Deconstructing Hekla – Hells Gate Revisited

Image by unknown. This I have been saving for sometime for you readers. I know that several of you are equaly interested in Auroras, as you are of volcanoes. Well at least almost. So here is alte christmas present for you.


A few days ago I had one of these moments when it felt like my brain was coming apart. As usual when this happens and volcanoes are involved, it also involves the eminent volcanologist Sturkell. For this reason I have rather grumpily renamed him from time to time into Sturkleton the Confuser.

Why then does he confuse me? For no other reason than that he has a wonderful knack of taking “well known” Icelandic volcanoes, and remakes them with an offhand remark that sends me reeling. This time he did it in style.

I have prided myself of actually knowing a bit about Hekla, and that I had read about every word ever written on the subject. Well, I had read almost everything, and understood less than I thought. All the facts were there for me to see, but apparently not to understand. Then I find a hidden away page with notes on Hekla by Sturkell in Swedish. And then I was thoroughly confused. By the way, the page is referenced below, so you can translate it at your own risk.

When I started learning about volcanoes I did so by reading Erik Klemettis blog, and the first thing I read from him was a hint that one should always start the learning process at Global Volcanism Programs list of Volcanoes. After a while I started noticing that it had its errors, but was good enough as a quick check. Guess if I got short-changed on Hekla by GVP.

My brain works in a way that demands that I try to understand what is going on, and how to expand my knowledge from what is known, otherwise I get bored. So I decided to learn how volcanoes work. I had by then found a volcano that kind of had every odd behavior a volcano can have. So I thought I could understand almost any volcano by studying one particular complex volcano. Of course I am talking about Hekla.

Classic Hekla

The image of Hekla is of a fissure volcano that is in an intermediary stage between becoming a stratovolcano and being a fissure. It erupts in a very complex pattern showing a wide variety of styles and behaviors. It erupts anything from acidic andesites to your basic unevolved basalt. Classic Hekla is truly the volcano of volcanoes. It beggars one to find even one volcanic behavior it has not exhibited true the ages.

Image: Sturkell

What is a volcano?

This is a question we should ask ourselves more often I think. I here admit that I am a bit damaged by Klemetti.  I tend to see a volcano from the perspective of what it erupts. This view comes from the opinion that a volcano quite often behave according to its type of magma. It is most often a good view, since often it is true. Yes the magma can change over time, but it almost never change from one basic type into a completely different.

Milkshake in Ireland

A few years ago I visited Dublin and saw a milkshake bar. I will use that experience as an analogy since they served Guinness.

Almost every volcano on Iceland produces the same general type of magma. It ranges from the normal Iceland basalts, to more evolved types of magma that through various processes have evolved out of the original basalts. Each volcano produces different lavas, but they are kind of versions of each other. And they are also slightly different in content of metals and rare earth minerals. So, one can discern one volcanoes magma from another volcanoes magma.

One could say that they all contain milk and ice-cream in various amounts and that they have different flavors. Yes, it can sometimes be hard to know if a milkshake tastes raspberry or strawberry, but it is no problem to taste the difference between chocolate and vanilla.

Katla and Eyjafjallajökull are examples of Strawberry and Raspberry, especially around Fimmvörduhals. But it is easy to taste the difference between Katla and Krafla. So you can see the Eastern Volcanic Zone as a long line of Milkshake machines.

Imagine than that right to the side of them you find a tap, pull it, and out comes Guinness. It contains neither milk, nor ice-cream. So what is it doing in a Milkshake Bar for kids?

The point of the analogy is to show in very easy terms how large the difference is between Hekla proper’s two types of discharged lavas, and the others. The first phase of Hekla proper is to violently erupt acidic intermediary ash and pumice; the second phase is a calcium-alkali andesite. So Hekla proper is your basic subduction volcano regarding its magmas.

Image: Sturkell. The plot is showing the cilicic content during eruptions. Please not the very small increases in cilicic content during the current unusual eruptive cycle in Hekla.

Eruption mismatch between Sturkell and GVP

Sturkells list of eruptions in Hekla proper is five eruptions shorter than GVP counted from the first day of settlement. The reason for Sturkell using the settlement as a starting date is most likely the Icelandic knack of writing things down. So we really know a lot of the post-settlement eruptions of Hekla.

Let us start with Lambafit and Lambafitjahraun in 1913 about 15 kilometers northeast of Hekla Proper. The eruption did not start with a large ashy explosive phase. It was slightly explosive throughout, but not that much, it produced a fissure eruption at Lambafit and Mundafit. The magma erupted was of the general Katla area lavas. We can also notice that Lambafit is outside of the known extension of the Hekla Proper fissures reach. Knowing that a milkshake does not come out of a Guinness tap we can safely assume that Lambafit is a different volcano with a different origin and build-up than Hekla.

Also the eruption before that at Krakagigar 1878 is not of Hekla type, also the 1725 eruption is going down the tube for the same reasons.

Then we are back at the eruption of Raudubjallar 10 kilometers southwest of Hekla. The same wildly different magmas, the same low explosiveness, but this time we got a crater row. Then we can cut away the 1440 eruption too while we are at it.

Between 1104 AD and 1100 BC (H3 Tephra) Hekla did probably not erupt. There are a few non-explosive eruptions attributed to Hekla from 650 AD up to 1104. But they should have their magmas checked really.

Now instead it is Vatnafjöll volcano that becomes the active member as we go backwards in time. Vatnafjöll is large volcanic band that starts to the southeast and then continues in southern trending band until it ends up to the SSW of Hekla. Vatnafjöll erupts the same magma that Hekla does under the second phase of the Hekla Proper eruption, but foregoes the initial explosive Hekla magma type entirely. It is also significantly lower in Fluorine content. But it is the closest in magma type to Hekla in the area. Instead of Guinness you could call Vatnafjöll a Killkenny Beer. Vatnafjöll then continues to be the main erupting volcano in the area back to de-glaciation. These eruptions from Vatnafjöll were interspersed by the vast tephra eruptions of Hekla that is used in to date Icelandic eruptions through tephrochronology.

Before that it seems like it was the badly known Búrfell volcano that was active during glaciation.

Image rights reserved by: Séverine Moune, Olgeir Sigmarsson, Thorvaldur Thordarson & Pierre-J. Gauthier


As you all know Hekla proper is more explosive the longer it has been dormant. The reason for this being that the cilicic content increases in the intial phase magma being erupted out of Hekla if she has not erupted for a long time. The longer the intermission, the higher the cilica grade, seems to be a good rule of thumb.

So if you remove the 1913 and 1878 eruption, then the intermission before the 1947 eruption becomes 102 years. Same goes in even greater aspect for the 1104 eruption of course. Not to mention the large tephra eruptions during the period of Vatnafjölls activity.

It also has an impact regarding Heklas explosivity. We are now left with only 1 eruption as small as a VEI-2 from Hekla Proper. The removal of the non-Hekla eruptions has significantly increased the explosive factor of Hekla. One should keep in mind that the large tephras from Hekla are VEI-5s and VEI-6s. So, the average becomes a rather hefty VEI-4. And it kind of explains how Hekla is able to keep up with delivering VEI-3s every ten year now.

And, this very high rate of eruptions that we are seeing now, it is in this new light unheard of in the history of Hekla. Also note the very small increase of cilicic content between eruptions.


It is likely that Vatnafjöll and Hekla share the same basic magmatic origin, but that Vatnafjöll lack one component or more in what is going on inside of Hekla proper. So, we can see those two as a volcanic field in their own right.

Then begs the question of the other volcanoes that almost surround Hekla Proper, and also has the nasty habit of erupting in between the two known volcanoes of the Hekla Field.

I have a problem believing that there can be one driving force behind these non-Heklic volcanoes. They are too distributed to be getting the magma from one single place. Also, there seems to be different flavors to the lavas being ejected.

I guess you could explain it with it being two or more surrounding volcanoes. The locations of Lambafit and Raudubjallar and the general look of them make them into most likely being volcanoes in their own right. One should though note that if you draw a line between them you get a kind of fissure line. So I guess those two could explain most of the eruptions.

Krakagigars flavor is of a third type, so perhaps one could just tentatively say that they are a second volcanic field that co-exists with the Hekla Volcanic Field.

As you notice the separation of the two types of erupted lavas like this reshape the image of Hekla a lot. It also at the same time simplifies the behavior of Hekla and quite frankly makes Hekla more believable and a tad more understandable. But at the same time it raises loads of new question about what is going on down below in the general Hekla area.



Sturkells online notes;

And of course Carmichaels groundbreaking work on Icelandic intermediary magmas. Good read for anyone who has been wondering what Thingmuli is about.

694 thoughts on “Deconstructing Hekla – Hells Gate Revisited

  1. From AVCAN’s FB:

    Note 418 AVCAN – earthquake activity – VOLCANIC – island of HIERRO – 14 January 2012-11: 00 pm peninsula – earthquake and volcanic activity continues moderada-alta, with a bit of seismicity in the interior of the island. Signal of volcanic tremor continues in the eruptive zone of la Restinga, stays more or less with an amplitude which in general terms this up a little bit, especially the last few hours. Magnitude between 2.5 and 1.3. New earthquakes 4. Depth to 10, 12, 16 and 16 km. yesterday 2. Yesterday (4). Today will be 1. In total van 11998 earthquakes located in El Hierro by IGN from 9: 00 pm of the day 19 July 2011 (Henry).

    A curious fact AVCAN Facebook has also 11998 people that I like… thanks to all those who have done so possible… ^ _ – (Translated by Bing)!/pages/Actualidad-Volc%C3%A1nica-de-Canarias-AVCAN/163883668446

    The ratio of friends to EQs is approx 1:1.

    • Also: Volcanic news de Canarias (AVCAN) new post for comments on WEBCAMS. La mancha in the water is perfectly, as well as the area of the bubbling in such a way that are occasionally pyroclastic floating in the form of black spots that Miss white steam to reach the surface, yesterday were some… (Henry)

      Pity they don’t give the time for their post.

  2. Avcan FB

    Here. also see a sismito before 15: 50, does the horizontal blue line on 11 Hz indicate that noise by the release of gases that you indicabais in the photos of AVCAN? Thank you.
    See Translation

    • I do not need more than Lurkings well written thingamabit on how a tsunami on a large scale could not happen.
      Guess what, Lurking is right (as he normally is…), a large landslide could not devastate the US, not even fairly close by places would be destroyed.
      But, I could never describe it as well as he did.

  3. Carl
    Just a point though one did happen hundreds of years ago from La Palma and it did reach the east coast of the USA.?

    • Yes, I have read about that one. But as far as I know the reaching Florida part is rather dubious. And even if it did it was not higher than a normal wave in the area…
      One should remember that the tsunami in japan, and the boxing day tsunami was enormously much larger than anything the canaries have ever put out.
      I would be more worried about a 30 kilometre cliff side under water breaking off… And that will not happen in the atlantic. Wrong geometry.

  4. “Luis Ignacio Gonzalez de Vallejo, Professor of Geological Engineering from the Universidad Complutense de Madrid said , that in the Canary Islands there have been waves of up to 50 meters created by large landslides” (La Vanguardia translated)

    • Possibly. But big landslips (slope failures) can’t happen in the same place twice because there is no mechanism to replace the land that has slipped.

  5. As far as I understand in the past there have been landslides on El Hierro at :

    San Andres
    El Julan
    Las Playas
    Henry Seamount

    • Judith, check this paper:
      Masson et al. 2002, Slope failures on the flanks of the western Canary Islands,
      It’s linked up in the page on “Wonder what Bob is up to?” with some other relevant papers.

  6. In the same way he did, a few days ago, responsible for disseminating scientific of the Instituto Volcanológico de Canarias, David Calvo, expert in geological hazards, Luis Ignacio González of Vallejo (Professor at the Complutense of Madrid) also confirmed the existence makes tens of thousands of years, even millions, of the ‘Megatsunami’ (large tsunamis, prefers to call them González Vallejo #.)
    “Tsunami must not have been, geologically speaking, a rare phenomenon, but rather occurred with some frequency in the last million years.” “Within this period there have been several”, explains Professor.
    However, this specialist in geological hazards ensures that “it is pure speculation that a tsunami originating in the Canary Islands to reach the Atlantic coast of the United States in the past.”United States “, as some urban legends are.” “There have been many landslides in the Canary Islands, but no one has said that the deposit found in the Atlantic coast of America reflects a tsunami caused by such landslides.” “It is pure speculation,” wields González Vallejo.
    They are various assumptions indicating that approximately 300,000 years ago, a landslide occurred in El Hierro ended up generating a megatsunami that swept the Atlantic coast of America # seriously affecting the place where today lies the city of New York #.
    There are even researchers, two Englishmen in particular, indicating that in a future eruption of the island of La Palma, the flanks of this will end up sliding by causing a large tsunami that, again, would a brutal force the United States causing havoc.
    This professor at the Complutense University of Madrid has eight years, together with other colleagues, investigating large landslides occurred during the geological history of the Canary Islands.
    “We have studied the landslides of Güímar and Orotava, absolute datings have been made, and we have a rough idea of which 600,000 years ago, in the case of la Orotava, a million years, in the case of Güímar and 150,000 years, in the case of Teno, there were great landslides.” The latter is associated, probably to sliding and explosive volcanic eruption of the pre-Teide, which had as consequences a large landslide which produced a large tsunami. “And we have found remains of this tsunami.”
    “We know that the slide of Güímar, possibly reached the island of Gran Canaria and produced a tidal wave #… # the wave overtook the 40/50 m”, sentence González Vallejo

    • 300,000 years ago a landslide on El Hierro that caused a tsunami that impacted NY.

      So the known collapse of El Golfo 130,000 years ago did not? So if you live in the US, it seems that you do not have too much to worry about from El Hierro.

  7. Megatsunami – would need a “mega volume” slide… And then there remains the geometry issue of the Atlantic. But no need for a 40m wave to see the eastern coast of the US in front of some trouble.
    Probably the next problems for the northern hemisphere will be economic / social rather that caused by natural hazards… Although a reminder that we are small and insignificant on the planetary scale could stop some modern behaviors that I can’t like… Look at how the blue ball looks like. There’s more that happened than a La Palma landslide… 🙂 As we just can’t predict the really fat stuff, we should just prevent what we can localize and manage, and learn to simply live with the risk for all the rest. Whatever happens – that’s life… I pick myself up and get back in the race… All you need is a Swiss knife and tape – and a Jeep. Wow, a McGyver flashback. I need shleep. Good night crazy bunch of good people.

    • Hi All, posted a few weeks ago, family and xmas got in the way of posting again. I have been lurking as the mother in law is now on holiday in the canaries, wanted to assure her she would be OK!

      Seems Bob is not done yet, his bowels are still playing up, much like mine after a heavy night last night, thanks for that 2nd bottle of red Paul!

      In order for the worlds financial system to become more stable, many things need to happen, the best of which would be one single worldwide currency. You can blame Iceland for kick starting the current situation, borrowing from Japan for low interest then loaning that money out, until the exchange rate went t!ts up this was a good plan. This was the impulse (sharp rise and fall in a signal) that rang around the worlds financial markets. Remove most of the variables, then it would become predictable. Not that this will happen.

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 )

Google photo

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

Twitter picture

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

Facebook photo

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

Connecting to %s