Cosmic microwave background birefringence is a tiny rotation of the sky’s polarized light. Measuring it is hard because the same rotated pattern can also come from a common misalignment of the detectors’ polarization angles. The Minami–Komatsu (MK) method separates a true cosmic rotation from an instrumental offset by using polarized emission from our Galaxy. The new paper introduces a different test that does not rely on that Galactic model and so provides a complementary check on the MK-based result.

The authors build a “birefringence-blind” differential estimator. It uses combinations of antisymmetric and symmetric cross-spectra between maps. For any pair of maps, a common rotation (what would include a cosmic birefringence signal) cancels exactly, and the estimator returns the angle difference Δαij = αi − αj between the two detector sets. By combining many map pairs and fixing a gauge (a reference choice for the overall angle), they reconstruct the pattern of detector-dependent calibration offsets across the instrument.

They applied this procedure to eight Planck NPIPE detector-set maps. The differential reconstruction of detector angles agrees with the MK reconstruction that used Galactic foreground polarization. Quantitatively, the two patterns give χ2 = 8.18 for seven degrees of freedom, corresponding to a probability-to-exceed (PTE) of 0.32. All shared-data residuals are below 1.8σ. These numbers indicate good internal consistency between the two analysis routes on the same maps.

Why this matters: EB-based searches for cosmic birefringence (EB means the cross-correlation between two polarization patterns called E and B) are sensitive to tiny angle shifts. Having an independent estimator that is blind to a common rotation gives confidence that the MK separation is not driven by a particular foreground model or by a single analysis choice. The new differential test can be added as a validation layer for future analyses that aim to detect or constrain cosmic birefringence.