Add envfit Results to an ordiggplot Graph — autolayer.envfit • ggvegan

Function adds layers of fitted vectors and centroids of factor levels in an ordiggplot() graph.

Usage

# S3 method for class 'envfit'
autolayer(
  object,
  text = TRUE,
  box = FALSE,
  arrow.mul = 1,
  arrow.params = list(),
  text.params = list(),
  ...
)

Arguments

object: vegan::envfit() result object
text: add text to plot (logical).
box: write text on a non-transparent label (logical).
arrow.mul: arrow multiplier.
arrow.params, text.params: List of additional parameters to ggplot2::geom_segment() for arrows, and to text labels of both arrows and centroids of factor levels (ggplot2::geom_text(), ggplot2::geom_label()).
...: Other parameters passed to all graphical functions geom_ordi_arrow(), geom_ordi_text() and geom_ordi_label().

Value

Returns ggplot2 layers geom_segment for arrows and geom_text or geom_label for their text labels (when appropriate), and another layer of geom_text or geom_label for the centroids of factor leves (when appropriate). The score type is called envfit.

Details

Function adds a previosly fitted vegan::envfit() model. This does not adapt to changes in the ordiggplot() object, for instance in scaling or axes, but it must re-fitted for a changed model.

Author

Jari Oksanen

Examples

library(vegan)
#> Loading required package: permute
library(ggplot2)
data(mite, mite.env, package = "vegan")
mod <- cca(mite)
## you must have same scaling in envfit() and ordiggplot()
ef <- envfit(mod ~ Shrub+Topo+WatrCont+SubsDens, mite.env,
  scaling = "sites")
ordiggplot(mod, scaling="sites") +
  geom_ordi_axis() +
  geom_ordi_point("sites") +
  autolayer(ef, arrow.mul=1.3, col="navy", box=TRUE,
    text.params=list(mapping=aes(fontface="bold")))