Volcanoes are mechanical as well as thermal anomalies in the Earth's crust. Because of their weakness, volcanoes may begin to deform earlier than the surrounding region in both volumetric and shear strain. A possible example of volumetric strain is the 1974 eruption of the Izu-Oshima volcano, which started two days earlier than the 1974 Izu-Hanto-Oki earthquake (M6.9). The location of the volcanoes and the earthquake, and sense of the faulting permit this interpretaion. Shear strain may also be concentrated in volcanoes, deforming them aseismically. The aseismic deformation of volcanoes may create stress concentration in the surrounding more brittle region, where small earthquakes will occur. They occasionally trigger a larger earthquake farther away from the volcano and are called foreshocks. This probably was the case for the foreshocks and main shock (M7.0) of the 1978 Izu-Oshima-Kinkai earthquake. Active faults having an active volcano at one end may be result of the recurrence of the above mechanism. Such examples include the Tanna and the Hirayama faults to the south and north of Hakone volcano and the Atotsugawa fault to the west of Tateyama volcano. Earthquake epicenters during the 1986 summit eruption of the Izu-Oshima volcano had a similar distribution to those during the period prior to the eruption, in that the summit region was rather aseismic and the epicenters were concentrated to the north and west of the summit. This pattern may be understood in the same way as the 1978 foreshocks, considering that the ambient stress field is of strike-slip type with the P axis in a NW-SE direction. Following the summit eruption, a fissure eruption took place on the flank of the Izu-Oshima volcano. The hypocenter distribution dramatically changed from the pattern during the summit eruption. Earthquakes are now distributed in a linear zone trending NW-SE, the same direction as the flank fissures and the P axis of the ambient stress field. Taking the vertical and horizontal deformation associated with the flank eruption into consideration, the linear erathquake distribution may be interpreted as indicating dike intrusions at depth.

抄録全体を表示

The geomorphological characteristics of volcanic products and its volume of the 1986 eruption of Izu-Oshima volcano were clarified as follows based on the careful interpretation from aerial photographs and large scale maps of pre- and post-1986 eruption, and photogrammetric measurement of topographical change : 1. Distribution of 1986 craters, cinder cones and lava flows were represented carefully on a map considering their accurate topographic positions and micro-landforms. 2. Arrangement of 1986 craters is in left lateral echelon, whose position is coincident with the zone of flank volcanos and craters. 3. LB I and III lavas, erupted and flowed by breaking the cinder cones, have distinguishing features. Their surface structure and color tone are different from those of LA lava. 4. LB II lava was effused by forcing through the agglutinated cinder cone two days after the fissure eruption. Other part of cinder cone was swelled out toward the lower slope. 5. The discolored water areas around the Izu-Oshima seem to have occurred repeatedly prior to the 1986 eruption, by the slope failures, heavy rainfall or turbulent sea. 6. The total mass of volcanic products through the 1986 eruption (except vapor-phase materials) is estimated to be about 7.9×10⁷ tons. The details are as follows : 3.5×10⁷ tons through Nov. 15～19, 4.4×10⁷ tons on Nov. 21. 7. Among the volcanic products mentioned above, the scoria fall deposit is distributed in the eastern part of the island. Its mass is estimated to be about 9×10⁶ tons, including those fallen of the sea surface. The volcanic ash deposited in very small amount, and the most was swept away within a few days. 8. The mass of volcanic products through the 1986 eruption slightly exceeds than that of the 1950・1951 eruption.

抄録全体を表示

Izu-Oshima Volcano, Japan, erupted early on the evening of November 21, 1986, to form fire fountains along two major fissures running linearly southeast to northwest, one on the caldera floor (eruption B) and the other on the flank outside the northern caldera rim (eruption C). During the eruption B, the individual vents opened in succession towards the southeast and the northwest with a velocity of 0.42 m/s. During the eruption C, the individual vents also opened in succession towards the northwest with a velocity of 0.23 m/s. An additional feature of interest is a temporal and spatial gap between the erupion C. If this gap was caused by subsurface magma movement, its velocity is estimated to be 0.29 m/s. This velocity is so close to the extension velocity of the eruptions B and C that bilateral magma flow over the length of 3.3 km is inferred, though the possibility of accidental coincidence cannot be excluded.

抄録全体を表示

Izu-Oshima volcano is an active insular volcano mainly of basalt and located 120 km to the south of Tokyo. There are two connected calderas, 3-4km across, on the summit, and Miharayama, a post-caldera cinder cone, occupies the southern part of the caldera. The eruption of 1986 began with splended fire fountaining on Nov. 15 at the pit crater in the summit crater of Miharayama (A vent), after 12 years of dormancy. Well vesiculated scoria and ash including Pele's hair were prduced from the fountain. Lava filled up the Miharayama crater and flowed down to the caldera floor on Nov. 19. fountaining activity stopped on the night of 19th. Intermittent explosive eruption began on 20th and bombs became the dominant ejecta. A violent eruption started at 16 : 15, Nov. 21 with opening of fissure vents (B) trending NNW on the caldera floor. Eruption column rose up to 16, 000m. One and a half hour later another fissure vent (C) opened on the upper slope outside the caldera rim. Eleven craterlets were opened and a lava flow (LCI) had almost reached to Motomachi, the largest town in the island. Until next morning most vents ceased their activity except A and B3 vents. A vent continued intermittent explosions until 23th, and B3 vent ejected fine ash until 22th. A large spatter rampart was formed along B fissure vent and two lave flows (LBI, LBIII) spread out on the caldera floor. Another small lava flow flowed out from unsolidified interior of the spatter rampart on Nov. 23. These lava flows are aa and transitive lavas to block lava. They are 5 to 10 meters thick and their surfaces were covered by reddish brown blocks of agglutinate from the collapsed spatter rampart. Thick deposit of air-fall scoria was distributed mainly on the east of the vents. This air-fall scoria is comparable in volume to the scoria ejected in the past large eruptions, but has different texture and vesicularity. Another small Strombolian eruption at A vent occurred late in the evening of Dec. 18 and lasted about three hours. The activity of 1986 erupted 6-8×10⁷ tons of magma. This quantity is nearly comparable to that of 1950-1951 activity. Change in the mode of eruption (continuous fountaining to intermittent explosive activity) at A vent indicates the viscosity increase of the magma. Microscope and SEM observations have revealed that the groundmass crystallinity of the scoria erupted from A vent markedly increased with time, less than 3% to about 15%. This increase in crystallinity raised the viscosity of the magma ten times or more. Temperature decrease and degassing during the early eruption phase caused the increase in crystallinity.

抄録全体を表示

Sequences and products of the Izu-Oshima 1986-1987 eruptions which started on November 15, 1986, were investigated tephrochronologically. The results are summarized as follows : 1) Summit eruptions (Crater A) During 15-20, Nov. 1986, Strombolian eruptions continued to make a lava lake from where lava flows spilt over and went down the slope of the central cone to the caldera floor (LA I～IV). Volcanic ash and scoria (TA-1～4) were dispersed to the eastern and western parts of the island. On 21 Nov., a little after the beginning of the fissure eruption (Craters B), Strombolian eruptions were reactivated and ejected large volcanic bombs and scoria (TA-5) from Crater A. On Dec. 18, 1986, small explosion occurred from the Crater A for three or four hours, ejecting a scoria fall (TA-6) and bomb. The level of the lava lake lowered about 5 meters. On Nov. 16, 1987, a phreatic explosion occurred to break the crust of the lava lake, and the lava drained back to the deep on Nov. 18. 2) Fissure eruptions in the caldera floor (Craters B) At 16 : 15, on Nov. 21, 1986, fissure eruptions (Craters B) started on the caldera floor and extended to the slope of the central cone. The eruptions became explosive one, generating lava fountains with the height of more than 1500 meters, with a high discharge rate of 8×10⁶ ton/hour, producing pyroclastic cones and rootless (clastogenic) lava flows (LB I and III). Subplinian scoria falls were dispersed to west (TB-1) and east (TB-2). About 5 hours after the beginning, the activity waned to produce only volcanic ash (TB-3 and -6) and finer scoria falls (TB-4 and -5) and ceased on Nov. 23. A rheomorphic lava flow (LB II) occurred from the edge of the deformed cone on Nov. 23. 3) Fissure eruptions on the somma slope (Craters C) At 17 : 45, on Nov. 21, 1986, fissure eruptions occurred on the somma slope, and produced two lava flows (LC I and II), scoria cones, and vesicular scoria falls (TC-1 and -3) from the 11 craters. 4) The 1986 eruptions ejected 0.053 km³, 7.9×10⁷ tons of lava and pyroclasts from A, B and C craters (Table 4).

抄録全体を表示

During the 1986 Izu-Oshima eruption, various types of lave flows were formed. Those are classified into four : coherent lava flows, overspill lava flows, clastogenic (rootless) lava flows and rheomorphic lava flows. Characteristics of each type are summarized as follows : a) Coherent lava flows. In the later stage of LCI lava flow, coherent lava was observed to be effused from C6 crater. b) Overspill lava flows. The summit eruptions formed a lava lake, from where LA I～IV lava flows spilt over and flowed down the slope of the cone. c) Clastogenic lava flows. Two clastogenic lava flows, LB I and LB III, were originated from high lava fountains from B fissure vents. Those show relict clast structures, covered by blocks of collapsed scoria cone. d) Rheomorphic lava flow. A rheomorphic lava flow, LB II, was observed to flow down from the scoria cone beside B7 crater two days after the eruption of B fissure vents. Chemical composition of LB II is uniform as SiO₂=57wt%, though LBI, LBIII and most scoria falls from B craters show SiO₂=54wt%, except for the earliest scoria fall showing SiO₂=57wt%. This suggests that the lowest part of the scoria cone was welded and deformed, and squeezed up to the surface.

抄録全体を表示

The 1986 lava flows extruded by the fissure eruption at Izu-Oshima were reproduced by a modified method of a numerical simulation. Lava flows are regarded as Bingham fluid in the calculation. The procedure of calculations includes successive increase in viscosity and yield strength due to cooling by radiation. Two initial values of temperatures of lava are assumed, 1100℃ (5×10⁴ poise) and 1000℃ (3×10⁶ poise). The lava flow LCI on the western flank is reproduced better in the case that the initial temperature is 1100℃. The lava flow LB III in the caldera is reproduced better in the case that the initial temperature is 1000℃. The results suggest that the temperature at the start of the flow LB III was lower by 100℃ than that of LCI. The calculated distribution of the lava flow LB I in the caldera did not coincide with the actual one because of the insufficient precision of the topographic map and the change of topography caused by the tephra or the upheaval near the craters immediately before the extrusion of the lava.

抄録全体を表示

The phase 2 eruption of the Izu Oshima volcano, November 1986, was remarkably observed not only from the ground but from many TV helicopters in the air. The eruption began at 16 : 15 on the 21st from a 1-km long fissure vent, B, on the summit caldera floor, which are composed of northwestern B_N fissure (560m long) and southeastern B_S fissure (460m long). By 16 : 55, more than 10-km high subplinian eruption column was established on the B fissure vent, while the B_N fissure fed 2B_N lava flow (1.3×10¹⁰kg) from the northwestern lowest point and the B_S fissure also fed 2B_S lava flow (6.6×10⁹kg) from the northwestern lowest point. These lava flows never originated as a so-called rootless lava flow or rheological flowage of agglutinate. They were formed by simple overflowing of magma to earth surface. Another lava flow 2C (4.6×10⁸kg) from C fissure on the northwestern flank is also an overflowing lava. The phase 2 eruption is interpreted as the surface expression of an intrusive event of three sheets of dikes from the magma chamber. A possible rootless lava flow was formed during the first 30 minutes of the eruption along the B_N fissure when the fountain height was still low. It was buried by the succeeding subplinian scoria. However, on the afternoon of the 23rd it again appeared on the scoria cone to form 2B_C lava flow. During several days after the eruption, a part of the scoria cone moved up to 30 m toward the down slope. This is due to rheological flowage of agglutinate layer which was a result of the oblique fountain from a vent on the B_N fissure.

抄録全体を表示

During the 1986 eruption of Izu-Oshima volcano, a drastic change of pattern in seismicity is observed. About one day after the summit eruption on Nov. 15, a very active earthquake swarm began at the northern part of the island. The correlation between the activitiy of the eruption and that of the earthquake swarm shows the stress concentration around the depressurized magma reservoir. After the fissure eruption on Nov. 21, hypocentral distribution dramatically changed. Just after the eruption, earthquakes began to propagate from the both tips of the fissure to NW and SE directions. The propagating velocities to NW and SE directions are 11 m/min. and 16 m/min., respectively. This is interpreted as the propagation of the change in the stress field owing to the fissure eruption. However, these two groups are very different from each other. The focal mechanisms of the events at the northwesten part are almost strike slip types, which are consistent with the typical stress field around Izu-Oshima Island. No clear subsidence is observed near the epicentral area. On the other hand, those of southeastern part are almost normal fault types, which shows NE-SW tension. The subsidence in the southeastern part takes its maximum just above the region of high seismicity.

抄録全体を表示

Izu-Oshima volcano erupted from the summit crater on November 15, 1986. The strong motion accelerograph station in Izu-Oshima recorded felt earthquakes soon after the eruption. We installed two additional temporary accelerograph stations in Izu-Oshima on November 20. The fissure eruption commenced on November 21. The eruption was preceded by intense earthquakes. The largest earthquake with M 6.0 occurred at 0041 UTC (Universal Coordinated Time) on November 22. More than 440 earthquakes were recorded at strong motion accelerograph stations from November 16 to December 7. Hypocenters were determined for about 310 earthquakes. Before the fissure eruption, epicenters were distributed in the northern and western parts of Izu-Oshima Island. Earthquakes prior to the fissure eruption were concentrated within a small area of about 2 km in diameter under the northern caldera rim. Focal depths were less than about 5 km. After the fissure eruption of November 21, epicenters were widespread and hypocenters depths were distributed between 0 km and 15 km. Low frequency earthquakes were observed from November 22 to December 7.

抄録全体を表示

Fan-shooting observations along the east coast of Oshima island were carried out to study, mainly, anomaly of wave propagation under the Volcano. Explosions used in these observations were carried out originally for the purpose of seismic wave velocity measurement [SP. Senbazaki], and of refraction survey across Oshima Island [SP. II]. SP. Senbazaki is located on the other side of the Caldera to the observation sites, and SP. II in the northern part of the Caldera. Amplitudes of the earliest phase and the largest phase in each seismic record, and also ground noise level at the station were examined. Seismic amplitude data were corrected for distance on the basis that the amplitude decreases as the second power of distance from the shot point. Amplitude data by SP. Senbazaki were also corrected for explosion efficiency because those were obtained in four separate observations. The results are summarized as follows. 1) First peak-to-peak amplitude of seismic wave passing througth the 5.5 km/s layer remarkably decreases on the other side of the Caldera. It seems that a large attenuation in the 5.5 km/s layer occurs under the Caldera. 2) Attenuations in other velocity layers are not so clear. 3) A very sharp decrease of amplitude at a narrow area (F₇, F₁, F₈] is not caused by the anomaly along the ray path, but by the characteristics of the sites.

抄録全体を表示

The 1986 eruptions of Izu-Oshima volcano were accompanied by volcanic tremors with large amplitudes. The tremors were observed by the 59-type seismometer with a proper period of 5 sec. at the Oshima Weather Station, situated at NNW 4.8 km distant from the central crater. The tremor occurrences were continuous or sometimes intermittent corresponding to eruptive durations. The amplitude variations and wave properties of the tremors were investigated by analyzing drum records of the seismometer. The amplitudes of the tremors did not always show a dependence on the discharge rates of magma, and increased from the morning of Nov. 17 when continuous lava fountain with rumbling sounds changed into that with explosive sounds. The tremors with lower frequency (ca. 0.5 Hz) were predominant from the evening of Nov. 17 to the morning of Nov. 18, and the tremors with higher frequency (ca. 0.8 and 1.0 Hz) were predominant since the morning of Nov. 18. As the result from particle motion diagrams, the tremors with lower frequency show the prevailing directions of vibration from NNE-SSW to NW-SE in horizontal plane, and seem to be composed of Rayleigh and SV types of wave motion. The wave frequencies of the tremors during the 1986 eruptions were lower than those during the 1950-51 eruptions. Released energy of the volcanic tremors during the 1986 eruptions were estimated to be 6×10¹⁹erg. This is three times of energy of the volcanic tremors during the 1950-51 eruptions, though the eruptive duration of the 1986 eruptions were very shorter than that of 1950-51 eruptions.

抄録全体を表示

After the 1986 eruption of the Izu-Oshima Volcano, we have been observing ground tilts of curious pattern which closely correspond to the occurrence of volcanic tremors. Each of the tilt events, which we refer to as “saw-teeth-shaped tilt change”, is composed of a set of two tilt stages ; slow tilt motion toward the direction of the central crater, and successive fast motion toward the opposite direction. (The amplitude of tilt motion is about 5・10^-8 radian.) The duration of each event is about 1-2 hours for the first stage, and 6-60 minutes for the second stage, which corresponds to the occurrence of volcanic tremor. In order to account for these characteristic features, we propose a magma contraction model. The model, implying step-wise decrease of internal pressure of magma reservoir, is harmonious with the observational fact that the central part of the volcano is subsiding after the 1986 eruption.

抄録全体を表示

Crustal deformations associated with the 1986 eruption of Izu-Oshima volcano were studied. A zone of relative depression with the strike of NW-SE, whose maximum reached about 60 cm and average 30 cm, was observed in the central part of the island, while uplift zones up to 20 cm were revealed in the northeastern and southwestern parts. The Okada tidal station, the levelling datum in Izu-Oshima, is considered to have subsided after the eruption on the basis of tide observations. Distance measurements after the eruption have revealed a large extension of the baseline which crosses a fissure. In order to explain these observations, a combined model of the pressure decrease in the magma chamber and the opening of a tensile fault is presented. A tensile fault, which runs across the island from northwest to southeast, is estimated to be 15 km long, 10 km wide and 2 m thick, and its upper margin should be 2 km deep. On the other hand, the depth of a pressure source is estimated to be about 10 km. This model suggests that not only a deflation of the volcano but also an intensive intrusion of the magma took place beneath the island at the time of the 1986 eruption of Izu-Oshima volcano.

抄録全体を表示

Repeated precise levelling has been carried out annually since August 1982, on the summit caldera area of Oshima volcano, Izu. From May 1986, levelling has been performed with a few months interval. In Novemver 1986, the summit as well as the flank-fissure eruptions occurred in Oshima volcano. The levelling in the summit caldera revealed some unusual relations between the vertical ground deformation and the eruption. Typical features of the summit caldera deformation are summarized as follows : 1. The inside of the caldera is continuously subsiding with respect to the outer somma of the caldera. The center of subsidence is situated at the central cone Mihara-yama. It is presumed that the subsidence is caused by compaction mainly due to the load of material which piled up in the past eruption. 2. Unlike the traditional idea of inflation-deflation process, the central cone did not inflate before the summit eruption. Following the summit eruption, subsidence was suddenly accelerated by the increasing load of newly erupted material. Subsidence rate was decreasing exponentially with time after the eruption. 3. At the margin of the caldera, a considerable amount of subsidence occurred during the eruption. Therefore, the summit caldera rim of Oshima volcano can be regarded as an active fault.

抄録全体を表示

Izu-Oshima volcano commenced a series of eruptions on November 15, 1986 after twelve years’ quiescence. The renewed volcanic activity started as a summit eruption, later developed a drastic fissure eruption on November 21, and ended as a small summit eruption on December 18. Since 1975 we have repeated distance-measurement survey using an Electro-Optical Distance-Meter (EODM) in and around the summit caldera of the volcano. As a result we succeeded in the detection of a significant horizontal crustal deformation that occurred across the sequence of eruptions. The deformation of the summit caldera comprises a NE-SW extension and a NW-SE contraction. We infer from other related observational data that the greater part of the deformation resulted within the period of activity of the fissure eruption on November 21. The horizontal deformation of the summit caldera is explained, in outline, by the dike-intrusion model that TADA and HASHIMOTO (1987) proposed on the basis of the results of presice levelling survey. However, further study is necessary to get a strict conclusion.

抄録全体を表示

Crustal tilt movements were observed before and during the 1986 eruption of Izu-Oshima Volcano by four tiltmeters installed in Izu-Oshima Island. A borehole-type tiltmeter settled at Habu area (HB), southern part of the island, recorded NNE ground-up tilt change up to 1 μ radian from November 4, 1986. The tilt change turned the direction on November 12. A bubble-type tiltmeter settled at Kita-Gairin (KG), located at the north apron, and a borehole-type tiltmeter settled at Seibu-Gairin (SG), located at the northwest apron, recorded tilt movements up to 10-20μ radian between November 14 and 19. The directions of movements seem to indicate the subsidence along the eruptive fissure on November 21. The movement turned around on November 19. The tiltmeters at Habu, Seibu-Gairin and Kita-Gairin indicated a precursory tilt change about 2 hours before the eruption on November 21. A bubble-type tiltmeter settled at Weather station (WS), located at the northern part of the island, also indicated an anomalous change 1.5 hours before the eruption. These tilt changes indicate a possibility to predict fissure eruption by tilt observations. The crustal tilt movements observed by the tiltmeters at Seibu-Gairin and Weather station turned the direction twice associated with the eruption activities before ceasing of the eruption on November 21. In the middle of this eruption, anomalous WSW ground-down tilt movements up to 20μ radian were observed by the tiltmeter at Habu. The tilt direction was changed twice clockwise at a right angle. The tilt movements towards the northeast ground-down direction attained more than 100μ radian in ten days. This crustal movement is attributed to a buried tensile movement in the southern part of the island.

抄録全体を表示

During the period from Nov., 1986 to Nov., 1987, the Izu-Oshima volcano erupted five times. The National Research Center for Disaster Prevention had been conducting continuous crustal tilt observations using the borehole-type tiltmeter. These observations were being made at Gojinka and Habu, located at the northwest caldera rim and the southeast part of the island, respectively. Tilt changes associated with a train of eruptions were observed at both stations and results were as follows. 1) Precursory changes just before the eruption were observed at Gojinka four times, except for the eruption which occurred on Nov.15, 1986. Also at Habu precursory changes just before the eruption were observed prior to the eruptions on Nov.21, 1986 and Nov.18, 1987. 2) At both stations, the directions of these precursory changes were mostly ground-up, towards the north-northwest. On the other hand, the direction of changes observed after the eruptions were ground-down, again towards the north-northwest. These tilt changes before and after the eruptions indicate an inflation-deflation process in north-northwest direction. 3) Utilizing these tilt data and the volumetric strain data we have made an estimation on the location of the pressure source based on the Mogi’s model. It was determined that the pressure source was located approximately at a depth of about 2 kilometers, less than 1 kilometer away from Gojinka towards north-northwest. 4) Most of the precursory changes are attributed to the inflation-deflation process of the magma reservoir, which we estimated above. However, direction of the precursory change about two hours prior to the fissure eruption on Nov.21, which was observed at Gojinka, was not north-northwest but east-northeast (towards the fissure). We presumed that this change was caused by the magma intruding on to the surface of the earth. 5) From the results of earthquake observations and precise levelling, it has been conjectured that after the eruptions on Nov.21, the magma intruded in the southeastern direction of the island (YAMAOKA, 1987 ; TADA and HASIMOTO, 1987). The tilt changes at Habu corresponded to the intrusion. From the results of continuous tilt observations, it can be conjectured that the greater part of the intrusion took place by Nov.28.

抄録全体を表示

Izu-Oshima Volcano erupted on Nov.15, 1986. Many open ground cracks were formed in a north-northwest trending broad zone across the island nearly concurrently with the fissure eruption which began at 16 : 15 Nov. 21. The formation of the open cracks indicates that extentional horizontal displacement in the NE-SW direction occurred near ground surface. After the eruption, we commenced the observation of ground surface deformation mainly across the cracks by many invar-wire extensometers. The extensometer is useful for rapid setting up. This paper reports the result of the extensometer observation until Nov. 1987. It is noted that remarkable ground surface deformation was observed from Nov. 18, 1987 and extentional step occurred on Nov. 21 in the southeastern part of the island. The ground surface deformation suggests that the activity of dike-form magma under the area is still continuing.

抄録全体を表示

We present a physical model that explains the gravity and elevation changes caused by the 1986 eruption of Izu-Oshima volcano. The model assumes (1) a spherical pressure source and (2) formation of a fracture zone where a large number of infinitesimal cracks open in a horizontal direction, most probably in the NE-SW direction. The density of the crack-filling matter is estimated to be 0.3 to 1.3g/cm³, far smaller than the value expected from the dyke-intrusion model.

抄録全体を表示

Every volcano would manifest particular activities whenever it erupts. It is rather difficult to presume patterns of the future activity but it is hopeful to find any tendency of possible activities. In this respect, any empirical laws deduced from the past activities of various volcanoes should be useful for appreciation of the progressing activities. First, empirical laws for deformations and gravity changes on volcanoes are proposed, and referring to these laws, elevation and gravity changes associated with the 1986 eruption of Ooshima volcano are discussed. The subsidence observed along the route of precise levels on the northern slope of the volcano may have been caused by magma intrusions thereabout while the subsidence observed along the route in the southeastern part induces doubts as to its volcanic origin ; the latter may have been probably due to earthquake vibrations of the ground. A possible distribution of elevation changes is presented : the changes are composed of a tilt of the island as a whole and a local depression on the northern slope. Generally, gravity would change in relation to ground deformations with the progress of volcanic activities. To discriminate any significant gravity changes associated with the 1986 eruption of Ooshima volcano, the gravity changes are plotted against the elevation changes : they have a poor correlation each other. This suggests that their origins are differently located. The gravity changes may be attributable to local changes superposed on those due to a tilt of the island as a whole. If more bench marks were effectively distributed on the volcano, we should have detected local gravity changes of volcanic origin, if any, more clearly.

抄録全体を表示

Hydrographic Department, M.S.A. Japan conducted an airborne magnetic survey on Osima Island by an aircraft, YS-11, to investigate magnetic structures and magnetic changes accompanied by volcanic activity after the eruption on Nov. 21, 1986. In airborne magnetic surveys, total intensities were measured by a proton precession magnetometer every two seconds. Positions just over the island were determined using video image and those over sea were measured by the VLF/OMEGA positioning system. Positioning error over the island was supposed to be less than 100 m in usual condition and the survey line spaces range from 300 m to 800 m. Detailed magnetic anomaly map at the altitude of 3000 ft was compiled based on above survey data. Using above survey data, three-dimensional magnetic analysis was conducted by Talwani’s method. Bulk mean intensity of the magnetization of the edifice of Osima amouts to 12 A/m reflecting a basaltic volcano. Direction of the magnetization vector varies depending on the method of approximation of a trend field in anomaly analysis. Under the condition that the bias level as well as the magnetization vector is calculated in an analysis simultaneously, direction of the magnetization vector becomes Dec=12°E and Inc=64°, respectively, where declination of 12°E is more close to the normal field direction than values calculated by the planar trend assumption (VACQUIER and UYEDA, 1967 ; KODAMA and UYEDA, 1978). Magnetic anomaly caused by the central cone of Mt. Mihara and its adjacent area is derived by subtracting the calculated anomaly field by the Osima edifice from the observed one, on the assumption that the Osima edifice is magnetized in present field direction. From above analysis, magnetic anomaly caused by themal demagnetization has been discovered just over the fissure of B-crater, created by eruption on Nov. 21, 1986. Amplitude of anomaly caused by thermal demagnetization amounts to 300 nT and the width of demagnetized part, which is running N60°W from the vicinity of B-crater, is ranging from 300 m to 400 m in horizontal distance. Residual field obtained above does not show apparent anomaly field originated from Mt. Mihara, which may imply that Mt. Mihara is thermaly demagnetized in 1～10^-1 A/m order. Two-dimensional anomaly analyses on two magnetic profiles in NE-SW direction show that volcanic basement is extending from the submarine Habu volcanic spur to the area beneath the edifice of Osima in NW direction. Such a two-dimensional magnetic structure is significant to understand a discrepancy of magnetization direction obtained by three-dimensional model analysis from present field one. The total intensity field derived from the airborne magnetic survey in December 1986 was compared with that in September 1964, which was compiled by UTASHIRO et al. (1972) to obtain magnetic change for these 22 years. This comparison shows a drastic magnetic decrease amounting more than 1000 nT in the vicinity of B-crater. The decreased area is consistent with the demagnetized area inferred from magnetic analysis, however, it’s amount is much larger than one of the latter demagnetization source. Results of repeated aeromagnetic surveys, which were conducted from December 24, 1986 to January 19, 1987 to detect a magnetic change accompanied by volcanic activity, are also reported together with some discussion of error estimate and model calculation of thermal demagnetization.

抄録全体を表示

Aeromagnetic surveys were carried out over the Izu-Oshima Volcano, before and after the November 1986 eruption. The comparison between the pre- and post-eruption aeromagnetic data showed no significant change in the magnetic anomaly. The magnetic subsurface structure of the Izu-Oshima Volcano was investigated on the assumption that the volcano is magnetized in the present Earth’s field direction. The topography magnetization was determined from correlation between the original aeromagnetic data and the magnetic anomaly produced by uniformly magnetized terrain. The estimated topography magnetization was 7.0 A/m. The analysis for the regional trend in the residual magnetic anomaly after the terrain correction represented the existence of a large-scale dike in the NW-SE direction. The local anomalies in the residual map after the removals of terrain and dike effects correspond to the distribution of craters and lavas of the Izu-Oshima Volcano.

抄録全体を表示

Bathymetric, seismic, geomagnetic and gravity surveys around Izu-Osima Volcano were carried out by the Survey Vessel Takuyo of the Hydrographic Department of Japan, after volcanic eruption of Osima volcano in 1986. The survey revealed several geological and geophysical characteristics at the adjacent area of Osima Island as follows : i) NW-SE topographic and geomagnetic elongations due to underwater fissure eruptions were recognized in the area off Tigasaki, northwestern end of Osima Island and off Habu, southeastern end of Osima Island. ii) U-shaped bathymetric high adjacent to the west coast of Osima having high magnetic anomaly, or the Senba Spur, is considered to be an old volcano edifice contiguous to Osima volcano. Subsequently, the Hydrographic Department surveyed in detail off the southeast of Osima, the Habu Spur, on February 5, March 3 and April 16 in 1987 by the newly-built Survey Vessel Tenyo. She is equipped with a narrow multibeam echo sounder (Hydrochart) for shallow water (up to 1, 000 m), so the very precise bathymetric chart of the Habu Spur (scale 1/20, 000) has been made and some specific topography has become clear. The results from the analysis of Hydrochart records are as follows : i) The Habu Spur is composed of many knolls which are arranged in two parallel lines trending the southeast to the northwest. ii) The direction of the lines of lateral volcanoes is coincided with the direction of knolls existing on the flank of Osima volcano. iii) Two lines of lateral volcanoes are just like the straight line and are composed of continuous narrow ridges, which suggest something like intrusive dikes. iv) A 185 m knoll below the sea surface is discovered in the middle of the northeastern line of lateral volcanoes. x) Some water depths surveyed in 1987 have become deeper than the water depths surveyed in 1954 on the northeastern line of the lateral volcanoes (maximum change is 100 m). vi) The change of water depths from 1954 to 1987 on the 223 m knoll in the southern end of the northeastern lateral volcano line is divided into two patterns. One, the southeastern part of the lateral volcanoes line, is remained unchanged and the other, the northwestern part, have become deeper. The cause is unknown.

抄録全体を表示

During the 1986 Izu-Oshima eruption, various kinds of lavas and ejecta were formed. From the central cone (A vent), plagioclase-phyric basalt (SiO₂, 52.5-53.2 wt%) was erupted as scoria, bombs and lava flows. From the fissure on the caldera floor (B vent), aphyric basaltic andesite and aphyric andesite (SiO₂ 54-58 wt%) were erupted forming scoria, bombs and lava flows. Along the fissure on the outer slope of the main stratovolcano (C vent), aphyric basaltic andesite (SiO₂ 54-55 wt%) formed many cones and erupted lava flows. Detailed analysis of both major and minor elements indicates that A vent and B and C vents belong to different magma plumbing system. The high alumina nature of the basalt from A vent is explained by flotation and accumulation of plagioclase phenocryst at the top of the magma reservoir. Based on the compositions of coexisting pyroxenes, the temperature of the magmas on eruption are estimated to be 1100-1150℃ for basalt from A vent, 1100℃ for basaltic andesite from B and C vents, and 1070-1100℃ for andesite from B vent. Crystallization fractionation at relatively shallow magma reservoir (less than 2 k bar or shallower than 8 km) can account for compositional variation from basaltic to andesite magmas. During the fractionation, H₂O increased from 0.7% or less in basaltic magma to about 1% in andesite magma. This low pressure crystallization fractionation can also account for the compositional variation within each major eruption of the recent 1500 years. Comparison of major and trace element concentrations in the effusive rocks of 1986 eruption and the recent 1500 years eruptions indicates that they were derived from a single parental magma. The parental magma has been continuously fractionated at least these 1500 years within a deeper and probably much larger magma reservoir than the shallow magma reservoir mentioned above. Because of this fractionated nature of the supplied magma, the andesite magma, which was very rare in Oshima volcano, was easily formed by the successive fractionation in the small scale magma reservoir. Analysis of dacitic pumice fragments from B vent indicates that they are not the fragments of the dacite layer of the Senzu group (Tatsunokuchi dacite), but are probably fragments of the dacite pumice of the Younger Oshima group, which is not yet found within Oshima Island.

抄録全体を表示

Major and trace element compositions and ⁸⁷Sr/⁸⁶Sr ratios were determined for 3 lavas and 2 scorias of the 1986 eruption of Izu-Oshima volcano. The magma from A crater has similar chemical and Sr isotopic compositions to magmas which flowed out many times in the post-caldera stage of Oshima volcano. While, magmas from B and C craters have different chemical compositions from the magma from A crater, in spite of similar Sr isotopic compositions. REE pattern, Sr/Ca-Ba/Ca systematics and other trace element distribution pattern indicate that the magma of B and C lavas are fractional crystallization products of the magma of A lava.

抄録全体を表示

Chemical and isotopic compositions of volcanic gases and rocks are compared between the samples collected immediately before and after the 1986 eruption of Izu-Ohshima volcano. Around the pit crater of Mt. Mihara, the CO₂ and H₂ contents in fumarolic gases increased before the eruption. The chemical composition of volcanic gases from Mt. Mihara (A crater area) indicates that they are mainly of magmatic origin. The gases from the craters B and C, both formed by fissure eruption, did not come from the magma but were expelled from the hot lava and scoria. δ¹³C values show that CO₂ in these gases were derived from organic matters. This is consistent with the very short duration of fumarolic activity at the B and C crater area. Volcanic sublimates found around the high temperature fumaroles at the A crater area were mainly sulfate minerals. At the B and C crater area, copper chloride (CuCl₂ etc.) and copper oxide (CuO etc.) dominated at the B crater area. Most of the sublimates from the C crater area were salammoniac (NH₄Cl). Chlorine and fluorine contents of lava and scoria from A crater are found to be relatively uniform. On the other hand, those from B and C crater groups are highly valuable, and strong correlation is found between chlorine and fluorine contents.

抄録全体を表示

Temporal variations in temperature and ³He/⁴He ratio of fumarolic gas emitted from a steam well located about 3 km NNE of the central cone of Izu-Oshima volcano have been monitored since October, 1986. Chemical compositions of hot spring waters collected at a well close to the steam well have also been measured since September, 1986. The temperature of the gas increased 5-6℃ after the eruptions of November, 1986, reached the maximum value in February-March in 1987 and then decreased gradually. The ³He/⁴He ratio was nearly constant (1.7-1.9 R_atm) before and immediately after the eruptions of November, began to increase in December and then has been keeping a constant value of 5 R_atm since May, 1987. These features are interpreted to be caused by the uprise of the magma. Concentrations of most chemical components in the hot spring water have also changed after the eruptions of November, 1986.

抄録全体を表示

Periodic observations were carried out for chemical compositions of water samples from springs and boreholes (13 observation sites) in Izu Oshima Island at intervals of about a month just before and after the 1986 eruption of Izu Oshima volcano. Based on periodic observations, the influences of the Izu Oshima volcanic activity on the subsurface water system and on its chemical compositions are summarized as follows. (1) The change in flow path of subsurface water by the ascent of the magma. (2) The intrusion of magmatic emanations such as carbon dioxide into subsurface water layer followed by the water-rock interaction. (3) Sea water intrusion and disturbance of stratification of subsurface water consequent on the ground crack formation and the heat supply from the magma (magmatic emanations). (4) Dilution of subsurface water with meteoric water because of the stop of drawing water during the period of the evacuation (flight) of habitants from the island.

抄録全体を表示

Detailed characterization of the whole rock composition of the ejecta of the 1986-1987 eruption of Izu-Oshima volcano by the XRF technique (FUJII et al., 1988) clearly indicates that the ejecta from the central crater of Miharayama (Crater A) are different from those erupted from the fissures (Fissures B and C) on the caldera floor and on the outer slope of the main stratovolcano. This suggests that the conduits which led the A and B, C magmas to the surface were separated physically from each other down to a certain depth. The ejecta from A crater resemble closely to those erupted during the past 1300 years (Y magmas) while the ejecta from B and C fissures are unique in composition among the Izu-Oshima magmas. The A and Y magmas are Fe-enriched island arc type tholeiites that must have been derived from the primary tholeiite magmas through crystal fractionation of olivine, pyroxenes and plagioclase. The B and C magmas can be derived simply through crystal fractionation of A or Y magmas leading to Fe-enriched basaltic andesites, andesites and dacites. This strongly suggests that an isolated pocket of magma starting with a composition of Y underwent strong fractionation to produce volatile enriched ferroandesitic magma. This body of magma was probably activated by the sudden depressurization caused by shattering and fissuring of the crust to form a 1500 m high fire fountains and to produce a sub-Plinian scoria fall deposit. The vent of A crater must have been stable at least during the last 1300 years and directly, or through the subsidiary magma chamber(s), connected to the main chamber above the Moho, where an extensive crystal fractionation has been taking place to produce Y magma from the parent tholeiitic magma. Marked ground depression and extension and migration of seismicity observed during the eruption suggest a possibility that a substantial amount of additional magma was intruded to form a NW-SE trending dike during the peak phase of the eruption. This is in harmony with the Nakamura’s model of the volcano with the NW-SE trending dike swarm which is controlled by the regional compressional stress field. However, gravity and some other grophysical data suggest that the deformation could have been the result of underground cracking without magma injection. Our model is not conclusive on this matter and the expectation that this eruption will eventually lead to the large-scale activity that has been recurring in every 130±50 years is yet to be tested.

抄録全体を表示

Various features of the eruption that began on November 15, 1986 in Izu-oshima volcano are examined to infer the underground magmatic activities and the mechanism of the eruption. A massive dike intrustion that is assumed to have happened at the same time as the fissure eruption cannot explain the seismicity, deformation and other evidence consistently. Another preferable model that gives a systematic explanation of the available data is proposed as follows. A magma reservoir was situated at a depth of about 5 km below a NW (north-western) flank of Izu-Oshima volcano. This magma reservoir supplied magma to the summit crater through a well-developed vent, and caused the first summit eruption under an enhanced NW-SE compressive component of the tectonic stress. As the magma reservoir deflated with the discharge of magma in this summit eruption, a compressive stress increased in the neighboring area centered at the northern rim of the caldera. The stress finally fractured this area with intense seismicity, and produced fissures. Along these fissures, the magma that had penetrated into interstital space between rocks effused explosively with bubbling of steam. The magma had been more or less cooled and chemically differentiated in the interstitial space so that the ejecta from this fissure eruption was more felsic. The deflation of the magma reservoir due to the summit and fissure eruption resulted in a significant subsidence of the NE part of the island. Strain associated with opening of the volcanic fissures was transmitted through a strike-slip fault to the SE part of the island, and caused a local extensional stress and graben there.

抄録全体を表示

For the major eruptions of Miyakejima Volcano in 1983 and of Izu-Oshima Volcano in 1986, long and short term predictions were made by means of eruptive regularity based upon causal relation between eruptive activities and regional tectonic stresses. Level change in the bottomfloor of the summit crater of Izu-Oshima Volcano, identification of P₂-phase of swarm earthquakes in and around both volcanoes and marked change of geophysical data before major eruptions were available to predict the major eruptions of both volcanoes. The regularity suggests that the great eruptive energy of Izu-Oshima Volcano since 1986 should not be released until a great earthquake along the Sagami Trough.

抄録全体を表示