Table Joins & Choropleth Plotting¶
This time we'll join a table of US Census data to a Census Tract shapefile and create a choropleth map of the results.
>>> import pandas as pd
>>> import geopandas as gpd
>>> import matplotlib.pyplot as plt
>>> import osIn [1]:
import pandas as pd
import geopandas as gpd
import matplotlib.pyplot as plt
import os
Important! Reset your working directory to your home folder. Only run this once!¶
>>> os.chdir('../')
>>> os.getcwd()In [2]:
os.chdir('../../')
os.getcwd()
Out[2]:
Start by importing the Boulder_Co_Tracts shapefile as a variable:¶
>>> BoulderTracts = gpd.read_file('workshopdata/Boulder_Co_Tracts.shp')
>>> BoulderTractsIn [3]:
BoulderTracts = gpd.read_file('Data/Boulder_Co_Tracts.shp')
BoulderTracts
Out[3]:
Now, we'll read the Boulder_Co_MHHI csv file into a non-spatial Pandas dataframe:¶
>>> BoulderMHHI = pd.read_csv('workshopdata/Boulder_Co_MHHI.csv')
>>> BoulderMHHIIn [4]:
BoulderMHHI = pd.read_csv('Data/Boulder_Co_MHHI.csv')
BoulderMHHI
Out[4]:
Which column should we base the join on?¶
Take a close look at GEOID10:
>>> print(BoulderMHHI.GEOID10)
>>> print(BoulderTracts.GEOID10)In [5]:
print(BoulderMHHI.GEOID10)
print(BoulderTracts.GEOID10)
Arrrgh! Classic problem. Stripped leading zeros when importing the csv.¶
Good new: this is easy to fix. Reimport the BoulderMHHI csv and tell pandas to read the GEOID10 column as a string instead of a number:¶
>>> BoulderMHHI = pd.read_csv('workshopdata/Boulder_Co_MHHI.csv', dtype={'GEOID10':str})
>>> BoulderMHHIIn [6]:
BoulderMHHI = pd.read_csv('Data/Boulder_Co_MHHI.csv', dtype={"GEOID10":str})
BoulderMHHI
Out[6]:
In [7]:
BoulderMHHI_Tracts = BoulderTracts.merge(BoulderMHHI, on = 'GEOID10')
Done! Now view the results:¶
>>> BoulderMHHI_TractsIn [8]:
BoulderMHHI_Tracts
Out[8]:
In [9]:
BoulderMHHI_Tracts.plot('MHHI2014')
Out[9]:
SWEET MAP! But maybe different colors would be cool too. Check this out:¶
>>> BoulderMHHI_Tracts.plot('MHHI2014', cmap='RdYlGn', figsize=(10,10))In [10]:
BoulderMHHI_Tracts.plot('MHHI2014', cmap='RdYlGn', figsize=(10,10))
Out[10]:
In [11]:
BoulderMHHI_Tracts.to_file('Data/BoulderMHHI_Tracts.geojson', driver='GeoJSON')