A comparison of different semantic methods for integrating thematic geographical information: the example of land cover

Abstract

There is a generic problem when analysing multiple thematic datasets that spatial data of the same phenomenon will vary. Such variation is problematic for analysis that seek to integrate multitemporal data such as land cover, in order to identify the extent and location of any land cover change, and for models that require land cover as part of their input (e.g. climate change models). This paper compares and evaluates the effectiveness of different methods for integrating land cover data based on analysis of their semantics.