The surface of Mercury tells a remarkable story of planetary contraction. Like a grape transforming into a raisin, the planet has wrinkled as its interior slowly cooled. When a planet’s interior loses heat, its rocky outer shell is forced to accommodate the shrinking volume beneath it. The crust responds by crumpling, forming dramatic geological features such as ridges, cliffs, and steep scarps that scar the planet’s surface. These features, known to scientists as shortening structures, are direct evidence of the immense compressive forces that have acted on Mercury’s crust as the planet gradually lost its internal heat.

For decades, planetary scientists have relied on mapping these shortening structures to estimate exactly how much Mercury has shrunk. The logic is elegant: the more the crust has crumpled and folded, the more the planet’s radius must have decreased. By measuring the total accumulated strain expressed in these tectonic features, researchers can work backward to calculate the planet’s loss of radius. Previous estimates based on this method, informed by data from NASA’s MESSENGER mission, suggested that Mercury’s radius had decreased by approximately 8.3 kilometers since the planet’s early history.

However, the new research reveals a fundamental flaw in that approach. According to the study, published in the journal Geophysical Research Letters, the visible record of Mercury’s tectonic activity is far from complete. The planet’s rough and pockmarked surface, bombarded by impacts over billions of years, has obscured many of the very structures scientists use to measure its contraction. In essence, the planet’s own violent history has hidden the evidence of its shrinkage.

“Mercury’s surface preserves a record of how the planet has cooled and contracted, but we found that this record is incomplete,” explains Gaku Nishiyama, lead author of the study and a researcher at Hokkaido University. “We compared a global map of Mercury’s surface roughness with maps of shortening structures and contraction. Once we account for the effect of rough terrain, Mercury appears to have shrunk considerably more than what the visible tectonic record alone suggested.”

The research team’s methodology centered on a systematic comparison between surface roughness and the distribution of tectonic features across the entire planet. By overlaying global maps of surface texture with maps documenting the locations of ridges and scarps, the researchers identified a striking pattern: shortening structures were far more likely to be identified in smooth terrain than in rough regions. This correlation was too strong to be coincidental. Instead, it pointed to a systematic bias in the geological record, where the roughest terrain was concealing tectonic structures from view.

A particularly compelling piece of evidence emerged when the researchers examined areas surrounding relatively young impact craters. When a large asteroid or comet strikes a planetary surface, the explosion excavates material from beneath the crust and throws it outward in a debris field known as ejecta. These rough, jumbled deposits can blanket the surrounding landscape, burying older geological features beneath layers of impact debris. Around the Rachmaninoff crater, a prominent impact basin on Mercury, the researchers observed that shortening structures were noticeably less common in areas covered by rough ejecta. Additionally, the structures became progressively less visible closer to the crater itself, suggesting they were partially buried under impact-related material. This spatial pattern provided a natural experiment, demonstrating exactly how impact debris can erase tectonic evidence from the scientific record.

By mathematically correcting for this obscuration effect, the researchers recalculated Mercury’s total radial contraction. The revised estimate increased from the previously accepted value of 8.3 kilometers to approximately 11.6 kilometers, an increase of roughly 10 to 30 percent compared to earlier figures. Importantly, the team acknowledges that even this new number may still represent an underestimate. Because the relationship between surface roughness and hidden tectonic structures cannot be perfectly quantified, the true magnitude of Mercury’s shrinkage could be even greater than the revised calculation suggests.

The implications of this finding extend far beyond Mercury itself. The same principle—that rough surfaces can obscure tectonic records and lead scientists to underestimate planetary contraction—could apply to other rocky worlds in our solar system. The Moon, whose surface is even rougher than Mercury’s due to its long history of impacts, is a prime candidate for similar reassessment. If the lunar tectonic record has been similarly compromised by impact ejecta and rough terrain, scientists may need to revise estimates of how much the Moon has contracted as its interior cooled. The finding could also influence our understanding of other airless, cratered bodies such as Mars, whose ancient surface preserves billions of years of geological history buried under layers of impact debris.

The timing of this research is particularly significant given the upcoming BepiColombo mission, a joint endeavor between the European Space Agency and the Japan Aerospace Exploration Agency (JAXA). Currently en route to Mercury, the spacecraft carries a sophisticated instrument called the BepiColombo Laser Altimeter, or BELA, which will measure the planet’s surface topography with unprecedented precision. BELA is expected to characterize surface roughness at much finer scales than previous instruments, potentially revealing subtle relationships between younger geological events and tectonic structures that have been difficult to detect. The new findings from Hokkaido University provide a conceptual framework that will help scientists interpret the data BepiColombo sends back, ensuring that the mission’s measurements can be used to build a more complete picture of Mercury’s thermal evolution.

The study also touches on deeper questions about planetary cooling processes. All terrestrial planets—Mercury, Venus, Earth, and Mars—have been losing heat since their formation roughly 4.5 billion years ago. The rate and magnitude of that cooling, expressed through planetary contraction, provides crucial insights into a planet’s internal structure, its thermal history, and even its potential for past geological activity such as volcanism. Mercury, lacking a substantial atmosphere and plate tectonics, preserves its tectonic record with exceptional fidelity, making it an ideal laboratory for studying how rocky planets evolve. But as this study demonstrates, even the best-preserved records contain gaps, and accounting for those gaps can fundamentally change our understanding of a planet’s history.

For the scientific community, the research serves as a cautionary tale about the limitations of surface observation. When scientists estimate planetary contraction from visible tectonic features, they are essentially reading an incomplete book. Pages have been torn out by impacts, buried by ejecta, and hidden by rough terrain. The Hokkaido University team’s approach—systematically comparing terrain characteristics with tectonic feature distribution—offers a template for identifying and correcting such biases on other worlds. As missions to Mercury, the Moon, and Mars continue to return increasingly detailed data, methods like this will be essential for transforming raw observations into accurate reconstructions of planetary history.

Mercury, it turns out, has been hiding its true transformation in plain sight. The little planet that has shrunk so dramatically over billions of years has done so even more extensively than anyone realized. With future observations from BepiColombo poised to refine these measurements further, scientists are now one step closer to understanding the full story of how Mercury—and potentially other rocky worlds—cooled, contracted, and evolved into the planets we see today.

The shrinking of Mercury is not a metaphor. As the planet’s massive iron core cools and solidifies, the interior loses volume, and the overlying shell must adjust. Because rock cannot simply fold the way a drying apple’s skin does, Mercury’s crust instead responds by thrusting enormous slabs of terrain over one another, producing some of the largest and most spectacular contractional landforms in the solar system. These lobate scarps — cliff-like structures that can stretch for hundreds of kilometers and rise more than a kilometer above the surrounding plains — are the visible fingerprints of a planet compressing itself. Measurements derived from NASA’s MESSENGER mission, which orbited Mercury from 2011 to 2015, and from Mariner 10 imagery decades earlier, have allowed researchers to estimate the total reduction in the planet’s radius since the waning of the heavy bombardment at roughly five to seven kilometers, with some studies suggesting even more.

Here lies the problem the new study confronts directly. When scientists add up the offsets, or cumulative strain, recorded by all the mapped thrust faults on Mercury’s surface, the numbers they obtain tend to fall short of the contraction predicted by thermal evolution models of the planet’s interior. Conversely, some tectonic inventories imply more strain than classical frictional mechanics would seem to allow at the pressures and temperatures of Mercury’s crust. One of the great unresolved tensions in Mercury science, in other words, is that the planet’s tectonic record and its thermal history do not quite agree. Models that match the inferred radial contraction struggle to reproduce the fault population, and models tuned to the faults strain plausibility elsewhere. Something fundamental about how Mercury’s crust deforms has been missing from the equations.

The missing ingredient, the authors argue, is graphite. Mercury is unusual among the terrestrial planets in hosting a carbon-rich crust. Data from MESSENGER revealed unexpectedly high concentrations of carbon at the surface, and subsequent analyses of the planet’s large, low-reflectance deposits suggested that this carbon was delivered in the form of graphite, which floated upward as a buoyant primary crust when Mercury’s original magma ocean crystallized. Unlike most silicate minerals, graphite is highly graphitizable, mechanically weak, and famous on Earth for one property above all: it is an extraordinarily effective solid lubricant. Its layered crystal structure, in which strong covalent bonds hold carbon atoms within sheets but only weak van der Waals forces connect the sheets to one another, allows the layers to slide past each other with minimal resistance. It is the same property that made graphite useful in pencils and industrial bearings — and, according to the new work, it may have shaped the tectonics of an entire planet.

The team’s approach combined geological mapping with mechanical modeling. Using images and topographic data from MESSENGER, the researchers examined contractional structures across Mercury’s surface, including prominent scarps such as those cutting through terrains underlain by the low-reflectance, graphite-bearing material. They then performed numerical experiments on fault behavior, testing how thrust faults in Mercury’s crust would grow, slip, and accumulate strain under different assumptions about the frictional properties of the fault material. When the faults were modeled as ordinary silicate rock sliding on silicate rock, the familiar mismatch emerged: friction was simply too high to permit the observed displacements without generating consequences, such as excessive frictional heating, that the planet’s thermal record does not show. But when graphite was introduced as a fault gouge — a thin, weak layer along the sliding surface — the calculations changed dramatically.

With graphite lubricating the fault planes, the modeled thrusts could accumulate far larger offsets, over longer geological timescales, while dissipating far less heat. Frictional heating during fault slip is a well-known feedback on Earth and elsewhere: as faults move, friction converts mechanical work into heat, which weakens or strengthens the fault and alters the surrounding thermal regime. On Mercury, where the lithosphere is thick and cold and contraction has persisted for billions of years, unmitigated friction along thousands of thrust faults should have left a distinctive thermal signature in the planet’s crust and would have stalled further slip. Graphite sidesteps this limit. By lowering the coefficient of friction along the fault surfaces, the carbon-bearing layers reduce the resistance to sliding, allow faults to keep accumulating displacement, and thereby let the crust record a degree of global contraction that would otherwise be mechanically inaccessible.

The implications extend beyond reconciling two sets of numbers. If graphite-bearing horizons genuinely control how Mercury’s largest faults move, then the distribution of scarps on the planet should correlate, at least in part, with the distribution of the ancient graphite-rich crust. The low-reflectance material that blankets substantial portions of Mercury, long interpreted as excavated or exposed primordial carbon, may therefore not be a passive geological relic but an active player in the planet’s ongoing deformation. This reframes Mercury’s carbon crust from a curiosity of accretion and differentiation into a structural component of the planet’s mechanical behavior — a lubricating envelope that has shaped how the planet has cooled, contracted, and fractured over more than four billion years of solar system history.

The findings also speak to a broader principle in planetary science: that the detailed mineralogy of a crust can govern planet-scale tectonics. For decades, models of planetary thermal evolution have treated lithospheric rock as broadly similar in its frictional behavior, with composition mattering mainly for density and rheology. The Mercury results suggest that this simplification can be badly misleading. A crustal composition rich in weak, low-friction phases like graphite can permit radial contraction amounts, fault geometries, and heat budgets that a purely silicate model cannot reproduce. That insight may matter for other airless, carbon-bearing bodies as well. Asteroids and small planetary embryos rich in carbonaceous material, and perhaps the icy-rocky hybrids of the outer solar system, could likewise host deformation governed by weak phases rather than by the bulk strength of their constituent rock.

There is also a forward-looking dimension. The European Space Agency and JAXA’s BepiColombo mission, currently en route to Mercury with orbital insertion expected later this decade, carries instruments capable of mapping surface composition, topography, and magnetic and gravity fields at resolutions beyond anything MESSENGER achieved. The new results generate concrete, testable predictions: faults that cut graphite-bearing terrain should show systematically larger displacements for their lengths; low-reflectance material should be preferentially associated with the largest scarps; and the thermo-mechanical structure implied by fault populations should match models in which carbon-rich horizons localize deformation. BepiColombo’s data will provide an independent check on whether graphite’s lubricating hand is as decisive as the new study suggests.

For now, the study resolves a long-standing discrepancy with a deceptively simple physical insight: a planet’s outer shell can only shrink as fast as its faults can slide, and sliding depends on what is between the sliding surfaces. On Earth, water and clays lubricate faults; on Mercury, where liquid water has never been possible and clay minerals are absent, graphite — a residue of the planet’s fiery birth — took on that role. A world with no oceans, no atmosphere, and no plate tectonics has nevertheless found its own way to keep its crust moving, thanks to the slipperiest mineral it owns. Mercury’s shrinking, it turns out, has been quietly greased from the very beginning, and the pencil-tip mineral familiar to every schoolchild may hold one of the keys to reading the thermal history of the solar system’s innermost planet.

The work, published as an open-access article in Nature Communications, is likely to prompt a re-examination of contraction estimates not only for Mercury but for any planet where crustal composition deviates from the silicate norm. As planetary scientists refine thermal evolution models to incorporate compositional lubrication, one lesson stands out: the smallest details of mineral physics, scaled up across billions of years, can leave marks as large as kilometers of planetary radius. Mercury’s lobate scarps, towering over its cratered plains, are no longer merely monuments to a cooling core. They are the product of a delicate mechanical partnership between stone and carbon — one that the planet struck at its formation and has maintained ever since.