epygram.spectra — Class and functions for spectra

This module contains:

• a class to handle variance spectrum;
• a function to compute DCT spectrum from a 2D field;
• a function to sort spectra with regards to their name;
• a function to plot a series of spectra.

Classes

class epygram.spectra.Spectrum(variances, name=None, resolution=None, mean2=None, **kwargs)

Bases: epygram.util.RecursiveObject

A spectrum can be seen as a quantification of a signal’s variance with regards to scale. If the signal is defined in physical space on N points, its spectral representation will be a squared mean value (wavenumber 0) and variances for N-1 wavenumbers. For details and documentation, see

Denis et al. (2002) : ‘Spectral Decomposition of Two-Dimensional Atmospheric Fields on Limited-Area Domains Using the Discrete Cosine Transform (DCT)’

dump(filename)

Writes the spectrum with formatted output in filename.

wavelengths

Gets the wavelengths of the spectrum.

wavenumbers

Gets the wavenumbers of the spectrum.

write(out)

Writes the spectrum with formatted output in out.

out must be an output open file-like object (out.write() only is needed).

Functions

epygram.spectra.dctspectrum(x, log=None, verbose=False)

Function dctspectrum takes a 2D-array as argument and returns its 1D DCT ellipse spectrum.

For details and documentation, see

Denis et al. (2002) : ‘Spectral Decomposition of Two-Dimensional Atmospheric Fields on Limited-Area Domains Using the Discrete Cosine Transform (DCT).’

log is an optional logging.Logger instance to which print info in verbose case.

epygram.spectra.sort(spectra)

Sort a list of spectra with regards to their name.

epygram.spectra.plotspectra(spectra, slopes=[{'label': '-3', 'exp': -3, 'offset': 1}, {'label': '-5/3', 'exp': -1.6666666666666667, 'offset': 1}], zoom=None, unit='SI', title=None)

To plot a series of spectra.

Args:

• spectra = a Spectrum instance or a list of.

• unit: string accepting LaTeX-mathematical syntaxes

• slopes = list of dict( - exp=x where x is exposant of a A*k**-x slope - offset=A where A is logscale offset in a A*k**-x slope;

  a offset=1 is fitted to intercept the first spectra at wavenumber = 2

  • label=(optional label) appearing ‘k = label’ in legend)

• zoom = dict(xmin=,xmax=,ymin=,ymax=)

• title = string for title