provenance: Statistical Toolbox for Sedimentary Provenance Analysis

Bundles a number of established statistical methods to facilitate the visual interpretation of large datasets in sedimentary geology. Includes functionality for adaptive kernel density estimation, multidimensional scaling, generalised procrustes analysis and individual differences scaling using a variety of dissimilarity measures. Univariate provenance proxies, such as single-grain ages or (isotopic) compositions are compared with the Kolmogorov-Smirnov dissimilarity and Sircombe-Hazelton L2-norm. Categorical provenance proxies, such as mineralogical, petrographic or chemical compositions are compared with the Aitchison and Bray-Curtis distances. Also included are tools to plot compositional data on ternary diagrams, to calculate the sample size required for specified levels of statistical precision, and to assess the effects of hydraulic sorting on detrital compositions.

Version: 0.6
Depends: R (≥ 3.0.2)
Imports: MASS, smacof, shapes, methods
Published: 2015-09-22
Author: Pieter Vermeesch [aut, cre]
Maintainer: Pieter Vermeesch <p.vermeesch at>
License: GPL-2
NeedsCompilation: no
CRAN checks: provenance results


Reference manual: provenance.pdf
Package source: provenance_0.6.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
OS X Snow Leopard binaries: r-release: provenance_0.3.tgz, r-oldrel: not available
OS X Mavericks binaries: r-release: provenance_0.6.tgz
Old sources: provenance archive