Do We Have Sufficient Knowledge of the Galactic Foreground Emission in Cosmic Microwave Background Science?

Jia-Rui Li, Peibo Yuan, Yi-Fu Cai, Hao Liu · 2026

Read Do We Have Sufficient Knowledge of the Galactic Foreground Emission in Cosmic Microwave Background Science? by Jia-Rui Li, Peibo Yuan, Yi-Fu Cai, Hao Liu (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-14287.txt.

Introduction

Li and co-authors ask whether knowledge of Galactic foreground emission is good enough for cosmic microwave background work. Foregrounds can fake signals, as the BICEP2 dust episode showed in 2014. Most estimates model one component per emission type along each line of sight. The authors argue that real skies need several components, from depth structure and pixel mixing alike. Even the one two-component dust estimate fails, they show, since its emission parameters are too simple. CMB searches for primordial waves thus need richer foreground models.

Catalog record

CREATORJia-Rui Li, Peibo Yuan, Yi-Fu Cai, Hao Liu
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE56848
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.14287

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Do We Have Sufficient Knowledge of the Galactic Foreground Emission in Cosmic Microwave Background Science? Jia-Rui Li, Peibo Yuan, Yi-Fu Cai, Hao Liu Abstract Galactic foreground emission plays a key role in cosmic microwave background (CMB) science, particularly for detecting primordial gravitational waves. A well-known lesson is the “dust wave” identified by BICEP2 in 2014, which was ruled out through a more careful analysis of foreground emission. To date, most estimates of Galactic foreground emission have relied on the assumption that for each line of sight, only one component is considered per emission mechanism. However, the results in this work suggest that more complex modeling—particularly involving multiple components arising from either line-of-sight complexity or pixel mixing—may be necessary to fully account for Galactic foregrounds, including dust and other components. More interestingly, the only available two-component dust estimate also fails due to oversimplified emission parameters, although it is conceptually superior to single-component alternatives. These results yield three key conclusions: (1) Due to the intrinsic three-dimensional complexity of the Galactic environment, where physical conditions vary with both distance and direction, the actual radiation from Galactic foreground components cannot be accurately characterized by single-component models. (2) Consequently, CMB experiments require more frequency bands to resolve these components. (3) Spatial variations of foreground emission parameters should not be simplified, because in this work, all such simplifications are found to degrade the estimates significantly. I Introduction The investigation of the cosmic microwave background (CMB) has been integral to the evolution of modern cosmology, playing a crucial role in milestones such as the validation of the Big Bang theory ( 9 ) , the progress in understanding primordial nucleosynthesis ( 96 ) , and the refinement of the \rm{\Lambda} CDM model ( 2 ; 86 ; 61 ; 33 ) . Notably, the primordial CMB B -mode polarization, which is exclusively produced by primordial gravitational waves, offers a distinctive avenue for testing the validity of inflationary cosmology ( 83 ) . Unfortunately, foreground contamination presents challenges in the critical frequency range (70—220 GHz) for CMB studies. For frequencies below 80 GHz, synchrotron and free-free emissions must be taken into account, while thermal dust emission becomes increasingly dominant above 80 GHz ( 7 ; 60 ) . Additionally, anomalous microwave emission (AME) affects frequencies around 70 GHz ( 58 ) , and the cosmic infrared background (CIB) introduces further interference at frequencies above 300 GHz ( 64 ) . In the context of polarization, free-free emission is generally regarded as negligible ( 30 ; 41 ) , whereas the polarization of thermal dust emission can exhibit significant complexity ( 6 ) . Due to the presence of atmospheric water vapor, the atmospheric windows available to ground-based CMB telescopes are discontinuous across the microwave frequency range ( 19 ) . Thus, space-based observatories offer a significant advantage in estimating foreground emission parameters, particularly those related to thermal dust emission. Following the RELIKT-1 project ( 34 ) , which initiated space-based observations of the CMB, the COBE satellite pioneered measurements of CMB anisotropies ( 5 ) , paving the way for subsequent missions such as WMAP and Planck to advance precision cosmology. WMAP operates across five frequency bands from 23 to 94 GHz, but lacks high-frequency coverage ( 3 ) . In contrast, Planck extends the frequency range up to 857 GHz with its six High-Frequency Instrument (HFI) bands (100—857 GHz, 62 ) and three Low-Frequency Instrument (LFI) bands (30, 44, and 70 GHz, 65 ), enabling more precise measurements of thermal dust and CIB emissions ( 64 ) . LiteBIRD, a next-generation mission, will further expand observations across 15 frequency bands (40—402 GHz) using three telescope sets: LFT, MFT, and HFT ( 88 ; 82 ). Its key channels (100, 119, and 140 GHz) are designed to overlap with each other, reducing systematic errors and enhancing constraints on additional dust components such as AME ( 42 ) . PICO, another proposed space mission, will feature 21 frequency bands spanning 21—799 GHz and is designed to deliver superior sensitivity and angular resolution ( 22 ; 94 ) . With its extensive frequency coverage, PICO can enable more detailed modeling of foreground components and their polarization properties. However, the modeling of Galactic foregrounds in CMB research still faces several challenges: (1) Multiple foreground components including synchrotron, free-free emission, and AME, exhibit similar spectral behaviors at low frequencies ( 54 ) , leading to degeneracies that hinder an accurate component separation. (2) Current models of AME heavily rely on empirical fitting, with limited grounding in well-established physical mechanisms ( 54 ) , leaving considerable uncertainties. (3) The polarization of synchrotron and dust emission is spatially complex and poorly constrained ( 66 ) , resulting in significant modeling uncertainties that challenge the extraction of faint CMB primordial B modes ( 28 ) . Furthermore, the issue of multi-component thermal dust and synchrotron emission remains critical in polarization. Unlike temperature mode where intensities add up monotonically, the polarization signal is highly sensitive to the superposition of multi-component if they have misaligned polarization angles. Recently, we introduced a novel approach to address the complexity of foreground emission along the line of sight ( 43 ) . This approach investigates the relationship between linear regression ratios (noted as R ) and cross-correlation coefficients (noted as C ) for sky patches observed at different frequencies. We pointed out that although the expectations of R and C are independent, the deviation of C from 1 and the error of R are tightly linked. This connection enables a purely data-driven test of the model-to-data mismatch, which has ruled out existing single-component estimates of Galactic thermal dust emissions. For convenience, a brief methodology explanation can be found in Appendix A . In this paper, we extend our analysis to all accessible foreground estimates in the CMB channels, including both two-component thermal dust emission models and available single-component estimates of other emission mechanisms, such as synchrotron, AME, and free-free emission. The goal is to determine whether any of them is consistent with the observational data, or at least approaches the acceptance threshold. The structure of this work is straightforward: We briefly review the methodology in Appendix A and present results in Sections II — IV (covering thermal dust, synchrotron, and other emission mechanisms). Conclusions are given in Section V , with reproducible software available at GitHub, 1 1 1 https://github.com/liuhao-cn/foreground_validation which can also be used to evaluate consistency between any foreground model and observational data. II Test of thermal dust emission Since thermal dust emission dominates above 150 GHz, Planck ’s primary data products for this emission mechanism are sky maps at 353, 545, and 857 GHz. They exhibit significantly higher signal-to-noise ratios for thermal dust than other components/bands, making thermal dust the most studied foreground emission mechanism to date. Consequently, we initiate our analysis by evaluating all thermal dust emission models that we can find. Compared to 43 that tested only single-component modified blackbody models, this section expands the evaluation to include all major thermal dust models in current use, particularly two-component and physical modeling approaches, as listed in Table 1 and explained below: 1. Analytical models: This class contains single- and double-component modified blackbody models. 2. Physical models:

Read the complete text