Geoscience Reference
In-Depth Information
9 Multivariate Statistics
Miocene sedimentary rocks near the village
of Nachola, to the east of the Suguta Valley in
northern Kenya. Unmixing the composition
of such deposits using multivariate statistical
methods helps to determine the provenance
of the sediments and their temporal variations,
and hence changes in the relief, the drainage
network and the rainfall patterns.
9.1 Introduction
Multivariate analysis aims to understand and describe the relationships
between an arbitrary number of variables. Earth scientists ot en deal
with multivariate data sets such as microfossil assemblages, geochemical
i ngerprints of volcanic ash layers, or the clay mineral content of sedimentary
sequences. Such multivariate data sets consist of measurements of
p
variables
on
n
objects, which are usually stored in
n
-by-
p
arrays:
h e rows of the array represent the
n
objects and the columns represent the
p
variables. h e characteristics of the 2nd object in the suite of samples are
described by the vector in the second row of the data array:
