Plate tectonics data processing

Plate Tectonics Data Processing

This skill covers the use of the PB2002 (Bird, 2003) dataset for tectonic plate analysis.

1. Loading PB2002 Data

The PB2002 dataset typically consists of:

• PB2002_plates.json: Polygons representing the extent of each tectonic plate.
• PB2002_boundaries.json: LineStrings representing the boundaries between plates.

Implementation

import geopandas as gpd

plates_gdf = gpd.read_file("PB2002_plates.json")
boundaries_gdf = gpd.read_file("PB2002_boundaries.json")

2. Filtering by Plate Name

The Plate column (or similar) in the plates dataset identifies each plate.

• Pacific Plate: Often identified as "PA" or "Pacific".

pacific_plate = plates_gdf[plates_gdf['Plate'] == 'PA']

3. Spatial Joins and Point-in-Polygon

To find which earthquakes occurred within a specific plate:

# Assuming earthquakes_gdf and plates_gdf are in the same CRS
earthquakes_in_plates = gpd.sjoin(earthquakes_gdf, plates_gdf, how="inner", predicate="within")

# Filter for a specific plate
pacific_quakes = earthquakes_in_plates[earthquakes_in_plates['Plate'] == 'PA']

4. Identifying Relevant Boundaries

Boundaries are often labeled with the names of the two plates they separate (e.g., "PA-NA" for Pacific-North America).

To get all boundaries associated with the Pacific Plate:

# Search for 'PA' in the boundary name/description column
pacific_boundaries = boundaries_gdf[boundaries_gdf['Name'].str.contains('PA')]

5. Geometric Cleanup

Tectonic datasets can sometimes have minor self-intersections or gaps.

• Use gdf.geometry = gdf.geometry.buffer(0) to fix minor validity issues.
• Use gdf.union_all() to create a single geometry representing the entire plate or boundary set.