It is a narrow but passable road, or at least it used to be. It begins at the Saddle Road which crosses Hawai’i between Mauna Kea and Mauna Loa. Both mountains have observatories at or near the summit, but while Mauna Kea is busy day and night, the Mauna Loa observatory is a much quieter affair. Observatory Road runs to the Mauna Loa Observatory. It begins on top of a lava flow and continues like that for 17 miles, while climbing from 2 km to 3.4 km altitude while the colourful lava of the start turns grey and dark. At first it runs up the slope towards the northeast ridge of Mauna Loa, before turning sharply right and continuing parallel to the ridge, ever climbing. The vviews are one-sided but spectacular. The road consists of only a single lane – in the unlucky case of meeting another vehicle, you may need to move over onto to lava. And if there is fog, there is only a white line to guide you – best to turn around. Continue, and you’ll find the end of the road at the Observatory, 5 kilometers from and 800 meters below the summit.
The road is now cut in three places, showing one of the disadvantages of routing it parallel to the eruption-prone ridge. The rifts follow the same direction, and as their lava finds the fastest way downhill, it crosses the road at right angles, for each eruption point in turn. The lava took out not only the access, but also the power to the observatory. There are actually three different observatories which share the site, but all have now paused. Best known is the atmospheric observatory where the famous Keeling curve of CO2 is measured. There is also a solar observatory here, and a radio telescope called the Yuan-Tseh Lee Array which is operated by Taiwan.
The eruption came in the night. On Sunday 27 November, 22:40 local time, the shaking started. It was already the morning of Monday 28 November in the UK. The first shake was a brief snap. Just over a minute later, a much larger shake followed and after that it became continuous. There was strong tremor for perhaps 20 minutes which became weaker. Around 23:30, lava reached the surface and the eruption began. After midnight it started to pulsate, with pulses lasting a minute and spaced 2 minutes apart. By half past midnight, the quaking calmed down. By this time, the lava was flowing freely.
The eruption started in 2004. After the 1984 eruption, Mauna Loa had shown little activity, and even slight deflation from 1990 to 2002. In 2004, an earthquake swarm , 50 km deep heralded awakening. A new magma batch had arrived and began its slow move up. It took 10 years for this to reach the upper magma chamber, some 4 km below the southwestern rim of the summit crater. In 2015 this caused a phase of rapid inflation. That stopped again There was more but slower inflation after 2018. In late September 2022, the earthquake swarm and inflation began in earnest. It very quickly became apparent that this was more like it and that an eruption was becoming possible.
The inflation had been centred underneath the southwestern side of the summit crater. This crater is called, in typical Hawai’ian fashion Mokuʻāweoweo, roughly meaning ‘burning island’. The association of a bone-dry crater on top of a 4-km tall mountain with an island sounds surprising, However, remember that the original inhabitants came acros a wide ocean where the first sign of land would be a towering cloud, caused by the sun warming the land and setting off air convection and cloud formation – and rain. They had become familiar with the smoke rising from the crater in its frequent eruptions. It is a nested crater, elongated along the main rift axis. Along this axis it is some 5 kilometers long.
As the inflation had shown, the main shallow magma is not underneath the centre of the summit but is offset to the southwest. It hides about 4 kilometers below the summit, more or less at sea level. This magma chamber is fairly small: it continuously receives new magma, but this magma does not spend long here. We know this from the lack of crystallization of the erupted lava. Perhaps this is the secret behind the frequent eruptions: a small magma chamber that quickly overflows, upward. This magma chamber had been depleted in the 1984 eruption but was now slowly growing. The earlier earthquakes swarms may mainly have come from the magma entering the chamber. But when the inflation resumed in October, the motion placed the magma closer to the centre of the crater. A dike was growing out of the magma chamber, along the line of least resistance which was angled towards the centre of the crater. This put stress on the rift zone and it began to open up. The lava baby was getting ready to arrive.
In most cases, though, events like this do not lead to an eruption. The dikes end before they reach the surface. Magma is pushed out from the chamber and is on the move, but the pressure may be insufficient to break the tough rock (basalt, of course, and unfaulted). At some point the magma pressure equals the strength of the overlying rocks and the dike stops in its tracks. This time, though, there was help. A larger earthquake cracked the rift running through the centre of the crater. Now the broken rock failed the stress test. Magma saw its chance and in less than an hour, it broke through to the surface.
There was no pre-eruptive activity at Mokuʻāweoweo: gas measurements showed a constant temperature. There is no shallow magma chamber. The opening of the dike on Sunday night created the connection. The pulsing could have been the flow from the magma chamber establishing itself through the connection.
Initially the eruption was from a short fissure within the summit crater. The crater began to fill with lava, and perhaps slightly overflowed towards the southwest. This was exactly as expected: eruptions at the summit tend to start as lava fountains from short fissures with lava flows covering the floor of the caldera. It is not always like this: there are several fans of explosive debris on and beyond the rim. The crater itself has no record of these: the entire surface is from recent lava flows, looking like a solidified black lake. Dating has shows evidence for 5 explosive events which ejected blocks over a meter in size across the rim, requiring ejection velocity of 200-300 m/s. The explosions were comparable to the explosion which started Kilauea’s activation in 2008. One explosion was much larger, and may have deposited ash over much of the southern flank. The two eastern fans are 150 and 200 years old. The western fan is about 850 years old. The large explosion is older. But this time Mauna Loa limited itself to fountaining.
Mauna Loa eruptions can quickly begin to migrate down one of the two rift zones, a behaviour that is very similar to that of Hekla, also a rifted shield with very short warnings before eruptions. This eruption mirrored this common behaviour. Within 7 hours the eruption moved to the upper northeast rift zone and the summit ceased erupting. There was a brief excursion to the southwest, probably because the short fissure extended in this direction. But the magma supply itself did not move in this direction: the flow from the centre of the crater. Instead the dike extended towards the northeast. There had been earthquakes at the northeastern end of the summit crater in the week before the eruption. These were likely related to the growing dike putting pressure on the rift. The widening of the rift had reduced the stress and allowed the dike to grow further in this direction. Think magma tube with a hole in the top at one place, underneath Mokuʻāweoweo. When the tube is blocked further down the line, the magma spurts out through the hole. But as a new opening formed further down (the fissures), the first hole lost its feed and magma flowed through the tube to the more distant and lower altitude hole.
The northeast rift was a good choice. The southwest rift zone is potentially more damaging, as lava can reach populated areas near the coast within a day. Northeastern lava flows tend to remain in areas already well covered by previous lava flows. They are not entirely without risk: in principal both coasts, including Hilo, are within reach. Since 1850, lava has reached near Hilo five times, but they always stopped before reaching the town. Half of Hawai’i is covered by lava flows from Mauna Loa: it is best not to underestimate its eruptions.
Over the next few days, four short fissures formed at different distances from the summit. The eruptions from each fissure did not last long. Again, imaging the magma tube. You need a lot of pressure to feed more than one hole simultaneously. The final fissure, fissure 4, was a step too far. Fissure 3 extinguished the other three and is now the only active one. The last active fissure is at the end of the earthquakes distribution of the past week. This was indeed the weakest link.
The radar image shows the roughness of the ground, as measured by the reflected radar signal. Blue is mostly fresh (smooth) lava. (Some of the pit craters have a blue rim on the right, due to shadowing of the radar signal.) There are flows on the southwest, a bit below the ridge. The original summit fissure possibly extended into this direction, aided by overflow from the summit crater. This was only a brief phase which ended when the northeastern rift opened. The main flows are from the northeaster rift, and are now purely from fissure 3.
A good view of the fissure come from a newly installed webcam. I don’t know exactly where this webcam is, but installing (and powering) is may have been an adventure!
The northeast rift has a series of cinder cones and the current eruption is likely to add to these. The presence of the cinder cones shows that eruptions tend to focus on a single point. The current eruption has created four separate fissures, but currently only one remains active. If the eruption continues long enough, the remaining fissure may leave such a lasting cone. In fact the webcam shows that it is already well on the way to do so.
The lava lobe has reached the saddle region where the land is almost flat. This has stopped significant expansion – for now. At the current rate, it could be weeks before the saddle road itself is reached and the eruption may not last that long. The main road seems safe. But things can change. The eruption could stop, only to resume on a lower fissure a week later. It is unpredictable. Keep an eye on the Saddle Road flow. If the pressure at the fissure decreases, the lava flow will slow down and the lava will no longer reach the end of the flow and instead there will be smaller break-outs further up stream. If that happens, it is a first sign that the eruption is running out of steam.
Mauna Loa eruptions can last anywhere from a day to a year. Volcanologists probably hope it will last long, so they get proper data with modern instruments – for the first time for Mauna Loa. The rulers of the land probably hope it will end before it destroys the Saddle Road, although a nice viewing area next to the road would be god for the economy. Astronomers and atmospheric scientists want their observatory back. Some will want them not to do so so that CO2 emission can continue without checks and balances. VC, on the other hand, just wants to get good views from a safe eruption – let it continue for as long as it is safe to do so. Cheers!
Albert, 4 December 2022