Extant Silvoarable Practices in Europe

Extant Silvoarable Practices in Europe Report

Milestone 2.1 of the SAFE European Research contract QLK5-CT-2001-00560 Silvoarable Agroforestry For Europe (SAFE)

Compiled by Pierluigi Paris September 2002

SAFE Project – Report: Extant Silvoarable Practices in Europe

2

ReportExtant Silvoarable Practices in Europe

Table of Contents

1. Introduction………………………………………………………………………….. pag. 3

2. Country Reports…………………………………………………………………….. .pag. 4

2.1. France…………………………………………………………………………… .pag. 4

2.2. Germany……………………………………………………………………… ....pag. 13

2.3. Greece…………………………………………………………………………….pag. 23

2.4. Italy……………………………………………………………………………….pag. 27

2.5. The Netherlands………………………………………………………………….pag. 33

2.6. Spain………………………………………………………………………………pag. 41

2.7. British Isles……………………………………………………………………….pag. 58

2.8. Switzerland……………… ……………………………………………………..pag. 64

3. Summary Conclusion…………………………………………………………………pag. 74

4. List of References……………………………………………………………………..pag. 76

5. List of Figures…………………………………………………………………………pag. 82

6. List of Tables……………………………………………………………………….....pag. 83

Appendix. Photos of Extant Silvoarable Practices in Europe

Report Coordinator: Pierluigi Paris Consiglio Nazionale delle Ricerche Istituto di Biologia Agro-ambientale e Forestale (CNR-IBAF) Via G. Marconi, 2 – 05010 Porano (TR), Italy Phone: +39 0763 374674; fax +39 0763 374330 E-mail: [email protected]

Porano (TR)-Italy, 24 Sept. 2002


3

1. Introduction

The Project Silvoarable Agroforestry For Europe (SAFE) deals with silvoarable agroforestry which comprises widely-spaced trees intercropped with arable crops. The SAFE Project has various objectives and one of these is to “reduce the uncertainties concerning the validity of silvoarable systems in Europe” (see Technical Annex). Silvoarable practices and systems as land use were widely used in Europe before industrialisation of agriculture took place in the last century in the entire continent. Agroforestry as integrated agricultural and forestry land use management had evolved in the past as a consequence of technological innovations and population dynamic. Key factor in this evolution was the need to maintain soil fertility. At the origin of agriculture, in shifting cultivation, relationship between agriculture and forests was very tight. In the stable agriculture, forest grazing and silvopastoral systems were very common and maintenance of soil fertility was also based on a strict connection among agriculture, husbandry and forests. Between forests and cultivated lands there was a continuous transport of fertility, mostly through forest grazing (by manure). Scattered trees in croplands and pastures were likely to improve soil characteristics underneath tree canopy. In the Middle Age, with the mass diffusion of crops rotation in the farming system, soil fertility of croplands had began to be less connected to forests/trees. This connection had a drastic reduction with the introduction of chemical fertilizers and the mechanization of agriculture. Nowadays, forestry, agriculture and husbandry are three separated activities with very few chemical and energetic relationship. Nowadays, information concerning the status of agroforestry, in general, and of silvoarable systems in particular, are quite poor in Europe. This is due to the fact that most of the research activities and Institutions interests have been mostly focussed on monocropping systems, which are currently the core of rural market economy. In order to fill this gap, within the 10 Workpackages (WP) of the SAFE Project, the WP9 “European Silvoarable Knowledge” has the objective to collect and analyse currently available and new information on European Silvoarable systems.

The first milestone of the WP2, as scheduled in the Technical Annex, is this Report entitled “Extent Silvoarable Practices in Europe”, which summarises the information on silvoarable agroforestry in most of the counrties involved in the SAFE Project. This Report collects contributions of 7 European countries (France, Greece, Germany, Italy, The Netherlands, British Isles –UK and EIRE, and Spain) plus a contribution on “European agroforestry historical information”. These single contributions are reported in the Chapter 2.


4

2. Country Reports

2.1 France

Author: Liagre F. Assemblée Permanente des Chambres d’Agriculture 9 Avenue Georges V - 75006 Paris, France Safe Participant 9: APCAE-mail : [email protected]: +33 4 99612640

Inventory methodology

The national census of silvoarable practices In 2000, a census of practices associating trees and crops on farmland, charged by the Environment Ministry, was realised both by the SOLAGRO Association and INRA of Montpellier. This study brought a lot of interesting data and contacts of key-persons on the theme of traditional practices and development projects in agroforestry. An inquest sheet has been sent to 500 organisms supposed to be implicated in agroforestry actions: Agricultural Chambers, regional and local administrations (ADASEA, DDAF, DRAF, …), associations and research and development organisms (CEMAGREF, ONF, CRPF…). More of 120 answers came back.

The agroforestry network For some years, an informal network was created in France, composed with agroforestry farmers, professional organisms (CRPF, CERPAM, SIME, Agricultural Chambers), public administration and research institutes (INRA and CEMAGREF). The main objective of this group was that the agroforestry systems became legally recognised, systems where tree is associated with annual or perennial crop (grazed or harvested). An agroforestry memorandum has been sent to the representatives in 1999. This work of lobbying permitted the rules reforms in 2001 of certain agricultural and forestry laws to take into account the technical peculiarities of the agroforestry systems. (circular DPEI/SPM/C2001-4008 from 8th March 2001, circular DERF/SDF/C2001-3010 from 7th May 2001 and circular DERF/SDF/C2001-3020 from 8th August 2001). An agri-environmental measure "Creation and management of agroforestry parcels" has been approved by the Star comity in November 2001. The reforms have been officially presented on 15th February 2002 in the Agriculture and Fishing Ministry. This network regroups today the majority of main actors working in agroforestry research development in France.


5

The inventory sheet used in France The initial sheet has been adapted to be close to the kind of tree/crop association we have in France:

Table 1. Inventory Sheet of Silvoarable Systems (version for France)


6

1. Name and date: – March 2002

2. Kind of system:

3. Geographical location and ownership

Village: Municipality: Region:

Private: Communal: Other:

Number of farmers involved :

About

Period of origin : Dynamic: Extension, Regression, Stagnant:

Type of plantation: Trial, Experiment, Demonstration, Commercial:

4. Physical characteristics

Area (ha): ha Altitude (m): m Bedrock: Annual rainfall (mm): Slope (%):% Soil type:

5. Trees

Main species: Other species: Scheme of plantation: m Density : trees/ha

Average height (m): m Average diameter (cm): cm Height of bole: m Age (years): years Distance tree/crop n=0: m Distance tree/crop mature:Spontaneous: Planted: Other: Used for fruit: For firewood: Other:

6. Crop (s)

Kind: Annual: Perennial: Duration: years In rotation: Irrigated: Annual yield: Manual ?

7. Management

Tillage (frequency, depth): cm Tree fertilisation (frequency, type): Weed control around trees (frequency, type): Tree pruning (frequency): Use of stubble (grazing? burning? other?): Other intervention:


7

8. Remarks

9. Contact

Local difficulties

Statistics and agroforestry systems There is no reliable statistical study on the agroforestry systems in France. The data concerning the crop associations are rarely mentioned. The only available data are deduced from some census mentioning pré-vergers, orchards or scattered trees in farmland. After crosschecking with historical and social studies, we have the possibility to estimate the areas. We have at our disposal: the agricultural census of 1929 (orchards and scattered trees), the TERUTI survey from 1981 (prés-verger) and 2 census from the National Forestry Inventory (scattered trees in farmland). The calculated estimations bring some information about the total area or about the number of trees but not in term of quality (sanitary status, productivity, age and rate of re-plantation).

The lack of professional recognition Agroforestry is usually bad known from the professional organisms and it’s difficult to get some technical data. Agroforestry is often consider as a simple forestry plantation on farmland and doesn’t benefit from a clear classification (land or fields). In fact, the main data sources are the research institutes (agriculture, forestry or social) or the arboricultural or environment associations.

An usual polemical subject At the moment of searching objective data about the importance and representativeness of the agroforestry systems, we come up against some emotive debate. The part in favour of agroforestry let themselves be tempted by exaggerating the old or present practices, increasing the positive arguments, while the opponents will see in the intercropping practices a system in terminal phase, some of them considering agroforestry like a medieval practice. Some technical fields contacted for this inventory have denied some still living practices. So for a given practice, you can get according to your data sources a total area changing between 100 and 10 000 ha! For each study of agroforestry system, the available data issued from the bibliography or interviews with technicians or administrative organisms, have to be tested on the field (interviews with farmers, archive research, statistical census and visual inventory) with the aim to estimate better the area and number of implicated farmers.


8

Main results

Historical of the practices The multipurposal tree was existing in main of cultivating systems. Tree was a symbol of wealth and technical progress. Improving fodder, fruit and timber production, it was easily integrated in a manual and subsistence agriculture. From the oldest systems, we can name : • Intercropping olive tree with cereals. The plantations are irregulars and poor dense, often below 50 trees by hectare. • The intercropping vine. The vine is associated with fruit trees. This system, named Joualle, is mainly composed of peach tree, walnut tree and olive tree. Tree can be inserted in the rows or it can be used as support for the vine (hautain). • The “plantades” are oak plantations of low density. Dated from XIX century, this public area responded to multiple purposes : fodder production in summer, grazing area for pig in autumn, and wood production. • The gardens with trees associated with vegetables. • The “pré-vergers” or grazing orchards : in the west region, the agricultural tree replaces the forest tree from 1600. A landscape of hedged farmland and spaced orchards take place and peak at the beginning of the twentieth century, with a strong dominance of apple tree, pear tree, plum and cherry trees. • The intercropping poplar orchards. The poplar grove with intercropping coming from the XVIIIth century is surely the only one example of traditional system associating forest trees and crops. • Cultivating in forest in the overseas regions (production of vanilla in the Reunion Island).

Since 1900, the agroforestry traditional systems decrease strongly in France. Various factors explain this evolution :

• The intensification of the agriculture and the input apparition reduced the value of agricultural sustainability of the tree. The scattered trees disturbed the machines operations. The machine injuries eliminated a big number of trees sensible to diseases. The timber commerce financed the machine purchase. The mechanisation of the orchards is often incompatible with the crop presence. • The labour diminution on the farms limits the maintenance or the development of manual system of production. The only grazed orchards or orchards with regular scheme of plantation succeed in adapting to the mechanical constraints. • The national Regrouping of Lands (“remembrement”) enlarged the parcels, eliminated a large part of hedges, the scattered trees and low-density orchards. The farmers recuperated the trees to sell it before interchanging the parcels.


9

• The agriculture orientation into specialised fields (crop or tree). The productivity of a crop association isn’t comparable to the productivity of a pure crops rotation. You have a few technicians or searchers to justify the whole productivity of the agroforestry parcels. The farms reduced the productions variability and the home consumption decreased. • The different agricultural rules concern the pure crops. The direct subsidies for crop from the CAP policy imposed a parcel specialization and encourage indirectly the elimination of the scattered trees. For example, these subsidies were excluded from the poplar or walnut groves with intercropping, or strongly reduced in presence of fruit trees. • The French Policy was separating clearly the forestry field from the agricultural field. The agroforestry plantations couldn’t allow the forestry subsidies. The conditions of contract weren’t available in agroforestry context.

Since the 80s, we assist to a slow “come back” of the tree following to the excess of deforestation. The risks according to an economically specialization, the diversification research and the environment consideration in the recent agricultural reintroduce the tree in the policy of sustainability development. INRA and CEMAGREF initiated research programs to understand the interactions tree/crop and the economical justifications of the existing systems. New projects associate precious species and annual or perennial crops, supported by important rules reforms elected in 2001.

Panorama of present practices The regular soil cultivating for the crop implantation is a specific parameter, which distinguishes intercropping system from the prés-vergers. Both systems have in common to present very low densities of trees, associated to an annual intercropping production. 42 inventory sheets present the variability of these associations in France.

The traditional practices

Nowadays, the main form of traditional agroforestry in France is the association between dual-purpose fruit tree (timber + fruits), planted at a very low density and associated to intercrop or to a grazing production (prés-vergers).

The “pré-vergers” The census counts 7 main systems of “pré-vergers” in France. The ‘pré-verger’ is a low-density (40-80 trees/ha) plantation of fruit-trees on grassland. The grass is the dominant production. The most common tree species are apple tree, pear tree but also walnut, and cherry plum or cherry tree. Certain fruit trees are dual-purpose trees, such as walnut, pear tree or apple tree that could be shaped with a bottom log with a timber end use. There are several variants of the ‘pré-verger’, according to tree density, tree distribution in the plot, and the plantation floor management Trees could be scattered on a part of the grassland, in a corner or along the edges of the grassland. Rows of fruits-trees along roads are very common in Alsace.


10

Since 1950, the development of intensive orchards has strongly competed for the regression of the pré-verger, in such a way that its cultivation area in 60 years decreased of 68% in France (148 000 ha, year 1999). Nowadays, the pré-verger isn't seen as a profitable branch of farming activities. Although that, the pre-verger represents still today sixty percent of the national production of cider's apple. Furthermore, ‘pré-verger’ has many very important traits to be implemented in relation to modern integrated development of rural areas, in terms of landscape enrichment for agricultural tourism, biodiversity and habitat protection, and production processes of quality food.

The intercrops The census named 14 main systems in France. In Dauphiné and Périgord, the nut producers cultivate between the walnut rows during the first 5 to 15 years, some of them cultivating sometimes between mature trees. We estimate the walnut orchards having been intercropped about 15 000 ha which 4 000 ha are still intercropping. The regular orchard appeared during the 50 and is compatible with the intercropping practice. More than an association aimed at waiting the nut production beginning, this system permit to develop the agronomical advantages of the association (soil operations, tree formation, inputs optimization). The associated poplar groves represent more of 6 000 ha in the regions of Nord Pas de Calais, Champagne Ardenne, Pays de Loire, Poitou Charente and Midi Pyrénées. The farmers cultivate essentially corn during the first 3 years (duration permitted before 2001 for the eligibility to the CAP subsidies). Through the association, the farmers look for a better cash flow (corn, saving money in the tree maintenance) and underline the best rate of growth and survival of the poplars in crop presence. The other associations, like olive tree / cereals or fruit tree / vine, draw a mosaic poorly representative (about hundred of hectares).

The moderns practices

The intercrops in forest trees plantations, is the main form of modern agroforestry, strongly inspired from the agroforestry poplar groves and pré-vergers. This inventory presents 23 examples de practices. We find again examples of associations with walnut trees for timber in the traditional regions of nut production of Dauphiné and Périgord, or in some poplar zones. Various agroforestry projects associating farmers are managed by INRA and CRPF from Montpellier or by the CEMAGREF of Clermont Ferrand. We can notice that these projects often give some ideas to the neighbourhood. For example the Department Gard will see in 2002, 3 news projects for one hundred of hectares. At least, we have to name some peculiar isolated cases, a part from any research programme. For example, the farm of Claude Jollet counts 56 ha of tree association with barley and sunflower. The walnut tree and wild cherry tree were planted 25 years ago and the farmer go on intercropping. An other example is the farm of François Gardey de Soos who associates robinia and honey locust of 10/15 years old with bread-making cereal to improve the protein quality of his cereals (20 ha).


11

Table 2. Synthesis of the agroforestry practices in France

System Regions EstimatedSurfaces (ha)

Number of farmers

Prés-vergersBretagne, Basse-Normandie, Haute-Normandie, Pays de Loire, Lorraine, Alsace, Rhône-Alpes

170.000 42 000

Dual purpose walnut orchard with intercropping

Rhône-Alpes, Midi-Pyrénées, Aquitaine

15 000 which 4000 ha in association

1 000

Olive trees with intercropping

Provence Alpes Côtes d’Azur, Languedoc Roussillon 3 000 1 500

Other dual purpose tree orchard with intercropping

Rhône-Alpes, Midi-Pyrénées, Normandie, Bourgogne 100 100

Fruit trees with intercropping

Languedoc-Roussillon, Rhône alpes, Guadeloupe, Martinique, Réunion

2 000 730

Forest trees Plantations in grazing system

Aquitaine, Limousin, Auvergne, Rhône-Alpes, Midi-Pyrénées, Nord, Languedoc-Roussillon

600 120

Forest walnut orchard with intercropping

Rhône Alpes, Midi-Pyrénées, Languedoc Roussillon, Poitou Charente

200 75

Poplar groves with intercropping

Nord Pas de Calais, Champagne Ardenne, Pays de Loire, Poitou Charente and Midi Pyrénées

6 300 1 000

Other forest trees plantations with intercropping

Languedoc-Roussillon, Poitou Charente, Midi-Pyrénées 150 10-20

TOTAL FRANCE 180 - 190 000 ha 46-47 000

Perspectives for the present practices The perspectives for the new millenary are encouraging for the development of agroforestry projects. The new policies take into account the presence of the agricultural trees and integrate them in the functioning of the farms. The news CAP orientations favour the consultation with the communities, often in favour to support this kind of patrimony, and underline the idea for a sustainable agriculture, respectful for the environment where the tree can play a role lost at the end of the last century.


12

The recent rules reforms Three fundamental reforms have been approved in 2001 : • The intercrop area is eligible to CAP subsidies (circular DPEI/SPM/C2001-4008 – 8th March 2001) • The agroforestry plantations are available to the forestry subsidies, keeping their agricultural status (circular DERF/SDF/C2001-3010 – 7th May 2001). • The tree area in an agroforestry plot is eligible to the European PCPR (subsidy to compensate the agricultural incomes lost) (circular DERF/SDF/C2001-3020 – 8th August 2001). These three reforms permit new options for the future, eliminating the main administrative reasons of the last ten years decline of the agroforestry systems.

A new tool : the CTE The CTE (Contrat Territorial d’Exploitation or Farm Territorial Contract) link the farmer to the whole society context. The measures of the contract reinforce the environmental practices and help the farm to diversify its incomes. The State and Europe finance them but also the Communities. In this context, a measure for creating and managing the agroforestry plots has been approved by the Star Comity from Brussels. This measure compensates the additional cost linked to the trees presence in middle of the crop area (additional time, tree protection, tree maintenance to facilitate the machine operations,…). A new measure is under consideration for creating and managing the crop production in a forestry context.

Other positive factors The following factors increase politic decision-makers awareness when they have to support or not an agroforestry project. They are factors that can play a role for the agroforestry development: • The research progress underline the profitability conditions of the agroforestry association and show the feasibility of such systems. • Blocking the carbon in the wood or in the soil, the agroforestry plots participate to the greenhouse effect control. • With the aim of high quality timber production, the agroforestry projects contribute to reduce the deficit linked to the timber importation from the tropical forest, which the resources decrease dangerously. • Recuperating in depth the nitrates that have escaped from the crop, the roots net developed by the trees roots have a role to play for the soil and water quality improvement. This factor doesn’t let insensible the water agencies, able to finance some experiences.


13

2.3 Germany

Authors:Schindler B1, Schumann F1, Burkart K 1 and Mayus M 2

1FINIS e.V., Freies Institut für interdisciplinäre Studien Alter Bahnhof, 19348 Baek, Germany Safe Subcontractor 2 (to WU): FINISEmail: [email protected]: +49 38782 40871

2 Crop and Weed Ecology Group, Wageningen University Haarweg 333, 6709 RZ Wageningen, The Netherlands Safe Partecipant 2: WU E-mail: [email protected]: + 33 (0) 4 99 61 26 40

Recent evolutions of silvoarable agroforestry practices and systems

Nowadays, agroforestry is not a common land use system in Germany, neither traditional nor modern systems. Most traditional agroforestry systems disappeared with the increasing industrialisation of agriculture. Already in the 19th/ early 20th century silvo-arable agroforestry with forest tree types had been abandoned. Silvo-arable streuobst1 persisted until the 1950s and in Eastern Germany several streuobstäcker have been remained (report Felix Herzog: Historical agroforestry experimentation in Europe). Streuobstwiesen (pastoral streuobst), however, are still numerous and widespread throughout Germany. In the 17th century, when fruit production became important, many fruit trees were planted along roadsides, and on favourable conditions such as river valleys and south oriented slopes with undercropping (Herzog, 1998). At such sides streuobtswiesen still exist in Alsace (roadsides), Rhine-Main area (river valleys) and Baden-Württemberg (Southern slopes). In 1998, the total area of streuobstwiesen was about 312,000 ha (the few area of streubstäcker not included), while fruit tree statistics of 1938 show that the area of streuobst was approximately 800,000 hectares (mostly silvoarable) in the 1930s (report Felix Herzog: Historical agroforestry experimentation in Europe).From the mid 50’s – till 1990, the streuobst area was heavily declined (Rösler, 1996), due to its replacement by intensive managed orchards with dwarf trees or other intensive farming practices. This process was stimulated by subsided eradication programmes and stronger quality norms for dessert fruits. Now, streuobst areas are highly appreciated, because of their ecological and socio-

1 Streuobst is defined as „tall trees of different types and varieties of fruit, belonging to different age groups, which are dispersed on cropland, meadows and pastures in a rather irregular pattern“ (translated from Lucke et al., 1992). A silvoarable (streuobstäcker) and a silvopastoral (streuobstwiesen) form of streuobst can be distinguished.


14

cultural benefits. Many support programmes exist by non-governmental organisations in addition to state conservation policies. Other traditional systems that existed in Germany, such as Haubergerswirtschaft (an agrisilviculture system) in Middle-Europe (Siegerland) (Kapp, 1984), are much less studied and documented than streuobstwiesen.No significant development of modern silvoarable practices are observed, but there are changes in land use management, which are favourable for silvoarable and agroforestry systems. In the last three decades, there has been a trend towards the association of trees and crops: Since the 1980s, there is a growing attention for the ecological and environmental problems of intensive and industrialised farms that led to alternative practices of land use management. Among other issues biodiversity, soil erosion and soil compaction gained importance. Windbreaks, “Knicks” (hedges of willow), individual trees and trees at field edges are no longer regarded as disturbing factors but rather as valuable ecological and aesthetic components in the landscape, which have to be preserved. Since then, the extant streuobst areas are no longer unreserved cleared but appreciated and maintained. Many studies, support programs and the activities of nature conservation organisations provided knowledge useful for alternative farming including silvoarable practices. Moreover, these projects enhanced the environmental recognition by farmers. For instance tree nurseries started to grow grass and cover crops between woody plants instead of weeding them. In the 1990s, the discussions on alternative energy sources included the idea of producing wood for fuel. For this aim fast growing trees as poplars, ashes and willows were selected. In Brandenburg, this idea was almost realised on a former mining area in combination with a revitalisation program. However, the estimated costs for fuel wood production could not compete with other methods of energy production. This stopped the project. But the discussions on wood for fuel continued on a more European and International level (Hall, 1997). Agroforestry research was performed during the early years of the 20th century (Kapp, 1984, report Felix Herzog: Historical agroforestry experimentation in Europe), while in the past decades till recently agroforestry research and education at German universities was restricted to tropical regions. At present, a number of universities perform studies on agroforestry for regions in Germany.

Difficulties to find information on the agroforestry situation in Germany

Agroforestry and even more silvoarable systems are rather unknown to the public, but also to farmers, foresters, public authorities. The public authorities did not perform any regular statistics of agroforestry in Germany, but occasionally the situation of streuobst was enquired and for the last time in 1998. Since agroforestry research concerning temperate regions is very limited, there are only very few scientific papers on agroforestry systems for the situation in Germany available. Moreover, the literature on history and on current agroforestry type issues is mainly found in local journals and magazine. This makes a literature research very time consuming. In the nearby future the internet might be used by local authorities more commonly, which could ease our search for extant


15

silvoarable practices or new agroforestry projects in Germany. During the course of this project and in particular through the silvoarable social experiment in Germany, further existing silvoarable practises might be discovered.

Methods used for the silvoarable inventory

As agroforestry does not appear in the official governmental statistics and publications are rare, the first step of the inventory of extant silvoarable practices was to find sources of information. The following sources were used: 1. Local governmental institutions for agriculture and forestry 2. An NGO office working a.o. on the distribution of agroforestry knowledge 3. University departments for agriculture and forestry 4. Nature conservation associations 5. University library of Bonn 6. Internet (see annex) 7. Farmers in Northern (personal communication Schindler, B. and Schumann, F.) and Southern (Keyser, B.) Germany. The internet was an appropriate resource to obtain information on publication and to check for existing agroforestry projects, networks etc.. Another important source of information was an NGO office, which is setting up an agroforestry network for Germany. Through this office, a farm with a system similar to silvoarable systems and a farm with a silvoarable field that was abandoned in 1990 has been found in the south of Germany. In the North of Germany, farmers were found by contacting local public authorities like farmer organisations, environmental associations, and regional project associations. The farmers have been interviewed by phone using the inventory sheet developed by project partner 10 AUTH (Papanastasis, V.). The University of Kiel, Bonn, Berlin, Eberswalde, Halle and Hohenheim were investigated for running projects using the internet.

Main Results

The inventory of extant traditional and novel silvoarable and agroforestry systems, revealed that agroforestry plays no role in Germany at present. The farmers and forestry associations know of the existence of silvoarabel and silvopastoral systems in former times but have no information on extant tradtiotional agroforestry systems, with the exception of knicks and streuobst. The pastoral form of streuobst, streuobstwiesen is a widespread and still known traditional agroforestry system. Governmental institutions collected statistical data, for streuobst, in 1934, 1938 and more recently, after the industrialisation of agriculture in 1951 and 1965. In 1998, an enquiry of the state of streuobstwiesen was performed; the area of streuobstwiesen comprises 312,000 ha. Whereas, the silvoarable form streuobstacker is limited to a couple of very small fields (<0.5 ha). Therefore, farmers managing streuobstäcker at present could not be found, only a farmer in Bavaria


16

who kept streuobstäcker till 1990. His system is described in the annex (Farm Oppermann, inventory sheet 1).

Although agroforestry is not a recognized land use system in Germany, at least three farmers shifted their management towards practices that are similar to those of silvoarable systems. Two of these farms were found in the north and one farm in the south of the country. In all three cases, the farm is an ecological or organic managed system including woody plants. The motivation of the farmers was to reduce environmental growth constraints, by the combination of trees and crops. Below, a brief overview of their silvoarable system is given.

Literature overview on streuobst

The overview of the streuobstacker is completed by literature data on streuobstwiesen, since it might help the understanding of the perspectives for silvoarable systems in Germany.

Streuobstäcker In particular in the region Unterfranken (Bayern), streuobstäcker was a typical traditional land use system and is still known by the population. However, the remaining streuobstäcker almost completely disappeared with the introduction of the set-aside programmes (1980s?). The grants for taking Streuobstäcker out of production exceeded the monetary value of the fruit harvest. Many farmers transformed the streuobstäcker into streuobwiesen, used as set-aside area. Several small plots of streuobstäcker (as landscape elements) can still be observed in the recreation region of Rheinland-Pfalz close to the border with Alsace. Small streuobstäcker also exist in Eastern Germany (e.g. Magdeburger Börde): Undercropping of cherry trees (Prunus avium) mainly with turnip (Beta vulgaris), but also with alfalfa (Medicago sativa), potatoes (Solanum tuberosum) and oat (Avena sativa) is still practised by a few farmers up to today for household consumption (cited from the report of Felix Herzog: Historical agroforestry experimentation in Europe). The authors could not find indications that silvoarabel systems exist or existed in other regions of Germany. In general, streuobstäcker were small fields (< 1 ha) with two rows of fruit trees intercropped till close to the tree trunk. Fruit trees were also planted on roadsides. The tree branches are kept rather low, to facilitate fruit harvest. Nowadays, this causes some technical problems for the crop management. In the region Unterfranken, the major intercrops were potatoes and vegetables. Some full time farmers grow cereals between the tree rows. 12 years ago, there were streuobstäcker intercropped with asparagus in the Magdeburger Börde. Those systems also disappeared due to the grants for taking fruit trees out of production. Streuobstäcker, is an intensive form of land use and, hence, was performed in region with a high value of arable land, e.g. Unterfranken and Magdeburger Börde. In Unterfranken, where arable land is scare and sol prices high, streuobst was a favourable subsidiary holding. In the Magdeburger Börde, during the DDR period there was not much private land as a result of expropriation. The few available land (generally 0.5 ha per farmer) was used intensively for instance by (re-introducing) silvoarable streuobst.


17

In Unterfranken, the agricultural authorities remember and appreciate streuobstäcker for their landscape character. Nevertheless, streuobstäcker have not been included in the statistical inventory of streuobst trees.

Streuobstwiesen The following information are based on Zander (1999) and Baumhof-Pretzingen & Langer (1997). The traditional “streuobstwiesen” are characterised by an extensive management of high stem apple, pear, cherry, plums and other fruit trees, which were planted for additional land use of cropland, meadows and pastures. On average, tree density varies between 70 to 100 trees/ha, but there are large regional differences. Most of the fruits are not harvested for direct consumption but are collected as for the production of fruit-vine, dried fruits or spirits. Trees are pruned every two or three years and the pastures are cut or grazed once or twice a year. Since the 1950s the role of streuobst is steadily decreasing, since local fruit production has been taken over by intensively managed orchards. The production of high outward quality dessert fruit, according to the current regulations, cannot be achieved on streuobst. However, streuobstwiesen still play a role in the production of the precursors for the fruit juice industry, although its market share is decreasing as a result of growing competition with Middle and East European countries. In the period from the end of the 50s until the beginning of the 70s of the last century the production of fruits from streuobst was considered to be an obstacle for the contested fruit market and the clearing of these fruit plantations was rewarded. Besides other factors, this bonus system reduced the total area of streuobst with 70-75% (Rösler, 1996) from the mid of the 50s till 1990. The reduction of streuobst has slowed down since the 80s of the last century. The most significant reason was that besides fruit production other positive aspects of streuobst were recognised and became a strong denotation. Most of all, the aesthetical landscape value that streuobst have with their high stem trees and their different colours of blossom and under-storey pastures, its high level of biodiversity, and the existing diversity of fruit cultivars that have become rare nowadays, have been recognized as valuable qualities. Due to the recognition of its environmental benefits, many initiatives have been undertaken to conserve streuobst and have been promoted with public financial support. In 1989, an inventory has been conducted with respect to the situation of streuobstwiesen in Germany. The federal ministries had to collect information on the dimensions of the extant streuobstwiesen and about the type and amount of promoting activities. According to this inventory a total aerial of 312,000 hectare was registered, but this number has to be taken with care, since it is not clear if fruit trees planted along roads have been included into the inventory and thus counted as streuobstwiesen. A large part of the total area of streuobstwiesen is located in the state of Baden-Württemberg (160,000 – 190,000 ha) and in the state of Bavaria. Traditionally, fruit production is most exposed in the south near the Alpes and in the so-called Rhine-Main area (the triangle made up by the cities: Mainz, Frankfurt and Karlsruhe). In these regions you will find most of Germany’s extant streuobstwiesen. There is no reliable knowledge about the exploitation of these streuobstwiesen. The last statistical inventory on the number of fruit trees in Bavaria has been conducted in 1965. From our research it became clear that the states of


18

Baden-Württemberg and Bavaria are the only states in Germany that consider streuobstwiesen. as an optional source of income. In most of the federal states, public programmes for the promotion of streuobst exist. Subsidies cover the planting, the management of the trees and understorey crop. The estimated amount of applied subsidies was about 11 Mio. Euro for the year 1998. Baden-Württemberg used with 6,7 Mio the major part.

Streuobstwiesen in Baden-Württemberg The following information and data are based on the booklet “Streuobst“ published by the foundation Naturschutzfonds, Federal Ministery of Rural Area, Baden-Württemberg in 1997. In 1990, an inventory of streuobst trees gave 11.4 Mio trees of streuobst in Baden-Württemberg. The inventory was performed by the Federal Ministery for Food, Agriculture and Forestry. In representative regions, trees on streuobst were counted per fruit type, apple (Malus domestica L.), pear (Pyrus communis L.), plum (Prunus domestica L.) and mazard cherry (Prunus avium L.)). Conform the average density of 60-70 trees per hectare, this agrees with an areas of 160.000 – 190.000 ha streuobst. Compared to 1965, this is a reduction of about 37% trees. In 1934, there were 40 Mio. fruit treesin Baden-Württemberg (Ulrich, 1987, booklet “Streuobst”). Thus, in 56 years there was a reduction by about 70 % in streuobst. Two main reasons can be given for this heavy decline: As a consequence of new regulations for fruit plantations in Baden-Württemberg and new tax laws of the EU, 15.700 ha streuobstwiesen had been cleared and were transformed into intensively managed orchard in the period from 1957 - 1974. The stronger quality norms excluded a large part of the harvest from streuobst from the market of dessert fruit and, hence, reduced teh Streuobst area. The area of streuobst had been influenced by the EEC regulations concerning the quality norms for dessert apples (EEC regulation 1641/71). The stronger quality norms excluded a large part of the harvest from streuobst from the market of dessert fruit. Further reasons are: overbuilding of the landscape, changes in agricultural practises, decline in cider-pressing and distillery and the transformation of streuobstwiesen into areas for recreation. Moreover, epidemic diseases caused lost of streuobst.The ecological importance of streuobst was recognised in Baden-Württemberg at the end of the seventies. Therefore, this system was included in the mapping of endangered habitats (gefährdete Lebensräume), which was conducted during the period of 1977 – 1989. They counted a total of 4,156 streuobstwiesen with 3,803 fields smaller than 50 ha. The total area was 74,508 ha. In the same period, new regulations became effective, which defined many of registeredstreuobstwiesen as nature reserve, landscape conservation area, natural monument or as protected grassland. Baden-Württemberg developed also a list of grants for streuobst plantation, maintenance and sale of products in the context of Markentlastungs- und Kulturlandschaftsausgleich (MEKA; compensation for market discharge and cultural landscape). In 1995, a total subsidy amount of 13 Mio. DM was used for streuobst trees within the MEKA programme.


19

In addition to the efforts of Baden-Württemberg, many public initiatives were set up for the conservation of streuobst. Three examples are described here.

Streuobst in the district Würzburg (South Germany) The following information is obtained by a study from the Bayrische Landesanstalt für Weinbau und Gartenbau (LWG: institute for viniculture and horticulture) (Degenbeck, 2000a and b; Degenbeck, 2001). The district Würzburg, characterised by a mean annual temperature of 6°C, an average yearly precipitation of 600 mm and many sites with a south facing slope, is favourable for vineyards and orchards. The major income of the region comes from the vineyards, which are traditionally linked with orchard fruit production. The streuobst landscapes of this region are the most important ones in Bayern. Streuobst had been largely replaced by arable production, beacuse of the good soils. With the decline in streuobst, the knowledge on species in fruit trees decreased too. From originally 1000 fruit tree species only 5 species are still known by the local fruit and horticulture association. In the region are about 30.000 fruit trees. The fruit trees comprise 40 % apple , 27 % plums 23 % mazard cherry and 10 % pears and walnut. The age structure is as followed: 47 % belongs to the age group of full production, 30% belongs to the age group with increasing productivity and 23% are in the age group with decreasing production. Within a streuobst field the age group differs a lot. In general fruit trees are younger in the rural regions compared to those in the neighbourhood of cities. In the rural regions available public subsidies seemed to be used more frequently. The use of the fruit varies a lot. The major part of the harvest fruit is pressed. Traditionally the production of fruit brandy has played an important role. In some regions of the rural area of Würzburg, 60-80 % of the plums are still used for distillery. However, the production of vine plays almost no role anymore. Most of the apples are used as dessert fruit, for the other fruit types this utilisation can be between 1 and 80%. Not all streuobstwiesen are harvested and approximately 15 – 20 % of the fruits are lost due to this. The major part of the streuobst is private property, only 6.8 % belongs to the rural commune. The maintenance of the streuobst areas is generally poor and dependent on their economic profitability. A revitalisation of streuobstwiesen is economically only interesting with public support, since the financial ouput is not sufficient for weeding and pruning. The weeding would be profitable for apple and pear if the major part of the fruits could be sold as dessert fruit. The revitalisation of walnut and cherry is most profitable considering the labour costs for pruning and the yield of wood. Mechanisation of streuobst systems can be economically interesting when the fruit-tree harvesters belong to a co-operative.

Inventory of novel silvoarable agroforestry systems


20

Mixed farm Melchhof (Brandenburg, North Germany) Melchhof, a mixed ecological farm (member of Demeter), was founded in 1991 in Brandenburg. The farm comprises 11.5 ha with 3.5 ha vegetables, 1 ha roses (Rosa villosa) and 3 ha orchard. Conform the regulations of ecological farming; no chemical or synthetically fertilisers are used. Every year, the different vegetable species are rotated and cover crops are sown as intercrop. Due to a low annual rainfall (mean: 500 mm) and its bad distribution (6-8 week drought period between June - July), water stress is a crucial growth constraint. Therefore, in 1993, the Melchhof farm joined an experiment of the Humboldt University (Berlin) on the effect of hedges on water conservation, in 1993. The hedges are supposed to reduce wind speed and thus soil evaporation and plant transpiration. Hedges of rose shrubs (Rosa canina) were planted at a 18 m distance. The distance between shrubs in a row is 0.5 m. During the first four years the hedges had to be weeded for their establishment. Fruits are harvested by cutting the branches which carry fruits. Through this harvesting method the shrubs are pruned at the same time. The farmer of Melchhof noticed two disadvantages of the hedges. First, the wind speed reduction by the hedges favoured the conditions for the pest “carrot fly” which became an increasing problem. Secondly, the weed couch-grass was spread from the hedges into the vegetable field.

Klosterhof GbR (Niederbayern, South Germany) The Klosterhof GbR is an organic farm founded in Bayern, in 1998. Klosterhof has the intention to become a model farm in terms of the optimisation of ecological and agronomic system qualities. Major crops on the farm are perennials as Rosa canina (roses), goat willow (Salix caprea L.) andhazel (corylus avellana L.), which are used for seed and plant propagation. The farm grows also special and permanent crops: wild fruit trees- as service tree (Sorbus domestica L), midland hawthorn (Crataegus laevigata), walnut (Juglans regia), wild cherry (Prunus avium), burgsd.(Pyrus pyraster), purple-leaved barberry (Berberis vulgaris) and mountain ash (Sorbus aucuparia).The fruits are used for the production of vine, dry fruit and jam. The farm has an agricultural area of 13 ha with 9 ha for seed production of stock roses (Rosa sp.), 2 ha for seed production for autochthonous woody species, i.e. 1.5 ha Hazels (Corylus sp.) and 0,5 ha Violet Willow (Salix daphnoides Hybr.) for propagation. In 1998, shrubs and trees were planted. The shrubs were planted in uniform rows (150 plants/ ha) and 70 trees were planted with alternately species in a row. In 1999, an intercrop was sown between the woody plantations to obtain an income during the first years when the woody plants were nor yet producing fruits. After soil preparation, cereals are sown with an undercover crop composed of wild vegetables and wild flowers, including herbs. The cereal is harvested in the first year and the undercover crop in the second year. The strips are grassed by Highland cattle after the harvest of the undercover crop. This cycle is repeated every 7 years. The agroforestry system of the Klosterhof produces an gross margin of 5000 Euro per hectare.


21

Nursery Lorenz von Ehren (Hamburg, North Germany) At the beginning of the 80s, when environmental problems has been recognised, many farms shifted towards ecological management practises. So did the nursery Lorenz von Ehren, which was founded in 1865. The 400 ha large farm is specialised in raising and sale of avenue, individual and trained trees. Four year old trees are bought and kept for at least four years before they are sold for park, garden and street plantations. All type of trees adapted to the local climate are planted. The distance between trees depends on tree size, and varies between 2,5 x 2,5 up to 8 x 8 m. In earlier years the tree plantation was kept free of any understorey plants, thus, the soil between trees was bare. This planting strategy had several disadvantages in a climate with strong winds and high annual rainfall amount. On the bare soil wind and water erosion occurred frequently and the upper tree roots became often uncovered with the risk to dry out. After rainfalls of high intensity the soil slipped, such, that the use of heavy machines led to soil compaction. Moreover, the reflection of the radiation on the soil was high, which is unfavourable for some woody species. Due to these bad experiences with a bare soil, the farm changed its farming system. Prior to tree planting, sub soiling is performed up to a depth of 70-80 cm and cover crops are sown between tree rows. Plant species used as cover crops are fast growing types such as mustard and buckwheat, which are characterised by good and fast establishment, fast growth, deep rooting system, and nitrogen fixation. At the end of the growing season the cover crop mulch is incorporated into the soil. In the second year, grass is sown and cut 4 – 5 times a year. The mulch remains on the soil. For the intercrop, no fertilisers and herbicides are applied. However, herbicides are applied on the bare soil within a tree row. Trees are fertilised according to soil samples and after long drought spells, the trees are irrigated individually. The trees are pruned every year, from the first pruning onwards. The nursery Lorenz van Ehren has made good experiences with intercropping the trees, i.e. erosion control, improved soil quality, reduction of chemical input for pests and weeds. The farmer got the impression that the frequency of disease attacks has been reduced, but this has still to be quantified.

Perspectives for extant traditional and novel silvoarable practices

In Germany, there are prospects for agroforestry systems. Major reasons are: • Silvoarable systems facilitate to meet the interest of the CAP regulations. • There is much land available as a result of set aside programmes. • Silvoarables systems can combine multiple advantages for the environment. • Increasing recognition by the public of the need that biodiversity, landscape should be considered. The market of streuobst juice indicate that there is a potential for market for product higher price additional value. There is trend towards agri-environmental payments. • At present agriculture is heavily financially supported by grants. For the nearby future, it is expected that the subsidies will be reduced. In this case, agroforestry becomes very attractive, due to risk diversification. Moreover, silvoarable systems increase the possibilities to adapt to the market situation by its flexibility to move the farming activities between agricultural and forestry activities. (In France, a novel agroforestry policy scheme make future easier regulations possible.


22

Currently, the establishment of agroforestry and silvoarable systems is hampered by: 1) The poor system knowledge 2) The fact that it almost disappeared and thus no example farms exists for demonstration purpose 3) The land use regulations, which are completely separated for agriculture and forestry 4) The poor economic situation of farmers hampers investments in long term returns

Solutions for these barriers are in sight: 1) There is a slow but increase in available information on agroforestry (projects) by universities and other institutions. A proposal for a European agroforestry project will be submitted for the 6th KE. 2) A silvoarable field will be set-up in the north of Germany by the SAFE project. 3) Recent changes in land use regulations in France that include agroforestry situations will path the way for changes in other European countries 4) Special grants for agroforestry (Germany, EU) could take away this barrier From discussions with farmers and foresters it appears that the step towards silvoarable systems is easier from a forestry, a farm combining arable and forest stands, tree nurseries or large farms with set-aside land, than from a pure and/or small arable farm.


23

2.2 Greece

Authors:Papanastasis V P, Ispikoudis I, Mantzanas K and Tsatsiadis E Laboratory of Rangeland Ecology, School of Forestry and Natural Environment, Aristotle University of Thessaloniki, 54124 Greece Safe Participant 10: AUTHE-mail: [email protected]: +30 31 998933

Recent evolutions of silvoarables Silvoarable agroforestry, that comprises widely-spaced trees intercropped with arable crops, is an old practice in Greece. In several areas, traditional silvoarable systems occur with different combinations of trees and crops depending on the ecological zone and the agricultural practices. The first inventory of agroforestry systems including the silvoarable ones was conducted in 1987 (Schultz et al. 1987). This inventory covered several areas of Greece with different climatic and ecological conditions. Some of the recorded systems are the following: a) Valonia oak with cereals that cover large areas in south and west Greece. b) Olive trees with various crops such as wheat, barley, corn, alfalfa, grape vines and vegetable crops, i.e. potatoes, melons, tomatoes, beans, onions or fava beans. The total area, that the olive trees cover are 650,000 ha and the number of trees are 120,000,000. c) Carob trees with broadbeans and other herbaceous crops, also grazed by animals. d) Walnut trees with several cereals, tobacco, alfalfa or grape vines. e) Mulberry trees with corn, alfalfa or beans. f) Figs with cereals located in Crete and eastern Aegean islands. g) Oak trees (Quercus pubescens, Q. sessiliflora, Q. cerris and Q. macedonica) with cereals, tobacco, sunflower, alfalfa or vineyards covering large areas in central and northern Greece. h) Poplar trees with vegetable crops and corn or alfalfa. i) Wild pear with cereals, tobacco, vineyards or vegetable crops. Most of these systems still live but the inventory has not been competed yet. Of those recorded so far, the still living are: 1. Olive trees with cereals, 2. Poplars with cereals or fodder species or vegetables, 3. Walnuts with cereals or vines, 4. Orchard trees with cereals, 5. Oak trees with cereals.

Local difficulties to find reliable data on silvoarables 1. The small area of silvoarable plots (). There are more than one tree species around the plots, 2. Difficult to find a clear tree spacing in the plots


24

3. Different crops in a small area with different management, 4. Poplar trees are always planted around the plots, 5. Short periods of rent. These prevent the farmers from applying management rules that need larger time periods.

Methods used for the silvoarable inventory Measurements so far were made in the municipality of Askio. Other areas are also investigated but the results will be reported in July. An inventory sheet for every plot was completed. Measurements on trees (tree height and diameter on breast height) were made in every plot. The other information that included in the inventory sheets was taken from the farmers. The silvoarable systems were grouped according to the tree species and the crop species.

Main results There were found 21 different systems and 15 sub-systems. The main ones are shown in Table 3 for the systems and 2 for the sub-systems. Some examples of systems are shown on Photos 1-5 (see Appendix).

Perspectives for the still living silvoarable practices Traditional silvoarable systems are gradually being abandoned in the marginal areas while in the more productive areas they are replaced by monocultures. There is no motivation for the farmers to maintain them who perceive trees as an obstacle to the mechanization of their crops. Besides, there is no regional or national policy to improve and make them viable. More analysis together with suggestions will be provided when the inventory is completed.

Tree species Crop species

Walnut Oak Poplar Pyrus spp. Salix spp. Orchards

Cereals 1 a1 2 a, b2 1 b 2 a, b

Medicago spp. 1 a 1 b 1 b 1 a, b

Vines 1 a

Vegetables 1 a 1 b 2 a, b

Table 3 Main silvoarable systems at municipality of Askio, Greece. 1 a indicates that trees are planted inside the plot, 2 b indicates that trees are planted around the plot


25

Crop species Main tree species

Other tree species

Cereals Medicago spp. Vines Vegetables

Walnut

Oak 1 a1, b2

Poplar 1 b 1 b

Pyrus spp.

Salix spp.

Walnut

Orchards 1 b 1 a 1 b 2 b

Walnut 1 a, b

Oak

Poplar

Pyrus spp. 1 b

Salix spp.

Oak

Orchards 1 a, b

Walnut

Oak

Poplar

Pyrus spp.

Salix spp.

Poplar

Orchards 1 b

Tab. 1 . Main silvoarable sub-systems at municipality of Askio, Greece. 1 a indicates that trees are planted inside the plot, 2 b indicates that trees are planted around the plot


26

Figure 1. Silvoarable plots in four plot area classes at the region of municipality of Askio.

0

5

1 0

1 5

2 0

2 5

0 .0 - 0 .3 h a 0 .4 -0 .7 h a 0 .8 - 1 .0 h a > 1 .0 h aP lo t a r e a

Nr o

f plo

ts


27

2.4 Italy

Authors:Pisanelli 1 A and Paris P Consiglio Nazionale delle Ricerche Istituto di Agro-Biologia Ambientale e Forestale (CNR, IBAF) Via G. Marconi 2, 05010 Porano (Terni), Italy Safe Participant 6: [email protected]: +39 0763 374674

Recent evolution of silvoarable agroforestry practices and systems In Italy agroforestry practices were widely common land use systems until the industrialising of agriculture that began in the 1950s. The multifunctional characteristic of traditional land use systems allowed farmers and landowners to obtain several kinds of product (fuelwood, timber, crop yields, fodder, fruits, etc.) in addition to environmental benefits (conservation of biodiversity, control of soil erosion, replenishment of soil fertility, etc.). Many tree species such as poplars (Populus spp.), maples (Acer spp.), elms (Ulmus spp.), ashes (Fraxinus spp.), walnut (Juglans regia L.) and wild cherry (Prunus avium L.) were frequent in agricultural fields as isolated trees, in boundaries or edges and their contribution to timber production was estimated to be more than that from forests (12,000,000 m3 against 10,500,000 m3 before the II World War- Mezzalira, 2001).

The spread of mechanisation and the increasing availability of external inputs (pesticides, fertilisers, etc.) promoted a dramatic reduction in agro-ecosystem diversity (Figure 2). Monoculture progressively replaced intercropping systems and the process caused the development of ecological and economic problems. This process has been particularly evident in the most favourable areas where fertile soils and flat topography allowed the widespread adoption of intensive agricultural practices. In contrast, in marginal areas traditional systems have been often preserved because of the difficulty of practising modern agriculture. For example Bertolotto et al. (1995) found that in Umbria Region (central Italy, hilly prevalent morphology, Mediterranean climate), traditional agroforestry practices are mainly located where farmers are older and farm area is smaller than the regional average.

Because of the ecological and economic problems associated with modern agriculture, agroforestry has recently acquired a strong interest from research institutions. Also in industrialised countries, as well as in developing countries, several studies and methodologies have been developed for demonstrating the ecological and economic sustainability of agroforestry practices in comparison with intensive agriculture.


28

During the last decade, several inventories of agroforestry practices and systems were carried out in different agricultural areas of Italy. In Umbria (Bertolotto et al., 1995) olive trees (Olea europea L.) and grape vines (Vitis viniferta L.) were found to be the main woody components of traditional agroforestry systems. The olive tree is often intercropped with cereals (wheat-Triticum spp., barley- Hordeum vulgare L.), fodder crops (lucerne-Medicago sativa L, clovers-Trifulium spp.), pasture (for sheep) and various horticultural crops. The arrangement of olive trees on the fields can be scattered randomly or in rows, with tree spacing varying between 5 to 10 m depending on the intercrop.

The practices with grape vines are less frequent because they have been almost completely replaced by specialised vineyards. Traditionally vines were planted alternately in the same rows with other woody species such as maples and willows and intercropped between the rows with cereals.

In Sicily Region (Cullotta et al., 1999), agroforestry practices involve typical Mediterranean species that can grow in difficult environmental conditions (e.g. prolonged summer drought, poor soils). The main woody components of agroforestry systems are Ceratonia siliqua trees (estimated area about 20,000 ha), almond trees (Prunu dulcis L.) (estimated area about 18,000 ha), olive trees (estimated area about 79,000 ha) and grape vines (estimated area about 153,000 ha). These species are often mixed and intercropped with cereals (wheat) or legumes.

In Northern Italy, favourable ecological (fertile and deep soils, flat topography, etc.) and economic (large farms, availability of mechanisation, etc.) conditions reduced biological diversity. A survey conducted in a Po plain estimated that from 1980 and 1997, rows and edges in agricultural lands reduced by 29% and 44% respectively (Groppali, 1999). Mezzalira (1999) estimated that in Italy during the last 40 years, 70-90 % of the edges were destroyed and their present extent can be considered residual and insignificant.

Other common agroforestry practices in Northern Italy involved the intercropping of arable crops with poplars. This system was particularly common in Piedmont where poplar plantations (mostly hybrid poplars) are very traditional. According with Lapietra et al. (1991) about 4.1% of the total poplar area of the north of Italy was intercropped and maize (Zea mays L.), soya bean (Glycine max(L.) Merr.) and wheat were the most common crops. For plantations in the first two years of the cultivation cycle (when tree canopy dimensions allow the sunlight to reach the soil) the proportion of intercropping reaches 15% of the total poplar area.

Method used for the silvoarable inventory Agroforestry practices are not recognised as land use systems in official statistics, consequently data on the effective area of silvoarable agroforestry are not available.

Because of the lack of statistical data on silvoarable systems, the inventory was mainly carried out by means of farm surveys conducted in representative agricultural areas. Sample areas in which


29

silvoarable systems are supposed to be representative land use systems have been selected in Central (Umbria and Latium), Southern (Basilicata and Campania) and Northern (Piedmont) Italy. In each area contacts have been established with local research institutions and with farmers’ organisations in order to find representative samples of silvoarable systems and to facilitate data collection.

On-farm surveys were conducted using the inventory sheet specifically created by the project partners. Additional information was collected by open discussion with the farmers directly involved in the management of the systems.

Main results The surveyed silvoarable systems are shown in Table 4. Oak, poplar, olive tree and walnut represent the most common woody components of the systems in the representative areas. The associated arable crops are cereals, legumes and vegetables.

Silvoarable systems with oaks (in particular Quercus pubescens and Quercus cerris) are usually present in marginal rural areas (Photos 6-7). Trees are scattered in the fields and the density ranges between 7 and 250 trees/ha. Trees are usually more than 100 years old and they can be considered the relicts of traditional silvoarable systems in which woody species were important in order to ensure timber and fuelwood production and protection for animals in silvopastoral systems. The need for modern agriculture caused a progressive reduction of tree components. At present trees are sometimes safeguarded for protecting soil and improving the rural landscape. Wheat in rotation with clovers represents the most common crops and their annual yields are estimated to be 3-5 tonne/ha and 2 tonne/ha respectively. The intercropping of oats (Avena spp.) between oaks is less frequent. No information are available on the role of oak trees in improving soil fertility and crop yields as well as other environmental benefits (biodiversity, soil nutrients and water cycle). Similar systems are present in Sardinia Island where the tree component is cork oak (Quercus suber L.) a large proportion of cork oak groves being either intercropped with cereals or managed as silvopastoral systems (Photo 8). The economic and environmental roles of these systems are not controlled under specific legislation and they are considered as forests.

The survey identified several kinds of agroforestry systems in which olive trees are intercropped with cereals (wheat, barley), fodder crops (lucerne, clovers), pasture (for sheep), horticultural crops and vine. These systems are particularly common in Central Italy (Umbria and Latium) where olive tree cultivation is mainly for oil production (Photos 9-10). The systems usually cover small areas (less than 1 ha) and olive trees are scattered in fields or arranged in rows with tree spacing ranging between 5 and 10 m. Although olive tree intercropping systems are traditional and are mainly located in marginal lands, they are of economic value to farmers with limited land. Cereals (wheat) and legumes (clover and vetches- Vicia spp.) are cultivated in rotation and their annual yields are estimated to be 3 tonne/ha and 2 tonne/ha respectively.


30

Silvoarable systems with walnut represent one of the most interesting cultural systems that have been inventoried (Photo 11-12). In Campania, walnut is traditionally cultivated both for timber and fruit and it is usually intercropped with vegetables. The association between walnut and hazelnut (Corylus avellana L.) is also common. The practice is adopted in fertile volcanic soils on plains so that tree growth is fast (annual diameter increment is estimated to be more than 2 cm). Trees are 10 years old and are arranged in rows and the low density (less than 50 trees/ha) allows the crop cultivation under the canopy.

Poplar is typically cultivated in the Northern Italian regions on flat fertile soils (Padana plain). Traditionally the intercropping of young trees with maize, soybean and wheat was common. At present, these silvoarable systems are disappearing as a consequence of the application of European Union Regulations e.g. because European grants for tree planting don’t allow intercropping. Maize and soya bean productivity are estimated to be about 7.5 tonne/ha and 2.5 tonne/ha respectively. A poplar farm of 100 ha with intercropping with mais and soybean for half of the poplar cultivation cycle 10 years) were inventoried in Pedimont Region, northern Italy (Photo 13)

Perspectives for extant historical/traditional silvoarable practicesSeveral kinds of traditional silvoarable systems that involve typical agroforestry species are still present in rural areas of Italy.

Some systems, such as those involving oaks and olive trees, are mainly located in areas where environmental and economic conditions (small farm area, steeply sloping land, etc.) constrained the spread of modern agriculture in the 20th century. Agricultural products (crop yields and olive production) represent the main output of the systems. In this context the systems can be considered relicts of traditional land use systems and in some cases are preserved for protecting soil and improving rural landscape.

The systems with poplar and walnut appear more economically profitable because of the favourable agro-ecological conditions (fertile soils, flat land, large farms). The short or medium length tree cycle allows timber production to be an important output of the systems as well as the crop yields.

At present, an impact assessment of agricultural European policy (PAC) on land use systems appears necessary in order to analyse its effect on the spread of silvoarable systems. The Reg. 2080/92 promoted the afforestation of agricultural lands with trees. Nevertheless intercropping with arable crops was not allowed. Thus, probably, the intercropping system of poplar and crops is disappearing because farmers often have the priority of increasing their income through grants.

On the other hand, other systems that involve tree planting in rows i.e. on the boundary of fields is expanding as a consequence of the application of the Regulation 2080/92. This trend should be confirmed by the adoption of “Agenda 2000” program.


31

In this context the European Union seems to recognise the environmental benefits but not the economic value of agroforestry systems in Italy, as well in other European Countries.

Table 4. Inventory of traditional silvoarable systems in Italy System Agro-ecological traits Tree arrangement

and density Crops and annual yields Other species

Olive trees with intercropping

Central Italy, hilly lands, altitude 200-450 m a.s.l., rainfall 550-800 mm/year

In rows and scattered, 25-400 trees/ha

Vegetables, wheat (30 q/ha), alfalfa, clover, vetch (2 t/ha)

Fruit trees

Oak trees with intercropping

Central and South Italy, marginal inner lands, altitude 400-800 m a.s.l., rainfall 700-800 mm/year

Scattered, 10-250 trees/ha

Wheat (3-5 t/ha), oat (1.2 t/ha), clover (2 t/ha)

Pyrus spp.

Poplar plantation with intercropping

North Italy, plains, altitude 0-250 m a.s.l., rainfall 800-1000 mm/year

In rows, 200-333 trees/ha

Maize (7.5 t/ha), soybean (2.5 t/ha)

Walnut trees with intercropping

South Italy, plains, altitude 0-700 m a.s.l., rainfall 500-800 mm/year

In rows, 25-625 trees/ha

Vegetables, wheat Hazelnut


32

Figure 2. Schematic representation of the evolution of agricultural landscape in Italy since 1950 as a consequence of agricultural mechanization and industrialization (drawing: Daniela Salvini). From P. Paris et al., 2001.


33

2.5 The Netherlands

Authors:Mayus M1, `Briel vd J 2

1Crop and Weed Ecology Group, Wageningen University Haarweg 333, 6709 RZ Wageningen, The Netherlands Safe Participant 2: WU Tel: + 33 (0) 4 99 61 26 40 Email: [email protected]

2Gelders Particulier Grondbezit Hommesleweg 490, 6821 LX Arnhem, The Netherlands Safe Subcontractor 1 (to WU): GPGE-mail: [email protected]: +31 (0) 317 466565

Recent evolutions of silvoarable agroforestry practices and systems In the Netherlands, there are no important developments of modern silvoarable practices and traditional agroforestry systems have been disappeared with the exception of some small traditional orchards (comparable with streuobst) and poplar pastures (Gijsbers, 1994). At present, a high level of specification of land use activities, its segregation into forestry and agriculture (Gijsbers, 1994) and monocultures are typical for this country. Before the industrialisation of agriculture, however, multifunctional land use was common on Dutch farms. Often trees were integrated into arable land and pastures or planted along fields and roads, for the production of fruit, timber, fodder and fuel (Gijsbers, 1994).. During the last decade, a modern type of multipurpose land use is getting attention (Vos et al., 1998) and in this context several studies and demonstration projects exist. In the Netherlands, agroforestry research is still focused on tropical areas. But due to the recognition of the environmental problems caused by the intensive and industrialised agriculture and the importance of agricultural land for landscape and nature conservation, several research projects have been set up to develop alternative and new multifunctional land use systems. These systems should offer additional income resources for farmers through practices that conserve (or improve) the environment, nature (biodiversity), landscapes and recreation areas (Loenen et al. 2001). The introduction of trees on farms fits very well in the philosophy of such projects and several of these projects already include the plantation of trees, often without using the term agroforestry, silvoarable or silvopastoral. The projects are set-up by research institutes and various other associations, occupied with land use systems that re-introduce trees into arable land. Moreover, there is a foundation for traditional orchards and a foundation that promotes durable agroforestry, in the Netherlands. Hence, agroforestry gets a new chance in the Netherlands.


34

Difficulties to find information on the agroforestry situation in the Netherlands Agrofororestry systems are rather unknown to the Dutch population, but also to farmers, foresters and public authorities. Governmental institutions did not collect statistical data or performed an inventory of agroforestry. With the exception of few reports on alternative land use projects that consider the plantation of trees on farmland, no publications could be found. However, thanks to the centralisation of agricultural research and the rather dense network of farmers’ organisation it was possible to obtain a rather good overview on still living practices in the Netherlands. In particular for the region Gelderland, the inventory is rather complete, but in the South of the Netherlands further inventories could be useful. The Dutch agricultural research centre WUR (Wageningen University and Research Centre), has a good expertise in agroforestry systems in the tropics, but for the temperate zone agroforestry research just started. The author could find only a proceeding paper on agroforestry in the Netherlands (Gijsbers, 1994). This paper announced an inventory of past, present and potentially promising agroforestry practices in the Netherlands, but this study has not been continued (van Wiersum pers. Comm., 2000). Literature on history agroforestry issues might be found in local journals and magazine, other than available in the WUR libraries. This makes the literature study very time consuming, and was, therefore, not yet continued by the authors.

Methods used for the silvoarable inventory For the inventory of still living silvoarable practices, the network of farmers’ (G)LTO and land owners’ organisation GPG (sub-contractor of partner 2), as well as organizations considering land use systems that introduce trees into arable land (Stichting Boslandbouw, Stichting Hoogstamfruit Rivierenland, Stichting Stimulant, Stichting Robinia) and research institutes(WUR-Alterra) have been consulted. In addition, internet and libraries (Wageningen University and Research Centre; Stichting Robinia) were used as information source. The survey was not conducted on pre-selected representative agricultural areas, but all available contacts were used. The personal contacts with farmers of the region and the contacts to the above mentioned associations and institutes formed a rather dense network within the Netherlands, in particular for the region Gelderland (mid Netherlands). It might be worthwhile to have a closer look in the South of the country. On farm surveys were conducted using the inventory sheet developed by the project partner 10 AUTH (Papanastasis, V.). With the exception of one farmer, all inventoried farmers were linked to associations or research institutes. The information was obtained directly from the farmers or the organisations behind the project.

Main Results Agroforestry is not a recognised or known land system in the Netherlands. Tree and crop plantations are largely segregated (Gijsbers, 1994). The public, but also farmers and foresters have generally no knowledge of agroforestry. The situation of agroforestry in the Netherlands is very similar to the


35

situation in Germany. With the industrialisation of agriculture, trees have been abandoned from arable lands and most traditional orchards have been replaced by modern orchards, which made the combination with crops and pastures impossible (Gijsbers, 1994). At present, changes of land use practices is again an issue, in the Netherlands, where the pressure on rural land is very high (Loenen, van et al. 2001, Vos et al. 1998, Edelenbosch and Dik 1995, Gijsbers, 1994). Multifunctionality is required with respect to sustainable agriculture, diversification of farmers’ income, new green use of set-aside land, biodiversity, attractive landscapes and recreation. Agroforestry, can met many of these demands and, hence, it is not surprising that research organisations and associations, which deal with land use systems, set up projects that include the (re)introduction of trees into farmland. In this context the following projects are active: • Maintenance of traditional fruit orchards; Stichting Hoogstamfruit Rivierenland • Multipurpose sustainable land use; WUR-Alterra et al. • Promotion of durable agroforestry systems in the Netherlands; Stichting Boslandbouw • Mixed plantation of poplars and sugar beet/ maize; WUR-Alterra (former IBN) two local

experimental stations and a nature conservation association • Flevohout: demonstration plot of timber production on arable land; Stichting Robinia

The inventory resulted in 50 traditional and 16 modern agroforestry fields. All fields make part of one of those projects with the exception of one modern silvopastoral system, which was set-up by the initiative of a farmer himself, in the south-west of the Netherlands. The major ”types of agroforestry plots” (differences in system design and/or motivation) are given in Table 1 and described in more detail in the inventory sheet (annex).

In former time, fruit orchards with cattle grazing were common in the Betuwe and Zuid –Limburg, the middle and South of the Netherlands (Boer and Ebbers, in work). Commercial fruit growers have replaced high stem trees by short type. Now, only few traditional orchards areas are left in these regions. The orchards are small, on average about 1 ha. The practical knowledge for maintaining orchards (i.e. tree species) is barely available any more. Often, the trees are poorly maintained and young trees are rarely found (Gijsbers, 1994), probably due to high costs. The foundation “Hoogstamfruit Rivierenland” promotes the proper use and maintenance of traditional (highstem) fruit orchards. There are about 50 small orchards brought under the activities of the co-operation. Ecological management is a prerequisite (EKO certification).

In Zeeland, a region in the South-west oft the Netherlands, a private farmer started a walnut-pasture plantation in 1996. The motivation for setting-up an agroforestry field on his arable farm (wheat and potatoes) is twofold. At the time, i) the field needed another use and ii) he was curious to try something new. Since, he is interested in nature friendly farming and a tree nursery is in his neighbourhood, the idea of a walnut-pasture plantation came up, i.e. walnut trees for fruit production and the pasture for sheep grazing. Important arguments for the trees were the fact that the tree planting material (8 cm perimeter) was covered by a grant for farm plantations (“erfbeplanting”) and that the farmer could find an intermediate trade for the sale of nuts. Nevertheless, the agroforestry


36

field is seen as an experiment, that from an economic point of view is not yet beneficial. (H. Almekinders, pers. comm., July, 2002).

The project on multipurpose land use (“Meervoudig Duurzaam Landgebruik”) through the combination of fruit trees and shrubs with pasture started in 1999 and will carry till 2003. It makes part of a large regional program, in which several organisations are participating, under the co-ordination of WUR-Alterra. In the region of Winterswijk (East of the Netherlands), 8 farmers joined the project. The total area of the 8 agroforestry fields comprises about 10 ha. The design of these fields varies in terms of distances between trees and distance between crop and tree line. The tree species is walnut used for fruit and timber. The farmers receive a grant for participating for the duration of the project, i.e. 5 years. In case of further subsidies, the interviewed farmer would like to continue the walnut –pasture plantation, otherwise he would remove the trees. So far a continuation of grants for participating farmers is not promised.

In 1995, the “Stichting Boslandbouw” (Foundation Forest-Agriculture), a non-profit foundation started the promotion of durable agroforestry systems in the north of the Netherlands. It adheres to biological land use systems. The foundation aims to reduce the import of tropical hardwood, to promote environmental friendly farming, and to enhance the quality of rural areas. The silvoarable field of three farmers participating in the project are described in the Table 5.

The Wageningen Research Centre Alterra (former IBN-DLO) in combination with two experimental stations and a nature conservation association performed an agroforestry experiment from 1989 – 1994. This study compared the benefits of a system of poplar forestation with a mixed plantation of a) poplars with sugar beet/ maize and b) poplars with grass. Three different silvoarable systems (trials) have been studied in the region of Drenthe (Table 1). The economic results are documented (Edelenbosch and Dik, 1995) and a detail report on the ”growth and yield of trees and crops in mixture” is in progress (Bervaes et al., in work). The growth of poplars between the crops was higher compared to poplars in monoculture. Apparently, trees benefit from the fertilisers applied to the crop and from the intensive weeding regime of the crop. This benefit weighed out the growth reduction due to competition for light and water. The estimated tree growth increase equals about 2 years compared to poplars in monoculture. Maize and sugar beet showed also a slightly enhanced crop yield in the first year. From the third to the sixth year crop yields decreased gradually for about 2 and 13 % to 27 and 40 % of the yield for sugar beet and maize, respectively. The combination of poplars with grass showed a lower growth decrease for the grass and a lesser growth gain for poplars (1 year) compared to the mixed plantations of poplars with maize or sugar beet. The economic evaluation showed that the combination of poplars with maize and/or sugar beets is more attractive than a monoculture of poplars, if the current grant systems are not considered. Otherwise a monoculture of poplars is more attractive. The economic benefit of a mixture of poplars and grass depends upon the market price for grass. The actual situation of subsidies is such that subsidies are available for poplars in monoculture but not for poplars in mixed growth.


37

Furthermore, ALTERRA wrote a manual for “farmers with trees” (“Boeren met bomen”) (Oosterbaan, 2000). This guide gives general information on the function of several tree species in agriculture, their arrangement and management, system benefits and costs, as well as information on current subsidy regulations. Agroforestry grants do not exist. However, grants with respect to i) regulations of nature conservation 2000 and regulations agricultural nature conservation, ii) landscape care, iii) motivation environmental friendly agriculture (Oosterbaan, 2000) might be considered for agroforestry types. Even than the grant systems has to change to make agroforestry economical attractive compared to forest or arable systems (Edelenbosch and Dik, 1995).

The foundation “Stichting Robinia”, a non-profit and non-governmental organisation, promotes the sustainable chain production of quality wood by developing projects, performing research and consulting. Flevohout, a 60 ha demonstration forest that should visualise the forest – to wood chain and different management types to growers and consumers, includes a 0.5 ha silvoarable field. The demonstration site is located in Lelystad (Northern-Netherlands) and is performed in co-operation with “Stichting Boslandbouw”. The purpose of the silvoarable project is to design an extensive land uses system that provides green services and fits into the landscape (W.J. Sanders, pers. Comm., July 2002). Therefore, the silvoarable field comprises various beautiful tree species (e.g. Catalpa bignonioides, Alnus glutinosa, Prunus avium, Gleditsia triacanthos) and hyacinth as crop. Trees are grown for timber and hyacinth for flowers and bulbs. The trees should provide wood or other products on short term. The hyacinth will be planted between the tree rows from 2003 onwards.


38

Table 5. Living agroforestry systems in the Netherlands

System; Start/tree age

Agro-ecological traits

Size Plot

Nofields;Total area

Treedesign

Use of trees (T), crops (C)

Motivation; Project

Fruit trees- pasture

Gelderland (mid NL) 850 mm, river clay

5 - 6 ha 50 fields; ca. 50 ha

100-200 trees/ha alternated

T:fruit,timber

C:sheep,cattle

Maintain traditional orchards; Foundation “Hoogstamfruit Rivierenland”2

Walnut – pasture

1996

Zeeland(South-west NL)

ca. 0,7 ha

1 field 159 trees/ha 7x 9 m in rows

T: fruit C: sheep grazing

Personal interest, private farmer

Walnut – pasture

1999

Gelderland (mid)

850 mm, sandy soil

2,5 ha 8 fields; 10 ha

10 x 10 20 x 20

T: fruit

C:hay,grazing

Multipurpose land use; Demonstration plot of regional project

Walnut – pasture

ca. 4 years

Overijssel (North)

800 mm, podzol

9 ha 1 field 69 trees/ha in rows

T: fruit

C: cattle

Promotion durable agroforestry in NL, Foundation “Boslandbouw” 3

Cedar – Cheanomeles

ca. 5 years

Fryslân (North)

800 mm, podzol

1 ha 1 field 100 trees/ha in rows

T: fruit,timber

C: fruit, flower


Gleditsia triacanthos – shrubs, perennials, annualsca. 6 years

Fryslân (North)

800 mm, clay

0,5 ha 1field 20 trees/ha in rows

T: timber, fodder, shade

C: sheep, fodder


Alnus cordata - shrubs, perennials 8 timber trees, 10 fruit trees ca. 1 year

Drenthe (East)

800 mm, peat/podzol

3 ha 1field

in rows

T: fruit, timber, C: grazing, fodder


Poplar – maize, sugar beet

1989 - 1996

Drenthe (East)

850 mm, podzol

1,2 ha 1 field 202 trees/ha 4,5 x 11 in rows

T: timber, paper C:

Experiment: Comparison monoculture poplars with poplars with crops WUR-ALTERRA

Poplar – grass

1989 - 1996

Drenthe (East)

850 mm, podzol

0,72 ha 1 field 202 trees/ha 4,5 x 11 in rows

T: timber, paper C:

Experiment: Comparison monoculture poplars with poplars with grass WUR-ALTERRA

Poplar – grass

1989 - 1996

Drenthe (East)

850 mm, podzol

0,5 ha 1 field 404 trees/ha 4,5 x 5,5 in rows

T: timber, paper C:

Experiment: Comparison monoculture poplars with poplars with grass WUR-ALTERRA

Timber trees* - Hyacinth 2003

Lelystad (North) 0,5 ha 1 field T:timber C:flowers, bulbs

FLEVOHOUT-Demonstration of timber production onarable land Stichting Robinia


39

Perspectives for extant traditional and novel silvoarable practices Agroforestry is a very interesting land use systems for the Netherlands. It addresses not only European issues like sustainable agriculture, overproduction, resources of quality timber, but can also meet national requirements. In a dense populated country as the Netherlands, where space is scarce, multi purpose land use is needed (Vos et al., 1998). This can be achieved by the introduction of trees in arable land (Oosterbaan, 2000). Within the context of the Dutch situation the following potential benefits of silvorable systems are to point at. The system can: • support sustainable and environmental friendly farming practices. In the Netherlands, most

arable farms are very intensively and industrialised causing large environmental problems. In silvoarable fields, chemical input can be reduced and nitrogen leaching lowered by its uptake by tree roots (Schroth, 1999).

• enhance biodiversity and provide aesthetic landscape and recreation areas. This point is very well recognised and studies on the possibilities of reimbursement for farmers providing green services are running (Vos et al., 1998).

• form a new purpose for set-aside programs. There is much land available, i.e. taken out of production due to overproduction and financial problems of farmers (Loenen et al., 2001). Agroforestry is an interesting alternative land use systems for set-aside land, since the forest area and is very low (less than 10 %) (Oosterbaan, 2000). Studies show, that forestry is attractive to farmers when subsidies are available (Edelenbosch and Dik, 1995). In the case that subsidies are available, agroforestry should be even more attractive because it fits better into the practices of a farmer than forestry. Besides agrofrestry is a cheaper way of afforestation (Edelenbosch and Dik 1995).

• diversify farmers’ income, and help risk diversification. Silvoarable systems increase the possibilities to adapt to the market situation by its flexibility to move the farming activities between agricultural and forestry activities. For the nearby future, it is expected that the subsidies will be reduced ().

• reduce import of hardwood. Due to the small forest area, only 10 % of the wood demand is produced in the Netherlands and the demand for wood is increasing here(Boer and Ebbers, in work). The major part is imported, often produced in non-sustainable ways. Therefore, there is a need to produce quality timber and resource for energy wood in the own country.

• provide control of wind and water erosion. Wind and water erosion is an important process in the North and the South of the country, respectively ().

The existing silvoarable projects, described above, underline that the awareness of potential benefits of mixtures of tree and crop plantations is going ahead in the Netherlands. However, at present, the development of novel silvoarable (agrofrestry) systems is hampered by: • a lack of available knowledge on the functioning of agroforestry. Potential users, i.e. farmers,

foresters, land owners and policy makers should be informed and /or educated in terms of silvoarable design and management are desired (Vergunst 1996).


40

• the current grant systems does not include agroforestry systems. This has to be changed to make agroforestry economical attractive compared to forest or arable systems (Edelenbosch and Dik 1995).The recent changes in land use regulations in France that include agroforestry situations will path the way for changes in other European countries.

Once these barriers are taken away, there will be a good chance for modern agroforestry and silvoarble systems in the Netherlands.

The maintenance of extant traditional fruit orchards is supported by the foundation “Hoosgstam Rivierenland”. Considering the recognition of their importance for biodiversity, cultural heritage, landscape aesthetic it is most likely that these orchards are continued to be highly valued and to be preserved. Similar is true for the traditional poplar pasture.


41

2.6 Spain

Author:Moreno G. Forestry School, Centro Universitario Univ. Extremadura, Plasencia 10600 Cáceres, Spain Safe Participant 7: UEX E-mail: [email protected]: +34 27 427000

Recent evolution of silvoarable agroforestry systems The evolution of the main agroforestry systems in Spain can be drawn from the four national agriculture census carried out in the years 1962, 1972, 1982, and 1989 (Ministery of Agriculture, (MAPA) and by the National Institute for Statistics (INE). Data from the last census (1999) are not available yet. Therefore, data from the national survey of the farm structure (1997) were also used. The figures of the census and survey are not directly comparable because the disparities in the clasification of the agroforestry systems. Moreover, this classification has changed also along the different census; so, a direct comparison of figures is not feasible. Besides, the type of the agroforestry systems documented in the data are very general; for instance, intercropped fruit trees, without specification of tree species or type of annual crop. Finally, two important type of agroforestry systems are ignored (only some spread data are available):

• Hedgerow around the small plots or farms. • Small orchards and kitchen gardens, where the combination of vegetables and fruit trees is

frequent. In spite of that, it can be observed that the extension of the main arable agroforestry sytems in Spain has diminished very significantly in the last decade (Table 6). In only seven years (1982 to 1989), a dramatic decrease was observed for most of the agfroforestry systems: the surface occupied by the agroforestry systems decreased to 68%. The most affected systems was the combination of fruit trees with annual crop and vineyard with annual crop (decreased to 34,9 and 37,7 %, respectively). The combination of permanent crop (olive, fruit or vineyard) with annual crop decrease to 40% in those seven years. This last combination decrease to 2,5% between 1962 and 1997. By contrast, the annual crop in forest tress (mostly dehesas) only decrease to 36,4% in the same period. The evolution of the combination of the two permanent crop followed a similar trend, with an important reduction: between 1972 and 1997 the total area was diminished to 34,1 %.


42

Table 6. Recent evolution in the surface of the main agroforestry system in Spain. Agroforestry system 1962a 1972 a 1982 a 1989 a 1997 b

Intercropped (silvoarable) systems

Fruit + annual crop 402005 78999 27562

Vineyard + annual crop 21677 8175

Olive + annual crop 242628 39092 20219

Permanent crop +

annual crop

15988

Forest tree + annual crop c 685893 478375 433000 357000 249437

Combination of woody crops

Olive + Vineyard 181866 67875 78270 39203

Fruit tree + Vineyard 57406 33058 14981

Olive + Fruit tree 217816 195566 107485

Fruit tree + Fruit tree 60913 85563 47650

Two

permanent crops

137686

Other 99729 104505 38552 a National Agriculture Census (INE 1963, 1975, 1985 and 1991).b National survey of the farm structure (INE 1999a).c Annual Report of Agricultural Statistics (MAPA, 1985 and 2000). Refers to annual crops with presence of some mature forest treescovering between 5 – 20 % of the surface (open woodland). This type of intercropped system refers mainly to dehesas; for instance, in 1997, 247031 ha were cropped in dehesas, while in other type of forest only 2406 ha were cropped.


43

Figure 3. Recent changes in the surface of the total and intercropped open woodlands in Spain.

0,0

0,5

1,0

1,5

2,0

2,5

3,0

3,5

4,0

4,5

5,0

1974 1977 1980 1983 1986 1989 1992 1995 1998

years

Ope

n w

oodl

and

(mill

ions

of h

ecta

res)

0

100

200

300

400

500

600

700

800

Cro

p in

ope

n w

oodl

and

(tho

usan

ds o

f hec

tare

s)

Open woodlandcrop in open woodland

Data adapted from the Annual Report of Agricultural Statistics (MAPA 1985 and 2000).

The most important arable agroforestry system in Spain and also the best documented is the intercrop of open woodland, with annual data from 1974 to 1998 (Figure 3). This system consists mainly of cereal crops (also legumes, like vetch, chickpea, lupin) in dehesas (generally holm oak ones; 80% in 1982).

The main causes of the decrease in the surface of the different agroforestry system have been:

a) Plot concentration, that is, the concentration of several small arable plots in only one large plot, was carried out during the sixties and seventies in order to increase the average size of the cultivated plots for facilitating their mechanisation. This concentration caused the elimination of thousands ok kilometers of hedgerow (Table 2).


44

Table 7. Diminution of the total length of hedgerow in the farms of the different regions/countries in Europe.

Region or country % of reduction Date

Catalonia (Spain) 46 1957 to 1985

France 43 1975 to 1985

United Kingdom 23 1984 to 1990

Ireland 14 1937 to 1984

Adapted from Miguel et al. (2000)

b) Irrigation projects, with the transformation of oak open woodlands (dehesas) in monocrops. The inclusion of dehesas with other increasing open woodland (e.g. mountainous shrubland with some spaced trees emerging in the burnt forest) in the statistical figures explain the maintenance of the total surface of the open woodland in Spain (Figure 3) in spite of the reduction of the surface of the dehesa (Elena 1987). Moreover, the tree density has diminished in a 23% between 1957 and 1981 (Miguel et al. 2000).

c) Abandonment of most small-sized farms in mountain areas because of their low profitability and the loss of human population in these areas. Most of these farms were combination of several crops, e.g. olive, fruit, vineyard with vegetables or cereals. This abandonment is reflected in the decrease of the number of fruit trees spread out in the farms (Figure 4).


45

Figure 4. Recent changes in the number of spread fruit trees in comparison with the total area occupped by fruit tree orchards (monocrop) in Spain.

0

1

2

3

4

5

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

year

Frui

t tre

e or

char

ds

(Mill

ions

of h

ecta

res)

0

5

10

15

20

25

30

Spre

ad fr

uit t

rees

(Mill

ions

of t

rees

)

Monocrop Sparsed

Adapted from Annual Report of Agricultural Statistics ( MAPA 2000).

NOTE: Fruit trees included are: Orange, Lemon, Mandarine, Apple, Peach, Cherry, Pear, Plumb, Fig, Almond, Hazelnut, Walnut and Carob.

Local difficulties to find reliable data on silvoarables The national census of farms initiated in 1962 and was repeated approximately every ten years. In order to homogenize with the statistical information of the European Union, a Spanish survey of the structure of the farms is made since 1993 every two years. Finally, to avoid two paralell database, in 1999 the last national census combining the required information in the previous census and surveys was initiated (data are not available yet). In this database, the farm is the unit of information, and the data inform us about the size of farms and plots and the surface occupied by the different type of crops, including different associations of crops (Table 8). On the other hand, the Ministry of Agriculture elaborates an annual database (Anuario de Estadística Agroalimentaria, MAPA), with the statistic results on cultivated surfaces with the different crops, and their average yields and prices. All these source of data allow us to know the importance and trends of the main groups of agroforestry systems in Spain.


46

General groups Type of association

Citric – Citric

Fruit - Fruit

Olive - Fruit

Olive - Vineyard

Two permanent crops

Fruit - Vineyard

With Vineyard

With Olive Annual crops with permanent crops

With Fruit trees

Corn with bean Two annual crops

Others

Annual crops with forest tress

Small orchards (generally combining trees and

vegetable)

Others

Data source: Project of the national census of farm (INE 1999b). Results are not available yet.

Table 8. Different types of crop associations considered in the last national agriculture census (1999) in Spain.

However many difficulties arise from the use of these statistical data: • Data from the last census (1999) has not been published yet. Only some preliminary and

very general data are available in the website of the National Institute of Statistic (www.ine.es).

• The defined agroforestry systems are very general, grouping some very different systems. For instance, intercropped poplar with corn and dehesas crooped with oats are grouped under crops with forest trees. Moreover, in the results of the last census presented up to now, the different system are grouped under three categories: a) Annual crop with forest trees, b) annual crop with permanent crop, and c) two permanent crop.

• Data on the different type of agroforestry systems are only available at the scale of provinces. The reports at county and local level do not inform about each specific system.

• The yield of the different crops are not discernible for monocrops and agroforestry system.


47

The present situation of agroforestry in Spain Since the results of the last national agriculture census are only partially available, data from the last survey on the structure of farm (INE 1999a) are used to describe the present situation of the agroforestry in Spain (Table 9).

Table 9. Surface occupied for different groups of agroforesty systems in Spain in 1997. Silvoarable systems:

annual crop with TOTAL

REGIONS

Open

woodland

Other forest

trees Woody crop

Two

permanent

cropsOthers

Silvoarable Agroforestry Cropland

Andalucía 153.652 1.357 3.972 30.008 5.630 158.981 194.619 3.930.308

Aragón 0 268 248 9.698 746 516 10.960 1.802.765

Asturias 0 0 0 201 796 0 997 29.849

Baleares 0 14 5.532 13.391 1.465 5.546 20.402 187.912

Canarias 0 0 254 444 101 254 799 94.335

Cantabria 0 0 31 35 152 31 218 15.533

Castilla-León 33.973 0 47 1.646 937 34.020 36.603 3.747.228

Castilla-Mancha 6.519 0 1.037 8.519 121 7.556 16.196 4.009.170

Cataluña 0 0 642 21.227 12.400 642 34.269 932.295

Extremadura 50000 30 2.656 13.051 4.423 52.686 70.160 1.366.863

Galicia 430 7 328 291 1.294 765 2.350 400.847

Madrid 2.457 0 573 1.194 68 3.030 4.292 244.821

Murcia 0 0 404 8.493 108 404 9.005 472.811

Navarra 0 0 24 734 88 24 846 352.633

C. Valenciana 0 730 192 28.344 9.650 922 38.916 91.900

País Vasco 0 0 17 268 565 17 850 159.760

Rioja 0 0 32 141 9 32 182 783.848

Total 247.031 2.406 15.988 137.686 38.552 265.425 441.663 18.622.878

Adapted from the National Survey of farm structure (INE 1999a).

From this data, it can be highlighted the low relative importance of the agroforestry respect to the monocrops (only 2,4%). Annual crop in open woodland is the most important type of agroforestry system (55,9% of total), reaching till 92,8% in Castilla-León. Others combinations of annual crops


48

with permanent vegetation (permanent crops or forest trees) have less relevance (3,6 and 0,5%, respectively). In the regions with highest percentages of agroforestry, C. Valenciana and Baleares (42,3 and 10,9%, respectively), the most of the surface is included in the group of the ‘two permanent crop’, which usually refers to different combinations of fruit trees (e.g. olive and almond, olive + fig, ...) and vineyard with olive.

The small orchards could also be considered a still very frequent agroforestry system in Spain, although usually they are not included into the statistic of combined crops. It consists of small plots (lower than 0,05 ha), where usually vegetables are cropped with different fruit tree species. They are very frequent in Northern Spain (humid Spain), in terms of number, but not in surface.

Table 10. Regional distribution of the number and surface of small orchards in Spain.

Region Number Ha Region Number Ha

Andalucía 15.006 427 C. Valenciana 9.298 280

Aragón 8.950 246 Extremadura 6.296 169

Asturias 17.830 402 Galicia 94.599 2.381

Balears 3.970 100 Madrid 595 17

Canarias 3.530 108 Murcia 2.089 67

Cantabria 4.979 120 Navarra 3.648 99

Castilla y León 24.880 673 País Vasco 21.992 609

Castilla-La Mancha 7.190 201 Rioja 1.541 45

Cataluña 12.397 303 TOTAL 238.796 6.246 Data from the preliminary results of National Agriculture Census 1999 (INE 2002).

Finally, although there are not systematic and periodical data about the importance of the hedgerow, most of the trees living in the hedgerow are included in the official statistic into the group of sparse trees. Therefore, the relative importance of the sparse fruit trees respect to the monocropped fruit trees, could indicate us the importance of this traditional agroforestry system in Spain, as aforementioned in Figure 4. The list of the tree fruit species still found sparse in many rural areas is shown in Table 11. Even some species are more abundant in a sparse distribution than in monocrop: quince, walnut and prickly pear. The region with the highest number of sparse trees is Galicia (Northwestern Spain, the region with higher precipitation values in Spain), where apple, pear, peach, plum and walnut are very common. Citrics are more frequent in the Balearic Islands, fisg in Canary islands, hazelnuts in Asturias


49

(Northern Spain), almonds in Andalucia and Castilla-La Mancha (Southern and Central Spain), carobs in the mediterranean regions (Cataluña, Valencia, Balear islands and Andalucia), and finally, olives in Madrid and Castilla-La Mancha (Central Spain). That is, all over Spain, there are at least a type of fruit tree sparsely distributed.


50

Fruit species Nº of tree in monocrop Nº of sparse tree %

Cítric 28.561.900 737185 2,5%

Apple 4.930.500 1680918 25,4%

Pear 4.070.300 1216541 23,0%

Medlar 318.300 145642 31,4%

Quince 62.700 158978 71,7%

Apricot 2.501.800 203857 7,5%

Cherry 2.859.900 757219 20,9%

Peach 5.229.900 624649 10,7%

Nectarine 1.470.300 72780 4,7%

Plumb 2.006.600 802078 28,6%

Fig 2.007.400 740567 26,9%

Custard apple 309.900 20854 6,3%

Pomegranate tree 299.900 79710 21,0%

Avocado 812.300 44147 5,2%

Banana 865.400 29177 3,3%

Kiwi 67.300 34540 33,9%

Date palm 58.500 12477 17,6%

Prickly pear 165.800 208317 55,7%

Almond 65.852.500 2721492 4,0%

Walnut 290.500 443531 60,4%

Hazelnut 2.868.400 488236 14,5%

Olive 234.642.700 436957 0,2%

Carob 6.906.600 194820 2,7%

TOTAL 367.159.400 11854672 3,1% Source: Annual Report of Agricultural Statistics 2000 (MAPA 2000).

Table 11. Comparison of the number of fruit trees planted in monocrop versus the spread tress into and between arable plots in Spain.


51

Intercropped dehesa: the main arable agroforestry system in Spain Since, the intercropped dehesas is the main agroforestry system in Spain (Table 9), we will make a more detailed description of this agrosilvopastoral system and the importance of cultivation on it. The dehesa is probably the larger agro-silvo-pastoral system in Europe, located in western and southwestern Spain and Portugal, with about 3 million hectares (Diaz et al. 1997) of widely spaced oak trees mixed with pastures or intercropped with fodder or cereals. Dehesa farms are generally private states, ranging in size form 100 to 10000 ha (Campos 1993). They are located generally on flat or gently sloping areas at 400-800 m a.s.l., with oligotrophic, acid soils, while in more eutrophic soils there have been a complete removal of tree cover (treeless dehesas). A notable decrease in the area of dehesa occurred in Extremadura (Elena et al. 1987) and in western Andalucia (Fernández-Alés et al. 1992) in previous decades (mainly sixties and seventies). Moreover, an important reduction in tree density of dehesas (23% between 1951 and 1981) has been also quoted (Miguel et al. 2000). This reduction may be attributed to changes in land-uses (removal of oaks for crop mechanisation, irrigation, and afforestation with pines and eucalyptus). Fortunately, the surface of dehesa seems established in the last years. In fact, in Extremadura the substitution of oak forest or open woodland is forbidden since 1984. Nevertheless, the lack of accurate figures makes difficult to make any generalization; for instance, some authors have quoted a very recent decrease in the surface of dehesa, from 2,3 millions ha in 1985 to 1,7 millions ha in 1998 (Miguel 2000). These contradictory figures could be explained by the lack of a unique criterium to define the concept dehesa and the lack of systematically collected data. The traditional uses of dehesas were maintained until the end of the 1950s (Campos 1984). This includes tree management: thinning, selection of the most productive trees, and punning, for the production of acorns, charcoal, firewood and cork; tillage for cereal cultivation; and scrub clearance. The dehesa is periodically cultivated with cereal for grain or other fodder species (legumes), as a complementary source of food (grain) for sheep and cattle. This practice, in conjunction with the livestock usage, has been carried out so as to maintain the agro-silvo-pastoral system free from the encroachment of woody vegetation (Montero et al. 1998) and to encourage the development of a grass ground cover (Díaz et al. 1997). The more frequent tree species in the intercropped dehesas is the holm oak (Quercus ilex subsp ballota), although other species like cork oak (Quercus suber) and a deciduous oak (Quercus pyrenaica) are also found. The main crops are different cereals (oats, barley, wheat, rye; sometimes vetch-oat). Initially, soil poverty only permits crop rotations in cycles of two to five years, successively repeated, which bring about the partition of the dehesa into zones called “Cuartos”. However, little by little, small modifications begin to be apparent, including particularly the gradual reduction of the encroachment of woody vegetation, and the improving of the grassland quality (Montero et al. 1998). Thus the cycles show a tendency to lengthen to as much as ten or twelve years, and even to eliminate the tilling stage when the livestock can by itself keep down the encroachment of woody vegetation (Montero et al.1998). Owing the high variability of the rotation cycle, it is difficult to know the exact cropped area each year, and to know the total arable surface in the dehesas. Despite of that, it is clear that the importance of crops in the dehesa has decrease in the last years (Figure 3). Anyway, the dehesa is


52

the most important intercropped system in Spain, and it is a still relevant activity in some areas of Spain (South and Central-Western), namely in six provinces (Table 12). Dehesas cover 3-12% of the annual cropped land in these six provinces. Moreover, the total arable surface in dehesa is much higher than the annually cultivated area. In this context, Leco Berrocal (1996) pointed out that the total arable surface of dehesas in Extremadura reached up to 302900 ha in 1982 (25% of the total surface of dehesa), while in that year only 29527 ha were cultivated (10,3% of the total arable dehesas) (MAPA 1985).

Provinces Cropland Crop in open woodland Percentage

Salamanca 372.717 32.205 8,6%

Badajoz 979.371 30.000 3,1%

Cáceres 362.584 20.000 5,5%

Córdoba 747.363 89.941 12,0%

Huelva 175.765 16.721 9,5%

Sevilla 926.972 28.000 3,0% Data source: National Survey of Farm Structure (INE 1999b).

Table 12. Comparison of the total cropped land and cropped open woodland in six provinces of Spain, where the agrosilvopastoral system of dehesa is common.

Recently, Escribano and Pulido (1998) studied the variability of the crops in the dehesa, and their final use. Results are summarized in the Figure 5. It can be highlighted that only a 30% of the cultivated land in dehesas is harvested, while a 70% is used as a supplementary food, grazed directly by cattle in the field.


53

Figure 5. Importance of the crops in dehesas of Extremadura (Spain).

Data adapted from Escribano and Pulido (1998).

Land use in dehesas

Fallow2%

Others1%

Improved pasture

17%

Natural grassland

54%

Shrubland10%

Cultivated16%

Crop use in dehesas

Grazed70%

Harvested30%

Grazed crops in dehesas

Chickpea2%

Oats-barley1%

Vetch-oats40%

Others23%

Ray-grass17%

Oats17%

Harvested crops in dehesas

Wheat-vetch2%

Oats4%

Barley34%

Sunflower10%

Wheat-barley

2%

Wheat40%

Others8%


54

Intercropped dehesas in the county of Cuatro Lugares We have surveyed 26 farms in the county of Cuatro Lugares (Annex A), covering 23263 ha (of a total of 100000 ha in the county, of which 70% are dehesas). The survey covered all the dehesas of this county with any type of intercropped cereal, in order to know the characteristics and aims of this type of silvoarable system. Cuatro Lugares is a county with around. Cuatro Lugares is located in the center of Cáceres province (Western Spain), at ca. 400 m a.s.l., with very gently slope (2% on average), with soils dominated by chromic Luvisols over tertiary cemented sands. The mean annual precipitation is 576 L m-2 y-1,annual mean temperature is 16ºC, and PET of 860 L m-a y-1. The mean size of the dehesas is 895 ha, ranging from 300 to 4600 ha.

One third (33,2%) of the dehesas in Cuatro Lugares had any type of intercrop in 2002. The arable surface in these dehesas reaches more than half the surface of those dehesas (2673 ha; 53,6%), although only a 11,5% of the surface was cultivated that year (12460 ha; Table 13). Considering the total area of dehesa in Cuatro Lugares, the total arable and the present-year cultivated areas were 17,8 and 3,8 %, respectively, which is still very relevant: 74,9% of the total annual crops of this county in 2002, with only 25,1% in monocrop. Most of the farmers did not know the actual production of their crops, but some available data indicate us that the production is very low. The cultivation of these dehesas is kept because it attends to several objectives. The sum of the objectives given by the farmers in Cuatro Lugares were:

• Supplement of fodder for animals. Here, there are three possibilities: o Directly grazed by cattle (4 farms) or by wild animal (for hunting, 2 farms). o Harvested still green for hay. o Harvested for grain to cattle. The election between the different possibilities depend on the yield.

• To sell the cereal (only two farmer). • To control the shrub encroachment (aimed by all of farmers). • To improve the pasture and the soil (aimed by all the farmers). • For the incentives (most of them).


55

Table 13. Main results of the survey on the intercropped dehesas in the Cuatro Lugares county (Extremadura, Spain). Data from the interview of 26 farmers. Question

Answers Tree species Holm oak, rarely some cork oak

Tree density 23 on average (Range: 5 – 50)

Tree height 8,3 m on average (Range 5 - 10 m)

Tree canopy diameter 10,7 m on average (Range 7,2 – 15,4 m)

Tree arrangement Scattered

Tree management There only mature trees, which are prunned every 15-20 years.

Tree production Fruit (1300 kg ha-1 on average). Firewood and Fodder: No data.

NOTE: Only one farmer could answer partially this question

Crop Oats (24 farms) Wheat (7 farms) Barley (1 farm) Vetch (1 farm)

Oats/vetch (1 farm) Chickpea (1 farm) Lupin (1 farm)

Total intercropped area 2673 ha in total 103 ha per farm on average (Range 20 – 500 ha)

Total arable area 12460 ha in total 498 ha per farm on average (Range 70 – 2000 ha)

Rotation cycle Each plot is cultivate every 6,8 years on average (Range 2-12 years)

Rotation with pasture

Fertilization Very variable: nothing, only manure or a combiantion of manure wih

some inorganic fertilzers (e.g. Nitrate, Superphosphate, N+P+K, ...).

Irrigation No

Crop yield 1500 kg of cereal. 15000 kg of forage.

Harvest No harvested (6 farms)

Harvested for hay or feed (18 farmers).

Harvested for selling the cereal (2 farmers)


56

Perspectives for the still living silvoarable practices

In spite of the many different incentives invested to improve farms and forests, very few are addressed to favour agroforestry systems, and none for arable agroforestry systems. Moreover, up to now, the different programs of the CAP policy, including the agroenvironmental measures, could be playing a negative role for the introduction of the new agroforestry in Spain. The first striking observation is that incentives are established for annual crops and not for orchard trees. Moreover, when there are trees into the crops, the incentive for the cultivated surface decreases in relation to twice the quantity of the surface covered by tree canopies. If the farm is used for grazing, the reduction of the incentive for the presence of trees is not applied, but if the tree cover is greater than 50%, there are no incentives. With respect to the afforestation program, most of the financial aids for tree planting in Spain has been spent in plantations in marginal pastures and degraded lands (into the set-aside of landcrop program). There had not been incentives to introduce any type of agroforestry system because the main objective of this program is the reduction of the agriculture production, and so the afforested lands cannot be cultivated nor grazed at least for the first twenty years. Other important restriction that limit the introduction of the new agroforestry plots are: • The high density of plantation (minimum 333 - 600 trees/ha, in function of region and especies)

prevents that the afforested land can be use as silvoarable systems even in the future. In some areas, it is easy to find some plots afforested with trees of rapid growth (mainly poplar) among many cultivated plots, but they are never cropped because of the high density.

• Orchards trees have no incentive, except walnut in some regions, where the plantation of this tree has an extra incentive (1653 Euros/ha versus 1533 for other species), and the olive, of great importance in Spain.

• The minimum surface of afforestation varies between regions, but it is always high (5 - 10 ha). The plantation of hedgerow is not financed.

• Even in some regions (e.g. Extremadura) the main agricultural area (e.g some irrigated zones) are excluded of most of the incentives for tree planting.

For the maintenance of the traditional agroforestry system there are two additional problems: the depopulation of the many rural areas and the slow growth rate of the most trees involved in the traditional agroforestry systems (oak, olive, carob, ...). Only dehesas are benefiting of a specific program (in Extremadura) to regenerate the very aged tree population, owing to dehesas have not regenerated in the last decades, probably due to an overgrazing and the abandoment of the traditional care fot tree regeneration. In spite of the absence of any specif program to the implementation of the modern agroforestry system, or for the potentiation and conservation of the traditional ones, two existing programs could contribute to the expansion of the agroforestry systems in Spain in the future:

- Incentives for plans of improvement of the agricultural exploitation. - Incentives for the novel farmers.

In both cases, any model of agriculture and/or cattle raising can be partially financed. Nevertheless, in the last decade, this way has been scarcely used to establish any type of the agroforestry systems.


57

We hope that the increasing knowledge on the profitabilty of the agroforestry systems and their role in the improvement of the environment and landscape, might contribute to reverse this situation.

As conclusion, in Spain there have been a great diversity of traditional agroforestry systems, and some of them are still very common, namely intercropped dehesas; but paradoxically, almost no attention has been paid to these systems in the different national programs of agriculture policy.


58

2.7 British Isles

Authors:Incoll 1 LD and Burgess2 PJ 1 School of Biology, University of Leeds Woodhouse Lane,Ls2 9 JT, Leeds, UK Safe Participant 4: UNIVLEEDSE-mail: [email protected]: +44 113 23332874

2 Cranfield University Silsoe, Bedfordshire, MK45 4DT, UK Safer Participant 5: CRANE-mail: [email protected] : +44 1525 863046

Introduction If the widest definition of agroforestry, the growing of trees on farms, is adopted for the British Isles, then we must include individual trees in hedgerows e.g. on field boundaries, individual trees in fields, shelterbelts, woodland, short rotation coppice, silvoarable and silvopastoral systems and orchards as examples of agroforestry. These modern ways of growing woody species all have historical antecedents. In some cases the historical methods and systems simply survive unchanged though they may well be less well managed than they were in the past.

Before we discuss traditional silvoarable practices we must define 'silvoarable'. The English word ‘arable’ is used as a noun (n.) and adjective (a.). The noun means ‘capable of being ploughed, fit for tillage; opposed to pasture or wood-land’ (Source: Simpson and Weiner, OED, 1989). The adjective means ‘[of crops] that can be grown on arable’. It originates from the French word ‘arable’ or from the Latin word ‘arabilis’ (from ‘arare’ - to plough) and first appeared in written English when used by Lambarde in 1576 to refer to ‘arable land’. In the previous century it had been spelt ‘earable’ from the verb ‘ear’ - to plough, to till. The term 'silvo-' from the Latin ‘silva’ means 'a wood'. Thus silvoarable (a.) means literally ‘trees on land fit for ploughing’. For the purposes of this paper we assume that silvoarable means the growing of trees with arable crops.

For some of the above examples of agroforestry, such as short rotation coppice (SRC) and woodland, there is no associated arable cropping, except in so far as they adjoin fields of arable crops. Although coppicing is a traditional woodland management practice, short rotation coppice is a monoculture on arable land developed in the late 20th century to provide woody biomass from coppiced willow or poplar for conversion to electrical energy and heat. Intercropping of SRC has been tried, but only experimentally by the Long Ashton Research Station near Bristol (Nichols etal., 2000). Traditional silvopastoral systems, called wood-pasture, were much more common in the


59

British Isles than silvoarable ones. Rackham (1976 a, b; 1980) has described these traditional silvopastoral systems in detail. However there is no real traditional silvoarable equivalent in Britain of the cultural savannas of the Iberian Peninsula, Dehesa and Montado.

The traditional agroforestry systems or practices that can be considered to be silvoarable or to have a silvoarable component, are orchards and individual trees in field boundaries and in fields. We now describe them in detail.

1. Cropping associated with trees in field boundaries and fields

Recent inventories of trees outside forests show that there are millions of trees in British field boundaries and fields. For example, in the most recent survey of such trees in England, there were estimated to be about 4½ million living "individual boundary trees", about 1.8 million living "individual middle trees" i.e. trees in fields but not on their boundaries, about 22 million trees in groups of less than 0.1 ha and about 60 million in linear features over 2 m high and under 16-m wide e.g. windbreaks (Anon, 2001). ("Individual" means the canopies of adjacent trees do not touch.) The most common species were the broadleaves - ash, oak, willow, elm and sycamore. Conifers only comprised 4.8% of the total.

The trees in field boundaries and fields can be discussed in three categories.

a) Individual trees in fieldsMiddle trees may be isolated individual specimen trees or a number of individual trees distributed within a field to give a park-like landscape. Parkland has a savanna-like appearance similar to Dehesa and Montado and is an important landscape feature on many aristocratic estates. Unfortunately the above inventory does not distinguish parkland trees as a census category. However the trees are placed in the fields, singly or in groups, they constitute part of an agroforestry system because the trees will either have crops, or more likely pasture with grazing animals, surrounding them. They may not be perceived by the landowner or farmer to be an agroforestry system with the trees having economic value. Although such trees will tend to be prone to damage from grazing animals or from cultivation, they are less likely to contain embedded metal like fencing wire, nails or staples than boundary trees. Enlightened owners will replace trees on death or even plant new specimen trees with fenced protection from large grazing animals, such as deer and cattle. In the primarily arable regions of Britain, the increasing mechanisation of agriculture in the second half of the 20th century has resulted in the felling of many middle trees giving a prairie-like aspect to large arable fields. Between 1986 and 1998 the numbers of middle trees dropped by 85% (whereas numbers of boundary trees fell by 34%) (Anon, 2001).

b) Individual trees in hedgerows


60

Hedgerows are a key feature of field boundaries in the British landscape. Some have existed for over 1000 years, and they usually include a variety of shrub and tree species. Hedges in England range in form including large banks in the South-West, single rows of hawthorn in Eastern areas, and beech hedges in parts of the Scottish Borders and Exmoor (Andrews and Rebane, 1994). In addition to constraining livestock, field boundaries may prevent soil erosion (by water and wind), and provide shelter, although their original function would have been to enclose land whether it be arable or pasture.

The majority of boundary trees will have been self-seeded or animals will have transported their seed actively or passively into the boundary. However some farmers may have deliberately included trees when planting or restoring a hedgerow (Kirby, 1992). Such activities are currently supported in the UK under the European Union co-financed 'Countryside Stewardship Scheme'.

Although boundary trees have been traditionally regarded as an economic resource they will probably not be seen as such by the farmer nowadays so they will not get any attention with the objective of increasing their economic value such as pollarding or branch and form pruning. The perceived lack of economic value may well explain the large numbers of dead trees counted in the above census viz. 91,200 individual boundary trees (1.9%) and 45,600 (2.5%) middle trees. During the 1970s and 1980s, Dutch elm disease contributed significantly to such numbers because it caused the death of most mature elms. In 1998, 66.4% of the dead trees in the landscape outside woodland, were elms (Anon, 2001). Dead trees are usually removed because falling branches can threaten the safety of people, animals, farm machinery or buildings. However conservationists maintain that such standing dead timber provides an important habitat for invertebrates, small mammals and birds.

c) Shelterbelts and windbreaks In many areas of the British countryside there is evidence of the efforts of previous generations to specifically provide shelter to farming enterprises through tree planting (Hislop et al., 1997). In part this is a result of the British Isles being subject to some of the highest windspeeds in Europe. In recent surveys of farmers' uses of farm woodlands, shelter is often ranked high on their list of priorities (Hislop et al., 1997). Thus in Kent in South-East England, Italian alder windbreaks have been planted at a spacing of 120 m to provide shelter for orchards and hop fields are bounded by tall poplar hedges.

2. Silvoarable agroforestry systems of full-standard orchards

In the last quarter of the 19th Century and the first quarter to half of the 20th, orchardists in England planted full standard trees i.e. trees where the lowest branches were 2-m above the ground so that sheep and cattle could graze unimpeded beneath them and without damaging the fruit. These 'grass orchards' of apples or cherries were planted on a grid of up to 40 feet square (i.e. 12.2 m x 12.2 m) and of cider apples on a rectangular grid of 36 feet x 30 feet (i.e. 11 m between rows and 9.1 m


61

between trees) (Barker, 1937; Roach, 1985). Consequently in the early years after planting, there was a lot of open ground between trees in dessert (eating) and culinary (cooking) apple and sweet cherry orchards and this ground was often planted with soft fruits e.g. blackcurrants, gooseberries, raspberries or strawberries, or vegetables including asparagus (Roach, 1985) or, in the cherry orchards, even with bush cultivars of apple and plum. Planting schemes for such cropping were published by Hoare in 1928 (cited and illustrated diagrammatically in Dupraz and Newman, 1997). These arable orchards constituted a silvoarable system, leading ultimately to a silvopastoral system as the canopies expanded and the soft fruit and the bush fruit trees were replaced by a grass sward after about 12 years. If the apple cultivars were grown on semi-dwarfing rootstocks in orchards at 3-m spacing, then the intercrop was gooseberries or currants. Intercropping was not used in cider apple orchards. It is difficult to estimate the area of orchard that was in silvoarable form at any one time. In 1906-1910 the orchard area in Britain was 250,000 acres (101,120 ha) rising to a maximum of 273,000 acres (110,480 ha) in 1951-1955 (Roach, 1985). If 25% of the 1906-1910 area was in cider orchards, then about 76,000 ha remains. If the intercropping lasted 12 years before grassing down then at any one time, if all orchardists used the system, 25% (i.e. 19,000 ha) might be intercropped per annum in a 50-year rotation, or 12.5 % (i.e. 9,500 ha) per annum in a 100-year rotation.

The traditional orchards of widely spaced large trees were relatively expensive to prune and to pick and it took several years before they began bearing fruit. After the Second World War with the consequent shortage of manpower, intensive and semi-intensive systems were adopted when new orchards were planted. Light pruning resulted in early cropping and pruning and picking could be managed from the ground. When apple, pear or plum trees are grown on dwarfing rootstocks, the 2-m-wide-strip under the trees has no vegetation because weeds and grass are controlled by herbicide. The alley between the rows of apples, cherries or plums is grassed to allow easy vehicle access but may be initially cultivated between pears. For this reason and because of the close spacing of the trees, intercropping followed by grazing is no longer possible.

Where there are remnants of the traditional orchards, these orchards can now be protected by the Countryside Stewardship Scheme (CSS). This is a scheme for protecting and restoring traditional and historic landscapes. Under it, owners or tenants of old orchards containing less than 150 trees per hectare can receive a grant of 250 pounds sterling per hectare per year towards their restoration by pruning, grafting, restocking and care of old trees and towards the management of the grass beneath the trees by grazing or cutting. An example of such a traditional orchard is at Milstead in Kent where a 50-year-old, 6.5-ha sweet cherry standard orchard, planted 30 feet square (9.1 m x 9.1 m), is grazed by sheep and receives a CSS grant. The fact that it was planted when intensive orchards of other fruits were being planted reflects the much later development of suitable dwarfing rootstocks for cherry. Such an orchard has a potential life of 100 years. These CSS grants do not allow the re-creation of such orchards on new land so the silvoarable phase of such orchards no longer exists.


62

Conclusion

As we intimated in the introduction, the extant traditional silvoarable agroforestry systems in the British Isles are limited to individual trees in the middle of arable fields, individual trees in boundary hedgerows of arable fields and shelterbelts and windbreaks. The only other example of a traditional silvoarable system, the silvoarable management of widely spaced orchards of full-standard cherries and apples (non-cider), is no longer practiced, as the alleys between the modern orchards with bush fruit trees of apple and cherry on dwarfing and semi-dwarfing rootstocks are now narrow and grassed (but ungrazed) to allow continued access of machinery.

Sources of information Personal communication

John Leigh-Pemberton, Milstead, Sittingbourne, Kent Alex R. Nichols, DEFRA, Bury St Edmunds, Suffolk Scott Raffle, ADAS Horticulture, Maidstone, Kent Karen Russell, Horticulture Research International, East Malling, Kent Andrew Tinsley, ADAS Horticulture, Tunbridge Wells, Kent Jenny Wong, School of Agricultural and Forest Sciences, University of Wales, Bangor, Gwynedd

LL57 2UW

Tabl

e 14

APP

END

IX L

ist o

f Spe

cies

Aut

hori

ty fo

r sc

ient

ific

nam

es: M

abbe

rley,

D.J.

(199

7) T

he P

lant

Boo

k. C

ambr

idge

Uni

vers

ity P

ress

, Cam

brid

ge.

Scie

ntifi

c na

me

Com

mon

nam

e: E

nglis

h C

omm

on n

ame:

Fre

nch

Com

mon

nam

e: It

alia

n

Woo

dy s

peci

es

Ac

er p

seud

opla

tanu

s L.

Syca

mor

e Ér

able

syco

mor

e A

cero

mon

tano

or L

oppo

ne

Alnu

s cor

data

Des

f.A

lder

, Ita

lian

Aul

ne à

feui

lles e

n cœ

ur

Ont

ano

napo

leta

no

Cra

taeg

us m

onog

yna

Jacq

.H

awth

orn

Aub

épin

e -

Fagu

s syl

vatic

a L.

Bee

ch

Hêt

re

Fagg

io

Frax

inus

exc

elsi

or L

.A

sh

Frên

e co

mm

un

Fras

sino

H

umul

us lu

pulu

s L.

Hop

H

oubl

on

Lupp

olo

Mal

us d

omes

tica

Bor

kh.

App

le

Pom

mie

r M

elo

Mal

us d

omes

tica

Bor

kh.

App

le, c

ider

Po

mm

ier à

cid

re

Mel

o da

sidr

o Po

pulu

s sp.

Popl

ar

Peup

lier

Piop

po

Prun

us x

dom

estic

a L.

Plum

Pr

unie

r Su

sino

Pr

unus

avi

um (L

.) L.

Che

rry

Cer

isie

r C

ilieg

io

Pyru

s com

mun

is L

.Pe

ar

Poiri

er

Pero

Q

uerc

us ro

bur L

.O

ak, E

nglis

h, F

renc

h C

hêne

ped

oncu

le

Rov

ere

Ribe

s nig

rum

L.

Bla

ckcu

rran

t C

assi

s R

ibes

ner

o

Ribe

s rub

rum

L.

Red

curr

ant

Gro

seill

e R

ibes

Ribe

s uva

-cri

spa

L.G

oose

berr

y G

rose

ille

à m

aque

reau

U

va sp

ina

Rubu

sspp

.R

aspb

erry

M

ûre

Lam

pone

Sa

lix sp

.W

illow

Sa

ule

Salic

e U

lmus

pro

cera

Sal

isb.

Elm

, Eng

lish

Orm

e O

ntan

o In

gles

e

Cro

p sp

ecie

s (n

on-w

oody

)

As

para

gus o

ffini

cina

lis L

.A

spar

agus

A

sper

ge

Asp

arag

o Fr

agar

ia X

Stra

wbe

rry

Frai

se

Frag

ole

Com

pile

rs: L

.D. I

ncol

l, A

. Sib

bald

, C. D

upra

z, F

. Ago

stin

i

2.8 Switzerland

Historical agroforestry experimentation in Europe

Author: Herzog F Swiss Federal Research Station for Agroecology and Agriculture Reckenholzstr. 191, CH-8046 Zürich, Switzerland Safe Participant 8: FALE-mail: [email protected]: +41 1 3777111

Introduction After a brief review of the history of European agroforestry, the results of some historical agroforestry experiments are presented and discussed in the context of the SAFE project. This report is preliminary and it is intended to complete this information by further literature search.

The history of European agroforestry At its beginning, European agriculture was based on shifting cultivation. Tree based agricultural sys-tems are reported from Roman times (Lelle & Gold, 1994) and until about two centuries ago, many European forests were significant sources of food and were grazed with ruminants and pigs (Brownlow, 1992). There actually was no distinct limit between forest and agricultural land and the input of organic matter and energy necessary to keep agriculture sustainable came from forests in the form of fodder, litter and wood (Haber, 1994). As an example, Eckert (1995) estimated that in the Neidlingen valley (Baden-Wurttemberg, Germany) until about 1500, the forest provided three quarters of the nitrogen and 90 % of the phosphorous available for the fertilisation of fields, vineyards and gardens. In the 18th and 19th century, intercropping on cleared forest land between rows of planted or sown forest trees was common practice in many forest districts in Austria, Belgium, France and Germany (Beil, 1839; Kapp, 1984). When industrialisation made labour more expensive and agronomic progress allowed to restore and maintain soil fertility without having to recur to reforestation, European agroforestry practices started to decline (Kapp, 1984). Trees were increasingly banned from agricul-tural land. This is mainly due to agricultural mechanisation that is linked to the pressure for increased labour productivity, to land re-allocations in the process of consolidations of fragmented holdings and to increasing specialisation of the farming enterprises (Herzog, 1998).

Arable Streuobst2 in Germany For farmers the combination of arable crops and fruit trees was of particular interest. In fact, the expansion of fruit production which occurred in the 19th century was only possible because these trees could be integrated in their farm and did not impede the rotation (Lott, 1993). Basically, rows of standard fruit trees were planted at different distances (Figure 6), distances were wide enough for crop production to dominate. This fundamental scheme, which was to remain valid for about a century, was varied in numerous ways: high and low stem trees could be altered within the rows, different types of

2 defined as “tall trees of different types and varieties of fruit, belonging to different age groups, which are dispersed on cropland, meadows and pastures in a rather irregular pattern” (translated from Lucke et al., 1992, p. 10)


65

arable crops from cereals, root crops to vegetables, strawberries, etc. were chosen, often fruit trees were also combined with berry production on bushes (currant, gooseberry, etc.). Arable undercropping was the pre-condition for the extension of fruit production: farmers would have uprooted the trees immediately if they would have had to abandon undercropping (Lott 1993, p. 98) and specialised fruit production was not possible because of the long period without fruit yield which had to be overcome. Apparently there was a conflict between scientists which aimed at improved and maximised fruit production and therefore recommended to restrict undercropping and farmer organisations which in 1910 reproached „the otherwise quite good specialists, who think that commercial fruit production cannot be combined with undercropping, are not able to calculate“ (Lott 1993, p. 100). It seems that the specialists lacked an appreciation of this agroforestry system and were biased towards increasing fruit production whereas farmers had a more comprehensive understanding and tried to optimise the overall financial return.

Figure 6. “Plan for a fruit orchard of one morgen with root crops or asparagus or raspberry” In the tree lines pomme fruit (“K”) and stone fruit (“St”) are alternated, tree planting distance is 15 x 10 m (from Lott, 1993, Fig. 12, Reference 229). Similar plans exist for combinations with strawberry, currant, etc.

From the fruit tree statistics of 1938 it can be assumed that in the 1930s there were about 800'000 hectares of (mostly silvo-arable) Streuobst in the German Reich of that time (relating to the boundaries before the second World War, SRA 1940)3. In a review Trenkle (1944) examined the impact of

3 78.9 *106 fruit trees, assuming 100 trees per hectare. In addition 110 *106 fruit trees in home gardens were counted which were often also underplanted.


66

understorey crops in fruit orchards on the nutrient and water balance and on the yield of the fruit trees. In the 1940s fruit production was mainly in the hands of small family farms which had to combine fruit and arable production in order to guarantee subsistence and sufficient fodder production for farm animals. Therefore fruit tree lines were distant enough to allow for cropping. These crops were part of the normal rotation which usually consisted of cereals, root crops (potatoes, sugar and fodder beet), vegetables, clover and grass. Trenkle (1944) insisted on the competition for water and nutrients, especially between May and July, which may reduce fruit production. Competition was judged higher with cereals, deep tooting clover and grass whereas root crops and vegetables were considered less demanding during that period and were seen to improve soil structure (provided sufficient availability of fertiliser). Trenkle (1944) concluded that in regions of lower rainfall (bellow 700 mm per year) undercropping should be abandoned, trees should be spaced more closely and only undercropped with root crops during the first years after planting. In regions with higher rainfall (above 850 mm per year) trees may be underplanted with grass. This may even be advantageous for wood production and quality because of the high evapotranspiration in autumn. Trenkle (1944) examined competition between fruit trees and crops from the point of view of fruit production. His approach is opposite to the one of Wahlen and Gisiger (1937) who estimated the annual loss of fodder production (quantity, quality) due to fruit trees on grassland in Switzerland at about 15 – 20 Million Swiss Francs. In neither of the two articles the total productivity of the system was assessed although Trenkle (1944) recognised the socio-economic justification for combining trees and crops (small holdings, scarcity of land, high share of subsistence). At the Berlin University the interactions between fruit trees and understorey crops were examined systematically in field experiments. Schulz (1936) presented results from the experimental years 1932 and 1933, they are summarised on the following pages.

The Berlin experiments (Schulz, 1936) Schulz (1936) conducted field experiments on silvo-arable agroforestry using stands of mature pear and apple trees in Berlin Dahlem. The trees were 25-30 years old, the experimental design in the pear orchard and in the apple orchard is shown in Figure 7. Control plots (no trees) were 70 to 300 m from the trees. The soil was relatively uniform with water holding capacity of about 28%, pH around 7.4.

Figure 7: Experimental design in pear and apple plantations (Schulz, 1936, Fig. 11, 12)

Standard apple trees (Sch.v.Boscoop), 9 x 10 m, 30 sqm per plot

Half-standard pear trees, 5 x 5 m, 6 sqm per plot / 20 sqm per plot


67

1. Pear plantation (5 x 5 m), kohlrabi on plots of 6 sqm Yields were between 64% (1st harvest) and 33% (2nd harvest) as compared to the control plots. This is mainly explained by the reduced availability of light (Figure 8). Plots were irrigated and sufficiently fertilised to exclude competition for these factors. In addition there was a control with artificial shading by covering the crops with textiles. Yields under artificial shading roughly corresponded to yields in the agroforestry plots.

Figure 8 . Yield of kohlrabi and light intensity in unshaded control plots (100 %) and in agroforestry plots (pear plantation) compared to the control in two subsequent years (Schulz, 1936, Fig. 19).

2. Pear plantation (5 x 5 m), head lettuce on plots of 6 sqm

Control plot

Controlplot

Agro-forestry plot

Agro-forestry plot

2nd

harvest1st

harvest

1932

1933

Control plot

Controlplot

Agro-forestry plot

Agro-forestry plot 2nd

harvest1st

harvest

Yield

Light


68

Yields were at 53% as compared to the control plots, again the proportion of light and yield reduction was similar. The quality of the lettuce was strongly reduced: the share of premium quality lettuce from the agroforestry plot was only 1.5% (1st year) and 17.2% (2nd year) as compared to the control plot where 79.3% (1st year) and 87.6% (2nd year) could be sold on the market.

3. Apple plantation (9 x 10 m), bush bean, knob celery, white cabbage on plots of 30 sqm Yields were reduced by around 50% as compared to the control (Table 15). Again the reduction is explained by the reduced availability of light.

Table 15. Vegetable yields of control and agroforestry plots (Apple trees, 9 x 10 m) (Schulz, 1936, collated from Tables 37a, 37b, 37c).

Agroforestry plot Control kg % kg %

Bush bean 30.81 53 58.32 100 Knob celery 28.90 54 53.10 100 White cabbage 130.05 47 278.40 100

Figure 9. Impact of apple trees (9 x 10 m) on light intensity and yield of bush bean at different distances from tree row (Schulz, 1936, Fig. 21)


69

The plots of 30 sqm were subdivided in 5 sections of 2 m width each which reflect different distances from the tree row. Yield as well as light availability were recorded individually. This allowed to assess the impact of different distances from the tree row. As an example yield and light availability for the

bush bean experiment are shown in Figure 9.

The possible reasons for reduced yield under trees were examined (Table 16):

- light: there clearly was a parallel development of light availability and yield (Figure 9).

- water: in the top layer of the soil there was no real difference in water availability between agroforestry and control. About 13% (26 mm) of the precipitation was intercepted by the tree crowns (Table 16). Schulz (1936) concluded that differences in water availability cannot be the major cause for yield reduction under agroforestry although he acknowledged that he did not test water availability bellow 25 cm.

- temperature: temperature was more equilibrated in the agroforestry plot but daily minimum and maximum temperatures differed only by about one degree Celsius. It was concluded that this could not explain the strong reductions of crop yield.

100% yield level

Yield and light intensity

Lig

Yiel


70

Agroforestry plot Control2 m 2 m 2 m 2 m 2 m Average

Yield: relative yield [%] 33.9 65.0 84.7 56.5 31.1 54.3 100

Quality [% of total harvest]

Premium quality 57.3 93.2

Light: Relative intensity of light [%] 48.2 68.2 83.8 73.6 53.3 65.7 100

Water: Relative humidity in top soil [%]

31. May 10.8 12.8 12.1 14.7 12.6 12.6 11.1

24. June 15.4 18.1 15.3 17.5 16.0 16.4 17.1

29. August 9.9 11.8 9.9 13.0 10.1 10.9 10.7

21. September 8.6 9.0 9.3 11.6 9.6 9.6 9.8

11. October 9.3 9.5 9.6 11.3 10.1 10.0 9.4

Water: precipitation measured at soil level, total of June – August [mm] 190.9 169.5 213.5 193.2 201.5 190.8 217.5

Temperature: average daily minimum and maximum temperature from May to September [°C]

Minimum 10.9 10.3

Maximum 24.4 25.4

Table 16. Values for yield, quality, light, water and temperature availability in the knob celery plots under agroforestry (apple trees) and in the unshaded control in 1933 (collated from Schulz 1936, Tables 39b, 40, 41, 42, 44). The agroforestry plots between tree rows were subdivided in sections of 2 m. See Figure 7 for the experimental design.


71

From these data Schulz (1936) concluded that in the agroforestry systems he investigated the observed reduction of yield is mainly caused by the reduced availability of light. He confirmed this by a parallel experiment where shading was provided by textiles and which resulted in similar values as the agroforestry plot.

4. Potatoes and undercrop In both plantations (pear and apple) experiments with potatoes (and onions) were conducted. The potatoes („Holländischer Erstling“, a variety of early potatoes) were planted at distances of 50 x 50 cm in the same experimental design as described above. Again there was a parallel evolution between yield and light availability which increased with increasing distance from the trees. Compared to the control, the average potato yield in the agroforestry plot was 39.6 % (pear trees, 5 x 5 m) and 45.3 % (apple trees, 9 x 10 m), respectively (Figure 10).

Figure 10. Relative yield and light intensity of early potatoes under pear trees (5 x 5 m) and apple trees (9 x 10 m) as compared to a control plot without trees (Fig. 24 in Schulz, 1936).

In addition to the reduction in yield, the potatoes’ starch content under agroforestry was 13-20 % lower than in the control plot.

5. Conclusion

Pear plantation 5 x 5 m Apple plantation 9 x 10

100% yield on control

100% yield on control

Yield and intensity of light as compared to control

Light intensity

Yield Yield

Light intensity


72

The reduced availability of light was identified as the main factor for the reduced yield (and quality) of understorey crops in an agroforestry setting as compared to open plantings, followed eventually by water stress as the 2nd most important factor. Schulz (1936) discussed his results in the context of the literature of his time (mainly from the US, the UK and France). The observed yield reductions differed quite significantly from earlier published indications of yield reduction (e.g. Janson, 1924). These, however, were not based on experiments but resulted from general experience and observations.

Schulz (1936) repeatedly pointed out the difficulties of agroforestry experimentation which either requires long term experiments over several decades or has to use existing tree plantings which eventually do not allow for an optimum experimental design. In his experiments the main drawback of the existing plantation was that tree distances were rather low (5 x 5 m for pears) which clearly limited the possibilities for cropping. From other literature (e.g. Lott, 1993) it can be concluded that farmers generally used larger spacings.

Present time silvo-arable agroforestry in Germany Whereas silvo-arable agroforestry with forest tree types had been abandoned already in the 19th / early 20th century, silvo-arable Streuobst persisted until the 1950s. Then it was abandoned as well due to the development of intensive fruit production with dwarf trees and to large scale mechanisation of agricultural crops. In eastern Germany, however, there was an interesting exception. When farmers were forced to join agricultural co-operatives they were allowed to keep only small plots of land for their own purpose. On some of these plots the ancient silvo-arable Streuobst revived. Undercropping of cherry trees (Prunus avium) mainly with turnip (Beta vulgaris), but also with alfalfa (Medicago sativa), potatoes (Solanum tuberosum) and oat (Avena sativa) is still practised by a few farmers up to today for household consumption (Photo 14).

Relevance for SAFE The experiments of the first half of the 20th century are of general interest for the SAFE project because they address the topic which is also at the centre of SAFE: interactions between trees and crops, impact on yield, identification of limiting factors. Of course the technology was less advanced. On the other hand, Schulz (1936) made his trials in fully developed stands of trees. Unfortunately the crops he examined – except for potatoes – differ from the crops which are investigated in SAFE.

Reading these historical articles with the today’s point of view it is interesting to observe that none of the authors tried to examine the entire system. They concentrated on either the crop or on the trees. It is not surprising, therefore, that most authors come up with more drawbacks than advantages (trees reduce crop yield, crops hamper tree management and reduce fruit yield). In contrast, farmers seem to have had a more comprehensive point of view of optimising (financial) yield by combining the two components.

The following observations are of particular interest for SAFE:

- yield reduction of crops in fully developed agroforestry stands may be significant, they were around 50% in the Berlin experiments (Schultz, 1936) – although at rather narrow tree spacing;

- not only the quantity but also the quality of the crop’s yield may be reduced; if possible this must be considered in the economic model;


73

- deep rooting crops (grass, clover, cereals) were considered less suitable than root crops which improve soil structure;

- light and water were identified as the major limiting factors;

- in humid regions intercrops (or grass) were considered beneficial for the “ripening” of the wood due to the increased evapotranspiration in autumn;

Based on a refined literature search more quantitative information could be gathered from historical long term experiments and from experiments in fully developed agroforestry plots. They might be used for plausibility checks of the HySAFE model.


74

3. Summary Conclusions

The single country contributions on the extant silvoarable agroforestry in Europe are the result of the collection of information from bibliography or other sources (personal contacts and internet sites), as well of national/local surveys of silvoarable systems conducted within the SAFE Project during the first year of activity. The inventoried silvoarable systems are collected in the WP2 Deliverable 2.1 “Database of current silvoarable systems in Europe”.

Many problems affected the collection of information on extant silvoarable systems in the studied countries:

• Lack of interest on the agroforestry issue that is often ignored by many Institutions; • Lack of official statistical data. In many of the countries, with the exception of Spain,

France and in minor extent UK and Germany, there are not reliable statistical data on silvoarable systems;

• Information are in most cases dispersed and often on local journals, which are not easy to reach with the usual tools of literature search;

• Lack of enough time. In Greece, The Netherlands and Germany the inventories of silvoarable plots were totally initiated within the SAFE Project; therefore, in less than one year it was not very easy to find out the silvoarable plots dispersed in the countryside;

Although the above mentioned limits, two main scenarios concerning silvoarable agroforestry in Europe can be drawn, considering two main geographical and climatic areas of the continent:

• Mediterranean Europe; • Northern Europe.

The first is much richer of silvoarable systems in comparison to the second. This probably reflects the higher biodiversity of the Mediterranean ecosystems in comparison to the northern ones. Basically in Northern Europe we were able to document two types of silvoarable systems: i) Streuobst (mostly in continental Europe) and ii) linear systems in the British Isles. In Southern- Mediterranean Europe there are much more systems, namely i) dehesas with holm oak and/or other evergreen or deciduous oaks, ii) intercropped olive groves, iii) intercropped walnut groves and plantations, iii) pre-verges (which are similar to streuobst), iv) intercropping in poplar plantations, v) other forest tree plantations with intercropping. Generally in Northern Europe, one of the main reason of the diffusion of the silvoarable streuobst in the 18th century was the need for the farmers to combine long term fruit production with annual income from the arable crops (Herzog, 2002). Now, with the wide diffusion of modern intensive orchards the traditional fruit trees groves have been almost completely neglected, apart some peculiar areas as in former Eastern-Germany or were local laws and regulations protects these systems as a sort of historical heritage. In Germany and the Netherlands, issues like sustainable agriculture and nature protection are increasing the interest on agroforestry, and numerous research projects dealing with silvoarable systems are undergoing. In most of the Mediterranean silvoarable systems the tree component is multi-purpose (wood as fuelwood and/or timber) and fruits for human consumption (e.g.: walnuts, olives, various fruit) and for animal grazing (acorns). Systems like olive grove intercropping and dehesa are residual of much more extended old systems and they have been strongly reduced by the large adoption by most of the farmers of the modern


75

monocropping agriculture. Although their past importance, in both cases they have been surprisingly very much poor studied and their productive and ecological role is not fully understood, also in relation to their present extent in fragile areas. Other Mediterranean systems like poplar plantations and walnut plantations or groves with intercropping have timber as main output and this has made them more interesting on the economical point of view, being timber a valuable products on the European market. These systems have been retained also in fertile agricultural areas. Generalisation on their actual economical importance is not possible, because this can been affected by local national market situations and rural policies. Also in this case, there is a tremendous lack of research activity.

As main conclusion, although extant silvoarable systems in Europe are in great part residual of widely old used systems, there is still a considerably richness of these systems in our continent. The quantification of extant silvoarable systems in Europe is extremely difficult for the above mentioned reasons. Their productive role in the studied European countries is not yet fully understood and it deserves much more attention, also in relation to diversification of farmers income and sustainable farming systems, two issues of strategic importance for the global development.

Furthermore, this report is a unique and valuable source of data and information on extant silvoarable systems in Europe, the first one collecting all together information from 7 different European countries, concerning system types and their description, the main regions were these systems are still used, yield performances of the herbaceous and woody components and also the areas covered by some of the systems in some countries.


76

4. References

1. Almekinders H (2002) Personal Communication July 2002. Olieslagersweg 2, 4525 LC Retranchement, the Netherlands.

2. Andrews J and Rebane M. (1994) Farming and Wildlife. A practical handbook for the management, restoration and creation of wildlife habitats on farmland. The Royal Society for the Protection of Birds, Sandy. 358 p.

3. Anon (2001) National inventory of woodland and trees: England. Forestry Commission, Edinburgh. 58 p.

4. Barker BTP (1937) Cider Apple Production. Bulletin No. 104. Ministry of Agriculture and Fisheries. HMSO, London. 75 p.

5. Baumhof-Pregitzer M. and Langer S (1997) Ideen-Aktionen-Konzepte zum Erhalt der Streuobstwiesen in Baden-Württemberg. In: Stiftung Naturschutzfonds beim Ministerium Ländlicher Raum Baden-Württemberg (ed). ISSN: 1430-984X, Stuttgart, Germany.

6. Beigel B (1995). Lebensraum Streuobstflächen. Bayrisches Staatsminsterium für Ernährung, Landwirtschaft und Forsten (ed)

7. Beil A (1839) Aphorismen über die Verbindung des Feldbaues mit dem Waldbaue oder Röder und Baumfeldwirtschaft. Frankfurt a.M., Andreaische Buchdruckereien, 89 p.

8. Bertolotto C, Pisanelli A and Cannata F (1995) Pratiche agroforestali nella regione Umbria. Monti e Boschi 22: 39-45.

9. Boer J de and Ebbers J Boslandbouw. Report in work. 10. Brownlow M J C (1992) Acorns and swine: Historical lessons for modern agroforestry.

Quarterly Journal of Forestry 86(3): 181 - 190. 11. Bünger L und Kölbach D (1995). Streuobst-Bindeglied zwischen Naturschutz und

Landwirtschaft. Dokumentation Natur und Landschaft, Sonderheft 23, Bibliographie Nr. 69, Bonn, Bundesamt fur Naturschutz, 166 pp.

12. Campos P (1993) Valores comerciales y ambientales de las dehesas españolas. Agricultura y Sociedad, 66: 9-41.

13. Campos P (1984) Economía y Energía en la dehesa extremeña. Madrid: Instituto de Estudios Agrarios, Pesqueros y Alimentarios.

14. Coulon F, Dupraz C, Fabien L, Pointereau P (2001). Étude des pratiques agroforestières associant des arbres fruitiers de haute tige à des cultures ou des pâtures, Solagro/Inra, Ministère de l’Aménagement et du Territoire et de l’Environnement, 199 p.

15. Cullotta S, La Mantia T and Barbera G (1999) Descrizione e ruolo dei sistemi agroforestali in Sicilia. In: Magni M, Giordano E and Di Girolamo C (eds) Secondo Congresso Nazionale di Selvicoltura, Proc of a National Conference in Venezia, Italy, 24-27 June, 1999, Vol. 4 pp 429-438. Consulta Nazionale per le Foreste ed il Legno; Ministero delle Politiche Agricole e Forestali (Mi.P.A.), Accademia Italiana di Scienze Forestali, EdAs, Italy.

16. Davies J C (1979) The modern Cider Apple Orchard. Ministry of Agriculture, Fisheries and Food. MAFF, Pinner. 13 p.

17. Degebeck M (2000a). Wirtschaftlichkeit von Pflegemaßnahmen zur Revitalisierung vernachlässigter Streuobstbestände. Bayrische Landesanstalt für Weinbau und Gartenbau (LWG), Abt. Landschaftspflege (ed)

18. Degebeck M (2000b) Umfrage Ergebnisse zum Thema Streuobst im Landkreis Würzburg. Bayrische Landesanstalt für Weinbau und Gartenbau (LWG), Abt. Landschaftspflege (ed)


77

19. Degebeck M (2001) Maschineneinsatz bei der Streuobsternte – Versuchsergebnisse aus Unterfranken. In: Bayrische Landesanstalt für Weinbau und Gartenbau (LWG), Abt. Landschaftspflege (ed): In Schule und Beratung 1/01

20. Díaz M, Campos P, Pulido FJ (1997) The spanish dehesas: a diversity in land-use and wildlife. In: Farming and Bird in Europe. Academic Press. Pp. 178-209.

21. Drissner D (1996) Faunistisch-ökologische Untersuchungen an Dipteren- und Lepidopterengesell-schaften verschiedener Streuobstwiesen. Dissertastion, Universität Ulm, Germany.

22. Dupraz C (1994) Les associations d’arbres et de cultures intercalaires annuelles sous climat tempéré. Rev. For. Fr., 46: 72-83.

23. Dupraz C (1994) Prospects for easing land tenure conflicts with agroforestry in Mediterranean France: a research approach for intercropped timber orchards. Agroforestry Systems 25 :181-192.

24. Dupraz C, Fournier C, Balvay Y, Dauzat M, Pesteur S, Simorte V (1999) Influence de quatre années de culture intercalaire de blé et de colza sur la croissance de noyers hybrides en agroforesterie. In : « Bois et Forêts des Agriculteurs», Actes du colloque de Clermont-Ferrand des 20 et 21 Octobre 1999, Cemagref Editions, Antony, pp. 95-114.

25. Dupraz C, Lagacherie M, Liagre F, Boutland A (1995) Perspectives de diversification des exploitations agricoles de la région Midi-Pyrénées par l’agroforesterie. Rapport de fin d’étude commandité par le Conseil Régional Midi-Pyrénées, Inra-lepse éditeur, Montpellier, 253 p.

26. Dupraz C, Lagacherie M, Liagre F, Cabannes B, (1996). Des systèmes agroforestiers pour le Languedoc-Roussillon. Impact sur les exploitations agricoles et aspects environnementaux. Inra-Lepse éditeur, Montpellier, 418 p.

27. Dupraz C and Newman SM (1997) Temperate agroforestry: the European way. In Gordon, A.M. and Newman, S.M. (eds) “Temperate Agroforestry Systems.” CAB International, Wallingford. pp. 181-236.

28. Eckert G (1995) Untersuchungen zur Geschichte der Landnutzung und zur Landschaftspflege auf brachgefallenen Wacholderheiden und Steinobstwiesen im Neidlinger Tal (Kreis Eßlingen). Stuttgart, Verlag Ulrich E. Grauer, 178 p.

29. Edelenbosch N E and Dik E J (1995). Mengteelt van populier met suikerbieten, snijmais en gras. Deel 1. Economische evaluatie van mengteelt van bomen met landbouwgewassen. Report IBN-DLO 181.Wageningen, the Netherlands, 99 pp.

30. Elena M, López JA, Casas M, Sánchez del Corral A. (1987) El carbón de encina y la dehesa. Madrid: Ministerio de Agricultura, Pesca y Alimentación.

31. Escribano M, Pulido F (1998) La dehesa de Extremadura: estructura económica y recursos naturales. Consejería de Agricultura y Comercio – Junta de Extremadura, Merida (Spain). 145 pp.

32. Etienne M, Valeix J (1995) Installation d’élevages ovins sur grandes coupures. In : Systèmes sylvopastoraux en région méditerranéenne française, INRA éditeur, Fr.

33. Etienne M & Rapey H (1999). Simulating integration of agroforestry into livestock farmers’ projects in France. Agroforestry Systems. 43 : 257-272.

34. Fernández Alés R, Martín A, Ortega F, Alés EE (1992) Recent changes in landscape structure and function in a mediterranean region of SW Spain (1950-1984). Landscape Ecology, 7: 3-18.


78

35. Gavaland A, Cabanettes A (1999) Production de biocombustibles en parcelle agroforestière : comparaison entre espèces ligneuses d’accompagnement et cultures intercalaires. Rapport final de convention ADEME n° 9601039, 71 p, Fr.

36. Gijsbers HJM (1994) Agroforestry in the Netherlands. In: Welte E., Szabolcs I., Huettl (eds). Agroforestry and Land Use Change in Industrialized Nations.p.123.

37. Groppali R (1999) Siepi, filari e biodiversità nella Valpadana interna: l’esempio di Cremona tra 1980 e 1997. Monti e Boschi, 2: 19-23.

38. Grubb NH (1949) Cherries. Crosbie Lockwood, London. 186 p. 39. Haber W (1994) Ist „Nachhaltigkeit“ (sustainability) ein tragfähiges ökologisches Konzept? In:

Pfadenhauer J. (ed.) Verhandlungen der Gesellschaft für Ökologie, Freising-Weihenstephan, Bd. 23, 7 - 17.

40. Hall DO (1997). Biomass energy in industrialised countries –a view of the future. Forest Ecology and Management 91: 17-45.

41. Herzog F (1998) Agroforestry in temperate Europe: History, present importance and future development. In: Keulen H. van, Lantinga E. A., van Laar H. H. (eds.) Mixed Farming Systems in Europe. Workshop Proceedings, Dronten, The Netherlands, 25 - 28 May 1998. A. P. Minderhoudhoeve-Series 2, 47 - 52.

42. Herzog F (1998). Streuobst: a traditional agroforestry system as a model for agroforestry development in temperate Europe. Agroforestry Systems 42: 61-80.

43. Herzog F and Oetmann A (ed.) 2000.Communities of Interest and Agro-Ecosystem Restoration: Streuobst in Germany. Interactions between Agroecosystems and Rural Human Community. Advances in Agroecology, Springer, New York

44. Hislop AM, Palmer HE and Gardiner, B.A. (1999) Assessing woodland shelter on farms. In: Burgess, P.J., Brierley, E.D.R., Morris, J. and Evans, J. (eds) “Farm Woodlands for the Future. ” BIOS Scientific Publishers, Oxford. pp. 31-43.

45. Hofbauer R (1998) Untersuchungen zur Ökologie von Streuobstwiesen im württembergischen Alpenvorland. Cramer Verlag. ISBN: 3-443-64204-7.

46. INE 1963. I Censo Agrario de España 1962. Instituto Nacional de Estadística - Ministerio de Agricultura, Madrid. Several volumes.

47. INE 1975. Censo Agrario de España 1972. Instituto Nacional de Estadística- Ministerio de Agricultura, Madrid. Several volumes.

48. INE 1985. Censo Agrario de España 1982. Instituto Nacional de Estadística, Madrid. Several volumes.



51. INE 1999a. Encuesta sobre la estructura de las explotaciones agrícolas. 1997. Instituto Nacional de Estadística, Madrid. www.ine.es/inebase/cgi/um.

52. INE 1999b. Censo Agrario de España 1999. Proyecto. Instituto Nacional de Estadística, Madrid. 166 pp. www.ine.es/proyectos/agra1999/proyec99.htm

53. INE 2002. Censo Agrario de España 1999. Resumen de resultados. Instituto Nacional de Estadística. Unpublished data. www.ine.es/daco/daco42/agricultura/cenagra99.xls

54. Janson A (1924) Der Grossobstbau. Berlin, Parey. 55. Kapp G (1984) Agroforstwirtschaft in Deutschland. Der Waldfeldbau im 18. und 19.

Jahrhundert. Allg. Forst- u. J.-Ztg. 155(2): 266 - 270.


79

56. Kapp G (1984). Agroforstwirtschaft in Deutschland. Der Waldfeldbau im 18. und 19. Jahrhundert. (A historical form of agroforestery in the Federal Republic of Germany: The Agrisilviculture in the 18th and 20th centurary). Allgemeine Forst – und Jagdzeitung 155 (12): 266-270.

57. Kirby K (1992) Woodland and Wildlife. Whittet, London. 128 p. 58. Lapietra G, Coaloa D and Chiarabaglio PM (1991) Rapporto annuale sulla pioppicoltura 1990.

Cellulosa e Carta, 3: 20-23. 59. Leco Berrocal F (1996) Delimintación y modelización de los espacios adeheados extremeños.

Universidad de Extremadura. Cáceres. 268 pp. 60. Lelle M A, Gold M A (1994) Agroforestry systems for temperate climates: Lessons from

Roman Italy. Forest & Conservation History 38, 118 - 126. 61. Lelle M, Gold M (1992) Agroforestry systems for temperate climates : lessons from roman

Italy, Michigan State University. June 1992.

62. Liagre F (1993). Les pratiques de cultures intercalaires dans la noyeraie fruitière du Dauphiné. Mémoire de Mastère en Sciences Forestières, ENGREF, Montpellier, 80 pages + annexes.

63. Loenen P, Sluiskes GJ and Woltering CM (2001). Een boer maakt nog geen landschap! Agrarisch landschapsonderhoud gefinancierd vanuit toeristisch/ recreatieve sector. Report WUR-Alterra 32.Wageningen, the Netherlands. 34 pp.

64. Lott K (1993) Der historische Obstbau in Deutschland zwischen 1850 und 1910. Geschichte,

Dokumentation, Aussagen für den aktuellen Streuobstbau. Dissertation, Band 1. Berlin,

Humboldt Universität, Fachbereich Agrar- und Gartenbauwissenschaften, 191 p.

65. Lott K (1993). Der historische Obstbau in Deutschland zwischen 1850 und 1910. Geschichte Dokumentation, Aussagen fur den aktuellen Streuobstbau. Dissertation, Band 1. Berlin Humboldt-Universität, Fachbereich Agrar- und Gartenbauwissenschaften, Berlin, Germany. 191 pp.

66. Lucke R, Silbereisen R, Herzberger E (1992) Obstbäume in der Landschaft. Stuttgart, Eugen Ulmer, 300 pp

67. Lucke R, Silbereisen R and Herzberger E (1992). Obstbäume in der Landschaft. Eugen Ulmer Verlag, Stuttgart, 300 pp.

68. MAPA 1985. Anuario de Estadística Agraria 1985. Ministerio de Agricultura, Pesca y Alimentación, Madrid. 660 pp.

69. MAPA 2000. Anuario de Estadística Agroalimentaria 2000. Ministerio de Agricultura, Pesca y Alimentación, Madrid. www.mapya.es

70. Mary F, Delannoy E, Liagre F, Dupraz C, Lifran R (1997). Les planteurs de noyer du Dauphiné : pourquoi optent-ils pour les cultures intercalaires et les noyers double-fin ? La Forêt privée, 40 : 236, 53-65. Fr.

71. Mezzalira G (1999) Le siepi, un gradito ritorno nelle campagne divenute troppo nude dopo la loro eliminazione. Vita in Campagna, 10: 8-34.

72. Mezzalira G (2001) La produzione di legno-energia nell’arboricoltura lineare. L’Informatore Agrario, 34: 21-25.

73. Miguel E, Pointereau P, Steiner C (2000) Los árboles en el espacio agrario. Importancia hidrológica y ecológica. Banco Santander Central Hispano. Madrid. 93 pp.

74. Montero G, San Miguel A, Canellas I (1998). System of mediterranean silviculture “La dehesa”. In: Agricultura sostenible, (R. Jiménez Díaz and J. Lamo de Espinosa eds), Mundi-Prensa, Madrid. Pp 519-554. (English version of the chapter).


80

75. Nevoux L, Knapp N, Richert A (1999. Protection des vergers de poiriers hautes tiges du Domfrontais, CAUE de l’Orne et de la Manche, 86 pages + annexes.

76. Nichols AR (2000) The agronomic, environmental and economic implications of a combined food and energy system. Unpublished Ph.D. thesis. University of Bristol, Long Ashton, UK. 203 p.

77. Nichols AR, Kendall DA and Iles DR (2000). The agronomic and environmental implications of a combined food and energy system. In: Boatman, N.D. (ed.) “Vegetation in changing environments”. Aspects of Applied Biology 58, 363-372.

78. Nill D and Ziegler B (199?) Naturerlebnis Streuobtwiese. DIGITALSKALAR, Gartenstr. 50, 72116 Mössingen, Germany.

79. Osterbaan A (2000). Boeren met bomen.Report WUR-Alterra, ISBN 90-327-02939, Wageningen, the Netherlands. 40 pp.

80. Paris P, Pisanelli A, Buresti E, Musicanti A, Cannata F (2001) Agroforestry in Italy: tradition of the practice and research indications on new models. In: Cannata F, He Guanseng, Raddi P, Zirilli S (eds) Forestry and agroforestry for environmental protection and rural development, Proc of the Sino-Italian workshop in Beijing-China, 2-3 November, 1999, pp 91-100, Centro Promozione Pubblicità, Florence, Italy.

81. Rackham O (1976a) Forests and chases: the King’s wood pasture. In “Trees and woodland in the British landscape”. Dent, London. pp. 152-165.

82. Rackham O (1976b) Trees on the farm: hedges and elsewhere. In “Trees and woodland in the British landscape”. Dent, London. pp. 166-172.

83. Rackham O (1980) Wood-pasture systems and products. In “Ancient woodland its history, vegetation and uses in England”. Edward Arnold, London. pp.173-202.

84. Report DLO Wageningen, the Netherlands. 56 pp. 85. Roach FA (1985) Cultivated Fruits of Britain. Their Origin and History. Blackwell, Oxford. 349

p.86. Rösler M (1993) Vielfalt in aller Munde; Perspektiven für die Bewirtschaftung und

Vermarktung im bundesweiten Streuobstbau. Evangelische Akademie Bad Boll. Gemeinde Boll.

87. Rösler M (1996) Marktwirtschaftliche Bedeutung des Streuobstbaus. Auswirkung des EU-Streuobtsforums im Rahmen der 2. Rhöner Apfelmesse, pp 3-10. Zella, Rhöner Apfelbüro.

88. Roux V (1996) Les formations boisées hors forêt: aspects juridiques et fiscaux. APCA et Ministère de l’Agriculture et de la Pêche, éditeurs, Paris, 144p + annexes.

89. Sanders WJ (2002) Personal communication July 2002.Stichting Robinia, Postbus 494, 6700 AL Wagenigen , the Netherlands.

90. SCAFR (1999) Mise en place d’un statut spécifique pour les parcelles agroforestières. Rapport final, Ministère de l’Agriculture, DERF, Paris, avril 1999.

91. Schroth G (1999) A review of belowground interactions in agroforestry, focussing on mechanisms and management options. Agrorestry Systems 43: p. 5-34.

92. Schultz AM, VP Papanastasis, T Katelman, C Tsiouvaras, S Kandrelis and A Nastis (1987) Agroforestry in Greece. Working document, Thessaloniki.

93. Schulz F (1936) Obstpflanzung und Unterkultur in gegenseitiger Beeinflussung. Landwirtschaftliche Jahrbücher 82(5), 697 – 776.

94. Segouin O, Valadon A (1997) Enquête sur les boisements récents de peupliers en Lot-et-Garonne, Analyse de pratiques agroforestières ; les cultures intercalaires. Cemagref, Nogent-sur Vernisson, 45 pages.


81

95. Simpson JA and Weiner ESC (1989) The Oxford English Dictionary. 2nd Edition. Clarendon Press, Oxford.

96. SOLAGRO (2001). Étude écologique des pré-vergers en Midi-Pyrénées. DIREN Midi-Pyrénées.

97. SRA (1940) Statistisches Reichsamt (Ed.) Obstbaumzählung 1938 und Obsternte 1938. Berlin, Verlag für Sozialpolitik, Wirtschaft und Statistik, Statistik des Deutschen Reiches, Band 541, 369 p.

98. SRFB Languedoc-Roussillon (1998). Recherche d’un statut pour les parcelles agroforestières. Rapport final du groupe de travail sur l’agroforesterie, 8p. + annexes.

99. TERUTI (1998). Evaluation de la répartition des utilisations des sols. Fr.

100. Trenkle R (1944) Einfluss der Unterkulturen im Obstbau auf den Nährstoff- und Wasserhaushalt im Boden und auf das Gedeihen der Obstbäume. Der Wegweiser im Obst- und Gartenbau 42, 16 - 17

101. Vergunst PJB (1996). Nieuwe bossen op landbouwgrond. IBN-DLO, SKR 290. Vakgroep Ecologische Landbouw, Wageningen University, the Netherlands. 57 pp.

102. Vos W, Douw L, Hoogendoorn J, Korevaar H, Pedroli B and Spoelstra SF (1998) Programmeringsstudie miltifunctionele landbouw. Nieuwe wegen in het onderzoek

103. Wahlen F T, Gisiger L (1937) Über den Einfluss des Obstbaumbestandes auf Höhe und Qualität des Ertrages von Wiesland und den Nährstoffzustand der Wiesenböden. Lanwirtschaftliches Jahrbuch der Schweiz 51(3), 274 - 296.

104. Wiersum van KF (2000) Personal communication, 2000, Sub-department of forestry, Wageningen University, P.O. box 342, 6700 AH Wageningen, the Netherlands.


82

5. List of Figures

Figure 1. Silvoarable plots in four plot area classes at the region of municipality of Askio. ................. 26 Figure 2. Schematic representation of the evolution of agricultural landscape in Italy since 1950 as a

consequence of agricultural mechanization and industrialization (drawing: Daniela Salvini). From P. Paris et al., 2001......................................................................................................................................... 32

Figure 3. Recent changes in the surface of the total and intercropped open woodlands in Spain. ......... 43 Figure 4. Recent changes in the number of spread fruit trees in comparison with the total area

occupped by fruit tree orchards (monocrop) in Spain. .................................................................... 45 Figure 5. Importance of the crops in dehesas of Extremadura (Spain). .................................................. 53 Figure 6. “Plan for a fruit orchard of one morgen with root crops or asparagus or raspberry” In the tree

lines pomme fruit (“K”) and stone fruit (“St”) are alternated, tree planting distance is 15 x 10 m (from Lott, 1993, Fig. 12, Reference 229). Similar plans exist for combinations with strawberry, currant, etc. ...................................................................................................................................... 65

Figure 7: Experimental design in pear and apple plantations (Schulz, 1936, Fig. 11, 12)...................... 66 Figure 8 . Yield of kohlrabi and light intensity in unshaded control plots (100 %) and in agroforestry

plots (pear plantation) compared to the control in two subsequent years (Schulz, 1936, Fig. 19). 67 Figure 9. Impact of apple trees (9 x 10 m) on light intensity and yield of bush bean at different

distances from tree row (Schulz, 1936, Fig. 21) ............................................................................. 68 Figure 10. Relative yield and light intensity of early potatoes under pear trees (5 x 5 m) and apple

trees (9 x 10 m) as compared to a control plot without trees (Fig. 24 in Schulz, 1936). ................ 71


83

6. List of the tables

Table 1. Inventory Sheet of Silvoarable Systems (version for France).................................................. 5Table 2 Synthesis of the agroforestry practices in France....................................................................... 11 Table 3 Main silvoarable systems at municipality of Askio, Greece. ..................................................... 24 Table 4. Inventory of traditional silvoarable systems in Italy................................................................. 31Table 5. Living agroforestry systems in the Netherlands........................................................................ 38Table 6. Recent evolution in the surface of the main agroforestry system in Spain. .............................. 42 Table 7. Diminution of the total length of hedgerow in the farms of the different regions/countries in

Europe. ............................................................................................................................................ 44 Table 8. Different types of crop associations considered in the last national agriculture census (1999) in

Spain................................................................................................................................................ 46 Table 9. Surface occupied for different groups of agroforesty systems in Spain in 1997. .................... 47 Table 10. Regional distribution of the number and surface of small orchards in Spain. ........................ 48 Table 11. Comparison of the number of fruit trees planted in monocrop versus the spread tress into and

between arable plots in Spain. ......................................................................................................... 50 Table 12. Comparison of the total cropped land and cropped open woodland in six provinces of Spain,

where the agrosilvopastoral system of dehesa is common.............................................................. 52 Table 13. Main results of the survey on the intercropped dehesas in the Cuatro Lugares county

(Extremadura, Spain). Data from the interview of 26 farmers........................................................ 55 Table 14 APPENDIX List of Species ................................................................................................... 63 Table 15. Vegetable yields of control and agroforestry plots (Apple trees, 9 x 10 m) (Schulz, 1936,

collated from Tables 37a, 37b, 37c). ............................................................................................... 68 Table 16. Values for yield, quality, light, water and temperature availability in the knob celery plots

under agroforestry (apple trees) and in the unshaded control in 1933 (collated from Schulz 1936, Tables 39b, 40, 41, 42, 44). The agroforestry plots between tree rows were subdivided in sections of 2 m. See Figure 7 for the experimental design. .......................................................................... 70

Documents

Extant Silvoarable Practices in Europe