The current implementation does not work very well for phases/lags close to 0 across most of the cross spectrum; at the very least, there should be an option to use the same colormap as the modulus, as well as a diverging colormap with more normalization options.
The current implementation does not work very well for phases/lags close to 0 across most of the cross spectrum; at the very least, there should be an option to use the same colormap as the modulus, as well as a diverging colormap with more normalization options.