You can use a stage vs. time hydrograph that reflects the local tidal cycle at the downstream (ocean) end of the reach. As the tidal variation observation from gages usually depicts the WSEL for a large area, suggest that your downstream-most cross section be large enough that the velocity head is negligible. As usual this downstream boundary condition should be located far enough downstream of the area of interest that uncertainties in the values have no impact on the solution in study area. This for an ocean boundary a large cross section beyond the stream mouth. The upstream boundary condition in most situations should be a discharge hydrogaph. As the tidal cycle repeats continuously on a daily basis, one should warm up the model for several tidal cycles to achieve a repeating solution. The time step needs to be small enough to resolve the boundary conditions and conform to stability requirements.
Michael