Astronomers Turned 27 Years of Radio Snapshots Into a Movie of a Black Hole Jet. The 'Shock Waves' Everyone Saw Were Moving at the Speed of the Plasma.

For decades, radio astronomers studying the jets fired out by supermassive black holes have watched bright blobs race outward at apparent speeds faster than light, and the standard explanation has been that those blobs are shock waves plowing through the jet plasma. A new study published in Nature takes that picture apart for one of the best-observed jets in the sky. Using an artificial-intelligence reconstruction method, a team converted 116 separate radio observations of the blazar 3C 345, spanning 27 years, into one continuous video, and found that the bright knots move at essentially the same speed as the plasma around them. Shocks, by definition, should not.

The technique is called kine, which its authors describe as a "video reconstruction algorithm for VLBI observations of variable sources." Very long baseline interferometry combines radio telescopes across continents to achieve extreme angular resolution, but each observing session yields a sparse, noisy snapshot, and researchers have traditionally reconstructed each epoch on its own and then compared the results by eye or by fitting simple model components. Kine instead represents the whole time series as a single neural network that learns how the source changes across space and time, so that information from every epoch helps sharpen every other. "Simultaneous imaging of all observations allows information to be shared across frames, improving resolution and dynamic range beyond what is achievable by frame-by-frame imaging," the authors write.

Applied to the MOJAVE monitoring program's archive of 3C 345 from the Very Long Baseline Array, kine produced a movie with about four times the resolution of conventional imaging and roughly 140 times the image contrast. Because the reconstruction is smooth in time, it can be sampled at any moment, which let the team apply optical-flow methods, the same family of tools used to track motion in ordinary video, and measure the local velocity of the plasma everywhere in the jet rather than only following a handful of bright features.

That measurement is what undercuts the shock interpretation. A shock compresses the plasma it passes through and has to travel at a speed different from the underlying flow. "We found no evidence that traveling bright components are strongly shocked regions, as previously proposed, because the component speeds are of the same order as the average flow speed in the same regions," the authors write. The polarization data point the same way: if the knots were shocks, the fraction of polarized light should peak where the brightness does, and it does not. Instead, the team argues, the knots are patches of plasma that are locally brighter because they are magnetically energized and because their motion is aimed nearly toward us, which relativistic beaming amplifies.

The researchers are careful to say the result applies to 3C 345 and not automatically to every jet. But the same archive holds decades of monitoring data on hundreds of active galaxies, and running kine across it could produce full motion maps of each, testing whether the shock model survives anywhere. The authors also note that the mathematical problem they solved, recovering a smoothly changing object from sparse, scattered measurements, is close to the one faced in magnetic resonance imaging of a beating heart or breathing lungs, and they suggest the method could travel from astronomy into medicine.