For a century and a half, physicists have known exactly what should happen when two fluids slide past each other at different speeds. The boundary buckles. Then it curls. Then it rolls into vortexes. That’s the Kelvin-Helmholtz instability, worked out in the late 1860s, and it’s why wind puts ripples on water and shears clouds into rows of curves.
Nobody had ever confirmed it happening on the Sun.
Now a team led by David Kuridze and Friedrich Wöger of the National Solar Observatory reports that Kelvin-Helmholtz instabilities aren’t just visible on the solar surface. They’re everywhere. And their study argues this may change how we think about heat, mass and magnetic energy moving through the Sun’s atmosphere.
They were hiding because they’re tiny
The reason plasma whirlpools stayed invisible for so long is unglamorous: they’re small. Smaller than what any telescope with a mirror under 2 meters can resolve, which for most of the history of solar physics meant every telescope on Earth.
That changed with the Daniel K. Inouye Solar Telescope, the 4-meter instrument the US National Science Foundation opened in Hawaii. It’s the largest solar telescope in the world and entered its operational phase in November 2021.
The vortexes were a byproduct of a test
Here’s the part that undercuts any tidy discovery narrative: Kuridze’s team wasn’t hunting for vortexes. During a three-minute window on April 14, 2025, they pointed the telescope at an active region near the center of the solar disk and recorded at a wavelength of 416 nanometers, using a diagnostic camera setup built jointly by the National Solar Observatory and the Max Planck Institute for Solar System Research.
“The main goal was to achieve diffraction-limited performance with the telescope,” Kuridze said.
Every telescope has a diffraction limit, the finest detail its optics can physically deliver. Two things set it: mirror size and observed wavelength. Bigger mirror, shorter wavelength, finer detail. The Inouye mirror was fixed at 4 meters, so the team pushed the other variable.
“416 nanometers is towards the smaller portion of the visible spectrum,” Kuridze said. “We selected it because we wanted to achieve a higher diffraction limit and higher resolution.”
740 frames a second, 19 kilometers of detail
The camera read out 740 frames per second at exposures of 100 microseconds. Two thousand of those frames went into each final image through multi-frame blind deconvolution, a technique that numerically models and strips out the blurring Earth’s atmosphere leaves behind after the telescope’s optics have done their work.
What came out was a movie of the solar surface with a new frame every two seconds and a spatial resolution of about 19 kilometers. That’s the theoretical limit of a 4-meter mirror at that wavelength. They hit the wall exactly.
“As a byproduct we got these amazing observations, which allowed us to see something which has never been seen before,” Kuridze said.
What a blurry smudge turned into
At 416 nanometers the solar surface is dominated by granules, the convection cells hauling heat up from the interior, interlaced with concentrated bundles of intense magnetic fields. In lower-resolution images, the boundary between a magnetic bundle and the surrounding granulation reads as smooth and slightly blurry.
In the DKIST data it looks nothing like that. The interfaces are made almost entirely of vortex-like structures and fine dark striations.
The team identified 47 vortex-bearing interfaces in the field of view and measured the spacing between adjacent curls: usually 60 to 100 kilometers apart, with individual vortexes running 25 to 170 kilometers in diameter. The smallest ones sat right at the 19-kilometer resolution limit, which means nobody knows how far down the sizes actually go.
They grow fast, too. Tracking them, the team found the vortexes can double in size in under a minute, and they propagate along the interfaces at 0.67 to 3 kilometers per second.
Two phases, and one of them is a mess
Ask what you’d see hovering above one of these boundaries and Kuridze’s answer depends on when you show up. The instability runs in two phases.
“The first is the linear phase, when things are more relaxed, very well organized, regular and beautiful,” he said. “If you see them from close range you will see something like cloud-type things, which are rolling.”
Then the calm ends. “At some point everything turns into a non-linear regime and then things get messy,” Kuridze said. “You are basically getting turbulence, very chaotic turbulence.”
Why the curls form there and nowhere else
Magnetic field lines behave like elastic threads running through the plasma, and they resist being bent. When they lie along the direction of flow, they yank a rippling boundary back flat and smother the instability before it can grow. When they run across the flow, they do nothing.
In the strong magnetic regions DKIST looked at, the field points almost straight up out of the surface while granular flows slide past it sideways. The threads are strung the wrong way to hold anything together. So the curls grow unchecked.
The simulations that made the case
Seeing something nobody has seen before is also a good way to fool yourself with an image-processing artifact, and the team knew it.
“Everything looked like it should be Kelvin-Helmholtz, but of course this is not enough,” Kuridze said. “You need theoretical proof to make sure that it is really Kelvin-Helmholtz.”
So they simulated 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 region they’d actually observed. Then they synthesized the images DKIST should have registered while staring at that simulated patch, computing 500 spectral points across the observed wavelength band, applying the transmission profile of the real interference filter and blurring the result to match the real telescope’s resolution.
The synthetic images reproduced the vortexes’ appearance and dynamics, growth rate, distribution and propagation speeds included. That’s what pushed the team to conclude the observations are most likely real.
A stirring mechanism nobody budgeted for
A strong magnetic field is supposed to hold plasma still. A sunspot is the clearest case: a patch where the field chokes off convection entirely, leaving the surface dark and cool.
Finding an instability that spins up vortexes along the edge of every magnetic element means finding a stirring mechanism where none was expected. Magnetized and unmagnetized gas can blend into each other. Cool material from the edges of convection cells can leak into the magnetic regions and change how heat moves just under the visible surface. Existing models of solar convection don’t account for any of that.
Then there’s the corona, the Sun’s million-degree outer atmosphere. It’s heated partly by field lines being shuffled at their anchor points until they braid into tangles that snap and dump energy. The mechanism doing the shuffling has never been observed.
“You have these twisting motions everywhere at the surface of the magnetic element, and this twisting motion is nothing else than braiding of the magnetic fields,” Kuridze said.
Three minutes is not a survey
All of this rests on a three-minute observation window and simulations that were themselves limited. Worth keeping in mind before anyone rewrites a textbook.
“One thing we just don’t know is how small these Kelvin-Helmholtz patterns get on the Sun,” Kuridze said. What DKIST can resolve sits right against the resolution limit, so smaller structures may be slipping through. The obvious workaround is running simulations finer than the telescope can see. The team tried. It hasn’t gone well.
“When you do this at higher resolution in the simulations, some extra physics needs to be involved, and we are not sure exactly how things work in computational simulations when you need to reach those resolutions,” he said. “This is a completely new area, and we need to do more investigation.”
There’s a second gap. DKIST mostly delivered high-resolution visual images, which serve as a decent proxy for where the magnetic fields are and how strong they are. A proxy, though. Everything the team knows about how fast the plasma actually moves comes from simulations, not the telescope, and the magnetic field driving the whole thing has never been measured directly at this scale.
Kuridze’s fix is more telescope time, not more cleverness. “If you want to quantify how the magnetic field is evolving in time, what sort of dissipation you have, how much energy is released, how much energy budget there is for eruptions and flares, you need much longer observations, and you need magnetic maps,” he said. “This is the next challenge and the next milestone.”
Nature, 2026. DOI: 10.1038/s41586-026-10871-3
