Spatiotemporal evolution of a marine caldera-forming eruption, generating a low-aspect ratio pyroclastic flow, 7.3 ka, Kikai caldera, Japan: Implication from near-vent eruptive deposits

Abstract

The VEI 7, 7.3 ka caldera-forming eruption of the Kikai caldera occurred in a shallow sea, and caused devastating damage to the prehistoric human settlements of southern Kyushu, Japan. This Holocene activity at this volcano records eruptions with compositional range of 56 to 74 wt.% in SiO 2 spanning with the climactic eruption. In this study, stratigraphy, component, and lithology of the pyroclastic deposits were analyzed at near-vent islands (Satsuma Iwo-jima and Take-shima) in order to reconstruct this eruption. Stratigraphic sections are characterized by plinian pumice-fall deposits (Unit A), intraplinian flow deposits (Unit B), climactic pyroclastic flow deposits (Units C1–C3), and co-ignimbrite ash-fall deposits (Unit D). In total the estimated magma volume in the system was 70–80 km 3 (DRE) and the eruption therefore represents the evacuation of a major silicic magmatic system. The plinian stage (Phase 1) is subdivided into an initial small phase and a second large one. The column height in the second phase was estimated to be 40–43 km. The total tephra volume of this stage was estimated to be 40 km 3. The magma discharge rate has been calculated from the column height data to be 2 × 10 8 kg/s. The eruption duration is also estimated to be a minimum of approximately 28 h. Collapse of the column (Phase 2) produced Unit B, which consists of multiple thin lithic-rich or pumice-rich layers or pods, including welded pumice-fall layers. The deposits are characterized by stratified or cross-stratified facies and display various degree of welding. These sedimentary characteristics indicate that, during the plinian column collapse, high temperature turbulent density currents were generated where dense pyroclasts were well segregated, resulting in the lithic-rich layers or pods. Phase 3 is characterized by Units C1–C3. Unit C1 shows non-welded stratified facies, which consist of lithic and crystals, including quenched juvenile materials as a minor constituent. Unit C2 displays welded stratified facies, which consist of lithic-rich layers and pumice-rich layers. These two subunits occur only in topographic lows in Satsuma Iwo-jima. Unit C3 is thickest and poorly-sorted non-welded massive deposit, which includes fragments of welded tuff from underlying units in proximal regions. These facts indicate that multiple pyroclastic density currents produced Units C1 and C2 in the near-vent area, and were followed by the main sustained current producing Unit C3, a low-aspect ratio ignimbrite, distributed over a wide area of southern Kyushu across the sea. Varying extents of magma–water interactions started during Phase 2, continuing during the early stages of Phase 3, and diminished during the climactic C3 ignimbrite stage. In addition, collapse of the caldera may have started before Unit C deposition, based on the evidence of a fault overlain by Unit C on the caldera rim. The collapse may have initiated water access to the magma. The source appears to have been biased toward the western side of the caldera. The Holocene evolution of the Kikai volcano records the existence of a large silicic magma system at depths of about 7 km that coexisted with or was regularly recharged with mafic magma.