Why Mars Lost Most of Its Atmosphere

Mars today has a surface atmospheric pressure of about 0.6 kilopascals, less than 1 percent of Earth's 101.3 kilopascals, thin enough that liquid water boils away almost instantly if it appears on the surface. Yet dried river channels, deltas, and mineral deposits mapped by orbiters and driven over by rovers like Curiosity and Perseverance show the planet once had a thick enough atmosphere, and a warm enough climate, to keep water flowing for extended periods. NASA's MAVEN spacecraft (Mars Atmosphere and Volatile EvolutioN), which has been orbiting Mars since September 2014, has measured the culprit directly: with no protective global magnetic field, the solar wind physically knocks Martian air into space, a process scientists call sputtering. This article walks through what MAVEN found, why Mars lost its magnetic shield in the first place, and what that combination of facts tells us about the planet's history and its prospects for future exploration.

The evidence that Mars used to have a thicker atmosphere

Before getting into why Mars lost its air, it is worth establishing that it had much more of it to begin with. The case does not rest on a single observation. It rests on several independent lines of geological and atmospheric evidence that converge on the same conclusion: early Mars, roughly 3.5 to 4 billion years ago, was a very different planet.

The clearest signal comes from the surface itself. Orbital imagery and rover traverses have documented branching valley networks that resemble river drainage systems on Earth, along with deltas where sediment fans out exactly the way it does when a river empties into a standing body of water. Perseverance is currently exploring Jezero Crater specifically because orbital images showed what looked like a delta feeding into an ancient lakebed, and the rover has since confirmed sedimentary layering consistent with that interpretation. Curiosity, working in Gale Crater since 2012, has found mudstones and lake-bed deposits that formed from fine sediment settling in still water, the kind of process that requires standing liquid to persist for a meaningful stretch of time rather than a single flash flood.

Mineralogy adds a second, independent line of evidence. Instruments on Mars orbiters and rovers have detected clay minerals (phyllosilicates) and certain sulfate and carbonate minerals that form through sustained chemical interaction between rock and liquid water. Clays in particular need water to persist for a long time at a stable, moderate pH to form in meaningful quantities, which is a different kind of proof than a photograph of a dry channel. It tells you not just that water flowed, but that it stuck around under chemically survivable conditions.

The atmospheric physics ties these two threads together. Liquid water cannot remain stable on a surface with almost no atmospheric pressure, because at 0.6 kilopascals, water's boiling point drops below its freezing point across most of the planet, so it either freezes solid or flashes to vapor rather than pooling and flowing. For rivers, deltas, and lakes to have existed for the extended periods the geology implies, Mars needed both a thicker atmosphere to hold pressure high enough for liquid water to be stable and, in most current climate models, a stronger greenhouse effect than today's thin carbon dioxide atmosphere alone can provide to keep surface temperatures above freezing at least seasonally. Estimating exactly how thick that early atmosphere was, and how warm the climate really got, remains an active area of research, with published estimates varying depending on which climate model and geological constraint researchers weight most heavily.

Key finding: MAVEN measurements published in 2017 found that roughly 65 percent of the argon gas ever present in Mars's atmosphere has been lost to space, calculated from the ratio of two argon isotopes preserved in the current atmosphere and matched against measurements taken near the surface (NASA, 2017).

The magnetic field Mars doesn't have

Earth's global magnetic field comes from the geodynamo, a term for the process by which convection currents in the liquid iron of Earth's outer core generate a self-sustaining magnetic field. That field extends far out into space, forming a magnetosphere that deflects most of the solar wind, the constant stream of charged particles the sun blasts outward at 300 to 800 kilometers per second, around the planet rather than letting it strike the atmosphere directly.

Mars does not have that shield anymore, and the reason comes down to size and heat retention. Mars is roughly half Earth's diameter and about one tenth its mass, which means it had far less internal heat to begin with and lost that heat to space much faster than a larger planet would. Once the core cooled and its convective churning slowed or stopped, the dynamo that generated a planet-wide magnetic field shut down. Data from the Mars Global Surveyor mission in the late 1990s found the direct fingerprint of that shutdown: patches of strongly magnetized ancient crust, mostly in the planet's southern highlands, that only make sense if a global magnetic field existed when that crust formed and cooled, then later switched off. Based on the age of the rock in those magnetized regions, researchers estimate Mars lost its global dynamo somewhere between roughly 4.1 and 3.6 billion years ago, though the exact timeline and whether the shutdown was gradual or abrupt remain subjects of ongoing research.

Without a global magnetosphere, the solar wind interacts directly with the top of the Martian atmosphere instead of being deflected around the planet. That direct contact is where the atmosphere actually starts leaving.

What ion pickup and sputtering actually do

Ion pickup is the process by which the sun's magnetic field, carried outward by the solar wind, sweeps up electrically charged atmospheric particles at the top of Mars's atmosphere and drags them away into space, the same way a river current will carry off anything floating loosely on its surface. Sputtering is a related but distinct mechanism: some of those picked-up ions get accelerated by the solar wind's magnetic field and slam back down into the upper atmosphere at high speed, physically knocking neutral gas molecules loose the way a cue ball scatters a rack of billiard balls. MAVEN's instruments have directly measured both processes happening in real time, and mission scientists have identified sputtering in particular as responsible for the majority of the argon and carbon dioxide lost from Mars's atmosphere over its history (NASA, 2017).

What MAVEN actually measured, mission by mission

NASA launched MAVEN in November 2013 specifically to answer the question this article is built around: what stripped away most of Mars's atmosphere, and how fast is it still happening? The spacecraft entered orbit around Mars in September 2014 and has spent more than a decade using instruments that measure the solar wind, the planet's weak and patchy remnant magnetic fields, and the composition of gas escaping from the top of the atmosphere.

The headline present-day number from the mission is that Mars is currently losing roughly 100 grams of atmospheric gas to space every second under typical solar conditions, a rate MAVEN's team calculated by directly sampling escaping ions and neutral particles as the spacecraft passed through Mars's upper atmosphere and the region just beyond it. On its own, 100 grams a second sounds modest, a little more than a stick of butter's worth of gas leaving the entire planet every second. But over the roughly 4 billion years since Mars lost its magnetic field, even a slow, steady leak compounds into the loss of most of an atmosphere, especially once you account for periods when the sun was significantly more active than it is today.

That last point matters, because MAVEN did not just measure a steady background rate. In March 2015, a series of solar storms hit Mars while MAVEN was watching, and the mission team found that atmospheric loss accelerated sharply during those events, confirming that stripping is not constant but spikes with solar activity (NASA/University of Colorado Boulder, 2015). Because the young sun, in its first billion years, is understood from solar physics to have been far more magnetically active and to have produced far more frequent and intense solar storms than today's sun, MAVEN's storm-driven measurements imply that early Mars would have lost atmosphere considerably faster than the present-day 100-gram-per-second baseline suggests. That is a key piece of why a planet that lost its magnetic shield roughly 4 billion years ago has ended up with less than 1 percent of Earth's atmospheric pressure today, rather than a slow trickle that would have left more air behind.

The 2017 isotope-ratio result adds the long-term historical piece MAVEN's real-time measurements alone cannot provide. Argon is a useful tracer because it is chemically inert, meaning it does not get locked into rocks or minerals the way carbon dioxide can; whatever argon Mars had, it either stayed in the atmosphere or escaped to space, with no third option. MAVEN's science team measured the ratio between two argon isotopes, one lighter and one heavier, in the current upper atmosphere. Because the lighter isotope escapes more easily than the heavier one, the process leaves what remains behind isotopically enriched in the heavier form, the same principle used to date how much has evaporated from a sealed sample. Comparing that enriched present-day ratio against the ratio measured at the surface let the team calculate that about 65 percent of all the argon Mars ever had is now gone (NASA, 2017). Because carbon dioxide is chemically involved in Mars's carbonate rocks and polar ice caps, its total loss history is harder to pin down precisely with this method, but MAVEN's team concluded the same sputtering process was the dominant loss mechanism for the bulk of the planet's carbon dioxide as well.

Why weak gravity makes the problem worse

Mars's surface gravity is about 3.7 meters per second squared, roughly 38 percent of Earth's 9.8 meters per second squared. That weaker pull lowers the escape velocity, the minimum speed an object needs to break free of a planet's gravity entirely, to about 5.0 kilometers per second on Mars versus 11.2 kilometers per second on Earth.

That difference matters most for light gas molecules. Thermal escape, sometimes called Jeans escape, is the process by which gas molecules moving fast enough due to ordinary heat-driven motion simply exceed a planet's escape velocity and drift off into space, no solar wind involved. Hydrogen atoms, being the lightest element, reach high speeds more easily than heavier molecules like carbon dioxide or nitrogen at the same temperature, so on any planet, hydrogen is the gas most vulnerable to thermal escape. Mars's lower escape velocity means a larger fraction of its hydrogen atoms cross that threshold on any given day compared to Earth, where the same hydrogen atoms would mostly stay gravitationally bound.

This is a genuinely separate mechanism from the solar-wind-driven sputtering and ion pickup MAVEN measured, and it is worth keeping the two straight. Sputtering, powered by the solar wind hitting an unprotected atmosphere, is understood to be the dominant reason Mars lost the bulk of its argon and carbon dioxide. Thermal escape, powered by weak gravity alone, works even on a planet with a magnetic field, and mission scientists have found it plays the larger role specifically in Mars's ongoing loss of hydrogen, mainly as water vapor breaks apart in the upper atmosphere. Both mechanisms drained Mars over billions of years, they just picked off different gases through different physical routes.

There is a third, slower factor that compounds both of these: Mars's size limited how much atmosphere and volatile material (water, carbon dioxide, and other gases that vaporize at relatively low temperatures) it captured and retained during its formation in the first place, compared to a larger planet in a similar orbit. A smaller planet also cools faster, which is the same underlying reason its dynamo shut down early. Every piece of the story, weak gravity, a small core, an early-dying dynamo, and direct solar wind exposure, traces back to the same root cause: Mars is roughly half Earth's diameter, and that size difference cascades through nearly every process that determines whether a planet keeps its atmosphere or loses it.

Earth vs. Mars: the numbers side by side

The table below lines up the key physical differences discussed above. Each of these factors independently makes Mars worse at holding onto an atmosphere than Earth, and together they explain both why Mars lost most of its air and why that loss was effectively permanent rather than something the planet could later reverse on its own.

Factor	Earth	Mars
Surface pressure	101.3 kPa	~0.6 kPa
Global magnetic field	Active (core geodynamo)	Lost ~4.1–3.6 billion years ago
Escape velocity	11.2 km/s	~5.0 km/s
Surface gravity	9.8 m/s²	~3.7 m/s²
Planetary diameter	~12,742 km	~6,779 km (~53% of Earth's)
Present-day atmospheric loss rate	Negligible, shielded by magnetosphere	~100 g/second (MAVEN, typical solar conditions)
Historical volcanic activity	Ongoing (plate tectonics recycles crust)	Largely dormant for hundreds of millions of years

Mars loses on every column that determines whether a planet keeps its atmosphere: weaker gravity, a dead internal dynamo, and no active volcanism to resupply gas lost to space.

Volcanic activity belongs on this list because it is the resupply side of the equation, not just the loss side. On Earth, ongoing plate tectonics and volcanism continually recycle material from the interior back into the atmosphere and oceans, offsetting some gas loss over geological time. Mars's largest volcanoes, including Olympus Mons, the largest known volcano in the solar system, were built by enormous eruptions in the planet's distant past, but the geological consensus, based on crater-counting age estimates of Martian lava flows, is that large-scale Martian volcanism slowed dramatically and mostly stopped hundreds of millions of years ago as the planet's interior cooled. A planet that is both losing atmosphere faster and no longer replacing much of it ends up in exactly the state Mars is in today.

What this means for Mars's climate history and future exploration

Piecing together the magnetic field shutdown, MAVEN's direct measurements of ongoing loss, and the geological evidence for a once-wetter surface gives scientists a coherent, if still incomplete, timeline. Early Mars, protected by a global magnetic field and likely warmed by a thicker atmosphere, appears to have supported at least intermittent surface water for a geologically meaningful stretch of time, long enough to carve valley networks and deposit clay minerals. Once the dynamo shut down roughly 4 billion years ago, the atmosphere began draining away through solar wind sputtering, a process that MAVEN's storm observations suggest would have run faster still while the young sun was more active. As the atmosphere thinned, the greenhouse warming it provided weakened too, cooling the surface and making it progressively harder for liquid water to persist, a feedback loop that likely accelerated Mars's transition from a marginally habitable world to the frozen desert it is now.

This history is not purely academic. It shapes how researchers think about the search for past life on Mars, since a period of stable surface water billions of years ago is exactly the kind of window astrobiologists look for, which is part of why Perseverance is specifically caching samples from Jezero Crater's ancient lakebed and delta deposits for eventual return to Earth. It also matters for any future human exploration or habitation of Mars. Any proposal to substantially thicken Mars's atmosphere artificially, sometimes discussed under the informal term terraforming, runs directly into the fact that Mars still has no global magnetic field and no functioning dynamo to restore one. Without that shield, any newly added atmosphere would remain exposed to the same solar wind stripping MAVEN has already measured stripping the existing thin atmosphere, meaning added gas would face the same long-term erosion Mars's original atmosphere did. That does not make the physics impossible over sufficiently long timescales, but it is a genuine, well-documented constraint that any serious proposal has to account for rather than skip past.

Common misconceptions about Mars's atmospheric loss

A few misunderstandings about this topic come up often enough that they are worth addressing directly, since getting the mechanism right actually matters for understanding the planet's history correctly.

The first is treating gravity alone as the explanation. It is a real and quantifiable contributor, particularly for light gases like hydrogen through thermal escape, but MAVEN's own findings point to solar wind sputtering as the dominant loss mechanism for the heavier gases, including carbon dioxide and argon, that made up the bulk of Mars's early atmosphere. A planet with Mars's exact gravity but an intact global magnetic field would likely have retained dramatically more of its original atmosphere, because the magnetosphere would have kept the solar wind from directly contacting the upper atmosphere in the first place.

The second misconception is imagining atmospheric loss as a single catastrophic event, something closer to an explosion than a process. The evidence points the other way. MAVEN's data describes a gradual, ongoing process that has run for roughly 4 billion years, punctuated by faster loss during solar storms rather than one dramatic stripping event. Mars is still losing atmosphere today, right now, at a measurable rate, which is part of why MAVEN remains operational: it is documenting a process that has not stopped, not one confined to the ancient past.

The third misconception, common in casual discussions of terraforming, is assuming a thicker atmosphere could simply be added to Mars and would then behave the way Earth's does. Because Mars still lacks the magnetic shield that protects Earth's atmosphere, any added gas would face the same long-term solar wind erosion that already stripped the original atmosphere, an obstacle that has nothing to do with how much gas gets added and everything to do with the planet's missing magnetosphere.

The takeaway

Mars lost most of its atmosphere because its smaller core cooled and its global magnetic field shut down roughly 4 billion years ago, and MAVEN has since measured the direct consequence: solar wind sputtering and ion pickup still strip away about 100 grams of Martian gas every second today, a process that accelerates sharply during solar storms and that mission data suggests ran considerably faster while the young sun was more active. Weaker gravity compounds the loss by making it easier for light gases like hydrogen to escape thermally, and the near-total shutdown of Martian volcanism means the planet stopped replenishing what it lost. Argon isotope measurements published by MAVEN's team in 2017 indicate roughly 65 percent of the planet's original argon is gone, a number that, combined with the geological record of ancient rivers, deltas, and clay minerals, paints a coherent picture of a once-wetter world stripped down to a thin, frigid desert by the loss of a single planetary feature: a functioning magnetic dynamo.

---

Read next:

• What makes a planet habitable
• Why Venus is hotter than Mercury
• Why the Moon has so many craters

FAQ

Q: How much atmosphere does Mars have compared to Earth?

Mars's surface atmospheric pressure is about 0.6 kilopascals, less than 1 percent of Earth's 101.3 kilopascals, making it far too thin for liquid water to remain stable on the surface today.

Q: Why did Mars lose its magnetic field?

Mars's smaller core, roughly half Earth's diameter, cooled and largely solidified an estimated 4.1 to 3.6 billion years ago, shutting down the convective geodynamo that had generated a global magnetic field. Magnetized patches of ancient crust discovered by the Mars Global Surveyor mission in the late 1990s only make sense if that global field existed and later switched off.

Q: How is NASA measuring Mars's atmospheric loss today?

NASA's MAVEN spacecraft, orbiting Mars since September 2014, directly measures solar wind stripping atmospheric particles through processes called ion pickup and sputtering, and mission data indicates Mars continues losing roughly 100 grams of gas to space every second under typical solar conditions, a rate that spikes during solar storms.

Q: Does weaker gravity alone explain why Mars lost its atmosphere?

No. Weaker gravity, about 38 percent of Earth's, lowers the escape velocity and makes thermal (Jeans) escape of light gases like hydrogen easier, but MAVEN's team identified solar wind sputtering, driven by the loss of Mars's global magnetic field, as the dominant mechanism behind the loss of the planet's heavier gases, including argon and carbon dioxide.

Q: Was Mars ever wet enough to have liquid water on its surface?

Yes. Ancient river valleys, deltas, and clay minerals studied by orbiters and rovers including Curiosity and Perseverance provide strong geological evidence that liquid water flowed and pooled on Mars's surface for extended periods billions of years ago, when the planet still had a thicker atmosphere and a global magnetic field.

Q: How much of Mars's original atmosphere has actually been lost to space?

MAVEN's science team calculated, using the ratio of two argon isotopes measured in the current atmosphere versus the surface, that approximately 65 percent of all the argon Mars ever had has escaped to space, a figure published in 2017 and used as a proxy for the broader scale of the planet's total atmospheric loss.

Q: Could Mars's atmosphere be artificially thickened for future habitation?

Any effort to substantially thicken Mars's atmosphere would face the same solar wind stripping MAVEN has already measured, because Mars still has no functioning magnetic dynamo to shield newly added gas. This does not rule out the idea outright, but it is a real physical constraint that serious proposals have to account for.