Mercury lost nearly 12 miles—and its roughest ground hid the proof
In the roughest parts of Mercury, the wrinkles go missing. That’s the puzzle Gaku Nishiyama and colleagues ran into when they laid a new global map of the planet’s surface roughness over maps of its shortening features—the markers of a world that has cooled and contracted for billions of years. The most rugged terrain showed fewer of those telltale lines.
Taking that absence seriously changed the story. By accounting for shrinkage features likely buried under jumbled impact debris in those rough regions, the team argues Mercury has been contracting between 10% and 30% faster than earlier estimates suggested, and the planet has shed nearly 12 miles (19 kilometers) from its diameter since it formed.
A battered world, and a number with consequences
Mercury took shape 4.5 billion years ago. Ever since, the planet has been losing the heat of its birth and crumpling as it cools. The new analysis says the missing, debris-hidden shortening features imply a total change in diameter of up to 14.5 miles (23 kilometers), rather than the 2.5 to 10 miles (4 to 16 kilometers) that had been the working range. Within that, the team’s calculation of nearly 12 miles (19 kilometers) of shrinkage sits as a concrete tally to test models against.
“30% is a little bit surprising, but the corrected amount of contraction actually makes sense to me,” Nishiyama said. It’s not just bookkeeping: how much a rocky planet has contracted is a direct clue to how much it has cooled, and to what that cooling says about the stuff inside.
What roughness hides
On a world where craters overlap craters, fresh impacts don’t simply dig new holes; they toss blankets of debris that can bury older structures. The team’s planet-wide comparison found that the most rugged spots on Mercury have fewer wrinkles from shrinking. That mismatch is the crux: if roughness correlates with impact-jumbled terrain, then the cleanest place to count contraction is where the surface is smoother and less reworked.
The researchers used regions less affected by debris to estimate how many shortening features are likely missing elsewhere, then scaled up. That adjustment—10% to 30% more contraction—reconciles the sparse wrinkle counts in rough terrain with the physics of a uniformly cooling interior.
Inside Mercury, a record of how it formed
Why fight for a few miles? Because the interior of Mercury would reflect its formation, as Nishiyama puts it. The updated contraction gives modelers a firmer target for the planet’s cooling history, composition and starting conditions.
“The amount of contraction — an indicator of the extent of cooling — is one of the most important observables,” he said. A larger shrinkage budget narrows the combinations of interior ingredients and temperatures that can reproduce Mercury’s long, slow exhale of heat.
One map, and the study behind it
The work builds on a simple but global insight: measure roughness everywhere, then ask where the expected signs of planetary shortening drop out. The study—First Global Map of Mercury’s Surface Roughness Down to Kilometric Baselines: Implications for the Planet’s Geologic Evolution—lays out that comparison and the revised contraction estimate in detail.
It’s an exercise in seeing through the noise of bombardment to the quieter, older signal of a planet cooling. That signal, the authors argue, has been underestimated precisely where Mercury has been most battered.
The deeper cut
Bias correction, planetary-scale
If shortening features are counted directly from imagery, any process that preferentially hides them will bias the tally low. Impact gardening is exactly such a process: ballistic sedimentation and regolith overturn roughen the surface and can mantle low-relief contractional landforms. The team’s move is to use surface roughness as a proxy for the intensity of that mantling. In smoother domains, where ejecta blankets are thinner and less jumbled, wrinkle counts should be closer to complete; in rougher domains, counts are censored. A first-order correction is then to calibrate a deficit-versus-roughness relation from the smooth regions and apply it globally. That’s what yields the additional 10–30% contraction and pushes total diameter change toward the 19–23 km end of the spectrum. The logic hinges on two assumptions: that contractional features formed broadly across the lithosphere during secular cooling, and that roughness at the mapped baselines is dominated by impact processes rather than tectonic ones. The former is anchored in thermal evolution expectations; the latter is the fragile part, because any endogenous roughening (e.g., late tectonism) would confound the proxy. The authors’ argument is that the global anti-correlation—roughness up, wrinkles down—is too strong, and too widespread, to be anything but impact masking.
What we still don’t know
Nishiyama cautions that the updated figures could still be an underestimate. The present global picture leaves fine-scale structures unresolved, and some fraction of the planet’s contraction budget could remain disguised under rough terrain or simply too small to have been tallied yet.
That’s where the next spacecraft comes in. The BepiColombo mission will start conducting scans of Mercury’s surface in much higher resolution than this, offering a stricter test of the roughness–wrinkle link and the room left in Mercury’s shrinkage budget.
A planet’s past, pinned to miles
The stakes are modest in size and large in meaning: a handful of miles in diameter change, integrated over 4.5 billion years, constrain how Mercury carries heat and what it was made from at the start. They also remind us of a methodological trap on every battered world: the hardest-hit ground can be the least honest witness to deep time.
If BepiColombo’s sharper view confirms more hidden wrinkles, Mercury’s interior story tightens again. If it does not, the “corrected amount of contraction” that already makes sense to Nishiyama will still have reset the baseline for how we read a cratered planet’s past.
The paper: First Global Map of Mercury’s Surface Roughness Down to Kilometric Baselines: Implications for the Planet’s Geologic Evolution (The Planetary Science Journal, 2026)
Related research: Mercury's Tectonic and Geodynamic History: 1. Contractional Tectonic Landform Analysis and Tectonic Strain Using Machine Learning (Journal of Geophysical Research: Planets, 2026)
Sources: Mercury is shrinking faster than we thought. What's going on inside of it? (www.space.com); Mercury may have shrunk much more than we thought, new study hints (www.livescience.com)
Images: Cover: NASA/JPL; Figure 1: NASA/JPL
How this article was made: Everyman Science uses AI tools to structure, format and optimise its articles, and occasionally to produce illustrations where no free photograph exists. The reporting these articles are based on is human-produced and cited above. Spotted an error? Write to [email protected] and we will correct it. — The editors How we work.
Science desk team of Everyman Science, curating and reporting on the day’s most significant developments in research, space exploration, and technology.
