The Surprising Culprit Behind Mars Gully Formation Is Not Liquid Water

Researchers were stunned by early photos revealing Martian gullies, then baffled as these features visibly shifted. Their striking resemblance to Eart...

By Dr. Stella Cosmos

September 13, 2026

What's Carving Active Gullies on Mars? It's Not Water

When the first images of Martian gullies arrived from orbit, scientists were genuinely stunned. Sharp, sinuous channels sliced across crater walls and polar slopes, bearing an almost uncanny resemblance to erosional features carved by flowing water right here on Earth. The shock deepened further when subsequent observations revealed that these gullies were actively changing — shifting, growing, and evolving over timescales of months to years. The implication seemed almost too extraordinary: could liquid water still be flowing, at least seasonally, on the surface of the Red Planet?

The answer, as planetary scientists quickly recognized, was almost certainly no. Mars is a frigid, near-airless world, with surface temperatures averaging around -60°C (-80°F) and an atmospheric pressure less than one percent of Earth's. Under these conditions, liquid water is thermodynamically unstable at the Martian surface — it would either freeze solid or rapidly boil away into vapor. So something else entirely must be responsible for sculpting these dynamic features. Now, a compelling new study is putting the most credible explanation yet on the table: not water, not geysers, but a process called CO₂ frost fluidization — a kind of planetary-scale air hockey effect driven by sublimating dry ice.

The Gullies That Shouldn't Exist

Martian gullies were first discovered in 2000 using images from NASA's Mars Global Surveyor, and they have fascinated and perplexed planetary scientists ever since. Morphologically, they are strikingly similar to the alluvial gullies and debris flows found in mountainous and polar regions on Earth — featuring characteristic alcoves at their heads, sinuous channels running down steep slopes, and fan-shaped aprons of deposited material at their bases. On our own planet, such features are almost exclusively carved by liquid water, making the Martian versions deeply puzzling.

Over the following two decades, orbiters monitored these gullies closely and confirmed the unsettling fact: they were not static, ancient relics. Some gullies appeared to form or significantly extend within a single Martian year. This active modification demanded a present-day process — something ongoing and seasonal. The scientific debate over what that process might be has raged ever since, with liquid water, brines, CO₂ frost, and various other mechanisms all competing for the title of prime suspect.

A New Study Targets a Peculiar Location

A new paper from Apolline Leclef of the Institut d'Astrophysique Spatiale at Université Paris-Saclay, along with her colleagues, available in pre-print on arXiv, takes a rigorous, multi-dataset approach to resolving this debate. Crucially, the team chose to focus on a particularly revealing location: a system of deep, ancient depressions near Mars' south pole known as Sisyphi Cavi.

This choice was deliberate and scientifically shrewd. Most of Mars' active gullies are found at mid-latitudes — temperate zones where multiple processes, including the possible persistence of briny liquid water, cannot be entirely ruled out. Sisyphi Cavi, by contrast, sits in the frigid polar region adjacent to the permanent southern ice cap. Conditions there are so extreme that liquid water of any kind becomes essentially impossible to invoke. Yet Sisyphi Cavi hosts active gullies too — making it an ideal natural laboratory for isolating the true causative mechanism, stripped of ambiguity.

To conduct their investigation, the researchers drew on data from two of the most prolific and long-lived observers of the Martian environment:

Mars Reconnaissance Orbiter (MRO) — specifically the Compact Reconnaissance Imaging Spectrometer for Mars (CRISM), which captures high-resolution spectral data across visible and near-infrared wavelengths.
ESA's Mars Express — particularly its OMEGA (Observatoire pour la Minéralogie, l'Eau, les Glaces et l'Activité) imaging spectrometer, a powerful instrument capable of mapping the mineralogical and icy composition of the Martian surface.

Together, these instruments provided the team with a rich, multi-year spectral record of Sisyphi Cavi across different seasons — exactly the kind of temporal coverage needed to track the freeze-thaw cycle of the Martian polar environment.

First, Eliminating Water

The researchers began by rigorously testing the most intuitive hypothesis: liquid water. Water ice and liquid water both possess a very distinctive spectral absorption feature at approximately 1.5 micrometers (μm) in near-infrared wavelengths — a reliable fingerprint detectable by both CRISM and OMEGA. If liquid water were driving gully formation at Sisyphi Cavi, its spectral signature should appear precisely during the period of active gully modification in spring.

It did not. The 1.5 μm water absorption band was completely absent during the critical spring ice-melt period, when the seasonal blanket of CO₂ ice retreats and gully activity peaks. While the team did detect traces of hydrated salts and clay minerals — compounds that typically require liquid water to form — these were found exclusively on high plateaus and crater rims far removed from the gully systems themselves. They likely represent ancient, relict deposits from Mars' warmer, wetter past, not evidence of any present-day liquid water activity.

"The spectral data are unambiguous: liquid water plays no role in the current modification of gullies at Sisyphi Cavi. The 1.5 μm absorption signature that would betray its presence is simply not there during the active season."

This result is consistent with the broader scientific consensus, recently reinforced by multiple orbital and landed missions, that liquid water on the present Martian surface is vanishingly rare, if it exists at all. Mars lost the bulk of its surface water billions of years ago as its magnetic field weakened and its thick early atmosphere was gradually stripped away by the solar wind.

Then, Testing the Geyser Hypothesis

With water ruled out, the researchers turned to the next leading candidate: CO₂ geysers. On Mars, geysers operate by a mechanism fundamentally different from their terrestrial counterparts, described by planetary scientist Hugh Kieffer and now commonly referred to as the Kieffer geyser mechanism.

Here's how it works: during Martian winter, the polar regions are blanketed by a transparent slab of seasonal CO₂ ice (dry ice), sometimes several meters thick. As spring approaches and sunlight intensifies, solar radiation penetrates this translucent ice layer and warms the dark soil beneath it. This heating causes CO₂ ice at the base of the slab to sublimate directly into gas, building pressure in a confined space beneath the overlying ice. Eventually, the pressure exceeds the structural strength of the ice, which ruptures — sending high-velocity jets of CO₂ gas and dark, entrained dust and sand erupting upward through cracks and vents. The result, visible from orbit, is a characteristic pattern of dark fans and spots radiating across the bright polar surface.

These geysers are a well-documented and genuinely spectacular Martian phenomenon, confirmed by both Mars Express and MRO imagery. And at first glance, they seem like a plausible gully-carving mechanism — energetic jets of gas and sediment erupting onto steep slopes could theoretically initiate downslope flows.

But the timing doesn't fit. According to Leclef and colleagues' analysis, geyser activity at Sisyphi Cavi peaks around the southern spring equinox, driven by the first significant solar input of the season. Gully modification, however, doesn't begin until mid-to-late spring — well after the geysers have largely subsided and the seasonal ice cover has substantially retreated. This temporal mismatch effectively rules out geysers as the primary driver of gully formation. Cause and effect are simply not synchronized.

The "Air Hockey Effect": CO₂ Frost Fluidization

Having systematically eliminated both water and geysers, the study's authors converge on a third mechanism — one that is elegant in its simplicity and powerfully supported by the observational data. The team calls it CO₂ frost fluidization, though one might informally think of it as the Air Hockey Effect.

Anyone who has played air hockey will intuitively grasp the core principle. On an air hockey table, a cushion of pressurized air injected through small holes in the surface supports the puck from below, dramatically reducing friction and allowing the puck to glide effortlessly across the table with almost no resistance. On Mars, a thermodynamic process achieves something remarkably analogous — but on a geological scale.

As the southern polar spring advances and solar energy increases, the base of the seasonal CO₂ ice sheet begins to sublimate. Rather than escaping through discrete vents (as in the geyser mechanism), this gas instead becomes trapped in a thin, pressurized layer between the bottom of the ice slab and the underlying soil. This gas layer acts as a lubricating fluid, effectively decoupling the heavy ice-and-sediment mixture from the bedrock below and dramatically reducing the friction that would otherwise hold it in place on a slope.

The result is a gravity-driven granular flow: a fast-moving avalanche of CO₂ ice, fine-grained sediment, and entrained material that rushes downslope, carving channels and trenches into the Martian landscape as it goes, and depositing characteristic alluvial-style aprons of material at the base of slopes. The timing of this process — beginning after the peak of geyser activity and coinciding with the progressive retreat of the seasonal ice sheet in mid-to-late spring — matches the observed pattern of gully modification at Sisyphi Cavi almost perfectly.

This mechanism also explains why gullies tend to appear on pole-facing slopes, where seasonal CO₂ ice persists longest and accumulates most thickly, providing the greatest reservoir of sublimating gas to power fluidized flows.

Broader Implications: Earth-Like Features, Alien Physics

If Leclef and colleagues are correct — and the evidence they present is compelling — the findings carry significant implications for our understanding of planetary geomorphology. For decades, the instinctive scientific response to seeing Earth-like landforms on other worlds has been to search for Earth-like processes. The Martian gullies triggered exactly this kind of reasoning: they look like water-carved channels, therefore water must have carved them.

This study suggests that assumption deserves scrutiny. As the authors note, Earth-like features do not always require Earth-like physics. The dynamic erosion documented at Sisyphi Cavi proceeds without a single molecule of liquid water, powered instead entirely by the seasonal freeze-thaw cycle of CO₂ in the Martian atmosphere — a process with no direct equivalent on Earth.

The findings reinforce the importance of multi-instrument, multi-season observations in correctly interpreting planetary surface processes.
They highlight the need for caution when applying terrestrial geomorphological intuitions to other planetary bodies.
They suggest that CO₂ frost fluidization may be widespread on Mars, potentially explaining active gullies across a range of latitudes — not just in polar regions.
They have implications for future surface missions to Mars, particularly any operating in polar or high-latitude regions, where such granular flows could pose hazards.
From an astrobiological perspective, the results are sobering — removing yet another potential niche for present-day liquid water on the Martian surface.

For planetary scientists and geologists, however, the picture that emerges is genuinely thrilling. Mars is not a static, dead world. It is a dynamically evolving planet, with active surface processes reshaping its landscape today — just not in the ways we might have initially expected. The Martian atmosphere, thin and cold as it is, drives a seasonal cycle of CO₂ condensation and sublimation that sculpts the terrain in ways we are only beginning to fully understand.

"While not the headline that astrobiologists were hoping for, this result is a reminder that Mars continues to surprise us. The planet is geologically active in ways that are entirely its own — and understanding those processes is essential groundwork for any future human or robotic exploration of its surface."

For the latest research on Mars surface processes and the ongoing search for signs of habitability, readers can explore resources from NASA's Mars Exploration Program and the ESA Mars Express mission page. The pre-print of the Leclef et al. study is available on arXiv, and broader context on Mars geomorphology can be found through the Lunar and Planetary Institute's Mars research archive.

Frequently Asked Questions

Quick answers to common questions about this article

1 What are Martian gullies and when were they first discovered?

Martian gullies are sharp, channel-like features carved into crater walls and slopes across the Red Planet. Scientists first spotted them in 2000 through NASA's Mars Global Surveyor imagery. They closely resemble water-carved gullies on Earth, complete with alcoves, sinuous channels, and fan-shaped debris deposits at their bases.

2 Why can't liquid water be responsible for the gullies on Mars?

Mars is simply too cold and thin-aired for liquid water to survive at the surface. Temperatures average around -60°C (-80°F), and atmospheric pressure sits below 1% of Earth's. Any liquid water exposed to these conditions would either instantly freeze solid or rapidly vaporize, making it physically impossible as a gully-forming agent.

3 What is CO₂ frost fluidization and how does it carve gullies?

CO₂ frost fluidization occurs when dry ice deposits sublimate — turning directly from solid to gas — beneath overlying soil and debris. The escaping gas acts like a lubricant, lifting and mobilizing surface material downslope in a process scientists compare to planetary-scale air hockey, carving gullies without a single drop of water.

4 Are Martian gullies still actively forming today?

Yes, and that's what makes them so scientifically exciting. Orbital observations over two decades confirmed that certain gullies visibly grow, shift, or extend within a single Martian year, roughly equivalent to 687 Earth days. This means an active, ongoing seasonal process — not ancient erosion — is continuously reshaping parts of the Martian surface right now.

5 Where on Mars are these gullies typically found?

Gullies appear most commonly on steep slopes inside craters and along polar-facing hillsides across Mars. These locations tend to accumulate seasonal CO₂ frost during winter months, which later sublimes as temperatures rise, supporting the dry ice theory. Their distribution strongly correlates with areas receiving particular patterns of seasonal frost deposition.

6 Who is currently researching the cause of Mars gullies?

Planetary scientist Apolline Leclef from the Institut d'Astrophysique Spatiale at Université Paris-Saclay, along with collaborating colleagues, recently published compelling new research supporting CO₂ frost fluidization as the primary culprit. Their pre-print study, available on arXiv, represents one of the most detailed investigations into these puzzling Martian features to date.