The data clarified the seasonal changes in nutrients, which was not precisely grasped before because the systematic monitoring programs did not have sufficiently fine interval. Figure 7 shows the time series of DIN (= NO₃+NO₂+NH₄), DIP (= PO₄), and DSi (= Si(OH)₄) between 134.6^oE and 134.8^oE in the period of April 1994 to March 2001 extracted from the total data. Concentration of DIN, DIP, and DSi decreases around February due to the uptake by phytoplankton composed mainly of diatom during their spring bloom (). At the end of spring bloom, they sink to the bottom, where they are decomposed to inorganic nutrients. In autumn, the nutrients are brought back to the upper layer by the vertical turnover caused by the cooling of surface water (). The amount of DSi recovers earlier in summer, however, due to the runoff from the monsoonal rain. As the nutrients are brought back to the upper layer, the condition becomes favorable for another bloom and when the light condition becomes adequate in the following year, the bloom of the phytoplankton starts again. This is the basic seasonal pattern of the nutrient cycle, however there are some difference depending on the local physical conditions.

Figure 7. A time series of nutrients obtained at the central part of the Harima-nada Basin (134.6o E ~ 134.8o E) in the period of April 1994 to March 2001. The diatoms are thought to absorb DIN, DIP, and DSi in the molar ratio of 16:1:16 (Redfield ratio). The DIP value is multiplied by 16 to evaluate the contribution of each nutrients to the plankton bloom.

How about the spatial distributions? All of the data were averaged for the period of April 1994 to March 2001 at every 0.2^o longitudinal section and plotted in Figure 8.
According to this plot, salinity is higher in west, while DIN, DIP, and DSi are higher in east. This spatial trend is reasonable because the highly populated and industrialized area, as well as the effluent of Yodo River, is located at the east end of the inland sea. It is noteworthy that the ratio of DSi/DIN is lower in the east. What causes this trend? And, how does it affect the marine ecosystem, such as bloom of the phytoplankton?

Recently, the "silica deficiency hypothesis" has been attracting attention⁷⁾. The anthropogenic loading of N and P and the artificial stagnant waters due to the damming of rivers have increased, whereas the supply of silica due to the natural weathering remains the same. The combination of such changes enhances the growth of fresh water diatoms, which absorb the DSi and sinks to the bottom, and will reduce the concentration of DSi in the water and will ultimately cut the amount of DSi reaching the coastal seas. This may cause the gradual shift from dominance of diatoms (siliceous) to that of flagellates (non-siliceous). Whereas the diatoms are the basis of a healthy ecosystem, the flagellates are potentially harmful in that it may cause the red tides, which damage the fisheries.

Figure 8. Horizontal distribution of nutrients values, which were obtained by averaging the values in the period of April 1994 and March 2001 for every 0.2o longitude and plotted against the middle point of longitudinal averaging range. Beppu is at near the left end, and Osaka is at right.

The upstream of Yodo River is Lake Biwa-ko, which has been eutrophicated.
The concentration of DSi in Yasu River, one of the rivers flowing into this lake (Fig.1), is around 200 µmol/L. However, it drops by one order of magnitude in the Lake Biwa-ko, implying the uptake of DSi by the fresh water diatom. As a result, the eastern part of the Seto Inland Sea, affected by the Yodo River, is characterized by the lowered ratio of DSi/DIN. This explains the fact that there are more occurrences of harmful red tide in the eastern part on Inland Sea than the west.

By looking at the details of the comparison between the Osaka Bay and Harima-nada Basin, both located at eastern part on Inland Sea, however, one will see that the number of occurrence of flagellates' blooms is higher in Harima-nada Basin in spite of the higher DSi/DIN ratio. This is presumably because the certain amount of DSi is supplied from Yodo River directly to Osaka Bay and enhancing the diatom bloom, even though the ratio of DSi/DIN has been reduced in the Lake Biwa-ko. There are limited numbers of marine monitoring programs dealing with DSi in Japan, and this data has contributed to many other researches.

Considering the significance of the silica deficiency issue, SCOPE (Scientific Committee on Problems of the Environment) has organized two International Workshop on the Land-Ocean Nutrient Fluxes: Silica Cycle⁷⁾ (Linköping, Sweden, October 3-5, 1999, and Nha Trang, Vietnam 25-28 Sept. 2000). Our monitoring data was one of the few data presented from the Asian regions.