Geopandas projections

Skill cxcscmu/SkillLearnBench/skills/b1-one-shot-claude-haiku-4-5/earthquake-plate-calculation/geopandas-projections

[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.

Install

npx -y skills add cxcscmu/SkillLearnBench --skill geopandas-projections

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its author says it does

Copied from the file, not written here

Use GeoPandas with coordinate projections to perform accurate spatial calculations and transformations.

SKILL.md

2.9 KB, as published. Nobody here has run it

GeoPandas Spatial Projections

Overview

GeoPandas is built on top of Shapely and Fiona, enabling geographic data manipulation with proper coordinate reference systems (CRS). Using correct projections is critical for accurate distance calculations and spatial operations.

Installation

pip install geopandas shapely fiona pyproj

Key Concepts

Coordinate Reference Systems (CRS)

EPSG:4326: WGS84 (lat/lon), commonly used for geographic data but NOT suitable for distance calculations
EPSG:3857: Web Mercator, used for web mapping
Regional Projected CRS: For accurate local distance calculations (e.g., UTM zones)

Distance Calculations

Always project to a projected CRS before calculating distances. Geographic CRS (like EPSG:4326) measure in degrees, not kilometers.

Code Examples

Creating GeoDataFrames from Points

import geopandas as gpd
from shapely.geometry import Point
import pandas as pd

# From earthquake data
earthquakes_df = pd.read_json('/root/earthquakes_2024.json')
geometry = [Point(xy) for xy in zip(earthquakes_df['longitude'], earthquakes_df['latitude'])]
gdf = gpd.GeoDataFrame(earthquakes_df, geometry=geometry, crs='EPSG:4326')

Loading GeoJSON with Boundaries

import json

# Load GeoJSON and convert to GeoDataFrame
with open('/root/PB2002_boundaries.json', 'r') as f:
    geojson_data = json.load(f)
boundaries_gdf = gpd.GeoDataFrame.from_features(geojson_data['features'], crs='EPSG:4326')

Projecting to Projected CRS

# Project to a suitable CRS for distance calculations
# Example: project to Mercator for global analysis
gdf_projected = gdf.to_crs('EPSG:3857')
boundaries_projected = boundaries_gdf.to_crs('EPSG:3857')

# Or use a specific UTM zone for a region
# EPSG:32633 is UTM zone 33N

Spatial Filtering (Point in Polygon)

# Check if points fall within polygons
earthquakes_in_plate = gpd.sjoin(gdf, plate_polygons, how='inner', predicate='within')

Distance to Nearest Geometry

# Calculate distance from each point to nearest boundary
def min_distance_to_boundary(row, boundary_geom):
    return row['geometry'].distance(boundary_geom)

# Distance in projected CRS (meters) or geographic CRS (degrees)
gdf['distance'] = gdf.geometry.distance(boundary_geometry)

Common Pitfalls

Calculating distances in geographic CRS - Use projected CRS for kilometers
Mixing CRS - Always ensure geometries have the same CRS before operations
Not checking polygon orientation - Invalid/reversed rings can cause issues

Best Practices

Always set CRS explicitly when creating GeoDataFrames
Project to appropriate CRS before distance calculations
Use gdf.to_crs() to transform, not gdf.crs = new_crs
Validate geometries: gdf.is_valid.all()