Mercator and Equal Earth maps side by side

A Mercator map (left) and Equal Earth (right). Graphic made with Natural Earth data.

I've recently had a few non-mapping friends ask me about the United Nations' recent, near-unanimous decision to support the use of the Equal Earth projection for world maps.

The African Union has been advocating for this change, motivated by a very longstanding criticism of the familiar Mercator map. Mercator is probably the most common way to visualize the world, but it contains a notable distortion: places near the poles (like Greenland) are too large, and places near the equator are too small. Consequently, regions near the equator, like much of Africa, are given short shrift: they look much smaller than they really are in relation to other landmasses. (My colleague at the Map and Data Library, Nick Field, wrote about this in a post last year.)

The news stories and social media discussions I've seen have framed the decision as a simple replacement of an inherently 'bad' map with a 'good' one (or, more rarely, vice versa). There's more nuance to how we use different maps than that, though and these nuances are worth exploring for anyone working with maps and spatial data.

What is a Map Projection?

The Earth is curved, while paper and computer screens are flat. To make a map, we squash the Earth's three dimensions into two using a mathematical translation called a projection. To our dismay, however, it's physically impossible to perfectly, accurately translate everything from those three dimensions onto a flat plane.

So, every map projection must compromise somewhere. Area, shape, distance, direction, scale, or some combination of these properties are going to be distorted. There's no way around this. Most projections prioritize certain properties at the expense of others. If you're working with spatial data, your job is to decide which properties matter most, and which trade-offs make sense.

Why Was the Mercator Projection Created?

1569 Mercator map

Composite view of the Mercator 1569 world map from the copy at the Bibliothèque Nationale de France. Public domain. https://en.wikipedia.org/wiki/File:Mercator_1569_composite.jpg (visit this link to see a larger version).

The Mercator projection was developed in 1569. Its creator wasn't trying to make the most visually accurate picture of the world. Instead, this projection was meant to solve a practical problem of its day. Using Mercator's innovative globe-squashing formula, angles and directions were accurately preserved. This meant that a ship maintaining a constant compass direction could plot its route as a straight line on a map. For marine navigation, this property was much more valuable than scale.

I'd still endorse this projection as an excellent choice for anyone planning a sixteenth-century ocean voyage (or studying the people who did). It's a poorer choice for a classroom wall map intended to show the relative sizes of the continents.

Why Equal Earth?

Equal Earth map

Equal Earth projection SW.jpg CC BY-SA 4.0 https://commons.wikimedia.org/wiki/File:Equal_Earth_projection_SW.jpg

Equal Earth prioritizes what Mercator does not: the relative sizes of landmasses. This type of projection is called an equal area projection: countries and continents appear in proportion to their actual surface area on Earth. Africa looks enormous, because it is.

I should mention: Equal Earth is not the first attempt to solve this problem! In particular, the Gall-Peters projection (originally developed in the nineteenth century) was marketed in the 1970s as a more equitable alternative to Mercator. The famous controversy around this shows us that map projections aren't just technical tools; they have a political and cultural significance that can prompt very emotional reactions.

The Equal Earth projection was introduced in 2018. As is the case with all projections, it's not perfect - in its case, shape is distorted at the expense of area. Its creators understood this, though, and it does a good job of meeting their intended goals: preserving the relative sizes of landmasses while being aesthetically pleasing.

And another note: the UN hasn't banned the Mercator projection or declared Equal Earth the only right way to map the world. All they've really done is voiced their support for it as a good projection choice for general-purpose world mapping.

There's No 'Correct' World Map

There are a theoretically infinite number of possible map projections, preserving and distorting different spatial properties, to choose from. In its maps of the world, National Geographic chooses to use the Winkel tripel projection. Winkel tripel is a compromise projection that attempts to minimize distortions to area, direction, and distance all at once, without being absolutely faithful to any of these (that's the 'compromise').

Many projections are optimized for particular regions. If your goal is to produce a map of Canada, you might choose a different projection than if you were mapping Africa, South America, or the entire globe. (Or even other planets, but let's stick to Earth for now!)

If you download GIS boundary files to spatially analyze and map Canadian census statistics, you'll likely be working with data projected to the Lambert conformal conic projection. That's because Statistics Canada has chosen Lambert conformal conic as a good option for mapping all of Canada.

Many GIS projects use the Universal transverse mercator (UTM) system for small study areas. In this system, the Earth is divided into narrow strips called zones. Each zone gets its own localized projection, making distortion pretty negligible for everything within that zone.

Dymaxion world map projection

Dymaxion projection.png. Public domain. https://commons.wikimedia.org/wiki/File:Dymaxion_projection.png

If you decide to get creative, you might come across some interesting novelties. An interrupted icosahedral projection (such as the 'Fuller' or 'Dymaxion' option available in ArcGIS), for example, unfolds the globe as a 20-sided shape. You can use this sort of projection to create really cool cartography where no direction is 'up'! However, their visual interruptions and unusual shape will present their own weird challenges. Anyway, the important lesson is that, as with any tool, different projections are suited to different tasks.

Choosing a projection for your own work

So now you know that, rather than simply being 'good' or 'bad', every map projection contains within it decisions about what to preserve and what to distort. If you're using spatial data in your work, how do you decide which projection to use?

The good news: you don’t need to have any knowledge of the mathematical formulas underpinning projections to use them effectively. GIS software will be happy to transform your data layers to different projections for you (see MDL's Projecting data in ArcGIS Pro and Projecting data in QGIS tutorials for a technical how-to).

All you need to do is think about your study area and what you’ll be measuring or analyzing. And if the visual presentation of your work is important, it is absolutely valid to consider the aesthetics of the projection too.

For relatively small areas – for example, mapping transit routes in a city – the location’s corresponding UTM zone is often a good choice. This will preserve shapes well and generally minimize distortion. For Southern Ontario, you’ll typically use Zone 17 (which you’ll see specified in QGIS or ArcGIS's options as NAD 83 UTM Zone 17N or similar). You can find out what UTM Zone any place is in using our web map.

If your study area is larger than the width of a single UTM zone, you have options! For mapping a province or state, you may find that projections have been developed specifically for that area – for example, Ontario MNR Lambert conformal conic is often used by provincial authorities here in Ontario. But, like Mercator, it preserves angles rather than areas (that’s what ‘conformal’ means). You could take a look at Albers equal area or another equal area projection if area calculations are important to your work.

In projects that involve mapping the entire earth, your goals might be analytical (examples: accurately quantifying different land cover areas, or calculating travel distances). If this is the case, choose a projection that preserves whatever spatial properties you'll be analyzing. Alternatively, if you simply need to display the earth graphically, you should choose a projection that meets your aesthetic goals – one that your intended audience can readily understand, that visually works as a communication tool.

When I first started learning GIS, I worried about how to choose the ‘objectively correct’ projection for my maps. I'd look at seemingly endless lists of possible projections and feel overwhelmed by the options. Now that I have a better understanding of how projections work, I’ll encourage you not to stress like I did: just choose a one that works well with your study area and your mapping goals, and you’ll be fine!

Further Learning

Books available from U of T Libraries

Map use: reading, analysis, interpretation https://librarysearch.library.utoronto.ca/permalink/01UTORONTO_INST/rj2ppj/alma991106048383706196

Making maps : a visual guide to map design for GIS https://librarysearch.library.utoronto.ca/permalink/01UTORONTO_INST/n5nrf0/alma991107441531106196 (see Chapter 5: Geographic Frameworks)

Websites

Esri tutorial: Choose the right projection https://learn.arcgis.com/en/projects/choose-the-right-projection/ (Note: this is designed for the ArcGIS Pro software, but nonetheless contains good general information on projections. You may especially be interested in the comparative chart of different projections).
Michael Minn’s list of world map projections: https://michaelminn.net/tutorials/gis-projections-world/
Projection Wizard: https://projectionwizard.org/
MDL's UTM Zone web map: https://utoronto.maps.arcgis.com/apps/mapviewer/index.html?webmap=fc3b38f3fbd34657866c6a45bde0f784&mapOnly=true

If you’d like help with map projections, or anything else related to maps or data, you’re very welcome to contact the Map & Data Library any time. We're happy to help!

You might also like:

Map(s) of the Month: Global Map Projections

Equal Earth, Mercator, and how to choose a map projection