Quantitative comparison of large sets of geochronological data using multivariate analysis: A provenance study example from Australia

A variety of modem isotopic analytical techniques now allow the relatively rapid acquisition of large sets of geochronological data, particularly in p rovenance studies. Traditional interpretation of this type of data often re lies on visual comparison techniques that are vulnerable to subjective bias and this problem becomes acute with increasingly large quantities of data. Multivariate techniques are presented in this article to objectively evalu ate age components within the detrital zircon age data from 31 beach sand a nd sedimentary samples in Australia for provenance studies. Using the SHRIM P Ion Microprobe each sample typically contains 60 or more individual age m easurements with a combined total of 2150 analyses. Principal components de rived from the data set allow the construction of provenance models that ar e independent of expected age components. In this instance the principal co mponents are interpreted in terms of existing knowledge about Australian ge ology, in particular the provenance relationship between "local" Phanerozoi caged and "exotic" late Proterozoic-aged protosources. Using principal comp onents analysis this relationship is seen clearly and objectively in one di agram illustrating provenance evolution along the eastern coastline. Copyri ght (C) 2000 Elsevier Science Ltd.