How to fix the limits of color scale in matplotlib?

Question

I am implementing a loop to produce contour plots using the function contourf in matplotlib. The objective of the study is to find out any moving patterns in the area. But, the plots produced are having different color scales. Some of them have -4 to 4 while others have -1.5 to 9.0 and so on which renders the interpretation pointless. How can I fix this color scale to -5.0 to 9.0?

Also, when I try to export the plots number of colorbars increases in each plot. For example the second plot in the loop has 2 colorbars and fifth plot has 5 colorbars and so on.

What I've done so far:

from mpl_toolkits.basemap import Basemap
import matplotlib.pyplot as plt
from numpy import linspace
from numpy import meshgrid

i=0
while i<len(inputdata):
    map = Basemap(projection='cyl', llcrnrlat=5.125, llcrnrlon=60.125, urcrnrlat=34.875, urcrnrlon=94.875)

    data = np.array(inputdata[i])

    x = linspace(map.llcrnrx, map.urcrnrx, data.shape[1])
    y = linspace(map.llcrnry, map.urcrnry, data.shape[0])

    xx, yy = meshgrid(x, y)

    map.contourf(xx, yy, data, cmap = 'summer_r')

    plt.colormap()
    plt.savefig('filename.jpg',dpi=300)
    i+=1

I meant plt.colorbar(). Sorry for the typo. Also, the data is huge and sort of an IP that inhibits me from sharing the same. Shape of the input data is (13,120,140) such that the loop produces 13 plots - but with different scales. — navajyothmp, May 10 '18 at 10:42
Just found another way. I added the parameter levels and set it to be from -5 to 9. That changed the color scale and colorbars of all plots. — navajyothmp, May 10 '18 at 11:22

DavidG · Accepted Answer · 2018-05-10T11:36:02.417

In order to change the limits of the colorbar, you can call plt.clim() before you call plt.colorbar and specify the range manually:

from mpl_toolkits.basemap import Basemap
import matplotlib.pyplot as plt
from numpy import linspace
from numpy import meshgrid

i=0
while i<len(inputdata):
    map = Basemap(projection='cyl', llcrnrlat=5.125, llcrnrlon=60.125, urcrnrlat=34.875, urcrnrlon=94.875)

    data = np.array(inputdata[i])

    x = linspace(map.llcrnrx, map.urcrnrx, data.shape[1])
    y = linspace(map.llcrnry, map.urcrnry, data.shape[0])

    xx, yy = meshgrid(x, y)

    map.contourf(xx, yy, data, cmap = 'summer_r')

    plt.clim(-5, 9)  # manually setup the range of the colorscale and colorbar
    plt.colormap()
    plt.savefig('filename.jpg',dpi=300)
    plt.clf()
    i+=1

You also may want to clear the figure after you have saved it to avoid multiple colorbars appearing using plt.clf()

How to fix the limits of color scale in matplotlib?

1 Answers1