AbstractDue to the joint constraints of boundary conditions, including seabed topography, driven water level at open boundary (DWLOB), and bottom friction coefficient (BFC), the accuracy of tidal numerical modeling in coastal and offshore waters is still relatively low. A two-dimensional tide numerical model (2D-MIKE21) based on synchronously optimized boundary conditions was used to simulate a tidal model of the Haizhou Bay of the Yellow Sea in China. The water depth with higher resolution and accuracy than the charted depth was used as the seabed topography. The DWLOB was calculated from 13 tidal constituents (including the two long-period constituents, Sa and Ssa) of the regional tidal model of China’s seas, CST1. The calculation of the BFC takes into account the spatial variation of water depth. For validation, we compared the simulated water level heights in this article and the tidal forecasts from the CST1 model with the 1-month observations at three tide gauges in Haizhou Bay and six points altimetry data of Jason-1, Jason-2, and Jason-3 satellites at the same time in 2007, which passed the modeling domain, and the former had a smaller root-mean square (RMS) than the latter. We compared the 13 dominant tidal constituents, which were from the 1-year simulated water level heights in this article by using tidal harmonic analysis and the one from the CST1 models, with that from the altimetry data of Jason-1, Jason-2, and Jason-3 satellites between 2002 and 2019. The root sum square (RSS) of the former was 8.16 cm, and that of the latter was 8.94 cm.