Improving models of the cosmic infrared background using CMB lensing mass maps

TL;DR

This paper demonstrates how cross-correlating the cosmic infrared background with CMB lensing mass maps significantly enhances the precision of star formation models and reduces sample variance in measurements.

Contribution

It introduces a method to improve CIB parameter constraints by using CMB lensing maps, reducing sample variance and refining redshift distribution estimates.

Findings

20-100% improvement in star formation model parameters

Enhanced constraints on CIB redshift distribution

Reduced sample variance through cross-correlation with CMB lensing

Abstract

The cosmic infrared background (CIB) sourced by infrared emission from dusty star-forming galaxies is a valuable source of information on the star formation history of the Universe. In measurements of the millimeter sky at frequencies higher than $\sim 300$ GHz, the CIB and thermal emission from Galactic dust dominate. A limited understanding of the CIB contribution at lower frequencies on the other hand can hinder efforts to measure the kinetic Sunyaev-Zeldovich spectrum on small scales as well as new physics that affects the damping tail of the cosmic microwave background (CMB). The Planck satellite has measured with high fidelity the CIB at 217, 353, 545 and 857 GHz. On very large scales, this measurement is limited by our ability to separate the CIB from Galactic dust, but on intermediate scales, the measurements are limited by sample variance in the underlying matter field traced by the CIB. We show how significant improvements (20-100%) can be obtained on parameters of star formation models by cross-correlating the CIB (as measured from existing {\it Planck} maps or upcoming CCAT-prime maps) with upcoming mass maps inferred from gravitational lensing of the CMB. This improvement comes from improved knowledge of the redshift distribution of star-forming galaxies as well as through the use of the unbiased matter density inferred from CMB lensing mass maps to cancel the sample variance in the CIB field. We also find that further improvements can be obtained on CIB model parameters if the cross-correlation of the CIB with CMB lensing is measured over a wider area while restricting the more challenging CIB auto-spectrum measurement to the cleanest 5% of the sky.