The maps showing how much of the United States is empty is visually interesting, but mostly demonstrates the importance of the data-aggregation area. A similar map for Canada is flawed by mistaking "no data available" for "no one is there."
Michael Chung's map follows the same colour scheme as Freeman's and Sherwood's: white are populated places, and green are empty. One small problem: to protect privacy, Statistics Canada does not release data in census areas with too few people. They also don't release data in areas where they think the data quality is too low, or for some First Nations reserves. A lack of data does not mean a census tract is empty — it could mean it is sparsely populated, it is First Nations territory, or that the data quality is too low — but those regions are all coloured green on the map instead of a third colour for "No Data Available."
Considering that the huge number of census tracts for which no or limited data were released are closely correlated with areas that have limited populations, it's pretty ingenuous to list all those areas as unpopulated. In addition to those, a big chunk of First Nations reserves have limited data releases — if you look at the map of First Nations lands, you'll notice a whole lot of dense red dots tracing the "uninhabited" central interior. Aggregating to a larger data-collection area (a census subdivision) reduces this problem, as in this Atlas of Canada map:
This presentation still has a data-aggregation problem, as does Chung's revision of his original map using dissemination blocks. If you click over to the Atlas of Canada you can check out the same data clumped to different scales, and a density-map with people per unit area.
My next problem with the map can't be fixed by more careful use of census data, and is related to the "Thar be dragons" wilderness interpretation of uninhabited regions. Canada is a resource-heavy country, with a large number of temporarily-inhabited locations. Mining camps, forestry camps, and exploration camps are all seasonally inhabited, with populations that wax and wane throughout the year. I once spent nearly 300 days in a single year bouncing around resource industry camps, yet my place of residence, my registered home for taxes and voting, was in the city, not some camp in the outback. Census data cannot capture the greater complexity of where my migratory co-workers and I spend our time.
We don't live here, we just work here. Photography credit: Mika McKinnon
Alice Arm, British Columbia, is a ghost town where the post office hasn't been open since the 1800s, and has a spring-through-fall population of 20 to 50 people before boarding up the windows each winter. Galore Creek, British Columbia has had annual exploration or development work since the 1950s, and at one point had its own camp-ambulance. I can't even locate a map of most of the resource-camps I've travelled through, yet they have a distinct annual population, frequently with the same core people returning every year. Do these temporary homes for really count as wilderness? With a third of a million workers in the mining industry, another quarter million in forestry, and the uncounted non-direct employees (every camp has a cook!), that's an awful lot of people to ignore when the national population tops out at under 40 million. Each of those camps aren't just seasonal settlements — they're also centers for exploiting and modifying the environment, shifting the surrounding location from "wilderness" to "resource."
Canada is big, with a very small population. It's true that the population really is centered around the southern border and the coasts. But this map over-simplifies and ignores large swaths of low-population-density areas, and that is far less interesting than the true story of our sparse and migratory population. It's a beautiful image, but it's one that visually deceives about places-that-aren't-cities being wilderness when so much is impacted by logging, mining, and exploration. Maps are propaganda for raw data, a way of telling stories with numbers. I'd love to see a map of Canada that reflects our complex interaction with temporary settlements, native land, resource development, and the true wild. More importantly, I'd hate to see this map or others like it spread the myth of a country of vast, untouched wilderness when our reality is far more fragile.
For more on maps, consider these historic maps of the Moon or of the Earth. Chung has joined in the Discussion if you have any questions for him directly. Here's some more Canadian content for you!