[Objective] The mountain flood runoff on the west bank of Dianchi Lake was simulated and calculated in order to provide a scientific basis for constructing flood interception facilities, reducing mountain flood disasters, and protecting the water environment of Dianchi Lake.[Methods] We used the SCS-CN runoff model with 33 field-measured rainfall-runoff datasets for the mountainous area on the west bank of Dianchi Lake in Kunming City in 2019 and 2020. The curve number (CN) value was optimized and set according to the slope combined with previous rainfall. The initial abstraction coefficient (λ) was optimized by the exhaust method. Simulation accuracy and parameter applicability for the optimized SCS-CN model were verified by 19 field-measured rainfall-runoff datasets in 2021.[Results] ① The CN value obtained by the slope and previous rainfall classification optimization was only suitable for small and medium rainfall amounts (<30 mm). For heavy rainfall (≥ 30 mm), the antecedent moisture condition (AMC) in the standard SCS-CN model should be modified by linear interpolation according to the total rainfall amount in the previous five days, and then the corresponding slope CN correction value should be determined. ② The optimal λ values for small and medium rainfall events and for heavy rainfall events in the mountainous area on the west bank of Dianchi Lake were 0.15 and 0.20, respectively. ③ The NSE values after model verification for small and medium rainfall and for heavy rainfall were 0.852 2 and 0.797 8, respectively, and the model accuracy rates were 93.33% and 75%, respectively.[Conclusion] The optimized SCS-CN model was considered to be feasible for calculating runoff under the conditions of small and medium rainfall (<30 mm) and heavy rainfall (≥ 30 mm) in the mountainous area on the west bank of Dianchi Lake, and therefore can provide a scientific basis and theoretical reference for the prediction of surface runoff and the further optimization of the SCS-CN model in this area.