We study the large-scale velocity fields traced by galaxy clusters in numerical simulations of a box of side 960 $h^{-1}$ Mpc, and compare them with available data on real clusters. In order to test the reliability of the simulations, which are based on an optimized version of the Zel'dovich approximation, we compare their cluster velocities with those of "exact" N-body simulations, and find a remarkable agreement between the two according to a variety of statistical tests. We analyse Cold Dark Matter models with density parameter in the range $0.2\le \Omega_0\le 1$, both with and without the cosmological constant term to provide a flat geometry. We also simulate a Cold+Hot Dark Matter model, with 30% provided by the hot component. Comparison with real data is performed by applying tests based on the cumulative velocity frequency distribution (CVFD) and bulk flow statistics. For the CVFD, we use observational velocity data from different authors. By merging all available data in a combined sample, the CVFD test is able to exclude only open models with $\Omega_0\le 0.4$ and flat models with $\Omega_0=0.2$. However, the analysis of individual observational samples gives contradictory results; in particular the recent Giovanelli data, which imply much lower velocities (with significantly smaller uncertainties) than the other data, are actually only consistent with the open $\Omega_0=0.4$ and the flat $\Omega_0=0.2$ models, i.e. the models which are excluded by the other data. Furthermore, CVFD and bulk flow analyses of the Branchini et al. reconstructed velocity data disfavour precisely those models accepted on the grounds of Giovanelli's sample. Finally, we confirm that the Lauer & Postman bulk flow determination is an extremely rare event in the cosmological models we have analysed.

We study the dipole for a redshift sample of Abell/ACO clusters. To elucidate the constraints it places on dark matter models, we use numerical simulations based on the Zel'dovich approximation. We run 20 realizations of each of six different dark matter models: four of these have a density parameter $\Omega_0=1$, while the other two have $\Omega_0=0.2$, one with and one without a cosmological constant term. For the Abell/ACO sample, we evaluate the parameter $\beta=\Omega_0^{0.6}/b_{ cl}$, which reaches its asymptotic value at $R=160 Mpc/h$. We find $\beta=0.15\pm 0.04$ and $0.25\pm 0.06$ in LG and CMB frames respectively. After identifying in the simulations LG-like observers, we construct mock cluster samples. We find that an alignment between the cluster dipole and observer velocity, such as that observed (about 20 degrees) should not be expected necessarily in the simulations. This, together with the large observer-to-observer variance estimates of $\beta$, makes it difficult to place any firm constraints on cosmological models. Our results demonstrate that the large amplitude and convergence depth of the cluster dipole cannot be taken as strong evidence either for or against a low--density Universe.