Date Published:

June 1, 2017

Abstract:

We present a Bayesian phase-space reconstruction of the cosmiclarge-scale matter density and velocity fields from the Sloan DigitalSky Survey-III Baryon Oscillations Spectroscopic Survey Data Release 12CMASS galaxy clustering catalogue. We rely on a given Λ cold darkmatter cosmology, a mesh resolution in the range of 6-10 h^-1Mpc, and a lognormal-Poisson model with a redshift-dependent non-linearbias. The bias parameters are derived from the data and a generalrenormalized perturbation theory approach. We use combined Gibbs andHamiltonian sampling, implemented in the argo code, to iterativelyreconstruct the dark matter density field and the coherent peculiarvelocities of individual galaxies, correcting hereby for coherentredshift space distortions. Our tests relying on accurate N-body-basedmock galaxy catalogues show unbiased real space power spectra of thenon-linear density field up to k ˜ 0.2 h Mpc^-1, andvanishing quadrupoles down to r ˜ 20 h^-1 Mpc. We alsodemonstrate that the non-linear cosmic web can be obtained from thetidal field tensor based on the Gaussian component of the reconstructeddensity field. We find that the reconstructed velocities have astatistical correlation coefficient compared to the true velocities ofeach individual light-cone mock galaxy of r ˜ 0.68 including about10 per cent of satellite galaxies with virial motions (about r = 0.75without satellites). The power spectra of the velocity divergence agreewell with theoretical predictions up to k ˜ 0.2 h Mpc^-1.This work will be especially useful to improve, for example, baryonacoustic oscillation reconstructions, kinematic Sunyaev-Zeldovich,integrated Sachs-Wolfe measurements or environmental studies.

Website