Japan is a major economic power in the modern world. Currently, its economy is the third largest in the entire world and exceeded only by those of the USA and The Peoples Republic of China. But as well as being one of the most economically and technologically advanced countries, Japan is also among the most volcanically active countries in the world.
Recently, the Japanese volcanoes have only produced smaller eruptions. The most recent VEI 5 was 250 years ago. The last VEI 7 (Kikai) occured 6,400 years ago. However, with a volcanic chain some 2,500 km long, fuelled by several subducting plates, the potential for large eruptions clearly exists. The 1792 eruption and flank collapse of Mount Unzen caused 15,000 casualties. Further evidence for destructive potential comes from a number of large calderas. On the world volcanic threat index, Japan is ranked third, after Indonesia (of course) and the Philippines. This index takes into account volcanic hazards, number of volcanoes, and population density within 30 km of a volcano.
The Five Calderas
Japan consists of three main islands: Hokkaido in the north, Honshu (the main island) in the middle, and Kyushu in the south. Kyushu has the most notable volcanic history. It shows a sequence of calderas, approximately along a north-south line. Southernmost, off the coast, is Kikai. Next are Ata, Aira and Kakuto. Northernmost in this arc of five calderas and 23 smaller volcanoes is Aso, located in central Kyushu. The volcanic zone continues to the southern edge of Honshu, but without further calderas, and terminates at the southern edge of Honshu. The five main calderas are:
- The submerged Kikai-Akahoya caldera is 17 by 20 km across: it was formed by a VEI 7 eruption around 4,550 BC which spread ash across Japan, and pyroclastics over much of Kyushu. Ignimbrite deposits dated at 140,000 and 95,000 BP show that there had been similar eruptions before.
- Ata caldera is a 25 by 12 km submarine caldera which has been dated at 105,000 BP. It contains a small caldera, 4 by 3 km, from an explosion only some 4,000 years ago.
- The Aira caldera, 80 km north of Kikai along the same graben, is 23 by 17 km in extent and formed at 22,000 BP. A second eruption 16,000 BP formed the 5-km Wakamiko caldera. Aira hosts the highly active volcano Sakura-Jima on its edge. Aira caldera is accumulating deep magma at a rate of 100 km3 over 10,000 yr (eg. a potential VEI 7 every 10,000 years). The shallow magma reservoir underneath Sakura-Jima, connected to the deeper reservoir, is gaining mass at about a third of this rate.
- Kakuto caldera is 15 by 10 km; it is approximately 110,000 years old but appears to be double: its western side is now thought to be a separate caldera called Kobyashi which may be 65,000 or even 120,000 years old. Kakuto contains the more recent Kirishima volcano on its southeastern edge, which itself has built a 20 km wide volcanic zone and should be considered a risk for a future caldera events.
- Aso caldera is the largest, at 25 by 18 km, and the oldest, formed by four eruptions dated between 270 to 90 thousand years ago. Unlike the other calderas, it is crossed by a major tectonic line, the Oita-Kumamoto Tectonic Line.
(Note: There is an uncertainty over the eruption volumes quoted in the papers refer to volume DRE or ash (tephra). If the latter, a rule-of-thumb is that tephra/ash is 2½ times as voluminous as magma/lava, hence the figures might need be down-sized to 40%.)
Together, the calderas have produced an estimated 1,580 km3 of magma over 300,000 years. About half of this was erupted from Aso. The average rate of magma eruption is 0.005 km3 per year, a little more than Yellowstone and a little less than Taupo. All calderas are visible in maps of heat flow on the surface: this shows that magma chambers still exist and thus, the possibility of further great eruptions.
The volcanic arc is caused by the subduction of the Philippine Sea plate underneath the Eurasian plate. This plate subducts in two zones, the southern Ryuku trench, with normal subduction, and the northern Nankai trough where subduction is at an angle with shear (sideways) movement. Kyushu lies near the junction of these two.
The subduction (re-)started about 6 million years ago when the Philippine Sea plate began to move north. This caused north-south extension (stretching) in central Kyushu, which resulted in a depression 70 by 40 km wide, called the Hohi volcanic zone. Mount Aso is located at the southern edge of this zone. Large scale fissure eruptions built up large lava plateaus. The Philippine Sea plate began to move more to the west about 1.5 to 2 million years ago, possibly related to the Eurasian plate beginning to move east at this time, pushed by the Baikal rift. This diminished the north-south extension and volcanism changed character. Volcanoes in Kyushu had begun to build from about 3 million years ago but when the plate turned to the west, they began to align with the subduction zone. Large pyroclastic eruptions started at about 1.3 million years ago.
To the south of Kyushu, rifting began, also about 2 million years ago (currently terminating in the Kagoshima graben, at the southern tip of Kyushu). This pushed the South Kyushu microplate into a counter-clockwise rotation. So far it has rotated by about 30 degrees. At the same time, shear developed to the north and east (the Median Tectonic Line) due to the oblique movement of the subducting plate. The two motions, rotation to the south and accelerating shear to the north, meet in central Kyushu. Volcanoes here are strongly affected by tectonic stress!
Activity at Aso began with basaltic eruptions, 2.2 million years ago. Intermittent eruptions continued but over time became more explosive, with pyroclastic and rhyolitic deposits. The major caldera explosions began 300 thousand years ago, when the crust began to be compressed by the tectonic stress. Previously crust extension had been more dominant.
The caldera formed in four huge eruptions dated at 270, 140, 120 and 90 thousand years ago. The first one was the smallest and the last one the largest. Two flank volcanoes, Akai and Omine, 20 km west and southwest of the caldera, lie along a fault extending from Aso towards Kumamoto City. Their lava shows composition indistinguishable from the caldera eruption and they were fed by the same magma chamber.
Since the caldera formation more than seventeen new eruption cones have formed in its centre. The five main cones are Neko, Taka (the highest, just under 1600 meters), Naka-Dake, Eboshi, and Mt. Kishima. Together these are called Mount Aso. Only Naka-Dake is currently active: it has erupted 167 times from seven different craters since 553 CE. The on-going eruption, from the most northerly of its craters, started in Nov 2014. Of the smaller cones, Kometsuka basaltic scoria cone, 10,000 years old, and Kishima-Dake, 3700 years, are also classified as active. The number of cones shows that the magma travels easily here.
The lavas from Aso show strontium-rich Adakites. These come from subduction where the subducting plate was itself young, and therefore still hot. The magmas are also very water-rich. The magma chamber underneath the main peaks is about 6 km deep, extends down to 10 km, and is 4-7 km wide. But there are deeper magma sources. A sill has been reported at 15 km depth, causing pressure over a large area, and a low velocity layer at 10-24 km depth, and finally a hot, melted region located at 30 km deep, which may be caused by a mantle wedge.
This wedge is reported as being “serpentinised”, a metamorphic mineralization process where mafic and ultramafic rocks mainly constituted of the minerals Forsterite and Fayalite metamorphose into Aqueous Silica, Serpentine and Brucite by the assimilation of water. Basically, every three molecules of Fayalite bind two molecules of water. Forsterite can either turn into Serpentine where every three molecules assimilate four of water, or Brucite where two molecules of Forsterite assimilates three of water.
Why is this so dangerous? Consider a slab of serpentinised rock several hundred if not thousands of cubic kilometres of these minerals and the location; in the top of the mantle at temperatures well in excess of 1200 degrees Celsius! The water liberated from its chemical bindings as the slab melts back into basalt act as a flux, melting the overlying continental crust and forming large quantities of silica-rich, evolved magmas (dacite, rhyolite). The explosive potential is tremendous as at those temperatures and pressures, superheated water that flashes into steam expands up to 10,000 times in volume. This is one reason why the volcanoes of Kyushu are so explosive and why there is a very real potential for very large, caldera-forming eruptions in the near geological future, especially at Aso.
The calderas of Kyushu tend to be multiple: a new centre develops, perhaps on the edge of the previous caldera, which goes through its own cycle of growth and collapse. This is most clearly seen in Aira, with a secondary caldera, and in the double caldera of Kakuto. It grows and falls – and falls again. A larger eruption can wipe out an older, smaller caldera: this may have happened in Aso where the most recent eruption at 90 kA BP was the largest.
Caldera eruptions have continued elsewhere in the arc until the present day. It would therefore be optimistic to assume that Aso is safely in its post-caldera phase. The large number of eruption sites over a large area within the caldera, and the continuing activity, show that the ingredients for large eruptions are still present. Each of the large caldera eruptions started with evolved, crystallized magma and moved to more juvenile magma. This suggests a large reservoir of magma. Seismology data suggests that beneath Aso is a “hot zone” of primed mantle. Whether this is part of an even larger magma chamber is an open question.
The long recurrence time is an indication of the potential. Statistically, the longer the time between repeat eruptions, the larger the eruption can be. Statistics of known eruptions indicate that for eruptions of size VEI7.5, the repeat period is 40,000 yr. Thus, an eruption larger than this is theoretically possible in Aso.
Kyushu is a densely populated area. With the large, caldera-forming eruption Aso is capable of, ignimbrites would travel in excess of 50 km from the vent and the shape of the terrain; open to the west, would act as a funnel so that not only would the entire Kumamoto province be within the zone of 0% survivability but also Nagasaki Prefecture at 80 km distance with its 446,000 inhabitants would also be at great risk as well as the Saga Prefecture (pop. 0.85 million) about 100 km to the northwest.
There has never been a VEI 7-sized eruption in a modern, developed country. The world has gone through a quiet spell, volcano-wise, over the past 150 years. We do not know how resilient our society is to a Tambora-sized eruption. That one will happen is certain. Japan is a likely place where this will occur, because of its sheer number of volcanoes. Mt Aso is a particular worry because of its history, location, and magma reservoir that one publication describes as ‘charged’.
It is well to remember that as a result of the Fukushima disaster in March 2011, an area of land some 310 square kilometres in extent was rendered uninhabitable, 160,000 people were evacuated permanently and the total cost estimated at between $250-300 billion. This pales into insignificance compared to the devastation a VEI 7 could cause. First of all, the death toll would be horrendous unless millions could be evacuated in advance – to where and for how long? Second, the area covered by ignimbrites would most likely exceed 7,500 square kilometres, 25 times the area rendered uninhabitable by Fukushima and be just as effectively rendered uninhabitable for the foreseeable future. The area covered by ash from pyroclastic flows or deposited aerially to depths between 1 and 100 cm, thus rendered useless for agricultural purposes for a period of time ranging from a year up to several decades, would be far greater. Just to provide a simple, everyday necessity such as potable water could turn out to be such a problem that a very large segment of the population of Kyushu might have to be evacuated on that count alone until infrastructures for water purification had been installed. There are other Japanese volcanoes which may be equally dangerous, but perhaps not recognized as such. Are we prepared?
Henrik & Albert
- Hunter, 1998. http://petrology.oxfordjournals.org/content/39/7/1255.full.pdf+html
- Hiroki Kamata and Kazuto Kodama, 2001, Volcanic history and tectonics of the Southwest Japan Arc http://onlinelibrary.wiley.com/doi/10.1046/j.1440-1738.1999.00241.x/abstract (paywalled – sorry!)
- Volcanic History of Kyushu https://www.numo.or.jp/en/reports/pdf/TR-09-02-5.pdf
- Velocity structure in the crust beneath the Kyushu area: Saiga et al. 2010. www.terrapub.co.jp/journals/EPS/pdf/2010/6205/62050449.pdf