Abstract: The Advanced Research WRF (ARW) version 3.6 has been used to simulate a heavy rainfall event that has occurred on 5-6 October 2010 over Hainan Island. The impacts of three cumulus parameterization schemes (KF, BMJ, TiedTke) and four microphysics parameterization schemes (Lin et al, WSM5, WSM6, Thompson) on heavy rainfall simulations are studied by means of precipitation, wind, reflectivity, and cloud structure. The results indicate that WRF model is sensitive to all combinations of cumulus parameterization schemes and microphysics parameterization schemes; each of them impacts greatly on time and location of rainfall intensity through regulating the structure of temperature and humidity fields. The combinations with Thompson scheme, which can catch the characterization of moisture transport, thermal and dynamic conditions, are more sensitive to the amount of precipitation than other combinational schemes during the development of heavy rainfall. The precipitation can be effected through changing the developing height and amount of cloud water mixing ratio and rain water mixing ratio. The Thompson scheme coupled with the TiedTke scheme can simulate characteristics of the tropical rainstorm event well and results in the most close values to the observed data. The simulated maximum cloud water mixing ratio is at the position where the vertical velocity and reflectivity is also maximized. Additionally, the production of water vapor mixing ratio is independent of cumulus parameterization schemes and microphysics parameterization schemes.