Digitization and geo-referencing of botanical distribution maps

Aim In many fields of research, valuable information is stored in atlases and maps which are only available as printed media. Scientists who are interested in computational analysis often seek to digitize the data to make it accessible for numerical calculations. In this paper, an approach using the example of digitizing distribution maps taken from plant-taxonomic atlases is described. For this purpose, a software tool has been built for application in the Sonderforschungsbereich 350 'Interactions between and Modelling of Continental Geosystems' at the University of Bonn. Its functionality has to be simple in use and capable of transforming fields drawn on geographical maps into grid data even where the type of map projection is unknown. Location As an example, plant distributions over Europe and Asia have been digitized. Methods To achieve this aim through an objective statistical analysis, the local deterministic approximation has been used for geo-referencing. Results For this reason the development of a new software tool was required. Although there is a large market for related software from Geographical Information Systems (GIS ), none of the available GIS-programs appeared to be capable of geo-referencing maps with unidentified projections. Rather than to give a product comparison, this paper aims on the methodology and mathematical formalism. Main conclusions The developed algorithm is a handy tool to capture data from maps based on obscure projections. Valuable historical maps, which are problematic for standard GIS-programs, can be made accessible for modern research work.