How old is plate tectonics?

Plate tectonics is the process that underpins much of our understanding of the Earth. It explains manymany aspects of the Earth, from magnetic patterns in oceanic rocks to the distribution of plants and animals. How unusual is it? Well, it doesn’t seen to be happening on other rocky planets in our solar system. Many geologists have argued that plate tectonics wasn’t active during the earth’s early history. As astronomers find many rocky planets in other solar systems, the question of understanding how ‘typical’ plate tectonics has implications beyond the earth. How long has it been going on – how old is it?

The Precambrian, is the span of earth’s history before the Cambrian. When geological periods were first defined, by largely-British geologists in the Nineteenth century, they were distinguished by the fossils they contained. Precambrian fossils are rare (hard shells only evolved in the Cambrian) and not until the Twentieth century could we calculate the absolute age of rocks. Precambrian rocks in Britain are fairly uncommon, mostly restricted to the Highlands of Scotland, so lumping them into one group made sense at the time.

Geological timescale clock from Wikipedia

Unfortunately, it turns out that the Precambrian covers the vast majority of earth’s history. Now that we can get an absolute age for many rocks, it’s possible to divide the Precambrian into smaller chunks. The next level of division consists of the (increasingly old) Proterozoic (“earlier life”), Archaean (“beginning”) and Hadean (“hellish”).

Precambrian rocks are often very different to modern ones. As well as the atmosphere being very different, the earth itself was hotter (younger radioactive isotopes give off more heat). Rocks like komatiite lavas which erupted at 1600 °C, far hotter than modern basaltic lava, suggest that mantle temperatures were then much higher. For this and other reasons, its often been assumed that plate tectonics was not active in earth’s early history.

A paper by Peter Cawood and two other Oz-based scientists called “Precambrian plate tectonics: criteria and evidence” (free!) addresses this question in a systematic way.  First they contrast plate tectonics and ‘plume tectonics’ as two different ways of transferring heat out of the earth. Both are active now (e.g. Hawaiian plume is a hotspot) but plate tectonics is dominant.

How to distinguish the two? One key difference, Cawood argues, is that plate tectonics involves “the differential horizontal motion of plates”. Plate tectonics is all about chunks of crust wandering about the place, so evidence of this is significant. How to track the ancient movements of the plates? Palaeomagnetology, or palaeomagic as it is jokingly referred to, is the study of earth’s ancient magnetic field. As magnetic minerals form, or cool down, they fix an impression of earth’s magnetic field within them. Heating samples in the lab allows us to measure the orientation of this fossil magnetic field or ‘palaeopole’. One way in which these palaeopoles can be useful is to tell you the latitude of the sample at that time. Plotting palaeopoles from different areas of Precambrian rocks at different times, Cawood demonstrate that they change latitude over time, both absolutely and relative to each other. If continents are drifting, then plate tectonics is responsible.

Greenstone Xenoliths in Archean Gneiss, near Sand River, Ontario

Greenstone Xenoliths in Archean Gneiss, near Sand River, Ontario, by Ron Schott:

Archean rocks often consist of distinctive granite-greenstone terranes that are not linked obviously to plate tectonic processes. Cawood lists evidence that while it may not be obvious, the link is there – in particular distinctive features such as ophiolites (slices of sea-floor on continents) and eclogites (very deeply buried metamorphic rocks) are increasingly being identified in very old rocks. Evidence from geochemistry and metal deposits is also brought to bare to argue that plate tectonics was active for most of the Precambrian and may have been active from the dawn of earth’s history. Precambrian rocks are distinctive, but the fundamental mechanism that drives the modern earth affected them too.

This paper is a great summary, but is not the final word (of course). Other scientists argue that plate tectonics wasn’t active and other processes were dominant. For example one groupuse numerical modelling and emphasise the importance of mantle temperature. If the mantle is too hot, then the lithosphere is weakened by melt and so not rigid enough to move as plate. An intermediate stage towards modern plate tectonics involves shallow underthrusting of oceanic lithosphere under continents. A very recent paper involving physical modelling of Archean crust provides an overview of alternative views. The paper focuses on explaining features of Granite-greenstone terranes such as “dome and keel” geometry in terms of channel flow. Channel flow is where soft squishy crust starts flowing sideways under pressure; today it happens (perhaps) only in thickened crust in mountain belts, like the Himalayas. In the hot Archean, it could have been a much more common process.

Whether or not the fundamental processes are the same, the Archean earth was very different to the planet we are sitting on now. It was frequently struck by large lumps of space debris, had a radically different atmosphere, no ‘visible’ life and weird geology. If we were suddenly transported to the Archean, we might (in the few moments before we suffocated) think we were on a different planet. When studying the remains of such a place, the uniformitarian idea that “the present is the key to the past” is stretched to breaking point. Understanding these extremely ancient rocks is very hard indeed, but it is one of the most interesting challenges in geology.


Cawood, P.A., Kröner, A., & Pisarevsky, S (2006). Precambrian plate tectonics: criteria
and evidence GSA Today DOI: 10.1130/GSAT01607.1

Open access link.

L.B. Harris et al. Regional shortening followed by channel flow induced collapse: A new mechanism for “dome and keel” geometries in Neoarchaean granite-greenstone terrains Precambrian Research 212-213 (2012) 139–154  Open access link.

Sizova et al., (2010) Subduction styles in the Precambrian: Insight from numerical experiments Lithos 116, 3-4 access link.


Categories: open access, tectonics

Comments (5)

  1. yodges says:

    i really love all your posts about continental and plate tectonics and the nature of the LAB and mountain building… this is easily the most interesting aspect of geology/geophysics to myself and i really enjoy your writing regarding these subjects.

  2. Metageologist says:

    Thanks. On Twitter @ferwen has just put a link to research suggesting plate tectonics is active on Mars. Not sure about it myself (not having read the paper) but interesting and relevant:

  3. Torbjörn Larsson, OM says:

    Thanks for this! Astrobiology interest here, so I was going to ask about the Mars p.t. suggestion. Haven’t read the paper, but the group has some videos and sure enough, the Valles Marineris looks shifted by way of fortuitous large impactors across the rift (plus stuff only an experienced geologist can see, I guess).

    Maybe not quite sufficient for “the _differential_ horizontal motion of plates”, I have to read both papers now. :-)

  4. Metageologist says:

    I agree. I’m sure they have decent evidence for offset, but to extrapolate from that to ‘discovering plate tectonics’ is too much. Plate tectonics with only two plates doesn’t make sense as there is only one possible sense of movement.
    Perhaps I’m overly influenced by this poor press release which has worrying statements about ‘how far below the surface the plates are located’. This is daft, as the plates *are* the surface.
    Anyway, we should wait for the paper.

  5. Dan McShane says:

    Just wanted to say nice write up.

Leave a comment:

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>