Advertisement

Models for Studying the Occurrence of Lead and Zinc in a Deltaic Environment

  • Richard A. Reyment
Part of the Computer Applications in the Earth Sciences book series (CAES)

Abstract

The relationships between the Pb and Zn content of sediments of the Niger Delta, West Africa, and Eh, organic content of the sediments, content of carbonate shell substance, Mn, P, S, and depth of origin of sample have been studied by the multivatiate statistical method of canonical correlation. The first model weighs the “predictor” variables of organic content and depth, in negative association, against a “response” vector dominated by P, negatively associated with Pb, Mn and S. The correlation of this vector with the original variables brings out the underlying relationship (Mn, -P, 2S). A second statistically significant, and almost equally important relationship for these data present the correlation between organic content and depth, in positive association, bound to a response vector dominated by Pb, Mn and P. The underlying relationship is shown to be between all elements in relation to depth and content of organic substance. The distribution of Pb and Zn is described by this vectorial association. The predictor and response structures of this root seem to express the distribution of the organic sedimentary component and the chemical constituents. A third, although nonsignificant, interrelationship weighs shells, with their soft parts, against Pb, negatively associated with P and S. Eh is of no account in the first two models but plays a certain role for the third root. A fourth set of non-significant canonical variates may represent random variation in the Eh determinations. Although the bivariate correlation coefficient between Pb and Zn is high, there is a difference in part of their patterns of distribution in our material.

Keywords

Organic Content Bivariate Correlation Canonical Correlation Canonical Correlation Analysis Canonical Variate 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Blackith, R. A., and Reyment, R. A., 1971, Multivariate morphometries: Academic Press, London, 412 p.Google Scholar
  2. Degens, E. T., 1968, Geochemie der Sedimente: Ferdinand Enke Verlag, Stuttgart, 282 p.Google Scholar
  3. Lee, P. J., 1969a, The theory and application of canonical trend surfaces: Jour. Geology, v. 77, no. 3, p. 303–318.CrossRefGoogle Scholar
  4. Lee, P. J., 1969b, FORTRAN IV programs for canonical correlation and canonical trend-surface analysis: Kansas Geol. Survey Computer Contr. 32, 46p.Google Scholar
  5. Lee, P. J., and Middleton, G. V., 1967, Application of canonical correlation to trend analysis, in Computer applications in the earth sciences: Colloquium on trend analysis: Kansas Geol. Survey Computer Contr. 12, p. 19–21.Google Scholar
  6. Love, W. A., and Stewart, D. K., 1968, Interpreting canonical correlations: theory and practice: Am- Inst. Res., Palo Alto, 75 p.Google Scholar
  7. Reyment, R. A., 1969, Interstitial ecology of the Niger Delta: Bull. Geol. Inst. Univ. Upsala, v. 1, p. 121–159.Google Scholar
  8. Reyment, R. A., and Ramden, H. A., 1970, FORTRAN IV program for canonical variates analysis for the CDC 3600 computer: Kansas Geol. Survey Computer Contr. 47, 40 p.Google Scholar

Copyright information

© Plenum Press, New York 1972

Authors and Affiliations

  • Richard A. Reyment
    • 1
  1. 1.Uppsala UniversitetSweden

Personalised recommendations