An exploration of the geographic balance of counties
In an area where the population is evenly and symmetrically distributed, the geographic center and the center of population will be close together. In an area where people live clustered on one corner or edge, these points will be far apart. This visualization uses the linear distance between these two points to explore the most balanced and the least balanced counties (and county-like units) in the United States.
See the blog post which explains more about this project's concept.
The population centroid data comes from the Census Bureau's Mean Centers of Population, 2010.
Geographic centroids were computed using QGIS on the Census Bureau's 20m County Boundaries shapefile.
Raw distance is calculated in meters by data/calculator.py using the Vincenty distance algorithm in geopy. Weighted distance is the raw distance divided by the square root of area.
This data was massaged and combined into data/output-data.csv, and an id field in the format SSCCC was generated, where SS = State FIP; CCC = County FIP.
The 20m County Boundaries shapefile was then transformed into a TopoJSON using ogr2ogr and topojson, again generating an id field in the SSCCC format.
- The Aleutians West unit causes all sorts of headaches for QGIS, d3, and Leaflet alike due to the fact that it crosses -180 latitude. It's been removed for now.
- Need to implement a rank-number in the list