Presentation Information
[T2-O-6]Geofluid mapping based on simultaneous analysis of seismic velocity and electrical conductivity: the connection between magmas, fluids, volcanoes, and earthquakes
*Hikaru IWAMORI1, Geofluid Mapping Team (1. Earthquake Research Institute, The University of Tokyo)
Keywords:
geofluid,magma,fluid,volcano,earthquake,seicmic velocity,electrical conductivity
Aqueous fluids and magmas within the Earth's crust and upper mantle (termed "geofluids") play a pivotal role in diverse geological processes across a broad range of spatial and temporal scales. These processes encompass hydrothermal, magmatic, and geodynamic phenomena that influence ore formation, volcanism, seismic activity, plate motion, and crust-mantle evolution. For example, fluids reduce frictional strength along faults, facilitating seismic activity (Hasegawa et al., 2012, EPSL; Sibson, 2009, Tectonophys.), while chemical interactions within subducting plates modify rock viscosity and contribute to subduction dynamics (Nakao et al., 2016, EPSL). Additionally, melt present at the lithosphere–asthenosphere boundary can influence plate motion (Kawakatsu et al., 2009, Science). Geofluids also drive geochemical differentiation by transporting volatile elements and shaping mantle heterogeneity (Iwamori & Nakamura, 2015, Gondwana Res.). Despite their significance, conventional approaches have faced challenges in accurately determining the distribution and quantity of geofluids within the solid Earth.
To address this, our research group recently developed a method that simultaneously analyzes seismic wave velocities (Vp, Vs) and electrical conductivity (σ) to estimate lithology–geofluid parameters, including subsurface lithology type, geofluid phase, geofluid volume, and geometrical parameters (aspect ratio and connectivity). This method consists of: (A) A forward model (Iwamori et al., 2021, JGR), which calculates Vp, Vs, and σ based on properties of solid–liquid mixtures, incorporating temperature, pressure, lithology, geofluid type and composition, volume, and geometrical parameters. (B) An inversion model (Kuwatani et al., 2023, JGR), which employs Bayesian inference to identify the optimal lithology, geofluid type, volume, and geometry that best reproduce observed Vp, Vs, and σ values. To improve the reliability of inferred parameters, prior constraints—such as surface heat flow, erupted lava composition, and hot spring water chemistry—are integrated into the analysis. Applying this method to datasets from the Japan arc, specifically northeastern Japan, we performed 3D geofluid mapping spanning approximately 80 km east–west, 50 km north–south, and 40 km in depth. This enabled identification and quantification of the distribution of aqueous fluids, basaltic magma, and andesitic magma (Iwamori et al., 2025, Communications Earth & Environment). The geofluid mapping revealed:
(i) Magmas are broadly distributed along the Moho beneath both volcanic and non-volcanic (forearc) regions.
(ii) Aqueous fluids are released from the magmas, forming a substantial reservoir at depths of 10–20 km.
(iii) Elevated fluid pressure is estimated at the top of the reservoir, exceeding lithostatic pressure by more than 200 MPa.
(iv) The highest seismic activity is observed in association with the elevated fluid pressure described in (iii).
(v) Andesitic magma is present beneath active volcanoes, possibly having ascended from the Moho.
Expanding geofluid mapping to wider regions may facilitate quantitative predictions of hydrothermal, magmatic, and geodynamic processes relevant to ore deposition, volcanic activity, and earthquakes. Among these, earthquake prediction remains an urgent global challenge. Geofluid mapping is expected to be particularly effective in regions where subsurface fluid activity is anticipated. Provided that Vp, Vs, and σ data are available—along with distributions and chemical compositions of heat flow, lava, and spring water—this method offers global applicability for estimating geofluid distributions.
To address this, our research group recently developed a method that simultaneously analyzes seismic wave velocities (Vp, Vs) and electrical conductivity (σ) to estimate lithology–geofluid parameters, including subsurface lithology type, geofluid phase, geofluid volume, and geometrical parameters (aspect ratio and connectivity). This method consists of: (A) A forward model (Iwamori et al., 2021, JGR), which calculates Vp, Vs, and σ based on properties of solid–liquid mixtures, incorporating temperature, pressure, lithology, geofluid type and composition, volume, and geometrical parameters. (B) An inversion model (Kuwatani et al., 2023, JGR), which employs Bayesian inference to identify the optimal lithology, geofluid type, volume, and geometry that best reproduce observed Vp, Vs, and σ values. To improve the reliability of inferred parameters, prior constraints—such as surface heat flow, erupted lava composition, and hot spring water chemistry—are integrated into the analysis. Applying this method to datasets from the Japan arc, specifically northeastern Japan, we performed 3D geofluid mapping spanning approximately 80 km east–west, 50 km north–south, and 40 km in depth. This enabled identification and quantification of the distribution of aqueous fluids, basaltic magma, and andesitic magma (Iwamori et al., 2025, Communications Earth & Environment). The geofluid mapping revealed:
(i) Magmas are broadly distributed along the Moho beneath both volcanic and non-volcanic (forearc) regions.
(ii) Aqueous fluids are released from the magmas, forming a substantial reservoir at depths of 10–20 km.
(iii) Elevated fluid pressure is estimated at the top of the reservoir, exceeding lithostatic pressure by more than 200 MPa.
(iv) The highest seismic activity is observed in association with the elevated fluid pressure described in (iii).
(v) Andesitic magma is present beneath active volcanoes, possibly having ascended from the Moho.
Expanding geofluid mapping to wider regions may facilitate quantitative predictions of hydrothermal, magmatic, and geodynamic processes relevant to ore deposition, volcanic activity, and earthquakes. Among these, earthquake prediction remains an urgent global challenge. Geofluid mapping is expected to be particularly effective in regions where subsurface fluid activity is anticipated. Provided that Vp, Vs, and σ data are available—along with distributions and chemical compositions of heat flow, lava, and spring water—this method offers global applicability for estimating geofluid distributions.