Hidemi Tanaka

Fault rocks and past to recent fluid characteristics from the borehole survey of the Nojima fault ruptured in the 1995 Kobe earthquake, southwest Japan

The mineralogy, fluid inclusions, and distribution of fault rocks of the Nojima fault were examined in the core recovered from a borehole drilled by the Geological Survey of Japan (GSJ) 12 months after the 1995 Kobe (Hyogo-ken Nanbu) earthquake (MJMA = 7.2) in southwest Japan. The borehole was drilled across a slipped portion of the fault to a depth of 746.7 m. Nearly continuous coring between 152.2 and 746.7 m recovered granodiorite protolith, porphyry dikes, and fault-related rocks. The fault zone was intersected at 426.2 m and is characterized by a greater intensity of brittle deformation and/or hydrothermal alteration than typical host granodiorite. The fault core consists of three types of fault gouge and occurs at the depth range of 623.1 to 625.3 m. The fault-normal thicknesses of the fault core and the fault zone are 0.3 m and >46.5 m, respectively. Three types of hydrothermal alteration are recorded by mineral assemblages and fluid inclusions. The first type is characterized by chloritization of mafic minerals at >200°C and occurred prior to the fault activity during the intrusion and cooling of the granodiorite. The second type occurred during faulting and is recorded by zeolite mineralization at <200°C. The third type is recorded by carbonate mineralization related to recent fluid flow. Although most of the second type of alteration occurred prior to the third type, repeated mineralization is recorded by mutually crosscutting relationships between zeolite and carbonate veins and between zeolite vein and carbonate-precipitated fault gouge. This may record repeated changes in fluid chemistry within the fault zone in connection with the seismic cycle. Although the Nojima fault slipped in the 1995 earthquake, ancient fault-related textures and mineral alteration are well preserved in the fault rocks.

Structural and chemical characterization of shear zones in the freshly activated Nojima fault, Awaji Island, southwest Japan

Behavior and role of each shear zone in a shallow fault zone of granitic origin during seismic cycles are revealed by comprehensive examinations of petrographic and chemical characterization on a fault zone in the Geological Survey of Japan (GSJ) drill core penetrating the Nojima fault which was activated during the 1995 Hyogoken-Nanbu earthquake (M = 7.2). The GSJ core consists of granodiorite and porphyritic intrusive rocks including a Nojima fault zone which involves seven thin shear zones: main shear zone (MSZ, 625 m depth), upper cataclasite zone (UCZ), upper shear zone (USZ), lower shear zones (LSZ-1 and LSZ-2), and lower cataclasite zones (LCZ-1 and LCZ-2). These shear zones are generally surrounded by weakly pulverized and altered (fault-related) rocks (WPAR) which generally show volume gain. The fault zone architecture is clarified as follows: (1) Total thickness of the Nojima fault zone is ∼70 m. (2) All shear zones except the older cataclasite zones (UCZ-1, LCZ-1, and LCZ-2) were evolved from WPAR, indicating that pulverization and alteration of recent activity were more diffused at the initial stage of faulting and gradually localized to each shear zone. (3) The MSZ (2 m thick) can be regarded as a high-velocity frictional zone with accompanying volume loss (compaction) and possibly with heat generation during coseismic periods. (4) The LSZ-1 (3.6 m thick), located just beneath the MSZ and typically showing explosion brecciation texture, is also regarded as a coseismic shear zone. This zone could function as a trap zone for fluid or gas during postseismic/interseismic periods. (5) The LSZ-2 (2.7 m thick), located around 710 m depth, contains foliated fault gouge enriched with clay minerals and characterized by a large degree of volume gain, possibly a result of slow velocity motion or creep during the interseismic and/or postseismic periods.

Timing of the cataclastic deformation along the Akaishi Tectonic Line, central Japan

This paper describes K-Ar ages of cataclasites and fault gouges from the Akaishi Tectonic Line (ATL), central Japan. Petrological and mineralogical features of these rocks are also examined. Based on the results, we test the hypothesis that the K-Ar ages of these rocks represent the age of hydrothermal alteration associated with the fault movement. Intensity of deformation and alteration increase passing from host rock into cataclasite and finally into fault gouge. This increase corresponds to an increase of the value of crystallinity index (Kübler index) of the micaceous minerals contained in these rocks. Furthermore, the degree of rejuvention of K-Ar ages of the micaceous minerals increases in the same sense. A correlation of K-Ar ages (plotted on the y axis) versus Kübler index (plotted on the x axis) yields a concave curve asymptotically parallel with the x axis at approximately 15 Ma. This curve is interpreted to represent the mode of decrease of the relative amounts of inherited argon in the K-Ar system of the micaceous minerals, corresponding to an increase in the intensity of deformation and alteration. Inclination of the curve becomes zero if all the inherited argon is lost from the K-Ar system of the micaceous minerals. Thus, it is concluded that the hydrothermal alteration occurred at approximately 15 Ma. The strike-slip basin along the ATL formed the middle Miocene. The K-Ar dating of ATL gouges indicates that the ATL was active simultaneously with formation of the strike-slip basin along its trace during the middle Miocene.

Dislocation creep of fine-grained recrystallized plagioclase under low-temperature conditions

Abstract Lattice preferred orientation (LPO) and dislocation substructure of fine-grained plagioclase were investigated in ultramylonite from a shear zone along the Hatagawa fault zone in northeast Japan. In the shear zone, the orthoclase mole fraction of the recrystallized alkali-feldspar is not less than 0.94. This composition suggests that deformation occurred below 500°C. Selected area diffraction by transmission electron microscopy (TEM) for an electron-transparent foil and an X-ray texture goniometer (XTG) were used to determine the LPO of plagioclase. Burgers vectors of dislocations were determined by invisibility criteria. The (100), (010), and (001) pole figures reveal strong preferred orientations. The XTG results are consistent with the LPO determined by TEM. The TEM observations confirmed the presence of at least six Burgers vectors. These Burgers vectors or the net slip directions of multiple systems are almost parallel to the lineation. Thus the present study indicates ductility by dislocation creep for fine-grained plagioclase in ultramylonite deformed at relatively low temperature. The LPO pattern and the Burgers vectors determined in this study are different from those of deformation at much higher temperature. Thus the present study also suggests a switching of the slip systems for plagioclase depending on the temperature.