On 26 September 1958, a severe typhoon advanced into eastern Japan and heavy rainfall occurred on the Izu Peninsula and in the southern Kanto District. Especially an extremely large amount of precipitation was concentrated to the upper part of the Kanogawa River Area, resulting in a violent and disastrous inundation of the middle part of the Izu Peninsula.
The purpose of this paper is to reveal the details of rainfall distribution over the Izu Peninsula and its adjacent area at that time. To this end, the pattern of time change in the precipitation was analized.
The locations and the numbers of the various kinds of observation stations are shown in Fig. 1. As for small scale analysis, available data are extremely sparse, so it was hardly possible to construct maps of small-scale wind field and hourly rainfall distribution.
Therefore, first of all, a detailed map of the total rainfall distribution of that area was drawn, using coaxial method, since the rainfall distribution in a mountainous region is greatly affected by the complexity of the topography. The topographical factors were thus represented as follows:
1. Heights.……The mean value of the elevations of eight points on the circumference of a circle, drawn around the observation station using a 5km radius.
2. Valley-direction, ……Equivalent to the direction from the center of the point of minimum elevation of the circumference of the same circle.
3. Mountain-derection (Maximum inclination).……The direction of maximum inclination is equivalent to the direction from the center of maximum elevation of the circumference of a similar circle with a radius of 2.5km).
4. Exposure.……Unshadowed parts by more than 400 meter heigher mountains than the elevation of the obsevation station. (expressed by the center-angle of a circle, drawn around the observation station with radius 20km)
5. Distance from the path of the typhoon center.……expressed in terms of the minimum distance from the track of the center of the typhoon.
6. Distance along the path of the typhoon center.……The initial point being mesuared from Nagatsuro Local Meteorological Observatory.
7. Region.……Determined by the results of the construction of coaxisual nomogramms, which describe two regions; 1. Most of Izu Peninsula 2. The northern patr of the Izu area.
The coaxial graph shows that the factors 1, 2, 3 and 7 were significant. The topographical factors were measured, and the rainfall amount were culculated graphically and plotted on the maps with each 4 km-meshed grid points. Thus the density of the places where the rainfall amount was estimated is three times greater than the actual rain-gage stations, and the isohyeth were drawn with ease and exact (Fig. 3).
As is seen in Figs. 2 and 3, a larger amount of rain was precipitated on the northern slope of Amagi Mountains than on the southern slopes and the belt of heavy rainfall crossed from southwest to north-east. In the central part of the Izu Peninsula as much as 700mm rainfall was recorded.
In order to get time chages of precipitation patterns which evolved with the approach of the typoon, hourly rainfall amounts were converted into percentages (assuming the total rainfall was 100 percent) at each hourly observation station. These percentage values were then plotted on the maps (scale 1/500, 000) and isohyeths were drawn. (An example was represented in Fig. 4) Next, using the Figs. 3 and 4, hourly precipitation pattern were drawn (Fig. 4) The results show that the heavy rainfall zone was located unwavingly in the middle of the Izu Peninsula on the afternoon of 26th September 1958. Particuarly unchanged the location of the maximum rainfall where 80mm of hourly rainfall between 8 and 9 p.m, and 120mm of rainfall between 9 and 10 p.m. were recorded. That was at Yugashima Village in the Middle Izu Peninsula.
View full abstract