Internal function to produce the final DML and WAIS spectra (signal, noise, SNR) used for Figs. (3) and (4) in Münch and Laepple (2018). The DML spectra are a combination of the results from the DML1 and DML2 data sets.
Usage
PublicationSNR(
spec,
data = c("corr.full", "corr.diff.only", "corr.t.unc.only", "raw"),
dml.knit.f = 0.1,
df.log = 0.125
)Arguments
- spec
spectral results for the DML and WAIS firn core data as output from
WrapSpectralResults.- data
character string naming the version of the spectral results in
specto use; one of "corr.full", "corr.diff.only", "corr.t.unc.only", and "raw"; defaults to "corr.full".- dml.knit.f
frequency at which to combine the spectra from the DML1 and DML2 data sets (defaults to 1/10 yr^(-1)); DML2 spectra are used for lower frequencies than
dml.knit.f, DML1 for the higher frequencies.- df.log
width of Gaussian kernel in logarithmic frequency units for additional smoothing for visual purposes;
NULLsuppresses smoothing.
Value
A list with the components dml and wais, where each is
a list containing three spectral objects (`?spec.object`) and one numeric
vector:
signal:the signal spectrum.
noise:the noise spectrum.
snr:the frequency-dependent signal-to-noise ratio.
f.cutoff:a two-element vector with the index and value of the cutoff frequency from constraining the diffusion correction.
References
Münch, T. and Laepple, T.: What climate signal is contained in decadal- to centennial-scale isotope variations from Antarctic ice cores? Clim. Past, 14, 2053–2070, https://doi.org/10.5194/cp-14-2053-2018, 2018.