Autoplot graphics for vegan rank-abundance models

ggplot2 graphics for Rank-Abundance Distribution models fitted with vegan functions vegan::radfit() or produced with vegan::as.rad().

Usage

# S3 method for class 'radfit'
autoplot(
  object,
  facet = TRUE,
  point.params = list(),
  line.params = list(),
  ...
)

# S3 method for class 'radfit.frame'
autoplot(object, point.params = list(), line.params = list(), ...)

# S3 method for class 'radline'
autoplot(object, point.params = list(), line.params = list(), ...)

# S3 method for class 'rad'
autoplot(object, point.params = list(), line.params = list(), ...)

# S3 method for class 'rad.frame'
autoplot(object, point.params = list(), highlight = NULL, ...)

Arguments

object

Result object from radfit.

facet

Draw each fitted model to a separate facet or (if FALSE) all fitted lines to a single graph.

point.params, line.params

Parameters to modify points or lines (passed to geom_point and geom_line).

...

Additional arguments passed to the functions.

highlight

Names of species that should be highlighted as coloured points.

Value

Returns a ggplot object.

Details

The ggplot2::autoplot() function draws graphics which are ggplot2 alternatives for lattice graphics in vegan. In addition, there are functions for vegan::as.rad() results which do not have dedicated graphics invegan.

Author

Jari Oksanen

Examples

library(vegan)
library(ggplot2)
data(mite)
m1 <- radfit(mite[1, ])

## With logarithmic y-axis (default) Pre-emption model is a line
autoplot(m1) +
  labs(title="log-Abundance: Pre-emption model is a line")


## With log-log scale, Zipf model is a line
autoplot(m1) +
  scale_x_log10() +
  labs(title="log-log Scale: Zipf model is a line")


## Show only the best model
autoplot(m1, pick = "AIC")


## Show selected models in one frame
autoplot(m1, pick = c("Z","M","L"), facet=FALSE)


## plot best models for several sites
m <- radfit(mite[1:12,])
autoplot(m) +
  labs(title = "Model Selection AIC (Default)")


## use BIC and reorder sites by their diversity
autoplot(m, pick="BIC", order.by = diversity(mite[1:12,])) +
   labs(title="Model Selection BIC, Ordered by Increasing Diversity")


## Plot RAD models without fits highlighting most abundant species in the
## whole data.
m0 <- as.rad(mite[1:12,])
dominants <- names(sort(colSums(mite), decreasing = TRUE))[1:6]
autoplot(m0, highlight = dominants)