dga {labdsv}

Direct Gradient Analysis

Description

Direct gradient analysis is a graphical representation of the abundance distribution of (typically) species along opposing environmental gradients

Usage

dga(z,x,y,step=50,pres="+",abs="-",labcex=1,
    xlab = deparse(substitute(x)), ylab = deparse(substitute(y)),
    pch = 1, title = "", ...)

Arguments

z

the variable (typically a species abundance) to be plotted

x

the variable to use as the x axis

y

the variable to use as the y axis

step

controls the grid density fed to the GAM surface fitter

pres

the symbol to print when a species is present (presence/absence mode)

abs

the symbol to print when a species is absent (presence/absence mode)

labcex

the character size for contour labels

xlab

the x axis legend

ylab

the y axis legend

pch

the symbol to print in continuous abundance plots

title

the title to print

...

miscellaneous arguments to pass to par

Details

‘dga’ interpolates a grid of x,y values from the supplied data and fits a GAM (from mgcv) of the z variable to the grid. For presence/absence data (enterd as a logical) it employs a binomial family, for species abundances a negative binomial is employed. The GAM surface is then represented by a contour map and abundance symbols as described above.

Value

a graph of the distribution of the z variable on a grid of x and y is displayed on the current active device.

Note

Direct gradient analysis was promoted by Robert Whittaker and followers as a preferred method of vegetation analysis.

Author(s)

David W. Roberts [e-mail]

See Also

gam

Examples

data(bryceveg) # returns a data.frame called bryceveg
x <- c(0.2,0.5,1.0,2.0,3.0,4.0,5.0,6.0)
y <- c(0.2,0.5,3.0,15.0,37.5,62.5,85.0,97.5)
cover <- abundtrans(bryceveg,x,y)
data(brycesite)
dga(round(cover$arcpat),brycesite$elev,brycesite$av)