Eruption has started

Screen Shot 2014-08-29 at 02.15.09View from Mila webcam

The IMO has confirmed that a fissure eruption has started north of Dyngjujökull.

Please do not link to and sites at this time

Please use the live Youtube stream for the webcam rather than the Mila link, as the latter is likely to be in use by emergency authorities and IMO.

This is just a holding post – more information will be provided later as it becomes available.



After reviewing images. This eruption seems to for the moment take the best possible solution way. Currently the true eruption has not started, but will do so in a few hours. What I mean by that is that currently we are mostly seeing steam coming out of the newly opened fissure, this is caused by rising magma heating away the water table on a local scale.

As the true eruption starts we will see fire fountains a few hundred meters high. I also expect that the fissure will widen during the day, and we might also see new fissures open up in the days and weeks to come. The ultimate scale of the eruption remains to be seen, and this eruption could continue for weeks, months, or even years.

I am though happy that the eruption started now instead of later since this will limit the amount of peak pressure in the volcanic system which should dampen the initial effusive rate. I do not currently expect a large explosive event, but if a fissure opens up closer to the volcano under the ice it could turn ashy temporarily.

Normally fissure eruptions tend to open new fissures closer to the parent volcano, so the current fissure is most likely going to be the furtherest point from Bárdarbunga that this eruption will reach. I do not expect a follow up eruption at Askja at this juncture.


this is a 3D view of the situation this morning. Note the seismic activity under Askja, and some activity aver the 64.7° zone. To my mind this should not be too far from the eruption site.

And this is a zoom on the earthquake front.
Note the reactivation around 64.8°N

it is a view from the east – all quakes over mag2 since the 16th – finishing 6H38 this morning

1,104 thoughts on “Eruption has started

  1. Hello Dragons,

    0 scientific background here.

    Is it possible that ongoing geomagnetic changes are effecting volcanic activity across the world? As the force of gravity shifts, it causes earth’s crust to adjust itself, spurring volcanic activities like what we’ve seen in Iceland and Papua New Guinea?

    see bbc article: ”Gravity satellite yields ‘potato earth’ view”

    Notice the areas where gravity field is currently strongest are Papua and Iceland. Rocks getting pushed down harder than ‘usual’ – a plausible cause for quake activity ?

    My apologies if this point has been addressed or if we were taught this in grade 1… I haven’t gotten a clear explanation as of yet nor have I spotted one on VC.

    Thanks for this great website !

    • Not entirely sure without having done any personal research, but if you’re referring to magnetic poles shifting, then I would say it has little to no impact on volcanism.

      Volcanism in Iceland and Papua New Guinea has not become stronger in recent times – they’ve both been extremely active volcanic areas over the last 1 million years along with other extremely active areas. Iceland sits above a very active hotspot and the mid atlantic ridge – this is the cause of the voluminous volcanism that has been ongoing for a very long time here. Papua New Guinea sits beside a very active subduction zone, although it’s not nearly as active as Iceland is.

    • Well… gravity itself does not change. One kg of material exerts the same amount of gravitational “pull” as it would anywhere else. So, how can a gravitational field get stronger? Simple, add more mass. So, if a field survey detects stronger gravity, it is because a greater concentration of mass. This could be indicative of denser material, or a fortuitous alignment of masses, but at the base of it all, only a greater concentration of mass can do that. In Iceland, you have pretty much the direct center of the North Atlantic Large igneous province. This has been erupting for millions of years. Down around Papua, you have billions of tons of oceanic crust diving (almost vertically in places) under the overriding plate. Oceanic crust is quite dense, approaching upwards of 3100 kg/m³ due to being composed mostly of basalt and gabbro intrusions. Continental crust on the other hand, averages about 2700 kg/m³. Given those structural variations. It’s not really that surprising that the field measurements come out the way that they do.

    • There could be a couple things you picking up and confusing, nothing to do with gravity though.
      First is the process known as glacial (isostatic) rebound. Where the weight of ice during glaciation forces the continental crust to sink into the mantle, and then when it melts the crust bobs back up like a cork due to its lower density.
      The second is the hypothesis that periods of low solar activity loosely correlate with periods of high volcanic activity (and I mean loooose!) It has been theorised that the magnetic changes to the sun somehow influence the magnetic field of the earths core resulting in more heating/hotspot activity and more volcanic activity. There is little evidence for this either way! 🙂

  2. Is the fissure swarm magma sourced from below the crust? YUP!
    Nope → “maybe” (GL edit) We need actual tomography to get a better idea.

    Looking east 10 km down

    Looking ESE 10 km down

    Looking east from 7.5 km down

    Bardarbunga – looking due south 5 km depth

    I can’t past the link to the dynamic model here, but as you can see the fissure is feed by the asthenosphere, not by Bardarbunga.

    And can we now please get past the repeated baseless claim that there are no valid deep earth quakes in the sequence given this is the IMO’s checked data plot.

    And no more pretending that large quakes are not occurring in the fissure swarms. There are astonishingly powerful quakes in those narrow confines.

    • The mean depth of the MOHO is around 40km under Bardabunga swarm. Its the thickest crust in Iceland:
      So your proposal that the swarm has accessed the mantle depends upon there being a rift/fissure that has a shallower base at 15km coincident with the plotted events. As far as I know rifting is only known from surface features, there is not sufficient information at depth to justify that conclusion.
      If we are seeing a repeat of Holuhraun, that lava has features that link its source to Bardabunga, not mantle.

      • Not sure his sources, but Carl has mentioned in multiple places that the mantle is comparatively shallow in the rift swarms, and typically is in the 20km depth range. Don’t quote me on that, but I tend to believe that.

        • I dont know the resolution of the gravimetric and seismic techniques used, but there seem to be agreement that the mean depth is 40km

          Within that mean there must be rifitng, but at density that leaves the mean value at 40km.
          Whether a 15m-wide intrusion has intersected a rift with a 15km deep MOHO is, I think, not possible to determine.

        • Perhaps this source:
          F.A. Darbyshire, etal., Structure of the crust and uppermost mantle of Iceland, from a combined seismic and gravity study. (2000)

        • That may be so cbus, you would certainly expect it to be, he’s probably right, or more than right enough.

          Please forgive me cbus and other if i tread on toes a bit, I have a sense that you have a good grip on geology from your comments, like some other ‘core’ people in here. As a poorly integrated noob here, I can’t immediately see who has a clue, and who is interested, and entertains a ‘view’. ;D But really, why wouldn’t they.

          But the reality check of swivel the head around at the countryside to see whats really there tells you what’s in fact possible below, especially in such a young terrain as Iceland’s. And what I see a LOT of volcanics, coming from above the moho, just as with everywhere else on earth. And somehow a

          The moho is not dry of melt, if it were dry and did not melt, it would not be a moho, would it? As it would not have the required seismic signature, unless it were also partially melted.

          QED 😀

          So the very act of admitting a deep moho exists is the same thing as saying a HUMONGOUS reservoir of melt is really to go very close to the surface!

          QED Oh the irony! 😀

          How did this get so screwed up? I mean, this stuff is so basic. I find it odd people don’t know these things. But there enough here who keep it interesting and funny. Maybe one day I;ll integrate 😀 But the truth is, I’m awed to see Iceland doing this in my lifetime. 🙂

          DragonEdit: Turn down the Snappishness Unmentionable.

      • Bard is on the triple point above the hotspot. Not sure how that effects gravity readings!
        The lack of inflation shows that this a tension driven rifting event. The rifting started at the weakest point and is spreading north till it stops (anyone know the size of the block and the location of the next transform fault?)
        What we don’t know is the effect of the Bard magma being fed into the fault. Will it act like silicone gel, filling up the gap and preventing decompressional melt like Laki, or will it act like a hot knife through crustal butter causing the rift to be even bigger?

        • Interesting question, got to go to bed but we can discuss this soon if you like, I’d love to hear your ideas on that.

      • @ Peter: “The mean depth of the MOHO is around 40 km under Bardabunga swarm. Its the thickest crust in Iceland:”

        So? I’m trying to work out why you think that’s significant then I realized you don’t know what you are talking about. Where did you get the entire false idea that it has to get to the ‘mantle’, to access a melt? Now I realize you know nothing, or you wouldn’t say such a thing. So this is an opportunity to educate those who are similarly confused about what the asthenophere is, and the role it plays in melt-sourcing.

        Igneous intrusions are sourced from three things broad categories that are really all just silicate rocks.

        The subcrust is about 15 km down in this case, it varies, it evolves, like everything everywhere else. Geology is just slow evolving.

        In this case the asthenosphere is almost always what melts, get this creates the melt that erupts in Iceland! You didn’t know that so whatever you’ve been reading that led you to think different, stop reading it, throw it in the bin, it’s total BS.

        I notice several people seem to be affected by this figment that somehow the precious asthenosphere doesn’t melt? phft! OK. But do you even know what it is, chemically, it’s mineralogically? Do you know where it came from? Or why it is a bit different now? Or why it is distinguished from the solid mantle below as a bit different acoustically and chemically?

        The vast bulk of rock in Iceland came from the partial melting of the asthenosphere.

        Sorry if you didn’t realize that. It’s the very reason why there is an asthenosphere! And the asthenosphere comes from the partial melting of the deeper parts of the mantle. But it is all mantle, the asthenosphere is actually the mantle. How people ended up with the view its is unable to melt is the result of no formal education or work in the field.

        The mantle creates both asthenosphere and crust, and continues to contributes to both. the asthenosphere contributes to the crust and doesn’t ever stop doing so. The crusts is also self reprocessing and sometimes resumes back into the asthenosphere. And the mid-crust can also melt and re-intrude at a higher level and a ‘new’ evolved crust. So the crust can come from ANY area below down to considerable depth.

        Anyway this could not be more misconceived and such basics should have been straightened out a long time ago. I noticed even Jon put to me the other day that the moho was 46 km deep. That was the argument for the impossibility of deep intrusion, now. So I was sitting there thinking, why is he saying that? Now I know, many of you seem to think its significant.

        No guys, the asthenosphere is what decompression melts, its how it becomes the asthnosphere and has a different seismic velocity.

        i.e. it’s partially melted already, its what people refer to it as ‘mush’ or ‘slush’, because it is often partially melted. And because it is partially melted seismic waves pass through it at different speeds to solid rock.

        But what would I know, Wiki is a far more credible source than any geo, apparently, so let’s confer:

        Mantle (geology) From Wikipedia, the free encyclopedia
        “Earth’s mantle is a silicate rocky shell about 2,900 kilometers (1,800 mi) thick[1] that constitutes about 84% of Earth’s volume.[2] It is predominantly solid but in geological time it behaves like very viscous liquid.”

        Asthenosphere From Wikipedia, the free encyclopedia
        “The lithosphere-asthenosphere boundary is usually referred to as LAB. Asthenosphere is generally solid although some of its regions could be melted (e.g. below mid-ocean ridge). The lower boundary of the asthenosphere is not well defined. The thickness of the asthenosphere depends mainly on the temperature. For some regions asthenosphere could extend as deep as 700 km (430 mi). It is considered the source region of mid-ocean ridge basalt (MORB).”

        MORBis the same stuff that makes up Iceland.

        Now, the Moho was discovered by a seismologist named Mohorovicic who determined that the seismic propagation properties were consistent with a partial solid, and a partial liquid. Egad! This was a very powerful evidence at that time for the mysterious source of volcanic magma.

        And he was right, that is exactly where >90% of volcanic melts come from, i.e. ABOVE the moho, in the asthenosphere!


        So if people are going to talk about Iceland’s vulcanology you do actually need to at least understand where the melt came from.

        It could do with a proof read but its almost my bed time, so we shall resume the banter later, good night.

        DragonEdit: Unmentionable, I think you could do with a lowering of your Snappishness level. I further seriously suggest that you google who Peter Cobbold is.

        • Thanks – good stuff. Very helpful. A while back I posted an opinion that Bardarbunga could not be the sole originator of this rift through basic mechanics. We’ve all but established magma enters the rift system through tension forces. These forces far more likely to *pull* magma out of Barda’s chambers rather than the idea Barda went into eruption mode and pushed it out sideways instead of erupting through its established weak point – its lid. Mechanically, I can’t work with that. The caldera lid is the easiest route for pressure to be released and it has recently shown it is flexibly responsive to pressure by its deflation, so therefore it is pretty ductile. So I can’t work with Barda being the sole source of magma for the rifting.

          Earlier diagrams showed a kind of giant magma pond feeding the bottom of the rifting. I don’t go with that, either. Given the stop-start nature of the rift and the wide range of quake patterns, I picture the supply from the asthenosphere as multiple feeders in sills and dykes reaching up into the rift area, drawn up by tension and decompression.

          Does this fit with your knowledge? I just can’t see the bottom of a rift like this happily open to some kind of giant swirling molten magma pond. That sort of scenario would have weird and wonderful fluctuations in the crust that aren’t there in the instrument readings.

          I’m a dumbo not an expert! But I’d appreciate opinions on this because if I’m wrong I’d like to know it.

          Oh – and I still stubbornly stick by my hypothesis that the rifting has NOT reached Askja, and is stalled by cold, hardened magma in Askja’s wider fissure system. The various diagrams offered show the leading edge of the rift nowhere near Askja. So why do people insist it is inside Askja’s system? It is only bumping up against Askja’s hardened ‘knot’ (like a knot in wood).

          But I may be wrong! 🙂

          • Well done! Not there’s a mind that actually want to understand, not just get told a story, but an actual understanding of Earth! I LOVE that! 🙂 Brilliant.

            “A while back I posted an opinion that Bardarbunga could not be the sole originator of this rift through basic mechanics.”

            Wow, I’m really astounded people thought that, it just takes you nowhere. I can see why you moved on. Well, we’ll talk again, you are doing great sorting this stuff out. 🙂

            • Thank you for the link Mopshell!

              I do not doubt mr Cobbold’s knowledge.
              It is just that the hot plume is a part of kindergarten geology 101 in Iceland.

              This was kind of like being told that 2+2 are not 4 😉

            • Oh ohr, I’m sorry – I didn’t do that very well at all. You’re always so very respectful too. I was, though, really excited to discover Peter’s professional background – you’re all, every one of you, such extraordinary people here! 🙂

            • I do not share Peters view on mantleplumes. He is though highly respected in here none the less.
              After all, a good and healthy polite discussion is good for the blood stream 🙂

              It was the slight lack of “polite” in one of the commenters answers that we took offense on. Not the opinions.

            • Now that sounds entirely reasonable to me. 🙂 Mind you, I was fascinated to read about Peter’s professional background – what interesting lives people here have! I’m a tad envious but, far more than that, very appreciative of the extraordinary company here. Wow! 😀

            • It is a pleasant and educating company in here
              – by the way, no need to be apologizing to me Mopshell. You did that very well – I did not know of Peters background (I am more used to first names being an Icelander – I have a feeling it is all right in here).
              Then again – my French aint that good, but it sure looks impressing 🙂

            • My French wasn’t good enough either – I had to resort to Gurgle Translate! 🙂 My but your English is SO GOOD!

    • “There are astonishingly powerful quakes in those narrow confines.”

      Well, lets look at that. What size quakes are you referring to? If one were to assume normal mode faulting, we can work out an approximation as to how much material could be at play, filling the void. Be forewarned that using this methodology is fraught with potential errors. It’s not a proven method, just something that I’ve come up with, and my expertise is definitely not geophysics.

      Also, a stated rule of thumb about Moho depth pertaining to rifts in general, does not mean that a heavily underplated crust over a hotspot is going to be exactly the same. In other words, a generalization does not always fit.

      Also, there is a possibility that there is another crust segment underlaying Iceland. I ran across that possibility while trying to puzzle out why Iceland is much more thick than normal volcanic islands. It could be a whole slab of continental crust sitting on top of a section of oceanic crust. There was a subduction zone in this area back when Avalonia was still in existence several million years ago. In essence, figuring out Iceland is like trying to decipher the parts of several million year old train wreck.

      Nearby, is the Jan Mayen microcontinent. Tiny Jan Mayen island is the only feature of it that is above water, but the fragment is continental crust likely sheared off of the Greenland craton and welded to the Eurasian continent. Usually, continental crust rides above water, but not in this case.

      My contention is that while the chaos was going on to produce that, somehow, a shard was left laying on top of oceanic crust and we now know that as Iceland.

      Caveat: The usual, not a geologist, not a seismologist, and I didn’t stay at a holiday in express last night. However, I have read “The Dancing Wu Li Masters

      • “Also, a stated rule of thumb about Moho depth pertaining to rifts in general, does not mean that a heavily underplated crust over a hotspot is going to be exactly the same. In other words, a generalization does not always fit.”

        If you think it’s not routinely derived from above the moho, then what’s with all the volcanics above you dash a toe on? At some point a basic reality-check must defeat a bad theories and conceptual bagage GeoL.

        Got to go, will discuss later with you if you wish to pursue it.

      • I don’t play that game. I never stated that “the mantle was not involved.” All I am saying is that in this instance, there is no direct evidence that is the case right now.

        However, if you have it, I am all ears.

        • um, still answering people, should be in bed. Not ‘playing’ a game Geo, I don;t know what you mean by that. However, I’m a geo and you are not. I say look around and try to deny the mechanism that put it there on the terrain in Iceland. I sure won’t. The fact of the terrain being covered in melted and erupted partial melt MORB type rocks from the asthenosphere should be a pretty strong hint about the real situation. No games, don’t think like that, this is a conversation. 😉 🙂

          DragonEdit: ‘I’m a geo and you are not.’ Take it down at least a notch.

          • “My contention is that while the chaos was going on to produce that, somehow, a shard was left laying on top of oceanic crust and we now know that as Iceland.”

            Sorry, forgot this bit. Look 95% or more of the ‘cute’ theories I was taught were all wrong, in part or in whole, so I have zero respect for the cute theory worship for what is most probably complete BS. Don’t mind me though mate, I am being always irrelevant because I watch people tie themselves in knots over that stuff and it’s just sad to see, while they ignore the reality in front of them. The unmentionable in my namesake is my hat-tip to the fact that to say such is often like tossing a pork chop into the synagogue. UNMENTIONABLE! 😀 So it’s just me, not you. 😀 Really, this time, bed for sure.

            DragonEdit: Check your email. You have a very short and simple answer in it. Ponder that answer well.

            DragonUpdate: Now two emails.

          • Eh.. matters not.

            There is enough energy release below 20 km to make the idea fully plausible. (>3.01008E+13 Nm in 15 quakes over the last 48 hours)

            I defer to the cloaked Phd.

            • Hmm, here are some plots showing the energy released per depth per day.

              First scale according to energy (25000GJ):

              Second scale set to max 1000GJ

              Third scale set to max 2GJ

              From the perspective of the amount of EQ’s per depth per day (activity) I got these two:

              And scaled so the deep ones show:

              To me this looks like all the action is rather shallow, mostly in the first 10km of dept.

              DragonNote: This was salvaged from the dungeon (Pending review). Please do not add any more than THREE links to any comment. It’s not our rule, just one of the vagaries of WordPress. Thanks for these lovely plots! Best/UKV

        • Think the evidence is in on that , def involvement of second feed not from barbarbunga to main dyke. See post below for links , images and data.

    • The raw earthquake plot has come slightly closer to Askja today. You can see the cluster of reds juuuust sticking over the bottom of the Myvatan grid on Vedur. That wasn’t there yesterday.

      • Tweet by Dave McGarvie @subglacial, 3 h ago:
        “The EQ swarm NE of Bárðarbunga extends through the E side of Askja which is dominated by old subglacial basalts.”
        I also thought, that the swarm now is passing by Askja and heading in direction of Herðubreiðartögl.

  3. Newbie here, just want to add my thanks for all the posts, commentary, speculations, and enthusiasm!! I happened across a link to this blog in my internet cruising and have become quite addicted the last few days. Not only are the posts immensely educational, but the mix of understandable explanations, total technical volcanic “chinese,” musings, hypothesizing, joking and sillyness is perfect 🙂 The community here is clearly tight knit, but not clique-y. You all rock 😉

    oh yeah, last night around 9pm ET, when you all first spotted the initial eruption, was AWESOME!

    • Being a also a newbie, I heartily second that comment. I stumbled on this site about a week ago and have gotten nothing done since then. Great job dragons and everyone else that contributes there.

  4. Evening all! What a day! I need a drink! I spend years stuck in front of my computer, then the one day I want to be stuck there, I have to spend the day rushing about doing other things. Then I get stuck in traffic on the way home – in the middle of nowhere, where there’s never been traffic – and so I’m late home. In short, I’m doomed to never be in the thick of it when something volcanic happens (remember I missed the entire Grimsvotn eruption due to a camping holiday). Sigh! At least I can now look at the extremely interesting fog. 😉

    • From IMO’s website:

      12:20 29/08/2014

      “At this moment it is unclear how the situation will develop. However, three scenarios are considered most likely:
      – The migration of magma could stop, resulting in a gradual reduction in seismic activity and no further eruptions.
      – The dike could reach the Earth’s surface north of Dyngjujökull causing another eruption, possibly on a new fissure. Such an eruption could include lava flow and (or) explosive activity.
      – The intrusion reaches the surface and an eruption occurs again where either the fissure is partly or entirely beneath Dyngjujökull. This would most likely produce a flood in Jökulsá á Fjöllum and perhaps explosive, ash-producing activity.

      Other scenarios cannot be excluded. For example, an eruption inside the Bárdarbunga caldera.”

      And at 18:30 the were reporting that there had been 1200 eqs today, which is roughly the same as yesterday.

      So you may not have missed the entire or even the main event …

  5. Not drawing any conclusions, but the swarm is starting to look like it’s cutting across eastern Askja and directly into Herdubreid. It may never do this, and I know Henrik has mentioned that this is very unlikely (but who knows, this swarm has already done quite a few unlikely things).

    I find this interesting since minor activity before the major swarms started on the 16th had been going on in the background both at Herdubreid and Bardarbunga, and they both started seeing an uptick in activity at roughly the same time a few months back. Just makes me wonder if it was an early sign of tension accumulating along the rift line, causing unrest across a wide area.

    • While it could stop at any time, the only thing guaranteed to stop it is the transverse fault at the end of the crustal block! Anyone know where that is?

  6. Tremor been rising steadily in the Sth of glacier (vot), also some in nth, but Sth glacier stations seem rising sharper

  7. Just one stupid question shall i go to bed or shall I stay up and see if anything happens ???

    Joking love the site learnt loads

  8. Well this may answer many questions

    Bearing in mind the current eruption “may not” be part if the larger dyke ? But a small branch from bardarbunga caldera. Current analysis will shortly show ” where the erupted lava has come from ” fairly shortly.

    If tweet is correct (follow link to earthsciences , some nice plots and imagery) then main dyke has separate feed from mantle.

      • Agreed!

        This stood out to me – “The model indicates that the eruption is coming from the patch with maximum opening (5m)”

        this provides two important things to me. 1 – there are areas where surface rifting has reached 5 meters. Given this is only localized, but I wonder if it’s that size as a result of the eruption, or if it reached 5 meters wide independent of the eruption. A 5 meter rift is much larger than most of us had estimated, even if it’s only a localized rift.

        The other thing that stands out to me, is that eruptions start when you get a 5 meter rift. Well, this is somewhat obvious, but it provides a good benchmark to guesstimate a larger eruption across the spreading zone.

      • The satellite deformation data plus GPS has allowed them to determine that the intrusion is narrower at depth, with max defomation of 5m in the top ca. 2 km.

  9. I don’t know guys, based on the EQ activity continuing around Bardar, and all the quakes over the entire area, I think there is still a lot more to come, maybe in days, weeks. I don’t see the “activity subsiding” scenario as very likely, except in immediate the short term. The amount of energy stored right now, it has to go somewhere.

    • The earthquake activity is the release of crustal tension! The stored energy is the Bard magma column feeding into the fissure plus the large temp/pressure gradient between mantle/crust and surface.
      The “Fun Level” is determined by the location where the dissolved gasses decide to come out of solution.

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s