Population Mountains(pudding.cool)
pudding.cool
Population Mountains
https://pudding.cool/2018/12/3d-cities-story/index.html
23 comments
That has to be some sort of arithmetic error, e.g. all of the 450 people assigned to a single square meter degenerate polygon in the underlying GIS data.
Wouldn't add up as the spikes are taller than in large cities which would have dense multistorey apartment blocks, and also there's quite a few spikes—they continue in a line along a major road, so I'm guessing they're likely the result of some convoluted story like corrupted gps points in csvs being corrected to the nearest boundary of a polygon by some intermediate processing or some similar craziness.
It's scary how populous the island of Java has become. Between Java and northern India, the way the map represents people, it looks like the land is overrun by people. Of course there are other high density areas but much of Africa and the Americas and obvs Siberia look like they have ample open lands without much people.
I hope economic growth and education kick in and allow some of these rapidly growing populations to stabilize. circa 100 million in the Pearl river delta --imagine a drought there.
I hope economic growth and education kick in and allow some of these rapidly growing populations to stabilize. circa 100 million in the Pearl river delta --imagine a drought there.
Or rising water levels from climate change. https://www.nytimes.com/interactive/2017/04/07/world/asia/cl...
The PRD probably isn’t growing all that much from births given the societal preference for not having too many kids. HK has a birth rate of 1.2.
Interactive version: https://pudding.cool/2018/10/city_3d/
Hold Ctrl while clicking/dragging to change 3D z-axis and rotation.
Hold Ctrl while clicking/dragging to change 3D z-axis and rotation.
Makes it very clear what underlies the "this circle contains half the world's population": https://brilliantmaps.com/population-circle/
The visualisation is really nice!
Is there any text or CSV formatted version of the source data? When I went to the GHSL website [1], it says:
Format: the grid data are distributed as raster files in TIF format. The ZIP files contain raster files together with pyramids (i.e., TIF and OVR files).
I made MOBAC [2] profiles for every country in the world, but that's using administrative boundaries (country borders). I'd rather increase the zoom level in built-up areas automatically.
https://github.com/peterburk/mobacProfiles
[1] https://ghslsys.jrc.ec.europa.eu/faq.php
[2] https://mobac.sourceforge.io
Is there any text or CSV formatted version of the source data? When I went to the GHSL website [1], it says:
Format: the grid data are distributed as raster files in TIF format. The ZIP files contain raster files together with pyramids (i.e., TIF and OVR files).
I made MOBAC [2] profiles for every country in the world, but that's using administrative boundaries (country borders). I'd rather increase the zoom level in built-up areas automatically.
https://github.com/peterburk/mobacProfiles
[1] https://ghslsys.jrc.ec.europa.eu/faq.php
[2] https://mobac.sourceforge.io
This is a fun visualization but it seems to me it achieves the opposite of what it hopes to - it hides much of the population from view. Vast regions of the world with low population density appear empty - but they aren't empty, they simply have low population density. In aggregate these regions might have more population than the cities, but this visualization highlights dense regions. An individual pixel with a height of 60 in a city might be surrounded by 800 pixels with a height of 1. This is visually indistinguishable from 800 pixels with a height of 0, eliding an obvious difference in meaning.
Over half the world's population is now urban.
http://www.un.org/en/development/desa/news/population/world-...
Cartgrams represent this fairly well.
United States: https://sites.google.com/site/xiangyilin1309/_/rsrc/14727681...
Australia: https://i.imgur.com/qIojvsU.png
China: https://i0.wp.com/www.p-wood.co/wp-content/uploads/2016/08/C...
World -- population & gdp: http://www.viewsoftheworld.net/wp-content/uploads/2016/05/Th...
http://www.un.org/en/development/desa/news/population/world-...
Cartgrams represent this fairly well.
United States: https://sites.google.com/site/xiangyilin1309/_/rsrc/14727681...
Australia: https://i.imgur.com/qIojvsU.png
China: https://i0.wp.com/www.p-wood.co/wp-content/uploads/2016/08/C...
World -- population & gdp: http://www.viewsoftheworld.net/wp-content/uploads/2016/05/Th...
Not sure I agree with this. The background color is still indicative, for example you can see how much more populated the rural areas in the eastern US are compared to the west, by the shades of white to brown over large areas.
I know a big part of this is getting the population data. But in the rare chance you do get your hands on geospatial data like this, Uber's Kepler tool is an easy way to start analyzing and visualizing the data with great results.
https://kepler.gl/#/demo
https://kepler.gl/#/demo
I'm amazed at how flat Atlanta is compared to other cities. Even having lived there, I didn't realize the population density was that much lower than cities in the Northeast and on the West Coast.
This is great! I'm so excited for the wave of geo-spacial data demos using Mapbox's amazing API and their turf.js computation library that accompanies it.
I'd love to see this animated over time
Dang - this is a really cool viz.
This is addicting <3
But... hrmmm... the small town of Nowogródek Pomorski in north-western Poland (population 450) appears to have the tallest spikes in all of Europe. I wonder what that's about.
Actually, the anomalies (or "noise" as the article calls them) seem some of the most interesting parts of the map—it certainly makes it very easy to readily identify patterns in the GHSL in terms of consistent/predictable -vs- possibly unreliable data.