基于GEOS-Chem V12.6.3的全球CO_(2)浓度同化系统的构建  

Construction of a global CO_(2)concentration assimilation system based on GEOS-Chem V12.6.3

作　　者：霍霄 王茂华 张钦伟 魏崇 黄永健 顾倩荣 HUO Xiao;WANG Maohua;ZHANG Qinwei;WEI Chong;HUANG Yongjian;GU Qianrong(Shanghai Carbon Data Research Center,Shanghai Advanced Research Institute,Chinese Academy of Sciences,Shanghai 201210,China;University of Chinese Academy of Sciences,Beijing 100049,China;Key Laboratory of Low-Carbon Conversion Science&Engineering,Shanghai Advanced Research Institute,Chinese Academy of Sciences,Shanghai 201210,China;Center for Excellence in Brain Science and Intelligence Technology,Chinese Academy of Sciences,Shanghai 200031,China)

机构地区：[1]中国科学院上海高等研究院碳数据与碳评估研究中心,上海201210 [2]中国科学院大学,北京100049 [3]中国科学院上海高等研究院低碳转化科学与工程重点实验室,上海201210 [4]中国科学院脑科学与智能技术卓越创新中心,上海200031

出　　处：《西安理工大学学报》2023年第2期184-197,共14页Journal of Xi'an University of Technology

基　　金：中国科学院重点部署资助项目(ZDRW-ZS-2019-1);国际碳卫星观测数据分析合作计划资助(131211KYSB20180002)。

摘　　要：为了研究同化OCO-2卫星柱浓度(XCO_(2))数据对于全球CO_(2)浓度模拟的影响,本文基于GEOS-Chem V12.6.3,采用四维变分(four dimensional variational,4D-Var)的方法,构建了一个同化OCO-2卫星XCO_(2)数据的全球大气CO_(2)浓度同化系统。首先,采用有限差分法验证了观测算子、积云对流、行星边界层和平流4个伴随模块计算结果的正确性。然后,以2018年为例,设计了模拟和同化两个实验,并利用TCCON、地面和航飞3种独立的观测数据进行对比验证。结果显示,同化实验结果与TCCON、地面和航飞观测数据之间的平均误差分别为0.37 mL/m^(3)、0.41 mL/m^(3)和0.51 mL/m^(3),相比于模拟实验,分别改善了40.32%、42.25%和45.15%,表明了同化OCO-2卫星的XCO_(2)数据能显著提高对全球大气CO_(2)浓度模拟的准确性。In order to study the influence of the assimilation of OCO-2 satellite column-averaged dry air mole fractions of CO_(2)(XCO_(2))data on global CO_(2)simulation,the study,based on GEOS-Chem V12.6.3,adopts the four dimensional variational(4D-Var)method to develop a global atmospheric CO_(2)concentration assimilation system that assimilates OCO-2 satellite XCO_(2)data.First,the finite difference method is used to validate the correctness of the calculation results of the observation operator,the cloud convection,the planetary boundary layer,and the advection modules.Then,two experiments on simulation and assimilation are designed taking the year 2018 as an example,with three observation data of TCCON,ground and aircraft used for comparison and validation.The results show that the mean errors between the assimilation experiment results and independent TCCON,ground and aircraft observations are 0.37 mL/m^(3),0.41 mL/m^(3),and 0.51 mL/m^(3),respectively,which are improved by 40.32%,42.25%,and 45.15%compared with the simulation experiment results.It shows that the assimilation system could significantly improve the accuracy of the global atmospheric CO_(2)concentration estimation.

关 键 词：CO_(2)浓度 同化系统 4D-VAR OCO-2 GEOS-CHEM 

分 类 号：P402[天文地球—大气物理学与大气环境]