Plate Nº 39 · recorded October 10, 2026

Earth & ClimateReported finding

Earth's Day Drifts by Milliseconds. The Culprit May Be the Inner Core.

A Nature study by Huifeng Zhang and Mathieu Dumberry ties multi-decade shifts in Earth's day length to gravitational torque from the inner core, which they say deforms viscously over about 10 years.

By Marcus Bennett3 min read636 words

In brief

  1. Study published Sept. 23, 2026 in Nature by University of Alberta's Huifeng Zhang and Mathieu Dumberry
  2. Earth's day length varies by a few milliseconds over multi-decadal timescales
  3. Liquid outer core and roughly 3,000-km-thick mantle exchange angular momentum in opposite directions
  4. Inner core appears to deform viscously on a timescale of about 10 years
  5. Day-length drift is explained by a balance between inner-core gravitational torque and opposing core-mantle boundary torque
Earth’s day isn’t exactly 24 hours. The reason may lie deep inside the planet
Plate Nº 39Earth’s day isn’t exactly 24 hours. The reason may lie deep inside the planet — AI-generated

Earth's day length drifts by a few milliseconds over multi-decadal timescales — and a study published September 23, 2026 in Nature now ties that drift to gravitational tugs from the planet's solid inner core.

The research, led by University of Alberta PhD student Huifeng Zhang and professor Mathieu Dumberry, identifies gravitational torque — a twisting force created when unevenly shaped masses pull on each other across distance — as the missing link between Earth's deepest interior and the rotation we measure at the surface.

What exactly is changing?

Earth's rotation does not stay perfectly constant. Over multi-decade timescales, some days grow a fraction of a millisecond longer, others slightly shorter.

These shifts, called "decadal variations in length of day," are too small for any clock a person would notice. Atomic clocks and satellite measurements track them routinely.

For about 30 years, scientists have known part of the story. Earth's liquid outer core — the molten iron-nickel layer surrounding the solid inner core — sometimes speeds up over several decades and then slows again. The roughly 3,000-kilometer-thick mantle responds in the opposite direction: when the liquid core accelerates, the mantle slightly decelerates, and vice versa.

This exchange preserves Earth's total angular momentum — the product of a body's rotational inertia and its spin rate, which must stay constant in the absence of external torques.

Why was the mechanism unclear?

The puzzle has always been how the liquid core and the rocky mantle exchange that rotational momentum. The two layers meet at a sharp boundary, and the friction or electromagnetic drag there seemed too weak to do the job alone.

Zhang and Dumberry argue that the inner core — the solid sphere at the very center of the planet — supplies a previously overlooked lever.

Because the inner core is not a perfect sphere, small changes in how it rotates generate faint gravitational pulls on mass variations inside the mantle. In effect, the inner core twists the mantle the way a slightly lopsided gear would twist a neighboring wheel.

What opposes the twist?

A second force works against the gravitational pull. Known as "core-mantle boundary torque," it produces friction and electromagnetic resistance at the interface between the liquid core and the solid mantle.

According to the researchers, decadal shifts in day length arise from a delicate balance between the gravitational torque from the inner core and this opposing core-mantle boundary torque. When one grows, the other pulls back, limiting how much any single day's length can change.

What does this tell us about the deep Earth?

Perhaps the most unexpected implication is that the inner core behaves less like rigid rock and more like a slow-flowing solid.

Zhang and Dumberry say the inner core appears to "deform viscously" — meaning it changes shape gradually under stress — on a timescale of about 10 years.

That finding challenges a popular image of the inner core as a permanently frozen ball. Instead, the deepest reaches of the planet may respond to surrounding forces far more dynamically than the word "solid" usually suggests.

What are the limits of this work?

The model is theoretical. It builds on observational data for decadal day-length variations and core rotation rates, but researchers cannot yet directly measure the inner core's own motion at such depths.

The authors also note that other contributors — such as shifts in atmospheric circulation and ocean currents — can affect day length on shorter timescales. The gravitational-torque mechanism they describe targets the multi-decadal pattern specifically.

The study does not predict future changes in day length. Its goal is to show that a roughly 30-year-old question now has a plausible physical answer — one that ties the planet's rotation to a feature buried about 5,000 kilometers beneath our feet.

via ualberta.ca (Original)

Filed under

  • inner-core
  • earth-rotation
  • day-length
  • geophysics
  • core-mantle-boundary
Share this article:

More from Marcus Bennett

Marcus Bennett

Show full bio

News editor covering marketplaces and e-commerce at SciBeat.

221 articles

Nearby plates

« Previous articleNext article »