Space—time correlations between inland earthquakes in central Japan and great offshore earthquakes along the Nankai trough: Implication for destructive earthquake prediction

Based on the tectonic framework of central Japan, including the surrounding submarine areas, the space-time relationship between destructive inland earthquakes of magnitudesM 6.4 or greater and great offshore earthquakes along the Nankai trough was examined. From east to west, four tectonic lines are defined as lines linking active faults: the Itoigawa-Shizuoka tectonic line (ISTL), the Tsurugawan-Isewan tectonic line (TITL), the Hanaore-Kongo fault line (HKFL), and the Arima-Takatsuki tectonic line (ATTL). The TITL divides central Japan into the Chubu and Kinki districts, and probably extends southward to the Nankai trough. The Chubu district is subdivided into four blocks by boundary lines linking NW-SE trending active faults having left-lateral strike slip. In the Kinki district, N-S trending, active reverse, steep-dip faults are dominant in the triangular region north of the Median Tectonic line, between the TITL and HKFL, forming a basin-and-range province.

Starting from 1586 A.D., a seismic space-time sequence of high seismic activity in the Chubu district in which earthquake occurrence migrates from the eastern to western tectonic lines of central Japan was identified. The sequence also revealed that inland earthquakes preceded great offshore earthquakes which occurred along the Nankai trough. It was also found that a destructive earthquake tends to occur on the HKFL within 30 years after the occurrence on the TITL, and that the western Nankai trough generated great earthquakes ofM≥7.0 at intervals ranging from 8 to 49 years after the HKFL earthquakes. If the eastern Nankai trough is coupled with the western Nankai trough, a forthcoming greater earthquake measuringM 8.5 may be expected. Since such great earthquakes are always accompanied by large tsunamis, much attention should be focussed on possible tsunami disasters along the Pacific coast of central Japan.

Based on its tectonic structure, a tectonic model of central Japan is proposed. The seismic space-time sequence, which attempts to explain the cause of the sequential earthquake generation, is also discussed.     

Application of a modified pattern informatics method to forecasting the locations of future large earthquakes in the central Japan

We propose a modification of the Pattern Informatics (PI) method that has been developed for forecasting the locations of future large earthquakes. This forecast is based on analyzing the space–time patterns of past earthquakes to find possible locations where future large earthquakes are expected to occur. A characteristic of our modification is that the effect of errors in the locations of past earthquakes on the output forecast is reduced. We apply the modified and original methods to seismicity in the central part of Japan and compared the forecast performances. We also invoke the Relative Intensity (RI) of seismic activity and randomized catalogs to constitute null hypotheses. We do statistical tests using the Molchan and Relative Operating Characteristic (ROC) diagrams and the log-likelihoods and show that the forecast for using the modified PI method is generally better than the competing original-PI forecast and the forecasts from the null hypotheses. Using the bootstrap technique with Monte-Carlo simulations, we further confirm that earthquake sequences simulated based on the modified-PI forecast can be statistically the same as the real earthquake sequence so that the forecast is acceptable. The main and innovative science in this paper is the modification of the PI method and the demonstration of its applicability, showing a considerable promise as an intermediate-term earthquake forecasting tool.     

The EEPAS forecasting model and the probability of moderate-to-large earthquakes in central Japan

The EEPAS (“Every Earthquake a Precursor According to Scale”) model is a space–time point-process model based on the precursory scale increase (Ψ) phenomenon and associated predictive scaling relations. It has previously been fitted to the New Zealand earthquake catalogue, and applied successfully in quasi-prospective tests on the CNSS catalogue for California for forecasting earthquakes with magnitudes above 5.75 and on the JMA catalogue of Japan for magnitudes above 6.75. Here we test whether the Ψ scaling relations extend to lower magnitudes, by applying EEPAS to depth-restricted subsets of the NIED catalogue of the Kanto area, central Japan, for magnitudes above 4.75. As in previous studies, the EEPAS model is found to be more informative than a quasi-static baseline model based on proximity to past earthquakes, and much more informative than the stationary uniform Poisson model. The information that it provides is illustrated by maps of the earthquake occurrence rate density, covering magnitudes from 5.0 to 8.0, for the central Japan region as at the beginning of year 2004, using the NIED and JMA catalogues to mid-2003.     

On the problem of destructive Iranian earthquakes and their causative faults

This article is devoted to evaluating destructive earthquakes (magnitude >6) of Iran and determining properties of their source parameters. First of all, a database of documented earthquakes has been prepared via reliable references and causative faults of each event have been determined. Then, geometric parameters of each fault have been presented completely. Critical parameters such as Maximum Credible Rupture, MCR, and Maximum Credible Earthquake, MCE, have been compiled based on the geometrical parameters of the earthquake faults. The calculated parameters have been compared to the maximum earthquake and the surface rupture which have been recorded for the earthquake faults. Also, the distance between the epicenter of documented earthquake events and their causative faults has been calculated (the distance was less than 20 km for 90% of the data). Then, the distance between destructive earthquakes (with the magnitude more than 6) and the nearest active fault has been calculated. If the estimated distance is less than 20 km and the mechanism of the active fault and the event are reported the same, the active fault will be introduced as a probable causative fault of that earthquake. In the process, all of the available geological, tectonic, seismotectonic maps, aerial geophysical data as well as remote sensing images have been evaluated. Based on the quality and importance of earthquake data, the events have been classified into three categories: (1) the earthquakes which have their causative faults documented, (2) the events with magnitude higher than 7, and (3) the events with the magnitude between 6 and 7. For each category, related maps and tables have been compiled and presented. Some important faults and events have been also described throughout the paper. As mentioned in this paper, these faults are likely to be in high seismic regions with potential for large-magnitude events as they are long, deep and bound sectors of the margins characterized by different deformation and coupling rates on the plate interface.     

Carbonaceous material in pelitic schists of the Sanbagawa metamorphic belt in central Shikoku,Japan

Carbonaceous material in pelitic schists of the Sanbagawa metamorphic belt in central Shikoku, Japan, was separated from the host rocks and its X-ray diffraction and chemical composition were studied. Its crystal structure and chemistry change continously with increasing metamorphic grade and approach those of well-ordered graphite near the biotite isograd. As graphitization is a rate process, the temperature of complete graphitization differs from one metamorphic terrain to another as a function of the duration of metamorphism. In an individual metamorphic terraan, however, the degree of graphitization is a useful indicator of relative metamorphic temperature in lower-grade rocks.     

Estimation of the spatial distribution of ground motion parameters for two recent earthquakes in Japan

A recent development in strong motion instrumentation in Japan provides an opportunity to collect valuable data sets, especially after moderate and large magnitude events. Gathering and modeling these data is a necessity for better understanding of regional ground motion characteristics. Estimations of the spatial distribution of earthquake ground motion plays an important role in early-stage damage assessments for both rescue operations by disaster management agencies as well as damage studies of urban structures. Subsurface geology layers and local soil conditions lead to soil amplification that contributes to the estimated ground motion parameters of the surface. We present a case study of the applicability of the nationally proposed GIS-based soil amplification ratios [J. Soil Dyn. Earthqu. Eng. 19 (2000) 41–53] to the October 6, 2000 Tottori-ken Seibu (western Tottori Prefecture) and the March 24, 2001 Geiyo earthquakes in Japan. First, ground motion values were converted to those at a hypothetical ground base-rock level (outcrop) using an amplification ratio for each 1×1 km area, based on geomorphological and subsurface geology information. Then a Kriging method, assuming an attenuation relationship at the base-rock as a trend component, is applied. Finally, the spatial distribution of ground motion at ground surface is obtained by applying GIS-based amplification factors for the entire region. The correlation between the observed and estimated ground motion values is reasonable for both earthquakes. Thus, the proposed method is applicable in near real-time early-damage assessments and seismic hazard studies in Japan.     

The higher grade metamorphic zonation of the Sambagawa metamorphic belt in central Shikoku, Japan

The higher grade metamorphic zonation of the Sambagawa (= Sanbagawa) belt is established for the first time for the whole area of central Shikoku. As discontinuous reactions to define the isograd are absent, the metamorphic grade is primarily determined by the Mg-Fe partitioning between garnet and chlorite along representative traverses. However, for regional mapping, mineralogical features of the pelitic schists, such as using mineral assemblages of more than divariant equilibrium, the modal garnet to chlorite ratio, and the optical properties of chlorite, are employed as auxiliary criteria.
The presence of the highest grade mineral zone in the middle of the structural level is confirmed, but its spatial distribution is far more complex than hitherto accepted. Thermal axes are now confirmed at three different structural levels. A model is presented in which the stacking of thrust sheets of different grade took place while metamorphic reactions were in progress. Thermal readjustment brought a continuous metamorphic temperature gradient across and within the thrust sheets. Tectonic blocks of metagabbro and ultramafic rock were emplaced synchronously with thinning and subsequently also re-equilibrated. Local anomalies of metamorphic grade, represented by mixing of schists of different metamorphic grade, exist, but they are due to a later stage event.     

Large-scale landslides in Toyama Prefecture,central Japan,and their probable relationship with earthquakes

Large-scale landslides along the Kubusu and Besso rivers in Toyama Prefecture are developed in the Miocene Iwaine Formation, which is composed of andesitic lava, tuff, and tuff breccia. In the middle member of this formation, the tuff is easily altered to montmorillonite-bearing rock, and subsequently plays an important role in the development of landslides events, which tend to be large-scale events, as the massive lava of the upper member forms a cap rock over the tuff. The Kiritani and Koinami basins, which are flat intermontane basins located along the Kubusu and Besso rivers, respectively, are interpreted as landslide-dammed lakes, later filled with sediment. Accelerator mass spectrometry ¹⁴C ages show that the landslides forming each dam occurred simultaneously, at approximately 2500 BP. These ages were measured from wood fragments embedded in the landslide material of Kiritani, and from an in situ stump drowned during the impoundment of Koinami. If the trigger of these landslides was an earthquake, it is most likely to have been the penultimate event along the Atotsugawa fault zone.     

Phase relations of pumpellyite-actinolite facies metabasites in the Sanbagawa metamorphic belt in central Shikoku, Japan

Sanbagawa metabasites metamorphosed at conditions near the upper limit of the pumpellytic-actinolite facies were examined in terms of phase equilibria in the five component system Al₂O₃---Fe₂O₃---CaO---MgO---FeO. The Fe³⁺ content of epidote measured as X^Ep_Fe (=Fe/(Fe + Al) of epidote) in the assemblage epidote-chlorite-actinolite-pumpellyite decreases gradually towards the higher-grade, pumpellyite-free areas. The progressive change in X_Fe^Ep can be detected within one metabasite bed 200 meters thick near the upper limit of the pumpellyite-actinolite facies. The Mg---Fe²⁺ substitution, as expressed by variation of Fe/(Fe + Mg) in chlorite (0.40–0.55) has little effect on the Fe³⁺ + Al) ratios of epidote and pumpellyite in the above-mentioned assemblage. The lowet X_Fe^Ep in the pumpellite-bearing assemblage is 0.15 and hence the upper limit of the pumpellyite-actinolite facies is defined by the appearance of an epidote-chlorite-actinolite assemblage with X^Ep_Fc = C.15     

Subduction related antigorite CPO patterns from forearc mantle in the Sanbagawa belt,southwest Japan

Antigorite (Atg) is stable throughout large parts of the wedge mantle of most subduction zones. Atg shows strong acoustic anisotropy and crystallographic preferred orientation (CPO) patterns of this mineral may contribute significantly to seismic anisotropy in convergent margins. Atg CPO patterns from the Higashi-Akaishi (HA) forearc mantle body of southwest Japan adds to the data set suggesting the most common Atg CPO pattern has a c-axis perpendicular to the foliation and a b-axis parallel to the stretching lineation. Statistical analysis using the eigenvector method of Atg CPO from two mutually perpendicular directions in the same sample (YZ-section and XZ-section) shows no significant differences implying sample preparation has no significant affect on the resulting Atg CPO. Reuss (uniform stress) averages of anisotropy for the Higashi-Akaishi samples are approximately treble the values for Voigt (uniform strain) averages. When comparing calculated anisotropy of hydrated mantle peridotite samples—such as the Higashi-Akaishi unit—with observed S-wave delay times in convergent margins, the appropriate averaging method needs to be considered.     

Occurrence of lawsonite in pelitic schists from the Sanbagawa metamorphic belt, central Shikoku, Japan

The occurrence of lawsonite is described from pelitic schists of the lower-grade part of the pumpellyite-bearing subzone of the chlorite zone in the Asemi River area of central Shikoku. The lawsonite-bearing parageneses are consistent with the generally accepted view that the Sanbagawa facies series represents higher pressures than the lawsonite-bearing facies series in New Zealand.     

Spatial distribution of earthquakes in the Kii peninsula,Japan, south of the Median Tectonic Line

Precise data obtained by a high-sensitivity micro-earthquake observatory network are used to determine simultaneously the crustal structure and the spatial distribution of small earthquakes in the Kii peninsula region, Japan. The spatial distribution of hypocenters thus determined clearly shows two distinct groups of earthquakes: (1) a group of shallow (H ? 10 km) earthquakes on the western coast of the Kii peninsula near Wakayama; (2) a group of mantle earthquakes, having a depth ranging from 30 to 70 km and trending NE-SW, in the central part of the Kii peninsula. Along the trend of the second group, a marked structural anomaly is found which suggests the presence of a high-velocity zone at depths below 20 km. A projection of the hypocenters of the earthquakes belonging to the second group onto a vertical plane strikingNW-SE shows a wedge-like distribution to a depth of 70 km. The spatial relation between this wedge-like distribution and the 1944 Tonankai earthquake (M = 8.0) suggests a common tectonic process which is now taking place in the Kii peninsula region. The activity of the earthquakes of the first group terminates abruptly to the north at the Median Tectonic Line. This activity is represented by numerous but relatively small events (M < 5) without any conspicuous major earthquakes in history. It is suggested that the strength of the crust in this region of shallow activity is too weak to sustain stresses large enough to be released in a major event; rather, the stresses which probably originate from the tectonic activity represented by the earthquakes of the second group are released by numerous minor fracturings of the low-strength crust. A possibility of using the weak crust for detecting a remote stress accumulation is suggested.     

Verification test of the mean event size model for moderate earthquakes in the Kanto region, central Japan

In the Kanto region, there have been reports of decreasing b-values prior to earthquakes (M ≥ 5.5). The change of b-values is defined as the difference between the long- and short-term average earthquake magnitudes. A hazard function for moderately large earthquakes has also been proposed. This model was based on earthquakes that occurred between 1982 and 1999, and its effectiveness is measured retrospectively. In order to evaluate the effectiveness of the model through the use of independent data, the verification test is based on earthquakes observed after January 2000. Through the end of 2004, the only event where there was a decrease in mean event size was observed on June 3, 2000. This decrease resulted in a log-likelihood for the proposed model 1.3 units larger than that of the Poisson model, supporting the validity of the proposed model. Without accumulating further examples, we attempted to improve the verification test by expanding the study volume and by lowering the target cutoff magnitude in order to overcome the small sample size. When two other targets from the expanded volume were added, the difference in the log-likelihood (ΔlnL) increased to 3.6. In this case, the information rate per event was about 1.2, larger than that of the model period. Lowering the cutoff magnitude increased ΔlnL to 2.5. These extended tests led to higher confidence in the model with the larger ΔlnL value than did the primary test. From the viewpoint of the per-event information rate, each case involving targets with a magnitude 5.5 and larger resulted in better performance than in the model period.     

Intra-plate seismicity and inter-plate earthquakes: Historical activity in Southwest Japan

Over the last four hundred years the spatial variation of intra-plate seismicity in Southwest Japan correlates well with the occurrence of great inter-plate earthquakes. For fifty years before an inter-plate earthquake the intra-plate seismic activity is highest along a belt inland. For ten years afterwards it falls off in this belt, but rises on both sides along the Philippine Sea and Japan Sea coasts. Then it becomes low and remains low throughout the whole region until fifty to thirty years before the next inter-plate event, as shown by Utsu in 1974. An intermittent underthrusting drag exerted by the Philippine Sea plate seems to control the intra-plate seismicity, which partly takes up the relative plate motion as internal deformation. When a great inter-plate earthquake occurs, tectonic stress is released and seismic activity falls off in the central belt. The breaking of the plate boundary temporarily weakens the coupling between the two plates along the shallower part of the interface, which gently dips toward the Japan Sea coast. The decoupling causes stress concentration in the deeper part and results in increased seismic activity along the Japan Sea coast. The activity along the Philippine Sea coast may be interpreted as aftershock activity.     

Seismic patterns in Northern China and the characteristics of the epicentral distribution of the medium strong earthquakes before the large events

In this paper, the seismic pattern in Northern China from 30 ° to 42 ° N latitude and 104 ° to 125 ° E longitude, and the characteristics of the epicentral distribution before large events are presented. The results suggest that:

• 1.(1) the earthquakes in the region are mainly located in the orthogonal curvilinear network formed by the seismic belts;
• 2.(2) the larger earthquakes (M ⩾6) occurred mainly in the nodal regions of this grid:
• 3.(3) the strike of the fracture planes of the earthquakes coincided with the directions of the seismic belts;
• 4.(4) the pattern of medium strong earthquakes (M ⩾ 4.7) prior to thirteen large earthquakes (M⩾ 7) are analysed to be of three types:
  • 4.1.(a) mainly arranged along the two intersecting belts,
  • 4.2.(b) randomly distributed,
  • 4.3.(3) forming seismic gaps.

A theoretical basis and rules for drawing the orthogonal grid is presented, and an idea for the prediction of the sites of future earthquakes in Northern China is suggested.     

Mineral distribution within polymineralic veins in the Sanbagawa belt,Japan: implications for mass transfer during vein formation

Pelitic schists of the Sanbagawa metamorphic belt contain several types of polymineralic veins that formed during the late stages of exhumation. The vein mineral assemblages are quartz + albite + K-feldspar + chlorite ± calcite (Type I, II) and quartz + albite + calcite (Type III). Type I and II veins contain quartz and albite with stretched-crystal and elongate-blocky textures, respectively. The mineral species within Type I veins vary with compositional bands within the host rocks. Type III veins are characterized by euhedral to subhedral quartz grains with concentric zoning and a homogeneous distribution along the vein length. The vein textures vary depending on the crack aperture during multiple crack-seal events: <0.08 mm for Type I, and 0.5–10 mm for Type III. Type II veins show intermediate features between Type I and III veins in terms of mineral distribution (weak dependence on the host rock composition) and apparent crack aperture (less than 1–15 mm). These observations suggest a transition in the dominant transport mechanism of vein components with increasing crack aperture, from diffusion from host rocks to fluid advection along cracks.     

Southward extrusion of eclogite-bearing mafic–ultramafic bodies in the Sanbagawa belt, central Shikoku, Japan

Several mafic rock masses, which have experienced eclogite facies metamorphism, are distributed in flat-lying non-eclogitic schists in an intermediate structural level (thermal core) of the Sanbagawa belt. The largest, Iratsu mass, and an associated peridotite, the Higashi-Akaishi mass, extend E–W for about 8 km, and N–S for about 3 km, and are surrounded by pelitic, basic and quartz schists. The Iratsu mass consists of metabasites of gabbroic and basaltic origin, with intercalations of ultramafic rocks, felsic gneiss, quartz schist and metacarbonate. The Iratsu mass can be divided into two layers along a WNW-trending metacarbonate layer. The Higashi-Akaishi mass consists of peridotite with intercalations of garnet clinopyroxenite. It is situated beneath the western half of the Iratsu mass, and their mutual boundary dips gently or steeply to the N or NE. These masses underwent eclogite, and subsequent epidote-amphibolite facies metamorphism as has been reported elsewhere. The Iratsu–Higashi-Akaishi masses and the surrounding rocks underwent ductile deformation under epidote-amphibolite facies (or lower P–T) metamorphic conditions. Their foliation generally trends WNW and dips moderately to the NNE, and the mineral lineation mostly plunges to the N and NE. In non-eclogitic schists surrounding the Iratsu–Higashi-Akaishi masses, the foliation generally trends WNW and dips gently or steeply to the N or S and the mineral lineation mostly plunges to the NW, N and NE. Kinematic analysis of deformation structures in outcrops and oriented samples has been performed to determine shear senses. Consistent top-to-the-north, normal fault displacements are observed in peridotite layers of the Higashi-Akaishi mass and eclogite-bearing epidote amphibolite layers of the Iratsu mass. Top-to-the-northeast or top-to-the-northwest displacements also occur in non-eclogitic pelitic–quartz schists on the northern side of the Iratsu mass. In the structural bottom of the Iratsu–Higashi-Akaishi masses and to the south, reverse fault (top-to-the-south) movements are recognized in serpentinized peridotite and non-eclogitic schists. These observations provide the following constraints on the kinematics of the rock masses: (1) northward normal displacement of Iratsu relative to Higashi-Akaishi, (2) northward normal displacement of non-eclogitic schists on the north of the Iratsu mass and (3) southward thrusting of the Iratsu–Higashi-Akaishi masses upon non-eclogitic schists in the south. The exhumation process of the Iratsu–Higashi-Akaishi masses can be explained by their southward extrusion.     

The Asama igneous complex, central Japan: An ultramafic-mafic layered intrusion in the Mikabu greenstone belt, Sambagawa metamorphic terrain

The Asama igneous complex comprises layered mafic and ultramafic plutonic rocks exposed over about 500×6000 m in the Mikabu greenstone belt, Sambagawa metamorphic terrain of Mie Prefecture; its margins terminate by faults, and there is no trace of chilled rocks. The exposed layered sequence is about 460 m thick, and includes dunite, plagioclase wehrlite, olivine gabbro and two-pyroxene gabbro. The crystallization sequence of essential cumulus minerals is olivine, followed by plagioclase and clinopyroxene together, and finally the appearance of orthopyroxene. Olivine systematically varies in composition from Fo₈₉ to Fo₇₈ with stratigraphic height in the lower to middle portion of the layered sequence. The composition of clinopyroxene changes from Ca₄₉Mg₄₆Fe₅ to Ca₄₀Mg₄₇Fe₁₃ upward in the layered sequence; cumulus orthopyroxene, which occurs at the top of the exposed layered sequence, has a composition of Ca₂Mg₇₄Fe₂₄. Cumulus chromite occurs as disseminated grains in peridotitic rocks, and tends to increase its Fe³⁺/(Cr+Al+Fe³⁺) ratio with stratigraphic height. The most aluminous chromite [Cr/(Cr+Al) = 0.48] occurs in dunite that crystallized shortly before plagioclase began to separate as an essential phase. The Cr/(Cr+Al) ratio of the most aluminous chromite, coupled with the crystallization order of essential minerals, suggests that the Asama parental magma was moderately enriched in plagioclase and clinopyroxene components in the normative mineral diagram plagioclase-clinopyroxene-orthopyroxene. It was similar to a Hawaiian tholeiite and different from the Bushveld and Great “Dyke” parental magmas that were more enriched in orthopyroxene component; it also differed from mid-oceanic ridge basalts that are more depleted in the orthopyroxene component. The Asama clinopyroxene and chromite show characteristically high TiO₂ contents and are also similar to those in Hawaiian tholeiites. The Asama igneous complex probably resulted from the crystallization of a magma of a Hawaiian (oceanic-island) tholeiite composition and formed in an oceanic island regime.