10:45 AM - 11:15 AM
[SVC33-07] Eruption mechanism and diverse periodicity of Geysers—Observational and Numerical studies of Shikabe Geyser in southern Hokkaido, Japan
★Invited Papers
Keywords:geyser, periodicity, eruption dynamics
Geyser has been regarded as an analogy of volcanic eruption, and its eruption mechanism has been investigated because it would provide helpful insights into understanding the eruptive processes of volcanoes (Hurwitz and Manga, 2017). Several conceptual models have been proposed to explain the periodic eruption based on observations and laboratory experiments (e.g., Kieffer, 1984; Belousov et al., 2013). However, they only partially explain observed phenomena and leave unclear questions about the physical processes controlling the periodic eruption. Geysers indicate a time-predictable system in which the eruption size controls the time to the next eruption. Nevertheless, the eruption intervals are not always regular but can vary complicatedly, and the physical background has been poorly understood. In this study, we conducted field observations and numerical experiments of the Shikabe Geyser in southern Hokkaido, Japan, to understand the eruption mechanism and critical factors controlling various changes in the eruption intervals.
We implemented the observations inside and outside the conduit at Shikabe Geyser. Comparison of the pressure, temperature, and video observations in the conduit, chemical analysis of the discharged gas and water, and surface phenomena indicated that the decompression boiling at the shallow part of the conduit triggered the eruption and dissolved CO2 gas promoted the gas generation at the deep depth. To investigate the detailed physical processes inside the conduit during the eruption, we conducted numerical experiments using the wellbore-reservoir coupled two-phase flow simulator, T2Well/ECO2N (Pan et al., 2011). Considering a one-dimensional wellbore and radially symmetric aquifer as the computational domain and assigning the atmospheric condition at the wellhead and the water supply condition at the terminus of the aquifer as the boundary conditions, the unsteady simulation was performed. The typical characteristics of the eruption cycle reproduced in this simulation corresponded to those observed at Shikabe Geyser and other geysers. Furthermore, the simulation results indicate that the CO2 gas dissolved in the water plays a role in deepening the flash depth while it does not contribute to the eruption explosivity. Based on the spatiotemporal variations in the parameters, we proposed two feedback: a self-enhancing process, which promotes the boiling at the shallow high-temperature gradient region, and a self-limiting process, which suppresses the boiling at the deep low-temperature gradient region, would control the periodic eruption of geysers.
Our continuous observation from November 2019 to the present revealed that the eruption intervals at Shikabe Geyser show long-term changes, which comprise extension in winter and shortening in summer, and short-term changes occurring in time range from hours to days. In addition, we observed interesting characteristics, such as the modes of the eruption intervals distribution transited from unimodal to multi-modal, which indicated a non-linear response of the system. Focusing on the short-term variation of the time series data obtained from 2022 to 2023, the eruption interval changes that occurred in the period longer than one day indicated a significant negative correlation with the changes in the barometric pressure. We investigated the responses of the eruption intervals to the barometric pressure variation by conducting the numerical simulation using the T2Well/ECO2N. The simulation results suggested that the barometric pressure loading would suppress the boiling during the eruption by raising the saturation temperature. The weakening of the eruption would decrease the discharged mass, shortening the eruption intervals.
We implemented the observations inside and outside the conduit at Shikabe Geyser. Comparison of the pressure, temperature, and video observations in the conduit, chemical analysis of the discharged gas and water, and surface phenomena indicated that the decompression boiling at the shallow part of the conduit triggered the eruption and dissolved CO2 gas promoted the gas generation at the deep depth. To investigate the detailed physical processes inside the conduit during the eruption, we conducted numerical experiments using the wellbore-reservoir coupled two-phase flow simulator, T2Well/ECO2N (Pan et al., 2011). Considering a one-dimensional wellbore and radially symmetric aquifer as the computational domain and assigning the atmospheric condition at the wellhead and the water supply condition at the terminus of the aquifer as the boundary conditions, the unsteady simulation was performed. The typical characteristics of the eruption cycle reproduced in this simulation corresponded to those observed at Shikabe Geyser and other geysers. Furthermore, the simulation results indicate that the CO2 gas dissolved in the water plays a role in deepening the flash depth while it does not contribute to the eruption explosivity. Based on the spatiotemporal variations in the parameters, we proposed two feedback: a self-enhancing process, which promotes the boiling at the shallow high-temperature gradient region, and a self-limiting process, which suppresses the boiling at the deep low-temperature gradient region, would control the periodic eruption of geysers.
Our continuous observation from November 2019 to the present revealed that the eruption intervals at Shikabe Geyser show long-term changes, which comprise extension in winter and shortening in summer, and short-term changes occurring in time range from hours to days. In addition, we observed interesting characteristics, such as the modes of the eruption intervals distribution transited from unimodal to multi-modal, which indicated a non-linear response of the system. Focusing on the short-term variation of the time series data obtained from 2022 to 2023, the eruption interval changes that occurred in the period longer than one day indicated a significant negative correlation with the changes in the barometric pressure. We investigated the responses of the eruption intervals to the barometric pressure variation by conducting the numerical simulation using the T2Well/ECO2N. The simulation results suggested that the barometric pressure loading would suppress the boiling during the eruption by raising the saturation temperature. The weakening of the eruption would decrease the discharged mass, shortening the eruption intervals.