Abstract: The reflectivity and radial velocity data from multiple Doppler radars were assimilated by the Community Gridpoint Statistical Interpolation (GSI) system, and a tornado occurred in Funing on 23 June 2016 was then simulated and studied by using the Weather Research and Forecasting (WRF) model. The main results are as follows. (1) The simulations can be improved to some extent by assimilating either radar reflectivity or radial velocity. The experiment with both the radar reflectivity and radial velocity assimilated show the best result, and the radial velocity plays a more important role than the reflectivity. (2) Assimilation of radar reflectivity by using complex cloud analysis system can efficiently adjust the initial hydrometeors and thermal fields, which makes an important contribution to the adjustment of water vapor and thermal conditions. Assimilation of radar radial velocity via the three-dimensional variational data assimilation system directly affects the wind field and subsequently affects the water vapor transport, hydrometeors and instability, which makes an important contribution to the adjustment of dynamic and thermal conditions. Assimilation of both the radar reflectivity and the radial velocity can combine their advantages to obtain a best balance among dynamic, water vapor and thermal conditions. (3) With assimilation of both the radar reflectivity and the radial velocity, the storm with the vortex structure of tornado near Funing can be simulated by the model. The simulated storm occurrence, development, track, surface hourly maximum wind speed and precipitation agree well with observations, although certain gaps still exist between the simulation and observations with respect to the strength and structure of the vortex.