Most planets orbit in roughly the same plane their star spins in, and in the same direction. It's a leftover habit from the flattened disk of gas and dust that built both the star and its planets together. When a planet breaks that habit badly enough to orbit backward relative to its star's spin, astronomers take notice — because something usually had to shove it there. A newly reported world does exactly that, and this time the usual suspects are missing from the scene.
The planet is GJ 3090 b, a sub-Neptune about 2.2 times Earth's radius and 4.5 times its mass, circling a red dwarf roughly 73 light-years away in the constellation Phoenix. It completes an orbit in just 2.9 days. A team led by Yann Carteret, a PhD student at the University of Geneva, used the NIRPS and HARPS spectrographs on the European Southern Observatory's 3.6-meter telescope in Chile to measure the planet's orbital tilt against its star's spin axis. The result, submitted to arXiv on September 21 and covered days later by Universe Today and Knowridge: a 3D obliquity of about 136 degrees (+24/-18). Since anything past 90 degrees counts as retrograde, GJ 3090 b is moving backward — the first confirmed case of a planet orbiting opposite to its star's rotation around an M-dwarf star, and the smallest exoplanet ever to have its orbital obliquity mapped around a red dwarf.
How do you even measure a planet's tilt?
You can't point a telescope at a planet 73 light-years away and watch it tilt. Instead, the team relied on the Rossiter-McLaughlin effect, a trick that turns a planet's transit into a natural experiment on the star's own light.
As GJ 3090 b crosses in front of its star, it blocks starlight from different patches of the stellar disk in sequence. Because the star is rotating, one side of that disk is moving toward Earth and the other away, which normally produces a smooth, symmetric wobble in the star's measured radial velocity as the planet transits. If the planet's orbit is tilted or backward relative to that spin, the wobble comes out lopsided or even reversed. By observing six transits and modeling the distortion in exquisite detail, Carteret's team reconstructed not just a rough tilt but a full 3D obliquity — a much harder measurement than the simpler sky-projected angle astronomers usually settle for, and one that's particularly difficult to pull off for a planet this small around a star this dim.
Why won't the numbers behave?
In our own solar system, the planets' orbits are misaligned with the sun's spin by only about 7 degrees — close enough to call it flat. A 136-degree tilt is a different category of problem entirely, and it's one that planet-formation theory usually explains with a wrecking ball: a massive companion planet or a binary star, gravitationally bullying a smaller planet's orbit out of the disk plane over millions of years, or a violent early instability that flung planets into chaotic, tilted paths.
GJ 3090 b doesn't have an obvious wrecking ball. The system does host at least one other planet, but the researchers found no massive companion and no binary star lurking nearby that could account for a tilt this extreme. That's the part making the paper's authors — and now other astronomers reading it — sit up. Without a perturber, the standard playbook doesn't have a chapter for this.
The team's alternative explanation is almost as strange as the observation itself: a "primordial disk flip." The idea is that the protoplanetary disk around GJ 3090's star could have taken on a fresh, misaligned dose of gas and dust late in its formation — material falling in from a different direction than the original disk, tilting the whole structure before the planet finished forming inside it. Rather than a planet getting knocked out of alignment after the fact, the disk itself would have been built crooked from the start, with disk-driven migration carrying the young planet inward along that tilted path. It's a proposal, not a confirmed mechanism, and the paper is explicit that it's the explanation left standing once the more familiar ones fail to fit.
Why It Matters
Retrograde orbits aren't unheard of — a handful of hot Jupiters around sun-like stars have shown similarly extreme tilts, usually blamed on a massive companion doing the shoving. What makes GJ 3090 b different is the combination: the smallest planet ever measured this way, the first confirmed case around a red dwarf, and a system that appears to lack the gravitational bully that's supposed to be a prerequisite for this kind of chaos.
Red dwarfs are the most common stars in the galaxy, and they host a disproportionate share of the small, close-in planets that current instruments are best equipped to find and characterize — including many of the "habitable zone" candidates that get the most public attention. If disks around these stars can end up misaligned without any companion planet to blame, that complicates a basic assumption baked into a lot of planet-formation modeling: that a system's history can be read off from its present-day architecture. A tilted orbit with no visible perturber means the tilt might record something that happened earlier and more quietly than astronomers assumed — infalling material, disk warping, or other formation-stage processes that leave no massive companion behind as a smoking gun. Sorting that out will likely require finding more examples like GJ 3090 b, which is exactly the kind of measurement NIRPS was built to make on faint, small red dwarfs where it was previously out of reach.
