The Kirishima Volcano Group is a volcanic field ideal for studying the mechanism of steam-driven eruptions because many eruptions of this type occurred in the historical era and geophysical observation networks have been installed in this volcano. We made regular geothermal observations to understand the hydrothermal activity in Ebinokogen Ioyama Volcano. Geothermal activity resumed around the Ioyama from December 2015. A steam blowout occurred in April 2017, and a hydrothermal eruption occurred in April 2018. Geothermal activity had gradually increased before these events, suggesting intrusion of the magmatic component fluids in the hydrothermal system under the volcano. The April 2018 eruption was a magmatic hydrothermal eruption caused by the injection of magmatic fluids into a very-shallow hydrothermal system as a bottom–up fluid pressurization, although juvenile materials were not identifiable. Additionally, the upwelling of mixed magma–meteoric fluids to the surface as a kick was observed just before the eruption to cause the top–down flashing of April 2018. A series of events was generated in the shallower hydrothermal regime consisting of multiple systems divided by conductive caprock layers.

No abstract

Recent seismic monitoring mainly by the Hi-net High sensitivity seismograph network in Japan for the last decade has been revealing the 3D structure of velocity anomaly in the crust and mantle due to heterogeneous presence of deep-seated fluids and fluid-related deep low frequency earthquakes in subduction zones of Japan arc. Petrological water circulation models combined with geophysical subduction models quantitatively show the water budget in the solid earth. The recent findings infer the importance of deep hydrothermal fluid on the occurrence of inland earthquakes. As the models are built with the result obtained by monitoring, experimental techniques or simulations, implications from evidence-based geochemical and geological studies are expected for the proof of water circulation models. Hence, we examined chemical features of deep groundwaters in SW Japan arc, and showed spatial distribution of deep-seated fluid mixed into groundwater. We found that the deep-seated fluid, whose isotopic composition is similar to magmatic, has the high Li/Cl ratio 0.001 in wt. ratio and concluded that Li/Cl is a good indicator for detecting the slab-related deep-seated fluid in groundwaters. Spatial distribution of Li/Cl reveals that slab-related deep-seated fluid upwells along the faults and tectonic lines, and at close to Quaternary volcanoes in SW Japan arc. In most cases, upwelling places are found close to the areas where deep low frequency DLF earthquakes are occurring, implying that deep-seated fluid causes DLF events.