As far as volcanoes go, we like to think of them as immutable giants that rarely if ever change. We like to see Grimsvötn as a glacier covered giant of a volcano, that almost always produce moderate ashy eruptions, that are relatively speaking short-lived.
Yes, once upon a blue moon it will do something big, like the Lakí-styled eruptions, or even a Saksunarvátn or five. But hey, those are 7 eruptions in 13 000 years. Otherwise it goes like a clockwork in that ice-cowered medium range.
Another thing “we know” about Grimsvötn is that it likes to send a few Amazon River’s worth of icy water down the range in the form of Jökulhlaups.
Let us sum Grimsvötn up to make it clear: Glacier covered frequent eruptor of medium sized eruptions that needs continent sized diapers to hold its water.
There are two fallacies here. The first one is a general one that is valid for most volcanoes. About one eruption in three will not be normal for your average volcano. They are far better at throwing oddballs than we like to believe.
The more Grimsvötn specific fallacy is that we just need to look back 24 years to see that half of the eruptions are oddballs.
1996 was a definite oddball where Grimsvötn suddenly interacted with Bárdarbunga and created the Gjálp eruption on a radial fissure emanating from Grimsvötn.
1998 and 2004 was back to the same large VEI-2 to VEI-3 that Grimsvötn has done since the last Lakí eruption.
So, it came as a surprise for those who relied on 221 years of docility as 2011 came roaring around the corner with a borderline VEI-5. Now, remember that the Icelandic scientists are made of stern stuff, so as they tallied up the ejecta, they came up just a tad short, and being out of stern stuff it is a VEI-4.
It did not really end there, there is still a heated (pardon the pun) debate if at least a part of the Holuhráun magma came from the northern end of the Grimsvötn eruption in some sort of weird volcanic mating with Bárdarbunga (they should really get a room).
The Glacially Impaired Grimsvötn
As the mantleplume driven romance under Vatnajökull continues it has made Grimsvötn suitably hot and it has picked up sun-bathing as a hobby. Okay, I will stop with the puns now. Let us instead get down to the facts.
After a normal eruption the hole in the glacier normally heals up pretty fast. The geothermal energy output is not large enough to keep the glacier at bay since it is barely large enough to keep a bottom lake just above the freezing point.
Actually, it takes a lot of energy to keep that lake at all. The geothermal output of Grimsvötn was staggering at 250 Watts per square metre in 2007.
As a reference Yellowstone caldera has a heat flow of 100 to 300 Watts per square metre with peaks reaching up towards 1600 Watts in parts of Yellowstone Lake. So, in other words, Yellowstone and Grimsvötn used to have similar outputs, and here the takeaway is that Yellowstone has no problem snowing over in winter. So, we do know that 250 Watts is not enough to stop snow or ice forming.
I should point out that the data for Grimsvötn is from before the 2011 eruption and that to the best of my knowledge and google skills, there are no new figures out.
In the first couple of years the scar after the eruption started to slowly heal up, it was slow but that can be expected after a larger than usual eruption.
But, as we the decade started to wane we instead saw a reversal of the process, the scar started to become larger, during summer there was a little bit of open water (packed with ice shards, but none the less), and even a piece of the caldera wall suddenly became ice free.
By now Grimsvötn has turned into a true caldera-lake with almost 10 percent of the surface being ice-free during the summer months and an entire new section of the caldera wall is ice free. The glacier itself on all sides is looking like it has been struck by a sizeable blow and it is crumbling towards this new gash in the ice cover.
At the same time stories started to leak out about visitors and scientists having their shoes melted (anecdotal), but they have measured temperatures ranging above 70 degrees Celsius on that shiny newly exposed caldera face.
At the same time gas measurements taken show that the volcano is degassing at a pace that would make most volcanoes jealous during an eruption. Icelandic volcanoes are famously gassy during eruptions but having one degas as much as one of its erupting brethren, that is something for the history books.
The question is if this is a new thing, or if there was evidence of a change prior? In a paper released in 2007 Jarosch and Gudmundsson found that between 1998 and 2004 a depression that was 100 metres deep in the ice layer above the lake on the eastern side had formed with an associated increase in geothermal output ranging up towards 390 watts per square meter.
It seems like this increase in geothermal activity increased towards the SSW of the caldera after the 2011 eruption and that we can safely assume the heat flux now is well above even the figures given by Jarosch and Gudmundsson.
For those not familiar with Grimsvötn the scale of melting is incomprehensible. The glacier above Grimsvötn varies between 600 and 800 metres thick. This means that between 2 and 4 cubic kilometres of glacial ice has disappeared, with the bulk of it in the last few years.
An interesting sidenote is that in a previous article here in Volcanocafé, I and Andrej Flis found evidence for magma chamber expansion in under the caldera wall next to the area studied by Jarosch and Gudmundsson.
We then hypothesized that this expansion was caused by fresh hot magma heating this side of the magma chamber above the solidus point so that brittle earthquakes could no longer form, creating an expanding radius of earthquakes that had moved as much as a kilometre in under the old caldera rim.
It is not impossible that the same process is now occurring more to the SSW along the caldera rim, and we should get around one day to study that.
As some have noticed, there is a discrepancy in the GPS data for GFUM-station and the tilt data given for the GRVC station. As such this is not evidence of intruding magma since this would give widespread signals.
I would give everyone the friendly hint that until we know more, it is not that fruitful to ponder why GRVC is seemingly shooting upwards like a bat out of hell. My guess, if the signal is even correct, would be that it is isostatic rebound caused by a few missing cubic kilometres worth of ice. But, as of now that is just an idle guesstimate.
A little on Jökulhlaups
At this point I feel that I should write a short thing on Jökulhlaups since Grimsvötn is rightfully famous for them.
They are caused as melting water accumulates under the ice and suddenly pushes up the margin of the glacier and the excess water rapidly pours out forming the largest river on the planet for a short time.
I think that most people believe that the risk for Jökulhlaups would increase if you melt off a few cubic kilometres of glacier.
Let us first look at a couple of things. The first part is that due to continued uplift the caldera lake has domed up and decreased the size of the water reservoir. At the same time the reservoir decreased in size during each eruption since emplaced lava on the bottom squeezes out water.
This means that the average jökulhlaup will be smaller over time.
Now, let us put another nail in the coffin lid over the reservoir. Previous the reservoir had a solid lid above in the form of a 600 to 800 meters thick glacier. That lid now has a fairly large hole in it.
This means that the water pressure is regulated through that hole, instead of out under the glacier. Obviously, enough water and it will come out either way, but for now this is creating a problem for sizeable hlaups to occur. Smaller hlaups can though occur and might even become more common.
The near future of Grimsvötn
That Grimsvötn will erupt in a not too distant future is not a big secret. But the question I have been asked is what these changes are and what the new data means for the upcoming eruption, and if it might be a sign that there could be a second Skaftár Fire around the corner.
I see the risk for a Skaftár Fire as slim to none currently. The reason is that the pent-up strain out in that part of the fissure swarm was well and truly released back when it happened, and that it will take a cycle or two more before it can rip apart again.
There is though a sizeable chunk of that fissure swarm between Lakí and Grimsvötn that did not rift in 1783 and several central volcanoes there. A sub-glacial mix between Gjálp and Skaftár Fires is not out of the question, but not the most likely option.
The second option is that we get a reactivation of one or more of Háabunga, Thordharhyrna, or that East or West Geirvörtur might erupt. Those might happen with radial fissure eruptions. These are fairly likely options. Either during the upcoming eruption, or during an eruption down the line.
Another less likely option is that Grimsvötn has a caldera extending eruption that is in the VEI-5 range (Icelandic sized VEI-5). It is in the light of the extending magma reservoir boundary a bit more likely than another Fire-style eruption happening.
Obviously Grimsvötn could pull of a standard sized VEI-3, and usually this would be the low odds option, after all it has erupted sizeably recently. I do not entirely buy into that, not when considering the changes in heat flux and magma reservoir expansion.
Now for the 3 options I do think are the most likely. These are in no particular order. The first one is that Grimsvötn suffers another radial Gjálp style eruption caused by weakening of the magma reservoir side, the most likely option for this is the case is down the radial fissure towards Hammarínn.
The second option is that we will see 2011 Redux with another brutal fast VEI-4. My favourite option would be that after a brief ashy Surtsey like opening of the eruption an island forms in the lake that transforms into nice shiny lava fountains.
Either way, Grimsvötn is quite able to do some more landscaping in the near future.