文章基本信息

标题：ON THE RELIABILITY OF DATA OBTAINED BY KRIGING
本地全文：下载
作者：Burkhard Schaffrin
期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2000
卷号：XXXIII Part B4 (/1-3)
页码：893-900
出版社：Copernicus Publications
摘要：Irregularly distributed data may be interpreted as a sample drawn from an underlying spatial process whose p roperties can be assumed to be known or unknown, depending on the particular situation. In case of a relatively smooth pro cess, one of the various Kriging methods co uld be employed to derive gridded point data with comparable accuracy provided that no outliers are present. Otherwise, they have to be eliminated beforehand or, at least, their influence must be reduced to the level of random uncertainty. Measures of reliability to describe the potential for identifying outliers in suspicious sample points and to quantify the effect of any undetected outliers – well-known for the Gauss-Markov Model – will be introduced for the case of a spatial process where the sampled data are suppo sedly correlated, at least in the spatial sense. In this study, we shall consider Simple as well as Ordinary Kriging which is essentially identical to " least-squares collocation" with (known, resp. unknown) constant trend
关键词：Outlier testing; reliability; Kriging; spatial d ata