Posted on by Jón Frímann

Signs that a eruption in Grímsfjall volcano might be close

Over the past two weeks there have been signs appearing that a eruption might be close in Grímsfjall volcano. But a GPS station named Skrokkalda. This GPS station has started to move north, but this indicates that inflation has started in Grímsfjall volcano at full power. This did also happen at the same GPS station in 2004 before the eruption in Grímsfjall volcano that same year. But it is debated if this movement was created by ice or by a dike intrusion in Grímsfjall volcano. It is my just my personal opinion on this movement that takes place at the GPS station.

It currently remains to be seen if this northerly movement continues or if it starts to deflate again. So far this is the only indicator that Grímsfjall volcano is getting ready for a eruption. When this did happen in the year 2004 it was almost three weeks after this process started until a eruption took place. But then a glacier flood did trigger the start of the eruption in the year 2004.

For the moment the only thing to do is to wait and see what happens with Grímsfjall volcano. Nobody can tell how long the wait is going to be.

44 Replies to “Signs that a eruption in Grímsfjall volcano might be close”

  1. Hi Jon, I note that the plot only seems to run to Julian date 2010.95 whilst we are at JD 2010.342 – or am I reading this wrong? I was wondering if you have any other info on this?

    1. Nope, you are getting it wrong. They divide the year from 0-100, so 2010.50 means the day, when half of the year is over. This is a silly unit in my eyes.

    2. Since my first comment got lost, I try it a second time:
      Nope, you are getting it wrong. They divide the year from 0-100, so 2010.50 means the day, when half of the year is over. This is a silly unit in my eyes.

      1. Thanks for that Chris. I have been labouring under a misunderstanding for some time then!

        What a daft way of doing it!

  2. I am looking at the map where both SKRO and GFUM stations are mapped out. I just cant get my head around it. According to the GPS plot both stations are moving quite visibly north.

    How is this possible? There is an inflation beneath Grimsfjall but wouldnt that result in SKRO movement westward? If they are both moving north wouldnt that mean a “peak” somewhere in the middle of those stations a bit to the south?

      2. There’s seven years data on that plot – the present northward inflation is a little kick up at the right end- just visible.

  3. Has any one noticed that while we were yammering about the three ongoing swarm regions that pretty much all activity under Vatnajökull has fallen silent?

    2. Been thinking the same… At the same time there have been no significant changes in GPS position, nor in tremor.

      1. As we’ve seen with the recent jökulhlaup and previous eruptions, tremor and earthquakes only start a few hours before an eruption.

      2. As Jon has said, it depends… Some icelandic volcanoes do not give any warnings (except a few hours before), some do give plenty of warnings before even thinking about an eruption.

    3. [at] David L

      Well, whether or not I consider it significant is irrelevant. I’m not a geoscience expert. I think it’s notable. I haven’t run a time plot for just the Vatnajökull area, but I have seen 19.8 day, 37 day, and 28.4 day components in the “all of Iceland” Fourier transforms. Not that it means anything, it tends to move around a lot.

      2. Not really. I did the FFT against the hourly quake count just to see what it spit out. It is usually never the same as you add more data. It’s mostly just, well, I don’t even want to say “interesting” since that infers some meaning. and I don’t think that it has any meaning. At least on a level that I can see.

  4. I have to disagree with the statement that this GPS-data changes in SRK are caused by inflation at Grimsfjall. If indeed Grimsfjall would be inflating we would see the following things:

    -SRK moving north west -> current movement is northeast
    -Clear changes in the vertical component of the GRF station -> which is not the case
    -Clear changes on other GPS stations in the same distance as GRF station -> also not the case
    -More frequent (deep) earthquakes within the Grimsfjall volcano, caused by the movement of magma -> not the case

    In my opinion, the sudden change detected by SRK station is not in relation with Grimsfjall, but rather with tectonical changes.
    An interesting feature is that station FJOC, located 50 km NE of station SRK, has also been showing a sudden change at exact the same time as the changes at SRK started. These movements are completely opposite. What means that FJOC and SRK are moving towards eachother. (FJOC is moving southeast)

    Link to data FJOC: http://hraun.vedur.is/ja/gps/predorb/allar_pred.html
    And for those who are too lazy to scroll down: http://img256.imageshack.us/img256/4972/fjoc.png

    My personal theory about this tectonical change is a small transformal shift. Explained in this nice drawing of mine ^^ :
    http://img251.imageshack.us/img251/3434/fjoc2.png

    These are just my amateur thoughts on this. Because I actually think a movement in Grimsfjoll volcano wouldnt be noticed by a GPS station more than 60km away.

  6. There seems to be a lot of low frequency (0.5-1 hz) spikes within the tremor plot for Grimsfjall the last 24 hours. Could that potentially be a sign of an impending eruption? I haven’t seen it in any of the other tremor plots nor previously at this location, usually the spikes at these places are all high or medium frequency.

    http://hraun.vedur.is/ja/oroi/grf.gif

  7. [at] Pieter

    I whittled away about an hour laying in the positions of those GPS sites… I was going to compare the two mentioned stations with the plate boundaries. Then I ran into the slight problem of not having a shape file for the boundaries that I had confidence in. The UTIG Plates shape file doesn’t cover all of Iceland, it gets through the Reykjanes and stops until it’s north of Iceland. I came to the conclusion that a fine grained definition of exactly where the boundary is in that region at is sort of up for grabs. I’ve seen USGS plots that put it headed south towards Katla then to the Reykjanes, and other USGS data the has it tracking inland pretty much between the two microplates, then over to Bardabunga/Askja.

    *sigh….

    So instead, I dug up info on the largest post glacial lava flows in Iceland. Moved the table data over to a plot just for the S&G aspect of it.

    ‘Yall enjoy.

    http://i51.tinypic.com/2mqqowg.png

    1. Lurking, have you noticed the “stress wave” in So Cal again in the last week starting from its typical point and moving up past the Laguna Salada?

      1. Yeah.. but these seem to be smaller than the set that got my attention a while back. With an already active region at the north end, dunno what will happen. If it travels as fast as the last one, it will be a bit before it gets there. What was it? 145 to 160 miles a month?

      2. It looks like the wave crashed on the shore of Laguna Salada again with a 4.0 there 🙂

  8. @Renato & everyone else:

    Since there seems to be no signs I can see on Grimsfjöll actually being about to erupt except a week north and east trend at SKRO and no or small inflation on GRIM or SKRO I will instead go back to talking about the silence of Vatnajökull during our hectic times.
    Remember that we have uplift at 4 points of the sub-plate. The active uplift points being right now is Krysuvík, Godabunga (Katla) and Askja/Herdubreid with the Grimsfjöll & Bardarbunga being inflated but resting right now.
    This uplift of 3 of the roughly four corners of the microplate creates a lot of tension when you factor in sideral and lateral tectonic stress factors. For instance the area northeast of Langjökull is probably being pressed down right now.
    And then we have shearing factors running north from Krysuvík towards Raudholar (Brennisteinsfjöll direction towards Reykjavik, thank you Jón) and upwards plus tectonic(?) quakes at Esjan. These ones show as few but persistant quakes in the area. Then we have the area running from Katla to Grimsfjöll/Bardarbunga that is just plain dead (due to the crust being more pliable). This area will sooner or later start to shatter to since it can only take so much tension. The area between Bardarbunga/Grimsfjöll/Kverkfjöll towards Askja is also dead, but here the crust is also pliable. Can also break in a large fissure eruptions.
    And then you get a shearing towards north from Askja. Krafla is still warm and the crust pliable after the Kraflafires so it is pretty quake dead, but there has been a couple in the last week. And then we have Theistareykjarbunga. And here I politely disagree with Jón. Even though the likelyhood of an eruption in our time is really small it has started to show small signs of new activity. What kind of activity? I don’t know (with a huge shoulder-shrug).

    Back to Vatnajökull, with 3 corners (Vatna is in a line between 2 of them so it is not a corner) lifting it is actually being releived of strain in a fluid dynamics context. I therefore hypothetise that an eruption there is not likely for some time (half a year to a year). But, here I am talking about a central went eruption. There might be a fissure eruption due to lateral and sideral stress factors in the “dead areas”. Ie, between Katla & Vatnajökull volcanos, or between Askja & Vatnajökull volcanos.

    1. You are misunderstanding how tectonics work in Iceland. Here are few points worth to keep in mind when looking at data about Icelandic tectonics, earthquakes and all that stuff.

      #1: The microplates appears to move slowly. The respective locations is unclear for most part. This specially in regards to Tröllaskagi Microplate. The Hreppar Microplate is locked between the two major plates that rift apart in Iceland.
      #2: Normally activity in one area of the rift zone is unlikely to start a activity in the other area of the rift zone. This is often the case when there is a great distances between the area. But this does not mean that activity can take place at many locations in Iceland at the same time. But it also does not mean that they are connected.
      #3: Vatnajökull and Reykjanes do not belong to the same rift zone. Reykjanes is on the western rift zone in Iceland. Vatnajökull is on the eastern rift zone in Iceland. Höfsjökull might sit on it’s own rift zone (based on the fissure markings on geological maps).
      #4: Earthquake activity in Iceland depends on how much spreading happens in the fracture zone or the fissure zones in Iceland. The most spreading takes place in TFZ (up to 2cm/year I think). But we also have the most activity there because of that fact.

      Here is a good map with all zone names in Iceland.

      http://6224544939848152310-a-1802744773732722657-s-sites.googlegroups.com/site/myvo90/Home/field-trips/iceland/Slide7.PNG?attachauth=ANoY7cpshVdKKvV02b5xMq24Yp6xa90MeUgzEZNXILAgYK03ewNmvS69ZOgiEg3viTiCxF95X6l4XMbiKBukUF1pCxh1BiHFNljFoMjfvmfdCGipkQf-GDTnEDWpISi_6vHK0yB_52Nf0ri3I2BcR-UIk1yPoKAS8tXORYlK-zC1ylplNoxZHCn03qy_luUi5jMGtIufl3tAbdvNc9rZw2DeoruFh2HLgftOWCGq2Y3G6znD7T4iogs%3D&attredirects=0

  9. As I said, I was unclear so it is entirely my fault for being misunderstood here 😉

    My point being was that this time it did influence it, even though it is normally unlikely.

    Remember that this is a microplate that is already pressurized and with 3 corners being inflated at the same time, you will have strain being transfered away from the 3 points of inflatement pressure. In this case it will be disipated in 3 ways shearing along the sides of the microplate, flexing of the microplate, and a depression of the northwest corner.
    And this is allready evidently taking place, something that is proven by the quake-pattern during the last days. This theory is the only one that holds up for now, and since there is evidence (corresponding quakes) it is validated untill proven incorrect.

    I will now comment on your comments:
    #2: I know that it normally does not, but this time the circumstances is a bit unusual even for Iceland. And these unusual effects added up enough for it to cause unusual effects at an unusual place as we have seen in the last couple of weeks northwest of Langjökull.
    #3: No, I know that Langjökull does not belong to the same rift zone, but the 2 different rift zones border the same microplates on 2 oposing sides. That was my point.
    #4: Yes, Tjörnes and Reykjaness Ridge moves with on average 2cm/year. Problem is that the rest doesn’t and hasn’t for some time. And that is the basis of my discussion. This lower movement in the area between these points is causing a lot of stress being built up, and that is being released as either large fractures ripping open during large quakes, or causes large fissure eruptions depending on where they happen. But over geological timespans the motion in the area between Tjörness and Reykjaness ridge is probably also 2cm. (If it is not… Help us all.)

    1. The current situation in Iceland is not unusual in geological terms. In terms if the rift zones you have to release that the eastern rift zone in Iceland is a opening rift zone, it is being formed at this moment. It is a long process that is not going to finish for next several million years or so.

      Intraplate earthquakes in Iceland appears to come in waves. I do not know the reason for it. But there is nothing unusual about it. As the crust has to adjust for the movement that takes place in this area.

      Here is a abstract from a study in the earthquakes in Iceland. They got all the numbers on the spreading correct.

      “The average spreading rate in south Iceland, 19 mm/yr, is distributed over two parallel rift zones. We measured a Global Positioning System (GPS) network of 42 stations in the eastern zone in 1994. This network consists of stations measured with GPS in 1986 and 1992, stations in a 60 km long distance profile measured several times since 1967, and a few new stations. The 1994 GPS data were processed using the Bernese software version 3.5, and the average position uncertainties are about 3 mm in horizontal components and about 8 mm in the vertical component for baselines up to 100 km. Comparison with results of former GPS campaigns gives a uniform extension, steady in time, perpendicular to the spreading axis. A strain rate of 0.12 ± 0.01 μstrain/yr is observed across the 100 km wide network, or an extension of about 12 mm/yr. […]”

      http://www.agu.org/journals/ABS/1997/96JB03893.shtml

      Here is a old study (1989) on earthquakes in Iceland.

      http://notendur.hi.is/~heidi/Data/Herdu-Askja-literature/Einarsson91.pdf

      When the ridge moves, it moves the plate boundary with them. That is how tectonics work. But this process is slow, so you don’t really notice.

      What I am trying say here in not so clear words is that your idea that you base your discussion on is flawed because it is based on the wrong idea that does not hold up to the scientific data. I have made this mistake my self many times when I was starting to learn about earthquakes and volcanoes. But it did not make me right, as I did so learn when my ideas failed completely and did not hold up to reality.

      I also want to point out that stress builds up in tectonic plates when they are not moving (no earthquakes). When there are earthquakes the plates are moving. This goes for even the smallest of earthquakes, they do mean that the plates in the area are moving. Even if it slow movement.

      1. I still feel like I am not good enough to explain myself here more than that I am wrong…

        My point was that the last series of quakes was due to pent up stress.
        And I was not talking about ridge-movement. I was instead talking about the stress being built up due to inflation in the corners (while the ridge-movement is doing it’s thing as usual),
        and that the motion in 3 out of 4 corners due to inflation shows up at the last corner in the form of our late swarm of quakes (while the ridges keap up moving happily).
        Or I might have missed that it is normal with all 3 corners being lifting at the same time? After all Katla started in 99(?) and Askja in 07 and Krisuvik in 08. So ten years ago no corner was lifted, 3 years ago 2 corners where lifting and 2 times the third corner have been lifting since 08. I do not know if I would call that normal (but it has probably happened before).
        What would be cool though would be to check if the same northwestern-corner quakes happened during the last inflation-period in Krysuvik last year. But I canot find those figures…

        I think we are talking around each other mightily here since neither of us are native english speakers. You are absolutly correct in what you are saying, I know that. And I have to start thinking about how I try to say what I try to say…

  11. I have just noticed it is the same at Myrdalsjokull (Katla/Eyja) Last quake went off the radar on the 7th. No new quakes since then.

    Is this something to worry about???????

      2. Not always. The earthquakes which are related to movements of the glacier should come to an end now. So there are less quakes.

  13. Looking at the Vatnajökull GPS data (and removing bumps by snow/ice on antenna)…

    GFUM station has moved slowly UP during the time span of the figure http://hraun.vedur.is/ja/vatnajokulsvoktun/gps_grimsfjall.html and had a suspiciously rapid movement to NORTH at 2010,83-23010,85 (just 2-5 days) looking like an effect of ice on the antenna, but the new level has stayed there. It has also moved slowly to EAST during whole of the time.

    According to http://hraun.vedur.is/ja/viku/2010/vika_39/ and http://hraun.vedur.is/ja/viku/2010/vika_40/ there were plenty of eqs at the Hamarinn-Bardarbunga system during those weeks.

    The rapid movements at SKRO during the last days look very much like effects of ice, too. We’ll see, if the new level stays there for e.g. weeks…

    Until we see, what’s going on there, I suspect that GFUM movements are really related to magma movements (inflation) at the Hamarinn-Bardarbunga system, not Grimsfjäll/Grimsvötn system.

Comments are closed.