Very Long-Period Seismic Signals and Collapse Events at the Kilauea Summit Crater in 2018
- 1Southern University of Science and Technology, Department of Earth and Space Sciences, Shenzhen, China (yell@sustech.edu.cn)
- 2Department of Earth and Planetary Sciences, UC Santa Cruz, Santa Cruz, CA, 95064 USA
- 3Seismological Laboratory, California Institute of Technology, Pasadena, CA, 91125 USA.
- 4Earthquake Research Institute, The University of Tokyo, Tokyo, Japan.
The 2018 Kilauea eruption and East Rift Zone dike intrusion have been accompanied by more than 60 earthquakes with magnitude 4.7 – 5.3 near the summit. Corresponding long-period moment tensor solutions (constrained to have no isotropic component) have nearly vertical P-axes with moderate to large non-double-couple components (~20%-80%). The predominantly normal faulting is consistent with summit deflation as magma drains to the rift zone. Ground velocities recorded by local broadband seismic stations (<5 km) along the edge of the crater show distinct behavior between mid-May events and subsequent events. From May 16 to 26, twelve M5 events with large eruptive plumes below the SE edge of Halema'uma'u crater generated very long-period (VLP) seismic pulses with durations of about 20-40 s at all azimuths, suggesting distinct static outward displacements. This group can be represented by either isotropic or CLVD source. Almost all VLP pulses ended with sharp arrivals that are likely from small collapses. Similar VLP signals had been observed during the 2000 Miyake-jima eruption (Kikuchi et al., 2001; Kumagai et al., 2001). After May 28, M5 events were located below both north and south edges of the crater and generated broadband ground motions at the same stations; with large amplitude fracture-generated high-frequency signals (often clipped). Seismicity at the summit was low during the VLP events but increased to have 20-40 events per hour before later M5 events followed by several hours of reduced activity. After May 28, seismicity has been quasi-periodic, with intervals from ~2 to ~1 day. Comparison with solid earth tides and volume strain variations suggests that they are not direct driving factors, because the time interval between collapse events varied significantly. We consider a model with two stages, stage 1 (from May 16 to 26) with inflation transients that causes VLP events with large ash/gas explosion, and stage 2 (after May 28) dominated by collapse with extensive shallow fracturing and weaker gas explosions. The rapid expansion of the VLP events may be due to water flashing to steam, or partial collapse into the magma causing abrupt inflation. We investigate these VLP events comprehensively with local broadband seismic data, infrasound signals, GPS and tilt signals in this study, along with comparisons with published studies for the 2018 Kilauea sequences.
How to cite: Ye, L., Lay, T., Kanamori, H., Brodsky, E., Ohminato, T., Ichihara, M., Yamakawa, K., Tsuruoka, H., and Satake, K.: Very Long-Period Seismic Signals and Collapse Events at the Kilauea Summit Crater in 2018, EGU General Assembly 2023, Vienna, Austria, 24–28 Apr 2023, EGU23-2598, https://doi.org/10.5194/egusphere-egu23-2598, 2023.