Peering Beneath Mars's Wrinkles: What Hidden Faults Tell Us About the Red Planet's Geology

If you've ever scrolled through images of Mars, you've probably spotted them: long, sinuous ridges snaking across the surface. We call them "wrinkle ridges," and from orbit they look like nothing more than gentle bumps. But they're really windows into some genuinely fascinating geology unfolding beneath the Martian surface.

In my earlier work, I studied how these ridges form and what they reveal about the ancient stresses locked into the Martian crust. One big question kept nagging at me, though: what do these structures actually look like underground?

A Rare Opportunity

Here's the obvious problem with planetary geology—we can't exactly fly out to Mars and dig trenches to peek beneath the surface. But every now and then, nature does the digging for us. Impact craters, collapse pits, and ancient valleys occasionally slice right through wrinkle ridges, leaving behind natural cross-sections that lay the subsurface bare.

So that's what I went looking for. I hunted across the Tharsis region of Mars for these rare exposures, zeroing in on twelve sites where deep cuts—some reaching 500 to 1000 meters below the surface—revealed the hidden architecture beneath wrinkle ridges.

What We Found Underground

Using high-resolution images from Mars orbiters, we mapped out the tangled patterns of faults beneath these ridges. What we found was far more intricate than simple surface buckling.

The Fault Patterns

The subsurface is dominated by thrust faults and reverse faults —fractures where one block of rock has been shoved up and over another. The distinction matters: we call it a "thrust" if the fault dips at less than 45 degrees, and a "reverse fault" if it's steeper than 45 degrees. Both turn up beneath wrinkle ridges all the time, and often you'll find both in the same structure.

One of the most consistent things we saw was that asymmetric wrinkle ridges have a dominant fault system on one side. The main thrust fault usually breaks the surface right at the base of the ridge's steeper slope. Picture the ridge sitting on top of a ramp that tilts everything to one side—that's roughly the geometry.

Wherever we could find planar sections exposed in crater walls or cliff faces, we measured the dip angles of these faults. On average, the primary thrust faults dip at about 37° ± 2° , while the associated backthrusts (faults dipping the opposite way) come in a touch shallower at 28° ± 2°.

The Listric Shape Discovery

One of my favorite findings was that many of these faults don't keep a constant angle as they descend. At nine of our twelve study sites, we watched faults start out fairly steep near the surface and then bend over to shallower angles at depth —typically somewhere between 500 and 1000 meters down.

That "listric" shape—curving from steep to shallow—really matters. It hints that these faults might be tying into a nearly horizontal weak layer deeper in the crust, what geologists call a detachment zone.

Complexity and Subsidiary Faults

Most wrinkle ridges aren't the work of a single, tidy fault. They're built from a multitude of subsidiary and splay faults —smaller faults that branch off the main structure. These are what give the features their characteristically "wrinkled" look, and their name. It's a bit like pushing a rug across the floor: you don't get one clean fold, you get a messy crowd of smaller wrinkles and creases.

Folding and Faulting Together

At one spectacular exposure along the wall of Ophir Chasma, we could actually trace individual rock layers straight through a wrinkle ridge. The layers are clearly folded—bent into a lopsided arch, what geologists call an anticline. Small thrust faults cut across the folded layers, and you can watch how each fault lines up with a little step in the ridge topography.

This site is a beautiful reminder that wrinkle ridge formation isn't purely about faulting or purely about folding. It's the two processes working hand in hand.

Polarity Changes Along Strike

Here's another curious one: some wrinkle ridges change character along their length. A ridge might have a steep eastern slope in one stretch, then flip to a steep western slope further along. These polarity changes track which fault system is doing most of the heavy lifting. Where the dominant thrust switches direction, the ridge's asymmetry switches right along with it.

What This Means: Testing Formation Models

For decades, planetary scientists have floated all sorts of models for how wrinkle ridges form. Our subsurface observations finally let us test which ones actually hold up against reality.

Simple Buckling: Doesn't Fit

The simplest idea—that wrinkle ridges are just buckled layers of rock, like a bent sheet of paper—doesn't survive contact with what we see. The clear presence of major thrust faults cutting through the structure rules out simple elastic bending.

Fault-Bend Folds: Unlikely

Another model casts wrinkle ridges as "fault-bend folds"—folds that develop where faults have a flat-ramp-flat geometry. This one predicts that the main fault should have a shallow dip where it reaches the surface. But we keep finding faults that break the surface at steep angles (often around 38°). That mismatch makes fault-bend folding an unlikely explanation for most wrinkle ridges.

Fault-Propagation Folds: Best Match

The observations fit best with fault-propagation fold models. The idea here is that a thrust fault propagates upward from a deeper detachment but bleeds off displacement as it climbs. The "missing" displacement gets soaked up by folding of the overlying rocks. And that produces exactly what we observe: an asymmetric fold with a steep forelimb, where a steep reverse fault reaches the surface at the base of the ridge.

One variant in particular—the Chester and Chester (1990) two-ramp model—matches the characteristic "stair-stepping" profile that so many wrinkle ridges show, where a narrower ridge perches asymmetrically on top of a broader swell.

Continuum Mechanics Models: Also Consistent

Numerical models that treat the crust as a continuum (like the elastic dislocation model by Watters, 2004, and the boundary element model by Okubo and Schultz, 2004) line up nicely with our observations too. They predict a lot of what we see: blind thrust faults, backthrusts forming in weak zones, bimodal ridge topography, and the way thrust faults pair up with the steeper ridge slopes.

The Detachment Question

One of the longest-running debates in wrinkle ridge studies is whether these structures are "thin-skinned" or "thick-skinned" deformation.

Thin-skinned means the faults connect to a shallow, nearly horizontal detachment—a weak layer that lets the upper crust slide along independently of whatever's beneath it. Thick-skinned means the faults plunge deep into the crust with no significant detachment to speak of.

Our observations point toward thin-skinned deformation, at least for the upper crust. The listric shape of the faults—steepening upward from shallower angles at depth—suggests they're joining onto something nearly horizontal below our level of observation.

In my earlier study on Lunae Planum, I argued that this detachment might sit at the base of the permafrost, where trapped liquid water under pressure creates ideal conditions for a weak, slippery layer. The subsurface observations from this study fit that interpretation, though I'll be honest—we can't directly see the detachment itself, since our exposures only reach 500-1000 meters deep while the detachment probably lies at 2.5-4 kilometers.

Conjugate Faults and Symmetric Ridges

Most wrinkle ridges are asymmetric, but a few wear more symmetric profiles. At one site in particular (Area 12), we found a gorgeous system of conjugate reverse faults —two fault systems dipping toward each other at similar angles and meeting at depth. That arrangement builds a "pop-up" structure with relatively even slopes on both sides.

Finding conjugate faults is a big deal, because it confirms one of the formation models proposed by earlier researchers (Allemand and Thomas, 1992; Mangold et al., 1998). That said, conjugate systems of equal weight seem to be the exception, not the rule. Far more often, if there are faults on both flanks of a ridge, one calls the shots while the other plays a supporting role.

The Role of Impact Craters

There's a fair methodological worry to address here: are the structures we see in crater walls really representative, or did the crater itself reshape the wrinkle ridge?

We can actually answer this with some confidence. In most cases, the wrinkle ridges clearly came after the craters—we can see uplift of the crater floors, which tells us faulting kept going long after the impact. The craters are viewing windows into pre-existing structures, not events that rewrote them.

A few ridges do show slight bending or deflection where they cross craters, but the overall fault patterns and geometries are just what we'd expect for these structures. The thinner overburden left by the crater cavity might make slip a little easier, perhaps nudging displacement slightly higher on faults inside the crater, but the basic architecture stays representative.

Connecting to Terrestrial Analogs

The structures we see beneath Martian wrinkle ridges look strikingly like features in certain fold-and-thrust belts here on Earth—especially the Yakima fold belt in Washington State. There, asymmetric anticlinal ridges grew by fault-propagation folding of Columbia River Basalts, a setting that's a close cousin of the Martian flood basalt plains where most wrinkle ridges live.

The Yakima folds show the same cast of characters: thrust faults reaching the surface at steep angles, folded basalt layers, detachments at depth (deeper than on Mars, admittedly), and subsidiary faulting that complicates the surface morphology. Having that terrestrial analog gives me real confidence that the fault-propagation fold models we're applying to Mars are on solid ground.

Implications for Martian Crustal Structure

So what do these observations tell us about Mars more broadly?

First, they confirm that the Martian crust went through significant horizontal shortening. The amount shifts from place to place—in my Lunae Planum study, I calculated shortening ranging from about 116 meters in areas closer to the Tharsis center to 56 meters further away.

Second, the apparent presence of shallow detachment zones points to mechanical layering in the Martian crust. You simply can't have thin-skinned deformation without weak layers that let the upper and lower crust decouple.

Third, the fact that we see brittle faulting and ductile folding happening side by side tells us something about the mechanical properties and conditions in the upper few kilometers of the Martian crust during the Hesperian period (roughly 3.6-3.9 billion years ago).

Fourth, if the detachment zone really does sit at the base of a permafrost layer (as I argued in my Lunae Planum study), then these structures offer indirect evidence for subsurface water distribution on ancient Mars. The depth at which the faults flatten out gives us clues about where liquid water might have lurked beneath a frozen surface.

Variability is the Norm

If there's one lesson that runs through this whole study, it's that wrinkle ridges are structurally diverse. They share a common toolkit—thrust faulting, folding, horizontal shortening—but the details swing wildly from one site to the next.

Some ridges are symmetric; most are asymmetric. Some show conjugate faults; most have a single dominant system. Fault dips range from fairly gentle (20-30°) to quite steep (50-60°). Some ridges show clear listric faults; others look more planar.

That variability probably comes down to differences in:

• The properties of the crustal materials being deformed
• The depth and character of weak layers
• The local stress conditions
• The amount of shortening accommodated
• The presence or absence of pre-existing structures

Rather than forcing every wrinkle ridge into a single formation model, I think we're better off accepting that these features formed under a whole range of conditions and can wear a whole range of structural styles.

Limitations and Future Work

Let me be upfront about what this study can't do. We can only see the uppermost 500-1000 meters of the wrinkle ridge subsurface. The crucial question of whether there's a regional detachment at 3-4 kilometers depth stays stubbornly out of our direct reach.

To really settle the thin-skinned versus thick-skinned debate, we'd need to see much deeper. Future missions carrying ground-penetrating radar or seismic instruments could potentially "see" those deeper structures. The InSight lander has already handed us our first detailed seismic data from Mars, and as more of that data rolls in from different regions, we'll be able to assemble a far more complete picture of Martian crustal structure.

There's another catch worth naming: the faults we see near the surface might be subsidiary structures rather than the main, deep-seated ones. We've done our best to focus on the most prominent, throughgoing faults, but there's always some uncertainty about which features are primary and which are secondary.

A Personal Reflection

Mapping faults on Mars is painstaking work. You stare at grayscale images, trying to trace faint lineations along crater walls, measuring angles from subtle shifts in slope, and forever second-guessing whether what you're seeing is a real geological structure or just an artifact of the lighting, the image processing, or your own wishful imagination.

But when the pieces finally click—when the fault trace you mapped on the northern crater wall connects perfectly with the one on the southern wall, when the dip angles you measured match what the kinematic models predict, when the ridge morphology suddenly makes complete sense given the subsurface structure—there's a deep satisfaction in it.

We're reading the tectonic history written into the Martian landscape, piecing together events from nearly four billion years ago on a planet none of us has ever set foot on. Each wrinkle ridge is a story of ancient stresses, of crust buckling under enormous forces, of rocks bending and breaking in ways that still leave their mark, visible from orbit, today.

Bringing It All Together

So what's the big picture? Wrinkle ridges on Mars formed through a combination of thrust faulting and folding, driven by horizontal compression of the crust. The structures are dominated by asymmetric fault-propagation folds, where thrust faults climb upward from deeper detachment zones and sculpt the ridge's characteristic shape.

The fault patterns are complex, with a crowd of subsidiary faults creating that wrinkled appearance. Faults tend to steepen upward, hinting that they connect to shallower-dipping structures at depth. The observations sit most comfortably with thin-skinned deformation in the upper crust, though we can't rule out deeper-seated faulting below our level of observation.

The subsurface architecture changes from ridge to ridge, mirroring the diverse conditions under which these features formed. Still, the same threads run through all of them: horizontal shortening, thrust faulting, asymmetric folding, and structural complexity.

Put alongside my earlier work on Lunae Planum—where I quantified shortening amounts, detachment depths, and proposed a mechanism involving water-lubricated décollement zones—we now have a fairly complete picture of how these enigmatic features came to be, and what they tell us about Martian crustal evolution.

The wrinkles on Mars's surface turn out to be anything but skin-deep. They're the fingerprints of a complex tectonic history, written in faults and folds that we're only now learning to read.

---

This research was published in Earth and Planetary Science Letters (2022) and builds on my earlier work on Lunae Planum wrinkle ridges published in Icarus (2022). The detailed structural analyses were conducted using high-resolution imagery from the Mars Reconnaissance Orbiter's HiRISE and CTX cameras, combined with elevation data from various Mars orbiter missions.

The study areas included wrinkle ridges at Solis Planum, Nilus Dorsa, Coprates Chasma, Lunae Planum, and Thaumasia Planum—all part of the circum-Tharsis wrinkle ridge system.