Plate tectonics: What set the Earth's plates in motion?

What set the Earth's plates in motion?
A snapshot from the film after 45 million years of spreading. The pink is the region where the mantle underneath the early continent has melted, facilitating its spreading, and the initiation of the plate tectonic process. Credit: Patrice Rey, Nicolas Flament and Nicolas Coltice.

The mystery of what kick-started the motion of our earth's massive tectonic plates across its surface has been explained by researchers at the University of Sydney.

"Earth is the only planet in our solar system where the process of occurs," said Professor Patrice Rey, from the University of Sydney's School of Geosciences.

"The suggests that until three billion years ago the 's crust was immobile so what sparked this unique phenomenon has fascinated geoscientists for decades. We suggest it was triggered by the spreading of early continents then eventually became a self-sustaining process."

Professor Rey is lead author of an article on the findings published in Nature on Wednesday, 17 September.

The other authors on the paper are Nicolas Flament, also from the School of Geosciences and Nicolas Coltice, from the University of Lyon.

There are eight major tectonic plates that move above the earth's mantle at rates up to 150 millimetres every year.

In simple terms the process involves plates being dragged into the mantle at certain points and moving away from each other at others, in what has been dubbed 'the conveyor belt'.

Plate tectonics depends on the inverse relationship between density of rocks and temperature.

At mid-oceanic ridges, rocks are hot and their density is low, making them buoyant or more able to float. As they move away from those ridges they cool down and their density increases until, where they become denser than the underlying hot mantle, they sink and are 'dragged' under.

The movie tells an 87 million year long story. It shows an early buoyant continent (made of a residual mantle in green and continental crust in red) slowly spreading toward the adjacent immobile plate (blue). After 45 million years, a short-lived subduction zone, where the plate goes under, develops. This allows the continent to surge toward the ocean, leading to the detachment of a continental block, the starting step in the movement of the continental plates or plate tectonics. Credit: Patrice Rey, Nicolas Flament and Nicolas Coltice

But three to four billion years ago, the earth's interior was hotter, volcanic activity was more prominent and did not become cold and dense enough to spontaneously sank.

"So the driving engine for plate tectonics didn't exist," said Professor Rey said.

"Instead, thick and buoyant early continents erupted in the middle of immobile plates. Our modelling shows that these early continents could have placed major stress on the surrounding plates. Because they were buoyant they spread horizontally, forcing adjacent plates to be pushed under at their edges."

"This spreading of the early continents could have produced intermittent episodes of plate tectonics until, as the earth's interior cooled and its crust and plate mantle became heavier, plate tectonics became a self-sustaining process which has never ceased and has shaped the face of our modern planet."

The new model also makes a number of predictions explaining features that have long puzzled the geoscience community.

Explore further

Faraway moon mimics Earth tectonics

More information: Spreading continents kick-started plate tectonics, Nature,
Journal information: Nature

Citation: Plate tectonics: What set the Earth's plates in motion? (2014, September 17) retrieved 15 July 2019 from
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.

Feedback to editors

User comments

Sep 17, 2014
This comment has been removed by a moderator.

Sep 18, 2014
Ha! And I complained the other day that the comparison between Jack Hill and Iceland zircons had killed Iceland as a prototype of the first continents and left us with a mystery.

I had forgotten this promising line of research (as well as the more mineral oriented "grinding finer, more malleable grains" starting mechanism). This is nicely in line with the zircon comparison result of a wet, cold Hadean/Archean Earth of ocean plates.

Of course, it still leaves the Icelandite-TTG-granite series unexplained, as well as the origin of the first continental crust.

@Aligo: Yes, convection, but not much of melt: "tectonic plates did not become cold and dense enough to spontaneously sank."

But as I describe above, these types of models leave much unexplained and (it seems to this layman) couple less with the mineralogy, plus constrains the timeline less, despite being consistent with what Hadean/Archean material we have.

Sep 18, 2014
Oops Even if this model doesn't take us back to the origin of the first continental crust , it _does_ couple with the geology (and so mineralogy). From the abstract: "Our model predicts the co-occurrence of deep to progressively shallower mafic volcanics and arc magmatism within continents in a self-consistent geodynamic framework, explaining the enigmatic multimodal volcanism and tectonic record of Archaean cratons6."

And after thinking about it, maybe they can constrain the timeline of plate tectonics, if not the appearance of first continental crust. I have to read the paper of course.

Conquer and divide has always been a good method in science. So I would be satisfied with a smaller mystery (of first continental crust and the Icelandite-TTG-granite series).

Sep 18, 2014
I think there's still some mystery in how the silica rich granite-type continental crust formed and why it is not found on other terrestrial planets. The reason why the iron rich oceanic plates slide under the continental plates lies within the inherent, temperature-independet density differences between those two main type rocks. If the minerals were uniformly distributed within the plates, plate tectonics might not exist and certainly there would be much less less variation in topology. What's more, if that were the case, practically our whole planet would be covered with oceans, and without massive continents, we obviously wouldn't be typing here. So besides geological history, we're basically talking about our own genesis as well.

I myself am quite fond of Minik Rosing's hypothesis that states that cyanobacterial activity helped give rise to continents due to the acceleration they gave to the dehydration reactions and diagenetic alteration that results in granitic melts.

Sep 19, 2014
@Rute: Interesting reference, thanks! Not very likely IMO, since cyanobacteria evolution can be stretched back to 3.0 billion years ago according to the first massive phylogeny on them ... but it's a stretch.

I think the granitoid-granite series is well understood, and granites has been found on Mars under similar conditions. (Curiosity in Gale crater IIRC; crustal rocks experiencing repeated wets and melts akin to plate tectonics processing.)

But we can agree on that this removes the constraints on the origin of first continental crust so we have to look elsewhere than start of plate tectonics.

Plate tectonics means rapid (kerogen) reprocessing of mainly carbon instead of slow (sedimentation + weathering), amps up biosphere productivity (NPP) something fiercely. Evolving language users would be rare anyway, like evolving the elephant trunk, according to most biologists. (Each evolved once.) I don't see plate tectonics as decisive as much as one enabler of many.

Sep 19, 2014
I found the beef on Rosing. He is very productive [ http://research.k...p;page=0 ] and one of my sources on how the faint Hadean/Archean Sun problem is solved.

So he has formed this whole model of Archean Earth. Already 2003 he put down his foot on that cyanobacteria was present already 3.7 billion years ago. [ ] But Roger Buick, associate professor of astrobiology at the University of Washington in Seattle, US, was cautious about the findings. "Anything of that sort of age has to be somewhat dubious. Those rocks have been put through the geological mill many times - it would be hard to say that anything you're seeing is primary," he said.

I can't find the plate tectonics references, so would appreciate a named reference.


Sep 19, 2014

My own reference for cyanobacteria was "Evolution of multicellularity coincided with increased diversification of cyanobacteria and the Great Oxidation Event", Schirrmeister et al, PNAS EE, 2012. As it happens I remembered _wrong_, that was the oldest dating node, my apologies. They can't exclude cyanobacteria evolved already 3.5 billion years ago (see their Fig. 1).*

So maybe.

* I remember now, my reason to doubt was that microanalysis of 3.8 Gyr Isua BIFs shows anoxic but oxidative photosynthesis. That would be a good test for late oxygen producing cyanobacteria, more connected to the GOA at 2.5 billion years than early BIFs.

But recently the BIFs have been removed to more plausibly abiotic origins around vents; note though the microanalysis and this notion both must obey Buick's concern. And there are plausibly biotic oxygen production earlier than GOE, depending on the still unclear sulfur record. As always, YMMV, but early cyanobacteria just maybe works.

Sep 19, 2014
Oops. Somehow I forgot to put quote marks around the Buick quote. It's from the linked article.

Sep 19, 2014
That is all very interesting information, thanks. I didn't know that granite had been found on Mars, just checked that out, very fascinating. My knowledge of geology is very limited so I'm always eager to learn something new, and actually I'm not a scientist at all, just a student at uni of applied sciences. But I do have some interests in specific geological topics, especially at the interface between biology and geology.

It seems to be that the dating for the beginning of oxidative photosynthesis divides opinions quite a lot, with the opposing camps over a billion years away from each other. I guess one reason for that is that the oxygen started accumulating on Earth in substantial degree only about 2,3 billion years ago as most of it was wasted on oxidation reactions, especially those producing iron oxides. Perhaps the snowball phases on Earth also contribute to the difficulty in deciphering ancient clues.

Sep 19, 2014
As for the reference for Rosing's hypothesis on continent formation, which you were apparently interested in, google "The rise of continents—An essay on the geologic consequences of photosynthesis"

Sep 20, 2014
Thanks, got it! Goes on my read list, I was swinging by the uni library today anyway.

That also netted this, , which reminded me that without life continental crust would be meager, at 3 % rather than 30+ (40) % of surface. It isn't enough to clinch the case that life started plate tectonics, but suggestive.

I'm a layman interested in astrobiology. (Depending on your definition, I am a "scientist" with a PhD and published papers. Not active in research and nothing in biology, mind...)

Sep 20, 2014
Nice, I found a very interesting paper from the references in the article you linked:

Too bad the subscriptions to scientific publications are pretty limited in my school, I might not be able to read that for free.

Speaking of which, there was actually a straight link to the Rosing et al. paper:

Google 'destroys' straight links to pdf files but I found out now that once you access the cached html version of a pdf file, you'll find the original URL.

Sep 22, 2014
Rute, thanks a bunch!

[Yes, paywalls... Anything in Geology (usually Hazen's papers) are not in the general uni license here in Uppsala. :-/]

Sep 22, 2014
The geological record suggests that until three billion years ago the earth's crust was immobile

Which suggest that the continents were together in pangea first. With the recently found "oceans" of water below the crust, one could postulate that it was the buildup of pressure that caused the initial fractures that now are the fault boundaries of our tectonic theory.

Sep 23, 2014
Pangaea existed 300 million years ago. 3 *billion* years ago there was very little continental crust; Butler's model (2011) suggests that it amounted to somewhere in the ballpark of 25% compared to current continental crust expanse.

Sep 23, 2014
How do we know that pangea was only 300 million years ago? If it's by estimating with current drift rates, then it's nothing but a guess.

Sep 23, 2014
How do we know that pangea was only 300 million years ago? If it's by estimating with current drift rates, then it's nothing but a guess.

The ancient drift rates leave multiple lines of evidence, such as the strips of magnetic rocks expanding from the the mid-ocean ridge whose magnetic minerals orient themselves as north-south or south-north while solidifying, according to the polar reversals.


Sep 25, 2014
"What set the Earth's plates in motion?" Well, obviously God did. God reached down with His mighty, all-noodly appendage and pushed the tectonic plates around -- and continues to do so to this very day.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more