敦煌戈壁气象塔站资料的质量控制  

A Quality Control Routine For Dunhuang Gobi Meteorology Tower Data

作　　者：王超[1] 韦志刚[1] 李振朝[1] 

机构地区：[1]中国科学院寒区旱区环境与工程研究所,甘肃兰州730000

出　　处：《干旱气象》2010年第2期121-127,共7页Journal of Arid Meteorology

基　　金：国家重点基础研究发展计划资助(2009CB421402);国家自然科学基金项目(40730952);中国科学院知识创新工程重要方向项目KZCX2-YW-220资助

摘　　要：讨论了气象观测资料的质量问题,并分析和强调了塔站资料质量控制的重要性,结合地面观测资料的质量控制方法,提出了逻辑极值检查、僵值检查、时间一致性检查、相似一致性检查、决策算法、质量控制码的标注和人工干预检查等组成的一套针对塔站资料的质控方法。利用提出的质量控制流程,对敦煌戈壁塔站2001年4月至2009年4月共9a的资料进行质量控制。结果表明:该流程能很好的找出缺测、错误、可疑的数据并给与标注,特别是在处理过程中,结合了综合判别法,使得该检验结果更为可靠。结果表明:塔站观测资料的质量是比较好的,正常数据占总数据的91.2%,非正常数据只是少数情况,其中大部分为缺测数据,占其78.7%。错误数据占总数据的1.93%,其中大部分为僵值数据,占其87.02%,其次分别为一致性和逻辑极值错误。This paper discussed the problem of quality control of meteorological observation data and emphasized its importance. Integrated with the ground quality control,a series of quality control methods were advanced to deal with observation data,such as logical extremum check,static check,time consistency check,similarity check,synthetic method,quality code and manual interference check. Especially in the treatment process,using the logical extremum check and synthetic method can make the test results more reliable. And the quality control flow method was applied on data from Dunhuang meteorology tower from April 2001 to April 2009. The results demonstrated that the method could differentiate the missing,error,dubious data and mark it. The tower data was reliable with normal data accounting for 91.2% of the total data. Most of the abnormal data were missing observations accounting for 76.8%; error data accounted for 1.93% of the total data. Most of error data are static data,accounting for 87.02%,the consistency error and logical extreme errors coming next.

关 键 词：敦煌 气象塔 质量控制 质量监视评估 

分 类 号：P413.2[天文地球—大气科学及气象学]