🔭 astrophysics

Status and future development of the COSMOCal Project for absolute CMB polarization calibration

This paper presents the status, updated timeline, and refined requirements of the COSMOCal project, which proposes a geostationary artificial calibration source to address instrumental systematics in CMB polarization measurements, while also investigating how residual calibration errors impact foreground cleaning and the recovery of primordial signals.

Original authors: Silvia Micheli, Alessia Ritacco, Alessandro Carones, Jonathan Aumont, Stefano Berta, Ludovico Bizzarri, François Boulanger, Andrea Catalano, Francesco Cuttaia, Anne Denis, François-Xavier Désert, Anne

Published 2026-07-15

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Silvia Micheli, Alessia Ritacco, Alessandro Carones, Jonathan Aumont, Stefano Berta, Ludovico Bizzarri, François Boulanger, Andrea Catalano, Francesco Cuttaia, Anne Denis, François-Xavier Désert, Anne Laure Fontana, David González-Ovejero, Benedetta Kalemi, Nicoletta Krachmalnicoff, Samuel Leclercq, Thibaut Louis, Juan-Francisco Macías-Pérez, Bruno Maffei, Tania Mcnamara, Marina Migliaccio, Ludovic Montier, Pascal Morfin, Michel Moulin, Louise Mousset, Matteo Murgia, Ioannis Myserlis, Federico Nati, Pierluigi Ortu, Michel Pérault, Giampaolo Pisano, Tonino Pisanu, Nicolas Ponthieu, Sofia Savorgnano, Luca Terenzi, Jeanne Treuttel, Léo Vacher, Christophe Vescovi, Mario Zannoni

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 universe as a giant, ancient radio station broadcasting a secret message from the very first moments of time. This message is the Cosmic Microwave Background (CMB), a faint glow of light left over from the Big Bang. Scientists are trying to tune into a very specific part of this broadcast: a whisper called "B-modes," which would prove that the universe expanded incredibly fast right after it began.

But here's the problem: trying to hear this whisper is like trying to listen to a single pin drop in a stadium full of screaming fans. The "fans" are the messy, bright emissions from our own galaxy, and the "pin drop" is so faint that even the tiniest error in your listening equipment can make you think you heard the pin when you didn't.

The Tuning Problem

To hear the whisper, scientists need to tune their instruments with extreme precision. Specifically, they need to know the exact angle at which their detectors are looking. If the angle is off by even a tiny bit—about 0.1° (which is like missing the bullseye on a dartboard from a mile away)—the signal gets scrambled.

Think of it like wearing 3D glasses. If your glasses are tilted just a little, the 3D effect breaks, and you see a blurry mess instead of a sharp image. In the universe, if the "glasses" (the detectors) are tilted, the "gradient" signals (E-modes) leak into the "curl" signals (B-modes). This leakage creates a fake signal that looks exactly like the secret message scientists are hunting for.

The Old Way vs. The New Plan

For a long time, scientists tried to fix their glasses by looking at the sky itself. They assumed that the universe shouldn't have a certain type of connection between its signals (called an EB correlation). If they saw that connection, they figured, "Oh, our glasses must be tilted!" and they would twist them back.

But this method has a big flaw: it's like trying to fix a broken compass by assuming the Earth's magnetic field is perfect. If the Earth's field is actually wobbly (which it might be, due to complex dust in space), your fix will be wrong. Plus, this method accidentally hides a different, real phenomenon called "cosmic birefringence," which is a mysterious rotation of light that could reveal new physics.

So, the COSMOCal project proposes a radical new idea: bring your own reference point.

The Cosmic "Calibration Star"

Instead of guessing based on the messy sky, COSMOCal plans to build a special, artificial lighthouse in space. This isn't a natural star; it's a man-made device that will be attached to a satellite sitting in a geostationary orbit (staying fixed over the same spot on Earth) by 2030.

This "lighthouse" will beam out a perfectly clean, stable, and known polarized signal. Ground-based telescopes in Chile, Spain, and Italy will point their dishes at this satellite. Because they know exactly what the signal should look like, they can check their instruments and say, "Aha! Our angle is off by 0.05 degrees. Let's fix it."

This turns the calibration from a guessing game into a precise science. Once these telescopes are calibrated, they can create a "gold standard" map of the universe that other telescopes (like the LiteBIRD satellite) can use to check their own work.

The Dusty Twist

The paper also tackles a tricky complication: space isn't empty; it's filled with cosmic dust. This dust doesn't just sit there; it glows and has its own complex patterns that can mess up the measurements.

To test if their new plan works, the authors ran simulations (computer models of the universe) using 100 different versions of the sky. They included very complex models of this dust, which have a hidden "twist" in their own signals (an intrinsic EB correlation) that makes them look like the very errors scientists are trying to avoid.

What did they find?
In these simulations, even with the messy dust and the complex models, the team showed that if they use the COSMOCal satellite to calibrate first, they can still figure out the exact angle of their instruments. They were able to recover the correct angle of 0.1° with high accuracy (within about 8% uncertainty in their models).

However, they also found a warning: if they only looked at the very largest, fuzzy parts of the sky (low numbers called "low multipoles"), the dust tricked them, and the angle looked wrong. But when they looked at the whole sky, including the smaller details, the math worked out perfectly.

The Bottom Line

The paper doesn't claim to have solved the mystery of the universe yet. Instead, it suggests that the COSMOCal project is a vital tool that could solve the calibration problem. By providing a stable, artificial reference point, it allows scientists to stop guessing and start measuring with the precision needed to finally hear that ancient whisper from the Big Bang. The simulations show that this approach is robust enough to handle the messy reality of cosmic dust, paving the way for future experiments to hunt for the truth without getting fooled by the noise.

Technical Summary: Status and Future Development of the COSMOCal Project for Absolute CMB Polarization Calibration

Problem Statement
Next-generation Cosmic Microwave Background (CMB) experiments, such as LiteBIRD and the Simons Observatory (SO), aim to detect primordial B-modes with a tensor-to-scalar ratio sensitivity of $r < 0.01$ . Achieving this requires controlling instrumental systematics to unprecedented levels, particularly the absolute calibration of detector polarization angles. Current calibration methods face significant limitations:

Astrophysical Calibrators: Standard celestial sources like the Crab Nebula (Tau A) have complex, frequency-dependent polarized emission with polarization angle uncertainties of $\sim 1^\circ$ , insufficient for the required $0.1^\circ$ accuracy.
Self-Calibration: The standard "EB nulling" technique assumes the intrinsic cosmological EB cross-correlation vanishes. However, this method cannot distinguish between instrumental mis-calibration ( $\Delta\alpha$ ) and cosmic birefringence (a parity-violating rotation of the polarization plane). Furthermore, Galactic foregrounds (synchrotron and thermal dust) exhibit intrinsic EB correlations and spectral complexities that can mimic instrumental leakage or birefringence, creating degeneracies that obscure the primordial signal.

Methodology
The paper addresses these challenges through two primary approaches: the definition of the COSMOCal project and a simulation-based analysis of calibration strategies under complex foreground conditions.

The COSMOCal Project:
- Concept: An artificial, stable, linearly polarized microwave source deployed as a guest payload on a geostationary satellite (Eutelsat), scheduled for launch by 2030.
- Instrument Design: The payload features eight fixed horns aligned with specific observatories: two pairs at 90/150 GHz for Chile (SO-LAT), one pair at 270 GHz for the IRAM 30m Telescope (Spain), and one pair at 90 GHz for the Sardinia Radio Telescope (Italy).
- Operation: The source emits fully linearly polarized signals with a $45^\circ$ relative offset between horn pairs. Its geostationary position ensures a fixed elevation for ground-based telescopes, allowing long integration times and stable background conditions.
- Role: It provides a primary absolute calibration reference for pilot observatories. These calibrated facilities will then generate reference maps to serve as secondary standards for cross-calibrating other experiments (e.g., SPT, LiteBIRD) that cannot observe the source directly.
Simulation and Analysis Framework:
- Foreground Modeling: The authors utilize the PySM package to simulate microwave sky data. They compare baseline models (d1, d10) against modified models (d1EB, d10EB) that incorporate intrinsic dust EB correlations. The dust emission is modeled using a "moment expansion" formalism to account for Spectral Energy Distribution (SED) distortions caused by line-of-sight integration and spatial variations in dust temperature ( $T$ ) and spectral index ( $\beta$ ).
- Component Separation: A blind component separation method, specifically a "realistic Multi-Clustering NILC" (rMC-NILC), is employed. This approach uses the Internal Linear Combination (ILC) technique with clustering algorithms to group sky patches with similar spectral properties, minimizing reliance on parametric foreground assumptions.
- Calibration Recovery: The study simulates a LiteBIRD-like satellite configuration (15 frequency channels) with an injected polarization angle mis-calibration of $\Delta\alpha = 0.1^\circ$ . The pipeline attempts to recover this angle by minimizing the residual EB cross-power spectrum using a $\chi^2$ approach, testing whether complex foregrounds degrade this recovery.

Key Contributions

Project Status: The paper outlines the current development status of COSMOCal, including the refined scientific and technical requirements, the instrument layout, and the timeline leading to a Phase A review in early 2027.
Foreground Complexity Analysis: It extends previous calibration studies by explicitly incorporating realistic, complex dust models (including intrinsic EB correlations and SED distortions) into the calibration pipeline validation.
Calibration Strategy Shift: The paper proposes a shift from standard EB-nulling self-calibration to an "EB matching" procedure. In this framework, the SO-LAT (calibrated by COSMOCal) provides a constraint on the cosmic birefringence angle, allowing residual EB signals to be attributed to instrumental systematics or foreground leakage rather than cosmological rotation.
Formalism for Dust EB: The work integrates a formalism describing non-zero dust EB correlations (based on filamentary models) into the moment expansion framework, allowing for the extrapolation of dust templates across frequencies.

Results

Recovery of Mis-calibration: The analysis demonstrates that the injected mis-calibration angle of $\Delta\alpha = 0.1^\circ$ $Δ α = 0. 1^{\circ}$ can be successfully recovered with high accuracy even in the presence of complex foregrounds.
- For the d1EBs1 model, the ensemble average recovered angle was $0.098^\circ$ with a $1\sigma$ statistical uncertainty of $7.7\%$ .
- For the d10EBs1 model, the recovered angle was $0.099^\circ$ with a $1\sigma$ uncertainty of $8.7\%$ .
Impact of Multipole Range: The study found that restricting the minimization to low multipoles ( $\ell < 40$ ) introduces a systematic bias in the recovered angle. However, utilizing the full multipole range ( $\ell_{max} = 1024$ ) stabilizes the likelihood, confirming that intermediate and high multipoles are crucial for robust calibration recovery.
Residual Contamination: While blind component separation absorbs frequency-dependent systematics, a global, frequency-independent rotation leaves a residual EB signal. The study confirms that this residual can be distinguished and quantified when the calibration reference is absolute rather than self-derived.

Significance
The paper argues that characterizing the complexity of Galactic dust is essential not only for cosmological signal extraction but also for the calibration strategies of future CMB experiments. The COSMOCal project offers a critical solution by decoupling instrumental calibration from the astrophysical sky, thereby breaking the degeneracy between instrumental leakage, cosmic birefringence, and foregrounds. By providing a stable, space-based reference, COSMOCal enables the transition from self-calibration (which risks absorbing cosmological signals) to absolute calibration, allowing for a robust detection of both primordial B-modes and potential parity-violating effects. The results suggest that with this framework, future experiments can achieve the necessary sub-degree calibration accuracy to meet their scientific goals.