Martian gullies have been the center of a debate about whether Mars ever has flowing water. Now, a comprehensive study examines the question.

Mars gully
The Mars Reconnaissance took this image of a gully on a crater wall from an altitude of about 150 miles (250 km).
NASA / JPL-Caltech / UArizona

Lots of Mars’s hillslopes have gullies, steep ravines that grow when we’re not looking. They look so much like Earth’s own gullies, formed when water and debris carve into steep slopes, that it’s easy to think that water must be involved on Mars, too. But physics says water shouldn’t ever be liquid anywhere on the Martian surface today. Many scientists therefore think gully formation must be triggered by some kind of dry process, involving ice (either water ice or carbon dioxide ice) that sublimates directly from solid to gas.

Gullies have been at the center of an ongoing debate about whether present-day Mars is ever wet or is always dry, ever since the formations were first seen in Mars Global Surveyor images at the turn of the millennium. In a new study, published in the August 2024 issue of Icarus, Axel Noblet (University of Western Ontario, Canada) and colleagues amass data on nearly 8,000 gullied slopes and come up with an answer to the “wet or dry” question: It’s both, and it depends. The surface is probably mostly dry now, but it could have been wet rather recently.

Map of where gullies are on Mars
This elevation maps shows the global distribution of hills with gullies.
Noblet et al. / Icarus 2024 August / CC BY-NC 4.0

If this seems like a cop-out, it’s not; Noblet’s paper articulates a “hierarchy of factors” that describes where gullies occur, with well-supported explanations as to why they form in one place and not another. None of the explanations in this paper are new. What’s new is how Noblet and coworkers reconcile apparent contradictions and inconsistencies among other researchers’ explanations of gully formation, explaining why an explanation that works for one spot on Mars doesn’t work in another.

The study includes four factors that control the location of Mars gullies. I’ll get to what the factors imply in a minute, but first, here’s the list, in order of their importance:

There are no gullies near the equator nor at the poles. There’s a very strong latitude dependence: they’re only found between 26° to 83°S and 28° to 76°N, and they are densest in a latitude band around 35° to 45° in both hemispheres.

Map of gullies
These maps show the density of gullies on Mars.
Noblet et al. / Icarus 2024 August / CC BY-NC 4.0

They usually occur on steep slopes. How many gullies there are in an area is strongly correlated to how much of the area has steep slopes. The median gradient for gullies is 15°, with 95% of them having slopes between 6° and 27°. (Note that the angle of repose for dry sand on Mars is 33°; at steeper angles, piles of sand would slump under the pull of gravity.) The preference for steep slopes explains why gullies are much more common in the southern hemisphere than in the north — there simply aren’t that many steep slopes in the north, which mostly hosts flat-lying lowlands. (Conversely, there’s no correlation between gully occurrence and elevation; whatever makes gullies happen doesn’t appear to care what the local elevation is.)

Gullies mostly form on equator-facing slopes nearer the poles, and pole-facing slopes nearer the equator. Also, there are more east-facing gullies nearer the poles, and no east-west preference nearer the equator. The switch between equator- and pole-facing slopes happens at a latitude of around 40° to 45° – note that this is the same latitude at which there are also the most gullies. However, there are lots of local exceptions to these general rules, which means that local geological conditions can override whatever causes this general preference.

Map showing how gullies are oriented
These maps reveal trends in the tendencies of gullies to either face the pole or equator (top map) as well as their tendency to face east or west (bottom map).
Noblet et al. / Icarus 2024 August / CC BY-NC 4.0

Gullies mostly form in areas where there is some ground ice, and not in areas where there is no ground ice, nor in areas with massive ice deposits. This one is especially intriguing — what is it about some ice that makes gullies more likely?

Map of gullies with respect to ice
Gullies tend to appear on the edge of ice distributions (yellow) rather than areas with deep ice (blue) or no ice at all.
Noblet et al. / Icarus 2024 August / CC BY-NC 4.0

What It All Means

And now for the interpretations. That the slopes of nearly all gullies are considerably shallower than the steepest angle at which Martian sand would slump says that gravity absolutely cannot be the only factor driving their formation. Something has to help reduce the friction between the particles of dust, sand, or rock that’s flowing downslope. But that something does not have to be liquid water; it could be a gas released from underground, for example, such as water vapor or carbon dioxide.

The east-facing preference at high latitudes suggests that when polar hillslopes are suddenly blasted with sunlight after a long, cold night, the gullies flow, depositing fresh, dark material at their tips. West-facing slopes don’t get such a temperature shock, warming slowly throughout the day. This effect would be especially pronounced in spring, when areas that have remained cold and dark throughout the polar night see sunlight for the first time in months.

The pole-facing preference at mid-latitudes is a bit harder to explain — unless you happen to know that Mars’s axial tilt has changed over time, repeatedly, in cycles lasting a couple hundred thousand years. (Which is to say, geologically recently.) The planet’s present tilt of 25° is pretty mild; in the past, it has been 45° or even higher.

Right now, most of Mars never sees ground temperatures above 0°C except for the top few millimeters of its surface. But with higher axial tilt, large areas of the mid-latitudes would receive enough sunlight in summer to melt the permafrost. Then, during winter, water vapor would become trapped as ice in the porous ground again, and the cycle would repeat. When the axial tilt becomes less extreme, areas that used to experience polar winters no longer would. Gully formation would gradually come to a halt, ending earlier at lower latitudes and later in higher latitudes.

The correlation with some ground ice strongly suggests that gullies have formed in places at the edges of ice deposits. That suggests they formed during a transition from a wetter period, favoring the formation of ground ice, to the present-day, dry period.

All that explains why gullies might have formed on Mars during periods of high axial tilt. But what about the gullies that have been seen to expand on Mars today, when the tilt is low? They really, really can’t result from liquid water unless it’s incredibly briny (to keep it from sublimating at the low surface pressure). And what about the gullies near the poles? How are they still changing over time?

A plausible answer to these questions involves carbon dioxide. While dry ice doesn’t expand as it freezes, like water does, condensation of dry ice can fill all the pores in the ground and glue particles together. If that ice suddenly goes away — say, when sunrise on a spring day lights up the ground for the first time in months — the sudden withdrawal of support can collapse the ground, and the rush of expanding gas can “fluidizing” the soil, reducing friction and encouraging flow even though no liquid is present.

So the advocates for dry gullies may be right about what’s happening on Mars right now, but it also appears that wetter processes have made most gullies across the globe not very long ago.

What’s most important about this study is that it’s not just a hypothesis — it’s a synthesis of 25 years of Mars science. It’s a theory with predictive power, and any Mars geomorphologist worth their salt should be running to the HiWish website to request that the Mars Reconnaissance Orbiter obtain the high-resolution images needed to test this theory.


Image of Anthony Barreiro

Anthony Barreiro

June 12, 2024 at 7:03 pm

Planetary science is getting more and more like geology, and less and less like astronomy.

In the last paragraph the phrase "worth their salt" is underlined as if it is a hyperlink, but it doesn't link to anything, at least in my firefox browser.

You must be logged in to post a comment.

Image of Peter Rowen

Peter Rowen

June 18, 2024 at 4:40 pm

Weird. Inspecting the HTML shows bare anchor tags with no link: salt. Maybe someone forgot to add the link to an earlier story about Martian brine or something.

You must be logged in to post a comment.

You must be logged in to post a comment.