Background colour when plotting masked Cube

[Simon Peatman]

I am using iris.plot to create a contourf of a Cube which has a masked array.  I would like all missing values to be greyed out.  Using matplotlib I can achieve this by setting axisbg when creating an axis object, but this has no effect in Iris:

import iris
import iris.plot as iplt
import matplotlib.pyplot as plt

cube = iris.load_cube('test.nc')

ax = plt.subplot(121, axisbg='0.5')
plt.contourf(cube.coord('longitude').points, cube.coord('latitude').points, cube.data)
ax = plt.subplot(122, axisbg='0.5')
iplt.contourf(cube)
plt.show()

As far as missing values are concerned, the first plot gives me exactly what I want but the second has the missing regions in white (see attached image).  Is there any way of setting the missing regions to grey in Iris plot?

[Andrew Dawson]

When you use subplot you are creating normal matplotlib axes, and when you use iris to plot a cube over a map (default for cubes with geographical coordinates) iris has to replace the provided axes with a GeoAxes instance from cartopy. When it does this replacement it doesn't carry all properties with it. The solution here is to set the background color *after* calling the iris plot function (probably something like ax.set_axis_bgcolor will do), or create GeoAxes from the get-go so iris doesn't have to replace them..