4
$\begingroup$

since there are topological isolation as well as prominence as measures to describe how "dominant" a summit is in comparison to the ones surrounding it, I was thinking about a similar measure to describe the dominance of a city in terms of population.

This would be a nice asset for an automatic map labeling system which labels cities on a map, omitting suburbs of a bigger city right next to them and featuring smaller towns which are the biggest cities in a certain radius, though.

I can think of it as a function which takes the population of a city and the distance to the next more populous city and generates some kind of "dominance score". The higher it is, the more likely it is to be featured on a map.

So are there already any algorithms covering this? Or maybe just "best practices" I can use?

Improve this question
$\endgroup$

1 Answer 1

Reset to default
1
$\begingroup$

Exciting question... it's something I've wondered as well, both for interest purposes and for utility in programs like you mention (I posed a fairly related question in the cartography SE site proposal). I've searched a long time but could not find any term/calculation for this... perhaps I too just did not know the correct term to use?

As a result, for a very basic calculation, I created an Excel spreadsheet and worked with US locations to formulate a simplistic prominence... calculating just the distance each city is from the nearest larger city. The outcome was this Sporcle quiz.

My results are likely not of great use for within a mapping program, as it takes no account of the degree of relative population size... but certainly would be interested in knowing more about that as well, as I've looked for a way to better prioritize the cities shown in a weather mapping solution I use. Not sure you'd be able to use contouring methods given the discretized nature of population. Instead I'd imagine perhaps iterating until you find the radius from the city border centroid where the surrounding population matches the population of the city. That sound like a more complex GIS problem, one I've considered asking before but haven't even been sure how to narrow down into a basic enough question for SE! Still, if there's no luck here of existing formulas, and you work out what way you want to calculate such a prominence (and can then boil it down to a basic question) GIS StackExchange may be the place to try. Hopefully you'll give us the link to any questions you propose there so users here can follow any resolution!

If the basic prominence I calculated is somehow useful, I could dig up the full spreadsheet. I believe I also might have found a potential global dataset as well, but never followed it up.

I cannot exaggerate how exciting it is to see someone else thinking about this topic, and hope your question brings some further better answers from those in the know!

Improve this answer
$\endgroup$
2
  • 1
    $\begingroup$ I took your quiz... didn't do well. There's a lot of Alaska cities! $\endgroup$
    – f.thorpe ♦
    Commented Nov 25, 2017 at 23:51
  • $\begingroup$ @farrenthorpe Likewise myself really. But seems some do like that mix of big and small towns. I guess it's a reminder that when you manage to meet someone in most of Alaska, it's a historic happening. $\endgroup$
    – JeopardyTempest
    Commented Dec 10, 2017 at 14:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.