🔭 astrophysics

All-sky modeling of Galactic emission at radio and microwave frequencies

This paper presents a new all-sky model of low-frequency Galactic emission that integrates recent radio and microwave surveys with Planck data via a Bayesian approach, yielding updated synchrotron and spinning dust maps with significantly higher amplitude estimates and reduced residuals compared to previous models.

Original authors: Gabriel A. Hoerning, Clive Dickinson, Stuart E. Harper, Roke Cepeda-Arroita, Hans K. Eriksen, Melis O. Irfan, J. Patrick Leahy, Jamie Leech, Michael E. Jones, Timothy J. Pearson, Michael W. Peel, Vasu

Published 2026-06-23

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Gabriel A. Hoerning, Clive Dickinson, Stuart E. Harper, Roke Cepeda-Arroita, Hans K. Eriksen, Melis O. Irfan, J. Patrick Leahy, Jamie Leech, Michael E. Jones, Timothy J. Pearson, Michael W. Peel, Vasundhara Shaw, Angela C. Taylor, Duncan J. Watts, Ingunn K. Wehus, Gilles Weymann-Despres

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the night sky not as a dark void, but as a bustling, noisy radio station. If you were to tune a radio to different frequencies, you wouldn't just hear static; you would hear a complex symphony of different "voices" coming from our own galaxy, the Milky Way. Some voices are the crackle of high-speed electrons (synchrotron), some are the hiss of hot gas (free-free), and others are the hum of tiny, spinning dust grains (spinning dust).

The problem for astronomers is that these voices all overlap. It's like trying to listen to a single violin in an orchestra where the drums, trumpets, and bass are all playing at once. For years, scientists have been trying to figure out which voice belongs to which instrument so they can hear the faintest signal of all: the Cosmic Microwave Background (CMB), which is the leftover echo of the Big Bang.

This paper presents a new, much clearer way to separate these voices. Here is how they did it, explained simply:

1. The Old Map vs. The New Map

Previously, scientists tried to map these radio voices using a very old, low-resolution map from 1982 (called the Haslam map) as a starting point. It was like trying to draw a modern city map using a sketch from the 1800s. They had to guess how the signals changed as they moved from low radio frequencies to higher microwave frequencies.

In this new study, the team didn't rely on that old sketch. Instead, they built a brand-new model using a fresh set of "ears" (telescopes and surveys) that are much more sensitive and cover the sky better. They used data from:

S-PASS and C-BASS: New surveys that act like high-fidelity microphones for the low-frequency radio sounds.
QUIJOTE: A telescope in Tenerife that listens to the middle frequencies.
WMAP and Planck: Space telescopes that listen to the higher microwave frequencies.

2. The "Mixing Board" Approach

The team used a sophisticated computer program called Commander. Think of this program as a master audio mixing board.

The Goal: The computer takes all the noisy recordings from the different telescopes and tries to figure out exactly how loud each "instrument" (synchrotron, free-free, spinning dust) is at every single spot in the sky.
The Method: Instead of guessing, the computer uses a "Bayesian" approach. This is like a detective who has a set of rules about how the instruments should sound (based on physics) and then adjusts the volume knobs until the combined sound perfectly matches the actual recordings from the telescopes.

3. The Big Discovery: The Radio Voice is Louder

The most surprising result of this new model is that the "synchrotron" voice (the sound of electrons spiraling in magnetic fields) is actually much louder than scientists previously thought.

At a specific frequency (4.76 GHz), the new model shows this signal is about twice as strong as the old Planck 2015 model predicted.
Why the difference? The old model relied on that 1982 sketch and assumed the sound got quieter in a predictable way as frequencies changed. The new model, using the fresh data from C-BASS and S-PASS, realized the sound doesn't fade as quickly as they thought. It's like realizing a singer is actually much louder than you thought because you finally put on better headphones.

4. Cleaning Up the Noise

The team also improved how they handle the other voices:

Spinning Dust: They found a better way to model the "hum" of spinning dust grains. Instead of assuming they all spin at the same speed, they allowed the model to show that in some parts of the galaxy, these grains spin faster or slower, changing the pitch of the hum.
Free-Free (Hot Gas): They found that in the crowded center of the galaxy, the hot gas signal is actually a bit quieter than previously thought, because the new model realized the other voices (synchrotron and spinning dust) were doing more of the work than they gave them credit for.

5. The Result: A Crystal Clear Sky

The final product is a new, all-sky map at 4.76 GHz.

Accuracy: The model is so good that the "leftover noise" (what's left after they subtract their model from the real data) is incredibly small—less than 10 micro-Kelvin over 95% of the sky. That's like hearing a whisper in a library and realizing 99% of the sound you hear is just the library's own quiet hum, not a secret conversation.
No More Guessing: Because they used data that covers the whole sky at these specific frequencies, they didn't have to guess or extrapolate as much as before. They have a direct measurement of the sky's "volume" at this frequency.

Why This Matters

This new map is like a high-definition reference guide for the radio sky.

For Cosmologists: It helps them subtract the galaxy's noise more accurately so they can hear the faint echo of the Big Bang (the CMB) without distortion.
For Simulators: It gives scientists a realistic "soundtrack" to use when they build computer simulations of the universe. If they want to test how their telescopes will work in the future, they can use this new, more accurate map to see what the sky actually looks like.

In short, this paper says: "We stopped using the old, blurry sketch of the radio sky. We used new, high-tech ears to listen to the whole galaxy, and we found out the radio signals are much louder and more complex than we thought. Now we have a much better map to help us understand the universe."

Technical Summary: All-sky modeling of Galactic emission at radio and microwave frequencies

Problem Statement
Accurate modeling and separation of Galactic foregrounds are critical for cosmological analyses, particularly the extraction of the Cosmic Microwave Background (CMB) and the study of the Interstellar Medium (ISM). At low frequencies ( $\lesssim 10$ GHz), the sky is dominated by synchrotron, free–free, and spinning dust emissions. However, separating these components is challenging due to degeneracies in their Spectral Energy Distributions (SEDs), especially in the radio-to-microwave transition region ( $5–30$ GHz). Previous models, such as the Planck 2015 analysis, relied on limited frequency coverage and simplified assumptions, such as a spatially constant synchrotron spectral index ( $\beta_s \approx -3.1$ ) and the use of the Haslam 408 MHz map as a primary synchrotron tracer. The Haslam map, while high signal-to-noise, contains known large-scale systematics and requires extrapolation across a wide frequency gap to microwave bands, introducing potential biases. Furthermore, the lack of complementary low-frequency surveys near CMB frequencies hindered the independent constraint of component SEDs.

Methodology
The authors present a new all-sky model of low-frequency diffuse Galactic emission using a Bayesian parametric component-separation approach implemented with the Commander code. The analysis focuses on total intensity (Stokes I) at a resolution of $1^\circ$ (NSIDE = 256).

Data Integration: The model incorporates a comprehensive set of 35 full- and partial-sky maps spanning 2.30 GHz to 857 GHz. Key datasets include:
- S-PASS (2.30 GHz): A southern-sky survey used to anchor the synchrotron component in the south.
- C-BASS (4.76 GHz): A northern-sky survey serving as the primary synchrotron anchor in the north, replacing the Haslam 408 MHz map.
- QUIJOTE (11–19 GHz): Used to constrain the peak frequency and spectral width of spinning dust, restricted to Galactic plane regions to mitigate radio-frequency interference (RFI) and large-scale filtering effects.
- WMAP and Planck (LFI/HFI): Reprocessed Cosmoglobe WMAP and Planck LFI data, along with Planck HFI channels, bridge the transition to thermal dust and provide high-frequency constraints.
Parametric Model:
- Synchrotron: Modeled with spatially varying amplitude ( $A_s$ ) and spectral index ( $\beta_s$ ) using a power-law SED. The model avoids the sub-gigahertz regime to prevent curvature complications, anchoring directly at 2.30 and 4.76 GHz.
- Free–free: Modeled using an emission measure (EM) and electron temperature ( $T_e$ ), constrained by an $H\alpha$ template (corrected for dust extinction) in high-latitude regions and pixel-by-pixel fitting in the Galactic plane.
- Spinning Dust: Modeled as a single-component log-normal SED with spatially varying amplitude, peak frequency ( $\nu_p$ ), and spectral width ( $w$ ). This replaces the two-component fixed-template approach of Planck 2015.
- Other Components: Thermal dust, Sunyaev–Zeldovich (SZ), and molecular line emissions follow the Planck 2015 strategy.
Inference: The analysis employs a Gibbs sampling chain of 40,000 samples to derive posterior distributions for amplitude and spectral parameters. Instrumental gains and monopole/dipole offsets are treated as nuisance parameters, with C-BASS and S-PASS defining the absolute amplitude scale for synchrotron in their respective hemispheres.

Key Contributions and Results

Revised Synchrotron Amplitude: The inclusion of C-BASS and S-PASS as direct anchors results in a synchrotron amplitude at 4.76 GHz that is approximately two times higher than the Planck 2015 prediction. A T–T analysis reveals the Planck 2015 model underestimated the amplitude by $\sim 50\%$ . This discrepancy is attributed to the flatter effective spectral index ( $\beta_{0.408-4.76} \approx -2.82$ ) observed between C-BASS and Haslam, compared to the assumed $\beta = -3.1$ , and the mitigation of Haslam-related large-scale systematics.
Spatially Varying Spectral Parameters: The model produces full-sky maps of the synchrotron spectral index ( $\beta_s$ ) and spinning dust parameters. The global median synchrotron index is $\tilde{\beta}_s = -3.18$ , with a dispersion of $\sigma_{sky} \approx 0.13$ . The spinning dust peak frequency ( $\nu_p$ ) is found to vary spatially, typically between 25–30 GHz in the inner Galactic plane and lower in diffuse regions, with specific sources (e.g., California Nebula) showing $\nu_p > 40$ GHz.
Improved Free–free Separation: The revised modeling reduces the free–free amplitude in the Galactic plane by $\sim 10\%$ compared to Planck 2015. This is a consequence of the increased flexibility in synchrotron and spinning dust descriptions, which reduces the need for free–free to absorb spectral mismatches.
Goodness-of-Fit: The model achieves RMS temperature residuals below 10 $\mu$ K over 95% of the sky up to 353 GHz. Fractional residuals are below 1.5% in the Galactic plane and below 5% across QUIJOTE bands. Residual angular power spectra lie more than two orders of magnitude below the CMB spectrum for multipoles $\ell < 192$ .
Reconstructed 4.76 GHz Map: A primary product is a reconstructed all-sky total-intensity map at 4.76 GHz. This map traces diffuse synchrotron emission with reduced systematics relative to the Haslam 408 MHz map and serves as a physically motivated intermediate-frequency anchor for sky simulations.

Significance
The paper claims that this work provides a new reference for foreground modeling and sky-simulation applications (e.g., Planck Sky Model, PySM). By leveraging contemporary, well-calibrated datasets (C-BASS, S-PASS, QUIJOTE) rather than relying on extrapolations from 408 MHz or simplified spectral assumptions, the model offers a more physically motivated description of the low-frequency sky. The results demonstrate that joint analyses combining multiple datasets are essential for breaking degeneracies and achieving robust component separation. The derived products describe the transition between radio and microwave emission, enabling more accurate CMB extraction and ISM characterization. The authors note that while the model is robust, the absolute sky zero level remains a relative framework, and future absolutely calibrated experiments will be required for a fully self-consistent determination of the sky's absolute temperature.