Physicists at the University of California San Diego have developed a way to check whether a faint twist in the oldest light in the universe is real or an artifact of the instruments measuring it. The method compares polarization maps made by different groups of detectors, looking for calibration differences that could imitate a cosmic signal. Tested on data from the European Space Agency's Planck satellite, it produced results consistent with an established analysis.

The question at stake is whether the polarization of the cosmic microwave background has rotated slightly during its nearly 14-billion-year journey. Such a rotation, known as cosmic birefringence, is not predicted by the Standard Model of particle physics. If confirmed, it could provide a new window onto dark matter, dark energy or other physics that current theories do not explain.

The obstacle is that the expected effect is extraordinarily small, and a telescope with a tiny error in the alignment of its polarization detectors can produce an almost identical signal. A common misalignment affecting all detectors in an instrument cannot be distinguished from a genuine rotation of the incoming light by looking at the combined measurement alone.

The new approach sidesteps that problem by comparing maps from separate detector groups against one another. A real cosmic rotation should appear consistently in all of them, so differences between maps point to relative calibration errors rather than to the universe. The method does not measure the overall rotation itself, but it offers an independent check on a critical part of the analysis.

Writing in The Astrophysical Journal Letters, Anto I. Lonappan, Brian Keating and Kam Arnold report that their reconstruction of relative calibration differences agrees with results from the Minami-Komatsu method, a widely used framework for separating instrumental rotation from possible cosmic birefringence. Combining the two approaches yields an estimated rotation angle of 0.37 ± 0.12 degrees, in line with earlier work.

The agreement is a validation of the technique, not a discovery. The authors emphasise that their method removes any rotation shared by all maps, so it cannot independently establish that the ancient light has actually turned. What it can do is give future experiments an additional way to test whether their instruments are trustworthy before interpreting a tiny signal as new physics.

The signal in the oldest light

The cosmic microwave background is radiation released when the universe was about 380,000 years old. Before that point, the cosmos was too hot and dense for light to travel freely; electrons scattered photons repeatedly. As the universe cooled, electrons and protons combined into neutral atoms, the fog cleared, and radiation began to move across space largely unimpeded.

That light now reaches us as a faint glow with a temperature of a few kelvin above absolute zero. Because it has travelled for nearly 13.8 billion years, it carries a snapshot of conditions in the early universe. Small temperature variations in the map correspond to density fluctuations that later grew into galaxies.

The radiation also carries a small amount of linear polarization. Polarization describes the direction in which the electric field of a light wave oscillates. In the early universe, photons scattering off free electrons acquired a preferred orientation, much as sunlight scattering through Earth's atmosphere becomes polarized.

Cosmic birefringence refers to a rotation of that polarization direction as the light travels. Some observations suggest the angle may have shifted by a fraction of a degree over the light's journey. A rotation of that size would be unremarkable in everyday optics, but in cosmology it could be consequential, because the Standard Model does not predict a mechanism that would rotate the polarization of the entire cosmic background in this way.

The potential payoff is large. Dark matter and dark energy together account for roughly 95 per cent of the universe's total energy density, yet their underlying nature remains unresolved. A confirmed birefringence signal would give physicists a new way to probe sectors of theory that have so far escaped direct detection.

Why a telescope can imitate the effect

The central difficulty is instrumental. A polarization-sensitive detector has an intrinsic angle, a direction against which it measures the orientation of incoming light. If that angle is miscalibrated by even a small amount, the measured polarization will appear rotated. The error has exactly the same mathematical form as cosmic birefringence.

This problem is especially stubborn when the misalignment is common to all detectors in an instrument. Suppose every detector in a telescope's focal plane is offset by the same angle. The combined data will show a uniform rotation, and there will be no internal reference that reveals whether the rotation came from the sky or from the detector assembly. Adding more data from the same instrument will not resolve the ambiguity.

Because the suspected cosmic signal is only a fraction of a degree, the required calibration precision is extreme. Small uncertainties in the orientation of a detector, or in the modelling of how the instrument responds to polarized light, can dominate the measurement.

Existing approaches try to break the degeneracy using assumptions about the statistics of the cosmic microwave background. The Minami-Komatsu analysis, for example, uses properties of the polarization patterns themselves to infer a common calibration angle. It has produced hints of a non-zero rotation, but the interpretation depends on the validity of its assumptions.

The UC San Diego team did not set out to replace that framework. Instead, they sought a complementary test that would examine a different aspect of the problem: whether separate groups of detectors within the same instrument agree with one another about polarization angles.

The new cross-check explained

The method is called differential polarization calibration. Rather than analysing one combined map of the sky, the researchers split the data into separate maps produced by different subsets of detectors. Each map is treated as an independent measurement of the same sky with its own possible calibration offset.

The logic is simple. If the universe has rotated the polarization of the cosmic microwave background, that rotation affects all the maps in the same way. When the maps are compared with one another, a common rotation cancels out. What remains are the differences between detector groups: the relative miscalibrations that the team wants to quantify.

Those relative differences are useful on their own. If one detector group appears rotated relative to another, that is a sign of an instrumental effect that needs to be understood. A well-behaved instrument should show consistent alignment across its subsets once known effects are accounted for.

The technique therefore functions as a consistency test. It cannot say whether the whole instrument shares an error, because a shared error is invisible to a comparison of its parts. But it can reveal disagreements between parts, which is exactly the kind of problem that could contaminate searches for a faint cosmological signal.

The authors describe this as a complementary way to examine the distinction between the universe and the instrument. It does not replace absolute calibration, which requires an external reference, but it adds an internal check that can be applied to existing and future data sets.

Eight Planck polarization maps

To test the method, the researchers used observations from Planck, the European Space Agency satellite that mapped the cosmic microwave background in unprecedented detail between 2009 and 2013. Planck's instruments recorded both temperature and polarization information across the sky, making its data a natural test bed for calibration studies.

The team analysed eight polarization maps constructed from Planck observations. Each map corresponded to a different grouping of detectors, allowing the researchers to compare their relative calibration angles. Planck data archives, including those served by NASA's Legacy Archive for Microwave Background Data Analysis, remain widely used for this kind of work.

The choice of Planck was practical rather than promotional. The satellite's data are public, well characterised and extensively studied, which makes them useful for testing whether a new analysis technique reproduces known results. The researchers were not claiming a new detection from Planck; they were checking whether their cross-check behaved as expected on a familiar data set.

They then compared the reconstruction of relative calibration differences with the result from the Minami-Komatsu analysis. The two methods rest on different assumptions: one uses internal comparisons between detector groups, while the other infers a common angle from the statistical properties of the polarization field. Despite that difference, they produced compatible results.

That agreement is the main technical finding of the paper. It suggests that the relative calibration differences reconstructed by the new method are consistent with the picture implied by the established analysis, strengthening confidence in how those differences are being assessed.

Agreement with the Minami-Komatsu method

The Minami-Komatsu approach, published in 2020, was a landmark in the study of cosmic birefringence. It used the fact that certain correlations in the cosmic microwave background should vanish if there were no rotation and no instrumental miscalibration. By measuring the departure from that expectation, the authors inferred a common polarization angle and found a hint of a non-zero value.

The UC San Diego team's method addresses a different quantity. Where Minami and Komatsu infer a common angle, the differential method measures how detector groups differ from one another. Because the two approaches probe different aspects of the same data, agreement between them is a meaningful cross-check rather than a circular result.

In the paper, the researchers report that their reconstruction of relative calibration differences, combined with the common calibration value from the established analysis, yields a cosmic birefringence angle of 0.37 ± 0.12 degrees. That matches the published Minami-Komatsu result.

The authors are explicit about what this match does and does not mean. It shows that the new framework can reproduce an existing result and that it can be used alongside that framework to examine calibration questions. It does not constitute an independent confirmation that the cosmic microwave background has rotated, because the new method cannot measure a rotation common to all maps.

That limitation is built into the design. A signal that appears equally in every detector group is indistinguishable from a common miscalibration when the maps are compared with each other. Determining whether the common component is cosmic or instrumental requires an absolute calibration reference, which is a separate problem.

What the method deliberately cannot measure

The inability to measure absolute rotation is not a flaw that the researchers tried to hide. It follows directly from the decision to compare maps with one another. Any rotation shared by all maps cancels in the comparison, leaving only the differences.

That design choice has consequences for interpretation. The new method can tell experimenters whether their detector groups agree. It cannot tell them whether the sky itself has turned. For that, they still need a way to anchor the absolute polarization angle to something external, or to rely on methods that make assumptions about the statistics of the cosmic background.

The team's estimate of 0.37 ± 0.12 degrees therefore depends on combining their reconstruction with the common angle inferred from the Minami-Komatsu analysis. The relative part comes from the new method; the absolute part comes from the established one. The resulting number is useful for comparison, but it is not a standalone measurement.

This distinction matters because the search for new physics demands a high standard of evidence. A rotation of a fraction of a degree could be a clue to physics beyond the Standard Model, but it could also be the residual of an instrumental effect that no one has fully characterised. Researchers need multiple independent checks before drawing that conclusion.

The authors position their technique as one of those checks. It narrows the space of possible instrumental explanations by testing whether detector groups are consistent with one another, and it gives future experiments a concrete diagnostic to apply to their own data.

The 0.37-degree estimate in context

The number that emerges from the combined analysis is small by everyday standards. An angle of 0.37 degrees is a tiny fraction of the arc across the sky, and the quoted uncertainty of 0.12 degrees is larger than a quarter of the central value. In cosmology, however, a rotation of this size applied to light that has travelled for nearly 14 billion years would be difficult to explain within the Standard Model.

The estimate matches the value reported by the Minami-Komatsu analysis. That agreement is reassuring about the internal consistency of the two methods, but the authors are explicit that it should not be treated as a second, independent detection. The absolute angle comes from the established framework; the new method supplies the reconstruction of how detector groups differ from one another.

Interpreting the number also requires care about what the uncertainty represents. It reflects the statistical and systematic information used in the combined analysis, not a complete accounting of every possible source of error in the Planck polarization data. Future instruments with different detector designs and calibration systems will provide their own tests.

If the rotation is real, its origin could lie in physics that affects light as it travels through the cosmos. If it is instrumental, the most likely explanation is a calibration effect that has so far resisted attempts to isolate it. The current work narrows the latter possibility by showing that relative calibration differences are consistent with what the established analysis assumes.

For now, the 0.37-degree figure is best understood as a benchmark for future work. Experiments now under construction will be able to measure polarization with greater precision, and they will be able to apply the same consistency tests to their own data. Whether the number holds up will depend on those measurements.

E modes, B modes and primordial gravitational waves

The calibration problem extends beyond cosmic birefringence. Cosmologists classify polarization patterns in the cosmic microwave background into two types, known as E modes and B modes. E modes, first detected in 2002, form symmetric patterns and arise from density variations in the early universe. B modes have a curl-like appearance.

One known source of B modes is gravitational lensing. As light travels from the early universe, the gravity of intervening matter bends its path and distorts the polarization pattern, generating a B-mode signal that has been observed. A second possible source has not yet been detected: primordial gravitational waves produced during a period of extremely rapid expansion known as inflation.

Finding those primordial B modes would be a major result. It would provide direct evidence for gravitational waves from the early universe and strong support for particular models of inflation. That is why several experiments are designed around the search, including the Simons Observatory, where Brian Keating is principal investigator.

The difficulty is that E modes are much stronger than the expected primordial B-mode signal. A small error in polarization calibration can cause some of the E-mode signal to appear incorrectly as B modes, creating a false signal that mimics the one researchers are looking for. The cleaner the calibration, the less likely that contamination becomes.

The differential method developed at UC San Diego is relevant here because it identifies differences in polarization angle calibration among detector groups. As instruments become more sensitive, such diagnostics will become increasingly important for deciding whether a faint signal is a genuine feature of the universe or an imperfection in the telescope.

Why calibration matters for future observatories

Cosmology is entering an era of larger and more sensitive polarization experiments. Instruments such as the Simons Observatory aim to measure the cosmic microwave background with precision far beyond earlier surveys. That sensitivity brings rewards, but it also raises the cost of small systematic errors.

A calibration error that was negligible in an older experiment can dominate a measurement in a newer one. As the statistical uncertainty shrinks, systematic uncertainties become the limiting factor. Understanding the instrument becomes as important as collecting more data.

The UC San Diego method is designed for this environment. It can be applied to any data set in which the sky has been observed by multiple detector groups that can be compared. It does not require a new telescope or a new observing strategy, only a willingness to split the data and test for internal consistency.

Keating has argued that the detection of primordial B modes would transform understanding of the early universe, and that such a measurement must survive rigorous calibration checks. The new framework addresses part of that requirement by giving experiments a way to test relative polarization angles before interpreting a result.

The approach is not a complete solution. Absolute calibration remains a separate challenge, and the method says nothing about other systematics such as beam shape, detector noise or foreground emission from our own galaxy. But it addresses a specific and troublesome degeneracy that has complicated the search for cosmic birefringence.

Possible physics beyond the Standard Model

The reason astronomers care about a fraction of a degree is that the Standard Model, despite its success, leaves major questions unanswered. It does not explain the identity of dark matter, nor the origin and observed magnitude of dark energy. Together those components make up about 95 per cent of the cosmic energy budget.

One theoretical proposal links cosmic birefringence to a field associated with axion-like particles. Such a field could interact with light and gradually rotate its polarization as it crosses the universe. Axions were originally proposed to solve a problem in the strong nuclear force, and they are also studied as dark-matter candidates, but the connection is not established. The researchers note that even if axion-like particles exist, they would not necessarily account for dark matter.

Other models connect a rotation of polarization to fields associated with dark energy, the phenomenon driving the accelerating expansion of the universe. In those scenarios, the birefringence would be a manifestation of physics that changes how light propagates over cosmological distances.

None of these ideas has been confirmed. The observational hint of a rotation remains just that: a hint, with a significance that researchers continue to debate. Turning it into evidence for new physics would require both a more precise measurement and a much stronger case that the signal is not instrumental.

That is why calibration work of the kind reported by the UC San Diego team matters. Before physicists can argue about axion-like fields or dark-energy couplings, they need to know that the angle they are measuring belongs to the sky rather than the telescope.

Limits of the current result

The paper reports a technical achievement rather than a cosmological discovery. The new cross-check works on Planck data, agrees with an established method and gives experimenters another diagnostic to use. It does not establish that the universe has rotated the polarization of the cosmic microwave background.

The estimate of 0.37 ± 0.12 degrees should be read in that light. It is a combination of the new relative-calibration reconstruction with the common angle from the Minami-Komatsu analysis, and it matches that analysis by construction in its absolute component. It is not an independent detection.

Several open questions remain. Experiments will need to apply the method to their own data, determine whether detector groups agree, and continue working on absolute calibration. The search for primordial B modes will require similar care, because the same class of error can contaminate that measurement too.

For now, the result is a reminder of how demanding cosmology has become. The signals that could reveal the nature of dark matter or the physics of inflation are so faint that researchers must interrogate their instruments as carefully as they interrogate the sky. The UC San Diego team has added one more way to make that interrogation rigorous.


Corrections and updates

Nexuswild welcomes factual corrections. Email contact@nexuswild.com with evidence and the article URL.