2021年夏季四川小时降水的不同模式预报对比分析

Comparative Analysis of Hourly Precipitation Forecasts from Different Models over Sichuan during Summer 2021

  • 摘要: 将2021年6—8月国家气象信息中心CMPAS融合降水实况分析产品作为实况降水,从降水量、降水频率、降水强度、峰值时间和降水日变化等方面,分析西南区域数值天气预报业务模式SWC-3km产品、中国气象局GRAPES模式3 km产品和欧洲中心中期数值预报模式产品对2021年夏季四川小时尺度降水的预报偏差特征。结果表明:SWC-3km模式对四川盆地夏季降水量的预报量级偏大较多,降水强度的预报大值区域与CMPAS较为符合,峰值时间的预报能够确定降水是在午后—夜间还是夜间—清晨,但日变化振幅与CMPAS的结果存在较大差异;GRAPES-3km模式对四川盆地夏季降水量的预报在宜宾、泸州均存在偏差,其他地区则较为一致;EC模式能较好地预报出四川盆地夏季降水的降水量,其中降水分布和大值中心、降水频率、降水强度的预报和CMPAS较为一致。

     

    Abstract: Using the China Meteorological Administration Multi-source Precipitation Analysis System(CMPAS) merged precipitation product from the National Meteorological Information Centre from June to August 2021 as the observed precipitation, the forecast deviation characteristics of three numerical models, including the Southwest Regional Numerical Weather Prediction Operational Model (SWC-3km), the GRAPES 3km model, and the European Centre for Medium-Range Weather Forecasts (ECMWF) model poduct, in predicting hourly-scale precipitation over Sichuan during summer 2021 are analyed. The analysis covered aspects such as precipitation amount, precipitation frequency, precipitation intensity, peak time, and diurnal variation. The results showed that the SWC-3km model significantly overestimated the summer precipitation amount in the Sichuan Basin, however, its forecast of high-intensity precipitation areas aligned well with CMPAS. The model could determine whether peak precipitation occurred in the afternoon-night or night-early morning period, the diurnal variation amplitude was quite different from the results of CMPAS. The GRAPES-3km model exhibited precipitation forecast biases in Yibin and Luzhou, while performing relatively consistently in other regions of the Sichuan Basin. The ECMWF model performed well in predicting summer precipitation in the Sichuan Basin, with its forecasts of precipitation distribution, high-value centers, precipitation frequency, and intensity showing good agreement with CMPAS.

     

/

返回文章
返回