Why can't one set to “False” specific axis ticklabels (ex: xlabels_top, ylabels_right) from a cartopy-geopandas plot?

问题

I am having serious difficulties in setting to False the xlabels_top and ylabels_right from my Geopandas plot.

This geopandas plot is made inside a Geoaxes subplot created with PlateCarree projection from Cartopy library.

My geopandas Geodataframe is in SIRGAS 2000 (units: degrees), EPSG: 4989. Therefore I created a Geodetic Globe object from the cartopy library.

Here is a code snippet:

import matplotlib.pyplot as plt
import cartopy.crs as ccrs
import geopandas as gpd

Geopandas_DF = gpd.read_file('my_file.shp')

# setting projection and Transform
Projection=ccrs.PlateCarree()
Transform = ccrs.Geodetic(globe=ccrs.Globe(ellipse='GRS80'))

Fig, Ax = plt.subplots(1,1, subplot_kw={'projection': Projection})

Geopandas_DF.plot(ax=Ax, transform=Ax.transData)

Ax.gridlines(crs=Projection , draw_labels=True, linewidth=0.5, 
             alpha=0.4, color='k', linestyle='--')

Ax.xlabels_top = False nn# It should turn off the upper x ticks
Ax.ylabels_right = False # It should turn off the right y ticks
Ax.ylabels_left = True
Ax.xlines = True

Fig.show()

Here is a figure example. One can notice that the xticks from the upper axis and the yticks from the right axis have not been turned OFF (False).

Therefore, I would like to know whether that is a problem between Cartopy and Geopandas, or am I doing something wrong in my code.

回答1:

The labels belong to the gridliner instance not the axes, you can turn them off there by storing the gridliner returned by the gridlines method and setting top_labels, right_labels as in:

import matplotlib.pyplot as plt
import cartopy.crs as ccrs
import geopandas as gpd

Geopandas_DF = gpd.read_file('my_file.shp')

# setting projection and Transform
Projection=ccrs.PlateCarree()
Transform = ccrs.Geodetic(globe=ccrs.Globe(ellipse='GRS80'))

Fig, Ax = plt.subplots(1,1, subplot_kw={'projection': Projection})

Geopandas_DF.plot(ax=Ax, transform=Ax.transData)

gl = Ax.gridlines(crs=Projection , draw_labels=True, linewidth=0.5, 
                  alpha=0.4, color='k', linestyle='--')

# For Cartopy <= 0.17
gl.xlabels_top = False
gl.ylabels_right = False
# For Cartopy >= 0.18
# gl.top_labels = False
# gl.right_labels = False

Fig.show()

回答2:

I finally managed to solve my problem. I used Ajdawson tips in order to do so. Here is the code with the solution:

import matplotlib.pyplot as plt
import cartopy.crs as ccrs
import geopandas as gpd

Geopandas_DF = gpd.read_file('my_file.shp')

# setting projection and Transform
Projection=ccrs.PlateCarree()
Transform = ccrs.Geodetic(globe=ccrs.Globe(ellipse='GRS80'))

Fig, Ax = plt.subplots(1,1, subplot_kw={'projection': Projection})

Geopandas_DF.plot(ax=Ax, transform=Ax.transData)

gl = Ax.gridlines(crs=Projection , draw_labels=True,    linewidth=0.5, 
                  alpha=0.4, color='k', linestyle='--')

gl.top_labels = False
gl.right_labels = False
gl.xlabels_top = False
gl.ylabels_right = False

Fig.show()

来源：https://stackoverflow.com/questions/54580278/why-cant-one-set-to-false-specific-axis-ticklabels-ex-xlabels-top-ylabels

标签

python-3.x

matplotlib

geopandas

cartopy