Hacker News Reader

So that region is not more densely populated than the rest of the world? That makes me a bit skeptical of your math because it obviously is. For one, you are using habitable land mass for the rest of the world while the circled area is mostly sea water. I wonder how that comment got to the top.

It's more densely populated because it has the highest concentration of habitable land.

Siberia to the north from there is not inhabited so densely because it has -50°C in the winter and +50°C in the summer.

Western Europe, US northeast, etc are also rather densely populated, but are smaller, with less-popuated areas, or with oceans.

I would love to know the region in Siberia in which there is +50 in the summer.

Oh, sorry, +50 is definitely a typo; +30. Say, Yakutsk is a major city in Siberia with a climate like that.

The circle contains 1/3 ocean and ocean is not landmass. You'll have to revise your calculation. Also, since the circle contains Tibet, desert in Xinjiang and Mongolia, your assumption that all of that is habitable land is even more wrong.

The water is a fair point, but GP didn't exclude Sahara, middle east or other deserts, so it would not be proper to exclude Xinjiang or Mongolia.

It clearly says " 2,500 miles in diameter" which makes it 1,250 mile radius.

pi * 1250 *1250 = 4,908,738 square miles or ~5 million square miles.

That is < 10% of the earths land mass.

Yeah this is the error in the parent calculation

It also clearly says "2500 radius circle" in the article title, and the submission title. The article is fast and loose with diameter vs radius later on as well. I can't blame the original commenter for using it as a radius when that's in the title.

But I agree, it's actually the diameter and changes the math a lot.

The distance from Mumbai India to Osaka Japan is ~4000 miles. Both of which are clearly inside the circle.

What are you using to assert that the 2500 mile number is the diameter?

This actually fits with the visual impression you get from the map. It's clearly far less than 50% of the inhabitable land mass

you also need to map the circle firs to the surface, a circle on a world map is not the same as a circle on the globe.

> Myers was amazed to find that a circle roughly 2,500 miles in diameter, when placed over the right part of Asia, could hold more than half the world's population.

The above is from the article. It says radius in the title and diameter in the article

A history of population within the circle would be interesting.

The population of China since 2000 years ago appears to have been a series of plateaus with brief periods of increase and a couple of brief decreases. The plateaus represent populations at the limit of agricultural systems. The increases are due to introduction of new systems, such as rice cultivation further north, double-cropping of rice and so forth. The decreases are periods of famine, disease and war. http://www.china-profile.com/data/fig_pop_0-2050.htm

Possibly other civilizations within the circle follow the same pattern of stability, innovation and crises.

If I recall correctly, rice can sustain a much greater number of people per unit of land than any other crop, and this explains why East Asia is more densely populated than any other part of the world.

Corn and potatoes each produce around 15 million calories per acre; rice only hits about 11 million. Wheat is significantly worse, at ~4 million.

The introduction of potatoes led to a rapid population increase in Europe during the first half of the 19th century.

https://www.history-magazine.com/potato.html

Sugar cane and sweet potatoes also produce similar calories per acre, depending on the soil, climate, etc. Sugar also has the advantage of being durable to ship. It could be produced in high volumes on tropical plantations and shipped to Europe.

Before bulk ocean shipping of food, population was generally limited by calorie production in the local area.

The big difference with rice is that you aren't dependent on the weather nearly as much as you are with corn and potatoes. Corn and Potatoes put out more calories per acre when the rains fall at the right times in the right amounts, but rice puts out those calories regardless of the weather.

Ultimately you can only sustain a population as large as your food output at the end of a two or thee year long drought.

Until the industrial revolution, the human population everywhere was in equilibrium with agricultural technology. Improvements to technology would lead to population growth until wealth per capita reached the same old level and people stopped having as many kids. Check out the book "A farewell to Alms" for a solid 10 chapters of historical records on this from around the world.

The first person to notice this universal truth about the human population was Malthus. He had the misfortune to make the observation right on the eve of era when it stopped being true.

A universal truth doesn’t stop being true... So it’s probably better to revise it to say that it behaves that way under certain conditions and those conditions have changed.

52M km^2 of land on our planet is habitable, after forests, shrubs, rivers and lakes [0]. Total land mass is 149M km^2. So only 1/3 of the world's land is habitable. That's actually ~20M square miles. Your habitable land estimate is already nearly 50% off according to the source I found.

The circle covers 50.2M km^2 (4000^2 * pi). I'm going to assume the 2/3 land quotes are good enough an estimate. That makes the land mass inside the circle 33.5M km^2 (2/3 * 50.2). That means that the circle contains just 22.5% (33.5M/149M) of ALL land mass.

I couldn't find any quick data source on how much land inside the circle is actually habitable. So I'll apply a few ratios:

If X% of the land in the circle is habitable:

  X = 33% --> 33% * 33.5M / 52M = 21.5%  
  X = 50% --> 50% * 33.5M / 52M = 32.2%  
  X = 75% --> 75% * 33.5M / 52M = 48.3%

Your estimations would hold iff >75% of the land in the circle is habitable.

[0]: https://ourworldindata.org/land-use

1/3rd of that circle is water. And a significant amount of it is also Siberia and Mongolia. Double check your math.

> 1/3rd of that circle is water.

Fine. Thrown in the saraha desert and a few other inhospitable areas and it'll make up that third.

> And a significant amount of it is also Siberia and Mongolia.

Are we looking at the same map? Hardly any part of siberia is included ( maybe the a tiny portion of the russian far east ). Feel free to take out the mongolian landmass if you want.

> Double check your math.

Did and my idea still holds.

| So in the end, about 50% of the world population lives in about 50% of the world's habitable land.

Your math may be accurate, but your assessment of habitable land-mass seems to be off.

I know and work with a number of people in Canada, and I've visited many, many times. I'd hardly consider the entire area of Canada "uninhabitable". (Per wikipedia [1], Canada has a total (non-water) area of 3,511,023 sq. mi.) 37 million people live Canada [1].

Also, just a quick visual check indicates that about 1/2 of the area within that circle is ocean. And, the population density of the Himalayas (which is in the circle) is arguably small as well.

[1] https://en.wikipedia.org/wiki/Canada

Doesn’t almost all of the population of Canada live along the southern border? My understanding is that the northern ~80% of Canada is very sparsely populated.

True [1]

But, still I wouldn't consider the entire country to be "uninhabitable"

[1] https://www.reddit.com/r/MapPorn/comments/d756ql/population_...

I think its something crazy like 90% of people live within 100 miles of the border

That is correct. Pretty much the all of the major cities are right around that 100 mile mark. On the coasts that’s because of geography (Victoria, Vancouver, Toronto, Ottawa, Montreal, and the Atlantic provinces). In the prairies we generally have one major city close to the border (Calgary, Regina, Winnipeg) and one further north (Edmonton, Saskatoon, none in MB)

You're still assuming ALL of that remaining land area in the circle is habitable.

Eg. India has 43% of it's land classified under "others" [0]. A large chunk of that 43% is not likely to be habitable.

0: https://en.wikipedia.org/wiki/Land_use_statistics_by_country

It's more than a third. Roughly half the circle is water. It also contains all of Tibet, the Hindu Kush and the Southeast Asian rainforest (except New Guinea). Furthermore it contains the highly mountainous islands of Japan, Taiwan, the Phillippines and Indonesia. Proclaiming that all or even most of this land is "habitable" is obviously silly if you just look at it:

https://media.cntraveler.com/photos/593eedf1f2125a2b5b9ed24d...

Any glance at a map of population density will reveal that all of the highest density areas (pink) are within the circle:

https://en.m.wikipedia.org/wiki/Population_density#/media/Fi...

This is definitely an interesting phenomenon and not just clickbait. The most likely explanation is that high rainfall in East Asia has allowed high-yield rice cultivation preceding the Green Revolution, allowing East Asia to support much higher population density in preindustrial times than any other region. Europe is too cold, Africa and West Asia too dry, and the Americas didn't have rice. (Although they did have the potato, but it didn't spread outside the Andes without European assistance.)

There's also the Gobi Desert.

> 2500 miles * 2500 miles * pi = 19,630,000 miles squared

> So that area represents about (19.63 / 57.51 ) 34% of the total world landmass. So 50% of humans live in 34% of the landmass.

I've been telling myself that I have to teach myself Non-Eucledian Geometry, but lacking any real motivation, I've successfully procrastinated for 6 years.

The unfortunate itch to be pedantic on the internet will hopefully drive me to learn how to correctly calculate the area of a circle on a sphere.

Anybody has a good textbook recommendation?

"So that area represents about (19.63 / 57.51 ) 34% of the total world landmass."

1/2 of said circle is ocean, not land. And some of said land is 'uninhabitable' i.e Gobi Desert.

And Canada is 3.8 miles squared, implying that 3.5 of those are uninhabitable is rather off. By that calculation, you'd be excluding most of Quebec and Ontario, which are 'inhabitable' just sparsely populated, in which case much of Tibet and China would be comparably 'uninhabitable' as well.