2.     Assessment of positional errors

Early maps of Orkney often show characteristic distortions of their outline which make recognition of closely related maps relatively easy. However, as the accuracy of the later maps of Orkney improved, using solely their overall appearance to determine the relationships between them becomes more difficult.  Also, although the general improvement in accuracy with time is readily apparent, it is less easy to provide a valid measure of it and identify the sources of that improvement. Outlined below is a strategy devised to tackle these problems.

It was assumed that the accuracy of a map can be assessed by the extent to which its outline differs from the modern outline and that the patterns of such differences can be used to determine relationships between maps.  Thus, although the relative positions of any two  places should be similar on different maps of Orkney, there is no obvious reason why the errors in their positioning should remain the same unless the maps have a common source.

It was decided to select a number of locations with features sufficiently characteristic to allow their identification even on greatly distorted old maps. The locations were also selected to be widely spaced across Orkney so as to minimize any impact on the estimate of overall accuracy from purely localized distortions. The errors in positioning of these locations were assumed to reflect the accuracy of  the map as a whole and be characteristic of it. Thirty five such locations were chosen and then a reference point was required from which the individual distances to these locations could be measured and their errors assessed. Kirkwall was chosen for this since it is centrally located, was prominent on virtually all maps and was always identifiable from its location on the isthmus between the east and west portions of the Orkney Mainland.