As the numbers for the Hunga Tonga eruption continues to come in it is becoming ever clearer that something truly momentous happened, something not seen or heard in 139 years.
With a columnar height of 55 kilometres, an explosive pressure wave travelling several laps around the planet, forming a medium sized deadly tsunami, gouging out a few cubic kilometres of rock, ash and silt, and so on and so forth it was to all points and purposes quite something.
Problem is that when we compare this eruption to a more common caldera forming event like for instance Pinatubo, we do not get the numbers to ad up. With such an enormous eruption column we should see very high amounts of SO2, and we should see a lot more erupted material.
The explosion does not seem to add up to the eruption. We need to figure out how such a comparatively small eruption could create such an over-sized explosion.
Judging from early satellite pictures we can estimate the eruption to have been between a mid-sized VEI-5, up to a miniscule VEI-6. So, we are back to 1991 and between Cerro Hudson and Pinatubo. So, the eruption is only small if we compare it to the explosion itself, in any other way it is the largest eruption in 31 years.
As explosions go it was the largest explosion witnessed by humans. Now remember that we are talking about the “boom” and not the eruption, humanity have witnessed quite a few larger eruptions.
Prior to Hunga Tonga the second largest explosion witnessed by humans was Krakatau in 1883, it has been estimated to have been between 20 and 30 megatons of TNT-equivalent.
It took the machinations of Beria, the insanity of Stalin, and the bizarre genius of Sakharov, but it in the early morning of the 30th of October of 1961 an enhanced Sluika with the project number AN602 was detonated with a yield of 50 megaton of TNT-equivalent.
The world named it the Tsar Bomba, and humanity for a brief moment in time grew a brain and stopped that particular direction of the atomic race.
Problem is that Nature is not that easily out staged. It was at that time busily priming a small non-descript volcanic caldera in the Kingdom of Tonga, named after the two small and equally non-descript caldera-rim islands, Hunga Tonga and Hunga Ha’apai.
After a bit of ever more impressive volcanic activity Nature was ready for the main show, and on the 15th of January it was showtime. Judging from audio tapes the main eruption was best counted in seconds, and in those few seconds Nature delivered an explosion in the range of 55 to 60 megatons of TNT-equivalent.
Incidentally, as I was writing this article the agency tasked with judging sizes of Nuclear Explosions just issued their calculations. It turns out I was spot on with my number of 55 to 60 megatons of TNT-equivalent. Just to be clear, I calculated this number on the 16th using barometric pressure readings, having confirmation of not being a “loon” is though always nice. Insert your favourite personal snicker sound here…
All we are left with are two miniscule remnants of the original islands, a gouged-out caldera, and a load of questions.
I will here try to answer two of them, and they are intimately inter-connected. Where did the SO2 go? And, what in the name of heck caused the explosion?
The missing SO2
This part is fairly easy to answer, and it has a profound effect on the second question. SO2 is a volcanic gas emitted from volcanoes together with fresh lava. Old lava contains comparatively little SO2 since almost all of it is transformed into all sorts of organic and non-organic sulphuric compounds, and those do not readily transform back into SO2 if lofted skywards.
We know that during the explosion a bulk portion of the bottom of the caldera decided to go and watch the movie The Adventures of Priscilla, Queen of the Desert at location in Australia.
In other words, it was a mixture of volcanic rock, old ash, old tephra, old pumice, and silt, that was exploded outwards, and that very little of fresh material was exploded skywards. And with little fresh material we get little SO2.
The Eruption Purist Interlude
There is a particular brand of stupid that now will start to argue that it was then a very small eruption, a VEI-2 or a VEI-3.
They will now categorically claim that the definition of every conceivable eruption scale is the amount of fresh material erupted.
I invite them to go and stand at a VEI-3 safe distance from the next Krakatau style eruption. Arguing pointless semantics is not your friend around volcanoes if you wish to remain alive.
The rest of us are probably happy with the total amount of ejecta, and the implications that have on life expectancy. So, let us move on to the main question. At least I am positively wet from anticipation by now.
The Other Volcanic Fluid
Most of our readers are familiar with the term “volcanic fluids”, but I should explain it the same. It is a term often used by volcanologists as they describe any unknown fluid inside of a volcano. Often in relation to inflation or deflation of various sections of very large calderas.
The fluid could be magma, water, molten sulphur, and so on. Basically, anything that is fluid enough to move around causing changes in a volcano, either by intrusion, or by being pressured into a new area.
In large dying calderas like Yellowstone this would be counted in as little as an inch or two over a year. But, in other livelier calderas it could be counted in metres in just a day or two. The latter holds true for calderas like Amatitlán, Campi Flegrei, and Tondano, just to name a few.
Typically, both laypeople and volcanologists are way more interested in the potential for magma movement, after all, that is what we perceive as the dangerous version of volcanic fluids. Turns out we may have been quite wrong about this.
Up until now the only type of volcanologist being arse bothered with the water part was geothermal volcanologists, and we are far and few in between.
So, to understand things we need to study water for a while.
Don’t tickle when wet
Water is an amazing and life-giving fluid existing in a surprising number of different forms. Let us get more intimate with those forms.
First of these forms we have the whisky cube form known as ice. The only version I like is the ice cubes in a drink, I come from Northern Sweden, and we hate Ice in all other forms. It has a magical property, and that is that it takes up a larger volume in frozen form and is thus lighter than water. This is why it is floating on top of lakes.
This is important to people from Northern Sweden since it makes it possible to drill holes in the ice to fish. It is a very miserable and cold version of fishing. It also gives us the opportunity to drive around on top of lakes with building material where there are no normal roads.
Then we have normal water existing between 0 and 100 degrees if we are at oceanic altitudes. We drink it, swim in it, drown in it, use it to cook, and so on. Most people use it to make coffee or tea. It is more often beneficial than not.
Above 100 degrees Celsius (there is probably some weird Imperial number for this that I do not know, and am to lazy to google) we enter the realm of steam. Victorians with impressively high hats understood steam, most of the rest of humanity do not. So, let us don a very high hat and barge onwards into our steamy business.
As your teakettle starts to boil you will see white stuff start to waft around, most people think this white stuff is steam. Alas, it is not. It is tiny water droplets that are suspended in the air. They form as the surface turns 100C and steam forms to instantly cool below 100C and the droplets form. The steam-layer is literally one molecule thick top-layer of water, so you will not be able to see it.
It is though amply possible to hurt yourself with your white fluffy wispy stuff. But, to get steam we need to raise the temperature beyond 100 degrees Celsius.
To go beyond we need to ad another force, more heat does not cut it, that would just produce more low energy suspended water droplets at a faster pace. We need to insert pressure into our equation.
Water is funny, it will boil at different temperatures depending on your elevation, this is due to the pressure dropping the higher we go giving us lower and lower temperatures for when the water droplets will form what we call “steam”.
If we instead increase the pressure, we well and truly reach into the realm of flash steam. This is any temperature between above 100 and 184 degrees. In the latter case we need a pressure of 10 Bar (atmospheric pressures, the atmospheric pressure is really 1.01325 Bar, but close enough). This is the same pressure you would find at 100 metres depth of water.
Now imagine being that 184 degrees Celsius pressurised water, as long as you are at or above this pressure you are water, but if you even drop a tiny amount below this pressure, you will instantly flash from 1 litre of water into 2100 litres of suspended water droplets. But if you are contained in a volume less than 2100 litres you will stay in true steam form above 100C.
If you are above that temperature you will get something called dry steam, it is an invisible mess of hotness. It is steam kept at high temperature and pressure, but below the flashover point.
Now, remember the poignant word for us is flashover, this is the point where any amount of water will explosively decompress from water into steam.
There is though one step up from this on the giggly tree of water. It is called Supercritical Fluid. It is water that is at such a high temperature and pressure that it simultaneously behaves as a fluid and a gas, while being neither of those things.
If pressure has the upper hand, it is behaving more like a fluid, and if temperature has the upper hand, it behaves more like a gas.
Steam and water up to this point has been manageable, you obviously need to treat it carefully, but you can safely build systems to power ships, steam locomotives, and all power generation plants with it.
Well, the safely part is not true. Those high hated Victorians literally blew up tens of thousands of workers during the industrial revolution before they mastered their craft, ho-hum history is fun.
Water in supercritical form is the suicidal psychotic giggling version of water, it not only wishes to not exist, but it at the same time also wants to take everyone with it in a big boom. If that was not enough, it likes to go through solids like they where made out of paper. It is also very good at dissolving solids. You need specialised materials to even contain it.
So, you want numbers on this branch of the giggly tree? Supercritical water forms at minimum of 373 degrees Celsius and 220 Bars of pressure. Yes, this is the minimum, in some volcanic systems the Supercritical water can be up towards 800 degrees with corresponding mind-numbing pressure values.
This is the only form of water that could have caused the explosion at Hunga Tonga. There may have been other more normal forms of water involved, but they lack the energy density to produce that kind of explosion.
Incidentally, 220 Bars of pressure is reached at less than 1 kilometre’s depth if we use the specific density of bog standard rock (2.3kg per litre of volume, it is probably something in Imperial that is really hard to calculate).
What was down there?
If you are like me, you will often and fondly imagine that you are a piece of drill steel being drilled into a volcano. It is not as bizarre as it may sound, after all my daytime job is to plan how to best drill into volcanoes for geothermal fluids to drive geothermal power plants with.
Even though we do not exactly know how Hunga Tonga’s geothermal system looked like, we know a lot about geothermal volcanic systems in general. After all, people have been drilling into them for quite some time.
Most often the drilling has been down into more normal forms of geothermal water. But, at times we have found Supercritical water down there.
A caldera is most often a closed off geothermal ecosystem. At the border of an active caldera, you often have an active ringfault, the inner side of it is often filled with material that is easily permeable to water, while the outside tends to be more sturdy.
Int the case of Hunga Tonga this geothermal ecosystem had a free and endless supply of surface water in the form of the Pacific Ocean. And over time that water slowly permeated down through the crushed up rock that constitutes the roof of the magma chamber.
As it went down it heated up and started to pool in layers (geothermal aquifers) depending on pressure and temperature. The upper ones would be the flash steam portion, but that is not as such interesting to us.
We are mainly interested in the water that climbed up the giggly tree hellbent on throwing itself down, aiming to hit every single branch, in other words let us stick to the Supercritical water.
There was probably one such supercritical layer slightly below 1000 meters depth. This would have been the prime driver of this layered volcanic cake (Sluika = Layered cake in Russian, just another hilarious detail).
Below that there was at least another layer of even hotter and more pressurised supercritical water. There is circumstantial evidence of a third such layer, so let us say that there was 2 to 3 supercritical geothermal layers inside the caldera.
These layers often move about upwards and downwards as they go, and the water is circulating around in them in unexpected and intricate ways.
At older more mature calderas like Yellowstone the supercritical water will find ways upwards with time, and they form some of the more spectacular features like geysers, hot pools, and all the other assorted geothermal joys in life.
Hunga Tonga was young and probably had not had time to become a truly circulating geothermal system. Well, not as far as we know that is.
As long as there was no eruption the supercritical water was happy where it was, it was of course trying to run away to the surface, but it seems like it was very slow going. And that was a good thing since there seems to have been 0.5 cubic kilometres of giggling hell-water down there.
The recipe for doom
The reason that these explosions are rare is because you need a large eruption to kickstart them. A normal eruption will just punch a hole straight through the aquifer with the lava quenching the hole as it passes through. It is like the lava is wielding the aquifers shut as it passes through them.
Most of the water that was seen in the vent came from much more shallow infiltrations of sea water, so let us not mix them together. What little came up from the depth just worked as an additional driver of the eruption.
All was well up until the 14th of January. It was a typical fairly benign Surtseyan eruption slowly building up to an archetypical VEI-4 eruption. On the 14th we had the main eruption, a VEI-4 of some magnitude that removed quite a bit of the island the vent was located at.
It was this excavation of material that spelled the doom of the caldera. Instead of rockmaterial of various sorts compressing the first supercritical layer we now had comparatively lighter water on top. Instead of the 220 Bar we might have had as little as 100 Bar.
I wish we would have had instruments and a camera there when what followed happened. It would have shown a wild ride.
As the pressure from above was lowered the Supercritical water started to move from a more fluidlike state into a more gaslike state, in turn pushing what was above upwards. In the first hour the uplift was probably just a few centimetres, but in the end the ground probably bulged several metres an hour.
In the end the rock layer could not hold it back, instantly all of the supercritical water transformed into dry steam (flashover) and as it did so, it flung up the rock above it in a process taking a second at most. With this layer removed there was not enough pressure left to contain the next layer, and that went up a few second later, and most likely the process had a final third explosive decompression.
All that now remained was for the ocean to fall into the hole that had been produced. At this point there was just a little bit of boiling of the sea water, and the eruption was halted for this eruption.
In a while an “Anak Hunga Tonga” will be born as the volcano rekindles its work, but that is a story for the future.
The question that remains is as follows: Have we just been lucky? Is this a style of eruption that could be far more common that we previously believed?
I think we have been lucky indeed. It is most likely far more common than previously believed, because this type of eruption has a much more common form named a Maar-formation.
Unlike a normal Maar-formation (that can be quite ugly) any calderas contain large amounts of supercritical water, and in my view, it is imperative that we upgrade our risk-assessments to also include this style of eruption.
For calderas near cities like Campi Flegrei it is important that we drill deep into the roof of the magma reservoir to build accurate maps of how much supercritical water there is, and where it is located.
This was a fortuitous wakeup call for science. We need to learn, and learn quickly, because having Tsar Bomba going off next to Naples is bad mojo indeed.