partvar: Variance partitioning accross nested scales

Description

Variance partitioning accross nested scales using a decomposition (varcomp function) of variance on restricted maximum likelihood (REML) method (lme function). See Messier et al. 2010 for more information.

Usage

partvar(traits, factors, printprogress = TRUE)

Arguments

traits

Matrix of traits with traits in column

factors

A matrix of factors with the first column corresponds to the higher level factor, the second row the second higher level factor and so on.

printprogress

Logical value; print progress during the calcul or not.

Value

An object of class "partvar" corresponding to a matrix of variance values with traits in rows and nested factors in column.

References

Messier, Julie, Brian J. McGill, et Martin J. Lechowicz. 2010. How do traits vary across ecological scales? A case for trait-based ecology: How do traits vary across ecological scales? Ecology Letters 13(7): 838-848. doi:10.1111/j.1461-0248.2010.01476.x.

Examples

Run this code

data(finch.ind)

	genus<-as.vector(unlist(strsplit(as.vector(sp.finch),"_")) 
	[seq(1,length(sp.finch)*2, by=2)])
	
	res.partvar.finch<-partvar(traits=traits.finch, 
	factors=cbind(sites=as.factor(as.vector(ind.plot.finch)), 
	species=as.factor(as.vector(sp.finch)), genus=as.factor(genus)))

	res.partvar.finch
	
	oldpar<-par()
	par(mfrow=c(2,2), mai=c(0.2,0.2,0.2,0.2))
	pie_partvar(res.partvar.finch, col=c("red", "green", "blue", "purple"))
	par(oldpar)

	bar_partvar(res.partvar.finch, col=c("red", "green", "blue", "purple"),
	leg=TRUE)