A closer look at the 100 mile 'border' zone using R and sf
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.



A short exploration of that 100 mile 'border' zone that the US Customs and Border Patrol get to ask you for your papers in.

Where is the 100 mile 'border' zone anyway?

Hawaii and Puerto Rico, you're definitely all in it, so I'm not going to plot you. Sorry.

More generally, how much of each US state or territory is 'in the zone'?

But that's just space. So here are the same proportions for with state populations (crudely estimated from county level data).

On the Twitter version of this graph I forgot to add in Puerto Rico, so the total percentage is a tiny bit higher now. Sorry Puerto Rico!

It's a bit hard to compare these graphs, so here's another way to look at these two sets of proportions together:

Over the diagonal states have more people than space covered by the zone. Under the diagonal, it's the opposite.

The population estimates used here are crude estimates. You should expect them to be underestimates of the true numbers affected. To see why this is true, consider the big populous counties in California overlapping the 'border' zone

Take San Bernardino County as an example.

From this census block data it's clear that most of the population is going to be in the south west - basically in Los Angeles.

Sure enough, 25% of the county but maybe 90% of its people are in the 'border' zone - around 1.5M more people than the simple interpolation would suggest. So, while quick to compute, the original numbers are on average too small.

More color you say? Well, ok. Here are two states up close with county populations and 'border' zone marked.

If you want to check my figures or ask more interesting questions than I did, the R that generated the first two maps is in border-maps.R. The R that generated the proportions is in border-states.R. And the code that did the California comparisons is in measurement-error.R. If you want to see other states besides Ohio and Pennsylvania, you'll want the function in state-plots.R.

So have at it - the license just says you shouldn't forget where you found the code.

Will Lowe. February 19, 2018