An Overview of Eruption Types

 Strombolian eruption of Kluychevskoy September 10, 2013 (Kvert webcam capture)

Strombolian eruption of Kluychevskoy September 10, 2013 (Kvert webcam capture)

From time to time when we discuss our beloved volcanoes, we get a mild-mannered enquiry – ”Excuse me, but what’s effusive? And that cryptodome you are all speaking of, what is that?” I thought I’d take the time to jot down a few notes trying to explain what is meant by the various types and what characterises each one of them.

First of all, the term eruption is sometimes used to describe any type of volcanic activity that results in a visible emission of ash and/or steam. In a comment on FB, Dr Boris Behncke recently pointed out that an eruption is when magma reaches the surface. From this definition, we can immediately distinguish between two main types – eruptions and failed eruptions.

Failed eruptions

A great example of a failed eruption; the Half-Dome pluton of granite revealed by millions of years of erosion. Yosemite National Park, California. (WikiMedia)

A great example of a failed eruption; the Half-Dome pluton of granite revealed by millions of years of erosion. Yosemite National Park, California. (WikiMedia)

When magma moves through the crust, it does so along conduits. These can be pre-existing or the magma may make new ones. If such a conduit is oriented sideways, i.e. the magma does not move upwards but sideways, we speak of it being either a dyke or a sill. The difference is that a dyke also extends in an up-down dimension like a book standing on its end and may cover quite large distances as evidenced by the 2014-5 Holuhraun eruption. The extension of a sill on the other hand is flat or horizontal like a book lying down.

Magma tends to collect in what with time becomes vast underground chambers. Logically the term for these is magma chamber. If there is no eruption for a long time, and we are speaking of a period of time many thousands if not tens of thousands of years, depending on the size of the chamber, the magma cools and solidifies to a point where it no longer is able to erupt. Once solidified, it is often referred to as a pluton. The Yosemite National Park in California is a splendid plutonic landscape where millions of years have eroded away the overlaying layers to reveal the ancient, solidified bodies of magma.

Err… How did she get here?!?

Err… How did she get here?!?

Because magma, being much hotter and gassy, has a lower density than that of the deeper bedrock, it moves or floats upwards to a point known as the point of neutral buoyancy where it’s density is at equilibrium with the surrounding rocks. On Earth, this point is reached at about four km below the surface. Sometimes, magma is energetic enough or contains so much volcanic gases under high pressure that it may move above this point. Most of the time, this does indeed lead to an eruption. But if it does not, i.e the magma moves to within ~1 – 1.5 km of the surface but does not erupt and begins to solidify, we speak of a cryptodome (which is similar to a lava dome, only that it is invisible below ground).


When magma reaches the surface, i.e. comes into contact with air as it moves above ground, the result is an eruption. The type of eruption, i.e. the manner in which the magma behaves as it comes up out of the ground and in our somewhat muddled terminology becomes lava = magma that has been erupted, is determined by two main considerations – its chemical composition and its relative content of volcanic gasses; primarily water but also carbon dioxide and sulphur dioxide.

Magma is usually described as being Basalt, Andesite, Dacite or Rhyolite, even if there are other sub-species such as Thrachytes and Phonolites. This division is based on the content of Silicon Dioxide, SiO2. Basalt has the lowest content of silica, in the region of ~40-50%. Andesite usually lies between ~55-60%, Dacite around 65% and Rhyolite greater than 70%. Intermediate magmas/lavas are describes as being basaltic-andesitic, andesitic-dacitic or rhyodacite. Generally speaking, erupting basalt has a temperature in the region of 1200C +/- 100C, whereas erupting rhyolite is cooler at about 800-850C.

The four main types of volcanic eruptions from left to right, top to bottom: – Effusive, sometimes also referred to as Hawaiian, eruption; lava flows from in rivers from a vent (uncredited image) – Effusive-explosive Strombolian eruption, Klyuchevskoy, Kamchatka (KVERT webcam) - Explosive Vulcanian (or Peléan) eruption, moderate eruption column accompanied by pyroclastic flows, Gunung Sinabung, Indonesia (FT Photo Diary, Jamie Han) - Explosive Plinian eruption, eruption column often greater than 20 km. Mount Pinatubo, Philippines, June 1991 (WikiMedia)

The four main types of volcanic eruptions from left to right, top to bottom:
– Effusive, sometimes also referred to as Hawaiian, eruption; lava flows from in rivers from a vent (uncredited image)
– Effusive-explosive Strombolian eruption, Klyuchevskoy, Kamchatka (KVERT webcam)
– Explosive Vulcanian (or Peléan) eruption, moderate eruption column accompanied by pyroclastic flows, Gunung Sinabung, Indonesia (FT Photo Diary, Jamie Han)
– Explosive Plinian eruption, eruption column often greater than 20 km. Mount Pinatubo, Philippines, June 1991 (WikiMedia)

While all types contain varying amounts of volcanic gasses, basalt generally contains more sulphur dioxide as it often comes more or less directly from the Earth’s mantle. Basalt usually does not contain much water and when it comes to how explosive an eruption is, water is by far the biggest factor. The greater the water content, the more explosive the magma tends to be. Because of their chemical composition, the most evolved magmas – dacite and rhyolite – can contain huge amounts of water.

The final aspect of how explosive an eruption can become is how “runny” the magma is. Just like hot syrup easily runs out of a warm bottle, hot basalt is far more runny than the sticky stodges at the other end. Rhyolite is extremely non-runny or viscous, as runny as thick oatmeal porridge in comparison to the more syrupy basalt if we continue the analogy.

Effusive Eruptions

If the magma erupted is basaltic, it is very runny and usually does not contain vast quantities of volcanic gas. At the vent, there is usually some spattering fountains of fire as the gas propels the erupting magma some tens to maybe hundreds of metres into the air. As the magma does not travel far in this manner, it forms a spatter cone in which it pools and sooner of later overflows, breaching the walls of the spatter cone. We get rivers of molten rock that can travel upwards of ten kilometres. The Hawaiians refer to this kind of lava as “Pahoe” or ropy. There is a second type of basaltic lava flow, the Hawaiian “A-aa”, which is more viscous and does not form rivers but up to 5-10 metres thick, somewhat blocky and slow-moving flows that spill forwards seemingly inexorably accompanied by a glassy crackle. Effusive eruptions are sometimes also referred to as Hawaiian.

Effusive a’aa lava flow from the 1984 eruption of Mauna Loa (WikiMedia)

Effusive a’aa lava flow from the 1984 eruption of Mauna Loa (WikiMedia)

But it is not only basalt that erupts effusively! Sometimes andesite, dacite and even rhyolite does not contain enough gas to erupt explosively, but oozes out of the vent just like the last dregs of toothpaste forced out of its tube. Commonly, this happens towards the end of an eruptive cycle and the extruded dome is often destroyed, either as it collapses to form a deadly pyroclastic flow such as the one at Unzen on June 3rd 1991 that killed 43 people including renowned volcanogists Katia and Maurice Kraft plus the American St Helens survivor Harry Glicken, or as the next eruptive cycle commences such as at Kelut in 2014.

Explosive Eruptions

The main types of explosive eruptions are usually sub-divided into Strombolian, Vulcanian, Peléan, Sub-plinian, Plinian, Ultra-plinian, Sub-glacial and Sub-acquatic (or Surtseyan). As the name implies, explosive eruptions are explosive and range from the very common small Strombolian type that is usually contained within the crater with spatter reaching the upper slopes to the extremely rare and violent ultra-Plinian ones which launch hundreds of thousands of tons of volcanic material per second as high as 30 km or more above the summit.

Night-time Strombolian eruption of Colima de Fuego, Mexico. (Webcams de Mexico)

Night-time Strombolian eruption of Colima de Fuego, Mexico. (Webcams de Mexico)

Strombolian eruption. The name derives from Stromboli; the poetically named “Lighthouse of the Mediterranean”, located among the Liparian Islands northeast of Sicily. Strombolian eruptions are characterised by short-lived explosive events that send a plume of ash and rocks many hundreds of metres above the summit from where they rain down on the uppermost slopes. During daytime, the eruptions appear to be grey but at night there is usually an intense reddish glow. The material that rains down on the upper slopes may coalesce to form lava flows and if such a flow encounters water, may cause what appears to be a pyroclastic flow as sometimes happens at Etna during the winter months. Apart from Stromboli itself, the Showa crater of Sakurajima is also known for its frequent bursts of Strombolian explosions as is Klyuchevskoy in Kamchatka where the eruptions typically cause lava flows that may travel all the way down the 4 km high edifice.

The 2001 vulcanian eruption of Mayon, Philippines (Maslog City Photos, R. Madronero)

The 2001 vulcanian eruption of Mayon, Philippines (Maslog City Photos, R. Madronero)

Vulcanian eruption. The term was coined by the Italian Giuseppe Mercalli who witnessed the 1888-90 eruptions of Vulcano, also one of the Liparian Islands. The Vulcanian eruption is characterised by a series of intermittent and irregularly spaced explosions in the crater as pulses of magma arrive which sends an eruption column kilometres into the air above the summit. As the eruption is not continuous, the column ranges in colour from the dark grey of magma-rich explosions to almost white when there is very little magma and thus mostly water.

The next step up in volcanic violence is the Peléan eruption, named after Mt Pelé in Martinique. This type of eruption is characterised by large pyroclastic flows, usually the result of the collapse of a recently extruded lava dome or spine, another characteristic of this type of eruption. The pyroclastic flows travel not only down the edifice of the volcano but may inundate the surrounding landscape as well. A good, recent example of this is the Soufrière Hills volcano on Montserrat, the pyroclastic flows of which caused first the evacuation of the capital Plymouth and later its complete destruction.

Sub-plinian, Plinian and Ultra-plinian. The term was coined in honour of Pliny the Younger who witnessed and described the 79 AD eruption of Mt Vesuvio. This is the most violent type of eruption where the main eruption launches an plume reminiscent of a Stone Pine tree tens of kilometres in the air within minutes. If the eruption column reaches an altitude of between 10 and 19 km, it is referred to as Sub-plinian, between 20 to 30 km Plinian and above 30 km as Ultra-plinian, the exact heights given for each types varies from source to source. The amount of magma ejected is in the case of the Sub-plinian eruption tons to tens of tons per second whereas the Plinian eruption launches hundreds to thousands of tons per second. In the case of an Ultra-plinian eruption, it may involve ejection rates well in excess of 100,000 tons per second and may reach as high as the stratosphere.

After a series of phreatic to phreatomagmatic precursor eruptions that in themselves can be quite spectacular – when Mt Pinatubo erupted in 1991 there were several Plinian precursor eruptions before the main one – there is a main eruption which lasts typically several minutes to several hours. Once the Plinian phase is over, the volcano usually shifts to Vulcanian or Peléan activity and this is usually the most destructive phase of the eruption with massive pyroclastic flows, such as Vesuvius in 79 AD and Pinatubo in 1991, that may continue for days or even weeks, and pyroclastic base surges such as Taal in 1965 where these were first identified.

Sub-glacial eruptions are as the name implies eruptions of a volcano covered by a glacier. As with all other eruptions, the smaller ones are far more frequent. These eruptions are not energetic enough to break through the several hundreds of metres to kilometres thick glacier but result in glacial melt, sometimes causing huge pits to form on the surface of the glacier, that result in a glacial flow or jökulhlaup. These are the most treacherous and in olden days, the deadliest ones as there was no visible warning that a hlaup was imminent. This caused several areas around primarily Katla to be abandoned and people avoided travelling there as much as possible. Sometimes, an eruption breaks through the glacier if it is either powerful enough (Katla 1918) or occurs in a spot where the glacier is thinner and weakened (Gjalp 1996).

If the covering glacier is several kilometres thick, the glacier won’t be breached. Instead the erupted magma collects around the vent to form a steep-sided, flat-topped mountain, a tuja which is revealed once the glacier eventually melts and recedes. There are plenty of such formations outside of Iceland; the type was first recognised in North America (in British Columbia and the Cascades.)

Sub-aquatic eruption of volcano off the coast of Nuku'Alofa, Tonga, in March 2009. (Dana Stephenson/Getty Images)

Sub-aquatic eruption of volcano off the coast of Nuku’Alofa, Tonga, in March 2009. (Dana Stephenson/Getty Images)

Just as the subglacial eruption occurs beneath a glacier, the Sub-aquatic eruption begins below the surface of a large body of water but may rise above the water to become sub-aerial as happened at Surtsey in 1963. This type of eruption is also referred to as Surtseyan.

Problems With the Current, “Magma-centric” Definition

Rescue workers at Japan’s Mt Ontake, September 2014 (Getty Images)

Rescue workers at Japan’s Mt Ontake, September 2014 (Getty Images)

At the beginning of this article, I referred to the current definition of an eruption; it begins when magma reaches the surface. On Saturday, September 27, 2014, at around 11:53 a.m. JST, there was a phreatic explosion at Mt Ontake which killed 54 people hiking to the summit of Japan’s second-highest volcano. As no magma reached the surface, it was a phreatic explosion that pulverised rock and remobilised old ash, this was not a volcanic eruption according to the definition.

This is where, in my opinion, the current definition is not only wrong but dead wrong! It is the result of a matter-centric view that ignores the driving force behind all of Earth’s geology; energy. Energy retained in the form of heat within our planet since it was formed ~4.5 BY by the collision of millions of asteroids and planetessimals. Energy resulting from the radioactive decay of Uranium 238 and Thorium 240.

One of the driving forces of our Universe is entropy; equalisation of energy potentials. The core of our planet is ~6,000K hot and with a mean radius of 6,371 km, this means that there is a temperature gradient of 1K per km. It is this energy difference that drives volcanism and plate tectonics, hence we should think of eruptions in the form of energy, and not matter (e.g. magma reaching the surface). Matter is only the vector.

If we do, we get a far better picture and understanding of what is going on. The lowest level of eruption would be gas emissions or volcanic degassing which would range from the carbon dioxide bubbles of the Laacher See to sulphur fumes of Solfatara in the Campi Flegrei. The next level would be hydrothermal energy such as Geysir or Old Faithful. From there we move on to the phreatic explosive events when superheated steam; a liquid under high pressure and temperature, reaches the critical point and flashes into steam explosively such as at Mt Ondake. If we adopt this energy-system instead of the current magma-centric view, Mt Ondake would have been listed as “active” and perhaps, 54 people might not have needlessly died. And with Solfatara listed as active, would those apartment blocks on the crater rim have been given the go-ahead?



74 thoughts on “An Overview of Eruption Types

  1. Very good Henrik, very good…
    Brought back some memories of St. Helens.
    “Can we get a little closer to the crater?”
    “No” I said.
    This was in April, 1980 on a USGS/Army Corps
    of Engineers survey flight.
    They decided I was right when this column popped
    up in a classic Strombolian manner. Not wanting
    to be brought down by a well tossed lava bomb or
    rock we got out of there…

    • Wasn’t this referred to by one of the volcanologists in a program about St Helens? It was something like “We wanted to get closer but the pilot said no and at that moment there was an explosion” or something like that.

      • Might it have been the Stoffels?

        Geologists Dorothy and Keith Stoffel had chartered a Cessna flight over Mount St. Helens the morning of May 18, 1980. As their plane flew overhead, the mountain began rumbling and suddenly erupted. Believed to have been the closest to the mountain during the time of the eruption, Keith Stoffel managed to capture the moments before and after the blast on film.

        • Not me -it was in April -before big event.
          It was clearing its throat..

          • I believe the Stoffel aircraft was chartered from the Portland Or.
            area..they were too close.
            Way too close…

  2. Great read, thank you Henrik. The downside of adopting the energy system is that Yellowstone becomes an active volcano again 😛

    • True. But then again, the upside is that many volcanoes that are now wishful-thinkingly classified as extinct or dormant get a far more accurate label; from “No worries, it’s dead. There hasn’t been an eruption in hundreds of years.” to “Sorry, but there are still signs of low-level activity even if an eruption is unlikely at present.”

  3. Very good article Henrik!
    I have a wish here. Could you write a similar article where you go through the phreatic detonation types.
    This is regardless of me agreeing with you on the matter versus energy issue of the definitions. But I still think it would be good to go through the diverse ways of phreatic detonations.

  4. HI Henrik .

    Long ago I had not commented on one of your post.
    And as always, it is clear, clean and precise … Everything I love. Photo of Half Dome is gorgeous …
    Return to our true values … Earth … LOL.
    Great friendships, George

    ” le Chaudron de Vulcain”

  5. thanks for the great article, Henrik.

    On another topic, I keep an eye from time to time on Iwo Jima webcam because, well, duh. It seems to always be steaming and sometimes you can’t see it for either weather or too much ash/steam??

    • If Iwo Jima would start steaming it would be worrisome. As far as I know it does not steam, so it should be clouds forming as the moist sea air is pushed over the mountain.

    • Might you be able to post a link to the Iwo Jima webcams, please? I’ve rummaged around to try and find them with no luck.


    Really good article and study about how earthquakes *may* affect volcanoes. In some ways, it’s similar to how liquification from earthquakes can affect soil, causing sinking and exaggerated shaking in some areas. In some ways, it only would make sense that a this same process could cause mixing in a magma chamber. Of course, you still need a magma chamber close to being ready to erupt along with quite a few other factors, but it’s interesting nonetheless.

    And for a pretty crazy video – here is liquification in action during the Tohoku 9.0 quake in 2011.

    • always had this nagging suspecion that if earthquakes *did* trigger eruptions (by assisting magma mixing, triggering nucleation or bubble formation, or whatever) it wouldn’t be an instant thing -hours or days- but rather on the lines of lighting the fuse for an eruption years or decades later. By which time that particular earthquake would be ancient history. Who could ever prove a delayed linkage of that sort, in an earthquake-prone region?

      • The fact that volcanic eruptions can be preceded by large but unrelated earthquakes seems well established. Krakatoa is a point in case: there was a major quake 3 years before and it was clear that the volcano activated after it. The how is less clear. At long distances, the surface waves dominate and they don’t penetrate the magma chamber. So it could only be the surface cracking. Closer to the quake other effects are possible, including the foam sloshing mentioned in the link above. Or perhaps the magma chamber can overturn, if it is already unstable. I don’t think the energy itself can come from the quake: it must be energy already inherent in the magma chamber.

        • In all likelihood, the earthquakes are indicators that something has happened below ground that may have changed the pressure within the magmatic system and that it is this this change in pressure (and temperature) that presages a future eruption. The quake that usually is connected with the 79 AD eruption of Vesuvius occurred about a decade before iIrc.

    • I don’t think earthquakes have that much of an effect. I suppose you may subconsciously be thinking about what happens when one shakes a bottle or can of fizzy and then open it?

      What really unbalances an existing magma chamber is when there is a large influx of thermal energy such as an infusion of juvenile, basaltic magma if the magma chamber is close to a critical temperature-pressure point for certain minerals, their mineralisation point. What happens when the temperature and pressure suddenly rises is that the the recently formed crystals topple above their remelting point and the reaction is exothermic, which releases more energy as the bonds are broken. Suddenly you have a sort of runaway chain reaction = boom! It seems to be what happened at Laacher See just to give one example.

      • Yeah, no doubt, you need a system that is at least somewhat ready to go for this to occur, but I can see how it can possibly be the straw that breaks a camel’s back. Definitely a slower process than many would think, but we do know that magma mixing is a common-found characteristic before eruptions, and this could be one way for that to occur.

        • Simply not enough energy, I’m afraid. Work the numbers! As an example, the chemical bond energy of SiO2 is estimated at 621.7 kJ/mol with a mol mass of 60.08 g/mol which means a m^3 of Quartz carries something on the order of 27.5 GJ and that’s the total energy released by a M 3.8 earthquake. Now take a full cubic kilometre. The molecular bonds contain ~27.5 EJ which would be the equivalent of a M 9.8 earthquake or eight times the Toba eruption.

          I think it’s safe to more or less dismiss the proposed destabilising effects on magma chambers by earthquakes except possible the very most powerful earthquakes.

          • Clearly it would need to be a very powerful quake, although I think you’re missing a few important aspects. Given, this is all somewhat theoretical, but I think it’s a bit more complicated than assuming you would need the quake to directly release the bonds.That isn’t really the idea.

            The main idea is that the quake would cause the mixing of magma, which could remobilize cold magma similar to how a basalt injection may do so (albeit, likely on a smaller scale). Areas that are cold could presumably come back into contact with hot areas of the magma, which would cause that zone of magma to heat up. The one caveat to this idea is that it would also cause the hotter magma to cool down, but I think in the end, it *could* release a solid amount of gas.

            Another important point is that large quakes could possibly open up pathways for basalt from depth to reach the primary magma chamber. Just as liquification caused by shaking can cause ground to behave more akin to a liquid, I would have to imagine there could be an affect on the deeper magmatic pathways as well. We know from history that water flows out of the ground during liquification events, so why wouldn’t this have the potential to affect magma as well? Now, I know there is a big difference between deeper crust and loose topsoil, and I know magma is way more viscous than water, but it could at least temporarily lower the pressure in the lower conduits, allowing magma to force an opening into the primary chamber.

          • That’s not my intention or idea Cbus. 🙂 I’m just making a comparison of energy levels in order to show how insignificant the energy released by earthquakes is compared to the amount of energy stored in bondings and crystal lattices. Now, imagine you have a phonolitic (more K/Na than SiO2) magma chamber which is at the point where feldspars begin to crystallise out. Add a 0,001 cubic kilometres of juvenile basalt at 1250 degrees C. As the temperature rises, the recently formed feldspar crystals begin to dissolve and the energy stored in their bondings and crystal lattices is released making the mush even hotter which causes yet more crystals to dissolve (Caveat: If I recall correctly, this is an exothermic reaction). What you get is a runaway chain reaction that dramatically increases the temperature but most of all the pressure within the magma chamber until the volcano goes BOOM in a major way.

            The amount of energy infused by earthquakes is not sufficient, not by a long chalk, to unstabilise a large body of magma in this fashion. Furthermore, by the time a magma chamber is at one of equilibrium points, the melt has been thoroughly differentiated so that the heavier stuff (metallic ions) sits at the bottom and sides. To all intents and purposes, such a magma chamber consists of a feldspar-quartz mush in solution with huge amounts of water.

          • Ok, now is the time to give up lurking, what I do regularly since Bardabunga went off…I was already tempted, when a lot of people wrote about lake Ijen witch revealed a common ignorance of chemistry. Or even the difference between chemistry and physics.
            And Henrik delivers the perfect example. Melting is a physical process. No chemical bond is braking during this. No reaction is taking place. The only energy released is the melting energy due to entropy difference.
            Next I would really love to know which density you used for the calculation?
            Pure SiO2 has a melting point of 1713°C. So what melts is never pure SiO2, but a multitude of chemically completely different crystals with their own enthalpy of fusion, compositions, melting temperatures. So forget about calculating melting energies.
            So back to topic…
            What an earthquake does is sending pressure waves through the magma body, which I assume could lead to local decompression melting, so an earthquake should transfer energy into a volcano and as cbus05 said if that is “the straw that breaks a camel’s back”…

            About me: lab technician, that quit her try at being a chemist after 4 semesters and works in chemical research for more than 10 years now, no big knowledge about geochemistry though. Loved to find so much valuable scientific information here and enjoyed the ruminations. :o)

            Released from the Dungeons!
            Welcome out of Lurkmode.

          • Hello Colloid!
            I hope I did not make to much of a bungle about the article I wrote about Kawah Ijen from a chemical standpoint. I am not a chemist so I tried to read up as well as possible upon the chemistry, but any errors would have been welcoming a correction.
            And mea culpa, I am a physicist so any physics getting into it is probably my fault.

            Please stay out of Lurkmode in the future!

          • Thanks for the welcome!
            No your artikel was alright and you didn’t mix it up to much, but the ideas that came afterwards…the battery-idea for example knocked me off my feet… ;o)
            Maybe I find time to write something there…
            It’s always the problem, that things are more complicated, than it looks from the view of a layman…which is very true for me if it comes to vulcanology…

          • Great read as always, this is such a good place to come to learn!

            If back in 2013 I would have said to everyone here that BB was going to experience around 30 M5+ quakes including an M5.8, all within a few weeks and all inside the caldera but the volcano proper would not erupt… would have laughed at me for sure.

            One would expect that all that concentrated energy and shaking would have caused something to pop, if it ever was going to pop??

          • Welcome out of lurking mode! I’m sorry if I gave an erroneous impression as well as a perfect example of ignorance – I only used SiO2 to arrive at a ballpark figure for the energy contained in bonds and lattices. The rho used was the 2.65 g per cm^3 of Quartz. You’re quite right about the temperature, 1713C, of Cristobalite (or perhaps more correctly, the SiO4 tetrahedron). But there is also the tridymite variety to consider.

            The exact processes are of course vastly more complicated with molten molecules of feldspar, quartz and other silicates combining with water, the water being “driven out” of the flux as the minerals crystallise and “re-entering” as the crystals get remelted. I’d really love to get my hands on a paper that explains this process within a magma chamber, complete with the relevant energy calculations. 🙂

          • The latent heat of melting of silicates is around 70 kcal/mole. You don’t break the molecular bond but the lattice. But you can only melt silicates this way if it is already at the right temperature. Earthquakes don’t do this. If they have any effect, it is to do with the gases or with an overturning (as in a lava lamp). In the case of a Vesuvius quake, a decade before the boom, or Krakatoa at three years, it is more likely that the quake opened a new conduit allowing deeper magma to rise.

          • The lava lamp actually has a name as an earthquake, non-double couple earthquake with zero net displacement. Not that catchy as a name I admit, but I did not name it.
            The most famous example of it is obviously Bárdarbunga that suffered a string of M5 N-DCEWZNDs… Not even catchy when abbreviated, let us rename them into Lavalamp-earthquakes and be done with it.
            Those earthquakes ment that it was hot fresh magma in the upper reservoir and that the old magma was reheated in the deep reservoir that is open towards the mantle.

          • Yes, that name could do with some improvement. Lava lamp quake? It is magma rather than lava and it is pitch dark down there so a lamp may not be the best word. but the idea seems the same. In Bardabunga, did the quakes cause the overturning or was it the other way around?

          • The earthquake is the waveform manifestation of the overturning, the same as a regular earthquake is the waveform manifestation of a fault-breach.

            To be even more precise.
            The overturning is the earthquake and the earthquake is the overturning. Remember that this is a magmatic earthquake only, as such it is the funkiest seismic event known to man. Well, together with the organ-pipe (magma-tube) slow-quakes known as tornillos.

          • Carl,

            From this paper, they suggest the older M5 quakes were in fact due to inflation along the ring faults and the recent ones are from caldera collapse.

            I have taken this from the paper to highlight the reference

            “Earthquakes with significant non-double-couple components in
            their focal mechanisms have occurred at Bardarbunga in the
            decades prior to the current event (Supporting Information Table S2;
            Nettles & Ekstrom 1998; Konstantinou et al. 2003; Tkalcic et al.
            2009). However, the polarities of the focal mechanisms for the
            current Bardarbunga eruption are opposite to those observed for
            the earlier earthquakes, that is vertical compressional axes for the
            former and tensional axes for the latter (Fig. 2f). One interpretation
            of these earlier events suggested that their faulting mechanisms
            are primarily due to rupture on outward dipping ring faults
            which are activated by inflation of a very shallow magma chamber
            (Nettles & Ekstrom 1998). However, this interpretation is highly
            non-unique, since a deflating magma chamber below the ring faults
            could also produce similar focal mechanisms (Ekstrom 1994). Our
            inversion results for the magma chamber predict a chamber radius
            smaller than the caldera radius for chamber depths less than 15 km
            (Supporting Information Figs S7 and S8), which implies inward
            dipping ring faults. The dip of the ring faults controls the rupture
            arc-length necessary to create the CLVD focal mechanisms. For
            this geometry, the earlier events can be explained by inflation of the
            magma chamber, leading to reverse motion on those ring faults.”

            taken from



          • Hello Ian!
            I would say that it is more than likely to be correct.
            The earlier earthquakes with non-double couple components happened during an inflation phase and entailed that cooler magma rapidly switched place with hotter magma.

          • I think what the article is suggesting is that the pre eruption quakes were caused by movement along the ring fault and not by magma rapidly shifting between chambers. You can model each large non double couple quake as a series of smaller double couple quakes progressing around the entire ring fault in rapid succession. If the ring fault is not vertical, but conical, it will give the same signature as the proposed magma shifting, but is, as I understand it, a different mechanism.

            To me it makes sense that opposite action happens during inflation and deflation: Reverse faulting during the inflation and normal faulting during the deflation.

          • I do not agree with that supposition.
            Remember that there was no net displacement, and there would have been with their proposed model. There would also have been a coda to the wave form as the pressure momentarily dropped. And, it was decidedly not made up out of many smaller quakes.
            They are trying to fit the data to model instead of making a model that supports the data.

      • Ehum…
        What happened in Laacher See is that a small dyke ended up in an aquiferous reservoir and it all went boom in huge steamdriven detonation.
        So, no gasses (only water) and no earthquake, no magma chamber, pretty much no nothing except this:

      • That’s a very impressive example of sieche waves. Possibly the best I’ve seen. Although I have heard that under the right circumstances they can cause inland tsunami.

      • It’s quite fascinating. But Sognefjorden is not a lake, it is the second longest fjord in the world. And it’s 50 meters from my house.

        Released from the Dungeons

      • Rayleigh waves. They can travel very long distances from the earthquake without losing much strength, but only over the surface so they won’t affect deeper magma chambers. Low frequency so they can set up resonances in water. You get these seiches in lakes and coastal waters – there were reports from Loch Ness after the Lisbon quake. Te coastal seiches are sometimes confused with small tsunamis but they are different.

        • Thank you for that, Albert.
          Yes, with my specific area of interest, I often notice how any unusual long period wave can be mistaken for a tsunami.

      • Got a YouTube link to that? The page you linked seems to use their own dodgy in-house video thing that doesn’t work in my browser, giving just a blank box, instead of using a bog-standard works-for-everyone youtube embed, for some silly reason.

        • I had that problem at first. They had a dodgy advert at the start which my browser blocked. After a while and some clicking the program showed.

          • No amount of clicking and reloading the page seems to work here. Do I have to whitelist a script at that site or something?

            A YouTube link would be so much easier, both for me and for my network’s security…

          • Nope. I get a click-to-play embed box, and when I click in it it “thinks” for a bit and then displays a black rectangle in the box with static text, no video. The text, if anyone can make heads or tails of it, says:

            “Du trenger javascript for å bruke denne tjenesten. Det virker som om Javascript er skrudd av i din nettleser.

            Du kan antagelig skru det på under valget innstillinger eller lignende i din nettleser.”

            So, how about a simple, works-for-everyone, no-mess-no-fuss YouTube link? :/

          • Doesn’t help. Allowing turns it blank black, adding puts an encouraging spinning “loading” wheel on there that never stops spinning and goes to a video, and adding then makes the wheel stop and change to another black box with different static text:

            “En uventet feil oppstod. Vennligst prøv igjen senere.”

            AFAIAC it’s simply hopelessly broken.

            Now please someone post a YouTube link. Really, in a world with YouTube there’s really no need to use any other video hosting solution unless it’s something like pr0n that YT doesn’t allow, and YT tends to be very reliable and scales well to very heavy use. I just don’t get why so many sites still insist on using small third-party video hosts that don’t work nearly as well, or even hacking their own together that tends to be even shoddier.

          • Sorry, VG or Verdens Gang is a newspaper and they tend to use proprietary tools. Might even be a licensing issue.

          • Nordic Newspapers do dable with video-photage quite a lot, so it is normal for them to do both shoot it and to use proprietary systems.

        • At least there’s a nice text article accompanying the broken video player at that page, which goes into some detail explaining the phenomenon. That’s good. 🙂

    • Which volcano would that be?
      It is more likely that it would be tectonic earthquakes.

  7. Total Energy release by day within the Bardarbunga Caldera since the end of the eruption with the KISA GPS plot superimposed.

    As the energy plot is going flat, so it appears the GPS plot is following. This is interesting enough to keep following along to see where it leads….if anywhere lol

  8. There is also the not so insignificant part that the latest set of earthquakes have not had clean breaks like the previous, this could indicate that there might have been fluid involved in the seismic events. And if dykes or sills have formed that would temporarily lower the pressure minutely.

Comments are closed.