双月刊

ISSN 1006-9895

CN 11-1768/O4

+高级检索 English
GRAPES区域集合预报对2019年中国汛期降水预报评估
作者:
作者单位:

1.中国气象局数值预报中心;2.浙江省气象科学研究所

作者简介:

通讯作者:

基金项目:

国家重点研发计划


Verification for GRAPES-REPS model precipitation forecasts over China during the flood season in 2019
Author:
Affiliation:

Numerical Weather Prediction Center of CMA

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
    摘要:

    2019年,数值预报中心开发了以GRAPES全球模式为驱动场,集合变换卡尔曼滤波为初值扰动方法,随机物理过程倾向项为模式扰动方法的10km水平分辨率GRAPES-REPS V3.0区域集合预报模式,并投入业务运行。基于该模式,作者开展了2019年7-9月夏季降水不确定性的集合预报实时试验,并从统计检验和个例分析角度,与GRAPES-REPS V2.0和ECMWF全球集合预报模式进行对比,由此对GRAPES-REPS V3.0区域集合预报模式的降水预报能力给予客观评价,并分析了引起中尺度强降水预报不确定性的物理机制,研究结论可为诊断集合预报模式及改进集合预报方法提供依据。结果表明:(1)GRAPES-REPS V3.0区域集合预报系统的降水ETS评分在所有预报时效和量级内均优于GRAPES-REPS V2.0区域集合预报模式,降水成员具有明显等同性,且概率预报技巧FSS评分较高,GRAPES-REPS V3.0区域集合预报模式降水预报效果全面优于GRAPES-REPS V2.0区域集合预报模式。(2)GRAPES-REPS V3.0区域集合预报的集合平均降水BIAS评分及小雨和暴雨ETS评分均明显优于ECMWF全球集合预报系统,降水概率预报与ECMWF降水概率具有一定可比性。(3)个例分析结果表明,不同集合预报模式通过刻画中尺度特征物理量不确定性来捕捉降水预报不确定性,初始时刻,GRAPES-REPS V3.0区域集合预报模式和ECMWF全球集合预报模式环流形势分布较为相似,随预报时效演变,GRAPES-REPS V3.0区域集合预报模式对中尺度动力、热力场捕捉更为准确,相应地对降水落区与量级预报较好,概率预报技巧较优。(4)与ECMWF全球集合预报模式相比,GRAPES区域集合预报模式集合成员能很好地预报降水发生、发展、消亡整个过程,故GRAPES-REPS V3.0区域集合预报系统对中国汛期降水具有较强的预报能力。

    Abstract:

    A regional EPS (Global and Regional Assimilation and Prediction Enhanced System-Regional Ensemble Prediction System (GRAPES-REPS V3.0) ) with a horizontal resolution of 10km has been developed and put into operation by the Numerical Weather Prediction Center of China Meteorological Administration in 2019, The background field of which comes from GRAPES global model, and initial perturbation and model perturbation method are Ensemble Transform Kalman Filter and Stochastic Perturbed Parameterization Tendencies respectively. The system was run in real-time for the 2019 summer season (July through September) and compared with GRAPES-REPS V2.0 and ECMWF global ensemble prediction system using statistical verification and case analysis methods, for an objective and comprehensive evaluation of precipitation forecast skill and forecast uncertainties for GRAPES-REPS V3.0 model, And We further analyze the physical mechanism leading to the forecast uncertainties of meso-scale intense precipitation, The results could provide basis for diagnosing regional ensemble prediction systems and developing ensemble forecast methods. The results are as follows: (1) GRAPES-REPS V3.0 model has a better precipitation ETS scores than GRAPES-REPS V2.0 model in terms of all the forecast lead times and rainfall classes with more equal rainfall members, And the probability forecast FSS scores are also better, So the precipitation forecast skills of GRAPES-REPS V3.0 model are totally better than those of GRAPES-REPS V2.0 model. (2) The ensemble mean precipitation BIAS and ETS scores of GRAPES-REPS V3.0 for light rain and rainstorm are better than ECMWF global ensemble forecast system, and the probability forecast skill of above two models is comparable. (3) The case studies show that, Different ensemble prediction systems capture precipitation forecast uncertainties by describing meso-scale physical quantities uncertainties, At initial lead time, The circulation patterns of GRAPES-REPS V3.0 regional ensemble prediction system and ECMWF global ensemble prediction system are similar, However, With the evolution of forecast lead time, GRAPES-REPS V3.0 ensemble prediction model could better describe meso-scale dynamic and thermal fields, with a more accurate location and magnitudes of rainfall and a better probabilistic forecast result. (4) Compared with ECMWF model, The ensemble members of GRAPES-REPS V3.0 model can well forecast the occurrence, development and extinction of rainfall processes, So GRAPES-REPS V3.0 model has shown a higher skill in forecasting the precipitation of Chinese flood season.

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2020-04-15
  • 最后修改日期:2020-06-15
  • 录用日期:2020-08-28
  • 在线发布日期:
  • 出版日期: