Taal eruption update

Image of how ash affects people inside the caldera. Image by WBEN Radio.

In my previous article I wrote that Taal most likely would not continue to erupt for a long period of time, based on the previous history of the volcano. Obviously Taal decided to go for something not recorded in the historic annals.
It is therefore time to take a deeper look at what is going on using an array of different data.

But, before that I need to straighten out two different things that have been circulating in the media, and on various internet sites.

The first part is that the BBC have misunderstood the volcano completely, they went so far that they called Taal “a small baby volcano”. It is kind of refreshing compared to the Daily Mail that calls every volcano a “supervolcano doom eruption”, but it is dangerously wrong and diminishes the possibility for a future mass-evacuation.

It is though equally wrong, Taal is after all a massive caldera volcano that have had confirmed VEI-7 eruptions, and numerous other caldera-extending events. There is a reason that Taal ended up on the New Decade Volcano Program.

The second thing is that some people have stated that the eruption is over. Yes, the volcanic activity on the surface has dropped significantly at the surface with steam and intermittent explosions creating columns reaching up to 1500 metres in height. But compared to the beginning there is definitely less surface activity. But as we will see below the activity has instead increased below surface.


The signals from the deep

On the left image one can see WSW/ENE compression caused by uplift, and on the right a similar trending uplift ridge caused be the intruding dyke. The widespread uplift is probably caused by an evolving sill system. Image by Sentinal-1 Geohazards.

INSar indicates WSW/ENE compression, but in reality, it is caused by magmatic central uplift pulling the rock in this direction towards the middle. INSar indicates that this is extending all the way to the ocean coastline.

At the same time there is according to INSar data a band of uplift ranging also in WSW/ENE. This is indicative of a dyke extending through the caldera trending in this direction.

Image along the Dyke, it seems to uniformly at 30km depth and upwards trending into the reservoir. Image by Andrej Flis.

GPS data corroborates this general trend, but interestingly the uplift pattern is broader than INSar gives at hand, in other words the entire shebang is uplifting, but there is a marked concentration along the dyke-line.

Water depth measurements in the lake shows us that the general water level is decreasing on all measured points around the lake. So much so that the river that normally is the outlet for the lake has dried up.
Parts of the dried-up river can be explained by uplift caused by the dyke, but it does not explain where the water has gone.

Probable magma movement route from the dyke. Centerpoint is the erupting vent on Volcano Island. Image by Andrej Flis.

During peak eruption the SO2 gas release was roughly 6500 metric tons per day, it then dropped to around 500 metric tons, before it rose sharply to the current 4500 metric tons per day.

Finally, we come to the seismic activity. It has been fairly constant since onset of eruption, but in the last couple of days it has increased markedly. The current count for today was more than 700 earthquakes in total and 26 of those was of volcanic type.

The volcanic type earthquakes ranged from M2.2 up to M3.8 during the last day. There have also been a number of tornillo type tremor events indicating increasing pressure in the magma reservoir.


Interpreting the signals
Collectively from INSar, GPS and seismic data we can infer that a dyke has formed transecting the caldera and that the central magma reservoir is being fed by this dyke, the general broader uplift indicates that horizontal pancake like sills have formed (see image for a visual explanation of pancake sills).

Very clear pancake sill. Sills typically cause widespread uniform uplift, and lateral dykes produce local uplift along the direction of the dyke. Photograph from Engineering Geology.

Due to the expanding dyke numerous faults and fissures has formed causing widespread destruction of houses and infrastructure.

From the seismic data we can surmise two things. One is that a ringfault has either started to form or, has reactivated from a dormant state. The ringfault runs partly along the caldera rim and partly outside of the caldera wall, indicating a complex magma reservoir topology.

There is a lot of crack in this image of nightlife around Taal. Photograph from Philstar.

Secondly it is evident that magma is influxing at a steady rate, larger than what erupted during peak eruption and that this influx has not stopped. Even though surface activity has decreased.

The cause for the drop in the caldera-lake water level is more uncertain, but it is likely that the water has infiltrated via fissures and faults into the active vents at Volcano Lake and was turned into steam. Another explanation might be that the water has disappeared due to evaporation, but I have so far not seen any data that the lake temperature has increased.


Possible consequences

Earthquake plot indicating the formation of a ringfault formation, low resolution due to low data quality. Image by Andrej Flis.

DOST-PHIVOLCS has stated that evacuation of up towards 1 million people might be necessary, but the result of the ongoing smaller evacuation has been so-so, indicating that a larger eruption might cause a high toll.
They have also warned that a larger explosive eruption might be hours to days away. The data analysed above is amply corroborating their assessment.

So far, the eruption ranges from a large VEI-3 to a small VEI-4, as such this is not uncommon for a Taal eruption.
Question is more about what is in the near future for Taal. I see three different options.

Option one is that the intrusion will decrease rapidly and that the current eruption ends in a few days. The intruded magma will cause further eruptions over the next couple of decades as the magma surfaces due to heat driven buoyancy. In other words, that Taal still is able to take care of business in the known historical way.

The second option is that the partially blocked pathway to the main vent will reopen, or that new vents will form, and that a second more vigorous eruptive phase will follow. Judging from the rate of intrusion and the size of the magma reservoir this second phase could be likely to reach VEI-4 or VEI-5 status.

The third option is that the intrusion of fresh hot magma continues without a corresponding eruptive outlet putting further strain on the various faults and fissures under the lake, or further invigorating the ringfault.

This could in turn lead to a larger eruption causing either a central caldera collapse, or a caldera extending event, partially fuelled by the lake water.

Depending on how much of the magma inside the reservoir will be reheated and made eruptible depends on how much hot fresh magma intrudes into it, the more that ends up above the eruptible limit, the worse the eruption would become.
If the entirety of the known magma reservoir would become eruptible (500km3) we would typically see a mid-range VEI-6 if it is at the centre of the caldera, or a large VEI-6 to a small VEI-7 if a ringfault caldera extension event occurs.

Option 3 is so far not likely; it is just a possibility further down the line. In my initial article about the eruption I believed that we would see option one as the by far most likely thing to happen. Now that more data has arrived, I am starting to think that option two is more likely, let us say that Taal currently hovers in between option 1 and option 2.

Also, remember that the level of fissuring and fracturing indicates that Taal might not be able to withstand the pressure involved in an option 3 scenario since something would break prior to it happening.


A word on volcano tourism to Taal

Do you feel more lucky than this horse? Photograph by EPA.

As everyone knows I love to watch volcanoes erupting. I am aware about the dangers involved, and I have the knowledge to make judgement calls on what is reasonably safe to do. Most people do not have that knowledge, instead they rely on so called “volcano-guides” to keep them safe.

In almost all cases the people working as guides do not know enough to keep you safe during an eruption, they are just out for your buck.

So, let me instead state this. There is not enough money on the planet to convince me to go inside the Taal caldera. This is based on the signals that I see coming out of the volcano.


191 thoughts on “Taal eruption update

  1. Wouldn’t be too happy about visiting Manila right now either

  2. Do you mean e-w compression or extension? Expansion of a dike makes me think of extension (locations on opposite sides of the dike would increase in distance from each other)

      • Sotiris Valkaniotis
        ‏ @SotisValkan
        Jan 18

        Correct, red (negative) values are towards west. Was trying an horizontal scalebar setting (that is more informative) and prob. left the text objects as they were..

    • Thanks for asking that, the first graphic had me scratching my head …

  3. Thank you Carl.

    I know it’s not the done thing to parp one’s trumpet, but I (for once) feel I got it right in my comment about dyke and sill formation a couple of days ago. Not based on experience, more just a gut feeling. But your scenario 2 matches how I felt when I wrote that.

    Unfortunately I believe scenario 2 could be quite ugly in its own messy way and I hope people are going to be able to get out swiftly.

    We’ll see. An interesting update and thank you!

  4. So, essentially the longer we go without any surface activity the bigger the bang? How long would it take for the intrusion to stress the faults and trigger and a VEI 6+ eruption?

    • This is only true assuming constant magmatic input. Without that, hard to say.

  5. Would you rather a. Jump around on top of Hekla, or b. jump around inside Taal caldera? ;-p

    Thanks as ever Carl. And ever grateful you didn’t stay away too long!

  6. Option 2 sounds eerily similar to the historical accounts of the 6-month long eruptive period in 1754 which included numerous VEI 3s and at least one 4 over that extended period.

  7. I would say a repeat of the 1754 eruption would be most likely unless the activity of last week resumes.

  8. Option 3 would be as a result of a caldera system failure and would likely require a sequence of unfortunate events maybe the result of a relatively large earthquake along an underlying fault, triggering a cascade of events, a larger version of MT St Helens. Perhaps .more option 2 going wrong and turning into 3. This situation is probably unlikely and impossible to predict based on observation, it will either go big or it won’t and as such if it did occur it would be a “black swan”. Hopefully this event won’t be it but it will happen somewhere either in this lifetime or some future lifetime and the world will change.

    • My guess is these numerous large caldera systems act as a relief valve of the earths crustal recycling system, through millennia they take this thermal energy and safely dissipate it in the shallow crust, but occasionally this goes wrong and the safety valve blows and as such it is a failure of design.

      • “Safely?” Sort of depends on where you are standing….

        • Remember most of these Caldera inflate and deflate again without even a murmur of and eruption in hundreds if not thousands of years, quite a few have extensive populations near them or in them and if you take a person’s lifetime, within that timeframe 99.99% safe, safer than flying.

          • That sounds like the number for Yellowstone (a million years between eruptions). Naples and Taupo are rather more active that that!

  9. Here’s a look at the uplift over a longer period

    Dorella Papadopoulou
    ‏ @dorella_pap

    #Sentinel1 LOS displacement time series (Dec 2018 – Jan 2020) using CNR-IREA P-SBAS service on @esa_gep The blue color over the #TaalVolcano shows uplift, that according to the time series, must have initiated in March-April 2019.

    And for a reference point on the island

  10. Forgive me if this is too simplistic.
    If the magma chamber is pressurised then some of that would be countered by the mass of water in the lake. If the water level is dropping say due to boiling away won’t this reduce the counter force on the chamber leading to some uplift? Or are the forces magnitudes apart?
    Secondly Carl states ‘it is likely that the water has infiltrated via fissures and faults into the active vents at Volcano Lake and was turned into steam.’ would the cooling effect of the water be enough to explain the lack of surface activity at the vent?
    Way outside my own discipline here, grateful for some education.

    • Wouldn’t the lake be steaming? the pressure of the water is lower by a considerable margin than the pressure from the crust below , so that water is not going to drain anywhere unless it finds away out the rim of the caldera, but it will heat up gradually and in the shallows may start to steam, a lot like the glacier bound complex in Iceland that forms cauldron in the ice from geothermal activity. If it starts having phreatic eruptions thru the lake then it could be getting a bit dodgy.

      • Phillipines is littered with caves large and small. Maybe as simple as a fissure cracked open one of them near the lake and some of the water drained out.

  11. https://www.philstar.com/headlines/2020/01/21/1986499/phivolcs-taal-recharging-powerful-eruption

    ““The island has slightly tilted while being pushed. Some portions in the northeastern side are sinking. Maybe the houses there are already under water,” Solidum said.

    Phivolcs volcano monitoring and eruption prediction division chief Mariton Bornas said the ground around the volcano has swollen except in the northeastern part. Swelling of a volcano signals an accumulation of magma near the surface.:

  12. A very interesting article indeed.
    I would like your work to be widely seen and please forgive my pedantry but, unless there’s evidence that the Daily Mail calls every volcano a “supervolcano doom eruption”, I don’t think quotation marks should be used.

    • I do. The mantra is popular in pretty much any “news” article.

      Proper use of quotes. 😀

    • The Mirror rather than the Mail, but you get the gist. Of course, Yellowstone is in no danger at all. It is unlikely to erupt in the next 10,000 years, possibly the next million years. That will not stop the UK tabloids. The Daily Mail is not the worst, admittedly. It has been several years since they last predicted a Yellowstone catastrophe. Who cares about science or truth when papers need selling?


      • One thing I could never understand, Albert. Does the Mail and Mirror have an audience in the UK who really cares if the northwest corner of Wyoming were to blow up that is large enough to sell copy? You would think that Campi would be their better StuporVolcano to stoke fears about – er, report on, as it is arguably closer and impacts a lot of popular holiday spots for UK locals, and thus would sell more papers. Certainly more so than a prairie of an ex-colony some 4,500+ miles away.

        What gives?

        • Interesting, isn’t it. No one ever really believes that predictions of catastrophe apply to them personally. If it is too big to deal with, the solution is denial. The climate crisis is an example: if it is presented as unsolvably bad, people will ignore it. But for Yellowstone, the US is far enough away that it seems a safe catastrophe, so it can be read without requiring denial as a coping mechanism.

          But it is weird that it now becomes an entertainment item. It is about getting a proper scare, like watching a horror movie, after which they turn off again and go back to their real life.

          Of course we volcano watchers would never look forward to a proper VEI 7..

          • A VEI 7 will have global repercussions, like the 2004 boxing day tsunami and the fukushima tsunami but much worse , loss of life during and post event could be millions, climate change, crop loss etc, disruption to airline travel, bankruptcy, even revolution and the fall of government and an upheaval of the current global order.They should be careful what they wish for!

          • You mean the you sane non misanthropic volcano watchers would never look forward to a proper VEI 7. Of course I am not sane and I am a misanthrope so I’ve got chips ready and a party planned for when it happens,

          • “Tallis Rockwell” isn’t an alias for “Nicolas Cage”, would it?

          • And no volcano in Texas. There is however a big one just across the border in New Mexico. Ideal for sorting your volcano-related issues..

          • You can hope for better than that. The Socorro caldera, ready while you wait.

          • Just don’t let the “Daily” tabloids find out… “Massive extinction-level eruption possible from “newly” discovered caldera in New Mexico!” I can see it now…

  13. https://www.philstar.com/headlines/2020/01/21/1986499/phivolcs-taal-recharging-powerful-eruption

    Phivolcs: Taal recharging for ‘powerful eruption’

    Volcano landscape altered

    MANILA, Philippines — Taal Volcano is swelling, a portion has sunk and the entire volcano island has tilted slightly, indicating a “resupply” of magma or molten rock rising to the surface that presages a powerful eruption amid the ongoing lull.

    This was according to the Philippine Institute of Volcanology and Seismology, whose official said Taal appeared to be “recharging.” Phivolcs has maintained Alert Level 4 meaning a destructive eruption is imminent within days or even hours.

    “There is resupply of magma from below based on earthquakes and ground deformation,” Science Undersecretary and Phivolcs officer-in-charge Renato Solidum Jr. said.

    Solidum also said the silent activity beneath Taal has significantly altered the landscape of the volcano island.

    He said some parts of the northeastern portion of the volcano have sunk while the entire volcano has swelled.

    “The island has slightly tilted while being pushed. Some portions in the northeastern side are sinking. Maybe the houses there are already under water,” Solidum said.

    Phivolcs volcano monitoring and eruption prediction division chief Mariton Bornas said the ground around the volcano has swollen except in the northeastern part. Swelling of a volcano signals an accumulation of magma near the surface.

    • i am a bit surprised/dissapointed PHIVOLCS hasn’t realized deformation stopped around the 15th… I guess there goes the second volcano-monitoring organization I no longer have much faith on.

      • Really? I think you might be in a rift zone mentality , you ain’t in Iceland anymore, this is the big boys club “club subduction” where we deal in highly evolved magma emplacements , you know nasty stuff like dacite and rhyolite. So i think the phillipine authorities are well aware of the situation , but not a lot they can do this is so unpredictable, they can say the sky is falling and this thing can easily just roll over and go back to sleep for decade it could go big in hours with not much warning, no win situation.

        • Although Taal does a range of magmas, it is mainly a basaltic volcano. It does up to dacite but not rhyolite.

          • The reference wasn’t just to taal but other subduction zone volcanoes that do have rhyolite , obviously the prehistory caldera forming eruptions involved something more evolved than basalt and so that potential is there, I feel some are painting this as a Icelandic type event and the location of this complex means it probably has a moreviolent potential than that.

          • “Early ignimbrites are predominantly calc-alkaline dacite,
            whereas more recent lava flows and tephra deposits are mostly basalts and basaltic andesites with tholeiitic afinity (Listanco, 1994; Knittel and
            Oles, 1995).”

        • I am in a rift zone mentality because this is a rift zone, that the Macolod Corridor is a rift is something geologists agree on and since you seem to be familiar with the Taupo Volcanic Zone you should be aware that rifting and subduction are not mutually exclusive but in fact often occuring at the same location.

          Reading about volcanoes and particularly Hawaii has made me come across tens of dike intrusions, I think I know well how to recognize them and how they work and evolve, so that it why almost ten days into the eruption seeing that PHIVOLS still hasn’t even used the word makes me question them being well aware of the situation.

          • I am going to defend PHIVOLCS. They have given correct warnings, although the speed of development a week ago surprised them, they had bene raising e warning level before. I agree that a dike or shape sill is likely. But that doesn’t help the warnings very much. Recent eruptions have always been on or near Volcano Island. There are no explosion scars elsewhere (ok, there is the old crater wall including Mount Maculod). The insar images are very nice but these are not easy to make; it takes a few days after the data becomes available, and it takes specialist knowledge and a dedicated computer. I would also warn that the images may not be full reliable in areas of significant ash cover. There seems to be some interpolation in the patterns.

            Volcanoes are not easily predictable. Taal could now go to sleep for 6 months and suddenly wake up. How do you deal with that s a volcano observatory? You can’t evacuate people for 6 months and you can’t not evacuate them. Can’t win.

            I also note that even in Hawaii, disaster was only narrowly averted when Kapoho was overrun. The evacuation was uncomfortably close. Even HVO could have been caught out.

          • So now you are comparing Hawaii to this, and PHIVOLS has no idea because they know what they are dealing with isnt Hawaii.
            This deformation could quite easily be due to expansion of evolved magma, this dyke idea because of the fluidity of the basalt will tend to make itself known quickly, the seismic activity is not forming a line it is more clustered like a balloon is being inflated beneath the lake.

          • And I do agree with the decisions they have taken until now and also that they sadly don’t have much data to work with. They are in a difficult situation with a difficult volcano but still I cannot agree with the interpretation they are giving.

            “I would also warn that the images may not be full reliable in areas of significant ash cover”. Yes, that is why I have tried coupling that with visual observations such as the cracking, changes in the lake shore and historic background.

            Taal is not like Hawaii, but many things I learned with Hawaii later have shown to be useful in other situations. And dike intrusions have some rules that do not change from one volcano to another. A dike is not really so weird, there are three general shapes of intrusions sills, pipes and dikes, and sills are rarely or never directly feeding an eruption so you are just left with the other 2 when it comes to an eruptive event.

          • A conduit is a vertical dike.. And a sill is horizontal, or at least tends to follow the contours of the surface, so it can’t easily erupt. Inflation is often due to a sill though.

          • Taals basalts are also much much more viscous than Iceland and Hawaiis

  14. Thanks for updating all of us with your data and experience!

    I don’t really understand why there is so much unknown about the size of “the” magma chamber or how “full” it is in such an ongoing situation?

    They were found by seismic wave messurements (slower/faster speed in the different media) and here we have now an event with 700 earthquakes and we do not really know if the input of fresh magma is in a steady state, pulsing or stopping – is it somehow like a singer on the stage, who doesn’t here his own voice? Would we need waves from a source fare away? Or just too many quakes, which results in too much and chaotic data?

    The company which paid my education, it’s a global cement producer, the chemistry of synthetic lava (~1400 °C) was my business in the lab, had the slogan: “A faster learning organization”, somehow I like to mention that…

  15. But if “The island has slightly tilted while being pushed. Some portions in the northeastern side are sinking. Maybe the houses there are already under water,” then there would be several points of inflation under and around the big lake?

    • The northeast coast of the island is sinking and the northeast coast of Taal Lake also doing so (which is what the interfegrogram shows), it means that a least the northeast half of the caldera is deflating.

      Intrusions of magma do not come out of air, and this one seems quite big, you need to drain a magma body, here being the storgage system of Taal.

      • Is there anywhere a Picture from the possiblle flooding at the lake???? Do you know something or anybody here??

        • Here is a link to a thread on @raffytima ‘s twitter with photographs of a property being flooded by the lake.


          Seems the water is rising on the northeastern (?) side of the lake. If the whole lake bed is tilting that would explain the change in water level seen on the south side of Lake Taal and the drying up of Pansipit River, which is clearly no longer the lowest point of the lake shore.

          • That is in Agoncillo which is located in the southwest coast of Lake Taal and it is very localized because in San Nicolas which is just 2-3 km away the water has receded (uplift). It seems that a small area above the dike has dropped into a graben (which is what would be expected) and this includes Agoncillo.

            This is the same pattern of deformation reported in 1911, when Agoncillo also flooded and the valley of Pansipit subsided.

            It is also to consider that the average level of the lake has likely also changed with all the deformation that has taken place.

          • This image shows the edge of the graben in Agoncillo:


          • Okay now you got me scouring google earth….I think the place in the photo is here:
            Yellow Coco – Batangas
            Agoncillo, Batangas, Philippines
            13.934920, 120.947354
            And what you say makes sense to me.

            I can’t find any info on lake levels on the northern shoreline, in that case.

  16. i would not count on using that horse to ride out of trouble. i hope the people are receiving the correct communication but i doubt it…..How fast could they leave and where would they go? and how fast could they get there. And experience tells me there would be many other people on the roads bringing the exodus out to a standstill. Lesson’s learned….. leave before You need to or better yet; like Lurk says…”Don’t be there in the first place.” Best!from motsfo watching DH shovel off the deck to the birdfeeder again. and it’s still falling. Not even going to try and shovel the driveway out to the road.Road not snowplowed anyway. Don’t need to go anywhere.

  17. Meanwhile Volcán de Fuego – rewind the stream if you’ve missed the action but live now

  18. TV news mentioned that plans are afoot to permanently re-settle a lot of people further away.

    Snag is you’ll just have to blink and there’ll be squatters, then shanties, then settlers in the empty ‘unsafe’ zone…

    I feel sorry for the guy who’d grazed his herd of cattle on the island, now faces a total loss…

    • Moving settlements has been the game plan for over 100+ years of dealing with Taal’s activity. Seems like an uphill battle. (literally) This is a country where a secondary concern of geophysicists is having their sensor gear wandering off. As noted. The problem will re-manifest itself as time goes on.

  19. I have a question about the origin of the lava seen erupting on the volcano island. Where is its source? Is this old magma re-mobilised from a previous eruption?

    • Always a possibility…. but there is no way to know for sure unless detailed geochemistry is performed.

  20. Amzing video showing the changes to Volcano Island after the eruption:

    • The south shore doesn’t show much of a change on the lake level but the eastern and northern shores of the island have underwent subsidence, particularly the northernmost point where many houses and trees are well below the water. They don’t fly over the west coast..

  21. Hi. I hope somebody can clarify. There’s been a bit of conflict with regard to the safety issue regarding the effects of possible explosion. An area near the volcano has a high location. Because of the height, some in the government say it will not experience any immediate danger (e.g. base surge, tsunami and ballistic projectile) and so, there’s no need for evacuation even if it is located within the 14km radius of the main crater. Others, perhaps out of caution, are pushing for mandatory evacuation.
    Cannot understand a lot of things so please kindly explain layman’s terms in terms of possible risk to the people living near but supposedly protected by the ridge. Thank you.

    • I am wondering which government this would be. Do they have scientific advisors or a volcano observatory?

      Some dangers lessen with height above the surroundings. Still, you would take an enormous risk guessing how high a pyroclastic flow will go. CO2 is a silent danger and that mainly affects low lying areas. Ash fall goes everywhere and it is also very unhealthy. Ballistic projectiles will go many kilometers – not even Everest would be safe (well, apart from being outside the 14-km zone). Even airplanes could get hit by flying rocks. No, if the volcano observatory recommends a 14-km evacuation zone, ignoring it because of some imagined immunity is not wise. There have not been major casualties from a volcanic eruption for a long time. That is in part because the most affected regions (Philippines, Indonesia) have volcano observatories which are on top of things and have proven their competence.

      • 14km radius assumes a modest eruption, but remember eruptions don’t have even effects out 14km, one could be ok at 10km and in severe danger at 30km depending on ash faĺ direction or pyroclastic flow direction, tsunami could have hazardous effects at quite some distance. A large VEI 6 eruption would need evacuation for a lot more distance than that. 14km is just a play it safe distance without uprooting to many people perhaps unnecessary if no larger event occurs, seems more political than anything else.

    • Well, CGN refers to an area called “Tagaytay” which sits right on the northern part of the outer caldera rim. Distance from the volcano 13-15km. Height 600m. So safety distance from a pyroclastic current would be ~555-300m (taal lake 5m above sea level, taal volcano island peak 300m). Today on twitter videos had been published that show ground cracks/fissures within the Tagaytay area. Based on the footage it looks like parts of the slope that goes down from the northern caldera to the lake is developing cracks or fissures.

      • Pyroclastic flow from the 186AD Taupo eruption was 3000 ft, tall has substantial igmbrite deposits, so can have voluminous eruptions.

      • I wonder if that could be to do with the east-west expansion Carl mentioned? Rather than the vertical. Hmm.
        Feels like it must be one heck of an intrusion.

  22. Carl,

    To your point about how dangerous Taal is at the moment. Have you seen this video yet? From a group that was hiking down into the volcano on the 12th. It’s a miracle that they made it out in time.


  23. Thinking on Carl’s scenario 2, I suppose so long as none of the cracks in the area of uplift start steaming, they remain purely dry ground movement (the Kilauea dyke comes to mind!). The conduit has ‘cleared it’s throat’ and is open and ready. Therefore I trust any new eruption will remain in the island in Taal Lake.

    I admit I had a horrible vision of a fracture at the caldera edge becoming an additional volcanic outlet in the south west. But so far as I can see there is no sign of any ‘modern’ eruptive history on the edge of the caldera to the south west apart from the old Bilibinwang Bay area to the west of the central island.

    Apart from that the nearest is the cone at Binintiang Munti on the south west of the Island, and externally the old stratovolcano, Mount Batualao way the the west. There are more ‘modern’ edifices on the eastern half of the caldera.

    I’d be interested if the experts agree that the upcoming eruption(s) will remain in the central island.

    How long it will be before the resurgent magma dyke system arrives at the volcano island, I suppose is anyone’s guess. I assume seismically we’ll get quite a bit of notice from the experts at Phivolcs!

    • Just a casual observation here, to me it seems like the original eruption has become just one actor in a large play , this actor may still have a central role in the production, but may not play as large a role in the final act as one may assume being lost in the overall grandeur of the resulting dramatic act.

  24. Interesting Specialist Remark on the IMO eq homepage:

    “At 7:55AM an M2.8 earthquake was detected in southern caldera of Katla volcano. No aftershocks have been measured. On 10th of January an M3.0 earthquake occurred in Katla, it was located further north in the caldera compared to the earthquake this morning. The earthquake this morning is located by one of the cauldrons in the caldera. The cauldrons are formed because of geothermal heat beneath the ice-cap, but no changes are seen in water-level or conductivity measurements in rivers around Mýrdalsjökull ice-cap. The earthquake this morning is the fourteenth earthquake larger than M2.0 in Katla caldera in the last 12 months.
    Written by a specialist at 21 Jan 11:10 GMT”

    Doesn’t seem to be anything significant on the tremor guages that I can see. I haven’t been watching Katla closely apart from its eq’s for a while, so perhaps IMO already has cause for extra vigilance. Seemed to have gone quiet after all the shallow activity a couple of years ago and a few of minor hlaups. Could the witch be stirring her cauldron?

    • I have a question for people who are watching the volcanoes of Iceland, in terms of deformation where is Katla? Is it inflating or not? Or at what rate? I remember seeing that 1 or 2 years ago Katla had one of the highest volcanic CO2 emissions of the planet which should mean a very high deep supply. But I have no idea right how to check GPS of Iceland.

      • Katla is indeed a monster of a basaltic volcano
        Capable of 18 km3 and likley larger basalt floods
        And did enromous pyroclastic events in early holocene

        Katla also haves an very large sufur content in its magmas
        Another sign of a prolific supply.
        Katla is one of the most beastly of basaltic volcanoes.

        Grimsvötn is even more monsterious, but its doing more frequent smaller eruptions
        But may have an even larger capacity when it comes to flood basalts

    • Is there a page from IMO (vedur.is) about katla with cumulative seismic moment, number of EQs and so on? Does anyone have a link?

        • I have been looking for this for months. This is exactly what I was looking for. Thanks Bjarki.

        • And thanks from me, too, Bjarki. (I should have inferred from the similar pages for Hekla and Barda that I have bookmarked.) It is amazing what IMO make readily and publicly available, so thanks to them aswell.

  25. Yes, I think she could be, but Helka has been acting strangely too.
    Has katla and Helka ever erupted at the same time, before?

    • From some studies I read the hydrothermal system is huge and is what causes the large column driven eruptions since the magma is mainly basaltic.

    • Lake is 234 km2 , so let’s assume intrusion is 20% of lake area, that is say 50km2 and say average depth of say 5km, so 250 km3 . But not an expert and just guessing but the size of the area being deformed well do the math?

      • But why 5 km!! not even magma chambers are that thick. And we have a dyke here…

        By looking at the amount of subsidence I think the volume of intrusion will be between 1 km³ and 0.01 km³ and that way I am relatively sure it will include the final value, but if you want something more specific in my opinion it may end up around 0.1 km³

        • Well i guess you have provided the answer 0.01 km, quite modest, a fraction of the magma erupted during the 2014 Iceland event, that is a powerful effect on 30 km of the earthy surface, as I said I was just doing a crude calculation on potential volume of material that could be erupted.

    • A second way I can think of estimating it is by assuming how much space has been created in the rift since the last intrusion in 1911. By considering the dyke would be filling a distance of 12 km, assuming a vertical dimension of 5-10 km and also assuming an opening rate of 3 cm/yr (based on how fast the Manila Trench does rollback) it would turn out at 0.2-0.4 km³. It has more assumptions than the other way I estimated it but at least gives a more specific number and also falls within the reasonable range I gave earlier.

    • I apologise but now all your posts speak to me In. Mr cage’s voice….

  26. Whether for technical failure or other reasons, all PHIVOLCS public seismograph pages stopped updating at 04:42 UTC today. The website is still up and you can still pull up the charts for all their instruments but the data ends at 04:42. There are some on the map in Indonesia on the PHIVOLCS site and they are still updating in real time so the website isn’t broken.

    If they do come back online they can be found here at the official site https://swift1.phivolcs.dost.gov.ph/seismograph/

    • Closest was PV VTBM, located on the island itself, and was actually still working up until 05:15 UTC yesterday – when it either stopped or the feed did.

    • I notice some commenting elsewhere that they think the government ordered it. Hopefully not true.

      Others have commented that probably not true as, for example, neither the USA or say Iceland ever cut feeds, even at times when events were ongoing that might (at worst) have gone decidedly pear-shaped themselves.

      Anyway hopefully it is just a technical issue.

        • “Why does this squiggly line keep going on past two squares”

          “should it be going up and down like that”

          “Why does that thing there also show up on”

          Just some comments in chat on streams


          • I love VolcanoYT’s live streams but the peanut gallery is sometimes really hilarious.

            The other day they were describing Taal as a bowl of really spicy soup.

  27. Just a reminder for people here as a way to understand the environment and nature of Taal. Taal is a large, very dangerous, and very prolific volcano for the following reasons:

    1. It sits in a highly active subduction region. This is an obvious requirement for most arc volcanoes to form, but is not the singular reason which makes it more dangerous than the “normal” stratovolcano.

    2. Taal sits in an active spreading zone. This is the predominant item that contributes to large caldera volcanism when combined with point 1. Remember – rifting means that rock can melt at a lower temperature. This makes it easier for the incoming magma from arc-subduction to melt the surrounding bedrock, and simply amplifies the overall magmatic output. With exception to Yellowstone, most volcanoes you could classify as “supervolcanoes” or quasi supervolcanoes exist in similar conditions.

    3. Taal has another potentially potent amplifier in that it has a large lake around it. Most of us here know that water can greatly increase explosivity.

    Now… that doesn’t mean things will necessarily be all bad or that this will positively be an enormous eruption (or even an eruption at all beyond the current). The above conditions create magma supply, but this doesn’t speak to pressure relief. If pressure gets relieved here due to the potentially semi-open nature of the volcano (as it has many times before), we likely end up with another eruption that’s perhaps a bit larger, with a lot of the more fresh basalt getting erupted. This is basically Carl’s scenario #2.

    Scenario 3 of a large explosive eruption (vei5+) may be a little bit different than people may be thinking here however. The mechanics of this aren’t simply just fresh magma pushing into the magma chamber, and reaching a threshold where the pressure exceeds the roof, and whatever magma went into the chamber leaves. The risk of a large explosive eruption will come more from whether the new basalt coming into the volcano right now can melt and rejuvenate any more viscous crystallized magma.


    – Most magma chambers only have a fraction of their magma sitting within an eruptible state.
    – Magma that is not eruptible (crystallized) keeps its gas contained internally. The gas content in uneruptible magma does not exert pressure on the caldera roof until it gets re-energized.
    – The pressure levels of this rock influence at what temperature the gases within this magma can be released.

    So the question becomes twofold:

    Will the basalt injection stay in the magma chamber long enough to rejuvenate crystallized viscous magma, and will that basalt injection be large enough to destabilize the equilibrium of the caldera roof?

    And secondly, if there IS an eruption, will that eruption damage the structure of the caldera lid enough to decrease the pressure of the magma below, causing a positive feedback loop to be put in place that feeds more explosivity.

  28. Hvo talk by Tina
    What we learned. what we think is happening, what we think will happening on the big island. There are more planned talks by other HVO scientists.

    • Very informative… Thanks … and she was doing so well up to the catch in her voice at the end when she was overcome with emotion about her relationship with the community… They have been thu so much together. Quite the volcanic history there. Quite the experience; even from a distance. Best!motsfo

    • You need to consider your rule “not to be there when it happens” 😁

    • I’m not. NW Florida for me. Our worst invasive species are tourists.

      • Hey that us Canadians… at worst you have a invasion of overly polite people 🙂

      • Problem is that just short of Canadia, you have some of the more surely people on the planet.

    • Thats right,,, i had forgotten about the hazard of falling iguanas. Hart Hat time! Best!motsfo

  29. New sentinel radar image. Now waiting for someone to turn it into a new inflation/deflation map.

    And here is a comparison of the southwest shore, Jan 9 vs Jan 21. You can see that the lake has lowered a bit

      • This is actually from a different (not public) satellite. But it will do.. The red line shows my best guess on where the intrusion is happening, based on the pattern of movements.

        • I think that would be the most easily interpretable chart I have seen so far for an interferogram in whole event, my thanks to the GSI.

          Those red lines within the zoomed area seem to correspond to the faults that have been activated by the dyke and opened so many fissures in the first days after the eruption. The fringes are not very well resolved within the faulted area but from the few that are identifiable it looks like it is moving away which is confirmed by visual observations of the Agoncillo area subsiding below the lake, it seems that the valley of Pansipit has dropped like a graben. This process seems to have taken place between the 12th and the 15th according to a visual observation of the lake level remaining stable after the 15th, also because the large earthquakes declined around that time.

          • Concur. The region between the red faults has dropped, but also has moved south-ish. On the northwest side of the fault, the area has come up and has moved away from the fault.

  30. So in the latest “TAAL VOLCANO ADVISORY: 22 JANUARY 2020 04:00 PM” are some key things: “Since 5:00 AM this morning, there were no ash emissions based on our seismic records and visual observations.” “For the past 24 hours, the Taal Volcano Network, which can record small earthquakes undetectable by the PSN, recorded four hundred eighty-one (481) volcanic earthquakes including eight (8) low-frequency earthquakes. Such intense activity likely signifies continuous magmatic intrusion beneath the Taal edifice, which may lead to further eruptive activity.”
    So, no visible things happening but ongoing activity. I hope the throat is not blocked and Taal is heading for a big hickup…

  31. One of the larger quakes for a while and the PHIVOLCS seismo network is back online.

    Date/Time : 22 Jan 2020 – 11:16:10 PM (15:16:10 UTC)
    Location : 14.11°N, 120.97°E – 001 km S 18° E of Tagaytay City (Cavite)
    Depth of Focus (Km) : 004
    Origin : VOLCANIC
    Magnitude : Ms 3.4

    • Checked it. That is exactly in the outer caldera rim. Where it goes down from Tagaytay to the lake. Probably near where they posted pictures of fissures on twitter yesterday. But sooo low that quake, wow…

  32. Big jump in the Taal thermal anomaly on last processed pass.

  33. Cascades Subduction Zone—What can the landscape tell us?

    IRIS Earthquake Science

    This animation, narrated by Nick Zentner, Central Washingon University, describes the geographic provinces of the Pacific Northwest, including the subducting plate, the subduction boundary, the Coast Range, the lowlands, and the Cascades mountain range.
    Written by Jenda Johnson, Earth Sciences Animated

    • Nice! However, one kibitz… they left off Mt. Lassen on the southern end of the Cascade Volcanoes… my favorite one!

      • ink they left out Meager Mountain at the northern (Canadfian) end as well

        • Cayley and Crater Lake too, to complete the list of the major CVs. Though not sure that Jefferson should really be included- no eruption on the mountain in over 15,000 years.

    • Sotiris Valkaniotis
      ‏ @SotisValkan
      45 minutes ago

      #Sentinel1 interferogram for Jan 15 – Jan 21 over #Taal volcano area show localized deformation & surface displacement along a SW-NE axis [main ones marked with a dashed line based on phase ifg]. Processed at @esa_gep

  34. https://news.abs-cbn.com/news/01/22/20/phivolcs-magma-wont-flow-from-batangas-fissures

    Phivolcs: Magma won’t flow from Batangas fissures

    MANILA – Amid posts about the possibility of magma flowing from fissures in Batangas towns due to volcanic quakes, the country’s chief seismologist on Wednesday assured the public that it was not the case.

    Philippine Institute of Volcanology and Seismology (Phivolcs) executive director Renato Solidum said molten rock under Batangas will flow towards Taal volcano, amid circulating social media posts saying the magma underneath Lemery, Agoncillo and San Nicolas were about to flow through ground fractures.

    “Wag kayong maniwala diyan,” he said.

    Solidum explained magma has always been under the surface of towns that surround Taal volcano, which has “magma chambers.”

    • This kind of explains what some seem to be missing that there is a lot of magma already in place before this current intrusion. Is this current intrusion mobilizing existing “cool” -more evolved magma in the system causing the island to tilt, sub surface dome building? Are there old extrusion type edifices at taal is the volcano island one of these, built up by viscous lava extrusion?

  35. Christina H Koch
    ‏Verified account @Astro_Christina
    3 hours ago

    Keeping watch on the volatile Taal Volcano in the Philippines from the @Space_Station. Just south of Manila, it stands out ash-covered and otherworldly in the middle of Taal Lake against the surrounding green highlands.

    • just north of the caldera lake, There is a wildfire going

  36. Are those ancient calderas to the south west? The two U shaped parts of the shoreline leading into the sea.

      • I would say yes. The lake to the north east seems to be a caldera as well. Wikipedia page:


        “The lake is shaped like a stylized ‘W’, with two peninsulas jutting out from the northern shore. Between these peninsulas, the middle lobe fills the large volcanic Laguna Caldera”

        • How much of the land in that area is igneous rock and how much of it is calcium-carbonate based? I forget the word for this, is it Karst? It is common in Vientam, China, Yucatan, and highly prone to caves and cave-ins.

        • As mentioned before, it reminds me of the Vesuvius (= current Taal eruption) and the Campi Flegrei (the “U shaped”).

    • There is some belief that the smaller “u” there to the south of Taal is a remnant submerged caldera that is now more or less extinct.

    • Thankfully, that footage does not appear to be from inside the crater as at the start it pans out to view lake Taal.

      Using the fishing pens/shoreline at the start as landmarks, i believe it has been shot on the path that leads from the north shore to the crater rim. About a third of the way along the path from the shore, there is a deep gully and i believe it is looking down in to this. What is rather interesting is the small crater/vent it shows does not appear on google earth so that is probably new. This will tie in really well with the extension line plotted on the Sentinel1 interferogram posted above by Squonk, and the excellent photo posted below by Dustdevil (NB – i’d say that the photo is looking almost due east rather than NE as stated). The fissure in the photo depicted by the dashed yellow line looks to extend beyond the crater and runs roughly along the gully/path to the north coast, with new vents along it.

  37. Nice “worst case scenario” video from Philippine volcanologist, risk assessment and evacuation zone seems to be based on 1754 eruption.

    • Just a comment from a Black Swan perspective.

      This is ASSUMING that 1754 is the worst case scenario. Much like the design criteria for the Fukishima seawall, which also ASSUMED that the largest recorded pacific basin tsunami in South America was the worst case scenario.

      To do a better estimation of the risk eruption size, a histogram would need to be constructed of every Taal eruption, but that is technically infeasible since detailed analysis of every eruption would need to be known and analyzed. Given that, using 1754 is a very good benchmark, but keep in mind that it has limitations. And… that volcanoes “hate” statistics and do pretty much anything that the prevailing physical processes dictate they do at that moment in time. PHILVOCS is very good at their job and 1754 is actually a really good baseline to assume worst case conditions. After all, volcanoes tend to stay within their historical eruptive patterns.

      • I also think that the label “worst case” is not really a good choice here. The caldera did come into existence somehow, and Taal is in the 2018 Newhall/Self/Robock paper. The real worst case thus would have to be a VEI7 eruption, I think. Not saying that this is at all likely or even possible now, that I obviously do not know. But I think “benchmark” would indeed be a much better term.

        • It seems that to get VEI7+ eruptions felsic magma is required (rhyolite, dacite, trachyte…). Taal (which erupts basalt and basalt-andesite) currently not possesing those magma compositions puts a limit on how large an explosive eruption can get. Basalt eruptions with water around can easily get to VEI4 and sometimes to VEI5, a VEI6 might also be feasible but I do not know of any really confirmed examples.

          • As said, for sure I am not nearly expert enough to weigh in on this. But just taking at face value that there is — it seems — no real consensus among experts (just compare your posting to Carl Rehnberg’s three scenarios — without saying that any one of you will be “right” or “wrong”).

            In my understanding, a “worst case” includes unlikely but not physically impossible events. Whereas a realistic scenario of course will include what is deemed likely.

          • There’s too much emphasis on the “E” in VEI in my opinion.

            By only including the potentially explosive scenarios, the VEI discounts the potentially even larger atmospheric impacts of relatively non-explosive basaltic eruptions.

            There should be something like a VEVI “Volcanic Eruption Volume Index” that would be more inclusive and representative of the impact potential.

            Taal’s VEI potential may not be technically explosive enough to qualify as a major but this discounts the substantial VEVI potential.

          • Some areas of the caldera have volcanic rock with silica as high as 68%. Perhaps all these historic eruptions are part of the same process of magma emplacement which eventually could lead to a much larger event?

          • True, there are couple of situations like lahars and jökulhlaups, lateral blasts and collapses, Taal-like base surges, Nyiragongoan style lava flows, voluminous effusive eruptions, sudden steam explosions… that pose their own hazards not very well covered by the explosivity index.

            Dominik, I guess that the physical limits are not very well established yet because the physics of volcanic processes still have a long way to go. For example the collapse events and system re-pressurization was mostly discovered during the collapse of Kilauea in 2018 but when comparing to caldera collapses in other volcanoes like Bardarbunga or Mijakejima it has been shown to be something fundamental and that always seems to take place. The first serious article on this was just published last December…

          • “Some areas of the caldera have volcanic rock with silica as high as 68%. Perhaps all these historic eruptions are part of the same process of magma emplacement which eventually could lead to a much larger event?”

            The early prehistoric ignimbrites were dacitic, currently Taal does not show to store silicic magmas anymore and it takes some considerable dormancy to produce it…

          • Dust Devil, do not get me wrong. I was actually quite impressed with how well you analysed the situation even before the precise interferogram analyses came out. I was sort of happy with understanding that the dyke intrusion now may end quite unspectacularly. It sort of fits what limited amount of understanding I believe to have.

            I was rather surprised with Carl’s much more cautious analysis. I still personally do not think we will see something very large here and now. My “criticism” here is really only the wording regarding “worst case”.

          • Maybe dacite doesn’t show because it is being stored and these eruptions of other types of magma are just blips in the storage process?

      • Worst case scenario in these situations is only what will be encountered in a lifetime of population or facility, so a hydroelectric dam might be designed to easily cope with a 100 yr flood or even a 500 year flood, but what happens if the 5000 yr flood happens. With the volcano they seem to be working on a 500 yr event which is allowing a good degree of leeway, but of course a Caldera forming event is possibly a 10000 year event, so likely to happen once in 150 lifetimes or less than 1 % chance in the rest of this century, no way of planning for that type of scenarios and living a sensible existence.

  38. Sulfur emission has gone down to 153 tons a day, could Taal have a plug? If I am not mistaken the deformation hasn’t ended.

  39. View of the Main Crater:

    The eruption seems dyke-fed and the fissure which goes up the north wall of the crater probably marks the path the intrusion followed to erupt in the north flank of the island.

    Also a video of Agoncillo in the southwest shore of the lake and effects of the subsidence:


    • I cannot any cooled dark lava flow
      Are you soure they got the information correct?

      • I does look more like an erosional structure than a flow, I hadn’t noticed that.

  40. https://www.gmanetwork.com/news/news/nation/723274/big-taal-volcano-eruption-30-possible-phivolcs/story/

    Big Taal Volcano eruption 30% possible —PHIVOLCS /b>

    There is still a 30-percent probability of a huge Taal Volcano eruption, according to the Philippine Institute of Volcanology and Seismology (PHIVOLCS) on Thursday.

    PHIVOLCS officer-in-charge Undersecretary Renato Solidum explained during a press conference that the experts have arrived at the said figure after various scientific considerations were factored in.

    “We look at the monitoring parameters that we have and binibigyan namin ng weight ito. We have experts that provide their opinion on what the data would be showing to us. Mayroon din tayong reference with the historical data that we have and the deposits that some of us have studied… kino-combine namin ‘yan,” Solidum said

    … Now that the magma has already moved up but has not been ejected yet, the PHIVOLCS is looking at many possible scenarios.

    …Solidum said that if the magma will not be ejected, it can solidify and turn into rocks. Yet, if a new supply of magma rises, it can push the old ones upwards too.

    From reading other reports, Solidum also said the reference for a huge eruption they were using was 1754.

  41. These graphs represent the past 2,000 earthquakes–not a complete set–recorded in the Taal area. The area is ~160 miles square +-5 miles. The program allows me to select an area but does not tell me how big the area is. The Excel file shows them going back to 2011. 84% of them are >= VEI 3. These earthquakes go down to over 300 km.




Comments are closed.