The Planet That Is Leaking Its Air Into Space

For years, the search for atmospheres on rocky exoplanets has mostly returned bad news. The small worlds that orbit red dwarfs sit close to stars that flare hard and blast them with radiation, and one nearby rocky planet after another has turned out to be bare, sun-scorched stone with no detectable air. So when Collin Cherubim, a graduate student at the Harvard-Smithsonian Center for Astrophysics, sat down to work out which of these worlds might still be holding on to a gaseous envelope, he was making a prediction that the field had learned to distrust.

His candidate was LHS 1140 b. It is a super-Earth, roughly 5.6 times the mass of our planet and 1.73 times its radius, circling a small red dwarf about 48 light-years away in the direction of the constellation Cetus. What makes it interesting is where it sits: in the habitable zone, the band around a star where a planet could, in principle, hold liquid water. Cherubim's models suggested that if LHS 1140 b had an atmosphere, helium should be leaking off the top of it, dragged away by the star's radiation. That leak should be detectable.

Catching a leak

To test the idea, the team turned to the WINERED near-infrared spectrograph on the Magellan Clay telescope at Las Campanas Observatory in Chile. Helium leaves a specific fingerprint in the infrared as it streams away from a planet, and in their 2024 observations the signature was there. The lead author described the detection as statistically rock solid, confirming the prediction.

Then came the complication. When the team looked again in 2025, the helium was gone. No leak, or at least none they could see. The escape, in other words, is not steady. It appears to switch on and off, which tells you the planet's outer atmosphere is a restless, changeable place rather than a fixed halo of gas.

That variability is the crux of what this finding is and is not. Detecting helium escaping from the top of an atmosphere is not the same as measuring the atmosphere directly. No one has read a full spectrum of the air on LHS 1140 b or determined what it is made of beneath the thin helium layer at the top. The escape is a signature, an indirect trace of a larger reservoir, and the honest reading is exactly the one the researchers give: this is the strongest evidence so far that a habitable-zone rocky world has retained an atmosphere, not proof of a stable, fully mapped one.

Bare rock or real atmosphere?

The open question is stark, and the headlines tend to flatten it. Jason Dittmann, an astronomer at the University of Florida who was not part of the study, framed the tension well: is LHS 1140 b "a bare rock with no atmosphere that sometimes burps up some gas," or is there a genuine, self-sustaining atmosphere that occasionally sheds material into space, the way Earth's own upper air slowly bleeds away? The on-and-off helium is consistent with either story. A rock that outgasses in fits would explain a signal that comes and goes. So would a real atmosphere buffeted by an active star.

The study's authors lean toward the more optimistic interpretation. If helium is escaping from the top, something has to be replenishing it from below, which points to a fuller atmosphere feeding the leak. From that reasoning, and from models of how atmospheres fractionate over time, they estimate the envelope may have persisted for more than three billion years.

Waiting on water

Settling the question will take a different kind of look. The team plans to use future space-based observations over the next several years to search for water vapor in the planet's atmosphere. If water turns up, that would be strong evidence for a stable, persistent atmosphere rather than an occasional belch of gas from bare rock. Until then, LHS 1140 b sits in an unusually honest scientific position: a promising target, close enough to study in detail, carrying a real signal that genuinely could go either way.

This work was published on July 16, 2026, in the peer-reviewed journal Science by Cherubim, David Charbonneau, and Robin Wordsworth of Harvard and the Center for Astrophysics. The result does not close the question of whether small worlds around small stars can keep their air. What it does is turn a discouraging pattern of empty rocks into something more interesting: a nearby planet that appears to be doing exactly what a world with an atmosphere should do, which is slowly, unevenly, lose a little of it to space.