Through the Haze: How Screening and Cosmic Birefringence Distort the Cosmic Microwave Background ✛
Context
The cosmic microwave background (CMB) is light born 380,000 years after the Big Bang. On its way to us, photons scatter off free electrons (Thomson scattering), which 'blurs' the anisotropy, an effect known as anisotropic screening. Another effect, cosmic birefringence, rotates the polarization plane due to hypothetical parity-violating fields. Both phenomena are secondary CMB anisotropies that have not yet been detected. Related works on arXiv explore similar topics: anisotropic birefringence (2409.01098, 2504.13154), constraints on cosmological models (2503.14454, 2503.14452), and early searches for birefringence (2001.10465). This work follows the same line but is the first to apply small-scale ACT DR6 data to screening.
Methods
Data from the sixth release of the Atacama Cosmology Telescope (ACT DR6), covering ~23% of the sky with a resolution of 1.4 arcminutes, were used. Quadratic estimators (similar to CMB lensing) implemented in the tempura and falafel pipelines were applied to reconstruct the screening field τ and the rotation angle α. For screening, the TT (temperature-temperature) estimator was used; for birefringence, the EB (E-mode and B-mode polarization) estimator. To avoid noise biases, a cross-split estimator was employed: the data were divided into four independent parts, and only cross-spectra between different parts were estimated. Biases were subtracted: the mean field, Gaussian noise N0, lensing leakage Nleak, secondary lensing N1, and residual Monte Carlo bias. For validation, 100 noiseless simulations with injected signal were used.
Results
For anisotropic screening, the power spectrum was obtained in the range 100 < L < 3000, consistent with zero (PTE = 0.68). The amplitude Aτ = 180 ± 157, giving an upper limit of Aτ < 450 at the 95% confidence level. This is roughly two orders of magnitude above predictions from reionization models. Significant contamination from extragalactic backgrounds was found: in the 90 GHz channel, an excess of power is visible, and WEBSKY simulations show a systematic bias up to 1.4σ. When the maximum multipole is reduced to 2000, the limit tightens to Aτ < 370. For anisotropic birefringence, ACB = 0.05 ± 0.06, with an upper limit of ACB < 0.15 (95%). This corresponds to a constraint on the Chern–Simons coupling constant gaγ < 4.9×10⁻²/HI. A conservative analysis with Lmin = 40 gives ACB < 0.52. The result is consistent with previous measurements but less restrictive than ACT DR4 due to larger polarization noise and limitations on large scales.
Implications
This work is the first to use small-scale ground-based telescope data to constrain anisotropic screening. Unlike neighboring studies that focus on large-scale Planck or BICEP/Keck data, this analysis uses a cross-split estimator robust to noise modeling errors. The main conclusion is that systematic errors from extragalactic backgrounds can be comparable to statistical errors, requiring polarization estimators or multi-frequency component separation. For birefringence, the work confirms the absence of a signal but highlights the importance of accounting for the global polarization angle as a systematic.
Future development
Future experiments such as the Simons Observatory and the South Pole Observatory will improve polarization sensitivity and allow the use of polarization estimators for screening, reducing the impact of backgrounds. Combining ACT data with Planck will expand the range of scales and reduce contamination. For birefringence, the key will be combining large-scale Planck data with high-sensitivity ground-based measurements. The next generation of experiments is expected to achieve the first detections of screening under realistic reionization models.
Impact
The results will impact cosmology (constraints on reionization and baryonic physics) and physics beyond the Standard Model (axion-like particles).
Next steps
Conduct an analysis with polarization estimators on Simons Observatory data. Combine ACT and Planck data to improve limits.
Key open problems
For cosmology, a key unsolved problem is the precise history of reionization (its midpoint, duration, inhomogeneity). Anisotropic screening is directly sensitive to these parameters, but current limits are two orders of magnitude above theoretical predictions. Another problem is the nature of dark energy and possible axion-like fields; birefringence could reveal them, but so far it has not been detected.
The knowledge machine's view
This section is addressed to the author of the paper and their colleagues. Neighbouring works were found by meaning-based search across our archive, not by word overlap. We do not review: we show the field around the work.
What the work does
This work analyzes ACT DR6 data using quadratic estimators to search for anisotropic screening and cosmic birefringence. Upper limits of Aτ < 450 and ACB < 0.15 are obtained, and contamination from extragalactic backgrounds is detected. This is the first time small-scale ground-based telescope data have been used to constrain screening.
Where its strength is
The strength lies in the use of cross-split estimators and careful bias subtraction, which allows control of systematics on small scales. This provides a solid foundation for further improving the constraints.
Where it could go next
- It might be interesting to compare the obtained constraints with approaches using polarization to search for isotropic birefringence, and to consider a combined analysis with other datasets.based on: Polarization of Light in Telescopes: A Hidden Twist arXiv:2601.03925
- There is a related work examining the connection between axion-like particles and cosmological observables; perhaps your limits on the Chern–Simons coupling constant can be translated into constraints on the parameters of such models.based on: Axions: How Small Can Their Share of Dark Matter Be? arXiv:2606.06410
- The archive has recent CMB measurements from the South Pole; comparing your results with their data could help assess consistency and potentially improve sensitivity through combination.based on: Cosmic Microwave Background: New Measurements from the South Pole arXiv:2601.20551
Worth trying
- Try applying your pipeline to data from another high-resolution experiment to test the robustness of the Aτ limit to instrumental systematics.
- Investigate how accounting for extragalactic backgrounds via models like WEBSKY affects the ACB constraint when using more aggressive masks.
- Consider adding polarization data on intermediate scales to improve sensitivity to birefringence.
Why this may matter
Constraints on screening and birefringence are important for testing physics beyond the Standard Model, including axion-like particles and modified gravity, and could guide future experiments.
Nearby works in our archive
- Dark Radiation and the Run of Inflation: What Hubble Is Hidingsimilarity 0.55
- Cosmic Wake Exposes Hidden Motionssimilarity 0.53
- Fingerprints in the Spectrum: Pulsars vs. Dark Mattersimilarity 0.53
- Polarization of Light in Telescopes: A Hidden Twistsimilarity 0.51
- Symphony of Shadows: Neutrino Telescopes Catch the Whisper of Dark Mattersimilarity 0.51
- Dark Forces Twisted the Light from the Pastsimilarity 0.51
- Axions: How Small Can Their Share of Dark Matter Be?similarity 0.50
- Cosmic Microwave Background: New Measurements from the South Polesimilarity 0.49
The archive is sparse around this work: the closest review is only 0.55, against a usual 0.64. It looks like few have drilled nearby.
The ACT telescope is located in the Atacama Desert at an altitude of 5190 meters, making it one of the highest observatories in the world. The indigenous Lican Antay communities have observed the sky from these same locations for centuries.
Key numbers
- upper limit Aτ: 450
- upper limit ACB: 0.15
- limit on coupling constant gaγ: 4.9×10⁻²/HI
- sky coverage: 23%
- telescope resolution: 1.4 arcminutes