The 6x2pt method: supernova velocities meet multiple tracers

TL;DR

The paper introduces the 6x2pt method, combining galaxy and supernova data with velocities to significantly improve cosmological parameter constraints over traditional approaches.

Contribution

It develops a comprehensive 6x2pt analysis framework that integrates multiple power spectra and demonstrates substantial improvements in parameter estimation accuracy.

Findings

Shrinks uncertainty in $\sigma_8, \gamma$ plane by over 50%.

Achieves constraints comparable to CMB data with orthogonal degeneracies.

Improves $H(z)/H_0$ constraints by over 70% across redshifts.

Abstract

We present a new methodology to analyse in a comprehensive way large-scale and supernovae (or any other distance indicator) surveys. Our approach combines galaxy and supernova position and redshift data with supernova peculiar velocities, obtained through their magnitude scatter, to construct a 6x2pt analysis which includes six power spectra. The 3x3 correlation matrix of these spectra expresses exhaustively the information content of the surveys at the linear level. We proceed then to forecast the performance of future surveys like LSST and 4MOST with a Fisher Matrix analysis, adopting both a model-dependent and a model-independent approach. We compare the performance of the 6x2pt approach to the traditional one using only galaxy clustering and some recently proposed combinations of galaxy and supernovae data and quantify the possible gains by optimally extracting the linear information. We show that the 6x2pt method shrinks the uncertainty area in the $\sigma_8, \gamma$ plane by more than half when compared to the traditional method. The combined clustering and velocity data on the growth of structures has uncertainties at similar levels to those of the CMB but exhibit orthogonal degeneracies, and the combined constraints yield improvements of factors of 5 in each of the five cosmological parameters here considered. Concerning the model-independent results, we find that our method can improve the constraints on $H(z)/H_0$ in all redshift bins by more than 70% with respect to the galaxy clustering alone and by 30% when supernova velocities (but not clustering) are considered, reaching a precision of 3-4% at high redshifts.