- Geological Survey of Japan, National Institute of Advanced Industrial Science and Technology, Tsukuba, Japan (marceau.gresse@aist.go.jp)
The Izu-Oshima volcano has been dormant for 38 years since its last eruption in 1986. Historical records indicate an average eruption interval of ~30 years, highlighting the importance of predicting changes in volcanic activity in the near future.
The water table of the Izu-Oshima volcano is at sea level, with a vadose zone about 500 meters thick in the summit caldera. Prior to the 1986 eruption, thermal and electromagnetic signals associated with volcanic fluids ascending through this vadose zone were observed (Kagiyama, 2018). We conducted numerical simulations of the hydrothermal system, based on electromagnetic and thermal observations, to analyze self-potential (SP) signal variations associated with fluid flow through the vadose zone.
Electromagnetic observations were conducted at 32 sites inside and outside the caldera using the Audio-Magnetotellurics (AMT) method in 2006, 2007, 2009, and 2022. A 3D resistivity structure was obtained using inversion analysis with the WSINV3DMT code. The results revealed low resistivities (1–10 Ω·m) below sea level, suggesting the presence of hot water containing dissolved components or altered minerals. In contrast, high resistivities (1,000–10,000 Ω·m) were observed above sea level, except beneath the crater, indicating unsaturated scoria or lava layers. The resistivity structure suggests that relatively stable hydrothermal activity has persisted below sea level through past eruptions.
Numerical simulations were conducted to investigate the effects of rainwater infiltration into the vadose zone and water vapor rising from a deep source. Hydrothermal convection was found to depend on the flow rate of the source. At low flow rates, water vapor condensed, limiting fluid flow to below sea level, which aligns with the observed low-resistivity structures. The SP distribution was calculated from fluid flow and compared with observed SP distributions from 2006 and 2018. While accounting for spatial heterogeneities in some areas is necessary, the observed SP distributions are generally reproduced by incorporating rainwater infiltration and resistivity structure (Onizawa et al., 2009).
Simulations indicated that hydrothermal convection below sea level is difficult to detect through SP distribution alone. However, if the flow rate of water vapor from the source increases, water vapor can rise to near-surface levels, crossing the water table. In such cases, the SP distribution exhibits a negative anomaly at the summit due to the counterflow of water vapor and condensed water in the vadose zone.
Continuous SP monitoring at 20 sites has been conducted since 2006, with data recorded at 1-minute intervals and transmitted to our institute daily. To date, the long-term trends have remained unchanged at most locations. We continue SP monitoring to detect signal changes anticipated by the simulations, which may provide critical insights into future volcanic activity.
How to cite: Matsushima, N. and Gresse, M.: Hydrothermal system of Izu-oshima volcano inferred from numerical simulation and field observations, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-3414, https://doi.org/10.5194/egusphere-egu25-3414, 2025.