"""
`chaosmagpy.plot_utils` provides functions for plotting model outputs.
.. autosummary::
:toctree: functions
plot_timeseries
plot_maps
plot_power_spectrum
nio_colormap
"""
import numpy as np
import matplotlib.pyplot as plt
import matplotlib.ticker as ticker
import matplotlib.dates as mdates
import warnings
from matplotlib.colors import LinearSegmentedColormap
from . import data_utils
from . import config_utils
try: # make cartopy optional
import cartopy.crs as ccrs
except ImportError:
warnings.warn('Could not import Cartopy package. Plotting data on maps '
'is not available in chaosmagpy.')
[docs]def plot_timeseries(time, *args, **kwargs):
"""
Creates line plots for the timeseries of the input arguments.
Parameters
----------
time : ndarray, shape (N,)
Array containing time in modified Julian dates.
*args : ndarray, shape (N, k)
Array containing `k` columns of values to plot against time. Several
arrays can be provided as separated arguments.
Other Parameters
----------------
figsize : 2-tuple of floats
Figure dimension (width, height) in inches.
titles : list of strings
Subplot titles (defaults to empty strings).
ylabel : string
Label of the vertical axis (defaults to an empty string).
layout : 2-tuple of int
Layout of the subplots (defaults to vertically stacked subplots).
**kwargs : keywords
Other options to pass to matplotlib plotting method.
Notes
-----
For more customization get access to the figure and axes handles
through matplotlib by using ``fig = plt.gcf()`` and ``axes = fig.axes``
right after the call to this plotting function.
"""
n = len(args) # number of subplots
defaults = {
'figsize': (config_utils.basicConfig['plots.figure_width'],
n*0.8*config_utils.basicConfig['plots.figure_width']),
'titles': n*[''],
'ylabel': '',
'layout': (n, 1)
}
kwargs = defaultkeys(defaults, kwargs)
# remove keywords that are not intended for plot
figsize = kwargs.pop('figsize')
layout = kwargs.pop('layout')
titles = kwargs.pop('titles')
ylabel = kwargs.pop('ylabel')
if layout[0]*layout[1] != n:
raise ValueError('Plot layout is not compatible with the number of '
'produced subplots.')
date_time = data_utils.timestamp(np.ravel(time))
fig, axes = plt.subplots(layout[0], layout[1], sharex='all',
figsize=figsize, squeeze=False)
for ax, component, title in zip(axes.flat, args, titles):
ax.plot(date_time, component, **kwargs)
ax.set_title(title)
ax.grid()
ax.set(ylabel=ylabel, xlabel='time')
fig.autofmt_xdate()
ax.fmt_xdata = mdates.DateFormatter('%Y-%m-%d-%h')
fig.tight_layout(rect=(0, 0.02, 1, 1))
[docs]def plot_maps(theta_grid, phi_grid, *args, **kwargs):
"""
Plots global maps of the input arguments.
Parameters
----------
theta_grid : ndarray
Array containing the colatitude in degrees.
phi_grid : ndarray
Array containing the longitude in degrees.
*args : ndarray
Array of values to plot on the global map. Several
arrays can be provided as separated arguments.
Other Parameters
----------------
figsize : 2-tuple of floats
Figure dimension (width, height) in inches.
titles : list of strings
Subplot titles (defaults to empty strings).
label : string
Colorbar label (defaults to an empty string).
layout : 2-tuple of int
Layout of the subplots (defaults to vertically stacked subplots).
cmap : str
Colormap code (defaults to ``'PuOr_r'`` colormap).
limiter : function, lambda expression
Function to compute symmetric colorbar limits (defaults to maximum of
the absolute values in the input, or use ``'vmin'``, ``'vmax'``
keywords instead).
projection : :mod:`cartopy.crs`
Projection of the target frame (defaults to
:func:`cartopy.crs.Mollweide()`).
transform : :mod:`cartopy.crs`
Projection of input frame (defaults to
:func:`cartopy.crs.PlateCarree()`)
**kwargs : keywords
Other options to pass to matplotlib :func:`pcolormesh` method.
Notes
-----
For more customization get access to the figure and axes handles
through matplotlib by using ``fig = plt.gcf()`` and ``axes = fig.axes``
right after the call to this plotting function.
"""
n = len(args) # number of plots
defaults = {
'figsize': (config_utils.basicConfig['plots.figure_width'],
n*0.4*config_utils.basicConfig['plots.figure_width']),
'titles': n*[''],
'label': '',
'layout': (n, 1),
'cmap': 'PuOr_r',
'limiter': lambda x: np.amax(np.abs(x)), # maximum value
'projection': ccrs.Mollweide(),
'transform': ccrs.PlateCarree()
}
kwargs = defaultkeys(defaults, kwargs)
# remove keywords that are not intended for pcolormesh
figsize = kwargs.pop('figsize')
titles = kwargs.pop('titles')
label = kwargs.pop('label')
limiter = kwargs.pop('limiter')
projection = kwargs.pop('projection')
layout = kwargs.pop('layout')
if layout[0]*layout[1] != n:
raise ValueError('Plot layout is not compatible with the number of '
'produced subplots.')
# create axis handle
fig, axes = plt.subplots(layout[0], layout[1], sharex=True, sharey=True,
subplot_kw=dict(projection=projection),
figsize=figsize, squeeze=False)
# make subplots
for ax, component, title in zip(axes.flat, args, titles):
# evaluate colorbar limits depending on vmax/vmin in kwargs
kwargs.setdefault('vmax', limiter(component))
kwargs.setdefault('vmin', -limiter(component))
# produce colormesh and evaluate keywords (defaults and input)
pc = ax.pcolormesh(phi_grid, 90. - theta_grid, component, **kwargs)
ax.gridlines(linewidth=0.5, linestyle='dashed',
ylocs=np.linspace(-90, 90, num=7), # parallels
xlocs=np.linspace(-180, 180, num=13)) # meridians
ax.coastlines(linewidth=0.5)
clb = plt.colorbar(pc, ax=ax, format=ticker.FuncFormatter(fmt),
extend='both')
clb.set_label(label)
ax.set_global()
ax.set_title(title)
fig.tight_layout()
[docs]def plot_power_spectrum(spectrum, **kwargs):
"""
Plot the spherical harmonic spectrum.
Parameters
----------
spectrum : ndarray, shape (N,)
Spherical harmonics spectrum of degree `N`.
Other Parameters
----------------
figsize : 2-tuple of floats
Figure dimension (width, height) in inches.
titles : list of strings
Subplot titles (defaults to empty strings).
ylabel : string
Label of the vertical axis (defaults to an empty string).
**kwargs
Keywords passed to :func:`matplotlib.pyplot.semilogy`
Notes
-----
For more customization get access to the figure and axes handles
through matplotlib by using ``fig = plt.gcf()`` and ``axes = fig.axes``
right after the call to this plotting function.
"""
defaults = {
'figsize': (config_utils.basicConfig['plots.figure_width'],
0.8*config_utils.basicConfig['plots.figure_width']),
'titles': '',
'ylabel': ''
}
kwargs = defaultkeys(defaults, kwargs)
figsize = kwargs.pop('figsize')
titles = kwargs.pop('titles')
ylabel = kwargs.pop('ylabel')
degrees = np.arange(1, spectrum.shape[0] + 1, step=1.0)
spectrum[spectrum == 0] = np.nan # remove non-positive values for log
# create axis handle
fig, ax = plt.subplots(1, 1, sharex=True, sharey=True, figsize=figsize)
ax.semilogy(degrees, spectrum, **kwargs)
ax.set_title(titles)
ax.grid(b=True, which='minor', linestyle=':')
ax.grid(b=True, which='major', linestyle='-', axis='both')
ax.set(ylabel=ylabel, xlabel='degree')
ax.xaxis.set_major_locator(ticker.MaxNLocator(integer=True))
ax.set_xlim((0, degrees[-1]))
fig.tight_layout()
def fmt(x, pos):
# format=ticker.FuncFormatter(fmt)
a, b = '{:.1e}'.format(x).split('e')
b = int(b)
if a == '0.0':
return r'${}$'.format(a)
else:
return r'${}$e${}$'.format(a, b)
def defaultkeys(defaults, keywords):
"""
Return dictionary of default keywords. Overwrite any keywords in
``defaults`` using keywords from ``keywords``, except if they are None.
Parameters
----------
defaults, keywords : dict
Dictionary of default and replacing keywords.
Returns
-------
keywords : dict
"""
# overwrite value with the one in kwargs, if not then use the default
for key, value in defaults.items():
if keywords.setdefault(key, value) is None:
keywords[key] = value
return keywords
[docs]def nio_colormap():
"""
Define custom-built colormap 'nio' and register. Can be called in plots as
``cmap='nio'`` after importing this module.
.. plot::
:include-source: false
import numpy as np
import chaosmagpy as cp
import matplotlib.pyplot as plt
gradient = np.linspace(0, 1, 256)
gradient = np.vstack((gradient, gradient))
figh = 0.35 + 0.15 + 0.22
fig, ax = plt.subplots(1, 1, figsize=(6.4, figh))
fig.subplots_adjust(top=0.792, bottom=0.208, left=0.023, right=0.977)
ax.imshow(gradient, aspect='auto', cmap='nio')
# ax.set_axis_off()
ax.get_xaxis().set_visible(False)
ax.get_yaxis().set_visible(False)
plt.show()
"""
cdict = {'red': ((0.0000, 0.0000, 0.0000),
(0.1667, 0.0000, 0.0000),
(0.3333, 0.0000, 0.0000),
(0.5020, 1.0000, 1.0000),
(0.6667, 1.0000, 1.0000),
(0.8333, 1.0000, 1.0000),
(1.0000, 1.0000, 1.0000)),
'green': ((0.0000, 0.0000, 0.0000),
(0.1667, 0.0000, 0.0000),
(0.3333, 1.0000, 1.0000),
(0.5020, 1.0000, 1.0000),
(0.6667, 1.0000, 1.0000),
(0.8333, 0.0000, 0.0000),
(1.0000, 0.0000, 0.0000)),
'blue': ((0.0000, 0.0000, 0.0000),
(0.1667, 1.0000, 1.0000),
(0.3333, 1.0000, 1.0000),
(0.5020, 1.0000, 1.0000), # >0.5 for white (intpl bug?)
(0.6667, 0.0000, 0.0000),
(0.8333, 0.0000, 0.0000),
(1.0000, 1.0000, 1.0000))}
return LinearSegmentedColormap('nio', cdict)
# register cmap name for convenient use
plt.register_cmap(cmap=nio_colormap())
if __name__ == '__main__':
pass