Flattening the Range: Map Projections and the Distortion No Flat Map Escapes

Opening Scene

The ranger station’s wall map covers a park that spans real, curved ground — ridgelines, valleys, a landscape that doesn’t actually lie flat. But the map on the wall is flat, because paper is flat and screens are flat. Somewhere between the real terrain and the printed map, someone made a choice about what to preserve and what to sacrifice: true shapes, true areas, true distances, or true directions — never all four at once. A ranger estimating how far a collared wolf traveled between two trail cameras needs a map that preserves distance in that region. A ranger comparing the total area of two habitat zones needs one that preserves area. Using the wrong one doesn’t just introduce a small error — it can flip the story the map appears to tell.

Every flat map is a compromise. The only question is which compromise, and whether the person reading the map knows which one was made.

In Plain English

A map projection is a mathematical method for representing the curved surface of the Earth on a flat plane. Because a sphere cannot be flattened without distortion, every projection sacrifices at least one of four properties — shape, area, distance, or direction — to preserve others. A projection that preserves area (equal-area) will distort shape near the poles; a projection that preserves shape locally (conformal, like the widely used Web Mercator) will badly distort area at high latitudes, making, for example, Greenland look larger than Africa when it’s roughly one-fourteenth the size. Choosing a projection isn’t a technical afterthought — it’s choosing which lie your map tells.

The Old Way

Before projection choice gets deliberate thought, common defaults include:

• Defaulting to Web Mercator for everything — because it’s the projection every web mapping library ships with, regardless of whether the map is meant to compare areas, in which case Mercator actively misleads.
• Never stating which projection was used — leaving a reader with no way to judge whether the map’s apparent proportions can be trusted for the comparison they’re about to make.
• Assuming distortion only matters at global scale — when in fact even regional and local maps carry projection-driven distortion that compounds when comparing zones across a wide area, like a large national park spanning several degrees of latitude.

None of these are usually intentional deception. They’re what happens when a default projection ships silently and nobody asks whether it fits the question being asked.

What’s Changing (and Why AI Is the Reason)

1. AI-assisted mapping tools can now recommend an appropriate projection automatically based on the analytical task — suggesting an equal-area projection when the visualization is about comparing region sizes, or a conformal one when local shape accuracy matters, rather than leaving every user to default to whatever ships out of the box.
2. This raises the importance of understanding what the recommendation is actually optimizing for. A tool can suggest a projection that fits a stated goal; it cannot know whether the analyst has correctly identified what the map needs to preserve in the first place. That framing decision stays human.
3. The rise of global-scale, AI-assisted spatial analysis has made projection mismatches more consequential, since automated pipelines comparing regions across continents will silently inherit whatever distortion the underlying projection carries, at a scale no single analyst would catch by eye.

The Metaphor, Fully Extended

Ranger Station Element	Geospatial Concept
The park’s real, curved terrain	The Earth’s actual curved surface
The flat wall map hanging in the station	A map projection — a flattened representation that must distort something
A distance estimate between two trail cameras, based on the wall map	Distance measurements taken from a projection that does or doesn’t preserve distance
Two habitat zones that look equal in size on the wall map but aren’t in reality	Area distortion from using a non-equal-area projection to compare region sizes
A note pinned to the map stating which projection and distortion tradeoffs apply	Explicitly documenting a map’s projection so readers can judge what comparisons it supports

Quick Recap

• Every flat map of the Earth distorts something — shape, area, distance, or direction — because a sphere cannot be flattened without tradeoffs.
• Web Mercator, the default in most web mapping tools, badly distorts area at high latitudes, making it unsuitable for area comparisons.
• AI-assisted tools can now recommend projections that fit an analytical task, but confirming the task is correctly framed remains a human judgment.
• A trustworthy map states its projection, especially when it’s being used to support a quantitative comparison.

Where This Fits in the Series

This article closes the foundational block of the series, following location as a data type, point maps, and choropleth maps with the distortion problem underlying all of them. Article 5 moves into core technique, starting with heatmaps and the challenge of aggregating dense point data into a readable grid.