Human land use practices are the single most important factor affecting and changes European landscapes. As a consequence, much attention has been devoted within Landscape Ecology to analyze changing patterns of land use and develop research strategies to understand the processes, behind these changes in order to inform policy makers. Models have become an increasingly important tool in this strain of research, partly due to the revolution in information technologies that has happened in the last 30 years. The, modeling of human decision making, however is often based on assumptions about the ability to identify structures and driving forces of a relatively stable nature, explaining or forecasting the process of change. This is not solely a modern phenomenon and in this paper we explore the way by the Danish military and government officials developed an early land use model to serve such purposes. At the end of the 19th century the Danish military realized, that they had a problem. The landscape surrounding the fortress of Copenhagen was too flat and didn’t have enough fixpoints in form of landscape elements and topography to ensure a proper direction of artillery fire from the batteries on the fortifications. To fix this problem, the Danish military planned a comprehensive survey of the agricultural landscape around Copenhagen in the year 1900, with the purpose of developing a model able of predicting agricultural land use, to be used to direct and control artillery systems on the land side of Copenhagen. The basic assumption behind the model was stated by the Artillery Command of the Army: “the field systems and rotational and use practices around Copenhagen are so stable that agricultural data could be used for directing artillery fire in the future.” The plan was to map every field within a perimeter of five kilometres from the front of the fortification in order to create a very detailed map in 1:10.000, complete with field boundaries, as well as register information about the rotational system. At first, the survey campaign seemed to be going very well, but the military quickly ran into problems. The rapid urbanization of the landscape north of Copenhagen meant, that farming was abandoned in preparation of development. On the island of Amager southwest of Copenhagen the farmers didn’t use any fixed rotational schemes because of the supply of fertilizer from the city of Copenhagen. And in addition to these obvious difficulties, the model was also challenged because the social and economic system which was supposed to be constant, as it had been for centuries. The urbanization led to urban sprawl and technological development in the agricultural sector fundamentally changed the landscape around Copenhagen in the period between 1900 and 1930. Modeling and efforts to understand the processes in the landscapes have developed tremendously since year 1900, but this story is a good reminder about potential pitfalls of modeling with a simplistic understanding of static human processes and patterns in the landscape.
Ideas and thoughts for the work with the paper to be discussed at the Copenhagen workshop:
Historical cartographic data produced for other purposes has been widely used within landscape ecology to set baselines for monitoring and analyze long-term landscape change. However relative more attention has been devoted towards the precision and digitization process of the data, than on understanding the perception of the landscape in which the map has been made and consequently the landscape data which has been represented at the map. Despite the growing awareness of the maps as constructed representation within a specific field or societal setting(Harley 1989)(Edney 1996), many papars within landscape ecology and landscape research seems to regard maps as objective.
The following quotes do more or less illustrates this point:
“Many human-made objects are missing in Swiss topographic maps because of resolution. The mapped ones generally lack information on use. For example, Swiss topographic maps do not indicate wjeter a building in a forest is used for forest management or recreation” page 302 (Hersperger, Langhamer, and Dalang 2012)
“Since all maps are abstraction of reality (Robinson et al. 1978), cartography has to cope with errors and uncertainty, which are mostly related to the scale of the map. The powerful capabilities of geographical information systems (GIS) generate a false perception of database accuracy as the precision of coordinates in GIS is by far better than the accuracy of the spatial data (Goodchild and Gopal 1992). Operational errors are often created through successive GIS data manipulations. Thus, one should identify the effects on spatial analysis of combining data with different levels of errors, and develop methods to integrate multisource data into homogeneous spatial databases. (Petit and Lambin 2002)
“The military survey maps provide a suitable basis for analyzing and evaluating the development trends in the landscape macrostructure. However, the main shortcoming of the First Military Survey Maps is that they suffer from some geodetic inaccuracy…”(Skaloš et al. 2011)
“Since topographic maps are perceptions of the environment and often rather a “text than a mirror of reality”(Harley 1989), definition of specific landscape elements or habitats may vary considerably over time and the location of features is often less accurate in earlier map editions compared with today’s standard…” (Kienast 1993)
So my idea is to expand the paper to also deal with the use of cartographic data as objective data, although many research paper recognize the “constructive” character of historical cartographic data, they spend more word on discussion the accuracy and digitization then on demonstrating real knowledge about the perception of landscape and space embedded in their sources. One point will be that GIS and models are used as instrument to invoke a sense of objective accuracy; this in some / most cases will be false if the landscape perception and representational practices of the map producer isn’t carefully analyzed. Thus, the point should be to raise the awareness about the influence of the social context of historical cartographic sources among landscape ecologist.
Edney, Matthew H. 1996. “Theory and the History of Cartography.” Imago Mundi 48 (January 1): 185–191. doi:10.2307/1151272.
Harley, J B. 1989. “DECONSTRUCTING THE MAP.” Cartographica: The International Journal for Geographic Information and Geovisualization 26 (2) (October 1): 1–20. doi:10.3138/E635-7827-1757-9T53.
Hersperger, Anna M., Dominik Langhamer, and Thomas Dalang. 2012. “Inventorying Human-made Objects: A Step Towards Better Understanding Land Use for Multifunctional Planning in a Periurban Swiss Landscape.” Landscape and Urban Planning 105 (3) (April): 307–314. doi:10.1016/j.landurbplan.2012.01.008.
Kienast, Felix. 1993. “Analysis of Historic Landscape Patterns with a Geographical Information System ? a Methodological Outline.” Landscape Ecology 8 (2) (June): 103–118. doi:10.1007/BF00141590.
Petit, C.C., and E.F. Lambin. 2002. “Impact of data integration technique on historical land-use/land-cover change: Comparing historical maps with remote sensing data in the Belgian Ardennes.” Landscape Ecology 17 (2) (June 1): 117–132. doi:10.1023/A:1016599627798.
Skaloš, Jan, Martin Weber, Zdeněk Lipský, Ivana Trpáková, Markéta Šantrůčková, Lenka Uhlířová, and Pavel Kukla. 2011. “Using Old Military Survey Maps and Orthophotograph Maps to Analyse Long-term Land Cover Changes – Case Study (Czech Republic).” Applied Geography 31 (2) (April): 426–438. doi:10.1016/j.apgeog.2010.10.004.