User Tools

Site Tools


Section: Ordination analysis

Variation partitioning (constrained ordination)

R functions

  • varpart (library vegan) - variation partitioning (using RDA, CCA or db-RDA) among up to four matrices of environmental variables. First argument (Y) is dependent variable, usually the matrix of species composition (the function calculates RDA, or, if chisquare = TRUE, CCA), but could be also only a single variable (in that case it calculates linear regression) or distance matrix (applying db-RDA using the function capscale). Further arguments (up to four) are (groups of) explanatory variables. The function uses either formula interface (with ~, see examples) or matrices. The interpretation should be based on adjusted R2, although raw R2 is also reported (for CCA, adjusted R2 is calculated by permutation method of Peres-Neto et al. 2006 and may slightly vary between re-analyses of the same data; the argument permutations specifies the number of permutations used to calculate adjusted R2 in CCA).
  • plot.varpart (library vegan) - draws Venn's diagram with fractions of explained variation. In default setting it doesn't show negative values of explained variation (argument cutoff = 0). The function can use arguments of showvarparts below, e.g. to add the labels for individual (groups of) variables (Xnames), or colors of the fractions (bg). Consult ?plot.varpart for more details.
  • showvarparts (library vegan) - draws schema of Venn's diagram with codes of individual fractions.
en/varpart_r.txt · Last modified: 2019/04/06 18:03 by David Zelený