41 comments on “Antarctica 3 – The Volcanoes of Marie Byrd Land

  1. Another fantastic article Agimarc! I’m flabbergasted at the extent of volcanism in Antarctica. These are really big edifices! And geologically speaking active very recently. I now have a new favourite volcano: Takahe! The takahe is an almost extinct flightless bird in New Zealand that my Dad and a few other rangers were given the task of finding in Fiordland (he was unsuccessful – they didn’t find it till much later) But look at the shape of it! And so many of these volcanos sport big calderas. Amazing.


    • South Island Takahe or North Island Takahe? Just teasing, I read that the northern type is extinct. What a colorful, strange bird!


  2. Another great post, thanks agimarc!

    A new report today for G. Agung:
    Some points:

    – Real Time Seismic Amplitude (RSAM) in the last 12 days has not increased but remains at a high level.
    – Earthquake activity is still dominated by high frequency events. This indicates that seismic activity at the volcano still represents the brittle failure of rock inside the volcano in response to magmatic intrusion.
    – Changes in seismic velocity indicate that pressurization under the volcano continues as the intruded volume of magma into the system increases and as magma moves towards the surface.

    – Visible emissions from the crater consist of 50-200m STEAM plumes (or better: water vapor, not smoke).
    – Water expulsion in the crater near the solfatara field has been observed by satellite. Water expulsion is thought to reflect a disturbance to the hydrologic system in response to intruded magma at depth.
    – Tiltmeter observations showed a sudden deflation on October 1, 2017, but following days until today showed continued inflation.
    – MultiGas measurements at Mt. Batur do not indicate the supply of new magma under that volcano. As of today, all indications suggest that all recent activity is related to Mt. Agung.


      • boy, that first one looks tantalising!! unfortunately it is paywalled. I remember Colin Wilson writing something about rapid rejuvenation of crustal plutons behind the Oruanui eruption. What intrigues me is that the crystals at depth are so cold and only heat up in the shallow storage. that turns my understanding of things completely on its head!


            • cool! just read the article. If I understand it correctly, their evidence shows that relatively cool bodies of magma (ie. below the solidus) located in a wider regime of sills and nearby chambers can be remobilised quickly upon injection of sufficient heat (basalt intrusion for example).. this makes sense to me.. an intrusion occurs and raises the proportion of melt in a crystal mush and if no eruption occurs, it cools down relatively quickly. Rinse and repeat, until one day a large enough intrusion comes along to mobilise the entire regime and it rises to shallow storage from where it gets erupted.


  3. I woke up this morning wondering whether the current intrusion under Agung is primarily transport of juvenile material from deep storage to the shallow chamber.. I guess this could also be a mechanism for a much larger eruption, i.e. if there is a large body of magma in the shallow chamber that is currently in a non-eruptible state but very close to tipping into an eruptible state then the eruptive potential of an intrusion would effectively be levered by a couple of orders of magnitude (this is the mechanism we talked about relating to Taupo back in the day at VC). Given the numerous recent eruptions I still think it is a safer bet to view Agung as an open system that has not had time to develop such potential (yet). The report from MAGMA also talks of decreasing seismic velocity in the system as it is infiltrated with magma… I would love to know what depths this LVZ is and whether they have actually identified the position and size of any shallow storage.


    • The main deep magma source was found at 18-22 km depth. Shallow level magma reservoirs (note the plural) exist at depths between 3 and 7 km (after the 1963 eruption).

      I also don’t think Agung could produce another VEI 4 or 5 because of the short repose time. Given that the eruptible magma in any chamber is usually less than 30% there shouldn’t be that much available. OTOH, in a short repose time, there may still be magma left hot, that would make for a greater amount of it being readily eruptible. In that case, it would not only be fresh basalt that comes up but also some of the leftover andesite from the last eruption.


  4. That ties in with paper on Zircon crystals, where a cool body can be reactivated at comparatively short notice so what was not eruptible before the intrusion is tipped into an eruptible state during it.

    Lordy, talk about fickle systems. I am glad it is not my job to make the calls on the response planning. It’s like planning for an incoming wave when you are at the beach, but it is midnight and you are blindfolded and you are not sure if the wave is going to make a sound or not as it approaches because it might not have broken yet.


  5. Wow, what a challenge! Geologists in Norway are pouring some 40 tsd. liters of water down the streams at Mannen mountain since midnight, to help loosen the rocks. Mannen has been unstable and threatening residents and roads for years. Choosing a particularly rainy period, they hope that the landslide will come down now. Waiting… here too. Watch the live webcam here:



  6. Interesting in view of the present situation at G. Agung: an investigation of the duration of eq swarms before volc. eruptions, based on statistics derived from the Global Volcanic Earthquake Swarm Database (GVESD) and geological considerations. (Benoit/McNutt, INGV, 1996)

    For example, the ascent of magma to the surface may express itself in longer lasting swarms, while intrusions or failed eruptions are manifested by shorter swarms. … swarm duration as a function of repose time … etc.

    Swarm durations found for VEI 4 eruptions with 24-85 years repose time was 24 – 90 days.

    Global volcanic earthquake swarm database and preliminary analysis of volcanic earthquake swarm duration (PDF can be downloaded)
    DOI: 10.4401/ag-3963

    I wonder if there are newer works based on this db?


  7. Wow Graniya, you’re fantastic at digging up this stuff! Very interesting article although it is now about 20 years old.. There are a couple of things I took from it: 1. mean swarm duration before an eruption is 8 days (which is surprisingly short suggesting either rapid ascent or that most eruptions occur from shallow chambers). 2. Silicic systems tend to have longer swarms than mafic systems (this might simply be to do with the rheology). 3. There is a strong correlation between swarm duration and repose time BUT ONLY for swarms that last four months or more. 4. As far as I could tell there is no correlation between swarm duration and eruption intensity. I would love to see if there is any correlation between swarm intensity and eruption intensity!

    Things to think about: Agung has been very active with 52 tephra forming eruptions over the last 5000 years, about 1 every 100 years. You would think the conduit would be pretty soft/broken with that sort of activity. Repose time since the last eruption is only 50 years. So why then such a vigorous swarm? Maybe cooling and solidification in the conduit occurs faster than thought? Maybe it is a sign that a lot of magma is on the move.. i dunno. I guess this is a pretty strong intrusion and suggests to me a fairly long magma column is on the move with VT events happening from 20 km depth to close to the surface (like at Fimmvorduhalsi). But if so I have no idea where the column has got to – it may only be from deep storage to shallow storage (I take your note about there being multiple shallow storage chambers). However, magma may also be rising from shallow storage towards the surface but I still don’t think this has happened yet as we don’t see any LP events in the seismic trace indicating movement of fluids or gases. However, we might not see HT until the magma gets within 600 to 800 m of the surface, i.e. just before the eruption starts, like at El Hierro.
    Alternative scenarios therefore range from the intrusion coming to a halt with shallow storage being rejuvenated and replenished after the 1964 eruption but no eruption happening (making the next eruption potentially bigger) through to the swarm continuing and an eruption occurring in the coming days/weeks. in other words it looks to me that just about everything is still possible. PVMBG say that the eruption is more likely than not and I imagine that assessment hasn’t changed. I wonder what caused the deflation a couple of days back? magma migrating “elsewhere”? – whatever that means!


  8. Image posted on Twitter this morning of Agung with an up to 1500m high plume:

    Image by @jannisdz, cropped.


    • I wonder what the composition of that is and of the water, how much is ground water and how much water is exsolved from the magma (AFAIK there is no way to make that distinction but it would be a good indication of how shallow the magma is). Most is probably CO2 and SO2 picking up ground water on the way. (Speaking out of my derrière here but I think that’s how it works). At White Island I was impressed how a tiny fumarole could generate such a large plume. There was also do much SO2 you had to wear a mask.


      • Yeah… this is the funny thing I have been poring about for a couple of days. It was said in the report three days ago that SO2 emissions have NOT been detected. But the gas could still be there, dissolved in the hydrothermal system.

        So, I understand they did not detect concentrations high enough to measure the amount of SO2. But even if it is dissolved in water, it should come out with the water vapor? If there is a lot of SO2 in the steam (which it should if the magma is quite near the surface) it should be measurable in the degassing area? Is SO2 easily broken up in S and O2 by heat? Might have to read up on this.


  9. If I only could figure out their various earthquakes: How are deep and shallow tectonic quakes different from Local tectonic quakes? Anyone know? And, if the magma is well up already, what happened to Hybrid and LP quakes? Are they not there, not mentioned, or mentioned under a different name?


    • Hi Granyia, I don’t think there is any difference that we could see in the seismic signal between a purely tectonic quake and a volcanic tectonic event, as both are created by brittle fracture. However, I imagine it should be technically possible to see some distinction if you got into the moment tensors (beachballs) as normal tectonic events involve slip and VT generally don’t. I guess the seismic trace of a VT event is more akin to an underground explosion (albeit a small one!) as cracks form. However, we don’t have that detailed information and I wouldn’t even know what to do with it if I did!


  10. From my observation on the ground over the last couple of weeks, I would make an educated guess that there are at least 5000 people well within the most dangerous zones at any given time during the day. That includes communities who refuse to evacuate, and people who come up during the day to look after their cows or search for fodder for their livestock. If the volcano erupts during the day the death toll could be a lot higher than 1963.
    From “News from under the Volcano – part 6”


  11. Gosh, this has been a volcanically active month so far – Agung keeps everyone on their toes, Ambae is still making eleven thousand people unhappy, and this morning, after a short period of raised seismicity, Shinmoedake (of Kirishima) made a comeback after its last show in 2011. The eruption started at 05:34 LT today and was said to be small, w/ ash emissions rising ~300m above the crater. Still ongoing though.
    This image is from Shinmoedake webcam at MBC, a few min ago:



Your thoughts?

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s