Testing the Velocity Field in Non--Scale Invariant Cold Dark Matter Models

Abstract

We analyze the cosmic peculiar velocity field as traced by a sample of 1184 spiral, elliptical and S0 galaxies, grouped in 704 objects. We perform a statistical analysis, by calculating the bulk flow, Cosmic Mach Number and velocity correlation function for this sample and for mock catalogs extracted from a set of N--body simulations. We run four cold dark matter (CDM) simulations: two tilted models (with spectral index $n=0.6$ and $n=0.8$), the standard model ($n=1$) and a ``blue" one ($n=1.2$), with different values of the linear bias parameter $b$. By means of a Maximum Likelihood analysis we estimate the ability of our models to fit the observations, as measured by the above statistics, and to reproduce the Local Group properties. On the basis of this analysis we conclude that the best model is the unbiased standard model $(n,~b)=(1,~1)$, even though the overall flatness of the joint likelihood function implies that one cannot strongly discriminate models in the range $0.8 \le n \le 1$, and $1 \le b \le 1.5$. Models with $b \geq 2.5$ are rejected at the $95\%$ confidence level. For $n=0.8$ the values of $b$ preferred by the present analysis, together with the {\em COBE} data, require a negligible contribution to $\Delta T/T$ by gravitational waves. Finally, the blue model, normalized to {\em COBE}, does not provide a good fit to the velocity data.