Wednesday, October 19, 2016

Trees, Trees, Trees!

Autumn is here — it's time for trees!

Three tree-related projects. First is a relatively simple (but data-intensive) rethinking of tree distributions. Instead of the usual species-level blob maps, I've made a series of maps showing the actual distribution of major tree types, plus some interesting higher-level patterns.

I've then used this same data as a starting point for a new kind of bioregionalist mapping: instead of a few discrete forest regions, I'm defining "arborregions" based on the similarity between a specific place and its wider surroundings. If bioregions are really meant as an alternative to the arbitrary lines of political jurisdictions, they should challenge not just the specific boundaries, but the hardness of those boundaries as well.

Finally, I've also taken a close look at every street tree in New Haven. After the ravages of Dutch elm disease, it turns out that Elm City doesn't actually have that many elms. Instead the city is a weave of a half dozen major tree types, with dozens of others scattered throughout.

And a quick bonus map, too: the apizza region of central Connecticut!

Friday, July 1, 2016


There is a book! At long last, there is a book!

After the Map: Cartography, Navigation, and the Transformation of Territory in the Twentieth Century has just been published by the University of Chicago Press. More info — along with high-res images, raw data, and a bibliography — is available on the book's website, I've spent almost a decade researching, writing, and revising this thing; I hope you like it!

To buy the book, the best prices are listed on

Monday, May 23, 2016

How high are the humans?

A quick bonbon! The global distribution of human beings by altitude: a histogram showing the number of people living at every elevation. Not surprisingly, coast-loving humans are a low-altitude species, and the distribution of humans is quite a bit lower than land in general — not even counting ice domes and barren deserts. Quick take-away: when you look out from the top of the Washington Monument, you are higher than half of everyone else in the world.

I also found some additional data for the early decades of American slavery: 1790 data for what's now Tennessee, plus small tweaks to coastal South Carolina and Indian lands in Kentucky before 1820.

Friday, May 6, 2016

Slavery in the North

The last of my trio of slavery projects: an interactive map of slavery in the north, town by town. Although it's easy to overlook northern slavery in comparison to its huge presence in the south, at the founding of the United States it was a serious part of the northern economy, especially in areas in New York and New Jersey first settled by the Dutch. Over two thousand slaves lived in New York City in 1790, and more than 60% of white families in what is now Brooklyn were slave-owners. Nearly every town in Pennsylvania, Connecticut, and Rhode Island had at least a few slaves.

The main task of this project was getting the data, but I'm also trying some new techniques for blending interactive and static mapping. Town-level data has always been available for the north (at least after a bit of math), but it has never before been mapped or digitized. Not surprisingly, disaggretating 90 counties — many huge and unhelpful — into 1,600 towns means that new patterns emerge, and it's possible to connect broad trends with local reality in a new way. The interactive map gives detailed information about every town, but I've also made sure that the project can be downloaded as a stand-alone digital poster.

Friday, April 22, 2016

Slavery in the United States

I'm pleased to share a major new project on the history of slavery in the United States. Even after 155 years of mapping slavery, there are still serious shortcomings in most typical maps. My strategy looks for a way around the straightjacket of county-based data and the false impression of spatial precision implied by sharp county boundaries. I incorporate historical data on more than 150 cities and towns; I also use dots instead of counties. Not only does this help to distinguish rural and urban areas (which often had sharply different levels of slavery), but it makes it possible to see population density and the predominance of slavery at the same time. I've posted a graphic explanation of my strategy here.

The project also includes a map of "peak slavery" that shows the maximum number of slaves that ever lived in an area, along with the year of the peak. In the vast majority of the south, slavery was booming right up to the Civil War; only in Delaware, Maryland, and eastern Virginia was slavery in natural decline.

Monday, April 11, 2016

Slave Insurance (and more railroads)

Two things!

First is a collaboration with Michael Ralph on the history of slave insurance in the US. Most insured slaves were highly skilled, and they were disproportionately urban. They were usually rented to others — especially on Ohio River steamboats, in Virginia coal mines, and in skilled trades in Atlantic port cities. In many ways, what we see on the map is an unfree version of the emerging relationship between life insurance and wage labor in the north. And we know their names.

Second is a new version of my map of world railways, updated with new data and a much-higher-resolution download!