Community Research and Development Information Service - CORDIS

Final Report Summary - AGFORWARD (AGFORWARD)

Executive Summary:
The AGFORWARD project (Grant N° 613520; 2014-2017) had the goal to promote agroforestry practices in Europe that will advance sustainable rural development. The four objectives (described on the next page) were to 1) determine the context and extent of agroforestry in Europe, 2) identify, develop and field-test agroforestry innovations through participatory networks, 3) evaluate innovative designs and practices at field-, farm-, and landscape-scales, and 4) promote agroforestry in Europe through policy development and dissemination. Agroforestry is defined as the practice of deliberately integrating woody vegetation (trees or shrubs) with crop and/or animal systems to benefit from the resulting ecological and economic interactions.

Context
European agroforestry has been estimated to cover 10.6 Mha (using a literature review) and 15.4 Mha using the pan-European LUCAS dataset (i.e. 8.8% of the utilised agricultural area). Livestock agroforestry (15.1 Mha) is, by far, the dominant type of agroforestry. The LUCAS analysis provides a uniform method to compare agroforestry areas between countries and over time.

Identify, develop and field-test agroforestry innovations
40 stakeholder groups (involving about 820 stakeholders across 13 European countries) developed and field-tested agroforestry innovations which have been reported in 40 “lesson learnt” reports, and in a user-friendly format in 46 “Agroforestry innovation leaflets”. The innovations for agroforestry systems of high nature and cultural value included cheaper methods of tree protection and guidance for establishing legumes in wood pastures. Innovations for agroforestry with timber plantations, olive groves and apple orchards include the use of medicinal plants and reduction of mowing costs. Innovations for integrating trees on arable farms included assessments of yield benefits by providing wind protection. Innovations for livestock farms included using trees to enhance animal welfare, shade protection, and as a source of fodder. Peer-reviewed journal papers and conference presentations on these and other related topics were developed.

Evaluation of agroforestry designs and practices at field- and landscape-scale
A range of publicly available field-scale analysis tools are available on the AGFORWARD website. These include the “CliPick” climate database, and web-applications of the Farm-SAFE and Hi-sAFe model. The results of field- and landscape-scale analysis, written up as peer-reviewed papers, highlight the benefits of agroforestry (relative to agriculture) for biodiversity enhancement and providing regulating ecosystem services, such as for climate and water regulation and purification.

Policy development and dissemination
Detailed reviews of existing policy and recommendations for future European agroforestry policy have been produced. The support provided is far wider than the single specified agroforestry measures. The recommendations included the collation of existing measures, and that agroforestry systems should not forfeit Pillar I payments. Opportunities for farm-level and landscape-level measures were also identified. The project results can be found on the project website (www.agforward.eu), a Facebook account (www.facebook.com/AgforwardProject), a Twitter account (https://twitter.com/AGFORWARD_EU), and a quarterly electronic newsletter (http://www.agforward.eu/index.php/en/newsletters-1514.html). The number of national associations in Europe was extended to twelve, and a web-based training resource on agroforestry (http://train.agforward.eu/language/en/agforall/) created. AGFORWARD also supported the Third European Agroforestry Conference in Montpellier in 2016 attracting 287 delegates from 26 countries including many farmers. We also initiated another 21 national conferences or conference sessions on agroforestry, made about 240 oral presentations, 61 poster presentations, produced about 50 news articles, and supported about 87 workshop, training or field-visit activities (in addition to the stakeholder groups).

Project Context and Objectives:
The European Union has targets to improve the competitiveness of European agriculture and forestry, whilst improving the environment and the quality of rural life. At the same time there is a need to improve resilience to climate change and to enhance biodiversity. During the twentieth century, large productivity advances were made by managing agriculture and forestry as separate practices, but often at a high environmental cost. In order to address landscape-scale issues such as biodiversity and water quality, we argue that farmers and society will benefit from considering land-use as a continuum including both agriculture and trees, and that there are significant opportunities for European farmers and society to benefit from a closer integration of trees with agriculture for a European bio-economy. This continuum of agriculture and trees can be termed “agroforestry”, and is the practice of deliberately integrating woody vegetation (trees or shrubs) with crop and/or animal systems to benefit from the resulting ecological and economic interactions.

The AGFORWARD project (Grant Agreement N° 613520) was co-funded by the European Commission, Directorate General for Research & Innovation, within the 7th Framework Programme of RTD, Theme 2 - Biotechnologies, Agriculture & Food. The overall goal of the project was to promote agroforestry practices in Europe that will advance sustainable rural development, improve competitiveness, and enable social and environmental enhancement.

The project had four objectives which were addressed through ten work packages:
1. To understand the context and extent of agroforestry in Europe (work-package 1);
2. To identify, develop and field-test innovations to improve the benefits and viability of agroforestry systems in Europe. This is being achieved through four participatory networks focused on four sectors described on the next page (work-packages 2 to 5);
3. To evaluate innovative agroforestry designs and practices for locations where agroforestry is currently not practised or is declining and to quantify the opportunities for uptake at a field and farm scale (work-package 6) and at a landscape scale (work-package 7);
4. To promote the wider adoption of appropriate agroforestry systems in Europe through policy development (work-package 8) and dissemination (work-package 9).
There was also a project management activity (work-package 10).
Project Results:
This section describes the main science and technology results from each of the research, technology and development work-packages.

1 OBJECTIVE 1
The first objective of the AGFORWARD project was to understand the context and extent of agroforestry in Europe, and this was addressed by work-package 1.

Work-package 1: Context
Deliverable 1.1 is a report describing successful agroforestry practices from Mediterranean areas bordering Europe that could be used to encourage agroforestry in Europe. Pagella et al. (2014) summarised four key areas of work. Firstly the report highlighted some of the agroforestry practices in the Mediterranean areas of North Africa and Western Asia such as systems with olives, argan, cork oak and alley cropping with saltbushes (Morocco). The second part used a climate analogue approach to look at potential climates for four existing dehesa sites in Spain. The predicted climate in 2050 and 2080 for some of the dehesa sites was predicted to resemble the current climate in some areas in Morocco. Part three included a participatory case study focused on an altitudinal transect in northern Morocco which showed that the best mechanisms to support increased tree use varied between farmers, and this variability required a certain level of institutional capacity. The last section described the use of the WaNuLCAS model to determine how trees and annual crops adapt to mean annual rainfalls varying between 550 to 1070 mm. The model predicted that the impact of trees on crops depended on the rainfall distribution in any particular season.

One of the objectives in work-package 1 was to advance the mapping and quantification of agroforestry in Europe. Milestone 1 (den Herder et al. 2015) provided a report that made a first attempt to stratify and quantify agroforestry practices within Europe using the categories used in the AGFORWARD project i.e. agroforestry of high nature and cultural value, agroforestry involving high value trees such as olive and fruit trees, and agroforestry involving arable and livestock production. The report was based on a preliminary review of literature and datasets describing agroforestry practices across Europe. Combining the published values suggested an agroforestry area of at least 10.6 million hectares excluding reindeer husbandry. This value was substantially larger than the 3.3 million hectares categorised as “agroforestry” by the CORINE Land Cover classification. It was argued that the difference can be explained by agroforestry also occurring in CORINE classes such as “annual crops associated with permanent crops”, and “agricultural mosaics with significant natural vegetation”.

Deliverable 1.2 (den Herder et al. 2016) is a report describing the current extent and trends of agroforestry use in the EU27 using the LUCAS Land Use and Land Cover data. The Copernicus Land Monitoring Survey (high resolution maps with tree cover density) was also used to estimate tree cover on agricultural land in seven countries (Austria, Switzerland, Estonia, Latvia, Lithuania, Sweden and Norway). Both sets of results were compared with a review of the literature and statistical inventories. Using the LUCAS database the total area under agroforestry in the EU27 was estimated to be 15.4 million ha which was equivalent to 3.6% of the territorial area or 8.8% of the utilised agricultural area. This was a larger estimate than the 10.6 million ha reported in the literature review. Livestock agroforestry was, by far, the dominant type of agroforestry in Europe accounting for 15.1 million ha. The area of silvoarable systems was estimated to be 358 thousand ha. The above totals include the grazing and intercropping of permanent crops (e.g. fruit trees and olives) (1.05 million ha) comprising 0.85 million ha of grazed systems and 0.22 million ha of intercropped systems. A hot spot analysis revealed that a high abundance of areas under agroforestry can be found in south, central and north-east Portugal, south-west, central and parts of north Spain, south of France, Sardinia, south Italy, central and north-east Greece, central and west Bulgaria, and an area in northern Romania.

LUCAS data were also used to estimate the extent of single trees and green linear elements such as hedgerows. Agroforestry involving single trees covers almost 300 thousand hectares corresponding to around 0.02% of the territorial area in the EU, with extensive areas in France, Spain, the UK, and Italy. Agroforestry involving hedgerows cover about 1.78 million hectares representing around 0.42% of the territorial area in the EU, with large areas in France, the UK, and Italy.

The higher estimate for the agroforestry area using the LUCAS data (15.4 million ha) than the literature review (10.6 million ha) can be partly explained by the inclusion of Bulgaria (+0.87 million ha) and higher estimates for Spain (+1.74 million ha), France (+1.05 million ha), Romania (+0.71 million ha), Italy (+0.44 million ha), the UK (+0.39 million ha), and Sweden (+0.37 million ha); there were lower estimates for Greece (-0.67 million ha) and Portugal (-0.48 million ha). When the LUCAS estimates were compared with national inventories, the LUCAS-based estimate of 5.58 million ha of agroforestry in Spain was within 10% of a national inventory value of 6.14 million ha. However, the LUCAS estimate for Portugal (1.17 million ha) was 22% lower than an estimate based on national inventories. Although there may still be some systematic errors, because the LUCAS data were collected and analysed in a uniform manner, it provides a method to compare agroforestry areas between countries and regions. It can highlight where agroforestry is currently practised and also where there are opportunities for increasing the area of agroforestry. Considering that agroforestry covers about 9% of the utilised agricultural area of the EU, agroforestry deserves a more prominent place on policy agendas.

Deliverable 1.3 is a report (Rois Díaz et al. 2017) describing the environmental and socio-economic framework conditions for agroforestry in Europe. It used two approaches: i) qualitative interviews with 183 farmers and ii) the development of a detailed multi-criteria model based on the responses of 18 participants working with European agroforestry. The report started with brief reviews of the agricultural sector in Europe (Section 2) and current published knowledge on the socio-economic and environmental drivers of agroforestry (Section 3). Section 4 describes the qualitative interviews carried out with 98 farmers practising agroforestry and 85 farmers not practising agroforestry across eight European countries. The approach used grounded theory to establish ‘Why is agroforestry accepted or not? The face-to-face or telephone interviewees were centred on eight questions, regarding their understanding of agroforestry, and their perceptions of the positive and negative aspects of agroforestry, which were reported and examined in turn. Section 5 describes the development of an Analytic Network Process (ANP) model based on the responses of 18 participants working with European agroforestry. The participants’ understanding of the interactions between the benefits, costs, opportunities and risk of five agroforestry options were examined for a typical farm enterprise scenario for five different European biogeographical regions. Across the five regions and five management options, in terms of environmental benefits, the strongest relationships occurred in terms of lower inputs of pesticides and/or fertiliser. In terms of economic benefits, the strongest relationships were associated with lower labour costs, production of higher quality crops and timber, or lower business risk due to diversification. For social benefits, the strongest relationship was with increased knowledge and information on agroforestry. In terms of costs, the strongest relationships were with increased labour costs or increased competition between crops, trees, and animals. In terms of opportunities, the strongest relationships were with the availability of subsidies or higher employment. Lastly, in terms of risks, the strongest relationships were with low market opportunities or lack of subsidies. Across the five biogeographical regions, the option of not implementing agroforestry received its lowest ranking in the Mediterranean region, which is also the region where the implementation of agroforestry is currently highest.

Further reading for work-package 1
den Herder M et al. (2015). Preliminary stratification and quantification of agroforestry in Europe. Milestone Report 1 (22 April 2015). 57 pp www.agforward.eu/index.php/en/preliminary-stratification-and-quantification-of-agroforestry-in-europe.html
den Herder et al. (2016). Current extent and trends of agroforestry in the EU27. Deliverable Report 1.2. (15 August 2016). 2nd Edition. 76 pp. www.agforward.eu/index.php/en/current-extent-and-trends-of-agroforestry-in-the-eu27.html
Pagella T et al. (2014). Agroforestry from Mediterranean Partner Countries: Report on possible technology transfer from Mediterranean Partner countries to European countries. Deliverable 1.1 (29 Dec 2014) 35 pp. www.agforward.eu/index.php/en/agroforestry-in-north-africa.html
Rois Díaz, M. et al. (2017). Environmental and socio-economic framework conditions of agroforestry in different regions in Europe. Deliverable Report 1.3 (10 April 2017). 100 pp. www.agforward.eu/index.php/en/environmental-and-socio-economic-framework-conditions-of-agroforestry-in-different-regions-in-europe.html


2 OBJECTIVE 2
The second objective of the AGFORWARD project was to identify, develop and field-test innovations to improve the benefits and viability of agroforestry systems in Europe. This was achieved through four participatory research and development networks (PDRN) focused on four sectors: agroforestry of high nature and cultural value, agroforestry for high value tree systems, agroforestry for arable farmers, and agroforestry for livestock farmers

The philosophy behind having four PDRNs was that drivers for agroforestry vary according to the existing farming system. Thus, there are existing systems already recognised for their high nature and cultural value such as dehesas, montados, other forms of wood pasture, and hedgerow systems (WP2), where any substantial change could result in a reduction in the quality of the landscape and its benefits. Then there are growers with high value tree systems who could benefit from intercropping and grazing (WP3). Finally, there are arable farmers (WP4) and livestock farmers (WP5) who could benefit from integrating trees in their systems. Forty stakeholder groups were established and grouped into the four PDRNs.

Work-package 2: High nature and cultural value agroforestry
The first stage was for the ten stakeholder groups to identify the key opportunities, constraints and potential innovations in their systems. In Milestone 2.2, Moreno et al. (2015a) found that the key areas for research were: 1) to improve understanding of the effect of the tree canopies on pasture production and carbon storage, and; 2) to undertake a number of experiments and field trials within six headings: i) the design, management and governance of hedgerow systems in France, ii) tree regeneration through the use of seeding and the use of dead branch and wood mulches, iii) livestock management through the use of GPS collars and “invisible fencing” in the UK and Sweden, iv) the selection of legume pastures and winter forages in Spain and Italy, v) new products and consumer acceptance in Spain, Portugal, and Greece, vi) the choice of livestock species and breed to support conservation in in Romania and Hungary. These research items were taken forward as described in the synthesis of research and development protocols by Moreno et al. (2015b) in Milestone 4.

In Deliverable 2.4, Moreno et al. (2016) provide a synthesis of the system descriptions of the components, structure, ecosystem services, and economic value of the agroforestry systems of high nature and cultural value. A synthesis of the reports is provided by Moreno et al (2016). An initial report on the evaluations was provided in Milestone 5 (Moreno et al. 2017) and a full synthesis of the lessons learnt was provided in Deliverable 2.5 by Moreno et al. (2018a). They are considered in terms of the implications for system design and management, tree regeneration, livestock management, fodder resources, new products, and conservation.

Deliverable 2.6 (Moreno et al. 2018b) summarises and presents 15 “Agroforestry Innovation” leaflets that comprise two pages each and provides a quick description of the innovation in a user-friendly format. The innovation leaflets are also available on the individual web-pages for each stakeholder group and on the Resources page of the AGFORWARD website. The innovations include methods to increase fodder quality and yields in the grazing systems. Others describe the adaption of grazing schemes and GPS-based technologies to improve livestock management. There are also leaflets describing low-cost practices to either promote tree regeneration (where there are too few trees) or to reduce tree and shrub cover (if tree cover is too high). One leaflet describes the opportunities to promote products made from Valonia oak acorns. Lastly, some leaflets describe how agroforestry systems can be managed to provide ecosystem services. The dissemination of these innovations is described in Milestone 7 (Moreno et al. 2018c).

Five of the main lessons learnt are outlined below:
Maintenance of old trees and establishment of new trees: many traditional wood pastures systems are either being degraded through the progressive loss of trees or plot abandonment. Research in Romania focused on the importance of protecting large old trees in wood-pastures and highlighted the importance of engaging local communities and securing funds to actively maintain the multiple values of these systems. Work in Spain highlighted the benefits of using repellents against rodents on seed acorns, pruned branches to protect seedlings, and thorny wire meshes to protect saplings.

Creating marketable products: work in Portugal, Germany and Hungary focused on the need to be pro-active on understory shrub management in order to maintain and rejuvenate encroached wood pastures and hedge-rich agroforestry landscapes. This included the sale of wood biomass as wood fuel and the need for grazing management. Research in Greece demonstrated a strong market for acorn cups for tanning and nuts for flour production derived from wood pastures.

New grazing schemes and GPS-based technologies: work in Spain showed that fast-rotational intensive grazing (very high instantaneous stocking rates and long recovering periods for paddocks), increased productivity and enhanced carbon sequestration, soil quality, biodiversity and natural regeneration of tree cover. In the UK, invisible fencing that operates with cattle collars that produce an audible or electric stimulus when cattle are close to a buried wire allows the control of cattle movement without affecting public access. In Sweden, GPS collars improved the effectiveness of reindeer herd management. Whilst these technologies are currently more expensive than other options, they offer social and management advantages and future price reduction are expected.

Increasing on-farm fodder production: making better use of on-farm resources can help reduce costs. In Spain and Italy, sowing self-seeding legume-rich mixes increased pasture quality and productivity without compromising the biodiversity of plant community. In Spain, the cultivation of triticale in wood pasture proved successful as a double-use fodder crop. Modelling work completed in Portugal demonstrated that intermediate levels of tree cover in Mediterranean wood pastures extended the number of days over which the system could support a given population of livestock.

Reinforcing the provision of ecosystem services: whilst there are opportunities to increase the marketable outputs, it remains important to emphasize that wood pasture and hedgerow systems are critical for the structure and ecological function, biodiversity, and soil health of large landscape areas. In many cases, the systems also support the conservation of traditional breeds of livestock.

Further reading for work-package 2
Moreno G et al. (2015a). Innovations to be examined for high nature and cultural value agroforestry. Milestone 2.2. 20 January 2015. 20 pp. www.agforward.eu/index.php/en/innovations-to-be-examined-for-high-nature-and-cultural-value-agroforestry.html
Moreno G et al. (2015b). Synthesis of the Research and Development protocols related to High Nature and Cultural Value Agroforestry. Milestone Report 4. 16 October 2015. 22 pp. www.agforward.eu/index.php/en/synthesis-of-the-research-and-development-protocols-related-to-high-nature-and-cultural-value-agroforestry.html
Moreno G et al. (2016). Agroforestry of high nature and cultural value: synthesis of system descriptions. Deliverable 2.4. 28 June 2016. 23 pp. www.agforward.eu/index.php/en/describing-high-nature-and-cultural-value-agroforestry-systems-a-summary.html
Moreno G et al. (2017). Initial report on studied innovations for agroforestry of high nature and cultural value. Milestone Report 5. 7 November 2017. 45 pp. www.agforward.eu/index.php/en/initial-report-on-studied-innovations-for-agroforestry-of-high-nature-and-cultural-value.html
Moreno G et al. (2018). Agroforestry of high nature and cultural value: Results of innovations. Deliverable 2.5. 12 January 2018. 15 pp. www.agforward.eu/index.php/en/agroforestry-of-high-nature-and-cultural-value-results-of-innovations.html
Moreno G et al. (2018b). Agroforestry of high nature and cultural value: Guidelines for farmers. Deliverable 2.6. 7 January 2018. 37 pp. www.agforward.eu/index.php/en/agroforestry-of-high-nature-and-cultural-value-guidelines-for-farmers.html
Moreno, G et al. (2018c). Agroforestry of high nature and cultural value: dissemination of results and recommendations. Milestone 7.. 4 February 2018. 26 pp. www.agforward.eu/index.php/en/agroforestry-of-high-nature-and-cultural-value-results-of-innovations.html


Work-package 3: Agroforestry for high value tree systems
Work-package 3 focused on agroforestry for high value tree systems such as high value walnut and chestnut trees, olive and citrus groves, and apple orchards. As explained in Milestone 9 by Pantera et al. (2015), 10 stakeholder groups were established during 2014 and each established a research and development protocol. Pantera et al. (2016) in Deliverable 3.7 provides a description of the selected agroforestry systems. The first group in Spain focused on the use of legumes and grazing within walnut and wild cherry plantations. The second Spanish group focused on mushroom production within a chestnut system and the use of grafted plants of selected varieties of chestnut. Three of the systems comprised the intercropping or grazing of olive groves. There was one group in Italy, two groups based in Molos in Central Greece and in Chalkidiki in Northern Greece, and a group focused on intercropping between oranges in Crete. Three of the systems comprised the grazing of apple orchards with a replicated field experiment in Northern Ireland, and field demonstrations and economic assessments in Normandie in France and in Herefordshire in England. The last group focused on the management and use of pollarded trees in the Pyrenees in South-West France.

A report (Milestone 11) by Pantera et al. (2017) described the initial results from the work of the ten stakeholder groups. Measurements within a walnut plantation being grown for timber in Spain demonstrated that tree diameter increments can be greater in grazed, than ungrazed, plots, and greater with a legume than an unfertiliser grassland understorey. Intercropping orange trees with chickpeas, also in Greece, appears to provide benefits for orange production. Lastly the grazing of apple orchards was considered a sustainable system where the tree trunks were pruned to a height of at least 1-2 m. Deliverable 3.8 (Pantera et al. 2018a) provides a synthesis of the lessons learnt across the ten stakeholder groups and Deliverable 3.9 (Pantera et al. 2018b) summarises and presents eleven “Agroforestry Innovation” leaflets that provide two page guidelines on each innovation. The relevant innovation leaflets are also available on the individual web-pages for each stakeholder group. The main results are summarised against six headings:

Sowing legume intercrops and grazing rather than cultivation of high value timber systems. It was often common practice to cultivate between the trees in high value timber plantations in Spain. Intercropping with a legume forage can reduce cultivation costs; although this reduced soil water content there was no measureable negative effect on the trees. Intercropping with a legume was shown to increase the level of soil nitrogen and potassium and decrease the level of phosphorus in the soil and in tree leaves. Whilst unmanaged pasture resulted in an annual yield of 3 Mg ha-1 y-1 and could support around 0.6 LU/ha/y, sowing rich-legume pasture in the alleys could roughly double the stocking rate without compromise the tree growth. Overall silvopastoral management was shown to be compatible with high quality timber production and sowing legume-rich forages did not reduce tree growth relative to the control unfertilized trees.

Thinning and pollarding: the research in Spain demonstrated a slight positive effect of thinning on walnut tree growth. Pollarding in Spain had neither positive nor negative effects on tree growth, but the incidence of cavitation was reduced. The stakeholder group in France sought ways of estimating and maximizing the cost-effective production of firewood from pollarded ash trees. The results demonstrated that the biomass of the crown was closely related to the circumference of the trunk, and that pollarding could stimulate trunk diameter growth and branch wood production.

Introducing pigs to chestnut woodlands: there are substantial areas of chestnuts in the Galicia region of Spain. The research demonstrated that these stands could be grazed, but the trees needed to be adequately protected and pro-active management was needed. Successful agroforestry systems provided additional revenue for the farmer, preserved biodiversity and increased nutrient cycling.

Development of new grafting techniques: the research in Spain demonstrated that micro-grafting techniques provided a quicker way of multiplying the stock of specific chestnut clones than conventional propagation.

Understorey crops for olive and orange groves: the research in Italy demonstrated that wild asparagus (Asparagus acutifolius) could be successfully grown as an understorey crop in olive groves. Although yields were lower beneath the olive trees than in an open field, the proportion of edible tender part of the spear increases and the overall revenue from the grove is increased. Narcissus and tulips were identified as two alternative understorey crops. In Greece, sowing nitrogen-fixing chickpeas or a mixture of barley with common vetch beneath olives was shown to reduce fertiliser requirement and provided an additional source of revenue. During parts of an orange tree rotation when the tree canopy was small (perhaps because of the grafting of new scions), sowing chickpeas reduced fertiliser requirements and enhanced soil health.

Grazing of apple orchards: such systems can be successful if due consideration is given to four points. Firstly, there is a need to consider the apple tree structure (as sheep are likely to damage the canopy of young bush orchards where the canopy starts below 1.2 m). Secondly, orchard systems with zero or minimal levels of spray are likely to be more suitable for grazing because a high level of spraying requires greater movement of the sheep. Thirdly, the behaviour of sheep breeds can vary substantially. Some lowland breeds are relatively sedentary whereas some upland breeds can behave like goats and are able to get on their two hind legs to reach browse. Lastly, a successful grazed orchard system requires a manager or a management arrangement that pays attention to the health of the apple trees, the daily monitoring of the sheep, and the availability of grass. Where the above requirements are met, farmers can benefit from orchard grazing because of the reduced mowing costs, reduced feed costs for sheep production, and the opportunity to use grassland elsewhere between April and July to produce, for example, a hay crop.

Further reading for work-package 3
Pantera A et al. (2015). Innovations to be examined for agroforestry for high value tree systems. Milestone 9. 20 January 2015. 14 pp. www.agforward.eu/index.php/en/agroforestry-innovations-to-be-examined-for-high-value-tree-systems.html
Pantera A et al. (2016). Agroforestry for high value trees: synthesis of system descriptions. Deliverable 3.7. 27 June 2016. 10 pp. www.agforward.eu/index.php/en/describing-agroforestry-systems-with-high-value-trees-a-summary.html
Pantera et al. (2017). Initial report on studied innovations of agroforestry with high value trees. Milestone Report 11. 7 August 2017. 26 pp. www.agforward.eu/index.php/en/initial-results-on-studied-innovations-of-agroforestry-with-high-value-trees.html
Pantera A et al. (2018a). Agroforestry for high value tree systems: results of innovations. Deliverable 3.8. 30 January 2018. 11 pp. www.agforward.eu/index.php/en/agroforestry-for-high-value-tree-systems-results-of-innovations.html
Pantera et al. (2018b). Agroforestry for high value tree systems: guidelines for farmers. Deliverable 3.9. 17 January 2018. 30 pp. www.agforward.eu/index.php/en/agroforestry-for-high-value-tree-systems-guidelines-for-farmers.html


Work-package 4: Agroforestry for arable farmers
Work-package 4 focused on the application of agroforestry in arable systems. Arable agriculture provides large quantities of food, but it can be associated with reductions in soil and water quality, reduced biodiversity, and the release of greenhouse gases. During 2014, as reported in Milestone 15 by Mirck et al. (2014), 12 stakeholder meetings were held across Europe to identify the key opportunities and constraints associated with integrating trees and arable systems. The research and development protocols were presented in Milestone 16 by Mirck and Burgess (2015). During the project, some of the French stakeholder groups were combined and the arable crop selection work below wild cherry in Spain (originally under work-package 3) was moved to work-package 4.

In Deliverable 4.10, Mirck (2016) describes the systems, which all involve alley cropping to allow the use of standard machinery, and had alley widths ranging from 6 m to 96 m. In Southern France, the systems include durum wheat production beneath walnut, poplar and Sorbus domestica with a focus on the selection of durum wheat varieties or weed control. In Picardy in Northern France, the focus was also on the weed effects of the tree alleys. In Western France, the focus was on a black walnut intercropping system. In Spain, there was an interest in cereal production beneath walnut and the cultivation of maize or medicinal plants between walnut and wild cherry. Other Southern European systems included field beans and aromatic plants between walnut and cherry in Greece and poplar and oak being grown in along ditches in arable fields in Italy. In central Europe the systems included growing paulownia with alfalfa and maize in Hungary, growing sugar beet with poplar and black locust short rotation coppice in Germany, and the systems in Switzerland had a particular focus on fruit trees. In the UK, the systems involved mixed broadleaf trees with organic vegetables and wheat variety selection between hazel and willow coppice.

An initial report on the research was presented in Milestone 17 by Mirck (2017). Deliverable 4.11 (Kanzler et al. 2018a) summarised the final results of the “lesson-learnt” across all but one of the stakeholder groups. Deliverable 4.12 by Kanzler et al. (2018b) summarized the results from 12 agroforestry innovation leaflets. The guidelines for farmers were initially summarized in an internal report (Milestone 18) and Milestone 19 (Kanzler et al. 2018c) describes the dissemination of the results. Seven groups of key lessons learnt are outlined below.

Range of spacings and time period for arable crop: the width of the arable alleys in the studied systems ranged from 6 m to 96 m. In understanding the interactions between trees, tree rows and arable crops, it was important to know the alley width. It was possible to distinguish between two types of alley cropping: i) closely-spaced tree systems where the primary focus may be on the trees and the effects of shade will eventually mean that arable cropping becomes unprofitable and ii) wide-spaced systems which will allowed continued cropping even when the trees were fully matured. In the first type of system (examined in Spain) integrating an arable and tree crop increased revenues during the early stages of the tree rotation.

Continuation of mechanised agriculture: the studied agroforestry systems for arable farmers were all relatively new with most trees being planted within the last 20 years in relatively straight lines to enable continued use of standard arable machinery to plant, manage and harvest the arable crops.
Selection of tree species: Unlike the investigation of the SAFE Project (2001-2005) which focused on timber trees, a number of the systems included apple trees or short rotation coppice, which can generate earlier sources of revenue.

Complementary tree-crop combinations: some crop – tree combinations are more complementary than others. Early-maturing barley gave higher yields than wheat beneath walnut in Mediterranean Spain. Medicinal plants proved to be a promising understory crop with higher concentrations of active ingredients under shade conditions. There was also evidence that some durum wheat, wheat, and barely cultivars were more suited to being grown in proximity with trees than other cultivars.

Positive effects of shelter or shade on crop yields: it can be difficult to design experiments that allow the correct statistical testing of a positive effect of the trees on arable yields, relative to an open control area. There was some evidence that the wind protection benefits of trees in eastern Germany and Hungary provided crop yield benefits relative to control areas due to the improved microclimate. There was also some evidence that in particularly hot years in Mediterranean Spain, the yield of barley between plantation trees was greater than in a control area. By contrast, reduced arable yields near the tree row were measured in six out of ten experiments.

Tree-row management: research in Hungary indicated that bio-mulches could suppress weed growth in the tree rows.

Wider ecosystem benefits: benefits of improved pest control, pollination, habitat, or wildlife were highlighted in France, the UK and Switzerland. An improvement in landscape aesthetics was implied in Spain, Greece and the UK, which could benefit rural tourism. Reduced soil erosion and/or improved soil structure was suggested in France, Greece and Switzerland. Reduced nitrogen leaching and improved nutrient cycling was indicated in Spain. Increased carbon sequestration was emphasized in Spain, Italy, Greece and the UK. Shelter for crops and soil from extreme weather events induced by climate change were highlighted in Spain, Italy, the UK and Germany.

Further reading for work-package 4
Mirck J et al. (2014). Agroforestry innovations to be evaluated for arable farmers. Milestone 15. 20 January 2015. 11 pp. www.agforward.eu/index.php/en/agroforestry-innovations-to-be-evaluated-for-arable-farmers.html
Mirck J, Burgess PJ (2015). Synthesis of the research and development protocols related to agroforestry for arable systems. Milestone 16. 1 October 2015. 13 pp. www.agforward.eu/index.php/en/synthesis-of-the-research-and-development-protocols-related-to-agroforestry-for-arable-systems.html
Mirck J (2016). Agroforestry for arable systems: synthesis of system descriptions. Deliverable 4.10. 13 May 2016. 15 pp. www.agforward.eu/index.php/en/describing-agroforestry-systems-for-arable-farmers-a-summary.html
Mirck J (2017). Initial report on studied innovations of agroforestry for arable farmers. Milestone 17. 31 March 2017. 31 pp. www.agforward.eu/index.php/en/initial-report-on-studied-innovations-of-agroforestry-for-arable-farmers.html
Kanzler M et al. (2018a). Agroforestry for arable farmers: results of innovations. Deliverable 4.11. 28 January 2018. 16 pp. www.agforward.eu/index.php/en/agroforestry-for-arable-farmers-results-of-innovations.html
Kanzler M et al. (2018b). Agroforestry for arable farmers: guidelines. Deliverable 4.12. 17 January 2018. 33 pp. www.agforward.eu/index.php/en/agroforestry-for-arable-farmers-guidelines.html
Kanzler M et al. (2018c). Agroforestry for arable farmers: dissemination of results and recommendations. Milestone 19. 7 February 2018. 18 pp. www.agforward.eu/index.php/en/agroforestry-for-arable-farmers-results-of-innovations.html


Work-package 5: Agroforestry for livestock farmers
Work-package 5 focused on the application of agroforestry in livestock systems across three sectors: i) poultry, ii) pigs and iii) ruminants. During 2014, eight stakeholder groups identified the key opportunities, constraints and potential innovations to promote the use of trees in their livestock systems. Milestone 21 (Hermansen et al. 2015) explains that a key area of focus was to reduce the knowledge gap between researchers and practitioners through “best practice guides’ and to collate the nutritional value of the components of the shrubs and trees being used in agroforestry systems. A second area of focus was the need for new experimental work and a range of research and development protocols were developed, as reported in Milestone 22 (Hermansen 2015).

In Deliverable 5.13, Hermansen (2016) summarised the systems studied by eight stakeholder groups across Europe. The integration of trees with egg production was studied in the Netherlands and the UK. The combination of trees with pig production was studied in Denmark, Italy, and Spain, and with cattle or dairy production in France, the UK, and the Netherlands.

The initial results from the experimental work was reported in Milestone 23 (Hermansen et al. 2017). The agroforestry innovations related to poultry production focused on the integration of chickens with apple orchards, biomass willows and miscanthus, and the use of improved understorey swards. The innovations related to the integration of trees with pigs focused on tree protection and fodder tree evaluation. Those focused on trees with cattle address tree protection, the spatial organisation of trees, and the use of willow and alder as fodder trees.

Deliverable 5.14 (Hermansen et al. 2018a) summarises the results of the eight “lesson-learnt”. In Deliverable 5.15 (Hermansen et al., 2018b) the results of eight agroforestry innovation leaflets are summarised. Across the group, six areas of innovation were identified:

Animal welfare and premium products: the inclusion of trees provided protection from solar radiation, from the wind, and in the case of poultry, a perceived protection from predators. The research in Denmark, Italy, France, and the Netherlands demonstrated that the provision of shade improved animal welfare, for example, supporting thermo-regulation and reducing sun-burn in pigs. This is increasingly recognized, and from 2018, it will be mandatory in Denmark for organic pig producers to provide access to shade, apart from huts, during summer months for the benefit of animal welfare. Tree cover, in particular for hens, trees also provided protection against predators, stimulating the hens to use a larger proportion of the outdoor range and thereby minimize infection risk and hot spots with excessive nutrient load. The stakeholder group in Italy indicated that whilst few consumers were aware of the welfare benefits of agroforestry, there were opportunities for livestock farmers practicing agroforestry to secure a premium price. The woodland egg system in the UK is an example of where the use of agroforestry has become a minimum standard to access some parts of the market.

Tree protection: a common challenge across the systems was the protection of the young trees against damage from the livestock. In Italy, a cage settled around the tree (about 70-80 cm high) made of thin metal wire net seemed to be the most effective method of tree protection from pigs. In France, electric fence, electric fencing tape and metal fences were very efficient in protecting trees from cow damage in grazing periods.

Tree fodder and an on-line database: the research in France, the Netherlands and the UK demonstrated that foliage from trees can be a significant feed resource depending on tree species, in particular in terms of energy, protein and micronutrients for cattle. In particular, white mulberry (Morus alba) and common ash (Fraxinus excelsior) showed high protein contents. Data on composition and feeding value of tree leaves can be found in an on-line fodder tree database.

System design, ranging of animals, and ground cover: the poultry-focused research in the UK focused on the establishment and maintenance of a healthy productive sward beneath the trees. Establishing a sward under the trees is possible with commercial available seed mixtures, but the challenge is to maintain the sward in the presence of chickens. There was a need to keep the hens away from newly established areas and to optimize chicken pressure over the season. It was also important to select an appropriate level of tree cover, and to manage the range of the hens to minimize their use of newly established areas. The research with pigs in Denmark also demonstrated the importance of the location of the trees, relative to resting and feeding areas, to minimize, for example, nitrate leaching.

Improving the environment: In the Netherlands, France and the UK, the inclusion of trees on livestock farms was shown to increase biodiversity including earthworm numbers. The research in the Netherlands and Spain also demonstrated increases in soil carbon sequestration and protection against soil erosion. The work with poultry in apple orchards indicated that the poultry could diminish the need for chemicals to control pests.

The need for additional skills and addressing trade-offs: the research in the Netherlands demonstrated that the skills to successfully manage poultry can be different from those to manage trees. If a poultry farmer wanted to start profitable tree production, he or she was advised to establish a partnership with a ‘tree professional’. Successful agroforestry systems also needed to address trade-offs, such as balancing the use of contractors versus the need to minimize additional people coming on-site.

Further reading for work-package 5
Hermansen JE et al. (2015). Agroforestry innovations to be evaluated for livestock farmers. Milestone 21. 27 January 2015. 10 pp. www.agforward.eu/index.php/en/agroforestry-innovations-to-be-evaluated-for-livestock-farmers.html
Hermansen J (2015). Synthesis of the research and development protocols related to agroforestry for livestock systems. Milestone 22. 2 October 2015. 5 pp. www.agforward.eu/index.php/en/synthesis-of-the-research-and-development-protocols-related-to-agroforestry-for-livestock-systems.html
Hermansen JE (2016). Agroforestry for livestock farmers: synthesis of system descriptions. Deliverable 5.13. 21 May 2016. 6 pp. www.agforward.eu/index.php/en/describing-agroforestry-systems-for-livestock-farmers-a-summary.html
Hermansen JE et al. (2017). Initial report on studied innovations of agroforestry for livestock farmers. Milestone 23. 20 January 2017. 12 pp. www.agforward.eu/index.php/en/initial-report-on-studied-innovations-of-agroforestry-for-livestock-farmers.html
Hermansen JE et al. (2018a). Agroforestry for livestock farmers: Results of innovations. Deliverable 5.14. 15 January 2018. 9 pp. www.agforward.eu/index.php/en/agroforestry-for-livestock-farmers-results-of-innovations.html
Hermansen JE et al. (2018b). Guidelines for improved agroforestry systems for livestock production. Deliverable 5.15. 15 January 2018. 23 pp. www.agforward.eu/index.php/en/guidelines-for-improved-agroforestry-systems-for-livestock-production.html
Hermansen JE et al. (2018c). Agroforestry for livestock farmers: dissemination of results and recommendations. Milestone 25. 26 January 2018. 16 pp. www.agforward.eu/index.php/en/agroforestry-for-livestock-farmers-results-of-innovations.html



3 OBJECTIVE 3
Objective 3 of the project was to evaluate innovative agroforestry designs and practices for locations where agroforestry is currently not practised or is declining and to quantify the opportunities for uptake at a field and farm scale (work-package 6) and at a landscape scale (work-package 7).


Work-package 6: Field- and farm-scale evaluation of innovations
A distinguishing feature of agroforestry practices and systems is that there is a wide range of possible tree, crop and livestock arrangements which cannot be simply addressed by experimentation within the constraints of a four-year project. Hence the adaptation and evaluation of the long-term implications can often be most effectively achieved through the use of models.

Milestone 26 (Palma 2015, Palma 2017) describes the creation and development of an innovative on-line tool call CliPick. It can provide a standardized daily climate data (e.g. maximum and minimum temperature, solar radiation, relative humidity, wind speed and rainfall) for any location in Europe (and parts of North Africa) in a 12 km grid by facilitating access to existing publicly-available climate datasets used in the International Panel on Climate Change assessment report 5 (IPCC AR5). This innovation has potential use beyond the agroforestry sector.

In Milestone 29, Palma et al. (2016) describe improvements to the daily-time step Yield-SAFE biophysical model of tree and crop growth. These improvements included new routines to model cork and fruit production, the inclusion of the effect of trees on temperature and wind speed, prediction of soil organic carbon turnover, and leaching of nitrate. New improvements in the user interface were made to improve linkage with Farm-SAFE, using a recoded Python version of Yield-SAFE, which also allowed for development of an initial web-version of the Eco-Yield-SAFE model (available at: http://www.isa.ulisboa.pt/proj/ecoyieldsafe ). Deliverable 9.27 (Graves et al. 2017) describes the development of the Farm-SAFE bio-economic model to compare the profitability and economic impact of arable, forestry and silvoarable systems at a farm level and to include short rotation coppice and fruit tree systems. The model is openly available on the AGFORWARD website. Milestone 30 by Lecomte et al. (2016) describes the development of Hi-sAFe which is a 3-D mechanistic simulation model representing tree and crop growth, taking into account light, water and nitrogen competition between trees and annual or perennial crops. The Hi-sAFe code is written in JAVA under the CAPSIS modelling platform and uses STICS as the crop model. The Hi-sAFe model has been made available on-line at www1.montpellier.inra.fr/wp-inra/hi-safe/en/

In Deliverable 6.16, Gosme et al. (2016) report some initial results of using the above models. The Hi-sAFe model was used to determine the optimum orientation of the tree rows in silvoarable systems to maximize crop irradiance during the grain filling phases. North-south tree lines were identified as being preferable at high latitudes (>50°) and east-west tree lines at low latitudes (<40°). For latitudes between 40° and 50°, the tree line orientation was reported to have no significant impact on crop irradiation at key phenological stages such as flowering or grain filling (Dupraz et al, accepted). The Yield-SAFE model was used to estimate the land equivalent ratios of a range of agroforestry systems in Switzerland at tree densities of 40 or 70 trees per hectare. In 12 out of 14 scenarios, integrating trees and crops was predicted to be more productive than growing them in separate forestry and arable systems i.e. a land equivalent ratio higher than 1 (predicted land equivalent ratios ranged from 0.95 to 1.30). In a third paper, the Yield-SAFE model was used to predict land equivalent ratios of simulated eucalyptus-ryegrass agroforestry systems in Portugal between 1 (irrigated system with 52 trees per hectare) and 1.2 (for rainfed-systems with 203 trees per hectare). The Yield-SAFE model was also used to look at the relative carbon storage benefits of encouraging agroforestry on soils with low and high water holding capacity. The results suggested that implementing agroforestry on 10% of the land with high soil water capacity could result in the same amount of carbon storage as using agroforestry on 50% of the land with a low water capacity. This type of analysis is of interest to policy makers who determine the type of financial incentives given to support carbon sequestration.

In Deliverable 6.17, Palma et al. (2017) describe some of the modelling workshops held in Portugal, Greece and the UK to collect data and understand the management dynamics of the modelled systems. The report also explains how the impact of tree, crop and grassland interactions on livestock carrying capacity have been developed in the modified Yield-SAFE model (Oliveira et al, accepted). Additional parameters are also presented for 10 tree species, three types of pasture and five crop species. The report also includes a paper describing how the updated Yield-SAFE model was used to predict the effect of tree density on food, biomass and bioenergy production in agricultural, agroforestry and forestry systems for the montado in Portugal, cherry tree pastures in Switzerland, silvoarable systems in the UK, and short rotation coppice systems in Germany. Increases in tree density were shown to increase the “extractable” energy in the Swiss, British and German systems; in the Portuguese system it was assumed that the bioenergy stored in the trees would not be harvested (Crous-Duran et al, accepted).

Deliverable 6.18 (García de Jalón et al. 2017b) explains how the Farm-SAFE model has been updated to allow the quantification of environmental externalities. In the UK, whilst the arable system was the most financially profitable land use, it also produced the most negative externalities. Including the economic value of greenhouse gas emissions, carbon sequestration and loss of soil, nitrogen and phosphorus meant that silvoarable agroforestry showed a similar societal benefit as arable cropping and a greater benefit than forestry. The report also describes the development and use of a new model called Forage-SAFE (Garcia de Jalon et al. 2017a) specifically constructed to examine the interactions between trees, the grass understorey, and livestock on the profitability of a wood pasture system. The last part of the report focuses on soil erosion in Britany in France, which demonstrated that increasing tree cover in treeless areas on pasture, but particularly on arable land, could provide significant economic benefits when externalities are evaluated at a regional level.

Further reading for work-package 6
Crous-Duran J et al (accepted). Modelling tree density effects on provisioning ecosystem services , Agroforestry Systems
Dupraz C (accepted) Influence of latitude on the light availability for intercrops in an agroforestry alley-cropping system. Agroforestry Systems
García de Jalón S et al. (2017a). Forage-SAFE model: management and economics of wood pasture systems. Model delivered by the AGFORWARD project (613520). www.agforward.eu/index.php/en/forage-safe-a-tool-to-assess-the-management-and-economics-of-wood-pasture-systems.html
García de Jalón S et al. (2017b). Modelling the economics of agroforestry at field- and farm-scale. Deliverable 6.18. 13 October 2017. 85 pp. www.agforward.eu/index.php/en/modelling-the-economics-of-agroforestry-at-field-and-farm-scale.html
Gosme M. et al. (2016). Initial modelled outputs at field scale. Deliverable 6.16. 23 August 2016. 29 pp. www.agforward.eu/index.php/en/initial-modelled-outputs-at-field-scale.html
Graves A et al. (2017). Web-application of the Yield-SAFE and Farm-SAFE Model. Deliverable 9.27. March 2017. www.agforward.eu/index.php/en/web-application-of-yield-safe-and-farm-safe-models.html
Oliveira TS et al (accepted). Using a process based model to assess trade-offs between different holm oak densities and livestock carrying capacity, Agroforestry Systems
Palma JHN (2015) CliPick: project database of pan-European climate data for default model use. Milestone Report 26. 10 October 2015. 22 pp. www.agforward.eu/index.php/en/clipick-project-database-of-pan-european-simulated-climate-data-for-default-model-use.html
Palma JHN (2017). CliPick – Climate change web picker. A tool bridging daily climate needs in process based modelling in forestry and agriculture, Forest Systems [S.l.], v. 26, n. 1, p. eRC01, May 2017. ISSN 2171-9845. (http://revistas.inia.es/index.php/fs/article/view/10251)
Palma JHN et al. (2016). Yield-SAFE model improvements. Milestone 29. 5 July 2016. 30 pp. www.agforward.eu/index.php/en/yield-safe-model-improvements.html
Palma JHN et al. (2017). Modelled agroforestry outputs at field and farm scale to support biophysical and environmental assessments. Deliverable 6.17. 18 October 2017. 162 pp. www.agforward.eu/index.php/en/modelled-agroforestry-outputs-at-field-and-farm-scale-to-support-biophysical-and-environmental-assessments.html
Lecomte I et al. (2016). Improvement of the Hi-sAFe model. Milestone 30. 21 July 2016. 7 pp. www.agforward.eu/index.php/en/improvement-of-the-hi-safe-model.html


Work-package 7: Landscape-scale evaluation of innovative agroforestry
As indicated under work-package 6, the wide range of possible tree, crop and livestock arrangements means that the evaluation of agroforestry system cannot be simply addressed by experimentation within the constraints of a four-year project. Work-package 7 focused on evaluating the effects of agroforestry at a landscape-scale using a range of tools including models.

Deliverable 7.19 (Plieninger et al. 2016) describes three papers that used systematic reviews and meta-analyses to improve our understanding of the extent of European agroforestry and the effects of agroforestry on biodiversity, ecosystem services and farm profitability. Fagerholm et al. (2015) report on a systematic review of 71 scientific papers which describe the effect of agroforestry in Europe on ecosystem services (i.e. the benefits provided to humans by agroforestry ecosystems). There were a large number of reports focused on quantifying the productivity and the regulation of environmental processes by wood pastures; there was less focus on cultural and economic effects. It is suggested that future research could put more emphasis on socio-cultural effects, should include greater stakeholder participation, and should make use of new tools, such as spatially-explicit mapping. Torralba et al (2016) describe a meta-analysis using 53 publications to examine the effects of agroforestry on ecosystem services and biodiversity, relative to conventional agriculture or forestry. The results indicated overall benefits from agroforestry in terms of erosion control, biodiversity, and soil fertility, whereas the effects on agricultural production were mixed. In the last paper, Plieninger et al. (2015) provided a pan-European assessment of wood-pastures in terms of area, their ecological and social value, management issues, and concluded with an examination of the implications for policy.

In Milestone 33, Moreno et al. (2017) describe 12 socio-cultural catchments that have been selected to determine the effect of agroforestry on ecosystem services provision at the landscape scale. Within each of the 12 catchments across Europe, “agroforestry”, “agriculture”, and “forestry” landscape test sites of a defined size were selected for detailed mapping and comparison of ecosystem services. This report specifically describes the location of the socio-cultural catchments and the location and land uses of the landscape test sites.

Deliverable 7.20 (Fagerholm et al. 2017) describe six novel studies examining the ecosystem services and profitability of agroforestry practices. Kay et al., (in review) provide a synthesis of different ecosystem service assessment approaches tested in different agroforestry systems identified for WP 7. García de Jalón et al., 2017 assessed the economic impact of trees in wood pastures for farm profitability using a new economic model called Forage-SAFE. Torralba et al., 2017 assess the co-production of ecosystem services in the Spanish dehesas by exploring the relationship between biophysical and sociocultural factors and farm management practices. Fagerholm et al., present a case study in the South-West of Spain in which by using a map-based survey, identifies ecosystem services and their spatial patterns and relationship to land properties. The study also finds linkages between ecosystem service provision and subjective well-being. Hartel et al. (2016) assess farmers’ multiple values of scattered trees (mature and old) from oak wood pastures in a traditional rural region of Romania. Finally, Garrido et al. (2017) perform face-to-face semi-structured interviews (n = 34) to describe stakeholders’ appreciation of ecosystem services from dehesa landscapes in Spain.

Deliverable 7.21 (Roces-Días et al. 2017) describes seven papers examining the effect of agroforestry on profitability, biodiversity and ecosystem services at a landscape scale. The studies encompass a wide range of sites and approaches. A version of the report, including the seven papers, has been submitted to the European Commission. However, as of December 2017, only one of the studies has been published as a peer-reviewed scientific paper. The report on this website briefly describes the seven studies and includes a version of the published paper. The plan is that the remaining papers will be added once they have been peer-reviewed and accepted. The paper led by Sonja Kay modelled regulating and provisioning ecosystem services across six landscapes in Portugal, Spain, Switzerland and the UK. Landscapes dominated by agroforestry were predicted to show lower nitrate leaching, soil loss, and higher carbon sequestration and storage, than landscapes not dominated by agroforestry.

Deliverable 7.22 (den Herder et al. 2018) describes four papers examining the potential consequences of a widespread uptake of agroforestry by farmers on the delivery of ecosystem services and farm profitability. The paper led by Sonja Kay provides an indication where agroforestry can mitigate multiple environmental problems and where the implementation of new agroforestry systems would be very beneficial. The two studies led by Michail Giannitsopoulos assessed the financial (farmer perspective) and economic performance (societal perspective) of agroforestry, conventional farming and forestry while also evaluating environmental externalities (carbon sequestration, nitrogen surplus, phosphorous surplus and soil erosion) by monetarizing their costs and benefits. In the three investigated countries (Spain, Switzerland, United Kingdom), agroforestry became more profitable than arable farming when a value was attributed to the environmental externalities. This was due to the greater net benefit in regulating ecosystem services of agroforestry compared to arable farming. At a national scale, the combined net benefit of provisioning and regulating services increased in all three countries by establishing agroforestry or forestry on arable land. The study led by Christos Damianidis evaluated the potential of agroforestry as a measure to reduce forest fire risk in the Mediterranean region. Forest fires are a big risk in Mediterranean countries and catastrophic wild fires present a great threat to societies causing large economic losses and loss of life. Forest fire data from 2008-2017 indicated that agroforestry areas had fewer fire incidents than forests or shrublands providing evidence of the potential of agroforestry to reduce wildfire risk.

Further reading for work-package 7
den Herder M et al. (2018). Predicting the impact of the widespread uptake of agroforestry on ecosystems and farm profitability. Deliverable 7.22. 26 February 2018. www.agforward.eu/index.php/en/landscape-scale-evalutations.html
Fagerholm N et al. (2015). A systematic map of ecosystem services assessments around European agroforestry. Ecological Indicators 62: 47-65. Doi: 10.1016/j.ecolind.2015.11.016
Fagerholm N et al. (2016). Assessing linkages between ecosystem services, land-use and well-being in an agroforestry landscape using public participation GIS. Applied Geography 74, 30-46. Doi: 10.1016/j.apgeog.2016.06.007
Fagerholm et al. (2017). Ecosystem services and profitability of agroforestry practices. Deliverable 7.20. 26 March 2017; updated 10 July 2017. 80 pp. www.agforward.eu/index.php/en/ecosystem-services-and-profitability-of-agroforestry-practices.html
García de Jalón S et al. (2018). Forage-SAFE: a model for assessing the impact of tree cover on wood pasture profitability. Ecological Modelling 372, 24-32. Doi: 10.1016/j.ecolmodel.2018.01.017
Garrido et al. (2017). Stakeholder perspectives of wood-pasture ecosystem services: A case study from Iberian dehesas. Land Use Policy 60, 324–333. Doi: 10.1016/j.landusepol.2016.10.022
Hartel et al. (2016). Valuing scattered trees from wood-pastures by farmers in a traditional rural region of Eastern Europe. Agriculture, Ecosystems & Environment 236, 304-311. http://dx.doi.org/10.1016/j.agee.2016.11.019
Kay S, Herzog F, Szerencsits E, Crous-Duran J, García de Jalón S (in review) Landscape-scale modelling of agroforestry ecosystems services: A methodological approach.
Moreno G et al. (2017). Spatial characterization of sample landscapes. Milestone 33. 4 January 2017. 53 pp. www.agforward.eu/index.php/en/spatial-characterization-of-sample-landscapes.html
Plieninger T et al. (2015). Wood-pastures of Europe: Geographic coverage, social-ecological values, conservation management, and policy implications. Biological Conservation 190: 70-79. Doi: 10.1016/j.biocon.2015.05.014
Plieninger T et al. (2016). Synthesis of existing European agroforestry performance. Deliverable 7.19. 13 November 2015 (updated 7 July 2016). 87 pp. www.agforward.eu/index.php/en/synthesis-of-european-agroforestry-performance-in-terms-of-biodiversity-ecosystem-services-and-profitability.html
Roces-Días JV et al. (2017). Profitability, biodiversity and ecosystem services of agroforestry at landscape scale. Deliverable 7.21. 31 December 2017. 29 pp. www.agforward.eu/index.php/en/profitability-biodiversity-and-ecosystem-services-of-agroforestry-at-landscape-scale.html
Torralba M et al. (2016). Do European agroforestry systems enhance biodiversity and ecosystem services? A meta-analysis. Agriculture, Ecosystems and Environment 230: 150-161. Doi: 10.1016/j.agee.2016.06.002
Torralba M et al. (2018). Exploring the role of farm management in the coproduction of ecosystem services in wood pastures. In Press. Rangeland Ecology & Management. Doi: 10.1016/j.rama.2017.09.001


4 OBJECTIVE 4
Objective 4 of the project was to promote the wider adoption of appropriate agroforestry systems in Europe through policy development (work-package 8) and dissemination (work-package 9). The outputs of work-package 9 are described in the potential impact section.


Work-package 8: Policy
Milestone 36 (Santiago-Freijanes et al. 2016) comprises a series of maps of rural development measures implemented across the EU from 2007 to 2013 which can be potentially be used to support agroforestry. It forms part of the background study for an analysis of the extent and success of policy measures to promote agroforestry completed in Deliverable 8.23 (Mosquera-Losada et al. 2016). Deliverable 8.23 is a 95 page report which describes the extent and success of previous and current policy measures to promote agroforestry in Europe. In addition to the 15.4 million ha of silvopastoral and silvoarable practices in Europe estimated from the LUCAS (2012) dataset by den Herder et al. (2017), the report highlights an additional 2.7 million hectares of grazed shrubland and 1.8 million hectares of homegardens which are both considered as agroforestry by FAO and ICRAF. The third section describes the main international policy framework for European policy and demonstrates how agroforestry can support the achievement of global and European policies to promote sustainable agriculture and rural development. This includes the role of agroforestry to reduce and counteract greenhouse gas emissions (e.g. climate-smart agriculture) and improve biodiversity.

The Common Agricultural Policy (CAP) in Europe is based on two pillars. Section 4 of the report focuses on agroforestry and the “first pillar” which supports payments to farmers provided they meet Statutory Management Requirements (SMRs) and maintain the land in Good Agricultural and Environmental Condition (GAEC). This can require the maintenance of landscape features such as hedges and isolated trees, and buffer strips along water courses. However Pillar I payments are only made on designated “agricultural” land defined as arable land, grassland, and permanent crops (e.g. fruit trees and short rotation coppice). The report explains how wide hedges or having more than a certain number of trees per hectare can mean that arable land (independent of the level of production) can become ineligible for payments. The eligibility of grassland areas with trees is more flexible as it can take account of locally established practices, but this depends on national or regional implementation of that option. The report argues that the uptake and maintenance of agroforestry practices (and the associated societal benefits) depend on agroforestry remaining eligible for Pillar I payments if grassland and arable lands are considered.

Section 5 of the report focuses on the second pillar of the CAP, which includes measures to support rural development such as agri-environmental payments. The report examines a full range of measures that supported or supports the integration of trees and shrubs with farming in the 2007-2013 and/or the initial activation (December 2015) of the 2014-2020 rural development programmes. This includes measures to support forest farming, silvoarable practices (forest strips and small stands, hedgerows, and isolated trees), and silvopasture practices (silvopastoralism and permanent crops, and mountain pastoralism). In total about 27 measures can potentially support agroforestry (in its broad definition). The section also includes a more detailed review of the specific “agroforestry” measures 222 (2007-2013) and 8.2 (2014-2020) to support the establishment of widely spaced trees on arable land.

Deliverable 8.24, which builds on Deliverable 8.23, is a 21 page report (Mosquera-Losada et al. 2017) which describes how policy can support the uptake of agroforestry in Europe. This report comprises seven sections and makes 15 recommendations. Section 1 explains that the objective is to provide guidance to policy makers in Europe on how modifications to policy can increase the uptake of agroforestry. Section 2 highlights why agroforestry should be supported as it is a sustainable land management option that delivers a large number of the market and non-market goods and services needed to address many high-policy-level societal goals. Section 3 defines agroforestry and the major types or practices of agroforestry in Europe. It defines agroforestry as “the deliberate integration of woody vegetation (trees and/or shrubs) as an upper storey on land, with pasture (consumed by animals) or an agricultural crop in the lower storey. The woody species can be evenly or unevenly distributed or occur on the border of plots. The woody species can deliver forestry or agricultural products or other ecosystem services (i.e. provisioning, regulating or cultural).” The five major types of agroforestry across Europe are: silvopasture; silvoarable; hedgerows, windbreaks and riparian buffer strips; forest farming and homegardens.

Section 4 focuses on cross-compliance and recommends that woody vegetation promotion and preservation linked to landscape features should be simplified with clearly stated objectives. Section 5 describes agroforestry in Pillar I of the Common Agricultural Policy and its relationship with cross-compliance, basic and greening payment. The report provides recommendations on how agroforestry should be supported on: i) arable land, ii) permanent grassland, and with iii) permanent crops. A recommended mechanism to allow farmers to establish, maintain and improve agroforestry practices, whilst retaining full Pillar I payments, is through the use of agroforestry management plans. It is also recommended that a fourth section called “agroforestry” is included as part of the future greening payment as it is one of the most effective ways of mitigating and adapting agriculture to climate change. Section 6 focuses on Pillar II measures related to agroforestry. Building on Deliverable 8.23, it recommends that the 27 measures associated with agroforestry practices should be presented as a single unique measure. It recommends that there should be Pillar II support for agroforestry establishment and management on both: i) agricultural land and ii) forest land. It also recommends the use of other activities or measures to encourage agroforestry through result-based payments at farm- and landscape-levels, through co-operation within the value chain, and the support of agroforestry knowledge at all education levels. A final global conclusion is that one way to advance the above to the wider benefit of Europe is through the development of a European Agroforestry strategy.

Further reading for work-package 8
den Herder M et al. (2017). Current extent and stratification of agroforestry in the European Union. Agriculture, Ecosystems and Environment 241: 121–132.
Mosquera-Losada MR et al. (2016). Extent and success of current policy measures to promote agroforestry across Europe. Deliverable 8.23. 8 December 2016. 95 pp. www.agforward.eu/index.php/en/extent-and-success-of-current-policy-measures-to-promote-agroforestry-across-europe.html
Mosquera-Losada MR et al. (2017). How can policy support the appropriate development and uptake of agroforestry in Europe? Deliverable 8.24. 7 September 2017. 21 pp. www.agforward.eu/index.php/en/how-can-policy-support-the-uptake-of-agroforestry-in-europe.html
Santiago-Freijanes JJ et al. (2016). Maps and indicators of rural development measures potentially related to agroforestry, across the EU (2007-2013). Milestone 36. 14 January 2016. 28 pp. www.agforward.eu/index.php/en/maps-and-indicators-of-rural-development-measures.html


Potential Impact:
This section describes the potential impact of the project, including the socio-economic impact and the wider societal implications of the project so far. It also describes the main dissemination activities and the exploitation of the results.

1.1 SOCIO-ECONOMIC IMPACT AND WIDER SOCIETAL IMPLICATIONS
The socio-economic impact of the project is discussed in terms of the four objectives of the project.

1. To understand the context and extent of agroforestry.
The pan-European LUCAS dataset can be used to provide a uniform systematic assessment of the extent of agroforestry. The analysis highlights that agroforestry is a significant European land use representing 15.4 Mha, and hence it is a valid focus for land use policy. The approach will allow the assessment of changes in land use over time, which in turn can help inform decisions regarding Land Use, Land Use Change and Forest (LULUCF) inventories in relation to climate change.

2. To identify, develop and field-test innovations.
The project has established 40 stakeholder groups, which are working with about 820 stakeholders across 13 European countries. These groups came together to provide solutions to practical issues. Each group produced a research protocol, a system description (which is publicly available), a "lessons learnt" report, and an "agroforestry innovation" leaflet. These resources provide a European-wide resource for stakeholders interested in establishing their own agroforestry systems.

3. To evaluate innovative agroforestry designs and practices.
As a result of the project, significant improvements and improved access have been provided for climate datasets and the Yield-SAFE and Hi-sAFe biophysical models of tree-crop interactions. Using these models, with the bio-economic Farm-SAFE model, allows assessment of the financial impact of agroforestry practices relative to conventional agriculture and forestry. The inclusion of societal values for regulating ecosystem services such as carbon sequestration, nutrient and soil retention, and runoff control typically increases the economic value of agroforestry and forestry systems, relative to agriculture. Hence the use of these models can help inform improved decision making. The impact of agroforestry can also be determined at a landscape-scale where it can have a significant positive impact on biodiversity and cultural ecosystem services such as aesthetics and recreational opportunity that enhance human well-being. Research from Spain highlights the importance of public access in maximising cultural services, and that ecosystem services are generally increased from a mosaic of landscapes.

4. To promote the wider adoption of appropriate agroforestry systems in Europe through policy development and dissemination.
The Common Agricultural Policy (CAP) (including rural development programmes) comprises about 39% of the annual EU budget and its effective use is of socio-economic importance. Agroforestry can offer social, environmental and animal welfare benefits whilst maintaining food production and hence it is a valid recipient of public support. The project proposes that the CAP adopts a broad definition of agroforestry and the project a means of categorising agroforestry types across a range of land uses. Because of the social and environmental benefits and continued food production, it is recommended that the managed developed of agroforestry on agricultural land remains fully eligible for Pillar I payments. A review of the 2007-2013 rural development programme highlighted that there is about 27 measures that can support agroforestry practices in their broadest sense. The project recommends that future programmes of the CAP uses a broad definition of agroforestry and brings together the current range of measures under the title of "agroforestry". There also appear opportunities to promote agroforestry through measures focused at farm- and landscape-levels.

The outputs of the project can continue to help with the dissemination of best agroforestry practice. The creation of 46 separate agroforestry innovation and 10 agroforestry best practice leaflets are a significant resource which have been made available in hardcopy and pdf formats. Much of the promotional work will continue at regional and national level. For example AGFORWARD staff were active in the initiation, by the French government, of a National Plan for the Development of Agroforestry in December 2015. AGFORWARD also supported the successful Third European Agroforestry Conference in May 2016 and a Fourth Conference is planned at Nijmegen in the Netherlands from 28-31 May 2018. Across Europe, there is increasing interest in agroforestry from farmers, policy makers, and advisors, who recognise that integrating trees with farming can be both financially and environmentally beneficial.

1.2 MAIN DISSEMINATION ACTIVITIES
The main dissemination activities within the AGFORWARD project took place within Work-package 9. The first dissemination task of the project was to create a dissemination protocol which was established in the initial months of the project. The second task was to disseminate much of the communication around an interactive internet-platform. The third task was to develop literature and guidance for specific user groups, the fourth task was “regular communication” during the project through the use of electronic newsletters, newspaper articles, and briefings from each WP. The fifth task focused on the provision of education tools, and the sixth task was to coordinate regional conferences and workshops, including a European presence at the World Agroforestry Congress in India in 2014 and a final project conference. These tasks are covered in turn.

Task 1: Establishment a dissemination protocol
A dissemination protocol was established in the initial months of the project.


Task 2: Dissemination around the website
A key avenue of communication, which comprises Deliverable 9.25, has been the AGFORWARD website (www.agforward.eu). The website was created within the first six months of the project in June 2014 and it has continued to evolve during the project. All of the deliverables produced during the project are available on the AGFORWARD website. The main pages of the website have been translated into 12 languages: English, French, Spanish, Greek, Romanian, Hungarian, German, Portuguese, Danish, Dutch, Italian and Polish. More than 40,000 visitors have visited more than 210,000 pages, and one third are returning visitors.


Task 3: Literature and guidance for specific groups
The literature and guidance included details of demonstration agroforestry sites, making available decision-making tools, supporting the creation of national associations, the production of leaflets on agroforestry innovations and best practice, and regional information on agroforestry policies.

Demonstration plots
Deliverable 9.26 is an on-line map showing the location of the stakeholder groups and selected agroforestry demonstration sites across Europe (Figure 3).

Decision making tools
Deliverable 9.27 (Graves et al., 2017) comprised the web-application of the Farm-SAFE bio-economic model for comparing the profitability of agricultural, forestry and agroforestry systems. The Hi-sAFe model has also been made available on-line through a registration process (https://www1.montpellier.inra.fr/wp-inra/hi-safe/en/). A new model called Forage-SAFE for examining the profitability of wood-pasture systems has also been made available on-line (García de Jalón et al. (2017). A new implementation of Yield-SAFE, the Eco-Yield-SAFE model became available online (http://www.isa.ulisboa.pt/proj/ecoyieldsafe).

Creation of national associations
Deliverable 9.28, completed in December 2015, was an expansion to twelve national agroforestry associations across Europe (Liagre et al. 2016). A webpage has been created to identify the contact details and the statutes of each association.

Booklets
Deliverable 9.30 (Liagre et al. 2018) combines a series of ten “Agroforestry Best Practice” leaflets based on substantial field experience. They were produced by Philippe Van Lerberghe with colleagues from the French Agroforestry Association, the World Agroforestry Centre and the other AGFORWARD partners. The first leaflet focuses on the importance of thinking through required objectives. Subsequent leaflets cover: the selection of appropriate tree species and planting material; appropriate designs; land preparation; tree planting and protection, and weed control. The last leaflet focuses on the role of pruning to increase the value of wood being grown for timber. The ten innovation leaflets together with the 46 agroforestry innovation leaflets produced from work-packages 2 to 5 were brought together in a folder (Blaguer et al., 2017). In parallel, a web version has been posted on the AGFORWARD website, from where they can be printed.

Task 4: Regulation communication through newsletters and social media
The AGFORWARD project produced a quarterly electronic newsletter which was sent regularly to 574 people with an interest in European agroforestry. Although not specifically mentioned in the Description of Work for the project, the AGFORWARD project has made extensive use of social media to promote the results of the project. A Facebook (https://www.facebook.com/AgforwardProject) page was created from the first month of the project, has over 1000 followers, and has been useful to generate interest in the project. In total 190 items have been posted since the creation of the Facebook page in Oct 2014. A Twitter account (@AGFORWARD_EU) was created in October 2014 which has produced 580 tweets (to February 2018) and has over 300 followers. There is also an AGFORWARD YouTube channel: https://www.youtube.com/channel/UCKMVZuFsp9qSfSHtUm0ubCQ that has over 300 followers.

Task 5: Educational tools
In June 2016, a training tool-kit (Deliverable 9.29) (Liagre et al. 2016b) was created on a new web-page domain was created (http://train.agforward.eu/language/en/agforall/. The training toolkit comprises 121 documents including videos, pdf, and slideshows. This task was completed in February 2018 with the lessons learnt reports being made available in a PowerPoint format. One of the main features of the site is the translation of all the documents on the stakeholder groups and lesson learnt innovations, into pedagogical slideshows. In addition, 25 videos of interviews undertaken during the European Agroforestry Conference in Montpellier in May 2016 have been added.

Task 6: Regional conferences and workshops
The sixth area of dissemination activity was the co-ordination of regional conferences and workshops. The AGFORWARD project has supported two European Agroforestry Conferences. The Second European Agroforestry Conference was held in Cottbus in 2014 and coincided with the first AGFORWARD General Assembly. The Conference led to the production of a book of proceedings extending to 278 pages (Palma et al. 2014). The Third European Agroforestry Conference was held in Montpellier in 2016 and coincided with the third AGFORWARD General Assembly. The book of Abstracts by Gosme et al (2016) extends to 466 pages and is available on the EURAF website. Both conferences led to published Proceedings, 4,100 and 1,800 downloads respectively to date (stats available at http://hdl.handle.net/10400.5/6764 and http://hdl.handle.net/10400.5/11690 respectively)

A final conference (Deliverable 9.31) was held in the European Parliament to discuss how to mainstream agroforestry in practice and through policy initiatives. The meeting entitled “1 + 1 = 3” was hosted by Paul Brannen MEP and examined how agroforestry can boost the revenues and resilience of Europe’s farmers (Figure 11). Paul Burgess, Co-ordinator of the AGFORWARD project, explained that agroforestry is an important European land use equivalent to almost 9% of the agricultural area. Agroforestry is dominated by silvopastoral systems (combining trees and shrubs with livestock) such as the dehesa and montado and wood pastures, but there were also distinct benefits from using more trees on arable farms. Paul explained that field research and farmer perceptions demonstrated that integrating trees with farming provides synergistic benefits including increased land use efficiency and income diversification, improved animal welfare, and increased biodiversity, soil conservation and carbon sequestration. He argued that these societal benefits warrant policies to encourage agroforestry, a reduction in the current administrative burdens, and there would be a net benefit if society compensated farmers for some of the additional on-farm labour and management costs. Fabien Balaguer described how agroforestry was being implemented in practice across a range of farm types in Southern France, with a particular focus on improving water and carbon management at catchment and landscape scales. Rosa Mosquera-Losada described how changes in policy within Europe could encourage the wider uptake of agroforestry. These include recognition of the wide range of agroforestry types (silvopastoral; silvoarable; hedgerow, windbreak and riparian buffer strips; forest farming and homegardens). She recommended that well-managed agroforestry on agricultural land should be fully eligible for Pillar I payments in the Common Agriculture Policy and that the wide and diverse range of agroforestry-related measures in Pillar II should be brought together.

After the presentations, an invited panel examined ways to mainstream agroforestry in Europe. The panel comprised Olivier De Schutter (Co-chair of the International Panel of Experts on Sustainable Food Systems), Valentin Opfermann (Policy Advisor on Agricultural and Environmental Research and Environmental Issues at COPA-COGECA), Patrick Worms (Senior Science Policy Adviser at the World Agroforestry Centre) and Frédéric Morand (Farmer and founder of Vert d’Iris International). There was a wide-ranging discussion and the responses to questions covered how current taxation systems (that often places high taxes on labour use) can penalise farm practices, such as agroforestry, where labour costs are often high. The panel also highlighted the very important role that agroforestry in Europe can and should play in terms of conserving soil and sequestering carbon. They recommended that agroforestry could and should play a role in enabling the Common Agricultural Policy to meet its environmental and economic objectives.

Further reading for dissemination activity
García de Jalón S et al. (2017). Forage-SAFE model: management and economics of wood pasture systems.. http://www.agforward.eu/index.php/en/forage-safe-a-tool-to-assess-the-management-and-economics-of-wood-pasture-systems.html
Gosme M et al. (Eds) (2016), 3rd European Agroforestry Conference Book of Abstracts. Montpellier, France 23-25 May 2016. ISBN: 978-2-87614-717-1, EAN: 9782876147171. http://hdl.handle.net/10400.5/11690
Graves A et al. (2017). Web-application of the Yield-SAFE and Farm-SAFE Model: Farm-SAFE_March17. Deliverable 9.27(9.3). March 2017. http://www.agforward.eu/index.php/en/web-application-of-yield-safe-and-farm-safe-models.html
Liagre F et al. (2016a). Expansion to 12 national agroforestry associations across Europe. Deliverable 9.28. 20 December 2015. 4 pp. http://www.agforward.eu/index.php/en/associations.html
Liagre F et al. (2016b). Training toolkit for farmers, technicians and students. Deliverable 9.29. 30 June 2016. 5 pp. http://train.agforward.eu/language/en/agforall/
Palma JHN et al. (2014). 2nd European Agroforestry Conference 04-06 June 2014 Cottbus-Germany Book of Abstracts. ISBN 978-972-97874-4-7. http://hdl.handle.net/10400.5/6764
Liagre F et al. (2018). Agroforestry folder for farmers and advisors. Deliverable 9.30. 26 January 2018. 45 pp. http://www.agforward.eu/index.php/en/best-practices-leaflets.html
List of Websites:
www.agforward.eu

Related information

Reported by

CRANFIELD UNIVERSITY
United Kingdom
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top