The result is:

Draw two axis to one colorbar using python and matplotlib.pyplot

In my post in 14th, Oct. 2018, I showed how to draw second tick axis on the colorbar. The method of plotting is not so good, so I'd like to show an improved way to draw second tick axis on the colorbar. This page shows how to draw two axis to one colorbar using python and matplotlib.pyplot. By adapting twinx to the colorbar axis and by shifting the position of labels and ticks, you can draw an figure like the example on this page.

See also:

Python Matplotlib Tips: Draw second colorbar axis outside of first axis

This code shows how to draw second colorbar ticks outside of first colorbar axis.

Python Matplotlib Tips: Add second x-axis below first x-axis using Python and matplotlib.pyplot

This page shows how to draw second x-axis below the first x-axis. Using "twinx", "set_ticks_position", and "set_label_position", and "spines['bottom'].set_position", you can move the second x-axis from the top of the figure to below the first x-axis.

In [1]:

import platform
print('python: '+platform.python_version())
import matplotlib.pyplot as plt
from matplotlib import __version__ as matplotlibversion
print('matplotlib: '+matplotlibversion)
import numpy as np
print('numpy: '+np.__version__)
%matplotlib inline

python: 3.6.3
matplotlib: 3.0.1
numpy: 1.15.4

In [2]:

import matplotlib.pyplot as plt
import numpy as np
%matplotlib inline

In [3]:

plt.figure(facecolor='w',figsize=(5,4))
# generate random data
x = np.random.randint(0,200,(11,11))
dmin,dmax = 0,200
plt.imshow(x,vmin=dmin,vmax=dmax)

# create the colorbar
# the aspect of the colorbar is set to 'equal', we have to set it to 'auto',
# otherwise twinx() will do weird stuff.
# ref: Draw colorbar with twin scales - stack overflow -
# URL: https://stackoverflow.com/questions/27151098/draw-colorbar-with-twin-scales
cbar = plt.colorbar()
pos = cbar.ax.get_position()
ax1 = cbar.ax
ax1.set_aspect('auto')

# create a second axis and specify ticks based on the relation between the first axis and second aces
ax2 = ax1.twinx()
ax2.set_ylim([dmin,dmax])
newlabel = [300,325,350,375,400,425,450] # labels of the ticklabels: the position in the new axis
k2degc = lambda t: t-273.15 # convert function: from Kelvin to Degree Celsius
newpos   = [k2degc(t) for t in newlabel]   # position of the ticklabels in the old axis
ax2.set_yticks(newpos)
ax2.set_yticklabels(newlabel)

# resize the colorbar
pos.x0 += 0.10
pos.x1 += 0.10


# arrange and adjust the position of each axis, ticks, and ticklabels
ax1.set_position(pos)
ax2.set_position(pos)
ax1.yaxis.set_ticks_position('right') # set the position of the first axis to right
ax1.yaxis.set_label_position('right') # set the position of the fitst axis to right
ax1.set_ylabel(u'Temperature [\u2103]')
ax2.yaxis.set_ticks_position('left') # set the position of the second axis to right
ax2.yaxis.set_label_position('left') # set the position of the second axis to right
# ax2.spines['left'].set_position(('outward', 50)) # adjust the position of the second axis
ax2.set_ylabel('Temperature [K]')

# plt.tight_layout()
# Save the figure
plt.savefig('color_ticks_right_left.png', bbox_inches='tight', pad_inches=0.02, dpi=150)

Draw two axis to one colorbar using python and matplotlib.pyplot