Introduction

Opportunities and Threats to Agroforestry in the Netherlands

There is a growing interest in agroforestry in the Netherlands. However, it is hard to find reliable data about the extent and growth of agroforestry (Den Herder et al. 2017). The total amount of hectares of agroforestry is unknown, but Van Dorp & Stobbelaar (2020) estimated the amount of food forests, as a specific form of agroforestry in the Netherlands as 0,1 promille of the Dutch agricultural area making it a niche activity compared to conventional agriculture. However, agroforestry practices are rapidly growing and all kinds of experiments are set up, especially alley cropping and silvopastoral systems. Plans are being made to speed up the growth; the idea is to have 25,000 hectares agroforestry, including 10,000 food forests, in 2030 in the Netherlands (Luske et al. 2020). To make agroforestry a viable technology, challenges should be mitigated and opportunities should be maximized. To support this, a national and several regional networks have been formed in the last few years (Agroforestry netwerk Brabant, n.d, agroforestry netwerk Gelderland, n.d., Agroforestry netwerk Nederland, n.d., http://www.europeanagroforestry.eu/countries/netherlands). This means that a basic infrastructure for change is in place. The current interest in agroforestry systems is mainly with agricultural producers who intend to convert (a part) of their agricultural land into agroforestry (Agroforestry netwerk Brabant, n.d, agroforestry netwerk Gelderland, n.d., Agroforestry netwerk Nederland, n.d.). This means that, if this transition will succeed, parts of the open agricultural landscape will transform into more forested areas. This change in landuse will have a profound influence on the landscape and the ecosystems.

According to Jose (2009) and Luske et al. (2020), the opportunities for a viable businessmodel in agroforestry lie in the potential to fulfill societal problems (biodiversity, water retention, food security, soil fertility, CO2 retention, etc. García de Jalón et al. (2018) indicated that improved biodiversity and wildlife habitats, animal health and welfare, and landscape aesthetics were seen as the main positive aspects of agroforestry. According to Luske et al. (2020) threats for agroforestry are among others: 1. Limited availability of agricultural land. Available land is scarce and very expensive in the Netherlands. There is a problem with finance in the first phase of an agroforestry system, when the trees are still developing and yields have not yet reached their full potential (Lovrić et al. 2018). 2. Unsupportive policies: the existing land use policies are not conducive for trees on farms (Ajayi & Place 2012; Urruth et al. 2022). 3. Holistic solutions are not considered: societal problems are usually divided into smaller problems, so the integrated value of agroforestry is not seen by policy makers and the larger public. 4. Agroforestry and its products are unknown by the public and retailers (Lovrić et al. 2018; Sollen-Norrlin et al. 2020) 5. The current agricultural system is in several ways locked in. This means that the current way of producing is set in its developmental path due to past developments. The policies encourage the existing system to develop further, making it difficult to change this existing system (Frison, n.d., Farstad et al. 2021).

So although agroforestry is growing in the Netherlands, and there is a lot of attention for it, there is still a threat of an impasse due to the many problems that are present. These problems can be understood in light of the larger forces acting on agriculture; the so-called drivers of change that bring about transformations in the agricultural system (Van Vliet et al. 2015; Plieninger et al. 2016, see also Sect. 2). By understanding the developments in, for example, technology, economy and/or institutions, ways out of the impasse can be found.

The question then is: how can a transition towards agroforestry be facilitated to overcome these threats and make use of its opportunities?. In this paper we will contribute to the understanding of the opportunities and threats regarding the transition to agroforestry, with a special focus on the drivers of change. Plieniger et al. (2016) describe how landscapes in Europe have been changing over time and what the driving forces of these changes were. As agroforestry is receiving more attention in the Netherlands, it is important to understand the drivers of change to be able to determine if they favour agroforestry.

Theory: Transformation towards agroforestry

In this section we introduce the transformation model we use to analyze the data (Fig. 1). It shows the transformation between two moments in time (T0 and T1), due to drivers of change influencing the social and physical realm simultaneously, to be able to explain that such changes don’t appear in isolation.

Fig. 1
figure 1

Transformation of realms through drivers of change (adapted from Van der Meulen et al. 2020)

Full size image

Social and physical realms

Land use and land use management are the result of many factors and influences. The owner/user of the land manages the area; decides which crops are grown, what investments are to be done etc. However, the landowner/-user is not an independent entity; the land has its own specific physical and locational characteristics. These characteristics are influenced by the environment around it, by factors that are not under direct control of the landowner/- user.

Stakeholders can perceive the impact of social factors differently: Laroche et al. (2019) indicated that stakeholders can perceive the impact of social factors to be greater than (bio)physical factors. We view the social and the physical realm co-evolve over time as subsystems of a land use system, influencing each other but not directly determining each other (Kemp et al. 2007; Carsjens 2009).

The social realm, such as laws, regulations an policies, encompasses behavioral and cultural aspects of the people living, working and interacting within a certain geographical area.

These relate to the users and the governance system as explained in Ostrom’s Social-Ecological System framework (Ostrom 2009). This formal context defines what is possible, what forms of land use are legal, how land use planning is done, etc.

By policy we mean a mechanism that indicates how the area is to be governed and how a landowner needs to act. This includes the steps to be taken when switching to a radically different farming system.

Logistics of the food chain are also important in this respect; food production in the Netherlands, even in extensive systems, is almost always directed toward a market. In most cases this is the world market, although in recent years, short value chains and local food markets have been gaining attention from consumers (Hardesty 2008; Schoolman et al. 2023).

Furthermore, the social realm can be regarded as a complex system (Moore et al. 2019), with interactions among diverse and constantly changing (human) factors and among its various stakeholders. Another characteristic of such a system is that new ways of working give rise to feedback loops that reinforce system behavior or lead to a discontinuance. Therefore new possibilities for agroforestry may influence the development of this farming system. Also efforts to introduce change can trigger stakeholders to self-organize and to develop a new state of relative stability (Moore et al. 2019), thereby making way for new systems e.g. new farming systems such as agroforestry.

The physical realm consists of the existing landscape, the soil conditions, local water tables, the local climate and the ecological conditions of an area. This is influenced by larger scale climate change, but also on a more local scale by the changes that occur due to land use. Potential land use is, influenced by what is possible within a given area, within a given physical realm. The physical realm can be also be further analyzed with Socio-Ecological Systems (SES) theories (Ostrom 2009).

The physical realm as a spatial element can also be defined by resource units and resource systems, with changes in spatial lay-out containing threats or opportunities for agroforestry. In SES studies these changes are not often incorporated (Stobbelaar 2020), therefore overlooking the important factor of time.

Drivers of change

To understand how and why agricultural land use changes, there is a need to take the factors into account that influence these changes, the so-called ‘drivers of change’. A driver of change is defined as ‘any natural- or human-induced factor that directly or indirectly brings about change in agricultural production systems’ (Hazell & Wood 2008, pg 501). Different drivers work at different scales such as at the global scale, the country scale, and the local scale, and even at the scale of the individual farm and landowner (Kristensen et al. 2016). The various effects and relative importance of these drivers of change can be difficult to distinguish. Interactions between drivers and other influential factors play an important role in the decision-making process at farm level (Hazell & Wood 2008; Mills et al. 2013).

We focus on what Kristensen (2016) calls the exogenous factors as drivers of change: the physical, socio-economic and policy environment (see also Table 1). These originate beyond the scope of the farm but have a large influence on the landowner/decision maker, and consist of aspects that are beyond the direct control of the landowner. These drivers are understood as factors that are relatively unaffected by other factors in the system (Van der Sluis et al. 2016). For example Quetier et al. (2005) describe s how changes in production rules within the Roquefort cheese protected geographical indication have caused farmers to develop an alternative land use strategy.

Table 1 Drivers of change related to agricultural production systems
Full size table

See Fig. 1 for a representation of the model as developed by Van der Meulen et al. (2020), relating the transformation of the social and physical realms to the drivers of change that influence both realms.

The temporal aspect is specifically presented in this model as the change or transformation over time. The drivers of change that influence this transformation, influence both the transformation of the social as well as the physical realms simultaneously. And as one of the realms changes, these changes will have an influence of the possibilities in the other reals as well. A change in the farmers’ behavior and with that, decisions regarding land use systems influences in the end the social and physical domains as well, in a continuous and dynamic feedback system (Feola & Binder 2010; Feola et al. 2015). The influence of small-scale developments by pioneers in the niche, how they influence general developments as well as how they are influenced by the general developments, is already described by Geels (2002). However, he doesn’t take the physical aspects of landscape, ecology and water into account, which are of crucial importance for agroforestry.

Specifically related to agricultural production systems, several authors have identified (exogenous) drivers of change at higher scales, and each of these drivers can be connected with the social and/or the physical realms. The drivers identified by Plieninger et al. (2016) and Van Vliet et al. (2015) operate at these larger scale levels and are a starting point for this research. In Table 1 the drivers as identified by these authors are listed. As depicted by the model in Fig. 1, the drivers of change have a direct and/or an indirect effect on both the social and the physical realms.

An important influence that can also play a role in these transitions are change agents. Van Vliet et al. (2015) mentioned that farmers can be important as moderators (or in other words, change agents) between drivers of change and the change in agricultural land use. Yet the role of actors, such as farmers and other individual stakeholders, have not been systemically considered in the past (Plieninger et al. 2016) and the role, position and knowledge of a change agent is important to understand as well (Cerf et al. 2011; Stobbelaar et al. 2022).

The actual land use is the visible outcome of the transformation of the social and the physical realm. It is important to realize that agricultural land use itself is not a fixed aspect of a certain location, but can change over the years. It is influenced by what is physically possible as determined by site characteristics such as topography and soil quality. However, agricultural land use is also determined by the most suitable production system and by what is and is not possible in the context of policy and laws (Van Vliet et al. 2015). At its very core, land use appears to be the aspect where the connection between the social and physical realm is most strongly felt.

The change from arable or dairy farming to agroforestry, is a very visible change and has a lot of impact in both the social and the physical realm. Therefore, we researched which drivers of change are present and in what way they influence the potential development of agroforestry. We used Lingezegen Park as a research case, since the establishment of the park was a result of land use planning, which in turn created new opportunities for new farming systems.

Methods

The case of lingezegen park

Lingezegen Park (Park Lingezegen, in Dutch) is a large public park with an area of 1700 ha between the cities of Arnhem and Nijmegen in the east of the Netherlands (see Fig. 2). The development of the park, which started in 2008, entailed a transition in the lay-out and in the use of the area, which was mainly under agricultural use (Park Lingezegen, n.d.). The area that forms Lingezegen Park was partly developed to support ecosystem services such as ecological corridors and partly developed to accommodate recreational opportunities, while part of the park remains agricultural, including six pilots with agroforestry initiatives, the total area of which is estimated to be 15–20 ha. Most of these initiatives are small scale pioneers. Two are actively inviting the general public on site by giving opportunities to visit and interact. Others are focusing on local marketing of their products. Depending on their success, these six initiatives could upscale or lead to more agroforestry initiatives, thus changing the characteristics of the area even further. It is important to know that five of the six initiatives are located in the northern area of the park, where the agricultural landscape is mainly open. The sixth initiative is in the south, where fruit production is part of the traditional farming system.

Fig. 2
figure 2

the location of Lingezegen Park (right) as situated in the country of the Netherlands (left)

Full size image

Pilot study

This study was organized as a pilot study to be able to get a first understanding of the opportunities and threats regarding the transition to agroforestry, with a special focus on the drivers of change.

In 2018 and 2019 rapid appraisals (Pretty 1995) were carried out in the area of Arnhem-Nijmegen, including the area of Lingezegen Park. The nature of a rapid appraisal is such that the approach is relatively quick and dirty, designed to get a first impression on stakeholders opinions on the research topic. This means that in a relative short time period, stakeholders are interviewed, the interviews analyzed and presented. The first rapid appraisal concentrated on the farming systems and the opportunities of the development of an agroforestry-based farming system within the area. For this rapid appraisal, 35 interviews were conducted with stakeholders who had an interest in agroforestry, foodforests and local production and marketing of food. The second appraisal was based on the first and geared more towards defining the drivers of change influencing the development of agroforestry. For this rapid appraisal, 21 interviews were conducted during one week of fieldwork and analyzed. Some of the stakeholders visited in 2019 were the same as in 2018. All of the stakeholders were selected from the area of Lingezegen Park and from the network build up during previous projects.

To develop an understanding of what the impact on the the area would be if agroforestry become a highly visible farming system, landscape design students analyzed the current landuse of the area and then visualized the potential changes of the area if agroforestry as a land use system would occupy larger areas in Lingezegen Park, up to 1,000 ha. This visualization was done with 3D software, with the use of the software programs Blender, Speedtree and Unreal Engine. Maps of the area were combined with an investigation of how an agroforestry system would look from various viewpoint (eye-level, birds’ eye view).

The key outcomes from these appraisals were that drivers that influenced the development of agroforestry, were not only related to the social realm (such as presence of a change agent and changing policies), but also to the physical realm (such as the landscape development in the area of Lingezegen Park). These results were used to develop the first version of the model as given in Fig. 1 (Van der Meulen et al. 2020).

The outcomes the visualizations were used as visual support for in-depth semi-structured interviews with three different stakeholders groups: a representative of Lingezegen Park, local producers (including an active member of the farmers’ union), and semi-governmental institutions. In total five interviews with individual stakeholders were held in the area of Lingezegen Park, to assess what they perceive as drivers towards the development of agroforestry.

The interviews were analyzed using a thematic analysis to determine what interviewees mentioned as drivers of change influencing the development of the area of Lingezegen Park, with specific reference (although not exclusively) to agroforestry. The number of times a driver of change was mentioned by the various stakeholders was taken as a rough estimate of the relevance for the individual stakeholder. With that, a rough comparison between the relevance of the various drivers of change between the different stakeholders could be made. This analysis was compared to the drivers of change found in the literature (Plieninger et al. (2016); Van Vliet et al. (2015) see Table 1) to determine if the drivers of change mentioned by the interviewees were comparable with those found in literature.

Results and analysis

Drivers of change in Lingezegen Park

From the in-depth interviews, a number of drivers of change came forward and were mentioned more than once (Table 2). what is noticeable, is that the importance of the different drivers varies with the type of stakeholders: the producers place more emphasis on the relations among stakeholders (scored +  + +), more so than the representative of the park board (scored +) or the semi-governmental agencies (scored +).

Table 2 Emerging themes as key drivers of change mentioned in the interviews
Full size table

Transition of the physical realm

The development of Lingezegen Park itself, or specifically the land use planning of the area, is one of the most mentioned visible drivers of change. Although the installment of the park was a political decision, the park itself has clear land use goals, such as water retention, sustainable land use and providing space for recreation.

In the past years, the land use in Lingezegen Park has changed considerably in some parts, due to the development of the park itself. The formal development of the landscape park in 2008, combined with land use planning in the period afterwards (Park Lingezegen, n.d.), made way for some changes that are clearly visible. There is more attention to tourism and touristic routes, to water retention at specific sites, to zones that ecologically connect various parts and to several small agroforestry sites found in the park.

An active change agent brought a group of innovators together, to explore more opportunities for agroforestry. They have started the first planting in 2020. At first they planted mainly hedgerows, which the open landscape of the agricultural part of Lingezegen Park, is a very a visible change. This shows that decisions taken in the social realm (i.e. deciding on a different type of farming system) will shape the physical realm (landscape and production method) and vice versa.

Most important aspects and changes seem to take place in the social realm, when looking at the interviews. However these changes affect the physical realm as well; the development of Lingezegen Park and the land use planning severely influenced the use of the area and the (agricultural) land use in some parts, affecting the physical aspects of the area. This functioned as a driver of change, for different forms of landuse (e.g. more recreation) as well as it opened options for further changes such as the transition to agroforestry.

Transition of the social realm

The relationships among stakeholders are mentioned as one of the most present forces in the area. Although in literature this is not usually named as a driver of change, it is an important force in the case study area. Producers, who have their farm in the area of Lingezegen Park, have clear views on what forces are operating in the area. They specifically mention the relationships among stakeholders quite often, while the institutions (Park board and semi-government) don’t see this as being as important.

During the first stages of the development of Lingezegen Park stakeholders from the areas were, according to themselves, not sufficiently taken into account. This has most likely contributed to the feelings and opinions articulated in the interviews. This process was perceived as quite top down and not very participative. The sentiments from the period of the park development’s initial stages could still be detected at moment of the interviews. Generally, the producers and other stakeholders express a need to cooperate with one another, as well as with new stakeholders such as inhabitants from Nijmegen and other local municipalities.

Furthermore social factors acting as drivers of change are the roles the various stakeholders within the area have. This complies with the findings of Laroche et al. (2019), who concluded that human factors related to social demands, policies, knowledge and profitability are more important than physical factors. These roles are mainly about who does what within the area, who takes responsibility and who takes the lead in developing initiatives. This ties in with the role of the individual change agent mentioned earlier. This person is mentioned in different interviews as the one who is able to organize new initiatives, mainly related to agroforestry. The chance agent acts, according to a narrative, strongly based on nature conservation in combination with sustainable agriculture. Agroforestry is a means to accomplish this, in this vision. (Rodriguez et al. 2009) stated that a change agent’s beliefs about sustainable agricultural practices (e.g. agroforestry) highly influences the support producers have for changes in their farming systems. This implies that this change agent has a good starting point for organizing the first steps of the further transition to an agroforestry system in the area.

Additionally with the development of the park, a vision on land use and maintenance was created for the area that is formally allocated to the park. This vision created opportunities for new initiatives; as even small adjustments of an agricultural production framework lead producers to rethink their system (Quetier et al. 2005).

In 2019, the aforementioned change agent had been able to motivate 17 other innovators. This group of innovators decided to apply for an agroforestry project with the NGO Natuur en Milieu Gelderland. Their effort demonstrated several actions that change agents can take to enable sustainable development, as identified by Mintrom & Rogers (2022); including articulating a clear vision; engaging others; and securing support from influential stakeholders.

Discussion

Developments related to threats and opportunities for agroforestry

Interestingly, not all of the themes coming from the interviews were identified from the literature as possible drivers of change (Plieninger et al. 2016; Van Vliet et al. 2015). Furthermore, not all drivers of changes from the literature where mentioned during the interviews. Social relations are not mentioned in the literature as a driver of change (Table 1), while in the interviews this is often mentioned. We suggest to add the driver local network to the literature to be able to take these social relations among stakeholders into account.

The interviews have shown that most important aspects and changes seem to take place in the social realm. This complies with the findings of Laroche et al. (2019), who concluded that human factors related to social demands, policies, knowledge, and profitability are more important than physical factors. However, these changes in the social realm also affect the physical realm; the development of Lingezegen Park and the land use planning. The outcome of this combined change meant chances for the growth of agroforestry in the area. Something that in itself also made changes in the physical realm (more woody elements) and social realm (new networks).

This study emphasizes the role of change agents. Rodriguez et al. (2009) stated that a change agent’s beliefs about sustainable agricultural practices (e.g. agroforestry) highly influences the support producers have for changes in their farming systems. This implies that this change agent has a good starting point for organizing the first steps of the further transition to an agroforestry system in the area.

The model we used (Fig. 1), adds the physical component to the traditional transition models (e.g. Geels 2002) and is more dynamic than the social-ecological system theory (Ostrom 2009). Therefore it is very helpful in describing the transition towards agroforestry as transition processes are very dynamic by nature; it gives a more exact description of what is happening.

Implications for transition towards agroforestry systems in the Netherlands

Institutional / political drivers

In Lingezegen Park a network of farmers interested in agroforestry has been established, besides and linked to the emerging agroforestry networks in the Netherlands (http://www.europeanagroforestry.eu/countries/netherlands,  n.d.). In addition, we have seen that there is a policy to change the use of a lot of land for water storage, nature development, recreation, etc. In our case it is the development of the Lingezegen Park. So there are changes in the social domain and changes in the physical domain, which can be forged together by change agents and where opportunities for agroforestry arise from it, especially in the availability of land.

Social-cultural drivers

Two important socio-cultural drivers that we found are motivation (to do things differently) and the role of the stakeholders. An interesting observation in our case study area is that part of the interest in agroforestry comes from stakeholders outside agriculture. These people are not yet locked in an existing system, they can think more freely about alternatives. In the network of stakeholders that has emerged in the region, they can inspire existing farmers to think differently.

Another important group of stakeholders that must be motivated for agroforestry and agroforestry products are the consumers. One of the problems of agroforestry is that it is too unknown to the public. To this end, the regional agroforestry network must be further expanded with, for example, restaurants that make delicious food from agroforestry products, such as De Nieuwe Winkel in Nijmegen (a restaurant with two Michelin stars).

Technological drivers

Switching to agroforestry has consequences for the physical system: the landscape will change considerably. It also has consequences for the social system, because working methods, regulations, etc. all change as well. Barriers to switching then lie in the unfamiliarity of actions and in the lack of mechanization (Van Dorp and Stobbelaar 2020). More attention will have to be paid to that. Another barrier, however, is that an agroforestry system needs time to mature and a financial gap can therefore arise between investments at the start and income that will only come much later. A solution that fits the character of agroforestry is the slow transition that many farmers use: not switching the entire farm at once, but plot by plot.

The role of change agents

For speeding up transitions, it is necessary that different sectors and levels in society collaborate (Stobbelaar et al. 2022). That is exactly the extra value of regional change agents. The case of Lingezegen Park shows that without change agents, sectors and (policy) layers would not have found each other in supporting integrated solutions like agroforestry. A change agent has the ability to find holistic solutions that are interesting for different parties. This allows the change agent to raise money from various sources and to align it with various policy goals.

Limitations of the study and future research directions

This study was a pilot study about the drivers of change in the transition to agroforestry. It draws upon a limited amount of data and is directed at a specific area in the Netherlands. Hence it is necessary that this study is expanded with more interviews and in other areas. Also the transition from existing forest towards agroforestry is interesting, whilst it has to take into account other types of policy.

Theoretically, it is interesting to further explore how drivers of change can be incorporated into the transition literature. From a practical point of view, it is interesting to study the influence of spatial planning on the possible development of agroforestry in a much broader sense.

Conclusion

For the transition towards agroforestry, in addition to all kinds of on-site innovations, it is also necessary to take due account of exogenous drivers of change, especially the functioning of local networks, land use planning and the availability of change agents. The combined effect of this largely determines opportunities for the growth of agroforestry. Furthermore, several threats have to be acknowledged, such as existing policies and the current locked-in development of agricultural production systems. Understanding the drivers of change can help to oppose these threats, and in the meantime taking advantage of existing opportunities. The case of Lingezegen Park clearly shows that the transformation of the social realm intertwines with developments in the physical realm. The development of the park, both a change in the physical as well as the social realm, made way for new initiatives in sustainable farming systems such as agroforestry. This was helped further along by a change agent in the park, operating in the social realm, with a coherent vision and action, making it possible for forerunners to organize themselves and take action, having an effect in the physical realm in turn. It is also important to realize that there were site-specific drivers present in the case of Lingezegen Park. This means that when working with a transition to agroforestry, or perhaps the transition of a farming system in general, it is necessary to understand the history of the social and physical realm (in our case the development of Lingezegen Park). The current stimulation of agroforestry through the national and regional agroforestry networks, will provide the opportunity to address these local issues.