The cmocean function returns perceptually-uniform colormaps generated by Kristen Thyng. A detailed description of the cmocean project is available at http://matplotlib.org/cmocean.
If you find an occasion to cite these colormaps for any reason, or if you just want some nice beach reading, check out the following paper from the journal Oceanography:
Kristen M. Thyng, Chad A. Greene, Robert D. Hetland, Heather M. Zimmerle, and Steven F. DiMarco. True colors of oceanography: Guidelines for effective and accurate colormap selection. Oceanography, September 2016. http://dx.doi.org/10.5670/oceanog.2016.66
cmocean cmap = cmocean('ColormapName') cmap = cmocean('-ColormapName') cmap = cmocean(...,NLevels) cmap = cmocean(...,'pivot',PivotValue) cmap = cmocean(...,'negative') cmocean(...)
cmocean without any inputs displays colormap options.
cmap = cmocean('ColormapName') returns a 256x3 colormap. ColormapName can be any of of the following:
cmap = cmocean('-ColormapName') a minus sign preceeding any ColormapName flips the order of the colormap.
cmap = cmocean(...,NLevels) specifies a number of levels in the colormap. Default value is 256.
cmap = cmocean(...,'pivot',PivotValue) centers a diverging colormap such that the point of color divergence corresponds to a specified value and maximum extents are set using current caxis limits. If no PivotValue is set, 0 is assumed. Early versions of this function used 'zero' as the syntax for 'pivot',0 and the old syntax is still supported.
cmap = cmocean(...,'negative') inverts the lightness profile of the colormap. This can be useful particularly for divergent colormaps if the default white point of divergence gets lost in a white background.
cmocean(...) without any outputs sets the current colormap to the current axes.
Using this sample plot:
Set the colormap to 'algae':
Same as above, but with an inverted algae colormap:
Set the colormap to a 12-level 'solar':
Get the RGB values of a 5-level thermal colormap:
RGB = cmocean('thermal',5)
RGB = 0.02 0.14 0.20 0.34 0.23 0.61 0.69 0.37 0.51 0.98 0.57 0.26 0.91 0.98 0.36
Some of those values are below zero and others are above. If this dataset represents anomalies, perhaps a diverging colormap is more appropriate:
It's unlikely that the center value of this color axis 1.7776 is an interesting value about which the data diverges. If you want to center the colormap on zero using the current color axis limits, simply include the 'pivot' option:
I was recently in a situation where I had a scatter plot of divergent data on a white background, but cmocean('balance') made the most important datapoints white and nearly invisible. Here's the situation:
% Some sample data with noise: x = 10*rand(300,1)-5; noise = 2*randn(size(x)); y = x.^2+noise; % A theoretical perfect x^2 line: x_theoretical = linspace(min(x),max(x),50); y_theoretical = x_theoretical.^2; % Plot the data: figure plot(x_theoretical,y_theoretical,'k-') hold on scatter(x,y,25,noise,'filled') cb = colorbar; ylabel(cb,'error relative to theory') box off axis tight
I wanted to show how far off my noisy data was from the perfect theoretical x-squared line, so the divergent cmocean('balance') map seemed appropriate:
But in the plot above, attention is drawn away from the theoretical line toward the dark red and dark blue outliers. In this case, a negative colormap may be preferred:
This function was written by Chad A. Greene of the Institute for Geophysics at the University of Texas at Austin (UTIG), June 2016, using colormaps created by Kristen Thyng of Texas A&M University, Department of Oceanography. More information on the cmocean project can be found at http://matplotlib.org/cmocean.