In this study, the dependence of typhoon-induced storm surge and wave setup effect on typhoon intensity and size in the South China Sea are examined using the coupled ADCIRC + SWAN model. The model is first verified based on in situ observations in the Pearl River Estuary during the passage of super-typhoon Hato (2017). The model output is in good agreement with the observations, with relative errors and correlation coefficients of 4.97% and 0.80 (wind speed), 1.88% and 0.89 (storm surge), and 5.89% and 0.94 (significant wave height, SWH). The coupled model results show that the wave setup has a nonnegligible effect on storm surge simulation. Two sets of numerical experiments are designed to evaluate the variations in wave setup contribution to storm surges upon different typhoon intensities and sizes. With the increase of typhoon intensity or size, both maximum values and extended spatial ranges of the simulated SWH and storm surge increase significantly. In particular, the contribution of wave setup to the maximum storm surge increases nonlinearly with increasing typhoon intensity and size. For typhoon intensities ranging from 25 to 70 m s−1, the wave contribution to the storm surge increases from 0.02 to 0.17 m (1.96–6.66%). For typhoons with sizes from 10 to 60 km, the wave contribution to storm surge increases from 0.05 to 0.13 m (4.89–6.53%). However, when typhoon intensity is above 55 m s−1 or typhoon size is above 40 km, the percentage of wave setup contribution to the total storm surge tends to be saturated robustly. This study underlines the importance of incorporating wave setup effects into simulations of typhoon-induced storm surges, which are very important in mitigating typhoon disasters and designing criteria related to societal infrastructure and coastal engineering.