Land/built-up

Revision as of 01:51, 30 August 2022 by Elie (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Using 'population' data to estimate built-up land

Some inner cities can be quite dense, but even the less-dense suburbs are still fully built-up, as none of the land is truly wilderness (...)( even the parks and nature trails are heavily touched by humans ), and none of it is considered farm land (...)( maybe this could change with suburban farming, but even then, maybe double-counting the same land (as both cropland and built-up land) is ok ).

Let's estimate the minimum population density that might be considered fully built-up land:

suburban_lot_size
5000 ft^2
"Large, 5,000 square foot lots, on the other hand, are the standard in US post-war suburbs."

Do Minimum Lot Size Rules Matter? - Strong Towns www.strongtowns.org › journal › do-minimum-lot-size-rules-matter
housing_fraction
40%
What fraction of the suburban land is housing
The rest would be roads, parks, parking lots, and a small number of commercial buildings.

This is just an educated guess, so if you know of actual data, please tell us in the discussion.
household_size
3 people
Average number of people living in a house, in our low-density suburb
This is intentionally a low estimate because we want to know the minimum population density for fully-built-up land.

household_size * housing_fraction / suburban_lot_size people per km^2 (calculation loading)

So, any land with more people than this, would be considered 100% built-up land in our analysis.

Any land with fewer people, will be counted proportionally. So if the population density is 10% of the threshold, we say the area contains 10% built-up land. This might be the case of a small family farm, where 90% of the lot is farm land, and the last 10% is housing and driveway.

Technically, there could be fully-built-up areas below this population threshold - such as industrial areas - but those are probably uncommon enough.

Let's test out our threshold using the image generator:

pop << data/population.data-float64-8640x4320 # population counts
pop @@ quantity_to_density # convert to 'people per km^2'
pop /= 2583.3385           # threshold for land to be considered 'fully built-up'
pop <= 1
pop @@ density_to_quantity
pop @@ stats

Which gives the result:

Dimensions: 8640 by 4320
Sum: 2708207.986793
Average (Mean): 0.072558
Standard Deviation: (+/-) 0.651528
Minimum: 0.000000 at [0,0]
Maximum: 21.466134 at [3093,2159]

The Sum: 2708207.986793 is the number of km2 of built-up land on Earth.

This is actually quite a lot higher than official estimates. OECD data says there's 784841 km2 of built-up land globallyprobably because their definition is more strict:

"'Built-up' is defined as the presence of buildings (roofed structures). This definition largely excludes other parts of urban environments and the human footprint such as paved surfaces (roads, parking lots), commercial and industrial sites (ports, landfills, quarries, runways) and urban green spaces (parks, gardens). Consequently, such built-up area may be quite different from other urban area data that use alternative definitions."
, and OurWorldInData says there's about 1.5 million km2.

Either way, built-up land is a tiny fraction of Earth's surface510 million km2, according to the same OurWorldInData source, especially compared to farm land43 million km2, according to the same OurWorldInData source.