Ancient Sea Levels Reveal Earth Tipped on Its Axis at Least Four Times in 320 Million Years

The Earth's geographic poles do not always stay where they are. A region that once sat near the North Pole can end up close to the equator, and the reverse can happen as well. This phenomenon is called true polar wander (TPW), and it occurs when the planet's rocky crust and mantle shift together relative to the spin axis. The core and the climate belts stay tied to that axis, but the solid outer shell can rotate around it.

For decades, many geoscientists treated true polar wander as either negligible or persistently slow, a background process too sluggish to reshape the planet in a human lifetime, or even across a few million years. Others argued that it might be episodic, with short bursts of rapid motion separated by long quiet intervals. The difficulty was that normal plate tectonics can produce signals that look like polar wander, and the paleomagnetic record alone has not been able to settle the question. Previous studies returned conflicting interpretations.

Now a study published in the journal Science on October 1, 2026 offers independent evidence from an unexpected direction: ancient sea levels. A small team led by Mathew Domeier, a geoscientist at the University of Oslo's Centre for Planetary Habitability in Norway, analyzed global continental flooding records spanning the last 320 million years. Their conclusion, reported in the paper with DOI 10.1126/science.aec8423, is that the solid Earth tipped relative to its spin axis in at least four rapid true polar wander episodes, with the strongest signals falling between 150 million and 140 million years ago and between 100 million and 90 million years ago.

The work matters because the oceans respond to a polar wander event far faster than the solid Earth can change its shape. When the crust and mantle reorient, relative sea levels rise in two opposite regions of the globe and fall in two others. That produces a distinctive quadrupolar pattern of highs and lows. Domeier's team searched for exactly that fingerprint in the flooding record, and they found it more than once.

Key Facts

The study used digital reconstruction maps of global continental flooding covering the last 320 million years, divided into 10-million-year snapshots. The researchers compared consecutive maps to track where seas advanced onto the continents and where they retreated. They then used computer models to test whether the observed changes matched the quadrupolar pattern expected from a rapid true polar wander event.

The analysis identified four possible episodes of rapid polar wander. Two of them had much stronger evidence. The first was between 200 million and 190 million years ago, during the Early Jurassic. The second was between 150 million and 140 million years ago, near the Late Jurassic to Early Cretaceous boundary. The third was between 100 million and 90 million years ago, in the mid-Cretaceous. The fourth was between 30 million and 20 million years ago, spanning the Oligocene to Miocene. The authors wrote that their inversions corroborate rapid TPW events in the mid-Cretaceous and near the Late Jurassic-Early Cretaceous boundary, and that similar episodes likely punctuated even earlier intervals of Earth history.

The findings cut against two other ideas. The team found little evidence for rapid TPW during most of the Cenozoic, the era that covers the last 66 million years. They also found no significant signal supporting the proposal that the rapid northward movement of the supercontinent Pangea was driven primarily by true polar wander during the late Carboniferous and Permian. That alternative explanation, which has been discussed for years, is not backed by the new sea-level reconstruction.

True polar wander is not only a thing of the distant past. Scientific American reported on October 2 that the process is still ongoing today, at a rate of about 10 centimeters per year. That figure exceeds the mean rate of differential plate motion. In the same report, Domeier said: 'Based on satellite measurements, we know that this is happening today, at about 10 centimeters a year, mainly because of melting ice caps.'

The authors state plainly that their results refute the view of TPW as negligible or persistently slow. In the paper, they write that the findings highlight the need to consider TPW as an episodic control on sea level change and likely other global environmental and biological dynamics. Because Earth is not a perfect sphere, its rotation axis is naturally balanced by the distribution of mass. When continental drift and mantle processes redistribute that mass, the balance can be disturbed, and the solid Earth reorients relative to its spin axis.

Analysis

The bigger picture here is that the solid Earth is not a stable platform. It is a body that can and does roll over relative to its own spin axis, and it has done so repeatedly during the time of the dinosaurs and beyond. That is a different mental model from the one most geology textbooks present, where the poles are essentially fixed and the continents do all the moving. The new study does not overturn plate tectonics, but it adds a second, globe-spanning motion on top of it.

The strength of the argument rests on the independence of the evidence. Paleomagnetic data have long been the primary tool for detecting true polar wander, but they are vulnerable to ambiguity: tectonic motion can mimic the TPW signature, which is why earlier studies disagreed. Sea level is a different physical system. Phys.org reported on October 2 that the oceans respond immediately when Earth shifts during a polar wander event, leaving traceable signatures in ancient flooding records. If both the magnetic record and the flooding record point to the same rapid episodes, the case becomes much harder to dismiss.

The Debrief reported on October 2 that the University of Oslo team found multiple robust quadrupolar patterns since 320 million years ago, with four rapid TPW signals that include separate events in the mid-Cretaceous and the Late Jurassic-Early Cretaceous. The researchers did not simply pick out any change in sea level; they looked for a specific geometric pattern that only a reorientation of the whole solid Earth should produce. The presence of that pattern in four separate intervals is the core of the claim.

There are still open questions. The two weaker episodes, in the Early Jurassic and the Oligocene to Miocene, have less support than the two strong ones. And a 10-million-year sampling interval is coarse: it can show that a rapid event happened within a window, but it cannot pin down its duration. The American Association for the Advancement of Science reported on October 1 that the findings corroborate earlier paleomagnetic and plate-motion studies suggesting rapid TPW, which is a helpful convergence, but the precise tempo of each episode remains a target for future work.

Why It Matters

If the solid Earth can tip by large amounts in geologically short times, then the consequences ripple through the entire Earth system. The crust and mantle move relative to the spin axis, but the core and the climate belts stay put. That means a continent can be carried from a polar latitude toward the equator, or the reverse, without any change in the planet's overall rotation. The resulting shifts in ocean basins, sea level, and regional climate would be enormous.

The authors suggest that rapid TPW could act as an episodic control on sea level change and likely on other global environmental and biological dynamics. That is a broad claim, and it opens the door to reexamining extinction events, ocean chemistry shifts, and magnetic field behavior during the Jurassic and Cretaceous. The study also notes that large or rapid episodes could produce major changes in climate, the biosphere, and Earth's magnetic field.

Closer to the present, the confirmation that true polar wander is happening now at about 10 centimeters per year, mainly because of melting ice caps, ties the deep past to an ongoing, measurable process. It is a reminder that the redistribution of mass on Earth's surface has consequences for the planet's orientation in space.

Next Up

The immediate next step for other researchers is to test the four episodes against independent records. Higher-resolution flooding maps, better paleomagnetic data, and tighter age constraints on the two weaker intervals could either strengthen or weaken the case. The team's own suggestion that similar fast TPW episodes likely punctuated earlier intervals of Earth history, before 320 million years ago, is an invitation to push the method deeper into the past.

The study also leaves the Cenozoic mostly quiet on the rapid TPW front, with little evidence for fast events during most of that era. That contrast, between a busy Jurassic and Cretaceous and a calmer recent past, is itself a puzzle worth pursuing. For now, the paper published in Science on October 1, 2026 stands as a strong, independent line of evidence that the planet's solid shell has rolled over at least four times in the last 320 million years.