Figures within the text

Fig. 1: Steps of the multi-scale, integrated participatory scenario development approach

Supplementary online material

Boundary conditions for GI implementation

Based on the information provided by the land use scenarios forCentral Germany (Priess and Hauck 2014), we asked the stakeholders of our regional scenario workshop to identify theframework conditions influencing farmers decisions regardingGI elements, e.g. whether or not they would retain certain elements, improve the existing ones or establish new elements. The conditions were summarised in the following clusters which were laterranked: financial incentives and income (selected 12 times); information and education (selected 11 times); regulatory guidelines (selected 7 times); value system and personal beliefs (selected 6 times); justice and equity (selected 4 times); technological advancements (selected twice). In the next step of the workshop, stakeholders were asked to makepredictionsabout how the framework conditions selected most often – “financial incentives and income”, “regulatory guidelines”and “information and education” – would develop up to2050 and would influence the ecological, economic and social consequences of GI implementation.

On the basis of input at the regional level, the following implementation scenarios (a., b., c.) were developed, in combination with keyquestions for the field visits:

  1. How would you change the GI on your farmland if
  2. no financial or any other (political) support for conserving or developing GI in the landscape wereprovided?
  3. GI weremaintainedand developed in the context of the EU Common Agricultural Policy’s greening measure “5 % ecological focus areas”?
  4. the EU were toset up a strong framework directive for the development of GI, with an independent programme and funding mechanism?
  5. Whatconsequences do you associate with a, b and c?
  6. What(potential) costs are associated with maintaining existing GI and establishing new GI (e.g. initial investments, opportunity costs)?
  7. What are other (potential) obstacles toestablishing new GI (e.g. information needs, ownership structures, etc.)?
  8. Whatbenefits doyou associate with GI which would be lost with option a?
  9. In reference to the three options above, whatfactorsplay a role in your farmmanagement?
  10. How should GIfunding measures be designed to suit your needs?

Finally, some background information was obtained regarding the type of farming business, the sizeof the holding in total, the leased area of the holding, and the main crops.

Full length methods section

This section appears as a summary in the printed article;the text here is the full-length version.

The aimof our study is to address the issue of perceived benefits and limitations ofGI implementation at four research sites in Saxony-Anhalt, Germany. To do this, we chose a participatory scenario developmentapproach thatlooks at possible future developments, as scenarios have proven to be useful forassessing the perceptions of local actors regarding landuse changes (e.g. Biggs et al. 2007; Reed et al. 2009; Plieninger et al. 2013).Specifically,“integrated scenarios […] may be thought of as coherent and plausible stories […] about the possible co-evolutionary pathways of combined human and environmental systems” (Swart et al.2004: 139). Based on overviews of scenario methods and techniques provided by Börjesonet al.(2006), Bishop et al. (2007), Mahmoud et al. (2009), Reed et al. (2009), March et al. (2012) and Priess and Hauck (2014), we combined the concept of multi-scale scenarios (Biggs et al. 2007) with the integrated scenario approach developed by Priess and Hauck (2014). This combination enabledus to conceptualiseand describe – in collaboration with regional and local actors– a range of possible future outcomes as well as the benefits and limitations of implementation scenariosfor this area.

Prior to the scenario development, we conducted a social network analysis (SNA) at the regional and local level (Figure 1) which served as an actor or stakeholder analysis (for a detailed description, see Hauck et al. 2016 and Werner 2013). The analysis revealed that at both levels farmers were perceived to have the largest influence on decisions relating to GIimplementation. Other important actors at the local level were the federal state agency for agriculture and forests, landowners, specialist journals and local communities. At the regional level administrative institutions were important, such as the federal state ministry for agriculture and the environment, both the federal state and regional nature conservation authorities, as well as farmers’ associations and landscape management associations. The networks were established for three categories: information, regulation and social pressure. This helped to reveal which kind of influence the actors have within the networks and what kinds of relationship exist. This was valuable information that helped in decidingwhom to invite to the participatory scenario development workshop and field visits.

The scenario development process involved the following steps:

Fig. 1: Steps of the multi-scale, integrated participatory scenario development approach

Step 1: Boundary conditions

In many scenario approaches “driver trajectories at the global scale are used as boundary conditionsto frame developments within the regional-scale scenarios” (Biggs et al. 2007: 8). We used Central German land use scenarios (Priessand Hauck 2014), which are in turn informed by the Global Environment Outlook (GEO4, UNEP 2007), as boundary conditions to inform the second step of the process, a regional stakeholder workshop. Literature on the history of the study sites was reviewed in order to gain an understanding of the development of agriculture in the area.

Step 2: Establishing framework conditions

For step two, all the regional stakeholders identified in the social network analysis were invited to a one-day workshop. The aim was to discuss the boundary conditions and to develop regional GI implementation scenarios. The stakeholders that were invited included biogas producers, farmers, landowners, mayors, agricultural consultants, and representatives of farmers’ associations, technical agricultural journals, a hunters’ association, the federal state agency for agriculture and forestry, the regional and federal state nature conservation authority, scientific institutions, environmental NGOs, the federal state office for agriculture, forestry and horticulture, the federal state department of environmental conservation, a landscape management association, the federal state ministry of agriculture and environment, the federalministry of state development and infrastructure, the conservation area administration, water and soil associations, and the regional planning office. Unfortunately,none of the actors representing agriculture outside of administrative bodies (such as farmers, farmers’ association or the agricultural journal) wished to attend the workshop or else they had to cancel at short notice. The 18 people who attended the workshop were given an introduction to the research design and then asked to identify the factors that potentially influence GIimplementation (for an overview of the results of these discussions, see “Boundary conditions for GI implementation” in this supplementary online material). We then presented two broad trends identified in step one: 1) a dominant focus on economic growth; 2) a growing focus on sustainability. We separated the participants into two groups and asked them to make a set of predictions about how the factors influencing GIimplementation would develop under one or the other trend. After the workshop, thesepredictionswere transformed into two compelling storylines: 1) Changes to green infrastructure under radical market forces and 2) Sustainable and inclusive development of green infrastructure.These were sent to the participants of the workshop with a request for feedback, in order to ensure the results had been correctly interwoven and condensed into coherent scenario storylines (see “Summarised scenario storylines” in thissupplementary online material).

Step 3: Eliciting perceived benefits and limitations concerning different GI implementation scenarios

In step three, the regional GI scenario storylines provided aframework for discussions with farmers and other local landusers, and helped in the development ofkey questions for field visits. As it happened, in the time between the regional workshop and the local level meetings, the CAPreform and the introduction of greening measures were made public. We therefore developed a third scenario in which the EFA greening measure was set as the GIimplementation pathway. This seemed appropriate, as EFA is believedto contribute to the implementation of the GIstrategy (Karhu 2011;Repohl et al. 2015) and the prospect of this scenario becoming reality was very concrete for farmers. This provided an incentive for them to participate in our exercise as they needed to consider including greening measures into their farm set-up soon anyway. The three scenarios used for the field visits can be summarised as follows: a) no financial or any other (political) support for conserving or developing GI in the landscape; b) GI is conserved and developed in the context of the CAP greening measure “5% ecological focus areas”, and c) the EU has set up a strong framework directive for the development of GI, with an independent programme and funding mechanism.

All the farmers within the research area were considered possible interview partners for the field visits and were contacted via telephone. The number of farmers contacted differed between the sites, ranging from two to eight farm businesses per research site. Not all landusers were willing to participate in our study, but for each research site at least one third of the farmed area was covered bythe participatory scenario exercise. The reasons mentioned for a lack of interest included time constraints and not being interested in environmental topics. Based on the results from the actor analysis, the other actors contacted were community representatives such as the mayor, the local branch of the farmers’ association, landscape management associations and the federal state agency for agriculture and forests. Only one mayor and one representative of the farmers’ association agreed to participate in our research.

The field visits were conducted during the wintertimeto reduce the likelihood of farmers not participating due to time constraints. Each research site was visited for about three hours, in two to three locations, depending on the willingness of the participants. The locations were chosen by the participants after being asked to select places where it was possible toview the research site from a higher vantage point or some other form of “lookout” point. In order to discuss the implementation scenarios, we asked the farmers how they would change the setup of their land based on the three scenarios and used a set of key questions to probethe associated benefits and limitations (see “Boundary conditions for GI implementation” in this supplementary online material).We used aerial photographs of the research sites where existing GI structures (other than the ones within the arable fields)werevisible. Light-coloured permanent markers were used tomark the structures and changes mentioned by the participants on two different versions of the maps, one for scenario a), and one for scenario b) and c); this was doneto derive spatially explicit scenarios. Highlightingexisting structures on the map and inthe surrounding landscape promptedmany points for discussion and was helpful forimagining concrete measures on and next to the fields.

We tried to organise the field visits in groups of farmers in order to give themthe opportunity to discuss with each other how they mightjointly develop GI across holdings. This approach worked only at Wanzleben and Friedeburg, however. At Wanzleben, one farmer and two representatives of the local landscape management associationjoined in the field visit. Afterwards more farmers were willing to participate, but they only agreed to an indoor discussion, which took place in a restaurant and lasted about two and a half hours,with the maps placed in the middle of a round table. At Friedeburg, two farmers participated in the field visit. At Greifenhagen only one farmer agreed to afield visit. In Schafstädt, one large farming business dominates the research site and refused to speak to us despite the efforts of the representative of the farmers’ association, who then agreed himself to a separate two-hour interview in a pub. We were eventually able to conduct the field visit in Schafstädt with the mayor, who had previously worked as a farmer on the research site fields.

Step 4: Analysing the benefits and limitations of GI implementation scenarios

Step four: All the field visit discussions were recorded digitally and later transcribed using the open access software F4. One exception was the group discussion in Wanzleben, where one participant did not agree to being recorded. There, the discussion was captured using written notes taken by the interviewer and another researcher. After each discussion, postscripts were written to record the most important observations. The analysis was done following the methodologyof “category-driven text analysis” (Mayring 2015: 13). The text was read numerous times and topics were marked with different colours and assigned to categories such as“financial incentives and income” (cf. Kuckartz 2016). This inductive coding (Schreier 2014) was repeated with each text until no new text passages remained to be marked. Relevant comments were extracted from the transcripts and language use was harmonised without changing the meaning. The relevant comments were grouped into the categories which were then further differentiated into subcategories. Because the amount of data was rendered manageable this way, the coded passages were copied into tables and the actual statements were used during analysis. This was done to avoid, as far as possible, the reduction and, with it, the loss of data (cf. Schreier 2014). The analysis was conducted qualitatively, partly because the data was collected to extract topics in an exploratory way and partly because the descriptive results were to be as close to the wording of the original data as possible (cf. Kuckartz 2016). The categories used for coding were then used as headings for the description of the data. This descriptive part was the main part of the results. Topics that were mentioned by all or most of the interviewees were deemed to be the most important ones (cf. Mayring 2015).However, other topics were also included in the results, as the aim of this study was to exploreall the factors that need to be considered in GI implementation, regardlessof representativeness or quantifiability. The categories derived from the data were used to structure the following results and discussion sections.

As respondents often explained their future decisions regardingGI by referring topast experiences, we present thesein combinationwith the results part below. Direct quotes within the results section derive from the respondents interviewed during the local level field visits.

Summarised scenario storylines

Here, we present summaries of the regional scenario storylines in orderto show which developments were predictedby the regional workshop participants.

Scenario 1: Changes togreen infrastructure under radical market forces

In order todeal with a severescarcityof resources and to easethe limited financial capacityof the state, markets are radically liberalised and globalisation is encouraged. The state interferesonly minimally with the free market economy. Regulations aimed at achieving ecological sustainability play a subordinate role and are no longer enforced. The market for agricultural products has been liberalised, and large scale farms operateaccording to short term, profitmaximising modes of production. There is no financial support available to protect or establish GI, including AEM. Farmers therefore implement GI measures only if they need to do so for economic reasons. A stark spatial divide exists between areas dedicated to nature conservation, such as national parks where tourism flourishes, and areas of high agricultural production where GI elements are sacrificed to expand production. One exception is organic agricultural production which is available for the wealthy sections of society only.

Scenario 2: Sustainable and inclusive development of green infrastructure

The severe scarcity of resources has led to a broadly held perception in society that “nature” deserves protection. There is a strong sense of responsibility and a commitment in society regardingecological concerns along withmany differentoptions for democratic participation. Agricultural subsidies are oriented towards ecological and social outputs. Regional processing and marketing of agricultural products is well established all over the country. Existing GI structures are established and maintained by farmers and local authorities alike. Also, citizens are actively involved. At EUlevel, a directive aimed atpreserving and re-establishing GI is mandatory for all Member States. A funding mechanism exists that enables extensive, participatory implementationof measures. Measures are longterm and partly mandatory, partly voluntary. Even though the regulations are widely accepted, effective control mechanisms are in place. Administrative structures exist toenable a swift response to funding applications. State-run competence centres enable knowledge exchange between science and practitioners and support thesharing ofbest practices.

Fig. 2 online: Location of study sites in Germany. 1) Wanzleben, 2) Greifenhagen, 3) Friedeburg, 4) Schafstädt (Mapof Germany by geodressing.de)

Fig.3 online: Major eras, events and changes inlanduse in the regionof what is today Saxony-Anhalt, Germany

Fig.4 online: Map of one research site showing 1) GI elements farmers would be likely to remove under scenario a(Aa-Ca) and 2) GI elements farmers would be willing to implement under scenario c(Ac-Cc) (Map: EUNIS2013 landuse classification (grouped), based on orthographic picture © GeoBasis-DE/BKG 2013.)