricultural regime has been able to maintain itself by posing partial solutions. However, the past years may have seen a signal for a coming turning point (Zwartkruis et al., 2020).
Food forests, productive ecosystems designed by humans in imitation of a natural forest (Green Deal Voedselbossen, 2018), have been proposed as an integral solution to M in the Netherlands. As opposed to partial solutions, integral solutions “[seek] to address various challenges and aspects with the same solution” (Elzen et al., 2012, p. 435-436). A food forest consists of a high diversity of perennial and woody species, of which parts serve as food for humans (Green Deal Voedselbossen, 2018). Rather than proposing a new agricultural system, food forestry makes the promise of reuniting two worlds that are currently separated: agriculture and nature (van Dinther, 2019). Although little peer-reviewed research has been conducted on food forestry, there are strong indications that food forests at least address all the facets of the MAC to some extent. Breidenbach et al. (2017, p. 93) investigated the biodiversity of an 8-year-old food forest and compared the data to a nearby Natura 2000 forest, for instance, and found that “the number of species is remarkably similar”. Another study by Rebisz (2019, p. 69) states that “food forestry […] may mitigate soil threats such as SOM [soil organic matter] decline, [soil] compaction and biodiversity loss”. Early results of the National Monitoring Program Food Forests (Nationaal Monitoringsprogramma Voedselbossen) suggest that a 20-year-old food forest could capture some 40 tons of carbon per hectare into above-ground biomass (Green Deal Voedselbossen, 2021),
• ▼ 25 percentage point between 1900-2010
Biodiversity 15% Mean Species Abundance (2010) Van Veen et al.(2008)
• ▼ 85% between 1950-2014
Social security 52.710 Number of farms (2020) Statistics Netherlands (2014); for farmers Statistics Netherlands (2020)
• ▲ ? (tempo uncertain)
Agricultural droughts ? Philip et al.(2020)
• ! 2 major outbreaks this century, only two in Europe
Zoonoses 2 Major outbreaks (2021) Abbott (2003); Cross et al.(2019)
• ▼10 percentage point between 1995-2018
Nitrogen 70% Share of natural area for which the Environmental Data Compendium critical nitrogen load is exceeded (2020)
• ▲ ? (tempo uncertain)
Soil compaction 50% Share of agricultural lands Van den Akker & Hoogland (2011) with overcompacted subsoils (2011)
which would be over two-thirds of the above-ground carbon sequestration capacity (CSC) of a mature temperate forest and a multitude of the above-ground CSC of temperate grasslands and croplands (IPCC, 2000).
Unlike most conventional agricultural systems, food forests only start to get significantly productive after approximately five years (Boulestreau & van Eck, 2016). By nature, food forests thus take time to prove themselves. At present, more than 90 hectares have been converted to food forestry in the Netherlands (Figure 2). The Ministry of Agriculture, Nature and Food Quality has adopted the goal of increasing that area to 1,000 ha ‘in the long-term’ (Ministry of Agriculture, Nature and Food Quality, 2020, p. 36). Considering this represents less than 0.1 percent of total agricultural land in the Netherlands, there is a need to accelerate the upscaling of food forestry for it to play a significant role in the transition towards sustainable agriculture. A relevant question then becomes what strategies actors with an interest in the upscaling of food forestry can apply in this endeavour. Moreover, despite ‘governance’ is explicitly mentioned as a research topic of interest in the Green Deal Food Forests (Green Deal Voedselbossen, 2018, p. 4), most ongoing academic research into food forestry is centred on ecological issues.
There is a rich literature on the governance of (sustainability) transitions in the Netherlands, and Dutch agriculture has been the subject of intensive study (El Bilali, 2019). However, several scholars have emphasised the need for a socio-spatial perspective on transitions and the current lack thereof (e.g. Smith et al., 2010; Coenen et al., 2012; Raven et al., 2012; Hansen & Coenen, 2015; Murphy, 2015). From the theoretical niche that developed from this discussion, ‘the geography of transitions’, two main themes are extracted with implications for upscaling mechanisms. First, it is expected that the suitability of upscaling strategies varies between different places. Second, governance issues in upscaling food forestry are presumed to relate to dynamics of scale. Vermunt et al. (2020) suggest that agroforestry systems, of which food forestry is a sub-category (see section 2.3), are a suitable object of analysis for the study of geographical dependencies in transition studies because of its presumed particular financial and ecological benefits.
To these ends, this study aims to enrich the knowledge on the role of place in sustainability transitions and to contribute to the emerging research agenda on food forestry. It identifies strategies for scaling up food forestry in the Netherlands beyond 1,000 hectares as well as the place-based conditions shaping the success of these strategies. For that purpose, ten top-level stakeholders from a variety of fields were consulted through focus group and in-depth interviews. Furthermore, document analysis was performed, with government documents as the main focal point. By presenting a comprehensive overview of upscaling strategies, the study contributes to the narrow literature on governance issues related to upscaling food forestry. The results deepen and broaden the understanding of the dynamics of place in transitions. Finally, it offers insights into the dynamics of scale in (agricultural) governance.
The study proceeds with outlining the theoretical and conceptual framework. Next, the methodology is described. The results are presented in section 3 and discussed in section 4. Lastly, section 5 concludes the study.
2. Theoretical & conceptual framework 2.1 The geography of sustainability transitions
As previously mentioned, several scholars have stressed the importance of spatial considerations in (sustainability) transition studies. In studies on the geography of sustainability transitions, most scholars distinguish between dynamics of scale and dynamics of place.
(1) Dynamics of place
First, as Coenen et al. (2012) note, place has theoretical implications for transition analyses:
“[T]ransition analyses have overlooked where transitions take place, and the socio-spatial relations and dynamics within which transitions evolve. This has the undesirable effect of neglecting the spatial contexts in which individual case analyses of sustainability transitions are situated thus reducing comparability between places which in turn limits particular transition analyses coalescing into a coherent body of theory.” (p. 969).
Place also has important practical implications for transition research. Institutional structures are highly place-dependent, as they are shaped by place-specific historical, economic and cultural contexts (Coenen et al., 2012). Across space, regime actors respond in different ways to exogenous developments, such as the MAC. Coenen et al. (2012) warn that neglecting these geographical considerations could result in “oversimplified conclusions”, wrongfully assuming generalisability of isolated success factors and barriers. In similar vein, from a perspective on the ecological embeddedness of agriculture, Vermunt et al. (2020) argue that:
“The importance of geographic embeddedness addresses the need for a fit between novelty and place due specific local social-ecological conditions. Models for change cannot easily be scaled, replicated or standardized. There is no ‘one size fits all’, implying that in every region, new approaches need to be developed with new stakeholders, new experiments and new institutions.” (p. 244)
Moreover, as Vermunt et al. argue, the role of place in agriculture deserves particular attention as farmers are confined to their land and, at the same time, the only actors with the ability to make direct changes to the agricultural landscape. The breakthrough of novelties in most sectors involves the ‘come and go’ of regime actors, also known as the technological substitution pathway (Geels et al., 2016). In contrast, Vermunt et al. (2020) found that the actor ‘constellation’ in agricultural systems is rather fixed and highly fragmented. Correspondingly, transitions in agricultural systems are thought to depend on endogenous change, whilst transition theory has primarily focused on exogenous change. According to Vermunt and colleagues, social-ecological dependencies are key in understanding the spatial stickiness of success factors for upscaling in agriculture. Their comparative case study in the Dutch dairy sector demonstrates that each place has unique ecological circumstances, resulting in particular ecological challenges and solutions.
Swagemakers et al. (2019) go even further and argues that place should be the very starting point of change in agriculture. They show how restorative farming practices, through a place-based approach, can be linked to place-embedded policies and markets, or a combination of both. Healey et
al. (2003, p. 86) support this approach: they suggest that “developing concepts of place-identity” could
increase support among stakeholders and invoke more inclusive governance. (2) Dynamics of scale
‘Scale’ is a broad concept. Across disciplines, it may be interpreted in different ways (Raven et al., 2012). Multi-scalarity is in fact at the core of transition theory, “but these are not geographical scales” (Coenen et al., 2012, p. 972). Thus far, physical scales have been largely neglected in transition studies: a relational view of scale has been preferred in order to gain a better understanding of interaction between actors (Raven et al., 2012). Smith et al. (2010, p. 443) notice that “the tendency has been to
audiences”. However, this has the disadvantage of losing sight of differences between scales (Coenen et al., 2012). In this study, a multi-scalar, geographical understanding of scale is therefore applied
(henceforward also simply referred to as ‘scale’). At different scales, actors strategise to accelerate or direct sustainability transitions (Smith et al., 2010) and, consequently, they do so in distinctive manners (Gupta et al., 2007).
As such, this study departs from the assumption that entry points and leverages for radical innovations vary between places, but also across multiple scales. It explores the possibility of embedding upscaling strategies in spatial contexts by examining strategies for scaling up food forestry in the Netherlands. 2.2 Strategies for upscaling
Following van Doren et al. (2018, p. 228), upscaling is here understood as “an increase in uptake,
growth, or replication” of a certain phenomenon, in this case, food forests. The notion of ‘strategy’ is
defined by van Assche et al. (2020, p. 696) as a “vision for a desirable longer-term future, coupled to
an idea of how to get there”. A strategy is neither a dream, which may be completely decoupled from
feasibility, nor a tactic, which focuses on short-term action (van Assche et al., 2020). The study depicts that a strategy works towards a desirable situation, within the boundaries of a given context. In addition, a strategy is defined as an intentional action aimed at addressing barriers or exploiting opportunities. 2.3 Food forestry
At present, food forestry (voedselbosbouw in Dutch) is a poorly established notion in academic literature. It is unclear where the term food forestry precisely stems from, but in recent years it has gained significant traction in the Netherlands. Food forestry is a form of agroforestry, the integration of trees and other woody perennial species in agricultural systems (Nair, 1985). However, food forestry is not the same as agroforestry, which also includes agricultural systems with a mix between woody perennial species and annual crops or cattle (e.g. alley cropping and silvopasture, see Figure 3). Because food forests are designed in imitation of a natural ecosystem – a young forest ecosystem – they are a form of agroecology. Hathaway (2015) defines agroecology as “an alternative paradigm of production
based on ecological principles” (p. 239). Still, agroecological systems do not always (exclusively)
cultivate woody and/or perennial species. The British notion of forest gardening is almost identical to food forestry, except it puts more emphasis on the cultivation of non-food produce (Crawford, 2010). The main reason why the concept of food forestry is adopted in this study is that it is embraced by relevant actors in the Netherlands. More exact, the study follows the definition of food forestry that was established in the Green Deal Food Forests (Green Deal Voedselbossen) in 2017, which consists of a set of intentions and arrangements as regards the upscaling of food forestry agreed upon by policy makers, governments, front-runners and researchers (see: Green Deal Voedselbossen, 2018). This definition is also attached to the Dutch crop code (1940) for food forests:
“ a specific ecosystem is qualified as a food forest on the basis of the following characteristics: - a productive ecosystem designed by humans in imitation of a natural forest, with a high diversity of perennial/woody species, of which parts (fruits, seeds, leaves, stems, etc.) serve as food for humans;
- presence of a canopy layer of higher trees;
- presence of at least three of the other niche vegetation layers of resp. smaller trees, shrubs, herbs, herbaceous perennials, root crops and climbers;
- presence of a healthy forest soil life;
- a robust size; i.e. a surface area of at least 0.5 hectares in an environment of high ecological richness; in severely depleted environments, a minimal surface area of 20 hectares is required.”
3. Methodology
This study used a theoretical lens on the place-embeddedness of transitions in agriculture to examine the spatiality of upscaling strategies. Food forestry was studied as a radical innovation in the Dutch agri-food sector. Challenging to studying such alternative agricultural practices is the fact that their development mostly takes place in protected environments (niches). For a real breakthrough to occur, however, the creation of new links with the regime is decisive (Elzen et al., 2012). In order to explore such linkage, this study, therefore, places emphasis on the long-term development of food forestry using the concept of ‘strategy’ (also see section 2.2). Hence, a mixed methods approach was adopted to incorporate both historical and future perspectives on strategising.
As a point of departure, a focus group was employed to explore expert knowledge on the upscaling of food forestry. The application of a focus group discussion is deemed valuable as it allows for collecting input from several experts efficiently within the practical constraints of this study. Besides, a focus group discussion allows for interaction between respondents and therefore provides them with the opportunity to challenge each other’s input. Taking into consideration the high level of involvement and knowledge of the participants in/on food forestry, the focus group followed a basic, loosely structured script, in accordance with methodological literature on focus groups (Bryman, 2016). Eight experts participated in the focus group, equal to the maximum focus group size proposed by Barbour (2007). The participants were enabled to present their view on upscaling strategies for food forestry after they had been briefed on the previously defined conceptual framework (see section 2.2 and 2.3). Table 1 shows an overview of the participants, who consented to the publication of their full names. In recruiting participants, three pursuits were considered (in order of decisiveness): i) selecting participants that are actively involved in the field of food forestry, at least in part through professional occupation, ii) selecting participants from a variety of disciplines and iii) creating a diverse group of respondents in terms of age and sex. This latter pursuit was failed to be met, as several contacted female stakeholders were unavailable at the date set for the focus group discussion. Due to Covid-19 restrictions, the focus group took place on a videoconferencing platform. The session was recorded and automatically transcribed through Word Online. The auto-generated transcript was manually edited and transformed into a non-verbatim transcript (see: Appendices).
Using only one focus group is believed to be unorthodox (Bryman, 2016), yet considered appropriate here for the reason that the network of experts on food forestry is highly interconnected and fairly small. To further achieve data richness and validity, document analysis was carried out and two
No. Full name Occupation Organization Sex
1 Jeroen Kruit Researcher WUR M
2 Kees van Veluw Researcher WUR; Louis Bolk Institute M
3 Isabelle van der Zanden Researcher NIOO-KNAW F
4 Stijn Heijs Board member, food forest practitioner Stichting Voedselbosbouw M
5 Marc Buiter Board member Stichting Voedselbosbouw M
6 Albert de Graaf Policy officer Province of Flevoland M
7 Ceriel Lucker Policy officer Ministry of Agriculture, Nature and Food Quality M
8 Michiel Philippart Lobbyist Stichting Voedselbosbouw M
Fig. 3. Various land management systems arranged by land use type (adapted from Green Deal Voedselbossen [2020]).
Abbreviations
MA = Ministry of Agriculture, Nature and Food Quality MIW = Ministry of Infrastructure and Water Management MIK = Ministry of the Interior and Kingdom Relation ME = Ministry of Economic Affairs and Climate Policy HoR = House of Representatives
in-depth interviews were conducted. The document analysis mainly focussed on government documents. The central database for Dutch government documents, Overheid.nl, was searched for documents containing the word ‘voedselbos’ (the Dutch word for ‘food forest’). This yielded 163 results (the database was accessed on 20 April 2021). All results were skimmed, after which 13 documents were considered relevant (Table 2). Relevance of documents was determined in line with the theoretical and conceptual framework, i.e. whether a document regarded a ‘strategy’ (not a tactic) and was concerned with food forestry as defined in section 2.3. In practice, this implied the exclusion of dozens of modifications in land-use plans from the analysis.
No. Title Document type Year Source/
Commissioned by Reference code 1 Vragen gesteld door de leden der
Kamer, met de daarop door de regering gegeven antwoorden
Answers to parliamentary questions
2020 HoR
ah-tk-20192020-3516 2 Nationale Bijenstrategie. Bed &
Breakfast for Bees
Strategic document 2018 MA blg-833946
3 Deltaprogramma 2020. Doorwerken aan de delta: nuchter, alert en voorbereid
Strategic document 2020 MIW; MA; MIK blg-901133
4 Het Europese Klimaatplan. Aandachtpunten voor de afstemming tussen Europees en nationaal Klimaatbeleid
Policy brief 2020 Netherlands
Enviromental Assessment Agency
blg-954639
5 Ruimte voor Biobased Bouwen. Strategische Verkenning
Strategic documents 2020 MIK; MA; The Province of Noord-Brabant; The Province of Zuid-Holland; Board of Government Advisors
blg-960349
6 Jong Leren Eten (terugblik). Voor een gezonde en duurzame generatie
Project report 2021 MA blg-974587
7 Green Deal Voedselbossen Green Deal 2017 ME;MIW; MA C-219
8 Jaarverslag en slotwet Minsterie van Economische Zaken en
Diergezondheidsfonds 2016
Annual report 2017 HoR kst-34725-XIII-1
9 EU-voorstellen: Gemeenschappelijk Landbouwbeleid COM (2018) 392, 393 en 394 Answers to parliamentary questions 2018 HoR kst-34965-3
10 Initiatiefnota van de leden Bromet en Futselaar over Nationaal Bomenplan
Report of written consultation
2020 HoR kst-35309-3
11 EU-voorstel: Mededeling van de Commissie aan het Europees parlement, de Raad, het Europees Economisch en Sociaal Comité en het Comité van de Regio’s inzake een ambitieuzere klimaatdoestelling voor Europa voor 2030: investeren in een klimaatneutrale toekomst voor ons allemaal
Letter to parliament 2021 Senate kst-35652-c
12 Natuurbeleid. Verslag van een schriftelijk overleg
Report of written consultation
2021 HoR kst-974564
13 Bos voor de toekomst. Uitwerking aambities en doelen landelijke Bossenstrategie en beleidsagenda 2030
Strategic document 2020 Association of Provincial Authorities (IPO); MA
kst-33576 Table 2. Documents studied in the document analysis.
Using ATLAS.ti 9 software, some 50 initial codes were developed based on the focus group transcript and documents through open and axial coding (Corbin & Strauss, 1990). Building on these early results, two in-depth interviews were conducted with an all-round food forest expert, Wouter van Eck, and the managing director of a major nature development fund of the province of Brabant (Groen
Ontwikkelingsfonds Brabant), Mary Fiers. Preliminary desk research provided additional input for the
interviews. This time, the interviews followed a detailed script focussed on validating early results, yet space for follow-up questions was left. The in-depth interviews were transcribed and connected to the initial codes (see: Appendices). From the coded transcripts and documents, upscaling strategies were extracted. In a similar manner to what Corbin and Strauss (1990) describe as selective coding, core categories were sought for in the upscaling strategies. Whereas the initial codes were obtained strongly inductively, this final step was carried out in an abductive way: the final classification presented in the results (section 4) builds on both the identified upscaling strategies and earlier classifications from van Doren et al. (2016) and van Doren et al. (2018), and Vermunt and colleagues’ (2020) findings on socio-ecological place dynamics in agriculture.
Reported findings in this study were based on consensus between at least two distinctive sources (i.e. two or more interviewees in the focus group and/or in-depth interviews, multiple documents studied in document analysis, or any combination of these). Notions that emerged from the focus group, yet were not sufficiently corroborated in the focus group discussion or through document analysis, were taken as a subject matter in the in-depth interviews. This was done to exclude any individual subjectivities from the primary results. Quotes of respondents/documents given in the following sections were translated to English.
4. Results
This section elaborates on the results through the agency of upscaling strategies identified in the focus group, in-depth interviews and studied documents. Table 3 provides an overview of these upscaling strategies. Findings are discussed along the lines of a fourfold classification of domains that emerged from the results: ecological, socio-cultural, policy and socio-demographic.
4.1 Ecological domain
The respondents stress the importance of linking food forestry to current ecological issues, most notably climate change, nitrogen pollution and water management, and related policy aims. Within the food forestry community, a much-debated question is whether food forests are economically viable without secondary activities. There is broad consensus, however, on the potential of deploying non-economic benefits of food forestry as entry points for upscaling. In light of the ongoing nitrogen crisis, a concrete suggestion is to plant food forests near Natura 2000 sites for their nitrogen-fixing ability. A motion calling to further explore this strategy has recently been adopted by national parliament (Tweede Kamer Der Staten-Generaal, 2021). Several respondents highlighted that the development of financing mechanisms for ecosystem services could drastically change the competitive position of food forests. Various indications were found for developments in this area, like this statement made in a letter to parliament by Vice-President of the European Commission Frans Timmermans:
“Moreover, initiatives in the area of food forests could also fall under other instruments that are developed by the Commission. More specifically, this relates to initiatives […] which should reward farmers and foresters for efforts in the area of carbon sequestration, including for example agroforestry”. (Document 11, 2021).
Nevertheless, at present, mandatory markets for ecosystems services are still in their infancy. In the near future, voluntary programs could play a significant role in the upscaling of food forestry. Some water authorities have started co-financing food forestry projects because of their benefits for local water management. Staatsbosbeheer, the Dutch government organization for forestry and nature management, has made available resources – most notably land holdings – to stimulate nature-based agriculture alike. The organization has signed Green Deal Food Forests and has been involved in the development of two food forests to this point.
Abbreviations
NG = national government EU = European Union RI = research institutes FF = food forest community Pr = Provinces
4.2 Socio-cultural domain
It is noted by respondents that aligning the development of food forestry with socio-cultural developments could support upscaling. One focus group participant “noticed that in the southern part
[of Limburg], due to [a history of] traditional fruit orchard farming, the threshold [for implementing food forestry] is many times lower, both with regard to land-use plans [‘bestemmingsplannen’] and the attitude of farmers towards food production through cultivation of woody species” (Interview 1, 2021).
Several respondents believe that, in parts of society, dietary changes could complement the development of food forestry, as food forests diversify domestic food production and are thought to produce crops with high nutritional value. Though the latter notion was a topic of discussion in the focus group: “for me, nutritional value is not the first argument that comes to mind. […] So, food is
emotion. Nutritional value is only of interest to a small group of ‘randstedelingen’ [inhabitants of the Randstad metropolitan area]” (Interview 1, 2021). This is in line with an overarching theme that
reoccurred in the focus group: several respondents drew attention to the need to foster a mind-shift that transcends rational arguments for food forestry. They feared that over-rationalising food forestry could harm its repute. “I think we will miss the boat if we do this. Politicians, multinationals, or I don’t know
who… We need to awe them in another way” (Interview 1, 2021). When asked how to concretise this,
No. Strategy Key actor(s) Applied/
Suggested 1 Eliminate regulations revolving around conservation of the ‘open cultural landscape’. Pr Applied 2 Develop (mandatory) financing mechanisms for ecosystem services. EU; NG; RI Suggested
3 Link food forestry to forest strategies (bossenstrategieën). NG; Pr Applied
4 Intensify (fundamental) research into food forestry. FF; RI Applied
5 Develop educational programmes on professional exploitation of food forests (i.e. in higher education).
RI Suggested
6 Develop educational programmes on agroecology in lower education. NG Applied
7 Employ food forestry for water management. Water authorities
(waterschappen)
Applied
8 Capitalise on place-specific agrarian histories. FF Applied
9 Shift food forestry from nature schemes to agricultural schemes. (In the long-term governments should pursue a more interdisciplinary approach to policy).
EU; NG Applied
10 Take over agricultural land of retiring farmers and connect the conditions for resale or tenancy to social policy aims.
Pr Applied
11 Include food forestry in regional development, especially in shrinking rural regions. NG; Pr 12 Employ food forestry for the restoration of nature reserves degraded by excess
nitrogen deposition.
NG; Pr Suggested
13 Connect food forests to regional cuisines. FF Applied
14 Advance the mind-shift to a new food system to transcend rational arguments for food forestry.
FF Suggested
15 Change subsidy flows in favour of food forestry (and agroforestry in general). EU; NG Applied
16 Stimulate dietary changes. FF; NG Suggested
17 Relax regulations on obligatory reforestation. NG; Pr Applied
18 Initiate food forestry projects on lands possessed by Staatsbosbeheer. Staatsbosbeheer (NG); FF
Applied Table 3. Strategies for scaling up food forestry in the Netherlands as derived through focus group interviewing, in-depth
interviewing and document analysis.
No. Strategy Key actor(s) Applied/
Suggested 1 Eliminate regulations revolving around conservation of the ‘open cultural landscape’. Pr Applied
2 Develop financing mechanisms for ecosystem services. EU; NG; RI Suggested
3 Link food forestry to forest strategies (bossenstrategieën). NG; Pr Applied
4 Intensify (fundamental) research into food forestry. FF; RI Applied
5 Develop educational programmes on professional exploitation of food forests (i.a. in higher education).
RI; NG Suggested
6 Develop educational programmes on agroecology in lower education. NG Applied
7 Employ food forestry for water management. Water authorities
(waterschappen)
Applied
8 Capitalize on place-specific agrarian histories. FF; Applied
9 Shift food forestry from nature schemes to agricultural schemes. (On the long-term governments should pursue a more interdisciplinary approach to policy).
10 Take over agricultural land of retiring farmers and connect the conditions for resale or tenancy to social policy aims.
Pr Applied
11 Include food forestry in regional development, especially in shrinking rural regions. NG; Pr 12 Employ food forestry for the restoration of nature reserves degraded by excess
nitrogen deposition.
NG; Pr Suggested
13 Connect food forests to regional cuisines. FF Applied
14 Facilitate the mind-shift to a new food system to transcend rational arguments for food forestry.
FF Suggested
15 Change subsidy flows in favour of food forestry (and agroforestry). EU; NG Applied
16 Stimulate dietary changes. FF; NG
17 Relax regulations on obligatory reforestation. 18 Staatsbosbeheer?
Table 2. Strategies for scaling up food forestry in the Netherlands as derived through focus group interviewing, in-depth interviewing and document analysis.
respondents point out that institutions in lower education can contribute to a mind-shift by adopting (aspects of) ecoliteracy in their curricula.
4.3 Policy domain
The case shows that capitalising on developments in the policy domain can support the institutionalisation of food forestry. The national bossenstrategie (forest strategy), which was established in the context of the Dutch Climate Agreement, has been an entry point for food forestry and resulted in the adoption of the 1,000-ha aim. It was found that some provinces are working on provincial forest strategies, in addition to the national forest strategy. Respondents note that these provincial strategies could contribute to further institutionalisation of food forestry.
Spatial policy appears to be a key barrier to the adoption of food forestry. More specifically, legislation and regulations revolving around obligatory reforestation (herplantingplicht), the protection of farmland birds and landscape openness can obstruct the establishment of a food forest. Municipalities determine restrictions to land uses, which are documented in land-use plans (bestemmingsplannen). However, provinces, which are united in the Association of Provincial Authorities (IPO), and to a lesser extent the national government, may play “an intermediary role” by promoting “invitational planning” (Interview 3, 2021). They could use instruments like vision documents (omgevingsvisies) and flexibility clauses in legislation and regulation to steer land use, or ultimately even overrule municipal land-use plans.
Formally, all provinces committed to these efforts by signing the national forest strategy. However, the extent to which provinces put these commitments into practice is context-dependent. To illustrate, several respondents mentioned that the continuation of stringent meadow bird conservation measures in the Province of Friesland partly explains the limited presence of food forestry in this region. Four provinces have underlined their pledge made in the national forest strategy by signing the Green Deal Food Forests. The Province of Brabant, which is considered a front-runner by respondents, deserves particular mention. In 2014, the Province of Brabant established a fund for the development of new nature, Groen Ontwikkelingsfonds Brabant. The fund was initiated by an environmental NGO, three nature conservation organisations and the regional farmers’ federation, in response to severe budget cuts in nature policy by the national government. This fund was given significant financial resources and the authority to acquire land and resell or lease it with land-use restrictions in favour of environmental policy aims. Section 4.4 further elaborates on this fund.
The power of subsidies as central instruments in agricultural economics was a theme that reoccurred throughout all research methods. Changing subsidy flows was mentioned as an intervention to improve the competitive position of food forestry in comparison to dominant agricultural systems, to incentivise advanced (experimental) research into food forestry and to enable a transformation towards food forestry systems in general. Subsidy schemes are perceived by respondents as rigid institutions, which are unlikely to be reformed overnight. Still, in regard to the national forest strategy, the Dutch Minister of Agriculture, Nature and Food Quality, wrote the following in a letter to parliament: “at the
moment, I am exploring the incorporation of agriculture-forest combinations, including agroforestry, as an eligible activity in the new National Strategic Plan [Nationaal Strategisch Plan] of the European Agricultural Policy” (Document 12, 2021).
Whether a water authority decides to adopt food forests as an instrument for water management, also depends on developments in their policies (also see section 4.1).
4.4 Socio-demographic domain
Respondents note that food forestry could also be tied to socio-demographic developments. Section 4.3 discusses the practice of land acquisition by the Province of Brabant for purposes of land-use change.
“There are a lot of farmers who would like to quit [farming]. So if land comes up for sale, we [the fund] buy[s] it” (Interview 2, 2021). In the past 20 years, the number of farmers in the Netherlands has almost
halved. Currently, the majority (59%) of farmers aged 55 or above has no successor lined up (Statistics Netherlands, 2021). “From an ideological stance point you can think this is good or not, either way it
is happening” (Interview 2, 2021). The numbers vary spatially, with Brabant and Gelderland being
outliers. In each of these provinces, some 3,000 farmers have no successor. Respondents affirm that this is a window of opportunity for governments to get involved in land acquisition. Furthermore,
respondents highlight the possibility of including food forestry in regional development, especially in areas of rural shrinkage. Food forests are low-maintenance systems, but harvesting in food forestry is time consuming due to its diversity and incompatibility with large harvesting machinery. As such, by some respondents, food forestry is believed to bring forward rural employment opportunities. “And you
could add to that […], that it will also result in many indirect jobs. Especially if you want to sell the products locally and regionally […], it will entail many indirect jobs in regions near food forests”
(Interview 1, 2021). Finally, respondents note that the presence of existing food forests in place also enhances the chance of upscaling. They believe that existing projects can inspire others to involve in food forestry, resulting in ‘hotspots’.
5. Analysis & discussion
This section further builds on the gathered results by discussing transition dynamics found across the four identified domains. In distinguishing between the dynamics of place and dynamics of scale, the main themes derived from the literature as described in the theoretical framework are revisited. This section concludes with limitations of the present study and recommendations for further research. 5.1 Dynamics of place
Across all four domains, place specificities were found affecting transition dynamics. The results therefore support the idea that place specificities determine where ‘fertile soils’ (Sekulova et al., 2017) for the breakthrough of niche innovations emerge. Firstly, some place specificities, such as the willingness of provinces to remove regulatory barriers, are relevant to nearly all food forestry projects. If such a place specificity favours the establishment of food forests over other development paths, it creates a ‘comparative institutional advantage’ (Coenen et al., 2012). Other place specificities could be an entry point for the adoption of food forestry, yet their absence may not rule adoption out. For example, the nitrogen-fixing ability of food forestry could be an entry point to upscale in areas affected by nitrogen deposition, yet there are other reasons food forestry projects might be initiated in these areas. Elsewhere, a forest strategy might be the direct cause to experiment with food forestry. These kinds of place specificities do not result in comparative institutional advantage per se. They rather bring about spatially bound institutional differentiation, on which actors with an interest in the upscaling of food forestry may anticipate. An obvious yet noteworthy place specificity of this kind is the presence of existing food forests. The results indicate that this is an important predictor for food forest emergence, resulting in an ‘oil spill dispersion’ effect. This is in accordance with earlier research conducted into the emergence of ‘grassroot innovations for sustainability’ in the U.S., among which permaculture initiatives, which suspects geographical proximity to existing innovations as an encouraging place specificity for the emergence of novel initiatives (Nicolosi et al., 2018). Lastly, part of the strategies that were suggested by respondents focused on place-based development. These strategies are aimed at linking up with place specificities which by themselves do not necessarily call for the adoption of food forestry. However, adapting to these local characteristics may support the embedding of novelties (Healey et al., 2003; Swagemakers et al., 2019). As such, “place-identity, local knowledge and
relational resources may be mobilised for innovative practices” (Smith et al., 2010, p. 444). One
example of this kind of place specificities from the empirical description includes regional cuisines, to which the selection and processing of crops in food forests could be tailored.
One important finding of the study regards the practice of land acquisition by governments and nuances earlier findings of Vermunt et al. (2020). They argue that “[f]armers are the only actors who
can implement direct agricultural landscape changes by adapting management practices” (Vermunt et al., 2020, p. 245). This characteristic of the agricultural sector is atypical to our main understanding of
transitions, which generally involve substitution of incumbent actors, and is supported by the results. Nevertheless, the idea that actor constellation in agriculture is “rather fixed” (Vermunt et al., 2020, p. 245) needs adjustment, taking into consideration the steady trend of a falling number of farming businesses. Governments have the opportunity to intervene in this process by taking over lands of retiring farmers and reselling or leasing these lands with restrictions. The assumption that change in the agricultural sector must be endogenous, risks that in pursuit of broad consensus, momentum in the transition towards sustainable agriculture is lost. In fact, the previous major transition in the Dutch agricultural sector, the agricultural modernisation, was met with considerable opposition. “It should be
noted that only a minority of farmers, especially young ones, wholeheartedly took part on the modernization programme. Others embarked more hesitantly” (Grin, 2010, p. 252). To cite a
respondent: “This is a new paradigm. We are moving from flat earth to round earth. And those who
don’t get this don’t have to take part, but we need people who do understand it” (Interview 3, 2021).
In all, the study corroborates the idea that “upscaling mechanisms […] should be based on a match between novelty and local social-ecological conditions” (Vermunt et al., 2020, p. 246), and makes the same argument for developments in the socio-cultural, policy and socio-demographic domain.
5.2 Dynamics of scale
Dynamics of scale were apparent in the results. It was found that governance processes relating to the upscaling of food forestry occur in at least seven tiers (Figure 4). The engagement of national government, provinces, water authorities and municipalities has extensively been discussed. In addition, the European Commission governs the agricultural sector through policy frameworks like the Common Agricultural Policy and the European Green Deal. It is also concerned with the creation of markets for ecosystem services. On the global level, arrangements such as the Paris Agreement are established that trickle down to other administrative levels. Finally, initiators of food forests themselves may stimulate change through self-governance (Runhaar et al., 2017): they can strategise through knowledge sharing, the collection and strategic expenditure of private financial support, conducting (civil) research, etc.
These seven tiers may relate to each other in three distinctive manners, for which the following classification is proposed. When governance practices at different levels are aligned, they complement (1) each other (Gupta et al., 2007). For instance, through signing the Green Deal Food Forests, four provinces have complemented national (food) forest targets, documented earlier by the central government in its forest strategy. Governance practices can also oppose (2) each other (Gupta et al., 2007). This can result in barriers to the upscaling of food forestry, yet it may also give rise to opportunities for upscaling, as occurred with Groen Ontwikkelingsfonds Brabant. The establishment of this fund was a contrary motion to policy developments at the national level, where economic hardship motivated drastic cutbacks in nature policy (Buijs et al., 2014). Lastly, an actor in one tier may ignore (3) governance efforts in another, such as the few provinces that have reportedly been ignoring some of the policy intentions expressed in the national forest strategy.
Developments in the four domains described in the results are not only often embedded in place, they are also reflected upon differently at different scale levels. What climate change means to the European Union, means something completely different to a minor municipality. Moreover, due to principles of subsidiarity and proportionality, is not only unlikely that actors at different administrative levels want to address developments in the exact same manners, they are simply unable to work in parallel because they are restricted by their competencies.
Alignment between all tiers is not always necessary for upscaling mechanisms to be effective. Sometimes, under circumstances of limited or absent obstruction in surrounding tiers, policy actions at one level can be enough to result in upscaling. However, it is evident that eliminating occurrences of obstruction and ignorance could accelerate upscaling. In addition to the intermediary role that provinces could play (see section 4.3), respondents in the focus group suggest that a national ‘transition team’, a
International arrangements European Union National government Provinces Water authorities Municipalities Self-governance
group of transition experts, could help to coordinate the implementation of upscaling strategies, so to navigate through differences between tiers.
In sum, governance involved in upscaling food forestry is multileveled, in part because it touches upon so many policy domains. This is in agreement with ideas of Gupta et al. (2007, p. 164), who suggested the presence of similar dynamics of scale in climate change governance: “[t]o see
policies as occurring within ‘nested territorial containers’ would be missing the point. The political dynamics and influences at each level are very different and the windows of opportunity may open up at different moments of time”. Upscaling strategies for food forestry must thusly not only be tailored to
place-specific conditions, but to scale-related conditions as well. 5.3 Limitations & recommendations for further research
The research design drafted for this study has proven to be suitable for research into upscaling strategies and the spatial dependencies thereof. It must be noted, however, that experts in this study had a limited ability to think in strategies (as defined in section 2.2). Their high level of involvement in the subject matter constrained them, which resulted in a tendency to talk about tactic. For instance, the economic viability of food forests is a defining component of the results, suggesting that respondents presumed
ceteris paribus. The majority of these ‘tactical’ suggestions were still valuable to the research, as they
continue to be important looking further ahead. Nonetheless, later study could consider involving ‘future thinkers’ in focus group discussions on upscaling strategies, such as philosophers, scenario planners or even artists.
Due to practical constraints, the number of in-depth interviews that was conducted is limited. This might have narrowed the scope of the study. From the focus group, municipalities did not appear to be a key actor in upscaling, yet this perspective may be further explored. Document analysis was performed mainly for purposes of data validity, but could also be used as a way to search for additional strategies. In respect to food forestry, a research method for collecting and analysing NGO documents could be valuable, as self-governance efforts by NGO’s in food forestry are ubiquitous. Moreover, the database that was consulted for government documents in this study, Overheid.nl, yielded relatively few documents originating from other administrative levels than the national government. This is a potential data gap as well. Provincial forest strategies, for instance, were not retrieved from the database, though they could hold valuable information. Considering the major influence of European Common Agricultural Policy on changes in agricultural land-use, more emphasis could be put on E.U. documents. One final comment that must be made in respect to the research methods is that the small size and interconnectivity of the network of experts on food forestry may have frustrated data triangulation. It happened to be the case that some interviewees also contributed to documents that were part of the document analysis.
Although the study provides insights into the way place specificities play a prominent role in sustainability transitions in agriculture, the genesis of these place specificities was beyond the scope of this study. This limitation has been found in more studies within the geography of sustainability transitions (Hansen & Coenen, 2015). “As a result, the consensus is still that place-specificity matters
while there is little generalisable knowledge and insight about how place-specificity matters for transitions” (p. 105). Whether the existence or even the emergence of place specificities can be
forecasted, is a relevant question for stakeholders and could be the subject of further research. The study helps to begin understanding spatial contexts relevant to food forestry, but particular domains or strategies should be studied in-depth to make comparative analyses between regions.
One particularity of food forestry that might complicate generalisation of the results, is its multifunctionality. Because of this multifunctionality, there is a broad and diverse spectrum of entry points for food forestry projects in the Netherlands, which magnifies spatial differences in upscaling strategies. Notwithstanding the need to develop a wider array of sustainability interventions in agriculture and beyond, the benefits of these interventions may be more one-dimensional, especially when it comes to interventions focussing on incremental change. To generalise results, future studies could focus on other kinds of radical agricultural innovations and novelties in other sectors or geographical areas.
6. Conclusion
The Dutch agricultural sector is suffering from an array of sustainability crises which are accumulating into a fundamental threat to its continuity. Whereas sustainability interventions in agriculture have focused on incremental change for decades, integral, more transformative interventions have received increasing attention in recent years. In order to explore the spatiality of transformative change in the Dutch agricultural sector, food forestry was studied through a theoretical lens on the geography of transitions. Ten top-level stakeholders were consulted through focus group and in-depth interviewing. Data richness and validity were further attained through document analysis focusing on government documents. Out of some 50 initial codes, eighteen strategies for upscaling and, subsequently, four domains (ecological, socio-cultural, socio-demographic, policy) emerged from the results.
The research reveals how upscaling strategies for food forestry, and presumably other radical innovations in agriculture, are affected by spatial contexts. Across the four domains, place specificities were found affecting upscaling strategies. As such, the study firstly further emphasises the critical role of place in agricultural transitions. Results nuance findings of earlier research by Vermunt et al. (2020) showing fixed actor constellation in the agricultural sector as a result of geographic embeddedness, as governments may guide land-use change by involving in land acquisition. Here it is asserted that scholars should be careful in assuming the exclusive possibility of endogenous change in agriculture, as this excludes potentially valuable contributions from initiatives like food forestry emerging outside the incumbent agricultural regime to a transition towards sustainable agriculture. The study also provides concrete insights into ways place-identity can be mobilised to strengthen local support for innovations as food forestry. The question to what extent the unfolding of place specificities is predictable is relevant for both stakeholders and transition theory and needs further research.
Second, the study provides insights into the different scale levels at which agricultural transitions are governed. It identifies seven tiers of governance involved in the upscaling of food forestry and three ways in which these tiers may relate to each other. These interactions are classified as complementation, obstruction and ignorance. Actors in different tiers reflect upon developments in the four domains in different ways and, above all, they are restricted by their (legislative) competencies. Respondents suggest that provinces and a yet to be established ‘transition team’ could mediate between different tiers to improve alignment. The upscaling of food forestry must be addressed through multilevel governance, in a similar way to how climate change is believed to be best governed by Gupta
et al. (2007).
In all, the study has identified upscaling strategies for upscaling food forestry and showed how these can be embedded in spatial contexts – both with respect to place and scale – to accelerate upscaling. The findings contribute to a better understanding of geography in sustainability transitions and the emerging research agenda on governance in food forestry.
Acknowledgements
First and foremost, I would like to thank my primary supervisor Jordy Willems for his support during this research and my entire bachelor’s programme. I also wish to express my gratitude to Crelis Rammelt and Katinka Wijsman for their supervision. I am grateful for the willingness of all interviewees to participate in the research. For reviewing my initial analysis, my thanks go out to Dorith Vermunt.
References
Abbott, A. (2003). Human fatality adds fresh impetus to fight against bird flu. Nature, 423(6935), 5-5.
Barbour, R. (2007). Doing Focus Groups. London: Sage.
Boulestreau, Y., & van Eck, W. (2016). Design and performance evaluation of a 1ha productive food forest model (Doctoral dissertation, Wageningen University and Research Centre [WUR]).
Breidenbach, J., Dijkgraaf, E., Rooduijn, B., Nijpels-Cieremans, R., & Strijkstra, A. M. (2017). Voedselbossen van belang voor biodiversiteit. De Levende Natuur, 118(3), 90-93.
Bryman, A. (2016). Social research methods. Oxford university press.
Buijs, A., Mattijssen, T., & Arts, B. (2014). “The man, the administration and the counter-discourse”: An analysis of the sudden turn in Dutch nature conservation policy. Land Use Policy, 38, 676-684.
Coenen, L., Benneworth, P., & Truffer, B. (2012). Toward a spatial perspective on sustainability transitions. Research
policy, 41(6), 968-979.
Corbin, J. M., & Strauss, A. (1990). Grounded theory research: Procedures, canons, and evaluative criteria. Qualitative
Crawford, M. (2010). Creating a forest garden. Cambridge: Green books.
Cross, A. R., Baldwin, V. M., Roy, S., Essex-Lopresti, A. E., Prior, J. L., & Harmer, N. J. (2019). Zoonoses under our noses. Microbes and Infection, 21(1), 10-19.
El Bilali, H. (2019). The multi-level perspective in research on sustainability transitions in agriculture and food systems: A systematic review. Agriculture, 9(4), 74.
Elzen, B., Barbier, M., Cerf, M., & Grin, J. (2012). Stimulating transitions towards sustainable farming systems. In Farming Systems Research into the 21st century: The new dynamic (pp. 431-455). Springer, Dordrecht.
Environmental Data Compendium. (2020, June 22). Geschiktheid stikstofdepositie stikstofgevoelige landnatuur, 2018. https://www.clo.nl/indicatoren/nl1592-kwaliteit-stikstofgevoelige-ecosystemen
Geels, F. W., Kern, F., Fuchs, G., Hinderer, N., Kungl, G., Mylan, J., ... & Wassermann, S. (2016). The enactment of socio-technical transition pathways: a reformulated typology and a comparative multi-level analysis of the German and UK low-carbon electricity transitions (1990–2014). Research policy, 45(4), 896-913.
Green Deal Voedselbossen. (2018, July). Green Deal Voedselbossen. Rijksdienst voor Ondernemend Nederland. https://www.greendeals.nl/sites/default/files/downloads/GD219-dealtekst-Voedselbossen.pdf
Green Deal Voedselbossen. (2020). Voedselbossen; kansen voor regionale overheden. Verslag online kennisbijeenkomst 27 oktober 2020. https://greendealvoedselbossen.nl/wp-content/uploads/2020/06/Verslag-kennisbijeenkomst-voedselbossen-27-oktober-2020-def.pdf
Green Deal Voedselbossen. (2021, February 25). Eerste resultaten uit Nationaal Monitoringsprogramma Voedselbossen. https://greendealvoedselbossen.nl/2288-2/
Green Deal Voedselbossen. (n.d.). Voedselbossenkaart. Retrieved 19 March 2021, from
https://greendealvoedselbossen.nl/voedselbossenkaart/
Grin, J. (2010). Modernization Processes in Dutch Agriculture, 1886 to present. In J. Grin, J. Rotmans, J. Schot (Eds.), Transitions to Sustainable Development. New Directions in the Study of Long Term Structural Change (pp. 223– 314). Routledge.
Gupta, J., Van Der Leeuw, K., & De Moel, H. (2007). Climate change: a ‘glocal’problem requiring ‘glocal’action. Environmental Sciences, 4(3), 139-148.
Hansen, T., & Coenen, L. (2015). The geography of sustainability transitions: Review, synthesis and reflections on an emergent research field. Environmental innovation and societal transitions, 17, 92-109.
Hathaway, M. D. (2015). Agroecology and permaculture: addressing key ecological problems by rethinking and redesigning agricultural systems. Journal of Environmental Studies and Sciences, 6(2), 239-250.
Interview 1 (2021). Interview 1. Appendix A.1
Interview 2 (2021). Interview 2. Appendix B.1
Interview 3 (2021). Interview 3. Appendix C.1
Janssen, R., & Rietveld, P. (1990). Multicriteria analysis and geographical information systems: an application to agricultural land use in the Netherlands. In Geographical information systems for urban and regional planning (pp. 129-139). Springer, Dordrecht.
Ministry of Agriculture, Nature and Food Quality. (2020). Bos voor de toekomst: Uitwerking ambities en doelen landelijke Bossenstrategie en beleidsagenda 2030. Ministry of Agriculture, Nature and Food Quality. https://www.rijksoverheid.nl/documenten/kamerstukken/2020/11/18/uitwerking-ambities-en-doelen-landelijke-bossenstrategie-en-beleidsagenda-2030
Murphy, J. T. (2015). Human geography and socio-technical transition studies: Promising intersections. Environmental
innovation and societal transitions, 17, 73-91.
Nair, P. R. (1985). Classification of agroforestry systems. Agroforestry systems, 3(2), 97-128.
Nicolosi, E., Medina, R., & Feola, G. (2018). Grassroots innovations for sustainability in the United States: A spatial analysis. Applied Geography, 91, 55-69.
Onwuegbuzie, A. J., Dickinson, W. B., Leech, N. L., & Zoran, A. G. (2009). A qualitative framework for collecting and analyzing data in focus group research. International journal of qualitative methods, 8(3), 1-21.
Philip, S. Y., Kew, S. F., van der Wiel, K., Wanders, N., & van Oldenborgh, G. J. (2020). Regional differentiation in climate change induced drought trends in the Netherlands. Environmental Research Letters, 15(9), 094081.
QS Quacquarelli Symonds. (n.d.). Agriculture & Forestry. Top Universities. Retrieved 28 May 2021, from https://www.topuniversities.com/university-rankings/university-subject-rankings/2021/agriculture-forestry
Raven, R., Schot, J., & Berkhout, F. (2012). Space and scale in socio-technical transitions. Environmental Innovation and
Societal Transitions, 4, 63-78.
Rebisz, S. (2019). Exploring temperate food forestry as a sustainable l and management practice: starting at the soil.
Runhaar, H. A. C., Melman, T. C., Boonstra, F. G., Erisman, J. W., Horlings, L. G., De Snoo, G. R., ... & Arts, B. J. M. (2017). Promoting nature conservation by Dutch farmers: a governance perspective. International Journal of Agricultural
Sustainability, 15(3), 264-281.
Sekulova, F., Anguelovski, I., Argüelles, L., & Conill, J. (2017). A ‘fertile soil’for sustainability-related community initiatives: A new analytical framework. Environment and Planning A, 49(10), 2362-2382.
Smith, A., Voß, J. P., & Grin, J. (2010). Innovation studies and sustainability transitions: The allure of the multi-level perspective and its challenges. Research policy, 39(4), 435-448.
Statistics Netherlands. (2014, June 26). Afname aantal boerenbedrijven zet door. https://www.cbs.nl/nl-nl/achtergrond/2014/26/afname-aantal-boerenbedrijven-zet-door
1 Interview dates and types are not specified for privacy reasons. Appendices are available in a secondary document upon
Statistics Netherlands. (2019, January 18). Agricultural export value over 90 bn euros in 2018. https://www.cbs.nl/en-gb/news/2019/03/agricultural-export-value-over-90-bn-euros-in-2018
Statistics Netherlands. (2020, May 6). Feiten en cijfers over de landbouw. https://www.cbs.nl/nl-
nl/achtergrond/2020/19/feiten-en-cijfers-over-de-landbouw#:%7E:text=in%20de%20landbouw.-,Hoeveel%20bedrijven%20zijn%20er%20in%20de%20landbouw%2C%20waaronder%20sierteelt%2C%20volgens,sni jbloembedrijf%2C%20in%20totaal%20856%20bedrijven.
Statistics Netherlands. (2021, January 13). Geen bedrijfsopvolger voor meer dan 16 duizend boerderijen. https://www.cbs.nl/nl-nl/nieuws/2021/02/geen-bedrijfsopvolger-voor-meer-dan-16-duizend-boerderijen
Stokstad, E. (2019). Nitrogen crisis threatens Dutch environment—and economy. Science, 366(6470), 1180–1181. https://doi.org/10.1126/science.366.6470.1180
Swagemakers, P., García, M. D. D., Milone, P., Ventura, F., & Wiskerke, J. S. (2019). Exploring cooperative place-based approaches to restorative agriculture. Journal of Rural Studies, 68, 191-199.
Tweede Kamer Der Staten-Generaal. (2021, April 21). Motie van het lid Grinwis c.s. over het inzetten van voedselbosbouw voor versterking van de biodiversiteit van Natura 2000-gebieden. Tweede Kamer.
https://www.tweedekamer.nl/kamerstukken/moties/detail?id=2021Z06662&did=2021D14771
Van Assche, K., Beunen, R., Gruezmacher, M., & Duineveld, M. (2020). Rethinking strategy in environmental governance. Journal of Environmental Policy & Planning, 22(5), 695-708.
Van den Akker, J. J. H., & Hoogland, T. (2011). Comparison of risk assessment methods to determine the subsoil compaction risk of agricultural soils in The Netherlands. Soil and Tillage Research, 114(2), 146–154. https://doi.org/10.1016/j.still.2011.04.002
Van Dinther, M. (2019, June 4). Als ons landbouwsysteem failliet is, is het voedselbos dan het antwoord? de Volkskrant. https://www.volkskrant.nl/kijkverder/v/2019/als-ons-landbouwsysteem-failliet-is-is-het-voedselbos-dan-het-antwoord/
Van Doren, D., Driessen, P. P., Runhaar, H., & Giezen, M. (2018). Scaling-up low-carbon urban initiatives: Towards a better understanding. Urban Studies, 55(1), 175-194.
Van Doren, D., Giezen, M., Driessen, P. P. J., & Runhaar, H. A. C. (2016). Scaling-up energy conservation initiatives: Barriers and local strategies. Sustainable Cities and Society, 26, 227-239.
Van Veen, M. P., ten Brink, B. J. E., Braat, L. C., & Melman, T. C. P. (2008). Halting biodiversity loss in the Netherlands: evaluation of progress. Netherlands Environmental Assessment Agency.
Vermunt, D. A., Negro, S. O., Van Laerhoven, F. S. J., Verweij, P. A., & Hekkert, M. P. (2020). Sustainability transitions in the agri-food sector: how ecology affects transition dynamics. Environmental Innovation and Societal Transitions, 36, 236-249.
Viviano, F. (2021, May 3). This Tiny Country Feeds The World. National Geographic Magazine. https://www.nationalgeographic.com/magazine/article/holland-agriculture-sustainable-farming
Watson, R. T., Noble, I. R., Bolin, B., Ravindranath, N. H., Verardo, D. J., & Dokken, D. J. (2000). Land use, land-use change and forestry: a special report of the Intergovernmental Panel on Climate Change. Cambridge University Press. Zwartkruis, J. V., Berg, H., Hof, A. F., & Kok, M. T. (2020). Agricultural nature conservation in the Netherlands: Three lenses
