Sapphire!

/ 03/08/2024

A sapphire (and diamond) ring, from Germany

Henrik taught us about what he called gemmology – the knowledge of gemstones, which he said was considered to be part of the geosciences and specifically a branch of mineralogy. It is also a subject closely related to volcanology. Many gemstones form deep underground, and rely on volcanoes to bring them to the surface and to our attention. Diamonds are a good example. They form in the mantle under tremendous pressure, and collect at the bottom of the deep cratons, the old thick cores of continents. There they sit, 150 or 200 km deep. But every now and then (quite rarely, sadly, it has not happened in living memory) a kimberlite eruption occurs. Those eruptions start that deep, move up to the surface on a wave of CO2 gas in a matter of days or even hours, and deliver their hoard of diamonds.

Most kimberlite eruptions fail in one of two ways. They may be too hot, and the diamonds perish on the way leaving us only their diamond-shaped carbon ashes. Or they may not make it all the way to the surface and leave a conduit underground, filled with diamonds (or with their ashes of the failed both ways). Erosion may later reveal these hidden treasures, and suddenly someone somewhere finds a diamond lying on the ground. People flock to these regions. This happened at the original Kimberley site – the location after which the kimberlite eruptions were named, and which is now only a deep hole in the ground. The Boer war was started about ownership of the Kimberley diamond mine. The ‘Boers’ (‘farmers’) had it and the English wanted it. Diamonds are hard won and can come with a deadly legacy. We still talk about ‘blood diamonds’ – although for some reason Kimberly itself is excluded from this term.

Once diamonds have been brought to the surface, they may not stay there. Rivers can move mountains – moving diamonds is peanuts, and the diamonds may end up well away from their source. Kimberley diamonds are now mainly mined on the Atlantic beaches, appropriately called the Skeleton Coast. But without volcanoes, they would not be there to be found, and many wedding rings and other jewelry would look very different. Your marriage may have a volcanic heritage.

There are other jewels. One of these is known as sapphire. Sapphire is a precious, deep-blue (although other colours exist) gemstone. The precious stones are not to be confused with the semi-precious ones: the only gemstones officially known as ‘precious’ are emerald (green), sapphire (blue), ruby (red) and diamond. (For some reason the Ring of Power was overlooked for this list, but then it did deceive.) The name ‘sapphire’ is apparently from a Greek-Persian origin. It may come from a word for lapis lazuli, the soft blue stone that was used in old paintings for the ultramarine used for painting deep-blue garments – if the painter could afford it! Lapis azuli is not in fact related to sapphire: it is a combination of different minerals.

Lapis azuli was so useful for paints because it is soft. Sapphire is not: it is only one step below diamond on the hardness scale. That scale is used to identify which mineral can scratch another mineral. It runs from 1 to 10. Diamond, a 10, can scratch sapphire since it is harder, but sapphire (9) will scratch almost any other natural material. (It is in fact the mineral which defines the ‘9’ class.) Obsidian was the volcanic tool of choice in the stone age, being able to cut through animal tissue with ease. But even this hardest volcanic material is only a ‘5’. (Don’t be fooled by the hardness: a hard mineral can still be brittle, and scratch resistant is not the same as chip resistant. A chip of the old block is still a chip.)

Looking at lists of material hardness, sapphire does not actually show up. That is because it is a variation of another mineral: corundum. Corundum is an aluminium-oxide (known as aluminum-oxide west of the Atlantic, thanks to the most spelling-prone dictionary in history, Websters). (An easier name is alumina). It exists in various forms: powders, fibres, cream or crystals. The first two are white and can be quite poisonous. The cream is a bit of a misnomer: I added it because corundum is a common ingredient of sunscreen. The crystal form of corundum is transparent in itself (as the word ‘crystal’ implies) and is poisonous mainly for your wallet. It is also hard: the name “corundum” probably is Sanskrit in origin, “kurunvinda”, meaning hard stone.

Corundum has a very high melting temperature, just over 2000 C. It is one of the minerals that can be found in space: it can form in the hot gas ejected by a star. In my night-time job, it is one of the things we study with JWST. (Don’t expect anything spectacular: these are mini-crystals less than a micrometer in size. We also find micro-diamonds in space – but we don’t advertise that. Anyway, buyer collects.)

The crystal form of corundum is rarely as transparent as the word implies. It can have quite a range of colours (or none). It comes as colourless, white, gray, blue, blue-green, green, violet, purple, orange, yellow, yellow-green, brown, golden amber, peachy pink, pink, black, and varieties of red. They are (artificially) divided into two categories. The red ones are called ruby, and anything else is called sapphire. (There is a bit of overlap around the colour pink with some would call a ruby and others a sapphire, depending mostly on whether you are the owner or the buyer.) Sapphires are known as blue gems, but that is not the case by definition!

Not all that glitters is gold (as Shakespeare states in the Merchant of Venice, followed by that shattering line ‘Gilded tombs do worms enfold‘). Henrik informed us that the famous Black Prince Ruby which at one time was mounted in the Tudor Crown and nowadays sits in Britain’s Imperial State Crown is not a ruby but a spinel. Spinel also contains aluminium and oxygen, but in addition contains magnesium which corundum does not – and spinel is not ‘precious’. It’s melting temperature is even higher than that of corundum, at 2300 C. Again, spinel is known in space and it has even been found in the worldwide iridium layer from the Chixulub impact. If you would like to test Henrik’s assertion, just do a scratch test on the Black Prince Ruby. Spinel is only hardness 8, so can be scratched by a sapphire. You can find the jewel in the Tower of London. You will probably spend the rest of your life there, but I am sure that is a sacrifice worth making in the name of Ruby Rebellion. (And that would make a good title for the next Doctor Who episode.) And it is not just Britain: the Cote de Bretagne Ruby of the French Crown Jewels is also a spinel. Whoever knew.

The Imperial Crown with the Black Prince ‘Ruby’ (red). The St. Edward’s Sapphire (blue) is a real sapphire – but the ruby is not a ruby

Why the range of colours? Corundum in itself is colourless. The colours come from impurities, where traces of other elements got themselves stuck inside the crystal structure. This is called ‘allochromatic’, where the colour comes from elements which are not a unusal part of the main crystal. (The opposite is called ‘idiochromatic’ and is the case for instance in lapis lazuli, which also contains aluminium and oxygen but combined with sulphur and either sodium or calcium; the intense blue colour comes from the sulphur which is included by default.) Rubies get their red colour from chromium which displaces some of the aluminium. Blue sapphires contain some titanium and iron. The iron or titanium changes the crystal structure such that it absorbs yellow. Seen under sunlight, the light that comes out now lacks the yellow, and it is this lack which causes the colour. If only iron is present, the sapphire appears green, or even green-yellow. Add the right amount of titanium, and it appears blue. The best gem-quality blue, requiring an ideal balance between iron and titanium is found in sapphire deposits from Kashmir. But be aware that sapphires are often treated in a heat bath to change their colour: before surrendering your fortune at the used-sapphire dealership, do make sure it has been independently tested so that you know what you buy. (Rubies are more resistant to heat-treatment.)

Sapphires can have actual uses beyond jewelry. Apple in particular is known for having used sapphire for the camera lenses in iphones. Expensive watches (remember those?) may have sapphire face coverings. Those sapphires are not coloured but are transparent (naturally) and are synthetic. No, you can’t take your iphone to the used-sapphire dealership to trade in the camera cover.

Growing sapphire

Sapphires are found in various locations across the world. Well known are Sri Lanka, Kashmir, Thailand, Myanmar, Australia and in the US, Montana. From this list, a volcanic connection is not obvious! None of these are in the top list for eruption tourism. But of course, almost any place on earth has a volcanic heritage if you go back far enough, just like any US president can seemingly trace their origin back to Ireland. Diamonds, where we know volcanoes have a crucial role, too are found in regions which nowadays are not particularly volcanic.

What do sapphires need in order to form? The recipe is simple: high pressure, high temperature, and aluminium. Aluminium can be found in sediments, as aluminium is not very solvable in water and has a tendency to end up in the muck at the bottom. The pressure and temperature are more of a problem: if you just take a rock, creating corundum (sapphire of ruby) within it may need a pressure and temperature only found below 30 km of depth. But add a bit fluid and chemistry can give a helping hand and allow them to form less deep, especially if a bit of heat remains available.

Mountains

Basically, you just pressure and heat applied to sedimentary layers. Mountain ranges sound like a good case, where the collision between two continents sweeps up the sediments of the ocean that was in between, and the mountains provide the depth and pressure. This is called the metamorphic deposits, where sapphires were formed by heat and pressure underneath mountains, at a depth of 10s of kilometers. Indeed, several of the main sapphire sites are related to the collision of India with Eurasia, with ages of roughly 40 million years. This has given us a sapphire belt running from Afghanistan to Vietnam, which includes the famous Kashmir sapphires.

It is not the only such sapphire belt. Another belt runs from Sri Lanka and southern India to Madagascar and Mozambique. If this seems a bit discontinuous (they are an ocean apart), it wasn’t in the past. Before the world fell apart (I remember it well), this was one continent, Gondwana, and all these sapphire (and ruby) deposits were adjacent. The sapphires here are dated to between 500 and 600 million years ago. This was the time when the east and west parts of Gondwana came together, in a collision which build mountains and made jewels.

Source: Giuliani et al., 2014; https://www.gia.edu/gems-gemology/winter-2019-geology-of-corundum-and-emerald-gem-deposits

There is a third such metamorphic region, but its origin is less clear because of its age. It is a rather shorter belt, is almost 3 billion years old and is located in southwest Greenland.

These metamorphic deposits are a major source of the world’s corundum. The mineral can form in this environment in three different ways. The first is directly in the metamorphic rock, purely through the pressure and temperatures on the original rock, similar to how slate forms out of mudstone. The second way is when the rock interacts with a fluid. For instance, the silica-poor rock may meet a rising fluid containing aluminium. ‘Fluid’ does not necessarily mean water and it may not even be a melt: solids too can flow under the right conditions. Granite intrusions do, and these are common below mountainous regions. The silica in the granite is absorbed by the rock, leaving the granite with a silica-poor composition that is perfect for the formation of corundum when it meets the aluminium-containing sediment layers. This ‘fluid method’ gives rise to thin layers of corundum, because the interaction happens only over small regions. The third way involves partial melting of the rock. Because silica melts at lower temperatures, it melts first, and the remaining rock becomes silica-poor and aluminium rich. It is now ready to form corundum. All three ways have contributed to the belts of metamorphic corundum.

Volcanoes

But there is another group of sapphires, not related to the three known belts of metamorphic deposits. These are found in many different places, along the Pacific from Tasmania and eastern Australia to eastern Russia, in France, Cameroon, and elsewhere. (Not in Canada: Canada has a Ruby mine near Ottawa, but this is a misnomer as it actually mined garnet.) This comes from the second route, a more volcanic one, and these are called the magmatic deposits.

The various sites have one thing in common: they are associated with magma, and specifically with a type called alkali-basalt. Basalt is poor in silica. Silica (SiO2) is important in continental crust, whilst basalt is sourced from the mantle.

Alkali-basalt is first of all a basalt: low silica, roughly in the range 40-50% of the mass of the rock. This makes the lava quite dark. (The silica fraction determines the colour of the lava: the higher the number the whiter the solidified lava.) The alkali part of the name means that it is relatively rich in sodium or potassium. This corresponds to the grey area of the basalt box in the picture. This is quite a common type which can be found in a variety of places, especially continental rifts and oceanic island arcs, but not in mid-oceanic rifts. It comes from relatively deep in the mantle, 150-200 km.

Alkali-basalts are associated with sapphires in Australia, China, Africa and Europe. Rubies are much rarer in these places. The sapphires have a range of colours from blue to yellow, but rarely pink or red (which would make it a ruby). These sapphires include more iron. Metamorphic corundum, on the other hand, tends to be poorer in iron.

The known deposits of magmatic sapphires date from the time after the dinosaurs: all the known older deposits are from ancient mountain belts.

A spiral time diagram of ruby (red) and sapphire (blue) deposits. Source: https://www.mdpi.com/2075-163X/10/7/597

But the sapphires cannot easily form in the magma itself. Instead, they are thought to be inclusions which are carried up with the magma. The origin of these inclusions themselves are still a matter of debate. One possibility is that a hot plume melts small amounts of the lithosphere, and that the corundum forms there. Other options are based around pre-existing sapphires, which are later carried up by the magma, much like the diamonds are already present at the bottom of the cratons before any volcanism occurs to carry them to the surface.

Transport

So we need not only make the sapphires, but also worry about the transport to the surface. In the case of mountains, this happens by removal of the overlying rock, i.e., erosion. After all, erosion, like faith, can move mountains. Mountains erode fast especially when still rising. Tens of kilometers of rock may disappear, enough to make the corundum appear.

Magmatic deposits take the hard route and bring it to the surface by hitching a ride with magma that is making the journey for you. If neither route is available, then the buried treasures remain hidden in the deep. As you read this, a royal (in potentia) sapphire might be sitting right below you. Better start digging!

The place where the gems first appear on the surface is called the primary location. They rarely stay put there for long. Water may move them from the original place to somewhere else. This is the secondary location, or the placer. If rivers are involved, then this secondary place may be far away on the flood plain. They may also just end up at the bottom of the slope, much closer to the primary location.

Source: https://www.gia.edu/gems-gemology/winter-2019-geology-of-corundum-and-emerald-gem-deposits

The Eifel

This brings me to the recent paper on this topic, involving sapphires in a thoroughly volcanic location. The paper was written by Sebastian Schmidt and others, and it appeared in May this year. It presented a study of some of the youngest natural sapphires known, in the heart of western Europe.

Europe has a surprising number of volcanic fields, separate from the well-known volcanics of Italy and Greece. They range from the Massif Central in France with its beautiful Puy’s and the field in the Western Carpathian Mountains between Poland and Slovakia to the domes and maars of the Eifel in Germany.

The Eifel region has been volcanically active over the past half a million years, including the VEI-6 eruption of the Laacher See. For a description of the region, see Gijs’ report in sleeping in our back garden the past present and future of the eifel volcanism part/. There are two main areas, the western and the eastern Eifel volcanic fields. They produce similar lava but magma in the western field is stored deeper (low crust, upper mantle) and erupts faster than the magma of the eastern field (mid crust, long storage time of 50,000 years). The Laacher See, source of a major eruption 12,000 years ago (and no, it is not the cause of the Younger Dryas), is in the eastern field.

Sapphires have been found in the volcanic deposits and in river sediment in the region. Don’t expect jewelry: they are not gem-quality and you may need a microscope to see them as they are less than 1mm in size. But they clearly are associated with the volcanism. And they exist across the region, including in the pyroclastics from the Laacher See eruption. We know in many cases even which eruption erupted the sapphires. And very important to scientists: these sapphires don’t have jewelry value, so it is possible (affordable, to be precise) to do the kind of analysis that scientists like, i.e., taking them apart and see what they are made of. It is called destructive analysis and it is not something we would be doing with the Black Prince Ruby. (Not for want of wanting, to be honest.) But to find out exactly which elements (and their isotopes) are in a gemstone, for instance to obtain its age from decay products of radioactivity, you’ll need to get its insides out. You can’t have your gem and date it. A date for a gem may not end well.

A number of sapphires were dated using uranium decay. All came from river beds in the eastern region, and cannot be traced to a particular eruption but the main eruption in this region is that of the Laacher see. Nine dates were obtained: all but one came out at around 60,000 years since they formed. The exception was clearly much older, and was dated to 25 million years, the sole grown-up in the crowd. The younger grains are about 50,000 years older than the Laacher eruption. However, they are clearly related to the volcanism here, and the age is about what would be expected from the age of the Laacher See eruption plus the storage time of magma in the magma reservoir. This is a clear argument for a volcanic origin of the sapphires. But they were not formed in the explosion itself: they were sitting around the magma chamber.

How about the single older grain? Is this the exception that makes or that breaks the rule? The river from which this grain was obtained contains some erosion from the ‘Siebengebirge’, a group of ‘seven hills’ (there are actually rather more) just just east of the Rhine which are quite a beautiful sight. These hills are volcanic, but they come from an earlier phase of volcanism which indeed occurred around 25 million years ago. So the grown-up also appears volcanic, just from an unintended volcano.

The scientists wanted to know more, so they kept crushing sapphires. Mass spectroscopy was used to identify the elements in the sapphires. (In fact, crushing is not enough for this. You need to ‘plasmafy’ them: turn the jewels into an ionized plasma.) (Please do ask permission from the owner first and do check they are not expecting their precious to ever be returned to them.)

Source: edited from https://link.springer.com/article/10.1007/s00410-024-02136-x#Sec11. These are sapphires in the volcanic ejecta. The different groups come from the different volcanoes in the region, each with slightly different iron abundance.

The results may seem a bit surprising for volcanic sapphires. The plot shows the amount of iron in the gems, versus the ratio between gallium and magnesium. Magmatic sapphires should be a bit higher in iron, but lower in magnesium. The iron abundance roughly fits, but the Ga/Mg ratio is what would be expected from metamorphic sediment – not magma. Gallium is lower in metamorphic rocks than in alkaline magma,and here the gallium is low. The sapphires formed in the magma, but apparently not from the magma! How is that possible? And why the large range in iron, which would indicate some involvement of the magma?

Other diagnostic diagrams can be made, and they all arrive at the same conclusion. The sapphires of the Eifel tend to the ‘metamorphic’ type, but some (especially those lower in the isotope 18O) are at the borderline between metamorphic and magmatic.

Digging for sapphires

What is the solution? Clearly, ‘magmatic’ sapphires still can have a range of origins. Sapphires found within volcanic rocks can be hitchhikers, picked up by the magma from the crust on the way. If the sapphires formed in the magma, it is likely to be from a syenite, which is a silica-poor magma formed from low degrees of melting: as this magma sits in the crust, the corundum would in time solidify (precipitate) as the magma cooled. If the sapphires formed in the continental crust, it will be a metamorphic (heated) rock (such as marble). Since in the Eifel, the age of the sapphires agree with that of the volcanism, the trigger must have been the same heat pulse that caused the volcanism. The same happened, apparently, in the nearby Siebengebirge, but much longer ago. But that doesn’t answer the question regarding how they formed.

The 18O fraction gives more information. Low 18O indicates an origin in the upper mantle or lower crust, while higher 18O indicates they formed in shallower crust. For this region, 18O values below 6 ‰ indicate an origin from crystallizing magma, while value above 10 ‰ represent the crust. Both are present here, and the sapphires with low 18O have composition of other elements which is closer (but not identical) to magmatic. On the other hand, although the sapphires occur in the pyroclastic flow, they are limited to inside solid ejected material, and are not found in the tephra which came from the liquid magma. Even though the Laacher See ejected various zones of its magma chamber with compositions ranging from relatively pure to quite evolved, none of those zones contained sapphires.

The easiest way to explain this dichotomy is if the sapphires formed on the edges of a cooling magma chamber. The lack of sapphires in the tephra suggest that they were not present in the liquid magma, so they must have been in the magma that had already solidified. If the magma chamber had partly melted a bit of the surrounding crust, then the interaction region would contain some magma with a bit of molten crust mixed in, but also some deriving only from a partial melt of that crust. If that crust is sedimentary with a bit of aluminium, the conditions would be there for the formation of our precious (albeit minute) stones, with some coming from the crust (high 18O) and some from the magma (low 18O). The intermediate values would come from sapphires that formed where the magma and molten crust mixed. Indeed, a few grains are known where the core of the grain has a composition consistent with crust, but the rim has a composition closer to that of a magma/crust mix. Clearly, this rim was added later, when the sapphire was within the magma. The fact that the difference between core and rim survived, without the 18O migrating between the core and rim, requires that the grain was kept at temperatures of no more than 700 C, i.e. in the cooler margins of the magma chamber.

The conclusion of the team is summarized in the following picture. The numbers refer to the 18O fractions in the various locations.

One study does not resolve major scientific controversies. But it is a brick in the wall, or at least a gem in the collection. In this location in the Eifel, the sapphires came not from the deep mantle, but formed in the crust in locations where the magma heated the local crust. Just like cultures can become much richer when interacting with (and challenged by) other cultures, so the magma picked up treasures and formed new ones. Precious stones formed by fire.

Albert Zijlstra, August 2024

The Hendrik original

Sebastian Schmidt et al: Petrologically controlled oxygen isotopic classification of cogenetic magmatic and metamorphic sapphire from Quaternary volcanic fields in the Eifel, Germany. Contributions to Mineralogy and Petrology, 195, 55 (2024). https://link.springer.com/article/10.1007/s00410-024-02136-x

A classification of gem corundum deposits aimed towards gem exploration.
https://www.sciencedirect.com/science/article/abs/pii/S0169136808000280

https://www.gia.edu/gems-gemology/winter-2019-geology-of-corundum-and-emerald-gem-deposits#part-i:-ruby-and-sapphire

Ruby Deposits: A Review and Geological Classification. https://www.mdpi.com/2075-163X/10/7/597

Berlin’s sapphire – the blurb says “the building is a sturdy, multi-sided thing, gleaming as it emerges from the cityscape. The colour of the ceramic covering even evokes the azure blue that corundum gives the precious stone itself.” But it is not volcanic.

Related posts

372 thoughts on “Sapphire!”

    • Quakes next to Grindavik, but not along the rift. I remember these quakes in this location before the last eruption, although there have been quakes here otherwise anyway so it isnt very reliable. But it probably is pressure increasing, I think this week will be the one.

      • This area is under a lot of stress. Crustal bending from the inflating sill directly north and it’s in the trigger quake zone from the dyke to the NE. There were lots of shallow quakes in this area also before the last eruption, but I don’t think the active part of the dyke stretches this far south.

        All eruptions have started with intense swarms near 63.881N 22.389W. I don’t expect the next one to be any different. There’s a small area void of quakes just south of this point. It splits the earthquake trace of the dyke in two segments and I suspect that’s where the feeder from the sill is located.

        • It ties in with vectors from the relative inflation too, just south of Sylingarfell, suggesting the same general site for the feeder.
          I think pressure is building more widely too, across Reykjanes, which also fits with earlier episodes.

        • Yes I meant to say these are stress quakes, not on the dike itself.

    • This may (in bad case) indicate an intrusion towards the SW like the January eruption. But it still doesn’t has to erupt there. The sill intruded there in November without an eruption. Is it expanding now due to rising magmatic pressure?

        • On that same note, I have studied the start of the last eruption and saw an abrupt rise in venting, sometimes in areas that haven’t seen any venting at all. The venting would get intense enough that it ‘outpaces’ the rest, minutes before an eruption. I think once an intense swarm starts, we should start looking for that.
          https://youtu.be/N3QXxqTHnIU?si=E2zoU-xuQ9tqkcRf

      • Ljosufjoll with another heavy (albeit deep) quake.

        Sundhnukar doesn’t look on the verge just yet, not really a high density of quakes over the past day or two, this is still in build up stage. Maybe 5/6 days away.

      • How did the earthquake swarm during the hours before the last eruption look like?

        • I’ve checked May 29th on https://skjalftalisa.vedur.is/#/page/map
          The first 0.x earthquakes began around 10:25 with three quakes between 10:00 and 10:30.
          10:30 to 11:00 the number of earthquakes increased to 30
          11:00 to 11:30 they counted 38
          11:30 to 12:00 40
          12:00 to 12:30 41
          12:30 to 13:00 53 (12:46 the eruption began)
          13:00 to 13:30 32

          With the first quakes around 10:25 to 12:46 it took 140 minutes from first innocent signs to the eruption. The warning before the eruption was published 11:40. Were the earthquakes between 10:30 and 11:00 enough to be cautious? How do/did they distinguish between a pre-eruptive swarm and a normal one?

          • Pretty sure they use real-time GPS data along with the earthquakes to tell when there’s an intrusion happening.

          • I was looking for those tremor charts for a while, then I recalled you can get them by clicking a Seismomenter (not enabled by default) on vafri.is.

            GRV has these bumps in the past days … construction?

          • Yes, on some webcams you can see that they are working on raising the berm between Hagafell and Sylingarfell.

  2. Just now I went to check the NASA Firms satellite and thermal anomalies show up again at Home Reef, Tonga-Kermadec arc, and with Iwo Jima I did create a movie of thermal activity for the month of May at Iwo Jima, it is rather startling to see the satellite flag so many hot spots in that month.

    I wish that there was someway to get some critical analysis of the NASA Firms data regarding Iwo JIma the whole past year because this satellite data might be telling us something important about the Iwo Jima volcano.

    • Its interesting this is nowhere near the recent vent, which isnt visible either. But theres nothing at the heat source that I am aware of or there would probably be a video of it.

      Maybe a simple answer but it might just be hot black sand? There is probably hydrothermal heating already so being baked in the sun might get things pretty high. It does seem to be only on the beach which would support this although not disprove other causes.

      • There are explosion craters in this region, and there has beeb recent activity, so this may be something

    • There haven’t been any press releases concerning this but Iwo-Jima has been having frequent small eruptions all over the island since 2023. Particularly in NW side where the fastest inflation is.

  3. News from RÚV:

    Eruption still expected as 500 quakes hit in one week (12 Aug)

    The number of small earthquakes in the Sundhnúks crater row is still increasing. All of these are below magnitude two and the majority of them below magnitude one.

    The tremors are scattered around the magma tunnel, south to Grindavík and north to Stóra-Skógfell.

    Minney Sigurðardóttir of the Icelandic Meteorological Office, says that currently there is no volcanic eruption to be seen. The situation is the same as it has been for the past few days.
    …
    “We’re expecting an eruption in the next few days,” Minney says.

    So worth watching the cams! But volcanoes have a bad habit of not doing what people expect…

    • And there is nice swarm on the peninsula. But in the wrong place

      • Reykjanestá (spoken as “Reykjanestå”?) is doing an offshore earthquake swarm now together with Eldey ridge. Maybe the Reykjanes swarm is an earthquake swarm like the Thorbjörn area had some years before. Does there have to be a sill formation like in November under Grindavik before eruptions can occur?

        • The icelandic letter á is a diphthong pronounced like the vowel sound in the english word “our”.

          Fagradalsfjall did not erupt from a shallow sill, so I’ll answer that question with a no, there does not have to be a sill formation before eruptions can occur.

          • So spelled phonetic to English, “Raykyanestah” ?

            I guess, depending on which English accent, that wont necessarily help much.

          • From what Tomas was saying, probably more like Raykyanestaw.
            I tend to try to pronounce Icelandic like Norwegian, but it isn’t always correct with some unexpected combinations and pronunciations.

          • If its a word not in English I just say it the way it is originally. Australian accent really doesnt work in any language except English. I get the impression that most people who learn English only after developing their native accent seem to naturally have a rotic pronunciation (‘american accent’). In Australian an R is only pronounced if it follows a vowel, and sounds like an American R. Rolled or tapped R in Japanese ir Spanish (or Icelandic 🙂 ) is actually a D or L sound to me naturally, although now I hear it as an R anyway.

            As for fow you spelled it Richard, to me that spells like ‘Raykyanestor’… I think it is by far easiest to just watch a video that has pronunciation. Back in 2021 was a video that did this for the volcanoes relevant now.

            https://youtu.be/fDll0rBRhDM?si=xeIvfRR96OrLmyea

            “Thraouenskyeldturhroyin”
            “Fagradahllsfyaktl”
            “Soondhngookurgigahrodth”

            🙂

          • Since English spelling is not phonetically consistent, it’s not easy to do phonetic spelling of words in other languages. To get it correct one should really resort to the International Phonetic Alphabet, IPA. The Icelandic word tá means toe and you can find the IPA representation here:
            https://en.wiktionary.org/wiki/t%C3%A1#Pronunciation_3

          • The fissure swarms of Reykjanestá and Svartsengi are not far away. Maybe the distribution of magma from Fagradalsfjall’s deep system to Svartsengi slowly spills over towards the west.

            The map layers with lavas show that both Reykjanestá and Svartsengi consist each of two seperate fissure lines. Svartsengi has I Eldvörp and II Grindavik/Sundhnukur. Reykjanestá has I Stampahraun and II Skálafell. During the Medieval Fires both I Eldvörp and Stampahraun erupted, while both II systems were quiet. This time it might be the opposite development. The present earthquake swarm of Reykjanestá points towards the II line of Skálafell.

          • Not only is English consistent generally, it also varies regionally, especially when dialects come into it. In my dialect R is pronounced very strongly, even in the middle or end of words.
            IPA is the way to go, but nativevEnglish speakers aren’t very aware of how tge combinations are pronounced in IPA.
            I did learn Norwegian, which is why I tend to use it as a guide. Icelandic is basically Old Norse, but modern Norwegian has changed to the point where Norwegians generally no longer understand Icelandic, even though they are language pairs, just like Fries and English are.

          • Icelandic is in fact to Norwegian/Danish/Swedish what Old English (Anglo-Saxon) is for English. With knowledge of one scandinavian language you can understand relatively well whole sentences in the other two languages. I also have learned Norwegian, but only understand single words in Icelandic sentences, not the whole sentence. Also the spelling of words and vocals is different to Norwegian/Swedish. That’s why I was uncertain about the spelling of the à and á.

            One difficulty in Icelandic is that they hate loanwords and instead use native expressions. This applies much to scientific loanwords in volcano news and makes them impossible to understand

          • Fries is very similar to Old English or Anglo-Saxon. It is where Middle and Modern English has developed from, just like Icelandic, Faroese and Norwegian developed from Old Norse. However, just like English has been greatly influenced and changed, so has Norwegian. English is a borrowing language, plus it has been heavily influenced by French (due to the Normans), white Norwegian has a heavy Danish influence and to a lesser degree, Swedish. When I was loving on Norway, I could understand written Danish as easily as Norwegian, even though the pronunciation is quite different. I could understand some verbal Danish and Swedish.

          • There was a lot of trading around the north sea coast. Frisia was used as name for the coastal population from northern France to Denmark, but it was not a uniform ethnic group. Rather than English developing from Fries, it is probably better to say that they influenced each other to form a language used for trading. There was close contact, with even part of the Beowulf story taking place in Frisia. There are also similarities in how runes developed in England and in Frisia, but they were not identical

          • A funny thing with Old Norse is that the written language is almost identical to Icelandic, but pronunciation has changed. How was á pronounced in Old Norse? Well, back then it was more like the modern day Swedish å. At least that’s what linguists studying Old Norse reconstructed pronunciation think. I’ve seen videos of people speaking reconstructed Old Norse and in my ears it sounds a lot like a Swedish person trying to speak Icelandic from written text 🙂

          • English still has some words in common with Dutch and Lower German dialects (f.e. “water”). It started in fact as a northern Lower German dialect and shifted in Britain gradually towards a new language. The biggest disruption was the influence of the Norman romance language which made English a half germanic and half roman language.

            The Scandinavian languages were heavily changed by the Hanseatic League. The Merchants introduced many Lower German nouns (mainly for business topics) into Swedish/Danish/Norwegian. The Scandinavian languages preserved a lot of written Lower German that got lost afterwards with the victory of High German (based on dialects in Martin Luther’s region).

          • I think I’ m going back a bit earlier in terms of Old English. Old English is probably pre-Norman, around the 8-10th centuries, when English was still largely the same language spoken by those who settled in the 5th and 6th centuries, the Jutes, Angles, Saxons and Frisians. I can’t say for sure, but it was probably Frisians from the more northerly islands, from modern day Netherlands to Denmark who settled in what is now East Anglia. That would make more sense linguistically and probably geographically.

    • At least White Island isnt a caldera. Maybe it could be one day, but presently it is just a stratovolcano. Apparently it has alternating periods of growth and we are between that now, when it grows it is mostly effusive. But given that it has a wide breached crater there seem to be occasional exceptions to that…

  5. Magma is definitely slowly moving down the ERZ of Kilauea, its past Pu’u O’o now. Probably why the POC tiltmeter is showing a declining tilt, it is now above the advancing magma not in front of it at least not completely. Im still predicting that around the end of the month to 1st week of September we will see a similar signal at JOKA, either that or more obvious uplift at Pu’u O’o area if magma cant push further east. Might be a month or two before we get an intrusion again somewhere, but this is the final part to fill from 2018 abd it already filled a bit before 2020, so I give it a year max before Pele is back to her old self. I dont think another lower ERZ eruption is immediate, but I also think at this point it is pretty foolish to think it will be another 50 years to the next one too…

    • Magnitude 3+ quake where the ERZ connector crosses the outermost caldera fault near Puhimau. There might be a surge of magma into the ERZ, will be interesting to see how the rest of it behaved.

    • The first eruption at Napau in September 1961 was preceded by summit eruptions (the last one in July). Also Pu’u O’o was preceded by a summit eruption in September 1982. Does the MERZ need the pressure by an ending summit eruption to do the first eruption? LERZ eruptions happen very often after the end of summit eruptions. Maybe the MERZ does it in weaker form with more distance to the preceding summit eruption. If so, we might expect a summit eruption first and then the onset of next MERZ fires.

      Recent examples Summit-LERZ “pair eruptions”: 1954-1955, 1959-1960, 2018
      Summit-MERZ “pairs”: 1961 (mentioned above), 1967/1968 summit – 1969 Mauna Ulu. 1971 summit -1972 Mauna Ulu. 1982 summit – 1983 Pu’u O’o.

      • There already was a series of summit activity that ended less than a year ago. The summit flank sequence is not very reliable, the magma erupts in the spot it is easiest to. In the mid 20th century Kilauea was fresh out of a quiet period that also followed (maybe coincidentally) a major drain. The summit was probably the easy option for a bit but even then, the ERZ was active all the way down within a few years. I think to be honest the 1961 summit eruptions were entirely because Halemaumau had a minor collapse in 1960 and probably had weak points to explout, as soon as it filled even a short way it was abandoned.

        The same thing on a way bigger scale happened since 2018, the summit collapsed so far the bottom was as low elevation as the lower ERZ, gravity favored an eruption there so it did for 3 years, and as soon as the floor got to over 950 meters elevation it was abandoned in favor of the SWRZ, and now the ERZ too. We can look at recebt behavior and it is pretty obvious. In late 2022 the stable summit eruption of over a year suddenly failed as a shallow intrusion and died slowly over the next weeks, just long enough to see Mauna Loa wake up but it probably was a complete coincidence. The lava was around 900 meters elevation. Since then there has been 3 eruptions from vents basically all over the 2018 caldera, with two of them even outside the degassed lake, and all of them failed abruptly after a few days. Then magma rushes into the SWRZ and fills it up to pre-2018 levels within 9 months, and causing two intrusions with one erupting at around 1000 meters elevation and also very short. Then it has another intrusion in the upper ERZ also at 1000 meters elevation, which fails. Now magma is flowing into the middle ERZ, and the rest of the volcano is now unable to pressurize enough to even quake much.

        I think tbe summit eruptions are over, we might get another brief one, but that could also just as easily be an eruption on the upper SWRZ ir upper ERZ again. But the most likely place now is the middle ERZ, probably a bit east of Pu’u O’o where the elevation is lower. If the summit was the easiest place to erupt then it probably wouldnt have stopped in 2022, and still be going now. But there are other areas to fill.

        It should be mentioned the middle ERZ is filling extremely fast. In only a few weeks MMAU has been uplifted about 1/3 of the total distance it subsided since the summit started refilling in 2020, and about the same elevation it was at a year ago. A year recovered in 2 weeks… The same is true at NUPM and MKAI north and south respectively. And KERZ station is also recovered to a year ago when it came online. Pu’u O’o is just starting to show it in the GPS although it is noisy so less obvious yet. But still at this rate the whole middle ERZ down to Heiheiahulu and the western 1955 vents will be full of magma by late September and probably erupting somewhere within a year. Its all a learning game really still. But so far the last few weeks is very similar to November 1982 about 2 months before Pu’u O’o started…

        • 2018 we observed how the draining of magma at the summit can increase magmatic pressure in the low east rift system and lead to an eruption. I imagine that the magma column below an erupting crater/caldera has a lot of weight inside like water in a vertical pipe. If the upward force to erupt ends, the mass falls back in the deeper magmatic system and increases pressure on magma there to move towards the rift system.

          • 1955 and 1960 eruptions better fit that sequenxe I think. Both of those started with summit activity then lower ERZ activity without an intermediate. 1955 took either 1 or 3 years, depending on where you stsrt. 1960 was only a few weeks, the magma supply in late 1959 was enormous, probably when Kilauea really became dominant.

            2018 was a lower ERZ eruption but it started from another ERZ eruption instead. I posted it a while back but the full tiltmeter record at UWE from 1956 to 2023 is available, and from 1975 to 2007 the summit was generally subsiding. Obviously, that was because there was a way out lower down. The inflation starting in 2007 first caused a new vebt east of Pu’u O’o with high flow rate, and the 2008 vent formed soon after. When inflation resumed in 2010 is when the summit vent went from a deep glowing pit to the lava lake. All of that entire time magma was filling the ERZ. 2007 might have been the time it actually filled, and was backing up. But the magma supply also went up too, as high as 0.4 km3 a year based on SO2. It is probably a similar value now based on deformation recently,but without an open vent SO2 is near 0 so it is hard to tell

            But basically 2018 started from overpressure in the ERZ itself. 1955 and 1960 started from summit overpressure. This isnt the same as those two eruptions actually intruding directly from the summit, they started locally but the pressure wasnt there.

            I think we cant really rely on the historical record though. Kilauea now is different to in the 60s and 70s. Even 80s to 2000s, up to then magma was variable chemistry, after 2012 it has been very homogeneous typically, 6.7% MgO, outside of early and some late 2018 lava that was <4% MgO and 9% MgO respectively. Still true now. 2012 is also when the ERZ seemed to start declining and the summit lake became higher and the dominant SO2 source.

    • Is there any significance to the higher level of negative earthquakes? There seem to be quite a few more compared to the last three eruptions.

      • Negative M are really tiny and hard to measure at the best of times. They get masked by bigger EQs/swarms/weather etc.

      • Scott:
        I think that the negative quakes (a valid reading) are subject to the inclement weather in Iceland. When the wind is blowing, the valid negative quakes become hard to distinguish, thus less of them would be recorded in the database. My best guess is that there is not too much significance in this.

    • I also noticed yesterday that the alert level for Reykjanesta has been raised to yellow, following the swarm.

    • It’s within the ‘rift zone’ but it’s actually a fair bit away from Ljosufjoll itself, around 20km. Be interesting to see if this is a reactivation. Albert did mention about some Hreppar plate reorganisation & magmatism around the edge of the ‘micropate’ coinciding with Reykjanes cycles.

  8. Checking back to Mount Ruang volcano, which blasted off last April 16th, it looks like the island is starting to slowly recover Picture taken by the Sentinel 2 satellites, July 31st.

    • I hope they do some seafloor mapping soon, the eruptive center in the middle of the sea at the same time as Ruang was popping off was quite mysterious.

      • Andy:
        I agree. There was anomalous activity and booming sounds which were NOT coming from Ruang Island, but seemed to indicate some submarine volcanic activity to the west of the island.

  10. Nice tremor going at the Grindavík/þorbjörn graphs other Raspberry Shake stations stream (began ~10 minutes ago).

    • 18:11- large tremor, a minute later, a larger tremor there.

    • If you want to see what the onset of an eruption looks like in the raspberry shake, you can go to:

      https://dataview.raspberryshake.org/#/AM/R252C/00/EHZ

      Then click the 24h view (sine wave icon in the top right corner), then select May 29. Now you can click anywhere in the drumplot to zoom in on a certain moment in time (use +- buttons to change the time span, the youtube channel shows 10 minutes of data), then click on the button with 3×3 small squares to enable the spectrogram. Check what it looks like before, during and after the swarm. The eruption started at 12:46.

      The eruption tremor really only shows up as an increase in power in the lowest frequencies. You might want to turn off the auto scaling to see the difference before and after. The actual onset is preceded by an absolute bombardment of quakes that show up as vertical lines in the spectrogram. This is the number one thing to look out for. If you see that forest of quakes in the spectrogram, then it’s time to get excited and sound the alarm.

    • Interesting at the end, I think the NEC became active. There was a strong glow within it after fountaining stopped. Either there was a vent in there too or the new Voragine cone is tall enough it has started flowing into the NEC. Considering only 40 years ago the NEC was the summit of Etna that really puts things into perspective for how much the summit has grown recently.

      Considering too, that apart from in 2001 and 2002 there has been no true flank eruptions of any respectable size, and those two eruptions were basically summit paroxysms in a different spot, Etna seems to have begun a new stage of stratovolcano growth. It is a gentle wide mountain with 4 craters now but at this rate it could be a 4 km tall perfect cone in a few hundred years time. Certainly extraordinary what is going on there now.

    • Etna does the paroxysms very suddenly. Is the rise of magma faster there than on hotspot volcanoes like Iceland and Hawaii? How much warning signs does Etna give before a paroxysm?

      • It’s a gas driven, open-vent volcano. The magma is always there, just beneath the surface. As pressure rises it starts to erupt out, then bubble growth feedback kicks in and blows like a geyser. I’m not sure what the warning is, I don’t think deformation or fracturing earthquakes provide any warning. Tremor and LP earthquake/strombolian activity seem to be the best pressure, but it only works shortly before the paroxysm starts.

        • This situation reminds me to Stromboli and similar volcanoes. I remember that sometimes Etna first did some mild Strombolian activity before the serious Paroxysm began. How/when do the Italian experts know that a Paroxysm is coming?

          • How/when do the Italian experts know that a Paroxysm is coming?

            I don’t think they do. No one does. Sometimes you can guess when they are happening regularly, and you can often guess a few hours beforehand, but they can be unpredictable until very shortly before eruption.

          • Did the last big eruptions of 1992, 2001 and 2002 start as suddenly as the recent Paroxysms or did they give more warning signals?

  12. Looking at Ioto on Google Maps satellite view there is a strikingly greenish pool of water on the beach just adjacent to most recent eruption site.

    Some of the seawater also appears to be discolored with a similar hue. Very interesting!

    • Probably a fire, an eruption there would be noticed and probably very fast and intense, the summit lake would probably sink, and founder revealing a 2 km wide incandescent surface, probably bubbling violently. A lava lake in a crater near Mauna Ulu drained in 1969 just after an eruption and the glow of the draining was so intense HVO thought high fountaining had resumed until they saw the real cause. Kilauea often surprise erupts during periods of unrest but when it starts everyone knows 🙂

      • Chad:
        I was wondering if the reading simply was indicating a hot spot, not a fire.

  14. Where’s the damn Sundhnukur eruption?

    Shouldn’t it have started half an hour to an hour after the jump in quakes and onset of tremor?

    • The previous events were preceded by many more quakes around 2+ just before they happened. There gave only been minor swarms of quakes below around 1.5 and most below 1 so far.

    • Surely you mean ‘dammed eruption’? Iceland has been very busy damming it in. Sundhnukur may have given up – nowhere for the lava to go!

      • Albert, the longer this next fissure eruption #9 waits, the more my concern that a very large voluminous flow will occur during the first hour after the lava breaks the surface. I admit being concerned that a gusher of lava will just sweep right into Grindavik.

    • Same as literally every other time, if it is about to erupt it will be obvious. We have seen it enough times now at both Kilauea and Mauna Loa, and now heaps of times exactly in the relevant location. Dikes intruding are very noisy but also very fast, there will be no doubt at all what is happening. You can see it really obviously now at Kilauea how different slow magma movement looks compared to a dike. Right now magma is flowing slowly down the ERZ, if there was an open vent it might be silent. But when a dike started a few weeks back it was 1000 quakes in a day and hundreds an hour at one point.

      Like, there have been 5 of these now, we know what it looks like.

    • I think the signs are, as Tomas Anderson linked to the Raspberry Shakestations site (thank you), are the the tremors have to at least be 1e-4 or greater ina secession in a time span of 10 minutes.
      Stations I used:
      R252C – Grindavík
      R5D8E – Þorbjörn
      RBDCF – Vogar
      (Note: all have to be EHZ to see it).
      There has been a few of such spikes I have seen, but not rapid or continuous enough.

    • The time between the beginnings of Sundhnukur episodes has since March expanded by +1 month with each episode.
      February to March 1 month
      March to May 2 months
      May to August 3 months
      August to December (Christmas?) 4 months

      If this tendency continues, 2025 may only get 1-2 eruptions at Sundhnukur. But they slowly grow towards more serious and dangerous eruptions. Will they align to Grimsvötn’s minor eruptions’ size (f.e. 1983)?

      • The eruptions now are probably already as big as standard Grimsvotn eruptions. Those are given as tephra volumes, which is about 3x the volume of lava rock. 2011 was 0.3 km3 DRE, about 10x that of recent Sundhnjukur eruptions, but it also took about 1.5-2 days to do that so was probably a similar intensity. Other eruptions are much smaller, a borderline VEI 4 is about 30 million m3 DRE, which is only slightly bigger than the curtain of fire stage at Sundhnjukur the last two times. Most Grimsvotn eruptions are below a 4 anyway, so are likely smaller DRE.

        Really, out of all the famous Icelandic central volcanoes only Katla and Hekla routinely do big eruptions at the central volcano itself. Vatnajokull volcanoes do small eruptions, usually, and rare huge rift eruptions. Typical rift is 1-3 or 4 km3 maybe every century or 2, scaling with size. Much larger are many millennia apart and probably not caused by shallow draining but magma accumulating in the crust under the shallow active system. Like is happening in Hawaii, at Pahala…….

        • The point about summit eruptions seems important. The Vatnajokul rift volcanoes have extensive rifts which drain magma from the summit reservoir (or from a common reservoir deeper below the summit. They also ‘talk’ to each other, or at least are aware what the others are doing, so there is a connection deep in the crust which transfers pressure. Oraefajokul has no rift and does summit eruptions. Hekla and Katla have limited rifts. Katla has one rift going a long way northeast and a short (inactive) one going south. It does both summit and rift eruptions. And Hekla is very young and hasn’t decided yet what it wants to be. It does summit eruptions and short rifts.

        • The average eruption of Grimsvötn is around VEI2 like 1983 with only a local tephra cover on Vatnajökull. VEI2 is 1 million to 10 million cubic km. A third of this is 300,000 to 3 million m³ DRE (or 0,0003 to 0,003 km³).

          Usually the average dormant time on Iceland was five years. During the Reykjanes Fires, this may be the aim for eruptions, if they keep going on long enough. Svartsengi eruption history mentions fewer eruptions than we’ve had until now. Maybe during Medieval time they united several episodes in one eruption.

          • I feel like its not unlikely the Middle Ages eruptions were significantly underreported, or the records were lost, or misinterpreted. Im not sure why that isnt considered more actually, even today most things are unrecorded, most of the Medieval eruptions were probably just glow in the distance or ashfall, and poorly located.

            Either that or there is a big difference in the amount of available magma this time. I think the fact that this whole cycle began with eruptions of exotic magma in a place that hasnt erupted since the late Pleistocene, probably was a more important detail than was appreciated for a long time.

  15. I think it is very interesting that the NASA Firms satellite does give us indications of submarine volcano eruptions. Apparently today one such eruption occurred at the Kyosei Seamount at 08:54 UTC 2024-Aug-14 Wed today. See .

  16. There were just 3 mag 3 quakes at Kilauea, upper ERZ connector where it crosses the outer caldera fault at Puhimau. Not a new intrusion but it does seem like magma is pushing that way pretty seriously now. I imagine some point we will see either increased uplift in the middle ERZ or magma will push further east faster. Or maybe both if supply is high.

    Worth noting the summit tilt is as high as it was to cause the Pauahi intrusion that started all of this activity. Only 3 weeks and magma is already inflating the middle ERZ 10× the rate it was subsiding before, and is getting past Pu’u O’o.

    • Is a repeat of Leilani likely? That end of the rift zone may still be hot & ductile enough to allow magma

      • Not likely, I guess not impossible. But not something anywhere near as big. 2018 was so big because of the whole ERZ being full, and a decade of the summit building up. I think to get something bigger again would need a similar setup. Although, its not impossible the early long fissure stage of a rift might have some vents in the lower ERZ, even if the central voluminous eruptions are much west. Like how 1961 was nearly in the lower ERZ byt the central vent of all of that sequence ended up being at Mauna Ulu, 20+ km west of there.

        But yeah another big lower ERZ eruption isnt my likely pick soon, but 10-20 years from now is a different story.

  17. How shallow has an intrusion to be to do kinds of phreatic/phreatomagmatic/hydrothermal/thermal/degassing volcanism on the surface? Eruptions like this can start without the rise of magma/lava until the surface. If a dike gets to the surface we usually get a lava eruption. But I imagine that a shallow dike that doesn’t do an effusive eruption, can still cause other types of volcanic activity.

  18. While we wait, it’s worth taking a look a this video from Just Icelandic. Some nice shots of the berms around the power plant.

  19. Sudden seismic noise at Kilauea, visible at RIMD and nearby seismometers

    • This station on Pu’u O’o’s lava field is remarkable with a strong S and E movement:
      ?fileTS=1723677142

      • South flank, definitely magma pushing that way. Although its only 1.5 cm so probably not the actual flank slipping much. Eastwards movement also means the deformation is strongest west of here, which is what HVO says, that the center of deformation remains west of Pu’u O’o. But that is the center, the leading edge of the magma does seem to be under or maybe even a little east of Pu’u O’o now. And it would make dense that inflation is strongest under the craters, they are craters for a reason, there is probably a magma chamber there. Both craters are right at the defined size transition of a caldera and a pit crater. Its very possible USGS made that boundary specifically based on Makaopuhi, to avoid having to call it a caldera.

        East of here the magma system is continuous but not as open, probably a sill complex above crystal mush cumulate. Pu’u O’o started as a dike from Napau but it probably lasted so long by getting fed from below by this complex. Twice after 1983 Napau and Makaopuhi had eruptions but neither was atypical, and more importantly Pu’u O’o drained out and then resumed after. Same thing may have even happened in 2018 if the south flank didnt slide and open up the whole ERZ.
        Maybe, in a really long way, we are about to see Pu’u O’o resume, although not likely at the actual original, at a new place somewhere else but with the same style. As Pu’u O’o did from Mauna Ulu. And Mauna Ulu possibly did from Kilauea Iki, which was an open lava geyser and persistent too, but never got to evolve into a shield.
        I guess this may just be what Kilauea does until its supply slows. Keeps building lava shields wherever is easiest, separated by sudden fast events.

  20. GeoNet has just posted a new White Island activity update. Still minor eruptive activity, but their aerial photo of it is just perfect. So I thought I’d put it up.

    • White Island is able to produce lava flows. Somewhere I’ve read that this is a possible scenario. But lava flows there would be very viscous and blocky Andesitic-Dacitic lava flows. Maybe something between blocky lava flow and dome?

      • I think I have read the same source, it is a peleean dome volcano, like Merapi but mostly underwater. Im not sure to what degree that presents a danger of a big eruption but based on other shallow water silicic eruptions im inclined to believe they are a lot less prone to water interaction. Home Reef and Lateiki, in Tonga, are two active and erupting dacitic stratovolcanoes making new islands by a lava dome directly extruding into the ocean. Compared to nearby Hunga Tonga Hunga Ha’apai, that is mafic and always had powerful surtseyan eruptions, and we all know what it did eventually…

        Not to say this is reliable or phreatomagmatic eruptions cant hapoen with silicic magma but but seems to be much more pronounced in fluid lava eruptions.

        • Last effusive (lava) eruptions of White Island were before 1826 (begin of British history record). I’d expect that effusive eruptions will be small, maybe a dome or short viscous lava flow, beautiful but harmless. But steam/phreatic explosions are a risk.

    • That is a very good photograph.

      I visited New Zealand in 1998, and remember seeing White Island smoking as we were flying in.

  22. The LFI feed is being very glitchy today. Buffering intermittently and pausing then lurching forward. Even if I set it to well behind live, it will every few seconds skip forward by as much as 10 seconds for no evident reason, and will soon catch back up to live.

    The MBL feed is behaving normally, at the same exact time.

    What would cause this behavior? Both are YT live streams, being viewed in the same browser with the same configuration and the same versions of everything at my end … but they are acting very differently to each other.

    • There was a discussion about it just before 🙂

      That video also does confirm something I was wondering about, that the eruption was not a singular point but a fissure that went towards and probably into the Northeast Crater.

      We have seen how the Southeast Crater grew another cone in 2011, and then after 2017 the two merged, making a single large SEC cone, that was of course very active in 2021. Seems the same thing is happening at the rest of the summit, the former central crater, becoming a single crater again.

      Will be interesting if the SEC becomes active again soon. It seems to be much more divergent at greater depth than the other 3 craters.

      • Isn’t it strange–upon watching an Etna paroxysm at night, you wish you could see it during the day–and vice versa?

  25. Fascinating article Albert, as always!

    In France, a deposit of magmatic sapphires has been found in a riverbed a few years ago. The river flows from the Monts Dore massif, an old stratovolcano a bit south from the Puys. The deposit has been the source of a legal conflict between two neighbors whose lands are separated by the river, as they each own half of the river’s width… Here is an article about this so-called “sapphire war”: https://france3-regions.francetvinfo.fr/auvergne-rhone-alpes/puy-de-dome/clermont-ferrand/il-trouve-des-saphirs-dans-une-riviere-ses-voisins-saisissent-la-justice-2842313.html

Comments are closed.