Earth’s forgotten youth – and beyond

A post by Chris RowanResearchBlogging.orgThe further back in time we go, the more and more fragmented the Earth’s geological record becomes. Whilst not exactly common, rocks with ages up to about 3.5 billion years old are found at multiple points on the Earth’s surface. However, rocks older than this are much less common. Extensive outcrops older than about 3.8 billion years are exceptionally rare, possibly because a series of very large meteorite impacts prior to this time – the Late Heavy Bombardment – largely destroyed any older bits of crust. The Acasta Gneiss in northern Canada, dated at around 4.03 billion (4030 million) years, is generally regarded as the oldest known outcrop of crust, although a recent study has claimed that the Nuvvuagittuq greenstone belt, also in northern Canada, may be as old as 4280 million years. The only known bits of the Earth that are older are 4.2-4.4 billion year-old zircon crystals found in the Jack Hills Conglomerate in Australia; the conglomerate itself was deposited about 3 billion years ago, but it contains debris eroded from much older, and now long-vanished, bits of crust.
Two or three data points is not a hell of a lot to go on when trying to reconstruct the evolution of the early Earth, especially when the material involved is far from pristine (the Acasta Gneiss, being a gneiss. has been partially remelted, for example). It is therefore no surprise that the geological timescale for the period between the Earth’s formation, about 4.56 billion years ago, and the start of the Archean Eon, usually pegged at about 3.8 billion years ago, is rather lacking in detail. This period is usually referred to as the ‘Hadean’, which is more of a reference to the presumed conditions on the Earth’s surface than a subtle pointer to the fact that we don’t know what the hell was really going on.
However, this has not stopped Colin Goldblatt and his co-authors having a go at adding a bit more structure to the Earth’s earliest days – and beyond. The ‘Chaotian’ eon at the start of their proposed new timescale is a common framework for the entire solar system, beginning with the gravitational collapse of the gas cloud that it would eventually form from. Key events – such as the initiation of solar fusion, or the first interactions between sizeable protoplanets that condensed from the protoplanetary disk – mark the boundaries between different eras and periods within the Chaotian. The start of the Hadean is marked by the collision of the proto-Earth, which the authors call Tellus, with another Mars-sized protoplanet, forming the Earth-Moon system.

Proposed new timescale for the formation of the solar system (Chaotian) and the evolution of the early Earth (Hadean). Click for a larger image

Thus, the Chaotian marks the time when solar system first became a distinct entity from the galactic neighbourhood; the beginning of the Hadean is when the Earth’s geological history begins to be shaped as much by internal processes, such as mantle convection, as by external events such as collisions with other protoplanets, Similarly, from beginning of the Archean, at the end of the Late Heavy Bombardment, internal processes start to completely dominate the Earth’s geological evolution; extraterrestrial collisions can still have significant geochemical and biological impacts, but they no longer melt the entire crust. Conceptually, I find this quite a nice way of looking at it.
The authors also attempt to subdivide the Hadean, but because we still don’t understand the key events in the Earth’s geological development over this period, it’s not quite as successful. The Hephaestean period probably covers the recovery from the Moon-forming impact. The Jacobian, Canadian and Acastan periods refer to the Jack Hills zircons, Nuvvuagittuq greenstone belt and Acasta Gneiss, respectively, but although these outcrops can give us clues about what the Earth was like when they first formed, it is a bit risky to try to characterise an entire planetary system from one sampling point. For example, the Jack Hills zircons tell us that granite – in other words, continental crust – was forming 4.4 billion years ago, but this is only a minimum age; we have no evidence that it wasn’t forming before that. Also, for all we know greenstone belts were also forming at exactly the same time, and have just not been preserved. The small amount of data we have available means that a single new outcrop might force the entire timescale to be redrafted.
It’s difficult to know if we’re ever going to be able to construct a truly robust, process based timescale for the first 700 million years of Earth’s history, because it’s unclear how much we’ll ever truly know about the Hadean. Still, this is an interesting attempt to set the story of our planet’s birth into a slightly more structured framework.
Goldblatt, C., Zahnle, K. J., Sleep, N. H., & Nisbet, E. G (2010). The Eons of Chaos and Hades Solid Earth, 1, 1-3

Categories: deep time, geology, paper reviews, past worlds

Comments (9)

  1. Can’t help thinking that calling the period between 4.4 and 4.3 billion years ago the “Jacobian” is going to cause confusion – that names already taken!

  2. Bob O'H says:

    Who is this Chris Rowan guy? Is he a new blogger here?

  3. Garry Hayes says:

    Thanks for the review! I’m talking about this time period in historical geology tonight, and I’ve been thinking the texts are not really up to date on the latest research.

  4. Lab Lemming says:

    Shouldn’t the naming rights for these time periods be reserved for the people who actually figure them out? Drawing a bunch of lines through early history based on geology that has yet to be done seems a bit presumptuous.
    Also, moving the Eoarchean/ Hadean boundary forward by 150 million years is going to confuse any future geologists reading current papers which use 4 GA.
    There’s also the problem that evidence of the LHB in rock that are 4.0-3.85 Ga is stubbornly unobserved, despite numerous attempts to do so. While the LHB is certainly important to the Lunar nearside, basing a fundamental division of the terrestrial timescale on a lunar event makes you look silly if subsequent research shows it had no effect on Earth.

  5. Lassi Hippel?§inen says:

    So a dozen new mock Greek names to remember? What’s the point?
    Even archeologists, who are humanists, are happy with numbers as subdivisions. (But being humanists, they prefer Roman numerals.)

  6. Chris Rowan says:

    So a dozen new mock Greek names to remember? What’s the point?
    Well, there are no real numbers here – the authors have just used rather arbitrary 100 million year chunks. The boundaries could quite easily be shifted in the light of new discoveries, so using names is a bit more flexible in this respect.
    I’d personally consider this a starting point rather than a finished product: it is reaching towards a process-based, phenomenological division of geological time, which I like, but runs into the wall of us not really knowing enough to do it properly.

  7. Not a big fan of subdivisions of geologic time without any record to speak of, but I do like the use of “Tellus” for “proto-Earth”.

  8. Popinga says:

    Time-scale is arbitrary, events are not certain, naming periods is presumptous. A good exercise for geologic nerds.
    Greetings from Italy!

  9. Chris says:

    I would have thought the formation of the earliest geological solids in the solar system (refractory inclusions) would be a good place to draw a line. 4.5672 billion years, or thereabouts. “Solar system and constituents cool enough, due to adequate radioactive decay, to allow solids to form” seems like a pretty important event to note. Pity they stuck to 100 MYr blocks.
    Things happened pretty quickly after that, in much the same way we mark the Cambrian as the starting point for multi-cellular life.