We apply a semi-analytic galaxy formation model to two high resolution cosmological N-body simulations to investigate analogues of the Milky Way system. We select these according to observed properties of the Milky Way rather than by halo mass as in
most previous work. For disk-dominated central galaxies with stellar mass (5--7) x 10d10Msun, the median host halo mass is 1.4 x 10d12Msun, with 1 sigma dispersion in the range [0.86, 3.1] x 10d12Msun, consistent with dynamical measurements of the Milky Way halo mass. For any given halo mass, the probability of hosting a Milky Way system is low, with a maximum of ~20% in haloes of mass ~10d12Msun. The model reproduces the V-band luminosity function and radial profile of the bright (MV < -9) Milky Way satellites. Galaxy formation in low mass haloes is found to be highly stochastic, resulting in an extremely large scatter in the relation between MV (or stellar mass) for satellites and the depth of the subhalo potential well in which they live, as measured by the maximum of the rotation curve, Vmax. We conclude that the too big to fail problem is an artifact of selecting satellites in N-body simulations according to subhalo properties: in 10% of cases we find that three or fewer of the brightest (or most massive) satellites have Vmax > 30 km/s. Our model predicts that around half of the dark matter subhaloes with Vmax > 20 km/s host satellites fainter than MV = -9 and so may be missing from existing surveys.
We measure the projected cross-correlation between low redshift (z < 0.5) far-IR selected galaxies in the SDP field of the Herschel-ATLAS (H-ATLAS) survey and optically selected galaxies from the Galaxy and Mass Assembly (GAMA) redshift survey. In or
der to obtain robust correlation functions, we restrict the analysis to a subset of 969 out of 6900 H-ATLAS galaxies, which have reliable optical counterparts with r<19.4 mag and well-determined spectroscopic redshifts. The overlap region between the two surveys is 12.6 sq. deg; the matched sample has a median redshift of z ~ 0.2. The cross-correlation of GAMA and H-ATLAS galaxies within this region can be fitted by a power law, with correlation length r_0 ~ 4.63 +/- 0.51 Mpc. Comparing with the corresponding auto-correlation function of GAMA galaxies within the SDP field yields a relative bias (averaged over 2-8 Mpc) of H-ATLAS and GAMA galaxies of b_H/b_G ~ 0.6. Combined with clustering measurements from previous optical studies, this indicates that most of the low redshift H-ATLAS sources are hosted by halos with masses comparable to that of the Milky Way. The correlation function appears to depend on the 250 um luminosity, L_250, with bright (median luminosity u L_250 ~ 1.6 x 10^10 L_sun) objects being somewhat more strongly clustered than faint ( u L_250 ~ 4.0 x 10^9 L_sun) objects. This implies that galaxies with higher dust-obscured star formation rates are hosted by more massive halos.
