Abstract:Warm-sector rainstorm events over the middle-lower reaches of Yangtze River (MLRYZ) usually have large prediction uncertainties due to that they are easily affected by the complex underlying surface, especially in the convection initiation stage. In this paper, high-resolution numerical simulation and convection-permitting ensemble simulations are carried out for a warm-sector rainstorm over MLRYZ on June 23, 2020 that was affected by the complex terrain to investigate the trigger mechanism and reveal the limited predictability of this event. The Lagrangian backward trajectories analysis of air parcels, sensitivity experiments of removing terrain and closing thermal effect, and ensemble sensitivity analysis are used to analyze the convection initiation stage. Results show that the lifted air parcels mainly come from planet boundary layer below 1.5 km. The valley wind driven by thermal effect of Xianxia and Shan Mountains is the dominant dynamic source that triggered local convergence and lifting. The divergence of high and low levels, the vertical configuration of moist potential vorticity and dipole potential vorticity anomaly have a good indication of the convection initiation. In addition, this event is highly sensitive to the 2 m temperature and apparent heat source at low level, indicating the importance of accurate underlying forcing to the warm-sector rainstorm prediction. The initial condition sensitivity experiments which reduce the initial errors gradually suggest that the predictability of the warm-sector convective event is significantly lower than that of the frontal event occurred in the north. The RMDTE (Root Mean Difference Total Energy) of frontal convection can be decreased continuously with the reduction of the initial errors, while the RMDTE curves of warm-sector convection still reach original level, showing a nonlinear convergence characteristic. Therefore, for the frontal convection with strong synoptic forcing, data assimilation technique may be prioritized to reduce the initial errors so as to further reduce the forecast errors. However, for the convection initiation process of warm-sector rainstorm under complex terrain, more attentions should be paid on ensemble forecasts to represent its uncertainty.