Gaussianization: Enhancing the Statistical Power of the Power Spectrum

TL;DR

This paper introduces a Gaussianization technique, specifically a log transform, to improve the statistical power of the power spectrum in analyzing non-Gaussian cosmological fields, leading to tighter parameter constraints.

Contribution

It demonstrates that applying a nonlinear log transform to density fields reduces covariance in the power spectrum, enhancing cosmological parameter estimation.

Findings

Log transform reduces power spectrum covariance.

Tighter constraints on cosmological parameters achieved.

Applicable to low-redshift matter density fields.

Abstract

The power spectrum is widely used in astronomy, to analyze temporal or spatial structure. In cosmology, it is used to quantify large-scale structure (LSS) and the cosmic microwave background (CMB). This is because the power spectrum completely quantifies Gaussian random fields, which the CMB and LSS fields seem to be at early epochs. However, at late epochs and small scales, cosmological density fields become highly non-Gaussian. The power spectrum loses power to describe LSS and CMB fields on small scales, most obviously through high covariance in the power spectrum as a function of scale. Practically, this significantly degrades constraints that observations can place on cosmological parameters. However, if a nonlinear transformation that produces a (more) Gaussian 1-point distribution is applied to a field, the extra covariance in the field's power spectrum can be dramatically reduced. In the case of the roughly lognormal low-redshift matter density field, a log transform accomplishes this. Applying a log transform to the density field before measuring the power spectrum also tightens cosmological parameter constraints by a factor of several.