A THEORY OF CARTOGRAPHIC ERROR AND ITS MEASUREMENT IN DIGITAL DATA BASES

The processes of map production are necessarily approximate, and thus the resulting map contains a variety of error effects. This paper develops a theory of information content that applies to geometric details on a map, based on the epsilon distance model. Examining map production technology, the epsilon model provides a reasonable approximation of the error expected. This error can be measured, and an example is worked using data from the GIRAS digital files produced by USGS. Under conservative assumptions, 7 percent of the selected study area around Pittsburgh lies in zones of potential error.