Article · measured areas, all 190 outlines
Mercator is lying about the size of these countries
Every map on your wall, every atlas spread and most of the web maps you have looked at have quietly resized the world before you saw it. The mechanism is not a conspiracy or an error — it is a mathematical consequence of keeping angles straight on a flat sheet, and it runs in one direction only: north and south get bigger, the equator stays honest. Below are the actual measured areas behind the 190 outlines on this site, with the size a Mercator canvas would show for each of them.
Africa holds 1.8 times the land of Russia
Put the two numbers side by side. Russia covers 17,098,242 km²; the 53 African countries in this dataset cover 30,215,078 km² between them, 3.2 times the land of the United States. Now draw the same two shapes on a Mercator canvas. Russia sits where the factor is 4.0, so it comes out at about 68,392,968 km² of canvas, while Africa — which barely leaves the tropics — comes out at 33,498,785. The projection draws Russia about 2.0 times the size of Africa. The land runs the other way by nearly a factor of two, and nothing about that is a matter of opinion: it is one number, applied everywhere north and south of the equator, and nowhere else.
The same test with the two big Pacific-edge states. Canada at 9,984,670 km² and China at 9,706,961 km² are within about 3% of each other in land. On a Mercator map Canada is drawn at roughly 39,938,680 km² and China at 14,466,193— about 2.8 times apart — and that gap, rather than the two countries, is what most people are recalling when they describe Canada as enormous. Brazil at 8,515,767 km² barely moves at all, because its centre sits at latitude 10°, where the factor is 1.03.
Greenland is the version everybody has heard, and the arithmetic is better than the slogan. Its land area is about 2,166,086 km²; at the latitudes it occupies the factor is roughly 10, so a Mercator drawing of it covers near 22,683,547 km² of canvas against drawn Africa's 33,498,785. The map brings the two within a factor of 1.5; the land differs by a factor of 14. Greenland is a territory rather than one of the 190 countries this site draws, which is why it appears in prose and not in the tables.
DR Congo is the cleanest single-country illustration in the dataset. It is the largest state entirely inside the tropics at 2,344,858 km², centred on latitude 0° — which is precisely where the projection leaves a shape alone. Every country above it on the sheet does not get that treatment, and that asymmetry, not any judgment about Africa, is why the continent reads small.
The equatorial band, which the projection leaves alone
For contrast, the largest countries whose outlines sit within 8° of the equator. Their Mercator factors are small enough to ignore, which means the relative sizes you see here are close to the sizes a web map shows — and the reason these shapes are hard to place is not that the map lied about their area, it is that everything around them was enlarged.
| Country | Continent | Centre latitude | True area (km²) | Area factor | Rank in dataset |
|---|---|---|---|---|---|
| DR Congo | Africa | 0° N | 2,344,858 | 1.00× | 10th of 190 |
| Indonesia | Asia | 5° S | 1,904,569 | 1.01× | 13th of 190 |
| Colombia | South America | 4° N | 1,141,748 | 1.00× | 24th of 190 |
| Tanzania | Africa | 6° S | 945,087 | 1.01× | 29th of 190 |
| Central African Republic | Africa | 7° N | 622,984 | 1.02× | 41st of 190 |
| South Sudan | Africa | 7° N | 619,745 | 1.02× | 42nd of 190 |
| Kenya | Africa | 1° N | 580,367 | 1.00× | 46th of 190 |
| Cameroon | Africa | 6° N | 475,442 | 1.01× | 52nd of 190 |
Why this matters for anyone learning the map by hand
If you only ever read a Mercator map, the proportions you internalise are the projection's, and the first time you have to draw something you reproduce them. That is not a soft criticism — on this site it is measurable. Scoring normalises where your sketch sits and how large it is, but it does not correct aspect ratio, because whether you remember a country as tall-and-thin or wide-and-shallow is the exact piece of knowledge being tested. A country drawn at its Mercator height rather than its true one is wrong in one dimension before you have put a single coastline in the wrong place.
Iceland is the cleanest single-country example in the dataset: drawn the way a Mercator map shows it, the outline needs its height cut by roughly 58% to match the real shape, and the score reflects the excess as lost overlap at both the top and the bottom of the sketch. The fix is not more care with the coastline — it is one deliberate correction to a number you already believe.
This is why the printable maps on this site are drawn in an equal-area projection rather than the one your browser's map uses. On the world blank map and the continental sheets, Spain visibly outweighs Sweden, Canada and China come out the near-equals they are, and Africa is the enormous thing it actually is. You get to learn the sizes off the page rather than off a correction afterwards — and when you then draw one of those countries, the proportions in your hand are already the true ones.
The two formulas, so you can check any map yourself
There is one mechanism here, not a list of exceptions, and it is a line of trigonometry. A Mercator map stretches a small figure at latitude φ equally in both directions — that is what keeps angles straight and the map usable for navigation — and the amount of stretch depends only on φ. Nothing below needs you to trust this site: take a country's centre latitude from any atlas and reproduce the numbers.
- 01
Lengths stretch by 1 ÷ cos(φ)
A kilometre of coastline at 60° latitude is drawn twice as long as a kilometre at the equator, in both directions. At 30° the factor is 1.15, at 45° it is 1.41, and at 60° it is 2.00 — the doubling everyone's wall map is built on.
- 02
Areas stretch by the square of that: 1 ÷ cos²(φ)
Because both dimensions are stretched by the same amount, the area goes by the square. That makes the inflation look mild at first and then runaway: 1.33× at 30°, 2.00× at 45°, 4.00× at 60°, and 10.5× at the 72° latitude that carries Greenland.
- 03
The correction a drawing needs is the square root
To un-Mercator a shape you divide its drawn height by the length factor, not the area factor. Iceland, shown 5.6 times its true area, is drawn 2.37 times too tall — which is why a sketch copied straight off a web map needs about 58% of its height taken back out.
- 04
Centre latitude is a good-enough proxy, not a perfect one
A country that spans thirty degrees of latitude — Russia and Canada both do — is not inflated by one single number; its northern edges are stretched far more than its southern ones. The tables here use each outline's measured centre latitude, which reproduces the ranking correctly and the individual figures approximately.
Applied band by band, with the largest outlines whose centres actually sit in each one:
| Latitude | Length factor | Area factor | Largest outlines centred here |
|---|---|---|---|
| 0° | 1.00× | 1.00× | DR Congo, Colombia, Kenya |
| 10° | 1.02× | 1.03× | Brazil, Indonesia, Peru |
| 20° | 1.06× | 1.13× | India, Mexico, Sudan |
| 30° | 1.15× | 1.33× | Australia, Argentina, Algeria |
| 40° | 1.31× | 1.70× | China, United States, Turkey |
| 50° | 1.56× | 2.42× | Mongolia, Ukraine, France |
| 60° | 2.00× | 4.00× | Russia, Canada, Sweden |
| 70° | 2.92× | 8.55× | Iceland |
| 80° | 5.76× | 33.16× | no outline centred here |
The twelve outlines the projection flatters most
Sorted by centre latitude rather than by area, so this is the list of shapes a Mercator map misrepresents hardest — and it is not the list of large countries. Only 2 of the twelve rank among the twenty biggest outlines in the dataset; the rest are small to medium states whose drawn areas are multiplied 2.8 times or more. If you learned the world off a wall map, these are the outlines you will be carrying around too big, and they are the ones to check first against a printed equal-area sheet.
| Rank | Country | Continent | Centre latitude | True area (km²) | Area factor | As drawn (km²) |
|---|---|---|---|---|---|---|
| 1 | Iceland | Europe | 65° N | 103,000 | 5.60× | 576,688 |
| 2 | Finland | Europe | 64° N | 338,424 | 5.20× | 1,761,072 |
| 3 | Norway | Europe | 62° N | 323,802 | 4.54× | 1,469,132 |
| 4 | Sweden | Europe | 62° N | 450,295 | 4.54× | 2,043,048 |
| 5 | Canada | North America | 60° N | 9,984,670 | 4.00× | 39,938,680 |
| 6 | Russia | Europe | 60° N | 17,098,242 | 4.00× | 68,392,968 |
| 7 | Estonia | Europe | 59° N | 45,227 | 3.77× | 170,498 |
| 8 | Latvia | Europe | 57° N | 64,559 | 3.37× | 217,640 |
| 9 | Denmark | Europe | 56° N | 43,094 | 3.20× | 137,814 |
| 10 | Lithuania | Europe | 56° N | 65,300 | 3.20× | 208,829 |
| 11 | United Kingdom | Europe | 54° N | 242,900 | 2.89× | 703,056 |
| 12 | Belarus | Europe | 53° N | 207,600 | 2.76× | 573,194 |
The ten largest outlines, true and as drawn
The ranking that matters most for anyone learning the world, put through the projection line by line. The true-area column is measured from the shipped geometry; the last column is the same shape re-drawn on a Mercator canvas at its own centre latitude. Read down the two area columns and the order changes; read the last column and you have the ranking most people carry in their heads.
| Rank | Country | Continent | Centre latitude | True area (km²) | Area factor | As drawn (km²) |
|---|---|---|---|---|---|---|
| 1 | Russia | Europe | 60° N | 17,098,242 | 4.00× | 68,392,968 |
| 2 | Canada | North America | 60° N | 9,984,670 | 4.00× | 39,938,680 |
| 3 | China | Asia | 35° N | 9,706,961 | 1.49× | 14,466,193 |
| 4 | United States | North America | 38° N | 9,372,610 | 1.61× | 15,093,719 |
| 5 | Brazil | South America | 10° S | 8,515,767 | 1.03× | 8,780,532 |
| 6 | Australia | Oceania | 27° S | 7,692,024 | 1.26× | 9,688,998 |
| 7 | India | Asia | 20° N | 3,287,590 | 1.13× | 3,723,111 |
| 8 | Argentina | South America | 34° S | 2,780,400 | 1.45× | 4,045,376 |
| 9 | Algeria | Africa | 28° N | 2,381,741 | 1.28× | 3,055,095 |
| 10 | DR Congo | Africa | 0° N | 2,344,858 | 1.00× | 2,344,858 |
What to do about it if you are learning the map
Three things, in order. First, print an equal-area sheet and keep it next to whatever you usually look at, the Europe blank map being the one where the distortion is worst, since the continent sits at 35–70° north and its northern countries are the ones people place highest. Second, draw the two or three countries you are most confident about and let the score argue with you: Norway and Finland are both long enough in latitude for a Mercator-trained hand to overshoot badly, and both are far smaller than the sheet they come off suggests. Third, memorise three pairings rather than a projection — Africa to Russia, Canada to China, Brazil to the country you think it should be next to. Getting those three relative sizes right fixes more drawing errors than any amount of coastline practice, because they set the scale the rest of the map is fitted to.
Related
Map projections and country sizes — questions
Why does Mercator make countries look bigger than they are?
Because Mercator keeps angles constant, which is what makes it usable for navigation, and the only way to do that on a flat sheet is to stretch everything more the further you get from the equator. The stretch is the same in both directions at any given point, so it multiplies lengths by one over cosine of the latitude and areas by the square of that. At the equator the factor is 1. Iceland, centred around 65° latitude, is shown 5.6 times its true size; Russia about 4.0 times. Near the equator almost nothing changes — Brazil is inflated by only 1.03.
Is Africa really bigger than Russia?
Yes, by a wide margin. The 53 African outlines here total 30,215,078 km², while Russia alone is 17,098,242 km² — Africa's total is 1.8 times that. On a Mercator map Russia dominates the top of the frame and Africa sits in the middle as a modest compact shape, because the projection inflates the first by about 4.0-fold and the second by almost nothing. That visual impression is wrong, and it is wrong in a way almost everybody reproduces when they draw either continent.
Is the Greenland-and-Africa comparison true?
It is the most quoted example of this distortion, and the underlying arithmetic is easy to check. Greenland's land area is about 2,166,086 km², roughly one fourteenth of Africa's 30,215,078. Plot it on a Mercator canvas at its real latitudes and the 10-fold inflation puts its drawn area near 22,683,547 km² — so on a classroom map the two look comparable, when Africa is about 14 times the size. Greenland is a territory rather than one of the 190 countries drawn here, which is why it is not in the tables above.
Which map projection shows the true sizes?
Any equal-area one. Lambert azimuthal equal-area is what this site uses for its continental blank maps, and it is the projection the European Environment Agency uses for Europe; Mollweide, Gall-Peters and Eckert IV are the other names you will meet. They all trade something else away — shape, angles, or the neat straight-line graticule a web map has — because you cannot have true areas and true angles on a flat sheet at the same time. Equal-area is the right trade for anyone trying to learn where things are and how big they are.
Does this actually change how I draw a country?
It changes the sizes you expect, which is the thing being measured. The accuracy score on this site scales your sketch and the real border into the same box and compares the overlapping areas, but it deliberately does not correct aspect ratio — because remembering that a country is wide and shallow rather than tall and narrow is exactly the knowledge under test. So a country you have only ever seen on Mercator comes out stretched: Iceland drawn the way a web map shows it needs its height cut by about 58% to get back to the real shape, before you have made a single other mistake.
Which countries does Mercator flatter the most?
The ones at high latitude, and in this dataset all of them are in the northern half of the world. Iceland is the most inflated, at 5.6 times its measured area on a Mercator canvas, followed by Finland and Norway. Every outline in that group of twelve sits above 53° latitude, they cover Europe and North America, and each is drawn at least 2.8 times its true area — which is why the top third of a Mercator world map carries so much of its apparent landmass.
Which countries does Mercator get right?
The ones on the equator. DR Congo, Indonesia, Colombia and the rest of the equatorial band are shown at close to their true relative size, since the inflation factor only becomes 1.02 even at 8° latitude. The practical consequence is a systematic bias in the other direction: the countries everyone finds hardest to place in proportion are the large ones near the equator, because they are the ones a wall map has been quietly shrinking next to the countries above them.
Why do people draw Scandinavia too big?
Because they learned the shape from a Mercator-style map and the relationship between the parts was never taught separately. On a Mercator canvas Iceland is shown 5.6 times its area and Sweden about 4.5 times, while DR Congo — 23 times the land of Iceland, and 5.2 times Sweden — is shown almost honestly. The result is a mental ranking that does not survive contact with an equal-area map. As a drawing exercise the fix is to size the northern states against the equatorial ones rather than against each other, because a group of neighbours at similar latitude is inflated at similar rates and their relative shapes are the one thing Mercator keeps roughly right.
Do equal-area maps distort anything?
Shape, and that is the point of the trade. In Lambert azimuthal equal-area the distortion grows as you move away from the point the map is centred on, so countries at the edge of a continental frame are stretched sideways rather than scaled up. It is why this site frames each continent on its own box instead of putting the whole world in one projection: an outline drawn near the centre of its own equal-area frame keeps its shape well enough to be recognisable, and keeps its area well enough to be honest.