cartopy

cartopy copied to clipboard

I need to align a cartopy 2D map subplot with a 1D line subplot (zonal-mean or meridional mean of the 2D data), using the following code:

import xarray as xr
import cartopy.crs as ccrs
import matplotlib.pyplot as plt


air = xr.tutorial.open_dataset('air_temperature').air[0]

fig = plt.figure(figsize=(14,5))

ax1 = fig.add_subplot(121,
                 projection=ccrs.PlateCarree(central_longitude=180))
air.plot.pcolormesh(ax=ax1,
               transform=ccrs.PlateCarree(),
               cbar_kwargs={'orientation':'horizontal'})
ax1.coastlines()
ax1.set_extent((air.lon[0], air.lon[-1], air.lat[-1], air.lat[0]),
               crs=ccrs.PlateCarree())

ax2 = fig.add_subplot(122)
air.mean('lon').plot(ax=ax2, y='lat')

plt.tight_layout()

This gives the plot as this.

Both subplots have the same y- (latitude-) range and I expect the boxes of the two axes could align with each other. Is there a simple way to do this? This has been proposed on stackoverflow.

I have not tested, but I would guess that you would have to set the aspect ratio of the plain Axes to something that would match the Cartopy Axes.

This example in the Iris gallery demonstrates how Cartopy and Matplotlib can be used to achieve this.

I would use compressed layout rather than tight layout. I added an answer on StackOverflow.

Edit: sorry I only just noticed how old this issue is. But now that compressed layout exists as a solution in mpl, perhaps this issue can be closed?

Thanks for your suggestions. I've already got this resolved using proplot as a workaround.

However, if one uses a projection (e.g., Robin) other than PlateCarree, then the latitudes would be slightly stretched. In this case can we still have a equivalent zonal-mean plot aligned with the projection?

Just tested with Mollweide and this is indeed a problem