Hiroyuki Hamaguchi

Degree one heterogeneity and hemispherical variation of anisotropy in the inner core from PKP(BC)–PKP(DF) times

We have investigated the relationship between heterogeneity and anisotropy in the inner core. Heterogeneity was examined using PKP(DF) and PKP(BC) phases traveling in the east-west direction (E-W paths). The geographical pattern of the PKP differential travel time (BC-DF) residuals was found to be a manifestation of degree one heterogeneity in the outermost 100 to 500 km of the inner core. Spherical harmonics analysis showed that the P velocities in the inner core are fast in the region from 43°E to 177°E (the quasi-eastern hemisphere) with a peak value of +0.29±0.15% and slow in the region from 183°W to 43°E (the quasi-western hemisphere) with a peak value of -0.66±0.20%. Regional variation of anisotropy in the inner core was examined using PKP phases traveling nearly parallel to the Earths spin axis (polar paths). The averages of BC-DF time residuals were 2.4±0.8 s in the quasi-western hemisphere and 0.5±0.4 s in the quasi-eastern hemisphere. We found that the inner core anisotropy predominates only in the quasi-western hemisphere in which the 0.66% slow P velocity is observed for the E-W paths. Geophysical implications of the degree one pattern in the inner core were briefly discussed. We propose that the hemispherical structure of the inner core reflects ancient core dynamics immediately after core formation.

Source process of very long period seismic events associated with the 1998 activity of Iwate Volcano, northeastern Japan

We observed very long period seismic events that are associated with the 1998 activity of Iwate Volcano, northeast Japan. The events show a dominant period of 10 s and duration of 30–60 s, often with accompanying short-period waves at the beginning and at the end of the long-period signals. By analyzing the broadband seismograms we find that the source elongates in the east-west direction for ∼4 km at a depth of 2 km beneath the western part of Iwate Volcano. Results of moment tensor inversions show a source mechanism of mutual deflation and inflation of two chambers located at the western and eastern edges of the source region. The source region coincides with the low seismic velocity zone detected by seismic tomography and is very close to the locations of pressure sources estimated from crustal deformation data. On the basis of these results we infer that the very long period seismic events are generated by transportation and movement of magmatic fluid (hot water and/or magma) in a shallow part of the volcano. We further present a simple source model of very long period seismic events based on one-dimensional flow dynamics and propose a new parameter to characterize the size of very long period event: the energy flow rate, which is obtained by dividing the seismic moment by the dominant period. The energy flow rate was estimated as 3.1×1012 J/s for the event on July 29, 1998.

Temporal changes of the crustal structure associated with the M6.1 earthquake on September 3, 1998, and the volcanic activity of Mount Iwate, Japan

We detected temporal changes of crustal structure by cross spectrum analyses of seismic waves excited by two artificial explosions that were carried out about one month before and two months after a M6.1 shallow earthquake. Phase spectrum analysis shows that seismic velocity of the upper crust around the focal area of the M6.1 earthquake and Mount lwate decreased 0.3 - 1.0 % during the three months. Since lager decreases of the seismic velocity were observed at the region close to the focal area and a volcanic pressure source, we conclude that stress change in the upper crust is a most plausible candidate to cause the velocity changes among other candidates. We further apply a coherence analysis to the observed seismic coda to locate regions where short- wavelength heterogeneity temporally changed. The result indicates a possibility of temporal changes of the heterogeneity beneath the volcano, where a magmatic fluid intrusion is inferred.

Earthquake generating stresses in a descending slab

Abstract A numerical study of stress within descending slabs beneath island arcs provides a general interpretation for the occurrence of deep- and intermediate-focus earthquakes. The computation of stress due to several factors, which have been proposed for the interpretation of the double-planed seismicity of intermediate-depth events, shows that thermal expansion and olivine-spinel phase change play an important role in earthquake generation. The stress field due to the combination of these two causes gives the following results; in the depth range 50–300 km, the compressional and tensional stress predominates in the upper and central part of the slab respectively. Maximum shear stress amounting to 1.6 GPa in compressional and 0.4 GPa in tensional fields occurs at a depth of 200 km. The stress is generally very small at around 300–350 km. Below 400 km, compressional stress exists at the central part of the slab and tensional stress appears near the upper and lower surfaces of the slab. Maximum shear stress is 2.8 GPa in compression and 1.6 GPa in tension at a depth of 500 km. The principal axes in these high stress fields align almost parallel to the dip direction of the slab. These stress features explain well such observed seismological facts beneath island arcs as the seismicity-depth relationship, planer configuration of seismic zone, existence or non-existence of double-planed seismicity and focal mechanisms of intermediate and deep focus earthquakes.

Crustal deformation associated with the 1998 seismo-volcanic crisis of Iwate Volcano, Northeastern Japan, as observed by a dense GPS network

Mt. Iwate (2,038 m) is an active volcano located in northeastern Japan. Unrest of the volcano started in September, 1995 with intermediate-depth tremors. The shallow seismicity gradually became active in February, 1998, accompanying the notable crustal deformation observed by a dense GPS network. The pattern of the horizontal displacements is characterized by radially directing outward from the volcano. We estimated the source position by inversion analyses for every two-months period, assuming two models; a point pressure source (Mogi model) and a tensile fault. The comparison of AIC’s for the two models indicates that the latter is proper from February to April, while the former is preferable afterward. The tensile fault was located at about 5 km WSW of the summit and 3 km in depth, then a Mogi source was estimated at the western neighbor of the tensile fault in the successive period and moved westward as far as 10 km W of the summit with shallowing its depth. It should be noted that the seismic area also expanded westward in the same period. This synchronicity suggests that the both phenomena were caused by a movement of magma from the deeper part beneath the summit to the western shallower part.

Scaling law of volcanic explosion earthquake

The aim of the present study is to clarify a scaling law for volcanic explosion earthquakes excited by a counter force of eruption. Based on the data of explosion earthquakes at three volcanoes, we derived the following results: The peak amplitude of the single force is proportional to the square of the pulse width of source time function. The initial pressure stored in a shallow region beneath a volcano before eruption is 1 MPa with a perturbation of one order. The kinetic energy of explosion earthquake is proportional to the cube of crater radius. The scaling law of the explosion earthquake is quite similar to that of the earthquake caused by a faulting. It is possible to predict the strength of explosion earthquake by measuring the crater radius beforehand.

Broadband Source Process of the 1998 Iwate Prefecture, Japan, Earthquake as Revealed from Inversion Analyses of Seismic Waveforms and Envelopes

An earthquake of M 6.1 occurred on 3 September 1998, along an active fault at the southwestern foot of Mt. Iwate, a volcano in northeastern Japan. Acceleration records of this earthquake were obtained at seven stations within 40 km of the epicenter. In order to investigate the source process of this earthquake in a broad frequency range, we simultaneously conducted inversion analyses of low-frequency seismic waveforms and high-frequency seismogram envelopes. First, executing the envelope inversion by using the envelope Green function derived from the radiative-transfer theory in the high-frequency band of 2-16 Hz, we estimated the spatial distribution of seismic-wave energy radiation on the fault plane of 10 km × 10 km. We found that seismic-wave energy was strongly radiated from the southwestern deeper part of the fault plane. By using data from the same stations and using the same fault geometry as the high-frequency analysis, we applied a waveform-inversion method to three-component displacement records in the low-frequency band of 0.1-0.33 Hz. This result showed that seismic moment was mainly released at the shallow part of the fault. Comparing these results, we found that high-frequency energy was strongly radiated from the deepest periphery of the region, where seismic moment was mainly released. This result implies that the radiation of high-frequency energy was associated with the arrest of rupture for this earthquake. Manuscript received 21 April 2001.

Migration of seismic activity during the 1998 volcanic unrest at Iwate volcano, northeastern Japan, with reference to P and S wave velocity anomaly and crustal deformation

Abstract We describe the seismicity at Iwate volcano, northeastern Japan, during the volcanic unrest of 1998 with reference to a three-dimensional P and S wave velocity model from tomographic analysis. The abnormal seismic activity beneath Iwate volcano started under the caldera in February, 1998 and migrated westward in the period February to August, 1998. Previous geodetic modeling [Sato and Hamaguchi, Chikyu Monthly 21 (1999) 312–317] suggested the growth of a dike in the time of the seismic activity. Comparing the seismicity and dike extension with the tomographic images of the P and S wave velocity structure, we find that the trace of the growing dike coincides with the region of the high Vp and high Vp/Vs ratio beneath the volcano. The seismic and geodetic data are consistent with an intrusion of magma or other fluid under the caldera in 1998. Another pressure source causing the predominant crustal deformation at Iwate volcano was detected from geodetic data, which was located in the region with high Vp/Vs ratio under the western end of the volcano through the period from February to August. It is suggested that the activation of the point pressure source probably associated with the inflation of a hot fluid reservoir relate to a geothermal region adjacent to the western edge of the volcano.

Process of the 1977 Nyiragongo eruption inferred from the analysis of long-period earthquakes and volcanic tremors

Abstract The lava lake present in the summit crater of Nyiragongo from 1928 to 1977 completely disappeared after the short-lived eruption of January 10th, 1977. The sequence of events during this eruption was elucidated by analysis of seven, long-period earthquakes (LP events) and volcanic tremors recorded at stations LWI, NAI and BUL. The seven LP events observed were composed mainly of Rayleigh waves with large amplitudes and were classified into two groups: one containing three and the other containing four events, based on waveform similarities. It was determined that the former events were excited by the vertical downward single forces and the latter events by the vertical upward ones. The short-period seismograms recorded at LWI and NAI showed that continuous volcanic tremors began roughly 6 min before the occurrence of the first LP event. After the appearance of seven burst-type tremor activities, tremor amplitudes suddenly increased and then gradually decreased. Taking the analytical results of seismic events and observed surface activities into consideration, we inferred the process of the 1977 Nyiragongo eruption to be as follows: 1. (1) quick draining of the liquid lava through the fissures, and disappearance of the long-lived lava lake (less than 6 min); 2. (2) large terraces collapsing onto the bottom of crater, causing the first three LP events during the next 8 min; 3. (3) intermittent vesiculation of lava at the top of magma chamber, exciting the last four LP events during the next 8 min; 4. (4) magma flowing in deeper parts of the conduit, generating the intense tremors measurable at NAI and lasting for roughly 30 min.

Broadband seismic signals associated with the 2000 volcanic unrest of Mount Bandai, northeastern Japan

Abstract We have observed volcano-tectonic (VT) earthquakes, volcanic tremor and very long-period events (VLPEs) associated with the volcanic unrest of Mount Bandai since 2000 using a broadband seismic network closely deployed around the volcano. VT earthquakes are characterized by high-frequency (>10 Hz) signals with clear onsets of P and S phases, whereas volcanic tremor is characterized by a long coda comprising frequencies ranging from a few hertz to more than 10 Hz. In contrast, waveforms of VLPEs consist of very long-period (∼10 s) signals preceded by short-period (