Cultural Landscapes Blog

WP4: Mapping Europe’s cultural landscapes and how they are changing

27 September 2016/by Tobias Kuemmerle & Christian Levers/Tags: land-use change, WP4, cultural landscape typology

At the onset of HERCULES, no consistent map was available of where cultural landscapes can be found in Europe as well as little information on what characterizes these landscapes across Europe. These knowledge gaps were real barriers for assessment of where and how recent land use changes impacts cultural landscapes, as well as for policies that could mitigate unwanted developments in order to effectively protect and manage the cultural, historical, and archaeological value of these landscapes.

A major part of the work in WP4 thus included developing the first typology of European cultural landscapes. The overarching goals of this work were to (i) capture the diversity of Europe’s heritage landscapes, (ii) support targeted policy making for protecting and investing in cultural landscapes, and (iii) provide a platform for empowering local communities through publically available information on the occurrence of cultural landscape types. To develop this typology, we relied on a wide array of spatial datasets, from farm account data over disaggregated Eurostat data to Panoramio pictures and maps of traditionally produced regional food (e.g., Buffalo Mozzarella from the Campania, blue cheese from Stilton, or Camembert from Normandy). We combined these datasets into three generic dimensions along which cultural landscapes can be characterized and mapped (A: land management intensity, B: landscape structure, C: landscape value and meaning) and integrated those into the HERCULES Knowledge Hub so that users can flexible define ‘their’ typology and map it out across Europe.

The resulting maps show how these different dimensions combine to shape cultural landscapes in Europe. Some agricultural landscapes, especially in the Mediterranean, were characterized by the presence of all cultural landscape dimensions while other areas, such as in eastern Germany and northern France, were characterized by the absence of cultural landscape dimensions. For forest landscapes, mountainous areas were characterized by the presence of all cultural landscape dimensions while intensively managed forest landscapes prevailed mainly in Scandinavia.

In parallel to the mapping of the cultural landscape typology, we gathered a wide range of indicators of land-use change processes that may affect cultural landscapes. These indicators were derived from satellite image time series, from ground surveys, and from agricultural and forestry statistics that we integrated with fine-scale spatial information. Importantly, our indicator database included information on changes in the area of broad land-use classes, such as when urban areas expand or when agricultural land becomes abandoned, but also changes in the management intensity within these classes, such as when agricultural intensification takes place and key landscape features such as hedgerows or low-intensity grasslands are lost. A particularly exciting example for the outcomes of these analyses was the first fine-scale map of farmland abandonment in Europe, mapped from MODIS satellite images, and highlighting that farmland abandonment since 2001 occurred mainly in Europe’s mountain regions and in its Eastern periphery (Figure).

By overlaying these land-change trends with our cultural landscapes typology, we could investigate the spatial concordance of cultural landscapes with past and future land-change trajectories and address the identification and classification of threats and opportunities to and for cultural landscapes throughout Europe. For example, intensification occurred mainly in Central and Western European countries and did not show large overlap with cultural landscapes of low management intensity, whereas Eastern European and the Mediterranean had widespread cultural landscapes and less intensification.

Together, these two key outcomes of WP4 – the cultural landscape typology and map on the one hand, and the land-use change indicators on the other – allowed us to reconstruct recent dynamics in cultural landscape and to detect hotspots and coldspots of cultural landscape loss. The outcomes of WP4, specifically the typology of cultural landscapes and the map of cultural landscape distribution, will then form the basis to study how cultural landscapes may change in the future in WP5.

Though not all results from WP4 are published already, key publications related to this work package include:

- Tieskens, K.F., Schulp, C.J.E., Levers, C., Lieskovský, J., Kuemmerle, T., Plieninger, T., Verburg, P.H. (2015): Characterization of European Cultural Landscapes: accounting for structure, land use intensity and value of rural landscapes. Under review.

- Levers, C. Butsic, V., Verburg, P. H., Müller, D., and Kuemmerle, T. (2016): Drivers of changes in agricultural intensity in Europe. Land Use Policy, 58, 380-393.

- Kuemmerle, T., Levers, L, Erb, K.H., Estel, S., Jepsen, M.R., Kroisleitner, C., Müller, D., Plutzar, C., Stürck, J., Verkerk, P.J., Verburg, P.H., and Reenberg, A. (2016) Hotspots of land use change in Europe. Environmental Research Letters, 11, 064020.

- Estel, S., Kuemmerle, T., Levers, C. and Hostert, P. (2016): Mapping cropland-use intensity across Europe using MODIS NDVI time series. Environmental Research Letters, 11, 024015.

- Levers, C., Müller, D., Erb, K. H., Haberl, H., Jepsen, M. R., Metzger, M. J., Meyfroidt, P., Plieninger, T., Plutzar, C., Stürck, J., Verburg, P. H., Verkerk, P. J., and Kuemmerle, T. (2015): Archetypical patterns and trajectories of land systems in Europe. Regional Environmental Change, in press.

- Estel, S., Kuemmerle, T., Alcantara, C., Levers, C., Prishchepov, A.V., and Hostert, P. (2015): Mapping farmland abandonment and recultivation across Europe using MODIS NDVI time series. Remote Sensing of Environment, 163, 312–325.

Figure: Farmland abandonment in Europe between 2001 and 2012, as mapped from MODIS satellite image time series (for details see Estel et al. 2015).