Abstract:
Based on the local classified average bias-adjusted algorithm, the Classified Quantitative Precipitation Estimation (CQPE) of six regional rainstorm occurred in Chongqing are calculated by using the data of SWAN-MCR and the surface rain gauges. On the basis of site inspection of CQPE and SWAN Quantitative Precipitation Estimation (SWAN-QPE). the precipitation forecast products of Chong Qing Storm-Scale Rapid Assimilation and Forcast System (CQSSRAFS) were tested and evaluated by using the surface rainfall data and CQPE. The main conclusions are as follows: (1) CQPE is slightly weaker and SWAN-QPE is stronger. From the evaluation test parameters, including ratio bias (BIAS), mean absolute error (MAE), relative error (RAE), root mean square error (RMSE) and correlation coefficient (CC), CQPE is superior to SWAN-QPE. (2) The verification results of CQSSRAFS precipitation forecast are consistent with the measured precipitation at the rainfall station and CQPE. In terms of effectiveness, the performance of CQSSRAFS precipitation forecast weakens with the increase of forecast time, especially after the 9th hour. At the precipitation level, the prediction ability of CQSSRAFS decreases with the increase of precipitation level. Compared with the “point-to-face” test of CQSSRAFS forecast precipitation by the measured precipitation of the rainfall station, the overall sample test of CQPE on its "face-to-face" is more representative.