CALSAGOS: Clustering ALgorithmS Applied to Galaxies in Overdense Systems

TL;DR

CALSAGOS is a Python package that uses clustering algorithms to identify galaxy cluster members and substructures with high accuracy, tested on mock catalogues, aiding astrophysical research.

Contribution

The paper introduces CALSAGOS, a novel Python tool for galaxy cluster analysis that effectively identifies members and substructures using clustering algorithms.

Findings

Member selection error of 1-6% depending on the function used

F1-score of 0.8, precision of 85%, and completeness of 100% for outer substructure detection

Reduced performance (F1-score 0.5) when identifying all substructures due to 2D projection limitations

Abstract

In this paper we present CALSAGOS: Clustering ALgorithmS Applied to Galaxies in Overdense Systems which is a PYTHON package developed to select cluster members and to search, find, and identify substructures. CALSAGOS is based on clustering algorithms and was developed to be used in spectroscopic and photometric samples. To test the performance of CALSAGOS we use the S-PLUS's mock catalogues and we found an error of 1\% - 6\% on member selection depending on the function that is used. Besides, CALSAGOS has a $F_1$-score of 0.8, a precision of 85\% and a completeness of 100\% in the identification of substructures in the outer regions of galaxy clusters ($r > r_{200}$). The $F_1$-score, precision and completeness of CALSAGOS fall to 0.5, 75\% and 40\% when we consider all substructure identifications (inner and outer) due to the function that searches, finds, and identifies the substructures works in 2D and cannot resolve the substructures projected over others.