For the observer hovering above one of these boundaries, Kuridze says, the view would depend on the time of arrival. The instability runs through two phases. “The first is the linear phase, when things are more relaxed, very well organized, regular and beautiful,” he says. “If you see them from close range you will see something like cloud-type things, which are rolling.”

But when the peaceful phase is over, it’s chaos. “At some point everything turns into a non-linear regime and then things get messy,” Kuridze explains. “You are basically getting turbulence—very chaotic turbulence.

Why do the curls form at the interfaces and not elsewhere? That’s due to the direction the magnetic field is pointing. Magnetic field lines act a bit like elastic threads running through the plasma, and they resist being bent. When those threads lie along the direction of the flow, they pull a rippling boundary back, which flattens and suppresses the instability before it can grow. When they run across the flow, they do nothing to stop it.

In the strong magnetic regions DKIST observed, the field points almost straight up out of the surface, while the granular flows sliding past it move sideways. The threads are strung the wrong way to hold the boundary together, so the curls grow with nothing to limit them.

Double-checking

Because Kuridze, Wöger, and their colleagues saw something nobody has seen before, they made sure the observation wasn’t just an artifact of image processing. “Everything looked like it should be Kelvin-Helmholtz, but of course this is not enough,” Kuridze says. “You need theoretical proof to make sure that it is really Kelvin-Helmholtz.”

To get that theoretical proof, researchers ran computer simulations modeling a patch of photosphere roughly 6 megameters on a side at a grid spacing of 3.2 kilometers, seeded with a magnetic field map of the actual observed region. Then, they synthesized images that the DKIST should have registered while looking at this simulated patch.