Minoru Takeo

Determination of constitutive relations of fault slip based on seismic wave analysis

Constitutive laws define the boundary conditions on fault plane and govern many aspects of earthquake failure. Although several constitutive laws have been formulated based on laboratory rock experiments and applied to theoretical studies in various fields, no actual relation during a natural earthquake has been determined. The 1995 Kobe earthquake is suitable for detailed kinematic analysis, and this enables the first evaluation of constitutive relations for a natural earthquake. In this study, we determine spatiotemporal slip distribution on an assumed fault plane of the 1995 Kobe earthquake by waveform inversion and then solve elastodynamic equations using a finite difference method to determine the stress distribution and constitutive relations on the fault plane. An inversion method based on Bayes theorem is employed to obtain a spatiotemporal slip distribution, and enables us to ensure the objective uniqueness of the solution with numerous parameters and smoothing constraints. This slip distribution is then used as part of the boundary condition in the finite difference calculation. The time histories of slip and shear stress obtained then provide a constitutive relation at each point on the fault plane. They show slip weakening relations almost everywhere on the fault plane, while slip rate dependency is not clear. The slip weakening behavior has a clear depth dependency indicating that the slip weakening rate (dτ/du) is smaller in the shallow crust than that in the deep crust. This may be associated with the paucity of shallow seismicity observed in the source region of this earthquake as reported for many mature fault systems.

Earthquake source asymmetry, structural media and rotation effects

Macroseismic Rotation Effects and Micromotions.- Development of Earthquake Rotational Effect Study.- Sources of Rotation and Twist Motions.- Some Examples of Rotation Effects: the Tulbagh Earthquake, South Africa.- Theory of Continua and Fields of Defects.- Deviations from Symmetry and Elasticity: Asymmetric Continuum Mechanics.- Degenerated Asymmetric Continuum Theory.- Continuum with Rotation Nuclei and Defects: Dislocation and Disclination Densities.- Towards a Discrete Theory of Defects.- Fault Dynamics and Related Radiation.- A Review on Friction.- Soliton Physics.- Rotation Motions, Seismic Source Models, and Asymmetry of Fracture.- Rotational Motions Excited by Earthquakes.- Ground Rotational Motions Recorded in Near-Source Region of Earthquakes.- Fracture-Band Geometry and Rotation Energy Release.- Rotation Motions: Recording and Analysis.- Glacier Motion: Seismic Events and Rotation/Tilt Phenomena.- Rotational Energy and Angular Momentum of Earthquakes.- Bend-Rotation Wave as a Mechanism of Macroseismic Effects.- Solitary Waves in Crustal Faults and their Application to Earthquakes.- Seismic Rotation Waves: Spin and Twist Solitons.- Earth Rotation, Elasticity and Geodynamics: Earthquake Wave Rotary Model.- Effects Related to Medium Structures and Complexity of Wave Propagation.- Seismic Rotation Waves in the Continuum with Nonlinear Microstructure.- Tectonic Solitons Propagating Along the Fault.- Complexity of Rotation Soliton Propagation.- Micromorphic Continuum with Defects and Taylor-Bishop-Hill Theory for Polycrystals: Anisotropic Propagation of Seismic Waves and the Golebiewska Gauge.- Seismic Ray Theory for Structural Medium based on Kawaguchi and Finsler Geometry.- From Non-Local to Asymmetric Deformation Field.- Earthquake Hazard in the Valley of Mexico: Entropy, Structure, Complexity.- Seismic Rotational Motions: Recording Techniques and Data Analysis.- Note on the Historical Rotation Seismographs.- Ring Laser Gyroscopes as Rotation Sensors for Seismic Wave Studies.- Rotational Motions in Seismology: Theory, Observation, Simulation.- Absolute Rotation Measurement Based on the Sagnac Effect.- Design of Rotation Seismometer and Non-Linear Behaviour of Rotation Components of Earthquakes.- Rotation and Twist Motion Recording - Couple Pendulum and Rigid Seismometers System.- Equation of Pendulum Motion Including Rotations and its Implications to the Strong-Ground Motion.- Strong Motion Rotation Sensor.- High-Resolution Wide-Range Tiltmeter: Observations of Earth Free Oscillations Excited by the 26 December 2004 Sumatra -Andaman Earthquake.- Fiber Optic Sensors for Seismic Monitoring.- Rotations and Engineering Seismology.- Deriving Seismic Surface Rotations for Engineering Purposes.- Effects of Torsional and Rocking Excitations on the Response of Structures.

Ground rotational motions recorded in near‐source region of earthquakes

Large rotational motions excited by earthquakes are recorded during an earthquake swarm of March, 1997, offshore Ito in Izu peninsula, Japan. The largest rotational velocity recorded at KAW is 2.6×10−2 rad/s around the north-south axis during an earthquake of magnitude 5.2 at 14:09 (GMT) on March 3. The rotational motions around the vertical axis excited by the largest event is 3.3×10−3 rad/s. Considering a spatial variation of slip velocity which directly relates to excitation of a rotational motion, we apply a simple point source model to the largest event and succeed in explaining the observed rotational motion around the vertical axis.

Mechanism of Phreatic Eruptions at Aso Volcano Inferred from Near-Field Broadband Seismic Observations

Broadband seismometers deployed at Aso volcano in Japan have detected a hydrothermal reservoir 1 to 1.5 kilometers beneath the crater that is continually resonating with periods as long as 15 seconds. When phreatic eruptions are observed, broadband seismograms elucidate a dynamic interplay between the reservoir and discharging flow along the conduit: gradual pressurization and long-period (approximately20 seconds) pulsations of the reservoir during the 100 to 200 seconds before the initiation of the discharge, followed by gradual deflation of the reservoir concurrent with the discharging flow. The hydrothermal reservoir, where water and heat from the deeper magma chamber probably interact, appears to help control the surface activity at Aso volcano.

Source Process of the Chi-Chi Earthquake: A Joint Inversion of Strong Motion Data and Global Positioning System Data with a Multifault Model

The Chi-Chi, Taiwan, earthquake of 21 September 1999 was a large thrust earthquake that caused disastrous damage. The surface fracture trace runs along the Chelungpu fault, which strikes N5� E along most of the faulting area but turns to the northeast while approaching its northern end. It finally presents a horsetaillike faulting system and bends to an east-northeast direction at the northern end. Both strong-motion and Global Positioning System (GPS) data recorded large displace- ments in the northern part. In this study, we use a multifault model to simulate the observed large strong-motion and GPS data. The results of the joint inversion reveal that in the southern part, slip occurred mainly in the shallow portion, reaching a maximum about 20 m on the ENE fault, where the rupture propagated to the deep part of the fault. There was obviously slip vector rotation from south to north, such that the slip on the branched fault plane was almost purely reverse faulting. The seismic moment, a total of 2.7 10 20 N m, was released in less than 40 sec during the mainshock. Aftershocks occurred mainly in areas where little or no slip occurred. Slip vector rotation can be explained by the complexity of regional tectonic stress. Changes in tectonic stress at the northern end of the fault inhibited the northward propagation of the rupture and favored the rupturing of the ENE fault.

The source of explosive eruptions of Sakurajima volcano, Japan

To understand the nature of explosive eruptions, in particular the role played by exsolution of magmatic gases, we studied the seismic waveforms produced by two such eruptions of Sakurajima volcano on October 30 and November 14,1986. An explosion begins when a mixture of magmatic gases and fragmented magma flows upward through a narrow vertical channel to the surface. The source region is a reservoir of gas-rich magma, where the pressure drops as gases flow out. The form of low-frequency seismic waves can be explained by a source that consists of two components. One component is a single force directed downward that represents the reaction to ejection of gas and magma. The other, more important component is a moment tensor that represents expansion or contraction of rock around a source. The first motion of seismic waves produced by an explosion is directed away from the volcano. Tilt and strain measurements show that the initial motion directed outward is soon followed by subsidence around the volcano. The eigenvectors of the seismic moment tensor are either nearly horizontal or vertical. The ratio of horizontal to vertical eigenvalues was 1.8:1 for the October 30 explosion. This suggests the geometry of the explosive source can be approximated as a buried vertical cylinder. The volume of contraction, 2×l03 m3, can be produced by a pressure change of 0.3 MPa over a cylinder with a radius of 160 m and a height of 300 m. We propose that an explosive eruption occurs after a small shallow source of high pressure forms within a vertical conduit, possibly exsolution of magmatic gases. Eventually, the small source bursts and pushes open a pathway to the surface, providing an upward escape for trapped gases. It is this burst that causes the initial outwardly directed motion at the surface. A deeper reservoir of gas-rich magma is then subjected to a lower confining pressure and also bursts, causing the explosion.

The dynamic rupture process of the 1993 Kushiro‐oki earthquake

We have constructed a dynamic rapture model of the 1993 Kushiro-oki earthquake (Mw = 7.6) that is consistent with the observed seismograms. This earthquake occurred beneath the city of Kushiro, Japan, at a depth of about 100 km, and the aftershocks are limited to a horizontal plane between a double seismic zone in the subducting slab. Excellent near-field strong-motion records are available. In our analysis, a kinematic inversion is followed by a dynamic crack calculation, and this pair of operations is repeated until convergence to the data is obtained. Two different conditions concerning the dynamic stress drop were considered in the dynamic calculation, and a dynamic model was constructed for each condition. The difference in these conditions has little effect on the results. The final dynamic model shows that this earthquake occurred with a localized high stress drop of about 40 MPa in the subducting slab, which indicates that the region between the double seismic zone is strong enough to sustain such a high stress. One of the two high strength regions in the final model lies near the upper plane of the double seismic zone and represents a barrier due to a substantial change of stress state in the slab. In the region where the initial breaks occurred 5 s before the main rupture, both the strength and the stress drop are smaller than in the adjacent region. This implies that before the main rupture the accumulated stress had already been released in the area of the initial breaks. The aftershocks tend to occur near the high strength area; in some of them, the P wave first-motions are opposite to those of the main shock.

Vulcanian eruptions with dominant single force components observed during the Asama 2004 volcanic activity in Japan

On September 1, 2004, Mt. Asama in central Japan erupted for the first time in 21 years. Between this moderate eruption and mid-November of the same year, 4 additional moderate eruptions occurred. We installed 8 broadband seismic stations in addition to the short period seismic network around the volcano and succeeded in recording the near-field seismic signals associated with the summit eruptions. The results of the waveform inversions clearly show that the force system exerted at the source region is dominated by vertical single force components. The source depths of the single force are shallower than 200 m from the bottom of the summit crater, and the order of magnitude of the single force is 1010–1011N. The source time history of each vertical single force component consists of two downward forces and one upward force. The initial downward force probably corresponds to the sudden removal of a lid capping the pressurized conduit. The drag force due to viscous magma moving upward in the conduit can explain the upward force. The correlation between the single force amplitudes and the amounts of volcanic deposits emitted from the summit crater are not necessarily positive, suggesting that the amount of deposits remaining within the summit crater may have played an important role in the excitation of the single force.

Source mechanism of seismic waves excited by pyroclastic flows observed at Unzen volcano, Japan

A number of pyroclastic flows have occurred at Unzen volcano (Unzendake) in Kyushu, Japan, since the end of May 1991. Seismic waves, which are excited by the pyroclastic flows, contain low-frequency components in their initial phases. They were recorded by two long-period seismographs in the seismic network deployed around the volcano. This paper is an attempt to clarify the source mechanism of the seismic waves triggered by the pyroclastic flows. Visual data show that the pyroclastic flow, which excites these seismic waves, occurs in three successive steps: first, fall of a lava dome front (dome collapse), then its collision with the mountain slope and its fragmentation into small blocks, and, finally, descent of the blocks along the slope. We can build a simple mechanical model which gives theoretical time histories of vertical and horizontal components of a force acting on the spot of dome collapse on the mountain slope. The model of a mechanism generating these seismic waves was constructed through analysis of the low-frequency components with the aid of inversion methods which can estimate source time function of the force acting on the source. The estimated time histories are close to those predicted by the simple mechanical model. We conclude that the three steps mentioned above are essential in exciting the low-frequency seismic signals. This idea is supported by the fact that the volume of fallen material estimated from the visual data is proportional to the maximum vertical amplitude of seismic waves triggered by the pyroclastic flow. This relation holds for the two pyroclastic flows of June 22,1991, and May 28,1992, which were studied in this paper. These findings seismically confirm a visual observation that the pyroclastic flow is a result of dome collapse.

The 1993 Kushiro‐Oki, Japan, Earthquake: A high stress‐drop event in a subducting slab

A spatial slip distribution of the 1993 Kushiro-Oki earthquake (Mw; 7.6) is undoubtedly constrained using near-field data and the detailed aftershock distribution. This earthquake has peculiar features in that the slip on the fault plane is restricted within an extremely small area (about 40 km × 20 km) between the upper and lower planes of the double seismic zone in the subducting slab beneath Hokkaido. The total seismic moment, averaged slip, and stress-drop in this area are 3.3 × 1020 Nm, 5.5 m, and 4.2 × 107 Pa, respectively. This earthquake indicates that the region of the slab between the planes of the double seismic zone beneath Hokkaido is strong enough to sustain a stress of about 4 × 107 Pa. The slip motion of this event is consistent with retrograde slab migration along the Southern Kuril trench, and may represent a process by which retrograde motion of subducting slabs could occur.