This is a strange end to a strange year. Kilauea erupting without even HVO noticing it until the lava was flowing, Etna exploding into life, and even New Zealand trying to get in on the action. And we were complaining that nothing was happening and we had to live off the glories of Christmas past and Christmas yet to come, with Christmas present nothing to look forward to. How wrong we were. Of course, in hindsight 2020 will be one for the history books: the year the world went mad. The expression ‘2020 vision’ will never mean the same after this.
Yesterday was the solstice, the Great Conjunction (800 years in the making: I managed to see it in a clear moment a day before), and a meteor swarm (none appeared trough the thick cloud): what more could one wish for as a portent of things to come? And indeed, the UK went into complete isolation with all border crossing closed, and even internal travel forbidden, with a new virus mutation being used as an excuse to do what should have been done weeks ago. (It is interesting that travel is forbidden but U-turns are still allowed.) We can now prepare for Christmas in isolation, Australia-style. (For non-UK readers, the term ‘Australia style’ is used here as a euphemism for not having something. An ‘Australia style trade deal’ means no trade deal. Abuse of language also is 2020 vision, in which reality has been replaced by advertisement slogans.)
In this darkness, out came Kilauea doing what it should have been doing weeks ago (December 2, to be precise. It had started in September when GPS measurements began to show inflation, and this became sustained over the next months. The inflation was centred underneath Halemau’au.
What caused this? There are three options: Deeper magma may have risen closer to the surface, the magma inflow may have increased, or the outflow was reduced. The rising magma is the easiest explanation. This gives inflation close to magma and deflation further out (as no new magma is involved, the inflation and deflation have to balance). But no such deflation was seen, at least on the public GPS data. My guess is therefore one of the other two. A notable feature of the inflation is that only the upper most part of the rift zone was involved with some earthquake activity. Further out, there was actually but of a reduction in pressure, at least judging from the GPS’s. (I should say this is purely my opinion and I may well be proven wrong. As we used to say – speculation alert!) If there was a higher magma influx, one might expect this to affect the rift zone as well. Therefore, it seems to me that a blockage in the rift zone was at least part of the reason. Magma continued to arrive in Kilauea, as it always does, but like to English borders, it couldn’t get out. So pressure increased.
On Dec 2, we all noticed a sudden jump in the tilt measurements. This was accompanied by earthquake activity. HVO reported a dike intrusion. Interestingly, the GPS was hardly affected and continued its steady upward trend. This is a good sign that in this case, the cause was upward movement of magma. The tilt excursion was large, but it is measured in micro radians which is a very small unit. So the actual amount of magma involved was not large. This must have come very close to a surface eruption, though.
A few days ago the earthquake swarm resumed, although with nothing out of the ordinary. We commented on this on the blog. In hindsight, with magma so close to the surface already, this was all it took. At 9:30 pm HST on Dec 20 (7:30am Dec 21 in GMT) the eruption started, taking everyone by surprise. It started less than 3 hours before the solstice (which was 10:02am GMT). Inconveniently, it started on the steep slope above the new lake, directly below the web cam. The camera could see it but the view would have been so much better from the other side. The eruption was effusive from the start. Perhaps the broken-up ground from the 2018 collapse had allowed gas to escape, preventing an explosive eruption. But no gas emissions had been noticed prior to the eruption. HVO saw the danger of explosions in the lake and raised the aviation level to red. But it remained effusive and the level has now been reduced to orange. But having to go from green straight to red shows how little warning this eruption gave in the hours before the event. That also shows that the conduit was already open (and filled) before the event. No braking of rock was required.
The eruption quickly formed three eruption sites, all on the side of the lake pit. The middle one was weak and soon died, but the other two both continued. They are about 400 meters apart. The three sites were approximately along a line, typically for a fissure eruption. Normally a fissure eruption eventually focuses on one location. This may still happen here but so far it hasn’t. Where did the magma come from? One way to find out is by imagining where the shortest distance under ground would lead to these three locations, rather than somewhere else on the slope. Draw the fissure line, and extend it perpendicular to the steep slope. This brings one to the edge of the collapse area. The fissure runs parallel to this edge. My guess is therefore (speculation alert!) that the magma came up through this ring fault. This says nothing about the location of the deeper conduit! It is about the last kilometer or so of travel. Perhaps the Dec 2 dike connected to this ring fault.
The eruption site is near the centre of the old Halemau’au crater. The deeper conduit seems to have survived from before 2018. It is notable that the fissure is close to (or the same?) as that of the 1954 summit eruption, again suggestion that deeper below the collapse area, the plumbing of Kilauea has survived the events of 2018.
The lava immediately started to flow into the lake. The lake, predictably, began to boil away. The heat capacity of water is four times that of lava, however, this was fresh, hot lava and the lake was already at 70C to begin with. So it didn’t take much to boil. The boiling is very energy intensive. Lava even at temperatures over 1000 C does not have the energy to boil its own volume of water. It can only come close by solidifying, and using its own latent heat of solidification. What happened was that the lava was flowing into the lake and going to the bottom. The lake started to both boil and rise, while the lava solidified. It looks to me though that there wasn’t enough lave to boil off the entire lake, and that would have taken more volume in lava than there was in water. The steam plume contained less water than had been in the lake: part of the water may have escaped through cracks in the rock, as the water level rose. Prior to the eruption the water lake had stopped rising, and it may have reached the level where it could drain away if going any higher. We ended up with a lava lake instead of a water lake, but the bottom is filled with solidified lava. The liquid lava was much less deep than the water had been. The escaping water may also explain the lack of gas emission: water may have taken some of the SO2 with it. (Speculation alert!)
The pit has kept filling up and the lava level is now tens of meters above the original water level, albeit still below the level of the fissures.
What will happen next? The tilt is showing a typical exponential, shallowing decline. The exponential-ness shows that the pressure behind the eruption is decreasing. The eruption rate is also likely going down. Initially it may have been 100 m3/s (I would guess not much higher because of the lack of strong fountaining), at the moment I guess it is below 50 m3/s. (but I am open to be corrected on this!) The fissuring may end quite suddenly when the pressure becomes too low to keep the fissure open. It may also continue for weeks at a lower rate. In the latter case the lava lake may drown the fissure and it becomes a more typical lava lake. It seems unlikely to me that the level will go much higher yet. But this lava lake is, in my opinion, going to be a feature of Kilauea for years to come, perhaps intermittent but always prone to return.
What happens next also depends on the rift. If it remains closed, we may eventually see the return of Kilauea lava lake of Christmas past, covering much of the caldera (speculation alert). This may take a decade or more. However, the changing pressure may re-open to rift zone leading to the end of the eruption and a fast draining. It is also possible that the southwestern rift zone re-opens. It is solidly blocked after many years of quiescence (sine 1974), but the fissure lines up with this rift zone. The most recent earthquake activity has been on the starting point of this rift. It is a low likelihood, but this eruption could be a stage in re-starting eruptions in the southwest. Any map of the region will show how badly lava-covered this region is – it has potential.
The current event may be a step on the way to re-establishing the Kilauea from before the never-ending Pu’u’O’o eruption. The eastern rift zone will erupt again. But it may not be as dominant as before. The summit has returned, and will show who is the boss. After all, you don’t get to be the summit if you let the rift do all the work.
And here ends my reading of the tea leaves. It is all speculation. But these safe eruptions are quite fun.
Albert, 22 Dec 2020