Journal of Seismology and Earthquake Engineering
Volume:4 Issue: 2, Summer and Fall 2002

Accurate focal depth estimates are essential for the correct interpretation of seismicity data in terms of regional tectonics and earthquake hazard assessment. Published global earthquake catalogues are a common source of focal depth information, but how accurate are they? We compare estimates of focal depths from the Harvard CMT catalogue and the Engdahl et al [8] relocations of the ISC catalogue with those determined by teleseismic waveform inversion methods, and find that the catalogues can be in error by up to 60km.

The specifications of the recorded strong motions are investigated based on some selected accelerograms obtained in the Iranian and Turkish networks. The data are all recorded in the distances less than 30km to their corresponding surface fault ruptures. These records are selected based on the amplitude of the recorded acceleration, their distance to the fault and the magnitude of the earthquakes. They show, in some cases, the directivity effects based on the multiple corner frequencies observable on the Fourier spectra of the acceleration and displacement time-histories. The cases for which the records are available in the beginning and the end of the rupture, were studied to show the possible differentiation of rupture and rise times, according to the azimuthal positions of the recording stations relative to the rupture locations. The displacement pulses (width and amplitude) are different accordingly; for instance the greatest displacement pulses of Tabas record, obtained in Tabas earthquake of 16 September 1978, Mw7.4, are representative for a position in which the rupture front is approaching towards the recording site. Different studied cases in Zagros and Central Iran, as well as the records of the Turkey earthquakes (that are obtained in the nearest distances to the fault) of 17 August 1999 (Kocaeli, Mw7.4) and 12 November 1999 (Duzce, Mw7.1) indicate the rupture times differing from 2 to 10 seconds and the rise times from 0.8 to 2.85 seconds for the earthquake magnitudes of 5.9 to 7.4 and the distances to the zone of energy release varying from 5 to 30km

A new scheme for the application of the intermediateterm medium-range earthquake prediction algorithm M8 is proposed.
The scheme accounts for the natural distribution of seismic activity, eliminates the subjectivity in the positioning of the areas of investigation and provides additional stability of the predictions with respect to the original variant. According to the retroactive testing in Italy and adjacent regions, this improvement is achieved without any significant change of the alarm volume in comparison with the results published so far.

The dynamic stability of shallow underground openings in jointed rock masses depends on several parameters that the most important of them for design engineers is the shear strength of rock joints. In this research in order to study the effects of cyclic loading on this parameter, about 50 joint replicas have been tested using triaxial and direct shear testing devices. The saw-tooth identical joint samples were prepared using silicon moulds and special mortar. The samples have been tested in different loading conditions to simulate the effects of small and large earthquakes. Variations of the shear strength with the parameters of cyclic loading and also degradation of asperities have been studied during the tests. The results show considerable changes of shear strength due to the rate of loading, number of cycles, frequency and amplitude of loading. Also the trends of wearing and asperity degradation have been studied in higher levels of cyclic displacement

Earthquake response of liquid storage tanks isolated by the sliding systems is investigated under bi-directional earthquake motion (i.e. two horizontal components). The frictional force of sliding systems is modeled by two ways referred as conventional and hysteretic model. The continuous liquid mass is lumped as convective mass, impulsive mass and rigid mass. The Corresponding stiffnesses associated with these lumped masses are worked out depending upon the properties of the tank wall and liquid mass. The governing equations of motion of the tank with sliding system are derived and solved by Newmark''s step-bystep method with iterations. The frictional forces mobilized at the interface of the sliding system is assumed to be velocity independent and their interaction in two horizontal directions is duly considered. For comparative study the earthquake response of isolated liquid storage tank obtained by conventional model is compared with corresponding response obtained by hysteretic model. In order to measure the effectiveness of isolation system the earthquake response of isolated tank is also compared with non-isolated tank. A parametric study is also conducted to study the effects of aspect ratio of the tank on the effectiveness of seismic isolation of the liquid storage tanks. It is found that the sliding systems are quite effective in reducing the earthquake response of liquid storage tanks. In addition, the same earthquake response of liquid storage tanks is predicted by conventional and hysteretic model of the sliding system.

In the seismic response analysis of large structures the effects of differential support excitation should be considered. The differential support excitation may be due to asynchrony in excitation of different supports, caused by the finite speed of travelling earthquake waves and/or due to non-uniformity of these excitations, resulting from a change in the properties of the earthquake waves as they travel. In this paper; due attention is given to the question of non-uniformity of support excitation and its effects on the response of such structures as gravity dams and cluster buildings. Non-uniformity of ground excitation is modelled using a set of simulated acceleration time histories obtained from a representative spatial variability model. Comparative studies on the effects of asynchrony and non-uniformity of supports'' motion on the structural response show that non-uniformity may, in some cases, amplify the effects of asynchrony and therefore should be considered in the analysis. It is also shown that for cluster buildings and buildings with large floor areas, the effects of differential support excitation could be considerable.