大数据 ›› 2022, Vol. 8 ›› Issue (2): 89-102.doi: 10.11959/j.issn.2096-0271.2022017

• 专题:空天大数据 • 上一篇    下一篇

数值天气预报对卫星大数据的需求分析

杨何群1,2, 王晓峰1,3, 高彦青1,3, 陆一闻1, 麻炳欣1, 王昕瑶1,3   

  1. 1 上海市生态气象和卫星遥感中心,上海 200030
    2 复旦大学大气与海洋科学系,上海 200438
    3 区域高分辨率数值预报创新中心,上海 200030
  • 出版日期:2022-03-15 发布日期:2022-03-01
  • 作者简介:杨何群(1981- ),女,上海市生态气象和卫星遥感中心高级工程师,复旦大学大气与海洋科学系博士生,主要研究方向为卫星遥感与生态气象、土地覆盖分类、城市热环境
    王晓峰(1975- ),男,博士,上海市生态气象和卫星遥感中心研究员、高级工程师,主要研究方向为区域高分辨率数值天气预报、中小尺度天气分析、台风预报
    高彦青(1994- ),男,上海市生态气象和卫星遥感中心助理工程师,主要研究方向为数值模拟
    陆一闻(1991- ),男,上海市生态气象和卫星遥感中心工程师,主要研究方向为卫星遥感反演、图像识别、水环境遥感监测
    麻炳欣(1986- ),男,上海市生态气象和卫星遥感中心工程师,主要研究方向为卫星遥感下垫面识别与生态评估
    王昕瑶(1992- ),女,上海市生态气象和卫星遥感中心工程师,主要研究方向为数值模式研发及应用
  • 基金资助:
    国家自然科学基金资助项目(41801367);上海市“科技创新行动计划”长三角科技创新共同体领域项目(20232410100)

Analysis of satellite big data requirements in numerical weather prediction

Hequn YANG1,2, Xiaofeng WANG1,3, Yanqing GAO1,3, Yiwen LU1, Bingxin MA1, Xinyao WANG1,3   

  1. 1 Shanghai Ecological Forecasting and Remote Sensing Center, Shanghai 200030, China
    2 Department of Atmospheric and Oceanic Sciences, Fudan University, Shanghai 200438, China
    3 Shanghai Innovative Center of Regional High Resolution NWP, Shanghai 200030, China
  • Online:2022-03-15 Published:2022-03-01
  • Supported by:
    The National Natural Science Foundation of China(41801367);Shanghai Science and Technology Commission Project in Yangtze River Delta Innovation Community Field of “Science and Technology Innovation Action Plan”(20232410100)

摘要:

多星协同的对地观测可提供多谱段、多时相、多要素、多尺度、多层次的遥感数据,为数值天气预报提供丰富的价值信息。为了支撑未来地球系统无缝隙精细化网格预报服务,从探测变量、时间分辨率、空间覆盖度和水平分辨率、垂直分辨率、精度与时效等方面探讨了数值天气预报对卫星观测大数据需求的应用现状。同时,为了使卫星大数据高受容于数值天气预报,总结了多星数据一体化一致性处理、全天候/耦合的资料同化方法、与人工智能深度结合、卫星观测与预报互动等方面面临的挑战和前景。

关键词: 数值天气预报, 卫星, 大数据, 特征变量, 时空分辨率, 精度

Abstract:

Multi cooperative satellites can provide multi spectral, multi temporal, multi factor, multi scale and multi-level remote sensing data, which is rich in valuable information for numerical weather prediction (NWP).In order to support earth system seamless fine gridded forecasting service in the future, the application status of satellite observation big data was discussed for numerical weather prediction from the aspects of detection variables, time density, spatial coverage, horizontal and vertical resolution, as well as accuracy and timeliness.At the same time, in order to make satellite big data be highly tolerant with NWP, the challenges and prospects were summarized, such as multi-satellite integrated and consistent processing, all-weather, coupled data assimilation methods, deep integration with artificial intelligence, and interaction between satellite observation and prediction.

Key words: numerical weather prediction, satellite, big data, feature variable, spatiotemporal resolution, accuracy

中图分类号: 

No Suggested Reading articles found!