Presentation #405.04 in the session Analysis of equilibrium collisionless systems: power and peril.
Constraining the connection between galaxies and their host dark matter halos is essential to our understanding of galaxy formation and evolution. Galaxy-halo connection is highly sensitive to cosmological parameters, particularly Ωm and σ8. However, there is an ongoing tension between these cosmological parameters inferred by Planck from CMB, and those inferred by usual probes of galaxy-halo connection like galaxy clustering and weak lensing. This tension might hint at new physics beyond ΛCDM, but could also be a result of systematics not properly accounted for in galaxy clustering and lensing, such as halo assembly bias.
Satellite kinematics provides a novel way to infer galaxy-halo connection that is independent of halo assembly bias. We build on the work by van den Bosch et al. (2019), to improve Basilisk (Bayesian hierarchical inference using satellite kinematics) into a competitive probe of galaxy-halo connection, at par with, and complementary to, clustering and lensing. We perform forward modelling of projected separation and line-of-sight velocity of satellite galaxies using spherical Jeans equation. Testing our method on SDSS-like mock surveys produce precise and unbiased constraints on input galaxy-halo connection and cosmology.
Applying Basilisk on SDSS DR7, we obtain extremely tight constraints on the distribution of galaxies in dark matter halos, as well as satellite velocity anisotropy, over a very wide halo mass and galaxy luminosity range. Our results are consistent with previous studies that used various probes of galaxy-halo connection. But, due to our improved constraining power, we note interesting deviations from the standard galaxy-halo connection model used in the literature. Finally, we combine satellite kinematics with galaxy clustering to put constraints on cosmological parameters. Our preliminary results are consistent with Planck, and could indicate some unaccounted-for systematics in clustering+lensing studies.