Spending hours of half-sleeping in front of Google Earth is always a source of inspiration and joy for geologists, and it also helps populate procrastinating periods.

For Martian geologists, we have the JMARS a planetary geographic information system application, where almost all the orbital images of Mars collected so far are available in one click.

A serendipitous finding

Surfing JMARS in Valles Marineris one day brought me in front of an image from the Mars Reconnaissance Orbiter camera HiRISE which had never been interpreted before. This image, on the floor of one of the main depressions of Valles Marineris, named Ophir Chasma, was highly unusual. With a structural geology background, I instantaneously identified a huge strike-slip shear zone similar to those that cross eroded continental shields: the Variscan Armorican Massif shear zones in France, Precambrian shear zones in the continental shields of Australia, Canada, Finland. The exposed magmatic intrusions next to the shear zone, as well as the sigmoidal deformation framed by linear shears, convinced me that deformation was in the brittle-ductile rheology field.

Exhumed brittle-ductile shear zones

I almost fell from my chair: there was obviously something big behind. Exhumation of deformation patterns found in the middle crust on Earth was highly unexpected on Mars, where an appropriate exhumation mechanism is unknown, in the absence of plate tectonics.

Another fascinating aspect was the width of the shear zone: 1 km at least, suggesting that its exposed length, a few tens of km², is not representative of the total length of the structure. At least hundreds of kilometers must be buried below the surface by more recent rocks.

Analyzing the area in detail from a structural point of view, it appeared that the best exposed part of the shear zone displays a large CC'S deformation pattern (main shears C, secondary shears C', schistosity S), implying a brittle-ductile rheology.

Which depth corresponds to this rheology in the Martian crust? It depends partly on the strain rate, but also a lot on rock type. For gypsum for instance, this rheology may correspond to the surface. For granites, several kilometers below the surface. Here, a granitoid-type composition was suspected because of the typical fracture patterns observed outside the shear zones, and because there are thick magmatic dykes not far from the shear zone on the canyon floor. From the LinMin algorithm, we developed a new, nonlinear spectral unmixing method to determine the mineral assemblage that fits the best the spectra of the sheared rocks - thanks to a CRISM hyperspectral cube very fortunately sitting there. We found that the sheared rocks have a gabbroic composition. 

One of the largest megabreccia ever found

We checked if similar deformation patterns are observed in other areas in Valles Marineris. In the northernmost trough of Valles Marineris, named Hebes Chasma, another shear zone was found. And probably others, in other places in northern Valles Marineris. In the Hebes trough, we additionally found a polymictic megabreccia in a 100 m thick fracture zone, with rock fragments tens of meters wide. Megabreccia are known from a range of environments on Earth, but the rock fragments in the breccia are so big (perhaps bigger than any other breccia fragments reported on Earth? Let me know if you have found bigger!) that whatever the origin, it testifies to exceptionally intense deformation.

So we found a HUGE brittle-ductile deformation system, mainly hidden in the crust, and a gigantic fault breccia. What to do with that? None of these things appear related to the deformation trends observed in Valles Marineris. The shears and the breccia formed before the Valles Marineris system formed, and even before most of the exposed rocks formed, meaning earlier than ~3.5 Ga. At that time, flood lavas and ash falls covered most of the region, then the troughs of Valles Marineris formed, and a new geologic story started, controlled by the deposition of volcano-sedimentary sequences and slope processes. No large deformation system remains from the ante-Valles Marineris period.

The structure of the dichotomy boundary exposed?

Here comes our colleague Richard A. Schultz, one of the sharpest planetary structural geology specialists. I showed him this mess. Fast answer. Easy. Are we not at the margin of the Borealis basin, this huge putative basin that perhaps generated the planetary dichotomy boundary? It is just aligned with the shear zones. Have you checked this?

Of course not. What an idea, the dichotomy boundary is much farther north.

Uups, I had forgotten those papers (first here, then there) from Andrews-Hanna and colleagues. The second one even models how the dichotomy boundary (aka the Borealis basin) influenced the tectonic development of Valles Marineris.

What else could it be?

Has anyone already seriously drawn the consequences of Andrews-Hanna's results for Valles Marineris? I am not sure. The models are based on the thin-shell modeling approach of crustal loading. Are they adapted enough to take the key elements of the evolution of a Martian crust (including tectonic deformation and magmatic processes in the whole Tharsis area) over one billion years or so into account?

Let's test the hypothesis. What if the shear zones and the megabreccia are related to the margin of the Borealis basin?  What else would we expect to find? What are the implications? 

If the margin of the Borealis basin is around here, one would expect the presence of ring faults, like for instance, around Mare Orientale on the Moon. Dike complexes and other intrusions would also form, and some of the dikes would follow the rings. Some of the faults would have been intensely brecciated. One would also expect that the impact would generate hydrothermal activity in a wet crust. The white matrix At that time, 4.5 Ga, the core dynamo was active, so why not think of hydrothermal activity along basin ring faults, generating linear magnetic anomalies? Topographic rings of impact basins are not well developed on Mars, in contrary to the one, but would some old mountains not fit the orientation of the expected rings?

We tested.

Valles Marineris before Valles Marineris

Such geological features are indeed found in the Valles Marineris region. The fault breccia would result from the impact, and the light-toned matrix between the breccia fragments would denote alteration by hydrothermal fluids.

But the Borealis basin scenario also explains the formation of other geologic features in the Valles Marineris region, as well as the orientation of the local crustal magnetic anomalies. These features were not much investigated earlier because taken alone, they do not tell a story. The Borealis basin provides the frame. The chronology below reveals what Valles Marineris may have been before its troughs formed: a brand new (pre-)Valles Marineris  history emerges!

Early life cocooning

An implication would be that the eastern part of Valles Marineris at least was a huge hydrothermal province at the beginning of the history of the planet. Conditions may have been stable for millions of years, perhaps enough for the emergence of life.

Gold and silver

Although because of the hydrothermal environment, the region might be dotted with metalliferous mineralizations, which are commonly found in large continental shear zones on Earth. The excellent book by Michel Jébrak and Eric Marcoux stresses that mesothermal gold deposits on Earth (30% of mined gold) are observed in secondary structures of exhumed brittle-ductile shear zones and their shallow extension to the surface. The modeled sheared rock composition is also where metals from the platinum group are found on Earth.

This is on Earth, not Mars, and Mars may be a totally different case. But we do not really know where to start seeking such minerals on Mars, why not start in the Valles Marineris region?  To draw our conclusions on metal occurrences, we only follow the standard strategy for finding mineral deposits, which starts from remote sensing.

Marsquakes and shear zone reactivation

Finally, large shear zones, like other large continental-scale structures, are known to be potentially reactivated, whatever the stress field. This is abundantly documented on Earth, as is apparent from focal mechanisms documented in the World Stress Map project, but this is not Earth-dependent: it is a consequence of a well-known rule in rock mechanics, the Byerlee's rule. Today, the Martian stress field is plausibly controlled by the body forces generated in the crust of the Tharsis bulge. And the InSight seismometer identified a quake whose epicenter, located in the Valles Marineris region, might correspond to reactivation of a fault from the identified shear zone system.

If our shear zone interpretations are correct, the perspectives open by the reported observations of the Valles Marineris crust early in the planet history are vast and fascinating. But we see an extremely narrow window to this history. What if the crust were exhumed over a much larger area? Perhaps would our interpretations join all the glorious science that fills the dustbins of history. But certainly, there is something big in Valles Marineris that would remain to be identified!

Poster image caption: CaSSIS image MY34_004174_182_1 (ESA/Roscosmos/CaSSIS) of part of the Hebes Chasma Shear Zone.