Multiscale Spatial Analysis of Young Stars Complex using the dbscan Clustering Algorithm

Clustering and spatial substructures studies of young stellar objects (YSOs) in star forming regions are key tracers to identify (1) the properties of the birth sites and (2) the dynamical evolution cluster structure with time. This work presents an approach identifying the multilevel topological substructures of star forming regions. In this preliminary work, we use recursively the dbscan density based clustering algorithm to identify the spatial multilevel substructures. From these multilevel substructures we build a clusterTree, the analogous of dendrogram, and an associated multilevel spectrum, and we show that they can be used to characterize the YSO spatial distribution. We first apply this procedure on different types of distributions (uniform, fractal, Plummer). We then apply this procedure to analyze the Taurus YSOs complex.