⚛️ phenomenology

Measuring Ultralight-Axion Coherence with Galaxy Polarization Correlations

This paper proposes using the three-dimensional correlation of galaxy polarization-rotation angles to measure the amplitude and spatial coherence scale of ultralight axion-like particles, offering a geometric late-time probe capable of detecting sub-degree correlated rotations and constraining axion mass and coupling parameters.

Original authors: Yasuo Doi

Published 2026-07-15

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Yasuo Doi

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 is filled with an invisible, ultra-light fog called "axion-like particles" (ALPs). These aren't just any fog; they are a special kind of dark matter that interacts with light in a very sneaky way. As light from distant galaxies travels through this fog, the fog acts like a giant, cosmic rotator, slowly twisting the polarization (the direction of the light's vibration) of that light.

For a long time, scientists have tried to measure this twist by looking at the light from the very beginning of the universe (the Cosmic Microwave Background) or distant quasars. They've been asking, "Is there a twist? How big is it?" But this is like trying to figure out the shape of a cloud just by looking at a single shadow. You might know the shadow exists, but you don't know if the cloud is a fluffy ball, a long snake, or a flat sheet.

The Big Idea: A 3D Game of "Hot and Cold"
In this paper, Yasuo Doi suggests a brand-new way to play this game. Instead of just measuring how much the light twists, we should measure how the twist changes as we look at pairs of galaxies at different distances from each other.

Think of the ALP fog like a giant, invisible ocean wave. If you dip two sticks into the water very close together, the water level at both sticks will be almost identical. If you move the sticks far apart, one might be on a crest while the other is in a trough. The distance where the water levels stop matching up is called the "coherence scale."

Doi proposes that by measuring the polarization of one million galaxies spread across a quarter of the sky, we can map out this "ocean wave." By looking at how the polarization twist correlates between galaxy pairs separated by different distances, we can find the exact point where the correlation "turns over" (stops matching).

What This Tells Us
If we find this "turnover," it's like finding the wavelength of the ocean wave. This measurement would tell us two incredibly important things:

The Momentum: It reveals the product of the particle's mass ( $m_a$ ) and its speed ( $v_a$ ). The paper suggests this method could detect these correlations for particles with masses between $10^{-29}$ and $10^{-27}$ eV (assuming a typical speed of $10^{-3}c$ ).
The Amount of Fog: The strength of the correlation tells us how much of this ALP fog exists compared to all the dark matter in the universe.

The Catch (and the Confidence)
It is important to remember that this is a forecast, not a discovery. The author has run simulations to show what would happen if we built a survey with specific capabilities.

The Goal: The paper simulates a survey with $10^6$ (one million) polarized galaxies.
The Noise: It assumes each galaxy's measurement has a "scatter" (error) of $10^\circ$ .
The Result: In these simulations, the method would be sensitive enough to detect a correlated rotation of less than one degree with 5 $\sigma$ confidence (a very high standard of certainty in physics).
The Precision: If a signal is found, the simulation suggests we could pin down the momentum scale ( $m_a v_a$ ) with a precision of about 0.1 in logarithmic terms (roughly 25% precision).

What This Method Rules Out (and What It Doesn't)
The paper is very clear about what this method is not doing.

It is not trying to measure the total amount of dark matter directly. It is looking for a specific pattern in how the light twists.
It does not assume the ALPs are clumped up inside individual galaxy clusters. The paper argues that because these particles are so light, their "waves" are much larger than any single galaxy, so they don't need to follow the local density of stars.
It explicitly rejects the idea that we can just measure the average twist of the whole sky. The paper argues that a simple average misses the most interesting part: the spatial structure of the twist. By subtracting the average, the method removes global calibration errors (like a ruler that is slightly bent) and focuses only on the local variations.

Why It's Cool
This approach is like a new kind of ruler. Previous methods were like measuring the total weight of a bag of marbles. This new method is like measuring the distance between the marbles to figure out how big the marbles are and how fast they are moving.

The paper suggests that this technique is complementary to other methods. While other experiments look at how dark matter pulls on things (gravity), this one looks at how it twists light. If the ALPs don't interact with light at all, this method sees nothing, but if they do, this method can see the "shape" of the dark matter field in a way gravity alone cannot.

In short, the paper proposes a clever geometric trick: use the 3D arrangement of galaxies to turn the universe itself into a giant detector for the "wavelength" of dark matter. If we can build a telescope capable of measuring the polarization of a million galaxies with the precision described, we might finally see the ripples in the dark matter ocean.

Technical Summary: Measuring Ultralight-Axion Coherence with Galaxy Polarization Correlations

Problem Statement
Ultralight axion-like particles (ALPs) are candidates for dark matter that couple to photons, inducing a rotation of the linear polarization of distant sources via the interaction $L_{a\gamma} = -\frac{1}{4}g_{a\gamma}a F_{\mu\nu}\tilde{F}^{\mu\nu}$ . While cosmic birefringence has been investigated using the Cosmic Microwave Background (CMB) and distant radio galaxies/quasars, existing searches primarily focus on measuring the amplitude of isotropic rotation. Current CMB analyses yield mixed results regarding isotropic birefringence, with some reporting rotations of $\sim 0.3^\circ$ and others finding values consistent with zero. Crucially, these methods do not directly probe the spatial coherence of the ALP field. The paper identifies a gap in utilizing the three-dimensional spatial correlation of polarization rotation angles to measure the ALP field's coherence scale, which is directly tied to the ALP momentum scale ( $m_a v_a$ ).

Methodology
The authors propose a statistical framework utilizing a catalog of polarized galaxies to measure the two-point correlation function of polarization rotation angles.

Theoretical Framework: For a nonrelativistic ALP field with an isotropic Gaussian velocity distribution, the equal-time field correlation function $\xi_a(r)$ follows a Gaussian form: $\xi_a(r)/\xi_a(0) = \exp[-(r/L_{phys}^G)^2]$ . The coherence scale is defined as $L_{phys}^G = \sqrt{6}/(m_a v_a)$ , where $m_a$ is the ALP mass and $v_a$ is the three-dimensional rms velocity dispersion.
Observable: The rotation angle $\alpha$ is estimated for each galaxy by comparing the observed polarization position angle with an independently reconstructed intrinsic angle. The analysis utilizes mean-subtracted rotation residuals ( $\delta \hat{\alpha}_i$ ) to project out spatially uniform additive offsets, such as global calibration errors or the observer-side endpoint of the rotation.
Estimator: A three-dimensional two-point correlation estimator is constructed for galaxy pairs binned by their comoving separation $r_{ij}$ . The model assumes a Gaussian template scaled by an amplitude $A$ (related to the rotation variance) and a coherence scale $L_G$ .
Survey Forecast: The authors present a fiducial forecast for a survey covering 25% of the sky ( $f_{sky}=0.25$ ) with $10^6$ polarized galaxies extending to redshift $z=2$ . They assume a total per-galaxy scatter of $\sigma_{tot} = 10^\circ$ (including measurement noise and intrinsic scatter) and a standard cosmology. The analysis accounts for redshift-space smearing due to peculiar velocities.

Key Results

Sensitivity: The proposed survey achieves a $5\sigma$ sensitivity to correlated rotations of sub-degree amplitude. The detectable range for the momentum scale is $m_a(10^{-3}c/v_a) \sim 10^{-29} \text{--} 10^{-27}$ eV.
Precision: For detected signals across much of this range, the marginalized precision on the logarithm of the momentum scale is $\sigma[\log_{10}(m_a v_a)] \lesssim 0.1$ . This corresponds to approximately 25% precision on the directly measured quantity $m_a v_a$ .
Amplitude Constraints: The amplitude threshold translates to constraints on the ALP abundance fraction $f_{ALP} = \Omega_a/\Omega_{DM}$ . For a benchmark coupling $g_{a\gamma} = 10^{-12} \text{ GeV}^{-1}$ , the method can probe abundance fractions as low as $f_{ALP} \sim 10^{-10}$ , depending on the mass.
Limiting Factors: The low-mass limit is constrained by the maximum survey baselines and the removal of the sample mean (which projects out modes coherent over the entire footprint). The high-mass limit is driven by the decreasing number of galaxy pairs within the shrinking coherence scale.

Significance and Claims
The paper claims that this method provides a "geometric late-time probe" that is qualitatively distinct from and complementary to CMB birefringence and structure-formation constraints.

Direct Coherence Measurement: Unlike amplitude-only measurements, the turnover in the spatial correlation function provides direct information on the ALP momentum scale ( $m_a v_a$ ). A detection would establish the existence of spatially varying birefringence and constrain the ALP coherence scale.
Systematic Robustness: The mean-subtracted estimator effectively removes spatially uniform additive offsets, including global polarization-angle calibration errors, which are a major systematic in other birefringence searches.
Null Tests: The full three-dimensional dependence allows for null tests; a genuine ALP signal should correlate primarily with three-dimensional pair separation and turn over near $L_G$ , whereas instrumental or foreground residuals are typically tied to angular position, observing conditions, or wavelength.
Complementarity: The method probes the late-time spatial phase structure of the birefringent field, whereas structure-formation constraints probe the gravitational impact of ultralight dark matter. The two approaches cut the parameter space in different directions: polarization requires photon coupling, while gravitational probes do not.

The authors note that the forecast relies on simplifications, such as approximating the covariance via pair counts rather than mock catalogs and using an effective equal-time template at a reference redshift rather than a full light-cone treatment. However, they argue the scaling relations are robust: the detection capability depends primarily on the effective weight $N_{gal}/\sigma_{tot}^2$ . Consequently, a survey with $10^6$ galaxies at $10^\circ$ scatter offers similar raw statistical weight to a survey with $10^4$ galaxies at $1^\circ$ scatter, provided the survey geometry samples the relevant three-dimensional pair separations.

Technical Summary: Measuring Ultralight-Axion Coherence with Galaxy Polarization Correlations

More like this