Open-System Magma Reservoir Affects Gas Segregation, Vesiculation, Fragmentation and Lava/Pyroclast Dispersal During the 1.2 km-deep 2007-2010 Submarine Eruption at West Mata Volcano

Abstract:

West Mata, a small, active rear-arc volcano in the NE Lau Basin, erupts crystal and gas rich boninite magma. Eruptions were observed at the summit (1.2 km water depth) during 5 ROV Jason dives in 2009 (the deepest erupting submarine volcano observed to date). Subsequent ROV and ship-based bathymetric mapping revealed that a pit crater formed and the summit eruption ceased in 2010, with roughly simultaneous eruptions along the SW rift zone. During the summit eruption, a combination of water depth, H₂O-CO₂-rich and high crystallinity magma, a split in the conduit to feed two vent sites, and waxing/waning magma supply led to a range of effusive/explosive eruption styles and volcanic deposit types. The 2-3 vent Hades cluster and the lone Prometheus vent had different eruption characteristics. Petrographic, petrologic and geochemical studies of erupted products indicate a change in magma composition in time and space over a period of 3.5 yrs, suggesting a small, open-system magma reservoir within the volcano. Prometheus (1174m depth) produced mostly pyroclastic material during our observations (e.g., highly vesicular glowing fluidal ejecta that cooled in the water column and rounded recycled dense clasts), but sampling and ²¹⁰Po radiometric dating show that several months prior pillowed lava flows, subsequently covered with cm-sized pyroclasts, had flowed >50m from the vent. In contrast, vents at Hades (1200m depth) cycled between lava production and vigorous degassing, 10-20m high fire fountains and bursts of glowing lava-skinned bubbles, the products of which froze/broke in the water column, forming unstable cones of spatter and scoria near the vents. We hypothesize that bubbles collapse rather than form lava balloons because of skin brittleness (from high crystal content) and hydrostatic pressure. Clast settling times and patterns suggest >100m water column rise height for 10+ cm-sized fragments. Pillow flows were also observed to be issuing from the base of the Hades cones some 30-50m below, and had traveled 100 m from the vent in the months before. This, plus hydrophone and water column data (Embley et al., G3, in review), and the occurrence of extensive deposits of young, glassy, identical composition cm-sized fragmental material 250 m from Hades suggest an earlier more vigorous phase of the eruption.