Fig.1 Area analyzed in this paper. We divided the area into 26 sub-domains A through to Z (shown in the upper row of each sub-domain) by longitudinal and meridional lines of 1 degree intervals. The total number of observations is given in the middle row. Numbers of observations where saline water higher than 34.95 and 35.00 was observed is shown in the bottom row without and with parentheses, respectively.
3. Occurrence Frequency of High Salinity Waters
We divided the area under consideration into 26 sub-domains, A through to Z, as shown in Fig.1, by longitudinal and meridional lines of l degree intervals. The number of observation points in our analyzed period is given in the middle row of each sub-domain by numerals. Numbers of the observation points where the saline water higher than 34.95 and 35.00 were observed are shown in the bottom row without and with parentheses, respectively.
We selected salinity values from 34.70 through to 35.00 at intervals of 0.05. The occurrence frequency of observations where salinities higher than each of these selected values was computed. The frequencies in % are shown in Fig.2 for each sub-domain. The frequency decreases with increasing of salinity value, but the manner of decrease is different between sub-domains. By using this decreasing manner, we classified the sub-domains into three groups：
(1)The frequency of waters with 34.70 or higher salinity is relatively high(say,＞80%). The rate of decrease is small for salinity values lower than 34.80, but decreases rapidly between 34.80 and 34.90 (full lines in Fig.2).
(2)The frequency at 34.70 has moderate values between 80 and 60%. The frequency decreases almost lineally from 34.70 to 34.90 (dotted lines in Fig.2).
(3)The frequency at 34.70 is lower (＜60%). The rate of decrease is rapid for lower salinities, but slow for higher salinities (dashed lines in Fig.2).
In Fig.1, the sub-domain belonging to group (1) are shown by a hatched area, those to group (2) by a dotted area, and those to group (3) by a white area, respectively. Group (3) is located near coast where the fresh water supply from land would be significant, The area of group (1) appears in the domain of the subtropical water, which is usually located to the south of the Kuroshio axis. The area of group (2) is located between those of (1) and (3).
The curves in Fig.2 belonging to groups (2) and (3) decrease towards nearly zero by 34.90. Though the occurrence frequency of 34.90 water is considerably low, the decreasing manner of these curves suggests that the salinity value of 34.90 is not unusual and may result from usual salinity fluctuations in the region under consideration. However, the occurrence of saline waters having salinity higher than 34.95 appears to be some what abnormal. Such abnormal high salinity waters may occur in region accompanying sporadic event-like phenomena.
4. Case Studies When the Abnormally Saline Water Higher than 35.00 was Found
4.1 R/V Wakayama and soyo-maru in August 1984.
The R/V Soyo-maru observed abnormally saline water higher than 35.00 on the same day as the R/V Wakayama (Table 1). Sites of these observations are shownin Fig.3. Among five observation lines running in a north-south direction, the westernmost line just off Cape Shionomisaki was surveyed by the R/V Wakayama on August 2, and the two middle and two eastern lines by the R/V Soyo-maru on July 23-27 and on August 1-2, respectively. The bold portions of the observation lines indicate that the abnormally saline water higher than 35.00 was found in the vertical section.
It should be noted that the abnormally saline water was found along the observation lines occupied on August 1-2, but not found along the lines occupied about one week before. Cross-marks (×) attached on the observation lines indicate the place where the highest salinity value was found and/or where the stratum of abnormally saline water was found at its thickest in each north-south cross-section.
