Python Geopandas: Exclude French Guiana from Europe Map

Question

I am writing a code for a map of europe with python geopandas.

I am currently facing a problem with French Guiana. I don't want it to display in the map, however, I don't find a way to detach it from France.

Here is my code:

import geopandas as gpd
import pandas as pd
import matplotlib.pyplot as plt


europe = gpd.read_file(gpd.datasets.get_path('naturalearth_lowres'))
europe = europe[europe.continent == 'Europe']
#europe = europe[europe.name != 'France']


data = pd.read_csv('HICP_EU_bycountry_12_2022.csv', delimiter=';')
data = data[['Area', 'Rate']]


merged_data = europe.merge(data, left_on='name', right_on='Area', how='left')


fig, ax = plt.subplots(figsize=(10, 6))

merged_data.plot(column='Rate', cmap='Reds', linewidth=0.8, ax=ax, edgecolor='0.8', legend=True)

ax.set_title('Inflation Rates in Europe', fontsize=16)
ax.set_axis_off()

for idx, row in merged_data.iterrows():
    rate = row['Rate']
    if not pd.isna(rate):
        ax.annotate(text=str(rate), xy=row['geometry'].centroid.coords[0], horizontalalignment='center', fontsize=8)

ax.set_facecolor('#f7f7f7')

plt.show()

Answer 1

As there is no separation between territories in this dataset, we can cut the polygons to your desired size using a bounding box (bbox) as a clip.

This is relatively simple, but we will need a small function and the Polygon class from Shapely to convert a bbox to a shape. In order to achieve this, we need to complete the following steps:

1. Get a bbox of your desired area
2. Convert it to a GeoDataFrame
3. Use it to clip europe.

From https://stackoverflow.com/a/68741143/18253502, we can convert bbox coordinates to a shapely Polygon. Bbox coordinates can be made at http://bboxfinder.com.

import geopandas as gpd
import pandas as pd
from shapely.geometry import Polygon
import matplotlib.pyplot as plt

# STEP 1 #
# Make polygon from bbox coordinates https://stackoverflow.com/a/68741143/18253502
def make_bbox(long0, lat0, long1, lat1):
    return Polygon([[long0, lat0],
                    [long1,lat0],
                    [long1,lat1],
                    [long0, lat1]])

# Coords covering Europe & Russia made with http://bboxfinder.com
bbox = make_bbox(-36.210938,28.304381,197.226563,81.361287)
## Alternatively, can clip to more standard European extent 
## with Central/Eastern Russia excluded
# bbox = make_bbox(-36.386719,29.228890,60.292969,74.543330)

# STEP 2 #
# Convert to gdf
bbox_gdf = gpd.GeoDataFrame(index=[0], crs='epsg:4326', geometry = [bbox])

# STEP 3 #
# Load europe
europe = gpd.read_file(gpd.datasets.get_path('naturalearth_lowres'))
europe = europe[europe.continent == 'Europe']

# Use bbox as clipping border for Europe
europe = europe.overlay(bbox_gdf, how="intersection")

Now europe has been clipped to the bbox extent:

# plot result
fig, ax = plt.subplots(figsize=(15,10))

europe.plot(linewidth=0.8, ax=ax, edgecolor='0.8', legend=True)
ax.set_title('Clipped Extent', fontsize=16)
ax.set_axis_off()
ax.set_facecolor('#f7f7f7')

plt.show()

Note that the Eastern tip of Russia is missing, as it wraps around to the other Longitude (see https://stackoverflow.com/a/71408567/18253502 for more details).

Answer 2

Alternatively, you can mess around with the MultiPolygon of France including Guiana and Corse. I extract the points from the polygon for France and make a new polygon to replace the multipolygon in the geo panda.

import geopandas as gpd
import matplotlib.pyplot as plt
from shapely.geometry import Polygon, MultiPolygon

# Load Europe map data from the geopandas library
world = gpd.read_file(gpd.datasets.get_path('naturalearth_lowres'))
europe = world[world.continent == 'Europe']

# Exclude French Guiana from the map (also Corsika though), 
tmp = [x.replace(')','') for x in str(europe.loc[43,'geometry']).split('((')[1:]][1]
tmp2 = [x.split(' ') for x in tmp.split(', ')][:-1]
tmp3 = [(float(x[0]),float(x[1])) for x in tmp2]
France_mainland = Polygon(tmp3)
europe.loc[europe['name']=='France','geometry'] = France_mainland