Abstract:To evaluate the performance of the DSAEF_LTP(dynamical–Statistical–Analog ensemble forecast model) during the impact of Super Typhoon Doksuri (2305) on the Fujian region, this paper conducted conventional and spatial verification of the typhoon's process precipitation forecasted by DSAEF_LTP, and compare it with the results of the Precipitation forecasts of the European Center for Medium-Range Weather Forecasting (hereafter referred to as ECMWF), and the Revised Precipitation Product of the Fujian Meteorological Bureau OTS (hereafter referred to as FZECMOS). The main findings are summarized as follows: The main findings are as follows: The DSAEF_LTP model's predictions for coastal intense precipitation distributions and the northeastern intense precipitation center in Fujian closely align with actual observations. The T_S values of the DSAEF_LTP model show significant improvement at the ≥100 mm and ≥250 mm thresholds compared to ECMWF and FZECMOS.However,the DSAEF_LTP model has shortcomings such as a significantly small forecast area for very heavy rainfall.At the ≥100 mm and ≥250 mm scale, verification results of MODE reveals that the DSAEF_LTP model outperforms ECMWF and FZECMOS in overall similarity, particularly in forecasting intense precipitation in isolated small areas. As the magnitude of precipitation increases, the overlap area between the DSAEF_LTP model forecast products and the observation also increases, indicating the model's effectiveness in forecasting extreme precipitation. The DSAEF_LTP model demonstrates its ability to flexibly adjust and screen historical similar TCs, allowing for the reasonable retention of similar TCs and their associated precipitation distributions, which enhances the ensemble forecasting effect.