A Morpho-Structural Analysis to Classify Building Centroids as Patterns of Urban Growth

This upload contains a dataset ready to be processed, an R script and a dataset with the associated results of the processing steps. The dataset is a GeoPackage file (EPSG:2154 - RGF93 / Lambert-93 - Projected) named “CV_PT_Ext_0217.gpkg” containing two layers (1) “CV_Building_PT”: a point feature class made of the centroids of residential buildings in 2002 and 2017 extracted from the French BD TOPO® (National Geographic Institute). This layer possesses two attribute columns indicative of the building presence for each period: “Pres_2002” and “Pres_2017” (modalities: “1” for building presence; “0” otherwise). (2) “Centre_Var_ext”: a polygon feature class related to the extent of the case study: a region in southern France named Centre-Var. The script performs a combination of variography analysis and morphological closings over the building centroids. The different bounding regions then allow classifying new residential buildings (the ones that appeared between 2002 and 2017) into different categories of patterns of urban growth according to their degrees of clustering/scattering and to their locations regarding existing urban areas. The GeoPackage associated with the results “CV_PT_Ext_0217_RES.gpkg” contains two additional layers (3) "PT_CLASS" a point feature class made of the new building centroids with an attribute column “cat” related to the classification outputs. The categories are as follows: 1 - clustered infill, 2 - scattered infill, 3 - clustered edge-expansion, 4 - scattered edge-expansion, 5 - clustered leapfrog and 6 - scattered leapfrog. (4) "BR_Clipping": a polygon feature class made of the different bounding regions clipped together.