The cold desert is certain. The path that led there is still being drawn.
Billions of years ago, Mars breathed a thicker air and ran with liquid water — then fell silent. Scientists have long told a tidy story of a dying magnetic field and a sun that stripped the sky away, but the planet's own rocks are now complicating that account, revealing that some of that lost atmosphere was not carried off into space but quietly buried in stone. The question of how Mars became what it is today turns out to be not a single cause but a convergence of pathways, and the full reckoning is still underway.
- The familiar narrative — magnetic field dies, solar wind scours the atmosphere, Mars freezes — is being quietly dismantled by new evidence from both orbit and the surface.
- NASA's MAVEN spacecraft has watched solar wind tear ions from the Martian sky in real time, and isotope ratios in argon confirm that escape to space is real and ongoing.
- A 2025 discovery of iron carbonate minerals in Gale crater suggests that a significant portion of Mars' lost carbon dioxide was not blown away but locked into the planet's own crust.
- The relationship between Mars losing its magnetic dynamo and losing its atmosphere is suggestive but not proven — Venus, with no magnetic field and a denser atmosphere, stands as an awkward counterexample.
- Resolving the balance between space loss and crustal sequestration now depends on better paleomagnetic dating and broader sampling of the sulfate-rich rock layers where carbonates hide.
Mars today is a cold, thin-aired desert — but it was not always so. Billions of years ago, rivers carved its surface and lakes pooled in its basins. The minerals locked in Martian rock leave no room for doubt: liquid water flowed there, at least intermittently, before roughly 3.5 billion years ago. What happened after is where scientific certainty begins to fray.
The standard explanation runs like this: Mars once had a magnetic field that shielded its atmosphere, the field collapsed between 4.2 and 3.7 billion years ago, and the Sun's solar wind gradually stripped the air into space, leaving the planet cold and bare. NASA's MAVEN spacecraft, orbiting Mars since 2014, has watched that stripping happen in real time — during a solar storm in 2015, escape rates spiked dramatically, and argon isotope measurements point in the same direction. The loss to space is real.
But the story is more tangled than the reflex version suggests. Whether early Mars was persistently warm and wet or mostly frozen with only episodic thaws remains genuinely unresolved — climate models support both pictures. The magnetic field's role is equally ambiguous: a planetary dynamo can open channels for atmospheric escape as well as close them, and Venus, with no magnetic field at all, somehow holds a dense atmosphere today.
The sharpest revision came in 2025, when Benjamin Tutolo's team reported that the Curiosity rover had found abundant iron carbonate minerals in the sulfate layers of Gale crater. Scaled across similar deposits on Mars, those rocks could account for several tens of millibar of carbon dioxide — a genuine piece of the missing atmospheric inventory, buried in the crust rather than lost to the void. Some of that carbon appears to have been released again later, hinting at a partial, imbalanced carbon cycle rather than a simple one-way trap.
Mars, it turns out, lost its atmosphere through more than one door. Some air escaped upward; some was swallowed by rock. What remains unknown is the proportion between those fates, and how much the dying magnetic field truly drove either. Better dating of the dynamo's collapse and wider sampling of the sulfate layers may yet close the gap. The cold desert at the end of the story is certain. The road that led there is still being mapped.
Mars today is a cold, thin-aired desert. But the planet was not always this way. Billions of years ago, water ran across its surface in rivers and lakes. Something changed. The standard explanation has become almost a reflex: Mars had a magnetic field that protected its atmosphere, the field died, the Sun's solar wind stripped the air away, and the planet froze. It is a clean story. It is also incomplete.
The water part holds up. Dried riverbeds, ancient lake basins, mineral deposits that only form in the presence of liquid water—these are written into the Martian rocks with enough clarity that no serious scientist disputes them. Liquid water flowed on Mars, at least in fits and starts, before roughly 3.5 billion years ago. That much is settled. Everything else gets murkier.
Take the question of warmth. Early Mars may have been persistently warm and wet, held in that state by a thick carbon dioxide atmosphere and other greenhouse gases. Or it may have been mostly frozen, a cold world that thawed only in episodes, brief windows when conditions allowed water to flow. Climate models support both pictures with equal seriousness. The argument has run for decades without resolution. "Once warmer" is shorthand that papers over a genuine scientific disagreement.
The magnetic field story is similarly complicated. Mars has no global magnetic field today, but the ancient rocks carry magnetic signatures locked in as they cooled—proof that a dynamo once ran in the planet's core. That dynamo shut down sometime between 4.2 and 3.7 billion years ago, around the same period the atmosphere appears to have thinned. The timing is not pinned down precisely. Recent paleomagnetic work led by Sarah Steele at Harvard, published in Science Advances in 2023, suggests the dynamo may have persisted longer and behaved in more complex ways than earlier reconstructions assumed. The field died. When, and how, is still being worked out.
NASA's MAVEN spacecraft, orbiting Mars since September 2014, has provided the clearest evidence that solar wind does indeed strip atmospheric gas into space. During a strong solar storm in March 2015, the spacecraft watched escape rates spike as charged particles were torn away. Measurements of argon isotopes point the same direction—argon does not cycle through rocks and volcanoes the way carbon dioxide does, so its isotope ratio records loss to space fairly directly. By the MAVEN team's reading, most of Mars' atmospheric loss went upward, into the void.
But here is where the story fractures. The assumption that a magnetic field acts as a simple shield—on means protection, off means loss—does not match the physics. A planetary field deflects some solar wind, yes, but it can also open channels for charged particles to escape, and in some configurations it may actually speed up atmospheric loss rather than prevent it. Venus complicates the picture further. It has no internally generated magnetic field and orbits closer to the Sun than Mars, yet it holds a dense atmosphere. The timing between Mars losing its dynamo and losing its air is suggestive, and the protective reading remains the leading one, but the causal link is not established. Some researchers have even argued that a longer-lived dynamo might have helped Mars lose water rather than preserve it.
The real surprise came in 2025. A team led by Benjamin Tutolo reported in Science that the Curiosity rover had found abundant siderite—iron carbonate—in the sulfate-rich layers of Gale crater. Scaled across similar deposits elsewhere on Mars, those rocks could hold the equivalent of a few to several tens of millibar of atmospheric carbon dioxide. That is a genuine reservoir, part of the missing inventory. It means some of the planet's lost air did not escape to space at all. It was locked into the crust as rock. Some of that buried carbon appears to have been released again later, suggesting a partial, imbalanced carbon cycle rather than a one-way trap.
So Mars lost its atmosphere by more than one route. Some was stripped away by solar wind, as MAVEN shows is still happening today. Some was sequestered in the rocks. What remains unknown is the balance between those two fates, what the early climate actually was, and how much the dying magnetic field truly mattered. Better paleomagnetic dating of the dynamo's shutdown and missions that sample more of the sulfate layers are likely to narrow the gaps. The cold desert is certain. The path that led there is still being drawn.
Notable Quotes
The outline is broadly right. The water was real, the dynamo did fail, and the solar wind does strip the atmosphere. But several links in that chain are less settled than the clean telling suggests.— Space Daily analysis
The match in timing between Mars losing its dynamo and losing its air is suggestive, and the protective reading is the leading one, but the causal claim is not established.— Scientific consensus on magnetic field protection
The Hearth Conversation Another angle on the story
The story we all learned is that Mars lost its atmosphere because it lost its magnetic field. Is that wrong?
Not wrong, exactly. But it's simpler than the evidence supports. The field did die, the solar wind does strip atmosphere, and those things happened around the same time. But the physics of how a magnetic field protects an atmosphere is messier than a simple on-off switch. And we've now found that a lot of the missing carbon dioxide didn't go to space at all—it's locked in rocks.
So how much of the atmosphere actually escaped to space?
That's the question. MAVEN and the argon isotope data show that solar wind loss is real and ongoing. But the 2025 discovery of all that iron carbonate in Gale crater means we have to account for a whole other sink. The balance between the two is still being worked out.
Why does it matter whether Mars was warm or cold early on?
Because it changes how we think about habitability and how planetary atmospheres evolve. A persistently warm, wet Mars is a different place than a frozen world with occasional thaws. And it affects how we interpret the magnetic field's role. A warmer climate might have needed more protection, or it might have been more fragile.
The magnetic field question seems especially tangled.
It is. Venus has no magnetic field and keeps its atmosphere fine. Some models suggest a longer-lived Martian field might have actually accelerated water loss rather than prevented it. The timing is suggestive, but causation isn't proven. That's what makes this story still open.
What would settle it?
Better dating of when the dynamo actually shut down, and more samples from those sulfate layers where the carbonate is hiding. The pieces are there. We just need to read them more precisely.