一种新的生态监测数据质量评估方法--以CERN乔木生长数据为例  

A new assessment method for the quality of ecological monitoring data:Taking CERN's tree growth dataset as a case

作　　者：颜绍馗[1] 吴冬秀[2] Singh AN 李媛良[1] 韦文珊[1] 崔杨[1] 汪思龙[1] 徐广标[1] 

机构地区：[1]中国科学院沈阳应用生态研究所会同森林生态实验站,沈阳110016 [2]中国科学院植物研究所植被与环境变化国家重点实验室/CERN生物分中心,北京100093 [3]Department of Botany, Panjab University, Chandigarh 160014, India

出　　处：《应用生态学报》2011年第4期1067-1074,共8页Chinese Journal of Applied Ecology

基　　金：中国科学院知识创新工程重要方向项目(KZCX2-YW-433-02)资助

摘　　要：将数据可靠性作为有序变量进行分级,在理论上使数据可靠性与主要生态过程、次级生态过程、外部过程等数据源建立关联,构建了一种生态监测数据质量评估方法,提供了一个新的数据质量指数.它通过观察记录的合格率来估计数据集的质量,其检测结果包括了每一条数据的可靠性级别、标记为离群或错误数据的原因,以及完整数据集的质量指数值.将该方法应用于CERN的两个乔木生长数据集,发现该数据质量指数可以定量评估乔木生长数据集的质量.该方法为相关软件的开发提供了基础.This paper presented a new and simple assessment method for the quality of ecological monitoring data.This method theorized the associations between the data reliability as an ordinal variable with different number of classes and the data sources such as natural main ecological processes,secondary ecological processes,and extraneous and exotic processes,and offered a new data quality index to estimate the quality of the whole dataset by using the reasonableness ratio of observations.The assessment results provided the reliability class of each dataset,good explanations for outlier（or error data） flagging decisions,and quality value of the whole dataset.The method was applied to assess two tree growth datasets from Chinese Ecosystem Research Network（CERN）,and the results demonstrated that the new data quality index could quantitatively evaluate the quality of the tree growth datasets.The new method would facilitate the development of corresponding software.

关 键 词：数据检测 信息系统 数据质量控制 离群数据 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]