Jezero’s False Beach Hid Three Later Waters

Perseverance found Jezero Crater’s Margin unit is slow-cooled olivine magma later flushed by three different waters. Orbiters had sold that carbonate-rich band as a lakeshore, a prime place to look for ancient life.

A SuperCam study published on 21 September 2026 in Communications Earth and Environment, volume 7, article 728, reconstructs that history from more than 185 bedrock targets. The beach NASA came to sample was mostly a later costume. The rock underneath recorded groundwater, lake or pore water, and a last pulse of hot fracture fluids, which is why the three cores already on the rover now look more useful than shoreline mud would have been.

Orbit Sold a Beach, SuperCam Found Magma

Jezero is a 45-kilometer bowl on the northwest side of the 1,200-kilometer Isidis impact basin. Perseverance landed there in February 2021 to read a vanished lake and cache rock for possible return to Earth. From orbit, a narrow inner-rim band carried a strong carbonate signal at about the old high-water mark, so the Mars 2020 team treated the Margin unit as a shore deposit that might lock in biosignatures.

The rover reached that inner edge in September 2023 and spent the fourth mission campaign walking it. Candice Bedford, a research scientist at Purdue University and the paper’s lead author, said the orbital story did not survive contact.

Before we arrived at the Margin Unit, the main hypothesis, derived from orbital observations, was that the carbonate seen from orbit formed from interaction with the lake that existed in Jezero Crater. But now we know that this location became a sort of crossroads for aqueous systems. The Margin Unit findings are important because Jezero Crater sits inside one of the largest exposures of carbonate on Mars, so what we learn here reaches well beyond this crater.

Candice Bedford, research scientist, Purdue University, NASA briefing

At the highest ground, SuperCam saw coarse, interlocking olivine that had cooled slowly in magma underground, then reached the surface only after the cover eroded away. Magnesium-iron olivine of that grain size is a cumulate, the same family of rock Perseverance already knew from the crater-floor Séitah formation. Bedford’s team now describes the outcrop as a sort of crossroads for aqueous systems built on that igneous body, not a pile of lake sand.

The eastern slabs still show cm-scale layering and possible cross-beds, and a separate stratigraphy study of Mastcam-Z outcrop models still reads those low benches as sediment reworked by wave action along a paleoshoreline near -2,400 meters. That shoreline is a veneer. The carbonate band that made the unit famous is the alteration of olivine, not a beach that grew in place.

Three Fluids Ran Through the Same Cracks

Once the magma body sat in the crater wall, water used its fractures as pipes. Bedford and colleagues group SuperCam chemistry into mineral clusters and tie those clusters to textures in rover images, which lets them order three main alteration episodes without assigning ages. The team can sequence the fluids. It cannot date them.

THE THREE WATER PASSES

Bedford put the first pass in household terms. “The fractures are like pipes, and the carbonate is the Mars ‘limescale’ that eventually blocked the ‘pipes’ up,” she said. Softer rock around those fills then wore away, so the carbonate still stands as raised ridges across the East and West Margin.

The last pass is the odd one. Fluorite on Earth usually means hot water moving through volcanic rock, and the vein chemistry matches that setting, yet the paper notes the local ground lacks the obvious volcanic plumbing that would make the comparison tidy. The hot water arrived. The heat source is still unsigned.

Silica Stops Below the Old Water Line

Perseverance walked about 265 meters (870 feet) of elevation across four informal stretches of the unit. Water damage is not spread evenly. It piles up below the hypothesized second terrace of the old lake, while the High Margin, above about -2,350 meters, still looks like dry, crystalline olivine.

FOUR STRETCHES OF THE MARGIN UNIT

• East Margin: Sols 911 to 1000, the lowest ground at -2,423 and -2,411 meters around Hans Amundsen Memorial Workspace, Turquoise Bay, and Gnaraloo Bay, with low slabs and possible cross-beds.
• West Margin: Sols 1021 to 1161, elevations -2,410 to -2,352 meters, thicker beds, pitted and wind-polished, including the Bunsen Peak workspace.
• North Margin: Sols 1192 to 1211, -2,350 to -2,352 meters on the north side of Neretva Vallis, remote sensing only, no abrasion patch.
• High Margin: Sols 1245 to 1280, -2,338 to -2,159 meters, fractured granular bedrock, one abrasion at Emerita Mesa, no core.

Eleni Ravanis, a planetary scientist at the University of Hawaii at Manoa and a coauthor, tied the silica to that waterline. “Some of the Margin Unit rocks also contain silica,” she said. “Turning olivine into carbonate can leave silica behind, and we see more of that silica in rocks that sat below the water line.”

The high-silica SuperCam cluster appears only in the East and West Margin, the stretches that sat below that second terrace. Near Neretva Vallis and the western fan, the same low rock also looks physically shuffled, so the lake did work the surface even though it did not build the unit.

A Drier Twin on the Crater Floor

If standing lake water were enough, the Séitah formation should be soaked. It sits lower in Jezero, an olivine cumulate that, on paper, spent as much time under the lake as the Margin unit’s low benches. SuperCam and earlier crater-floor work show it did not. Séitah carries far less alteration, and it lacks the carbonate ridges that mark old fractures in the East and West Margin.

Earlier papers suggested later-eroded cover might have shielded Séitah from the lake. Bedford’s team adds a simpler hydraulic reading. Séitah did not have the same fracture network when groundwater was moving, so water-rock reaction stayed limited. Cracks decided the chemistry. Depth in the old lake did not.

High Margin olivine is not a copy of Séitah either. Séitah holds both high-calcium and low-calcium pyroxene with its olivine. High Margin olivine sits with low-calcium pyroxene only, and SuperCam chemistry shows more MgO and less CaO than the crater-floor rock. Related magmas, different bodies, and very different plumbing after they cooled.

How SuperCam Read Rocks From 6.5 Meters Away

The sequence comes from SuperCam, the mast instrument co-led by Purdue, Los Alamos National Laboratory, and France’s IRAP and CNES. When the team flags a target, the rover can fire the laser from as far as 6.5 meters (21 feet). The spark becomes a brief plasma, and the light from that plasma lists the elements. The same head takes close images and reflectance spectra, so chemistry and texture can be matched without parking the arm on every rock.

NASA’s instrument pages describe that head as a laser, camera, and spectrometers package that can read a spot as small as a pencil point. For this paper the team ran a hierarchical cluster analysis on the Margin bedrock set and got eight compositional groups. Dust coatings, about 20 percent of the points, were pulled out because they sit on many Jezero rocks and would blur the unit’s own story. A mixed bin of 356 points, 22 percent of the set, sat too weakly in any group and was set aside.

SUPERCAM’S MAIN ROCK GROUPS

• Olivine cluster: 25 percent of the set, mean SiO2 44.1 wt%, MgO 31.6 wt%, FeOT 24.1 wt%, scattered around the SuperCam olivine standard.
• Fe/Mg carbonate: 17 percent, low SiO2 at 16.3 wt% with high MgO at 35.1 wt% and FeOT at 31.6 wt%.
• Silica-rich phase: 7 percent, mean SiO2 69.3 wt%, matching the glassy fill seen in abrasion patches.

That remote survey is why the waterline pattern holds. The rover did not have to grind every bench. It could shoot ridges, pores, and veins along the traverse and still keep the elevation stacked against the chemistry.

Pelican Point, Lefroy Bay, and Comet Geyser

The mission already pulled three cores from this unit for a possible future Earth laboratory, which is the practical punch of Bedford’s sequence. Each tube now holds more than one fluid chapter, not a single beach layer.

THE THREE MARGIN CORES

East Margin abrasions at Amherst Point and Bills Bay sit with the first two cores. Castle Geyser was abraded at Bunsen Peak before Comet Geyser was sealed. Old Faithful Geyser was abraded later on the descent toward Neretva Vallis. The High Margin abrasion at Emerita Mesa has no matching core, so the driest, coarsest olivine in the traverse is documented in place and not in a tube.

On Earth, water attacking olivine can free hydrogen that some microbes use as food, and it leaves carbonate and silica that can trap chemical leftovers of that metabolism. Three fluids through cracked, olivine-rich rock multiply those traps. The paper’s last line is blunt about the cache: the sequence “cementing the Margin unit and the samples collected by Perseverance as primary targets of astrobiological interest.”

Bedford has spent a decade on Mars rover data and still gets the same lesson. “If there is one thing I have learned after 10 years working with Mars rovers, it is that Mars constantly throws surprises at you,” she said. “It is very rare that things are as we expect them to be from orbital data.” The three tubes will not settle the life question on the rover. They are the reason this outcrop, which failed as a beach, is now a better place to have drilled.

Frequently Asked Questions

What is the Margin unit in Jezero Crater?

It is a light-toned, olivine- and carbonate-bearing band lining the inner western rim of Jezero, mapped from orbit as “marginal carbonates” because of a deep 2.5-micrometer absorption. The Mars 2020 mission made it the fourth science campaign because those carbonates sat at the estimated lake highstand. SuperCam now treats the high ground as an in-place igneous body and the low ground as that body plus later water and some physical reworking near the inlet channel.

Why does olivine matter for the search for ancient life?

When water meets olivine, the reaction can release hydrogen, which some Earth microbes use as food, and it dumps carbonate and silica that can lock in chemical traces of that activity. The Margin unit still holds leftover olivine beside those products, so the alteration did not run to completion. That leftover mix is what makes a returned core useful, because a lab can test both the original grains and the three later fills.

How is SuperCam built, beyond the laser?

The instrument is split into a Los Alamos body unit inside the rover, a French-built mast unit, and a Spanish calibration target with 36 individual standards. Besides laser-induced breakdown spectroscopy, it carries remote Raman, an infrared spectrometer, a microphone timed to the laser spark, and a color imager, a stack that grew out of Curiosity’s ChemCam. The mast-mounted sensor head masses 5.6 kilograms and draws 17.9 watts.

Did Perseverance collect any High Margin rock?

No core was taken above the old waterline. The High Margin work, sols 1245 to 1280, produced the Emerita Mesa abrasion patch and remote SuperCam points on granular, ventifacted bedrock, then the rover moved on. The three sealed Margin samples all come from East and West stretches that sat below the second terrace, so the driest olivine end-member is recorded only in images and spectra.

What does the Jones stratigraphy paper still claim about a beach?

Alexander Jones and colleagues spent about 350 sols of outcrop geology on the same unit and split it in two. They read the western rocks as a variably carbonated olivine cumulate and the eastern rocks as a sandy veneer reworked from that cumulate by waves at about -2,400 meters. Bedford’s SuperCam paper does not erase those eastern textures. It shows the carbonate that made the unit a target is mostly in situ alteration of olivine, with the lake as one of three later fluids rather than the rock’s parent.

Why are the late fluorite veins still a problem?

The 25-centimeter eastern veins carry calcium sulfate and fluorite, a pairing that on Earth usually means hydrothermal brine in volcanic rock, but the paper says the surrounding ground lacks the volcanic features that would explain that pairing in a textbook setting. A large vein informally called Point Cloates is also partly hidden by loose debris, so its link to the bedrock is still uncertain. Those gaps are why the cores, not the orbital map, are the next measurement.