I love large caldera volcanoes. The first reason is that they are always very beautiful, especially if they have lakes in them. The second reason is that most of them are of a class of volcanoes that leave their pointier cousins far behind. Size matters after all.
At one end of the beauty spectrum, you have Amatitlán with its lush greenery, flowers, small towns, lake with traditional fisherman, and stratovolcanoes poking out of the morning mist. It is a very human type of beauty but turned to the maximum end of the spectrum.
At the other end of the spectrum, you have Askja, it is stark and otherworldly. The two lakes are different shade of blue, one is a deep blue, and the other is turquoise. It is a landscape of ash and jagged rocks, instead of flowers hiding the raw power, you get the beauty of raw power undiluted.
I do not know if it is my knowledge about volcanoes causing it, or if it is some sort of weird affinity, but at calderas I can always feel the monster deep below murmuring to me when I visit them. Only one other volcano has ever talked to me like that, and that is Hekla. After all it is in the pre-stage of going caldera.
Regardless of type of caldera, or location, they can and often do quite a bit of alterations in the landscape without even erupting. Something that I will discuss more below.
What is a caldera?
I will here simplify things a bit, but generally a caldera volcano is a volcano with an underground volcanic system that is too big to be contained by the edifice above it. The feed rate from the depth is to large over time and magma reservoirs build up over time to prodigious sizes.
For a subduction volcano the calderification is often explosive, and for mantleplume volcanoes the calderification is often effusive. Exceptions do obviously exist, some mantleplume volcanoes explode, and some subductions effuse themselves into big holes in the ground.
In Iceland the large calderas are mainly caused by effusive eruptions draining large magma reservoirs, causing the overburden to drop down into them. Either in a blocking calderification where the overburden breaks up into blocks and drop down (Askja for instance), or as a piston (Bárdarbunga), where the magma reservoir is so deep that the entire overburden moves down in one piece (more or less).
Katlas calderification is less well known, but it is presumed to have been mainly effusive in origin. Grimsvötn is the oddball here, it has 3 calderas that formed during the Saksunarvatn tephras, a series of 5 very large explosive eruptions.
Obviously all four volcanoes have dabbled all over the spectrum, but as a general outline this is good enough.
The Icelandic Calderas
Icelandic volcanism fill the entire spectrum. At the lowest end you have tourist eruptions like Fagrafjall (the mountain formerly known as Fagradalsfjall). After that you have central volcanoes like Eyjafjallajökull that can at times be a spot of bother.
Next in order are the bigger central volcanoes, and minor caldera volcanoes. Here we find Hengill, Hekla, Krafla, Öraefajökull, Torfajökull. There are around 10 of these that can cause some serious problems for air traffic and the local population.
And at the far spectrum you have the Big Four. Askja, Bardarbunga, Grimsvötn & Katla, these are the volcanoes that potentially can cause problems on a global scale. All four of them have deep magma reservoirs, upper magma reservoirs, fissure swarms, and in some cases even satellite central volcanoes on their fissure swarms drawing from the same deep magma reservoirs.
As volcanoes go, they are very versatile and can do a wide number of different types of eruptions at different scales of destruction. From minor mild effusive eruptions to big climactic eruptions altering world weather patterns.
Askja is an old volcano, it is at least 200 000 years old, so it has had ample time to evolve into the volcano that we see today with all its sub-features. It is a nested caldera, and the original caldera event is unknown. One guess is that it occurred during the last ice age, we do though know that it did not happen post-glacially.
The outer caldera has been infilling with fresh effusive lava, as well as deepening at least once post-glacially during the 10 000-year-old large explosive eruption. This eruption was either a very large VEI-5, or a VEI-6 eruption.
It is also possible that a couple of large (5 cubic kilometre sized) effusive eruptions increased further the depth of the caldera.
The inner caldera we can though date with a fair bit of accuracy. It happened during the 1874-1876 volcano-tectonic event (it was so complex that the word eruption is not sufficient, hence the event moniker).
It all started in 1873 with the unrelated eruption of near Holuhraun I. Up until a decade ago it was believed to be an eruption in the same sequence as Askja, but we now know that the lava came out of Bardarbunga due petrochemical testing as well as having experienced Holuhraun II.
I am here leaving out that Holuhraun I (1797), would imply that the 1873 eruption would be Holuhraun II and the 2014 event would be Holuhraun III (aka, Nornahraun which is the official, but never used name for what should be Holuhraun III but is known as Holuhraun II). It is Iceland, it is confusing, and the recently erupting mountain should be renamed into Fagrafjall since it is a pretty mountain. I digress as per usual.
The only reason that I mention this eruption is that there might be a regional eruptive circle going on here. If so, the Holuhraun II eruption in 2014 might be related to an upcoming eruptive cycle at Askja. But many might bees, ifs, and buts… So, please take it with a huge pinch of salt.
Prior to 1874 Askja was relatively unknown, just one minor volcano causing minor mischief in the hinterlands of Iceland. With Katla, Öraefajökull, Hekla and Grimsvötn doing their best to kill people in Iceland nobody noticed it to any degree.
In 1874 a rifting fissure eruption started 45 kilometres north of Askja along the Sveinagja Graben, with epicenters slowly opening up closer and closer to the volcano, with the last one only 25 kilometres north of the volcano. A couple of hardy Icelander’s hiked up there and noted that it was impressive, but nothing like their grandfathers had seen during the Skaftár Fires (Lakí) 101 years prior.
An impressive rift eruption in the hinterlands was though not much to write home about in Iceland.
Inside the caldera all was calm on the surface, well probably since nobody bothered with checking. With modern equipment the picture would probably have been different. What we now know is that hot fresh material was coming up from the deep reservoir, and it went into an old stale disconnected magma reservoir. That magma reservoir was filled with evolved magmas, and it was rapidly heated by the fresh intruding material.
On the 28th of March 1875 a sub-plinian eruption started from that old magma reservoir, it rapidly grew in strength and on the 29th of March it turned into a climactic phreatoplinian eruption. This part of the eruption was a VEI-5 event causing so much distress in Iceland, due to killed livestock and poisoned water wells, that many had to leave the island permanently.
The ash travelled over to Scandinavia where ash was found all over the place, quite to the initial surprise of everyone there. This part of the eruption left a crater behind that later on filled with turquoise water, today it is known as Víti.
The eruption continued until 17th of October 1875 in abated form. At this point the calderification began as the magma reservoir near Víti started to block up and drop down. After more than a year of blocking a caldera had formed that was more than 12 square kilometres in size.
It started to rapidly fill with water, and it is today known as Lake Öskjuvatn, it is 220 metres deep, and from time to time suffers from caldera wall collapses. At the beginning it was quite warm, but nowadays it freezes over during the winter.
After this climactic eruption came a series of seven eruptions between 1919 and 1938 where dregs from the large intrusions was pushed out. And in 1861 there was a small rift eruption at Vikraborgir that came from the deep reservoir, but that seems to have been unrelated to the Askja Caldera.
The problem with Iceland is that Askja’s 3 siblings are all iced over. This means that we do not have any good data from inside the calderas of Bardarbunga, Grimsvötn and Askja. We can only infer from distal data what might be going on, but we do not see what the caldera floors are up to at any given point in time.
Instead, we will have to look at historical data from other large calderas around the world that are active. What we find there is that they can perform some serious landscaping without even being close to erupting.
Even the comatose Yellowstone can cause quite a bit of uplift, earthquake swarms, and new geothermal features and geothermal blowouts, without it being a sign of an upcoming eruption.
The King of Landscaping is obviously Campi Flegrei that can do several metres of uplift without erupting. And let us not forget Amatitlán that is slowly tilting to its side, drowning entire towns on one side, and uplifting other towns on the other side. One entire Mayan city is now under 30 meters of water.
There is something to learn here. Big calderas equal big landscaping. And even big landscaping does not equate an eruption. At Campi Flegrei and Amatitlán we have very long historical records, but at Askja we do not.
For Askja we have data from 1961 and onwards, from the beginning the data was simple tiltmeters, and today we have a good GPS-network combined with all sorts of other equipment. During these 61 years we have seen several minor deflation and inflation events, but nothing major.
Let us begin with what we see at Askja, what the data gives at hand. We see a well-defined area of inflation inside the caldera that is nearing 40 cm in upwards motion. We know that the area affected is small for being inside a large caldera and situated NW of Lake Öskjuvátn.
The distal impact of the inflation is very limited outside of the caldera, in fact it is less than the inflation rate post the 1961 eruption judging from the limited data available back then. We also know that the intrusion is shallow in origin at between 2 and 3km depth.
We also know that the intruding volume is small, between 0.001 and 0.003 cubic kilometres, and we all know that this is not something to write home about in Iceland. If this would erupt it would be an Icelandic equivalent of a shart.
What is not being seen at Askja is probably more interesting. And there is a lot of nothing to be had. Let us start with what the intrusion is not.
We are not seeing any evidence of an increased rate of magmatic inflow from depth, nor are we seeing a deep intrusion taking place, this explains in turn the minute volume of intruding volcanic fluids. This is not good news if you want a big boom at Askja.
We also see background levels of seismic activity. Prior to an eruption being close we would see quite a bit of activity on this front. It is even probable that we would see several large earthquake swarms spanning years prior to the start of an intra-caldera eruption.
We are also seeing no increase in volcanic gases, nor do we see an increase in geothermal heat output in Lake Öskjuvátn, except for a minor heating event a few years ago heating parts of the lake a bit early in the spring.
So, if it is not Hellmageddon 3000 that we are seeing, what then is it? Let me speculate here a bit, but it is speculation based on fact and data from other similar volcanoes, so I do not expect to be far off the mark.
What we are seeing is accumulation of volcanic fluid in a limited reservoir. I do not think these fluids are magma, instead I think it is water moving into hot ambient rock, causing the water to expand further as it heats up. Basically, we are seeing a deep geothermal field being born.
As the size increases of the field there is a minor risk that this will cause a phreatic upwelling and a Maar will be born. It is not as of now a sign of an upcoming major eruption.
This being said, sooner or later Askja will erupt again, after all it is what large volcanoes do on a regular basis. Prior to that we will see a lot more activity. For now, sorry to be the party-pooper.