应用气象学报
應用氣象學報
응용기상학보
QUARTERLY JOURNAL OF APPLIED METEOROLOGY
2010年
1期
70-75
,共6页
漆成莉%刘辉%马刚%张鹏%吴雪宝
漆成莉%劉輝%馬剛%張鵬%吳雪寶
칠성리%류휘%마강%장붕%오설보
大气廓线%拓扑学%大气样本库%取样
大氣廓線%拓撲學%大氣樣本庫%取樣
대기곽선%탁복학%대기양본고%취양
atmospheric profiles%topology technique%sampled atmospheric database%sampling technique
在模拟大气辐射传输过程、发展大气透过率模式及反演大气参数等许多应用中,大气状态廓线是重要的输入参数,为方便使用,建立具有一定代表性的样本数据库十分必要.目前,国际上有几十种各具特色的大气样本库,但包含中国区域的样本极少.为了弥补国际大气样本库中缺少中国区域大气廓线的不足,以中国区域2002年气象探空资料为基础,利用拓扑学方法建立了包含30条大气温度和湿度廓线的中国区域大气廓线样本库2.通过对所选样本的覆盖区域、观测时间、海拔高度、廓线垂直分布等特征分析,结果表明:该样本库具有样本时空分布均匀、样本独立性和气候代表性好等特点,符合大气样本库建立的基本原则,可为辐射传输研究和业务应用提供大气样本支持.
在模擬大氣輻射傳輸過程、髮展大氣透過率模式及反縯大氣參數等許多應用中,大氣狀態廓線是重要的輸入參數,為方便使用,建立具有一定代錶性的樣本數據庫十分必要.目前,國際上有幾十種各具特色的大氣樣本庫,但包含中國區域的樣本極少.為瞭瀰補國際大氣樣本庫中缺少中國區域大氣廓線的不足,以中國區域2002年氣象探空資料為基礎,利用拓撲學方法建立瞭包含30條大氣溫度和濕度廓線的中國區域大氣廓線樣本庫2.通過對所選樣本的覆蓋區域、觀測時間、海拔高度、廓線垂直分佈等特徵分析,結果錶明:該樣本庫具有樣本時空分佈均勻、樣本獨立性和氣候代錶性好等特點,符閤大氣樣本庫建立的基本原則,可為輻射傳輸研究和業務應用提供大氣樣本支持.
재모의대기복사전수과정、발전대기투과솔모식급반연대기삼수등허다응용중,대기상태곽선시중요적수입삼수,위방편사용,건립구유일정대표성적양본수거고십분필요.목전,국제상유궤십충각구특색적대기양본고,단포함중국구역적양본겁소.위료미보국제대기양본고중결소중국구역대기곽선적불족,이중국구역2002년기상탐공자료위기출,이용탁복학방법건립료포함30조대기온도화습도곽선적중국구역대기곽선양본고2.통과대소선양본적복개구역、관측시간、해발고도、곽선수직분포등특정분석,결과표명:해양본고구유양본시공분포균균、양본독립성화기후대표성호등특점,부합대기양본고건립적기본원칙,가위복사전수연구화업무응용제공대기양본지지.
Atmospheric profiles are important input parameters in simulation of atmospheric radiative transfer,emulation of new satellite instruments spectral characteristics,development of transmittance model,retrieval of atmospheric parameters and the assimilation of satellite data in numerical weather prediction.It is very necessary that built a sampled atmospheric profiles database to meet many kinds of application requirements.The quantity,covering region,and source of atmospheric profiles affect the efficiency and results of application a lot.A major attempt to sample atmospheric profiles on a global scale has been the constitution of successive versions of the Thermodynamic Initial Guess Retrieval database(TIGR),despite the high dimensionality.Each version groups together hundreds of profiles sampled from mass soundings observations of the atmosphere.There are dozens of atmospheric profiles database currently,but profiles in China region are very deficient.An atmospheric profiles database with samples in China region can enhance the accuracy in calculation of radiative transfer and profiles retrieval.61308 profiles of year 2002 are collected as a preparatory database that includes 124 radiosond station measurement distributing in China region through Meteorological Data Storage System,and a database is set up comprising 30 atmospheric temperature and humidity profiles and named it CRASD-2(China Regional Atmospheric Sample Database 2) with topology method.It is classified by surface pressure and divided into 5 groups and their respective surface pressure are 500 hPa,700 hPa,850 hPa,925 hPa,1000 hPa. The CRASD-2 covers the four seasons of one year and the two radiosond observation time of 00:00 and 12:00,so that it has characteristics of well-proportioned time and season range.Considering the vertical height range it contains the samples located in stations with the altitude of several to several thousand meters.Besides,the profiles in this database distribute in large overcast geographical region of China,and the profiles with large discrepancy distribute in north-east,north-west and south region,while samples with relative less difference mostly are located in south of the Yangtze River and Yellow and Huai River region.The diversity between profiles mainly exist in the different height of tropopause,discrepancy in atmospheric temperature and humidity condition of the same pressure level,and the existence of temperature inversion layer.With the independency between profiles and good climate representative,the CRASD-2 database is accord with the principle of a sampled database,so it can support the application of atmosphere samples in research of radiative transfer and remote sensing retrieval.This elementary employment of topology sampling technique in extracting regional profiles database implies it may support sampling database from longer time-serial observation data,and adding more profiles variables in sampling method in the future work.