*Dan Muramatsu1, Mie Ichihara1, Osamu Kuwano2
(1.Earthquake Reserch Institute, The University of Tokyo, 2.Japan Agency for Marine-Earth Science and Technology)
Keywords:Volcano, Acoustic wave, Rheology
Pulse-like acoustic signal is commonly observed during diverse volcanic phenomena including explosions, but their mechanisms are non-unique and not fully understood. During Strombolian eruptions, large bubbles bursting at the magma surface generate pulse-like signals, and similar phenomena occurring at mud suspension have been reported in hydrothermal eruptions (Jolly et al., 2016, JVGR). Muramatsu et al. (2022, JVGR) found a temporal change in the waveform and spectral content of pulse-like signals observed during the 2018 phreatic eruption of Kirishima Iwo-Yama volcano. Based on the rheological property of mud suspension sampled from the crater, they suggested that variations in yield stress and viscosity of the mud suspension depending on water fraction caused the temporal change in pulse-like acoustic signals. We validate this idea by laboratory experiment using analog material of the mud suspension. The mud sample from Iwo-Yama has yield stress and exhibits shear-thinning at high stress and strain rate. Water dispersion of the chemically synthesized clay (SUMECTON-SA, Kunimine Industries) was used as an analog material of the mud suspension. The experimental setup was based on Sánchez et al. (2023, GRL), where air bubbles are generated by injecting a constant flow of air from the bottom of a glass container with an injector. We measured bubble bursting at the surface with broadband microphones and high-speed cameras. The bubbles always spread horizontally and did not rise to the surface at the first injection. Bubbles rose and burst at the surface if strain was artificially applied from the tip of the injector to the surface before injection or if there were satellite bubbles trapped in the fluid. The waveforms and dominant frequencies of the pulse-like acoustic signals varied even for the same sample concentration and injection flow rate and sometimes showed temporal changes during an experiment. These behaviors can be attributed to yield stress and memory effects (strain history dependence) of the fluid (Vidal et al., 2009, PRE). Water splashing was also observed with bubble rupture, suggesting that separation of the liquid phase occurs in the bubble passage (conduit). We suggest that the temporal change in pulse-like acoustic signals observed at Iwo-Yama may be due not only to the variation of water fraction but also to the time-dependent rheology and liquid phase separation.