Mtabs: Structuring vegetation data tables

In dave: Functions for "Data Analysis in Vegetation Ecology"

Description

Mimics traditional manual ordering of vegetation data table by (i) clustering rows and columns (hclust), (ii) rearranging the resulting groups according to the first AOC axis (aocc), (iii) rearranging rows and columns inside groups based on CA (cca), (iv) Putting high resolving species on top of the table (aoc). Also offers variants for ordering.

Usage

Mtabs(veg, method = "raw", y.r, y.s, k.r, k.s, ndiffs, ...)
mtab(veg, method = "raw", y.r, y.s, k.r, k.s, ndiffs)
plottab(veg,rorder=NULL,sorder=NULL,grr=NULL,grs=NULL,y=0.5)
plottabl(veg,rorder=NULL,sorder=NULL,grr=NULL,grs=NULL,y=0.5)
setgroupsize(vec)

## Default S3 method:
Mtabs(veg, method, y.r, y.s, k.r, k.s, ndiffs, ...)
## S3 method for class 'Mtabs'
plot(x,...,method="normal")
## S3 method for class 'Mtabs'
summary(object,...,range=NULL)

Arguments

veg	This is a vegetation data frame, releves are rows, species columns
method	The method used for ordering: "raw", "sort", "ca", "clust", "aoc" or "mulva"
y.r	Transformation of species scores when clustering releves (rows): x'= x exp(y.r)
y.s	Transformation of species scores when clustering species (columns): x'= x exp(y.s)
k.r	The number of releve groups
k.s	The number of species groups
ndiffs	The number of (high resolving) species used for top portion of the table
...	Use method="normal" for conventional display, "compressed" for very large tables
rorder	The order of releves (rows) for printing
sorder	The order of species (columns) for printing
grr	The group labels of releves (rows) for printing
grs	The group labels of species (columns) for printing
x	An object of class "Mtabs"
object	An object of class "Mtabs"
range	A subset of species to be displayed in summary table, e.g., c(1,10) for the first 10.
vec	A vector of group labels, analyzed similar to function table(), but without sorting
y	Transformation of species scores: x'= x exp(y)

Details

Function plottab() and plottabl() are for internal use only

Value

An object of class "Mtabs" with at least the following items:

method	The method used for ordering
transf.r	Argument y.r
transf.s	Argument y.s
order.rel	The resulting order of rows
order.sp	The resulting order of columns
order.relgr	The resulting order of releve groups
order.spgr	The resulting order of species groups
MSCC	Mean square contingency coefficient
CAeig.rel	Eigenvalues of correspondence analysis
AOCeig.rel	Eigenvalues of analysis of concentration
veg	The input vegetation data frame
centroids	The matrix of groups centroids (see summary.Mtabs

Note

This extremely complex procedure accords with conventions used in vegetation ecology. It assumes that the vegetation data frame has many zero entries (plots in which species are not found). The summary method displays a frequency table (relative frequency of all species within the releve groups, centroid).

Author(s)

Otto Wildi

References

Wildi, O. 1989. A new numerical solution to traditional phytosociological tabular classification. Vegetatio 81: 95–106.

Wildi, O. 2017. Data Analysis in Vegetation Ecology. 3rd ed. CABI, Oxfordshire, Boston.

Examples

y.r<- 0.5 ; y.s<- 0.2                 # defining transformations used
k.r <- 3 ; k.s <- 4                   # row- and column numbers
ndiffs <- 18                          # no. of columns used to show pattern
o.Mt<-Mtabs(nveg,"mulva" ,y.r,y.s,k.r,k.s,ndiffs)
plot(o.Mt,method="normal")
# to see the original order simply replace "mulva" by "raw"

Example output

Loading required package: cluster
Loading required package: labdsv
Loading required package: mgcv
Loading required package: nlme
This is mgcv 1.8-33. For overview type 'help("mgcv-package")'.
This is labdsv 2.0-1
convert existing ordinations with as.dsvord()

Attaching package: ‘labdsv’

The following object is masked from ‘package:stats’:

    density

Loading required package: vegan
Loading required package: permute
Loading required package: lattice
This is vegan 2.5-7
Loading required package: nnet

Attaching package: ‘nnet’

The following object is masked from ‘package:mgcv’:

    multinom

Loading required package: tree
Call:
Mtabs.default(veg = nveg, method = "mulva", y.r = y.r, y.s = y.s, 
    k.r = k.r, k.s = k.s, ndiffs = ndiffs)

CA eigenvalues, %:       52.82 16.61 9.78 
AOC eigenvalues, %:      87.76 12.24 NA 

Mean square contingency coefficient: 0.20763 
Table split into 1 by 1 plots

dave documentation built on May 2, 2019, 1:30 p.m.