sample_to_population_partition {partitionBEFsp}

Calculate population-level partition

Description

takes a random but incomplete sample of species of size N from a larger community Q, and estiamtes population-level selection and complementarity effects

Usage

sample_to_population_partition(DRY, M, N = length(M), Q,
  smallQ_correction = TRUE, uncorrected_cov = FALSE, nboot = NA)

Arguments

DRY

change in relative yield, as calculated by the calculate_DRY function

M

monoculture biomass

N

number of species in the sample of the full community (i.e. the "sample") - defaults to length(M)

Q

total number of species in the full community (i.e. the "population")

smallQ_correction

tells whether to apply the correction for small Q, as shown in Eq. 3c in the main text - defaults to TRUE

uncorrected_cov

A character, which can be TRUE, FALSE, or COMP. Tells whether to use the standard sample-size corrected covariance function (FALSE), or

nboot

Number of bootstrap iterations to run for estimating confidence intervals for selection and complementarity effects. Defaults to NA - i.e. no bootstrapping. a covariance function that is not corrected for sample size (TRUE), or a "compromise" function that resembles the standard function for N < Q, and that resembles the non-corrected function for N ~ Q If TRUE, then SS + CS = YO - YE, sensu Loreau and Hector 2001 defaults to FALSE note - we do not recommend setting this to TRUE or "COMP", unless you require SS+CS=YO-YE

Value

a list with elements SS (the sample-level selection effect), CS (the sample-level complementarity effect), SP (the population-level selection effect), CP (the population-level complementarity effect), and confint, which is a list that includes summary data and the full bootstrapped for estimates of the confidence intervals (if nboot != NA)

Examples

# Please see package help file (?partitionBEFsp) for examples.