Shapes of clusters and groups of galaxies: comparison of model predictions with observations

PAZ, D. J.; MERCHÁN, M.; PADILLA, N.; LAMBAS, D. G.

Pucón

Congreso; XI Reunión Regional Latinoamericana de Astronomía; 2005

Unión Astronómica Internacional

We analyze the 3-DIM dark matter shapes of groups and clusters in a high resolution cosmological simulation to investigate into detail the dependence of the ellipticity distribution of groups on characteristics such as mass and number of members. We find that the triaxial nature of groups, derived from our analysis of the principal axes, is a strong function of the number of members. Groups tend to be prolate when they contain more particles. We find that this effect is mainly due to an artifact of the low number statistics. That means that the general triaxial shape of halos resolved with less particles have a systematic bias toward the oblate limit. This issue is particularly important in observed systems which have a low number of detected members, and low resolution simulations. Despite this resolution effect, we find a significant correlation between the mass and the halo shape. Masive systems are still more prolate than the smaller ones. We have also studied the distribution of ellipticities of groups of galaxies selected from the 2PIGG (Eke et al. 2004, MNRAS, 348, 866), and SDSSDR3GG (Merchan & Zandivarez 2005, ApJ, in press, astro-ph/0412257). The distribution of observed ellipticities are found to be a strong function of number of members, so that poor groups are more elongated than rich ones. However, this is again an artifact caused by poor statistics and not an intrinsic property of the groups as it has been suggested previously (Plionis et al. 2004, MNRAS, 352, 132). In order to confront the observational data with numerical models, we construct a mock 2PIGG catalgue from our numerical simulation box, populated by galaxies with properties obtained using a semianalytic model (Cole et al. 2000, MNRAS, 319, 168). By comparing the properties of the ellipticity distributions of the mock and real catalogues, we find that the mock catalogue remarkably reproduces the characteristics of the real data, in excellent agreement with the observations. We do not find any significant correlation, in real catalogues, between the shape of the observed groups, and properties of the member galaxies like color or spectral index.