Uncertainty investigation in historical maps for retrospective landscape assessments

Historical maps such as the Siegfried map from the 19th century are not only fascinating witnesses over the past of our landscape; they also hold useful information for scientific evaluation of changes over time within this landscape.

To incorporate the historical information into Geographical Information Systems (GIS) the needed information has to be extracted. Therefore, pattern recognition methods are applied.

In using geographic data from historical map for subsequent target applications different sources of uncertainty occur. The nature of uncertainty has to be understood in its entire complexity to realize a systematic investigation.

Uncertainty arises in different stages of map production and field survey, during transformation, digitization, processing and preprocessing of the data and due to the intended application. The latter requires a historical investigation by using different sources of information such as literature, historical photos, local reference maps and knowledge of influences from politics, science and society onto the topographic work.

Leyk S, Zimmermann NE, 2007 Improving land cover change analysis based on uncertain survey maps: predictive map correction and area estimations using fuzzy sets. Landscape Ecology 22: 257-272.

Leyk S, Boesch R, Weibel R, 2006. Saliency and semantic processing: Extracting forest cover from historical topographic maps. Pattern Recognition 39, 953-968.

Leyk S, Boesch R, and Weibel R, 2005. A conceptual framework for uncertainty investigation in map-based land cover change modeling. Transactions in GIS 9(3), 291-322.

Leyk S, Zimmermann NE, 2004. A predictive uncertainty model for field-based survey maps using generalized linear models. Lecture Notes in Computer Science 3234, 191-205.

Leyk S, Koehl M, Poncét F, 2002. Application of Future TerraSAR Data for Improvement of Forest Resource Assessments. Proceedings of the ForestSAT Symposium, Heriot Watt University, Edinburgh