Nothing Fancy → Ulawun


Long time no sea. 😀

Actually, I haven’t gone anywhere and I apologize for the sailor joke. What I really mean is that I haven’t done much other than lurk in and around the comments section. My last actual post was a rehash of an ancient topic that I had covered some time ago (done by request). Recently, Ulawun “blowed up real good” and mentioned in the comment chain. Turtlebirdman and I had a bit of back and forth about the plume height and the needed mass ejection rate needed to reach specific VEI levels. Since I was quite tired while discussing it, I will go ahead and concede defeat due to the fact that I have an obvious error in measuring the mass in some ultra-hot peppers I am growing. In the back channel, it was noted that I could eliminate all questions about the apparent mass gain (Pot+soil+water) that happened overnight if I just incinerate the whole pepper and measure the remaining carbon. Technically, this is doable, I ran across an article in SciAm {Amateur Scientist series} years ago about how to construct a “kiln” using an old metal thermos bottle, however my weight scales don’t have the precision to get super accurate measurements and that would leave me with no plant. (The intent of the series was geared towards determining the organic content of a sample of soil.) Enough about that.

One of my pet peeves is the term “Supervolcano.” This is nothing but a marketing word manufactured to drum up interest in a program. The term has no actual basis in science. Scientists use the term out of deference to their audience when talking about it since it has become quite common. My prefered term is “Large Caldera.” It is just as artificial, but at least you can determine a definition for it (sort of). “Large” being anything in the upper half of the size population of all calderas. Whether you choose median, mode or average as your lower cutoff point is really the only issue. Ulawun is by no means in the “Large Caldera” family. The issue is that it is surrounded by Large Calderas on the same island, meaning that it has a similar geologic setting to it’s cousins. To the northeast is Rabaul, and to the west is Dakataua. I don’t have an immediate reference, but if I remember correctly Rabaul has a handful of additional large calderas in it’s general vicinity. Does this mean Ulawun could potentially do something similar, well, yes, but also remember that there are just as many old non “Large Caldera” volcanoes on New Britain as there are spooky ones. In fact, likely more.

Not meaning to “beat a dead horse” but the issue with Turtlebirdman and my discussion was what VEI Ulawun was going to be. Darwin Volcanic Ash reporting center does not retain archived reports very well in the public domain, but over at the Volcano Discovery site, you can find them if you rummage around.

As many of you know, I lean heavily on Mastin et al for estimating mass ejection rates. As Turtlebirdman noted previously, Ulawun can and does generate a flowing magmatic component. That is not considered in my estimate. Due to Wellington and Darwin advisory centers both issuing VAAC reports, my spreadsheet skews my result towards the high end. Also consider that a VAAC report is naturally going to err towards over reporting since their main concern is safety of flight. By no means should you consider my results as authoritative.

Plagiarizing Mastin et al directly,

…within this error envelope the eruption rate varies from about 20,000,000 to 400,000,000 kg/s — a factor of four in either direction. The scatter in the data limits the accuracy with which one might constrain eruption rate given an observed plume height. (my bold)”

About Mastin et al. The design purpose of that paper, as I understand it, is to afford a way to get an empirical measure on eruptions for which little to no field data is available. For our purposes, we can “sneak a peek” at what the VEI may turn out to be. It’s not going to be highly accurate, but it will put us in the ball park with a sane estimate. In the past, I have found that it is a reasonable estimate and came quite close to Carl’s isopatch ash-fall methods on a previous eruption. Even getting close to my estimate of Kelud based on an alternate “milling formula” method of calculating energy release. Even getting close to Albert’s estimate based on the acoustics of that same eruption. I was quite quite pleased and more that a bit surprised that it got that close, and all 3 of our methods came reasonably close to what the actual volcanologists came up with in their final reports. Yes, “reasonably close” is just hand-waving, but for a bunch of aficionados operating outside of our normal skill sets, we did pretty well. {And you’ll find that my methods are bit unorthodox anyway. Blame the USN for that. In my specialty, you make it work even if it doesn’t want to cooperate.} (I had two separate incidents where I wound up doing “depot” level repair though we did not have “depot” level tools and equipment.) Even had an NSWC Tech come out to check my work and we had it within factory specs. Caveat: Had we known beforehand that the maintained we were doing was categorized as “depot level” only, we would have aborted our efforts and put it on the work list for the next availability. Do note that the re-coring of our liquid to liquid heat exchanger was done by the USS Shenandoah, a depot level certified facility. It turned out removing and re-installing it was also supposed to be relegated to specialists but we did it anyway. Nothing a semi-competent plumber couldn’t take care of. And you would be surprised of how many normal people have hands-on experience fighting with the plumbing under a kitchen sink.

Note that as of this moment, the eruption has not been declared over. This is just plotted to the point where the VAAC reports stopped noting an ongoing ash-cloud over the edifice.

Personally, I don’t think it is going to do anything unexpected. Note the deviation in the ejected mass accumulation rate. The first portion is a classic “throat clearing” followed by the main event where it really got energetic. The peak rate was on the order of 6958 m³/s. This is a fairly high rate and likely illustrative of why Ulawun has large caldera neighbors. There is also an unnamed (on Google Earth) 10km caldera only 34 km to the southwest of Ulawun. { 5.337122°S – 151.136495°E }. If you can find a name for it, please add it to the comment chain. Think of this as your riddle for the week. 😀 (No points for correct or incorrect, I honestly don’t know it’s name.) Your neighboring visitors will determine the accuracy.

On behalf of VolcanoCafe I wish to extend our thanks to Volcano Discovery for maintaining an archive of the Darwin and Wellington VAAC reports.

For those of you discussing magma temperatures, from a paper discussing Quaternary volcanoes on the Talasea Peninsula. (where Dakataua is located);

“…indicate equilibration temperatures in the range 920° to 860° C”

A reminder. The melt generation zone for a subducting plate approximately coincides with the 110 km contour of the subducted plate. This area can be found along most subduction zones in alignment with the associated volcanoes (For example, the Cascade volcanoes). Another thing that tends to occur is thinning of the over riding plate. This promotes extension and the formation of a back arc basin as rifting comes into play. (Rift zone paralleling the Tonga trench roughly 100 km west of the island chain. The Taupo volcanic region is on the southern end of this in New Zealand. The Bay of Plenty is roughly where it comes ashore.) The plate configuration in and around Ulawun is a veritable train wreck as the Solomon plate is jostled among the other fragments and plates of the area. The Solomon plate has the feature of being subducted on 3 sides with a transform as it’s southeastern boundary. (blue – convergent red – transform green – divergent) This is a screen-cap from Google Earth using the USGS kmz file as an overlay. The over riding plate in this image that New Britain is on is the South Bismark Plate. According to the Wikipedia entry, the merge rate is on the order of 86 mm/yr. (African Plate frame of reference)


A multidisciplinary effort to assign realistic source parameters to models of volcanic ash-cloud transport and dispersion during eruptions  Mastin et al (2009) – Journal of Volcanology and Geothermal Research, Volume 186, Issues 1–2, 30 September 2009, Pages 10-21

The volcanoes and caldera of Talasea, New Britain: Mineralogy Lowder (1970) -Contributions to Mineralogy and Petrology, December 1970, Volume 26, Issue 4, pp 324–340

Volcano Discovery VAAC archive. Data retrieved 27 June 2019.

Additional information can be found in Extensional and vertical tectonics in the New Guinea islands: Implications for island arc evolution  David Lindley (2006) Annals of Geophysics.  Here he states; “High-angle structures predominate throughout the region and indicate major vertical movements of crustal blocks

And, melt generation in New Britain may not be as simple as just dehydration melt from a subducted slab.  Other processes may be at play.

Magma Genesis in the New Britain Island Arc: Further Insights into Melting and Mass Transfer Processes – Woodhead et al (1998) Journal of Petrology, Volume 39, Issue 9, September 1998, Pages 1641–1668,





262 thoughts on “Nothing Fancy → Ulawun

  1. earthquakes seem to be spattering across a wider area of S Ca. now….. interesting times we live in… 😉 i’m not liking earthquakes after our 7.1

    • and i wonder how long it will take others to find me on this new page…….. 🙂

      • Gonna be fun when the mellinium tower’s soil liquidfies. It’s built on what used to be part of the bay.

        • It is already tilting. And thus the problem of building new housing in SF.

  2. found this guy on twitter…. very close to the action… (language) Danielle Madugo

    • things which have just less than 10% probability still happens about one tenth of the time 🙂

    • Latest aftershock (foreshock?) forecast

      The USGS estimates the chance of more aftershocks as follows: Within the next 1 Week until 2019-07-13 07:40:00 (UTC):

      The chance of an earthquake of magnitude 3 or higher is > 99 %, and it is most likely that as few as 360 or as many as 660 such earthquakes may occur in the case that the sequence is re-invigorated by a larger aftershock.

      The chance of an earthquake of magnitude 5 or higher is > 99 %, and it is most likely that as few as 0 or as many as 11 such earthquakes may occur.

      The chance of an earthquake of magnitude 6 or higher is 39 %, and it is most likely that as few as 0 or as many as 3 such earthquakes may occur.

      The chance of an earthquake of magnitude 7 or higher is 5 %, such an earthquake is possible but with a low probability.


      Naval Air Weapons Station China Lake is responding to a 5.0 magnitude earthquake at 8:16 p.m., and a 7.1 magnitude earthquake at 8:19 p.m pacific time and multiple, consistent aftershocks on July 5 following a 6.4 magnitude earthquake and numerous aftershocks on July 4.

      NAWS China Lake is not mission capable until further notice; however, security protocols remain in effect.

      Safety of personnel is currently the highest priority.

      NON ESSENTIAL active duty, drilling reservists, civilian employees, and dependents are authorized to evacuate to a radius of 100 miles from safe haven Naval Base Ventura County (NVBC).

      NVBC and its surrounding area is the preferred location for authorized evacuation on the basis of Installation Support Services.

      RDML Bette Bolivar, Commander Navy Region Southwest, has issued this evacuation order.

      A Family Information Call Center number has been established for questions regarding how to obtain support. The Information Hotline number is 1-844-523-2025.

      NAWS China Lake access remains Mission Essential Personnel only. Do NOT attempt to access the Installation unless you are Mission Essential Personnel.

      Further information will be distributed as it develops.

      • Just as a general and off-topic comment, given that the annual US military budget is about 5x greater than any hypothetical “enemy of the US”, the fact that one US military base used for weapons development and testing isn’t “mission capable” won’t cause me any lack of sleep. As for the personnel, if they reckon quakes are bad, they enlisted for war…

        • FYI, military personnel have no control over where they ultimately get stationed. Sure, they can request a billet. But their detailer makes the decision. That’s how I wound up in the cesspool of Southern California for 5 years.

      • anyone see an opportunity for a great movie???? i vote for the Rock and Kevin Sussman. So heads will spin with alternate theories….

    • Ouch!

      If you find yourself in that unfortunate situation, should you run like crazy, or is it better to stay in place and face the explosion so you can keep your eye on and possibly dodge any larger oncoming lava bombs? Running down a steep slope is a good way to fall and hurt yourself and even if you’re an olympic sprinter you wouldn’t make it more than maybe a couple of hundred meters before ejecta starts falling around you. Don’t be there in the first place is of course the safest bet every time, as I’m sure Lurking will tell you.

      • Problem is that there was virtually no warning. Sure, Stromboli had been a bit more active than normal, but nothing out of the ordinary. It started acting up at under 2 minutes before the main explosion. So…don’t be there doesn’t really apply. Or, you just can close off Stromboli alltogether. I think running is the only option btw. Staying and seeking shelter might protect you against bombs and falling debris, but the gases will get you. Apparently that’s what happened to the man who died (who apparently was the second person you hear talking in this video, don’t know if that’s true or not, there’s a lot of speculation online). He was not killed by lava bombs or rocks, as was reported earlier, but likely suffocated due to being exposed to high levels of volcanic gases.

  3. I’ve been seeing reports about the 6.4 earthquake and aftershocks, then SF had the 7.1 quake – and some articles are calling it an aftershock – at what point does a subsequent quake become the ‘main shock’ and the original quake become a ‘foreshock’ – or how far apart do they need to be so they don’t count as ‘linked’ ?

    • sorry southern california not sf – oops (hope no alarm caused)

    • aftershocks can happen a long time afterward…. depends on the area and depth involved…. even months or years afterward… ask me how i know…. 😉 Best!motsfo

      • Well, as you asked, how do you know?

        I guess you’ll have a lot of activity up your way as well. (Counting myself lucky I live in safe-as-it-gets Sydney, Australia)

    • The typical focal mechanisms are extensional/normal earthquakes in the Basin and Range, which is technically east of the Sierra. However, there is tectonically-driven extension taking place to the west, and the Owens Valley earthquake in 1872 (well north of last night’s 7.1 mw) was such an event where the eastern edge of the Owens Valley dropped nearly 20′ and shifted laterally almost 40′ as it slid down the western flank of the Sierra.
      For this latest Ridgecrest sequence though, all these quakes, now numbering in the thousands, have been either left-lateral or right lateral strike slip (depending on which fault is involved) and are occurring at very shallow depths within unusually brittle crust.
      Here’s the drumplot from nearby Coso Junction bracketing 7.1 mw shock about 12 hours ago.
      If I didn’t know better, I’d say there was a volcano going off. Amazing how active the aftershocks are.

      • The Mono Craters, a line of volcanic domes and craters south of Mono Lake in eastern California, represent the youngest rhyolitic volcanoes in the western United States. Rhyolite is a magma that is viscous and prone to explosive eruption.

        This field is next to Coso.

        Also in Death Valley just east of this current activity (north end of the valley)

        Yesterday’s Volcano
        Signs of recent volcanic activity exist in northern Death Valley at Ubehebe Crater. Caused by violent steam explosions, the craters may have formed as recently as 800 to 1300 years ago when hot, molten material came in contact with groundwater. These large depressions show that Death Valley’s geology is dynamic and ever changing.


        • Had a hard time reading Craig’s plot.

          This is for the last 24 hours.

        • I heard that Kevin Bacon has already looked into the cause of these “earthquakes” up in Lone Pine. I believe there’s already a movie about it called ‘Tremors’

    • In California the San Andreas does not normally take part in such events. It only does the biggest quakes, while all other faults in the state do the lesser ones. Around Palmdale, the San Andreas fault has been locked for well over a century. It is quite far from the recent events (50 miles), but there is a risk of increased strain on the San Andreas in this area. Over the next 6 months, there may be a slightly enhanced risk of an event.

      • You can have a 7.1 or smaller event on the San Andreas. The southern section is locked and has been for 150 years but the area around Palmdale is creeping, and the northern section is not as jammed up.

        I think the Loma Prieta earthquake was on the San Andreas or very close to it.

        • Loma Prieta wasn’t on the San Andreas but rather a blind splay fault.
          The focal mechanism was oblique and clearly different than most/all other San Andreas shocks.

          • Did not know that. Very close to San Andreas though, it goes right though the Santa Cruz mountains.

      • Here’s an article somewhat less optimistic than USGS:

        “This Temblor app map with another 2 hours of events gives a different impression of the M 7.1 aftershocks than the initial USGS map, suggesting that the rupture does not simply extend to the northwest. Based on these aftershocks it appears ‘bilateral’, meaning that the fault unzipped both to the northwest and southwest, for a total length of up to 55 km. This would be more consistent with its magnitude, as a strike-slip M 7.1 typically has a length of about 50 km. If this is correct, then parts of the Garlock Fault might also be brought closer to failure.”

    • It is interesting, but they do also forget that loihi is still a baby volcano, it still has a long way to go to reach the level of activity kilauea has now and will take longer to recover. If you go to the vc bar I put a comment on the subject of kilaueas probable recovery time(well, another one but with an independant source this time) but in short it is on track to be fully recovered in 3 years time from now.

      I think mauna loa in 1984 would be a better comparison to loihi’s 1996 activity, apart from the collapse the two have very similar precursors, last year was mostly the inevitable failure of the open magma system feeding puu oo as the pressure grew too high after many decades of eruptions and the high elevation of all the easy vents, it became easier to go downrift than anything else and once that started it just kept going until things evened out. By comparison loihi 1996 and mauna loa 1984 were more derived out of a new pulse of magma with more negligible summit drainage, which results in a more limited but intense eruption.

  4. And in other news: Vatnajokull has been awarded the status of world heritage site. Congratulations to Iceland!

    On the flip side, such sites are protected against changes, so further eruptions within the national park may not be allowed.

    • And hot of the press: the Jodrell Bank Observatory with its iconic Lovell Telescope has also just been added to the UNESCO world heritage list.


        Jodrell Bank Observatory has been declared a Unesco World Heritage Site.

        It has been at the forefront of astronomical research since its inception in 1945 and tracked US and Russian craft during the space race.

        The site in Cheshire is part of the University of Manchester. It is dominated by the landmark Lovell Telescope.

        … It is so sensitive that mobile phone use on the site is normally forbidden and the staff microwave oven is shielded by a metal box to prevent interference.

        The site featured in BBC’s Stargazing Live series, Douglas Adams’ Hitchhiker’s Guide to the Galaxy and Doctor Who

  5. This plot shows why Mauna Loa’s status was upgraded to yellow. Inflation is not yet at danger levels (it inflated faster in 2015) but overall activity is higher now.

    • A lot of that sudden jump on the red line is from the station recording quakes from kilauea last year, in june july and august, we all know what was going on in those months.
      But yes there is more happening now than before. Will be interesting which of the two happen first, both are capable of big things in the next 5 years. Maybe they will even co-erupt again or within the same year.
      2002 and 2015 didnt result in eruptions but theres only so many attempts that can be made before the only outcome is to breakout somewhere from overpressure, theres already a lot more magma intruded than was needed to recharge from 1984 (about 0.01 km3 a year, so 0.35 km3, 1984 was 0.23 km3). Apparently mauna loa erupted in 1750, 1780, 1803 and 1832 before its first well documented eruption in 1843, so it going 36 years without erupting is not unusual if including these, but not expected either. It does indicate it probably wont take much to tip things over now though, and HVO directly compared to 1984 which was not done in 2002 or 2015.

  6. I found the GPS units for the Long Valley network, p594 is one of the instruments on the link below. You can change instruments by clicking on the instruments on the map, it also shows the quakes.

    I looked at several locations (note some show on map but are no longer working) and can see some movement but with out spending time studying the instruments I can not make any conclusions. I pasted in the last 30 days for three of the instruments, p594, p595, and cccc

    Oh there are lots of instruments so take it easy. Maybe there is a paper that takes all this information and makes it easy (there is a paper listed on the network but it is pay walled)?? Ridgecrest quakes for Dummies?


  7. Perhaps a bit more pertinent to the blog….a new EQ swarm has started near the Coso Volcanic Field north of the 7.1 shock. No reason for any real concern ATTM, but definitely something to monitor.

    36°1’48” N 117°49’12” W, Summit Elevation 7874 ft (2400 m)
    Current Volcano Alert Level: NORMAL
    Current Aviation Color Code: GREEN

    An earthquake swarm started on the evening of July 5 at the southern margin of Coso Volcanic Field in Inyo County, California. The swarm activity was triggered by a magnitude M5.4 earthquake at 9:19 PM PDT located 20 km (~20 miles) ESE of Little Lake, which itself was likely an aftershock of the M7.1 earthquake that occurred about an hour earlier 17 km NNE of Ridgecrest, and south of the Coso area. An average of about 30 earthquakes per hour have been detected since, most within the range of magnitude M1 to M3. No ground deformation indicative of volcanic activity has been detected, and there is no imminent threat of an eruption. The California Volcano Observatory will continue to monitor the situation for any sign of volcanic activity and provide updates as warranted.

  8. The area IS volcanic. See the “Lava Mountains” south of Ridgecrest.

  9. This is a USGS model that shows which faults are ‘primed’, i.e. have a greater chance than normal to rupture. It is a few years old and may have been updated since. The southern section of the San Andreas has not ruptured since 1857 and is likely to fail within the next 25 years, although ‘likely’ is not the same as ‘will’. The state is riddled with faults. Most ‘smaller’ quakes (which can be well above M6) are on the other faults. The San Andreas itself is heavily locked and tends to give way in very large quakes only.

    • Notice all the faults. Pretty much all of California is made up of the old accretion prism scraped off of the Farallon when it went under. Some parts well sutured, others less so.

      • And that’s only a fraction of true number of faults. In addition to the main faults shown here,
        there are lots of others, including many blind-thrust faults where there are no seismic data that reveals their location.
        Notably (but not limited to) the 1989 Loma Prieta and Northridge quakes occurred on essentially unknown/minor faults, with both having significant vertical displacement(s) consistent with thrust fault activity.

      • I was wondering how so much land was on the Pacific plate side of the San Andreas.

Comments are closed.