SJPlot: A Fidelity Plot

In MJKowalewski/PaleoFidelity: Analyze and Visualize Agreement between Live and Dead Communities

A Fidelity Plot

Description

SJPlot function produces a bivariate plot of a correlation measure (x axis) and similarity measure (y axis) to depict live-dead fidelity at one or more sites.

Usage

SJPlot(
  x,
  bubble = FALSE,
  xlim = c(-1, 1),
  ylim = c(0, 1),
  trans = 0.3,
  cex = 1,
  cex.mean = 2,
  cex.model = 1,
  cex.model.mean = 1,
  cex.bubble = 2,
  cex.legend = 0.8,
  axes = T,
  pch = 21,
  col = "black",
  col.model = "green",
  col.mean = "white",
  col.model.mean = "green4",
  gpcol = NULL,
  pch2 = 21,
  pch3 = ".",
  pch4 = 21,
  PF = FALSE,
  CI = TRUE,
  adjF = FALSE,
  ssF = FALSE,
  unadjF = TRUE,
  addlegend = FALSE,
  addinfo = FALSE,
  addbubble = FALSE,
  info.y = 0.5,
  bubble.y = -0.15,
  xlab = NULL,
  ylab = NULL
)

Arguments

x	An object of the class 'FidelityEst' returned by FidelityEst function.
bubble	Logical (default=FALSE): to produce a bubble plot with symbols scaled by N-min (the number of observations in the smaller of the two (live vs. dead) compared samples).
xlim	A vector with two numerical values defining x axis limits (default = c(-1, 1)).
ylim	A vector with two numerical values defining y axis limits (default = c(0, 1)).
trans	A numerical value (default = 0.3) defining transparency of background fill for symbols.
cex	A numerical value (default = 1) defining symbol size (applicable if Bubble = FALSE).
cex.mean	A numerical value (default = 2) defining symbol size for means across all samples or sample groups.
cex.model	A numerical value (default = 1) defining symbol size for individual model estimates.
cex.model.mean	A numerical value (default = 1) defining symbol size for mean sample model estimates.
cex.bubble	A numerical value (default = 2) defining scaling factor for bubbles.
cex.legend	A numerical value (default = 1) defining text size for legend.
axes	Logical (default = TRUE): to determine if axes should be plotted
pch	An integer or a single character (default = 21) specifying symbol type.
col	A character string (default = 'black') defining symbol color.
col.model	A character string (default = 'gray') defining symbol color for model points.
col.mean	A character string (default = 'black') defining symbol color for the mean.
col.model.mean	A character string (default = 'white') defining symbol color for model sample means.
gpcol	Color names (default=1:length(levels(x$gp))) defining colors for sample groups, applicable when 'gp' factor was provided in FidelityEst function. If custom colors are provided, the number of colors must match number of levels in a 'gp' factor.
pch2	An integer or a single character (default=21) specifying symbol type for grand mean or group means.
pch3	An integer or a single character (default='.') specifying symbol type for simulated sample values.
pch4	An integer or a single character (default=21) specifying symbol type for simulated mean sample values.
PF	Logical (default=FALSE): adds a scatter plot of resampled fidelity estimates for the null model of perfect fidelity.
CI	Logical (default=TRUE): adds confidence intervals to corrected fidelity estimates
adjF	Logical (default=FALSE): plots adjusted fidelity estimates
ssF	Logical (default=FALSE): plots sample-standardized fidelity estimates
unadjF	Logical (default=TRUE): plots non-adjusted fidelity estimates
addlegend	Logical (default=FALSE): adds legend to the plot, if 'gp' factor is provided.
addinfo	Logical (default=FALSE): prints parameter values above the plot
addbubble	Logical (default=FALSE): prints legend relating bubble sizes to N-min, which denotes the number of specimens in the smaller of the two compared samples (usually, 'live' sample is the smaller of the two)
info.y	Numeric (default = 0.5): specifies how far above the plot should "addinfo" legend (see above) be plotted. This is "line" parameter value in mtext function)
bubble.y	Numeric (default = - 0.15): specifies how far above the plot should "addbubble" legend (see above) be plotted. This is "inset" parameter value in legend function)
xlab	Character string that provides x-axis label. The default value is inherited from FidelityEst object based on a correlation measure used there.
ylab	Character string that provides y-axis label. The default value is inherited from FidelityEst object based on a similarity measure used there.

Details

SJPlot function uses the object of the class 'FidelityEst', produced by FidelityEst function, to generate a live-dead fidelity plot. If default arguments for fidelity measures are used in FidelityEst function, a Spearman vs. Jaccard-Chao fidelity plot (as in Kidwell, 2007) is produced. If a grouping factor is provided, symbols are color-coded by levels and group means are plotted. Bivariate distributions produced by a resampling model can be included.

NOTE: This function utilizes plot.default function, including some of its arguments. It allows to explore visually various alternative estimates of fidelity.

Value

A single bivariate plot.

References

Kidwell, S.M., 2007, Discordance between living and death assemblages as evidence for anthropogenic ecological change. Proc Natl Acad Sci USA 104(45): 17701–17706.

Examples

out1 <- FidelityEst(FidData$live[6:9,], FidData$dead[6:9,], FidData$habitat[6:9],
                   n.filters=30, iter=100, t.filters=1)
SJPlot(out1)

MJKowalewski/PaleoFidelity documentation built on Aug. 25, 2024, 8:27 p.m.