Astrophysical aspects of dark matter direct detection: from the use of
the escape speed estimates to tests on halo models.

The WIMP velocity distribution, and in particular its high velocity
tail, has a strong impact on the way direct detection can constrain or
discover light WIMPs in the GeV mass range. Recently, there have been
important observational efforts to estimate the Galactic escape speed
at the position of the Earth, in particular the analysis published in
2014 by the RAVE Collaboration. Nevertheless, these new estimates
cannot be used blindly as they rely on assumptions in the dark halo
modeling, which induce tight correlations between the escape speed and
other local astrophysical parameters (e.g. the local circular speed
and the local dark matter density). We make a self-consistent study of
the implications of the RAVE results on direct detection assuming
isotropic dark matter velocity distributions, both Maxwellian and
ergodic, building in the latter case the dark matter phase-space
distribution from the Milky Way mass model via Eddington equation. We
evaluate the astrophysical uncertainties on direct detection limits
arising from the RAVE results. We finally consider anisotropic
velocity distributions, and we test our procedures on cosmological
simulations, which we analyze also to gain insights on the
astrophysical properties of the WIMP distribution which have an impact
on direct detection.