fspa: Flexible shortest path adjustment

Description

Flexible shortest path adjustment is a heuristic ordination method attempting to adjust pattern to ecological situations. It erases long distances in the resemblance matrix and replaces these by the sum of intermediate steps. Subsequent ordination uses function pco.

Usage

1
2
3
4
5
6
7
fspa(veg, method, d.rev, n.groups, ...)
fspa2(veg,method,d.rev=0.5,n.groups=3)

## Default S3 method:
fspa(veg, method, d.rev, n.groups, ...)
## S3 method for class 'fspa'
plot(x,...,axes=c(1,2))

Arguments

veg

A data frame of vegetation releves (rows) by species (columns)

method

The method used for calculating distance as available in function vegdist of package vegan, for instance method = "bray".

d.rev

The percentage of distances revised, for instance 0.5 (50 percent, the default).

n.groups

This classifies the data points for illustrative purposes (uses Ward's method).

...

Variable axes=c(1,2) (default), the axes to be plotted

x

An object of class "fspa".

axes

A vector of length two, assessing the axes used for plotting. Default is c(1,2).

Value

An output list of class "fspa" with at least the following intems:

oldpoints

Ordination scores before adjustment

newpoints

Ordination scores after adjustment

symbols

The symbols used for classified plot

nline

The number of lines to be drawn in the graph

startline

Coordinates where the lines start

endline

Coordinates where the lines end

dmat.before

Distance matrix before analysis

dmat.after

Distance matrix after analysis

endline

Coordinates where the lines end

d.rev

proportion of distances revised

Author(s)

Otto Wildi

References

Bradfield, G.E. & Kenkel, N.C. 1987. Nonlinear ordination using flexible shortest path adjustment of ecological distances. Ecology 68: 750–753.

Wildi, O. 2013. Data Analysis in Vegetation Ecology. 2nd ed. Wiley-Blackwell, Chichester.

Examples

1
2
o.fspa<- fspa(sveg,method="euclid",d.rev=0.75,n.groups=6)  # sveg is vegetation data
plot(o.fspa,axes=c(1,2))                                   # plots axis 1, 2

Questions? Problems? Suggestions? or email at ian@mutexlabs.com.

All documentation is copyright its authors; we didn't write any of that.