USGS photo, M Zoeller, 11 May 2025
Rainbow’s end
It is an amazing and powerful image. The wide lava fountain in the caldera like a wall flower in full bloom, with the cloud of volcano seeds above, the lava flowing from the bleeding flower, slowly re-filling one of the largest holes on Earth, and the distant rainbow linking the lava, the caldera, the surroundings and the clouds. The sky, the ground and the Earth all connected in a fiery and colourful dance.
Tradition has it that there is a pot of gold at the end of the rainbow. Which end is always left open. In this case, it is obvious. It has to be the end within the caldera, and the gold is being brought by the lava flow which conveniently terminates at the end of the rainbow. The lava even has the colour of gold. There is no doubt about it: this is the volcano which lays the golden egg. Kilauea is a gold mine and the rainbow is the tool which the miners are using to get the gold from the hole into their pockets. USGS has caught the gold bug and has gone full-Trump.
The reality is a bit more prosaic, of course. The lava is liquid rock, shining like gold because of its heat and carrying at most traces of gold. Archimedes in his Eureka-moment is measuring the volume of the lava to decide what part is gold – and what part is rock. The golden dreams evaporate as quickly as the falling rain: there’s no gold in them thar hills. Mark Twain visited Kilauea and wrote about the lava flows in the cavernous caldera: ‘ streams met other streams, and they mingled with and crossed and recrossed each other in every conceivable direction, like skate tracks on a popular skating ground. Sometimes streams twenty or thirty feet wide flowed from the holes to some distance without dividing – and through the opera-glasses we could see that they ran down small, steep hills and were genuine cataracts of fire, white at their source but soon cooling and turning to the richest red, grained with alternate lines of black and gold’. His ‘vision of hell and its angels’ ended in gold. But when Mark Twain coined the expression ‘there’s gold in them thar hills’, he rightly did not refer to Kilauea where the gold is fool’s gold. (As an aside, there may be some gold in the Thar desert – but that is another story.)
But why our obsession with gold? It is just one of many elements, and to be honest, it is not one of the most useful. It does have its uses. Especially electronics can benefit from gold. Gold conducts electric currents and does not corrode. This makes it useful in applications which require very low voltage and/or current, where any corrosion could be fatal. Your phone or even laptop (remember those?) will contain some gold. The metal is also soft and pliable, ideal for making wires. Gold can easily be alloyed with other elements to make it harder, if needed. As an aside, that may changes its colour, as shown in the diagram. The colour of magma is clearly not pure gold.
A second use of gold is for fillings in dentistry, although this is much less common than it used to be. But of all the gold in circulation, only around 10% is used for practical applications. 90% of its use is based on its perceived value. Gold is pretty, does not tarnish and is rare: what better metal is there to show off your worth? It is the ultimate crypto currency, the tulip bulb mania of today. Wedding rings, medals, oscars, gold fillings, gold coins, it is all based on showing value and status. If you really want to show off, be seen eating food covered in gold. It has no taste and is harmless, so this really is about show, not sustenance. The gold reserves of the central banks fall in the same category. The word ‘gold’ comes from Sanskrit, with as original meaning ‘shiny’. It was always about perception.
And this is not particular to modern western culture. Buddha statues were often covered in gold, in order to reproduce the reported colour of his skin. Buddhist sacred texts often use gold calligraphy against a blue background. The Incas called gold ‘tears of the Sun’. The Romans and Greeks had gold jewelry. In China and Japan, the use of gold was at times only allowed for high ranking officials. But none of these cultures found a convincing use of gold other than display. Perhaps its value made actual usage impractical even in those days.
Gold wreath, Corinth. (Source: Brooklyn museum)
Striking gold
Gold is found in many location on Earth, but there are some major deposits that have affected nations. The best-known is the Witwatersrand in South Africa. When England let the Afrikaners depart on their ‘Great Migration’, the plan was to let them have land of little value. As it turned out, milk and honey may have been in short supply but the Afrikaners were gifted enormous deposits of gold and diamonds, with minerals thrown in for good measure. Hence the Boer Wars when Musk Rhodes tried to gain control of the riches. It became a war of starvation which would haunt England’s conscience for decades. This is what gold does to people.
Gold is one of the densest metals: it weighs 19 grams per cubic centimetre. In comparison, lead is only 11 grams per cubic centimetre: many applications where lead is used would benefit from gold – if only we could afford it! The densest metal, osmium, is 22 grams per cubic centimetre, so gold is not far off. Of the usable metals (so not counting uranium and plutonium), only osmium, platinum and tungsten beat good old gold.
What all these metals have also in common is their rarity. That is not entirely accidental.
The fraction of the various elements in the Earth’s crust is shown in figure above, sourced from wikipedia. Many elements hover around 100 on the scale, where for every 1 million atoms of silicon, there is one atom from that element. Nine elements are well below that, highlighted in yellow (or mustard, if you prefer): rhodium (Rh), ruthenium (Ru), palladium (Pd), tellurium (Te), rhenium (Re), osmium (Os), iridium (Ir), platinum (Pt) and gold (Au): they are each a thousand times less abundant than, say, europium (Eu). Tellurium is a bit the exception in this list, but the other 8 are all so-called noble metals, resistant to corrosion. They don’t react with oxygen but do dissolve into molten iron. These elements are therefore known as ‘siderophiles’, literally ‘iron loving’.
And that is the reason for their rarity in the crust. Once upon a time, there was a prince the Solar System was new and the planets hadn’t formed yet. The raw material for the planets was floating around in space, around the new-born Sun. Now some elements exist in space as solid particles even at high temperatures: iron, silicon, magnesium, while some other elements evaporate easily, such as oxygen and carbon. So the latter were only solid far from the Sun and the former could exist as solid particles much closer in. The inner planets formed when these solid particles came together, while far out in the Solar system worlds formed from frozen water and carbon dioxide. This is why the inner planets are rocky: Mercury to Vesta, and the outer planets are icy: the comets, icy moon and ice giants.
As the Earth assembled itself, it grew from the infalling particles and later from gobbling up larger fragments including other protoplanets. (The Earth has a dark history. There is a reason it is the largest of the terrestrial planets.) The energy of these collisions melted the entire planet. And in this magma planet, the heavier elements began to sink to the bottom. This was mainly the ubiquitous iron, but it carried with it those high density iron lovers. Gold, ruthenium, palladium, down they went to the bottom. After some tens of millions of years, much of the Earth solidified, but not before almost all the iron had gone into the (still-molten) core. The Earth’s gold ended up with the iron in the deep, down into the core.
But not all. The deep mantle still contains a fraction of those noble elements. Deep mantle plumes can bring them up to the crust. A large intrusion from a plume can form enormous sills of mafic magma, sitting there for millions of years, unable to break through. Slowly, very slowly, the sills solidify. That goes so slow that the elements and minerals separate, the heavier ones sinking to the bottom and the lighter ones rising to the top. When the sill finally becomes solid, it has also become stratified. There are now layers upon layers containing high fractions of particular elements.
More time passes. The overlying crust is very slowly removed by erosion, possibly over billions of years. What once was tens of kilometers below the surface now is coming closer and closer. And finally, people come and find those layers. The most impressive of these is the so-called Bushveld igneous complex, a series of intrusions in short succession, 2 billion years ago, covering a region 300 by 200 km wide, north of Pretoria. The Bushveld complex in the Afrikaner heartland is now the source of 75% of the world’s platinum and 50% of the palladium.
South Africa’s gold has a different origin. There is some in the Bushveld, but the mineable deposits are found in other regions nearby. A major locality is of course Johannesburg, home of the earliest gold rush in South Africa, which led to this city developing in a locality where you might not expect one. This is the Witwatersrand Basin, an old sedimentary basin. The gold here was deposited by water, as much as a billion years before the Bushveld intrusion. If your continental plate is as old as South Africa, it may have seen more than one special event!
If you are interested in reading more about this ancient land, there are several posts on South Africa on VC, such as https://www.volcanocafe.org/the-drakensberg-and-the-storm-that-ended-gondwana/
The golden road
Water plays an important role in gold. People still pan for gold along rivers. The gold particles are eroded from their origin, and, being heavy, slowly make their way downstream, perhaps long after the original deposit has gone. In the Witwatersrand, this played out to an extreme degree where the gold collected in large reefs, thin but kilometers long. That required a rich source of gold, but also a mechanism to bring the gold to this reef. Could rivers do that? And why do most of the gold particles show little evidence for rolling along river beds (‘mechanical erosion’)? It appears that they formed in situ.
It was a different world, 3 billion years ago. There was no free oxygen in the air. Massive volcanic eruptions (you were waiting for these, weren’t you?) were bringing sulphur into the atmosphere. There is evidence for major flood basalts in the region, 2.87 billion years ago, at the same time that much of the gold was being deposited. A second deposition, 2.71 billion years ago, was covered by floor basalt shortly afterwards. What did the eruptions do?
The volcanoes brought the sulphur into the atmosphere as SO2 and H2S. The latter would not get oxidized for lack of oxygen. It ended up in the rain and in the water courses. Rivers would have been more acidic than nowadays, with pH of 4.5 to 6. In the presence of H2S in the water, gold dissolves as AuHS and Au(HS)2–. It still would have eroded from a gold source up in the long-gone mountains, but dissolved into the rivers. It would not remain as nuggets in the stream but quickly end up in the lake downstream.
Source: Heinrich, Nature Geoscience, 8, 206 (2015)
Sulphur is removed from water by reactions with iron, which forms pyrite – fool’s gold. As this occurred, gold would come out of the solution and quickly rain down on the lake bed. The idea is that every major volcanic pulse would lead to this process, ending up with a layer of gold at the Witwatersrand lake, accompanied by pyrite – exactly what is found. It is not a fast process: concentrations of dissolved gold remained low. But for an eruption like the Deccan traps, the released sulphur over a river catchment area could have formed a gold reef with a million years. The flood basalts in those archaean days may have been rather larger than nowadays and more sulphur-rich. It wouldn’t work in our days anyway because of the presence of O2 in our atmosphere. That gold in them thar hills – they don’t make it anymore.
But where did the original gold source come from? The gold reefs came from an eroding mountain – how did that golden mountain form? It turns out, it is the same process. Gold brought to the crust in an intrusion stays locked in the deep crust – until it meets water. Wait long enough, and a subduction zone will come along. The descending plate releases water and sulphur, and both percolate up. The water-soaked rock melts, and a subduction volcano forms. But in the deep melt, the gold meets the hot sulphuric water. Gold, normally reluctant to join in with chemistry, forms an AuS3 complex, and this form of gold travels more easily with the melt. The gold now ends up near the surface in veins. There are other models for how gold gets to the surface while other heavy noble metals do not, but this one seems to work well. This process is the goose that lays the golden egg.
But where did the gold come from originally? How deep does the plume need to be to form a gold-rich intrusion? Gold is heavy and likes iron: can it remain in the deep mantle or did it all end up in the core of the Earth? That is the golden question.
Ruthenium
The answer, it now seems, has come from one of those other funny elements: ruthenium. Like the other dense siderophiles, it is expected to have been taken into the core in the first 60 million years of the Earth’s existence. But the rain of meteors continued and this brought in some more ruthenium, which remained in the crust. There was however a difference. The new ruthenium had a slightly different ratio of the different isotopes. Ruthenium has 5 different isotopes, ranging from 98Ru to 102Ru, each accounting for 10-30% of the lot. The crucial isotopes are 100Ru and 102Ru which are a bit higher in the earlier ruthenium, i.e. what ended up in the core.
Source: Nils Mesling et al, Nature, May 2025, https://doi.org/10.1038/s41586-025-09003-0
Mesling measured these isotopes in basalt from various locations. The results are shown above for the 100Ru isotope. The modern upper mantle shows values around zero (meaning no enrichment or depletion), represented by the purple bar in the figure. That is seen in the Eifel in Germany, indeed a shallow hot spot. La Reunion and the Galapagos show similar values although with a larger uncertainty. The Bushveld intrusion shows depletion whilst ancient basalt from Greenland shows strong enrichments. But very notably, various samples from Hawaii (including Kilauea Iki) show enrichment.
What does this tell us? First, the Greenland values are very high and represent the oldest material. It came from a deep layer which was pure stuff from the oldest days. The Bushveld values are similar to those of the later addition of ruthenium, so represent the pure younger material. The older and younger material had not yet mixed to any significant degree: magma showed either one or the other. In the modern upper mantle the two are mixed, and this mix is shown in the Eifel. But Hawaii shows enrichment. It is not just getting bulk mantle, it has enrichment from the oldest stuff, although it is not pure old: there is some mixing. The extra, old stuff must come from the deep.
The authors argue that the Hawaiian ruthenium is a mix from material from the mantle and from the core. To get this, there must be some core material that is getting out at the core-mantle boundary, creating a layer with a mixed composition. The Hawaii plume is reaching into that layer. It is bringing core material to the surface!
The mixing of core and mantle may have happened at some time in the past, or it may still be happening: that would be hard to tell. But at some point, core material has been leaking back into the mantle.
Gold from the core
The article discusses other elements but not gold. News reports, however, jumped on the golden angle. For if ruthenium managed to escape the core, it could be expected that related elements could also do this. And gold is very much related: it too sank into the core, so did it too find its way back up? Does the gold that we value so highly contain material that long ago, as the Earth formed, sojourned to the core – and came back? As the authors commented in news reports, at least some of the small supplies of gold and other precious metals may have come from Earth’s core.
Gold is precious to us. People who have it, treasure it. And now we have another reason to admire it. This shiny, pliable metal has been on a journey. It has been carried by volcanoes – perhaps more than once. And now we know that that ring on your finger may even have seen the core of the Earth. A heart of gold.
Albert, May 2025
Sources
Ru and W isotope systematics in ocean island basalts reveals core leakage. Nils Messling et al. 2025, Nature, https://doi.org/10.1038/s41586-025-09003-0
Witwatersrand gold deposits formed by volcanic rain, anoxic rivers and Archaean life. Christoph Heinrich, 2015, Nature Geoscience, 8, 206
Before Ep. 4 starts, I would like to share an map I’ve made of what the field looks like (each brighter shade = >100 feet)
Also, here’s a poorly photoshopped image of what a potential record-breaker could look like (it might be a little smaller than this…)
Sorry, Ep. 24 (forgot the two)
Very nice. It will be great when ew topo maps update to show elevation change of the new stuff. Although that might only be possible if tall foutaining stops or dies down for a while.
That map is just speculative, eyeballing the features and using thermal maps from other episodes. Here’s more of which I play around with if the episodic nature of this eruption continues:
A few more. Not really noticeable, but you could see the lava field grow a bit. I just drew them willy-nilly, so I don’t really know how many episodes it did at that point.
The lava field isnt really going to change until it overflows the rim somewhere, the inly changes likely before that are the cone growing on the rim which will be dependant on how long and intense the fountaining stage is. I would be surprised if it is still doing high fountains at the point of overflow but at the same time the only way that would stop is by lava being able to erupt elsewhere. That is either by lava pushing into the lava lake below the new lava surface formed since December, or by a dike going directly down the SWRZ like in 1919 or 1971, but neither option looks particularly likely to happen within the next year at least.
There is also the option that the fountains just stop on their own, but that isnt actually very consistent with how Pu’u O’o evolved. Its fountains were only ever getting more powerful on average until pressure caused magma to intrude into the nearby ERZ instead, but actually there were numerous times before 1986 that actually failed to end the episodic eruptions, so even this method is unreliable. So long as the magma supply is high enough to keep the vents open, and the pressure stays at the summit, these eruptions will continue in my opinion. There still might be 500 meter fountains every few weeks in 2030 sending long a’a flows down the south flank. Notably most of the SWRZ underneath the Observatory overflows is actually a’a all the way to the ocean so there could be precedent, and the next few decades could be spectacular.
Looks like we are on again.
Perfect breakfast viewing! V3cam looks good at the moment
And E24 has started 🙂
The tremor is still only barely above background but is increasing. Still it shows there isnt actually a high eruption rate yet but fountains are still tens of meters tall, the magma seems to be very gas rich now in a way it wasnt before.
Tremor seems to be ramping up now
That might be the strongest I have ever seen it get, wow.
The fountain isfalling on the rim in huge volume, there might well be some lava flows up outside the caldera if it keeps going that way. There did look to be lava flowing fown the cliff from the rim when it is visible.
The north vent became the north vents, the neat cone and spillway was blown apart before the high fountains began. The south vent is still only weak but will probably join in soon. The glowing tephra cloud is enormous easily more than twice the height of the caldera wall.
Pele was really set on destroying the bluff around 22:30 local (V1 cam). She must have seen a certain hog centaur wandering around.
Im looking forward to the pictures of the aftermath. Going back an hour on tbe V3 stream, the foreground has been built up an enormous amount from all the tephra that fell there, like probably tens of meters, its level with the vents now…
At this rate within a few weeks the caldera wall will actually be buried, just a steep ashy slope probably ruddled with crevasses, its almost unbelievable except its literally forming on camera as we speak… Maybe by the end of the year Kilauea will even have a new summit peak.
V2 cam is insane! Looks like a lava heart?
V1 shows the fountain going outside the caldera rim now!
While this is still the relevant article, I think it needs to be put into perspective that the fountain now wpuld probably go far past to to of the image at the top of the article.
The south vent is also starting to get serious, it is easy to ignore but last episode it was only slightly shorter than the north vent fountain and kept that up for longer too. If it was day time basically everything in the left side of the V3 livestream would be grey too… 🙂
HVO reports 300 m high fountaining
Tiltmeter at UWD has fallen 3 microradians in about an hour, which is an eruption rate of 1.5 million m3 an hour roughly. At this rate E24 will be over before dawn, for only around 4.5 hours of activity. It will probably slow and tail off past 5 or 6 hours but still the eruption rate is starting to reach almost unbelievable figures. The most powerful eruptions at Svartsengi have been around 2000 m3/s but that is from a fissure that is many km long, I think over 10 in the relevant example. Kilauea is getting to within a factor of 4 of that right now with only two vents within about 30 meters of each other…
Seems roughly equivalent to 23 in terms of how fast the tilt is falling.
Yes it looks similar too, although the south vent hasnt evolved into the taller fountain despite high output of about half of the lava now, and the north vent fountain is still huge.
Svartsengi reached 1.500m³/s on average during the first 4 hours on may 29 with 4km fissure length. I dont know the average for the August eruption but it probably peaked at more than 3000m³/s on roughly 7km fissure length, though the last 2km only opened after that peak was reached, so during its peak it also had 4-5km length. No svarts eruption fissure of the current cycle reached 10km yet 🙁
2000 m3/s was for the eruption last August, maybe it was more but probably not 50% more at least not for long. I thought it was over 10 km long too, I guess not.
Lava jet again: https://www.youtube.com/watch?v=oG5zz9Sjw3E
Episode 24 and 23 are related by this type of lava fountains. How far can we compare the strength of the episodes?
View from Uēkahuna bluff on NW side of the caldera (near UWD tiltmeter):
Between episodes 23 and 24 there were around 10 micro cycles of deformation. The same number as between episodes 22 and 23. Maybe this help to get a prediction about the time for the the 25th episode (if the behaviour doesn’t change).
Also the pauses between the episodes has extended to ~10 days. The pauses are longer, and the episodes stronger than before episode 23.
The micro cycles I think are just daily interferance, some tiltmeters seem to show it more than others. But otherwise I agree, episodes are stronger and more intense, with a bit more of a gap. I wouldnt be surprised if gaps end up weeks long between huge episodes soon, although its hard to trust that something wont change suddenly.
The sun heats the ground unevenly inflating some areas and tips the tiltmeter which is super-sensible, so the micro cycles are the day-night cycles casting deformation on the instrument.
Yes, that is correct. It is diurnal variation caused by the heating of the surface by the afternoon sun. It is not always present, presumably because not every day is sunny. Rain can also induce tilt variations. Tilt measurements are rarely as clear-cut as in the current weekly Kilauea series!
OK, then it’s just the 10 days time frame between the episodes. We’ll see if this continues. If yes, then next episode will be on 15th June. After episode 22 the average duration of breaks was ~10 days (16th May, 25th May, 5th June).
I went back and found my oldest saved picture of tge eruption that can be compared. It is from January 27, so a month after the start.
https://i.imgur.com/iAkpkJ1.jpeg
Compared to this one which is from now… I tried to zoom it in to get similar scale for the back wall, it shows how much bigger it is now…
https://i.imgur.com/kYbkJFi.jpeg
I have a feeling that this steep area might collapse in a few fountaining episodes…
At about 3:33:30 on the V3 cam, I saw a molten (or hot) chunk far off to the south fall down the slope into the crater. I wasn’t up when the episode started, but was the fountain directed that far off the vertical so that a rootless flow started outside the crater?
Wow, never mind–I rewound and there was a lot of spatter welding to the crater wall at the start of the episode. Should be interesting to see once the sun comes up.
Rootless flow is hard to say but yes it went way up and rained hell onto the rim. I wouldnt be surprised if that rim is at least 10 meters taller and did form a rootless flow of some sort.
Yes, initially the fountain was blowing over the rim. But it was probably just hot tephra sliding back into the crater.
Sadly webcam B1 broke down one hour before the episode began … it had a good view over the down-dropped block and the lava field from the east side.
North vent has shut down a fewminutes ago.
And then eruption is over about 10 minutes ago.
From 8:55 pm to 4:25 am, 7.5 hours more or less. A bit lonver than I predixted earlier but the second half slowed down a lot
The episode lasted around 7 hours, including weak beginning/end. That’s close to the 6 hours of episode 23. So a similar duration of both the eruption and the breaks between the episodes.
The first 17 episodes until April 7-9th lasted relatively long. The shortest episodes were 13 hours: Episodes 6, 10, 11, 13. After the 17th episode no episode lasted as long as these short episodes. In April there was a significant change towards shorter and more energetic episodes.
The shortest episode of all episodes was a relatively weak one on 6th May (20th episode) that lasted 4.5 hours: https://www.youtube.com/watch?v=PvWp-W2o3mQ
After episode 22 there was obviously a new significant change towards short and high-rate eruptions. It’s uncertain, how long this behaviour is going to last, but episodes 23 and 24 belong to this group.
All in all we can discuss about phases of the current summit eruption:
I December to early April (1st to 17th episodes) introductory phase
II End of April to Mid of May (18th to 22nd episodes) increasing phase: Higher rate, shorter usual episode duration
III End of May until future (since 23rd episode) strong phase: Even Higher rate than in II, longer breaks, shorter duration
Is it accidentally that tiltmeters and many other graphs aren’t accessible now?
The NE part of the Kilauea Caldera towards Visitor Center had recently a number of earthquakes:
What causes these earthquakes? A tectonic inbalance of the caldera block by the mass of the fast growing lava shield on the SW part of the caldera? Or is there a magmatic reason?
I can’t remember back that far, but were there similar quakes in the same area when the summit collapsed in 2018?
Fuego is having a major paroxysm:
https://www.youtube.com/live/XRtEHKa2vmQ
Yes, looks like Etna, but more aggressive.
Pyroclastic flow: https://www.volcanodiscovery.com/fuego/news.html
Took a snap shot of the livestream from V3. Red represents “dark falls” (which lava/molten tephra might’ve fallen back) and the blue representing scarps, of which material fell from. A few more episodes like Ep. 23/24 and the wall might get covered up.
Here’s the original just in case.
And here’s my over-amateur analysis on the terrain. Purple represents the slime cracks from the forming tephra cone (or plateau), orange or vents, green are lava flows (M= massive or ā’ā; P= pāhoehoe; S= sheet) and the black line represents roughly the pre-eruption top of the caldera.
Slump, not slime.
Ah, I was just looking for a new volcanic life form to emerge from the ash
You’ve just reminded me of The Presence by John Saul. It’s a sort of action-mystery thing around some biological discovery in Hawaii. It has volcanoes (Kīlauea and Mauna Loa, maybe Haleakala or another Hawaiian volcano that starts with an H), but those are the only spoilers I’ll give. They had a good concept, only to sort of madden me with some of its plot elements, but overall a meh read.
It’s a really nice high fountain cone. The wide, smooth horseshoe shape and the concentric cracks are very typical of these cones.
Yeah, surprising that it turned from a few fissures to this. I always wondered what this terrain will look like by the end of this year. I think it could go many ways, but these are the ways I could think of by the end of the year:
1. Episodic fountaining continues – this is the “exciting route”, shooting out fountains on a sort of natural yet unpredictable clock. It seems to be pausing longer while getting more intense with each episode (maybe some record breakers, hopefully). This is the one route I admittedly wanted to go.
2. Turns to shield building – either the gas runs out or the vent is getting too wide to make tall fountains. By this point, it might be Fagradalsfjall 2021 2.0. I think this might be likely.
3. Eruption just ends.
4. Something breaks out on either rift zone – this will divert the magma from the summit vents, maybe even collapsing them.
These are just all the possibilities that, personally, I think might happen.
Anyways, here’s the elevation map for what the summit might look like if it took the exciting route.
And another map to detail the type of terrain here.
Dark red means the cinder cone, black for tephra blanket, brighter red for the tephra/lava mix, grey for debris/tephra mix, bright red for the lava and orange for the vents. This might look different by the end of the year but it is fun to speculate.
Regarding which is most likely:
1 and 2 are about equal chance. 3 is less likely but grows more so over time.
To be honest, I dont actually know how 4 can happen without either 3 happening first or Mauna Loa taking over.
Is the bay in the SW corner filled completely? The recent maps look indeed like this:
The SW bay is/was behind a steam cloud. It’s difficult to see this part well:
Google Maps still shows the bay in the SW corner of the caldera that gets buried slowly: https://www.google.com/maps/place/K%C4%ABlauea+Crater/@19.4163883,-155.2861544,4285m/data=!3m2!1e3!4b1!4m6!3m5!1s0x7953d0a313a8642d:0xc482c92398afd2da!8m2!3d19.4163889!4d-155.2758332!16s%2Fg%2F121n9f8h?entry=ttu&g_ep=EgoyMDI1MDYwMy4wIKXMDSoASAFQAw%3D%3D
Howe long before Halemaumau ceases to exist?
I’ve heard from Chad that the volume of Halema’uma’u stands at 0.7 km³, but I’m not quite sure. If the 0.7 km³ is the case, it might be about 10 to 15 years if the output is 0.1 km³ a year. If it’s 0.2 km³ a year, then it’ll be half that. Basically, according to Chad, you had to take into account the fact the future shield might overflow before it even fills the 2018 collapse. This is all only if it doesn’t stop.
Albert Halemaumau will still exist if the caldera fills, it refers to the home of Pele, not specifically to a geological formation. Otherwise you could argue Halemaumau was destroyed in 2018, or in 1924.
I think we will know how consistent this is soon though. Probably by the end of the year there will be a significant tephra cone on the rim that could get a name.
There is about 270 million m3 in the caldera now, and if the caldera had the highest published 1 km3 volume it still has about 0.68 km3 to go, which is about 3-4 years away at this rate. Still before 2030. The google AI immediately gives 0.8 – 1 km3.
About the vent overflowing the south rim before the north, after recent episodes im not so sure about that now. If the fountains stay this big then by the time the vents are tall enough to overflow the original tim there will be 50-100 meters of tephra blocking it… Lava leaking through under that tephra, through cracks, or flowing over the 1974 flows to the southeast might happen though.
The southern caldera rim is relatively low. I’d expect that lava flows exit here, before the whole Kaluapele caldera can be filled.
In the long run the current eruption may become a SWRZ version of the Ailā‘au eruption. Imagine that the whole eruption is going to last for 60 years. Sooner or later the eruption would change to a continuous lava tube eruption like Pu’u O’o. If the volume is like Ailā‘au, then lava would cover most of SWRZ and do an ocean entry somewhere down there. Although the current eruption is no SWRZ eruption, a big volume is able to cover a lot of the SWRZ region.
It looks as if the cone’s peak is only ~50m below the caldera rim. Because of the night during episode 24 it’s difficult to say where the “dark falls” came from: Tephra or lava? I’d rather say tephra, because the lava fountain took a different direction.
This video shows the lava fountain of E24: https://www.youtube.com/watch?v=sVGlmQPL_kA
The immediate cliff above the cones is dotted with lava bombs/drops and short flows that originated there. But above the cliff, there is probably only tephra.
Just after the post I’ve clearly seen how lava fountain dropped lava on the caldera rim. So my guess was wrong. The video shows from 0:48 how the lava fountain covered the caldera rim and lava flows run back into the caldera. The speaker of the video also describes it clearly as a “glowing lava cascade back down the cliff”.
HVO observed that the eruptive plume reached at least 5,000 meters high. This applies to a VEI2 or 3, if we only look at the vertical size of the ash plume.
B1 webcam finally published a single photo yesterday:
Does it show a progress of the lava field towards the east?
GPS data show a slow and flat inflation of the summit since February:
Does this indicate that the inflow of magma is bigger than the erupted volume?
Most of the GPS stations on the HVO web have not been updating for the past few months, which is a big nuisance, but the few active GPS in the Halema’uma’u and south of the caldera areas show slight deflation, so the eruption is most likely erupting a little more than is being supplied.
The UWEV-CRIM extension likely reflects the slow spreading of Kilauea, which might well be taking up some magma too, as deep, perpetual dike bodies expand.
If the eruption erupts more than it receives, can we exclude an escalation of the eruption? Maybe the extension of the pause between episodes to 10 days mean that the output per month hasn’t increased, although the individual episodes had high rates during their short lives. What is the reason for the recent quakes in the northeastern part of the caldera?
The deflation is so slight compated to observed magma supply its close to negligible in comparison, its possible that deflation on this scale isnt even really related to the deep supply but just the fact there is an open vent relieving pressure that reached a high last year.
I think too, the open vent being now very robust and most likely a permanent feature of decades to come, that will have strong influence. Magma is a liquid and so incompressible, if there is no outlet the supply will slow and pressure increase, which is where it broke end of last year. In theory supply is near zero when an eruption happens although probably not quite in reality.
There is one key factor though. With an open vent at the hottest and most direct path in the magma system, its not unlikely that the magma supply now is almost entirely unrestricted. The only thing at odds is gravity but that only really applies as an obstruction if magma is able to flow into other parts of the volcano which was true last year but not since September.
Basically all this is to say there is a very good chance a very large area of Kilauea will be resurfaced by 2050…
Usually an episodic eruption like this precedes a longer period of steady eruptions. Maybe this is the way also the Ailā‘au began in the first year. We might see a similar eruption like Ailā‘au, but on the SW corner of the Caldera, where future lava flows would rather run over SWRZ towards the coast.
Imgine this eruption for 60 years, maybe a bit changing pattern over time, but more or less similar eruption rate. This have much more volume than Pu’u O’o.
You are almost perfectly describing the Observatory shield 🙂
4 microradians in 1.5 days, recovery is faster than after E23 so far… At this rate E25 will be only in 3 days, not the over a week expected.
Recovery right now is notably faster than after any episode of the past 3 months; it will likely slow down, but nonetheless, it’s impressive.
Coincidentally, roughly the day of Ep. 24, a year ago, started the massive inflation period of Kilauea volcano that uplifted the summit 40 cm in 50 days as an area 20 km long was spanned by deformation, and swarmed both connectors and the summit at the same time.
After deep deformation, the recovery afterwards has usually a concave growth. First steep inflation, later decreasing inflation.
The fast recharge of the summit and high rate of the overall eruption resembles a bit the 1790-1840 period, but I have the impression that we see the beginning of something bigger than that. The inflation last year … and the Pahala swarms since 2019 were probably signs for a massive development we see now.
Mauna Ulu did episodic spectacular eruptions for seven months. Then it switched towards effusive lava flows without lava fountains. Our eruption now lasts around seven months like the first stage of Mauna Ulu. https://www.nps.gov/havo/learn/nature/mauna-ulu-eruption.htm
We’ll see whether our eruption follows Mauna Ulu’s pattern or stays episodic for longer time.
I think at least part of the Pahala swarm is the deep.section of a fault going all the way to the surface at Nali’ikakani point. The same fault bounds the region of Kilaueas south flank, east of that is Kilauea while west of that is Mauna Loa mostly.
But, the flare up in 2019 is way to coincidental to the 2018 events, and after 2022 its pretty obvious at least normal Mauna Loa eruptions have basically nothing to do with it and Mauna Loa isnt taking over, if anything it is more quiet now than most other times in the last 25 years. Kilauea by contrast basically refilled almost all the subsurface deformation of 2018 in a single year last year, which might amount to anywhere between 0.3 and 0.7 km3, and is now 1/3 of the way filling the physical caldera, with 0.1 km3 in 6 months.
There is on other thing though. The Pahala swarm has had flare ups, the first obvious one after 2018 was a year later in August? 2019, but there was an even bigger flare up in 2022 if I remember right. If these did represent magma surges starting, then which one arrived last year, was it the 2019 pulse or 2022? If we only just saw the 2019 pulse last year and it did all this, then actually no we havent seen the peak yet and that wont happen for 2 years. Or whatever number it is if I got the second year wrong.
If we get another surge though, double the rate it is today, then to be honest the caldera might be filled much faster than expected. 0.6 km3 to go but it could potentially even do that within a year. It sounds unbelievable but then so did everything else Kilauea has done in the last decade and we got to watch that all happen 🙂
If we have the onset of a great shield building eruption like Obversatory shield eruption, it is something completely different to anything we’ve experienced since 1800.
Maybe we historically underestimated the force of Hawaii to let Kilauea grow vertically. Since 1790 most eruptions let lava flow down, but didn’t rise the peak of Kilauea. A voluminous shield eruption like Obversatory shield is the way Kilauea growths towards a 3 km high peak one day.
I wouldn’t look too much at the past. Kilauea has a range of eruption types, durations and locations. Expect something new.
Yes it would be different, but Pu’u O’o would have filled the caldera if it was in Halemaumau too so not really unprecedented.
I think people do underestimate it though, I see a lot of stuff about it probably taking centuries to fill…
From HVO:
‘Data analysis has confirmed that lava fountains from episode 24 reached heights of approximately 1,200 feet (365 meters), which were slightly higher than episode 23, and a new record for the current eruption.’
What do we know about the transition from the Observatory Shield eruption to Ailā‘au eruption?
Old post is up! We republished a post on when the world was young and hot-headed. In a way, this was our oldest post – we have hesitated to go further back in time than this!
https://www.volcanocafe.org/time-for-komatiite-2/