dplR (version 1.7.0)

wavelet.plot: Plot a Continuous Wavelet Transform

Description

This function creates a filled.contour plot of a continuous wavelet transform as output from morlet.

Usage

wavelet.plot(wave.list,
             wavelet.levels = quantile(wave.list$Power,
                                       probs = (0:10)/10),
             add.coi = TRUE, add.sig = TRUE,
             x.lab = gettext("Time", domain = "R-dplR"),
             period.lab = gettext("Period", domain = "R-dplR"),
             crn.lab = gettext("RWI", domain = "R-dplR"),
             key.cols = rev(rainbow(length(wavelet.levels)-1)),
             key.lab = parse(text=paste0("\"",
                                         gettext("Power",
                                                 domain="R-dplR"),
                                         "\"^2")),
             add.spline = FALSE, f = 0.5, nyrs = NULL,
             crn.col = "black", crn.lwd = 1,coi.col='black',
             crn.ylim = range(wave.list$y) * c(0.95, 1.05),
             side.by.side = FALSE,
             useRaster = FALSE, res = 150, reverse.y = FALSE, …)

Arguments

wave.list

A list. Output from morlet.

wavelet.levels

A numeric vector. Values for levels of the filled contours for the wavelet plot.

add.coi

A logical flag. Add cone of influence?

add.sig

A logical flag. Add contour lines for significance?

x.lab

X-axis label.

period.lab

Y-axis label for the wavelet plot.

crn.lab

Y-axis label for the time-series plot.

key.cols

A vector of colors for the wavelets and the key.

key.lab

Label for key.

add.spline

A logical flag. Add a spline to the time-series plot using ffcsaps?.

nyrs

A number giving the rigidity of the smoothing spline, defaults to 0.33 of series length if nyrs is NULL.

f

A number between 0 and 1 giving the frequency response or wavelength cutoff for the smoothing spline. Defaults to 0.5.

crn.col

Line color for the time-series plot.

crn.lwd

Line width for the time-series plot.

coi.col

Color for the COI if add.coi is TRUE.

crn.ylim

Axis limits for the time-series plot.

side.by.side

A logical flag. Plots will be in one row if TRUE.

useRaster

A logical flag. If TRUE, the filled contours are drawn as a raster image. Other parts of the plot are not affected. useRaster=TRUE can be especially useful when a pdf device is used: the size and complexity of the PDF file will probably be greatly reduced. Setting this to TRUE has negative effects when used with a bitmap device such as png. If NA, plotting of a raster image will be attempted if and only if the name of the graphics device is "pdf" or "postscript". The default is FALSE: draw directly to the graphics device without using an intermediate raster image.

res

A numeric vector of length 1. The resolution (pixels per inch) of the filled contours when a raster image is used. See useRaster.

reverse.y

A logical flag. If TRUE, the Y-axis will be reversed, i.e. period increasing towards the bottom. The default is FALSE.

…

Arguments passed to rasterPlot. Only relevant when the filled contours are drawn as a raster image. See useRaster.

Value

None. This function is invoked for its side effect, which is to produce a plot.

Details

This produces a plot of a continuous wavelet transform and plots the original time series. Contours are added for significance and a cone of influence polygon can be added as well. Anything within the cone of influence should not be interpreted.

The time series can be plotted with a smoothing spline as well.

References

Torrence, C. and Compo, G. P. (1998) A practical guide to wavelet analysis. Bulletin of the American Meteorological Society, 79(1), 61<U+2013>78.

See Also

morlet, ffcsaps

Examples

# NOT RUN {
library(stats)
library(utils)
data(ca533)
ca533.rwi <- detrend(rwl = ca533, method = "ModNegExp")
ca533.crn <- chron(ca533.rwi, prefix = "CAM", prewhiten = FALSE)
Years <- time(ca533.crn)
CAMstd <- ca533.crn[, 1]
out.wave <- morlet(y1 = CAMstd, x1 = Years, p2 = 9, dj = 0.1,
                   siglvl = 0.99)
wavelet.plot(out.wave, useRaster = NA)
# Alternative palette with better separation of colors
# via: rev(RColorBrewer::brewer.pal(10, "Spectral"))
specCols <- c("#5E4FA2", "#3288BD", "#66C2A5", "#ABDDA4", "#E6F598", 
              "#FEE08B", "#FDAE61", "#F46D43", "#D53E4F", "#9E0142")
wavelet.plot(out.wave, key.cols=specCols,useRaster = NA)

# fewer colors
levs <- quantile(out.wave$Power, probs = c(0, 0.5, 0.75, 0.9, 0.99))
wavelet.plot(out.wave, wavelet.levels = levs, add.sig = FALSE,
             key.cols = c("#FFFFFF", "#ABDDA4", "#FDAE61", "#D7191C"), useRaster = NA)
# }