For those who have missed the previous 3 parts and the prequel. Here is the link to the last part. In it is links to the prequel.
The nuclear background
To be able to understand large eruptions you need to understand a bit about nuclear weapons. In this case there are several reasons why we need to do analogies with the darker side of physics and the volcanoes that we love to learn about.
But first we have to become archaeologists. Some of you might have noticed that when using carbon-14 dating you get a result with the time set to BP (Before Present), and if you have studied archaeology you will know that this means the year 1950. One might think this is just an arbitrary date, but it is not. This is the last year where you can do any atomic dating due to nuclear fallout from aerial nuclear weapons testing. It will take about 100 more years before it is possible to test our present time. Imagine, from the year 1950 up to 2100 there will forever be a hole in the chronology.
Now, those with a bit of pangeant for Cold War history will believe that the spike in the picture above comes from the Tzar Bomba test, nothing could be more false. The Russian Sakharov-designed nuclear bombs were much better designs than the American Ulam-Teller bombs. The Tzar Bomba used up 98 percent of the available fissionable material, the largest American bomb (Castle Bravo) only used up 13 percent. For all points and purposes every American hydrogen bomb up until the late sixties was a dud, albeit effective enough to kill people on a horrifying scale.
The highest point of nuclear material fallout is in 1963, and that date is one of the two used to calibrate ice core samples. It is easy enough to find. The other year used to calibrate the measurements is the Lakí event. The main reason for this is that those two show up in every ice core drill-sample in the northern hemisphere. It is a pretty handy tool for scientific comparisons between different glaciers and for calibrating timelines within the respective ice core.
The Lakí fallout trace is in two parts. First you will find an ash layer, the second part will be emplaced on top of the ash, and that is a sulphate layer, residual acidity from the sulphurous gasses released by the eruption that both stayed airborne for a longer period, and also was released up to a year after the eruption had stopped.
I have for quite some time known that something was wrong with the modeling of the amount of ash and dust ejected by the Lakí eruption. With the current figures of explosively expelled volcanic ash and blast dust (together with larger local fallout debris) we have a figure running around 1.3 cubic kilometers of Dense Rock Equivalent.
This figure is problematic since it cannot account for the fact that we can find traces of it in every ice core and at the same time account that ashes was collected as far south as Venice in Italy.
The current amount of ash is based on calculations with figures derived from the amount of local larger debris. As such the equations pan out, but it cannot explain the amount of ash in other parts of the northern hemisphere. There should in many places be no ash, or significantly less. Remember here that the amount of aerial fallout is just slightly higher than it was from Eyjafjallajökull eruption, and that is not possible to find across the globe. So something else must be at play here.
Calculating Nuclear Fallout
As any good scientists Volcanologists steal what they need from other scientific disciplines. In physics we steal a lot from the mathematical department. So, it is befitting that the Volcanologists raided us for the formulations to calculate nuclear fallout patterns when they needed to calculate volcanic fallout. The difference is normally so small that it should not matter really. But there are differences, some small and some so large that it becomes a problem.
First of all, as a physicist we differentiate nuclear dust (any dust that became radioactive during the detonation) and true radioactive byproducts from the detonation. The first part will behave like your average volcanic ash. It will drift around for a while depending on size and how high it was lofted into the atmosphere. The second part is basically atoms of fusionable or fissionable material that was not used up in the detonation. This will drift for a much longer time.
This is not so important really; it will though give a negative margin of error. Volcanic fallout normally travels shorter than the model gives at hand since you do not have traceable individual atoms.
There is though a real problem when using the model. The basic model was developed for the original small scale fusionbombs used over Hiroshima and Nagasaki. When we started to detonate the far larger hydrogen-bombs it was rapidly discovered that the formulations had less and less with reality to do. This was foremost a problem with the dirty American Ulam-Teller bombs.
Let us now for a moment get back to the picture in the beginning. The top of the mushroom cloud is at 56 kilometers height, and that is a full stratospheric injection of fallout material. At those heights any particles will pass several times around the globe before they drift to the surface.
We now have to answer a question. What is it that lifts ash up into the higher parts of the atmosphere? Most people believe it is the explosive force that lifts ashes up into the atmosphere, but that is not true. What lift the ashes are convective currents of air. In simpler terms, ashes ride upwards on the hot air rising.
The Tzar Bomba was an almost instantaneous event where a fireball 10 km across was created, and it was at its hottest as hot as the core of the star. In an instant flash of terrible beauty humanity had made a star on planet Earth. It created a cloud of ash that rose higher than anything in human history, but it contained only what it lifted during the first few minutes, and for a couple of months it drifted across the entire northern hemisphere before falling down, almost uniformly spread.
A volcano has a much lower power output than that, but if the volcano is big enough, and heats a large enough area and the eruption last for a sufficient time it will sooner or later rival, or even surpass, the amount of ash ejected into the higher atmosphere.
From Ashes to Ashes
The Skaftár Fires in its initial stage caused an 1800 meter high fire curtain, and it also heated a large area, this 1200 degree air and gasses sucked in air from all sides and created a tremendous upwelling of heated gas and air mixture. This created a bubble of denser than normal air to form above the eruption and into the bubble ash, dust, and shards of tephra was lifted. Anything heavy quickly fell down again as rubble, but the finer particles smaller than 2 millimeter hung there suspended at an altitude ranging from 13 to 18 kilometers. The suspended particles were then moved onwards on the higher air currents at high speed.
The particles then quickly started to fall downwards as the heated air and gas cooled at altitude, and they fell according to size and weight. What is interesting is the time it took. Before this it was considered to have happened rather fast and according to the basic function of the nuclear fallout model due to the assumption that the atmospheric injection stopped at a maximum altitude of 10 km (boundary level between troposphere and stratosphere).
How do we then know that the ashes reached a higher altitude? We know this from the fact that we can find ashes in all ice core samples across the northern hemisphere. We also know that the particle density is almost the same in the direction of the pass around the hemisphere as it is after circumnavigating the globe. The difference is that the heavier and larger particles are gone in the Greenland samples, but not entirely in the Svalbard samples. Incidentally, that also gives us the direction it travelled.
We know that the particle count in Svalbard is on average 250 particles per square meter, and that sums up to less than a gram per square meter. That does not sound a lot, until you start a bit of calculation. Now you say that less than 1 gram per square meter is not a lot to hang up in a Christmas tree. Only problem here… there are a lot of square meters if you sum up the polar zone and the temperate zone in the northern hemisphere. This is not a mathematics site, so I did the heavy math for you on Mathematica. The result is that the minimum weight given by the mathematical model is 25 920 000 000 tones. With an average density of 3 that gives an absolute minimum of 8.64 cubic kilometers of expelled rock in various forms recalculated into dense rock equivalent. Please note, this is the lowest value the modeling gives. This gives that the Skaftár Fires was ashier than previously believed. This may in turn have had an added effect on the climate beyond the gas emissions.
For those who are interested in ice core samples, ashes of Lakí and the gasses trapped in the ice cores of Svalbard I recommend the paper “The Icelandic Laki volcanic tephra layer in the Lomonosovfonna ice core, Svalbard” by Kekonen et al. Also, it holds a nice proof of the temperature shift going down in the year during and after the eruption.