Towards new approaches to detect extragalactic globular clusters : the fornax cluster of galaxies as a study case

Abstract

Abstract: Globular clusters (GCs) are very old and dense stellar systems, ubiquitous across a variety of galaxies. The assembly history of GC systems is tightly related with the merger history of their host galaxies. Thus, they may provide extensive insights about the history of formation and evolution of galaxies. In this context, a proper identification of extragalactic GCs is of utmost importance. Furthermore, the new era of photometric surveys, represented for instance by the Vera Rubin Observatory’s Legacy Survey of Space and Time (LSST), calls for improvements in our capability of effectively classifying objects. It is essential to develop more versatile models that don’t rely on heavy case-by-case tuning. Traditional methods for selecting extragalactic GC candidates rely on cuts within certain color-color diagrams. However, the most promising color combinations require the use of infrared bands (e.g K band), not available in LSST, nor in the recently laucnched Euclid Space Telescope. Still, we expect that multi-band photometric data in the ugrizy LSST bands might hold intricate patterns which have note been fully explored yet. Hence, as a preparation for the LSST era, this work aims to assemble an LSST-like photometric catalogue to investigate the distribution of spectroscopically confirmed GCs across different diagrams of interest, making comparisons between the traditional color-color diagrams and other ones built with statistical tools. To achieve this, we shall study the GC system of the Fornax Cluster. That is because publicly available data from the Dark Energy Survey (DES) and Fornax Deep Survey (FDS) can be combined to obtain photometry in ugrizY bands. Moreover, the Fornax Cluster has a sample of spectroscopically confirmed GCs with about 1300 objects. Once an LSST-like catalogue is assembled, properly organized and filtered, we apply Principal Component Analysis to construct principal component diagrams, which help constrain the region occupied by the confirmed GCs. In order to quantitatively evaluate the effects of PCA in contrast with the distribution of points in color-color diagrams such as u - i vs i - Y , we fit 2D elliptical Gaussian distributions over the diagrams. The fitted spreads of the Gaussians express how compact the sample of confirmed GCs is. Our results suggest that the confirmed globular clusters in principal components diagrams are indeed grouped in a more compact region when compared with the most promising color-color diagrams (given the available ugrizY bands). We emphasize that this result points in the direction of the use of more general statistical methods to help us better utilize the available astronomical data to select cleaner samples of extragalactic GC candidates

Não inclui resumo.