青藏高原及周边UTLS水汽时空特征的多源资料对比  

Comparison of upper troposphere and lower stratosphere water vapor spatial and temporal distribution between reanalysis data and MLS observations over the Tibetan Plateau

作　　者：唐南军 任荣彩[2] 吴国雄[2] TANG Nanjun;REN Rongcai;WU Guoxiong(Center for Data Assimilation Research and Applications,Nanjing University of Information Science&Technology,Nanjing 210044,China;State Key Laboratory of Numerical Modeling of Atmospheric Sciences and Geophysical Fluid Dynamics,Institute of Atmospheric Physics,Chinese Academy of Sciences,Beijing 100029,China)

机构地区：[1]南京信息工程大学资料同化研究与应用中心,江苏南京210044 [2]中国科学院大气物理研究所大气科学和地球流体力学数值模拟国家重点实验室,北京100029

出　　处：《大气科学学报》2020年第2期275-286,共12页Transactions of Atmospheric Sciences

基　　金：中国科学院战略性先导科技专项(A类)项目(XDA17010105);国家自然科学基金资助项目(91437105,91837311);中国科学院前沿科学重点研究项目(QYZDY-SSW-DQC018)。

摘　　要：利用Aura卫星微波临边观测仪(Microwave Limb Sounder,MLS)数据,评估了ERA-I、MERRA、JRA-55、CFSR和NCEP2等5套再分析资料的水汽数据在青藏高原及周边上对流层-下平流层(Upper Troposphere and Lower Stratosphere,UTLS)的质量,然后选取其中质量较好的两套水汽数据,分析它们对青藏高原及周边UTLS水汽的时空分布和演变的表征能力。结果表明,与MLS数据相比,5套再分析资料中在UTLS普遍偏湿,最大偏湿在上对流层215 hPa,约为165%,而在下平流层,ERA-I和MERRA与MLS的差异相对较小。总的来看,ERA-I和MERRA表征的水汽与MLS更为接近。进一步的对比表明,ERA-I和MERRA中青藏高原及周边水汽含量的时空分布与MLS较为接近,夏季能够表征青藏高原在纬向和经向上的水汽高值区,冬季能够表征对流层顶、西风急流中心附近的水汽梯度带,而且MERRA的结果要好于ERA-I。ERA-I、MERRA和MLS中青藏高原地区的水汽季节演变都表现为冬季1-2月水汽含量低,夏季7-8月水汽含量高,水汽的季节变化在200~300 hPa最大。MLS资料显示,在青藏高原地区对流层顶附近,存在随时间向上向极的水汽传输信号。相较而言,ERA-I对向上水汽传输信号的表征更好,而MERRA对下平流层(100 hPa)向极水汽传输信号的表征更好。Understanding the bias in reanalysis water vapor data is very conducive to studying the stratosphere and troposphere water vapor transport,particularly over the Tibetan Plateau( TP),which is an important water vapor source in the extratropical region where it is located.In this paper,the Aura MLS satellite observations taken over the TP are used to evaluate the upper troposphere and lower stratosphere( UTLS) water vapor data from the ERA-I,MERRA,JRA-55,CFSR and NCEP2 reanalysis data taken during the period of August 2004 to December2013.The results reveal that the water vapor data from ERA-I and MERRA are superior to the other reanalysis data.In addition,the water vapor spatial and temporal distributions in ERA-I and MERRA are further compared by MLS satellite observations.The contrast results showthat all of the reanalysis data used are wetter in terms of UTLS compared to the MLS satellite observations. Specifically,the ERA-I,MERRA,JRA-55 and CFSR reanalysis data overestimated the upper tropospheric( 215 hPa) water vapor content by up to 165% compared to the MLS observations,while the water vapor data from ERA-I and MERRA bore small differences with the MLS observations.Overall,the water vapor data from ERA-I and MERRA were closer to the MLS observations compared to the other reanalysis data. Further contrast results showthat the water vapor horizontal and vertical structures in ERA-I and MERRA are consistent with the MLS observations over the TP in winter and summer.In the MLS observations,the water vapor maximum in the latitude and longitude occurred over the TP in summer,and that there is a strong meridional gradient of water vapor near tropopause and westerly jet in winter.These observations were also made in ERA-I and MERRA,yet the spatial distribution is better in MERRA.The water vapor content over the TP exhibited significant seasonal changes,being lowest in January and February,and highest in July and August,while the seasonal variability was greatest at 200 ~ 300 hPa.The seasonal changes in ERA-I and ME

关 键 词：青藏高原 上对流层-下平流层 水汽 对比分析 

分 类 号：P426[天文地球—大气科学及气象学]