Volcanology is filled with moments when you look at your favourite volcano doing something interesting, and you hope that it will erupt. There is no shame to admit it, we are secretly cheering our favourites on towards the inevitable eruption.
Time and time again we are though let down by our volcanoes, and if you are like me, you will start to ponder if they are broken somehow. And, knowing you my dear readers, I know that you are quite like me in this regard.
So, let us use 3 volcanoes in Iceland as examples to study for the most common reason of our disappointment, water. I should though say that this is also a common fault in other volcanoes around the world.
Volcanologists at the fault
First, let me start with taking a bit of the blame for this. I and many others tend to use the rather generic term of “volcanic fluids” a tad to often. Like all generic terms it is just either used by someone who is incredibly lazy, or when you are not have an inkling about what is happening, and you do not want to admit it.
So, what are volcanic fluids then? Well, it is most often water, but it can also be gasses in fluid form, or anything else that is floating. Problem is that unless we take a good look we do not know if it is root beer or magma.
So, when a volcanologist uses the term volcanic fluids, you can be almost certain that there will not be an eruption in the near future.
If you instead see your favourite volcanologist use highly specific terminology, like for instance “rapidly propagating dyke formation”, it is time to be exited. Because trust me, that is what your favourite volcanologist is.
Volcanic fluids are used when writing a stale report on something rather mundane that needs explaining for the general population. And writing water expansion is not as scientific sounding as “intrusion of volcanic fluids”.
Problem is compounded by the not so small part that volcanologists firmly believe that every Billy-Joe Bob know that they are mostly talking about water when using the term volcanic fluids. Most readers here do though know the difference, and since many volcanologists read what is written in the comments, they get positive feedback loops in their usage of generic terms. Rinse and repeat.
And then comes The Hysterati*, they do not understand it at all. All fluids are magma according to them, and all magma will immediately erupt, and all volcanoes are Hypercanoes™.
Let me just say that it has been giggle worthy reading what The Hysterati has been writing in the last few weeks in regards of three of my favourite Icelandic volcanoes.
So, let us at a leisurely pace go through what Askja, Grimsvötn and Katla have been up to, look at the signs given, and most importantly let us looks at the signs not given. After that it will be clear why they are broken down and not erupting their magmatic hearts out.
First, I need to introduce a very specific term. A supercritical fluid is when something is neither a gas, nor a fluid, and at the same time.
Let us ponder tap water. As you pour it into your tea kettle and set it to boil it will emit steam. Both water and steam (water vapour) are just two different states of the old H2O, same goes with ice.
These three states are the ones we encounter in our daily lives. But there is a fourth one that we rarely if ever encounter unlike you are like me, it is called a supercritical state. And trust me on this, you do not want to wake up with a few cubic metres of that knocking on your door, it is waters psychopathic cousin.
We all know that steam forms at about 100C at sea level. It can minutely vary depending on extraneous mineral content and local gravimetric anomalies. If you instead boil your water in a pressure cooker, you can achieve higher steam temperatures since increased pressure requires higher temperature to create steam.
So, what happens to water that comes into close contact with magma at let us say 3km depth? If it meets lava at the surface it would convert into steam that is 100C, and then convert back into water.
Problem is that at 3 000 meters depth the pressure is 810 Bar (that is probably something in US imperials, but I am to lazy to convert it into non-metric values, Google is your friend). The temperature of the water will most often be well above 270C or more.
At that point it will want to be steam, but is too squeezed to become steam, so the water turns itself into a supercritical fluid.
Problem is that 1kg of water has the density of 1, whereas supercritical water has the density of 0.326 kg per 1 cubic decimetre of volume. In other words, the water will expand as it is converted by a factor of roughly three.
Anything expanding in fairly solid bedrock will cause all sorts of interesting phenomena, so let us dig a bit deeper into this.
During the last year Askja has inflated with more than 40 centimetres locally inside the caldera. This has led many people to state that an eruption is close. I seriously doubt that it is the case due to what is missing in the picture.
It can be summed up in a single question. Where are the earthquakes? After all, Icelandic volcanoes are famously noisy prior to eruption with two glaring exceptions to the rule, Grimsvötn and Hekla. But they are very special exceptions caused by an open constant feeder mechanism from the mantle that is held open due to them frequently erupting and that they have a steady supply of fresh magma slowly entering the system.
Askja on the other hand is as far as known more episodic, with large intrusions happening, and after that follows diminishing eruptions of ever more stale magma due to lack of influx. Other such volcanoes are quite noisy prior to eruption.
There is also a lack of deep earthquakes and no clearcut deep feeder that is visible on the earthquake data from the MOHO up to the magma reservoir.
What has been detected is though a deep accumulation of magma below the crust trying to find its way upwards, this was especially evident in 2013 prior to the eruption at Holuhraun. This caused a broad and widespread movement upwards of the entire volcanic system of Askja.
Instead, we have a highly localized spot that is inflating quietly at a depth of 2.5 to 3 kilometres depth.
Let us now look at the caldera inside the caldera rim. We know that it has suffered from at least 3 major calderification events, ranging from a mid-sized VEI-5 up to a good sized VEI-6. These eruptions did not directly cause the formation of the caldera due to massive explosive eruptions, or technically it is calderas in plural.
Instead, the void created by profuse effusive eruptions caused the roof to crack and fall into the magma reservoir in the form of blocks of rock. Unlike Bardarbunga that is acting like a piston due to the depth of the magma reservoir, it is easier for Askja to just drop down piece by piece.
This has created a complete mess down to the shallow magma reservoir of crushed up rocks, and crushed rocks are fairly aseismic. It has also created a good ground for water to percolate down through.
We also know that there is a lot of ground water at Askja, we know this from the two caldera lakes inside the volcano. We also know that Askja is hydrothermally active.
So, let us now toy with the thought that water has slowly percolated down through the jumble of rocks until it came near the magma reservoir. Here the rocks are more stable and can create a non-permeable lid.
Under that local lid the water would happily expand as it converts into supercritical fluid and lift the lid upwards. Water that is not under the lid will just move upwards as ordinary geothermal water and heat up the area at the surface, something that we know has happened at the lake.
The lake used to freeze over, now not so much. So, we know there is an increase in hydrothermal water circulation.
This is how you get a mysterious quiet large uplift without any trace evidence of fresh magma intruding into the magma reservoir.
Potentially in the end we might end up with a Maar formation, and those can indeed be spectacular in their own right, but we will need quite a bit more inflation before that happens.
In the last few weeks there has been much activity among The Hysterati due to new hydrothermal openings being discovered in the ice above the caldera lake. It has been well known that the hydrothermal output into the lake has increased significantly after the last eruption in 2011.
We can see that from the steaming ground on the Southern caldera wall, and in a small part of the lake being held open by upwelling warmer water.
Nobody as far as I know have remeasured the thickness of the ice on top of the lake, but it is a fair assumption that the thickness has decreased in the last 11 years.
During the last eruption a portion of the caldera roof was significantly damaged, so it is now far more permeable for water to seep down and get near the magma, but here it has caused a rapid circulation of large amounts of hot water that is slowly increasing the temperature of the lake, and that in turn is melting the ice in interesting ways.
Grimsvötn will obviously erupt sooner or later anyway, but the increased hydrothermal activity is a result of the last eruption and not a sign of the upcoming eruption. Causation is important after all.
No volcano in Iceland is causing so much feverish activity among The Hysterati as Katla. And, for once they sort of have backing from the scientific community.
It is a well-known big hitter when it erupts. A track record containing several VEI-6 eruptions and numerous VEI-5 eruptions is hard to argue with. Even harder to argue with is the Eldgjá eruption.
So, let us look at what signs there are. We do have a record of deep earthquakes leading all the way from the mantle to the magma reservoir. We do have earthquakes around the magma reservoir indicating increasing pressure in the system, we also have a broad long-term indication of steadily ongoing inflation.
In other words, we have all that we want to see a volcano doing prior to erupting. We know that prior to onset of eruption we will see large earthquakes lasting for days, if not weeks, and we know that what we will see will be very noisy indeed. An upcoming eruption will start with earthquakes at M4, or even larger.
If we then see M4 earthquakes it is easy to rejoice that our not so little volcano is about to erupt again. And since repose time often equate to size of the eruption, we would naturally expect something impressive, so we go and buy more popcorn and sit down to wait.
And then nothing… is our volcano truly broken somehow. Sadness sets in, or even perhaps a little bit of anger over our impotent volcano.
But alas, the fault is ours once more. This time around we forgot that not all earthquakes are equal. And once more we forgot the adage of every successful shopkeeper in history. “Location, location, location…”
We also forgot our nemesis, water.
Katla has a relatively shallow and very large hydrothermal system created by the same processes as in other large-scale calderas like Askja and Grimsvötn. All those large eruptions have created a roof above the magma reservoir that is constituted by severely mulched up rock.
Water is easily percolating down through cracks, and then it does the same expansion as at Askja, but this time the orientation is different. Here dykes of rapidly expanding water are cracking the rock sideways as pressure increases.
And when the pressure is high enough a very shallow earthquake will happen, typically at depths of around 100 to 500 metres. After the earthquake a new fumarole forms and the water is vented out over time.
Instead, look for a swarm of earthquakes starting between 3 to 5km that are progressing upwards in a steady manner. That would be magma moving upwards.
How do we fix our volcanoes?
Well, it is easier to fix ourselves. Instead of jumping every time it is far better to be relaxed and check if the signs are indeed produced by magma, or if it is just pesky water.
There is though hope, most volcanoes will erupt again, and these 3 volcanoes are certainly going to erupt again.
For Askja and Katla we can even state that we will see a lot of earthquakes and other activity that we can interpret well enough to make a forecast in time before they erupt.
Grimsvötn is not so easy, it is teetering on the brink as things stand now. It can erupt after just a brief runup tomorrow, or in years to come.
There is after all a general rule that the longer repose a volcano has had, the more impressive the signs will be. Grimsvötn has erupted too much lately to give off a lot of signs prior to an eruption.
I will leave you with this thought. Here we have 3 volcanoes in the same country, erupting similar base magma, and still water is producing three distinctly different false positives. The lesson here is that you need to know your volcano very well before you can say what is what, and that a sure-fire thing at one volcano can be completely misleading at the next volcano over on the same chain of volcanoes.
*=The Hysterati is my own colloquialism for a certain type of YouTube producers who have found out that they can live handsomely by screaming loudly that a volcano is going to destroy the Universe, and that intentionally misinterprets any volcanic news to draw in punters. There is a certain space reserved in Purgatory for them.