fuzzyq {FuzzyQ}

Fuzzy Quantification of Common and Rare Species in Ecological Communities

Description

Perform fuzzy clustering of each species based on their abundance and occupancy.

Usage

fuzzyq(
  M,
  diss = "gower",
  rm.absent = FALSE,
  sorting = TRUE,
  keep.Diss = FALSE,
  std = FALSE,
  wgts = c(1, 1),
  ...
)

Arguments

M

A matrix or data frame of species abundances (columns). Each row represents a site.

diss

String. Specify the dissimilarity coefficient to be used. Default is "gower". See daisy in package cluster for other choices.

rm.absent

Logical. Whether or not absent species are to be removed from the calculations.

sorting

Logical. If TRUE (the default) species are sorted in the output by ascending silhouette widths within each cluster, else species are arranged in the same order as in the input matrix or data frame.

keep.Diss

Logical. Whether or not the species dissimilarity matrix shoudl be returned. The default is FALSE.

std

Logical. Whether or not the measurements of occupancy and abundance are to be standardized before calculating the dissimilarities. Measurements are standardized for each variable (column), by subtracting the variable's mean value and dividing by the variable's mean absolute deviation. It only takes effect if diss is different from "gower".

wgts

an optional numeric vector of length 2. To be used if diss = "gower", specifying weights for occupancy and abundance, respectively. Default is 1 each as in Gower's original formula.

...

Arguments to be passed to function fanny in package cluster.

Value

A list of class fuzzyq containing the following:

A_O

Abundance-occupancy information for each species.

Diss

Object of class dist with pairwise dissimilarities among species based on A_O. (only if keep.Diss = TRUE).

spp

Clustering metrics per species: Cluster membership (where 0 and 1 denote allocation to the rare and common category, respectively), Silhouette Widths and Commonness Indices).

global

Community level clustering metrics: Average silhouette widths per cluster and globally, Mean commonness indices per cluster and Normalized Dunn's coefficient.

See Also

fanny and daisy in package cluster

Examples

data(antsA)
FQAnts <- fuzzyq(antsA, sorting = TRUE)