2:23 PM - 2:42 PM
[SVC30-03] Investigation of hydrothermal structure and short-term transition in Noboribetsu geothermal area based on a numerical simulation
Keywords:cap structure, fluid transport numerical simulation, unrest
Background
Recent electromagnetic surveys at volcanoes with well-developed hydrothermal systems have reported clay-rich, low-permeable cap layers in the shallow part (e.g., Yoshimura et al., 2018). Such low-permeable layers are known to act as a seal against hydrothermal water ascending from depth. In some cases, partial rupture of the seal with an increased permeability leads to rapid decompression and vaporization of the fluid underneath, resulting in a phreatic eruption (e.g., Tsukamoto et al., 2018). Even though an eruption does not occur, the pattern of fluid transport and the associated mechanical and thermal changes in the system may exhibit various responses (e.g., Tanaka et al., 2018). We conducted a parametric study using numerical experiments to investigate the temporal transition of volcanic activities associated with changes in permeability and/or fluid supply in a hydrothermal system with a sealing structure. We chose the Noboribetsu geothermal area, Kuttara Volcano, as a model field, where numerous phreatic eruptions are recognized from geological records (Goto et al., 2013). In recent years, some elevated hydrothermal activities occurred, such as abrupt blowing events of hot water from the thermal pond locally called Taisho-Jigoku (JMA, 2019).
Method
We used the TOUGH2 code to simulate a multi-component and multi-phase groundwater flow in a porous medium. The model space has a horizontal range of 10 by 10km and a depth range of 0 to 1500m b.s.l., centered on the Jigokudani steaming ground. In our model, we incorporated topography and subdivided the model space into five domains with different permeabilities: "Host rock" as surrounding media, "Conduit" as a primary pathway of hydrothermal water, "Cap rock" as a sealing structure, "Surface layer" covering from the top of surface to approximately 250 m depth, and "Fracture zone" with an intermediate permeability on the intersection between the Conduit and Cap rock. First, we reproduced a quasi-steady state by introducing a meteoric water supply from the top, injection of thermal salt water from the bottom of the Conduit, and a constant heat flow at the bottom of the Host rock. Next, we calculated the bulk resistivity of each cell based on the simulation output: temperature, pressure, and NaCl concentration of pore water. Then we compared it to the resistivity model obtained from the magnetotelluric surveys in the previous studies. In the second step, we searched for the conditions that reproduce the reported transient surface manifestations, such as an elevated ground temperature and expansion of steaming ground from 2013 to 2022 in the Kasayama area, north of Jigokudani. To this end, we imposed instantaneous changes in the permeability of the Fracture zone and the injection rate of hydrothermal water. Then we investigated the system's response for ten years with a focus on the pore pressure, temperature, and heat discharge from the ground surface.
Results
The quasi-steady state simulation well reproduced the location of hot zones corresponding to Jigokudani, the Oyunuma thermal pond, and Noboribetsu Spa, as well as the surface heat loss estimated by Terada et al. (2012). We also confirmed that the low resistivity anomaly of our model below the Noboribetsu geothermal area generally agreed with those estimated from the magnetotelluric surveys (Goto et al., 2015; Hashimoto et al., 2019). Furthermore, the transient experiments indicated that the combination of changes in the permeability of the Fracture zone and the injection rate of hydrothermal water could either increase or decrease the pore pressure right beneath the cap rock, even though the change in the surface heat discharge was similar. Our results showed that detailed knowledge of sealing structures is necessary to understand or predict the behavior of volcanic hydrothermal systems.
This research was supported by the Integrated Program for Next Generation Volcano Research and Human Resource Development.
Recent electromagnetic surveys at volcanoes with well-developed hydrothermal systems have reported clay-rich, low-permeable cap layers in the shallow part (e.g., Yoshimura et al., 2018). Such low-permeable layers are known to act as a seal against hydrothermal water ascending from depth. In some cases, partial rupture of the seal with an increased permeability leads to rapid decompression and vaporization of the fluid underneath, resulting in a phreatic eruption (e.g., Tsukamoto et al., 2018). Even though an eruption does not occur, the pattern of fluid transport and the associated mechanical and thermal changes in the system may exhibit various responses (e.g., Tanaka et al., 2018). We conducted a parametric study using numerical experiments to investigate the temporal transition of volcanic activities associated with changes in permeability and/or fluid supply in a hydrothermal system with a sealing structure. We chose the Noboribetsu geothermal area, Kuttara Volcano, as a model field, where numerous phreatic eruptions are recognized from geological records (Goto et al., 2013). In recent years, some elevated hydrothermal activities occurred, such as abrupt blowing events of hot water from the thermal pond locally called Taisho-Jigoku (JMA, 2019).
Method
We used the TOUGH2 code to simulate a multi-component and multi-phase groundwater flow in a porous medium. The model space has a horizontal range of 10 by 10km and a depth range of 0 to 1500m b.s.l., centered on the Jigokudani steaming ground. In our model, we incorporated topography and subdivided the model space into five domains with different permeabilities: "Host rock" as surrounding media, "Conduit" as a primary pathway of hydrothermal water, "Cap rock" as a sealing structure, "Surface layer" covering from the top of surface to approximately 250 m depth, and "Fracture zone" with an intermediate permeability on the intersection between the Conduit and Cap rock. First, we reproduced a quasi-steady state by introducing a meteoric water supply from the top, injection of thermal salt water from the bottom of the Conduit, and a constant heat flow at the bottom of the Host rock. Next, we calculated the bulk resistivity of each cell based on the simulation output: temperature, pressure, and NaCl concentration of pore water. Then we compared it to the resistivity model obtained from the magnetotelluric surveys in the previous studies. In the second step, we searched for the conditions that reproduce the reported transient surface manifestations, such as an elevated ground temperature and expansion of steaming ground from 2013 to 2022 in the Kasayama area, north of Jigokudani. To this end, we imposed instantaneous changes in the permeability of the Fracture zone and the injection rate of hydrothermal water. Then we investigated the system's response for ten years with a focus on the pore pressure, temperature, and heat discharge from the ground surface.
Results
The quasi-steady state simulation well reproduced the location of hot zones corresponding to Jigokudani, the Oyunuma thermal pond, and Noboribetsu Spa, as well as the surface heat loss estimated by Terada et al. (2012). We also confirmed that the low resistivity anomaly of our model below the Noboribetsu geothermal area generally agreed with those estimated from the magnetotelluric surveys (Goto et al., 2015; Hashimoto et al., 2019). Furthermore, the transient experiments indicated that the combination of changes in the permeability of the Fracture zone and the injection rate of hydrothermal water could either increase or decrease the pore pressure right beneath the cap rock, even though the change in the surface heat discharge was similar. Our results showed that detailed knowledge of sealing structures is necessary to understand or predict the behavior of volcanic hydrothermal systems.
This research was supported by the Integrated Program for Next Generation Volcano Research and Human Resource Development.