The aim of this paper is to make a statistical investigation of the pressure oscillations occurring near the center of a typhoon. An analysis of the barograms recorded on the Southwest Islands, Japan, during the decade from 1956 to 1965 shows that the occurrence frequency of the oscillations with the maximum double amplitude exceeding 3 mb is approximately 10%, and the frequency of those with the maximum amplitude of 2 mb or larger is about 30%.
In order to get a clue as to the relationship of pressureos cillations to an elliptical eye, the reported frequency of an elliptical eye is also examined in this paper, based on the data obtained from flight observations. The reported frequency of an elliptical eye is estimated to be about 10%, and is considered to be fairly independent of the central pressure.
Considerable agreement of the occurrence frequencies of both pressure oscillations and an elliptical eye suggests that pressure oscillations are closely related with an elliptical eye. From an examination of reports on an eye shape in cases where pressure oscillations were recorded, it is inferred that some of oscillations are associated with an elliptical eye.

View full abstract

The Japan Meteorological A gency is responsible for the issuance of the warning for near and distant tsunamis. The time limit of the warning issuance is twenty minutes after a tsunami occurring in a region within 600 km from the Japanese coasts.
The number of data sent through the exclusive telegraph of the Agency by seismic monitoring stations amounts to more than one hundred within ten minutes after the occurrence of a large earthquake. It is not an easy task to process manually all the information in the short period of time, and it is urgently required that this task to be automated.
The telegram data, however, are less accurate than those measured in a normal condition, and sometimes the data are in confusion. The most difficult problem in the automatic processing of the tsunami warning data is the rejection of incorrect data in the course of computation. The difficulty, however, will be overcome by making use of a man-machine communication system in the data processing.
From this viewpoint, a program for analyzing data for the tsunami warning is written for a computer of small capacity consisting of a video display and other devices.
The program is tested using some earthquakes occurring in and near Japan and the results prove successful. The system will reduce remarkably the time limit of the issuance of the tsunami warning.

View full abstract

The Owase area is well known to Japanese meteorologists as the heaviest rainfall region in this country. The Owase weather station recorded a maximum annual rainfall over 6,000 mm and a maximum daily rainfall over 400 mm. Nevertheless, it was pointed out by the weather radar meteorologists at Nagoya that the radar 142 km apart from Owase is unable to show the heaviness of the Owase rain. This suggests that either the height of the rainclouds are so low that the radar beam misses them or the raindrop spectra have particular characteristics.
Therefore, m a ny cloud physicists has been interested in the rain mechanism at Owase. The present field investigation was planned primarily to provide data for giving an answer to the question raised in radar utilization. The instruments used were two x-band radars, a vertically pointing Doppler radar and a short range RHI radar, and auxiliary ground instruments such as a raindrop recorder and raingauges. The data collections made successfully during the operational period were for the typhoon rainband on 30th to 31st Aug. and for showers developed along a stationary front on 16th to 21st, Sept.1971. The present paper mainly describes the case study on the latter shower, though it was not a case of heaviest rain typical of the area. This paper is also composed of two main parts: the first part, Section 2 through 6, aims to construct the cloud structure, and the second part, Sections 7 through 10, deduces a probable precipitation mechanism which is capable of generating rainfall of reasonable intensity.
A quasi-statio n ary front moving slowly eastwards existed during the period from 16th to 18th, Sept. as shown in Fig.1. Radar observation was made on the showers which had been developing along the stationary front around the Kii-peninsula, Honshu, Japan.
The Doppler radar data of height-time section and the RHI echo video of three dimensions are as shown in Figs.3,4 and 6a, and they were compared with the PPI gain step echoes of the Nagoya radar which is shown in Fig.2a. Thus the three-dimensional structures of the elementary convective cloud and their complex echoes were constructed as shown schematically in Figs.5b and 6b. Each convective cloud is consisted with a tower part and the precipitation streak from the tower which spread out at the surface. The two parts, the tower and the precipitation streak, are also exhibited primarily in Figs.3 and 4. The shower complex was composed of such convective clouds whose towers were horizontally spaced by several to ten kilometer intervals, as revealed in Fig.2a.
The Doppler radar also provides the vertical distr i butions of up- and down-drafts within the rainclouds, as given in Figs.7 to 9. Especially in the present case, considerably steady updrafts about 1 m/s on average (2-3 m/s in the maximum) have been found within the low level streaks, as shown by the second mode in Fig.9. The low level cloud thus involving a relatively steady updraft is noticed to play a significant role in the development of heavy rain. As the generated raindrops fall from the tower cloud into the low cloud in a seeding style, they develop in the form of a trapezoidal spectrum as shown schematically in Fig.14. Analyses of the observed raindrop spectra show that a rapid increase in the rainfall rate is associated with increase mainly in concentration of raindrops rather than in the spectral width until the rate reaches a certain characteristic magnitude, say 9 mm/hr in the present case, but after that the rainfall increase is associated with increase mainly in the spectral width, i. e. Dm, rather than in concentration. The latter growth pattern of the above, which is specified in the present case as “trapezoidal growth of raindrop spectrum”, is considered to occur in relation with the double layer structure of raincloud as revealed in Fig.9and by echo analysis.
Under the assumption of this drop growth mechanism,

View full abstract

In order to investigate the structure and mechanism of occurence of the heavy rainfall around Owase, a vertically pointing Doppler radar, a short range RHI radar and a 8.6 mm cloud detecting radar were operated in the yard of the Owase weather station together with a raindrop recorder and other weather instruments, for two months from August to September,1971. Analyses were made by adding the data taken by the Nagoya weather radar of 5 cm wavelength and the radio-sonde sounding at Shionomisaki. The present paper describes the rainband associated with Typhoon 7123 which brought a heavy rainfall at Owase.
Four hundred km apart from the Owase on the eastside of the typhoon, a broad rainband about 500 km in width was developed and travelled ENE-wards with the typhoon (see Fig.11). At Owase rainshower began at about 0700; on 30th August, and the peak rainfall, nearly 200 mm, was observed between 1500 and 1800, and the daily rainfall amount recorded 500 mm (Fig.6).
From the PPI observation, it is shown that the broad band consists of many spiral bands about 50 km in width. The stratified video rainbands contained many intense convective cellular echoes and travelled eastwards successively with certain intervals. Each individual cellular echo developed successively on the sea south of Owase and moved to the north bringing heavy rainfall at Owase (see Figs.11,18 and 20).
From the echo analy sis of the RHI radar echo, it was found that a bright band developed at 4.5 km in height through the rainband (as shown in Figs.12-14) showing that the upper part of the spiral band clouds was formed of snow crystals. Relatively few of the intense convective echoes penetrated the bright band as thunderstorms, but in most convective echoes the updrafts were restricted within the layer below the bright band as revealed in the Doppler radar data. Such convective cloud had an updraft less than 4 km in height and about 2 km in width and with a maximum magnitude of 10 m/s, and such updrafts were distributed within intervals of about 5 km (see Figs.15,17 and 19).
Further, the Doppler data showed many updraft bubbles as shown in Fig.15 with a considerably low layer, lower than 1 km, and indicated that the average vertical current for a certain duration during which the intense rainfall peaks were associated with the time near the strong updraft was combined with a relatively steady updraft about 1-2 m/s which developed within the low layer. (see Figs.15and 16).
From the above results it is concluded that, although the rainfall rate is usually of the order of 10 mm/hr when the clouds develop mostly stratified with a bright band, those which develop within the circumference of a cumulonimbus and combined with lower cumuli can cause the rainfall rate to intensity to a heavy grade, say of the order of 100 mm/hr, and also to continue more persistently.

View full abstract