Human made earthquakes in Hengill volcano

It appears that operations of Orkuveita Reykjavíkur are creating the earthquake swarms that have been seen in Hengill volcanoo. But as part of operations in Hengill geothermal plant they pump down cold water into the bed rock under considerable pressure. This is according to Sigþrúður Ármannsdóttir a geologist at Icelandic Met Office. But Ví quoted here on this earthquakes swarms in Hengill volcano.

This type of earthquake swarms are going to repeat them self for as long Orkuveita Reykjavíkur is pumping down cold water into the bed rock in Hengill volcano. As the water creates a pressure change in the surrounding rock.

The most intresting earthquakes that happened as because of this swarm where 3.8 km NE of Hveragerði. But the size of that earthquake was ML2.6 (automatic size). and it was considerable distance from Hengill volcano and appears to have been triggered by the earthquake swarm in Hengill volcano (unclear however and just my own observation).

News about this.

Orkuveitan framkallar jarðskjálfta í Henglinum
(Ví, Icelandic)

108 Replies to “Human made earthquakes in Hengill volcano”

  1. It can also be seen, that these earthquake swarms near the powerstation appear only during the working hours. The stopped last friday afternoon and continued on monday morning.

  2. Hmmm! I wouldn’t like to have a drilling station like that in my backyard.
    Who knows what consequences it would bring?
    If there are intrusions (which seems to be the case) and fissures, what kind of stresses the swarms could bring to other areas, like Krisuvik and Hekla which are not so far from this spot?
    I am not saying there is any connection, what I am saying is: if you start gently shaking a box of fizzy bottles, it’s likely that one of them might go boom.
    107 earthquakes is more than a gentle shaking. No good for areas that are in the fringe of an eruption. Hmmmm…. (OK: CAVEAT – Stupid amateur remark)

  3. Well if the incident a while back when drilling stumbled across some rhyolite magma the results could be…interesting. 🙂
    When the magma is contained it is quite docile and the buildup towards an eruption takes a longer time.

    If you by mistake poke a hole and release the pressure the magma will collapse and there will be room for serious gas buildup within the magma..Just like the “fizzy bottle” thing mentioned by Renato. Magma + gas = no afterwork that friday. 😉
    All this depending on the size of the drilling of course and if there is several drill sites and so on…

    CAVEAT: you know what i mean..Not vulcanologist and so on..yadyada…

  4. Either it is not true at all, or they have in secret done something so vastly dangerous that I would be running for the hills if I was in that part of Iceland.
    I will write to long posts about it, one why I do not think that this swarm is entirely made by the cold water. The second one will be about if it is and what the consequnces could be.
    At the time being, contemplate cold water under pressure hitting a magma-chamber that is located on an active tectonic faultline.

    1. Kind of scary scenario: Every time there is a quake swarm going on, those poor IMO seismologists have to call all the thermal power plants in the whole island just to make sure, they are not doing those little Monday-afternoon-fun-lab-runs again…

  5. Part 1:

    Here are the reasons for me not believing that the swarm is manmade. Yes, some of the quakes might have been that. And with that I mean the Hellisheidi quakes of less then 2 km depth. The others might have been induced by them, but I do not think so, and if they are they should emediatly suspend pumping operations.

    1. Sofar as I know the boreholes at Hellisheidi are not deep enough to cause the deep quakes, nor is the volume pumped large enough to affect anything outside of the powerplant. I will return to this point in the next post.
    2. Timing is wrong, the quake swarm started at 2 in the morning and ended at 12 today. That is not consistent with working hours.
    3. The area affected is quite simply to large.

    Remember that what they are saying is that the water pumped down has spread 8km down (into the MOHO) and covers more than 40 square kilometres. That is one hell of a lot of water needed to seep that far down and away… If it is true after all they have lost controll of the process.

    1. Totally agree and look how fast they were to “accept responsibility” in the paper for all the quakes.

  6. Do they line their pumping holes of is this filling rock fissures with high pressure steam?

    1. This is the rock’n’roll version.
      They do not drill two holes, line them with pipe, connect them at depth and pump water around in a closed cycle.
      The version we are planning/planned was to drill two holes into a cavity, pumping water down one tube and having steam coming up another.
      If this is what happens we will scrap our plans, or go for a closed system with lower efficiency.
      I stated that this system is dangerous 2 years ago when we looked at the Hellisheidi and its future plans, but not even I didn’t see this one coming.

      1. @Carl: Hey, it’s impressive that you even brought it up 2 years ago, let alone got so darn close. Who knows what’s next, these volcanoes are so unpredictable.

    2. This is cold water, no steam. The cold water goes into the bedrock and warms up because of the magma below. Then it rises up again and then it comes out as hot springs and steam that powers the geothermal plant in the area.

  7. Part 2:

    The one of two possible reasons for a human-made quake swarm is that they unbeknownst to everyone have drilled into semi-sollid cooling magma (rhyolitic mush?) like they did at the Iceland Deep Drilling Project at Kraflavirkjun.
    As I have stated before we have judged it to dangerous to do that when the chamber is situated ontop of a faultline. We where very carefull to find a volcano that did not do that since the risks of water running away would be to large. And we are actually considering to scrap the project even before one borehole is drilled due to the dangers with this.
    But if they have drilled that deep they are right now hitting the cooling magma deep under Hengill with cold water. But, I do not think they are that stupid and I have seen nothing indicating that they are doing it. But, they might have hit very close to a pocket of magma that seeped up during the last infiltration phase.
    Back to point one in the post before. What might have happened is that the faultline and the surrounding bedrock is very fragile due to it pretty much being a triple-junction separating the MAR. Then the water pumped down could have seeped away and started to widen the cracks of the existing dyke that is allready infused with a new infiltration at root-level. It seems to also have spread away upwards to Hengill proper.
    If this is the case they might be inadvertently on the brink of blowing open the conduit downwards with clear signs of it.
    I would seriously say that they should suspend pumping and thoroughly analyze what is happening together with MET. Then they could see if A) Quakes die off after suspension, and B) after analyzing the quakes and determining that some of them arent magmatic, start to pump again, but slowly increasing pressure to find the treshold of stability. Because right now they are playing with the dices.

    I think I will reschedule and continue to Iceland on my way home from Shanghai to see what the hell they are doing. And there went my weekend to hell.

    1. @Carl: “playing with the dices” is probably closer to Russian Roulette!
      This is all very frightening.

  8. OK, Carl.
    We’ll get you some Müsli for compensation.
    Very good reasoning there, I think.
    I suggest that you also post your comments over EB, because this is todays subject over there. 🙂

  9. What kind of working hours do they have on this project? 24hrs / day in different shifts? Would be plausible since an operation like this would probably have a high cost / hour.

    Anyway that was a side track. I still see some tremors going on at both sites. Hengill and Krysuvik.

    I tend to go with Carl´s theory that it might have been man made tremors at the beginning but I am starting to worry that they might have set off a chain reaction that is accumulating. Have there been sufficient investigations as to how this operation would affect the existing fissures, dikes or any such geological structure? Is the area well known when it comes to these features?

    My point beeing that these are volcanoes that has not erupted in a very long time which makes the knowledge less than that of more active areas. I have no doubt that there has been investigations on how these work and I am in no way diminishing the work of the icelandic geologists.

    But when push comes to shoves all they have is rock, old lavafields and topography to look at. And based on that build theories. Who is to say that the magma hasnt changed characteristics? Exactly how are the dikes spread out? What kind of fissures are there down below?

    Many questions and i truly hope they have all the answers. 🙂

    1. Don’t I remember that the first of the earthquake swarming in this general area last year started just out to sea, then drifted northwards and a wee bit eastwards. In which case it is just that enormous area of shattered fault boundary which stretches up iceland from the ridge shifting a bit and nothing to do with the power plant.

  10. Also I don’t believe this explanation because there are two swarms, one just south east of each of those two lakes many miles apart. They are not drilling at both are they? To me it looks like that giant faultline of fissured area is creaking a bit due to quite natural processes.

    1. The other one is Krysuvik which is not new. This area has been inflating and probably has a magma intrusion in it’s magma chamber, but this has been going on for some months. (years actually)

      The only connection between these two is that volcanism near the Reykjanes peninsula happens in cycli. Last one was during the Middle Ages and there are signs that a new cyclus is about to start.

      1. They seem to be on the same mass of parallel faultlines stretching virtually right across iceland on this map:-

        Click here

        The whole lot seems to creak on and off all the time, irrespective of pumping stations.

  11. Strange place for an EQ 6,1 km SSV af Seljalandsfossi, out by the shore “sandur”. It´s neither big nor deep, I just got curios.

    1. Could be explosion, they are moving how the glacier water flow from the (glacier) river down to the shore to avoid sand collecting in Landeyjarhöfn (a harbor on the south coast).

  12. I remember reading about similar earthquake swarms at geothermal plants in California and Colorado here in the US, as well as I think in Switzerland. It may be interesting to look up articles on those events to read more about it. I’ve asked my brother who’s a geologist working in the natural gas industry here to see if he can dig up any information on the events here.
    Also for those doubting if it could be due to pumping water down. He did point out some news about earthquakes in Arkansas and Texas caused by hydraulic fracturing of natural gas wells. I am not sure how deep or how much water they are pumping down there at Hengill. In Texas they pumped down some several million gallons of water, around 2.5-3.5km deep. Several earthquakes up to 3.6 mag were determined to be caused, and there’s no volcanoes and no active faults in the area — unlike Iceland of course. So it doesn’t seem implausible that we can see this kind of activity in a much more active area.

    Now, if it’s safe or not, is a totally different question.

    1. There is a nearly endless earthquake swarm at The Geysers in the Clear Lake volcanic field north of San Francisco, CA. There are also frequent swarms at Coso Field in southern California. Both of these locations have hydrothermal power plants where they pump water in and get steam out.
      These quakes rarely exceed 3.0 in magnitude, and are relatively shallow. (<6km deep)
      The Long Valley Caldera and Yellowstone often experience similar swarms, probably due to natural ground water. I believe many of Iceland's quake swarms occur through the same mechanism.
      It is also well known that draining an aquifer can cause quakes.

  13. The EQ is probably going to be moved. The accuracy is at 33 or so. When someone takes a look at it from IMO it will get revised. If someone will look at it that is. 🙂

    Just a “ghost” in the system. 🙂

  14. What’s up with the strain gage at HEL ? And does anybody have a map showing the locations of the strain gages? I’m assuming that HEL is Hekla.

    Note by Jón Frímann: I had to make a minor edit to this comment to allow it to pass from pending status in WordPress. Sorry for that to the person how made this comment.

      1. I just got it started and just after the earthquake in NZ. But I would not think so. As NZ is on the other side of the planet. In Iceland and Denmark I am not going to record anything below Mw7.8 that happens on the other side of the planet with high frequency geophones like the one I have. I am going to need at least 10 second period seismometer to see earthquake like the one in NZ here in Denmark and Iceland.

        Thanks! I it good to have something up and running. Now I just need the PC monitor for my desktop computer and my server computer. I got the monitor for the server computer, but no power converter for it.

  15. LSF is a mess. Hard to tell what it’s doing. I can take a look at it if you like, but I doubt I will be able to glean anything useful.

    1. Thanks but no need to work on this. Only thing you get from seismologist studying these faults is the same conclusions…its a mess. Aftershock pattern to 7.2 is anomalously high almost 1 year after and that this is “worrisome”

  16. i think this one was missed somehow… ok, it has a low quality, but it is still something, evn if its a shaddow. most intersting for me is the deept of the quake

    22.02.2011 01:43:42 64.000 -19.473 14.0 km 0.9 50.11 9.5 km E of Hekla

    1. Now… is it deflation? Or is it the crust segments are no longer wedged as tightly together? (stress moving on to some other area)

    2. If you look at the trend, it has had similar “deflation” periods before (on top of the general inflating trend).. So this is very likely a similar one “bump” in the trend.

      I’d not talk about deflation. I’d rather talk about “inhaling in-between to continue blowing with greater force again”.

      1. I do agree with you that there is that possibility also. I was talking about a possible deflation-period.
        Time will tell. 🙂

    1. That’s no deflation, just a very very very tiny inimini winkipinky ground deformation but I’m pretty sure this isn’t gonna stop Grimsfjall from inflating further.

      The quakes are rather interesting though, they seem to get more and more common. Still no large quantities but over time we can see that the quantity and the magnitude of the quakes at Grimsfjall slightly increase.

      (Cumulative seismic moment)

      1. The red and blue line look almost like twins or look like two drops of water! The red line is (almost) an exact copy of the blue one!
        Very interesting!

    2. Statistically speaking, this is not even a fly’s fart! Within error bounds it’s the same as yesterday.

      1. I have to get them from my geophone close to Hekla volcano. I did record them at that station, at least the larger earthquake. That is how I get the wave from data. From my own hardware.

      2. Jon: Thank you for your Reply to my question.

        But: Is there any way that IMO – or the Icelandic Institute which checks the earthquakes – has a webpage or something similiar where they provide the waveform of icelandic quakes?

        like the official webpage of the Swiss Seismic Service from the ETH Zürich.

        Index Earthquakes Switzerland

        You can then click on a quake, on the new page on science and technical and you can view the waveform of the quake.


      3. Yeah… Jón is the best source… the nearest station available via IRIS is in Reykjavik.

  17. Hmm! Could there be a small earthquake swarm starting in Eyjafjallajökull’s crater? I guess we will have to watch to see if there are more quakes. They are all very very small. Are they earthquakes?

    1. I think these are quakes caused by the cooling magma and/or ice quakes. (Since they are very shallow and the small magnitude)

      1. Hello Pieter,

        If it is magma which has been cooling down since the end of the eruption last year, why don’t we see earthquakes at lady Eyja more frequently, say every day. I won’t believe that on several days the lava at Eyjafjalla stops cooling for a while (when there are no quakes).
        I saw 2 quakes there: one at magnitude 1.1 on 1,4 km deep and the other one at magnitude 1,5 on 4,1 km deep. I don’t know if size and depth still has to be checked, but you say it’s cooling magma in the crater. Surely the crater isn’t 4,1 km deep, is it?
        Also 2 quakes at lady Kat: one 2,1 magnitude on 1 km and one 1,3 magnitude at 2,9 km depth.
        Do you or does anybody have an idea what kind of quakes they are (for example ice?)
        Kind regards,


      2. Magma cooling itself isnt creating the earthquakes, it’s the mountain itself resetting itself into a natural position after alot has blown out. This does goes deep down into the roots which also cool down and settle into place.
        (This is my guess)

  18. I would be more worried about Grimsvotn. I don’t think the crust is able to take the pressure from beneath anymore, and I think the increase of earthquakes is a result of that. My amateur guess would be less then one month, but hey who am I. 😀

  19. I believe that all those small quakes we’ve been seeing recently are due to overall improvements in Iceland’s monitoring system .
    They are but small creaks and squeaks from frost-defrosting within unimportant cracks.

    1. I beg to differ. We have now two deep quakes at 4.1 and 3.7km. The last one hasn’t been reviewed yet, but they are too deep for frost quakes. The are all almost on the same spot. It would be interesting if these quakes appear on Jóns helicorder or if they are too weak.

      1. They did not appear on my helicorder at Hekla geophone. The lowest limit that I can detect for Eyjafjallajökull volcano is ML0.9 at distance of 43 km. But then the weather has to be really good.

  20. I wanted to let you know (my readers) that I have decided to move to Germany when the following criteria is met.

    1. I have enough to pay for the move. Both the rent insurance and the firsts months of the rent (2 or more months?).
    2. Found a apartment on the ground floor. Any help with this problem are welcomed. Just send me a email or comment here. Thanks! 🙂
    3. When I know some german. I currently live close to the German borders so this is a less of problem but when I was living in Iceland.

    I plan moving to a town named Daun in Germany. But this town is on top of West Eifel Volcanic Field (

    I am going to start collecting money for this in April and after that I am going to collect the donations and the ads money that I get into this fund so it can collect the money that I need as fast as possible. The current time frame should give me about two years until I can move to Germany and start to monitor the West Eifel Volcanic Field.

    I know that I just moved to Denmark. But I work in the long planning idea. So now I can setup my self for a short stay in Denmark and make sure that I don’t get focus on what I need to do so I can move to Germany.

    I also want to thank for the support that I have gotten so far. You are the best. 🙂

    1. Hi Jon,

      Eifel is an area quite near to where I live here in Cologne, so if you have questions about the area just ask. Daun is a nice small town but if you don’t have a car it may be better to settle down in a town or village near the railroad Cologne – Trier. Good luck in your plans.

      1. I don’t have a car and I don’t plan to get one. So your suggestion is welcomed on this issue. I do not know how public transport is in smaller towns in Germany.

        If you have any ideas on how to find a apartment to rent I would welcome any such suggestion.

        This plan is a subject to change (town location, not area) so Daun is just a idea at the moment.

      2. In that general area the town of Gerolstein has probably the best train connections both to Trier and Cologne. I don’t know people living in that area now, but when your plans come closer to the point I can look for newspapers from that area. I go for mineral collection trips some times per year in that area, so contact me when the time of your move comes closer.

      3. Thanks! I just hope that you will be around in one to two years time (at the longest five years).

        In Denmark they have renting companies (or something like that) that rent apartments out to people. Are no renting companies like that in Germany ?

        It is also nice to know that the area has interesting minerals. Does it have fossils also ?

      4. Its actually more common in Germany to rent an appartment than to buy it. A lot of appartments are owned by private persons, who rent them, others are rented by or through companies.

      5. @Chris, Here in Denmark they have house renting companies called bolig. They rent many apartments to people.

        I have no means to buy apartment yet. So that is not a option for me.

        Here is the company that I rent from in Denmark.

        I would be looking for some similar in Germany if it exist.

      6. I’ll probably be here for many years. There are fossils in that area, but I don’t know much about that, sorry.

  21. That wagging bit is interesting. Thanks.

    (From a related article on the same subject)

    “For minutes to weeks before eruptions, tremors in nearly all volcanoes stay in a narrow band of frequencies from about 0.5 to 2 HZ. Just before and during the eruption, the frequency climbs to a higher pitch, and the range spreads out to between 0.5 and 7 HZ. This similarity in tremors has been hard to explain because each volcano differs in many variables such as physical structure, magma composition or gas content.”

  22. There are also a lot of tremors there in all frequencies

    Adjacent stations also show some but less tremoring so it looks trustworthy. But not earthquakes so far.

    Are there any indications what signs Katla would give before an eruption? Is it like Hekla that just goes boom without a warning or are there many signs?

    1. It did this last autumn (October I think) too. It did not lead to eruption then. It looks a bit scary though!

    2. Morten, Katla from my readings would likely have some larger quakes over 3.0 with smaller quakes becoming shallower over a 12-24 hour period as the magma breaks free. Once the large quakes show up, things escalate pretty quickly like line outside the huer house on dollar day.

  23. I don’t think the run up to Katla’s last significant eruption was documented unfortunately. Maybe someone else here will correct me.

    1. Treacleminer:
      I’ve seen this pattern before.But at the time Jón saw something peculiar in it too.

      1. The tremors are dying down again. Hopefully nothing nasty will happen. We could do with Jon and Carl (no offense to anyone I have forgotten to list) to interpret what happened.

      2. Only for a short while, they are back again now, but the do not seem to be increasing. Are there any strain meters nearby?

      3. Tremors are rising in many other stations like Krisuvik, Vestmaneyer and Grimsey.
        I don’t see much wind right now. Could be ocean waves?
        Well, as I said, not new to my lay eyes.

      4. and as Sander says, there is a sharp apparent movement Upwards and to the South East,, in fact it is off scale on the upward axis. Of course it could be an equipment fault/ice or whatever.

  24. I’ve been noticing a pattern of deeper quakes around Katla for quite some time now but they are usually small and I think it was mentioned they could be attributed to icing.

    1. At the bottom of Sigrun’s chart is the line:

      White points are probably affected by ice and/or snow on antenna and should be ignored.

      All the last points in this chart are white points…

  25. Tremor increasing again now. Perhaps there is something real and is not yet over. Perhaps not anything too worrying though.

  26. Good news. I did order a ADSL2+ connection today. So everything is going to back to normal soon. But I hope to buy the tables for the computers and the tv after the weekend.

    I am going to get the ADSL2+ connection in about two weeks time or so.

Comments are closed.