Surface pressure patterns of summer around Japan were classified objectively by use of correlation method. Surface pressure at 36stations, not at grids point, was used. Although 8 types of mappattern were found, difference among them is not always clear, and 30% of the entire samples can not be classified into the 8 types. However, an improvement of this method may find the relation. between some weather phenomena and pressure patterns, with reference to the other meteorological elements.
This is the introductory note to the three successive papers on the unusual rise of sea level in September,1971. They form a sequel to the cooperative study on the physical mechanism of the unusual sea level variations (December 1971March 1972, Representative: Prof. K. YOSHIDA).
High sea levels with time scale more than one week appear sometimes along the Pacific coast of West Japan when the meteorological conditions are rather gentle. If the phenomena appear associated with the spring tide in late summer or early fall, the sea water floods over the land at some places on the coast. In this paper, eight cases of the h igh sea level selected from the tide records of five years from 1964 to 1968 and one case in September 1971 are investigated in connection with meteorological and oceanic conditions. The high sea level considered here is not the so-called “storm surge” caused directly by meteorological disturbances. It may be due to the variation of the Kuroshio which is induced by some meteorological conditions. The slow westward propagation of phase of the high sea level may be explained by considering the continental shelf waves.
A few typical cases of unusual high or low mean sea level are selected from the tidal data for Tokyo Bay by the following ways: (1) Unusual days are selected at each station under the condition that the magnitude of daily mean tidal residual be mo r e than twice its standard deviation. (2) We take up those cases in which more than three consecutive unusual days occur simultaneously in the whole area of Tokyo Bay. Both meteorological and oceanographic factors are considered in connection with the occurrence of unusual high or low sea level, which takes place once a year or so. At such times the weather condition is rather gentle in general, and the oceanic condition seems to be responsible for the unusual rise or fall. Warmer sea surface temperature is likely to be associated with the rising of the sea level and colder temperature with the falling. In the inner part of the bay, the land and sea breezes are related to these variations, so that the former (southerly wind) I serves to raise up warmer water, and the latter (northerly wind) to drive out colder water.
This study aims to simulate the unusual rise of sea level in September,1971, by the barotropic motion on the continental shelf to the south of Japan based on storm surge equations. We chose two numerical models. In the first model, we only considered the effects of Typhoon 7123. In the second model, the approach of the Kuroshio axis was taken into account. Results of computations show that the long-lasting rise of sea level is only possible in the second model. T he continental shelf wave moving to the west was also simulated in that model.
It was found that the water-soluble polyvinyl alcohol (poval)film is available for the size measurement of water droplets about 10to 300μ in diameter. The relation between the diameters of droplets (d) and those of their traces on the film (D) was expressed by the following empirical equation D=4.54 d0.8, and the calibration curve was presented in Fig.4. It was found that the calibration curve could be extended to larger drops of about 4 mm in diameter. The poval film is further available as a specimen supporter for the electron microscope. For example, the size of a trace of small droplets on the film can be measured with an electron microscope and besides some materials contained in a droplet can be identified by the electron diffraction method.
Examining the data of field experiments of elevated-source diffusion at some coastal regions of Japan and paying special attention to the analysis of vertical concentration profile, we have obtained the following results: 1. Sometimes a remarkable rise of tracer cloud occurs over a hill or a ridge, which makes the surface concentration pattern much different from that typical of the elevated-source diffusion over a flat terrain. 2. The sea breeze, usually stably stratified over the sea in the daytime of warm season, is subject to the heating from below inl a nd and a convective layer develops with the downwind distance from the coastline. Accompanying the modification of sea breeze inl a nd, the diffusion rate changes from one over the sea to that in the neutral layer. This is well reflected in the change of Pasqu i ll stability in the vertical spread of concentration. 3. Two peaks are found in the vertical concentration profile, when the source height is below the base of the inversion or s t able layer. One corresponds to the source height and the other coincides with the base of the inversion or stable layer. The concentration of the latter is much larger than that of the former, which makes the surface concentration decrease.
The present paper presents the local differences in the seasonal variation of stroke mortality and the relation between stroke mortality and temperature in Japan where deaths have got concentrated particularly in winter. The statistical analysis has been made prefecture by prefecture and for the six big cities. And deaths from stroke are compared with those from heart diseases. 1. The death index for stroke shows a sharp peak, regardless of temperature changes, from the northern part of the Tohoku dis trict through the southern tip of Kyushu. In all other prefectures, seasonal variation is bigger than in Hokkaido. The w inter peak is higher for mortality from heart diseases than that from stro ke, and seasonal variation is again the smallest. This is attributed to the difference in the heating system between Hokkaido and other prefectures. 2. The coefficient of regression of stroke mortality on temperature is generally negative and, moreover, it is higher in the northern prefectures except Hokkaido, where the seasonal variation of stroke mortality is slightly smaller than in the urban areas. Among the six big cities, the death rate is higher in Tokyo and Yokohama than in Kyoto and Osaka, and Tokyo shows the bigge s t coefficient of correlation between mortality and temperature for both stroke and heart diseases, but no significant difference can be seen among the six cities. Due consideration has to be given to the marked differences in death causes between the urban areas and the northern districts for both stroke and heart diseases and to the differences i n living customs between the Kanto and Kansai regions. 3. In the winter of 1962, the death rate was exceptionally high due probably to the country-wide spread of influenza. It is interesting to note the exceptional peak only for heart diseases in the relatively warm areas and for both stroke and heart diseases in the cold districts. 4. In the districts where temperature exceeds 24°C in the season, the death rate declines along with the rise of temperature b u t begins an upcurve after the temperature rises higher than 24°C. This phenomenon calls for careful studies from the viewpoin t o f biometeorology and human ecology.
We developed a new method of computing wave propagation to be used in our numerical model of ocean wave prediction (ISOZAKI and UJI,1973). This is a combination of the finite difference meth od and “jumped” technique. Several numerical experiments support that our method will give reasonable wave propagation.
Because of various advantages, such as high recording density, versatility of processing of recorded waves, and possibility of repeating processes, magnetic tapes have been widely applied to recording seismic waves recently. On the other hand, they have the disadvantage that the information contained in them cannot be utilized without the process of playing them back. On the whole, however, exploitation of magnetic tapes for seismometry has bright prospects. Having this aim in mind, we are investigating methods of processing seismic waves recorded on magnetic tapes. The procedures of analysis we are using now are as follows: First: Estimation of apparent directions of epicenters of earthquakes making use of P wave train. Second: Co-ordinate transformation of horizontal components of seismic waves to this apparent direction. Third: Use of these compound components in combination with vertical components in order to discriminate S and other phases. At present, the analogue processing is mainly used, but the digital processing is also being investigated for the purpose of ensuring more accuracy and precision and of repeating processes. In this paper (Part I of the study), processing of seismic waves by means of co-ordinate transformation of horizontal components of seismic waves and by means of multiplication of the transformed components with vertical components is treated. The purpose of coordinate transformation in this report is to seek the incident direction of P wave by changing the direction of the seismometer equivalently in the electric circuit.
Instead of L(t)Z(t) in Part I, -∫L(t)Z(t) dt is analyzed in this paper. The positive directions of rectangular co-ordinate (N, E, Z) are taken north, east and upward directions respectively. θ is measured clockwise from the north. The blockdiagram of the analyzing circuit is shown in Fig.1. The output of a magnetic tape recorder, three components of seismic waves, are processed through band-pass filter. Then w(t)=∫L(t)Z(t)dt where L(t)=N(t)cosθ+E(t)sinθ is analogically obtained at θ=0° and 90°, and recorded with a penrecorder. When the following two wave forms x(t)=-∫E(t)Z(t)dt y(t)=-∫N(t)Z(t)dt in adequate time duration from the initial P phase, are synthesized on the x, y co-ordinate plane, the co-ordinate point (x, y) moves along an orbit in the direction of the epicenter. As easily seen from the above definition, the x and y directions correspond to the east and north directions respectively. The direction of the orbit is measured with the eye, and called the apparent direction of the epicenter. One example of the above procedure is shown in Figs.3 to 4. Table 1 gives the list of earthquakes and the apparent directions of epicenters obtained in the above-mentioned procedure. The mark (i) means the case where three components of the initial motion of P waves are measured. Then accumulated frequency graphs for the difference between apparent and true directions of epicenters are shown in Fig.5. The frequency here means the relative frequency (maximum: 100). The conclusions drawn from this figure are as follows: (1) As for near earthquakes, the differences of both directions are generally small in the case where seismic waves are analyzed in the frequency ranges 1.0 to 2.0 Hz. (2) As for distant earthquakes, the difference of both directions are generally small in the case where seismic waves are analyzed in the frequency ranges 0.25 to 0.5 Hz. (3) The lower the frequency of seismic waves analyzed, the greater the number of shocks whose epicenter directions are difficult to be determine. This implies the frequency dependence of the energy of seismic waves. Fig.6 shows the deflections of the apparent direction of the epicenter to the true direction. From this figure, the following tend- ency is clearly observed. As for earthquakes whose epicenter directions are northeast from Matsushiro, including shocks of Japan to shocks of Alaska, the apparent directions of epicenters deflect north from the true direc- tion. The deflections of directions turn out to be greatest in cases where seismic waves are analyzed in the frequency ranges 0.5 to 1.0 Hz. The lower the frequency of seismic waves, the smaller the deflection of direction. On the other hand, deflection is small in cases where seismic waves are analyzed in the frequency ranges 1 to 2 Hz.
The method reported in Part II of this study is improved on. The resultant wave forms of the operational circuit are recorded directly on an; X-Y recorder. Though the principle is the same as in Part I, the experiment was done using two pairs of multipliers and integraters made by authors (Fig.1). The up-down motions of the recording-pen of the X-Y recorder was made at regular time intervals for the purpose of time mark. The processing was done in the two frequency ranges of 0.5 to 1.0 Hz and of 0.07 to 0.14 Hz. The list of analyzed earthquakes and estimated di r e ctions of epicenters are given in Table 1. Directions of epicenters obtained from the three components of the initial P phase are also given in this table. Examples of records are shown in Figs.2 to 12. Figs.13 and 14 are drawn in accordance with the procedure described in Part II. The conclusions drawn from this analysis are as follows: (1) Absolute values of deflections of directions of epicenters of near earthquakes are relatively small in cases where seismic wa v es of frequency ranges of 0.07 to 0.14 are analyzed, while t hose of distant earthquakes are relatively small in the case where seismic waves of frequency ranges of 0.07 to 0.14 Hz are analyzed. Due to slowness of response of the X-Y recorder, the frequency range of 1 to 2 Hz of seismic waves, where most favorable results were obtained for near earthquakes in Part II, is not analyzed in this case. (2) As in Part II of this study, apparent directions of epicenters obtained by this analysis of earthquakes whose epicenters are northeast of Matsushiro deflect to the north from the true direction. Estimated epicenter directions from three components of the initial P phases show a similar tendency. (3) Earthquakes whose initial P phases are observed in three components are picked up. Then directions of epicenters estimated from the initial P phases are compared with the direc t i ons of epicenters estimated with the method described in this report. These two kinds of estimated directions show no significant difference between them (Fig.15). (4) Examples of results of the same procedure for the PP waves are shown in Fig.16. Deflections of estimated epicenter d irections from true directions are greater in PP waves than in P waves, the resultant curves of both types of waves show similar features. (5) The S mark in Fig.2 shows the point corresponding to the initial S phases. If S waves consist of SH components, the direction of progress in orbit should be reversed and go towards the origin of the co-ordinate, at the initial Sphase. But as observed in Fig.2, the direction of progress is changed but not reversed in the actual orbit. This means S waves have not only SH components but also SV components. (6) One of the practical applications of this study is the azimuth finder for the epicenter. We want to try to make this instrument for practical use.
It is suggested that, if the crater bottom shows a tendency to upheave owing to underground stress, the eruption at the crater is approaching. At the Mihara Crater of Izu-Oshima Volcano, the crater bottom was usually deep during the dormant periods, and rose up to shallow levels before eruptions. Furthermore, great earthquakes in 1923(the Kanto Great Earthquake. Magnitude=7.9) and 1953 (the BOSOoki Earthquake. Magnitude =7.5), both of which had epicenters located near Izu-Oshima Island, occurred when the Mihara Crater bottom was shallow. Therefore, survey of the crater bottom should be available for prediction of volcanic eruptions or great earthquakes. At the present, the Mihara Crater bott o m is too deep to be seen from the crater rim. So topographical survey on an aeroplane was carried out on March 16 in 1972. The accuracy of location calculated for the base points of the present survey was less than ±30 cm. As the result of the present survey, a new topographical m ap was made, and the following facts were revealed: 1). Diameter of the crater mouth was 360 m across from north to south, and 400 m from east to west. 2). Diameter of the crater bottom was about 150 m. 3). Depth of the crater bottom was about 240 m from the crater rim. The bottom is located 448 m above the sea level. The crater bottom has changed in height since the 1950-51 Great Eruption as in the following: Apr.1951.........680m above sea level Nov.1951.........640m above sea level Feb.1955......620m above sea level Jan.1958........660m above sea level Mar.1968...... (about 400m)above sea level Mar.......1972-448m 4). The red-hot lava stayed at the crater bottom. The size of the lava pool was 6 m in length,3 m in width. 5). The volume of the crater (from rim to bottom) was estimated at 8.4X106 m3. 6). A sink hole,30 m in diameter and 15 m in depth, was found newly at the southwest cinder cone of the crater rim. This hole is the trace of the active pit in the 1950 Great Eruption.