Deriving physical parameters of unresolved star clusters

VIII. Limits of aperture photometry for star cluster studies

Center for Physical Sciences and Technology, Saulėtekio av. 3, 10257 Vilnius, Lithuania
e-mail: vladas.vansevicius@ftmc.lt

Received: 21 February 2024
Accepted: 22 May 2024

Abstract

Context. Recently, it has been noticed that the discrepancies in the integrated colour indices (CIs) between star clusters and models are mostly due to the projection of bright stars in the apertures. In order to reduce this problem, the method of adaptive aperture photometry has been proposed. This method has been applied to star clusters from the M 31 Panchromatic Hubble Andromeda Treasury (PHAT) survey, and studies show that the adaptive aperture photometry performs better than the conventional approach.

Aims. The aim of this study is to determine the best achievable limits on the accuracy and applicability of the aperture photometry method for studying star clusters in the local Universe.

Methods. We computed a large network of artificial 3D star clusters spanning the parameter space of the M 31 clusters. We then simulated images of these clusters by projecting each onto a 2D plane from 100 directions. Star cluster images were generated in six passbands to match the PHAT survey. To investigate the limiting accuracy of aperture photometry and the limits of its applicability to star cluster studies, we measured the simulated images and performed parameter determination tests.

Results. We demonstrate that star clusters with and without post-main-sequence stars have significant photometric differences. We show that in order to obtain reliable physical parameters of star clusters, the CIs must be measured using an aperture with a radius larger than the cluster’s half-light radius. Furthermore, we demonstrate that the parameter determination of young clusters (∼10 Myr) is problematic regardless of the aperture size used. Therefore, it is advisable to determine the parameters of these clusters using colour-magnitude diagram fitting methods, when possible. We also show that the randomness of the viewing angle can lead to a CI uncertainty of up to 0.1 mag, depending on cluster parameters and aperture size.