A short city population/datanerdfest coming up in this thread, beginning with a little panoramic tour of the Tokyo metro area - city as far as the eye can see
(I made this using Google Earth Studio btw, a very useful tool)
background: I have been using/playing with/analysing global city population data quite a bit over the years, most recently in an academic paper on the topic with @CJHoole and @stehincks
we looked at population density for the whole world, using 1km resolution data from the EU's GHSL project - e.g. here's a little snapshot of how it looks across part of West Africa, from our paper
are there really a zillion people in Tokyo and not a zillion in London? That question can be answered using a simple but powerful tool like NASA's population estimator - e.g. here's what you get of you draw roughly 50km radius circles in Tokyo and London
it can be quite addictive, but also useful - e.g. here's New York's wider metro area (using a boundary that doesn't officially exist, of course)
fair to say that geographers, planners and others have been looking at this kind of thing for zillions of years, of course, but it is interesting and important: e.g. take a look at the UN's 'World's Cities in 2018' report and their 'what is a city' page
here's one table from our paper showing the 'city' populations of different urban areas based on one particular boundary definition - we also added population density figures to the table
there's a much shorter extract of our work at p. 69 in the European Commission's Atlas of the Human Planet 2020
The Stoke-on-Trent Megaregion is 19 million people huge
(it can be quite amusing/enlightening drawing these arbitrary 100km circles on the estimator map, but watch out for varying area calcs)
more seriously, you can actually get boundaries from the GHSL dataset as well (their GHS-SMOD 'urban centre' definition), which is what we used to compare density in the 'cities' below
- I would not be doing my job if I didn't mention the fact that the wonderful team behind the European Commission's GHSL data have released the 'urban centre' boundary file as a GeoPackage that can be used in your GIS of choice
I've added this list of places in the Tokyo area - these are from the GHSL 'urban centres' boundary file - I've added all the places within the 'Tokyo' area, plus satellite background to show the extent of the urban fabric
and finally* - here are three different 'Tokyos', but as you can see, the '23 Special Wards' would be too tight a definition for a true 'urban area' population - for comparison, the 23 Special Wards of Tokyo cover 620 sq km and Greater London covers just under 1,600 sq km
*maybe
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Here's the final version of the global population density graphic I made, because another one that is circulating was less well defined, and someone cropped NZ off it
this is the little 'About this print' sheet that goes along with it
and, should you have a burning question like 'how can I make one of these?' or 'where do I get the data?' then see my blog for all the answers
Possibly of use for UK boffins, data journalists and the like: our (beta) version of a single UK local authority file with insets included (+ all area codes) - join data once, turn insets on/off as required
the bounding boxes are included as well, in a separate file
If you just wanted main map + an inset for London, you could filter layer like this:
"inset_area" IN ('Greater London','None')
you get an inset, you get an inset, everyone gets an inset - except they don't - we may do a version 2 at some point with (e.g. South Wales, Teesside) but for now this is inset in stone
Short map thread, inspired @ScienceAndMaps earlier this week. Many people are familiar with @OrdnanceSurvey grid references (e.g. summit of Ben Nevis is NN 166 712) - those two letters you can see on the map of Great Britain below (2 letters = 100km x 100km - known as a 'myriad')
baffling combination of letters, no? Well, to crack that code you need to zoom out and see the 500km x 500km tiles this is based on (there's no letter I, so we have 25 squares) - image is from @EDINA_Digimap
isn't it nice that the N and the S align roughly with north and south? Ordnance Survey is the mapping agency for GB rather than the whole UK, hence no Northern Ireland on the grid. So, first letter from the 500km part, second letter from the 100km part (then we have 20km squares)
Just had a behind the scenes tour at @NorthendPrint in Sheffield as they finalise our global population print and was blown away by the tech. Amazing stuff. Prints look super sharp too.
I could watch this all day, but probably shouldn't
Anyone needing ready-made UK geodata for local authorities that includes the latest ONS mid-year population estimates (2019): I've shared it here in shp, gpkg and geojson formats (you can also grab place labels, buildings and IMD files while you're at it)