Estimation of sampling bias between different sampling protocols on contaminated land

摘要: A method has been devised for the estimation of sampling bias between different protocols on contaminated land. pilot study was conducted at one site land near Wirksworth in Derbyshire. Four schemes (simple regular grid taking single auger samples, five-fold composite herringbone and stratified random) were applied to sample topsoil. The samples decomposed with nitric perchloric acids analysed using ICP-AES 24 elements. interpretation results focused two elements contrasting properties: lead, which shows high geochemical variability across field copper, very low variation. Although mean lead concentration vary markedly (e.g., 5680–6868 µg g–1) differences are not statistically significant owing quantified as variance. For copper much lower variance a 2.45 g–1 detected designs (i.e., simple random design). methodology could be more demanding studies such interlaboratory trials estimating organizations this site.