Abstract: To address the problems of insufficient data in mountainous basins, low model practicality, and inadequate accuracy in peak flow forecast, a runoff coefficient-based mountain flood forecast model was proposed by establishing the relationship between disaster-causing factors of mountain floods and the rainfall-runoff response mechanism, as well as the quantitative function between underlying surface parameters and runoff coefficients. Twelve representative small watersheds prone to mountain floods in the Haihe River, Yellow River, Huaihe River, and Changjiang River basins were selected as study areas to analyze model applicability. The results showed that the model achieved a qualified peak flow rate of over 60% in all basins during validation period, with an average peak time difference of less than 3 hours and a Nash efficiency coefficient above 0.6, which can accurately simulate the flood process. Moreover, its simple parameter calibration significantly improved the accuracy and applicability of the forecast. Particularly in humid regions, the model s performance was significantly better than in semi-humid and semi-arid regions, demonstrating a good applicability in areas with similar runoff generation and concentration mechanisms. The research indicates that the model has the advantages of being simple, easy to use, and effective, showing a broad application prospects and promotion value in mountain flood forecast work.