The Unicorn of Map Projections
Cartographers have long discussed the challenges of distortion in map projections and how they may influence our understanding of the world around us. Some have even suggested that map projections like Mercator and web Mercator, and their extreme exaggeration of land areas in the non-equatorial regions (for example, just look at gigantic Greenland and Antarctica!) are a sinister and intentional attempt to manipulate global world views! For instance, the Boston Public School system recently changed the maps in their classroom to use the Gall-Peters equal-area map projection instead of Mercator to present a less “Eurocentric” worldview to students.
Changing the map projection changes the way that people can see and interpret spatial relationships in the world — take a look at the difference between the Mercator (left) and Gall-Peters projections (right):
While (I think) maps are always awesome; they are also always wrong! Though there are many reasons behind this, such as the selective sampling done when collecting data, simplification of real-world locations into points, lines, and polygons, etc. one of the biggest problems with global-scale maps is the fact that we are squishing the 3-dimensional world into a flat map using a map projection. And there is no perfect unicorn map projection that can solve all of the problems that come up when we squish the world flat. I recently wrote about this problem in a paper on “The Unicorn of Map Projections” in a special issue of short papers in the International Journal of Cartography where “Cartographers Write About Cartography.”
Map projection is the process of transforming the angular measurements of the globe (e.g., latitude and longitude) into planar coordinates.
The process is sort of like when you peel an orange (except way more mathematical, and way less tasty) — you have to tear the skin of the orange, and if you want it to really be flat and in an attractive shape, you also need to squish and stretch the peel until it looks how you want.
Take a look at some of the wide variety of options! Here is an assortment to consider:
This not only shows the varied shape of different projections, but also visualizes the distortion in each of them. The red shades are angular distortion and the green shades are areal distortion — darker colors show more distortion, lighter shows less distortion. Note that every projection shown has at least one of these types of distortion!
With respect to different map projections and their distortion — they are definitely not all created equal, and picking the right map projection for a specific project involves careful weighing of tradeoffs between the distortion in the projection, the map type, and the tasks that a map reader needs to do in interpreting the map. For instance, if it’s really important to compare areas, an equal area projection is great; if it’s important to understand distance from a specific location, an equidistant projection is great; etc.
In my recent paper on the Unicorn of Map Projections, I discuss five map projections that have all attempted some claim on perfection — such as being ‘the world’s most accurate map,’ ‘the most accurate flat map of the Earth yet,’ and ‘the perfect map.’ But, none of these are really perfect or 100% accurate as all map projections distort the world in some way! However, many of them look really cool and can add some eye-catching bling to your next mapping project.
If you’re interested in learning more about these unicorn map projections, and the trade-offs on using them, you’ll find additional discussion, and a bit of great map projection humor, in the paper. If you really want to know more about map projections, I wrote a book a few years back with Fritz Kessler (Penn State University) on Working with Map Projections: A Guide to Their Selection. And if you find other great maps in the wild that are trying to earn unicorn status, let me know!
Articles sur des sujets connexes
Abonnez-vous à notre blog
Obtenez les dernières nouvelles de Tableau dans votre boîte de réception.