The North Pacific Rim is a segment of the circum-Pacific orogenic belt lying along the great circle between Mesoamerica and Indochina. Paleotectonic reconstructions rely upon integration of information about rocks exposed on land, crustal thicknesses, paleolatitudes of crustal blocks, sediment layers cored at sea, and geomagnetic anomalies. Continental margins have been modified by accretion of oceanic materials during subduction, suturing of continental blocks by collision, and opening or trapping of marginal seas. Prior to the breakup of Pangaea, a vast Paleopacific seafloor was built by spreading coeval with the subduction that elsewhere assembled Pangaea. After the breakup of Pangaea, circum-Pacific subduction accreted deformed increments of the Paleopacific seafloor to the edges of continental blocks now along the North Pacific Rim. Cretaceous crustal collisions closed the North Pacific Rim and isolated the Arctic Ocean. Paleogene accretion of the continental Okhotsk block caused subduction to shift from the Bering shelf edge to the Aleutian chain. The elbow in the Emperor-Hawaii hotspot track records a change in Pacific plate motion at about the same time. Current circum-Pacific arcs include east-facing island arcs and west-facing continental arcs in a consistent pattern that implies net westward drift of continental lithosphere with respect to underlying asthenosphere.

View full abstract

Extensive geological and geophysical evidence suggests that numerous fragments of continents, miniplates, and so called "island arcs" have been incorporated into the Circum Pacific continents. The old rocks exposed on these bodies bear strong evidence for continental origins. This leads to the speculation that a large continental mass existed once in what is now the Pacific Ocean. This mass-which we call the Pacifica continent-could have been part of the Pangea Super Continent, adjacent to Australia and Antarctica. When this continent broke into fragments, they drifted toward continental collision in South America, North America, Alaska, Kamchatka, Japan and East Asia. Submerged platforms in the Pacific Ocean, such as the Ontong Java area, the Shatsky rise, and the Manihiki plateau, may also be remnants of Pacifica. The thick crusts of these plateaus, with velocities typical of continents, are thus predicted to be continental crusts.
Although the details of the breakup and collision of Pacifica cannot be resolved very well at present, the postulated existence of this continent supports a large generalization: We suggest that all spreading centers on earth may originate underneath continental masses. Without Pacifica of course, the present day east Pacific rise is without associated continents. If continents account for all spreading, it may be because the continental crust acts as thermal blanket, warming the lower lithosphere and upper asthenosphere.
Adding to this the further hypothesis and that subducted ridges are responsible for back arc rifting and spreading, we obtain the typical trench-continent-ridge sequences, containing volcanism, uplifted blocks, metamorphism, and rifting. Multiple collisions involving several continental slivers and ridges may result in several consecutive sequences juxtaposed on one another. Such complexities with geological record are common in belts such as Western North America.
On the basis of the similarities of (a) the geophysical aspects-seismicity and crustal thickness, (b) morphology, and (c) geological complexities, we propose in this paper that the circum Pacific mountain belts may be at least in part the result of past continental collisions, quite similar to those associated with the Alpine belts.
The notion of Pacifica and its breakup may provide an explanation for the similarities of flora, fauna, and rock sequences in widely separated locations in the mountain belts across the Pacific, and may tie in divergent paleomagnetic data.

View full abstract

While in the Honshu region, the Benioff zone exhibits a straight profile, dipping west at about 30°, the Mariana arc has a curved zone changing into a very steeply dipping zone at depth, and the Peru-Chilean arc has a very extended and gently dipping zone at 16°. These can be interpreted as three type-cases where, correspondingly, the trenchline is at rest with respect to the deeper mantle and the trenchline is moving in the dip direction with respect to the deeper mantle. Where the direction of relative motion has changed, we can then expect sharper bends as in the cases of Central Bonin and the Peru-Chilean arcs.
This interpretation is consistent with the concept of the decoupling of plate motion across the asthenosphere and enables us to use the Benioff Zone to decipher the recent history of absolute plate motion.
The steepening of the Marianas Zone is closely related to the westward motion of the Philippine Sea Plate. The separation of the plates due to anchoring of the trenchline, after the subducted lithosphere reaches the mesosphere, and the continued westward motion of the Philippine Sea Plate is probably responsible for the formation of the back-arc basin and the lack of large shallow thrust-type earthquakes there. This could be one of the several mechanisms for the formation of the back-arc basin.

View full abstract

Some of the fundamental features of plate tectonics are interpreted in connection with the behavior of oceanic crust. It is shown to be likely that the oceanic crust which is produced at the mid-ocean ridge by chemical differentiation may be removed from the downgoing slab by melting at the depth of asthenosphere behind the deep-sea trench. The melting of crustal material after the subduction is made possible by an efficient supply of heat through the well-developed asthenosphere with a low-velocity and high-attenuation of seismic waves. The removal of subducted oceanic crust from the slab is consistent with the positive gravity anomaly behind trenches and the double Benioff zone recently discovered. We propose new type of driving forces of plate motion, which arises from the density contrast between the crust and mantle when the oceanic crust is either created or destructed. The proposed driving mechanism is consistent with the non-uniform size and shape of individual plates, the migration of mid-ocean ridges and compressional intraplate stress, while these facts are difficult to understand in the framework of conventional models. A continuous accumulation of basaltic magma beneath the trench-arc system results in a catastrophic overflow of material, which corresponds to back-arc spreading. The picture presented in this paper explains the evolution of marginal basins that is characterized by the presence of remnant arcs, the changes in stress field and the dip angle of the slab, and the anomalous depth-age relationships.

View full abstract

Near the junction of two arc systems, the state of stress within the earth's crust often appears to differ from the regional trend. The internal deformation of the continental plate at arc-arc junctions is probably controlled by geometrical configurations and the forces acting on the plate edges. If forces acting on the two neighbouring plate boundaries which intersect at a junction are not parallel to each other, convergence (or divergence) or shearing would take place at the junction. At a plate edge, four types of force (or displacement) can be assumed, whose direction is constrained by the strike of the edge. Thus four types of deformation can be expected within the continental plate near an arc-arc junction. One of these four basic types appears to be dominant in the real earth, probably because of the limitation on possible combinations of force (or displacement) types acting on the two neighbouring plate boundaries. The anomaly in the state of stress found near the junctions of the Kurile and Japan arcs and of the Aleutian and Kamchatka arcs may be related to a shearing force acting on the obliquely converging plate boundaries.

View full abstract

The mechanical behavior of a plate entering a trench has been investigated through comparison between the amount of faulting present in the outer trench wall and the maximum bending stress, as inferred from fitting bathymetry data seaward of the trench axis to the theoretical deflection curve for the bending of an elastic plate. Out of 20 circum-Pacific trench profiles examined, 16 satisfy a quantitative test for elastic behavior seaward of the trench axis and exhibit a consistent relationship between maximum inferred bending stress and the roughness of topography on the outer trench wall. This result, in conjunction with the distribution of earthquake focal mechanisms within the subducting plate, is consistent with a model of an 80km thick lithospheric plate having an elastic core approximately 30km thick, outside of which brittle failure occurs.

View full abstract

We have examined agencies inducing flow in the sub-Pacific mantle. The capable agencies are buoyancy forces (causing convection), and passage of the tidal bulge. Bulge passage induces rotational flow dimensionally identical to that induced by buoyancy, and interacting with it. The stress imposed by bulge passage is additive across the width of the sub-Pacific flow cell, in the fashion that the buoyancy stress is additive in thickness. Each stress builds thus to values in excess of 108dyn/cmcm2, overcoming the flow resistance provided by mantle viscosity. The tidal flow contribution favors the development of the west limb of convection cells, because these represent rotational flow in the tidal direction. East limbs are inhibited.
Under these forces the Pacific plate has developed as the cool, rigid surface of mantle convection surfacing in the eastern Pacific, and foundering at the western margin. The west, co-tidal limb of the East Pacific Rise now extends clear across the Pacific. The counter-tidal east limb is dwindling through its replacement by the extending west limb of the Atlantic spreading center, bearing the Americas plates.
The unbroken line of gravity highs at the western Pacific margin marks excess material accumulating where the west limb of the East Pacific Rise encounters the less mobile Asian continental mass. Material added to the asthenosphere causes disruption of the Asia margin, secondary sea-floor spreading, and the formation of arc-bounded basins. At the termination of the east, counter-tidal limb (overlain by the Cocos and Nazca plates) this process is not operative. Marginal basins, if they form, are destroyed along the Andean orogenic front by the westward-advancing Americas plates.

View full abstract

The Eastern Outer Islands form a group of small islands situated in the south-west Pacific Ocean, to the south-east of the main Solomon Islands chain, and to the north of the New Hebrides chain.
They are founded upon the northern part of the submarine Fiji Plateau and flanked by deep sea trenches on the western, northern and eastern sides, but by an east-west trending fracture zone on the southern side. Two south-easterly trending island chains can be recognized, the western chain includes the islands of Tinakula, Nendö, Vanikolo and Utupua, and the eastern chain includes the islands of the Duff Group, Anuta and Fatutaka. The island of Tikopia is situated midway between both chains. The Torres and Vitiaz Trenches form the western and eastern flanks respectively of the north Fiji Plateau, which is itself bordered to the west and east respectively by the Indo-Australian and Pacific lithospheric plates. A westerly extension of the Vitiaz Trench, known as the Cape Johnson Trough, forms the northern boundary. The south side of the region is delimited by the Hazel Holme Fracture Zone.
Volcanic rocks within the group range from picrite basalts, through basalts and andesites, to dacites. Mafic lavas predominate in the older islands of the western chain, two islands of the eastern chain and in samples dredged from the Vitiaz Trench. More sialic lavas are exposed in two islands of the eastern chain, the isolated island of Tikopia and also in the active volcanic island of Tinakula.
Petrologic, petrochemical, seismic and heat flow evidence suggests that the Eastern Outer Islands represent two, discrete south-easterly trending island arcs, each associated with the adjacent trench. The sequence of volcanic episodes can be best explained in terms of plate tectonics, in which the islands of the western chain were produced during two episodes of volcanic activity associated with an easterly inclined Torres subduction zone. Late Oligocene to early Miocene subduction produced the island of Nendö, but the islands of Utupua, Vanikolo and Tinakula were produced during the late Pliocene to Recent. The westerly-dipping Vitiaz subduction zone is postulated as a source from which the east facing Duffs-Anuta-Fatutaka island arc was produced during the Middle Miocene to late Pliocene.

View full abstract

Upper mantle velocity structure in the New Hebrides island arc region has been determined to a depth of 240km from the analysis of P and S wave travel times of 39 deep earthquakes using KAILA's (1969) analytical method. The present analysis reveals a linear increase of P wave velocity from 8.06km/sec at a 40-km depth to 8.19km/sec at a 240-km depth with a gradient of only 0.06±0.01km/sec per 100km. For S waves also, the velocity increases linearly from 4.55km/sec at a 40-km depth to 4.64km/sec at a depth of 220-km with a gradient of only 0.04±0.02km/sec per 100km. The velocity gradients for P and S waves, from 40 to 240km depth in the upper mantle beneath the New Hebrides arc are found to be extremely small as compared to those at similar depths in the adjacent Tonga-Kermadec-New Zealand region. Consequantly, the P and S velocities, to a depth of 240km., beneath the New Hebrides arc are found to be about 6% lower, on the average, than those in the Tonga-Kermadec-New Zealand region. This velocity difference is attributed primarily to very high upper mantle temperatures in the New Hebrides region which may be about 1, 000°C higher than those in the Tonga-Kermadec-New Zealand region. An alternative explanation based on existence of high temperatures in this region is also presented which explains deep seismic activity at 600km depth in the New Hebrides region without invoking the concept of a detached lithospheric slab proposed by some of the earlier workers.

View full abstract

Upper mantle velocity structure in the Tonga-Kermadec island arc region has been studied in great detail to a depth of 665km from the analysis of P and S wave travel times of 130 deep earthquakes using KAILA'S (1969) analytical method. The velocity function for P waves determined from the present study reveals a velocity of 7.97km/sec at a 40-km depth which increases linearly, with a high velocity gradient of 0.52±0.02km/sec per 100km, to 8.96km/sec at a depth of 230km. From 230km downwards, the velocity increases with a considerably smaller gradient of 0.23±0.02km/sec per 100km, reaching a value of 9.37 km/sec at a depth of 410km. At this transition depth of 410km, there is a first order velocity discontinuity-the velocity increasing from 9.37 to 9.85km/sec. In the depth range from 410 to 600km, the P velocity gradient is found to be extremely small, velocity increasing only to a value of 9.90km/sec at a depth of 600km. At this depth of 600km there is again a first order velocity discontinuity-the velocity increasing from 9.90 to 10.64km/sec. Below 600km depth, P velocity increases linearly from 10.64 to 10.72km/sec at a depth of 660km.
The S velocity function, determined to a depth of 665km, also reveals similar features as the P velocity function. However, a decrease in the S velocity gradient is found to occur at a smaller depth of only 130km. The S velocity determined as 4.58km/sec at a 40-km depth increases linearly with a gradient of 0.28±0.02km/sec per 100km, to 4.83km/sec at a depth of 130km. From 130 to 410km depth S velocity is found to increase linearly from 4.83 to 5.11km/sec with a low velocity gradient of only 0.10±0.01km/sec per 100km. At the transition depth of 410km there is a first order velocity discontinuity for S waves-the velocity increasing from 5.11 to 5.32km/sec. Below 410km depth the S velocity gradient is also extremely small, velocity increasing only to a value of 5.39km/sec at a depth of 600km. At this depth of 600km, there is again a first order velocity discontinuity for S waves-the velocity increasing from 5.39 to 6.21km/sec. Below 600km depth, S velocity again increases linearly from 6.21 to 6.34km/sec at a depth of 665km. The decrease in the velocity gradient at a depth of 230km for P waves and 130km for S waves is interpreted as a second order low velocity channel in the upper mantle beneath the Tonga-Kermadec-New Zealand region.
The P and S velocities in the Tonga-Kermadec-New Zealand region are about 6% higher, on the average, than those in the adjacent New Hebrides island arc region to a depth of at least 240km. These velocities are also much higher, by about 3 to 6%, than those in the Japan region to a depth of almost 600km; and also substantially higher than those in the northeastern Australia, the western United States and those for the average upper mantle structure valid for the Pacific ocean basin and its surroundings. It is thus found that the P and S velocities in the inclined seismic zone beneath the Tonga-Kermadec-New Zealand region are probably the highest in the Pacific region and that these velocity differences are prevailing almost to a depth of about 600km. These differences of about 6% in the velocities can be explained primarily by differences in temperatures of the upper mantle material of the order of 1, 000°C. It is therefore inferred that the mantle material in the inclined seismic zone beneath the Tonga-Kermadec-New Zealand region is relatively much colder than that in the Circum-Pacific region.

View full abstract

A morphological study of the Southern part of the New Hebrides island arc system makes it possible to define the different structural units:
1) An accretionary prism with a constant width of approximately 75km does exist;
2) The morphology of this prism varies rapidly along the arc and is linked, firstly to the morphology and structure of the upper part of the dipping plate and, secondly to the presence of volcanic island acting as sediment sources;
3) The island arc and its connecting structural features such as troughs at the rear of the arc, are disrupted by transversal discontinuities, the largest of which could be related to fractures of the oceanic crust of the dipping plate.

View full abstract

Upper Triassic shale from central Seram gives a direction of magnetization (D=82, I=23) which (assuming, as seems probable, that it was acquired before folding, and during a period of reversed magnetization) indicates Seram lay in 12±7°S latitude and has rotated anticlockwise 98° since the late Triassic. The equivalent (south) palaeomagnetic pole is at 8°N, 207°E.
Pillow basalt of late Miocene age, from Kelang Island, west Seram has a magnetic vector D=106, I=9, indicating extrusion in 5°S latitude, and anticlockwise rotation of 74°. The equivalent (south) palaeomagnetic pole is at 16°S, 214°E.
These results, although only of a reconnaissance nature, are consistent with the postulations of various authors that Seram has rotated anticlockwise in late Cainozoic.

View full abstract

K-Ar determinations for ten samples of the basaltic lavas of Pulau Kelang, Seram (for which palaeomagnetic measurements indicate a reversed palaeomagnetic pole at 16°S, 214°E) yield a mean age of 7.6±1.4Ma. Direct evidence for the age of the lavas is lacking. This K-Ar date confirms a previous inference based on young topographic features of the island that the lavas are of late Cainozoic age and allows a more precise stratigraphic assignment to the Upper Miocene.

View full abstract

A compilation is given of all the pole positions either calculated from or given in several paleomagnetic studies carried out so far on Mexican regions. These are compared with the paleomagnetic data from North America (N.A.). The results (Oligocene data) clearly show the presence of relative tectonic movements of the Western Cordillera. The data for Recent to Miocene age seem to be consistent with N.A. data except Pliocene pole positions for Baja California which is taken as an evidence of movement of Baja California relative to Mexico or North America. The Mesozoic pole positions are also considered to show the possibility of tectonic instability of Mexico relative to 'stable' North America. Practically no work has been done so far on older rocks (Pre-Mesozoic) of Mexico.

View full abstract

The three major tin granitoid belts are, from west to east:
a) Western: From Phuket to Tenasserim. The tin deposits are associated with highlevel Cretaceous adamellite, granite and pegmatite. The mineralization extends over a wide vertical extent.
b) Main Range: From Bangka to South Thailand. The tin deposits are associated with deep-seated large-microcline granite and adamellite of late Carboniferous and late Triassic age. The mineralization is confined to the roof zones of the batholith.
c) Eastern: From Billiton to Pahang-Trengganu. The tin-tungsten deposits are associated with adamellite to granite of Permian to mid Triassic high-level plutons. Mineralization extends over a wide vertical extent and there is an important Fe-Sn association.
Only the Eastern Belt can be classified as Circum-Pacific type. It represents an epizonal volcano-plutonic arc characterized by plutonic rocks ranging from gabbro, through tonalite, granodiotite, adamellite to granite. Rhyolitic ignimbritic volcanic rocks are important.
The Western Belt has some of the Eastern Belt characteristics, but lacks the range of plutonic rocks, and volcanic rocks are absent. However, the Tertiary opening of the Andaman Sea requires that the Burmese-Indonesian volcanic arc formerly was adjacent to the adamellite-granite belt.
The Main Range Belt is interpreted as resulting from crustal anatexis of the leading edge of the western craton as it attempted to subduct beneath the Eastern volcano-plutonic arc following the late Triassic closure of the central marginal basin.

View full abstract

An analysis of the available bathymetric, gravity and seismicity data of the Andaman Sea that lies between the latitudes 4° to 16°N and longitudes 91° to 97°E has been carried out. A seismicity map of the Andaman Sea has been prepared using all available data for the period 1916-1975. The epicentral distribution parallels the structural lineation of the Andaman Arc showing larger activity under the Andaman Basin. An inclined seismic zone extending at least up to a depth of 150km is found to be present underneath the Andaman Basin. The seismic zone dips towards the continental side, attaining its deepest part not right below the axis of the negative gravity anomalies, but is rather deflected further east towards the volcanic islands. For a more complete understanding of plate motions in Andaman Sea area, all available focal mechanism solutions incorporating 12 new ones have been analysed. It is observed that normal, thrust as well as strike-slip faulting take place in the area. The direction of seismic slip vectors for thrust-type solutions is generally eastward over the areas of outer sedimentary arc. In the southern part of the Andaman Sea, the direction of seismic slip vectors is towards south. Over the southeastern part of the Andaman Sea, in continuation of the Semangko Fault (rift) of Sumatra, strike-slip mechanism predominates. This gives a clear evidence of the continuation of the Semangko Fault beyond Sumatra underneath the Andaman Sea up to 9°N. The results obtained through the focal mechanism studies further suggest that thrust faulting is prevalent between 30 to 90km depth under the Andaman Sea. Normal faulting prevails at shallow as well as at greater depths (more than 90km). The transcurrent type of movement appears to affect a considerable thickness of the lithosphere in the area, at least up to a depth of 150km. On the basis of these results it is suggested that an active subduction zone is present underneath the Andaman Basin. The present structural form as well as seismicity of the Andaman and the west-Indonesian Arc seems to have resulted as a result of subduction of the Indian Plate at the continental margin of the SE-Asian Plate.

View full abstract

Seismic activity and focal mechanisms in the vicinity of Taiwan and the Philippines are studied to elucidate the tectonics in this complex region. Epicentral distribution and vertical profiles of earthquake foci support the idea that the entire Philippines is not a part of the Eurasian plate, but another block of lithosphere and the relative motion between the Philippine Sea and the Eurasian plates is shared by the subduction along the two boundaries, west and east of the Philippines. Thrust type of mechanisms are the dominant mode of deformation along the eastern margin of the Philippines; in contrast, no thrust type of solutions are obtained along the western margin of this islands. Between Taiwan and Luzon, mode of plate consumption is most complex. Seismic activity is mostly shallow and diffuse in a 200km wide zone. Reverse faultings along the eastern margin of Taiwan, strike-slip faultings off the southeastern coast of Taiwan, and normal faultings between the Manila trench and the North-Luzon trough are the major mode of deformation. We believe that the region between Taiwan and Luzon constitutes a left-lateral shear zone due to the gradual transition of site of plate consumption from eastern margin of Taiwan to east Luzon. Subduction under the west-facing Luzon arc in this region is likely to be ending now; this might be causing the opening of the sedimentary wedge behind the Manila trench. Normal faultings in this study probably indicate the initial phase or prespreading phase of opening of the area behind the Manila trench.

View full abstract

Taiwan represents a very young arc-continental margin collision zone in a long subduction boundary. The collision started in Late Pliocene and is still vigorously taking place. Off coast of NE Taiwan a northward subducting slab, extending west at depth to northern Taiwan, can clearly be defined; although most parts of Taiwan have been rising steadily at about 5mm/year for the last 8, 500 years, northern Taiwan has had periods with no uplift. The intensity of the collision decreases toward the south off the island, and an east-dipping subduction zone can be delineated there. Thus Taiwan can be viewed as a transform zone in between two subduction zones with quite different geometries.
Seismically Taiwan is much more active than its neighbors, the Ryukyus and Luzon. Large earthquakes reveal the nature of the intense on-going intra-plate deformation; on land, EW compression or left-lateral shear occur along NNE faults and right-lateral shear occur along nearly EW faults; offshore to the southeast of Taiwan, left-lateral shear along NWW or NW faults and thrusts in several directions coexist; to the northeast, the focal mechanisms agree well with other subduction zones. The Ryukyus are terminated at about 123° E by a number of NNE striking right-lateral faults. Focal mechanisms to the southeast of Taiwan are consistent with a tectonic stress direction of S46° E to S76° E and plunging at -2°to 15°.

View full abstract

The geological and structural contrast between the north and central Ryukyus and the south Ryukyus has been significant since the Late Mesozoic. The difference seems to correspond to that of the nature of the Philippine Sea floor facing the Ryukyus, i.e. the Daito Ridges and Amami plateau to the north and deeper basin to the south. The north and central Ryukyus were a separate tectonic unit from the south Ryukyus from the Late Mesozoic to Middle Tertiary. Subsequently they have united to form an island arc as the island groups shifted southeastwards with different rates in the Late Tertiary to Quaternary.

View full abstract

A series of seismological observations of explosions from 1972 to 1977 has been worked out to study underground structure and possible anomaly of wave propagation beneath and around the Sakurajima Volcano and the Aira Caldera as well as regional structure of south Kyushu. Results described in this report are summarized as follows:
1. Name, velocity and bottom depth of the identified layers are, L-1, 3.7-3.8km/sec, 0.7-1.8km; L-2, 4.7-4.9km/sec, 3.3-5.6km; L-3, 5.6-6.1km/sec, 22km; L-4, 7.0km/sec, (40km); L-5, (7.8km/sec).
2. Two fan-shooting observations revealed following anomalies of wave propagation.
1) A large attenuation of the amplitude of seismic waves occurs under the Sakurajima Volcano and the Aira Caldera.
2) Wave velocity is likely to decrease under the Sakurajima Volcano.
3) These anomalies of wave propagation occurs at the 6km/sec layer.

View full abstract

Sedimentary and structural characteristics in two types of accretionary fold belts in central Japan are introduced and the tectonic significance are discussed. The Shimanto Fold Belt, grown up from the Cretaceous arc-trench gap and trench slope sediments, has the collisional features of large scale folds. The fold styles in the belt are differently developed in inner and outer parts. The former is characterized by a series of shear folds and the latter is by a series of lens folds. This difference may be caused chiefly by the different geothermal gradients of the area. The Miocene Miura Basin is influenced by lateral compression of strike slip field at the time of sedimentation. The development of the basin was related to the strike slip motion at the plate boundary between the Philippine Sea and Asian Plates.

View full abstract

The Honshu Geosyncline of middle Paleozoic to early Mesozoic age suffered a strong diastrophism in late Permian to early Triassic as shown by many stratigraphic breaks in the Maizuru and other Belts of Southwest Japan. This geosyncline survived through the Triassic Period in the Tamba and other Belts. The Tamba Belt had many troughs, submarine volcanism and also some tectonic lands. Abundant detrital materials had been supplied from both northern uplifting land, and southern tectonic lands. Detailed nature of sediments and their facies distribution and paleogeography of the middle to late Triassic Period are discussed.

View full abstract

A detailed metamorphic zonal mapping is being in progress on the Sanbagawa metamorphic belt in central Shikoku. The mapping is based upon the distribution of index minerals, garnet and biotite in pelitic schists, and on the sliding equilibrium among silicate and oxide minerals. The distribution of mineral zones has revealed a peculiar thermal structure of the metamorphic complex that the highest-grade rocks occur in the middle of apparent stratigraphy. A large scale recumbent fold, with south vergency and extending for more than 20km, is postulated as a possible structural interpretation.
It is concluded, as the most probable model we could imagine at the moment, that before the maximum temperature of metamorphism was reached, the Sanbagawa schists had been metamorphosed in more or less normal thermal regime that the temperature had increased downwards. Then a large scale recumbent fold took place, separating the higher-grade rocks from the heat source and bringing them in between the lower-grade ones. This recumbent fold was accompanied by the start of the uplift of the whole metamorphic complex, while continuing metamorphic reactions with decreasing temperature and pressure.
The fact that the Sanbagawa belt is overturned suggests that a very distinctive crustal shortening took place in the present day Sanbagawa terrain in the Mesozoic time, and that the present day distribution of pre-Tertiary geologic units in the outer zone of the south-western Japan can hardly be in situ.

View full abstract

The late Mesozoic to early Neogene geosyncline in the Outer zone of Southwest Japan has been studied in detail in the Kii Peninsula by the Research Group for the Shimanto Geosyncline. The existence of the Kuroshio Paleoland to the south of the geosyncline was inferred by various sedimentologic evidences. The Shimanto belt in the Kii Peninsula is divided from north to south into three zones of Cretaceous, Eocene and Oligocene to lower Miocene. In these belts thick geosynclinal sediments were accumulated showing coarsening upward. The southward migration of the basin occurred in Cretaceous/Eocene, Eocene/Oligocene, and in early Miocene. In the present paper the reconstruction of paleogeography of the Shimanto geosyncline was attempted and the Kuroshio Paleoland was discussed in relation to the geohistory of the Philippine Sea. In spite of the detailed geologic survey in the Kii Peninsula there is no evidence of large exotic blocks nor tectonic melanges, and this does not support the plate tectonic model of the Pacific-type orogeny for the Shimanto belt.

View full abstract

Detailed correlation and chronology of the Neogene and Quaternary marine sediments in the Japanese island arc area have been established by the application of magnetostratigraphic methods supplemented by microbiostratigraphic data. The sedimentation rates of several sedimentary sequences show a similar pattern among the studied areas, and two drastic synchronous changes in the rates of sedimentation are recognized. Thus the Japanese Neogene and Quaternary can be divided into three major time intervals, named I, II, and III in increasing age. The boundaries between these three intervals are 4.7 mybp (base of the Gilbert Epoch; magnetic anomaly3) and 10.4mybp (Epoch 9; magnetic anomaly 5). The geographic distribution of the land area during the time interval I and II was similar to the present; however, in the time interval III, it is completely different but similar to the present Bonin-Mariana arc area. It has been documented by Hawaiian hot spots and spreading features on the East Pacific Rise that the plate motion in the Pacific Ocean area has also changed drastically. The time interval I is the period of high rate of sedimentation (several hundreds cm/1, 000 years) and moderately increasing plate motion; the time interval II extremely low rate of sedimentation (less than several cm/1, 000 years) and slow plate motion, and at the same time land areas were expanded; the time interval III moderate rate of sedimentation (several tens cm/1, 000 years) and high rate of plate motion, and land areas were reduced. These drastic changes can be explained by the "cyclic evolutionary model", originally proposed by Kanamori, and Forsyth and Uyeda's slab-pulling driving force of the oceanic plate motion as follows. The drastic change from the time interval III to II is ascribable to detachment of the down-going slab from the ocean plate. The reduction in plate motion may also be triggered by the detachment, which releases the ocean plate from the down pulling force. The high rate of sedimentation in the time interval I is resulted from the steepening in the topographic relief and the increase in the amplitude of tectonic deformation, which should be related with the horizontal compressional stress caused by the coupling between the continental and oceanic lithospheres.

View full abstract

In the Japanese coastal area, deformation patterns deduced from the height of former shorelines are classified into four types, A, B, C and D, each reflecting different response of tectonic regions to island arc movements. Each area has been progressively and acceleratedly deformed in a same pattern during the late Quaternary. Maximum rate of average uplift is 1.5m/1, 000 years for the Last Interglacial terrace and 4m/1, 000 years for the Holocene terrace.
The landward tilting, type D, on the Pacific coast of Southwest Japan has been associated with great earthquakes occurring below the inner slope of the Nankai Trough. Type D area is separated from upwarping mountains by hinge lines along which subsidence has accumulated. Tectonic basins filled with younger sediments on the continental slopes are assumed to be depressed zones along the former hinge lines. Ages of deformed shorelines suggest that until the early Pleistocene seismic deformation had affected only the continental slopes and later propagated onto the coastal area in the late Pleistocene at the latest, resulting in the intensification of deformation of former shorelines. In contrast, type C deformation has predominated on the Pacific coast of Northeast Japan, which are 200km distant from the Japan Trench and seem to be still located landward of hinge lines of seismic crustal movement.

View full abstract

Detailed analysis of magnetic anomalies has revealed a clear pattern of symmetric lineations in the Shikoku Inter-arc Basin, northern Philippine Sea. Amplitudes of anomalies are in general a few hundred nannotesla (gammas, peak to peak), which are moderate compared to those of the normal ocean basins accreted from the mid-oceanic ridges and are relatively larger than those of some other inter-arc basins such as the Parece Vela Basin, Mariana Trough and West Philippine Basin. Correlation of anomalies is usually so good that age identification can be convincingly performed except for the axial irregular zone. Mode of opening derived from the distribution of magnetic anomalies as well as the topographic features provides the evolutionary history of the Shikoku Basin in the following manner:
1) The Kyushu-Palau and Shichito-Iwojima Ridges began rifting at their northern end at about 30 mybp. The rifting propagated towards south at a speed of about 10cm/year.
2) After the whole basin was rifted at about 25 mybp, it continued to open symmetrically from the central spreading axis at a half rate of nearly 4cm/year until about 22 mybp.
3) In the latest stage of opening the spreading became slower and even irregular. The spreading axis jumped in some parts of the basin. A chain of seamounts was formed and widespread off-ridge intrusions occurred in the eastern portion of the basin.

View full abstract

Magnetic data obtained during the period of the Geodynamics Project by various institutions are compiled and a new distribution map of magnetic anomaly lineations is proposed for the northwestern Pacific Basin and in the north Philippine Sea.

View full abstract

The Izu-Bonin arc joins with Honshu at the Fossa Magna where pre-Miocene terrains bend in a cusp form. The Miocene terrains in the region also have a northward-convex structure north of the Izu Peninsula. Moreover, highly compressive deformation, Quaternary strong uplift and anomalous trajectories of crustal stress axes also characterize this region.
These features of central Honshu at the junction are explained well by assuming that a north-south trending plate boundary has been located off central Honshu since the late Cretaceous. The bend of the terrains was formed for the most part in the early Tertiary by buoyant subduction of aseismic ridges lying along the north-south trending transform fault.
The Izu-Bonin arc, which was developed along this transform fault, has been dragged northward by oblique subduction of the Pacific plate and underwent subduction beneath central Honshu during the late Tertiary. In the early Quaternary, the Izu Block (the Izu Peninsula) of the Izu-Bonin arc collided with central Honshu and is pushing it north-northwestward. It is very likely that the triple junction off central Honshu has been located at its present position relative to Honshu since the late Mesozoic.

View full abstract

Dislocation densities of olivine grains in peridotite nodules from Ichinomegata (25 samples), Sannomegata (18 samples), Oki-Dogo (13 samples), Hamada (3 samples) and Takashima (10 samples) were measured for estimating differential stress in the upper mantle. The density was in the range of 10⁶-10⁷cm^-2. The differential stress is estimated as 100-300 bars using KOHLSTEDT and GOETZE'S empirical relationship (1974) between the dislocation density and the differential stress. Strain rate is inferred from the geotherm of the island arc and the flow law of olivine single crystal proposed by KOHLSTEDT and GOETZE (1974) and DURHAM and GOETZE (1977). Strain rate in the upper mantle from 30 to 100km depth is less than 10^-15sec^-1, but it is nearly constant around 10^-13-10^-12sec^-1 from 100 to 200km depth.
The convective flow induced by descending oceanic plate is suggested and the velocity of the return flow is estimated to be 5cm/year. The return flow of the upper mantle toward the trench causes the lateral variation of magma chemistry and the upper mantle materials, if partially molten liquid continuously flows out from the moving upper mantle. The model in this study gives a relation between the horizontal distance from the volcanic front, X, and the concentration ratio of C_f¹/C¹ (C_f¹; concentration at volcanic front) as follows;
C_f¹/C¹=exp(-βφX/v),
in which β and v are the rate of outflow of the liquid per unit volume of the upper mantle and the mean velocity of the return flow, and φ is a constant.

View full abstract

Recent seismic researches have revealed the variation of shallow seismic activity across the trench-arc system of Japan, which is characterized by, A) a high activity throughout the crust and upper mantle between trench axis and "aseismic front", B) a low activity between "aseismic-front" and volcanic front, and C) a moderate activity within the upper crust behind volcanic front. This variation of shallow seismicity across the island arc is explained by the effects of pore pressure in the crust, based on laboratory experiments of stick-slip and stable-sliding movements combined with temperature distribution estimated from terrestrial heat flow and thermal model of subducting slab. High pore pressure is expected in the low activity area (region B). Pore pressure is controlled mainly by 1) fluid supply from below (released from subducting oceanic crust), and 2) permeability distribution in the crust and upper mantle. Dehydration process in subducting slab has a great influence on shallow seismic activity.

View full abstract

Along arcs, a belt-like area can be identified where shallow seismicity within the continental plate is extremely low compared with other parts of the plate margin. Examples are in the Tohoku, Hokkaido, Kurile, Kamchatka, Aleutian, Peru-Chile, New Hebrides and Tonga arcs. The present paper proposes to call this inactive (or less-active) area the "aseismic belt, " which seems to be a typical feature of arcs. The aseismic belt is some tens of kilometers wide and is located, generally speaking, along the frontal non-volcanic arc between the volcanic front and the aseismic front. This belt can be explained as a mechanically unstrained area on the basis of a plausible model of plate subduction. Geodetic data and seismological results obtained in Japan are incorporated into this model in the framework of plate tectonics.

View full abstract

The tsunamicity of the Sanriku Coast has been shown to have an anomalous gap in energy, or tsunami magnitude. Furthermore, this gap in magnitude correlates directly with epicentral distance from shore-there also is a gap in the epicenter distribution versus offshore distance. The large tsunamis are generated on the east side of the trench. The gap is for intermediate tsunamis, of which there are none. These observed features can be explained by considering the spatial distribution of shear stress about a subduction zone. The stress attains maxima at two locations: one is on the upper interface between the mantle and the downgoing slab; the other is within the oceanic lithosphere, at the flexure. The observation rationalizes the observed anomalies of tsunamicity.

View full abstract

Distribution of intermediate and deep earthquake hypocenters beneath the Hokkaido corner, one of the notable arc-junctions in Japan, is investigated. A specially designed projection of the hypocenters is made to demonstrate the configuration of the seismic zone beneath the arc-junction. The configuration is characterized by a peculiar folding in the depth range between 80 and 150km, which seems to be caused by plastic deformation of the descending lithosphere.
Attenuation of seismic waves of local earthquakes is also studied. Q values for S waves fall in the range between 50 and 200 in the upper mantle beneath Hokkaido. The distribution of Q values shows that the high attenuation zones correspond to the areas of high heat flow.

View full abstract

In northwestern Hokkaido, where the Honshu and Kuril arcs intersect, an enormous pile of Neogene sediments (about 10km thick) is developed in close relation to the Cenozoic orogenesis. The Kotanbetsu Formation, representing the lower to middle Miocene sequence of these sediments, exceeds 3km. in thickness and is characterized by gravity-flow deposits. The strata are divisible into three major facies: 1) chaotic deposits or olistostrome facies, 2) graded-bed or turbidite facies, and 3) ripple-bed or contourite facies.
The chaotic deposit facies, rather restricted in distribution in the proximal parts of the basin, is further subdivided into three sedimentary types: pebbly mudstone, chaotic breccia, and matrix-deficient conglomerate. The graded-bed and the ripple-bed facies are predominant in the relatively distal parts of the basin. The chaotic deposits were generated by intense tectonic movements of the 'Mesozoic' basement rocks thrust up to towards the west. The graded beds were deposited from turbidity currents, although their current directions are not clearly defined, whilst the ripple beds were formed by southward-flowing contour currents.
These characteristic sediments are comparable in their tectonic position of an apparent back-arc belt to the Onerahi Chaos-Breccia of Northland, New Zealand, related to the Miocene to Pliocene Kaikoura Orogeny. The Kotanbetsu and its equivalents, however, may have been the deposits in basins produced by collision of two continental blocks, known generally as the Hidaka Orogeny, but not in the ancient back-arc belt.

View full abstract

Seismic waves generated by two explosions of dynamite, 5 tons each, in the Sea of Japan off the coast of northern Honshu were observed at more than 100 temporary and permanent seismological stations in the Hokkaido, Honshu, Sado, and Oki islands. A purpose of these measurements was to extend our investigation of lateral variation in Pn velocity which has been found around northeastern Japan in the previous explosion experiments. In fact, a lateral variation in Pn velocity by about 5% was confirmed in regions of the uppermost mantle below the Sea of Japan and the Honshu island, although the boundary where the velocity change takes place was not determined.
The measurements have also revealed an indication that the upper mantle just beneath the Moho interface under the area in the southeastern half of the Sea of Japan is anisotropic with respect to P-wave velocity. The velocity variation in the anisotropy is approximately 0.4km/sec (i.e., 5%) about a mean velocity of 7.94km/sec. The direction of the maximum velocity is 141°E of north which corresponds roughly to a direction perpendicular to the general trend of northern Honshu as well as to magnetic lineations.

View full abstract

Geodynamics and tectonic evolution of the North-Eastern Asia was determined by the establishment of those structural elements in the geological sections which are typical of the modern active continental margins.
The Neogene island arcs on the majority of their strike coincide with the modern ones (the Kuril-Kamchatka) but locally they are broken up (the Western Sakhalin). They are conjugated with the back, front and interarc troughs. The Paleogene island arc occurs, probably, in the northwestern Kamchatka and on the Academy of Sciences uplift in the Sea of Okhotsk. The Early Mesozoic Uda-Murgal island arc marks the southeastern boundary of the ancient Eastern Siberia megablock. Along its southern and northern boundaries the island arc complexes of the same age are distinguished. Position of Eugeosynclinal zones on the outer side of the island arcs indicates the isolation of the Eastern Siberia megablock from the Bureya-Ehanka, Okhotomorskiy and Chukotka ones the basement of which is composed by the Early Precambrian metamorphic complexes of sialic composition and granitoids. Tectonic movements at the end of Neocomian are expressed in the formation of the Andian type active continental margin with the Okhotsk-Chukotka margin-continental volcanic belt originated on a site of the Uda-Murgal island arc. This process was accompanied by the geometrical changes of the Benioff zone determined by the analysis of K₂O content in volcanites on the base of Dickinson and Hatherton's diagrams -the angle of dip became more gentle, the width along the dip and depth of the magma active part of the Benioff zone were increased. Simultaneously with the development of the Okhotsk-Chukotka volcanic belt there was formed the island arc on the East of the Sikhote-Alin. Owing to the Pre-Senonian movements on a site of this arc the Eastern Sikhote-Alin margin-continental volcanic belt similar to the Okhotsk-Chukotka one originated.
Many of the island arcs and margin-continental volcanic belts occur along the margins of the ancient sialic megablocks. Paleotectonic reconstructions prove the fact that there were significant horizontal displacements of the sialic megablocks causing the crash of the oceanic basins situated between them. The enlargement of the eastern part of the Asian continent took place not only at the expense of the island arcs displacement towards the ocean, but also due to the sialic blocks joining.
Geodynamics and tectonic evolution of the continental margin is revealed by the determination of those structural elements in the geological sections which are typical of the modern and Late Cenozoic geosynclinal systems (island arcs). The most significant attention is paid to the tectonic position and nature of volcanic belts of different types and ages widely distributed in the North East of Asia. They help to reconstruct the former plate boundaries and to determine the kinematics of their movements accroding to the plate tectonics models. Among them we distinguish: (1) the inner island arcs or volcanic geanticlines, (2) margin-continental volcanic belts, (3) plateau-basalts coinciding with the margin-continental belts in the space but being the independent formations.

View full abstract

The reconstructions of Mesozoic pre-drift arrangement of tectonic units, grounded on the geological and geophysical data and on the fit of continental slope contours, were compiled for the regions of the sea of Japan, the South China sea, the Tasman and Caribbean seas and the Gulf of Mexico. These reconstructions confirm a supposition that the basins of marginal seas were formed by the tension and break of the continental crust. In the deep basins a crust of oceanic type was formed by the sea floor spreading. The tension and rupture of the earth's crust was due in most areas of marginal seas to the drift of island arcs from the continent towards the Pacific, but in other areas (the Gulf of Mexico, the Caribbean and Tasman seas) it was connected with the removal of adjacent large continental blocks (N. and S. Americas, Australia and Antarctica).

View full abstract

The authors emphasize the important role of the large scale strike-slip faults in the tectonic history of the circum-Japan Sea region including the origin of the Japan Sea Basins. NE-SW to NNE-SSW trending faults in the region were formed by the intense compression from the south which corresponds to the "pulse" suggested by LARSON and PITMAN (1972). NS to NNW-SSE trending faults in Northeast Japan were formed one after another from east to west by the left-lateral simple shear force induced by the subduction of the peculiar transform fault between the Kula and Tethys plates. Subduction of the seamount chain on the peculiar transform fault formed NW-SE trending faults and the cusp structure in the Kanto region. The subduction of the Kula-Pacific and Tethys ridges produced the acidic igneous activities of the Cretaceous to Paleogene in the Asiatic continental margin. Japan Sea Basins were formed in the Paleogene time by the southward drift of the western part of Japan bounded on the east by the Tanakura shear zone and on the west by the Tsushima fault, both of which had already been formed as left-lateral faults in the Cretaceous time.

View full abstract

Definite spatial and temporal relations were found between large interplate earthquakes and eruptive activities along the arc systems in the northwestern Pacific margin; Kurile-Kamchatka and Japan areas. A major pattern of the relation between both phenomena reveals that eruptive activity increases during the preseismic stage to the landward of the rupture zone in the direction parallel to the convergent plate motion. Increased eruptive activity had occurred from between 2 to 27 years prior to the large earthquakes. The volcanoes usually cease their activities and sometimes decrease them after seismicity. The intense activity generally migrates to adjacent areas, namely to the landward extension of seismicity gaps. These observations suggest that the eruptive activity is strongly influenced by regional tectonic stresses, and that the accumulated crustal strain may squeeze up magmas. Support for this conclusion was obtained by observing the level change of the magma head at Mihara-yama in Oshima Volcano, Japan. Some of the volcanoes also erupted immediately after the large earthquakes. One explanation for the post-shock eruption may be provided by considering the possible effect of seismic shocks on the catastrophic seismic events. This study may provide an important key for predicting large interplate earthquakes and major eruptions in the arc areas.

View full abstract

Partial melting at the hypocentral depths (30 to 40km) is closely related to great earthquakes occurring in comparatively high heat flow zones such as the sea just south off the southwestern Japan. It is obvious from the thermal gradient curves in such high heat flow zones that temperature at the hypocentral depths approaches the melting point of wet peridotite. Under such circumstances the partial melting proceeds by the effect of some pressure decrease or temperature increase associated with the huge tectonic force acting on the rocks. The partial melting will yield some increase in volume. Consequently some stresses will be originated and added to the tectonic forces due to the mantle convection and other origins to trigger earthquakes. Changes in seismic wave velocities owing to the partial melting is calculated in this paper using the Lindemann's equation based upon the Debye's theory of specific heat.

View full abstract

The intermediate and deep earthquake zone comes into existence under the condition that the continental side upheaves and the oceanic side subsides, judging from the geologic development of East Asia including the Japanese Islands since the Mesozoic age.
The authors try an experiment by F.E.M. to examine the internal condition accompanied by the vertical movement in the deeper part of the earth and explain the geologic development of East Asia including the Japanese Islands and the earthquake zone.

View full abstract