Hi, Brent.

In your original message, you mentioned the cb_2020_us_all_500k.zip
boundaries, which are clipped to the shoreline.

You might also be interested in looking at the Tiger/Line shapefiles
that are available here:
https://www.census.gov/cgi-bin/geo/shapefiles/index.php

These boundaries are not clipped to the shoreline, so the polygons for
counties like San Juan and Whatcom extend out into the water and
actually share a border, and thus are truly adjacent polygons.  I'm
attaching a map image to illustrate -- the thick borders are the
Tiger/Line shapefiles.

The US Census Bureau also publishes a county adjacency file which I
think is based on the Tiger/Line polygons.
https://www.census.gov/geographies/reference-files/time-series/geo/county-adjacency.html

Keith


On Wed, Mar 11, 2026 at 3:18 AM Brent Blumenstein via QGIS-User
<[email protected]> wrote:
>
> Thank you for the response. Yes I agree that some "handwork" could be used 
> here but I think I should pursue the automated solution until I cannot 
> implement it. I have found that the using the census data does indeed find 
> counties where the coordinates indicate isolation, so my plan to focus on 
> isolated counties is what I am pursuing at the moment.
>
> However, I found that the census data does not provide a consistent 
> identification of isolated counties across states. There is a huge difference 
> between Masachuchetts and Washington states. The census data does identify 
> Dukes County as isolated but does not identify San Juan County as isolated. 
> In fact, San Juan county is connected to Whatcom and Skagit Counties but does 
> not identify San Juan County as connected to Island, Clallam and Jefferson 
> Counties, which are clearly neighbors even though a little further away. 
> Dukes County and San Juan Counties have roughly the same distance to the 
> nearest county, so there must be some distance parameter at work when the 
> dataset was created, but this theoretical parameter seems to implemented 
> differently in these two states.
>
> This finding complicates the algorithm I dream about refining.
>
> BTW: The coordinate files from xxx.yyy say that San Juan County has five 
> neighbors {Whatcom, Skagit, Island, Clallam and Jefferson Counties}. This 
> result seems correct to me geven the goals of the project. Similarly, xxx.yyy 
> identifies Dukes County as having mainland counties as neighbors.
>
> I have attached maps for reference.
_______________________________________________
QGIS-User mailing list
[email protected]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Reply via email to