Mercator preserves local shape, not relative area.
Precise claimMercator is conformal, not equal-area. Its point scale grows as sec(latitude), so drawn area inflation grows as sec²(latitude); Africa is roughly fourteen times Greenland’s true area.
Applies
The live demo applies the point-scale law at a simplified landmass center latitude to expose the mechanism. Area figures are rounded reference values.
Does not prove
Simplified outlines and center-latitude scaling are not survey-grade area measurement. Other projections have different trade-offs, and Mercator remains useful where conformality matters.
Portable rule
If the map is Mercator, then compare latitudes before comparing areas.
Declare the discussion context, then copy a stable link. No account or personal data is attached.
Is Greenland Really the Size of Africa?
On the map below they look like a fair fight. Before anyone corrects you: how many Greenlands do you think make one Africa?
Your system prefers reduced motion, so this page skips its entrance animations and smooth scrolling. Nothing here runs on animation — the map still drags and the numbers still compute.
Interactive
Simplified outlines on a Mercator-style grid — schematic, not a survey. Greenland highlighted.
The reveal
Mercator is conformal, not equal-area. Its point scale grows as sec(latitude), so drawn area inflation grows as sec²(latitude); Africa is roughly fourteen times Greenland’s true area.
You locked in: —.
≈ 30.2Mkm² — Africa
≈ 2.16Mkm² — Greenland
≈ 14×Greenlands in one Africa
you: —truth: ≈14
051015
On a Mercator map every feature is stretched by sec² of its latitude. Africa straddles the equator, where the stretch barely exceeds ×1. Greenland sits near the top, where the stretch averages about ×10. The map didn't mismeasure Greenland — it applied a rule you weren't told about.
Interactive
See it — drag Greenland home to the equator
72°N — the map draws Greenland ×3.2 (linear), ×10.5 (area) vs. its true size.
The outlines are simplified; the shrink is not drawn in advance. It is sec(latitude), computed live from wherever you leave the block.
The map was drawn to steer by. You were reading it to size by.
The rebuilt model
You assumed:
A familiar world map shows countries at their true relative sizes.
The actual model:
A flat map cannot preserve every property. Mercator preserves angles and local shape by increasing scale toward the poles, sacrificing area comparison.
The variable that failed you:
Latitude-dependent scale, hidden behind a familiar outline.
One more pair — same map
No slider this time. Picture the Mercator wall map, then commit.
You picked: —.
Right — and you likely had to override the map to say it. Russia ≈ 17.1M km²; Africa ≈ 30.2M km², about 1.8×. On the wall map Russia sprawls across the whole top of Asia like the biggest thing on Earth. It's the Greenland trick, one latitude band south.
The map got you again — or the memory of it did. Russia ≈ 17.1M km², enormous, and stretched further by sitting at 50–70°N. Africa ≈ 30.2M km²: about 1.8× bigger. Same rule, second pair.
Find any Mercator world map — a classroom poster, or the default view of most web maps.
Compare Greenland and Africa by eye. Write down your gut ratio before measuring anything.
Check the stretch where Greenland sits: 1 ÷ cos²(72°) ≈ 10.5. The playground below does the math for any latitude.
Now look the areas up in any atlas: Africa ≈ 30.2 million km², Greenland ≈ 2.16 million km². The ≈14× was never drawn on the map — only its distortion was.
Interactive
Playground — the Mercator area tax, by latitude
60° latitude → ×4.00 area
Every value is 1 ÷ cos²(φ), computed as you move the slider — nothing pre-rendered. At 60° the factor is exactly 4; at 80° about 33. The bar saturates at ×100; the formula doesn't — the pole itself sits at infinity, which is why no Mercator map contains it.
Scope: The live demo applies the point-scale law at a simplified landmass center latitude to expose the mechanism. Area figures are rounded reference values.
Does not prove: Simplified outlines and center-latitude scaling are not survey-grade area measurement. Other projections have different trade-offs, and Mercator remains useful where conformality matters.
Take it with you
The portable rule
If the map is Mercator, then compare latitudes before comparing areas.
Data visualization
Use an equal-area projection when region size is the quantity being compared.
Design review
Name the property a map preserves before approving spatial comparisons.
News graphics
Treat high-latitude visual dominance as a projection effect, not geographic magnitude.
Interactive
Apply it — five seconds
Right. On a Mercator choropleth the blob is mostly latitude. The figure is the data — and Y's is about 15× X's.
The blob is mostly latitude. Compare the numbers: Y reports about 15× the cases of X.