Kiyoo Mogi

Earthquakes and fractures

Abstract The high degree of similarity between earthquakes and fractures strongly supports the fracture hypothesis of earthquakes. From this standpoint the stochastic process of earthquake occurrence, the relation between magnitude of earthquakes and their number, the mechanism of various patterns of earthquake sequences, the relation between the patterns of earthquake sequences and geotectonic structures, and the mechanism of foreshocks and aftershocks are discussed based on laboratory experiments on the brittle fracture of materials. The fracture theory seems to give a systematic explanation of these individual features of earthquake phenomena. This theory seems also to have application to the earthquake prediction problem.

Effect of the triaxial stress system on the failure of dolomite and limestone

Abstract Failure under conditions of triaxial stress, in which two of the principal stresses are equal, have been studied intensely, but few investigations of the effect of the more general combined stress system have been made because of the experimental difficulties of achieving conditions of true homogeneous triaxial stress. A new triaxial apparatus has made possible the study of failure of brittle rocks under a general triaxial stress system in which all three principal stresses are different. By this method, the effects of the stress system to the fracture stress, yield stress and the ductility in several rocks. Dunham Dolomite and Solenhofen Limestone were experimentally studied. The fracture strength increases markedly not only with the increase of the least compression σ 3 but also with the increase of the intermediate compression σ 2 , except for very high σ 2 . The influences of σ 2 and σ 3 on the ductility and the stress drop, however, are just the opposite. A new empirical failure criterion, which corresponds to the modified von Mises theory, correlates the present results satisfactorily.

Relationship between shallow and deep seismicity in the western Pacific region

Abstract The two types of relationship between the shallow and deep seismic activities in island-arc regions are described: 1. (1) A great shallow earthquake in a region is preceded and sometimes followed by the marked increase of deep seismic activity in the same down-dip seismic zone perpendicular to the trend of an arc structure. The increase of the deep seismic activity before the great shallow earthquake cannot be understood as an accidental event nor as a mere trigger of the great shallow earthquake, but as an essential forerunning phenomenon of it. This regularity is found in the Kurile-Kamchatka and northern Japan island-arc regions, particularly near the ends of these two arcs and their junctions, where great shallow earthquakes frequently occur. 2. (2) The seismic activity tends to progress in time from a shallow to a deep region within a descending lithosphere. This regularity is found in the Mariana and Tonga arcs, where the seismic activity is nearly continuous from the surface to the deep region and no very great shallow earthquake occurred in the considered period. The rate of migration in these earthquakes along the descending lithosphere is estimated at about 50 km per year, nearly similar to that in shallow earthquakes. These regularities in the space-time distribution of shallow and deep earthquakes in island-arc regions can be explained from the viewpoint that the oceanic lithosphere is pushed downward under the continental lithosphere by external forces, probably due to the mantle convection. The present discussion suggests that the descending lithosphere does not act as a rigid plate but as a visco-elastic body under gradually increasing loading.

Temperature changes in an artesian spring at Usami in the Izu Peninsula (Japan) and their relation to earthquakes

Abstract Temperature measurements over quite a long period (since 1976) at an artesian spring at Usami Hot Springs, which is located on the northeastern coast of the Izu Peninsula (Japan) where there has been considerable seismic activity recently, have revealed that the temperature of this hot spring varies in close relation to the occurrence of earthquakes and volcanic eruptions in this region. The highly accurate measurements that have been carried out continuously since 1982 show the nature of temperature changes in this artesian spring at normal times. Based on this, an examination of the long-term temperature change curve produced the following results: (1) normally, the temperature of this artesian spring falls slowly in a more or less linear manner: (2) at the time of a strong earthquake the temperature jumps suddenly in a step pattern; (3) both the long- and short-term precursory changes in temperature have been observed before large shallow earthquakes that have occurred relatively near to the well. This paper also considers the mechanism behind these temperature changes.

Fracture and Flow of Rocks

Mogi, K., 1972. Fracture and flow of rocks. In: A.R. Ritsema (Editor), The Upper Mantle. Tectonophysics, 13(1–4): 541–568. Recent laboratory studies of the fracture and flow of rocks are reviewed. New developments in laboratory apparatus have made it possible to discuss quantitatively the stress–strain relation under a wide range of geologically realistic environmental conditions. Particular effort has been devoted to strength and ductility of rocks under combined stresses. The effect of confining pressure on the fracture strength of rocks follows closely the Coulomb criterion, except for the extremely high pressure dependence at low confining pressure in some rocks which can be attributed to crack closure. However, from results of a new triaxial compression test in which all three principal stresses are different, the intermediate principal stress σ2 has an important influence, so that current fracture criteria, such as Coulomb, Mohr or Griffith, are not entirely applicable. The observed behavior under combined stress is adequately described by a new fracture criterion, which has the form of a generalized Von Mises criterion. Not only fracture stress, but also the yield stress is affected by both the least compression σ3, and by the intermediate compression σ2. The other formula, obtained by generalization of the Von Mises criterion, can also be applied to this case. The ductility of rocks strongly increases with the increasing σ3, but decreases with increasing σ2. The transition from brittle to ductile state can be explained by the Orowan frictional hypothesis, when deformation in the post-yield region takes place by cataclastic flow, but the frictional hypothesis cannot be applied when large homogeneous deformation takes place in the post-yield region before fracture. Recent studies of other environmental parameters are reviewed. Effects leading to weakening and to embrittlement of rocks under conditions expected in deep regions are discussed, because of possible application to the fracture hypothesis of earthquakes.

On the pressure dependence of strength of rocks and the coulomb fracture criterion

Abstract The observed microfracturing before faulting in rocks provides a basis for a physical model for the internal friction and the cohesive stress of the Coulomb criterion. Based on recent experimental results in the literature, the applicability of the Coulomb criterion is reexamined. The pressure dependence of strength of brittle rocks can be predicted approximately by the Coulomb criterion using measured values of sliding friction, except for a low-pressure region for certain rocks and near the brittle-ductile transition pressure. Deviation at low pressure seems to be explained either by the effect of cracked state or by different fracture mechanisms at low confining pressures, and deviation near the transition pressure is attributed to the increase of local yielding or fracturing before faulting.

Temporal Variation of Crustal Deformation During the Days Preceding a Thrust-type Great Earthquake — The 1944 Tonankai Earthquake of Magnitude 8.1, Japan

The temporal variation in precursory ground tilt prior to the 1944 Tonankai (Japan) earthquake, which is a great thrust-type earthquake along the Nankai Trough, is discussed using the analysis of data from repeated surveys along short-distance leveling routes.Sato (1970) pointed out that an anomalous tilt occurred one day before the earthquake at Kakegawa near the northern end of the focal region of the earthquake. From the analysis of additional leveling data, Satos result is re-examined and the temporal change in the ground tilt is deduced for the period of about ten days beginning six days before the earthquake. A remarkable precursory tilt started two or three days before the earthquake. The direction of the precursory tilt was up towards the south (uplift on the southern Nankai Trough side), but the coseismic tilt was up towards the southeast, perpendicular to the strike of the main thrust fault of the Tonankai earthquake. The postseismic tilt was probably opposite of the coseismic tilt. The preseismic tilt is attributed to precursory slip on part of the main fault. If similar precursory deformation occurs before a future earthquake expected to occur in the adjacent Tokai region, the deformation may help predict the time of the Tokai earthquake.

Active periods in the world's chief seismic belts

Abstract The existence of active or inactive periods in the worlds chief seismic belts, which is one of the fundamental problems in space—time seismicity, is discussed based on space—time diagrams of the worlds large shallow earthquakes in the past seventy-year period. In those seismic belts which correspond to zones of convergence between two large rigid plates, even if the seismic belt is very long, a number of large shallow earthquakes occur in succession throughout the belt within a limited period (the active period). This seismic activity seems to change periodically. Except for the northern circum-Pacific belt, the worlds chief seismic belts of this type have nearly common active periods. In other seismic zones, which correspond to triple junctions of rigid plates or complex plate-boundary zones, the space—time distribution of large shallow earthquakes does not show such a regular pattern. The activity of these belts is rather continuous. The present results; suggest that large shallow earthquakes are strongly coupled on a global scale. These global-scale interactions of large shallow earthquakes support the hypothesis of rigidplate tectonics.

A new method for estimation of the crustal stress from cored rock samples: Laboratory study in the case of uniaxial compression

Abstract We present a new method by which the maximum previous stress in rocks can be estimated. The Kaiser effect in acoustic emission is often used for an estimation of the stress to which rocks have been subjected. However, there are cases in which the Kaiser effect is not clear. In such cases, estimation of the previous stress is difficult by the conventional method which is based on the acoustic emission activity during one stress cycle. Using a new method of cyclic loading, we can estimate the stress in cases where the conventional method does not work. In the new method, the difference of the acoustic emission activity between the first and second reloading begins to increase markedly at the previous maximum stress. Even if the Kaiser effect is not clear, we can estimate the previous stress by observation of this difference. Our new method may be applicable for estimating the crustal stress. In this paper, we study the influence of water and temperature under uniaxial compression, because rocks in the earths crust are saturated and subjected to high temperatures and they are disturbed by water and temperature in the sampling process. The present experiment shows that the Kaiser effect is often masked by changes in water content and temperature. However, we can estimate the previous stress by the new method even after the rock has been disturbed by water and temperature. The accuracy in the stress estimation by this method is satisfactory for the ratio of the previously applied stress to the fracture strength of the rock less than 40%.

Global variation of seismic activity

Abstract The global variation in the space—time distribution of great shallow earthquakes is examined using the surface-wave magnitude M and the magnitude Mω obtained from the seismic moment. In the past several decades, the seismic activity was complementary in the regions of low latitude ( 40° ). During the roughly twentyyear period around 1960, the seismic activity was very high in the high-latitude regions and appreciably low in the low-latitude regions. In particular, it is pointed out that the greatest earthquakes occurred roughly simultaneously in the north and south regions of high latitude around 1960. However, the present time seems to be an active period in the low-latitude regions. These regions account for the larger part of the earths surface, and sustain a degree of destruction and loss of life that is far out of proportion with the relative energy release in the low and high latitude regions.