Stunning photograph of two fisherman on Taal Lake inside Taal Crater inside Bulkang Taal. Photograph taken by Michael Angelo Luna, Wikimedia Commons.
I am taking this opportunity to look at 5 volcanoes that at least I find interesting. I am doing this out of the perspective of the life-curve of an eruption. I find this perspective to be interesting, and I think that it is something that would be both entertaining and informative for our dear readers.
I have noticed that there are a few misconceptions constantly going around about volcanic eruptions that I would like to clear up, and these five volcanoes gives an ample opportunity to do so. Each of these volcanoes gives an opportunity each to clear up one of these misconceptions.
There is though a general point I would like to make. For most volcanoes it is possible to, with a fair degree of accuracy, forecast an upcoming eruption. Mind you, I am not talking about predicting an eruption. Predictions are made in teacups filled with fish-entrails, or by hysteric YouTubers filled with the Dutch Incense.
A forecast is made upon verifiable data, scientific theory-building, historic data of previous eruptions, modelling of data, and so on. It is in other words the volcanic equivalent of weather forecasts and is nowadays done with a fair degree of accuracy. For those who are familiar with my writing, they know that this is something that I love to do for volcanoes I know well, and with a fair degree of accuracy.
The reason behind my interest is that back in 2010 when I said that it was possible the dogma was that it was not possible. So, I got a fair bit of slack for me stating that it was possible. The interesting part was that Professor Páll Einarsson had already stumped the world with his insanely accurate forecast of the 2000 eruption of Hekla where he missed the time of the eruption with a single minute.
What I have been interested in lately is the possibility of forecasting the end of an eruption. To my surprise I found that I got even more slack for that assertion then I got back in 2010.
So, there will be a bit about forecasting in this article, it will though not be a complete article about it. I will get around to doing that one of these days, I hope.
Just a few final words about my interest in eruption forecasting. Well, if you plan to build billions worth of geothermal plants it is incredibly wise to know where and when a future eruption will occur. Simple as that, and admittedly a tad of professional pride also.
Now, time for me to stop being an incredible windbag and get into the volcanoes at hand.
Death of Fagradalsfjall
How we remember Fagradalsfjall. Image captured from the live feed.
As an eruption begin it heralds the end of itself. Well, at least for normal and sensible volcanoes that are not Stromboli.
I like to look at eruptions like the profile of a volcano. At the left side you have the intruding magma and the steady increase of earthquakes, onset of degassing, inflation evident on GPS-stations, possibly a few phreatic blasts, and so on. I used to find this part of the eruptive curve to be the most interesting.
At the peak you have the onset of the eruption. After that you look at how much magma is incoming, and how much is moving out, normally there is a deficit in the incoming magma versus the outgoing magma.
If there is a net deficit of incoming magma you will see deflation on the GPS-station, and if this is the case the earthquakes are associated with shrinkage of the magma reservoir, and not a sign of new magma arriving. Many people make the mistake of believing that renewed earthquake activity is a sign of magma arrival and believe that the eruption will continue for a long time, or even increase.
If incoming magma was the case for the earthquakes, it would quickly show up on the GPS-station(s). If it does not and deflation continues the earthquakes are the sad gong-blows of the impending end of the eruption.
During the first 30 something days of the eruption at Fagradalsfjall the GPS-signal kept surprisingly stable and did not show any deflation. The roughly 10 cubic meters per second that was incoming from the mantle was ejected at the top of the vent(s).
Station just south of Fagradalsfjall, it is the one showing best what is happening at the previously erupting vents. Image from Icelandic Met Office.
This led to a forecast from the Icelandic authorities that the eruption would last for years. I did not have any large opinion against this assumption of a steady state eruption more than that it might be a bit premature to state that. But, at that point the forecast was underpinned by available data.
In all honesty, at that point I agreed with them. I might argue with other volcanologists, but I never argue with data. Data always win those arguments.
Later as the rate of the eruption increased the data started to show a slow deflationary trend. This perked my interest, and I followed this slow decline for a while. To my surprise the “years of eruption”-meme took on a life of its own, despite what the data was evidencing.
And since I get an almost perverse joy out of being an iconoclast against the powers that be when I find that they are disregarding the available data, I sat down and went through oodles of data and made a forecast based on what was evident from the data at hand.
I published this forecast over on our Facebook group during the summer. In it I stated that the eruption would end within 90 days, with the most likely time for cessation would be at the end of the time period.
This exploded out over the internet and took on a life of its own. Even Páll declared that I was ejecting lactose out of my nostrils, well he used different words, but the sentiment was there. I just reiterated the facts, stated that he would owe me a beer soon, and sat down to follow the data as it came into being before our eyes.
So, what will happen at Fagradalsfjall in the future? It seems like a good question, after all there are still a lot of earthquakes there.
The possible future of Fagradalsfjall
Snowdusted Keilir in the distance. Photograph by Soffía Snaeland, Wikimedia Commons.
Caveat: Currently the data available is unclear and any changes are minor, so this is based on less data than I would like to have at hand.
The question here is if there will be a new eruption at a different location, if there will be a new eruption at the old vents, or if this was it for the foreseeable future.
This is a tricky question. Yes, the eruption has ended, we can all agree on that. But the data is not completely excluding the possibility for a new eruption in the future.
Often after an eruption the GPS-trajectories will after a while start to show a slow deflation-curve. This is in turn caused by cooling of the residual magma. In this case so far, the trajectories are remarkably steady, or even showing that a very small inflation is occurring.
Here it is good to remember that what set off the eruption was magma pooling below Reykjanes, that in turn set off pent up tectonic strain. First the magma caused a fault to break open under Thorbjörn, but the pent-up strain was not great enough to form a dyke all the way up to a surface vent.
As the strain there was spent the magma instead released the pent-up strain at Fagradalsfjall, and a rapidly dilating dyke formed as magma rushed into the voids created by the earthquake activity. The magma entered from below just south of Keilir and travelled laterally over to Fagradalsfjall where it erupted.
By now that strain is well and truly spent, and a new rifting fissure eruption there is about 800 years away, which seems to be the time needed for that much strain to accumulate.
The station that is showing Keilir best. Image from Icelandic Met Office.
The strain is fully released at the site of the eruption, but there might be a bit of strain remaining over at Keilir.
That just leaves magmatic influx as a potential driver for a new eruption. And there seems to a be a small amount of that occurring. There is though far less magma entering into the system, and the system had a magma deficit as the eruption stopped.
So, a new eruption is right now a bit away if it will happen (unless inflation picks up again). Right now, the data is hanging in the balance. Over at Keilir there are earthquakes slowly moving upwards, this might indicate a formative new conduit there.
Over at the original vents the earthquakes more seems to substantiate conduits failing as magma is withdrawn. The GPS trajectories is simultaneously indicating minor dyke dilation just SSW of Keilir, and at the same there is slow compression of the dyke over at Fagradalsfjall itself.
Currently I see a very low probability of a new eruption near the previous vents, and a low but existing risk of a new eruption 1km SSW of Keilir.
Most likely though the next eruption from this volcanic system will be at least 800 years into the future.
Regardless, I am going to Iceland late spring or early summer to collect my beer from Páll, and it would be nice if there was a volcanic bonfire there to celebrate the occasion. It is though unlikely that the bonfire will be from this volcano.
The Dying of La Palma
Image of the ash covered cemetary of La Palma. Photograph borrowed from Daily Sabah under fair usage.
The Canaries and La Palma is thankfully less complicated since there are no pesky effects of the Mid Atlantic Rift to contend with. Here it is a simple case of magma going up into the island, forming a new dyke, and then erupting.
And as the lava rushes out it is happily deflating spelling out the upcoming end of the eruption. The eruption has come with some unexpected turns and twists that are well worth mentioning.
First surprising thing is the size of the eruption. If this had been in Iceland it would have been a small eruption and would most likely have ended up as a footnote eruption. But, for an eruption at La Palma it is quite sizeable, in fact it is by now the largest eruption there witnessed by humans. Turns out that size is relative after all, at least if the receiving end is comparatively small.
It is a part of the human psyche to have a good thing last for a very long time, and it is also a part of the human psyche to have our good things ever getting “better” and larger. This is especially true for volcano afficionados, both for laymen and professionals. Deep down in our grubby little hearts we wish for our favourite eruption to increase over time until the point of Greenland flying towards the moon. Yes, I am indeed implying a fair bit of addiction here, I have dubbed us “volcanoholics” for a good reason.
Problem is just that nature seems to disagree with our wishes. I am equally affected, so this is not me pointing any fingers. But I am well aware that what the data tells, is what will happen.
The amazingly colourful impending death of an eruption. Image from IGN.
I am mentioning this since many people tend to misinterpret the data at hand, or disregard at least parts of the data, in the hope of a bigger upcoming volcanic fix. Instead, remember that as one eruption wanes, nature is busy preparing the next eruption some place else. Then the fun begins anew and fresh, something that I find pleasing.
So, what is the data telling us? Simple, the larger than average eruption at La Palma will end soon. There is clear deflation visible, the northern end is rebounding upwards again after having been pushed down by the initial inflation, the dyke is narrowing. Death is nigh to the delight of the residents of La Palma.
Another unusual thing is that this eruption have been unusually explosive for being a basalt eruption. We have all seen the pictures of houses totally enveloped in ash and lapilli. There is still an ongoing discussion about what has brought this about.
Some state that it is due to large amounts of volcanic gas. Yes, there has been quite a bit of gas, but less than what comes out during an Icelandic eruption. Others have put forth that there is a high content of water in the magma, either from deep down or from aquifers in the conduits. I have so far not seen any data substantiating this.
The third option is that the eruption is partially drawing on remobilized magma from older dykes, that in turn is containing volatiles like water and volcanic gases. This idea would imply that there would a part of the lava that is more evolved than the basalt from depth.
Until we have data from samples taken, we will not know with any certainty what is going on, but if I was pinched, I would favour the volcanic dregs out of dyke’s theory. After all, that is happening quite often at other basalt volcanoes when they are unusually explosive.
So, the big question. When will the fun end? It is already showing signs of sputtering, so I would say anything from today and 90 days forward. Currently I favour a shorter option judging by the data from the GPS-stations. A few days to a few weeks are most likely.
Return to Sleep
It is hard to grasp the size of Askja. Photograph by Wolfgang Beyer, Wikimedia Commons.
A while ago the collective eyebrows of every volcanologist moved in the direction of Venus as we followed the onset of rapid inflation at Askja. The reason for this is that new eruption cycles of Askja comes about 150-230 years apart and can be quite a handful.
Askja unlike the previous volcanoes is one of those big ones that can wake up and do a number of different styles of eruptions, some styles come with a definite flair for the dramatic.
Since the last ice age, it has done everything from small eruptions that are best described as cute and cuddly, via multi-cubic kilometre rifting fissure eruption, and large caldera-forming VEI-5 eruptions.
What we do know is that the first eruption of a new cycle tends to be bigger than the following eruptions. This is due to the initial one being driven by fresh and hot magma arriving up from the mantleplume, and the subsequent ones are driven by leftovers from the big show.
There is also quite a bit of old evolved magma in a fairly sizeable magma reservoir, and if that is remobilized by the hot magma this can cause a powerful explosive eruption. This last happened in 1875 when hot basalt remobilized rhyolite in a way that is not conducive to the life expectancy of volcano tourists.
Unlike during the runup to Fagradalsfjall I started to pack my things, about 200kg worth of equipment, but as I sat down and waited for the signs that would indicate that an eruption was a week away or so, things became boring as the uplift started to tail off.
OLAC-station at Askja. Image by the Icelandic Met Office.
Wondering what had happened I took a deeper look at the GPS-trajectories and discovered that we all had been fooled a bit. Yes, the trajectory at the Ólafsgígar GPS (OLAC) had been substantial with 12mm south, 20mm west and a whopping 160 millimetres upwards in 2.5 months, there was not much happening at other nearby stations.
Yes, the other stations showed a bit of movement, but not nearly as much as at OLAC. This means that the intrusion was localized and small for such a large volcano.
There was also the problem that the rate of inflation rather unceremoniously declined and reverted into deflation. So, for the time being there will be no pictures of a swearing Carl carrying 200 kg of equipment up the side of a volcano.
In the end I know that this was the first sign of the beginning of an eruptive cycle at Askja, there will be more intrusions, more inflation events, and one of those will be the real deal. We also known that we are nearing the peak of the mantleplume cycle, and that will set off Askja in the end. So, I am keeping the gear at the ready. I give it a decade or so at most.
The snails trace of Grimsvötn
The Cumulative Seismic Moment Chart. Icelandic Met Office.
If you wish for something large in Iceland Grimsvötn is by far the best bet currently. Hekla seems to be sleepy right now, and Katla seems to be saving its energy for something a tad larger down the line.
All the way since the larger than average 2011 eruption Grimsvötn has had a slow but steady increase in average numbers of monthly earthquakes and also the average size of the earthquakes as they occur.
The open waters of Grimsvötn. Photograph borrowed from Morgunbladid.
The eruption of Holuhraun in 2014 slowed down the proceedings a bit due to decrease of external pressure from Bardarbunga, but from 2017 and onwards the increase has been slow and steady.
The same goes for the GPS-trajectory, it is showing slow and steady inflation and dilation of the magma reservoir as magma is entering at a steady pace from down below. There is also an increase in thermal energy output into the lake, so much so that it is showing open water during the summer near the last eruptive went over at the caldera wall.
By now both me and Albert have technically lost the bet on who could best forecast the upcoming eruption, but since Albert had a later date than me, I will still buy him a right nice pint when I can nail him to the chair in a pub again.
Even though it is moving along nicely towards the upcoming eruption it is starting to feel like it is doing so at a snail’s pace. But there is a clear snail’s trace left in the data, so I am happy to wait a little longer. Currently I would say April, but that could be wrong once again.
The Swansong of an Island
The phreatic detonation in the boiling lake at Crater Island in Taal Lake in Taal Caldera inside Bulkang Taal. Image from Phivolcs.
This part is something that I have hoped that I would never have to write, and I hope with all of my heart that I am wrong in this case.
The 234 square kilometre caldera of Bulkang Taal (Taal=Pure or True) would to most people not look like a volcano at all, instead it looks like a beautiful and peaceful lake with an island in the middle. If you look at the island you will find another lake, and that lake would definitely give things away since it is merrily boiling all the time.
Nobody who is reading this will have missed that Bulkang Taal, The True Volcano, have been active since 2020. The local authorities have at least twice raised the alert level to 3 on a 5-degree scale, and with good reasons.
There have been minor phreatic blasts in the lake as water have come into contact with fresh intruding magma, the gas flow has increased at times to levels where the capital of Manilla was inundated with hazardous vog, there have been widespread and substantial uplift and caldera extension.
And there have been numerous earthquake swarms, both of volcanic type earthquakes, and tectonic rock breaking types.
What is giving me reasons to be less than happy here is that the activity has been going on since March of 2019 and since then the activity and the data at hand has slowly increased in severity.
For a lack of better words, it has been like seeing a giant slowly wake up, put on the clothes, having coffee, and then going out in the garden starting to build up a mountain of gunpowder. The longer it is building up that metaphorical mountain of gunpowder, the worse it could be when the fuse is finally lit.
Long runups like this is generally not a good thing around volcanoes of this size and type. If it erupts fairly fast after onset the eruption will most often be manageable, but years of building to an eruption? Rarely a good sign.
Currently I would not be surprised if the eruption when it finally occurs will be in the range of the eruptions during the 18th century, or even as big as the 1754 eruption. As such they would be a disaster for the area in and around Bulkang Taal.
What could we expect from such an eruption? At least 1 meter of ash covering the downwind shores of Lake Taal and quite a bit of disruption to the economy of the Philippines.
A more unlikely scenario, but that is well within the capacity of Bulkang Taal, is a larger eruption. It is here prudent to remember that the entirety of the scenic lake has been created by a number of eruptions ranging from VEI-6 to VEI-7. The sad part is that the longer the runup-phase last, the greater the risk increases of something like this happening.
Currently the risk of something horrendous happening is perhaps 1 percent, but even that is an uncomfortable number indeed.
I do not like being restless volcano calderas like The Pure Volcano. Not at all.