佐賀県唐津市高島のアルカリ玄武岩中の捕獲岩．かんらん岩の一種で，上部マントル最上部を代表すると考えられる．ダナイト捕獲岩は西南日本に多産し，かんらん岩捕獲岩としては最も一般的なものである（例えば, Arai et al., 2000）．ダナイトは，ほとんどマントルの代表的鉱物であるかんらん石より構成される単純な岩石である．ダナイトは，マグマが関与した深部過程（マグマからの結晶集積，輝石を含むかんらん岩とマグマの反応など）で形成されると考えられるが，その詳しい成因を特定するのはなかなか難しい．マグマ過程で生成された後，岩石は流動等で変形するが，かんらん石にしばしば見られるキンクバンドと呼ばれる折れ曲がり（中心よりやや左上）はその産物である． かんらん岩を厚さ0.03 mm 程度の薄片にして偏光顕微鏡の直交ポーラー（試料を2枚の偏光板で挟む）で観察するとこの写真のような極彩色のステンドグラスの世界が広がる．偏光顕微鏡の直交ポーラーでの色を干渉色と呼ぶが，0.03 mm の厚さでこのような鮮やかな干渉色が観察されるのは，かんらん石の特徴である． （写真・解説：荒井章司）
Recent active-source seismic studies on the old Pacific Plate (125-121 Myr old) off the Kuril Trench provide new insights into formation/alternation processes of the crust and uppermost mantle structures. These studies show high P-wave velocities of 8.5-8.6 km/s and strong azimuthal anisotropies of 8.5-9.8% in the uppermost mantle, immediately below the clearly imaged Moho and the lower crustal ridge-ward dipping reflectors (LCDR), which are observed with constant dip angles (20-25°) and intervals (2.5 km). From the observations, we conclude that the LCDR and the strong anisotropic mantle were formed by an active-mantle flow near a paleo-spreading center, which drags the base of the crust away from the spreading center. This could be direct evidence of an active mantle flow at a spreading center. Another important observation is a systematic structural change toward the trench from the outer rise region. Recent seismic studies found that P-wave velocities decrease and VP/VS ratios increase toward the trench from a distance of 200 km. The observations suggest that the water content within the incoming ocean crust increases toward the trench, accompanied by the development of bending-related fractures at the top of the oceanic crust. P-wave reduction (2-5%) in the uppermost mantle is observed at the near trench. This might suggest serpentinization of the mantle of ～20%, but it is necessary to have precise S-wave information to further interpret the mantle velocity reduction near the trench region.
Abyssal peridotites recovered directly from mid-ocean ridges (mid-ocean ridge peridotites hereafter) are generally interpreted to be formed as a residue after partial melting and melt extraction in the adiabatically upwelling mantle beneath mid-ocean ridges. Osmium and some other isotopic characteristics of some mid-ocean ridge peridotites, which were formed by partial melting events, are much older than present-day mid-ocean ridge systems (ancient melting residual peridotites), however, suggesting ancient origins. Ancient melting residual peridotites might have been incorporated into the oceanic asthenosphere beneath mid-ocean ridges. If this is the case, the nature of the oceanic plate and its formation processes need to be reconsidered because the asthenospheric mantle, which has abundant ancient melting residual peridotites, is not sufficient to create basaltic oceanic crusts, followed by formation of the serpentinized peridotite crust by hydrothermal circulation along faults developed around the mid-ocean ridges by continuous spreading. Comprehensive geochemical studies on depleted mid-ocean ridge peridotites are required to detect and verify ancient melting residual peridotites from mid-ocean ridge peridotites. It is crucial to examine the effects of minor Os-rich phases on 187Os/188Os isotopic compositions to understand the meaning of their model ages. Precise chemical analyses of fluid mobile elements including light elements, such as H, Li, Be, and B, on clinopyroxene, as well as orthopyroxene, provide clues about differences in geochemical signatures and thermal histories between present-day mid-ocean ridge residual peridotites and ancient melting residual peridotites.
Chemical differentiation from pyrolitic lherzolite to harzburgite due to partial melting and melt extraction process causes the chemical heterogeneity in the Earth's upper mantle. Phase relation and chemical variations of residues obtained with melting experiments of dry and hydrous pyrolitic lherzolite are compared to understand the effects of water on chemical differentiation. In dry conditions, orthopyroxene/olivine ratio decreases with increasing degree of melting. In hydrous conditions, the stability field of residual orthopyroxene, however, expands relative to olivine above solidus, and the harzburgitic residue contains a large amount of Mg-rich (Mg# > 0.93) orthopyroxene above 4 GPa in pressure. The chemistry of residues obtained from hydrous experiments agrees well with chemical variations of continental cratonic garnet harzburgite. This observation indicates that cratonic harzburgite with a high orthopyroxene content possibly reflects formation by melt depletion under various water contents from almost anhydrous to 2 wt% in the upper mantle at depths of about 100 to 200 km. The orthopyroxene-rich harzburgite, similar to continental cratonic harzburgite, may be formed at deep mantle wedges in the present Earth because water is being dragged into the deep mantle wedge by subducting slabs. Orthopyroxene-rich harzburgite may be detected in seismological observations because a jump in elastic wave velocities occur at 9-10 GPa (270-300 km in depth) in the harzburgite due to the orthorhombic to high-pressure monoclinic phase transition in (Mg, Fe) SiO3 pyroxene. A small jump in seismic velocities at about 250-300 km in depth, the X discontinuity, has occasionally been observed in seismic profiles at some subduction zones in southern Africa and the southern Pacific. The phase transition of (Mg, Fe) SiO3 pyroxene in orthopyroxene-rich harzburgite correspond to the X discontinuity.
Serpentine minerals (lizardite, chrysotile, and antigorite) are a major group of hydrous phyllosilicates resulting from the hydrothermal alteration of mantle peridotite. Their distinct rheological properties mean that serpentine minerals have a strong influence on the mechanical and seismogenic behavior of faults and plate boundaries in both continental and oceanic settings. In this paper we review the results of laboratory experiments performed to understand the frictional and mechanical properties, and deformation mechanisms of serpentinite. Frictional sliding experiments at low slip rates show that antigorite exhibits velocity-strengthening behavior (a−b > 0) over a wide range of temperature (25-400°C) , while values of (a−b) for chrysotile become negative as temperature increases (25-281°C) . This indicates that the stability of slip along serpentinite-bearing faults depends on the serpentine species and fault depth. Frictional sliding of antigorite at seismic slip rates leads to weakening by flash heating. Axial compression experiments at confining pressures of up to 4 GPa show that antigorite is stronger than lizardite by at least a factor of two. The flow law for dislocation creep of antigorite based on stress values at 〜15% strain also predicts differential stresses that are substantially lower than those for the dislocation creep of olivine at natural strain rates (10−10 to 10−14 s−1) . This suggests that the viscosity of serpentinite promotes slab–mantle decoupling. However, the antigorite flow law should be used with caution because antigorite starts to deform by semi-brittle flow after 〜20% strain. Large-strain simple-shear deformation of antigorite aggregates at high pressure (1 GPa) results in a strong alignment of antigorite c-axes normal to the shear plane. This observation explains the trench-parallel anisotropy beneath the Ryukyu subduction zone. Although dehydration embrittlement is considered a primary cause of intermediate-depth earthquakes, recent high-pressure experiments on antigorite show stable sliding behavior or detect no acoustic emissions during dehydration reactions. We emphasize that the presence of talc derived from the metasomatic alteration of serpentine further weakens and stabilizes the slab–mantle interface and promotes long-lived ( > 1 Ma) detachment faulting.
Olivine crystal grains have various crystallographic orientations within a peridotite, resulting in a crystallographic fabric as well as a texture. Six types of fabrics have been identified in mantle peridotites: A, B, C, D, E and AG types. These fabric types have unique seismic properties such as P-wave and S-wave velocity anisotropy. A new method is proposed to classify olivine fabrics based on P-wave velocity structure on a Vp-Flinn diagram. Three P-wave velocities (V1, V2, V3) are used, two of which are the maximum (V1) and minimum (V3) velocities, and the third (V2) is the velocity perpendicular to the orientations of these two velocities. Relationships between V1/V2 and V2/V3 classify fabrics into three types: A type (equivalent to B, C and E types), D type and AG type. Moreover, taking into account structural framework such as foliation and lineation, the Vp-Flinn diagram can be expanded to identify all types of fabrics. This method is successfully applied to fabrics previously studied in Oman ophiolite.
This paper reviews research on ultramafic pseudotachylytes. Shand coined the name pseudotachylyte in 1916 from his studies on enigmatic vein networks developed in Precambrian granitic rocks, South Africa. Ultramafic pseudotachylyte was first reported from Balmuccia peridotite, N. Italy in 1995. Significant research progress has been made on researching this controversial rock type, coupled with advances in rock mechanics, rock rheology, and seismology. The difficulty and complexity of the studying ultramafic pseudotachylytes is analyzed referring mainly to case studies on the Balmuccia ultramafic pseudotachylytes. It is emphasized that studies of ultramafic pseudotachylytes are closely linked to understanding the mechanics of deep earthquakes occurring in the upper mantle, and also to understanding physicochemical behavior under the extreme conditions that may occur at very sites where seismic waves are generated.
Mantle xenoliths provide snapshots of lithospheric mantle processes at the time of an eruption. Their fluid inclusions have the potential to enhance the resolution of images. Identifying major volatile compositions of fluid inclusions provides accurate figures for mantle fluid and various aspects of the lithospheric mantle. Fluid inclusions in mantle xenoliths are generally composed of CO2 coexisting occasionally with smaller amounts of CO. We can calculate the oxygen fugacity of a host xenolith from the relative abundance of CO. Isotopic characteristics of elements in fluid inclusions are useful when considering the origins of fluid inclusions. The fluid density of a fluid inclusion is also a useful depth probe for mantle xenoliths. The internal pressure of fluid inclusions was equilibrated with ambient pressure before the mantle xenoliths were trapped by magma. Thus, we can estimate the depth provenance of mantle xenoliths from fluid density. Combining fluid-inclusion geobarometry with some other characteristics of both fluid inclusion and host mantle xenoliths enables us to assess the three-dimensional structure of the lithosphere from various viewpoints. Fluid inclusions in mantle xenoliths serve as an index for investigating both the elastic and plastic strength of mantle minerals. Changes in ambient pressure and temperature of mantle xenoliths during entrainment by host magma engender elastic changes in the volumes of constituent minerals, thereby inducing small changes in the fluid density of fluid inclusions. Thus, we can investigate elastic properties of mantle minerals from the gradation of fluid density among minerals in a mantle xenolith. In addition, the dislocation density around fluid inclusions in minerals reflects the degree of plastic deformation of the host crystal in response to the overpressure of fluid during entrainment. Thus, observations of fluid density and dislocations are an effective new probe for elucidating rheological properties of the lithospheric mantle. Consequently, fluid inclusions in the mantle xenoliths provide a unique way of exploring the lithospheric mantle.
The distribution and behavior of halogens in the mantle are still poorly understood. Chlorine, bromine, and iodine are highly incompatible and strongly partitioned into fluids, whereas fluorine is partitioned into melts rather than fluids. As a result, during differentiation of the Earth, the former halogens became more abundant in surface reservoirs and scarcer in the mantle. Because their geochemical properties differ, each reservoir has a distinct composition. Therefore, the halogen compositions of mantle-derived materials are expected to enable material exchanges to be traced between the surface of the Earth and its interior, especially water transportation into the mantle accompanying a subducting slab. Exhumed peridotites in metamorphic belts and mantle-derived xenoliths in volcanic rocks can provide intrinsic halogen information on the mantle, whereas halogen compositions in volcanic rocks are changed from those of their mantle source during melting and degassing processes, in which halogen behavior is not well constrained. Very low halogen concentration in mantle-derived materials hampers analyses using conventional methods. Recent progress of analytical techniques involving the noble gas method overcomes such difficulties, making it possible to intensively investigate halogens in mantle-derived materials during the last decade. Presented here is a review of studies of halogens in volcanic and mantle-derived rocks and minerals. Recent findings of distinctive Br/Cl and I/Cl ratios in exhumed peridotites in metamorphic belts, mantle xenoliths from arc volcanoes and kimberlites suggest subduction of halogen acquired from sedimentary pore water or crustal brine into the mantle.
Subduction-zone magmatism is triggered by the addition of H2O-rich slab-derived flux: aqueous fluids, hydrous partial melts or supercritical fluids from the subducting slab through reactions. Whether the slab-derived flux is an aqueous fluid, a partial melt, or a supercritical fluid remains an open question. In general, with increasing pressure, aqueous fluids dissolve more silicate components and silicate melts dissolve more H2O. Under low-pressure conditions, those aqueous fluids and hydrous silicate melts remain isolated phases due to the miscibility gap. As pressure increases, the miscibility gap disappears and the two liquid phases becomes one phase. This vanishing point is regarded as critical end point or second critical end point. X-ray radiography experiments locate the pressure of the second critical end point at 2.5 GPa (83 km depth) and 700°C for sediment-H2O, and at 2.8 GPa (92 km depth) and 750°C for high-Mg andesite (HMA)-H2O. These depths correspond to the depth range of a subducted oceanic plate beneath volcanic arcs. Sediment-derived supercritical fluids, which are fed to the mantle wedge from the subducting slab, may react with the mantle peridotite to form HMA supercritical fluids due to peritectic reaction between silica-rich fluids and olivine-rich mantle peridotite. Such HMA supercritical fluids may separate into aqueous fluids and HMA melts at 92 km depth during ascent. HMA magmas can be erupted as they are, if the HMA melts segregate without reacting to the overriding peridotite. Partitioning behaviors between aqueous fluids and melts are determined with and without (Na, K) Cl using synchrotron X-ray fluorescence. The data indicate that highly saline fluids effectively transfer large-ion lithophile elements. If the slab-derived supercritical fluids contain Cl and subsequently separate into aqueous fluids and melts in the mantle wedge, then such aqueous fluids inherit much more Cl and also more or less amounts of large ion lithophile elements than the coexisting melts. In contrast, Cl-free aqueous fluids can not effectively transfer Pb and alkali earth elements to the magma source. Enrichment of some large-ion lithophile elements in arc basalts relative to mid-oceanic ridge basalts has been attributed to mantle source fertilization by such aqueous fluids from a dehydrating oceanic plate. Such aqueous fluids are likely to contain Cl, although the amount remains to be quantified. If such silica-rich magmas survive as andesitic melts under a limited reaction with mantle minerals, they may erupt as HMA magmas having slab-derived signatures.
The mantle is one of the largest reservoirs of deeply buried carbon and nitrogen in the Earth. Currently, mantle xenolith, which potentially originated from indigenous mantle, is the sole sampling access for mantle material from which to understand the enigmatic deep carbon and nitrogen cycles. Based on Deines (2002), there are three major origins of deep carbon in mantle xenolith: i) pristine mantle carbon, ii) sedimentary organic carbon, and iii) oceanic limestone, including inorganic carbon. We can estimate how those three end-members combine using the CO2/3He indicator. In contrast, some laboratory-based experiments demonstrate abiotic formation of hydrocarbon (up to C32n-alkane: McCollom and Seewald, 2006) and other relevant molecules during the Fischer-Tropsch Type reaction. We discuss the origins of carbon and nitrogen from the viewpoint of bulk geochemistry and molecular-specific organic geochemistry. Important concepts of the geochromatography of crustal fluid coupled with the mantle refertilization process need to be shared to obtain a further understanding of deep carbon and nitrogen dynamics.