Choropleth map with a customized legend

We will create a Choropleth map using the Matplotlib library to visualize data about CO2 emissions per capita across Europe.

This chart features a unique legend, distinctive annotations, and a vibrant colormap, resulting in an attractive output.

This post provides a step-by-step guide to reproduce the map, starting with a basic background map.

About

This plot is a choropleth map. It show the CO2 emissions per capita in 2021 for each country.

The chart was made by Joseph Barbier. Thanks to him for accepting sharing his work here!

Let's see what the final picture will look like:

choropleth map with custom legend

Libraries

First, you need to install the following librairies:

matplotlib is used for creating the chart and add customization features
pandas and geopandas are used to put the data into a dataframe and manipulate geographical data
drawarrow for the arrows
pyfonts for the fonts
geopandas: for the geographical data
highlight_text: to add beautiful annotations to the chart

And that's it!

import pandas as pd
import geopandas as gpd
import matplotlib.pyplot as plt
from pypalettes import load_cmap
from matplotlib.font_manager import FontProperties
from drawarrow import fig_arrow
from pyfonts import load_font

from highlight_text import fig_text, ax_text

Dataset

Creating a choropleth map necessitates a dataset with geographical information. For this example, we need to load the world map from the Gallery's repo.

Next, we should load the dataset containing the values we want to represent on the map, such as CO2 per capita by country. This dataset is also accessible in the Gallery's repo.

url = "https://raw.githubusercontent.com/holtzy/the-python-graph-gallery/master/static/data/europe.geojson"
world = gpd.read_file(url)
world.head()

	pop_est	continent	name	iso_a3	gdp_md_est	geometry
0	144373535.0	Europe	Russia	RUS	1699876	MULTIPOLYGON (((180.00000 71.51571, 180.00000 ...
1	5347896.0	Europe	Norway	NOR	403336	MULTIPOLYGON (((15.14282 79.67431, 15.52255 80...
2	67059887.0	Europe	France	FRA	2715518	MULTIPOLYGON (((-51.65780 4.15623, -52.24934 3...
3	10285453.0	Europe	Sweden	SWE	530883	POLYGON ((11.02737 58.85615, 11.46827 59.43239...
4	9466856.0	Europe	Belarus	BLR	63080	POLYGON ((28.17671 56.16913, 29.22951 55.91834...

Now we merge this dataset with the world map with a merge function. This function will match the country names in the two datasets and add the CO2 values to the world map dataframe.

url = "https://raw.githubusercontent.com/holtzy/the-python-graph-gallery/master/static/data/co2PerCapita.csv"
df = pd.read_csv(url)

# merge data
data = world.merge(df, how='left', left_on='name', right_on='Country')

# filter to keep only specific countries
data = data[data['continent'] == 'Europe']
data = data[~data['name'].isin(['Russia', 'Iceland'])]
data = data[data['Year'] == 2021]
data = data[['name', 'Total', 'geometry']]
data = data.dropna()
data.head()

	name	Total	geometry
543	Norway	7.573273	MULTIPOLYGON (((15.14282 79.67431, 15.52255 80...
815	France	4.741312	MULTIPOLYGON (((-51.65780 4.15623, -52.24934 3...
1087	Sweden	3.424918	POLYGON ((11.02737 58.85615, 11.46827 59.43239...
1359	Belarus	6.222741	POLYGON ((28.17671 56.16913, 29.22951 55.91834...
1631	Ukraine	4.637058	POLYGON ((32.15944 52.06125, 32.41206 52.28869...

Background map

Thanks to the geopandas library, we can easily add a background map to our plot by simply calling the plot() function on our geo dataframe.

With just a few lines of code, we can create a synthetic map that displays a European map.

# initialize the figure
fig, ax = plt.subplots(figsize=(10, 10), dpi=300)

# create the plot
data.plot(ax=ax)
plt.show()

Custom axis

The next step is to adjust the plot's ranges (latitude and longitude) using the set_xlim() and set_ylim() functions on the axis object.

We also remove the spines around the map since they are not particularly useful in this context.

# initialize the figure
fig, ax = plt.subplots(figsize=(10, 10), dpi=300)

# create the plot
data.plot(ax=ax)

# custom axis
ax.set_xlim(-11, 41)
ax.set_ylim(32, 73)
ax.set_axis_off()

plt.show()

Choropleth map

A choropleth map is distinct from a background map because it employs color gradients to depict data, with each country shaded according to its data value.

We initiate by loading a colormap via the load_cmap() function from the pypalettes library, subsequently integrating this colormap into the plot by including cmap=cmap in the plot() function.

To enhance readability, we also define the edgecolor and linewidth of the countries.

# load colormap
cmap = load_cmap('BrwnYl', cmap_type='continuous')

# initialize the figure
fig, ax = plt.subplots(figsize=(10, 10), dpi=300)

# create the plot
data.plot(ax=ax, column='Total', cmap=cmap, edgecolor='black', linewidth=0.5)

# custom axis
ax.set_xlim(-11, 41)
ax.set_ylim(32, 73)
ax.set_axis_off()

plt.show()