Bayesian cosmic density field inference from redshift space dark matter maps.

Area of Science:

• Cosmology
• Astrophysics
• Statistical Physics

Background:

• Accurate reconstruction of primordial density fields is crucial for understanding cosmic evolution.
• Previous methods often neglected or iteratively corrected for redshift space distortions.
• Dark matter tracers like galaxy clusters and the intergalactic medium provide observational data.

Purpose of the Study:

• To develop a self-consistent Bayesian formalism for sampling primordial density fields.
• To incorporate redshift space distortions directly into the reconstruction process.
• To provide an unbiased and isotropic reconstruction of the dark matter density field.

Main Methods:

• Utilized a Hamiltonian Monte Carlo posterior sampling approach.
• Derived an analytic solution for coherent flows, including tidal fields and spherical collapse.
• Introduced novel algorithmic implementations for mass assignment kernels and time-step optimization.

Main Results:

• The developed algorithm, 'barcode', reconstructs isotropic primordial density fields.
• Power spectra of reconstructed fields are unbiased compared to mock observations.
• The method is tested within the Zel'dovich approximation and shows promising results.

Conclusions:

• The new Bayesian formalism accurately reconstructs dark matter density fields, accounting for redshift space distortions.
• The 'barcode' algorithm is suitable for analyzing the cosmic web down to megaparsec scales.
• This method can be applied to various cosmological surveys and serve as a baryon acoustic oscillation reconstruction algorithm.