An alkali-rhyolite tuff-ring is newly identified in the western end of the Oga Peninsula and named as Toga volcano in this paper. The existence of this maar-type volcano at the Toga Bay has been suspected for a long time because of the elliptical embayment reminiscent of a maar and the distribution of the Toga Pumice localized along the bay coast. The Toga Pumice is cornposed mainly of pumice and non- to poorly-vesicular glass shards, but many pumices of lapilli size are rounded and fines are poor giving a sandy epiclastic appearance to the deposit. In our latest survey along the bay coast, the Toga Pumice is found to be in direct contact with the basement rocks. The contact steeply inclines at 40-50° and envelopes an elliptical area 2.0 km×2.4 km covering the bay and bay coast to form a funnel-shape structure. The basement rocks at the contact are brecciated to a depth of several tens of centimeters, or collapsed into fragments to be contained in the Toga Pumice. The beds inside the inferred crater incline toward the center of the crater at 10-30° or much smaller angles, presumably reflecting a shallow concave structure infilling the more steeply sided crater. The deposit is thinly to thickly bedded to be parallel- to wavy- or cross-stratified, inversely to normally graded with many furrows, rip-up clasts and load casts, and is sorted as well as fines-depleted pyroclastic flow deposits and/or pyroclastic surge deposits. These features are characterisitic to turbidites and indicate the place of emplacement was filled with water. Constituent glass shards are, however, commonly platy or blocky and likely to be phreatomagmatic in origin, and pumice lapilli are interpreted to have been originally angular but rounded by repeated entrainment and abrasion in multiple phreatomagmatic eruptions and succeeding emplacement in the crater lake. A pyroclastic surge deposit (Oga Pumice Tuff) correlative in composition and age to the Toga Pumice occurs at Anden and Wakimoto, 11 km and 15 km east of Toga, respectively. The juvenile pumice lapilli are angular to subrounded, in contrast with the pumice lapilli of the Toga Pumice.

抄録全体を表示

Recent research by Kano et al. (2002) confirmed the existence of a maar-type crater in and around Toga Bay. They named it Toga volcano. In this paper we reveal features of the above crater by analyzing its gravity anomaly. For this purpose we combined gravity data from two sources; AU (Akita University) and SK (Sekiyushigen-Kaihatu Co.). AU data was acquired mainly in 1981 and additionally in 1987. SK data was acquired for oil exploration in 1957. For this publication all data were reprocessed into a unified system. An isolated negative anomaly which was clearly detected on the local anomaly map was interpreted as evidence of low density pumiceous sediment filling the crater. Peak value of negative anomaly is -5.7 mGal. Basement is mainly volcanic rocks. A two-layer structure was assumed in the inverse analysis, which is based on iterative accommodation. On the assumption of 0.6 g/cm³ in density difference, the crater was analyzed as about 380 m in maximum depth and about 1.8-2.2 km in diameter. The mass deficiency was estimated as 2.9×10⁸ ton by Gauss's theorem. Possible variations of the structure are discussed in relation to allowable modifications of the density difference. Some related comments are given on the effects of measurement error and ill-balanced distribution of gravity stations.

抄録全体を表示

A three-dimensional density structure in the shallower crust beneath the Hida Mountains, central Japan, is estimated by using gravity data. We employed seismic velocity structure models both as an initial condition and constraints of gravity structure models. We estimated that an extremely low-density body (density is smaller than 2.1 g/cm³) exists at 4-8 km depth (2 to 6 km below the sea level) beneath Mt. Tateyama (3,015 m) and also along the Hida mountains. Estimated horizontal extent of the body is about 14 km in east to west, about 28 km in north to south directions, respectively, and about 4 km in thickness. The volume of the body is about 1,000 km³. Spatial distribution of the extremely low-density body is well consistent with a low velocity region estimated from studies of seismic tomography.

抄録全体を表示

Erupted magma of the 2000 eruption of the Miyakejima volcano changed from basaltic andesite to basalt during the caldera formation, from aphyric basaltic andesite with SiO₂=54 wt.% to plagioclase-phyric basalt with SiO₂=51.5 wt.%. Whole-rock compositions of the basaltic andesite of the June and July eruptions are plotted on the extension of the temporal variation of the previous eruptive materials, suggesting that the andesitic magma erupted in June and July eruptions were driven from the magma system worked for the last 500 years. Petrological character of the basalt in the eruptive materials of August, by contrast, is different from the previous lavas of the Miyakejima volcano. This shows that a new basaltic magma ascended to the shallow magma system after the caldera collapse. Identical ratio of the incompatible elements among the eruptive materials of the 2000 eruption and the recent eruptions suggests that they were driven from a common parental magma.

抄録全体を表示

Directions, intensities and stability of the natural remanent magnetization of the Yatsugatake Volcanic Chain, central Japan were measured by the conventional paleomagnetic method. We investigated about 600 samples of the volcanic rocks with ages ranging from 0.01 to 1.3 Ma together with basement rocks younger than 3 Ma collected at 130 sites covering the entire period of the Yatsugatake activity. Total 320 reliable paleomagnetic directions from 76 sites were determined using the Zijderveld diagrams for progressive alternating-field demagnetization. Paleomagnetic polarities thus obtained are correlated to the standard magnetic chrons and subchrons with the aid of K-Ar radiometric ages of rocks. Results are summarized as follows in the order of decreasing ages; (1) Kiretto andesite and Minoto porphyrite representing the earliest volcanic products have the normal polarity correlatable to a subchron older than the Jaramillo event in the Matuyama reversed polarity chron, whereas their basement (Shiga welded tuff, Yashigamine lava and Mizuochi-kannon lava) has reversed polarity. (2) Kasuga volcanics, Kyowa-bokujyo lava, Tatamiishi lava and Honmagawa lava composing the lower layers of the old Yatsugatake stage are all reversedly magnetized and correlatable to the Matuyama reversed polarity chron. (3) Most of the upper layers overlying the Kasuga volcanics have the normal polarity corresponding to the Brunhes normal polarity chron. (4) Reversed polarities are found in several lava flows of the younger volcanic stages. They appear to be records of the Blake event and the Laschamp excursion in the Brunhes chron. (5) Most of the paleomagnetic directions of the Yatsugatake samples are clockwisely deflected by approximately 10° to 15° from the present geographical north. These deflections seem to be caused by the tectonic movement of the Fossa Magna region, since similar deviations of the paleomagnetic directions are also observed in the Quaternary volcanics of the western Izu Peninsula adjacent to the Yatsugatake region.

抄録全体を表示

To investigate possibilities for ‘forecasting’ the potential for explosive eruptions at active lava domes a method using time-series analysis of satellite shortwave infrared images was studied. The rational behind this is that the potential ‘explosiveness’ of an active lava dome seems likely to have a close relationship to the state of gas accumulation in the volcanic edifice, which can be inferred from the correlation between the discharge rates of magma and gas, i.e. the gas/magma discharge rate ratio. To assess this we used ten night-time Landsat 5 Thematic Mapper images of Unzen Volcano, taken between October 1991 and January 1993 during Phase 1 of the volcanoes 1991-1995 eruption. Our hypothesis is that the surface thermal state of the endogenous and exogenous dome growth areas is at least partly controlled by the magma discharge rate, whilst the thermal state of the fumarolically active areas is similarly controlled by the discharge rate of magmatic gas. If this hypothesis is correct then we can indirectly infer the gas/magma discharge rate ratio from analysis of the thermal anomalies detected on the satellite infrared images. In the Unzen time-series, the size of the thermally radiant areas in the dome growth and fumarolic areas were significantly positively correlated (r²=0.56, n=10). This suggests a mechanism for effective gas release existed at the volcano over the time-period studied, with the rates of magma and gas discharge varying in proportion. Through this mechanism gas was efficiently vented from the edifice and major gas accumulation prevented, thus helping to minimise the number of large explosive events that occurred during the Unzen eruption.

抄録全体を表示

Unzen eruption starting from 1990 prompted us to put as much constraint as possible in the velocity structure of whole Shimabara peninsula, and we re-analyzed a controlled source seismic experiment data obtained on a seismic line with the strike of N-S in 1986. An airgun was shot repeatedly as the controlled source at Port Kuchinotsu, Bessho dam and Port Taira. These shots were recorded by 114 stations. A 2-D velocity model to a depth of 20 km was obtained by the 2-D ray tracing technique and reflectivity method using not only the traveltimes and the amplitudes, but also a relationship of the amplitudes versus the offsets from each shot position to each receiver station. We identified four layers beneath the Shimabara peninsula, (1) an uppermost layer (V_p=2.6-3.3 km/s), (2) a basement (V_p=3.6-3.9 km/s), (3) an upper crust (V_p=5.9-6.3 km/s), and (4) a lower crust (V_p=7.0-7.3 km/s). The upper crust consists of two parts, one is the upper part with large vertical velocity gradient (0.1s^-1) and the P-wave velocity of 5.9-6.3 km/s, and the other is the lower part with almost constant velocity of 6.3 km/s. The lower crust underlaying the upper crust also consists of two parts. The P-wave velocities of the upper and lower parts are inferred to be 7.0 km/s and 7.3 km/s from amplitude analyses of the reflected waves. Beneath the central part of the peninsula, the upper crust becomes thinner and the lower crust swells. The boundary between the upper and the lower crusts is located at a depth of 9-11 km. The bottom of the lower crust with high velocity contrast is located at a depth of 19-20 km.

抄録全体を表示

Spatial distribution of b value in the frequency-magnitude relation is investigated for upper plane events of the double-planed deep seismic zone beneath the northeastern Japan arc, Locally high b value areas are detected at three locations west of the volcanic front. Inclined low S-wave velocity zone in the mantle wedge has extremely low values at locations just above these three high b value areas in the subducted Pacific slab. The close spatial relation between S-wave velocity distribution and earthquake size distribution suggests along-arc variation of magma source within the slab beneath this volcanic arc.

抄録全体を表示

The volcanic front in the Southwest Japan Arc formed relative to the subducting Philippine Sea Plate. In the west Chugoku district, the volcanic front changes direction from E-W to NNE-SSW. Twenty-two Quaternary lava domes in this area can be topographically classified into two arrays (East and West arrays) based on the degree of dissection. K-Ar age results show that domes of the East array formed earlier (0.6-0.3 Ma) than those of the West array (0.2 Ma-recent). The two arrays in different ages may result from the migration of volcanic front caused by westward migration of the PSP slab.

抄録全体を表示

The three-dimensional P- and S-wave velocity structures around Usu volcano, Japan are investigated using traveltime data of earthquakes during the eruptive activity in 2000, in order to clarify magma plumbing system and mechanical properties of host rocks which can affect magmatic activities. A tomographic inversion is started from a three-dimensional initial model based on a density structure inferred from a gravitational study, because strong lateral heterogeneity is expected in this region and final results can depend on the initial model. As a result, the following features are obtained. (1) Deepening of upper crustal layer from the southern coast of Lake Toya to Uchiura Bay at depth of 2～4 km is found for P- and S-wave velocity structures. This feature is consistent with density and apparent electrical resistive structures revealed by previous studies and the layer is interpreted as identical to Pre-Neogene basement by comparing with surface and borehole geological information. (2) Precursory earthquakes and earthquakes after the beginning of the eruption in 2000 mainly occurred within the Pre-Neogene basement or along the uppermost depth of the basement. This implies the seismic activity was strongly constrained by the structure. (3) V_p/V_s of greater than 2.0 is detected at the surface level and the V_p/V_s decreases with increasing the depth. The decrease of the V_p/V_s within Quaternary and Neogene layer can be due to the reduction of the porosity. A fairly small value of about 1.5 is obtained at depths greater than 4 km within the Pre-Neogene basement. The low V_p/V_s is possibly explained by the existence of a quartz-rich rock such as quartzite or hot-water under pressure and temperature condition near water-steam transition.

抄録全体を表示

Usu volcano has erupted at 13:07 JST on 31st of March, 2000, about 23 years after the 1977 eruption. Many precursory earthquakes occurred since March 27, as in the case of the former eruption. In order to reveal characteristics of these precursory earthquakes, we performed temporal strong-motion observations at two stations near Usu volcano for the period of March 29 to April 1. Wide band and wide dynamic range accelerometers were installed and more than 300 earthquakes (M〓4) were obtained by them. Low frequency earthquakes are picked out based on frequency contents of seismic waves. The number of the low frequency earthquakes is 51, that is much less compared with that of the high frequency earthquakes. The low frequency earthquakes having a dominant frequency of 0.6 Hz show a complex waveform and long duration in comparison with the volcanotectonic earthquakes. At least eight earthquake families are found among the low frequency earthquakes. Although it is somewhat difficult to identify P phases of these earthquakes, hypocenters of about half of them are determined based on 1D velocity structure and a few P-wave arrival times. Most of the epicentres are scattered around the new craters, Nishi-yama and Kompira-yama, but some of them are located around Meiji-Shinzan, east of the craters. The focal depths are less than 3 km. Activity of these earthquakes makes a peak in and around the eruption in contrast to the peak activity of volcanotectonic earthquakes on 30th of March, one day before the eruption. We performed the moment tensor inversion to determine the source mechanism of a part of earthquake families. It seems that the result of the inversion can be explained by a tensile-share crack model and that the source process is complicated.

抄録全体を表示

There was intense acoustic activity associated with the eruption of Mount Usu, which began on March 31, 2000. Repetitive phreatic explosions generated many isolated infrasonic signals, which were observed at plural acoustic stations. During the periods when acoustic activity was high, infrasonic pulses as many as 200 were identified every 10 minutes. Source location of infrasonic signals could be well identified from the records of the low frequency microphone network. Two active craters, Nishiyama and Konpirayama craterlets, are clearly distinguished by sound source determination analysis in spite of their proximity (～1 km). To investigate temporal change in the acoustic activity from April to June, 2000, we developed a method to detect arrival and amplitude of infrasonic signals automatically. The number of automatically identified infrasonic signals exceeds 1.46 million during three months. It seems that there is a good correlation between acoustic activity and seismic signal amplitude. Patterns of acoustic activity and infrasonic pulse shapes observed at Usu volcano are very similar to those observed at Stromboli volcano, Italy. We named the acoustic activity attended with phreatic explosion which scatters a lot of clods, “mud-pool Strombolian type”. Phreatic explosion excites not only infrasonic pulses but also seismic signals observed before the arrival of an infrasonic pulse. Existence of a Rayleigh wave phase with a large amplitude suggests that the seismic wave is excited at a shallow part of the crater.

抄録全体を表示

A helicopter-borne electromagnetic survey (HEM) with simultaneous measurements of the Earth's magnetic field was conducted to better understand the subsurface structure of Usu volcano, Hokkaido, Japan, with a special reference to the 2000 eruption. The HEM system is a modified DIGHEM-V model that is operated at five frequencies ranging from 220 Hz to 137,500 Hz with five-fold frequency increments. The survey was flown in late October 2000 at an altitude of 70-100 m above terrain along north-south survey lines and east-west tie lines, spaced 100 m and 1,000 m apart, respectively. A diiferential GPS system was employed for flight path recovery. The observed electromagnetic data were processed and apparent resistivity maps were created for each five frequencies. Cross-sections of apparent resistivity profiles were also calculated from an interpolation of apparent resistivity data at five frequencies. Aeromagnetic anomalies were reduced onto a smoothed observed surface and a reduction-to-the-pole anomaly map was created. The characteristics of the distribution of apparent resistivities and magnetic anomalies were summarized as follows:

抄録全体を表示

A continuous GPS observation network with 5 stations using single frequency receivers around Usu volcano was installed two days before the 2000 eruption, which took place at the western part of the volcano on 31 March 2000. The cumulative displacements in the radial direction from the center of the volcano reached 1.2 m before the eruption at station HKG located north of the central Usu volcanic massif and 2.3 m at station KNP located 500 m northeast from one of the new craters formed by this eruption. Both displacements were monotonously increasing with time. However at station IZM located closest to the crater, the displacement was only 0.2 m in the NW direction up to the noon 30 March and then changed its direction of SE. After the eruption till the end of April, the cumulative displacements at KNP and IZM exceeded about 1.8 m and 8.9 m, respectively, while the small amount of about 0.2 m was observed at HKG. New 20 GPS stations were installed by early May around the Nishiyama crater in order to monitor the remarkable doming phenomena which started near Nishiyama crater after the eruption. The observation up to the end of May 2000 shows displacements at all stations characterized by both upheaval and outward radial direction from the Nishiyama crater. By August 2000 the displacements at all stations had turned in the direction towards the Nishiyama crater and started to subside.

抄録全体を表示

The eruption of Mt. Usu started in its north-western part on May 31, 2000. The first eruption occurred at the western flank of Mt. Nishiyama, which was named “Nishiyama vents group”, where many small craterlets appeared. In this local area the cryptodome has been growing since the first eruption in 2000. We observed the crustal deformation in this area by a theodolite after the first eruption from May to September, 2000 and June, 2001. The theodolite measurement revealed that the skyline in Nishiyama vents group had risen until July, 2000. The accumulated uplift around N-B (Nishiyama-B) crater could be as high as 5.5 m from May 11 to August 23. The rate of the upheaval was decayed gradually, and turned to a subsidence in August, but subsidence rate became smaller gradually after summer, 2000. We estimated the amounts of inflation volume in three periods, in May, June and July to August, 2000. The inflation rates were 74,000 m³, 41,000 m³ and 8,000 m³ per day, respectively. Their pressure sources are estimated to be located 30(h400 meters beneath N-B crater that is the main crater in Nishiyama vents group.

抄録全体を表示

From 31 March 2000, a series of phreatomagmatic and phreatic eruptions occurred at Usu volcano, located at the southwestern Hokkaido, northeast Japan. Surface ruptures, such as ground tilting, fault, crack, damage of structure, and surface muss movement occurred near and around the volcano from just before the initial eruption, and their formation continued for three months. Characteristics of surface ruptures are subdivided into compressional, tensional, and strike-slip regimes, based on by their deformation styles. Compressional deformations are most dominant in western flank of the volcano, and characterized by, buckling of kerbs, shortening of V-shape drain, asymmetrical pressure ridge of asphalt, and ground tilting. Tensional deformations consist of graben, ground subsidence, and open crack at road. Strike-slip deformation had appeared regionally. These are right-lateral or left-lateral strike slip faults, en echelon clacks and caused horizontal offset of tunnels. Four main zones and four sub-zones are recognized based on styles and superiority of the deformation. Compressional deformations are dominant in Zone 1-3. On the other hands, strike-slip and tensional deformations with ground uplift are remarkable in Zone 4. Zone 1 is characterized by severe compressional deformations. Eespecially, at far side from the volcano in Zone la, bending of rail, damages of house and bridge, and obvious ground uplift marked "Deformation Front." In this zone, first defonuations had appeared between 30 March evening and 31 March morning, prior to the first eruption at 31 March afternoon. Deformation rate decreased rapidly until the middle of April, and obvious deformation continued until the beginning of June. Although Zone 2 is also characterized by severe compressional deformations, deformations are more intense toward the volcano. Deformation rate was still in high level at the middle of June, and continued until July. In Zone 4, strike-slip fault and normal-fault grabens had grown from the last of March-April to June-July. Deep ground block (at least 60 m under the surface) also deformed by horizontal ground movement.

抄録全体を表示

The temporal variation of the ground deformation in case of the 2000 eruption of Volcano Use has been detected by the GPS observation conducted by Geographical Survey Institute, Geological Survey of Hokkaido and Hokkaido University. The process is modeled considering a system composed of magma intrusion as a source and crustal structure as a complex medium applying the theory of the intercellular response of the nerve cell. The system response indicates the time evolution of apparent ground deformation started at about 9 o'clock in the morning of March 29, 2000 and lasted until the time of the eruption on March 31. The deformation continued as the relaxation process with a decay time constant of 1.2 to 1.5 days and never has been excited again. This gradual relaxation harmonizes with our repeated gravity measurement surrounding Volcano Usu. The deformation in the vicinity of the Nishiyama crater lasted more or less longer until April 2. After that, the relaxation of the deformation with a decay time constant of 3 to 3.5 days has been observed near the crater region, suggesting two different temporal variations of the deformation processes.

抄録全体を表示

Mt. Usu began erupting Mar. 31, 2000. The activity of the eruption during the first two to three days has about a half day period. I construct a dynamical model which conforms the periodic nature and other observations. The physical process of this model is the formation of a low density layer by vesiculating the medium over the intruded magma, which causes Rayleigh-Taylor instability. The instability develops to form a conduit and results in the outflow (eruption) of the vesiculated material. The mechanical responses of the related materials (the magma and the surrounding medium, i.e., ground) are assumed to be that of viscous fluid. The model so constructed is a dynamical system consisting of a coupled first order non-linear differential equations with two unknowns, the radius of the conduit and the thickness of the vesiculated layer. The calculated outflow rate have the characteristics of the dumped oscillation, the period and dumping of which depend on the parameters such as the viscosities and a threshold pressure to keep opening the conduit. The oscillation is identified as that of the observed eruption activity with period of about 12hours. Adjusting parameters so as to have the same period as the observed, we obtain rather low viscosity of the ground of 10^9 Pas and horizontal dimension (radius) of the intruded magma of 20 m.

抄録全体を表示

Usu Volcano cornmenced the first eruption with a phreatomagmatic explosion at the western foot of the volcano on 31 March 2000. The total eruptive products during the March 31 eruption attained to 2.4×10⁵ tons, which was the largest among the numerous number of eruptions in the following months. In this paper, transport and depositional processes, aggregate structure and whole deposit grain size population were investigated for the March 31 tephra of the phreatomagmatic eruption. Each sample of March 31 tephra dispersed to NE, reaching to Sapporo 80 km away from the vent, was analyzed in grain size. The grain size distribution of each sample was separated into a, b and c subpopulations, which are characterized by Log-normal distribution. The subpopulation-a (in sand and granule) decreases gradually in mean grain size with distance. Even in proximal area, however, subpopulations-b (coarse silt) and -c (fine silt to clay) occupy about 30 wt.%. Subpopulations-b and -c tend to increase gradually with distance. This is because the tephras fell out mainly as aggregate ash, as eyewitnessed on the day of the eruption and observed in the deposits.

抄録全体を表示

The volcanic activities of the Usu 2000 eruption were monitored and reported by many scientists and the mass media. Summarizing these observation results, most of relatively large explosion events occurred during March 31 to April 7, 2000. Around the Nishiyama and Kompira crater groups, the pyroclastic deposits with multi units can be divided into 19 layers on the basis of their visible color, grain size and sedimentary structure; from Layer A to Layer S in ascending order. The eruptive dates of each layer inferred from the wind directions, the eruptive sequence, and the distribution of deposits are summarized as follows; the Layer A, characterized as the light gray color ash fall deposit including in some pumice layers, was generated by March 31 phreatomagmatic explosions occurred at Nishiyama crater group. The Layer B, composed poorly sorted breccia and ash layer with gray color, was generated on March 31 p.m. at Nishiyama craters. The Layer C to the Layer G, dark brown-gray aggregate ash, were derived from the volcanic eruptions occurred on April 1 to 2 in Nishiyama and Kompira crater group. The Layer H to Layer M and Layer O, mainly consist with gray and reddish brown aggregate ash including in lithic fragments, were generated during April 3 and 4 in Kompira crater group. The Layer N, which distributes around N19 crater, generated on April 4. The Layer P, massive ash with gray color, was generated on April 6 in Kompira crater group. After April 7, the Layer S, characterized as light brown aggregate ash, has been generated from the recent minor activities around limited craters. The amount of Layer A fallen in the range from the source to Toyako Onsencho Town is estimated at 1.2×10⁸ kg, and total amount of Layer A including in the distal area is 2.4×10⁸ kg. On the other hand, amount of other deposits generated during April 1 to 6 (e.g. Layer B, N, and Q) is an order of 10⁶-10⁷ kg. Total amount of the pyroclastic deposits erupted from the Usu 2000 eruption is more than 6.4×10⁸ kg.

抄録全体を表示

Events of the March 31, 2000 phreatomagmatic eruption at Usu Volcano can be subdivided into two stages: (1) large-scale black plume stage having 6 major events from 13:07 to 15:28, and (2) relatively small-scale, black plume and jet-type plume stage having 5 major events from 15:28 to 17:07. The total thickness of the March 31 deposit are 100 cm at a site 90 m north from the vent, 9 cm at 550 m NNE from the vent, and 2.5 cm at 1,030 m NE from the vent. The March 31 deposit at the proximal site (90 m north) comprises 10-70 mm thick 20 layers, consisting of sandy coarse 11 layers and sandy to silty matrix-supported 9 layers. The 20 layers at the proximal site can be correlated with each eruption event observed from 13:07 to 17:25 on March 31. In the proximal deposit, essential white pumice ratio is high in the early stage (42%), then the ratio decreases toward the final stage (up to 21%). Essential pumice fragments are white and subrounded, and contain many 0.05-0.2 mm vesicles. The pumice is relatively poorly vesiculated. Grain-size distributions of all layers in the proximal deposit show bimodal distributions, with peaks at -0.5φ to 1φ and 6.5φ to 7.5φ. Poorly-sorted muddy layers are observed in post April 1 deposits. These layers formed during repeated phreatic explosions associated with abundant mud-clast spattering. The proximal area data show that the discharged mass of the March 31 deposit was 2.2×10⁸ kg in total and 1.0×10⁸ kg in the proximal area to 8 mm isopach.

抄録全体を表示

The March 31 eruption was the first and largest event in the 2000 eruption of Usu volcano. The eruptive products (ash and pumice) contained considerable amount of accidental or accessory fragments, however, we identified the essential materials among the products. Firstly, we concluded that the pumice (Us-2000pm) is essential, rather than reworked material of the past pumices. It is because Us-2000pm contains magnetite phenocrysts whose Mg/Mn ratio and Al component are different from those in any other historic pumices of Usu volcano. Secondly, we found microlite-rich vesiculated fresh glass (Us-2000g) in the ash and concluded that it is also essential. It is because magnetite in Us-2000g is compositionally identical to that in Us-2000pm. Therefore, the March 31 eruption was phreatomagmatic, rather than phreatic. The 2000 eruptive product (Us-2000pm) has similar petrological feature to the historical ejecta, especially the 1977 eruptive product. Zoning patterns of plagioclase phenocrysts in Us-2000pm represent crystal growth since the 1663 eruption, as well as in other historical ejecta. This means that the 2000 magma was derived from the same magma-feeding system of the historical eruptions, and that the magma-feeding system has existed since the 1663 eruption.

抄録全体を表示

Study on products of phreatomagmatic eruption on 31 March 2000 at Usu volcano gives insight into timing of bubble nucleation and gas escape from system during magma ascent from the reservoir. The essential fragrnents (pumice and micropumice) vary in vesicularity which is negatively correlated to water content in the rhyolitic groundmass glass. This implies that variable vesicularity resulted from different degree of water exsolution from melt to bubbles, rather than different degree of gas escape. The difference in water exsolution probably reflects the different timing of quenching, where micropumices were quenched in various depths by the aquifer whereas pumice was not quenched. Number density of bubbles in essential fragments changes with increasing vesicularity; it once increases with vesicularity and then decreases. The vesicular texture indicates that the decrease of the number density resulted mainly from bubble coalescence. Bubble size distributions (BSD) for less vesiculated micropumices are composed of two linear parts in log-linear plot. The slope and intercept values for smaller bubbles are higher than those for larger ones, indicating that nucleation rate has increased during a continuous nucleation. This is in accordance with the change in BSD with increasing vesicularity; number of small bubbles increases rapidly, while that of larger bubbles is nearly constant. Because micropumices were quenched around the aquifer, nucleation rate is inferred to have increased before magma's reaching there.

抄録全体を表示

We studied high-frequency earthquake swarm associated with the eruption of the Usu volcano in 2000 using the data observed by a national seismic network in southern Hokkaido. To get a precise hypocenter location, we applied the double-difference method and station correction to hypocenter determination. Systematic shift of epicenters possibly caused by heterogeneous velocity structure of the upper crust is needed to be consistent with the initial motions of the seismograms at the nearest station. Concentration of hypocenters under the northern flank of the volcano in the initial stage suggests that the magma started its activity at about 5 km in depth at the region. Concentric expansion of swarm area occurred before the eruption and formed doughnut pattern of which center is located near the summit of the volcano. Doughnut pattern may represent relaxation of stress under the volcano which is caused by magma movement and pore pressure change under the volcano.

抄録全体を表示

The eruption of Usu volcano started on 31 March, 2000, caused remarkable crustal deformation. In order to monitor the on-going crustal deformation around the volcano, the Japan Meteorological Agency deployed a new quasi-real time GPS system. The system includes 19 GPS stations and one analysis center. Each station consists with a cellular-phone and an single frequency GPS receiver driven by solar batteries. All the data from each station were automatically transferred to the analysis center every 2 hours. The relative co-ordinates of all stations were immediately calculated and displayed on the screen at the center. Using this system, we could detect a crustal deformation indicating a gradual decay of the magma activity of the Usu 2000 eruption on a quasi-real time base.

抄録全体を表示

Remarkable crustal movement occurred near the eruption area in the Abuta town associated with the 2000 eruption of Usu volcano, Japan. In order to clarify the regional crustal movement, we measured the intervals between the poles of aerial cables two times after the eruption. As a result, we observed significant shortening of the intervals near the eruption point. Distribution of the interval changes of contraction or expansion suggested heterogeneity of the crustal movement due to several kinds of magma movements at different magma reservoirs.

抄録全体を表示