- Email: [email protected]
Mapping and analysing historical indicators of ecosystem services in Germany
Ecological Indicators 75 (2017) 101–110
Contents lists available at ScienceDirect
Ecological Indicators journal homepage: www.elsevier.com/locate/ecolind
Mapping and analysing historical indicators of ecosystem services in Germany Andreas Dittrich a,∗ , Henrik von Wehrden b,c , David J. Abson b , Bartosz Bartkowski d , Anna F. Cord a , Pascal Fust b , Christian Hoyer a , Stephan Kambach e,f , Markus A. Meyer a , ¯ e˙ g,h , Marta Nieto-Romero i , Ralf Seppelt a,j , Michael Beckmann a Rita Radzeviˇciut a
UFZ - Helmholtz Centre for Environmental Research, Department of Computational Landscape Ecology, Permoserstr. 15, 04318 Leipzig, Germany Leuphana University Lüneburg, Faculty of Sustainability, Scharnhorststr. 1, 21355 Lüneburg, Germany c Leuphana University Lüneburg, Centre of Methods, Scharnhorststr. 1, 21335 Lüneburg, Germany d UFZ - Helmholtz Centre for Environmental Research, Department of Economics, Permoserstr. 15, 04318 Leipzig, Germany e UFZ - Helmholtz Centre for Environmental Research, Department of Community Ecology, Theodor-Lieser-Str. 4, 06120 Halle, Germany f Martin-Luther-University Halle-Wittenberg, Department of Biology/Geobotany and Botanical Garden, Am Kirchtor 1, 06108 Halle (Saale), Germany g University of Leipzig, Faculty of Biosciences, Pharmacy and Psychology, Institut of Biology – Molecular Evolution and Animal Systematics, Talstr. 33, 04103 Leipzig, Germany h ˇ Vilnius University, Faculty of Natural Sciences, Department of Zoology, M.K.Ciurlionio str. 21/27, LT-03101, Vilnius, Lithuania i University of Aveiro, Department of Social, Political and Territorial Sciences, Campus Universitario de Santiago, 3810-193 Aveiro, Portugal j Martin-Luther-University Halle-Wittenberg, Institute of Geoscience and Geography, 06099 Halle (Saale), Germany b
a r t i c l e
i n f o
Article history: Received 15 June 2016 Received in revised form 22 September 2016 Accepted 3 December 2016 Keywords: Cultural landscapes Hot spots Landscape identity Literature review Spatial analysis Traditional land-use systems
a b s t r a c t In recent ecosystem service studies, historical data have gained importance as basis for analysing temporal trends and for adapted land management strategies; however, the total number of such studies remains small. Contributing to recent efforts, the primary objective of this study was to assess local ecosystem service products historically used in Germany and to link their distribution patterns to environmental gradients and traditional land-use systems. From maps and detailed regional descriptions of regionally distinct historic farmsteads, building materials used and village types we extracted information on ecosystem service products appropriated in 1950 and before. A spatial model was used to test the derived ecosystem service diversity against topo-climatic conditions. Regional service richness was further compared to the type of traditional land-use system (i.e. focus on crops, focus on livestock or mixed systems). We were able to identify hot spots of historical ecosystem service provisioning in Northern and Southern Germany, whereas significantly lower service numbers were recorded in Eastern Germany. The strong spatial differences in the diversity of historical service products could be explained best by (high) precipitation during the vegetation period. Furthermore, traditional livestock keeping, which relied on various fodder sources and fertilisation techniques to improve poor soil quality, and mixed systems mostly co-occurred with higher regional ecosystem service richness. The baseline of historical ecosystem service provisioning analysed here aids our understanding of current land-use patterns in Germany. Furthermore, a change of perception for specific landscape elements became apparent from our analyses. For example, hedges planted to separate livestock and to provide fuel in the past are today appreciated as important elements for biodiversity conservation. Furthermore, our study helps to preserve knowledge about locally sourced ecosystem services thereby increasing the understanding of cultural landscapes which may help to maintain their remnants. © 2016 Elsevier Ltd. All rights reserved.
1. Introduction
∗ Corresponding author. E-mail address: [email protected] (A. Dittrich). http://dx.doi.org/10.1016/j.ecolind.2016.12.010 1470-160X/© 2016 Elsevier Ltd. All rights reserved.
In recent ecosystem service (hereafter ES) studies, historical data have gained increasing importance in determining tradeoffs and synergies among multiple ES and as basis for adapting ˜ et al., 2013; Renard land management strategies (Morán-Ordónez
102
A. Dittrich et al. / Ecological Indicators 75 (2017) 101–110
et al., 2015; Tomscha and Gergel, 2016). However, the total number of such studies remains small (Plieninger et al., 2016) and the published research are either based on public statistics and infrastructure indicators (Renard et al., 2015), land cover maps and historical aerial photography (Lautenbach et al., 2010; Tomscha and ˜ Gergel, 2016) or literature and technical reports (Morán-Ordónez et al., 2013). Despite the variety of methods applied, they commonly conclude that (i) better baseline information on the past provisioning of ES is needed and that (ii) ignoring time will limit the understanding of complex ES dynamics and interactions (e.g. Renard et al., 2015; Tomscha and Gergel, 2016). The use of historical data may provide insights into the diverse sets of ES that have shaped and maintained agricultural cultural landscapes (Antrop, 2005), which are the outcome of the co-evolution between human society and the environment over time (UNESCO, 2014). Such cultural landscapes are the result of human-induced changes through traditional land-use systems (i.e., practices that are not part of modern, intensive agriculture; Bignal et al., 1995) intended to fulfil societal demands for agro-ecological (ecosystem) products and services (Antrop, 2005; Fisher et al., 2009). Traditional cultural landscapes, found across the globe, contribute to aesthetic qualities (Hartel et al., 2014) and foster genetic, organismal and ecological diversity (e.g. Heath and Tucker, 1995; Herzog, 1998). Furthermore, such landscapes preserve regional agricultural knowledge and the diversification of management systems which provide a buffer against unforeseen stochastic events or disturbances, thereby increasing landscape resilience (Barthel et al., 2013). Multifunctional cultural landscapes are valued for their ecological, social and historic functions (Barthel et al., 2013; Plieninger et al., 2013), yet they are vulnerable to the twin threats of agricultural intensification and abandonment due to their low economic returns and changing perceptions of their value (Hanspach et al., 2014). Both abandonment and intensification have led to a loss of numerous provisioning ES and the related agro-biodiversity worldwide (von Wehrden et al., 2014) and fundamentally altered traditional cultural landscapes. In order to establish strategies to maintain and protect cultural landscapes and their related agro-biodiversity, a better understanding on how such landscapes developed as a result of their environmental conditions and anthro˜ et al., 2013). pogenic use is needed (Farina, 2000; Morán-Ordónez Compiling data on historical provisioning of ES in cultural landscapes as a starting point for detailed (temporal) analyses poses challenges, mainly due to the fact that historical data cannot be directly collected but has to be derived from existing data sources. In particular, spatially explicit data on the provisioning of ES are hard to come by. Historical aerial photography is probably most promising (e.g. Lautenbach et al., 2010; Tomscha and Gergel, 2016) in this context but also restricted by data availability when working across broad spatial scales or over long periods of time. Therefore, there is a need to find proxy indicators that can capture historic ES provision and to relate those services to cultural land forms and environmental conditions. In this study, we apply the ES concept to a dataset on the distribution of historical farmhouses, construction materials used, village and farm types throughout Germany in 1950 and before (Ellenberg, 1990). We extracted information on ES products, which are defined as the goods and benefits derived from ES (HainesYoung and Potschin, 2013). While we identify and investigate some of the interdependencies of ecological and human systems that shape cultural landscapes, our study has two main aims: first, we analyse whether different traditional land-use systems can be related to differences in regional ES richness; second, we explore if the spatial distribution of service diversity can be explained by
environmental conditions such as precipitation, temperature and terrain ruggedness. 2. Data and methods 2.1. Study area Germany is characterized by large topo-climatic gradients (altitude: −3–2962 m a.s.l., mean annual precipitation: 483–2340 mm, mean annual temperature: −3.7 – 11 ◦ C, mean annual sunshine duration: 1376–1873 h) which can be related to the various forms of cultural landscapes and rural construction methods found (Ellenberg, 1990). Traditional land-use systems in Germany until approx. 1800 mainly aimed at the continuous supply of multiple products, rather than on optimizing yield (Beck, 1986). While land close to villages was mainly used for crop production, grasslands and more distant forests were used for livestock keeping (SchulzeHagen, 2004). While the change from natural to rural landscapes was gradual, two major periods of land-use change have been described (e.g. Antrop, 2005; Haase et al., 2007). Within the first period (19th century until the Second World War, Fig. A1), common land was increasingly privatized and used for crop production, chemical fertilizers were first introduced, mechanization of agricultural production began and livestock housing systems became more popular. The second period of land-use change (post-World War landscapes) can be characterized by (1) the intensive use of chemical fertilizers and plant protection products (Spielman and Pandya-Lorch, 2009), (2) land consolidation (Biˇcı´ık et al., 2001), the exchange of small and scattered agricultural areas between different farmers in order to form larger, continuous fields with a single owner (FAO, 2015), (3) further mechanisation and specialisation of agricultural systems and (4) industrial livestock keeping with intensive grassland management (Schulze-Hagen, 2004). Overall, competitive advantages due to environmental conditions, economies of scale in production and the use of external inputs to bolster production led to increased land-use specialization and landscape homogenization (Blaxter and Robertson, 1995). Both periods of change have fundamentally transformed cultural landscapes in Germany and led to a degradation of many ES and a severe loss of biodiversity (e.g. Poschlod et al., 2005). 2.2. Overview of data sources and methods applied Within the study at hand, we used different data sources and methods to answer our two main questions (Fig. 1). Further details on each of the steps are described in the following (see section Maps and Regional descriptions for more information about the data recorded by Ellenberg; see section Spatial regression of historic ecosystem service diversity and environmental variables for environmental variables analysed). 2.2.1. Historical information about rural landscapes in Germany Heinz Ellenberg (1913–1997) was a German botanist, who mainly conducted research in the field of vegetation ecology and developed a 9-point scale in order to rate the preferences of plants for environmental factors (Leuschner, 1997). Beside this work, he was also interested in the temporal evolution of housing types in traditional cultural landscapes. This interest resulted in the publication of his book “Bauernhaus und Landschaft” (“Farmhouse and Landscape”, 1990). By compiling maps, notes and photographs, Ellenberg collected this rich and unique source of data about historical building types, construction materials, farm types, village forms, and landscape elements and their spatial distribution in Germany. He gathered data between 1932 and 1988 and explicitly stated that his aim was to provide historical information about the rural landscapes in 1950 and before throughout Germany. We
A. Dittrich et al. / Ecological Indicators 75 (2017) 101–110
103
Fig. 1. Schematic overview of the data used to address the two aims of the study, including the applied methods. CAR = Conditional Autoregression Model.
used this collection of data as source of information for historical ES products used by the rural population at this time (see below). 2.2.1.1. Maps. The central element of Ellenberg (1990) are 80 maps (Fig. 2), based on either the German topographical map (spatial resolution of approx. 10 × 12 km) or the UTM (Universal Transverse Mercator) coordinate system (10 × 10 km). All observations per grid cell were recorded by Ellenberg on an ordinal scale with five categories: very often, often, medium abundant, rare, and very rare (see Fig. 2). All grid cells which were not surveyed were labelled as ‘missing information’ and were hence excluded from our analysis. 2.2.1.2. Regional descriptions. Ellenberg used the spatial information gathered in his maps for delineating homogenous landscape zones, which differ from each other regarding their building types, material usage and type of villages. In doing so, he differentiated nine regions, 45 subregions (the main reference unit of Ellenberg’s landscape zones) and 97 sub-areas (with distinct differences in minor important characteristics). To further characterise these regions and to complement the information from the maps, he used notes taken in the field and information from the literature that resulted in very detailed regional descriptions including information on e.g., landscape scenery, agricultural management practices, climate and soil conditions. 2.2.2. Identification of indicators for historical ecosystem service products We carefully reviewed both maps and regional descriptions to identify indicators for historical ES. All ES products identified were independently verified by two of the authors. Identification of service products (we applied the CICES v4.3 classification, HainesYoung and Potschin (2013)) was often directly possible, e.g. for hedges used to separate livestock, trees planted around houses as wind protection, orchard meadows, organic material used in walls to apply clay (see Fig. 3), and stable wood in half-timbered constructions (see Fig. 3). In addition, some descriptions and mapped information were indirectly linked to certain service products. As an example, the farm house types (see Fig. 2) were used as proxy for the importance of either crop production or livestock keeping system or the occurrence of houses with certain roof pitches was used as an indicator for the use of straw and reed as roof covering material (see Fig. 3). In total, we found indication for eight ES classes covering 32 ES products. A detailed overview about the identified service products is given in Table 1.
2.2.3. Distribution of traditional land-use systems and ecosystem services richness To get the most complete picture of ES products provided, we compiled a complete list of all ES products for each of the 45 sub-regions based on the regional descriptions (which summarized information from the maps but also provided additional information as some ES were not represented in the maps, see Table 1 and section Regional descriptions). As opposed to the maps, regional descriptions provided only binary information on presence/absence of ES products. We hence determined the richness of ES products at the sub-region level and used flower plots (R package: graphics, R Core Team, 2013) to illustrate spatial differences in ES richness in relation to the dominant traditional land-use system by means of visual comparison (Fig. 5). To identify the dominant traditional land-use system for each of the 45 sub-regions, we analysed those maps providing detailed data on the relative importance of livestock keeping, crop production or mixed land-use systems (as defined in Baldock et al., 1994). In order to finally differentiate between the three main land-use systems, we applied the following rule: if farm houses and village forms typical for livestock keeping (see Fig. 2) were on average more abundant by at least one ordinal scale category as compared to those typical for the production of cereal crops, wine or fruits, the respective sub-region was assigned to the group of livestock keeping systems (analogously for crop production systems). If the difference in abundance was less than one category, sub-regions were assigned to mixed systems (Fig. 5). 2.2.4. Spatial regression of historic ecosystem service diversity and environmental variables For the spatial regression, we had to restrict our analysis to the 15 ES products represented in Ellenberg’s grid-based maps (see Fig. 1, marked ‘M’ in Table 1) to extract spatial explicit accounts of those ES products. By taking into account the relative abundance measurements (ordinal scale from one to five) recorded for every grid cell, we calculated the Shannon diversity (Shannon and Weaver, 1949) of ES products per cell. To ensure spatial compatibility between the two types of maps used by Ellenberg (1990), we resampled the data based on UTM raster maps to match the grid cells of the topographical maps of Germany. For analysing the relationship between these 15 ES products and environmental gradients, we selected soil quality (Roßberg et al., 2007), mean precipitation and temperature during the vegetation period (DWD, 2014), slope, altitude as well as terrain ruggedness (BKG, 2014) as potentially meaningful environmental predictors. The arithmetic mean of each environmental variable was calculated per grid cell
104
A. Dittrich et al. / Ecological Indicators 75 (2017) 101–110
Table 1 Indicators for historical ecosystem service products and descriptions extracted from Ellenberg (1990). Classes, divisions and groups refer to the CICES classification (version 4.3; Haines-Young and Potschin, 2013). Data source refers to either spatial data based on maps (M) or data based on regional descriptions (R), shown in the last column. As more detailed information about the different types of fodder were provided in the regional descriptions than recorded spatially explicit ‘Ma ’ indicates the map combining all types of fodder. Class [Division] (Group)
Short name
Service product
Biomass [Nutrition] (Cultivated crops)
Cultivated crops
Provisioning Cereal crops Vegetables Orchards Orchard meadows Wine Hop
Description
Data source
E.g. wheat and rye E.g. potatoes, pumpkins and vegetables in the garden Fruit production on a big scale Extensive fruit production
M, R R
Vineyards Hop production
R M, R
M, R M, R
Biomass [Nutrition] (Reared animals and their outputs)
Reared animals Wild animals
Honey Fish
Bee keeping Inland fishery
R R
Biomass [Materials] (Fibres and other materials from plants, algae and animals for direct use or processing)
Materials for direct use
Carrier material
Material in wall used as base to apply clay (e.g. straw, bendable wood) Plant material applied between straight wood (e.g. moss) Hemp was used to produce ropes Stable wood that was used for half-timbered constructions (mainly oak, partly spruce) Straight wood that was used for blockhouses (mainly spruce) Timber used for building without specific purpose Weatherproof wood used to protect walls (boards, planks and window shutters) Plant material used to cover roofs (straw, reed)
M, R
Insulation Ropes Stable wood
Straight wood Timber Weather protection Weather protection roofing Aesthetic appreciation Ornamental use Biomass [Materials] (Materials from plants, algae and animals for agricultural use)
Materials for agricultural use
Mulching
Peat Plaggen
River sediments Fodder-hay Fodder-oak Grazing Grazing/fodderhay Hedges
Biomass-based energy sources [Energy] (Plant-based resources)
Plant-based energy
Fire wood-hedges Fire wood-trees Fuel-peat
Gaseous/air flows [Mediation of flows] (Storm protection) No equivalent CICES class
R R M, R
M, R R M, R M, R
Flowers used to decorate house
R
Special wood used for handcrafts
R
Plant material was used as mulch for livestock bedding and put back to arable land Peat was added to soil to increase productivity Calluna vulgaris was extracted from common land and mixed with animal faeces Regular floods increased soil productivity in the floodplains Hay as fodder for livestock (mainly in livestock housing systems) Acorns were used for pig fattening (mainly in pig-housing systems) Livestock was fed with grass (mainly kept on pastures) Livestock was fed with grass (kept on pastures) as well as hay production as fodder (livestock housing systems) Hedges used to separate allotments for livestock keeping
R
Hedges were cut down or pruned regularly, material used as fuel Parts of trees extracted regularly and used as fuel Peat was extracted and used locally as well as transported to bigger cities
M, R M, R
Ma , R Ma , R Ma , R Ma , R
M, R M, R R M, R
Storm protection
Regulation/Maintenance Wind protection Trees and hedges around houses planted as wind protection
R
Fire protection
Fire protection
R
Big trees near houses attracted lightnings and hence protect nearby houses
A. Dittrich et al. / Ecological Indicators 75 (2017) 101–110
105
Fig. 2. Exemplary maps compiled by Ellenberg. Left: UTM raster map with information on the frequency of occurrence of separate barns typical for regions with high importance of crop production. Right: German topographical map with abundance information on certain house forms (so-called “Einfirst-, Winkel-, and T-Höfe”) which are typical farm houses in regions with high importance of livestock keeping. We added the red box to highlight the ordinal scale with five categories: biggest dot = very often and smallest dot = very rare. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) Source of the images: Ellenberg (1990).
Fig. 3. Examples of building materials identified as ecosystem service products. Left: use of reed as roof cover was defined as “Weather protection roofing” and timber used for frame construction of half-timbered construction was classified as “Stable wood”. Right: different organic materials were used in walls as base to apply clay, in this case bendable wood and wooden poles are shown, these materials were classified as “Carrier material”.(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) Source of images: Ellenberg (1990), right panel modified after Bedal (1985).
of the topographical map of Germany (R package: raster; Hijmans et al., 2014). Prior to analysis, all variables were tested for collinearity and due to their strong correlation with precipitation (|r| > 0.7; Dormann et al., 2013) the variables slope, ruggedness and altitude were identified as redundant (see Fig. A2). However, we retained ruggedness as predictor as it was not highly correlated with variables other than precipitation and we assumed that at least one measure of topography may be important for our analysis. Furthermore, ruggedness integrates information captured by slope and elevation change over a defined area (in our case: 1000 m2 ) and is therefore a proxy for the accessibility of a landscape.
The selected variables (soil quality, mean precipitation, temperature and ruggedness) were then standardized in order to allow better comparability of the model outputs (Schielzeth, 2010). To test Shannon diversity of ES against the standardized environmental predictors while accounting for spatial autocorrelation, we applied a spatial conditional autoregression model (CAR, R package: spdep; Bivand et al., 2014), with a neighbourhood size of four (“Rooks case”). This type of model is appropriate for identifying drivers on spatial patterns of ES (Mouchet et al., 2014). The residuals of the model were tested for remaining spatial autocorrelation.
106
A. Dittrich et al. / Ecological Indicators 75 (2017) 101–110
3. Results 3.1. Regional richness of historical ecosystem service products The review of ES products assessed within the scope of this study revealed that the northern sub-regions III-1, I-1 and II-3 as well as the central and southern regions V-6, VI-1, VI-3 and VIII-2 had the highest richness ranging from seven to 14 products (Fig. 4a). With eight ES products, sub-region IV-5 had the highest richness in the eastern part of Germany. In the proximity of northern and southern sub-regions characterised by high numbers of ES products, areas are located with similar average numbers of products (five to six); the same is true for sub-regions in the central part of Germany. The lowest number of ES products was identified for the eastern sub-regions IV-1 and IV-2, most parts of region IX and sub-region III-8 in the south as well as the western subregions V-9 and III-5 with a maximum of three different ES products. “Materials for direct use”, “Materials for agricultural use” and “Cultivated crops” were mentioned in the regional descriptions for nearly all regions. However, despite the comparatively high number of ES products belonging to the class “Materials for direct use” (10 in total, see Table 1), this class contributed more than 50% of the ES products assessed in only half of all sub-regions, indicating large spatial differences in the use of the related products (Fig. A3). Compared to other areas, sub-region VIII-2 was characterized by a high richness of ES products which were used as building materials and belonged to the service class “Materials for direct use” (see flower plots in Fig. 5 and Table A1). The second highest number of such ES products (four) was recorded in the sub-regions V-6, VI-1, VI-2, VI-4 and VII-3 (Fig. 4a). Sub-regions with two to three ES products belonging to the class “Cultivated crops” were mainly located in region III and V. ES products belonging to the classes “Reared animals”, “Wild animals”, “Storm protection” and “Fire protection” appeared to be spatially restricted as they were referenced only in 16% of all sub-regions. For example, the ES product assigned to “Storm protection” could only be identified in the northern regions I and II as well as in VII-2 in the west of Germany (see Fig. 5 and Table A1).
3.2. Distribution of traditional land-use systems Livestock keeping was dominant in 45% of sub-regions and most prominent Northern and Southern Germany whereas crop production was dominant in 21% of sub-regions that were mostly distributed in Eastern and Central Germany. The remaining 34% of sub-regions that were classified as mixed systems were mainly found in Central Germany (see Fig. 5). The most diverse livestock keeping system was found in sub-region III-1 with four ES products referring to different types of fodder (see Table A1) within the class “Materials for Agricultural Use”. In this sub-region, grazingforests for pigs (indicated by “fodder-oak”, see Table 1), livestock housing systems (indicated by “fodder-hay”) as well as outdoor keeping systems (indicated by “grazing”) were utilized. More than one fodder-related service product was also recorded for II-3, V-6 and VI-3. Sub-regions with either high (e.g. I-1, III-1, II-1, VII-2, V-6, and II3) or low (e.g. III-8, IV-2, IX-3, IX-4, and IX-5) numbers of ES classes and products mostly co-occurred with livestock keeping or mixed systems. The only exceptions were IV-5, V-3 and V-7, for which we found high numbers of ES products in crop production systems. Overall, sub-regions with average to lower numbers of ES products tended to be crop production systems, e.g. IV-1, IV-3, V-2, and V-9 (see Fig. 5).
Table 2 Relationship between Shannon diversity of historical ecosystem services and environmental variables (z-transformed) and their interactions at map sheet scale (∼130 km2 ) based on results of the spatial conditional autoregression model (CAR). Coefficients
Estimate
Std. Error
p-value
Soil Quality Ruggedness Mean Temperature during Vegetation Period Mean Precipitation during Vegetation Period Soil Quality: Ruggedness Soil Quality: Mean Temperature Soil Quality: Mean Precipitation Ruggedness: Mean Temperature Ruggedness: Mean Precipitation Mean Temperature: Mean Precipitation
0.026 −0.002 0.017 0.083 0.041 0.003 −0.013 −0.028 −0.027 0.025
0.017 0.030 0.025 0.039 0.026 0.016 0.027 0.014 0.012 0.017
0.135 0.952 0.482 0.032 0.114 0.841 0.633 0.045 0.030 0.127
3.3. Relationship between diversity of historical ecosystem service products and environmental conditions The CAR model was highly significant (p < 0.001) with a R2 of 0.410. In general, precipitation had the strongest positive influence on the service diversity (Fig. 4b, Table 2). Also, the interactions between ruggedness and temperature as well as ruggedness and precipitation were significant (p < 0.05). This means that in areas with flat terrain but with high temperatures or high precipitation, the number of services was greater compared to other regions. The extent of spatial autocorrelation further suggests that other region-specific drivers not specifically considered as predictors in the model were affecting the diversity of provisioning service products. 4. Discussion This study provides the first spatially explicit account of the provision of ES products historically used at a national scale in Germany. By basing our analyses on regional descriptions and maps of farmsteads, village forms, agricultural practices and specific landscape features, we were able to provide further insights on the characteristics and historical configuration of cultural landscapes. We found that a high richness of these products was associated with distinct types of traditional land-use systems (in particular livestock keeping and mixed farming systems). Furthermore, spatial differences in service diversity could be partially explained by environmental gradients. The present study contributes to the restricted numbers of studies that investigated temporal aspects of ˜ et al., ES at local/regional scales (Haase et al., 2007; Morán-Ordónez 2013; Palomo et al., 2014; Renard et al., 2015; Tomscha and Gergel, 2016). 4.1. Historical distribution of ecosystem service products In the past, when traditional land-use systems were more abundant in Germany, large regional differences in the number of service products existed. Based on both the regional descriptions and the maps, we found a hot spot of ES products in North-western Germany (Fig. 4), where livestock keeping and mixed farming systems dominated agricultural production (Fig. 5). These systems relied, for example, on shrubs as fences and sources of fuel, as well as acorns for pig fattening. Since soil fertility in these areas was generally low, heather was used as bedding and converted into fertilizer over winter time. Further, extracted peat served as fertilizer. While these different land-use types certainly represent more multifunctional landscapes, the higher numbers of ES in North-western Germany may also partly have arisen from delayed agricultural intensification up until 1950 (Büssis, 2006) and an overall slower population growth (Ehmer, 2004).
A. Dittrich et al. / Ecological Indicators 75 (2017) 101–110
107
Fig. 4. Historical distribution of ecosystem service products identified based on data provided by Ellenberg (1990) and classified according to CICES v4.3 (Haines-Young and Potschin, 2013). Panel a: Richness in ecosystem service products per sub-region. Services (in total 29) were identified from regional descriptions (including notes and photographs). Panel b: Diversity in ecosystem service products per grid cell (∼130 km2 ). Spatial accounts of services (in total 15) were extracted from maps. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
The spatial model helped us explain the lower numbers of ES in the central part of Germany, as precipitation during the growing season had an overall positive influence on ES numbers. Drier areas can be found in the eastern part of this region, which was dominated by crop production − also due to its extremely high soil fertility. Here, most of the building types relied on bricks and stones and not on organic material, which led to a lower number of ES implicitly recorded by Ellenberg. Precipitation also determines the range of oak species (Ellenberg and Leuschner, 2010) needed for half-timbered constructions, which resulted in slightly higher numbers of ES in the western part of this region, characterized by more rainfall. In contrast to our assumption that in areas with a complex topography the number of ES would be higher, terrain was not a good predictor for the distribution of ES. To improve the spatial model, further predictors such as socioeconomic data that capture human and cultural developments could be included. Specifically, data on population density, migration and growth, income or the distribution of ethnic minorities (e.g. Sorbs, Frisians) could prove particularly valuable in future analyses, e.g. such as done by Dunford et al. (2015). This is supported by Hartel et al. (2014) who showed that poverty is the main driver for the existence of traditional land-use systems that feature high ES diversity in Romania. However, to our knowledge, spatially explicit data on socio-economic variables between the 19th century and the Second World War in Germany are not available.
4.2. Challenges of mapping historical ecosystem service indicators Spatially explicit data are needed to establish causal relationships between historical ES provisioning and local environmental and socio-economic variables (Mouchet et al., 2014). Aerial photography, which assists topographic mapping since the 1920s, is obviously a promising approach and widely used in historical landscape analysis (e.g. Dittrich et al., 2011). However, data availability at larger scales is limited and data processing is time consuming (e.g. geographical rectification, digitizing by hand). Other promising approaches are intensive literature reviews as conducted by ˜ et al. (2013) which can provide detailed informaMorán-Ordónez tion on how differently cultural landscape sustained the livelihood of their rural populations. Even though this study assessed a total of 32 service products historically used in Germany, this set is far from being comprehensive. While provisioning services could be considered extensively, cultural, regulation and maintenance services were only scarcely accounted for. Due to this inherent bias in the data source, we were not able to establish a causal relationship between the type of traditional land-use system and the number of ES. To achieve a more complete account of ES, other data sources would have to be considered. For example, to identify regulating services such as flood mitigation, historical land cover data could be analysed (e.g. Früh-Müller et al., 2014).
108
A. Dittrich et al. / Ecological Indicators 75 (2017) 101–110
Fig. 5. Spatial distribution patterns of ecosystem service products (grouped by ecosystem service classes after CICES; see Table 1 for further details) for each of the sub-regions defined by Ellenberg (1990). Petal length in the flower plots indicates differences in service richness per service class among sub-regions. Colour codes for the flower plots refer to the main service classes. Grey bars show abundance of livestock keeping and crop production systems.
A. Dittrich et al. / Ecological Indicators 75 (2017) 101–110
4.3. Relevance for today’s ecosystem management and environmental policies Despite the limitations noted above, the data source and analysis presented here provided insights in the provisioning of ES in historical landscapes partly not existent anymore due to intensive land-use changes since then. This baseline of ES provisioning aids our understanding of the development of current land-use patterns, as for example North-western Germany is still known for intensive livestock production whereas crop production is important for Eastern Germany (Dittrich et al. unpublished). Furthermore, by comparing past and present utilization of services a better understanding regarding the processes leading to the generation and ˜ et al., 2013; Renard extraction of ES can be gained (Morán-Ordónez et al., 2015) making the loss of certain ES traceable. In that respect, understanding the historical co-evolution of cultural landscapes and ecological resource use (here conceptualized in terms of ES) may be important to preserve or regain multifunctional landscapes. If we wish to maintain the diversity and unique character of cultural landscapes in the face of increasing temporally and spatially homogeneous agricultural land-use practices then there is a need for these cultural landscapes to replicate the multi-faceted and multifunctional uses from which such landscapes arose. Therefore, it would be highly relevant to compare historical and legally-proposed land-use options for their impact on the multifunctionality of landscapes. A large number of subsidy schemes (e.g. Thomas et al., 2009; Schleyer and Plieninger 2011; Ulrich and Riecken, 2012) for historical land-use options equally show that knowledge on their impact on ES is necessary to identify more effective and regionally adapted options. A better understanding of the ES such landscape provided and consideration of how these unique landscapes could provide similar suites of services suitable for the 21st century would help to restore cultural landscapes. Their values come not only from their cultural importance, but also form the flows of services they provide. However, most of today’s subsidies and ‘greening’ measures (Common Agricultural Policy; BMEL, 2015) aim at protecting biodiversity aspects (e.g. hedges planted to separate livestock in the past are important for biodiversity today; Ponti et al., 2005) that co-evolved with cultural landscapes. As societal demands on landscapes are changing continuously (e.g. Haase et al., 2007) it is very likely that “new” regulating/maintenance as well as cultural services are wanted by society, which in turn may create trade-offs with the conservation targets of the present subsidies. Balancing these trade-offs would require workshops with all stakeholders involved (Förster et al., 2015; Seppelt and Cumming, 2016).
5. Conclusion and outlook The countryside in Europe is becoming less and less a place that provides a livelihood for the majority of the inhabitants, thereby “landscape and rural life are becoming ominously disjoined” (Lowenthal, 1997). The presented study for historical ES provisioning helps to understand how landscapes have been used and developed in the past in Germany, a task for which the ES concept proved to be a useful lens through which to assess the benefits for society provided by cultural landscapes. While the present study focuses on Germany, it may be regarded as representative for many regions in Europe. We further suggest that the investigation of historical ES could be a valuable approach to clarify the identity and functionality of cultural landscapes in the changing spatial context of society (Antrop, 2005). As the existence of cultural ˜ et al., landscapes depends on their active use (e.g. Morán-Ordónez 2013), assessing historical ES may also provide incentives for the maintenance of such landscapes. This is partly realized by current
109
subsidy programs, which aim at protecting former ES in cultural landscapes to foster the conservation of agro-biodiversity and landscape aesthetics. However, more sustainable options to reconnect people to the rural landscape should also be explored. To support acceptance and application of these incentives we advocate to also appeal to the history and former diversity of ES that shaped the homeland of today’s residents. Acknowledgements The presented study is the outcome of a Ph.D. synthesis workshop of the Helmholtz Research School for Ecosystem Services under Changing Land Use and Climate (ESCALATE) at the Helmholtz Centre for Environmental Research – UFZ and the Leuphana University Lunenburg. The work was funded partly by the Helmholtz Programme ‘Terrestrial Environment’. We thank Ludwig Ellenberg who approved the usage of the data gathered by his father Heinz Ellenberg. We further thank Wulf Jung from GISCON for providing digital versions of the maps published in the book as well as Dietmar Roßberg for providing a Germany-wide soil quality map. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ecolind.2016. 12.010. References Antrop, M., 2005. Why landscapes of the past are important for the future. Landsc. Urban Plann. 70, 21–34, http://dx.doi.org/10.1016/j.landurbplan.2003.10.002. Büssis, H., 2006. Die Theorie von der extensiven Allmendenutzung-eine ükologische Fehlinterpretation! Charadrius 42, 1–8. BKG (Bundesamt für Kartographie und Geodäsie), 2014. Digitale Geländemodelle [WWW Document]. https://upd.geodatenzentrum.de/geodaten/gdz rahmen. gdz div (Accessed 1 December 2014). BMEL (Bundesministerium für Ernährung und Landwirtschaft), 2015. Umsetzung der EU-Agrarreform in Deutschland - Ausgabe 2015 [WWW Document], http://www.bmel.de/SharedDocs/Downloads/Broschueren/ UmsetzungGAPinD.html (Accessed 21 July 2015). Baldock, D., Beaufoy, G., Dark, J., 1994. The Nature of Farming: Low Intensity Farming Systems in Nine European Countries. Institute for European Environmental Policy, London. Barthel, S., Crumley, C., Svedin, U., 2013. Bio-cultural refugia—safeguarding diversity of practices for food security and biodiversity. Glob. Environ. Change 23, 1142–1152, http://dx.doi.org/10.1016/j.gloenvcha.2013.05.001. Beck, R., 1986. Naturale Ökonomie: Unterfinning, bäuerliche Wirtschaft in einem oberbayerischen Dorf des frühen 18. Jahrhunderts. Deutscher Kunstverlag, München. Bedal, K., 1985. Häuser aus Franken. In: Museumsführer Fränkisches Freilichtmuseum in Bad Windsheim, second ed. Delph’sche Verlagsbuchhandlung, München. ˇ epánek, V., 2001. Land-use changes and their social driving Biˇcı´ık, I., Jeleˇcek, L., Stˇ forces in Czechia in the 19th and 20th centuries. Land Use Policy 18, 65–73, http://dx.doi.org/10.1016/S0264-8377(00)00047-8. Bignal, E., McCracken, D., Corrie, H., 1995. Defining European low-intensity farming systems: the nature of farming. In: McCracken, D., Bignal, E., Wenlock, S. (Eds.), Farming on the Edge: The Nature of Traditional Farmland in Europe. Joint Nature Conservation Committee, Peterborough, pp. 29–37. Bivand, R., Altman, M., Anselin, L., Assunc¸ão, R., Berke, O., Bernat, A., Blanchet, G., Blankmeyer, E., Carvalho, M., Christensen, B., Chun, Y., Dormann, C., Dray, S., Gómez-Rubio, V., Halbersma, R., Krainski, E., Legendre, P., Lewin-Koh, N., Li, H., Ma, J., Millo, G., Mueller, W., Ono, H., Peres-Neto, P., Piras, G., Reder, M., Tiefelsdorf, M., Yu, D., 2014. spdep: Spatial Dependence: Weighting Schemes, Statistics and Models. Blaxter, K., Robertson, N., 1995. From Dearth to Plenty: The Second Agricultural Revolution. Cambridge University Press, Cambridge. DWD (Deutscher Wetterdienst), 2014. GISC (Global Information System Centre) [WWW Document]. http://gisc.dwd.de/GISC DWD/start js JSP.do (Accessed 12 January 2014). Dittrich, A., Buerkert, A., Brinkmann, K., 2011. Assessment of land use and land cover changes during the last 50 years in oases and surrounding rangelands of Xinjiang, NW China. J. Agric. Rural Dev. Tropics Subtropics 111, 129–142 http://nbn-resolving.de/urn:nbn:de:hebis:34-2011052737598. Dormann, C.F., Elith, J., Bacher, S., Buchmann, C., Carl, G., Carré, G., Marquéz, J.R.G., Gruber, B., Lafourcade, B., Leitão, P.J., Münkemüller, T., McClean, C., Osborne, P.E., Reineking, B., Schröder, B., Skidmore, A.K., Zurell, D., Lautenbach, S., 2013.
110
A. Dittrich et al. / Ecological Indicators 75 (2017) 101–110
Collinearity: a review of methods to deal with it and a simulation study evaluating their performance. Ecography 36, 27–46, http://dx.doi.org/10.1111/ j.1600-0587.2012.07348.x. Dunford, R.W., Smith, A.C., Harrison, P.A., Hanganu, D., 2015. Ecosystem service provision in a changing Europe: adapting to the impacts of combined climate and socio-economic change. Landsc. Ecol. 30, 443–461, http://dx.doi.org/10. 1007/s10980-014-0148-2. Ehmer, J., 2004. Bevölkerungsgeschichte und Historische Demographie 1800–2000. Oldenbourg, München. Ellenberg, H., Leuschner, C., 2010. Vegetation Mitteleuropas mit den Alpen. In: ökologischer, dynamischer und historischer, sixth ed. Ulmer, Stuttgart. Ellenberg, H., 1990. Bauernhaus and Landschaft: in ökologischer und historischer Sicht. Ulmer, Stuttgart. Förster, J., Barkmann, J., Fricke, R., Hotes, S., Kleyer, M., Kobbe, S., Kübler, D., Rumbauer, M., Seppelt, R., Settele, J., Spangenberg, J.H., Tekken, V., Václavík, T., Wittmer, H., 2015. Assessing ecosystem services for informing land-use decisions: a problem-oriented approach. Ecol. Soc. 20 (3), 31, http://dx.doi.org/ 10.5751/ES-07804-200331. FAO (Food and Agriculture Organization of the United Nations), 2015. What is land consolidation [WWW Document]. http://www.fao.org/docrep/006/y4954e/ y4954e06.htm (Accessed 21 July 2015). Farina, A., 2000. The cultural landscape as a model for the integration of ecology and economics. BioScience 50, 313–320, http://dx.doi.org/10.1641/00063568(2000)050[0313:TCLAAM]2.3.CO;2. Fisher, B., Turner, R.K., Morling, P., 2009. Defining and classifying ecosystem services for decision making. Ecol. Econ. 68, 643–653, http://dx.doi.org/10. 1016/j.ecolecon.2008.09.014. Früh-Müller, A., Wegmann, M., Koellner, T., 2014. Flood exposure and settlement expansion since pre-industrial times in 1850 until 2011 in north Bavaria, Germany. Regional Environ. Change 15, 183–193, http://dx.doi.org/10.1007/ s10113-014-0633-9. Haase, D., Walz, U., Neubert, M., Rosenberg, M., 2007. Changes to Central European landscapes—analysing historical maps to approach current environmental issues, examples from Saxony, Central Germany. Land Use Policy 24, 248–263, http://dx.doi.org/10.1016/j.landusepol.2005.09.003. Haines-Young, R., Potschin, M., 2013. Common International Classification of Ecosystem Services (CICES): Consultation on Version 4, August–December 2012. EEA Framework Contract No EEA. Contract No EEA/IEA/09/003. Hanspach, J., Hartel, T., Milcu, A.I., Mikulcak, F., Dorresteijn, I., Loos, J., von Wehrden, H., Kuemmerle, T., Abson, D., Kovács-Hostyánszki, A., Báldi, A., Fischer, J., 2014. A holistic approach to studying social-ecological systems and its application to southern Transylvania. Ecol. Soc. 19 (4), 32, http://dx.doi.org/ 10.5751/ES-06915-190432. Hartel, T., Fischer, J., Câmpeanu, C., Milcu, A.I., Hanspach, J., Fazey, I., 2014. The importance of ecosystem services for rural inhabitants in a changing cultural landscape in Romania. Ecol. Soc. 19 (2), 42, http://dx.doi.org/10.5751/ES06333-190242. Heath, M.F., Tucker, G.M., 1995. Ornithological value and pastoral farming systems. In: McCracken, D., Bignal, E., Wenlock, S. (Eds.), Farming on the Edge: The Nature of Traditional Farmland in Europe. Joint Nature Conservation Committee, Peterborough, pp. 54–59. Herzog, F., 1998. Streuobst: a traditional agroforestry system as a model for agroforestry development in temperate Europe. Agrofor. Syst. 42, 61–80, http://dx.doi.org/10.1023/A:1006152127824. Hijmans, R.J., Etten, van, J., Mattiuzzi, M., Sumner, M., Greenberg, J.A., Lamigueiro, O.P., Bevan, A., Racine, E.B., Shortridge, A., 2014. Raster: Geographic data analysis and modelling. R Package Version 2, 2–12 http://CRAN.R-project.org/ package=raster. Lautenbach, S., Kugel, C., Lausch, A., Seppelt, R., 2010. Analysis of historic changes in regional ecosystem service provisioning using land use data. Ecol. Indic. 11, 676–687, http://dx.doi.org/10.1016/j.ecolind.2010.09.007. Leuschner, C., 1997. Heinz Ellenberg (1913–1997). Phytocoenologia 4, 457–462, http://dx.doi.org/10.1127/phyto/27/1997/457. Lowenthal, D., 1997. European landscape transformations: the rural residue. In: Groth, P.E., Bressi, T.W. (Eds.), Understanding Ordinary Landscapes. Yale University Press, Yale, pp. 180–188. ˜ Morán-Ordónez, A., Bugter, R., Suárez-Seoane, S., Luis, E., de Calvo, L., 2013. Temporal changes in socio-ecological systems and their impact on ecosystem services at different governance scales: a case study of heathlands. Ecosystems 16, 765–782, http://dx.doi.org/10.1007/s10021-013-9649-0.
Mouchet, M.A., Lamarque, P., Martín-López, B., Crouzat, E., Gos, P., Byczek, C., Lavorel, S., 2014. An interdisciplinary methodological guide for quantifying associations between ecosystem services. Glob. Environ. Change 28, 298–308, http://dx.doi.org/10.1016/j.gloenvcha.2014.07.012. Palomo, I., Martín-López, B., Zorrilla-Miras, P., Del Amo, D.G., Montes, C., 2014. ˜ Deliberative mapping of ecosystem services within and around Donana National Park (SW Spain) in relation to land use change. Reg. Environ. Change 14, 237–251, http://dx.doi.org/10.1007/s10113-013-0488-5. Plieninger, T., Bieling, C., Ohnesorge, B., Schaich, H., Schleyer, C., Wolff, F., 2013. Exploring futures of ecosystem services in cultural landscapes through participatory scenario development in the Swabian Alb, Germany. Ecol. Soc. 18 (3), 39, http://dx.doi.org/10.5751/ES-05802-180339. Plieninger, T., Draux, H., Fagerholm, N., Bieling, C., Bürgi, M., Kizos, T., Kuemmerle, T., Primdahl, J., Verburg, P.H., 2016. The driving forces of landscape change in Europe: a systematic review of the evidence. Land Use Policy 57, 204, http:// dx.doi.org/10.1016/j.landusepol.2016.04.040. Ponti, L., Ricci, C., Veronesi, F., Torricelli, R., 2005. Natural hedges as an element of functional biodiversity in agroecosystems: the case of a Central Italy vineyard. Bull. Insectol. 58, 19–23. Poschlod, P., Bakker, J.P., Kahmen, S., 2005. Changing land use and its impact on biodiversity. Basic Appl. Ecol. 6, 93–98, http://dx.doi.org/10.1016/j.baae.2004. 12.001. R Core Team, 2013. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. Renard, D., Rhemtulla, J.M., Bennett, E.M., 2015. Historical dynamics in ecosystem service bundles. Proc. Natl. Acad. Sci. 112, 13411–13416, http://dx.doi.org/10. 1073/pnas.1502565112. Roßberg, D., Michel, V., Graf, R., Neukampf, R., 2007. Definition von Boden-Klima-Räumen für die Bundesrepublik Deutschland. Nachrichtenblatt des deutschen Pflanzenschutzdienstes 59, 155–161. Schielzeth, H., 2010. Simple means to improve the interpretability of regression coefficients. Methods Ecol. Evol. 1, 103–113, http://dx.doi.org/10.1111/j.2041210X.2010.00012.x. Schleyer, C., Plieninger, T., 2011. Obstacles and options for the design and implementation of payment schemes for ecosystem services provided through farm trees in Saxony, Germany. Eviron. Conserv. 38, 454–463, http://dx.doi. org/10.1017/S0376892911000361. Schulze-Hagen, K., 2004. Allmenden und ihr Vogelreichtum – Wandel von Landschaft, Landwirtschaft und Avifauna in den letzten 250 Jahren. Charadrius 40, 97–121. Seppelt, R., Cumming, G.S., 2016. Humanity’s distance to nature: time for environmental austerity? Landsc. Ecol. 31, 1645–1651, http://dx.doi.org/10. 1007/s10980-016-0423-5. Shannon, C.E., Weaver, W., 1949. The Mathematical Theory of Communication. University of Illinois Press, Urbana and Chicago. Spielman, D.J., Pandya-Lorch, R., 2009. Millions Fed: Proven Successes in Agricultural Development. International Food Policy Research Institute, Washington D.C. Thomas, F., Denzel, K., Hartman, E.L.R., Schmoock, K., 2009. Kurzfassung der Agrarumwelt- und Naturschutzprogramme. In: Darstellung und Analyse der Entwicklung von Maßnahmen der Agrarumwelt- und Naturschutzprogramme in der Bundesrepublik Deutschland. BfN –Skripten 253. Bundesamt für Naturschutz, Bonn. Tomscha, S., Gergel, S., 2016. Ecosystem service trade-offs and synergies misunderstood without landscape history. Ecol. Soc. 21 (1), 43, http://dx.doi. org/10.5751/ES-08345-210143. UNESCO (United Nations Educational, Scientific and Cultural Organization), 2014. Biosphere Reserves [WWW Document], Accessed 1 December 2014 http:// www.unesco.org/new/en/natural-sciences/environment/ecological-sciences/ biosphere-reserves/. Ulrich, K., Riecken, U., 2012. Moorschutzstrategien, –initiativen und –programme in Deutschland. Natur und Landschaft 87, 81–86. von Wehrden, H., Abson, D.J., Beckmann, M., Cord, A.F., Klotz, S., Seppelt, R., 2014. Realigning the land-sharing/land-sparing debate to match conservation needs: considering diversity scales and land-use history. Landsc. Ecol. 29, 941–948, http://dx.doi.org/10.1007/s10980-014-0038-7.
Contents lists available at ScienceDirect
Ecological Indicators journal homepage: www.elsevier.com/locate/ecolind
Mapping and analysing historical indicators of ecosystem services in Germany Andreas Dittrich a,∗ , Henrik von Wehrden b,c , David J. Abson b , Bartosz Bartkowski d , Anna F. Cord a , Pascal Fust b , Christian Hoyer a , Stephan Kambach e,f , Markus A. Meyer a , ¯ e˙ g,h , Marta Nieto-Romero i , Ralf Seppelt a,j , Michael Beckmann a Rita Radzeviˇciut a
UFZ - Helmholtz Centre for Environmental Research, Department of Computational Landscape Ecology, Permoserstr. 15, 04318 Leipzig, Germany Leuphana University Lüneburg, Faculty of Sustainability, Scharnhorststr. 1, 21355 Lüneburg, Germany c Leuphana University Lüneburg, Centre of Methods, Scharnhorststr. 1, 21335 Lüneburg, Germany d UFZ - Helmholtz Centre for Environmental Research, Department of Economics, Permoserstr. 15, 04318 Leipzig, Germany e UFZ - Helmholtz Centre for Environmental Research, Department of Community Ecology, Theodor-Lieser-Str. 4, 06120 Halle, Germany f Martin-Luther-University Halle-Wittenberg, Department of Biology/Geobotany and Botanical Garden, Am Kirchtor 1, 06108 Halle (Saale), Germany g University of Leipzig, Faculty of Biosciences, Pharmacy and Psychology, Institut of Biology – Molecular Evolution and Animal Systematics, Talstr. 33, 04103 Leipzig, Germany h ˇ Vilnius University, Faculty of Natural Sciences, Department of Zoology, M.K.Ciurlionio str. 21/27, LT-03101, Vilnius, Lithuania i University of Aveiro, Department of Social, Political and Territorial Sciences, Campus Universitario de Santiago, 3810-193 Aveiro, Portugal j Martin-Luther-University Halle-Wittenberg, Institute of Geoscience and Geography, 06099 Halle (Saale), Germany b
a r t i c l e
i n f o
Article history: Received 15 June 2016 Received in revised form 22 September 2016 Accepted 3 December 2016 Keywords: Cultural landscapes Hot spots Landscape identity Literature review Spatial analysis Traditional land-use systems
a b s t r a c t In recent ecosystem service studies, historical data have gained importance as basis for analysing temporal trends and for adapted land management strategies; however, the total number of such studies remains small. Contributing to recent efforts, the primary objective of this study was to assess local ecosystem service products historically used in Germany and to link their distribution patterns to environmental gradients and traditional land-use systems. From maps and detailed regional descriptions of regionally distinct historic farmsteads, building materials used and village types we extracted information on ecosystem service products appropriated in 1950 and before. A spatial model was used to test the derived ecosystem service diversity against topo-climatic conditions. Regional service richness was further compared to the type of traditional land-use system (i.e. focus on crops, focus on livestock or mixed systems). We were able to identify hot spots of historical ecosystem service provisioning in Northern and Southern Germany, whereas significantly lower service numbers were recorded in Eastern Germany. The strong spatial differences in the diversity of historical service products could be explained best by (high) precipitation during the vegetation period. Furthermore, traditional livestock keeping, which relied on various fodder sources and fertilisation techniques to improve poor soil quality, and mixed systems mostly co-occurred with higher regional ecosystem service richness. The baseline of historical ecosystem service provisioning analysed here aids our understanding of current land-use patterns in Germany. Furthermore, a change of perception for specific landscape elements became apparent from our analyses. For example, hedges planted to separate livestock and to provide fuel in the past are today appreciated as important elements for biodiversity conservation. Furthermore, our study helps to preserve knowledge about locally sourced ecosystem services thereby increasing the understanding of cultural landscapes which may help to maintain their remnants. © 2016 Elsevier Ltd. All rights reserved.
1. Introduction
∗ Corresponding author. E-mail address: [email protected] (A. Dittrich). http://dx.doi.org/10.1016/j.ecolind.2016.12.010 1470-160X/© 2016 Elsevier Ltd. All rights reserved.
In recent ecosystem service (hereafter ES) studies, historical data have gained increasing importance in determining tradeoffs and synergies among multiple ES and as basis for adapting ˜ et al., 2013; Renard land management strategies (Morán-Ordónez
102
A. Dittrich et al. / Ecological Indicators 75 (2017) 101–110
et al., 2015; Tomscha and Gergel, 2016). However, the total number of such studies remains small (Plieninger et al., 2016) and the published research are either based on public statistics and infrastructure indicators (Renard et al., 2015), land cover maps and historical aerial photography (Lautenbach et al., 2010; Tomscha and ˜ Gergel, 2016) or literature and technical reports (Morán-Ordónez et al., 2013). Despite the variety of methods applied, they commonly conclude that (i) better baseline information on the past provisioning of ES is needed and that (ii) ignoring time will limit the understanding of complex ES dynamics and interactions (e.g. Renard et al., 2015; Tomscha and Gergel, 2016). The use of historical data may provide insights into the diverse sets of ES that have shaped and maintained agricultural cultural landscapes (Antrop, 2005), which are the outcome of the co-evolution between human society and the environment over time (UNESCO, 2014). Such cultural landscapes are the result of human-induced changes through traditional land-use systems (i.e., practices that are not part of modern, intensive agriculture; Bignal et al., 1995) intended to fulfil societal demands for agro-ecological (ecosystem) products and services (Antrop, 2005; Fisher et al., 2009). Traditional cultural landscapes, found across the globe, contribute to aesthetic qualities (Hartel et al., 2014) and foster genetic, organismal and ecological diversity (e.g. Heath and Tucker, 1995; Herzog, 1998). Furthermore, such landscapes preserve regional agricultural knowledge and the diversification of management systems which provide a buffer against unforeseen stochastic events or disturbances, thereby increasing landscape resilience (Barthel et al., 2013). Multifunctional cultural landscapes are valued for their ecological, social and historic functions (Barthel et al., 2013; Plieninger et al., 2013), yet they are vulnerable to the twin threats of agricultural intensification and abandonment due to their low economic returns and changing perceptions of their value (Hanspach et al., 2014). Both abandonment and intensification have led to a loss of numerous provisioning ES and the related agro-biodiversity worldwide (von Wehrden et al., 2014) and fundamentally altered traditional cultural landscapes. In order to establish strategies to maintain and protect cultural landscapes and their related agro-biodiversity, a better understanding on how such landscapes developed as a result of their environmental conditions and anthro˜ et al., 2013). pogenic use is needed (Farina, 2000; Morán-Ordónez Compiling data on historical provisioning of ES in cultural landscapes as a starting point for detailed (temporal) analyses poses challenges, mainly due to the fact that historical data cannot be directly collected but has to be derived from existing data sources. In particular, spatially explicit data on the provisioning of ES are hard to come by. Historical aerial photography is probably most promising (e.g. Lautenbach et al., 2010; Tomscha and Gergel, 2016) in this context but also restricted by data availability when working across broad spatial scales or over long periods of time. Therefore, there is a need to find proxy indicators that can capture historic ES provision and to relate those services to cultural land forms and environmental conditions. In this study, we apply the ES concept to a dataset on the distribution of historical farmhouses, construction materials used, village and farm types throughout Germany in 1950 and before (Ellenberg, 1990). We extracted information on ES products, which are defined as the goods and benefits derived from ES (HainesYoung and Potschin, 2013). While we identify and investigate some of the interdependencies of ecological and human systems that shape cultural landscapes, our study has two main aims: first, we analyse whether different traditional land-use systems can be related to differences in regional ES richness; second, we explore if the spatial distribution of service diversity can be explained by
environmental conditions such as precipitation, temperature and terrain ruggedness. 2. Data and methods 2.1. Study area Germany is characterized by large topo-climatic gradients (altitude: −3–2962 m a.s.l., mean annual precipitation: 483–2340 mm, mean annual temperature: −3.7 – 11 ◦ C, mean annual sunshine duration: 1376–1873 h) which can be related to the various forms of cultural landscapes and rural construction methods found (Ellenberg, 1990). Traditional land-use systems in Germany until approx. 1800 mainly aimed at the continuous supply of multiple products, rather than on optimizing yield (Beck, 1986). While land close to villages was mainly used for crop production, grasslands and more distant forests were used for livestock keeping (SchulzeHagen, 2004). While the change from natural to rural landscapes was gradual, two major periods of land-use change have been described (e.g. Antrop, 2005; Haase et al., 2007). Within the first period (19th century until the Second World War, Fig. A1), common land was increasingly privatized and used for crop production, chemical fertilizers were first introduced, mechanization of agricultural production began and livestock housing systems became more popular. The second period of land-use change (post-World War landscapes) can be characterized by (1) the intensive use of chemical fertilizers and plant protection products (Spielman and Pandya-Lorch, 2009), (2) land consolidation (Biˇcı´ık et al., 2001), the exchange of small and scattered agricultural areas between different farmers in order to form larger, continuous fields with a single owner (FAO, 2015), (3) further mechanisation and specialisation of agricultural systems and (4) industrial livestock keeping with intensive grassland management (Schulze-Hagen, 2004). Overall, competitive advantages due to environmental conditions, economies of scale in production and the use of external inputs to bolster production led to increased land-use specialization and landscape homogenization (Blaxter and Robertson, 1995). Both periods of change have fundamentally transformed cultural landscapes in Germany and led to a degradation of many ES and a severe loss of biodiversity (e.g. Poschlod et al., 2005). 2.2. Overview of data sources and methods applied Within the study at hand, we used different data sources and methods to answer our two main questions (Fig. 1). Further details on each of the steps are described in the following (see section Maps and Regional descriptions for more information about the data recorded by Ellenberg; see section Spatial regression of historic ecosystem service diversity and environmental variables for environmental variables analysed). 2.2.1. Historical information about rural landscapes in Germany Heinz Ellenberg (1913–1997) was a German botanist, who mainly conducted research in the field of vegetation ecology and developed a 9-point scale in order to rate the preferences of plants for environmental factors (Leuschner, 1997). Beside this work, he was also interested in the temporal evolution of housing types in traditional cultural landscapes. This interest resulted in the publication of his book “Bauernhaus und Landschaft” (“Farmhouse and Landscape”, 1990). By compiling maps, notes and photographs, Ellenberg collected this rich and unique source of data about historical building types, construction materials, farm types, village forms, and landscape elements and their spatial distribution in Germany. He gathered data between 1932 and 1988 and explicitly stated that his aim was to provide historical information about the rural landscapes in 1950 and before throughout Germany. We
A. Dittrich et al. / Ecological Indicators 75 (2017) 101–110
103
Fig. 1. Schematic overview of the data used to address the two aims of the study, including the applied methods. CAR = Conditional Autoregression Model.
used this collection of data as source of information for historical ES products used by the rural population at this time (see below). 2.2.1.1. Maps. The central element of Ellenberg (1990) are 80 maps (Fig. 2), based on either the German topographical map (spatial resolution of approx. 10 × 12 km) or the UTM (Universal Transverse Mercator) coordinate system (10 × 10 km). All observations per grid cell were recorded by Ellenberg on an ordinal scale with five categories: very often, often, medium abundant, rare, and very rare (see Fig. 2). All grid cells which were not surveyed were labelled as ‘missing information’ and were hence excluded from our analysis. 2.2.1.2. Regional descriptions. Ellenberg used the spatial information gathered in his maps for delineating homogenous landscape zones, which differ from each other regarding their building types, material usage and type of villages. In doing so, he differentiated nine regions, 45 subregions (the main reference unit of Ellenberg’s landscape zones) and 97 sub-areas (with distinct differences in minor important characteristics). To further characterise these regions and to complement the information from the maps, he used notes taken in the field and information from the literature that resulted in very detailed regional descriptions including information on e.g., landscape scenery, agricultural management practices, climate and soil conditions. 2.2.2. Identification of indicators for historical ecosystem service products We carefully reviewed both maps and regional descriptions to identify indicators for historical ES. All ES products identified were independently verified by two of the authors. Identification of service products (we applied the CICES v4.3 classification, HainesYoung and Potschin (2013)) was often directly possible, e.g. for hedges used to separate livestock, trees planted around houses as wind protection, orchard meadows, organic material used in walls to apply clay (see Fig. 3), and stable wood in half-timbered constructions (see Fig. 3). In addition, some descriptions and mapped information were indirectly linked to certain service products. As an example, the farm house types (see Fig. 2) were used as proxy for the importance of either crop production or livestock keeping system or the occurrence of houses with certain roof pitches was used as an indicator for the use of straw and reed as roof covering material (see Fig. 3). In total, we found indication for eight ES classes covering 32 ES products. A detailed overview about the identified service products is given in Table 1.
2.2.3. Distribution of traditional land-use systems and ecosystem services richness To get the most complete picture of ES products provided, we compiled a complete list of all ES products for each of the 45 sub-regions based on the regional descriptions (which summarized information from the maps but also provided additional information as some ES were not represented in the maps, see Table 1 and section Regional descriptions). As opposed to the maps, regional descriptions provided only binary information on presence/absence of ES products. We hence determined the richness of ES products at the sub-region level and used flower plots (R package: graphics, R Core Team, 2013) to illustrate spatial differences in ES richness in relation to the dominant traditional land-use system by means of visual comparison (Fig. 5). To identify the dominant traditional land-use system for each of the 45 sub-regions, we analysed those maps providing detailed data on the relative importance of livestock keeping, crop production or mixed land-use systems (as defined in Baldock et al., 1994). In order to finally differentiate between the three main land-use systems, we applied the following rule: if farm houses and village forms typical for livestock keeping (see Fig. 2) were on average more abundant by at least one ordinal scale category as compared to those typical for the production of cereal crops, wine or fruits, the respective sub-region was assigned to the group of livestock keeping systems (analogously for crop production systems). If the difference in abundance was less than one category, sub-regions were assigned to mixed systems (Fig. 5). 2.2.4. Spatial regression of historic ecosystem service diversity and environmental variables For the spatial regression, we had to restrict our analysis to the 15 ES products represented in Ellenberg’s grid-based maps (see Fig. 1, marked ‘M’ in Table 1) to extract spatial explicit accounts of those ES products. By taking into account the relative abundance measurements (ordinal scale from one to five) recorded for every grid cell, we calculated the Shannon diversity (Shannon and Weaver, 1949) of ES products per cell. To ensure spatial compatibility between the two types of maps used by Ellenberg (1990), we resampled the data based on UTM raster maps to match the grid cells of the topographical maps of Germany. For analysing the relationship between these 15 ES products and environmental gradients, we selected soil quality (Roßberg et al., 2007), mean precipitation and temperature during the vegetation period (DWD, 2014), slope, altitude as well as terrain ruggedness (BKG, 2014) as potentially meaningful environmental predictors. The arithmetic mean of each environmental variable was calculated per grid cell
104
A. Dittrich et al. / Ecological Indicators 75 (2017) 101–110
Table 1 Indicators for historical ecosystem service products and descriptions extracted from Ellenberg (1990). Classes, divisions and groups refer to the CICES classification (version 4.3; Haines-Young and Potschin, 2013). Data source refers to either spatial data based on maps (M) or data based on regional descriptions (R), shown in the last column. As more detailed information about the different types of fodder were provided in the regional descriptions than recorded spatially explicit ‘Ma ’ indicates the map combining all types of fodder. Class [Division] (Group)
Short name
Service product
Biomass [Nutrition] (Cultivated crops)
Cultivated crops
Provisioning Cereal crops Vegetables Orchards Orchard meadows Wine Hop
Description
Data source
E.g. wheat and rye E.g. potatoes, pumpkins and vegetables in the garden Fruit production on a big scale Extensive fruit production
M, R R
Vineyards Hop production
R M, R
M, R M, R
Biomass [Nutrition] (Reared animals and their outputs)
Reared animals Wild animals
Honey Fish
Bee keeping Inland fishery
R R
Biomass [Materials] (Fibres and other materials from plants, algae and animals for direct use or processing)
Materials for direct use
Carrier material
Material in wall used as base to apply clay (e.g. straw, bendable wood) Plant material applied between straight wood (e.g. moss) Hemp was used to produce ropes Stable wood that was used for half-timbered constructions (mainly oak, partly spruce) Straight wood that was used for blockhouses (mainly spruce) Timber used for building without specific purpose Weatherproof wood used to protect walls (boards, planks and window shutters) Plant material used to cover roofs (straw, reed)
M, R
Insulation Ropes Stable wood
Straight wood Timber Weather protection Weather protection roofing Aesthetic appreciation Ornamental use Biomass [Materials] (Materials from plants, algae and animals for agricultural use)
Materials for agricultural use
Mulching
Peat Plaggen
River sediments Fodder-hay Fodder-oak Grazing Grazing/fodderhay Hedges
Biomass-based energy sources [Energy] (Plant-based resources)
Plant-based energy
Fire wood-hedges Fire wood-trees Fuel-peat
Gaseous/air flows [Mediation of flows] (Storm protection) No equivalent CICES class
R R M, R
M, R R M, R M, R
Flowers used to decorate house
R
Special wood used for handcrafts
R
Plant material was used as mulch for livestock bedding and put back to arable land Peat was added to soil to increase productivity Calluna vulgaris was extracted from common land and mixed with animal faeces Regular floods increased soil productivity in the floodplains Hay as fodder for livestock (mainly in livestock housing systems) Acorns were used for pig fattening (mainly in pig-housing systems) Livestock was fed with grass (mainly kept on pastures) Livestock was fed with grass (kept on pastures) as well as hay production as fodder (livestock housing systems) Hedges used to separate allotments for livestock keeping
R
Hedges were cut down or pruned regularly, material used as fuel Parts of trees extracted regularly and used as fuel Peat was extracted and used locally as well as transported to bigger cities
M, R M, R
Ma , R Ma , R Ma , R Ma , R
M, R M, R R M, R
Storm protection
Regulation/Maintenance Wind protection Trees and hedges around houses planted as wind protection
R
Fire protection
Fire protection
R
Big trees near houses attracted lightnings and hence protect nearby houses
A. Dittrich et al. / Ecological Indicators 75 (2017) 101–110
105
Fig. 2. Exemplary maps compiled by Ellenberg. Left: UTM raster map with information on the frequency of occurrence of separate barns typical for regions with high importance of crop production. Right: German topographical map with abundance information on certain house forms (so-called “Einfirst-, Winkel-, and T-Höfe”) which are typical farm houses in regions with high importance of livestock keeping. We added the red box to highlight the ordinal scale with five categories: biggest dot = very often and smallest dot = very rare. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) Source of the images: Ellenberg (1990).
Fig. 3. Examples of building materials identified as ecosystem service products. Left: use of reed as roof cover was defined as “Weather protection roofing” and timber used for frame construction of half-timbered construction was classified as “Stable wood”. Right: different organic materials were used in walls as base to apply clay, in this case bendable wood and wooden poles are shown, these materials were classified as “Carrier material”.(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) Source of images: Ellenberg (1990), right panel modified after Bedal (1985).
of the topographical map of Germany (R package: raster; Hijmans et al., 2014). Prior to analysis, all variables were tested for collinearity and due to their strong correlation with precipitation (|r| > 0.7; Dormann et al., 2013) the variables slope, ruggedness and altitude were identified as redundant (see Fig. A2). However, we retained ruggedness as predictor as it was not highly correlated with variables other than precipitation and we assumed that at least one measure of topography may be important for our analysis. Furthermore, ruggedness integrates information captured by slope and elevation change over a defined area (in our case: 1000 m2 ) and is therefore a proxy for the accessibility of a landscape.
The selected variables (soil quality, mean precipitation, temperature and ruggedness) were then standardized in order to allow better comparability of the model outputs (Schielzeth, 2010). To test Shannon diversity of ES against the standardized environmental predictors while accounting for spatial autocorrelation, we applied a spatial conditional autoregression model (CAR, R package: spdep; Bivand et al., 2014), with a neighbourhood size of four (“Rooks case”). This type of model is appropriate for identifying drivers on spatial patterns of ES (Mouchet et al., 2014). The residuals of the model were tested for remaining spatial autocorrelation.
106
A. Dittrich et al. / Ecological Indicators 75 (2017) 101–110
3. Results 3.1. Regional richness of historical ecosystem service products The review of ES products assessed within the scope of this study revealed that the northern sub-regions III-1, I-1 and II-3 as well as the central and southern regions V-6, VI-1, VI-3 and VIII-2 had the highest richness ranging from seven to 14 products (Fig. 4a). With eight ES products, sub-region IV-5 had the highest richness in the eastern part of Germany. In the proximity of northern and southern sub-regions characterised by high numbers of ES products, areas are located with similar average numbers of products (five to six); the same is true for sub-regions in the central part of Germany. The lowest number of ES products was identified for the eastern sub-regions IV-1 and IV-2, most parts of region IX and sub-region III-8 in the south as well as the western subregions V-9 and III-5 with a maximum of three different ES products. “Materials for direct use”, “Materials for agricultural use” and “Cultivated crops” were mentioned in the regional descriptions for nearly all regions. However, despite the comparatively high number of ES products belonging to the class “Materials for direct use” (10 in total, see Table 1), this class contributed more than 50% of the ES products assessed in only half of all sub-regions, indicating large spatial differences in the use of the related products (Fig. A3). Compared to other areas, sub-region VIII-2 was characterized by a high richness of ES products which were used as building materials and belonged to the service class “Materials for direct use” (see flower plots in Fig. 5 and Table A1). The second highest number of such ES products (four) was recorded in the sub-regions V-6, VI-1, VI-2, VI-4 and VII-3 (Fig. 4a). Sub-regions with two to three ES products belonging to the class “Cultivated crops” were mainly located in region III and V. ES products belonging to the classes “Reared animals”, “Wild animals”, “Storm protection” and “Fire protection” appeared to be spatially restricted as they were referenced only in 16% of all sub-regions. For example, the ES product assigned to “Storm protection” could only be identified in the northern regions I and II as well as in VII-2 in the west of Germany (see Fig. 5 and Table A1).
3.2. Distribution of traditional land-use systems Livestock keeping was dominant in 45% of sub-regions and most prominent Northern and Southern Germany whereas crop production was dominant in 21% of sub-regions that were mostly distributed in Eastern and Central Germany. The remaining 34% of sub-regions that were classified as mixed systems were mainly found in Central Germany (see Fig. 5). The most diverse livestock keeping system was found in sub-region III-1 with four ES products referring to different types of fodder (see Table A1) within the class “Materials for Agricultural Use”. In this sub-region, grazingforests for pigs (indicated by “fodder-oak”, see Table 1), livestock housing systems (indicated by “fodder-hay”) as well as outdoor keeping systems (indicated by “grazing”) were utilized. More than one fodder-related service product was also recorded for II-3, V-6 and VI-3. Sub-regions with either high (e.g. I-1, III-1, II-1, VII-2, V-6, and II3) or low (e.g. III-8, IV-2, IX-3, IX-4, and IX-5) numbers of ES classes and products mostly co-occurred with livestock keeping or mixed systems. The only exceptions were IV-5, V-3 and V-7, for which we found high numbers of ES products in crop production systems. Overall, sub-regions with average to lower numbers of ES products tended to be crop production systems, e.g. IV-1, IV-3, V-2, and V-9 (see Fig. 5).
Table 2 Relationship between Shannon diversity of historical ecosystem services and environmental variables (z-transformed) and their interactions at map sheet scale (∼130 km2 ) based on results of the spatial conditional autoregression model (CAR). Coefficients
Estimate
Std. Error
p-value
Soil Quality Ruggedness Mean Temperature during Vegetation Period Mean Precipitation during Vegetation Period Soil Quality: Ruggedness Soil Quality: Mean Temperature Soil Quality: Mean Precipitation Ruggedness: Mean Temperature Ruggedness: Mean Precipitation Mean Temperature: Mean Precipitation
0.026 −0.002 0.017 0.083 0.041 0.003 −0.013 −0.028 −0.027 0.025
0.017 0.030 0.025 0.039 0.026 0.016 0.027 0.014 0.012 0.017
0.135 0.952 0.482 0.032 0.114 0.841 0.633 0.045 0.030 0.127
3.3. Relationship between diversity of historical ecosystem service products and environmental conditions The CAR model was highly significant (p < 0.001) with a R2 of 0.410. In general, precipitation had the strongest positive influence on the service diversity (Fig. 4b, Table 2). Also, the interactions between ruggedness and temperature as well as ruggedness and precipitation were significant (p < 0.05). This means that in areas with flat terrain but with high temperatures or high precipitation, the number of services was greater compared to other regions. The extent of spatial autocorrelation further suggests that other region-specific drivers not specifically considered as predictors in the model were affecting the diversity of provisioning service products. 4. Discussion This study provides the first spatially explicit account of the provision of ES products historically used at a national scale in Germany. By basing our analyses on regional descriptions and maps of farmsteads, village forms, agricultural practices and specific landscape features, we were able to provide further insights on the characteristics and historical configuration of cultural landscapes. We found that a high richness of these products was associated with distinct types of traditional land-use systems (in particular livestock keeping and mixed farming systems). Furthermore, spatial differences in service diversity could be partially explained by environmental gradients. The present study contributes to the restricted numbers of studies that investigated temporal aspects of ˜ et al., ES at local/regional scales (Haase et al., 2007; Morán-Ordónez 2013; Palomo et al., 2014; Renard et al., 2015; Tomscha and Gergel, 2016). 4.1. Historical distribution of ecosystem service products In the past, when traditional land-use systems were more abundant in Germany, large regional differences in the number of service products existed. Based on both the regional descriptions and the maps, we found a hot spot of ES products in North-western Germany (Fig. 4), where livestock keeping and mixed farming systems dominated agricultural production (Fig. 5). These systems relied, for example, on shrubs as fences and sources of fuel, as well as acorns for pig fattening. Since soil fertility in these areas was generally low, heather was used as bedding and converted into fertilizer over winter time. Further, extracted peat served as fertilizer. While these different land-use types certainly represent more multifunctional landscapes, the higher numbers of ES in North-western Germany may also partly have arisen from delayed agricultural intensification up until 1950 (Büssis, 2006) and an overall slower population growth (Ehmer, 2004).
A. Dittrich et al. / Ecological Indicators 75 (2017) 101–110
107
Fig. 4. Historical distribution of ecosystem service products identified based on data provided by Ellenberg (1990) and classified according to CICES v4.3 (Haines-Young and Potschin, 2013). Panel a: Richness in ecosystem service products per sub-region. Services (in total 29) were identified from regional descriptions (including notes and photographs). Panel b: Diversity in ecosystem service products per grid cell (∼130 km2 ). Spatial accounts of services (in total 15) were extracted from maps. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
The spatial model helped us explain the lower numbers of ES in the central part of Germany, as precipitation during the growing season had an overall positive influence on ES numbers. Drier areas can be found in the eastern part of this region, which was dominated by crop production − also due to its extremely high soil fertility. Here, most of the building types relied on bricks and stones and not on organic material, which led to a lower number of ES implicitly recorded by Ellenberg. Precipitation also determines the range of oak species (Ellenberg and Leuschner, 2010) needed for half-timbered constructions, which resulted in slightly higher numbers of ES in the western part of this region, characterized by more rainfall. In contrast to our assumption that in areas with a complex topography the number of ES would be higher, terrain was not a good predictor for the distribution of ES. To improve the spatial model, further predictors such as socioeconomic data that capture human and cultural developments could be included. Specifically, data on population density, migration and growth, income or the distribution of ethnic minorities (e.g. Sorbs, Frisians) could prove particularly valuable in future analyses, e.g. such as done by Dunford et al. (2015). This is supported by Hartel et al. (2014) who showed that poverty is the main driver for the existence of traditional land-use systems that feature high ES diversity in Romania. However, to our knowledge, spatially explicit data on socio-economic variables between the 19th century and the Second World War in Germany are not available.
4.2. Challenges of mapping historical ecosystem service indicators Spatially explicit data are needed to establish causal relationships between historical ES provisioning and local environmental and socio-economic variables (Mouchet et al., 2014). Aerial photography, which assists topographic mapping since the 1920s, is obviously a promising approach and widely used in historical landscape analysis (e.g. Dittrich et al., 2011). However, data availability at larger scales is limited and data processing is time consuming (e.g. geographical rectification, digitizing by hand). Other promising approaches are intensive literature reviews as conducted by ˜ et al. (2013) which can provide detailed informaMorán-Ordónez tion on how differently cultural landscape sustained the livelihood of their rural populations. Even though this study assessed a total of 32 service products historically used in Germany, this set is far from being comprehensive. While provisioning services could be considered extensively, cultural, regulation and maintenance services were only scarcely accounted for. Due to this inherent bias in the data source, we were not able to establish a causal relationship between the type of traditional land-use system and the number of ES. To achieve a more complete account of ES, other data sources would have to be considered. For example, to identify regulating services such as flood mitigation, historical land cover data could be analysed (e.g. Früh-Müller et al., 2014).
108
A. Dittrich et al. / Ecological Indicators 75 (2017) 101–110
Fig. 5. Spatial distribution patterns of ecosystem service products (grouped by ecosystem service classes after CICES; see Table 1 for further details) for each of the sub-regions defined by Ellenberg (1990). Petal length in the flower plots indicates differences in service richness per service class among sub-regions. Colour codes for the flower plots refer to the main service classes. Grey bars show abundance of livestock keeping and crop production systems.
A. Dittrich et al. / Ecological Indicators 75 (2017) 101–110
4.3. Relevance for today’s ecosystem management and environmental policies Despite the limitations noted above, the data source and analysis presented here provided insights in the provisioning of ES in historical landscapes partly not existent anymore due to intensive land-use changes since then. This baseline of ES provisioning aids our understanding of the development of current land-use patterns, as for example North-western Germany is still known for intensive livestock production whereas crop production is important for Eastern Germany (Dittrich et al. unpublished). Furthermore, by comparing past and present utilization of services a better understanding regarding the processes leading to the generation and ˜ et al., 2013; Renard extraction of ES can be gained (Morán-Ordónez et al., 2015) making the loss of certain ES traceable. In that respect, understanding the historical co-evolution of cultural landscapes and ecological resource use (here conceptualized in terms of ES) may be important to preserve or regain multifunctional landscapes. If we wish to maintain the diversity and unique character of cultural landscapes in the face of increasing temporally and spatially homogeneous agricultural land-use practices then there is a need for these cultural landscapes to replicate the multi-faceted and multifunctional uses from which such landscapes arose. Therefore, it would be highly relevant to compare historical and legally-proposed land-use options for their impact on the multifunctionality of landscapes. A large number of subsidy schemes (e.g. Thomas et al., 2009; Schleyer and Plieninger 2011; Ulrich and Riecken, 2012) for historical land-use options equally show that knowledge on their impact on ES is necessary to identify more effective and regionally adapted options. A better understanding of the ES such landscape provided and consideration of how these unique landscapes could provide similar suites of services suitable for the 21st century would help to restore cultural landscapes. Their values come not only from their cultural importance, but also form the flows of services they provide. However, most of today’s subsidies and ‘greening’ measures (Common Agricultural Policy; BMEL, 2015) aim at protecting biodiversity aspects (e.g. hedges planted to separate livestock in the past are important for biodiversity today; Ponti et al., 2005) that co-evolved with cultural landscapes. As societal demands on landscapes are changing continuously (e.g. Haase et al., 2007) it is very likely that “new” regulating/maintenance as well as cultural services are wanted by society, which in turn may create trade-offs with the conservation targets of the present subsidies. Balancing these trade-offs would require workshops with all stakeholders involved (Förster et al., 2015; Seppelt and Cumming, 2016).
5. Conclusion and outlook The countryside in Europe is becoming less and less a place that provides a livelihood for the majority of the inhabitants, thereby “landscape and rural life are becoming ominously disjoined” (Lowenthal, 1997). The presented study for historical ES provisioning helps to understand how landscapes have been used and developed in the past in Germany, a task for which the ES concept proved to be a useful lens through which to assess the benefits for society provided by cultural landscapes. While the present study focuses on Germany, it may be regarded as representative for many regions in Europe. We further suggest that the investigation of historical ES could be a valuable approach to clarify the identity and functionality of cultural landscapes in the changing spatial context of society (Antrop, 2005). As the existence of cultural ˜ et al., landscapes depends on their active use (e.g. Morán-Ordónez 2013), assessing historical ES may also provide incentives for the maintenance of such landscapes. This is partly realized by current
109
subsidy programs, which aim at protecting former ES in cultural landscapes to foster the conservation of agro-biodiversity and landscape aesthetics. However, more sustainable options to reconnect people to the rural landscape should also be explored. To support acceptance and application of these incentives we advocate to also appeal to the history and former diversity of ES that shaped the homeland of today’s residents. Acknowledgements The presented study is the outcome of a Ph.D. synthesis workshop of the Helmholtz Research School for Ecosystem Services under Changing Land Use and Climate (ESCALATE) at the Helmholtz Centre for Environmental Research – UFZ and the Leuphana University Lunenburg. The work was funded partly by the Helmholtz Programme ‘Terrestrial Environment’. We thank Ludwig Ellenberg who approved the usage of the data gathered by his father Heinz Ellenberg. We further thank Wulf Jung from GISCON for providing digital versions of the maps published in the book as well as Dietmar Roßberg for providing a Germany-wide soil quality map. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ecolind.2016. 12.010. References Antrop, M., 2005. Why landscapes of the past are important for the future. Landsc. Urban Plann. 70, 21–34, http://dx.doi.org/10.1016/j.landurbplan.2003.10.002. Büssis, H., 2006. Die Theorie von der extensiven Allmendenutzung-eine ükologische Fehlinterpretation! Charadrius 42, 1–8. BKG (Bundesamt für Kartographie und Geodäsie), 2014. Digitale Geländemodelle [WWW Document]. https://upd.geodatenzentrum.de/geodaten/gdz rahmen. gdz div (Accessed 1 December 2014). BMEL (Bundesministerium für Ernährung und Landwirtschaft), 2015. Umsetzung der EU-Agrarreform in Deutschland - Ausgabe 2015 [WWW Document], http://www.bmel.de/SharedDocs/Downloads/Broschueren/ UmsetzungGAPinD.html (Accessed 21 July 2015). Baldock, D., Beaufoy, G., Dark, J., 1994. The Nature of Farming: Low Intensity Farming Systems in Nine European Countries. Institute for European Environmental Policy, London. Barthel, S., Crumley, C., Svedin, U., 2013. Bio-cultural refugia—safeguarding diversity of practices for food security and biodiversity. Glob. Environ. Change 23, 1142–1152, http://dx.doi.org/10.1016/j.gloenvcha.2013.05.001. Beck, R., 1986. Naturale Ökonomie: Unterfinning, bäuerliche Wirtschaft in einem oberbayerischen Dorf des frühen 18. Jahrhunderts. Deutscher Kunstverlag, München. Bedal, K., 1985. Häuser aus Franken. In: Museumsführer Fränkisches Freilichtmuseum in Bad Windsheim, second ed. Delph’sche Verlagsbuchhandlung, München. ˇ epánek, V., 2001. Land-use changes and their social driving Biˇcı´ık, I., Jeleˇcek, L., Stˇ forces in Czechia in the 19th and 20th centuries. Land Use Policy 18, 65–73, http://dx.doi.org/10.1016/S0264-8377(00)00047-8. Bignal, E., McCracken, D., Corrie, H., 1995. Defining European low-intensity farming systems: the nature of farming. In: McCracken, D., Bignal, E., Wenlock, S. (Eds.), Farming on the Edge: The Nature of Traditional Farmland in Europe. Joint Nature Conservation Committee, Peterborough, pp. 29–37. Bivand, R., Altman, M., Anselin, L., Assunc¸ão, R., Berke, O., Bernat, A., Blanchet, G., Blankmeyer, E., Carvalho, M., Christensen, B., Chun, Y., Dormann, C., Dray, S., Gómez-Rubio, V., Halbersma, R., Krainski, E., Legendre, P., Lewin-Koh, N., Li, H., Ma, J., Millo, G., Mueller, W., Ono, H., Peres-Neto, P., Piras, G., Reder, M., Tiefelsdorf, M., Yu, D., 2014. spdep: Spatial Dependence: Weighting Schemes, Statistics and Models. Blaxter, K., Robertson, N., 1995. From Dearth to Plenty: The Second Agricultural Revolution. Cambridge University Press, Cambridge. DWD (Deutscher Wetterdienst), 2014. GISC (Global Information System Centre) [WWW Document]. http://gisc.dwd.de/GISC DWD/start js JSP.do (Accessed 12 January 2014). Dittrich, A., Buerkert, A., Brinkmann, K., 2011. Assessment of land use and land cover changes during the last 50 years in oases and surrounding rangelands of Xinjiang, NW China. J. Agric. Rural Dev. Tropics Subtropics 111, 129–142 http://nbn-resolving.de/urn:nbn:de:hebis:34-2011052737598. Dormann, C.F., Elith, J., Bacher, S., Buchmann, C., Carl, G., Carré, G., Marquéz, J.R.G., Gruber, B., Lafourcade, B., Leitão, P.J., Münkemüller, T., McClean, C., Osborne, P.E., Reineking, B., Schröder, B., Skidmore, A.K., Zurell, D., Lautenbach, S., 2013.
110
A. Dittrich et al. / Ecological Indicators 75 (2017) 101–110
Collinearity: a review of methods to deal with it and a simulation study evaluating their performance. Ecography 36, 27–46, http://dx.doi.org/10.1111/ j.1600-0587.2012.07348.x. Dunford, R.W., Smith, A.C., Harrison, P.A., Hanganu, D., 2015. Ecosystem service provision in a changing Europe: adapting to the impacts of combined climate and socio-economic change. Landsc. Ecol. 30, 443–461, http://dx.doi.org/10. 1007/s10980-014-0148-2. Ehmer, J., 2004. Bevölkerungsgeschichte und Historische Demographie 1800–2000. Oldenbourg, München. Ellenberg, H., Leuschner, C., 2010. Vegetation Mitteleuropas mit den Alpen. In: ökologischer, dynamischer und historischer, sixth ed. Ulmer, Stuttgart. Ellenberg, H., 1990. Bauernhaus and Landschaft: in ökologischer und historischer Sicht. Ulmer, Stuttgart. Förster, J., Barkmann, J., Fricke, R., Hotes, S., Kleyer, M., Kobbe, S., Kübler, D., Rumbauer, M., Seppelt, R., Settele, J., Spangenberg, J.H., Tekken, V., Václavík, T., Wittmer, H., 2015. Assessing ecosystem services for informing land-use decisions: a problem-oriented approach. Ecol. Soc. 20 (3), 31, http://dx.doi.org/ 10.5751/ES-07804-200331. FAO (Food and Agriculture Organization of the United Nations), 2015. What is land consolidation [WWW Document]. http://www.fao.org/docrep/006/y4954e/ y4954e06.htm (Accessed 21 July 2015). Farina, A., 2000. The cultural landscape as a model for the integration of ecology and economics. BioScience 50, 313–320, http://dx.doi.org/10.1641/00063568(2000)050[0313:TCLAAM]2.3.CO;2. Fisher, B., Turner, R.K., Morling, P., 2009. Defining and classifying ecosystem services for decision making. Ecol. Econ. 68, 643–653, http://dx.doi.org/10. 1016/j.ecolecon.2008.09.014. Früh-Müller, A., Wegmann, M., Koellner, T., 2014. Flood exposure and settlement expansion since pre-industrial times in 1850 until 2011 in north Bavaria, Germany. Regional Environ. Change 15, 183–193, http://dx.doi.org/10.1007/ s10113-014-0633-9. Haase, D., Walz, U., Neubert, M., Rosenberg, M., 2007. Changes to Central European landscapes—analysing historical maps to approach current environmental issues, examples from Saxony, Central Germany. Land Use Policy 24, 248–263, http://dx.doi.org/10.1016/j.landusepol.2005.09.003. Haines-Young, R., Potschin, M., 2013. Common International Classification of Ecosystem Services (CICES): Consultation on Version 4, August–December 2012. EEA Framework Contract No EEA. Contract No EEA/IEA/09/003. Hanspach, J., Hartel, T., Milcu, A.I., Mikulcak, F., Dorresteijn, I., Loos, J., von Wehrden, H., Kuemmerle, T., Abson, D., Kovács-Hostyánszki, A., Báldi, A., Fischer, J., 2014. A holistic approach to studying social-ecological systems and its application to southern Transylvania. Ecol. Soc. 19 (4), 32, http://dx.doi.org/ 10.5751/ES-06915-190432. Hartel, T., Fischer, J., Câmpeanu, C., Milcu, A.I., Hanspach, J., Fazey, I., 2014. The importance of ecosystem services for rural inhabitants in a changing cultural landscape in Romania. Ecol. Soc. 19 (2), 42, http://dx.doi.org/10.5751/ES06333-190242. Heath, M.F., Tucker, G.M., 1995. Ornithological value and pastoral farming systems. In: McCracken, D., Bignal, E., Wenlock, S. (Eds.), Farming on the Edge: The Nature of Traditional Farmland in Europe. Joint Nature Conservation Committee, Peterborough, pp. 54–59. Herzog, F., 1998. Streuobst: a traditional agroforestry system as a model for agroforestry development in temperate Europe. Agrofor. Syst. 42, 61–80, http://dx.doi.org/10.1023/A:1006152127824. Hijmans, R.J., Etten, van, J., Mattiuzzi, M., Sumner, M., Greenberg, J.A., Lamigueiro, O.P., Bevan, A., Racine, E.B., Shortridge, A., 2014. Raster: Geographic data analysis and modelling. R Package Version 2, 2–12 http://CRAN.R-project.org/ package=raster. Lautenbach, S., Kugel, C., Lausch, A., Seppelt, R., 2010. Analysis of historic changes in regional ecosystem service provisioning using land use data. Ecol. Indic. 11, 676–687, http://dx.doi.org/10.1016/j.ecolind.2010.09.007. Leuschner, C., 1997. Heinz Ellenberg (1913–1997). Phytocoenologia 4, 457–462, http://dx.doi.org/10.1127/phyto/27/1997/457. Lowenthal, D., 1997. European landscape transformations: the rural residue. In: Groth, P.E., Bressi, T.W. (Eds.), Understanding Ordinary Landscapes. Yale University Press, Yale, pp. 180–188. ˜ Morán-Ordónez, A., Bugter, R., Suárez-Seoane, S., Luis, E., de Calvo, L., 2013. Temporal changes in socio-ecological systems and their impact on ecosystem services at different governance scales: a case study of heathlands. Ecosystems 16, 765–782, http://dx.doi.org/10.1007/s10021-013-9649-0.
Mouchet, M.A., Lamarque, P., Martín-López, B., Crouzat, E., Gos, P., Byczek, C., Lavorel, S., 2014. An interdisciplinary methodological guide for quantifying associations between ecosystem services. Glob. Environ. Change 28, 298–308, http://dx.doi.org/10.1016/j.gloenvcha.2014.07.012. Palomo, I., Martín-López, B., Zorrilla-Miras, P., Del Amo, D.G., Montes, C., 2014. ˜ Deliberative mapping of ecosystem services within and around Donana National Park (SW Spain) in relation to land use change. Reg. Environ. Change 14, 237–251, http://dx.doi.org/10.1007/s10113-013-0488-5. Plieninger, T., Bieling, C., Ohnesorge, B., Schaich, H., Schleyer, C., Wolff, F., 2013. Exploring futures of ecosystem services in cultural landscapes through participatory scenario development in the Swabian Alb, Germany. Ecol. Soc. 18 (3), 39, http://dx.doi.org/10.5751/ES-05802-180339. Plieninger, T., Draux, H., Fagerholm, N., Bieling, C., Bürgi, M., Kizos, T., Kuemmerle, T., Primdahl, J., Verburg, P.H., 2016. The driving forces of landscape change in Europe: a systematic review of the evidence. Land Use Policy 57, 204, http:// dx.doi.org/10.1016/j.landusepol.2016.04.040. Ponti, L., Ricci, C., Veronesi, F., Torricelli, R., 2005. Natural hedges as an element of functional biodiversity in agroecosystems: the case of a Central Italy vineyard. Bull. Insectol. 58, 19–23. Poschlod, P., Bakker, J.P., Kahmen, S., 2005. Changing land use and its impact on biodiversity. Basic Appl. Ecol. 6, 93–98, http://dx.doi.org/10.1016/j.baae.2004. 12.001. R Core Team, 2013. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. Renard, D., Rhemtulla, J.M., Bennett, E.M., 2015. Historical dynamics in ecosystem service bundles. Proc. Natl. Acad. Sci. 112, 13411–13416, http://dx.doi.org/10. 1073/pnas.1502565112. Roßberg, D., Michel, V., Graf, R., Neukampf, R., 2007. Definition von Boden-Klima-Räumen für die Bundesrepublik Deutschland. Nachrichtenblatt des deutschen Pflanzenschutzdienstes 59, 155–161. Schielzeth, H., 2010. Simple means to improve the interpretability of regression coefficients. Methods Ecol. Evol. 1, 103–113, http://dx.doi.org/10.1111/j.2041210X.2010.00012.x. Schleyer, C., Plieninger, T., 2011. Obstacles and options for the design and implementation of payment schemes for ecosystem services provided through farm trees in Saxony, Germany. Eviron. Conserv. 38, 454–463, http://dx.doi. org/10.1017/S0376892911000361. Schulze-Hagen, K., 2004. Allmenden und ihr Vogelreichtum – Wandel von Landschaft, Landwirtschaft und Avifauna in den letzten 250 Jahren. Charadrius 40, 97–121. Seppelt, R., Cumming, G.S., 2016. Humanity’s distance to nature: time for environmental austerity? Landsc. Ecol. 31, 1645–1651, http://dx.doi.org/10. 1007/s10980-016-0423-5. Shannon, C.E., Weaver, W., 1949. The Mathematical Theory of Communication. University of Illinois Press, Urbana and Chicago. Spielman, D.J., Pandya-Lorch, R., 2009. Millions Fed: Proven Successes in Agricultural Development. International Food Policy Research Institute, Washington D.C. Thomas, F., Denzel, K., Hartman, E.L.R., Schmoock, K., 2009. Kurzfassung der Agrarumwelt- und Naturschutzprogramme. In: Darstellung und Analyse der Entwicklung von Maßnahmen der Agrarumwelt- und Naturschutzprogramme in der Bundesrepublik Deutschland. BfN –Skripten 253. Bundesamt für Naturschutz, Bonn. Tomscha, S., Gergel, S., 2016. Ecosystem service trade-offs and synergies misunderstood without landscape history. Ecol. Soc. 21 (1), 43, http://dx.doi. org/10.5751/ES-08345-210143. UNESCO (United Nations Educational, Scientific and Cultural Organization), 2014. Biosphere Reserves [WWW Document], Accessed 1 December 2014 http:// www.unesco.org/new/en/natural-sciences/environment/ecological-sciences/ biosphere-reserves/. Ulrich, K., Riecken, U., 2012. Moorschutzstrategien, –initiativen und –programme in Deutschland. Natur und Landschaft 87, 81–86. von Wehrden, H., Abson, D.J., Beckmann, M., Cord, A.F., Klotz, S., Seppelt, R., 2014. Realigning the land-sharing/land-sparing debate to match conservation needs: considering diversity scales and land-use history. Landsc. Ecol. 29, 941–948, http://dx.doi.org/10.1007/s10980-014-0038-7.