Is there a forest transition outside forests? Trajectories of farm trees and effects on ecosystem services in an agricultural landscape in Eastern Germany
Highlights
• Significant and unexpected 24.8% increase of farm tree cover from 1964 to 2008. • Overall gains in tree cover similar in the socialist and in the post-socialist era. • Accompanied by increases in agriculture-related ecosystem service provision. • First evidence of forest transition outside closed forests in industrialized country. • Drivers of change mainly related to farm policies and environmental consciousness.
Introduction
During the 19th and 20th centuries, most industrial countries experienced a nationwide transformation of land use: from decreasing to expanding forest areas, a phenomenon commonly termed ‘forest transition’ (Lambin and Meyfroidt, 2010). Forest transition theory postulates that a country's forest cover generally declines as it develops socially and economically, but this trend will eventually be reversed and forest cover may actually expand (Mather, 1992). This pattern results in a ‘U-shaped curve’ for forest cover over time (Barbier et al., 2010) and is composed of two separate, overlapping trajectories of land use: (a) a gradual reduction of deforestation and (b) the recovery of forest area after transition (Grainger, 1995). There is ample empirical evidence, both historical and current, for the halting of deforestation and a trend reversal through reforestation and afforestation in several European countries, including Austria (Krausmann, 2006), Denmark (Mather et al., 1998), France (Mather et al., 1999), and Switzerland (Mather and Fairbairn, 2000). In Germany, both in the East and in the West, forest surface has slightly but permanently increased from 1950 to the 2000s. More significant has been a rapid 19% increase of growing stock (and subsequently of carbon stocks) from 1987 to 2002, to an average of 320 m3 ha−1 (Kauppi et al., 2006). The drivers of this process are strongly place-specific, but include technological changes (which reduced pressure to clear additional forest for livestock or crops), the substitution of wood fuels by fossil fuels, the introduction or strengthening of forest legislation, and the abandonment of marginal farmland, which made way for reforestation or afforestation processes in central Europe (Mather, 2001).
Outside closed forests, European rural landscapes exhibit a diversity of tree-based land-use systems (Auclair et al., 2000), but the question of a forest transition within these systems has rarely been studied, and there are no consistent global and (referring to developed countries) national data on the coverage or extent of these systems (Shvidenko et al., 2005). They have been conceptualized as ‘trees outside forests’ (FAO, 2001), ‘farm trees’ (more related to single trees; Arnold and Deewes, 1997), or ‘farm woodlands’ (more related to small forest stands; van der Horst, 2006). The FAO has defined them indirectly as ‘all trees excluded from the definition of forest and other wooded lands’ (FAO, 2000, p. 40). Farm trees and woodlands are common throughout the world, both in traditional cultural landscapes (e.g. hedgerows or wood-pastures) and in recently modified landscapes (e.g. rainforest remnants in Central America) (Manning et al., 2006). They can be the product of spontaneous regrowth or have been planted, domesticated, and cultivated. Farm trees and woodlands represent seminatural habitats within the farmland mosaic and form part of a ‘high-quality agricultural matrix’, which is increasingly coming to the attention of researchers in landscape ecology (Grashof-Bokdam et al., 2009) and conservation biology (Vandermeer and Perfecto, 2007). In this paper, we relate the term ‘farm trees’ both to single trees and small woodlands in agricultural landscapes.
Farm trees have been conceptualized as ‘keystone structures’, because their effect on ecosystem functioning is believed to be disproportionately high relative to the small area occupied by any individual tree (Gibbons et al., 2008). Their ecological importance is particularly high in intensively farmed landscapes (Skaloš and Engstová, 2010). Farm trees create high structural diversity in agricultural landscapes, thereby providing a great number of microhabitats and permitting multi-directional movements of biota across landscapes and ecological networks (Manning et al., 2009). They are also important ecosystem service providers in, for example, buffering groundwater pollution (Ryszkowski and Kedziora, 2007), controlling surface-runoff and soil erosion (Pattanayak and Mercer, 1997), sequestering carbon (Plieninger, 2011), providing biofuels (Plieninger et al., 2006) and fulfilling cultural ecosystem services (McCollin, 2000).
A number of case studies from around the world have reported that farm trees and woodlands are in serious decline (Plieninger and Schaar, 2008, Gibbons et al., 2008). Among the most common threats are direct legal or illegal clearing, a lack of tree regeneration, and pathogens degrading tree health (Manning et al., 2006). Central Europe's agricultural landscapes (in which trees and woodland fragments are embedded) have experienced critical landscape changes in the past fifty years, which have often resulted in a degradation of ecological resilience (Rescia et al., 2010). A series of drivers of rural landscape change has been identified (Verburg et al., 2010), of which the following have potential negative impacts on farm trees: the intensification and upscaling of agriculture through, among other factors, land consolidation programs, increased mechanization, drainage, irrigation, and inputs of agrochemicals (on the most productive agricultural lands); agricultural extensification, marginalization, and overgrowth of abandoned agricultural lands (in areas where socio-economic conditions for agriculture are less favorable); and urban and infrastructural sprawl. A set of countervailing drivers that lead to the appearance of ecologically valuable elements in agricultural landscapes has been termed ‘greening’ (Hersperger and Bürgi, 2009). It includes the maintenance and plantation of trees, which has often been enabled through public funds, nature conservation acts, and compensating measures as well as through the activities of farmers, extension services, local conservation organizations, and land development agencies (Kristensen and Caspersen, 2002, Schleyer and Plieninger, 2011).
In Eastern Germany, farm tree trajectories are assumed to have been critically impacted by the legacy of ‘socialist agriculture’. This model of land use in the former German Democratic Republic (GDR) ideologically and practically promoted agro-commodity production at large scale and triggered extraordinarily comprehensive and rapid pushes for agricultural intensification (Schmidt, 1990). In this context, large-scale landscape interventions (so-called complex meliorations) were performed between 1965 and the first half of the 1970s to allow heavy machinery to operate (Gross, 1996, Philipp, 1997, Schmidt, 1990). Measures included elimination of ‘disturbing’ and ‘inoperable’ hedgerows, groups of trees, alleys etc. and upscaling of plot size up to 300 ha. The 1980s experienced a certain ‘correction stage’, in which – besides other measures – establishment of shelterbelts as a measure against soil erosion was promoted. In 1990, the breakdown of the socialist regime and German unification imposed a profound structural change for agriculture (Philipp, 1997). Property rights on land were privatized or restituted to the legal owners. This step revived a rather fragmented land-ownership structure. Most of the restituted landowners quickly leased their land to restructured and reorganized agricultural cooperatives, whose production goals had been determined before 1990 by the central planning system. The transition process has also resulted in substantially smaller agricultural production units – being now joint stock companies, limited liability companies, or producer cooperatives – with heterogeneous interests, different production portfolios, and economic potential (Laschewski, 1998). The 1992 MacSharry reforms of the Common Agricultural (CAP) of the European Union marked a transition from price support for agricultural produce to direct income support. Since then, farmers in the EU received substantial amounts of direct subsidies that were paid per crop area. For larger cereal farmers, the payments were conditional on setting-aside at least 15–10% of their croplands from 1996 onwards, a regulation that has been suspended since 2007 (Silvis and Lapperre, 2010).
While the overall change of Europe's cultural landscapes has been frequently studied at different scales (see Antrop, 2005, for an overview), the trajectories of farm trees and woodlands in terms of land cover and landscape structure remain little understood. Among the existing studies, most have focused on the effects of hedgerows of different densities, spatial structures, and management intensities on the composition of several plant and animal species groups (for example LeCoeur et al., 1997, Wehling and Diekmann, 2009). Information about the spatial and temporal dynamics and land-use determinants is thus available for hedgerows (e.g. Burel and Baudry, 1990, Deckers et al., 2005, Kristensen, 2003, Kristensen and Caspersen, 2002, Petit et al., 2003), but not for other classes of farm trees and woodlands. Most studies have not distinguished between the effects of tree gains and losses, making the analysis of a potential forest transition difficult. There is a particular lack of studies relating farm tree trajectories with the provision of ecosystem services. The aim of the present study is to fill these gaps by analyzing the spatial–temporal dynamics of farm trees and woodlands in an agricultural landscape in Eastern Germany, from 1964/1965 (henceforth ‘1964’) to 2008, based on aerial photographs and digital orthophotos. Taking a landscape ecological perspective, we: (1) quantify spatial–temporal dynamics of farm trees over a period of 44 years; (2) disentangle processes of gain and loss in the socialist and post-socialist periods of Eastern Germany; (3) assess differences in ecosystem services provided by farm trees between 1964 and 2008, using a set of landscape metrics. The results are interpreted in the light of forest transition theory, focusing on agricultural intensification and extensification in the socialist (between 1964 and 1990) and post-socialist (1990–2008) periods as potential drivers of change. As it takes some time before policy changes translate into observable landscape changes (especially in the case of trees), the studied periods of 1964–1992 and 1992–2008 can be considered congruent with these eras. We aim to provide a perspective on the dynamics of farm trees and woodlands that may be valuable for evaluating woodland patch dynamics in other agroecosystems beyond our study area.
Section snippets
Study area
An area of 28,050 ha within the Upper Lusatia region in Eastern Germany was selected as a case study (Fig. 1). The landscape is dominated by arable land (77% cover), while forests (10%), pastures (5%) and areas of arable land interspersed with significant areas of natural vegetation (5%) occur in fragments. Settlements and traffic infrastructure covered 3% in 2000 (European Environmental Agency, 2011a). Mean elevation is around 170–200 m asl. The area is drained through the Spree river system and
Datasets used
The number, area, and spatial configuration of a population of tree patches was examined through analyses of aerial photographs and digital orthophotographs for the years 1964/1965 (the earliest set of photographs with broad coverage available for the study area) and 2008, as well as a collection of aerial photographs from 1992 that was only available for a part of the study area. The earliest available set of aerial photographs, dated from 1964, was acquired as grayscale imagery at a scale of
Spatial–temporal dynamics, 1964–2008
With a mean cover of 6.9% and 8.7%, farm trees and woodlands covered important parts of the agricultural landscape both in 1964 and 2008. In both periods, there was at least one tree patch in each of the 100 plots studied. However, the proportions of individual farm tree classes were distributed quite heterogeneously (Fig. 2): In 1964, woodlots accounted for only 7.6% of patches in terms of number, but represented 58.9% of total farm tree area. Vice versa, alleys and tree rows covered only 6.4%
Trajectories of change in landscape structure and ecosystem services provision
In this study we observed a substantial increase of overall farm tree cover between 1964 and 2008, but trajectories have been very disparate across the different tree classes. In particular, scattered fruit trees showed an opposite trend and decreased strongly. There was a stronger increase in tree cover in the steep valleys than in the hills and plateaus, indicating a significant interdependence between topography and trajectories of change (Deckers et al., 2005). Most of these valleys are
Conclusions
This study has assessed changes in small landscape elements over two major political periods. In revealing a considerable increase of farm trees and woodlands in an intensive farming region, we seek to put in a new light the prevailing view that scattered trees are in general decline. Rather, we stress that coarse-grained and long-term intensively cultivated agricultural landscapes may have had low tree cover throughout their younger history. Increases in farm tree cover were followed by
Acknowledgements
This study was supported by Grant FKZ 01UU0904A of the German Federal Ministry of Education and Research (BMBF). The funding source was not involved in the study design; the collection, analysis and interpretation of data; the writing of the report; or in the decision to submit the paper to this journal. Chris Hank greatly improved the language of the paper. Fig. 4 is based on 1964 aerial photographs 74-64-097 and 74-64-082 (provided by the German Federal Archive) and on 2008 digital
References (72)
Why landscapes of the past are important for the future
Landscape Urban Plan.
(2005)- et al.
The forest transition: towards a more comprehensive theoretical framework
Land Use Policy
(2010) - et al.
Patterns and drivers of post-socialist farmland abandonment in Western Ukraine
Land Use Policy
(2011) - et al.
Institutional incentives for managing the landscape: inducing cooperation for the production of ecosystem services
Ecol. Econ.
(2007) Farming in former East Germany: past policies and future prospects
Landscape Urban Plan.
(1996)- et al.
Going beyond landscape change description: quantifying the importance of driving forces of landscape change in a Central Europe case study
Land Use Policy
(2009) - et al.
Dynamic land use and land cover changes and their effect on forest resources in a coastal village of Matemwe, Zanzibar, Tanzania
Land Use Policy
(2011) Multivariate analysis of landscape changes and farm characteristics in a study area in central Jutland, Denmark
Ecol. Model.
(2003)- et al.
Analysis of changes in a shelterbelt network landscape in central Jutland, Denmark
J. Environ. Manage.
(2002) - et al.
Land use transitions: socio-ecological feedback versus socio-economic change
Land Use Policy
(2010)