LHS 1140b may be rebuilding its own atmosphere, astronomers find

A rocky exoplanet 48 light years from Earth may be doing something no world in the Solar System appears to do: pulling gas out of the space around it to top up its own atmosphere. The conclusion, drawn from a reanalysis of nearly a decade of transit observations of LHS 1140b, was published on 16 July 2026 and offers the clearest evidence yet that atmospheric loss on a habitable-zone rocky world can be partially reversible.

The result matters because LHS 1140b sits in the so-called habitable zone of a faint, cool star called LHS 1140 in the constellation Cetus. It was first announced in 2017 as a promising target for atmospheric characterisation. The new work suggests the planet is not just losing gas into space but also accreting fresh helium from a surrounding reservoir, a process the study's authors frame as the first plausible detection of atmospheric replenishment on an exoplanet of this size.

A decade of transits, re-read

The team reprocessed data from multiple instruments, including archival Hubble Space Telescope spectra and ground-based transit photometry gathered between 2017 and 2025. By stacking observations across many orbits, the authors could separate the faint, time-varying helium signal at the planet's limb from stellar noise and instrumental drift. The signal they isolated is unusual: a deep absorption feature in the near-ultraviolet helium line at 1083 nanometres, persisting for longer than the transit itself, and recurring with the planet's 24.7-day orbital period.

That pattern, the authors argue, is consistent with an extended envelope of helium gas surrounding the planet. As the world passes in front of its star, the envelope imprints a longer, shallower absorption signature than the planetary disc alone would produce. The team estimates the effective radius of the helium halo extends well beyond the planet's solid surface, suggesting a substantial reservoir of light gas.

Why the gas keeps coming

The replenishment mechanism is the central claim. On most known exoplanets, atmospheric escape is a one-way street: stellar radiation, especially in the extreme ultraviolet, heats the upper atmosphere and drives gas out into space faster than it can be replaced. For LHS 1140b, the new modelling suggests the host star's relatively low ultraviolet output and the planet's appreciable surface gravity create a window in which the outflow cools and partially condenses before it escapes the planet's Hill sphere. Some of that material, the study proposes, can fall back along the magnetic field lines and rejoin the upper atmosphere, while a steady trickle of fresh gas is drawn in from the interplanetary medium.

The authors are careful to note that this is a model-dependent interpretation. The observed helium signature could also be produced by a static, primordial envelope that has been slowly leaking since the planet formed. Distinguishing between the two scenarios will require catching a bright stellar flare in real time and watching how the helium signal responds; a flare should temporarily strip a replenishing envelope but only modify a static one.

A counter-narrative: ancient air, not new air

The competing reading is straightforward and not unreasonable. LHS 1140b is older than the Earth. It may have begun life with a thick primordial hydrogen-helium envelope that has been slowly eroding ever since, much as a comet loses mass on each perihelion. In that view, what the team is seeing is the tail end of an escape process, not a feedback loop. The extended absorption signature would then reflect a still-large but shrinking atmosphere, and the "replenishment" framing would be a misreading of a slow leak.

The study acknowledges this alternative and tries to rule it out with two arguments. First, the inferred mass-loss rate is too low for a steady-state hydrodynamic escape model, given the star's current X-ray and ultraviolet output. Second, the helium signal shows modest variability over the decade of observation that is more consistent with an active inflow-outflow balance than a monotonically declining reservoir. Neither argument is decisive on its own, and the paper explicitly calls for sustained monitoring to settle the question.

What this changes about the search for habitable worlds

If the replenishment interpretation holds, it complicates one of the central assumptions of exoplanet habitability research: that a planet's atmosphere at any given moment reflects a long-term balance between delivery and loss, with little feedback between the two. A world that can top itself up from its local space environment has more atmospheric stability than the standard picture suggests. That has practical consequences for target selection on the upcoming Habitable Worlds Observatory and for the European Space Agency's Plato mission, both of which will need to decide which rocky planets in nearby M-dwarf systems merit expensive follow-up spectroscopy.

It also redraws the line between "worlds with air" and "worlds without." If LHS 1140b can keep a measurable helium envelope through a dynamic inflow, then the absence of an atmosphere on a similar planet cannot be read as evidence of catastrophic loss. It may simply mean the inflow has not yet been caught in the act, or that the star has been unusually quiet for the window of observation.

What remains uncertain

The single largest caveat is sample size. The helium signature is detected on one planet, in one system, with one instrument suite. Stellar activity models for M-dwarfs carry systematic uncertainties of a factor of two or more, and those uncertainties propagate directly into the escape and inflow rates. The authors' replenishment model is a fit to those rates, not an independent measurement of gas motion. A second epoch of Hubble or James Webb observations, timed to overlap with a stellar flare, would be the cleanest test.

The broader question is whether this is the rule or the exception among rocky habitable-zone planets. Current facilities cannot answer that; only a dedicated ultraviolet spectrograph on a future flagship mission could survey enough targets to put LHS 1140b in context. Until then, the safest reading is that the planet is doing something interesting and possibly novel, but that the field has not yet built the instruments to confirm it.

Desk note

Wire coverage of the study, carried through scix and Phys.org on 16 July 2026, framed the result as evidence of an "atmosphere" on LHS 1140b. This publication treats that framing as provisional: the underlying signal is real and reproducible, but the interpretation as dynamic replenishment rather than residual primordial gas is model-dependent and will need a triggered flare observation to confirm. We have leaned on the cautious reading offered by the paper's own authors rather than the more enthusiastic paraphrase that some headlines have used.