Lava lake tectonics

As cool geo-imagery has poured in for the upcoming Accretionary Wedge*, I have been pondering what I should throw into the collective album. Eventually, I decided that when it comes to dramatic imagery that ties into my geological interests (albeit somewhat obliquely), you can’t beat a lava lake.

Erte_Ale_1.jpg
The Erta Ale lava lake, Ethiopia. Source

A vat of bubbling molten lava seems more like something from a bad disaster movie than real life, but sometimes there is the right balance between vent shape and eruption rate for a lava lake to form. The one shown above is found in the crater of Erta Ale, a basaltic shield volcano found in the Afar rift region of Ethiopia. What is really cool about this picture is what is going on at the lake surface. The uppermost layer of lava has cooled and solidified, forming a thin crust on top of the molten lava beneath. This surface rind has the same composition, but different mechanical properties, than the fluid it covers, much like ice on a winter lake. The churning of the convecting lava beneath has broken apart this crust into a number of different slabs; in the picture above, the boundaries are marked by the lines of bright molten lava, creating a dramatic natural jigsaw puzzle. The same convection currents jostle the slabs against each other, moving them apart in some places, with new upwelling lava cooling and solidifying in the gap, and in other places pushing them back into the depths of the lake to be remelted.

Erta_Ale_2.jpg
Source

Erta_Ale_3.jpg
Source

Although you should all know by now that the Earth’s mantle is not molten, it is divided mechanically in the same way as the Erte Ale lava lake. The upper mantle forms the colder, stronger, lithosphere, rigid enough to hold its shape; the lower mantle forms the warmer, weaker, asthenosphere. Convection currents in the asthenosphere move heat from the Earth’s interior towards the surface, and in the process also drive the motion of the overlying lithospheric plates. So the Erta Ale lava lake is in fact a fabulous natural analogue for plate tectonics, and because in this case the convecting medium is molten, moving at centimetres per second rather than centimeters per year, we can see processes that take tens of miliions of years on a global scale happening in just a few hours. It’s quite compelling to watch.

*it’s my carnival and I’ll submit after the deadline if I want to.

Categories: geology, tectonics, volcanoes

Comments (8)

  1. Birger Johansson says:

    A question: As I have been told, ice floats because water is *one of the few* substances whose solid form is less dense than its liquid form. Presumably the solidified lava floats because some of its components also have this rare quality, and in that case, which substance/element ?

  2. Birger Johansson says:

    BTW, speaking of plate tectonics, here is a link to Pysorg.com about the volcanism occurring when the Appalachians were formed, and the connection to the end-Ordovician mass extinction:
    “Volcanoes played pivotal role in ancient ice age, mass extinction” http://www.physorg.com/news175785444.html
    I include the link because your American readers will be familiar with the Appalachians; this is yet another way to see how intimately your nearby geological features are linked to large-scale, ancient phenomena

  3. Torbj??rn Larsson, OM says:

    ice floats because water is *one of the few* substances whose solid form is less dense than its liquid form.

    It will be interesting to know what the experts say. Meanwhile, some caveats:
    - While it is “unusual” that the crystallized form floats, it isn’t unique. However, that ice expands when freezing is, for non-metallic compounds. (Due to its hydrogen bonds.) [Wikipedia]
    Caveat to caveat: Another probably unique property of water is that it seems it contains the crystalline phase as (likely dynamic) micro-associations in a matrix of the liquid phase at all temperatures. (Tending towards less liquid at lower temperatures.) The difference in thermal behavior between these two phases is what makes 4 degC water densest around atmospheric pressures.
    So the floating ice is likely unique in a more fuzzy way.
    - You are assuming that the surface solid is the same composition as the not yet atmosphere exposed liquid. I wouldn’t.
    - You are assuming that the remelt described is not gravitationally helped. I wouldn’t.
    - What is the viscosity of lava? That will influence solid lava sink time.

  4. Kyle says:

    “we can see processes that take tens of miliions of years on a global scale happening in just a few hours.”
    Throughout the video above there are several instances in which the rate of “plate movement” increases dramatically, as anyone who views it will see. Give this, can we expect the same type of violent and sudden movement of our own plates?

  5. Torbj??rn Larsson, OM says:

    OK, no expert answer yet. I looked into this (good thing too, since I have attend a course that touches this subject) and indeed crust in general, whether the global or local, doesn’t have the same composition as the melt.

  6. Eamon Knight says:

    Has anyone ever tried to come up with a tectonics demo using some sort of wax/oil mix? I you could get it right, that would be awesome.

  7. VolcanoMan says:

    I never liked the lava lake analogy to plate tectonics. Does a lava lake exhibit slab pull? Convection yes, but I am convinced by the evidence that mantle convection is more complex than the lava lake analogy implies, and that plates are both pulled apart by the subduction process as well as pushed apart by upwelling magma at the rift zone. As the properties of lava in a lava lake, and the “plastic” asthenosphere are different (not just compositionally – the pressures in the asthenosphere must be considered as well), the only thing a lava lake demonstrates to me is crustal formation (and in that impressive second photograph, deformation – that’s the first time I’ve seen evidence of transform faulting in a lava lake crust). Just my opinion, given what I know about geology (less than you bloggers, and certainly less than I’d like, but more than most people, although, as my name suggests, volcanism is my primary subject of fascination and man, I’m on the first plane to the DRC when people stop killing each other there to visit the Nyiragongo lava lake).

  8. Chris Rowan says:

    @Kyle: Throughout the video above there are several instances in which the rate of “plate movement” increases dramatically, as anyone who views it will see. Give this, can we expect the same type of violent and sudden movement of our own plates?
    Some plates do move faster than others, and at varying rates over time. It depends on what process is causing the acceleration in the video (which, to me, seems to be related to accelerated ‘subduction’, or foundering of slabs)
    @Torbjorn Larsson: crust in general, whether the global or local, doesn’t have the same composition as the melt.
    Yes, but tectonic plates do not consist of just (or even mainly) the crust, but also the lithosphere, which does have pretty much the same composition as the underlying asthenosphere. The crust is just the scum on the scum, if you like.
    @VolcanoMan – analogies are, by their nature, imperfect. Their usefulness depends on what you are using them for. In this case, the ideas of mechanical vs compositional boundaries, and the effects of underlying convection on a rigid (but not too rigid) boundary layer, can be appropriately highlighted and linked to global tectonic processes. If you want to talk in detail about things like the balance of driving forces, then the comparison probably does break down somewhat. Note I wasn’t trying to do that…
    As for slab pull – I’m not so sure that some equivalent process is not going on, if intermittently (see 1st part of this comment)