Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
An interesting exercise is to find your city on here, and then look up a census tract that has that same density. Chances are it'll seem like a pretty "average" neighborhood in your metro as it represents the kind of place the "average" person lives in.
Houston is pretty dense inside the inner loop and is much denser than many cities outside of the first 100 sq. mile. For example:
Inner loop has 96 sq. mile and 30% of it is undeveloped or not livable such as heavy industrial area, ship channel, huge parks (Memorial , Herman & Buffalo bayou park. 469,051 people live in about 67 sq miles = 7k/sq. mile. Boston Suffolk Co. is similar in size with about 121 sq. mile but half of it is water. 757,318 population in 59 sq. mile = 12,836/sq. mile. Boston is quite denser but Houston inner loop is developing fast and catching up. Part of Houston around downtown is starting to look like northern cities. Lots of townhomes, mixed used projects, urban style apartment building and mass transit rail. Our metro area is growing by 27% or 1,620,000 people in next 10 years. Lots of it will be inner loop.
At 2,257,926 in 579 sq. mile, Houston average 3900/sq. mile. For Boston to reach 2,262,324 population it would take the combine population of Suffolk (757,318 & 59 sq. mile) + Middlesex Co (1,505,006 & 823 sq. mile) = 882 total sq. mile and 3101/sq. mile. 900/sq. mi less than Houston.
Any way you mix the combination of the 4 surrounding counties to equal Houston city population: Suffolk (757,318 & 59 sq.mi) + Norfolk Co (666,303 & 400 sq.mi) + Essex Co (742,582 & 501 sq.mi) = 2,166,203 & 960 total sq. mile = 2,256/sq. mile. Still less dense than Houston first 579 square miles.
Going out further, Harris county has 1729 sq. mi & 4,092,459 population = 2,367/ sq. mile. Combining Boston 4 surrounding counties: Suffolk 757,318 & 59 sq.mi + Norfolk Co 666,303 & 400 sq.mi + Middlesex Co 1,505,006 & 823 sq.mi + Essex Co 742,582 & 501 sq.mi = 1783 & 3,671,209 population = 2059/sq. mile. About 15% less dense.
Both area density even out around 5000 square miles. Harris Co 4,092,459 @ 1729 + Fort Bend Co 585,385 @ 875, Montgomery Co 447,718 @ 1044, Brazoria Co 313,166 @ 1386 + Galveston Co 291,309 @ 398 = 5,730,027 & 5,432 sq.mi with 1,055/sq.mi. Keep in mind, most of these surrounding counties have rural area farther away from the city but are added to land area.
Houston urban/suburban population ends around this point. You might add another 300k with another 4000 square miles. Boston on the other hand will add another 1.6 million with providence and its suburbs totaling more than 7.5 million in 10,200 sq. miles. Houston 5 counties will be close to Boston CSA around the next Census in 2020 with half the square miles. Boston area is growing only around 5% so Houston will catch up in density. Houston grew big in the 50's around the time US gov. built all the freeways and residential airconditioner got popular. Our city is more spread out like L.A which grew after the Aquaduct was built around the automobile age. People couldn't live without A/C or water. Boston and many northern city grew back in the horse and buggy days. Good transportation were not available so everything had to be close together like older European cities. Now that the new trend is to develop back into the inner city, Houston is building mass transit rails and pedestrian friendly neighborhood. We are fortunate to be growing at a time of new urbanism. I just want to point out apple to apple comparison with either similar population and total land area. There's a big difference between what is actual and what's on paper.
Actually some interseting stuff, another way to look at this is the area covered by UA (Census defined, not the Int'l crap that goes down to 200 ppsm) and population. On this metric a lot of what you say holds true in that a Houston or Dallas has higher density in their UA than do a Philly or Boston. That being said the UA population is significantly smaller for either Dallas or Houston when compared Philly as an example.
Also UA takes out the counties and combines the populated space not areas in counties hardly populated and far from the core. The Philly UA as an example covers 1,300 sq miles (is weird a smaller footprint than Harris County, Harris county is huge) with a population of 5.4 million as an example. Houston is somewhere around 950 sq miles and 4.3 million (more dense with less population the inverse of comparing the city borders).
But just a couple more points; you discuss the inner loop and industrial space (ALL cities have this and water and this and that and parks etc.) and the loop honestly is at best half as dense as Boston in your comparison. Houston will never achieve that level of density or even close; they just dont build that level of density in Houston. But one thing Houston and Dallas and in the LA/SD model does is build with pretty consistent dense "suburban" (I will take heat on this term on here but there is a boat load of say 3K development and consitent morse than in older metros as you move away from the core) style development thoughout a lot of its developed space. Though this space is roughly hald as dense as the development predecessor LA.
Weighted density is the best way to determine what density people actually live at in a city. It weights each census tract density by the population and so what you end up getting is the average density experienced in the city (what the average resident lives at, if that makes sense).
Weighted density thus controls for vast stretches of unpopulated land and gives a much more accurate representation of how a city is really "lived."
I saw and posted some commentary on this a while back. A lot of this makes sense but to me this does not take into scale; honululu as an extreme example; but overall this does make sense.
As an aside this also excludes more connected areas for a place like Philly where the census draws a line and asigns intro a new metro - this cuts out density at the behest of sprawl and under represents this metric for the Philly MSA; but off that soap box - on the whole this makes a lot meore sense that many metrics I see on here
I was just thinking that to approximate how "big-city" a place is, you could multiply the weighted density by the population (or create an appropriate index with the two). It wouldn't be perfect, but would take into account probably the two most important things we consider when we call a place "big."
NY would be way above everything else. (I always consider it a US outlier and never use it in comparisons of american cities. Nothing ever compares.)
Then LA would be pretty far above the rest.
Then Chicago, Philly, then probably SF, Boston,
Miami, DC, Baltimore
After that, not sure.
But in other words, the sunbelt cities wouldn't even begin to rank in "bigness."
I was just thinking that to approximate how "big-city" a place is, you could multiply the weighted density by the population (or create an appropriate index with the two). It wouldn't be perfect, but would take into account probably the two most important things we consider when we call a place "big."
NY would be way above everything else. (I always consider it a US outlier and never use it in comparisons of american cities. Nothing ever compares.)
Then LA would be pretty far above the rest.
Then Chicago, Philly, then probably SF, Boston,
Miami, DC, Baltimore
After that, not sure.
But in other words, the sunbelt cities wouldn't even begin to rank in "bigness."
My only argument here is that this methodology more gets at the metro than city. LA is hard to catagorize; not as dense or core centric but has more sustained/consitent density than just about any other city (maybe not NYC); an argument could made that on this aspect it would surpass NYC though
I think there a few ways cities can be compared; one is the true core, typically 10-30 sq miles and the UA or this density weighted metric is also an interesting way
You can even employ simple radius metrics, to me they hold weight because even if you say there is water or whatever, the fact of the matter is once you get past the measurement, these people are still that far away
But there is no perfect metric or way to compare cities and how they developed. Comparing boston to LA for example presents many challenges as there are disparate municipality sizes and vastly different development styles
Edit: on your list
I also think you are missing Detroit; it is actually still a very large area and has pretty densly developed burbs even as the city has lost a lot of density
The estimates I have are based on 2009 estimates and a radius sample that counts the people density closest to the core - the 2010 numbers will be a little different but the numbers in the link and radius thread are pretty close, SF is an outlier as the water does cause a lower number but extropolating the SF city density to 100 miles is wildly innaccurate as the inverse
LA in the core 100 miles is actually pretty dense and maintained with continuity
Only NYC and Chicago maintain a 10K density over 200 sq miles (Philly come closest at about 9,400 at 200 miles)
Only NYC/LA/Chicago/Philly/Boston do so to 100 miles in real population; that I can quantify for you. SF would be close if you removed the water but would fall a little short (9,600). No other cities come close including DC at 100 miles and Miami would also miss 10K denisty more significantly utilizing the densest 100 miles. Menaing acyually how many REALLY do live in that space.
Some of the Sunbelt cities make up significant ground extending out to 10 and 25 mile in radius with more continuity in the suburban/multimodal development.
this... only actually useful data ^. other has no real life meaning and cherry picked.
So one of the constant debates here in population is how some cities encompass larger areas so they contain more population but have a much lower density, so they are not truly as big as the numbers imply. So I went through and took the top 60 largest cities in population based on the 2000 Census (will update with 2010 numbers when they are all released) and worked out their densities based on their square miles. Then I put them all at exactly 100 square miles and came out with new figures on what the population would be and where cities would truly rank.
So here they are. This is the new ranking as well as their overall position change in the top 60.
Top 10 Gainers in Position
1. Santa Ana, CA: +51
2. Anaheim: +40
3. Cincinnati: +36
5. Miami: +35
6. Minneapolis: +35
7. Toledo: +34
8. St. Louis: +33
9. Oakland: +32
10. Long Beach: +30
Top 10 Losers in Position
1. Jacksonville: -45
2. Fort Worth -36
3. Indianapolis: -35
4. Phoenix: -35
5. Nashville: -32
6. Houston: -29
7. Charlotte: -29
8. San Antonio: -29
9. Oklahoma City: -28
10. Memphis: -26
Certainly some surprises. Not a single Texas city, for example, moved up. All of them instead moved much further down the list.
I'm confused. If you just took all of the city's densities and plugged them into 100 sqm blocks, doesn't that mean this is just a list of cities by density? What is different about this methodology?
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.
Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.
Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
