Indexing, Selecting and Exploring spatial data in GeoPandas

 Indexing, Selecting and Exploring spatial data in GeoPandas

After reading this tutorial practice, you can have a full basic spatial data analysis in your pocket.

You can do different style of selection of geodataframe in GeoPandas. If you want to select only one column for example:

In [38]:

import geopandas as gpd
import matplotlib.pyplot as plt

In [24]:

world = gpd.read_file(gpd.datasets.get_path('naturalearth_lowres'))
world.head()

Out[24]:

	pop_est	continent	name	iso_a3	gdp_md_est	geometry
0	920938	Oceania	Fiji	FJI	8374.0	MULTIPOLYGON (((180.00000 -16.06713, 180.00000...
1	53950935	Africa	Tanzania	TZA	150600.0	POLYGON ((33.90371 -0.95000, 34.07262 -1.05982...
2	603253	Africa	W. Sahara	ESH	906.5	POLYGON ((-8.66559 27.65643, -8.66512 27.58948...
3	35623680	North America	Canada	CAN	1674000.0	MULTIPOLYGON (((-122.84000 49.00000, -122.9742...
4	326625791	North America	United States of America	USA	18560000.0	MULTIPOLYGON (((-122.84000 49.00000, -120.0000...

In [39]:

world.columns.values

Out[39]:

array(['pop_est', 'continent', 'name', 'iso_a3', 'gdp_md_est', 'geometry'],
      dtype=object)

In [40]:

el_mondo = world.copy()# I made a copy of our main world data
el_mondo = el_mondo.set_index('iso_a3') # using a specific column as an index
el_mondo.head()

Out[40]:

	pop_est	continent	name	gdp_md_est	geometry
iso_a3
FJI	920938	Oceania	Fiji	8374.0	MULTIPOLYGON (((180.00000 -16.06713, 180.00000...
TZA	53950935	Africa	Tanzania	150600.0	POLYGON ((33.90371 -0.95000, 34.07262 -1.05982...
ESH	603253	Africa	W. Sahara	906.5	POLYGON ((-8.66559 27.65643, -8.66512 27.58948...
CAN	35623680	North America	Canada	1674000.0	MULTIPOLYGON (((-122.84000 49.00000, -122.9742...
USA	326625791	North America	United States of America	18560000.0	MULTIPOLYGON (((-122.84000 49.00000, -120.0000...

My country is Senegal and let's extract Senegal and visualized it by using .loc function in geopandas

In [41]:

senegal = el_mondo.loc['SEN', 'geometry']
senegal

Out[41]:

In [ ]:

# Plotting the southern part of the world
souther_world = el_mondo.cx[:, :0] # slice with bounding box
souther_world.plot(figsize=(10,3))

Analyzing geospatial data

Find gpd of Africa and Europe, then compare them

The second part of this geospatial practice, i will try to find the gdp of African countries and later on compare it to gdp of European Countries. My comparison is not equal from the first observation. We know already the European gdp is much higher than the African gdp, but we are just trying to process, explore and analyze data here.

In [42]:

world.head()

Out[42]:

	pop_est	continent	name	iso_a3	gdp_md_est	geometry
0	920938	Oceania	Fiji	FJI	8374.0	MULTIPOLYGON (((180.00000 -16.06713, 180.00000...
1	53950935	Africa	Tanzania	TZA	150600.0	POLYGON ((33.90371 -0.95000, 34.07262 -1.05982...
2	603253	Africa	W. Sahara	ESH	906.5	POLYGON ((-8.66559 27.65643, -8.66512 27.58948...
3	35623680	North America	Canada	CAN	1674000.0	MULTIPOLYGON (((-122.84000 49.00000, -122.9742...
4	326625791	North America	United States of America	USA	18560000.0	MULTIPOLYGON (((-122.84000 49.00000, -120.0000...

In [43]:

# Extract Africa
africa = world.query('continent == "Africa"')
africa.head()

Out[43]:

	pop_est	continent	name	iso_a3	gdp_md_est	geometry
1	53950935	Africa	Tanzania	TZA	150600.0	POLYGON ((33.90371 -0.95000, 34.07262 -1.05982...
2	603253	Africa	W. Sahara	ESH	906.5	POLYGON ((-8.66559 27.65643, -8.66512 27.58948...
11	83301151	Africa	Dem. Rep. Congo	COD	66010.0	POLYGON ((29.34000 -4.49998, 29.51999 -5.41998...
12	7531386	Africa	Somalia	SOM	4719.0	POLYGON ((41.58513 -1.68325, 40.99300 -0.85829...
13	47615739	Africa	Kenya	KEN	152700.0	POLYGON ((39.20222 -4.67677, 37.76690 -3.67712...

We need to add a GDP column

In [44]:

africa.columns

Out[44]:

Index(['pop_est', 'continent', 'name', 'iso_a3', 'gdp_md_est', 'geometry'], dtype='object')

In [45]:

africa['GDP'] = (africa['gdp_md_est'] / africa['pop_est'])
africa.columns

d:\programm files\python\lib\site-packages\geopandas\geodataframe.py:853: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  super(GeoDataFrame, self).__setitem__(key, value)

Out[45]:

Index(['pop_est', 'continent', 'name', 'iso_a3', 'gdp_md_est', 'geometry',
       'GDP'],
      dtype='object')

In [46]:

#africa.drop(['GDP_africa'], axis=1, inplace=True)
africa['GDP_africa'] = africa.gdp_md_est / africa.pop_est
africa.columns

Out[46]:

Index(['pop_est', 'continent', 'name', 'iso_a3', 'gdp_md_est', 'geometry',
       'GDP', 'GDP_africa'],
      dtype='object')

Ooh, we have a GDP per capital for every country in Africa. Note: We use only african countries stated on geopandas datasets, there might be some other states that are dependency.

In [ ]:

africa.GDP_africa.max()

In [ ]:

# Another method to find the GDP, note i've wrote GPD insted of GDP, just ignore it
maxClm = africa.max()['GDP_africa'] 
maxClm

In [ ]:

# Finding the five largest GDP for african countries
africa.nlargest(5, ['GDP_africa']) 

Our result is a big challenge, because the countries stated here are mostly a little richer than other african countries except Libya for nowadays

Filtering the largest 5 African countries, i'm still looking how to do that

In [ ]:

#africa.groupby("name")['GDP_africa'].mean()
#When you plot the above line, you get gdp mean for every african country

In [47]:

africa.groupby("name")['GDP_africa'].mean().plot(kind='bar', figsize=(20,12))

Out[47]:

<AxesSubplot:xlabel='name'>

Oooh lovely, it's not my country 'Senegal' the top one, it's Equatorial Guinea, followed by Gabon and Botswana. No one of these countries are in West Africa. Another analyze can be for example, what's make them different from the others. A simple answer can be an oil production for Gabon but that can't be only reason because there are other countries in Africa who produces much more oil than Gabon. I'm here for technical data analysis, so far i will continue on my technical data analysis. To see the figure better, you can download a Magnifying glass from google app extension.

In [48]:

# Another plotting method which is little faster
#africa.groupby('name').plot()
# if you plot the line above, it will print out every african country a part from another 

In [49]:

# Going deeper
africa.groupby("name")['GDP_africa'].describe()

Out[49]:

	count	mean	std	min	25%	50%	75%	max
name
Algeria	1.0	0.014875	NaN	0.014875	0.014875	0.014875	0.014875	0.014875
Angola	1.0	0.006448	NaN	0.006448	0.006448	0.006448	0.006448	0.006448
Benin	1.0	0.002202	NaN	0.002202	0.002202	0.002202	0.002202	0.002202
Botswana	1.0	0.016209	NaN	0.016209	0.016209	0.016209	0.016209	0.016209
Burkina Faso	1.0	0.001641	NaN	0.001641	0.001641	0.001641	0.001641	0.001641
Burundi	1.0	0.000688	NaN	0.000688	0.000688	0.000688	0.000688	0.000688
Cameroon	1.0	0.003090	NaN	0.003090	0.003090	0.003090	0.003090	0.003090
Central African Rep.	1.0	0.000570	NaN	0.000570	0.000570	0.000570	0.000570	0.000570
Chad	1.0	0.002533	NaN	0.002533	0.002533	0.002533	0.002533	0.002533
Congo	1.0	0.006109	NaN	0.006109	0.006109	0.006109	0.006109	0.006109
Côte d'Ivoire	1.0	0.003602	NaN	0.003602	0.003602	0.003602	0.003602	0.003602
Dem. Rep. Congo	1.0	0.000792	NaN	0.000792	0.000792	0.000792	0.000792	0.000792
Djibouti	1.0	0.003866	NaN	0.003866	0.003866	0.003866	0.003866	0.003866
Egypt	1.0	0.011387	NaN	0.011387	0.011387	0.011387	0.011387	0.011387
Eq. Guinea	1.0	0.040817	NaN	0.040817	0.040817	0.040817	0.040817	0.040817
Eritrea	1.0	0.001549	NaN	0.001549	0.001549	0.001549	0.001549	0.001549
Ethiopia	1.0	0.001658	NaN	0.001658	0.001658	0.001658	0.001658	0.001658
Gabon	1.0	0.020302	NaN	0.020302	0.020302	0.020302	0.020302	0.020302
Gambia	1.0	0.001651	NaN	0.001651	0.001651	0.001651	0.001651	0.001651
Ghana	1.0	0.004393	NaN	0.004393	0.004393	0.004393	0.004393	0.004393
Guinea	1.0	0.001295	NaN	0.001295	0.001295	0.001295	0.001295	0.001295
Guinea-Bissau	1.0	0.001591	NaN	0.001591	0.001591	0.001591	0.001591	0.001591
Kenya	1.0	0.003207	NaN	0.003207	0.003207	0.003207	0.003207	0.003207
Lesotho	1.0	0.003074	NaN	0.003074	0.003074	0.003074	0.003074	0.003074
Liberia	1.0	0.000828	NaN	0.000828	0.000828	0.000828	0.000828	0.000828
Libya	1.0	0.013661	NaN	0.013661	0.013661	0.013661	0.013661	0.013661
Madagascar	1.0	0.001471	NaN	0.001471	0.001471	0.001471	0.001471	0.001471
Malawi	1.0	0.001104	NaN	0.001104	0.001104	0.001104	0.001104	0.001104
Mali	1.0	0.002130	NaN	0.002130	0.002130	0.002130	0.002130	0.002130
Mauritania	1.0	0.004446	NaN	0.004446	0.004446	0.004446	0.004446	0.004446
Morocco	1.0	0.008321	NaN	0.008321	0.008321	0.008321	0.008321	0.008321
Mozambique	1.0	0.001317	NaN	0.001317	0.001317	0.001317	0.001317	0.001317
Namibia	1.0	0.010460	NaN	0.010460	0.010460	0.010460	0.010460	0.010460
Niger	1.0	0.001047	NaN	0.001047	0.001047	0.001047	0.001047	0.001047
Nigeria	1.0	0.005713	NaN	0.005713	0.005713	0.005713	0.005713	0.005713
Rwanda	1.0	0.001846	NaN	0.001846	0.001846	0.001846	0.001846	0.001846
S. Sudan	1.0	0.001603	NaN	0.001603	0.001603	0.001603	0.001603	0.001603
Senegal	1.0	0.002708	NaN	0.002708	0.002708	0.002708	0.002708	0.002708
Sierra Leone	1.0	0.001726	NaN	0.001726	0.001726	0.001726	0.001726	0.001726
Somalia	1.0	0.000627	NaN	0.000627	0.000627	0.000627	0.000627	0.000627
Somaliland	1.0	0.003500	NaN	0.003500	0.003500	0.003500	0.003500	0.003500
South Africa	1.0	0.013477	NaN	0.013477	0.013477	0.013477	0.013477	0.013477
Sudan	1.0	0.004721	NaN	0.004721	0.004721	0.004721	0.004721	0.004721
Tanzania	1.0	0.002791	NaN	0.002791	0.002791	0.002791	0.002791	0.002791
Togo	1.0	0.001458	NaN	0.001458	0.001458	0.001458	0.001458	0.001458
Tunisia	1.0	0.011470	NaN	0.011470	0.011470	0.011470	0.011470	0.011470
Uganda	1.0	0.002146	NaN	0.002146	0.002146	0.002146	0.002146	0.002146
W. Sahara	1.0	0.001503	NaN	0.001503	0.001503	0.001503	0.001503	0.001503
Zambia	1.0	0.004080	NaN	0.004080	0.004080	0.004080	0.004080	0.004080
Zimbabwe	1.0	0.002052	NaN	0.002052	0.002052	0.002052	0.002052	0.002052
eSwatini	1.0	0.007538	NaN	0.007538	0.007538	0.007538	0.007538	0.007538

In [51]:

total_gdp_africa = africa['GDP_africa'].sum()
total_gdp_africa

Out[51]:

0.2672931076038051

In [54]:

africa.to_file('african_continent.gpkg')
africa.to_file('african_continent.geojson')
africa.to_file('african_continent')

Let's stop here with our analyzing for Africa, now let's focus on Europe which is the comparison

As talking about Europe here, my heart is going to Scotland, i love you Alba, i remember you Lucy, the lovely child, great child in great place. As my daily routine was waiting my lovely daughter Aysha after dropping her at school, one day, i stopped somewhere and a great man over there came to me and said gently, did you sow your car tire condition? ooh my dear, it was totally broken, thank to him always and God bless you wherever you are.

In [55]:

europe = world.query('continent == "Europe"')
europe.head()

Out[55]:

	pop_est	continent	name	iso_a3	gdp_md_est	geometry
18	142257519	Europe	Russia	RUS	3745000.0	MULTIPOLYGON (((178.725 71.099, 180.000 71.516...
21	5320045	Europe	Norway	-99	364700.0	MULTIPOLYGON (((15.143 79.674, 15.523 80.016, ...
43	67106161	Europe	France	-99	2699000.0	MULTIPOLYGON (((-51.658 4.156, -52.249 3.241, ...
110	9960487	Europe	Sweden	SWE	498100.0	POLYGON ((11.027 58.856, 11.468 59.432, 12.300...
111	9549747	Europe	Belarus	BLR	165400.0	POLYGON ((28.177 56.169, 29.230 55.918, 29.372...

In [56]:

# Extracting all countries in Europe
[i for i in europe['name']]

Out[56]:

['Russia',
 'Norway',
 'France',
 'Sweden',
 'Belarus',
 'Ukraine',
 'Poland',
 'Austria',
 'Hungary',
 'Moldova',
 'Romania',
 'Lithuania',
 'Latvia',
 'Estonia',
 'Germany',
 'Bulgaria',
 'Greece',
 'Albania',
 'Croatia',
 'Switzerland',
 'Luxembourg',
 'Belgium',
 'Netherlands',
 'Portugal',
 'Spain',
 'Ireland',
 'Italy',
 'Denmark',
 'United Kingdom',
 'Iceland',
 'Slovenia',
 'Finland',
 'Slovakia',
 'Czechia',
 'Bosnia and Herz.',
 'Macedonia',
 'Serbia',
 'Montenegro',
 'Kosovo']

In [59]:

# You can extract european countries names in different ways
#europe['name']
# If you print above line, you will get all european countries names also

In [60]:

# Creating a GDP for european countries
europe['GDP_europe'] = europe.gdp_md_est / europe.pop_est

d:\programm files\python\lib\site-packages\geopandas\geodataframe.py:853: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  super(GeoDataFrame, self).__setitem__(key, value)

In [61]:

europe.head()

Out[61]:

	pop_est	continent	name	iso_a3	gdp_md_est	geometry	GDP_europe
18	142257519	Europe	Russia	RUS	3745000.0	MULTIPOLYGON (((178.725 71.099, 180.000 71.516...	0.026325
21	5320045	Europe	Norway	-99	364700.0	MULTIPOLYGON (((15.143 79.674, 15.523 80.016, ...	0.068552
43	67106161	Europe	France	-99	2699000.0	MULTIPOLYGON (((-51.658 4.156, -52.249 3.241, ...	0.040220
110	9960487	Europe	Sweden	SWE	498100.0	POLYGON ((11.027 58.856, 11.468 59.432, 12.300...	0.050008
111	9549747	Europe	Belarus	BLR	165400.0	POLYGON ((28.177 56.169, 29.230 55.918, 29.372...	0.017320

In [62]:

europe.nlargest(5, ['GDP_europe']) 

Out[62]:

	pop_est	continent	name	iso_a3	gdp_md_est	geometry	GDP_europe
128	594130	Europe	Luxembourg	LUX	58740.0	POLYGON ((6.04307 50.12805, 6.24275 49.90223, ...	0.098867
21	5320045	Europe	Norway	-99	364700.0	MULTIPOLYGON (((15.14282 79.67431, 15.52255 80...	0.068552
133	5011102	Europe	Ireland	IRL	322000.0	POLYGON ((-6.19788 53.86757, -6.03299 53.15316...	0.064257
127	8236303	Europe	Switzerland	CHE	496300.0	POLYGON ((9.59423 47.52506, 9.63293 47.34760, ...	0.060258
130	17084719	Europe	Netherlands	NLD	870800.0	POLYGON ((6.90514 53.48216, 7.09205 53.14404, ...	0.050970

In [67]:

europe.nlargest(5, ['GDP_europe']).plot()

Out[67]:

<AxesSubplot:>

In [68]:

europe.groupby("name")['GDP_europe'].mean().plot(kind='bar', figsize=(20,12))

Out[68]:

<AxesSubplot:xlabel='name'>

The highest gdp in Europe in Luxembourg and lowest is Kosovo, but both country are having a little similarity about area

In [73]:

total_gdp_europe = europe['GDP_europe'].sum()
total_gdp_europe

Out[73]:

1.314297770189312

The total gdp of europe is higher than africa with asimple eye observation. Let's calculate the difference

In [74]:

total_gdp_europe - total_gdp_africa

Out[74]:

1.0470046625855067

In [ ]:

After analyzing European GDP, let's see the world GDP

In [26]:

world_without_antarctica = world[(world.pop_est>0) & (world.name!="Antarctica")]
# world.name != means all world map except Antarctica, != means not equal
# Let's verify, first find the continent in world then find if the Antarctica is in world_without_antarctica

set(world.continent) # i used set to avoid repetition 

Out[26]:

{'Africa',
 'Antarctica',
 'Asia',
 'Europe',
 'North America',
 'Oceania',
 'Seven seas (open ocean)',
 'South America'}

In [27]:

# Verification
'Antarctica' in world.continent # world.continent is the column where Antarctica was belonging too

Out[27]:

False

In [28]:

world_without_antarctica.plot()

Out[28]:

<AxesSubplot:>

In [29]:

world.plot()

Out[29]:

<AxesSubplot:>

The last two figures are different, you can see it by yourself

Just a small notification for readers: White analyzing any data, there are thousands of ideas coming around and sometimes you need to show up your experiences and skills

In [34]:

world_without_antarctica['gdp_per_cap'] = world_without_antarctica['gdp_md_est'] / world_without_antarctica['pop_est']
world_without_antarctica.head()

d:\programm files\python\lib\site-packages\geopandas\geodataframe.py:853: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  super(GeoDataFrame, self).__setitem__(key, value)

Out[34]:

	pop_est	continent	name	iso_a3	gdp_md_est	geometry	gdp_per_cap
0	920938	Oceania	Fiji	FJI	8374.0	MULTIPOLYGON (((180.00000 -16.06713, 180.00000...	0.009093
1	53950935	Africa	Tanzania	TZA	150600.0	POLYGON ((33.90371 -0.95000, 34.07262 -1.05982...	0.002791
2	603253	Africa	W. Sahara	ESH	906.5	POLYGON ((-8.66559 27.65643, -8.66512 27.58948...	0.001503
3	35623680	North America	Canada	CAN	1674000.0	MULTIPOLYGON (((-122.84000 49.00000, -122.9742...	0.046991
4	326625791	North America	United States of America	USA	18560000.0	MULTIPOLYGON (((-122.84000 49.00000, -120.0000...	0.056823

In [37]:

world_without_antarctica.plot(column= 'gdp_per_cap', figsize= (20, 8))

Out[37]:

<AxesSubplot:>

Exploring the World Population density with color

In [85]:

fig, ax = plt.subplots(1,1)
world.plot(column='pop_est', ax=ax, legend=True)

Out[85]:

<AxesSubplot:>

China seems to be the most popular

In [90]:

fig, ax = plt.subplots(1,1)
africa = world.query('continent == "Africa"')
africa.plot(column='pop_est', ax=ax, legend=True)

Out[90]:

<AxesSubplot:>

Nigeria is the most popular country in Africa

In [87]:

fig, ax = plt.subplots(1,1)
europe.plot(column='pop_est', ax=ax, legend=True)

Out[87]:

<AxesSubplot:>

Russia seems to be the most populest country in Europe

In [97]:

from mpl_toolkits.axes_grid import make_axes_locatable
fig, ax = plt.subplots(1,1)
divider = make_axes_locatable(ax)
cax = divider.append_axes("right", size="20%", pad=0.1)
africa.plot(column='pop_est', ax=ax, legend=True, cax=cax)

Out[97]:

<AxesSubplot:>

In [98]:

fig, ax = plt.subplots(1,1)
africa.plot(column = 'pop_est', ax=ax, legend=True,
           legend_kwds={'label': 'Population by Country', 'orientation': 'horizontal'})

Out[98]:

<AxesSubplot:>

In [105]:

world['gdp_per_cap'] = world.gdp_md_est / world.pop_est
world.plot(column = 'gdp_per_cap', cmap='OrRd') # choosing color

Out[105]:

<AxesSubplot:>

The most south part of the globe is less with gdp except Australia and Golf countries

In [107]:

africa['gdp_per_cap'] = africa.gdp_md_est / africa.pop_est

d:\programm files\python\lib\site-packages\geopandas\geodataframe.py:853: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  super(GeoDataFrame, self).__setitem__(key, value)

In [108]:

africa.columns

Out[108]:

Index(['pop_est', 'continent', 'name', 'iso_a3', 'gdp_md_est', 'geometry',
       'gdp_per_cap'],
      dtype='object')

In [111]:

africa.plot(column='gdp_per_cap', cmap='OrRd')

Out[111]:

<AxesSubplot:>

Here, we can see clearly that the northern part of Africa and Southern Part are much richer than the rest.

In [113]:

africa.boundary.plot() # To visualize better without any color

Out[113]:

<AxesSubplot:>

In [114]:

world.plot(column='gdp_per_cap', cmap='OrRd', scheme='quantiles')

Out[114]:

<AxesSubplot:>

In [115]:

africa.plot(column='gdp_per_cap', cmap='OrRd', scheme='quantiles')

Out[115]:

<AxesSubplot:>

Exploring missing data in Geopandas

Exploring spatial data is just fantastic, however, when we have a spatial data that contains a missing values. Geopandas from the version 0.7 to higher version ignores it automatically.

We will create a data from world['pop_est'], 40 rows and transform them to nan or NAN, sometimes NaN. We used .loc to concentrate in one specific column that will not affect other columns values.

In [117]:

import numpy as np
world.loc[np.random.choice(world.index, 40), 'pop_est'] = np.nan
world.head()

Out[117]:

	pop_est	continent	name	iso_a3	gdp_md_est	geometry	gdp_per_cap
0	NaN	Oceania	Fiji	FJI	8374.0	MULTIPOLYGON (((180.00000 -16.06713, 180.00000...	0.009093
1	53950935.0	Africa	Tanzania	TZA	150600.0	POLYGON ((33.90371 -0.95000, 34.07262 -1.05982...	0.002791
2	603253.0	Africa	W. Sahara	ESH	906.5	POLYGON ((-8.66559 27.65643, -8.66512 27.58948...	0.001503
3	35623680.0	North America	Canada	CAN	1674000.0	MULTIPOLYGON (((-122.84000 49.00000, -122.9742...	0.046991
4	NaN	North America	United States of America	USA	18560000.0	MULTIPOLYGON (((-122.84000 49.00000, -120.0000...	0.056823

In [120]:

null_data = world[world.isnull().any(axis=1)]
null_data

Out[120]:

	pop_est	continent	name	iso_a3	gdp_md_est	geometry	gdp_per_cap
0	NaN	Oceania	Fiji	FJI	8374.0	MULTIPOLYGON (((180.00000 -16.06713, 180.00000...	0.009093
4	NaN	North America	United States of America	USA	18560000.0	MULTIPOLYGON (((-122.84000 49.00000, -120.0000...	0.056823
7	NaN	Oceania	Papua New Guinea	PNG	28020.0	MULTIPOLYGON (((141.00021 -2.60015, 142.73525 ...	0.004055
16	NaN	North America	Haiti	HTI	19340.0	POLYGON ((-71.71236 19.71446, -71.62487 19.169...	0.001817
19	NaN	North America	Bahamas	BHS	9066.0	MULTIPOLYGON (((-78.98000 26.79000, -78.51000 ...	0.027474
...	...	...	...	...	...	...	...
161	NaN	Asia	Cyprus	CYP	29260.0	POLYGON ((32.73178 35.14003, 32.91957 35.08783...	0.023953
167	NaN	Africa	Somaliland	-99	12250.0	POLYGON ((48.94820 11.41062, 48.94820 11.41062...	0.003500
169	NaN	Africa	Rwanda	RWA	21970.0	POLYGON ((30.41910 -1.13466, 30.81613 -1.69891...	0.001846
170	NaN	Europe	Bosnia and Herz.	BIH	42530.0	POLYGON ((18.56000 42.65000, 17.67492 43.02856...	0.011029
173	NaN	Europe	Montenegro	MNE	10610.0	POLYGON ((20.07070 42.58863, 19.80161 42.50009...	0.016512

66 rows × 7 columns

In [124]:

null_data.plot(figsize=(20,8))

Out[124]:

<AxesSubplot:>

In [122]:

world.plot(column='pop_est')

Out[122]:

<AxesSubplot:>

Mapping with Layers

I might have a different geometry points locations and want to map them in specific map. To be clearer, i have bike sations addresses in Paris and want to map them (display) on Paris map or France map. The Point locations are represented with layers. It's very important to make sure that the data are always having similar crs (coordinate reference system).

In [127]:

import geopandas
cities = geopandas.read_file(geopandas.datasets.get_path('naturalearth_cities'))
cities.head()

Out[127]:

	name	geometry
0	Vatican City	POINT (12.45339 41.90328)
1	San Marino	POINT (12.44177 43.93610)
2	Vaduz	POINT (9.51667 47.13372)
3	Luxembourg	POINT (6.13000 49.61166)
4	Palikir	POINT (158.14997 6.91664)

In [128]:

cities.plot(marker='*', color='green', markersize=5)

Out[128]:

<AxesSubplot:>

In [129]:

type(cities.crs)

Out[129]:

pyproj.crs.crs.CRS

We make sure that we have full world map first then check if the cities crs is same with world crs.

In [130]:

# i will replace the world with globe
globe = geopandas.read_file(geopandas.datasets.get_path('naturalearth_lowres'))
globe.head()

Out[130]:

	pop_est	continent	name	iso_a3	gdp_md_est	geometry
0	920938	Oceania	Fiji	FJI	8374.0	MULTIPOLYGON (((180.00000 -16.06713, 180.00000...
1	53950935	Africa	Tanzania	TZA	150600.0	POLYGON ((33.90371 -0.95000, 34.07262 -1.05982...
2	603253	Africa	W. Sahara	ESH	906.5	POLYGON ((-8.66559 27.65643, -8.66512 27.58948...
3	35623680	North America	Canada	CAN	1674000.0	MULTIPOLYGON (((-122.84000 49.00000, -122.9742...
4	326625791	North America	United States of America	USA	18560000.0	MULTIPOLYGON (((-122.84000 49.00000, -120.0000...

Note: Our globe data is not having gdp_per_cap column

In [131]:

globe.columns

Out[131]:

Index(['pop_est', 'continent', 'name', 'iso_a3', 'gdp_md_est', 'geometry'], dtype='object')

In [133]:

cities = cities.to_crs(globe.crs)
type(cities.crs)

Out[133]:

pyproj.crs.crs.CRS

In [134]:

globe.crs

Out[134]:

<Geographic 2D CRS: EPSG:4326>
Name: WGS 84
Axis Info [ellipsoidal]:
- Lat[north]: Geodetic latitude (degree)
- Lon[east]: Geodetic longitude (degree)
Area of Use:
- name: World
- bounds: (-180.0, -90.0, 180.0, 90.0)
Datum: World Geodetic System 1984
- Ellipsoid: WGS 84
- Prime Meridian: Greenwich

In [135]:

cities.crs

Out[135]:

<Geographic 2D CRS: EPSG:4326>
Name: WGS 84
Axis Info [ellipsoidal]:
- Lat[north]: Geodetic latitude (degree)
- Lon[east]: Geodetic longitude (degree)
Area of Use:
- name: World
- bounds: (-180.0, -90.0, 180.0, 90.0)
Datum: World Geodetic System 1984
- Ellipsoid: WGS 84
- Prime Meridian: Greenwich

I think they have same crs.

In [140]:

base = world.plot(color = 'white', edgecolor='black'); # i could just do this facecolor = 'none'
cities.plot(ax=base, marker='o', color='red', markersize=5)

Out[140]:

<AxesSubplot:>

I will experiment the last code with Senegal and see how it will look like

In [142]:

senegal = globe[globe['name'] == 'Senegal']
senegal

Out[142]:

	pop_est	continent	name	iso_a3	gdp_md_est	geometry
51	14668522	Africa	Senegal	SEN	39720.0	POLYGON ((-16.71373 13.59496, -17.12611 14.373...

In [146]:

senegal.plot(color='white', edgecolor = 'black')

Out[146]:

<AxesSubplot:>

I will upload some region in Senegal and try to map them

In [148]:

import pandas as pd
location = pd.read_excel('D:/DATA/villes_senegal.xlsx')
location.head()

Out[148]:

	city	Region	Long	Lat
0	Bigno	Dakar	-16.22639	12.81028
1	Dakar	Dakar	-17.44406	14.69370
2	Daara	Louga	-15.47993	15.34844
3	Diawara	Matam	-12.54374	15.02196
4	Diofior	Fatick	-16.66667	14.18333

As i don't have geometry column ('its only a pandas dataframe not geodataframe since the columns does not include geometry column'), i will create it. To do that, this is an idea: polygon_geom = zip(lon_list, lat_list)

In [155]:

from shapely.geometry import Point, Polygon
location['geometry'] = [Point(xy) for xy in zip(location['Long'], location['Lat'])]
location.columns.values

Out[155]:

array(['city', 'Region', 'Long', 'Lat', 'geometry'], dtype=object)

In [156]:

location.head()

Out[156]:

	city	Region	Long	Lat	geometry
0	Bigno	Dakar	-16.22639	12.81028	POINT (-16.22639 12.81028)
1	Dakar	Dakar	-17.44406	14.69370	POINT (-17.44406 14.6937)
2	Daara	Louga	-15.47993	15.34844	POINT (-15.47993 15.34844)
3	Diawara	Matam	-12.54374	15.02196	POINT (-12.54374 15.02196)
4	Diofior	Fatick	-16.66667	14.18333	POINT (-16.66667 14.18333)

In [189]:

# Coordinate reference system : WGS84
correct_crs = {'init': 'epsg:4326'}

After creating crs for our excel loaded data, we still don't have a geodataframe data, for example if we plot it, we get something different.

In [159]:

location.plot()

Out[159]:

<AxesSubplot:>

In [176]:

# Creating a Geographic data frame 
location_gdf = geopandas.GeoDataFrame(
    location, geometry=geopandas.points_from_xy(location.Long, location.Lat))

In [178]:

location_gdf.head()

Out[178]:

	city	Region	Long	Lat	geometry
0	Bigno	Dakar	-16.22639	12.81028	POINT (-16.22639 12.81028)
1	Dakar	Dakar	-17.44406	14.69370	POINT (-17.44406 14.69370)
2	Daara	Louga	-15.47993	15.34844	POINT (-15.47993 15.34844)
3	Diawara	Matam	-12.54374	15.02196	POINT (-12.54374 15.02196)
4	Diofior	Fatick	-16.66667	14.18333	POINT (-16.66667 14.18333)

In [179]:

type(location_gdf)

Out[179]:

geopandas.geodataframe.GeoDataFrame

In [180]:

location_gdf.plot()

Out[180]:

<AxesSubplot:>

Oooh that's it, let's map them in Senegal map.

In [181]:

ax = senegal.plot(color='white', edgecolor='black')
location_gdf.plot(ax=ax, color='red')

Out[181]:

<AxesSubplot:>

In [167]:

senegal.crs

Out[167]:

<Geographic 2D CRS: EPSG:4326>
Name: WGS 84
Axis Info [ellipsoidal]:
- Lat[north]: Geodetic latitude (degree)
- Lon[east]: Geodetic longitude (degree)
Area of Use:
- name: World
- bounds: (-180.0, -90.0, 180.0, 90.0)
Datum: World Geodetic System 1984
- Ellipsoid: WGS 84
- Prime Meridian: Greenwich

In [186]:

print(location_gdf.crs)

None

In [192]:

location_gdf.set_crs("EPSG:4326")

Out[192]:

	city	Region	Long	Lat	geometry
0	Bigno	Dakar	-16.22639	12.81028	POINT (-16.22639 12.81028)
1	Dakar	Dakar	-17.44406	14.69370	POINT (-17.44406 14.69370)
2	Daara	Louga	-15.47993	15.34844	POINT (-15.47993 15.34844)
3	Diawara	Matam	-12.54374	15.02196	POINT (-12.54374 15.02196)
4	Diofior	Fatick	-16.66667	14.18333	POINT (-16.66667 14.18333)
5	Diourbel	Diourbel	-16.24363	14.64792	POINT (-16.24363 14.64792)
6	Fatick	Fatick	-16.58583	14.35806	POINT (-16.58583 14.35806)
7	Foundioungne	Fatick	-16.46667	14.13333	POINT (-16.46667 14.13333)
8	Gandiaye	Kaolack	-16.26667	14.23333	POINT (-16.26667 14.23333)
9	Goléré	Saint-Louis	-14.10165	16.25575	POINT (-14.10165 16.25575)

In [196]:

location_gdf.info()

<class 'geopandas.geodataframe.GeoDataFrame'>
RangeIndex: 10 entries, 0 to 9
Data columns (total 5 columns):
 #   Column    Non-Null Count  Dtype   
---  ------    --------------  -----   
 0   city      10 non-null     object  
 1   Region    10 non-null     object  
 2   Long      10 non-null     float64 
 3   Lat       10 non-null     float64 
 4   geometry  10 non-null     geometry
dtypes: float64(2), geometry(1), object(2)
memory usage: 344.0+ bytes

In [199]:

# Another method to map 
base = senegal.plot(color='white', edgecolor='black')
location_gdf.plot(ax=base, marker='o', color='red', markersize=5)

Out[199]:

<AxesSubplot:>

Just to notify that party in Senegal map look like an open river or mouth, it's Gambia

Again, another method to display the main capital city of every country in world (mapping layers)

In [211]:

fig, ax = plt.subplots()
ax.set_aspect('equal')
world.plot(ax=ax, color='white', edgecolor='black')
cities.plot(ax=ax, color='white', edgecolor='black')
cities.plot(ax=ax, marker='o', color='red', markersize=5)
plt.show()

In [212]:

# displaying every city name 
ax = cities.plot(figsize=(20,8))
for x, y, label in zip(cities.geometry.x, cities.geometry.y, cities.name):
    ax.annotate(label, xy=(x, y), xytext=(3, 3), textcoords="offset points")

In [210]:

ax = plt.figure().add_subplot(1, 1, 1)
senegal.plot(ax=ax, color='white', edgecolor='gold')
location_gdf.plot(ax = ax)

Out[210]:

<AxesSubplot:>

In [215]:

ax = cities.plot(color='k')
world.plot(ax=ax)

Out[215]:

<AxesSubplot:>

This method of displaying layers is based on geometric polygon order in the datasets, to change the order, we need to use 'zorder' attribute

In [218]:

ax = cities.plot(color='k')
world.plot(ax=ax, zorder=1) # on bottom
plt.title('Dots on bottom of lines')
plt.tight_layout()

Conclusion:

At the end of this practical tutorial, we could explore geopandas datasets, analyzing spatial data and interact with a specific area in the world, explore GDP for any country and many more.
While writing this tutorial, i remember you Lucy from Alba, my great love to you a shiny child.