Geospatial Analysis with GeoPandas#

Overview#

When working with geographic data (earthquakes, plate boundaries, etc.), using geopandas with proper coordinate projections provides accurate distance calculations and efficient spatial operations. This guide covers best practices for geospatial analysis.

Key Concepts#

Geographic vs Projected Coordinate Systems#

Coordinate SystemTypeUnitsUse CaseEPSG:4326 (WGS84)GeographicDegrees (lat/lon)Data storage, displayEPSG:4087 (World Equidistant Cylindrical)ProjectedMetersDistance calculations

Critical Rule: Never calculate distances directly in geographic coordinates (EPSG:4326). Always project to a metric coordinate system first.

Why Projection Matters#

python

# ❌ INCORRECT: Calculating distance in EPSG:4326
# This treats degrees as if they were equal distances everywhere on Earth
gdf = gpd.GeoDataFrame(..., crs="EPSG:4326")
distance = point1.distance(point2)  # Wrong! Returns degrees, not meters

# ✅ CORRECT: Project to metric CRS first
gdf_projected = gdf.to_crs("EPSG:4087")
distance_meters = point1_proj.distance(point2_proj)  # Correct! Returns meters
distance_km = distance_meters / 1000.0

Loading Geospatial Data#

From GeoJSON Files#

python

import geopandas as gpd

# Load GeoJSON files directly
gdf_plates = gpd.read_file("plates.json")
gdf_boundaries = gpd.read_file("boundaries.json")

From Regular Data with Coordinates#

python

from shapely.geometry import Point
import geopandas as gpd

# Convert coordinate data to GeoDataFrame
data = [
    {"id": 1, "lat": 35.0, "lon": 140.0, "value": 5.5},
    {"id": 2, "lat": 36.0, "lon": 141.0, "value": 6.0},
]

geometry = [Point(row["lon"], row["lat"]) for row in data]
gdf = gpd.GeoDataFrame(data, geometry=geometry, crs="EPSG:4326")

Spatial Filtering#

Finding Points Within a Polygon#

python

# Get the polygon of interest
target_poly = gdf_plates[gdf_plates["Name"] == "Pacific"].geometry.unary_union

# Filter points that fall within the polygon
points_inside = gdf_points[gdf_points.within(target_poly)]

print(f"Found {len(points_inside)} points inside the polygon")

Using .unary_union for Multiple Geometries#

When you have multiple polygons or lines that should be treated as one:

python

# Combine multiple boundary segments into one geometry
all_boundaries = gdf_boundaries.geometry.unary_union

# Or filter first, then combine
pacific_boundaries = gdf_boundaries[
    gdf_boundaries["Name"].str.contains("PA")
].geometry.unary_union

Distance Calculations#

Point to Line/Boundary Distance#

python

# 1. Load your data
gdf_points = gpd.read_file("points.json")
gdf_boundaries = gpd.read_file("boundaries.json")

# 2. Project to metric coordinate system
METRIC_CRS = "EPSG:4087"
points_proj = gdf_points.to_crs(METRIC_CRS)
boundaries_proj = gdf_boundaries.to_crs(METRIC_CRS)

# 3. Combine boundary segments if needed
boundary_geom = boundaries_proj.geometry.unary_union

# 4. Calculate distances (returns meters)
gdf_points["distance_m"] = points_proj.geometry.distance(boundary_geom)
gdf_points["distance_km"] = gdf_points["distance_m"] / 1000.0

Finding Furthest Point#

python

# Sort by distance and get the furthest point
furthest = gdf_points.nlargest(1, "distance_km").iloc[0]

print(f"Furthest point: {furthest['id']}")
print(f"Distance: {furthest['distance_km']:.2f} km")

Common Workflow Pattern#

Here's a complete example for analyzing earthquakes near plate boundaries:

python

import geopandas as gpd
from shapely.geometry import Point

# 1. Load data
earthquakes_data = [...]  # Your earthquake data
gdf_plates = gpd.read_file("plates.json")
gdf_boundaries = gpd.read_file("boundaries.json")

# 2. Create earthquake GeoDataFrame
geometry = [Point(eq["longitude"], eq["latitude"]) for eq in earthquakes_data]
gdf_eq = gpd.GeoDataFrame(earthquakes_data, geometry=geometry, crs="EPSG:4326")

# 3. Spatial filtering - find earthquakes in specific plate
target_plate = gdf_plates[gdf_plates["Code"] == "PA"].geometry.unary_union
earthquakes_in_plate = gdf_eq[gdf_eq.within(target_plate)].copy()

# 4. Calculate distances (project to metric CRS)
METRIC_CRS = "EPSG:4087"
eq_proj = earthquakes_in_plate.to_crs(METRIC_CRS)

# Filter and combine relevant boundaries
plate_boundaries = gdf_boundaries[
    gdf_boundaries["Name"].str.contains("PA")
].to_crs(METRIC_CRS).geometry.unary_union

# Calculate distances
earthquakes_in_plate["distance_km"] = eq_proj.geometry.distance(plate_boundaries) / 1000.0

# 5. Find the furthest earthquake
furthest_eq = earthquakes_in_plate.nlargest(1, "distance_km").iloc[0]

Filtering by Attributes#

python

# Filter by name or code
pacific_plate = gdf_plates[gdf_plates["PlateName"] == "Pacific"]
pacific_plate_alt = gdf_plates[gdf_plates["Code"] == "PA"]

# Filter boundaries involving a specific plate
pacific_bounds = gdf_boundaries[
    (gdf_boundaries["PlateA"] == "PA") |
    (gdf_boundaries["PlateB"] == "PA")
]

# String pattern matching
pa_related = gdf_boundaries[gdf_boundaries["Name"].str.contains("PA")]

Performance Tips#

1. Filter before projecting: Reduce data size before expensive operations
2. Project once: Convert to metric CRS once, not in loops
3. Use .unary_union: Combine geometries before distance calculations
4. Copy when modifying: Use .copy() when creating filtered DataFrames

python

# Good: Filter first, then project
small_subset = gdf_large[gdf_large["region"] == "Pacific"]
small_projected = small_subset.to_crs(METRIC_CRS)

# Avoid: Projecting large dataset just to filter
# gdf_projected = gdf_large.to_crs(METRIC_CRS)
# small_subset = gdf_projected[gdf_projected["region"] == "Pacific"]

Common Pitfalls#

IssueProblemSolutionDistance in degreesUsing EPSG:4326 for distance calculationsProject to EPSG:4087 or similar metric CRSAntimeridian issuesManual longitude adjustments (±360)Use geopandas spatial operations, they handle itSlow performanceCalculating distance to each boundary point separatelyUse .unary_union + single .distance() callMissing geometriesSome features have no geometryFilter with gdf[gdf.geometry.notna()]

When NOT to Use Manual Calculations#

Avoid implementing your own:

• Haversine distance formulas (use geopandas projections instead)
• Point-in-polygon checks (use .within())
• Iterating through boundary points (use .distance() with .unary_union)

These manual approaches are slower, more error-prone, and less accurate than geopandas methods.

Best Practices Summary#

1. ✅ Load GeoJSON with gpd.read_file()
2. ✅ Use .within() for spatial filtering
3. ✅ Project to metric CRS ( EPSG:4087) before distance calculations
4. ✅ Combine geometries with .unary_union before distance calculation
5. ✅ Use .distance() method for point-to-geometry distances
6. ✅ Use .nlargest() / .nsmallest() for finding extreme values
7. ❌ Never calculate distances in EPSG:4326
8. ❌ Avoid manual Haversine implementations
9. ❌ Don't iterate through individual boundary points