On the accuracy of old maps
Cartographica Helvetica 8 (1993) 47–49
Recently a new method for analyzing old maps was published by Peter Mekenkamp. The author succeeds in visualizing the positional accuracy of selected points by comparing all map distances between these points with their 'true' values reduced to map scale. For each point the standard deviation of the differences between map distances and 'true' distances to all other points gives a measure for the positional accuracy of that point. This value is taken as the radius of a circle drawn around the point, thus making its positional accuracy visible.
However, this simple and efficient procedure leaves the questions unanswered of how these points are incorrectly located and how their position should be corrected. In the present paper this problem is solved by an adjustement using the least-squares method, a well-known algorithm in geodesy. The false points are brought into an optimal position with respect to the 'true' location by minimizing the sum of the squares of the required displacements.
The method by Mekenkamp and its extension by a geodetic adjustement procedure are demonstrated on the general sheet of the Atlas tyrolensis (1774) by Peter Anich and Blasius Hueber.