多源降水数据融合研究及应用进展

Advances in Multi-Source Precipitation Merging Research

查看全文

作者:

  • 潘旸 国家气象信息中心 北京 100081
  • 谷军霞 国家气象信息中心 北京 100081
  • 徐宾 国家气象信息中心 北京 100081
  • 沈艳 国家气象信息中心 北京 100081
  • 韩帅 国家气象信息中心 北京 100081
  • 师春香 国家气象信息中心 北京 100081

中文摘要:

 融合降水产品结合了不同来源降水资料的优势,在天气气候监测、气候变化研究、模式检验及水文预报领域得到了广泛应用。介绍了适用于不同需求的国际主流融合降水产品,从产品应用的角度介绍了融合技术发展。重点介绍了国家气象信息中心在降水数据融合方面取得的一些研究进展。如基于“PDF(概率密度函数)+BMA(贝叶斯模式平均)+OI(最优插值)”和降尺度技术研制的高分辨率的地面-雷达-卫星三源降水融合产品,采用红外冷云外推技术研制的东亚多卫星集成降水(EMSIP)产品,采用多重网格三维变分(STMAS)方法制作的1 km陆面数据同化分析(HRCLDASv1.0)的降水驱动,以及引进美国CFSR/Land降水驱动融合技术研制的全球融合降水产品等。其中,基于中国多源降水融合分析业务化系统(CMPAS-Hourly V2.1)的高分辨率三源融合降水产品在智能网格预报、GPAPES检验评估等业务中发挥积极作用。计划未来还将在融合算法优化、分钟级降水产品研制、机器学习新方法应用等方面开展研究工作。

中文关键词:

降水产品,偏差订正,融合技术,研究进展

KeyWords:

precipitation product, bias correction, merging technique, research advances

Abstract:

 The multi-source precipitation products take the advantages of every single source precipitation data, and they have been applied in many respects such as weather and climate monitoring, climate change research, model evaluation, and hydrological prediction. For those applications, there are many prevailing products. CMA/NMIC also developed many multisource precipitation merging techniques and products, such as the high resolution radar-satellite-gauge merged precipitations based on the “PDF (Probability Density Function) + BMA (Bayesian Model Averaging) + OI (Optimal Interpolation)” and downscaling technique, the East-Asia multi-satellite integrated precipitation (EMSIP) using IR moving vector extrapolating technique, the hourly/0.01° merged forcing data for the high-resolution CMA land data assimilation system (HRCLDASv1.0) using STMAS (Space Time Multiscale Analysis System) technique, and a global merged precipitation based on the advanced satellite-retrial precipitation and merging algorithm, in research of CRA-Land reanalysis. At present, the operational CMA multisource precipitation analysis system (CMPAS-Hourly V2.1) was constructed to produce the real-time high-resolution and highquality merged precipitation, and well applied in GRAPES model evaluation. In future, the merging algorithm optimization, the minute precipitation products development, and ML (Machine Learning) technique application will be conducted.

版权所有:《气象科技进展》编辑部

联系电话:(010)58993291 68409927

Email:qxkjjz@163.com

气政邮:《气象科技进展》编辑部/中国气象局图书馆(气象科技史研究中心)/干部学院/中国气象局
地址:北京市中关村南大街46号中国气象局气象干部陪训学院 邮编:100081