US Congress District Shapes

An aggregation of data modeling the United States Congress district boundaries, including a full history of past districts.

File format

GeoJSON, in the style of Who's On First records (pretty-printed, properties sorted alphabetically, with the geometry last). Note that records lack a Who's On First numeric ID; these records have not been imported into the Who's On First gazetteer (yet), I am just using the WOF encoding style.

File paths

The directory structure and filename are constructed using the following:

• 2-char state (e.g., ca for California, or dc for District of Columbia, from Python's us package)
• When the district started, by congressional session
• When the district ended, by congressional session
• District number (0 refers to at-large, -1 is for "shapes describing Indian territories within states during the 18th and early 19th centuries," from Lewis, et al.)

data/[state]/[state]_[start session]_to_[end session]_[district].[display|lookup].geojson

For example, the 1st congressional district for Idaho, from the 66th to 89th session is saved at these locations:

data/id/id_66_to_89_1.lookup.geojson
data/id/id_66_to_89_1.display.geojson

The two versions correspond to:

• .lookup.geojson includes the original geometry
• .display.geojson includes a simplified geometry

The simplified geometries are optimized for end-user display, and take up less space. We use Mapshaper for simplification (mapshaper -filter-islands min-area=500000 -simplify resolution=300).

Building from scratch

You can rebuild the data from the original sources, by using make data.

Spatialite index

You can build a SQLite/Spatialite index with: make spatialite.

PostGIS index

You can build a PostgreSQL/PostGIS index with: make postgis.

Database URL

The database configuration is controlled by the DATABASE_URL environment variable.

export DATABASE_URL="sqlite://us-congress.db"

For PostGIS:

export DATABASE_URL="postgres://user:pass@hostname/dbname"

Or a simpler version, for local PostGIS development:

export DATABASE_URL="postgres://dbname"

Dependencies

• make, curl, unzip
• Python 2.7
  • us
  • py-mapzen-whosonfirst-sources
  • py-mapzen-whosonfirst-geojson
  • py-mapzen-whosonfirst-utils
• GDAL
• Mapshaper
• Spatialite

Sources

• Jeffrey B. Lewis, et al.
  Jeffrey B. Lewis, Brandon DeVine, Lincoln Pitcher, and Kenneth C. Martis. (2013) Digital Boundary Definitions of U.S. Congressional Districts, 1789-2012.
• United States Census TIGER/Line®
• The Unified Judicial System of Pennsylvania
• The United States Senate
• Washington Post