A Black Hole Grew a Jet. Now Astronomers Can Measure Its Magnetic Field.
In May 2023, astronomers pointed the Very Large Array at a galaxy whose black hole had just begun doing something it had never been seen to do. Every radio look at 1ES 1927+654 before that year had found a faint, compact source of the kind a quiet galaxy makes. Then its radio core started to brighten, and kept brightening, until it stood about 60 times its old level. A jet was coming out. What the observation was after was the lean of the radio waves, the tilt a magnetic field imprints on the light it helps create. The lean was not there. The polarization came in below 0.6%, an upper limit and nothing more.
In August 2024, the same measurement worked. The jet's radio light came back polarized at 2.1%, and the angle of that polarization would later do something stranger still. Onic I. Shuvo and Eileen T. Meyer, both at the University of Maryland, Baltimore County, and their colleagues report both results in a preprint posted Sept. 8 and submitted to The Astrophysical Journal. It has not been through peer review.
Electrons spiraling through a magnetic field give off radio waves, and those waves come out leaning in a direction the field sets. Read the lean and you have the field's orientation on the sky. Watch the lean move and you are watching something rearrange itself. Astronomers have been doing this on young radio jets for decades. The claim here is narrower: the first such detection during the formation of a jet in a changing-look galaxy, one that alters its appearance over months, and the first for this object.
The jet itself is not in doubt. Meyer and colleagues reported the radio flare and a resolved, two-sided structure a fraction of a light-year across in The Astrophysical Journal Letters, and Sibasish Laha and colleagues followed the launch across wavelengths in The Astrophysical Journal, both in 2025. Those two papers share most of their author list with this one, so the new polarization result has a second dataset behind it.
The August 2024 run spanned a wide range of radio frequencies, and the share of light that was polarized rose roughly thirtyfold from the lowest to the highest. That points to a layered outflow. An ordered magnetic field sits in the compact core, wrapped in a turbulent, magnetized sheath that scrambles the longer wavelengths on their way out. At the top of the range, where the sheath matters least, the angles agree closely with one another, the signature of a field that is combed rather than tangled. That field does not line up with the jet's axis, which the authors read as a sign of shocks or shear near the base.
Then, beginning in March 2026, the angle started to travel. Over 81 days it turned smoothly through about 137 degrees, always in the same direction. Then it reversed, giving back about 74 degrees by July 28, a little over half of what it had gained. Two things stayed put while it happened. The jet's total radio brightness held steady to within a few percent, and the polarized share of it sank to the lowest values of the entire campaign.
That combination is the argument. Two explanations are on the table, and the authors prefer the second. A passing change in how opaque the emitting region is could rotate the angle and suppress the polarization at once, but it would be expected to move the total brightness as well, and the brightness did not. The alternative is that something physical crossed the core: a newly ejected clump of plasma, or a shock inside the jet. Its own magnetic field would add to the field already there and swing the sum. The paper's abstract puts it as "consistent with a propagating disturbance crossing the compact emitting region," which is a good deal weaker than "caused by." On that reading, the partial return is the disturbance leaving.
The measurement also puts a number on the magnetic field, and the number needs care. How much the angle shifts with wavelength records how much magnetized gas the light crossed on its way here. That yields an average field along the line of sight of roughly 1 to 100 milligauss, thousandths of a gauss, in the material around the jet. Extrapolate inward to where the jet is launched and the figure becomes 10 to 1,000 gauss, comfortably inside the range that magnetically driven jet models call for. That second one is an extrapolation and not a measurement. It assumes the intervening gas sits about three light-years from the black hole, and that the field weakens in proportion to distance, on top of a gas density the authors say is uncertain by two orders of magnitude.
Circular polarization, the other kind, showed up in none of the observations. Whether the magnetic field settled back to where it started is beyond what this campaign can say, and the follow-up that could answer it has already been granted: a repeat of the multi-frequency run, which will show whether the jet's field was permanently rearranged.