SPIDER-IX. Classifying galaxy groups according to their velocity distribution

Abstract

We introduce a new method to study the velocity distribution of galaxy systems, the Hellinger Distance (HD), designed for detecting departures from a Gaussian velocity distribution. Testing different approaches to measure normality of a distribution, we conclude that HD is the least vulnerable method to type I and II statistical errors. We define a relaxed galactic system as the one with unimodal velocity distribution and a normality deviation below a critical value (HD < 0.05). In this work, we study the Gaussian nature of the velocity distribution of the Berlind group sample, and of the FoF groups from the Millennium simulation. For the Berlind group sample (z < 0.1), 67 per cent of the systems are classified as relaxed, while for the Millennium sample we find 63 per cent (z = 0). We verify that in multi-modal groups the average mass of modes in high-multiplicity (N ≥ 20) systems are significantly larger than in low-multiplicity ones (N < 20), suggesting that groups experience a mass growth at an increasing virialization rate towards z = 0, with larger systems accreting more massive subunits. We also investigate the connection between galaxy properties ([Fe/H], Age, eClass, g-r, Rpetro and μpetro) and the Gaussianity of the velocity distribution of the groups. Bright galaxies (Mr =-20.7) residing in the inner and outer regions of groups do not show significant differences in the listed quantities regardless if the group has a Gaussian (G) or a Non-Gaussian (NG) velocity distribution. However, the situation is significantly different when we examine the faint galaxies (-20.7