Structural parameters of 389 local Open Clusters

Abstract

Context. The distribution of member stars in the surroundings of an open cluster can shed light on the process of its formation, evolution and dissolution. The analysis of structural parameters of open clusters as a function of their age and position in the Galaxy brings constraints on theoretical models of cluster evolution. The Gaia catalogue is very appropriate to find members of open clusters at large distance from their centers. Aims. We aim at revisiting the membership lists of open clusters from the solar vicinity, in particular by extending these membership lists to the peripheral areas thanks to Gaia EDR3. We then take advantage of these new lists of members to study the morphological properties and the mass segregation levels of the clusters. Methods. We used the clustering algorithm HDBSCAN on Gaia parallaxes and proper motions to systematically look for members up to 50 pc from the cluster centers. We fitted a King's function on the radial density profile of these clusters and a Gaussian Mixture Model on their two dimensional distribution of members to study their shape. We also evaluated the degree of mass segregation of the clusters and the correlations of these parameters with the age and Galactic position of the clusters. Results. Our methodology performs well on 389 clusters out of the 467 selected ones, including several recently discovered clusters that were poorly studied until now. We report the detection of vast coronae around almost all the clusters and report the detection of 71 open clusters with tidal tails, multiplying by more than four the number of such structures identified. We find the size of the cores to be on average smaller for old clusters than for young ones. Also, the overall size of the clusters seems to slightly increase with age while the fraction of stars in the halo seems to decrease. As expected the mass segregation is more pronounced in the oldest clusters but a clear trend with age is not seen. Conclusions. Open Clusters are more extended than previously expected, regardless of their age. The decrease in the proportion of stars populating the clusters halos highlights the different cluster evaporation processes and the short timescales they need to affect the clusters. Reported parameters like cluster sizes or mass segregation levels all depend on cluster ages but can not be described as single functions of time.

Unravelling UBC 274: A morphological, kinematical, and chemical analysis of a disrupting open cluster

“…To define the members of each cluster, they chose to use 0.5 as a probability cutoff, which was considered the best compromise between contamination and recovery rates for the 389 clusters they analysed. However, UBC 274 is quite different from that sample of clusters, particularly because it is located further away than the analysed clusters in Tarricq et al (2021b), and this option is not well adapted in this case. We based our probability cutoff on the analysis of the performance of the algorithm on simulations of synthetic clusters.…”

“…As noticed in Tarricq et al (2021b), the distribution of the member's probability depends on multiple factors, including the heliocentric distance of the cluster and the relative density of the cluster with respect to the surrounding field in the proper motion and parallax space. To define the members of each cluster, they chose to use 0.5 as a probability cutoff, which was considered the best compromise between contamination and recovery rates for the 389 clusters they analysed.…”

“…We obtained a list of members of the cluster using the same method as Tarricq et al (2021b) and taking the mean cluster position and velocity computed by Cantat-Gaudin et al (2020) as a starting point. We did a cone search in the Gaia EDR3 archive centred on the cluster with a 50 pc radius, and up to magnitude G = 19.…”

“…We used the Python implementation of Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDB-SCAN) (Campello et al 2013;McInnes et al 2017) to identify the cluster members in the (µ α * , µ δ , ) space. We used the same configuration of the HDBSCAN hyperparameters as Tarricq et al (2021b). To compute membership probabilities, we performed 100 runs of the algorithm, using a random sampling on the proper motions and parallax according to their full covariance matrix.…”

“…Thanks to observations at MIKE on the 6.5m Clay Magellan Telescope in Las Campanas Observatory tidal structures, sometimes extending more than 100 pc from the cluster centre (Meingast & Alves 2019;Meingast et al 2021). Recently, Tarricq et al (2021b) used EDR3 data to perform a larger scale study of the 2D morphology of the OC population (up to 1.5 kpc from the Sun), and they found remarkable elongations and tidal tails in the plane of the sky for 70 OCs.…”