Dark matter in Draco and the Local Group: Implications for direct detection experiments

We use a cosmological simulation of the Local Group to make quantitative and speculative predictions for direct detection experiments. Cold dark matter (CDM) halos form via a complex series of mergers, accretion events and violent relaxation which precludes the formation of significant caustic features predicted by axially symmetric collapse. The halo density profiles are combined with observational constraints on the galactic mass distribution to constrain the local density of cold dark matter to lie in the range $0.18\lesssim {\rho }_{{}_{\mathrm{CDM}}}{(R}_{\odot })/({\mathrm{GeV}\mathrm{}\mathrm{}\mathrm{cm}}^{-3})\lesssim \mathrm{0.30.}$ In velocity space, coherent streams of dark matter from tidally disrupted halos fill the halo and provide a tracer of the merging hierarchy. The particle velocities within triaxial CDM halos cannot be approximated by a simple Maxwellian distribution and are radially biased at the solar position. The detailed phase space structure within the solar system will depend on the early merger history of the progenitor halos and the importance of major mergers over accretion dominated growth. We follow the formation of a “Draco” sized dSph halo of mass ${10}^8{M}_{\odot }$ with several million particles and high force accuracy. Its internal structure and substructure resemble that of galactic or cluster mass halos: the density profile has a singular central cusp and it contains thousands of subhalos orbiting within its virial radius demonstrating a self-similar nature to collisionless dark matter subclustering. The singular cores of substructure halos always survive complete tidal disruption although mass loss is continuous and rapid. Extrapolating wildly to earth mass halos with velocity dispersion of 1 ${\mathrm{m}\mathrm{}\mathrm{s}}^{-1}$ (roughly equal to the free streaming scale for neutralinos) we find that most of the dark matter may remain attached to bound subhalos. Further numerical and analytic work is required to confirm the existence of a detectable smooth component.