Spectral Properties of Anomalous Microwave Emission in 144 Galactic Clouds

TL;DR

This work delivers the largest compilation of AME spectra in compact Galactic clouds by combining unprecedented low-frequency coverage from S-PASS, C-BASS, and QUIJOTE with 21 ancillary maps. The authors fit a data-driven SED using MCMC without informative priors, capturing AME amplitude, peak frequency, and width with improved precision and identifying 83 new AME sources. They show that AME spectra are generally broader than single-phase spinning-dust predictions, though the narrowest widths align with theory, and reveal a strong link between AME and thermal dust radiance as well as PAH tracers. The positive correlation between AME peak frequency and thermal dust temperature challenges current spinning-dust models, suggesting the need for self-consistent dust evolution and radiative-transfer treatments to realistically describe environmental dependencies and grain populations.

Abstract

Anomalous Microwave Emission (AME) is a diffuse microwave component thought to arise from spinning dust grains, yet remains poorly understood. We analyze AME in 144 Galactic clouds by combining low-frequency maps from S-PASS (2.3 GHz), C-BASS (4.76 GHz), and QUIJOTE (10-20 GHz) with 21 ancillary maps. Using aperture photometry and parametric SED fitting via MCMC methods without informative priors, we measure AME emissivity, peak frequency, and spectral width. We achieve peak frequency constraints nearly three times tighter than previous work and identify 83 new AME sources. AME spectra are generally broader than predicted by spinning dust models for a single phase of the interstellar medium, suggesting either multiple spinning dust components along the line of sight or incomplete representation of the grain size distribution in current models. However, the narrowest observed widths match theoretical predictions, supporting the spinning dust hypothesis. The AME amplitude correlates most strongly with the thermal dust peak flux and radiance, showing $\sim30$% scatter and sublinear scaling, which suggests reduced AME efficiency in regions with brighter thermal dust emission. AME peak frequency increases with thermal dust temperature in a trend current theoretical models do not reproduce, indicating that spinning dust models must incorporate dust evolution and radiative transfer in a self-consistent framework where environmental parameters and grain properties are interdependent. PAH tracers correlate with AME emissivity, supporting a physical link to small dust grains. Finally, a log-Gaussian function provides a good empirical description of the AME spectrum across the sample, given current data quality and frequency coverage.

Figures (10)

• Figure 1: SEDs for two sources: G159.02-33.88 (top) and G195.90-02.60 (bottom). In each panel, the solid black line shows the best-fit model, with $\chi^2_{\rm{red}} = 0.8$ and $0.7$, respectively. Individual realizations from the converged MCMC chain, illustrating model scatter, are shown in blue. Color-corrected flux densities are plotted as orange points, with hollow markers indicating data points excluded from the fit due to residual CO contamination or excessively high frequencies. Each individual best-fit model component is displayed in gray, with the dotted line representing the AME component. The lower sub-panels display residuals in units of statistical deviation from the fit, with the $1\sigma$ region shaded in light blue.
• Figure 2: Morphology of high-latitude source G159.02-33.88 in selected maps, consisting of a collection of dark clouds including LDN 1454/53/58 and DOBASHI 4162. The solid circle ($72'$ radius) marks the primary aperture, while the dashed annulus ($96'$–$120'$) represents the background region. The grid spans $6.7^\circ$ per side. The AME component is visible above 5 GHz, with a strong correspondence between the WMAP K-band (22.8 GHz) near the AME peak and the DIRBE 240 µ m (1249 GHz) map near the thermal dust peak. The reprocessed WISE 12 µ m PAH-dominated map from Meisner2014 is also shown. The color scale is linear, normalized to the pixel range of each image.
• Figure 3: Locations of the 144 AME sources with $>2\sigma$ detection, represented by green stars, on top of the DIRBE 240 µ m map.
• Figure 4: Distributions of AME parameters in the source sample. Each panel includes an inset in the top-right showing the median (black) and smallest (green) 1$\sigma$ uncertainties for that parameter among all sources. For $A_{\rm{AME}}$, the median and smallest uncertainties are 0.6 Jy and 0.1 Jy respectively, too small to be visible on A4 scale. A best-fit Gaussian is drawn as a red-dashed line on the $\nu_{\rm{AME}}$ and $W_{\rm{AME}}$ distributions. The red shaded region in the $W_{\rm{AME}}$ panel highlights the range of effective widths predicted by single-phase SpDust2 and SpyDust templates (see Table \ref{['tab:spdust_vs_lognormal']}).
• Figure 5: Comparison of this work (y-axis) with previous studies (x-axis) for $A_{\rm{AME}}$ (top panel) and $\nu_{\rm{AME}}$ (bottom panel). The dashed gray line represents a 1:1 relation.