The use of multivariate statistics to resolve multiple contamination signals in the oxygen isotope analysis of biogenic silica

Wilson, Katy E.; Leng, Melanie J. ORCID: https://orcid.org/0000-0003-1115-5166; Mackay, Anson W.. 2014 The use of multivariate statistics to resolve multiple contamination signals in the oxygen isotope analysis of biogenic silica. Journal of Quaternary Science, 29 (7). 641-649. https://doi.org/10.1002/jqs.2729

Before downloading, please read NORA policies.

Preview

Text
Wilson et al_JQS.pdf - Accepted Version
Download (1MB) | Preview

Official URL: http://dx.doi.org/10.1002/jqs.2729

Abstract/Summary

Analysis of the oxygen isotope composition (δ18O) of diatom silica is a commonly used tool for palaeoclimate reconstruction that recent studies have demonstrated may be complicated by the presence of non-diatom detrital material. Such contamination can mask any true climate-driven signal, leading to spurious results. Analysis of the 2.6-Ma Barsemoi diatomites from the East African Rift Valley highlights the presence of both tephra and clay in purified samples. Here we present a new method for assessing the relative contribution and geochemical composition of contamination components where sedimentary samples may be affected by more than one type of contamination. This approach shows that the incorporation of analytical techniques, such as X-ray fluorescence spectrometry, coupled with statistical modelling, can be used to develop a three end-member model to successfully resolve climate-driven changes in δ18Odiatom. Mass-balance corrections made to δ18Odiatom data demonstrate the importance of adopting quantitative geochemical analysis in tandem with the δ18O analysis of biogenic silica, to obtain accurate and meaningful results for palaeoclimate reconstruction.

Item Type:	Publication - Article
Digital Object Identifier (DOI):	https://doi.org/10.1002/jqs.2729
ISSN:	02678179
Date made live:	14 Aug 2014 13:53 +0 (UTC)
URI:	https://nora.nerc.ac.uk/id/eprint/508104

Item Type:

Publication - Article

Digital Object Identifier (DOI):

https://doi.org/10.1002/jqs.2729

ISSN:

02678179

Date made live:

14 Aug 2014 13:53 +0 (UTC)

URI:

https://nora.nerc.ac.uk/id/eprint/508104