The use of minimal spanning tree to characterize the 2D cluster galaxy distribution

We use the Minimal Spanning Tree to characterize the aggregation level of g iven sets of points. We test 3 distances based on the histogram of the MST edges to discriminate between the distributions. We calibrate the method by using artificial sets following Poisson, King or NFW distributions. The di stance using the mean, the dispersion and the skewness of the histogram of MST edges provides the more efficient results. We apply this distance to a subsample of the ENACS clusters and we show that the bright galaxies are si gnificantly more aggregated than the faint ones. The contamination provided by uniformly distributed held galaxies is negligible. On the other hand, w e show that the presence of clustered groups on the same cluster line of si ght masked the variation of the distance with the considered magnitude.