A method for computing focal mechanisms by using polarities and amplitudes of the P wave first motions was designed. In the first step, some suitable solutions were picked up by using only polarities. A method based on the ‘Fourier method’ designed by AOKI (1986) for fast computation was adopted. In the second step, the most suitable solution was selected by using amplitudes. In order to avoid the path effect on the attenuation of the P wave amplitude, an earthquake cluster occurring in a very small region was analyzed and the amplitude correction for each station was made.
This method was applied to the analysis of the aftershock sequence following the earthquake with M4.9 which occurred in Kyoto-Osaka border region on May 28, 1987. As a result of this analysis, some characteristic features of the distribution of focal mechanism types were found. Most of focal mechanisms of aftershocks showed the same type as that of the main shock, that is, east-westward P axis and thrust type focal mechanism. But those of the aftershocks which located at the edge of the aftershock area or slightly distant from the fault plane of the main shock almost showed strike-slip type or intermediate type focal mechanism. Such focal mechanisms may be related to the stress field disturbed by the dislocation of the main shock.
It is conculuded that this method is very effective to calculate focal mechanisms for a group of earthquakes whose polarity data are scarce with sufficient accuracy.

抄録全体を表示

The observation of the concentration of ²²²Rn in air in tunnels have been carried out using flow-type ionization chambers in Tokai area. Observations started in April, 1977 at Toyohashi (TY), June, 1980 at Kikugawa (KI) and August, 1986 at Toyone (TN), Inabu (IB) and Asahi (AS). The ²²²Rn concentration observed at Toyohashi and Kikugawa shows pulse-like increase after heavy rainfall and is also affected by the atmospheric pressure change. It shows seasonal variation: it increases in summer and decreases in winter.
Measurements of ²²²Rn concentration in groundwater flowing in the tunnel at Toyohashi had also been carried out using the method of a toluene extraction-liquid scintillation counting for 13 months from May, 1986 to May, 1987 in order to compare it with the ²²²Rn concentration in air in the same tunnel. The ²²²Rn concentrations in groundwater also showed seasonal variation, though its amplitude was about a half that in air.
We assumed that the ²²²Rn rich soil gas supplied the tunnel with ²²²Rn. We estimated the pulse-like disturbances of ²²²Rn concentration in air observed after rainfall from local strain variation caused by rainfalls. In this calculation, it was supposed that the amount of the ²²²Rn rich soil gas squeezed out from the basement rock, increases in proportion to the contraction rate of the basement rock. Relatively good agreement was obtained between observed and calculated values. We infer that the seasonal variations of ²²²Rn concentration in air and in groundwater in the tunnel are due to the seasonal variation of the ²²²Rn concentration in the soil gas in the basement rock.

抄録全体を表示

The height of a sea wall is usually designed based on water levels of large tsunamis experienced in the past. Behaviors of tsunamis exceeding the design height are examined by means of numerical experiments taking the sea walls at Matsuzaki, Shizuoka prefecture and at Taro, Iwate prefecture. In the case of Matsuzaki, the inundated area reduces to 70 percents of the original one by the presence of the wall although the invasion of tsunami through the river mouth, where no water gate is constructed, is noticeable. In the case of Taro, the sea wall height is designed at 10 meters that equals the height of the 1933 Sanriku tsunami. According to our experiment, however, if a tsunami has the same height as that of the 1933 tsunami, the water level close to the wall front becomes higher than the height of the sea wall. In this case, the water level at some places in the town becomes so high as to give rise to human and material damage in spite of the construction of the sea wall. It is recommended, therefore, that local inhabitants take refuge onto hilltop, unless a small height is predicted by the tsunami warning.

抄録全体を表示

Non-uniform slip on a heterogeneous fault is investigated to describe the detail of earthquake rupture processes. The fault is parameterized by average stress drop and fault size to model the deterministic part of the faulting process. The size of small scale fault heterogeneities and the variance of local stress drop are introduced to describe the stochastic part of the faulting. Nonuniform slip (or non-uniform stress drop) tends to increase the high frequency contents of S-waves, and thus controlls the excitation of strong ground motion. Short-waves from the heterogeneous fault are generated as an energy-additive form by the stochastic fracturing of random fault heterogeneities. This brings us a seismic directivity effect particular to the incoherent short-waves, which is different from the seismic Doppler effect on the long-waves. This short-period seismic directivity predicts the azimuthal variation of short-wave amplitudes as
(1-v cos θ/β)^-1/2
for a unilateral faulting and
{1-(v cos θ/β)2}^-1/2
for a bilateral faulting, where v is an average rupture velocity, θ is the station azimuth from the fault-propagating direction, and β is S-wave velocity. Maximum and root-mean square accelerations are derived theoretically in terms of the variance of local stress drops, number density of fault heterogeneities, and fault size considering the short-period seismic directivity.

抄録全体を表示