Agung volcano nearing an eruption? Here is what to expect from the Bali volcano.

Agung, a volcano on the popular tourist destination of the Indonesian island of Bali has recently been upgraded to the highest alert level (4 of 4 to the “red” level) possible on the Indonesian volcanic alert scale. As a result, the highly experienced and competent Indonesia volcano authorities have started evacuating locals who live near the volcano while establishing a 9 kilometer exclusion radius. This exclusion radius may expand over time as conditions change for the famous volcano.

For Official Information, Please Refer to the Indonesian Volcano Authorities, Who are Some of the Best In the World at Their Job. Visit their Website (use Google translate to read the language) at .

EDIT: As of 9/30/2017, a live seismogram has been added at the URL: 

There is also a youtube webcam link here: 

I also suggest reading this interview (use Google translate) with Dr. Surono, who is one of the leading experts on Indonesian volcanoes, and the former head of the Indonesian Center for Volcanology and Geological Hazard Mitigation (PVMBG).

Progression of seismicity at the Agung Volcano. Image from the Ministry of Energy and Mineral Resources at, through the  Center for Volcanology and Geological Hazard Mitigation (PVMBG) .

Will Agung Erupt?

This is the ever-present question volcanologists always ask themselves. As I  have always mentioned, there are no sure things in volcanology, and conditions are always subject to change. What we do know is that there is elevated signals of risk that signal a high possibility of an eruption here. As of the most recent reports I’ve found, Agung has been experiencing heightened and increasing earthquakes, and more importantly, the start of harmonic tremor.

Heightened Earthquake Activity

The volcano tectonic earthquakes do not seem to be of a particularly high energy with the most energetic quakes being in the range of m 3.0 to m 4.0 on the richter scale. For perspective, Bardarbunga, the large volcano in Iceland which erupted in 2013 had many earthquakes as large as 5.5+ on the richter scale prior to it’s effusive eruption. Mt. St. Helens similarly had some very powerful earthquakes before it’s signature 1980 blast.

Earthquake power does not necessarily signal potential power or energy of a future eruption

While there may be some correlation between earthquake energy prior to an eruption and the subsequent scale of an eruption, some volcanoes can have a lighter load of earthquakes before they erupt.

On average, more energetic earthquakes can signal that the magma conduits are more brittle or hardened. This means that for an eruption to occur, new and rising magma would need to fracture that rock to expand upwards, which results in more energetic and powerful earthquakes. Clearly, additional magma pushing up from below would equate to an increased eruptive capacity, but the strength of the overlying magma conduits also plays a significant role here.

For Agung, it may be a more “open” system as it last erupted in 1963, which for a volcano, is not all that long. As a result, the conduits and magmatic system may not require tons of energy for fresh magma to open up the dikes as it pushes towards the surface.

Since Agung has not yet erupted, some of the strongest earthquakes may still be yet to come, although it could just as easily fizzle out and stall if the magma pushing from below does not have the requisite power to push through the solidified roof.

The above video is a post-eruption video of the Indonesian volcano of Merapi, which had a vei-4 eruption in 2010. Merapi is similar in size, and also is located near highly populated regions of Indonesia, although it does not have quite as explosive of a history as Agung has.

Harmonic Tremor at Agung Signals Degassing and High Possibility of an Eruption

More important than the currently occurring volcano-tectonic earthquakes is the presence of what is known as Harmonic tremor. Harmonic tremor is an acoustic pattern that is found to occur in volcanic systems when magma is moving through a closed chamber, or when magma is degassing. Indonesian volcanologists have noticed an increase in the harmonic tremor at Agung, which is what prompted them to raise the alert level to red.

It’s important that degassing (which is signaled by the harmonic tremor readings)  indicates that fresh magma has pushed its way up towards the surface to a place where depressurization starts to occur. In short, when magma travels upwards through the crust, the pressure tends to decrease. As this pressure decreases, gases (such as water, co2, and sulfur dioxide) that are previously dissolved within the magma start to bubble, expand, and get released from the magma. This results in a potential runaway effect, and is also why certain volcanoes tend to erupt in a more explosive fashion than others (such as Hawaii). Volcanoes like Agung which seem to have a lot of water dissolved in their magma tend to be extra explosive as water is an extremely explosive gas when it flashes to steam.

Exercising Caution in Regards to Predictions

While the harmonic tremor may indicate a high likelihood of an eruption, there still is a chance nothing will happen. Every year, many volcanoes exhibit similar activity with no eruptions as there just is not enough energy coming through the volcanic system to break the overlying rock. As an example of this, the volcano Cerro Negro on the Colombia / Ecuador border experienced over 130,000 earthquakes in the year 2014 with an earthquake as large as 5.6 on the richter scale with no eruption. But for a more active and open volcanic system, there may not be a similar amount of energy required to start an eruption.

What Will Happen if an Eruption Starts?

I will state this clearly, Agung is a very dangerous volcano. The 1963 eruption, a VEI-5 eruption that occurred here was one of the larger eruptions in the past century, and it killed over 1000 people despite the fact that the eruption did not occur as a surprise to locals.

Agung’s problems start with the basic fact that there is a high population living within a close proximity to the volcano. Within just 30 kilometers of the volcano, there is nearly one million people according to the Global Volcanism Program’s estimations. The city Kubu is a tourist destination city that sits within 15 kilometers of the volcano. Its population is approximately 53,000 permanent residents.  A simple look at the map shows that past eruptive activity has sent lahars and other volcanic products on paths in the general vicinity of this city all the way to the northern coast.

I highlighted a path north of the volcano that shows where past lahars and volcanic products have traveled. There is a clear path in the earth that is still visible from the 1963 eruption. A new eruption could result in similar effects.

While there is no guarantee that ashflows, lahars, and other eruptive activity would follow the same path, it can be reasonably assumed that there would be a strong likelihood for this to occur if an eruption of a similar size to the 1963 eruption happens here.

A more proper mitigation map can be seen below, which shows the propensity for Agung to affect even larger cities that are further away.

Threat map from Agung. Red indicates risk from lava flows and ash. Yellow indicates lahar risk.

Agung Has a Relatively Unknown, yet Explosive History

Agung is a volcano that seems to have had a very explosive history. The 1963 eruption is very well known as a large eruption, being that it was a vei-5 blast in historical times. Prior to this eruption however, the volcano’s history is far less known. Like most Indonesian volcanoes, it is likely that Agung experienced many smaller             eruptions between the larger blasts. The GVP also has listed  but  another VEI-5 blast in just 1843, but there is very little information on this eruption. Regardless, two large eruptions in such a short period of time indicates a volcano that is primed with a very large magma storage capacity that is capable of producing big eruptions without needing a long time-period to recharge itself.

Quick Information and Background on Agung

  • Agung has multiple magma chambers, with one being a deep chamber that sits around 18-22 kilometers of depth, and a much more shallow magma chamber that occupies the depth range between 3 and 7 kilometers deep.
  • Insar satellite data which has recorded somewhat recent inflation suggests that the primary upper magma chamber sits around 2-4 kilometers.
  • The deep magma chamber sits close to the boundary between the mantle and the upper crust, which is often referred to as the MOHO, and the upper chamber is presumed to sit in a region which occupies a boundary between oceanic-like crust and an upper sedimentary layer.
  • Like most other volcanoes in this region, the magma composition is predominantly basaltic-andesitic.

Read more in depth scientific information about Agung at 

Agung’s Explosive Neighbors

While not all volcanoes are similar, and even proximal volcanoes can behave wildly differently, looking at nearby volcanoes that share similar traits can be a good way to gauge the potential of a volcanic system. For Agung, things can get a little bit scary. On the island of Bali, it’s closest neighbur Batur has had multiple large caldera forming blasts, each of which were likely of a vei-7 scale based on caldera size.

The other main volcano, Bratan, has a similarly explosive history. Things get even more interesting when you look at adjacent islands. the two Indonesian islands directly to the east of Bali are the homes of the volcanoes Rinjani/Samalas, and Mt Tambora. Both of these volcanoes produced VEI-7 eruptions within the past 1000 years.

While neighboring volcanoes are not predictive of a singular volcano’s future, this indicates that the volcanoes of this region are extremely active, and have a long history of highly explosive activity. 

In between the big blasts, it is quite likely that there were also many many smaller eruptions, so we shouldn’t predict that with every tremor or eruption that something catastrophic will happen.  There are two reported small eruptions of a small VEI-2 size in the early 1800’s, but there likely have been many more small eruptions over the course of the past 10,000+ years. With that said, it’s important to be aware of the risks and potential of a volcano like Agung, which shares many traits such as magma composition, size, and crustal formation with its highly explosive neighbors.

How the 1963 Agung Eruption Progressed

Agung’s VEI-5 eruption in 1963 did not progress in a manner that many would otherwise expect if they were to know that the volcano experienced a large eruption. As the eruption started in February of 1963, it started out without a major blast as it emitted a lava flow that slowly pushed down the upper slopes. It wasn’t until nearly a month later in March that a major eruption ocurred. A second explosive eruption occurred later on in May of that year.

Looking back on the occurrence and sequence of events, it seems that the primary explosive eruption occurred after old degassed magma stuck in the upper conduits was slowly pushed out. Once that magma was clear, the primary eruptive event was clear to begin as the primed and explosive gas-rich magma was free to make it to the surface.

There is a decent chance we could see similar behavior in the event of a new eruption. Of course, as a caveat, we need to keep in mind that volcanoes often erupt in very different styles, changing how they erupt with each subsequent eruption.

Summing Things Up

Agung is an explosive volcano that has a short history of very strong eruptions on a highly populated island. Prior to the 1800’s, we have very little information to discern what has occurred at Agung, so our ability to discern all the possibilities here is a little bit limited.

Speculating on Agung Eruption Possibilities

  • No eruption occurs as the magma pushing from below can’t produce enough energy to reach the surface.
  • An eruption occurs, but it only produces a small eruptive event or lava flow.
  • An eruption occurs, starting with throat clearing and potentially pushing out a new lava flow. Once the de-gassed magma is cleared from the upper conduits, the gas-rich magma starts to erupt in a much more explosive manner, creating an eruption in the range of VEI-4 to VEI-5.This eruption results in damaging and potentially deadly lahars, bothersome ashfall in areas, and a lot of destruction of property and crops. A few may be caught in pyroclastic flows if they have ignored the evacuations warnings and exclusion zones.
  • Not super likely, but an eruption larger than VEI-5 potentially occurs. At this point, all bets are off as to what would or could happen here.

Overall, I will not try to predict an eruption as that would be silly. But I encourage all to listen to authorities, and lean heavily on the Indonesian authorities for up-to-date information about Mt. Agung and any potential eruption here.

13 thoughts on “Agung volcano nearing an eruption? Here is what to expect from the Bali volcano.”

  1. Possibility #3 should say VEI 3 to 5.
    Agung ’63 got VEI 5 due to large total material erupted, which is a lot but spread in 11 months time. In away, it’s like a VEI 3 at energy level but went on for quite a while that enough material produced to put it in VEI 5.
    Mt. St. Hellens (1980) for example, is a VEI 5 within hours. Much bigger explosion energy, total amount material produced was similar to Agung’63 but in much shorter time.

  2. Hi cbus, I’m responding to your thoughts about the conditions for a caldera forming eruption:
    Assuming the caldera forms from explosive evacuation of the chamber (as opposed to subsidence after effusive evacuation of the chamber as in a shield volcano) I think a volcano needs to check a couple of boxes first (all of the figures below are very approximate rules of thumb)

    1. the chamber width relative to depth needs to be at least roughly 1:1
    So at Agung, presuming the shallow storage is 5 km deep, the chamber will need to be 5 km in diameter or more.
    2. To get sufficient evacuation of the chamber it will need to be preceded by a voluminous eruption before collapse would be initiated.
    3. To get such a large eruption it will need to have a sufficiently primed chamber. Primed here mainly means being saturated with gas.
    4. The saturation of a magma depends on the type of magma and the depth. According to Parfitt and Wilson, the solubility laws for H2O are:
    n = 0.1078 P^0.7 for basalt ranging through to
    n = 0.4111 P^0.5 for rhyolite
    (in other words rhyolite can contain a lot more water before it becomes saturated – one reason why rhyolite eruptions can be very explosive)

    My take on the above is that the large calderas in the Lesser Sunda Arc have resulted from a history of frequent basaltic/andesitic eruptions which have established a feeder system and shallow chambers that keep getting replenished with primitive magmas. Over time the eruptions become less frequent and the gas that would otherwise be erupted is absorbed in increasingly evolved magmas in shallow storage until these become saturated and a large eruption occurs.This implies a longer repose time before a big eruption than we see at Agung. Admittedly this is a very simplistic interpretation and I haven’t checked the composition of the Batur/Tambora/Rinjani eruptions to see if they all fit with this interpretation (will do when I get some time!) so I could be wildly wrong on this. But if I am right I don’t think we are going to see a really big eruption at Agung just yet.

    1. Just realised that I didn’t explain why I think the chamber is not primed:
      The main reason in my view is that we don’t see any HT or LP events in the seismogram which are more reliable signs of exsolution of gas and gas-driven movement of magma (remember the paper on “magma wagging”?) If these signals were to be seen, particularly at depth (which we can’t determine but I imagine PVMBG can), then I think you could take a safe bet that the chamber was full of saturated magma. But at Agung we only see VT events (admittedly a lot) which are attributed to brittle failure in advance of a rising magma column (i.e. buoyancy driven). When this column gets shallow enough for bubbles to form, the event will probably tip into eruption mode. It’s not there yet though.

      1. .. and of course I should add we don’t have anything like the information at our disposal as the experts so maybe there are signs of a pending bigger eruption that we just don’t see (I say this just in case anyone else is reading this – always trust the local authorities more than any amateur hack like me on the internet!)

        1. Thanks bruce, very valid points. I had read initially that there was harmonic tremor, but that clearly is not the case as more information has been released.

          It’s interesting that this seems to be a very significant intrusion event, yet hasn’t triggered that type of tremor at all at this point. One wonders how long this can occur until something may actually get fully set off.

          1. Agreed. The trouble is having so little experience in reading these things. Fimmvorduhalsi had a couple of long and sustained swarms prior to the eruption and when the eruption did occur it was not all that voluminous in the wider scale of things. OTOH, the length and vigor of an intrusion is not necessarily any indication of the size of the ensuing eruption – if one even occurs!
            Imagine if we were able to follow the earthquake swarm prior to the 1886 Tarawera eruption.. it was primitive magma so it probably started from the mantle/crust boundary and rose towards what is commonly known as one of the more prolific rhyolitic centers in the world that had erupted a fairly voluminous lava dome 300 years earlier.
            The last thing I would have expected in that scenario would have been a plinian basaltic eruption!
            The same thing applies here. What is the intrusion doing? Is the magma influx getting accommodated in shallow storage? (should be visible in inflation). If so, will it stay there? Or is it a magma column with enough buoyancy to make it to the surface (via shallow reservoirs or perhaps even bypassing them??) 2. How evolved is the magma in the shallow storage? And how much gas does the shallow storage contain? There are so many variables here it is impossible to make any firm guess of what is going to happen.

Leave a Reply