Magnitude scale and quantification of earthquakes

Despite various shortcomings, the earthquake magnitude scale is one of the most fundamental earthquake source parameters to be used for catalogs. Although use of a uniform scale is desirable, it is not always possible because of changes in instrumentation, the data reduction method and the magnitude formula, the station distribution, etc. As a result, various magnitude scales have been developed and are currently in use. Recent developments in seismometry and earthquake source theories provide more quantitative source parameters than the magnitude. In order to maintain continuity and uniformity of the data, it is important to relate these magnitude scales and the new parameters. In view of this importance, relations between different magnitude scales are examined with an emphasis on the difference in the period of the waves used for the magnitude determination. Use of several magnitude scales determined at different periods provides a convenient method for characterizing earthquakes. The moment magnitude can be used to quantify both shallow and deep earthquakes on the basis of wave energy radiated, and provides a uniform scheme.     

Complexity of rupture in large strike-slip earthquakes in Turkey

Complexity of rupture propagation has an important bearing on the state of stress along the earthquake fault plane and on the prediction of strong ground motion in the near-field. By studying far-field body waveforms recorded by WWSSN long-period seismograms it has been possible to investigate the degree of complexity of several Turkish earthquakes. The results, which are obtained by matching synthetic P waveforms to observed data indicate that the July 22, 1967 Mudurnu Valley earthquake (M_s = 7.1) is a complex event which can be explained by the superposition of elementary sources with variable amplitudes and source time sequence history. In this regard, it is very similar to the February 4, 1976 Guatemala earthquake (M_s = 7.5). A comparison of these two events indicates that their source-time series ranges from 5 to ca. 20 s and, regardless of the total moment of the earthquake, the moment of the individual events is bounded at around 5 × 10²⁶ dyn cm. The November 24, 1976 E. Turkey earthquake (M_s = 7.3), on the other hand, has a complexity which cannot be explained by such a simple model; in this respect, it may be more similar to the Tangshan, China, earthquake and as such, may involve significant thrust, normal or other complications to its faulting mechanism than the strike-slip mechanism of the P-wave first-motion data. The source time history for the 1967 Mudurnu Valley event is used to illustrate its significance in modeling strong ground motion in the near field. The complex source-time series of the 1967 event predicts greater amplitudes (2.5 larger) in strong ground motion than a uniform model scaled to the same size for a station 20 km from the fault. Such complexity is clearly important in understanding what strong ground motion to expect in the near-field of these and other continental strike-slip faults such as the San Andreas.     

Groundwater level and chemistry changes resulting from tunnel construction near Matsumoto City, Japan

 Before tunnel construction began, the groundwater chemical compositions and levels around the tunnel were studied to determine if water compositions could predict whether surface water will be influenced by tunnel construction. When the chemical composition of the well and springwater was similar to that of the tunnel seepage water, and the altitude of the well and spring was above the tunnel level, the groundwater level in the well and spring was influenced by draining tunnel seepage water. Therefore, comparing the chemical compositions of surface water and groundwater may be used for predictive purposes. However, the results of this study showed there was no noticeable chemical composition change in springwater prior to changes in groundwater level at a particular site. The changes in the hydrology of the plateau caused by tunnel construction were also studied, using measurements of the changes in groundwater chemistry as well as changes in groundwater levels. Prior to tunnel construction, river discharge was greater. Following tunnel construction, some river discharge decreased because springwater was drained as tunnel seepage water and the spring in the catchment dried up. Tritium concentration indicated that 3 years after tunnel construction, surface water did not reach tunnel levels in spite of groundwater level lowering and remaining unconfined groundwater being drained. Received: 17 January 1996 · Accepted: 10 July 1996     

从悬浮质和沉降质的观点描述中国云南湖泊的化学动态变化

在对云南抚仙湖、星云湖和洱海3个高原湖怕进行地球化学和湖沼学调查，确定湖水中微粒的溶解浓度和化学成分，对沉降质和沉积物进行分析的基础上，对这些湖泊的化学动态变化进行了讨论。     

Large-scale ripple marks on the shelf margin of the Northern Okinawa Trough

Trains of large-scale ripple marks (megaripples and sand waves) were found on the Amakusa and East China Sea shelves bordering the northern Okinawa Trough. Side-scan sonar surveys were carried out in 1974 and 1976 to investigate sea-floor features lying along a proposed submarine cable line. Megaripples were found on the outer margin of the Amakusa shelf between depths of 140 and 200 m. The megaripples were especially well developed at a depth of 167 m. They were typically straight-transverse crested with asymmetrical profiles, and measured 7 to 15 m in wavelength and 0.4 to 1.4 m in waveheight. Formation of the megaripples on the Amakusa shelf is probably controlled by relatively complex oceanographic conditions. A secondary circulation associated with the Gotô-nada clock-wise Current may be responsible for formation of the ripple marks. Local vorticities generated in the coastal boundary layer as a result of curvature of the Gotô-nada Current are known to cause the complex flow pattern at the Gotô and Amakusa shelf margins. The main semidiurnal (M₂) tidal current may also interact with these fluid processes.On the East China Sea shelf, megaripples and sand waves were found between depths of 140 and 220 m. Sand waves (200 m in wavelength) were observed in seismic reflection profiles. Large-scale lunate megaripples were observed at a depth of 154 m by the side-scan sonar. They had wavelengths of 10 to 30 m and waveheights of 1 to as high as 3 m. It appears from the types and nature of distribution of the megaripples that they are responding to the present-day flow regime, and it is partly ascertained from our observations over an interval of two years that the megaripples appear to be short-term response elements compared wit hteh sand waves. We conclude that the megaripples on the East China Sea shelf are current-formed during peak typhoon flow in August to November. From their distribution, the long term path of the main flow of the Tsushima Current is inferred at the edge of the East China Sea shelf. An area of low sediment mud content (less than 20 per cent) coincides with this path giving further support to our interpretation.     

Geologic and metamorphic evolution of the basement complexes in the Kontum Massif, central Vietnam

This paper presents a regional scale observation of metamorphic geology and mineral assemblage variations of Kontum Massif, central Vietnam, supplemented by pressure–temperature estimates and reconnaissance geochronological results. The mineral assemblage variations and thermobarometric results classify the massif into a low- to medium-temperature and relatively high-pressure northern part characterised by kyanite-bearing rocks (570–700 °C at 0.79–0.86 GPa) and a more complex southern part. The southern part can be subdivided into western and eastern regions. The western region shows very high-temperature (> 900 °C) and -pressure conditions characterised by the presence of garnet and orthopyroxene in both mafic and pelitic granulites (900–980 °C at 1.0–1.5 GPa). The eastern region contains widespread medium- to high-temperature and low-pressure rocks, with metamorphic grade increasing from north to south; epidote- or muscovite-bearing gneisses in the north (< 700–740 °C at < 0.50 GPa) to garnet-free mafic and orthopyroxene-free pelitic granulites in the south (790–920 °C at 0.63–0.84 GPa). The Permo-Triassic Sm–Nd ages (247–240 Ma) from high-temperature and -pressure granulites and recent geochronological studies suggest that the south-eastern part of Kontum Massif is composed of a Siluro-Ordovician continental fragment probably showing a low-pressure/temperature continental geothermal gradient derived from the Gondwana era with subsequent Permo-Triassic collision-related high-pressure reactivation zones.     

Shear wave velocity measurement during a standard penetration test

An experiment was carried out to develop a technique to measure shear wave velocity simultaneously with the standard penetration test popular in soil engineering. In the standard penetration test an impact at the bottom of a borehole is produced by weight dropping and may be expected to generate seismic waves. A three-component geophone was set on the ground surface near the borehole and the waves generated were recorded with a magnetic recorder at successive depths of the penetration test. The predominance of the SV wave obtained with this simple method was assured by measurement of the particle orbit. Signal amplitudes decrease with depth and become less than the noise level at a certain depth. Therefore records from deeper sources must be processed to disclose the shear waves. Since waveforms of SV events generated by blows of the penetration test at a given depth are very similar, the signal to noise ratio would be expected to be improved by a stack of wave trains. A paste-up of the radial component after stacking was compared with that before stacking and a refinement was clearly recognized. A vertical distribution of shear wave velocity was obtained by reading the onset time at each depth. Shear wave velocities thus obtained were compared with N values from the standard penetration test and specific resistivities from electrical logging in the same borehole. The data were mutually consistent. This experiment showed that a convenient, precise shear wave velocity measurement can be conducted during the routine work of a standard penetration test.     

Shear velocity and density of an attenuating earth

The dispersion that must accompany absorption is taken into account in many recent body-wave investigations but has been largely ignored in surface-wave and free-oscillation studies. In order to compare body-wave and free-oscillation data a correction must be made to travel times or periods to account for absorption-related physical dispersion. The correction depends on the frequency and Q of the data and can be as high as 1% which is much larger than the uncertainty of the raw data. Corrected toroidal mode data is inverted to obtain shear velocity and density versus depth. The average shear velocity in the upper 600 km is ～2% greater than obtained from the uncorrected data. The resulting shear-wave travel times oscillate about the Jeffreys-Bullen values with an average baseline of only +0.5 second. Thus, the discrepancy between body-wave and free-oscillation studies is eliminated.     

Variation of seismic source parameters and stress drops within a descending slab and its implications in plate mechanics

A least-squares searching technique has been developed to estimate the source dimensions of intermediate and deep focus earthquakes using azimuthal variations of body wave pulse-widths. With this method and also amplitude data, modes of rupture propagation, seismic moments, and stress drops of 17 intermediate and deep focus earthquakes in the Tonga-Kermadec region have been determined in order to investigate variations in source properties and the state of stress within the descending slab there. Three different modes of rupture; unilateral, bilateral, and circular faults, are compared and tested against observations. Results indicate that the unilateral fault is the best model for most of the earthquakes studied. Stress drops of the 17 events vary within a very large range, from 20 bar to about 4.6 kbar, and change significantly with depth. The magnitude of stress drops for depths between 220 and 430 km is very much higher than at shallower depths. This change in stress drop magnitude at about 220 km-depth seems to reflect a change in material properties both in the mantle and within the slab. Two regions of high stress drop are observed at depths of about 360 and 640 km. A relative minimum of stress drop is found at about 450–560 km where the earthquake frequency is particularly high. Earthquakes at the northern end to the Tonga arc, where the Benioff zone is laterally bent, show systematically higher stress drops than other events at comparable depths, but away from the bend. Also, events in regions of low seismicity appear to have higher stress drops than those in regions of high seismicity. The upper bound of seismic efficiency is found to decrease with depth, implying an increase of frictional force with depth at the earthquake source.