Tidal modeling experiments have been carried out using 2-dimensional POM (Princeton Ocean Model) which covered the area of 115∼150°E, 20∼52°N of northwestern Pacific with 1/12° horizontal resolution. ETOPO5 bathymetry has been chosen as the depth data. Harmonic constants of eight tidal components, M2, S2, K1, O1, K2, P1, N2, Q2, which was result of tidal numerical model calculated in global ocean by Matsumoto et al. (2000) were inputted at open boundaries. Several tests have been performed with varying the minimum depth, the bottom friction coefficient and the horizontal diffusion coefficient in order to find the best solution in good agreement with the observations.
The results showed that the improvements for M2, S2, K1 tide were made up to 42%, 32%, 26% as the minimum depth increase from 10m to 25m. The discrepancy between model results using different minimum depth is found to be up to 20cm for M2 tidal amplitude around Cheju Islands and the positions of amphidromes are dramatically changed in the Bohai Sea. The calculated ARE(Averaged Relative Error) values have been minimized when the bottom frictional coefficient and the minimum depth is 0.0015 and 25m, respectively. RMS errors were reduced when the horizontal diffusion coefficient was the constant, 2000. The effect of the horizontal diffusion coefficient, however, was relatively small(less than 7% for amplitude) compared with the effects of minimum depth and bottom friction coefficient.
To calculate the residual current from the 8 tidal components, a sensitivity test for the averaging period has been carried out. It is suggested that the proper averaging period for residual currents calculation is about 15 lunar days. The residual current speed is strong at the southwestern tip of Korea Peninsula, Gyoung gi bay, Hang ju bay and the northern tip of Taiwan. The prominent features of this residual currents are the northward currents at the western coast of Korea Peninsula and the two eddies at the northeastern part of Yangtze Bank.