About polygon area uncertainty in GIS and its implications on agro-forestry estimates

Abstract

Error affecting calculation of a polygon area from a digital map is an issue that is commonly neglected by remote sensing and Geographic information system (GIS) users. In this work, a method is presented aimed at estimating the uncertainty related to area calculation of polygons in a vector map. Additionally, area error relationship with polygon geometric features was analyzed, as well. A multivariate regression-based approach was applied for this task. After presenting the method, to demonstrate its operational capabilities, it was applied to 2 case studies corresponding (i) to a forest map and (ii) to the map from the Geo Spatial Aid Application used for in the framework of the EU Common Agriculture Policy. Estimated uncertainty (percentage) median values were found to be 0.01% and 0.02% for (i) and (ii), respectively. It was also demonstrated that the same polygon shape built using less vertices (longer vectors) generates area estimates that are less accurate than the ones from polygons built using a higher number of vertices (shorter vectors).