plot.plcfit {fitplc}

Plot a fitted vulnerability curve

Description

Standard plots of fitted curves (objects returned by fitplc, fitplcs, fitcond or fitconds), with plenty of options for customization.

Usage

## S3 method for class 'plcfit'
plot(x, xlab = NULL, ylab = NULL, ylim = NULL,
  pch = 19, plotPx = TRUE, plotci = TRUE, plotdata = TRUE,
  plotfit = TRUE, add = FALSE, multiplier = NULL,
  px_ci = c("bootstrap", "parametric", "none"),
  px_ci_type = c("vertical", "horizontal"), px_ci_label = TRUE,
  plotrandom = FALSE, pointcol = "black", linecol = "black",
  linetype = 1, linelwd = 1, linecol2 = "blue", pxlinecol = "red",
  pxcex = 0.7, citype = c("polygon", "lines"),
  cicol = alpha("lightgrey", 0.8), what = c("relk", "PLC", "embol"),
  selines = NULL, xaxis = c("positive", "negative"), ...)

## S3 method for class 'manyplcfit'
plot(x, what = c("relk", "embol", "PLC"),
  onepanel = FALSE, linecol = NULL, pointcol = NULL, pch = 19,
  legend = TRUE, legendwhere = "topright", ...)

Arguments

x

A fitted curve returned by fitplc

xlab, ylab

Optionally, X and Y axis labels (if not provided, a default is used).

ylim

Optionally, Y-axis limits.

pch

Optionally, the plotting symbol (default = 19, filled circles)

plotPx

Logical (default TRUE), whether to plot a vertical line for the P50.

plotci

Logical (default TRUE), whether to plot the confidence interval (if computed with bootci=TRUE).

plotdata

Logical (default TRUE), whether to add the data to the plot.

plotfit

Logical (default TRUE), whether to add the fitted curve to the plot.

add

Logical (default FALSE), whether to add the plot to a current device. This is useful to overlay two plots or curves, for example.

multiplier

Multiply the scaled data (for plotting).

px_ci

Option for the confidence interval around Px, either 'parametric' (confidence interval computed with confint), 'bootstrap' (computed with non-parametric bootstrap) or 'none' (no plotting of the confidence interval) (formerly argument was called selines)

px_ci_type

Either 'vertical' (default), or 'horizontal', to plot confidence limits for Px.

px_ci_label

Logical (default TRUE), whether to write a label next to the CI for Px.

plotrandom

Logical. If TRUE (the default is FALSE), plots the predictions for the random effects (only if random effects were included in the model fit).

pointcol

The color(s) of the data points.

linecol

The color(s) of the fitted curve (or color of the random effects curves if plotrandom=TRUE).

linetype

Line type for fitted curve (see options for lty in par).

linelwd

Width of the line (see options for lwd in par).

linecol2

The color of the fixed effects curve (if plotrandom=TRUE; otherwise ignored).

pxlinecol

The color of the lines indicating Px and its confidence interval

pxcex

Character size for the Px label above the Y-axis.

citype

Either 'polygon' (default), or 'lines', specifying formatting of the confidence interval in the plot.

cicol

The color of the confidence interval band (if plotted).

what

Either 'relk' or 'PLC' (or synonym 'embol'); it will plot either relative conductivity or percent loss conductivity (percent embolism).

selines

Obsolete; use px_ci

xaxis

Either 'positive' (default), so that water potential is plotted as positive values, or 'negative', plotting negative-valued water potentials.

...

Further parameters passed to plot, or points (when add=TRUE)

onepanel

For plotting of many curve fits, plot all curves in one panel (TRUE) or in separate panels (FALSE)

legend

(for fitconds and fitplcs only) Logical (default TRUE), whether to include a simple legend when plotting multiple fits

legendwhere

(for fitconds and fitplcs only) As in legend, specification of where to place legend (e.g. 'bottomleft'; coordinates not accepted)