Road Map for the Next Decade of Earth System Reanalysis in the United States
Road Map for the Next Decade of Earth System Reanalysis in the United States
复制标题
美国未来十年地球系统再分析路线图
DOI：
10.1175/bams-d-23-0011.1
复制
发表时间：
2023
影响因子：
8
通讯作者：
Slivinski, Laura
中科院分区：
文献类型：
--
作者：
Frolov, Sergey;Rousseaux, Cécile S.;Auligne, Tom;Dee, Dick;Gelaro, Ron;Heimbach, Patrick;Simpson, Isla;Slivinski, Laura
再分析将历史观测与现代地球系统模型（ESM）相结合，以生成地球系统在空间和时间上的完整历史。它是一个重要的基础设施，支持跨多个美国机构的关键任务活动（包括国家海洋和大气管理局（NOAA）、国家航空航天局（NASA）、国家科学基金会（NSF）、能源部（DOE）和国防部（DOD））、工业（包括能源、资源管理、农业、基础设施、保险、信息技术和金融）和学术界。特别是，再分析产品可以用作初始条件，以评估和校准NOAA，NASA和DoD以及研究机构所产生的环境预报，这些机构调查了亚季节到十年时间尺度的可预测性。再分析还可以提供过去天气状况、极端情况和趋势的基本气候记录，并可作为紧急措施的验证数据集。再分析可以量化对生计和商业至关重要的地球系统组成部分（如热、辐射、水、空气质量和碳）的储存和通量。最近，再分析数据集也被用于机器学习模型的训练，并作为地球系统新兴数字孪生模型的关键组成部分。
Reanalysis combines historical observations with modern Earth system models (ESMs) to generate a spatially and temporally complete history of the Earth system. It is an essential infrastructure that supports mission-critical activities across multiple US agencies (including the National Oceanic and Atmospheric Administration (NOAA), National Aeronautics and Space Administration (NASA), National Science Foundation (NSF), Department of Energy (DOE), and Department of Defense (DOD)), industry (including energy, resource management, agriculture, infrastructure, insurance, information technology, and finance), and academia. In particular, reanalysis products can be used as initial conditions to evaluate and calibrate environmental forecasts produced by NOAA, NASA, and DoD and research institutions that investigate predictability on sub-seasonal to decadal timescales. Reanalyses can also provide an essential climate record of past weather conditions, extremes and trends and can serve as verification datasets for ESMs. Reanalyses can quantify storage within and fluxes across the Earth system components essential to livelihood and commerce, such as, heat, radiation, water, air quality, and carbon. Most recently, reanalysis datasets are also used for training of the machine learning models and as critical ingredient of emerging digital twins of the Earth system.