中西 正男

Abstract To obtain novel information regarding normal faults that could cause tsunamis and large earthquakes on the outer slopes of the Kuril Trench, recently acquired multichannel seismic reflection (MCS) and bathymetric data are processed in this study to interpret the normal faults on the outer slope of the western Kuril Trench from near the Erimo Seamount to near the Nosappu Fracture Zone. This work also assesses the distribution and characteristics of the faults through comparisons with previous studies on the outer slope faults of the Japan Trench. The results of this study show that faulting starts approximately 100 km seaward of the trench axis, and a comparison with the normal fault distributions on the outer slope of the Japan Trench presented in previous studies reveals that the outer slope region of the Kuril Trench is narrower and more densely faulted than that of the Japan Trench. Additionally, the seismic and bathymetric data show that the Erimo and Takuyo-Daiichi Seamounts are deformed by normal faults. However, the number of faults in the seamount areas is lower than that in areas without them, which also affects the scale of the fault throws. The results of the MCS survey reveal that the thickness of the oceanic crust varies between 6 and 10 km due to the presence of these seamounts. Considering the positions of the outer swells presented in previous studies, the results suggest that the crustal and plate structures in the study area influence the distribution and formation of faults on the outer slope of the Kuril Trench. Graphical Abstract

Abstract Areas of old and cold oceanic plate lack conventional volcanism and have been assumed to be devoid of submarine hydrothermal activity. However, petit-spot volcanoes are common in areas of flexure of such oceanic plates. Here, we report hydrothermal ferromanganese oxides dredged from the vicinity of a petit-spot volcano at 5.7 km water depth in an area of oceanic plate flexure east of the Japan Trench. The bulk chemical, lead isotopic and mineralogical compositions of the samples indicate their formation by low-temperature hydrothermal activity, which can be interpreted as being caused by fluid–rock interactions at <200 °C. We propose that interaction of local marine sediments with volatile-rich petit-spot magmas may produce hydrothermal fluids containing not only iron and manganese but also enough amounts of carbon dioxide and methane to have implications for the global carbon cycle. However, contemporary hydrothermal activities at petit-spot volcanoes have not been confirmed yet.

Abstract The formation of the Ontong Java Nui super oceanic plateau (OJN), which is based on the model that the submarine Ontong Java Plateau (OJP), Manihiki Plateau (MP), and Hikurangi Plateau (HP) were once its contiguous fragments, could have been the largest globally consequential volcanic event in Earth’s history. This OJN hypothesis has been debated given the paucity of evidence, for example, the differences in crustal thickness, the compositional gap between MP and OJP basalts and the apparent older age of both plateaus relative to HP remain unresolved. Here we investigate the geochemical and ⁴⁰Ar-³⁹Ar ages of dredged rocks recovered from the OJP’s eastern margin. Volcanic rocks having compositions that match the low-Ti MP basalts are reported for the first time on the OJP and new ~ 96–116 Ma and 67–68 Ma ⁴⁰Ar-³⁹Ar age data bridge the temporal gap between OJP and HP. These results provide new evidence for the Ontong Java Nui hypothesis and a framework for an integrated tectonomagmatic evolution of the OJP, MP, and HP. The isotopic data imply four mantle components in the source of OJN that are also expressed in present-day Pacific hotspots sources, indicating origin from (and longevity of) the Pacific Large Low Shear-wave Velocity Province.

The stress field of oceanic lithosphere controls the distribution of submarine petit-spot volcanoes. However, the eruption sites of these petit-spot volcanoes are considered to be limited to concavely flexed regions of lithosphere off the outer rise. Here, we present new data for a recently identified petit-spot lava field on a convexly flexed section of the lithosphere adjacent to the subduction zone offshore of northeast Japan in an area containing more than 80 volcanoes. This area is marked by strongly alkaline lavas that were erupted on the convexly flexed region. As for the concavely flexed region where the petit-spots previously reported, the base of the lithosphere beneath the eruption sites is under extension, whereas the upper part of the lithosphere is under compression. This change in the stress field, from the lower to upper lithosphere, causes ascending dikes to stall in the mid-lithosphere, leading to metasomatic interaction with the surrounding peridotite. The new geochemical data of rocks and xenocrysts presented in this study indicate that strongly alkaline magmas erupted on the convexly flexed region would have ascended more rapidly through the mid-depth of lithosphere because of the extensional regime of the upper lithosphere and decreasing the degree of metasomatic reaction with the surrounding mantle peridotite. The results indicate that the degree of metasomatism and the compositional variations of petit-spot magmas are controlled mainly by the stress field of the lithosphere.

<p> It is generally accepted that hydration due to plate bending-induced normal faults (bend-faults) occurs in the region between a trench and an outer rise. Hydration of the oceanic plate has played a major role in global deep water circulation, as well as co-seismic megathrust slip at subduction zones. It is, however, emphasized that little is known yet about the degree and mode of hydration in the oceanic plate at outer rises. Investigating several subduction zones with various conditions is crucial to expand our knowledge of bend-fault hydration processes. The northwest Pacific (NW Pacific) region is one of the oldest, thus coldest, and most studied oceanic plates. Water circulation and hydration through bend-faults in the NW Pacific region are supported by the results of extensive recent geophysical surveys: (1) best-developed horst and graben structures, (2) high V_P/V_S ratio beneath the outer rise region, and (3) anomalously high heat flow. It is noted that tensional stress in the incoming plate is assumed to have extended to depths of 40 km during and after the 2011 Tohoku Earthquake, whereas it is shallower than 40 km for previous earthquakes. The projections of epicenters of microearthquakes after the 2011 Tohoku Earthquake are aligned with topographic lineation of horst and graben structures in the outer rise region. These lines of evidence suggest that hydration has progressed extensively to deeper sections of the oceanic plate since the 2011 Tohoku Earthquake. To address (a) hydration processes and their extents along the bend-fault and (b) physical properties of the bend-fault, in-situ physical properties and lithofacies are best obtained by ocean drilling in the NW Pacific region.</p>

<p> A newly compiled bathymetric map including parts of the Kuril, Japan, and northern Izu–Ogasawara trenches in the northwestern Pacific Ocean demonstrates that most bending-related topographic structures are limited to less than 80 km from the trench axis. This observation contrasts with one that bending-related structures of eastern Pacific trenches are limited to less than 50 km from the trench axis. The discrepancy may be due to differences in the ages of subducting oceanic plates. Bending-related topographic structures of the western Kuril and southern Japan trenches are not parallel to the trench axis, but instead are parallel to magnetic anomaly lineations. Those of the northern Izu–Ogasawara Trench are parallel to fracture zones. These observations indicate the rule that the inherited seafloor spreading fabric is reactivated instead of forming new faults when the degree of obliquity between inherited seafloor spreading fabric and trench axis reaches about 30°. This rule is applicable to most trenches around the Pacific Ocean, except for some parts of curved trenches and trenches near seamounts or other volcanic edifices constructed by off-ridge volcanism. Most bending-related topographic structures near off-ridge volcanic edifices are parallel to the trench axis. This observation suggests that inherited seafloor spreading fabric around the volcanic edifies was disrupted by volcanism.</p>

Shipboard multibeam survey is powerful tool to locate submarine volcanoes especially having small volume. Small submarine volcanoes may represent the initial stages of hotspot activity, but they may also form via litho spheric flexing, regional convection of the mantle, and the presence of fracture zones. Here we describe several volcanoes, flood lavas, and volcanic clusters in French Polynesia using data from archives of multibeam data. The clusters of small volcanoes are similar to petit-spots, and they are not considered to represent the initial stages of a hotspot as they are composed of both young and old edifices, and because the sites are located far from any known hotspot. These newly discovered submarine volcanoes are located in areas with low-velocity seismic shear waves at depths of 60 and 100 km. These lava fields will therefore facilitate geochemical mapping of the mantle in areas unrelated to hotspots, because these lavas may have developed from melts in the shallow mantle beneath French Polynesia. (C) 2016 Elsevier B.V. All rights reserved.

The few geological and geophysical studies of the Lyra Basin at the western margin of the Ontong Java Plateau (OJP Pacifi c Ocean) revealed that it is underlain by thicker than normal oceanic crust. The unusually thick oceanic crust is attributed to the emplacement of massive lava fl ows from the OJP. Dredging was conducted to sample the inferred OJP crust on the Lyra Basin but instead recovered younger extrusives that may have covered the older plateau lavas in the area. The Lyra Basin extrusives are alkalic basalts with (&lt sup&gt 87&lt /sup&gt Sr. &lt sup&gt 86&lt /sup&gt Sr. &lt inf&gt t&lt /inf&gt = 0.704513-0.705105, (&lt sup&gt 143&lt /sup&gt Nd/&lt sup&gt 144&lt /sup&gt Nd)&lt inf&gt t&lt /inf&gt = 0.512709-0.512749, ε&lt inf&gt Nd&lt /inf&gt (t) = +3.0 to +3.8, (&lt sup&gt 206&lt /sup&gt Pb/&lt sup&gt 204&lt /sup&gt Pb)&lt inf&gt t&lt /inf&gt = 18.488-18.722, (&lt sup&gt 207&lt /sup&gt Pb/&lt sup&gt 204&lt /sup&gt Pb)&lt inf&gt t&lt /inf&gt = 15.558-15.577, and (&lt sup&gt 208&lt /sup&gt Pb/&lt sup&gt 204&lt /sup&gt Pb)&lt inf&gt t&lt /inf&gt = 38.467-38.680 that are distinct from those of the OJP tholeiites. They have age-corrected (&lt sup&gt 187O&lt /sup&gt s/&lt sup&gt 188&lt /sup&gt Os)&lt inf&gt t&lt /inf&gt = 0.1263-0.1838 that overlap with the range of values determined for the Kroenke-type and Kwaimbaita-type OJP basalts, but their (&lt sup&gt 176&lt /sup&gt Hf/&lt sup&gt 177&lt /sup&gt Hf)&lt inf&gt t&lt /inf&gt = 0.28295-0.28299 and ε&lt inf&gt Hf&lt /inf&gt (t) = +7.9 to +9.3 values are lower. These isotopic compositions do not match those of any Polynesian ocean island volcanics. Instead, the Lyra Basin basalts have geochemical affi nity and isotopic compositions that overlap with those of some alkalic suite and alnoïtes in the island of Malaita, Solomon Islands. Although not directly related to the main plateau volcanism at 120 Ma, the geochemical data and modeling suggest that the origin of the Lyra Basin alkalic rocks may be genetically linked to the mantle preserved in the OJP thick lithospheric root, with magmatic contribution from the Rarotongan hotspot.

The Lyra Basin is believed to be a contiguous part of the Ontong Java Plateau (OJP), based on geophysical studies. Volcaniclastic rocks dredged at two sites in the Lyra Basin document another post-plateau episode of magmatism on the OJP they are olivine-titanaugite-phyric alkali basalts with as much as ~30% modal phenocrysts. Lyra Basin basalts have compositions that vary from picritic (MgO ~22 wt%) to more evolved (MgO ~5 wt%) and have low SiO&lt inf&gt 2&lt /inf&gt (41-46 wt%), high TiO&lt inf&gt 2&lt /inf&gt (2-4 wt%), and high Na&lt inf&gt 2&lt /inf&gt O + K&lt inf&gt 2&lt /inf&gt O (1-5 wt%) contents that are distinctly different from tholeiites that compose the main OJP. The &lt sup&gt 40&lt /sup&gt Ar-&lt sup&gt 39&lt /sup&gt Ar weighted mean age of Lyra Basin basalts is 65.3 ± 1.1 Ma, determined using a single-grain laser fusion method of the groundmass from the least altered alkali basalt and of biotite separates from differentiated samples. This age is interesting because it is much younger than the main stage of OJP formation (122 Ma) and no ca. 65 Ma alkaline basalts have been found previously near or on the OJP. Incompatible trace element modeling suggests that the volcanic rocks of the Lyra Basin may have been formed by a low degree of partial melting (~3%), predominantly at the garnet-lherzolite stability fi eld from the same OJP mantle source preserved in its thick lithospheric root. However, major and trace elements and isotopic compositions can be better explained by magma mixing of Rarotongan alkali magma and magma derived from OJP-source mantle melting (12% partial melting at garnet stability fi eld) in the ratio of 1:2. Although the trace element compositions of Lyra basalts can be reproduced by OJP-source mantle melting with or without contribution from the Rarotongan hotspot, the lower potassium content of the calculated Rarotongan hotspot-infl uenced melt is more compatible with that of an average composition of Lyra basalt. These results are consistent with previous reconstruction of the OJP path from 120 Ma to its present position, indicating that it may have passed over the Rarotongan hotspot at 65 Ma. In either case, the petrogenesis of Lyra Basin basalts highlights the role of the thick lithospheric root of the OJP in the late-stage development of the plateau. Additional evidence for episodic late-stage magmatic activity on the OJP helps to elucidate the magmatic evolution of the plateau and may provide insights into the origins of other large igneous provinces.

The Manihiki Plateau in the western equatorial Pacifi c Ocean is a Cretaceous Large Igneous Province. Several studies have proposed that the Manihiki Plateau was formed by the same mantle plume that formed the Ontong Java and Hikurangi plateaus ca. 125 Ma. Recent multibeam bathymetric surveys of the Manihiki Plateau reveal the morphology of the Danger Islands Troughs (DIT), Suvarov Trough, which are systems of deep troughs within the plateau. The troughs divide the Manihiki Plateau into three distinct provinces, the North Plateau, the Western Plateaus, and the High Plateau. The DIT between the High Plateau and Western Plateaus comprises four en echelon troughs. With one exception, all segments of the DIT are bordered by steep escarpments, to 1500 m high. The basins of the DIT are smooth. Elongated northeast-southwest-striking scarps are common in the southernmost DIT and at the junction between the DIT and Suvarov Trough. The features revealed by the new bathymetric data indicate that a sinistral strike-slip tectonic environment formed the DIT during the break-up into the Manihiki and Hikurangi plateaus, whereas the Suvarov Trough developed after the formation of the DIT.

It has been suggested that the Shatsky Rise oceanic plateau formation began simultaneously with a reorganization of spreading at a triple junction bordering the northern Pacifi c plate, and this coincidence has led to speculation about the connections between the two events. We present new marine geophysical data that constrain the seafl oor spreading history of the Pacifi c-Izanagi-Farallon triple junction just before the birth of the Shatsky Rise. Bathymetric data reveal en echelon, abandoned spreading centers trending northwest-southeast located adjacent to the southwest fl ank of the Shatsky Rise. Magnetic anomalies and bathymetry are interpreted to indicate that segments of the Pacifi c-Farallon Ridge near the triple junction propagated northwest from chron M23 (153 Ma) to chron M22 (151 Ma) during a spreading ridge reorganization at the edge of a likely microplate. Our detailed examination of bathymetric and magnetic anomaly lineations also shows that the strike of the Pacifi c-Izanagi Ridge changed gradually on the west side of the triple junction around chron M22. Our observations indicate that the plate boundary reorganization began several million years before the formation of the Shatsky Rise, implying that the eruption of the plateau did not cause the reorganization.

Elongated topographic structures associated with bending of the subducting oceanic plate along the western Kuril, Japan and northern Izu- Ogasawara trenches, were investigated using available multibeam bathymetric data. Magnetic anomaly lineations were also reidentified using available geomagnetic data to reveal controlling factors for strikes of bending-related topographic structures. The new bathymetric map demonstrates that most of bending-related topographic structures exist in the oceanward trench slopes deeper than 5,600 m. The map reveals that bending-related topographic structures are developed parallel to the trench axis or inherited seafloor spreading fabrics. Detailed identification of magnetic anomalies reveals curved lineations and discontinuity of lineations associated with propagation ridges. A trough with elongated escarpments associated with the propagating ridge in mid- Cretaceous Quiet Period was discovered near the trench-trench-trench triple junction. Comparison between the detailed bathymetric and magnetic anomaly lineation maps elucidates that abyssal hill fabrics were reactivated where the angle between abyssal hill fabrics and trench axis is less than about 30°. The topographic expression of bending-related structures are classified into two types according to whether new faults develop parallel to the trench axis or inherited seafloor spreading fabrics reactivate.

Detailed swath mapping of topography associated with other geophysical measurements has indicated that the outer wall of the western Kuril Trench is dissected by a number of normal faults which are parallel to the linear magnetic anomalies in the northwestern margin of the Pacific Ocean but slightly oblique to trend of the axis of the Kuril Trench by roughly 10°. It is inferred that about 130 million years old tectonic fabrics in the oceanic crust are rejuvenated when the lithosphere is bent downward under extensional stress at the trench outer wall to be subducted beneath the landward wedge. This pattern of faults shows a sharp contrast to alignment of faulted scarps found in the outer wall of the Japan and Izu-Ogasawara Trenches both trending by 60° oblique to the magnetic lineations. Majorities of outer wall scarps there are parallel to the trench axis, whereas some scarps trend subparallel to the axis forming zigzag pattern of escarpment. As direction of the extensional stress caused by downward bending of the lithosphere in these trench walls is nearly perpendicular to the old tectonic stresses, normal faultings seem to occur under the influence of the ongoing stress distribution alone irrespective of their old history.

The discovery of the magnetic anomaly lineations that can give ages of ocean floor is a very important role for establishment of the plate tectonics theory. The magnetic anomaly lineations also give us information of a history of movements of the oceanic plates. However, the origin of magnetic anomaly lineations still been obscured, that is, we can not clearly answer for the following questions: How thick is the source layer of magnetic anomaly lineations? How strong is the intensity of the magnetic source layer? In this paper we examined the relevant information concerning the magnetization of the oceanic crust from studies of observed marine magnetic anomalies and from rock -magnetism of oceanic basalts to get a goal of these questions in this paper.<BR>The skewness parameter that is deduced by precise magnetic anomaly lineations is important to identify marine magnetic anomalies. The magnetic polarity transition width is also important to do, though the parameter associated with this transition width has not almost utilized in the previous works. The anomalous skewness and the skewness discrepancy are often observed over the oceans. These observations might be explained not by a single-layer model but a two-layer model for magnetic source layer. The polarity transition width is defined the width which 95.4 % of the change from normal to reversed polarity occurs within. This width increases monotonically with spreading rates of ridges and/or with ages of ocean floors. This increasing is considered to be a manifestation of a more complicated crustal source consisting of two discrete layers. The analysis of the skewness parameter and transition width strongly supports that the sourc elayer of marine magnetic anomalies has a two-layer structure. The upper layer, consisting of surface lava flows of layer 2 A and possibly the sheeted dike complex, hasdi stinct and approximately vertical magnetic in the vicinity of opposite magn etized region boundaries. The lower layer, consisting of intrusive and gabbroic layers, has the boundaries gradually sloping downward away from the spreading center.<BR>Many detailed survey are carried out to reveal the structure of magnetic source layer by the multi narrow and the deep-towed magnetometer near active ridges. Inversion of magnetic layer using results of detailed surveys concluded that the magnetic source layer near the active ridges is less than 1 km in thickness. The polarity transition width of the relatively young layer is narrower than that of older oceanic floor, and the magnetic intensity of the relatively young layers higher (more than 10A/m) than that of older one. These conclusion indicate that the magnetic source layer near the active ridg es consists of a single layer structure. It is thought that the magnetic source layer grows with ages asoceanic crust by results of analysis of skewness and polarity transition width and inversion of magnetic source layer near active ridges.<BR>Several previous paleomagnetic studies indicate that intensity of natural remanent magnetization (NRM) of basaltic rocks composing the ocean crust rapidly decreases with ages in the past 10 to 20 Ma, and gradually increases older one. This change in NRM intensity is roughly proportional to the changes in intensity of saturation magnetization of the rocks and possibly due to sea-water alteration (low-temperature oxidation) of the primary ferromagnetic minerals contained in the rocks. NRM of the oceanic rocks is initially thermoremanent magnetization acquired at the time of formation of the oceanic crust. In accordance with progressive oxidation, fraction of TRM to bulk intensity decreases, while that of the secondary magnetization increases.