Grimstone vs Yellowvötn: Battle of the Giants

Carl – May 9, 2014 (re-post)

West Thumb Geyser Basin, Yellowstone lake (Yellowvötn). Photograph by unknown.

In the west corner of the ring you find Yellowstone wearing blue, red and white striped trunks and in the east corner you find Grimsvötn wearing Fire & Ice colored trunks. Welcome to a spectacular fight about who is the largest, meanest volcano on the block. As the fighters are squaring off we eagerly await for the first blow from this formidable match, and there it came, it is a stunning early knock and we have a countdown to ten. The new world champion is surprisingly Grimsvötn…


I read everything that Erik Klemetti writes and have been a big fan of his since he started the concept of volcano blogging. This week he wrote a splendid article on Yellowstone (link below) based on a paper by Jamie Farrel et al (link below) who used the earthquake data from 1984 up until 2011 to make the best mapping of the innards of Yellowstone so far. The technical part of it is described by Erik so I do not have to plow that ground here (read his piece instead).

What is interesting is that Yellowstone’s magma reservoir is now deemed to be larger than previously believed. It is now rated at between 200 to 600 cubic kilometers. Problem is just that most of it is not in the form of melt, between 85 to 95 percent is solid. Only 5 to 15 percent is now believed to be molten. This means that Yellowstone can’t erupt in its current state.

And if we assume a maximum intrusion rate of 1 cubic kilometer per year it would still take between 68 years and 228 years at best until Yellowstone reached the presumed barrier for when an eruption can occur. On top of that there is no evidence at all that magma is intruding into the magma reservoir at that speed. If it had done that Yellowstone would erupt quite often (every few years or so). In reality the magma influx seems to be miniscule at Yellowstone, if it is even influxing at all.

The Castle Geyser, Yellowstone National Park, U.S.A

This is really bad news for Yellowstone, but I guess that the Bison’s will be happy to know that they will not be blown off the face of the planet.

Now, let us look at a volcano that has a comparative magma reservoir and a proven influx of magma rated at the astonishing speed of 0.1 cubic kilometer per year and compare how that volcano behaves. Remember, this volcano is de facto more active and is today a far larger and more dangerous volcanic system than Yellowstone, so let us travel to the land of Ice and Fire.


Close up of the 20 kilometer high ash column during the 2011 eruption. Photograph by unknown.

First of all let us start with what Grimsvötn is not, and that is a Supervolcano. It has never suffered a VEI-8 eruption, nor will it ever. But it is a supererupting volcano none the less. In 8230BC the volcano suffered the VEI-6 Saksunarvatn tephra, in an explosive caldera forming eruption. It was the third known caldera forming eruption of Grimsvötn.

In that eruption large parts of the magmatic reservoir was destroyed, but the deep feeder system was still intact. And during the last 10 000 years the reservoir has built up to a prodigious size. The volume above the Curie-point is 400 cubic kilometers.

Now, the Curie-point is where a material becomes too hot to be magnetic. That does not in and of itself state that it is a melt. But it says something about the temperature the magnetic elements are in. Let us dwell a bit on temperatures.

For instance we know that artificial Yttrium Iron Garnet has the lowest known Curie-point at 287 degrees Celsius, but that one does not exist in Grimsvötn or any other volcano. The lowest temperature compound that could be found in Grimsvötn is Fe2O3 (hematite) with a Curie-point of 675 degrees Celsius. At that temperature some of the magma would be solid but melt would most likely be higher than at Yellowstone.

We do though know that the heat influx is great and that the volcano has ready access to nicely melted material in abundance. In fact, we know that the volcano is well above the required 40 percent melt for buoyancy driven eruptions. In most likelihood the Curie-point is the same as, or even above, that of pure iron (770C).

Now, how come that a non-supervolcano can have more eruptible magma than a VEI-8 brute like Yellowstone? Well, that answer is trickier to give. Basically it all boils down to the amount of energy entering the system, and that energy arrives in the form of heat carried by arriving magma from the mantleplumes residing under Yellowstone and Grimsvötn.

In Grimsvötns case that is 0.1 cubic kilometer of basalt at a temperature of 1 150 degrees Celsius per year and that is quite a lot of energy. If we compare that to Yellowstone it seems like almost no new heat is arriving, or a very tiny amount of it.

The evidence of this is that there is very little evidence of Yellowstone suffering from a resurging dome (inflation of the center of the caldera). In all likelihood Yellowstone has not received any new magma in the last 70 000 years. Remember that it takes quite a while to cool down a very large magma reservoir until it is almost all solid.

If we compare that to Grimsvötn who has received in the neighborhood of 800 cubic kilometers of fresh hot magma in the last 10 000 years you can understand the difference in energy input into that volcanic system. Those of you who are into math will now be collecting your jaws from the floor. Yes, Grimsvötn has erupted in the order of 400 cubic kilometers in 10 000 years, or on average 0.04 cubic kilometers per year (as Tephra and lava).

But, the figures grow even more stumping if we look at the amount erupted in the last 250 years. Without counting the Skaftár Fires (Lakí eruption) Grimsvötn has had 28 confirmed eruptions ranging from VEI-2s to the large VEI-4 2011 eruption. Together with the 3 cubic kilometer Dense Rock Equivalent and 15 cubic kilometer flood basalt of Lakí we get a volcano that has erupted as much or even more lava than the received magma influx.

This is of course not Grimsvötn 2011 as I wrote. I can only blame mounting senility. It is Eyjafjallajökull 2010 seen from farm Ormskot in Fljótshlíð. Thanks to commenter Jonbragi for setting me straight.

The 2011 eruption caught most scientists with their pants down since they did not believe that the eruption would be so big, but in hindsight it is easy to see that a magmatic system of that caliber can erupt a large volume without a long repose time.

This tells us that the magma reservoir is now at its limit of what it can hold, everything that goes in comes out as quickly, and as time goes by the risk of another caldera forming event increases, and most likely Grimsvötn is already teetering on the brink for one. The other risk is obviously a repeat of the Lakí-event, and that is not such a small risk. After all Iceland regularly suffer those at a rough 270 year interval (sometimes Iceland skips a cycle, remember that, nothing is ever certain around volcanoes).

Of course a caldera forming explosive eruption or a large scale flood basalt would impact global weather, especially in the northern hemisphere. But for those who Dream Dark Dreams of Destruction, it will be a large nuisance and not a catastrophe.

I am terribly sorry that I once again spanked the Doomsayers favorite volcano. Those people should understand that there is no cover-up, nor am I and every other scientist on the planet trying to fool them. It is nature herself that is against their Dark Dreams; Yellowstone is for all points and purposes no longer a volcano. End of discussion.


31 thoughts on “Grimstone vs Yellowvötn: Battle of the Giants

  1. Love this article. Fairly simple look at why the big nasty isn’t so big or particularly nasty anymore.

    Separate discussion I have been having with a freind is about Katmai and the valley of ten thousand smokes. Following the eruption, Katmai’s peak collapsed into the emptied chamber even though the actual eruption was several km away. Bardarbunga is doing the exact same thing but in slow motion. Holuhraun drained the chamber (not completely, but enough) and it was also several km away.

    • I know me, Carl, and a few others on here a while back reached a somewhat similar thought that there is likely a significantly larger central magma chamber beneath the Novarupta region that feeds roughly 5 nearby peaks. I personally see Novarupta as a region that in the geological future will have a much larger caldera eruption that drains more than just the smallish magma chamber beneath one of the feeder volcanoes.

      With that said, I see a distinct difference when comparing Novarupta to Bardarbunga.

      In Bardarbunga’s example, the magma was being originally stored below the volcano where the caldera collapse occurred. The magma then drained into the side-fissures and more or less “leaked” out to the gravitational low-point, causing the slow collapse of the central caldera.

      In Novarupta’s case, the magma was never stored beneath the Katmai volcano where the collapse occurred, at least not primarily. The magma was predominantly stored in a more centralized location that likely feeds multiple volcanoes (Griggs, Katmai, Trident, Mageik, possibly Martin). When the central larger eruption occurred, the depressurization of the central chamber likely led to under-pressure there, in which case the katmai magma feed collapsed in on itself into that central chamber.

      • Just to add a visual to the central magma chamber theory of Katmai / Novarupta, here is a quick screencap and annotation via Google earth.

        Novarupta in a lot of ways resembles a post-caldera volcano with somma volcanoes occupying a ring-fracture system. But the difference here is that there is no major caldera event of a central magma chamber. Despite this, we know that the primary vent was 8km away from Katmai, and the magma composition was a bit different than what was primarily seen at Katmai.

          • I don’t think there is – wouldn’t write off the possibility, but I have never heard anything. Alaska volcano observatory has a good amount of information out there regarding most of the volcanoes, especially those that are closer to the mainland.

  2. Good article. It’s because of VC I no longer worry about the doom and gloom of Yellowstone. Back before I started lurking on here, you’d here about how bad it would be and would have you think it’d happen in the near future. Then there’s the movie on the Syfy channel of an eruption at Yellowstone. Volcano porn? Maybe but not likely to happen when or the way they make you feel.

    Now for checking out cbus’ post. 🙂

  3. Grimsvotn 40km3 per 1000 years (about 20km3 of Laki and a combined 5km3 for a few several larger ones like the 2011 eruption). Question is: the remaining 15km3 is still stored in the chamber. It could unleash a Laki-sized event again, or perhaps some millenia go by, with accumulating such extra and then resulting in a 100km3 caldera-forming VEI7 eruption. Otherwise, I would estimate an average of 25km3 erupted material per 1000 years.

    Hekla has erupted roughly 1km3 per century (two VEI4+) (plus one or two 5km3 like 1104), this gives up to 20km3 per 1000 years approximately. Almost as much as the rate of Grimsvotn.

    Katla erupts around 1.5km5 per century, in two large VEI4 average eruptions. This totals 15km3 per millenia, but if one includes Eldgjá, then the total output for Katla is around 30km3 per 1000-1100 years.

    Bardarbunga has erupted probably a couple of small sized 1-2km3 lava floods (like Holuhraun) is the past millenia, plus two 10-15km3 large lava floods (Vatnaoldur and Veidivotn). This gives a total output of 30-35km2 per 1000-1200 years.

    Some regions around Langjokull show a handful of 5-10km3 lava fields, in the past 10.000 years. I estimate this to be a total output of 50km3 per 10.000 years, or 5km3 per millenia. About five times less output than the other big Icelandic volcanoes.

    • One can use the average rates of eruption per millenia to estimate how much magma can a volcano erupt in the future. And whether such volcano is pretty on the brink or not,

      If one takes for instance Grimsvotn. In the past 250 years since Laki, the volcano should have erupted around 7.5km3 (2.5km3 rate per century). Obviously Grimsvotn is collecting an increasing large pool of magma, ready for a big show. Another 2011-like eruption is quite easy thing. But a Laki repeat would still lie some centuries in the future.

      For Hekla, the output per 100 years is an average of 2km3. I think Hekla has done that since 1947, so I think Hekla is pretty relaxed.

      Bardarbunga. In the 500 years since Veidivotn, the volcano should have erupted on average 15km3. I think Holuhraun (3km3) was only a small demonstration. Other than that, I think Bardarbunga might have erupted at most a combined 2km3 in other smaller eruptions since Veidivotn. Bardarbunga has a potencial 10km3 due to erupt. I think it will continue to store, until it reaches the 15km3 threshold, probably sometime in the next couple of centuries. Or it could erupt again.

      Katla erupts on average 25km3 per millenia. It probably erupted 15km3 in all eruptions that occurred twice per century. With a 10km3 difference, Katla seems on the way to another big eruption, anywhere between 3 to 10km3.

      • Does that take Eldgjá into account? It’s supposed to be related to Katla’s system. That’s 18km3 right there. (roughly 934AD or 939AD depending on source)

        Nevermind. While adding superscript tags to our conversation I notice that you did take Eldgjá into account. I withdraw my question… but thank you for the insight!

        BTW, Katla, the housemaid, wasn’t evil. She was just ornery. Don’t cross her and you’ll be fine. I can fully understand her not wanting anyone to use her clothing (trousers, shoes, depending on whatever the particular version of the tale is that you hear or read.). This was a time of parasites and having your clothing being worn by someone else is a sure fire way to get something you don’t want. The magic trousers were probably a difficult thing to come by in the first place and likely were (in her opinion,) irreplaceable. Bardi, dealing with sheep all day, might have been prone to pick up any ticks or parasites that they may be carrying in their fur. So yeah, Katla is gonna be highly irate if he used her trousers when she specifically told him not to touch them, so she killed him when she found out what had happened. If Bardi had followed her orders, he probably would have lived through the ordeal. Maybe bruised, but bruises heal. Choking is a different matter. My impression of the Katla in the story is that of a highly independent and probably strong if not outright tough woman. That she had a volatile temper is a matter of record.

        Yeah, I’m defending a murderous “pagan” woman. But she deserves a fair shake. She can’t have been particularly evil if she was working for the local clergy. She was faced with the threat of getting lice from Bardi and it probably pissed her off to no end.

  4. … and the vapid idiots are still peddling their tripe.

    “And scientists currently believe there’s a 10% chance that a “supervolcanic Category 7 eruption” could take place this century, as pointed out by theoretical physicist Michio Kaku.”

    • That estimate is too low. It is closer to 30%, based on the number of such eruptions over the past 1000 years

      • I think ya missed the hilarity. It appears that they are mingling the VEI scale with the BBC term “super-duper”. Logically that means that a VEI-0 is a level of super-duper eruption. Though prolific, I don’t think Sakurijima has had as many super eruptions as that implies. Kagoshima Bay is in the the parent caldera (Aira Caldera).

  5. Anyone happen to have a link to his peer reviewed paper where he supports his claim?

    … waiting…

    • This map of the galloping hotspot is interesting-from
      that link I gave-btw when the paper was written they
      were of the school that the hotspot was still active..

    • Personally, I am of the opinion that the Columbia Basalts were formed from an eddy current of up-welling magma behind the detaching Farallon plate. Other papers have stated that the Yellowstone hotspot plume sliced the Farallon in two like a cutting torch as it slid down past it.

      • I have a lingering theory that the Yellowstone hotspot is also a remnant eddy of the detached Farallon plate, and not a deep mantle plume like Hawaii. One of the main differences is that the eddy currents crossed into thicker continental crust, whereas the crust in Washington is largely accretionary terrains that are easier for magma to find its way to the surface.

        Just a personal pet theory however.

    • The statement ‘shared a similar mantle source’ does not say that the Yellowstone plume was not the source of the Colombia basalts. It just states the facts, which is that the lavas are similar. Lurking may be right. Others have suggested that the high plateau of the Steens mountain (where the yellowstone hot spot started) pushed out magma northwards, similar to how the weight of Bardarbunga pushed the magma out to Holuhraun (although obviously on a different scale!).

      • Yep, I agree with Lurk on this. One thing is that the crust in the area of the various vents for the
        whole of the Columbia formations is full of sills and dikes, Imnaha Canyon is a good example..
        Albert you have a point, too the Steens Mountain area (also the Summer lake rim )-something happened to cause the plateaus to drop…
        Like movement of lava……

  6. Pingback: Spotlight on green news & views: Larsen C iceberg roams free; 2nd California wolf pack appears - LiberalVoiceLiberalVoice — Your source for everything about liberals and progressives! — News and tweets about everything liberals and progres

Leave a Reply