利用冬季（2009年12月-2010年2月）和夏季（2010年6-8月）6 h一次T639和NCEP 2 m温度分析场作为背景场， 采用EOF（Empirical Orthogonal Function）方法对我国及周边地区地面观测2 m温度进行了质量控制研究。结果表明， 使用NCEP资料， 观测资料剔除率比T639高， 而资料剔除率高的区域正好是观测增量双权重标准差较大的区域； 冬季观测增量双权重标准差比夏季大， 导致冬季资料的剔除率比夏季高； 与传统OMB（Observation-Minus-Background）质量控制方法对比， 发现两种质量控制方法都能使观测增量更接近正态分布， 但使用传统OMB质量控制方法会把一些能反映大气异常状态且正确的资料剔除， 而EOF质量控制方法则能较好地保留这些资料。
Using six hourly 2 m temperature of T639 and NCEP analysis in winter (from December 2009 to February 2010) and summer (from June to August 2010) as background fields, by EOF (Empirical Orthogonal Function) quality control method, 2 m temperature in China and its neighboring regions was studied. The results show that the outliers identified with NCEP background field is more than that of T639 background field. Meanwhile, the areas with more outliers overlap are the areas where the biweight standard deviations of observation increment higher. The biweight standard deviations in winter are higher than in summer, so does the number of outliers. Besides, the comparising between traditional OMB (Observation-Minus-Background) and EOF quality control methods reveale that some correct observations reflecting unusual states of atmospheric are removed by traditional OMB quality control method, but are retained by EOF quality control method, even though both methods could make the distribution of observation increments more Gaussian-like.
[1]冯伍虎, 邱崇践. 变分四维同化方法若干问题的数值试验[J]. 高原气象, 1999, 18(2): 138-146.
[2]龚建东, 丑纪范. 论过去资料在数值天气预报中使用的理论和方法[J]. 高原气象, 1999, 18(3): 392-399.
[3]刘黎平, 钱永甫, 吴爱明. 初始条件和边界条件对区域模式模拟青藏高原区域气候的影响[J]. 高原气象, 1999, 18(1): 20-27.
[4]王可丽, 江灏, 陈世强. 青藏高原地区的总云量——地面观测、 卫星反演和同化资料的对比分析[J]. 高原气象, 2001, 20(3): 252-257.
[5]Gandin L S. Complex quality control of meteorological observations[J]. Mon Wea Rev, 1988, 116: 1137-1156.
[6]Collins W G. Complex quality control of significant level radiosonde temperatures[J]. J Atmos Ocean Technol, 1998, 15: 69-79.
[7]Collins W G. The operational complex quality control of radiosonde heights and temperatures at the national centers for environmental prediction. Part I: Description of the method[J]. J Appl Meteor, 2001a, 40: 137-151.
[8]Ingleby N B, Lorenc A C. Bayesian quality control using multivariate normal distributions[J]. Quart J Roy Meteor Soc, 1993, 119: 1195-1225.
[9]Andersson E, Jarvinen H. Variational quality control[J]. Quart J Roy Meteor Soc, 1999, 125: 697-722.
[10]王少影, 张宇, 吕世华, 等. 金塔绿洲湍流资料的质量控制研究[J]. 高原气象, 2009, 28(6): 1260-1273.
[11]李茂善, 杨耀先, 马耀明, 等. 纳木错（湖）地区湍流数据质量控制和湍流通量变化特征[J]. 高原气象, 2012, 31(4): 875-884.
[12]Wade C G. A quality control program for surface meteorological data[J]. J Atmos Ocean Tech, 1987, 4: 435-453.
[13]Feng S, Hu S Q, Qian W H. Quality control of daily meteorological data in China, 1951-2000: A new dataset[J]. Int J Climatol, 2004, 24: 853-870.
[14]任芝花, 熊安元. 地面自动站观测资料三级质量控制业务系统的研制[J]. 气象, 2007, 33(1): 19-24.
[15]陶士伟, 仲跻芹, 徐枝芳. 地面自动站资料质量控制方案及应用[J]. 高原气象, 2009, 28(5): 1202-1209.
[16]Zou X,Qin Z K. Time dependence of diurnal cycle errors in surface temperature analyses[J]. Mon Wea Rev, 2010, 138: 1137-1156.
[17]Qin Z K, Zou X, Li G, et al. Quality control of surface station temperature data with non-Gaussian observation-minus-background distributions[J]. J Geophys Res, 2010, 115, D16312, doi:10.1029/2009JD013695.
[18]Xu Z, Wang Y, FAN G. A two-stage quality control method for 2-m temperature observations using biweight means and a progressive EOF analysis[J]. Mon Wea Rev, 2012, doi: 10.1175/MWR-D-11-00308.1(in press).
[19]Lanzante J R. Resistant, robust and nonparametric techniques for the analysis of climate data: theory and examples, including applications to historical radiosonde station data[J]. Int J Climatol, 1996, 16: 1197-1126.
[20]Kalnay E. Atmospheric modeling, data assimilation and predictability[M]. Cambridge: Cambridge university press, 2003: 199.
