Calculate bootstrapped IndVal z scores

Share:

Description

This function implements resampling (with replacement) of the observed environmental gradient and site-by-taxon matrix, and then calls the function b.getivz to obtain bootstrapped scores.

Usage

1
2
tboot(bSeq, env, taxa, ivTot = ivTot, minSplt = minSplt, nPerm = nPerm,
  memory = memory, imax = imax)

Arguments

bSeq

An index used to determine the sequence number of the current bootstrap replicate.

env

An environmental gradient.

taxa

A site-by-taxon matrix of taxa counts at each sampling location.

ivTot

A logical indicating whether IndVal scores should be calculated using total relative abundance or the mean relative abundace originally proposed by Dufrene and Legendre (1997). The default is to pass on the argument from the original TITAN funtion call.

minSplt

The minimum bin size for partitioning along the environmental gradient. The default is to pass on the argument from the original TITAN funtion call.

nPerm

The number of replicates used by the permutation procedure (not to be confused with the number of bootstrap replicates). The default is to pass on the argument from the original TITAN funtion call.

memory

A logical indicating whether scratch files should be used to store temporary data in order to preserve RAM during bootstrapping of large data sets. The default is to pass on the argument from the original TITAN funtion call.

imax

A logical indicating whether taxon-specific change points should be determined by IndVal maxima or z-score maxima (as in Baker and King 2010). The default is to pass on the argument from the original TITAN funtion call.

Details

Four pieces of information are obtained from every taxon during each bootstrap replicate. If the argument 'imax' is TRUE, bootstrapped change points are identified based on IndVal maxima, whereas if 'imax' is FALSE z-score maxima are used instead. In addition to the IndVal or z score maxima, the value of the environmental gradient, the indicator direction, and the p value are also retained for that point.

In addition to the above metric matrix for each taxon (1), the z scores across all candidate chnage points are retained from each replicate (2), as well as the response direction (maxgrp) (3) and a sorted version of resampled environmental values (4). These four items are combined as a list object.

Value

A list of four elements:

  • bt.metrics A matrix with nrow equal to number of taxa where the first column is the bootstrapped IndVal or z score maximum, the second is the environmental value, the third is the indicator direction, and the fourth is the p value at that point.

  • ivzs Z scores for all taxa across candidate change points in the replicate sample

  • bsrti A sorted version of the bootstrapped environmental gradient

  • rspdr Response direction (1 or 2) for all taxa across candidate change points in the replicate sample

Author(s)

M. Baker and R. King

References

Baker, ME and RS King. 2010. A new method for detecting and interpreting biodiversity and ecological community thresholds. Methods in Ecology and Evolution 1(1): 25:37.

See Also

b.getivz, ivzsums