wavelet.plot {dplR} | R Documentation |
This function creates a filled.contour
plot of a continuous wavelet
transform using the Morlet wavelet.
wavelet.plot(crn.vec,yr.vec,p2,dj=0.25,siglvl=0.99,...)
crn.vec |
a vector of values for the wavelet transform. |
yr.vec |
a vector of values giving the years for the plot. Must be the
same length as length(crn.vec) . |
p2 |
the numbers of power of two to be computed for the wavelet transform. |
dj |
sub-octaves per octave calculated. |
siglvl |
level for the significance test. Defaults to 0.99. |
... |
other arguments to pass to filled.contour. |
This produces a plot of a continuous wavelet transform. Its implementation
very closely follows Torrence and Compo (1998). The user provides a tree-ring
chronology (although detrended series are conceivably useful as well),
the years for the plot, the powers of two (for the scale parameter), and the
confidence level for the significance test. The function assumes that the data
are yearly and defaults to calculating four sub-octaves per octave (four
voices per power of two). The input crn.vec
is padded up to the next
power of two before the transform and the padding is removed before plotting.
Currently the Morlet wavelet is the only wavelet implemented; the wavenumber
(k0
) is fixed at six. In future releases, other wavelets will be
available (Dog, Paul, etc.). Similarly, a chi-square distribution is
used to assess significance at the level indicated. In future versions,
significance will be calculated against the global wavelet spectrum, or a
red-noise background.
The filled.contour levels are determined using
quantile(Power,probs=seq(0,1,0.1))
. A contour for significance is
displayed as is the cone of influence. Anything within the cone of influence
should not be interpreted.
Refer to Torrence and Compo (1998) for details on the transform, significance, etc.
None. This function is invoked for its side effect, which is to produce a plot.
The functions wavelet
and morlet
are ports of Torrence's
IDL code available at
http://atoc.colorado.edu/research/wavelets/software.html
Andy Bunn
Torrence, C. and Compo, G.P. (1998) A practical guide to wavelet analysis. Bulletin of the American Meteorological Society, 79: 61–78.
data(ca533) ca533.rwi <- detrend(rwl = ca533, method = "ModNegExp") ca533.crn <- chron(ca533.rwi, prefix = "CAM", prewhiten = FALSE) Years <- as.numeric(rownames(ca533.crn)) CAMstd <- ca533.crn[,1] wavelet.plot(CAMstd,Years,p2=9,siglvl=0.99,main="CAMstd")