How Can An Earthquake Shorten The Day? The Physics Of A Spinning Planet

Table of Contents (click to expand)
Yes: a magnitude 9.1 earthquake, like the one off Japan in 2011, drops a huge slab of rock a little closer to the line the planet spins around, so Earth turns a hair faster, the way a skater speeds up by pulling their arms in. NASA scientist Richard Gross put the change at about 1.4 millionths of a second per day, far too small to feel or even measure directly. The quake also nudged the point Earth's mass balances on by about 13.7 centimeters (5.4 inches), but the north pole and the seasons stayed exactly where they were.

Here is a fact that stops people cold. On March 11, 2011, a giant earthquake struck off the coast of Japan. It moved the island. It shifted a point deep inside the planet. And it made the day a tiny bit shorter.

That last part sounds made up. An earthquake changing the clock? It reads like a tabloid headline. Yet it is real, and you can work out why from one idea you already understand.

The trouble is that the popular version gets three things slightly wrong. It quotes the wrong number. It says the axis "tilted," which is not what happened. And it mixes up what was measured with what was only calculated. So let us build the physics first, then check the story piece by piece.

Why Does A Spinning Object Speed Up When It Pulls Itself In?

Picture a figure skater in a spin. Arms out wide, they turn at a steady pace. Then they pull their arms in tight against their body. At once they whip around faster, with no push from anyone. You have seen this a hundred times, and it looks a little like magic.

It is not magic. It is a rule called conservation of angular momentum. Think of spinning motion as a bucket of "turning." Once something is spinning and nothing outside is shoving it, the amount of turning in the bucket cannot change. It is locked in.

Turning depends on two things. One is how fast you spin. The other is how spread out your mass is. That second thing has a name: moment of inertia. Spread your mass out wide and it is large. Pull your mass in close and it drops.

Here is the trade. The bucket of turning is fixed. So pull your mass in, and your moment of inertia shrinks. Your spin speed then has to climb to keep the total the same. Arms in, spin up. Arms out, spin down. A tightrope walker's long pole uses the same idea, and so do your own arms when you wobble and try to stay balanced.

Hold on to that one sentence: pull the mass in, and the spin speeds up. The whole article rests on it.

Arms in, and the skater whips around faster. Earth does the same trick, with rock instead of arms. (AI-generated image)
Arms in, and the skater whips around faster. Earth does the same trick, with rock instead of arms. (AI-generated image)

Is The Earth A Figure Skater?

The Earth spins once a day, and nothing outside is pushing it around. That is exactly the skater on the ice. So the same rule applies. Move mass on Earth closer to the spin line, and the planet must turn a touch faster.

Now, what could shift that much mass? A big earthquake can.

Most giant quakes happen where one slab of the planet's crust dives under another. Geologists call this a subduction zone. The 2011 Japan quake was this kind, a "megathrust." A slab of crust the size of a country slipped along the fault. It settled lower, and the mass packed in closer to the center. A vast amount of rock ended up a little nearer the line Earth spins around.

You already know what has to happen next. Mass moved inward. The planet's moment of inertia dropped by a hair. The bucket of turning is fixed. So Earth's spin sped up by a hair. You could have called the direction of the effect yourself, before anyone quoted a single number. That is the nice part: the physics comes first, and the measurement only confirms it.

In a megathrust quake, one slab of crust slips under another and settles lower. Mass moves inward, and the planet spins a touch faster.
In a megathrust quake, one slab of crust slips under another and settles lower. Mass moves inward, and the planet spins a touch faster.

How Much Did The 2011 Earthquake Shorten The Day?

Now the number. And this is where almost everyone gets it wrong.

Within days of the quake, Richard Gross at NASA's Jet Propulsion Laboratory ran the math. His first pass said the day should have grown shorter by about 1.8 millionths of a second. That figure raced around the world. You can still find it everywhere today.

But it was a first pass. Later, using a better model of the fault from the U.S. Geological Survey, Gross redid the sum and got about 1.4 millionths of a second. That is the current figure. Nearly every news story still quotes the old one.

A millionth of a second is hard to picture. A single second holds a million of them. Your fastest reaction, about a quarter of a second, holds nearly two hundred thousand. The quake changed the day by 1.4 of those. It is real, and it is almost nothing.

One more thing matters, and Gross said it plainly. This number is a calculation, not a reading off a dial. The change is far smaller than the wobble in Earth's spin from wind and ocean. Gross searched the actual rotation records for the quake's fingerprint and could not find it. It was buried in the noise. The physics says the day got shorter. The math says by how much. No clock on Earth is sharp enough to catch it in the act.

Did The 2011 Earthquake Really Shift Earth's Axis?

The other viral claim is that the quake "tilted Earth's axis." This is the part that gets mangled the most, because Earth has two different axes, and people mix them up.

The first is the rotation axis. That is the line the planet actually spins around, the one that points near the North Star. It did not move in space. The seasons come from that axis, so the seasons did not change. Your compass still points the same way. Sunrise still happens on time.

The second is the figure axis. That is the line the planet's mass balances around, like the balance point of a slightly lumpy top. It is not the same as the spin line. The two sit about 10 meters (33 feet) apart at the pole. This balance point never sits still anyway. It traces a slow circle a few meters wide about every 14 months, a natural sway called the Chandler wobble. The 2011 quake shifted the balance point by about 13.7 centimeters (5.4 inches), toward 132 degrees east longitude. Roughly the width of your hand.

That is the whole "axis" story. A balance point buried in the physics moved by a hand's width. The pole you picture, the one at the top of every globe, did not budge.

Two axes people mix up. The quake nudged the red one (the balance point), not the black one (the spin line). The gap is drawn far wider than real life.
Two axes people mix up. The quake nudged the red one (the balance point), not the black one (the spin line). The gap is drawn far wider than real life.

What Did We Actually Measure? Japan Moved East

So the rotation figures are calculated. Was anything measured, though? Yes, and it is the strongest part of the whole story.

Japan sits on a dense grid of GPS stations. When the quake hit, they all lurched at once. The Geological Society reports that the northeast coast of Japan slid east by up to 4 meters (13 feet). That is not a model. That is thousands of receivers recording the ground moving under their feet. The popular "8 feet" figure is a fair, if modest, round number for it.

Out at sea it was far larger. Right above the break, instruments on the seafloor recorded the ground jumping about 24 meters (79 feet) sideways. The two crustal slabs slid past each other by roughly 50 meters. That is the raw mass shift that fed the tiny change in spin.

So the viral bundle mixes two different things. The ground movement was watched, live, by satellites. The shorter day and the shifted balance point were worked out on paper afterward. Both are true. But they are not the same.

A GNSS ground station like the ones across Japan. Thousands of them caught the coast jumping east in real time. (AI-generated image)
A GNSS ground station like the ones across Japan. Thousands of them caught the coast jumping east in real time. (AI-generated image)

Is A Day On Earth Getting Longer?

Here is the reframe, and it is the best part. If a magnitude 9 quake barely moves the needle, what actually changes the length of the day? Almost everything else, and by a lot more.

Start with the Moon. The Moon's gravity drags the oceans into tidal bulges, and that drag acts like a brake on the spinning Earth. Scientific American notes this adds about 1.7 milliseconds to the day every century. In the same trade-off as the skater, the Moon slowly drifts away as it steals that spin, by about 3.8 centimeters a year. Over a single century the Moon changes the day roughly a thousand times more than the 2011 quake did in an instant.

Then there is the weather. Winds and shifting water pile up and thin out across the year. Per the EarthScope Consortium, seasonal wind patterns drive around 90 percent of the year-to-year swing in day length. That swing runs close to a full millisecond, hundreds of times bigger than any single quake. And it happens every year, back and forth, as the seasons turn.

The chart below stacks them up. Notice where the earthquakes land.

The Moon and the weather move the day's length far more than the biggest quakes. The bars are on a log scale, so each gridline is ten times the last.
The Moon and the weather move the day's length far more than the biggest quakes. The bars are on a log scale, so each gridline is ten times the last.

Did The 2004 Earthquake Shift The Axis Too?

It did, and by more. The 2011 quake was not even the biggest recent case.

Gross ran the same math for other giants. Take the 2004 Sumatra quake, the one behind the Indian Ocean tsunami. It shortened the day by about 6.8 millionths of a second. It also shifted the balance point by about 7 centimeters (2.76 inches). The 2010 Chile quake trimmed the day by about 1.26 millionths of a second. Bigger quakes, closer to the equator, tend to have a larger effect, but all of them are still tiny.

The 2004 quake holds another record worth knowing. It ripped along its fault for around 10 minutes, over a rupture more than 1,200 kilometers long. Seismologists count it as the longest earthquake rupture ever recorded on instruments. Most quakes are over in seconds. That one kept going long enough to boil a kettle.

How Do You Even Measure A Millionth Of A Second Of Spin?

If no clock on Earth can catch the change directly, how do we track the day's length at all? With some of the most precise instruments ever built.

The main trick is called Very Long Baseline Interferometry, or VLBI. Radio dishes scattered across the planet all lock onto the same distant quasar, a bright beacon billions of light-years away. As Earth turns, the signal reaches each dish at a slightly different instant. Compare those instants and you can read the planet's spin to a sliver of a millisecond. Networks of GPS satellites and laser-ranging stations pitch in too.

A body called the International Earth Rotation Service gathers it all. It keeps the world's official time in step with the wobbly real Earth. When the two drift apart, it adds a "leap second." So the length of the day is not a fixed thing we assume. It is a number a whole global system measures, over and over, to a precision that would have seemed absurd a century ago.

So, Can An Earthquake Really Shorten The Day?

Yes. And now you know exactly what that sentence means, and what it does not.

A giant quake drops a slab of crust closer to the spin line. The planet's mass pulls in, its moment of inertia falls, and, just like the skater, it turns a hair faster. For 2011, the best figure is about 1.4 millionths of a second per day, not the 1.8 the headlines still repeat. The number is a careful calculation, because the change is too small for any clock to catch in the act.

The "axis" moved too, but not the one you picture. The balance point deep in the planet slid about a hand's width. The pole, the seasons, and your compass never noticed. What we did watch, in real time, was Japan's coast lurching several meters east.

And the honest punchline is the humbling one. Of everything that tugs on the length of a day, a magnitude 9 earthquake is the smallest item on the list. The Moon does far more. So does next season's weather. The Earth is not a clock ticking out perfect days. It is a lumpy, bendy spinning rock. And we have learned to measure its every twitch to a millionth of a second.

References (click to expand)
  1. Japan Quake May Have Shortened Earth Days, Moved Axis — NASA JPL
  2. Japan earthquake shortened Earth's day by 1.4 millionths of a second (interview with Richard Gross) — EarthSky
  3. On This Day: 2011 Tohoku Earthquake and Tsunami — NOAA NCEI
  4. Tohoku Earthquake, Japan — The Geological Society
  5. Displacement Above the Hypocenter of the 2011 Tohoku-Oki Earthquake — Science (Sato et al., 2011)
  6. Fact or Fiction: The Days (and Nights) Are Getting Longer — Scientific American
  7. A Day is Not Always 24 Hours: How Earth's Shifting Systems Cause Day Length Variation — EarthScope Consortium
  8. Extent, duration and speed of the 2004 Sumatra–Andaman earthquake imaged by the Hi-Net array — Nature (Ishii et al., 2005)
  9. Chandler Wobble — geodesy.science (International Association of Geodesy)

How this article was made. It was researched from the sources cited above and drafted with the help of AI, then fact-checked, edited and approved by Abhishek Jain before publication. Illustrations that are not credited to a photographer are generated diagrams or illustrations, not photographs.