Package 'ggvegan'

Description

Adds a biplot arrow layer to an existing plot

Usage

add_biplot_arrows(object, plt, vars, arrow.col = "navy")

Arguments

Description

Adds a biplot arrow layer to an existing plot

Usage

add_biplot_centroids(object, plt, vars, arrow.col = "navy")

Arguments

Description

Adds species and site score layers to an existing plot

Usage

add_spp_site_scores(object, plt, vars, geom, draw_list, arrows)

Arguments

Description

Scale vector arrows to fill proportion of the data.

Usage

arrow_mul(arrows, data, at = c(0, 0), fill = 0.75)

Arguments

Value

a numeric multiplier that will scale the arrows

Author(s)

Gavin L. Simpson

Description

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

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.

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.

Author(s)

Jari Oksanen

Examples

library(vegan)
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")))

Description

Function adds vegan::ordisurf() results to an graph.

Usage

## S3 method for class 'ordisurf'
autolayer(
  object,
  fill = FALSE,
  contour.params = list(),
  fill.params = list(),
  ...
)

Arguments

Details

Function draws contours of fitted vegan::ordisurf() surface and (optionally) a raster of the surface using ggplot2::geom_contour() and ggplot2::geom_raster().

Function uses gridded values of ordination plane saved within the vegan::ordisurf() result instead of scores data from an ordination object. Therefore its results can be added to any ordination graph, also from autoplot methods. However, you must be careful with using exactly the same axes and scaling in vegan::ordisurf() object and ordination.

Surface fill is non-transparent and will paint over all previous layers. Filled surface should be used before other visible layers, or fill should be made transparent with alpha in argument fill.params passed to ggplot2::geom_raster().

Value

Returns ggplot2 layers geom_contour and geom_raster (optionally).

Author(s)

Jari Oksanen

Examples

library(vegan)
library(ggplot2)
data(mite, mite.env, package="vegan")
mod <- cca(mite)
surf <- ordisurf(mod ~ WatrCont, mite.env, plot = FALSE)
ordiggplot(mod) +
  geom_ordi_point("sites", size = mite.env$WatrCont/100) +
  autolayer(surf) +
  autolayer(envfit(mod ~ WatrCont, mite.env, permutations=0),
    arrow.mul = 1.3)

Description

Produces a multi-layer ggplot object representing the output of objects produced by vegan::anosim().

Usage

## S3 method for class 'anosim'
autoplot(
  object,
  notch = TRUE,
  varwidth = TRUE,
  xlab = NULL,
  ylab = NULL,
  title = "Analysis of similarities",
  subtitle = NULL,
  caption = bquote(R == .(r) * "," ~ P == .(p)),
  ...
)

Arguments

Value

A ggplot object.

Author(s)

Didzis Elferts. Modifications by Gavin L. Simpson.

Examples

library("vegan")

data(dune)
data(dune.env)
dune.dist <- vegdist(dune)
dune.ano <- with(dune.env, anosim(dune.dist, Management))

autoplot(dune.ano, notch = FALSE)

Description

Produces a multi-layer ggplot object representing the output of objects produced by vegan::cca() or vegan::capscale().

Usage

## S3 method for class 'cca'
autoplot(
  object,
  axes = c(1, 2),
  geom = c("point", "text"),
  layers = c("species", "sites", "biplot", "centroids"),
  legend.position = "right",
  title = NULL,
  subtitle = NULL,
  caption = NULL,
  ylab = NULL,
  xlab = NULL,
  arrow.col = "navy",
  ...
)

Arguments

Details

TODO

Value

Returns a ggplot object.

Author(s)

Gavin L. Simpson

Examples

library("vegan")
data(dune)
data(dune.env)

sol <- cca(dune ~ A1 + Management, data = dune.env)
autoplot(sol)

Description

Produces a multi-layer ggplot object representing the output of objects produced by vegan::dbrda().

Usage

## S3 method for class 'dbrda'
autoplot(
  object,
  axes = c(1, 2),
  geom = c("point", "text"),
  layers = c("sites", "biplot", "centroids"),
  legend.position = "none",
  title = NULL,
  subtitle = NULL,
  caption = NULL,
  ylab = NULL,
  xlab = NULL,
  const = NULL,
  arrow.col = "navy",
  ...
)

Arguments

Details

TODO

Value

Returns a ggplot object.

Author(s)

Gavin L. Simpson

Examples

library("vegan")

data(dune, dune.env)

dune_dbrda <- dbrda(
  dune ~ A1 + Moisture + Use + Management,
  data = dune.env
)
autoplot(dune_dbrda)

Description

Produces a multi-layer ggplot object representing the output of objects produced by vegan::decorana().

Usage

## S3 method for class 'decorana'
autoplot(
  object,
  axes = c(1, 2),
  geom = c("point", "text"),
  layers = c("sites", "species"),
  legend.position = "right",
  title = NULL,
  subtitle = NULL,
  caption = NULL,
  ylab,
  xlab,
  ...
)

Arguments

Details

TODO

Value

Returns a ggplot object.

Author(s)

Gavin L. Simpson

Examples

library("vegan")
data(dune)

sol <- decorana(dune)
autoplot(sol)
autoplot(sol, layers = "species", geom = "text")

Description

Produces a multi-layer ggplot object representing the output of objects produced by vegan::envfit().

Usage

## S3 method for class 'envfit'
autoplot(
  object,
  geom = c("label", "text", "label_repel", "text_repel"),
  line.col = "black",
  xlab = NULL,
  ylab = NULL,
  title = NULL,
  subtitle = NULL,
  caption = NULL,
  ...
)

Arguments

Value

A ggplot object.

Author(s)

Gavin L. Simpson

Examples

library("vegan")
data(varespec, varechem)
ord1 <- metaMDS(varespec)
fit1 <- envfit(ord1, varechem, perm = 199)

autoplot(fit1, geom = 'label_repel')

data(dune, dune.env)
ord2 <- cca(dune)
fit2 <- envfit(ord2 ~ Moisture + A1, dune.env, perm = 199)

autoplot(fit2)

Description

Draws a bar plot of species rank abundance with Fisher's log-series superimposed.

Usage

## S3 method for class 'fisherfit'
autoplot(
  object,
  show.fitted = TRUE,
  xlab = "Abundance",
  ylab = "Number of Species",
  title = "Fisher's log-series distribution",
  bar.col = NA,
  bar.fill = "grey35",
  line.col = "red",
  linewidth = 0.7,
  subtitle = NULL,
  caption = NULL,
  ...
)

Arguments

Value

A ggplot object.

Author(s)

Gavin L. Simpson

Examples

library("vegan")

data(BCI)
mod <- fisherfit(BCI[5,])
autoplot(mod)

Description

Produces a multi-layer ggplot object representing the output of objects produced by vegan::isomap().

Usage

## S3 method for class 'isomap'
autoplot(
  object,
  axes = c(1, 2),
  geom = c("point", "text"),
  network = TRUE,
  line.col = "grey85",
  linewidth = 0.7,
  xlab = NULL,
  ylab = NULL,
  title = "Isometric feature mapping",
  subtitle = NULL,
  caption = NULL,
  ...
)

Arguments

Value

A ggplot object.

Author(s)

Gavin L. Simpson

Examples

library("vegan")

data(BCI)
dis <- vegdist(BCI)
ord <- isomap(dis, k = 3)

autoplot(ord)

autoplot(ord, geom = "text")

Description

Produces a multi-layer ggplot object representing the output of objects produced by vegan::metaMDS().

Usage

## S3 method for class 'metaMDS'
autoplot(
  object,
  geom = c("point", "text"),
  layers = c("species", "sites"),
  legend.position = "right",
  title = NULL,
  subtitle = NULL,
  caption = NULL,
  ylab,
  xlab,
  ...
)

Arguments

Details

TODO

Value

Returns a ggplot object.

Author(s)

Gavin L. Simpson

Examples

library("vegan")

data(dune)

sol <- metaMDS(dune)
autoplot(sol)

Description

Alternatives for lattice graphics functions vegan::densityplot.permustats(), vegan::densityplot.permustats() and vegan::boxplot.permustats().

Usage

## S3 method for class 'permustats'
autoplot(
  object,
  plot = c("box", "violin", "density", "qqnorm"),
  scale = FALSE,
  facet = FALSE,
  gg.params = list(),
  ...
)

Arguments

Details

Function fortify() returns a data frame with variables permutations (numeric) and term (factor labelling the permutation). The result of fortify() can be used to custom build diagnostic plots. autoplot() provides basic plots of vegan::permustats() objects with limited flexibility.

Value

Returns a ggplot object.

Examples

library("vegan")

data(dune, dune.env)
mod <- cca(dune ~ A1 + Management + Moisture, dune.env)
(ano <- anova(mod, by = "onedf"))
pstat <- permustats(ano)
head(fortify(pstat))
autoplot(pstat, "box")


if (requireNamespace("ggplot2")) {
  library("ggplot2")

  # avoid overplotting x-axis text
  autoplot(pstat, "violin") +
    scale_x_discrete(guide = guide_axis(n.dodge = 2))

  autoplot(pstat, "qqnorm", facet = TRUE) +
    geom_qq_line()
}

autoplot(pstat, "density", facet = TRUE)

Description

Produces a multi-layer ggplot object representing the output of objects produced by vegan::poolaccum().

Usage

## S3 method for class 'poolaccum'
autoplot(
  object,
  facet = TRUE,
  ribbon = facet,
  ncol = NULL,
  ribbon.alpha = 0.3,
  xlab = "Size",
  ylab = "Richness",
  title = "Accumulated species richness",
  subtitle = NULL,
  caption = NULL,
  ...
)

Arguments

Value

A ggplot object.

Author(s)

Didzis Elferts & Gavin L. Simpson

Examples

library("vegan")

data(BCI)
pool <- poolaccum(BCI)

autoplot(pool)

## Turn off facetting; turns off ribbon too
autoplot(pool, facet = FALSE)

Description

Produces a multi-layer ggplot object representing the output of objects produced by vegan::prc().

Usage

## S3 method for class 'prc'
autoplot(
  object,
  select,
  xlab,
  ylab,
  title = NULL,
  subtitle = NULL,
  caption = NULL,
  legend.position = "top",
  ...
)

Arguments

Details

TODO

Value

Returns a ggplot object.

Author(s)

Gavin L. Simpson

Examples

library("vegan")

data(pyrifos)
week <- gl(11, 12, labels=c(-4, -1, 0.1, 1, 2, 4, 8, 12, 15, 19, 24))
dose <- factor(rep(c(0.1, 0, 0, 0.9, 0, 44, 6, 0.1, 44, 0.9, 0, 6), 11))

## Fit PRC model
mod <- prc(pyrifos, dose, week)

## plot
want <- colSums(pyrifos) >= 300
autoplot(mod, select = want)

Description

Draws a bar plot of species rank abundance with Preston's log-normal model superimposed.

Usage

## S3 method for class 'prestonfit'
autoplot(
  object,
  show.fitted = TRUE,
  xlab = "Abundance",
  ylab = "Number of Species",
  title = "Preston's lognormal distribution",
  subtitle = NULL,
  caption = NULL,
  bar.col = NA,
  bar.fill = "grey35",
  line.col = "red",
  linewidth = 0.7,
  ...
)

Arguments

Value

A ggplot object.

Author(s)

Gavin L. Simpson

Examples

library("vegan")

data(BCI)
pfit <- prestonfit(colSums(BCI))
autoplot(pfit)

Description

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

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.

Value

Returns a ggplot object.

Author(s)

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)

Description

Produces a multi-layer ggplot object representing the output of objects produced by vegan::rda().

Usage

## S3 method for class 'rda'
autoplot(
  object,
  axes = c(1, 2),
  geom = c("point", "text"),
  layers = c("species", "sites", "biplot", "centroids"),
  arrows = TRUE,
  legend.position = "none",
  title = NULL,
  subtitle = NULL,
  caption = NULL,
  ylab = NULL,
  xlab = NULL,
  const = NULL,
  arrow.col = "navy",
  ...
)

Arguments

Details

TODO

Value

Returns a ggplot object.

Author(s)

Gavin L. Simpson

Examples

library("vegan")

data(dune)

pca <- rda(dune)
autoplot(pca)

## Just the species scores
autoplot(pca, layers = "species")

Description

Produces a multi-layer ggplot object representing the output of objects produced by vegan::pco().

Usage

## S3 method for class 'vegan_pco'
autoplot(
  object,
  geom = "point",
  legend.position = "right",
  title = NULL,
  subtitle = NULL,
  caption = NULL,
  ylab = NULL,
  xlab = NULL,
  ...
)

Arguments

Details

TODO

Value

Returns a ggplot object.

Author(s)

Gavin L. Simpson

Examples

library("vegan")

data(dune)

sol <- pco(dune)
autoplot(sol)

Description

Check user-supplied layers against list of valid layers

Usage

check_user_layers(user, valid, message = FALSE)

Arguments

Description

Prepares a fortified version of results from vegan::anosim() objects.

Usage

## S3 method for class 'anosim'
fortify(model, data, ...)

## S3 method for class 'anosim'
tidy(x, data, ...)

Arguments

Value

A data frame with columns Rank and Class with ranks of dissimilarity.

Author(s)

Original implementation by Didzis Elferts. Modification to tibbles by Gavin L. Simpson.

Examples

library("vegan")

library("ggplot2")
data(dune, dune.env)
dune.dist <- vegdist(dune)
dune.ano <- with(dune.env, anosim(dune.dist, Management))

df <- fortify(dune.ano)

ggplot(df, aes(x = class, y = rank)) +
    geom_boxplot(notch = FALSE, varwidth = TRUE)

Description

Fortifies an object of class "cca" to produce a data frame of the selected axis scores in long format, suitable for plotting with ggplot2::ggplot().

Usage

## S3 method for class 'cca'
fortify(model, data, axes = 1:6, layers = c("sp", "wa", "lc", "bp", "cn"), ...)

## S3 method for class 'cca'
tidy(x, data, axes = 1:6, layers = c("sp", "wa", "lc", "bp", "cn"), ...)

Arguments

Details

TODO

Value

A data frame (tibble) in long format containing the ordination scores. The first two components are score (the type of score in each row) and label (the text label to use on plots for this row). The remaining columns are the extracted ordination axis scores.

Author(s)

Gavin L. Simpson

Examples

library("vegan")

data(dune)
data(dune.env)

sol <- cca(dune ~ A1 + Management, data = dune.env)
head(fortify(sol))

Description

Fortifies an object of class "dbrda" to produce a data frame of the selected axis scores in long format, suitable for plotting with ggplot2::ggplot().

Usage

## S3 method for class 'dbrda'
fortify(
  model,
  data = NULL,
  axes = 1:6,
  layers = c("wa", "lc", "bp", "reg", "cn"),
  const = NULL,
  ...
)

## S3 method for class 'dbrda'
tidy(
  x,
  data = NULL,
  axes = 1:6,
  layers = c("wa", "lc", "bp", "reg", "cn"),
  const = NULL,
  ...
)

Arguments

Details

TODO

Value

A data frame (tibble) in long format containing the ordination scores. The first two components are score (the type of score in each row) and label (the text label to use on plots for this row). The remaining columns are the extracted ordination axis scores.

Author(s)

Gavin L. Simpson

Examples

library("vegan")

data(dune)
data(dune.env)

sol <- dbrda(dune ~ A1 + Management, data = dune.env)
head(fortify(sol))

Description

Fortifies an object of class "decorana" to produce a data frame of the selected axis scores in long format, suitable for plotting with ggplot2::ggplot().

Usage

## S3 method for class 'decorana'
fortify(model, data, axes = 1:4, layers = c("sites", "species"), ...)

## S3 method for class 'decorana'
tidy(x, data, axes = 1:4, layers = c("sites", "species"), ...)

Arguments

Details

TODO

Value

A data frame in long format containing the ordination scores. The first two components are the axis scores.

Author(s)

Gavin L. Simpson

Examples

library("vegan")

data(dune)

sol <- decorana(dune)
head(fortify(sol))
head(fortify(sol, layers = "species"))

Description

Produces a tidy data frame from the results of an vegan::envfit() object.

Usage

## S3 method for class 'envfit'
fortify(model, data, ...)

## S3 method for class 'envfit'
tidy(x, data, ...)

Arguments

Value

A data frame with columns label, type, containing the label for, and whether each row refers to, the fitted vector or factor. Remaining variables are coordinates on the respective ordination axes returned by vegan::scores.envfit().

Author(s)

Gavin L. Simpson

Examples

library("vegan")

data(varespec, varechem)
ord <- metaMDS(varespec)
fit <- envfit(ord, varechem, perm = 199)

fortify(fit)

data(dune, dune.env)
ord <- ca(dune)
fit <- envfit(ord ~ Moisture + A1, dune.env, perm = 199)

fortify(fit)

Description

Prepares a data frame of results from a vegan::fisherfit()) object suitable for plotting with ggplot2::ggplot()).

Usage

## S3 method for class 'fisherfit'
fortify(model, data, ...)

## S3 method for class 'fisherfit'
tidy(x, data, ...)

Arguments

Value

A data frame with columns 'Rank' and 'Abundance'. Additionally, Fisher's $\alpha$ and the nuisance parameter are returned as attributes 'alpha' and 'k' respectively.

Author(s)

Gavin L. Simpson

Examples

library("vegan")

data(BCI)
mod <- fisherfit(BCI[5,])
head(fortify(mod))

Description

Prepares fortified versions of results from vegan::isomap() objects.

Usage

## S3 method for class 'isomap'
fortify(model, data = NULL, axes = NULL, what = c("sites", "network"), ...)

## S3 method for class 'isomap'
tidy(x, data = NULL, axes = NULL, what = c("sites", "network"), ...)

Arguments

Details

Two different objects can be created from the results of an vegan::isomap() object. The first is the standard scores representation of fortified ordinations in vegan, which results in a wide data frame where rows contain observations and column the coordinates of observations on the MDS axes. Because ISOMAP also produces a network or sorts, the coordinates of the edges of the network can also be returned in a tidy format using what = "network".

Value

A data frame. For what = "sites", the data frame contains one variable per dimension of the multidimensional scaling embedding of the dissimilarities. Variables are named "DimX" with "X" being an integer. An additional variable is label, containing a label for each observation. For what = "network", the data frame contains four variables containing the coordinates in the chosen MDS axes for the start and end points of the network edges.

Author(s)

Gavin L. Simpson

Examples

library("vegan")

library("ggplot2")
data(BCI)
dis <- vegdist(BCI)
ord <- isomap(dis, k = 3)

df <- fortify(ord, data = data.frame(Richness = specnumber(BCI)))
net <- fortify(ord, what = "network", axes = 1:2)

ggplot(df, aes(x = dim1, y = dim2)) +
    geom_segment(data = net,
                 aes(x = xfrom, y = yfrom,
                     xend = xto, yend = yto),
                 colour = "grey85", size = 0.8) +
    geom_point(aes(size = Richness)) +
    coord_fixed()

Description

Fortifies an object of class "metaMDS" to produce a data frame of the selected axis scores in long format, suitable for plotting with ggplot2::ggplot().

Usage

## S3 method for class 'metaMDS'
fortify(model, data, axes = 1:2, layers = c("sites", "species"), ...)

## S3 method for class 'metaMDS'
tidy(x, data, layers = c("sites", "species"), ...)

Arguments

Details

TODO

Value

A data frame in long format containing the ordination scores. The first two components are the axis scores.

Author(s)

Gavin L. Simpson

Examples

library("vegan")

data(dune)

ord <- metaMDS(dune)
head(fortify(ord))

Description

Fortify permutation statistics

Usage

## S3 method for class 'permustats'
fortify(model, data, scale = FALSE, ...)

## S3 method for class 'permustats'
tidy(x, data, scale = FALSE, ...)

Arguments

Value

A tibble with columns permutations, and terms containing the values of tests statistics under the null hypothesis, and a factor labelling the permutation, respectively.

Examples

library("vegan")
data(dune, dune.env, package = "vegan")
mod <- adonis2(dune ~ Management + A1, data = dune.env)
## use permustats
perm <- permustats(mod)

Description

Prepares a fortified version of results from vegan::poolaccum() objects.

Usage

## S3 method for class 'poolaccum'
fortify(model, data, alpha = 0.05, ...)

## S3 method for class 'poolaccum'
tidy(x, data, alpha = 0.05, ...)

Arguments

Value

A data frame with columns index, size, richness, lower, upper, and std_dev, containing the richness index, permutation richness estimator, sample size, upper and lower $1 - \alpha$ quantile interval, and standard deviation of permutation estimates, respectively.

Author(s)

Gavin L. Simpson & Didzis Elferts

Examples

library("vegan")
library("ggplot2")
data(BCI)
pool <- poolaccum(BCI)
df <- fortify(pool)
df

ggplot(df, aes(x = size, y = richness, colour = index)) +
    geom_ribbon(aes(ymin = lower, ymax = upper, x = size, fill = index),
                alpha = 0.3, inherit.aes = FALSE) +
    geom_line() +
    facet_wrap(~ index)

Description

Fortifies an object of class "prc" to produce a data frame of the selected axis scores in long format, suitable for plotting with ggplot2::ggplot().

Usage

## S3 method for class 'prc'
fortify(model, data, scaling = "symmetric", axis = 1, ...)

## S3 method for class 'prc'
tidy(x, data, scaling = "symmetric", axis = 1, ...)

Arguments

Details

TODO

Value

A data frame in long format containing the ordination scores. The first three components are the Time, Treatment, and associated Response. The last two components, score and label are an indicator factor and a label for the rows for use in plotting.

Author(s)

Gavin L. Simpson

Examples

library("vegan")
data(pyrifos)
week <- gl(11, 12, labels=c(-4, -1, 0.1, 1, 2, 4, 8, 12, 15, 19, 24))
dose <- factor(rep(c(0.1, 0, 0, 0.9, 0, 44, 6, 0.1, 44, 0.9, 0, 6), 11))
ditch <- gl(12, 1, length=132)

pyrifos_prc <- prc(pyrifos, dose, week)
fortify(pyrifos_prc)

Description

Prepares a data frame of results from a vegan::prestonfit() object suitable for plotting with ggplot2::ggplot().

Usage

## S3 method for class 'prestonfit'
fortify(model, data, ...)

## S3 method for class 'prestonfit'
tidy(x, data, ...)

Arguments

Value

A data frame with columns ⁠'Octave'`` and ⁠'Abundance'“.

Author(s)

Gavin L. Simpson

Examples

library("vegan")

data(BCI)
pfit <- prestonfit(colSums(BCI))
fortify(pfit)

Description

Prepares a fortified version of results from vegan::radfit() or vegan::as.rad().

Usage

## S3 method for class 'radfit'
fortify(model, data, pick = NULL, ...)

## S3 method for class 'radfit.frame'
fortify(model, data, pick = "AIC", order.by = NULL, ...)

## S3 method for class 'radline'
fortify(model, data, ...)

## S3 method for class 'rad'
fortify(model, data, ...)

## S3 method for class 'rad.frame'
fortify(model, data, order.by = NULL, ...)

## S3 method for class 'radfit'
tidy(x, data, pick = NULL, ...)

## S3 method for class 'radfit.frame'
tidy(x, data, pick = "AIC", order.by = NULL, ...)

## S3 method for class 'radline'
tidy(x, data, ...)

## S3 method for class 'rad'
tidy(x, data, ...)

## S3 method for class 'rad.frame'
tidy(x, data, order.by = NULL, ...)

Arguments

Value

A data frame or tibble which in all cases has columns species (character), for species names, with rank and abundance for data, and may also have columns site (factor) for site names when several sites were analysed, fit of fitted abundances when evaluated, and model (factor) when several models were fitted.

Author(s)

Jari Oksanen

Description

Fortifies an object of class "rda" to produce a data frame of the selected axis scores in long format, suitable for plotting with ggplot2::ggplot().

Usage

## S3 method for class 'rda'
fortify(
  model,
  data = NULL,
  axes = 1:6,
  layers = c("sp", "wa", "lc", "bp", "cn"),
  const = NULL,
  ...
)

## S3 method for class 'rda'
tidy(
  x,
  data = NULL,
  axes = 1:6,
  layers = c("sp", "wa", "lc", "bp", "reg", "cn"),
  const = NULL,
  ...
)

Arguments

Details

TODO

Value

A data frame (tibble) in long format containing the ordination scores. The first two components are score (the type of score in each row) and label (the text label to use on plots for this row). The remaining columns are the extracted ordination axis scores.

Author(s)

Gavin L. Simpson

Examples

library("vegan")

data(dune)
data(dune.env)

sol <- rda(dune ~ A1 + Management, data = dune.env)
head(fortify(sol))

Description

Prepares a fortified version of results from vegan::renyiaccum() objects.

Usage

## S3 method for class 'renyiaccum'
fortify(model, data, ...)

## S3 method for class 'renyiaccum'
tidy(x, data, ...)

Arguments

Value

A data frame (tibble) is returned. What is returned depends on how vegan::renyiaccum() was called. If raw = FALSE, then a data frame with columns site, scale, mean, std_dev, min, max, q2.5, and q97.5, containing the accumulated sites, the Rényi scale, and summary statistics of the Rényi accumulation curves. An additional column collector will be present if collector = TRUE was used in the vegan::renyiaccum() call. If raw = TRUE, then a data frame with columns site, permutation, scale, and diversity, containing the site and permutation identifiers, Rényi scale, and the Rényi diversity, respectively.

Author(s)

Gavin L. Simpson much rewritten from an original by Didzis Elferts.

Examples

library("vegan")
library("ggplot2")
data(BCI)
pool <- renyiaccum(BCI)
df <- fortify(pool)
df

ggplot(df, aes(x = site, y = mean)) +
    geom_ribbon(aes(ymin = q2.5, ymax = q97.5, x = site),
                alpha = 0.2, inherit.aes = FALSE) +
    geom_line() +
    facet_wrap(~ scale)

Description

Fortifies an object of class "vegan_pco" to produce a data frame of the selected axis scores in long format, suitable for plotting with ggplot2::ggplot().

Usage

## S3 method for class 'vegan_pco'
fortify(model, data = NULL, axes = 1:6, ...)

## S3 method for class 'vegan_pco'
tidy(x, data = NULL, axes = 1:6, const = NULL, ...)

Arguments

Details

TODO

Value

A data frame (tibble) in long format containing the ordination scores. The first two components are score (the type of score in each row) and label (the text label to use on plots for this row). The remaining columns are the extracted ordination axis scores.

Author(s)

Gavin L. Simpson

Examples

library("vegan")

data(dune)

sol <- pco(dune)
head(fortify(sol))

Description

Function adds layers of arrows (ggplot2::geom_segment()) and (optionally) their text labels (ggplot2::geom_text() or ggplot2::geom_label()) to an ordiggplot() graph. Typically these arrows are biplot or regression scores, but the function allows drawing species arrows for PCA biplots, or adding fitted environmental vectors of vegan::envfit().

Usage

geom_ordi_arrow(
  score,
  data,
  text = TRUE,
  box = FALSE,
  arrow.mul = 1,
  arrow.params = list(),
  text.params = list(),
  ...
)

Arguments

Value

Returns ggplot2 layers geom_segment for arrows, and geom_text or geom_label (optionally) for their names.

Author(s)

Jari Oksanen

See Also

Underlying functions are ggplot2::geom_segment(), ggplot2::geom_text() and ggplot2::geom_label().

Examples

library(vegan)
data(varespec, varechem, package="vegan")
## PCA biplot
mod <- rda(varespec)
ordiggplot(mod, scaling = "sites", corr = TRUE) +
  geom_ordi_axis() +
  geom_ordi_point("sites") +
  geom_ordi_arrow("species")
## RDA
mod <- rda(varespec ~ N + K + Al + Baresoil, varechem)
ordiggplot(mod) +
  geom_ordi_axis() +
  geom_ordi_point("sites") +
  geom_ordi_arrow("biplot", box = TRUE)

Description

The origin (coordinates 0,0) has a special interpretation in eigenvector methods: sites, species or variables at the centroid are average cases, and the point is more exceptional the further it is from the origin. Therefore crosshair of axes through the origin should be added to the ordination graph with eigenvector methods. Often it is forgotten, though. See the Example in geom_ordi_arrow() where the length of PCA biplot arrow shows how exceptional the species is in ordination.

Usage

geom_ordi_axis(lty = 3, ...)

Arguments

Value

Returns ggplot2 layers geom_hline and geom_vline.

Author(s)

Jari Oksanen

See Also

The underlying functions are ggplot2::geom_hline() and ggplot2::geom_vline().

Examples

library(vegan)
library(ggplot2)
data(dune, package = "vegan")
mod <- cca(dune)
## simple "autoplot" using ggplot2::geom_text
ordiggplot(mod) +
  geom_ordi_axis() +
  geom_text()

Description

Function adds “labels” or text in a box with non-transparent background using ggplot2::geom_label() to an ordiggplot() graph. These can help in congested plot since uppermost labels are readable (but cover lower ones), or to emphasize text. Some ggvegan functions call ggplot2::geom_label() with argument box = TRUE.

Usage

geom_ordi_label(score, data, ...)

Arguments

Value

Returns a ggplot2 layer geom_label.

Author(s)

Jari Oksanen

See Also

ggplot2::geom_label().

Examples

library(vegan)
library(ggplot2)
data(dune, dune.env, package = "vegan")
mod <- cca(dune ~ Moisture, dune.env)
ordiggplot(mod) +
  geom_ordi_axis() +
  geom_ordi_point("sites") +
  geom_ordi_label("species", size = 3,
    mapping = aes(fontface = "italic")) +
  geom_ordi_label("centroids", size = 5,
    mapping = aes(fontface = "bold"))

Description

Function adds a point layer using ggplot2::geom_point() to an ordiggplot() graph.

Usage

geom_ordi_point(score, data, ...)

Arguments

Value

Returns a ggplot2 layer geom_point.

Author(s)

Jari Oksanen

See Also

ggplot2::geom_point().

Examples

library(vegan)
library(ggplot2)
data(mite, mite.env, package = "vegan")
mod <- metaMDS(mite, trace = 0)
surf <- vegan::ordisurf(mod ~ WatrCont, mite.env, plot = FALSE)
ef <- vegan::envfit(mod ~ WatrCont + Topo, mite.env)
## Change sizes and colours of points
ordiggplot(mod) +
  geom_ordi_point("sites",
    mapping=aes(colour = mite.env$Topo, size=mite.env$WatrCont/10)) +
  autolayer(surf) +
  autolayer(ef, box = TRUE)

Description

Function adds points to the exact position of the ordination score with a label, but labels are repelled from each other to avoid over-plotting. In very congested areas, the labels are completely omitted.

Usage

geom_ordi_repel(
  score,
  data,
  box = FALSE,
  points = TRUE,
  point.params = list(),
  text.params = list(),
  ...
)

Arguments

Value

Returns ggrepel layers geom_text_repel or geom_label_repel and ggplot2 layer geom_point (optionally).

Author(s)

Jari Oksanen

Examples

library(vegan)
data(mite, mite.env, package = "vegan")
mod <- cca(mite)
ordiggplot(mod) +
  geom_ordi_axis() +
  geom_ordi_point("sites") +
  geom_ordi_repel("species",
    text.params = list(size=3, fontface = "italic"))

Description

Function adds a text layer to an ordiggplot() graph using ggplot2::geom_text().

Usage

geom_ordi_text(score, data, ...)

Arguments

Value

Returns a ggplot2 layer geom_text.

Author(s)

Jari Oksanen

See Also

ggplot2::geom_text().

Examples

library(vegan)
library(ggplot2)
data(dune, package = "vegan")
mod <- metaMDS(dune, trace = 0)
ordiggplot(mod) +
  geom_ordi_text("sites", size=3) +
  geom_ordi_text("species", mapping=aes(fontface="italic"))

Description

Find the character vector of names for the two dimensions of data to be plotted.

Usage

get_dimension_names(object)

Arguments

Value

A length 2 character vector of dimension names.

Author(s)

Gavin L. Simpson

Description

Adds labels to a plot using one of ggplot2::geom_label(), ggplot2::geom_text(), ggrepel::geom_label_repel() or ggrepel::geom_text_repel().

Usage

label_fun(data, geom = c("label", "text", "label_repel", "text_repel"), vars)

Arguments

Author(s)

Gavin L. Simpson

Description

List of layers to draw for a given vegan object

Usage

layer_draw_list(valid, layers = NULL, available = NULL)

Arguments

Description

Function ordiggplot sets up an ordination graph but draws no result. You can add new graphical elements to this plot with ⁠geom_ordi_*⁠ functions of this package, or you can use standard ggplot2 ⁠geom_*⁠ functions and use ggscores as their data argument.

Usage

ordiggplot(model, axes = c(1, 2), legend.position = "right", ...)

ggscores(score)

Arguments

Details

The ggvegan package has two contrasting approaches to draw ordination plots. The autoplot functions (e.g. autoplot.rda(), autoplot.cca(), and autoplot.metaMDS()) draw a complete plot with one command, but the design is hard-coded in the function, and it may be difficult to add new layers to the graph.

In contrast, function ordiggplot() only sets up an ordination plot, but allows you to add layers to the graph one by one with full flexibility of the ggplot2 functions. It initializes the ggplot2 graph with idiom ggplot(data, mapping) which allows other functions to use the ordination data and axis mapping in adding new layers. The included mapping specifies coordinates x and y and variables label for text labels and a score-type specific colour. Any ggplot2 function that can use this mapping will work automatically. Support function ggscores selects a slice of ordination scores that can be used as a data argument in the geom function. To plot the site scores of ordination result mod as points, you can use idiom ordiggplot(mod) + geom_point(data = ggscores("sites")).

There are specific functions to ease adding layers to an ordination graph. See the documentation of geom_ordi_arrow(), geom_ordi_label(), geom_ordi_point(), geom_ordi_repel(), geom_ordi_text(). These correspond to similarly named ⁠geom_*⁠ functions. For instance, geom_ordi_point adds very little to geom_point, and you can use all arguments of the standard geom function. These functions are of type ⁠geom_ordi_*(score, ...)⁠ which is similar to ⁠geom_*(data = ggscores(score), ...)⁠. Some functions were adapted to typical ordination data and return compound geometries. For instance, geom_ordi_arrow() returns layers geom_segment for the arrows, and geom_text for their name labels. In addition, there are functions to add previously fitted results of vegan::envfit() and vegan::ordisurf() (see autolayer.envfit(), autolayer.ordisurf()).

The ordiggplot() function extracts results using fortify() functions of this package, and it accepts the arguments of those functions. This allows setting, e.g., the scaling of ordination axes. The ordiggplot skeleton sets up data used in plotting, and you should define axis scaling, axes etc in the ordiggplot call and they will be used in all added layers.

Value

Returns a ggplot object with slot data for full ordination data and slot mapping with ordination axes mapped to x and y, label to text labels for each row and factor score type mapped to colour.

Author(s)

Jari Oksanen

See Also

geom_ordi_arrow(), geom_ordi_axis(), geom_ordi_label(), geom_ordi_point(), geom_ordi_repel(), geom_ordi_text(), autolayer.envfit(), autolayer.ordisurf().

Examples

library("vegan")
library("ggplot2")
data(dune, dune.env, varespec, varechem)
m <- cca(dune ~ Management + A1, dune.env)

## data and mapping of the ordiggplot object
ordiggplot(m)@data
ordiggplot(m)@mapping

## use geom_ordi_* functions
ordiggplot(m) + geom_ordi_axis() +
  geom_ordi_point("sites") +
  geom_ordi_repel("species", col = "darkblue",
                 text.params = list(mapping = aes(fontface = "italic"))) +
  geom_ordi_label("centroids") +
  geom_ordi_arrow("biplot")

## use ggscores + standard geom_* functions
ordiggplot(m, scaling = "sites") +
  geom_point(data = ggscores("sites")) +
  geom_text(data = ggscores("species"),
            mapping = aes(fontface = "italic")) +
  geom_label(data = ggscores("centroids"), fill = "yellow") +
  geom_ordi_arrow("biplot")

## Messy arrow biplot for PCA
m <- rda(dune)
ordiggplot(m, corr = TRUE) +
  geom_ordi_axis() +
  geom_ordi_point("sites") +
  geom_ordi_arrow("species")

Description

Returns the number of scores returns in object 'x“.

Usage

scores_length(x)

Arguments

Value

a numeric vector of length 1 with the number of scores.

Author(s)

Gavin L. Simpson

Description

Fits arrows to show the direction of fastest increase in continuous environmental variables in ordination space.The arrows are scaled relative to their correlation coefficient, and they can be added to an ordination plot with geom_ordi_arrow().

The arrows are fitted to the data and mapping of ordiggplot(), and they will adapt to changes in the parameters of ordiggplot(). This is in contrast to similar autolayer.envfit() command or to ⁠geom_ordi_arrow(data=<envfit-object>)⁠ which will use a previously fitted ⁠<envfit-object>⁠ and will not not change if the ordiggplot() definitions change.

Usage

stat_vectorfit(
  mapping = NULL,
  data = NULL,
  geom = "text",
  position = "identity",
  na.rm = FALSE,
  show.legend = FALSE,
  inherit.aes = TRUE,
  edata = NULL,
  formula = NULL,
  arrow.mul = NULL,
  ...
)

Arguments

Value

Returns a layer that containts a StatVectorfit object that is responsible for rendering the fitted vectors in the plot.

Author(s)

Jari Oksanen

Examples

library("vegan")
library("ggplot2")

data(mite, mite.env)
m <- metaMDS(mite, trace = FALSE, trymax = 100)

## add fitted vectors for continuous variables
ordiggplot(m) +
  geom_ordi_point("sites") +
  geom_ordi_arrow("sites", stat = "vectorfit", edata = mite.env)

## can be faceted
ordiggplot(m) + geom_ordi_point("sites") +
  geom_ordi_arrow("sites", stat = "vectorfit", edata = mite.env) +
  facet_wrap(mite.env$Topo)

Description

Valid layers for vegan objects

Usage

valid_layers(object, ...)

## S3 method for class 'rda'
valid_layers(object, ...)

## S3 method for class 'cca'
valid_layers(object, ...)

## S3 method for class 'dbrda'
valid_layers(object, ...)

Arguments

Value

A character vector of valid ordination layers for the provided object.