Using the Urban Atlas dataset for estimating spatial metrics. Methodology and application in urban areas of Greece

The paper discusses the use of the Urban Atlas land use dataset for estimating spatial metrics to analyse the form of urban areas. Using a cartographic generalization process the original Urban Atlas data are transformed by merging neighboring patches of the same land use class to a single patch thus forming larger land use patches. The resulting database is used for estimating spatial metrics and fractal dimension of nine Greek cities at the class and landscape level. Metrics estimated with this procedure reflect the land use distribution and sidestep the problem introduced by the high resolution of Urban Atlas data in which every land use patch is a city block. The metrics are used to analyze the form of the various cities and pinpoint differences and similarities among them. The results demonstrate that with spatial metrics and particularly with class metrics, it is possible to quantify the spatial heterogeneity and fragmentation of urbanization patterns, as well as, the structure and composition of land uses, thus obtaining insight on the main morphological characteristics of each city.