Full-sky, arcminute-scale, 3D models of Galactic microwave foreground dust emission based on filaments

TL;DR

The paper introduces DustFilaments, a comprehensive 3D full-sky model of Galactic dust emission based on filament populations, capable of reproducing observed power spectra and non-Gaussian features, aiding CMB foreground analysis.

Contribution

The novel DustFilaments model simulates Galactic dust emission with millions of filaments, capturing small-scale structures and frequency decorrelation effects for improved CMB foreground testing.

Findings

Reproduces Planck 353 GHz dust power spectra

Simulates small-scale non-Gaussianities and frequency decorrelation

Matches Minkowski functional measurements of dust non-Gaussianity

Abstract

We present the DustFilaments code, a full sky model for the millimeter Galactic emission of thermal dust. Our model, composed of millions of filaments that are imperfectly aligned with the magnetic field, is able to reproduce the main features of the dust angular power spectra at 353 GHz as measured by the Planck mission. Our model is made up of a population of filaments with sizes following a Pareto distribution $\propto L_a^{-2.445}$, with an axis ratio between short and long semi-axes $\epsilon \sim 0.16$ and an angle of magnetic field misalignment with a dispersion RMS($\theta_{LH}$)$=10$ degree. On large scales our model follows a Planck-based template. On small scales, our model produces spectra that behave like power-laws up to $\ell \sim 4000$ or smaller scales by considering even smaller filaments, limited only by computing power. We can produce any number of Monte Carlo realizations of small-scale Galactic dust. Our model will allow tests of how the small-scale non-Gaussianity affect CMB weak lensing, and the consequences for the measurement of primordial gravitational waves or relativistic light relic species. Our model also can generate frequency decorrelation on the Modified Black Body (MBB) spectrum of dust, and is freely adjustable to different levels of decorrelation. This can be used to test the performance of component separation methods and the impact of frequency spectra residuals on primordial $B$-mode surveys. The filament density we paint in the sky is also able to reproduce the general level of non-Gaussianities measured by Minkowski functionals in the Planck 353 GHz channel map.