The relation between the Pinatubo eruption of 1991 and the Philippine earthquake of 1990 is studied on the basis of strain changes calculated from a fault model of the Philippine earthquake. At Pinatubo volcano, which is located 100 km away from the earthquake fault, the volumetric strain change induced by the earthquake is about 10-6. This caused gradual squeeze up of magma to the surface and the Pinatubo volcano erupted eleven months after the earthquake. The volumetric change calculated by a fault model of the 1990 earthquake is, however, several orders of magnitude smaller than the total volume observed during the eruptions. The magma squeezing model alone cannot explain the whole volume of ejected magma. A possible interpretation is that the volumetric strain change could have squeezed up magma in the magma reservoir, resulting in lowering the density of magma and enhanced the magma to rise further more. Such a positive feedback process could have occurred after the Philippine earthquake of 1990. It is possible that the Philippine earthquake triggered the activity of Pinatubo volcano.
The central Kyushu is a north-south spreading region characterized by east-west trending grabens and abundant active volcanoes. Available data indicate that the width of graben is 11-12 km apart from eruption centers of Unzen volcano, whereas it narrows down to about 7 km dose to the eruption centers. Focal mechanisms of earthquakes suggest that north-south tensional axis is not altered by the presence of volcanic conduit. Width of graben in general is pos tively correlated with thickness of lithosphere, and the width of graben is narrowed locally in accordance with the thinning of lithosphere beneath active volcanoes by the presence of magma chamber in middle crust. Evidences supporting this idea include shallow cutoff depth of seismicity and mid-crust low velocity area beneath the active volcano, shallow Curie temperature distribution, and high oxygen isotopic composition of the volcanic rocks.
The volcanoes of Kirishima, located in southern Kyushu, consist more than 20 volcanoes occupying an area of about 20 x 30 km elongated northwest-southeast. At least three volcanoes have historic records of eruptions ; Shinmoe-dake, Ohachi and Iwo-yama, and more than 10 volcanoes were active in the past 22, 000 years. This indicates that Kirishima is a multi-active volcanic group. Intense geothermal activity and frequent earthquake swarms are also characteristic features of Kirishima. Through seismological and electromagnetic observations, the nature of the Kirishima 'multi-active volcanic system' has been revealed. According to the seismometrical observation by the Kirishima Volcano Observatory, the majority of the hypocenters are distributed along several trends that are elongate either NE-SW or NW-SE. The dominant focal mechanism solutions indicate normal faulting with a NW-SE tensile axis in the NE-SW segments and of strike slip in the NW-SE segments. Active volcanoes are located along these lines, and extremely shallow earthquakes are distributed just beneath the craters of these volcanoes. The seismicity and focal mechanism indicate that the Kirishima area is 3-subject to NW-SE extensional stress. The slight extensional stress is favorable for a fault system that allows magma to ascend at various points. This may be the reason why a multi-active volcanic group was generated in this area instead of a large stratovolcano. According to the VLF-MT survey, the southwestern side of Kirishima is covered with low resistive surface layer, while the northeastern side is covered with high resistive layer. It is consistent with the fact that hot springs, fumaroles and steaming ground are scarce on the northeastern side of Kirishima, but are abundant on the southwestern side. It is also found that low resistive zone distributed along the trends of hypocenters ; from Shinmoe-dake to Takaharu. In the Iwo-yama region, which has the most active geothermal features at Kirishima, precise geothermal and electromagnetic surveys have been carried out. Through an inversion analysis using VLF, ELF and ULF-MT, a low resistivity second layer is found throughout the Kirishima area beneath the more resistive surface layer, which is about 100 m thick. The resistivity of the second layer decreases dramatically just around the summit of Iwo-yama. This is interpreted as being caused by a porous layer saturated with water which has wide distribution around Kirishima ; high temperature volcanic gas supplied from beneath Iwo-yama creates the extremely low resistive hydrothermal zone around the summit. The depth of the heat an dgas source is estimated to be about 10 km. Similar result was obtained at Shinmoe-dake, while no such low resistive zone was found beneath Ohachi. This result and other seismological and petrological evidences indicate that magma is stored at about 10km depth and rises several kilometers beneath northern volcanoes in Kirishima, such as Iwo-yama and Shinmoe-deke, while beneath southern volcanoes such as Ohachi, magma is supplied directly from the deeper part. Precise model of magma supply system in Kirishima as a multi-active volcanic system will be presented in the near future.
Time-space relation between major eruptions in Izu region and large earthquakes along the Sagami Trough shows the following correlation : T= 49.88-21.04 logD, where T is the time interval from beginning of a major eruption to the occurrence of the large earthquake in year and D is the distance in km from the epicenter to the volcano. The relationship reveals that the time interval is shorter as the distance between the epicenter and the volcano is smaller. This suggests that a future shock area gives a stress increase to the volcano and a major eruption occurs by squeezing up of magma. Then, it would be followed by a large earthquake along the Sagami Trough to release a compressional strain accumulation. Subsequently, the magma reservoir is expanded and the magma head subsides down, and a series of the volcanic activities ceases. This change is closely related to a level change of the magma head in the summit crater of Izu-Oshima Volcano. Thereby, Izu-Oshima Volcano has a role as a very sensitive strain gauge. The time-space relationship suggests that the strain in the region along the middle Sagami Trough has been accumulated since the major eruption of Izu-Oshima Volcano in 1986.
Recent progress in studies on magma plubming system beneath mid-ocean ridges is reviewed. Mid-ocean ridges are not continuous, homogeneous series of crests, but are segmented by various topographical offsets of several orders of magnitude. Such topographical ridge segmentation is a surface manifestation of along-strike variations in the pattern of mantle convection and the supply of magma. Upwelling of the mantle material beneath the spreading center takes the form of a diapir, rather than sheet-like flow. Such diapiric upwelling has an along-axis dimension of several hundred to several tens of kilometers, which corresponds to first-and second-order segments, but some are as small as the scale of third-order segments. Where magma upwelling is intense, hot lithosphere and high magmatic productivity produces thick crusts, resulting in low gravity anomaly. When the spreading rate is low, mantle flow is cooled on route to the bottom of the lithosphere and ceases to melt at depths. Thus magmas produced beneath slow spreading ridges have low degrees of melting. On the contrary, fast spreading ridges are followed by intense mantle upwelling which melts significantly the mantle column from deeper part of the upper mantle up to just beneath the crust. Magmatic budget is so large at fast spreading ridges as to maintain large, steady-state magma chambers along the ridge axes with extensions comparable to second-and third-order segments. Such magma chambers have a very small melt pocket one to two kilometers wide and up to several hundred meters thick, underlain by a large mush of crystals and melt. Cumulate layers on the top of the mush sink and deform to S-shaped or downwarping layers as seen in large layered plutonic bodies in some ophiolites. However, low spreading ridges do not have enough supply of magma but only possess short-lived small magma chambers consisting of crystal mush. Volcanic landforms consist of major topographic features associated with small volcanic edifices : the former is a large shieled volcano several tens of kilometers long and is probably formed by episodic eruptions every several tens of thousand years ; the latter is a small conical lava cone comprising pillow flows which is constructed during a short-period eruption but some are active as long as 1.8 million years which are unequivocally central volcanoes. Rifting episodes at spreading axes depend on the spreading rate and the width of the volcanic zone. Slow spreading ridges such as Iceland have rifting every 100 years, but fast spreading ridges such as East Pacific Rise rift every year or two. Such rifting episodes continue for several years and are associated with multiple dikes with an injection interval of a month up to a year. Diapiric mantle upwelling, structure of magma chambers and volcanic landforms seen in mid-ocean ridges resemble those in some large ophiolites such as the Troodos, Cyprus and the Semail, Oman, and post-Teritary volcanism in Iceland.
Following an episode in Northeast Iceland known as the “Krafla Rifting Episode” 1975-1981, a transient accerelation of the spreading rate between the North American and Eurasian plates was observed by geodetic surveys 1987-1990 using Global Positioning System (GPS). This post-rifting crustal deformation can be interpreted as the response of the shallow elastic layer to the rifting episode (dyke intrusions) delayed by mechanical coupling with an underlying viscous layer. We also found a smaller amount of radial displacements possibly caused by the inflation of the magma chamber beneath the Krafla Caldera.
We have studied recent crustal activities in and around the Izu Peninsula on the basis of crustal movements detected by geodetic surveys. Among them the 1986 eruption of Izu-Oshima and crustal uplift in the eastern Izu Peninsula since 1978 are characterized by magma intrusion. Trilateration and leveling revealed a graben formation associated with the 1986 eruption of Izu-Oshima. Subsidence of average 30 cm relative to the Okata tide station, the northern tip of the island, was observed in the southeastern and northwestern parts of the island, while uplift20 cm was seen in the northeastern and southwestern parts. Furthermore the uplifted parts moved apart and the subsided parts were displaced toward the central cone, Mt. Mihara. We interpret these movements by a vertical tensile fracture model and obtain an opening of 2.7 m of 12 km × 10 km fracture running through the island from northwest to southeast. This model suggests a large scale magma intrusion at the eruption. An anomalous uplift in the northeastern Izu Peninsula has been observed since 1976 accompanied by intensive seismic swarms. Since the occurrence of the Izu-Oshima earthquake on January 14, 1978, swarm activities have been concentrated east off the city of Ito. However the center of uplift was located south of the city of Ito. On the other hand, large crustal extension was observed across the source area of swarm. Submarine eruption occurred in the middle of the source area in 1989. We can interpret these events by a vertical tensile fracture model trending NW-SE as well. W e estimate an opening of 1 m of 20 km × 15 km fracture in the source area of swarm to explain the observed uplift and extension during 1978-1988. The top of the modeled fracture is estimated to be 3 km deep. We consider a 5 km × 5 km fracture situated closer to the land opened by 75 cm associated with the 1989 eruption. Its top margin is as shallow as 1 km for the 1989 eruption. Associated with the 1993 activity, the upward shift of the top margin of tensile fracture from 3 km to 2 km or less was suggested on the basis of temporal variations in side of the network near the source area. These change of depth of top margin of the fracture may mean the upward movement of magma. Therefore we would be able to evaluate the possibility of eruption by means of observation of crustal movements.
A kinematic mechanism of earthquake swarms induced by dyke intrusions was investigated for the seismo-volcanic activity off the east coast of the Izu Peninsula in July 1989. In this activity an intensive earthquake swarm composed of tectonic earthquakes continued for one week with large crustal deformation. From the crustal deformation data Okada and Yamamoto revealed that a dyke with 3 km width intruded from 7 to 1 km depth in the first day of the intensive swarm and then thickned to be 1 to 1.5 m thick. Relocated hypocenters show that the seismically active region moved from deep to shallow during the priod of the ascent of the dyke, suggesting the earthquakes occurred around the top of the intruded dyke. On the basis of the seismic data and the Okada and Yamamoto's dyke model, we examined three possible explanations or the cause of dyke-induced-earthquake swarm, that is, the stress change due to dyke intrusion, open crack-shear crack interaction (Hill's model) and the effect of pore-fluid-pressure increase, In order to explain the focal depth migration, the stress change due to dyke intrusion is the most favorable cause of the earthquake swarm, because the dyke intrusion increases differential stress above the top of the dyke and decreases that below the top. The observed focal mechanism type is consistent with the one expected from the ambient tectonic stress and the stress change due to the dyke intrusion.
In the coastal region of the State of Rio de Janeiro, Brazil, Jurassic (120 Ma) basaltic dykes are present intruding in late Precambrian (600 Ma) gneissic rocks. They are typically 10 m wide, constituted by thoreiitic gabbro, and frequently have sinistral en-echelon setting. Intrusion depth of the present surface exposure is estimated to be 4 km based on apatite fission track dating and regional uplift history. The en-echelon dyke terminals are often linked by a short oblique path, showing extended “Z” -shaped dyke appearance on outcrops. Some of the link paths of wide dykes (>4 m) are accompanied by well-consolidated narrow (<1 m) brecciated zone on both sides, composed uniquely of angular host gneiss fragments. The fragment size is relatively fine (1 cm) on dyke link path side (inner), and gross (10 cm) on host rock side (outer). Large fragments are very angular, and show auto-brecciated texture. The lithologic characteristics of the breccia correspond to those of fault breccia. Geometric setting and kinematic relation between these fault breccia zones and the dykes are similar to those of mid-ocean ridges and transform faults. Dyke dilatation due to magma injection may cause shear stress at the dyke junction zones. If the stress is sufficiently great, the host rock will rupture to form fault fracture zones, and they are successively filled by magma to result the link paths. If the stress is insufficient, there small cracks may be created. Such type of faults, namely dyke linkage faults, are primary and monogenic ones, which are generated in uncracked host rock body with no repeated stick slips. The movement mechanism of dyke linkage faults is similar to the Hill's model, but differs a little from it especially in case of the faults with link path filled by magma. Total displacement of dyke linkage fault is defined by dyke width, Even after the magma fill on the fault plane, fault displacement will continue aseismically. The basaltic dykes in this region commonly have well-developed horizontal columnar joints, but the cooling joints at link paths are pertubated and platy-like joints take place, indicating that the link paths were submitted to shear deformation during magma cooling. In this sense, the earthquakes are more likely to occur only at the top of the intruding dykes. The fault length is defined by en-echelon dyke off-set at junction zones, being several times more than the dyke width. In comparison with tectonic faults and earthquakes, the diplacement is very large relative to the length. The earthquakes generated by dyke linkage fault formation are expected to have strike-slip focal mechanism, linear epicentre trend, nodal plain oblique to the trend, and a high frequency spectrum. Some Japanese volcanic earthquake swarms have such characteristics.
Viscous magma flow in a fissure capable of changing its width is analyzed to understand magma transport in various tectonic conditions. The analysis predicts that the fissure widens only when the magma pressure of the chamber plus twice the tensile stress component normal to the fissure is positive. The effect of the tectonic stress is simulated by use of a simple model of magma plumbing system. For a moderate stress range, magma that has accumulated in the chamber first fills space in a widening fissure and then extrudes out of the surface. When the stress is compressional and strong enough to suppress the effect of accumulation, the fissure cannot open. When the stress is extensional and exceeds a critical value, the fissure widens up to an ultimate finite width with magma kept in it. Recent eruptions in eastern Izu peninsula and Izu-Oshima volcano are interpreted in relation to the difference of the local stress condition.
We propose a model for the cause of peculiar shape of the slab beneath Tokai region. The characteristics in the shape of the subducted slab of Philippine Sea plate in this region are ; (1) an aseismic region to the northwest of Izu peninsula, (2) overlapping of seismic zones beneath Nobi plane around Nagoya, (3) sharp bent of the trough axis between Nankai and Suruga trough. The shape can be interpreted as an effect of the shear deformation around Izu Peninsula. The deformation is produced by both the collision of Izu Peninsula to Eurasian plate and frequent dike intrusions in the volcanic region from Izu-Oshima to eastern Izu Peninsula. Consequently the moving direction of the slab rotates anti-clockwise and the slab moves over the neighboring slab subducted from Nankai trough. This directional change also produce a we ge-shaped gap to the northwest of Izu Peninsula. The shear deformation is consistent with the focal mechanisms and the result of GPS measurement in this region.
Izu Peninsula and adjacent areas, which are located on the northern tip of the Izu-Bonin volcanic arc, are characterized by intense crustal movements and volcanic activity. Many geomorphological, geological, and geophysical data were collected from this area and various tectonic models were proposed to explain them systematically. These tectonic models can be classified into two categories : models 1 and 2. While model 1 regards the area as a single tectonic province, model 2 proposes two or more tectonic provinces, which are bounded by tectonic lines. Models 1 and 2 can be classified into models 1A and 1B, models 2A, 2B, 2C, and 2D, respectively. Model 1A hypothesizes an anticlinal bend of the Philippine Sea plate, which is being generated by the subductions along the Suruga and Sagami troughs. Model 1B emphasizes the crustal stress field generated by the collision of the Izu-Bonin volcanic arc with Japan arc and the slab-pull force along the Sagami trough. Since there are many local tectonic features that cannot be explained by model 1A or 1B, models 2A-2D were proposed. Model 2A divides the study area into two tectonic provinces : the northern province of compressive deformation by conjugate faults and the southern province of right-lateral shearing deformation. Model 2B divides the area into the eastern and western provinces, which are defined by sharp contrasts in the geologic structure, seismicity, crustal stress field, crustal movements, focal mechanisms of earthquakes, paleomagnetic directions, and volcanic activity. Many observations support the validity of the model 2B tectonic provinces. Model 2C introduces the hypothesis that the Izu-Bonin arc is being fractured into the inner and outer arcs because of a contrast in buoyancy. The existence of the estimated model 2C fracture, W est Sagami Bay Fracture, is still under debate. Model 2D regards the Higashi-Izu monogenetic volcano field, located in the eastern Izu Peninsula, as a field of crustal spreading. Model 2D proposes a key to understanding a sharp contrast in tectonic features between the eastern and western provinces of model 2B as well as the complex geometry of the Philippine Sea slab beneath the Japan arc.