The eruption continues. There is so much hiding behind such an easy sentence. It continues – but always changes. It is not life as we know or understand it is the second most memorable phrase (at least in the paraphrased version) from Star Trek. This eruption is like that. You forget that this is actually a very small eruption, because you never know what to expect; it never does ‘more of the same’. You’ll love it, you’ll hate it, but you’ll never get bored.

At first there was the fissure which became the twin cone, slowly eradicating the valley of Geldingadalir. The fissure followed (approximately) the line of what I believe is a small pressure ridge, and ran along its peak. Just when we thought the lava would manage to escape the valley, it erupted a fissure in a much better location and the lava went straight into the valley of Meradalir, the next target on the list. The lava river was spectacular. But it turns out this was just a diversion. The fissure build a series of cones, but eventually the one closest to the original valley won the competition and exterminated the others. Now the lava was undecided, flowing in random directions, but soon found its way into the original valley – and now our favourite camera was the first real casualty of the eruption, sneakily destroyed from behind. The lava sneaked behind the original cones and continued (out of our sight) the onslaught on the escape route. This time it succeeded, and it entered a flattish area with a swampy pond, a rocky outcrop, and two exits – one north into Meradalir, and one south towards Nátthagi. It was a close call. The swamp became a lake of liquid rock, the rocky outcrop became swamped, and (perhaps hours before escaping south) a lava river developed into the Meradalir – absolutely spectacular but out of sight for all of our public cameras. We did wonder where all the Icelandic antics had gone, where were all the people who had been making a fool out of themselves in front of the lava and camera, but it turned out they were at the action we could not see, and no longer had any need to be the action themselves. And so we thought the action would repeat itself, with a new valley to fill and hopefully a new fissure to form. And so we became the fools.

What happened instead was that the victorious cone shape-shifted into a mini-Pinatubo. The change happened just after midnight on May 2^nd. The tremor suddenly went through the roof, and the eruption went out – and then boomed. And went out – and again went boom. It became a ‘boom and bust’ volcano exploding into action five times or more per hour. The booms were very impressive and our Reykjavik commentators saw one particular lava geyser from their apartments and gardens. What a view it was. The boom and bust culminated two days ago when after a particular vicious boom the geyser reached 400 meters. We were quite concerned for the people watching from a nearby hill who on camera seemed to be overtaken by the flaming cloud. In reality they were well away, in front of the eruption, but they looked very small compared to the spectacle they were watching! The falling debris set fires in many places. Iceland has had a very dry spring, and the moss was defenceless against the heat of the bombs that were raining down. I am amazed that no one was hurt. After this, the eruption went back to stable effusion, though with some huge lava boats blocking the outflow channel, the lava flooded the lava flood plain and almost found its way back south. After half a day, the cone went dark again and soon the boom and bust cycle restarted.

The evolution is beautifully captured in the time lapses made by Virtual.

The big fountaining is seen here.

The lava field in Meradalir is captured by a webcam accessible at http://brunnur.vedur.is/myndir/listi/webcam_meradalahnukurSSV.html Note that it is fairly slow to load and requires a 4k screen to do it justice. On the right is a menu where you can look through the most recent set of frames. It is the kind of view that makes you want to take the next flight to Iceland.

While the volcano boomed, the lava flow seemed to remain fairly constant. There were spurts following a boom, and there were moments when the river ran a bit dry, but it did not seem that the new behaviour affected the overall lava output very much. That remains true. The lava still reached Meradalir and the edges of the flow are still expanding, but there are no major new developments down there. It suggests that te lava flow rate is still in the range 5 to 7 m³/s. It is a small but remarkably stable eruption. The effusion rate is the only thing that is stable about it.

What caused the recent change in behaviour to boom and bust? There are two clues. One is that this is in fact normal behaviour for eruptions, just normally on a time scale of months to years (or longer) rather than hours. You do have to understand this volcano. It has been bottled up for the past 800 years, and now it is like a hyperactive child in the morning wanting to pack in all the playtime it has been missing into this one eruption, and needing to get rid of a lot of energy. This is the Calvin eruption. The second clue is that change in activity was picked up in the tremor plots. They became very noisy just before the boom started, and in the day when it went back to normal behaviour, the tremor charts also calmed down, before resuming its noisy trend.

What makes a tremor chart noisy? There are two main reasons. (Actually four, but I will ignore the third one, windy weather, and the fourth, human activity.) The first is a swarm of small earthquakes, as happens when magma breaks through rock. Such tremor can precede an eruption. The second one is turbulence. Magma flowing through a conduit is actually remarkably silent. The flow is laminar (like honey) and it flows fast in the middle but very slow at the edges. So there is not much friction with the rock, and it is that friction which can generate noise. Just before an eruption, tremor charts can go quiet. This happens when the conduit is ready and fully filled, and all that remains is cracking open the surface. It is the silence before the lava storm. A full conduit is silent.

But what if the conduit is only partly filled? In that cases it is interacting not with rock but with gas, and that can be noisy. Think intestines after a good lunch, with lots of gurgling showing the gas-rich passage of the much appreciated food. Lava is in effect digested rock (life – but not as we know it), and it too produces gas. Magma can become frothy, filled with small gas bubbles. As the magma rises and the pressure decreases, more gas comes out of the liquid and the bubbles can multiply. Tremor charts pick this up as a continuous background.

In most eruptions, the lava is exposed to the open air, and this allows the gas to escape. This is the sulfur smell of volcanoes. If nothing else convinces you, the smell is the final warning of a volcano that says ‘danger – stay away’. (Current evidence suggest this doesn’t work on Icelandic people who seem to be attracted to the heat of lava as moths are to light.)

But sometimes the gas can’t get out. In that case, the pressure on the gas can rise and rise until finally not even thick rock can contain it. Once the lid breaks, the gas comes out and the pressure inside drops instantly, This forces the remaining volatiles in the magma to turn to gas, they expand their volume 1000 times and what started as a break-out attempt turns into a full-blown explosion. The mountain goes boom, and a Pinatubo is born. From Vesuvius to Tambora, this is one of the main ways volcanoes explode (the second way is flank collapse).

Eruptions can enter a late phase where this kind of behaviour becomes repetitive. Agung did this a few years ago. It erupted explosively, then the lava began to flow but only in the crater. A lava lake developed. As the eruption tapered down, the lava lake developed a crust and this became eventually too thick for the magma below to break through. Kilauea’s lava lake should soon enter this phase. But there is still gas below, and it can no longer get out. So it builds up underneath the crust. Eventually it breaks the crust and blows, and the volcano has a late eruption, which may be months (or more) after the main event had ended. Agung did this after its recent activity, and it had done the same (but more powerfully) after the 1963 eruption. It is also a risk at Kilauea, when the current eruption ends. Even though an explosion down its 200 meter deep hole may be well contained, rocks can fly far. You have been warned.

Let’s go back to the Reykjanes eruption. For now we can make a guess on why it went boom and bust, and why it temporarily went back to steady growth.

The tremor shows that there was gas underground which could not get out. At the same time, the lava glow in the cone had gone dark. Where had it gone?

It seems that the lava had retreated below a lid. A crust had solidified, and the magma was suddenly caught below it. Iceland had tried to put the Reykjanes genie back in the bottle. Calvin did not take kindly to this. The internal pressure build up, as gas collected below the lid. After a while (10 minutes or so) it overcame the lid and the kid exploded back on the scene. The exploding gas took the lava with it, in a tall red geyser. After the geyser, the remaining gas bubbled out causing a phase of turbulent lava flow in the cone. But the lava itself lacked sufficient pressure to make this permanent, and once all the gas had gone the lava retreated, waiting for the next cycle.

Why did it change after the massive explosion? That explosion happened after a collapse of the walls of the cone. Two things may have happened. First, the rubble may have blocked the exit even more than it already was, and in consequence the gas pressure build up higher before it could overcome the weight. Second, the collapse cracked the lid and provided a hole which triggered the eruption, but the hole it made was very narrow. In either case, it made quite an entrance. The geyser went hundreds of meters tall, aided by the narrow nozzle (we know the opening was narrow for this eruption because the lava geyser was well focussed: the width of a jet is closely related to the size of the exit hole).

But this explosion was big enough to do real damage to the plumbing of the volcano. The back pressure wave moved through the magma conduit, and was big enough to even force out a bit of magma from the original cone – which had been dead for weeks. The zombie apocalypse produced a small lava flow. The fact that one cone could damage another shows that the fissure is fed by a single conduit, by the way. If each cone had its own conduits to the dike, then the pressure wave could not easily find its way to the other cones.

Now two things may have happened which changed the shape of the eruption to strong and stable. (The UK does not like stability, by the way. The last prime minister going to the polls on a record of ‘strong and stable’ lost her job. The current incumbent is about as unstable as they come, and has two clear election victories as reward. Life in the UK is an adventure. But back to Iceland.) The pressure wave may have pushed open the feeder channel and increased the magma flow to the cone: the flow rate increased enough for the magma to stay above the obstructing lid in the cone. Or perhaps somewhere, perhaps through the zombie cone or through the side of the active cone, a crack was formed in the rock which allowed the gas to escape.

The phase of stable eruption did not last long. Either the crack closed again (a 5-cm wide crack filled with lava could solidify in half a day), or the feeder conduit narrowed again and the magma flow went back to the old level. Gas began to collect again below the lid, and boom and bust restarted. Volcanoes will not be denied.

So this is my guess on what happened. It is speculation!

What will happen next? I can safely predict that something will happen. The tremor charts show that the tremor has steadily (linear, in fact) increased since May 2^nd. The current phase of instability is itself unstable. It is heading for another change. The only uncertainty is what change it will go for.

Perhaps the cone will collapse (it is looking pretty unstable already). Perhaps a new fissure will open. Perhaps the eruption will end. Perhaps it will pretend to end while in fact preparing for bigger boom. Or perhaps it will fool me and the current phase will last weeks, with fountains visible from Reykjavik. Or perhaps Hekla will blow. Predictions are best made in hindsight.

But this volcano has become much more dangerous to sight seeers. The warning signs are clear.

Albert, May 2021