*Tongjun Sun3,1, Sachihiko Itoh3, Eisuke Tsutsumi3, Miho Ishizu2, Kiyoshi Tanaka3
(1.Graduate School of Agricultural and Life Sciences/Faculty of Agriculture,The University of Tokyo, 2.Japan Agency for Marine-Earth Science and Technology, 3.Atmosphere and Ocean Research Institute, The University of Tokyo)
![](https://confit-sfs.atlas.jp/customer/jpgu2021/web/OSPA.png)
Keywords:Otsuchi Bay, tide, internal wave, coastal trapped wave, spring tide
In the Sanriku coastal areas, various current and waves vigorously interact on the narrow shelf and likely have major influences on the water exchanges between inshore and offshore areas. The inertial period dividing internal waves and coastal trapped waves is approximately 18.9 h at Otsuchi Bay in the middle of the Sanriku coast, and semidiurnal and diurnal frequencies become superinertial and subinertial, respectively. Consequently, both of these waves and their interactions are thought to have impacts on the inflow to and outflow from the bay. In this study, based on the mooring observations at the two stations at the north and south of the bay mouth during 2012–2016, we aim at showing the variability of internal waves forced by tidal flows and coastal trapped waves, and clarifying water exchange processes between the inshore and offshore areas. The mooring observation revealed that flow velocities were excited in summer at semidiurnal and diurnal bands, where the former was markedly strong. By contrast, the velocities of the semidiurnal and diurnal flows were comparable during winter, whereas their amplitude decreased down to ~1/5 of that during summer. Velocity data during July–October 2013 where both of the two stations are available showed that low-frequency background flow (period > 40 h) entered from the north and exited from the south of the bay mouth, respectively. Although the spring tide should appear approximately 2 times in one month due to its inherent fortnight period, semidiurnal and diurnal tidal flows had a shorter period and their amplitude showed irregular fluctuations. To investigate the cause of this fluctuation, harmonic analysis was applied to tide gauge records at the Kamaishi tide station, extracting semidiurnal and diurnal components, and were compared with the velocity fluctuations recorded at the mooring stations. The sea level fluctuations were dominated by the semidiurnal component during this period. The peaks of semidiurnal and diurnal tide almost coincided in the early July, but gradually separated after August, and then approached in the late October and finally coincided again. On the other hand, during February–May 2013, the sea level records showed spring/neap cycle and the fluctuation in the peak timing, but the velocity amplitude was very weak and the difference in the peak timing was hardly detected. As these phenomena cannot be explained solely by the local tides, interactions between semidiurnal and diurnal tides, wave propagations from remote areas, influences of low-frequency flows and changes in the stratification are thought to play some roles.