At the heart of GeoPandas’ functionality is the GeoSeries, a specialized data structure designed to store geometric objects. These objects can represent points, lines, or polygons, and come with a wide array of built-in methods and attributes for spatial analysis.

GeoPandas provides numerous methods to work with these geometric objects, ranging from basic operations like calculating areas and centroids to more complex analyses like spatial joins and overlays. Here, we will focus on three fundamental geometry methods: boundary, centroid, and buffer.

Exploring Boundaries

The boundary method returns the limits of each geometry in a GeoSeries. For polygons, this means the outer edge or perimeter.

import geopandas as gpd
import geodatasets

# Load the New York boroughs dataset
path = geodatasets.get_path('ny.bb')
nyc = gpd.read_file(path)

# Plot the boundaries
ax = nyc.geometry.boundary.plot(figsize=(5, 5))
ax.set_title('Borough Boundaries');

This code snippet loads our familiar New York City boroughs dataset and plots the boundaries of each borough. The boundary method extracts just the outer edges of each polygon, giving us a clear outline of the boroughs.

The resulting plot shows the perimeters of each borough, highlighting the shape and relative positions of New York City’s administrative divisions. This method is particularly useful when you want to focus on the edges or borders of geographic areas.

Finding Centroids

The centroid method calculates the geometric center of each shape in a GeoSeries. For complex shapes, this point might not always fall within the shape itself.

# Calculate and plot centroids
ax = nyc.plot(color='lightblue', edgecolor='black', figsize=(5, 5))
nyc.geometry.centroid.plot(ax=ax, color='red', markersize=50)
ax.set_title('Borough Centroids');

print("Borough Centroids:")
print(nyc[['BoroName', 'geometry']].centroid)

Here, we’re plotting the boroughs in light blue and then overlaying their centroids as red points. The centroid method returns a point geometry for each borough, representing its center of mass. Each point represents the centroid of a borough, with two numbers inside the parenthesis. These numbers are the x and y coordinates, respectively. 

This visualization helps us understand the spatial distribution of the boroughs. The centroids can be useful for various analyses, such as finding the most central location in each borough or as a simplified representation of the borough for certain calculations.

Creating Buffers

The buffer method generates a new geometry that includes all points within a specified distance of the original geometry. This is incredibly useful for proximity analyses.

# Create and plot buffer
buffer_distance = 5280  # 1 mile in feet
ax = nyc.plot(color='lightblue', edgecolor='black', figsize=(5, 5))
nyc.geometry.buffer(buffer_distance).boundary.plot(ax=ax, color='red')
ax.set_title('Boroughs with 1-mile Buffer');

In this example, we’re creating a 1-mile buffer around each borough. The buffer method expands each geometry by the specified distance in all directions. We then plot the original boroughs in light blue and the buffer boundaries in red.

This technique is often used in GIS for tasks like determining areas within a certain distance of a feature, creating zones of influence, or approximating service areas. 

Conclusion

These three methods—boundary, centroid, and buffer—are just a small sample of the powerful geometric operations available in GeoPandas. By leveraging these and other built-in methods, you can perform complex spatial analyses with just a few lines of code.

As you continue to explore GeoPandas, you’ll discover many more methods for manipulating and analyzing geometric data. Whether you’re working on urban planning, environmental science, or any field that deals with spatial data, understanding these geometric operations can be helpful for effective geospatial analysis. I encourage you to try applying these methods to different datasets, experiment with parameters, and see how you can combine multiple operations to gain deeper insights from your spatial data.