Thursday

Technological triumphs have led to observational datasets of unprecedented size and accuracy becoming available for young star clusters in recent years. Consequently, we are in a better position than ever before to profoundly improve our understanding of the formation, evolution and interactions of massive stars and their host clusters. However, this advancement has come at a price: the datasets suffer from significant incompleteness due to the clusters’ natal nebulosity and the wavelengths they are observed at (e.g. the Gaia survey is optical-wavelength), leading to patchy 2D, and scarce 3D, positional star data being available for most clusters.

Valuable insights can be gained through measuring the relative observed positions of stellar members to each other inside clusters as a function of cluster age and comparing these to cluster models. Unfortunately, while the full impact of dataset incompleteness on the conclusions drawn from spatial distribution studies is unclear, previous studies suggest this is not trivial problem. Potential issues are whether (1) identified differences in the spatial behaviour of stars in an observed cluster are real or owing to members’ absence; (2) 2D perspective effects are hindering correct interpretation of the true 3D behaviour of a cluster.

Buckner et al. (2019) developed and released novel N-dimensional spatial analysis tool INDICATE to the community - a robust local statistic to study the intensity, correlation and spatial distribution of point processes, for use on discrete astronomical datasets in any parameter space. Independent of the shape, size and density of a sample, results obtained with INDICATE for astronomical observations and models are directly comparable without conversion. In this talk we discuss INDICATE’s ability to overcome observational biases and perspective effects to produce reliable and consistent results for datasets that are up to 83.3% incomplete.