Statistical study of trace elements in the soluble organic and amorphous Fe-Mn phases of surficial sediments, Sudbury Basin. 1. Multivariate and spatial analysis

摘要: Abstract Samples of humus and B-and C-horizon till collected from the north east margins Sudbury Basin were analyzed for about 30 trace elements. The data consist two sets analyses, one being a conventional “near total” analysis following an aqua regia digestion, other “partia” selective leach designed to extract metals bound soluble organic fraction amorphous Fe Mn oxides B- till. Statistical processing, using both standard approaches some newly developed methods spatial analysis, applied evaluate geochemical patterns their relationship geological factors such as proximity known mineral occurrences. In particular, information content partial analyses was evaluated determine whether reveal interpretable not detected in analyses. From visual appraisal dotplots Ni Cu till, it is difficult add anything already seen “total” data. However, when residuals obtained by removing that part variation explained are mapped, distinctive produced. These enhanced applying adaptive filtering technique U-statistic. Multielement groups (“total” data, elements till) determined principal components show first due Ni-Co-Cr-Cu-Zn Pd-Au-Pt-Sb associations. A multifractal area-concentration model fitted component scores allows multielement anomalies be mapped. shown spatially associated with Ni-Cu deposits, second Au Ni-Cu-Au deposits. Analysis variance, “total”, residual taken turn dependent (“response”) variable, shows significantly “explained” (underlying rock type, contacts mineralization). Neither nor on own significant associations these either element This demonstrates value where they carried out addition this dataset. multivariate variance if all three surfical media (humus, used, extraction so advantageous, although still adding predictive capability, because additional “explanatory” comes multimedia rather than multiphase sources.