Published online 26 November 2008 | Nature | doi:10.1038/news.2008.1256


Minerals yield signs of early plate tectonics

Evidence of 4-billion-year-old subduction points to an early start for modern-Earth geology.

4.03 billion-year-old zirconJack Hills, Australia, the discovery site for the oldest known zircons on Earth. Inset shows a 4.03 billion-year-old zircon.B. Watson & M. Hopkins

Minerals preserved within ancient grains of rock, like flies within amber, have provided the earliest hint of plate tectonics on Earth. The minerals suggest that subduction — the process by which one tectonic plate slides beneath another — may have been active more than 4 billion years ago.

The result adds to the growing evidence that Earth was 'modern-looking'— with plate tectonics, mountains, oceans and blue skies — earlier than previously thought.

Little is known about the so-called Hadean or the 'hell-like' period of Earth's history, which spans the time from Earth's formation 4.6 billion years ago until about 3.8 billion years ago. The best geological clues of this era come from hard minerals called zircons, which have survived weathering and can be found within younger rocks.

Mark Harrison of the University of California, Los Angeles, and his colleagues previously found a so-called 'titanium thermometer' that can be used to determine the temperature at which a zircon crystal formed — the higher the temperature, the more titanium the mineral contains. They found that almost all 4-billion-year old zircons formed at about 700 ºC — much lower than the usual temperatures at which these minerals form. That suggested the zircons were saturated in water when they formed. "That was wild," says Harrison, as it indicated that Earth contained water on its surface at a time when it was once thought to be a hellish pit of molten lava.

Cool under pressure

Harrison's team has now taken a closer look1 at some 4.0–4.2 billion-year-old zircons from the Jack Hills of Australia, investigating the tiny inclusions of other minerals trapped inside them. The titanium thermometer results confirmed that these zircons, too, formed at 700 ºC. The researchers surveyed hundreds of zircons, and found that the most common mineral they contained was muscovite. Muscovite contains water, and today forms when two plates slam together and melt shale, such as where the Pacific plate sinks beneath the west coast of North America.

The team next checked the amount of aluminium present in the mineral inclusions — an indicator of the pressure at which the zircons formed. Only six of the mineral inclusions were big enough to analyze; from these, they found that the zircons must have formed at a minimum of 7,000 atmospheres of pressure — equivalent to being some 25 kilometres beneath the surface.

Given what is known about Earth's temperature at that time, it should have been about three times hotter than 700 ºC at that depth — so something must have been cooling the rocks down. "There's only one place on the planet today where you make magma and the heat is three times below average: subduction zones," says Harrison. Rock in subduction zones is cooled by ocean water.

If there was subduction, Harrison says, there may also have been a global system of plate tectonics. And that has implications for what the planet was like at that time. "The simplest view right now is that plate tectonics produces a planet that looks like it does today," he adds.

Subduction, but not as we know it?

Others agree with Harrison's conclusion that there were some cool patches of deep Earth in which these zircons formed, but point out that this doesn't necessarily mean plate tectonics existed as we know it today.


"Right now the only place that things go down is in subduction zones. But one can imagine other kinds of downwelling," says Robert Stern, a geologist at the University of Texas, Dallas. "On Venus, for example, there might be downwelling in the form of crust dripping downwards, driven by gravity rather than plate tectonics." Stern thinks that plate tectonics didn't get going on Earth until just a billion years ago, although he admits this is a minority view. Most others think that plate tectonics was probably active, in some form, as early as 3 or 4 billion years ago.

Harrison admits that his study cannot prove there were global plate tectonics so early. But he points out that the debate has moved from calling such a suggestion "impossible" a few years ago, to now calling it "inconclusive".

Many geologists are passionate about the early beginnings of plate tectonics because it seems to carry a message about conditions for life on Earth, although the exact links between the recycling of surface rock, water and evolution are unclear. "There's no question that Earth is a special planet," says Stern. "It is unique. Life, water, plate tectonics. You have to think that they're all somehow related." 

  • References

    1. Hopkins, M., Harrison, T. M. & Manning, C. E. Nature 456, 493–496 (2008).
Commenting is now closed.