Below 0.8 Earth radii, atmospheres don’t survive: a new model draws the line in the search for life

A planet’s size may be the single most important filter in the search for extraterrestrial life. That is the central conclusion of a new study published in The Planetary Science Journal on June 4, 2026, which introduces the Smaller Than Earth Habitability Model — STEHM — a computational framework that predicts whether rocky exoplanets smaller than Earth can hold onto an atmosphere long enough for life to emerge.

The work, led by Michelle Hill, a postdoctoral researcher in Laura Schaefer’s Planetary Modeling Group at the Stanford Doerr School of Sustainability, provides the most detailed treatment yet of the lower size boundary for atmospheric retention in the habitable zone. Its practical conclusion is blunt: planets with a radius smaller than 80 percent of Earth’s — below 0.8 R⊕ — are overwhelmingly likely to lose their atmospheres within a billion years. Life, as far as we can detect it remotely, requires an atmosphere. And an atmosphere, it turns out, requires a planet large enough to keep it.

The problem with small worlds

Since 1992, NASA has confirmed more than 6,000 exoplanets, with over 7,000 additional candidates awaiting confirmation. These worlds orbit roughly 4,700 stars across a galaxy that contains at least 100 billion of them, and current estimates suggest approximately one planet per star. The field has gone from questioning whether exoplanets exist at all to cataloguing thousands faster than they can be characterized.

That abundance creates a triage problem. Telescope time is finite. Which planets deserve the closest scrutiny? The most productive approach is to identify physical criteria that reliably separate candidates likely to be habitable from those that are not. Size is one such criterion, and until now it had not been quantitatively anchored to atmospheric survival timescales with sufficient physical rigor.

An atmosphere is not merely a feature of a planet — it is its first line of defense against the hostility of space. It shields the surface from high-energy stellar radiation, regulates temperature, and, in the case of life-bearing worlds, carries the chemical signatures that remote spectroscopy can detect. Without an atmosphere, there is nothing for a telescope to analyze. The question of whether a planet is habitable collapses, observationally, into the question of whether it has held onto its air.

How STEHM works

Hill built STEHM on top of ExoPlex, a Python-based code that calculates a planet’s mass and internal structure from its radius and internal pressure profiles. She generated six distinct planetary models spanning radii from 0.5 to 1.0 R⊕ — half Earth’s radius up to Earth’s exact size — all configured as rocky, CO2-bearing worlds with what geophysicists call stagnant lids: rigid, unmoving crusts that, unlike Earth’s plate-tectonic system, do not recycle material between the surface and mantle.

Stagnant lid worlds represent the most common rocky planet configuration. Earth, with its dynamic plate tectonics, is arguably the exception, not the rule. Choosing this configuration for the model makes its results directly applicable to the majority of terrestrial exoplanets astronomers are likely to encounter.

For each planetary profile, STEHM integrates several coupled processes over geological timescales, tracking how the atmosphere evolves across billions of years under the competing forces that build it up and tear it down.

The physics of atmospheric loss

A planet’s ability to retain its atmosphere is fundamentally a competition between what it produces and what it loses.

On the production side, volcanic outgassing is the primary replenishment mechanism for CO2 atmospheres. When the mantle is hot and active, volcanoes release carbon dioxide, maintaining the atmospheric reservoir against ongoing losses. The heat that drives this activity comes from two sources: residual heat from planetary formation, and the radioactive decay of elements including thorium, uranium, and potassium embedded in the mantle. These radiogenic elements are slow-burning internal furnaces that extend volcanic activity billions of years beyond what formation heat alone would sustain.

The size of the core relative to the mantle matters significantly here. A smaller core with a proportionally thicker mantle can accommodate higher concentrations of carbon and radiogenic elements. Larger inventory of those materials means longer volcanic activity and longer atmospheric replenishment.

On the loss side, stellar radiation — particularly extreme ultraviolet and X-ray photons, collectively called XUV radiation — is the primary destroyer. When high-energy photons strike atmospheric molecules, they can split CO2 into lighter atomic oxygen and carbon ions, which the planet’s gravity can no longer hold. Worse, escaping particles sometimes carry additional molecules with them through hydrodynamic escape, amplifying the loss beyond what photodissociation alone would produce.

The position of a planet within its star’s habitable zone modulates this bombardment. Too close to the star and the radiation is too intense; too far and the surface freezes. But even within the habitable zone, the XUV flux matters, and it is not constant: young stars are far more radiationally active than mature ones, meaning the first billion years of a planet’s life are the most dangerous for its atmosphere.

The 0.8 Earth radius threshold

The results from STEHM are clear and quantitatively grounded. Planets at or above 0.8 R⊕, positioned at a comfortable distance from a sun-like star, can maintain their CO2 atmospheres for 10 billion years or more — comparable to the current age of the solar system. Below that threshold, atmospheric survival drops off sharply.

Planets at 0.7 R⊕ may sustain their atmospheres if conditions are exceptionally favorable: low stellar radiation, high initial carbon content, a thick mantle rich in radiogenic elements, and a cool starting temperature. But these are narrow, contingent windows. For the typical small rocky world, the prognosis is grim.

The mechanism is straightforward. Smaller planets have weaker surface gravity, which means a lower escape velocity and less capacity to hold atmospheric molecules against thermal and radiative losses. They also cool faster, which extinguishes volcanic activity — and therefore CO2 replenishment — earlier. Once the mantle cools below the threshold for sustained volcanism, the atmospheric reservoir drains without replacement.

Hot starts and cold starts

One of the more nuanced findings from STEHM concerns the thermal history of planet formation. Hill identifies hot-start planets — those that formed with higher internal temperatures, often due to more energetic accretion — as particularly vulnerable. A hot-start planet melts its mantle early, disrupting the heat-regulating mechanisms that would otherwise prolong volcanic activity. This exposure occurs precisely when the host star is at its most radiationally active phase, early in its main-sequence life. The double jeopardy — early atmospheric vulnerability coinciding with peak stellar assault — dramatically shortens the lifetime of the atmosphere.

Cold-start planets, by contrast, preserve their internal heat reservoirs more efficiently, extending the window of volcanic activity into epochs when the star has calmed down. The timing of atmospheric exposure relative to the star’s activity cycle turns out to be as important as the absolute quantity of radiation received.

Validation against Venus and Mars

Before applying STEHM to hypothetical exoplanets, Hill tested it against the two nearest examples of rocky planets in the solar system. Venus — 0.95 R⊕, well above the threshold — was correctly predicted to maintain a thick, persistent CO2 atmosphere. Mars — 0.53 R⊕, well below it — was correctly predicted to have shed its atmosphere to the point of near-total depletion. The model reproduces the observed atmospheric fates of both planets, providing confidence in its physical assumptions.

The Mars result has a particular resonance. Public discussion of terraforming Mars often glosses over the fundamental question of why it lost its atmosphere in the first place. STEHM’s answer is that Mars never had a favorable starting position: its small size, combined with the absence of plate tectonics and rapid mantle cooling, meant that the atmospheric odds were stacked against it from the beginning, regardless of starting conditions.

What this means for the search for life

The practical implication of STEHM is that it redraws the relevant search space for exoplanet characterization. With observational resources finite and telescope time at a premium, a robust size filter changes which targets deserve spectroscopic follow-up.

«The only way that we’re going to ever find out if there are signatures of life out there is by observing the atmosphere of these planets,» Hill said. Remote spectroscopy — splitting a planet’s atmospheric light into its component wavelengths to identify chemical fingerprints — is the only tool available for characterizing worlds that no spacecraft will ever reach. That technique requires an atmosphere to exist. STEHM tells astronomers which planets to prioritize for that search.

The model is also timely with respect to upcoming observational infrastructure. ESA’s PLATO mission, designed to discover Earth-like planets around sun-like stars and characterize their properties with high precision, will generate a census of small rocky worlds. STEHM provides a physical framework for interpreting that census and triaging targets. Similarly, future spectroscopic campaigns with the James Webb Space Telescope and its successors will benefit from a principled size floor below which atmospheric follow-up becomes statistically wasteful.

Looking ahead

Hill’s next step is to extend STEHM to mobile-lid planets — worlds with tectonic activity similar to Earth’s. Plate tectonics fundamentally alters the carbon cycle, enabling the carbonate-silicate feedback loop that has regulated Earth’s atmospheric CO2 for billions of years. A mobile-lid planet at 0.7 R⊕ might behave very differently from the stagnant-lid equivalent that STEHM currently characterizes. Quantifying that difference will refine the lower bound and potentially open the door to additional planetary architectures worth investigating.

The broader question the model touches without fully answering is temporal: not just where life could exist, but when. The universe is 13.8 billion years old. Earth is 4.5 billion years old. The first generation of rocky planets in the galaxy may have formed billions of years before Earth. If life requires time — enough time for chemistry to organize, for evolution to produce complexity — then the question of our cosmic uniqueness may be partly a question of timing.

«Maybe the answer to why we haven’t found any life yet is that we’re so early in the grand scheme of what has been created through the lives and deaths of stars,» Hill said. «Maybe we’re one of the first.»

That possibility is not pessimistic. It is, depending on how one holds it, the most extraordinary idea in the history of science.

This article was written by Homer Dávila Gutiérrez, FRAS, based on the peer-reviewed study: Hill et al. (2026), «Smaller Than Earth Habitability Model (STEHM): The Lower Size Limit for Atmosphere Retention in the Habitable Zone,» The Planetary Science Journal. DOI: 10.3847/psj/ae6804