Population density patterns in cities (NYC and Chicago as a starter)

[memph]

Originally Posted by CyberEric

Interesting numbers regarding Bronx v Brooklyn.
Is there a list of cities in the US by weighted density?

Chris Bradford made a list for the biggest urban areas. I guess city propers could be interesting too though.
http://www.austincontrarian.com/aust...ed-densit.html

[M II A II R II K]

[Crawford]

Originally Posted by memph

One thing that's interesting is that the Bronx's weighted density is higher than Brooklyn's even though the Bronx's gross density is lower. This means most of the Bronx's residents live on a small part of the land while Brooklyn is more evenly spread out.

The Bronx actually feels far denser than Brooklyn. Most of the Bronx feels more like Manhattan than Brooklyn. Fewer skyscapers than Manhattan, of course, but the street level feel is almost indistinguishable, especially in the West Bronx.

Yes, Brooklyn and the Bronx have about the same density (residents per square mile), but the Bronx has a far lower % of land dedicated to housing (Bronx is 35% parkland and has far more space dedicated to industry and roadways than Brooklyn).

IMO, weighted density is a better comparison than strict density, because you really get to comparative differences in urban form.

[memph]

Originally Posted by memph

Chris Bradford made a list for the biggest urban areas. I guess city propers could be interesting too though.
http://www.austincontrarian.com/aust...ed-densit.html

Here are a few city proper weighted densities (population in brackets).

New York City: 64,025 ppsm (8.008 million)**
San Francisco: 30,005 ppsm (0.805 million)
Montreal: 22,064 ppsm (1.648 million)
Vancouver: 20,397 ppsm (0.618 million)
Toronto: 20,124 ppsm (2.615 million)
Chicago: 19,826 ppsm (2.703 million)*
Mississauga :12,736 ppsm (0.712 million)


*I think I added an extra census tract but it doesn't make much of a difference and would take too long to find
**2000 values for NYC, Canadian cities use 2011 census values and SF/Chicago use 2010 census values

While I haven't looked at Chicago's suburbs yet, I doubt there are many high density census tracts there, so here are the number of Chicagoans living above x density

100,000 ppsm: 1,630
75,000 ppsm: 22,785
50,000 ppsm: 89,369
40,000 ppsm: 130,900
30,000 ppsm: 380,879
25,000 ppsm: 654,242
20,000 ppsm: 1,078,359

In terms of sheer numbers, Chicago is behind Toronto but ahead of Montreal for the >50,000 ppsm densities. For >40,000 ppsm Chicago is behind both Montreal and Toronto, although Montreal has barely more people living at those densities. For >30,000 ppsm Chicago barely edges out Montreal and is close behind Toronto. For >20,000 ppsm and >25,000 ppsm Chicago is ahead of both Toronto and Montreal in terms of sheer numbers.

[Segun]

Got figures for Boston, Philadelphia, Washington DC, Los Angeles?

[memph]

Originally Posted by Segun

Got figures for Boston, Philadelphia, Washington DC, Los Angeles?

No, at least not yet. I did all of SF's urban area (doesn't not include Vallejo, Concord and Pleasanton areas, nor San Jose's MSA) since there is a fair bit of density outside SF proper in places like Oakland, Berkeley and Daly City.

2010 Population: 3,306,927
2010 Weighted Density: 14,740 ppsm

So the weighted density went down a little in the past decade, falling just below Toronto's UA weighted density of 14,853 ppsm and making it #3 in North America.

Population above:

100,000 ppsm: 20,725 (0.63%)
75,000 ppsm: 51,254 (1.55%)
50,000 ppsm: 93,803 (2.84%)
40,000 ppsm: 141,171 (4.27%)
30,000 ppsm: 328,494 (9.93%)
25,000 ppsm: 492,259 (14.89%)
20,000 ppsm: 746,826 (22.58%)
15,000 ppsm: 1,083,584 (32.77%)
10,000 ppsm: 1,742,547 (52.69%)
5,000 ppsm: 2,587,600 (78.25%)

edit: ok calculated DC and Old Toronto (pre-amalgamation of a bunch of suburbs). Old Toronto has 736,775 people and a weighted density of 30,792 ppsm, so it's basically like San Francisco.


New York City: 64,025 ppsm (8.008 million)**
San Francisco: 30,005 ppsm (0.805 million)
Montreal: 22,064 ppsm (1.648 million)
Vancouver: 20,397 ppsm (0.618 million)
Toronto: 20,124 ppsm (2.615 million)
Chicago: 19,826 ppsm (2.703 million)*
DC: 17,459 ppsm (0.602 million)
Mississauga :12,736 ppsm (0.712 million)


*I think I added an extra census tract but it doesn't make much of a difference and would take too long to find
**2000 values for NYC, Canadian cities use 2011 census values and SF, Chicago and DC use 2010 census values

[memph]

Alright, I finished Boston and Philly.

New York City: 64,025 ppsm (8.008 million)**
San Francisco: 30,005 ppsm (0.805 million)
Boston: 24,543 ppsm (0.617 million)
Montreal: 22,064 ppsm (1.648 million)
Vancouver: 20,397 ppsm (0.618 million)
Philadelphia: 20,283 ppsm (1.526 million)
Toronto: 20,124 ppsm (2.615 million)
Chicago: 19,826 ppsm (2.703 million)*
DC: 17,459 ppsm (0.602 million)
Mississauga: 12,736 ppsm (0.712 million)


*I think I added an extra census tract but it doesn't make much of a difference and would take too long to find
**2000 values for NYC, 2010 values for other US cities, Canadian cities use 2011 census values

[lawfin]

I wonder if you could do Chicago lakefront...especially its North lake front from Loop to Rogers Park.

I am surprised the NYC is as high as it is weighted; most of the other cities fall a little above or below roughly double their std density. NYC is 2.5 or so

[CyberEric]

Thanks a ton Memph, this is very interesting stuff.

[dave8721]

I did Miami and it came out pretty unusual compared to all the others. It has much more of a flat density rather than huge peaks and valleys. For example only 32% of the population lived at densities above 15,000 ppsm but 73% lived above 10,000. The weighted density came out to about 12,450 ppsm which is almost exactly the actual density (simple population / area). I guess this means the density curve was rather flat with a ton of people living at around 12,000 ppsm but fewer living in densities higher or lower.

**I would have loved to have seen Miami's numbers in 2012 or even 2011 as opposed to 2010. There was a huge increase in the ultra high densities in just a years time after the census was taken (downtown condo's filled up).

[memph]

Originally Posted by lawfin

I wonder if you could do Chicago lakefront...especially its North lake front from Loop to Rogers Park.

I am surprised the NYC is as high as it is weighted; most of the other cities fall a little above or below roughly double their std density. NYC is 2.5 or so

Can you draw the area you want to include? Mostly so that I know how far West you want me to go.
http://www.citypopulation.de/php/usa-chicago.php

Originally Posted by dave8721

I did Miami and it came out pretty unusual compared to all the others. It has much more of a flat density rather than huge peaks and valleys. For example only 32% of the population lived at densities above 15,000 ppsm but 73% lived above 10,000. The weighted density came out to about 12,450 ppsm which is almost exactly the actual density (simple population / area). I guess this means the density curve was rather flat with a ton of people living at around 12,000 ppsm but fewer living in densities higher or lower.

**I would have loved to have seen Miami's numbers in 2012 or even 2011 as opposed to 2010. There was a huge increase in the ultra high densities in just a years time after the census was taken (downtown condo's filled up).

So that was Miami proper for 2010, not including Miami Beach and what not? And is 12,450 ppsm rounded or exact? Mississauga's weighted density is much higher than the gross/standard density, even though most people live at around the same densities (except a few condos in the central part), however, Mississauga includes a huge amount of land around the airport where no-one lives.

[10023]

The other key difference between New York and Chicago, even where housing types are similar, is that apartments in NYC (particularly Manhattan) tend to be smaller.

[Crawford]

Originally Posted by 10023

The other key difference between New York and Chicago, even where housing types are similar, is that apartments in NYC (particularly Manhattan) tend to be smaller.

And ground coverage is much higher.

No alleys, front lawns, parking podiums, and lots are fully built out.

[dave8721]

Originally Posted by memph

Can you draw the area you want to include? Mostly so that I know how far West you want me to go.
http://www.citypopulation.de/php/usa-chicago.php

So that was Miami proper for 2010, not including Miami Beach and what not? And is 12,450 ppsm rounded or exact? Mississauga's weighted density is much higher than the gross/standard density, even though most people live at around the same densities (except a few condos in the central part), however, Mississauga includes a huge amount of land around the airport where no-one lives.

Yes just City of Miami and yes rounded (does + or - 3 or 4 people per square mile make much of a difference?)

[memph]

Originally Posted by dave8721

Yes just City of Miami and yes rounded (does + or - 3 or 4 people per square mile make much of a difference?)

I have to say it's hard to believe Miami's weighted density would be so close to the gross/overall density. Did your total population add up to 399,000? What was your methodology?

There's basically two ways of calculating the weighted density that I can think of.

1. Take the population of each census tract and divide by the population of the city. Multiply this number by the census tract's density to get the "weight". Do this for all the city's census tracts and add up all the "weights".

2. Take the population of each census tract and multiply by the density to get the "weight". Add up all the weights and divide the sum of all the weights by Miami's total population.

[iheartthed]

Originally Posted by Crawford

The Bronx actually feels far denser than Brooklyn. Most of the Bronx feels more like Manhattan than Brooklyn. Fewer skyscapers than Manhattan, of course, but the street level feel is almost indistinguishable, especially in the West Bronx.

Yes, Brooklyn and the Bronx have about the same density (residents per square mile), but the Bronx has a far lower % of land dedicated to housing (Bronx is 35% parkland and has far more space dedicated to industry and roadways than Brooklyn).

IMO, weighted density is a better comparison than strict density, because you really get to comparative differences in urban form.

There is also a lot of single family housing in the Bronx too. It has the dense areas that developed as psuedo extensions of Manhattan and then you have the post-war single family developments that resemble outer Queens or southern Westchester (I suspect it was an attempt to stem the losses of families to suburbanization). Brooklyn is more evenly developed with multi-unit housing, but relatively few high rises, so the density isn't as high as Manhattan (but still far above any other large American city).

[dave8721]

Originally Posted by memph

I have to say it's hard to believe Miami's weighted density would be so close to the gross/overall density. Did your total population add up to 399,000? What was your methodology?

There's basically two ways of calculating the weighted density that I can think of.

1. Take the population of each census tract and divide by the population of the city. Multiply this number by the census tract's density to get the "weight". Do this for all the city's census tracts and add up all the "weights".

2. Take the population of each census tract and multiply by the density to get the "weight". Add up all the weights and divide the sum of all the weights by Miami's total population.

Using that method (I did #2) I got ~15,900.

The method I had done to get my number before was simply to get the actual density number at which 50% of the population lives at or above (select the tracts above this density, what percentage of the population is this, play around with a few numbers until you get to 50%).

[memph]

Originally Posted by dave8721

Using that method (I did #2) I got ~15,900.

The method I had done to get my number before was simply to get the actual density number at which 50% of the population lives at or above (select the tracts above this density, what percentage of the population is this, play around with a few numbers until you get to 50%).

Ah ok, what you did was mean density, which can also be an interesting statistic but weighted density is an average, not a mean. The mean density would almost always be lower than the weighted density. I also finished Los Angeles, which doesn't appear that dense, but includes a lot of somewhat low density in the West San Fernando Valley, West LA and the area West of Long Beach. The sheer number of people living at high densities is rivalled mostly just by NYC and Toronto with SF coming out ahead only in the 75,000 ppsm+ range.

New York City: 64,025 ppsm (8.008 million)**
San Francisco: 30,005 ppsm (0.805 million)
Boston: 24,543 ppsm (0.617 million)
Montreal: 22,064 ppsm (1.648 million)
Vancouver: 20,397 ppsm (0.618 million)
Philadelphia: 20,283 ppsm (1.526 million)
Toronto: 20,124 ppsm (2.615 million)
Chicago: 19,826 ppsm (2.703 million)*
DC: 17,459 ppsm (0.602 million)
Los Angeles: 16,964 (3.791 million)
Miami: 15,900 ppsm (0.399 million)
Mississauga: 12,736 ppsm (0.712 million)


*I think I added an extra census tract but it doesn't make much of a difference and would take too long to find
**2000 values for NYC, 2010 values for other US cities, Canadian cities use 2011 census values

[CyberEric]

I know this is out of the scope a bit, but would you be able to do London?

[memph]

Originally Posted by CyberEric

I know this is out of the scope a bit, but would you be able to do London?

I don't think so, as far as I know they don't use census tracts. I would need information on something roughly the size of a census tract, boroughs are to big to make a useful comparison.