Small City, Big Metro

[Minneapolitan]

The following are the top 15 cities with the smallest population that have MSAs over 1,000,000 residents.

Rank; City; City Pop; MSA Pop; % MSA Pop living in city

1 Hartford, CT 124,848 1,184,564 10.54

2 Providence, RI 176,365 1,628,808 10.83

3 Richmond, VA 192,494 1,154,317 16.68

4 Orlando, FL 205,648 1,861,707 11.05

5 Rochester, NY 212,481 1,041,499 20.40

6 Birmingham, AL 233,149 1,082,193 21.54

7 Buffalo, NY 282,864 1,154,378 24.50

8 Riverside, CA 288,384 3,793,081 7.60

9 Cincinnati, OH 314,154 2,058,221 15.26

10 Tampa, FL 321,772 2,587,967 12.43

11 Pittsburgh, PA 322,450 2,401,575 13.43

12 St. Louis, MO 343,279 2,764,054 12.42

13 Minneapolis, MN 373,943 3,116,206 12.00

14 Miami, FL 379,724 5,361,723 7.08

15 Atlanta, GA 419,122 4,708,297 8.90

Minneapolis doesn't fit in with list very well being there are two central cities in the Twin Cities metro area. St. Paul literally borders Minneapolis and its CBD is only 10 miles from downtown Minneapolis. The two combine for a central city population of 670,000.

[Minneapolitan]

Well it doesn't surprise me. If we are talking about city propers then Atlanta is only 132.4 sq miles. The city limits are tiny in comparison to Dallas' 342.5 sq miles and Fort Worth's 292.5 square miles.

Just for numbers sake, lets say that Atlanta's municipal boundary was the size of Ft Worth or Dallas. The population numbers would be:

Atlanta @ 292.5 sq miles = 927,870

Atlanta @ 342.5 sq miles = 1,086,480

292.5 and 342.5 sq miles! That's crazy.

Minneapolis and St. Paul combined - 107.7 sq miles = 670,000

Minneapolis-St. Paul @ 252 sq miles = 1,120,787. The total includes all cities within the I-494/I-694 beltway.

[MadVlad]

Well it doesn't surprise me. If we are talking about city propers then Atlanta is only 132.4 sq miles. The city limits are tiny in comparison to Dallas' 342.5 sq miles and Fort Worth's 292.5 square miles.

Heh, only? Hartford is 17 (or 19, depending who you ask) square miles, that's why a lot of lists are crap since it's really apples and oranges...

[hudkina]

If Atlanta were the size of Houston, it woud have 1.89 million in 601.3 sq. mi. However, that's actually tiny compared to most cities. Atlanta would barely rank in the top 20.

1. New York - ~10 million

2. Los Angeles - ~6 million

3. Chicago - ~ 5.5 million

4. San Francisco - 4.3 million

5. Philadelphia - ~4 million

6. Miami - 3.8 million

7. Washington - 3.1 million

8. Detroit - 3.0 million

9. Boston - 2.8 million

10. Dallas - 2.8 million

11. Houston - 2.7 million

12. Phoenix - 2.6 million

13. San Diego - 2.5 million

14. Seattle - 2.3 million

15. Denver - 2.1 million

16. Cleveland - 2.1 million

17. San Bernardino - 2.1 million

18. Minneapolis - 2.0 million

19. Baltimore - 2.0 million

20. Atlanta - 1.9 million

21. Tampa - 1.8 million

22. St. Louis - 1.8 million

23. Portland - 1.6 million

24. Salt Lake City - 1.6 million

25. Sacramento - 1.5 million

26. Pittsburgh - 1.5 million

27. Cincinnati - 1.4 million

28. San Antonio - 1.4 million

29. Norfolk - 1.4 million

30. Orlando - 1.3 million

31. Las Vegas - 1.3 million

32. Kansas City - 1.3 million

33. Providence - 1.3 million

34. Milwaukee - 1.3 million

35. New Orleans - 1.2 million

36. Columbus - 1.2 million

37. Indianapolis - 1.1 million

38. Buffalo - 1.0 million

[Lady Celeste]

If Atlanta were the size of Houston, it woud have 1.89 million in 601.3 sq. mi. However, that's actually tiny compared to most cities. Atlanta would barely rank in the top 20.

1. New York - ~10 million

2. Los Angeles - ~6 million

3. Chicago - ~ 5.5 million

4. San Francisco - 4.3 million

5. Philadelphia - ~4 million

6. Miami - 3.8 million

7. Washington - 3.1 million

8. Detroit - 3.0 million

9. Boston - 2.8 million

10. Dallas - 2.8 million

11. Houston - 2.7 million

12. Phoenix - 2.6 million

13. San Diego - 2.5 million

14. Seattle - 2.3 million

15. Denver - 2.1 million

16. Cleveland - 2.1 million

17. San Bernardino - 2.1 million

18. Minneapolis - 2.0 million

19. Baltimore - 2.0 million

20. Atlanta - 1.9 million

21. Tampa - 1.8 million

22. St. Louis - 1.8 million

23. Portland - 1.6 million

24. Salt Lake City - 1.6 million

25. Sacramento - 1.5 million

26. Pittsburgh - 1.5 million

27. Cincinnati - 1.4 million

28. San Antonio - 1.4 million

29. Norfolk - 1.4 million

30. Orlando - 1.3 million

31. Las Vegas - 1.3 million

32. Kansas City - 1.3 million

33. Providence - 1.3 million

34. Milwaukee - 1.3 million

35. New Orleans - 1.2 million

36. Columbus - 1.2 million

37. Indianapolis - 1.1 million

38. Buffalo - 1.0 million

Technically it would NOT be tiny compared to MOST cities. If it is 20th out of all the cities in the US wouldn't that make it larger than MOST cities. I guess it's all in how you look at it....glass half full, glass half empty. Either way, being a southerner since birth, I would loathe Atlanta every getting too populated. While yes I am on an urban development site, I do not take too kindly to being packed in like sardines. Just my opinion. I would much rather have a nice manageable density, great amenities and low cost of living as opposed to a high population count.

Ohhhhhhh, I remember the days when metro Atlanta was much smaller.

[hudkina]

I meant of comparable metro sizes. Metro Atlanta had 4.2 million people, yet only 1.9 million with a 600 sq. mi. area around downtown. That's low. Tampa's metro population is nearly half that of Atlanta's yet the two have a comparable population within 600 sq. mi. Boston's metro is nearly the same population as Atlanta's yet the 600 sq. mi. core has a million more people.

Atlanta is the only metro (of the top 50) where less than half the population lives within a 600 sq. mi. central core. I haven't actually completed Los Angeles' numbers, but it's 600 sq. mi. population may fall within that category as well with 6 million vs. 12.4 million. But Atlanta is a completely different story than Los Angeles.

[Lady Celeste]

I meant of comparable metro sizes. Metro Atlanta had 4.2 million people, yet only 1.9 million with a 600 sq. mi. area around downtown. That's low. Tampa's metro population is nearly half that of Atlanta's yet the two have a comparable population within 600 sq. mi. Boston's metro is nearly the same population as Atlanta's yet the 600 sq. mi. core has a million more people.

Atlanta is the only metro (of the top 50) where less than half the population lives within a 600 sq. mi. central core. I haven't actually completed Los Angeles' numbers, but it's 600 sq. mi. population may fall within that category as well with 6 million vs. 12.4 million. But Atlanta is a completely different story than Los Angeles.

Ohhhhhhhhhhhhhh, okay. I totally understand what you mean. Thanks for clearing that up Hudkina. Very well explained.

[Aporkalypse]

I meant of comparable metro sizes. Metro Atlanta had 4.2 million people, yet only 1.9 million with a 600 sq. mi. area around downtown. That's low. Tampa's metro population is nearly half that of Atlanta's yet the two have a comparable population within 600 sq. mi. Boston's metro is nearly the same population as Atlanta's yet the 600 sq. mi. core has a million more people.

Atlanta is the only metro (of the top 50) where less than half the population lives within a 600 sq. mi. central core. I haven't actually completed Los Angeles' numbers, but it's 600 sq. mi. population may fall within that category as well with 6 million vs. 12.4 million. But Atlanta is a completely different story than Los Angeles.

Interesting data compilation, I wish there was a website that published that kind of data. It's so interesting to know. I wonder how closely that correlates to commute times. I know Atlanta and LA are the nation's worst metros for that.

[MadVlad]

If Atlanta were the size of Houston, it woud have 1.89 million in 601.3 sq. mi. However, that's actually tiny compared to most cities. Atlanta would barely rank in the top 20.

1. New York - ~10 million

2. Los Angeles - ~6 million

3. Chicago - ~ 5.5 million

4. San Francisco - 4.3 million

5. Philadelphia - ~4 million

6. Miami - 3.8 million

7. Washington - 3.1 million

8. Detroit - 3.0 million

9. Boston - 2.8 million

10. Dallas - 2.8 million

11. Houston - 2.7 million

12. Phoenix - 2.6 million

13. San Diego - 2.5 million

14. Seattle - 2.3 million

15. Denver - 2.1 million

16. Cleveland - 2.1 million

17. San Bernardino - 2.1 million

18. Minneapolis - 2.0 million

19. Baltimore - 2.0 million

20. Atlanta - 1.9 million

21. Tampa - 1.8 million

22. St. Louis - 1.8 million

23. Portland - 1.6 million

24. Salt Lake City - 1.6 million

25. Sacramento - 1.5 million

26. Pittsburgh - 1.5 million

27. Cincinnati - 1.4 million

28. San Antonio - 1.4 million

29. Norfolk - 1.4 million

30. Orlando - 1.3 million

31. Las Vegas - 1.3 million

32. Kansas City - 1.3 million

33. Providence - 1.3 million

34. Milwaukee - 1.3 million

35. New Orleans - 1.2 million

36. Columbus - 1.2 million

37. Indianapolis - 1.1 million

38. Buffalo - 1.0 million

Where did you compile that list from?

[unusualfire]

Where did you compile that list from?

I know Salt Lake City don't even have that much in it's CSA or MSA.

[hudkina]

That's because Salt Lake City is comprised of three metros. Salt Lake, Ogden, and Provo. The three metros have a combined population of 1,788,288. The 600 sq. mi. core has 1,552,673, which is 86.8% of the population of the three metros.

The List was compiled using Census Tracts. Basically I ranked all of the Census Tracts in a metro (and sometimes multiple metros) and whittled them down to an area that was closest to 600 square miles based on density.

The first thing I did was find the population, area, and density of each census tract within the counties that make up the three metros in the Salt Lake region. There are a total of 385, and they have a combined population of 1,788,288 in an area of 16,413.65 sq. mi.

The next step is to determine what the theoretical 600 sq. mi. population is. The reason is I need a base density to apply to a map. I add up the densest census tracts until they reach an area of about 600 sq. mi. In Salt Lake's case, the theoretical 600 sq. mi. city comprises 335 census tracts, of which the lowest tract has a density of just over 420 ppsm.

The next step is to create a map of the census tracts in the region with those with a density of at least 420 ppsm being highlighted.

Finally I start at the bottom of the list and delete the census tracts that don't isolate any highlighted census tracts. When a tract will isolate a highlighted tract if deleted, I add together all of the affected tracts and find a common density. I then rank that density in with the list as a whole.

For example in the picture above there are five highlighted census tracts in Tooele County that are separated from the main cluster by two census tracts, the one that surrounds them and another that is inbetween.

Those census tracts are:

Tooele 1307

Tooele 1308

Tooele 1309

Tooele 1310

Tooele 1311

Tooele 1312

Utah 1131.04

The seven tracts have a combined population of 43,175 in an area of 1,297.81 sq. mi. The combined density is 33.3 ppsm, and is then ranked in the list accordingly.

In the end, using this process, I find the densest combination of contiguous census tracts with an area near 600 sq. mi.

This is a map of the finished product:

Obviously the larger the metro, the harder it is, and I still haven't gotten around to doing the largest metros (New York, Los Angeles, Chicago, Philadelphia, Dallas). Ironically the western cities are much easier to do because there is a lot less exurban and rural sprawl surrounding the cities.

[MadVlad]

That's because Salt Lake City is comprised of three metros. Salt Lake, Ogden, and Provo. The three metros have a combined population of 1,788,288. The 600 sq. mi. core has 1,552,673, which is 86.8% of the population of the three metros.

The List was compiled using Census Tracts. Basically I ranked all of the Census Tracts in a metro (and sometimes multiple metros) and whittled them down to an area that was closest to 600 square miles based on density.

The first thing I did was find the population, area, and density of each census tract within the counties that make up the three metros in the Salt Lake region. There are a total of 385, and they have a combined population of 1,788,288 in an area of 16,413.65 sq. mi.

The next step is to determine what the theoretical 600 sq. mi. population is. The reason is I need a base density to apply to a map. I add up the densest census tracts until they reach an area of about 600 sq. mi. In Salt Lake's case, the theoretical 600 sq. mi. city comprises 335 census tracts, of which the lowest tract has a density of just over 420 ppsm.

The next step is to create a map of the census tracts in the region with those with a density of at least 420 ppsm being highlighted.

Finally I start at the bottom of the list and delete the census tracts that don't isolate any highlighted census tracts. When a tract will isolate a highlighted tract if deleted, I add together all of the affected tracts and find a common density. I then rank that density in with the list as a whole.

For example in the picture above there are five highlighted census tracts in Tooele County that are separated from the main cluster by two census tracts, the one that surrounds them and another that is inbetween.

Those census tracts are:

Tooele 1307

Tooele 1308

Tooele 1309

Tooele 1310

Tooele 1311

Tooele 1312

Utah 1131.04

The seven tracts have a combined population of 43,175 in an area of 1,297.81 sq. mi. The combined density is 33.3 ppsm, and is then ranked in the list accordingly.

In the end, using this process, I find the densest combination of contiguous census tracts with an area near 600 sq. mi.

This is a map of the finished product:

Obviously the larger the metro, the harder it is, and I still haven't gotten around to doing the largest metros (New York, Los Angeles, Chicago, Philadelphia, Dallas). Ironically the western cities are much easier to do because there is a lot less exurban and rural sprawl surrounding the cities.

Just wondering, I was going to try to figure out Hartford on my own, but I'm not sure where I'd get the data to try it...

[ctman987]

In Hartford and Providence it least it is a combination of sprawl and lack of large city borders that make these cities at the top of the list. The Hartford metro has definitly increased in the last ten years and not just population wise but town wise as towns that were thought to be rural have seen rises in population and are now considered metro Hartford with their somewhat easy access to highways (Example: Marlborough, CT Hebron, CT)

Hartford is only about 17 square miles. West Hartford has about another 60,000 people and is known for its highly successfull West Hartford Center but is a seperate town. East Hartford has about another 50,000 people, is home to Rentschler Field and soon 2 billion dollars worth of redevelopment...seperate town. In my view this is the problem but sadly it is to late to fix it b/c these towns dont want their cities problems

[unusualfire]

Well for one it's not even a single urbanized area. In 2000 The Salt lake City urbanized area had 887,650 people.

It seems you are jumping and adding any adjacent urbanized area that would fit into Houston's 601.3 square miles.

[hudkina]

Yes, that's exactly what I'm doing. To be fair to each city, I'm taking the densest census tracts surrounding the central city. Granted, Salt Lake City is a unique situation. Most of the other cities are contained within the main MSA, but the three MSAs in the Salt Valley were close enough that they could have easily been combined. I also did the same with Denver, Cleveland San Francisco, and a few others.

For instance, here is the core for San Francisco. It wraps around the bay and includes Oakland and San Jose.

Here's Miami:

There are other non-linear cities like Detroit, Houston, etc.

[Aporkalypse]

Yes, that's exactly what I'm doing. To be fair to each city, I'm taking the densest census tracts surrounding the central city. Granted, Salt Lake City is a unique situation. Most of the other cities are contained within the main MSA, but the three MSAs in the Salt Valley were close enough that they could have easily been combined. I also did the same with Denver, Cleveland San Francisco, and a few others.

For instance, here is the core for San Francisco. It wraps around the bay and includes Oakland and San Jose.

Here's Miami:

There are other non-linear cities like Detroit, Houston, etc.

Impressive.

Just out of curiosity, what do you do for a living?

[hudkina]

Nothing that has anything to do with this. I know I'm nerdy...

[Lady Celeste]

Hudkina, I am totally impressed. I just love raw data. Can't argue with numbers....well you could but why. You are the coolest. Thanks for the maps and graphs.

[MadVlad]

Can I see the numbers you got for Hartford?

[hudkina]

Hartford, combined with Springfield, Waterbury, and Meriden would have a population of about 1.4 million in an area of 602.94 sq. mi. That would actually rank it ahead of San Antonio and not far behind Pittsburgh.

[EbonyGardens]

Good maps, they do a good job of explaining the metros square mileage.

[Recchia]

Hudkina, is that ArcMap you're using? How long does it take you to do each metro, i.e. can you show us a bunch more maps??

[MadVlad]

Nice, let's pop us on that list big guy!!

[teshadoh]

Hudkina - I regret missing this thread. I've made several other attempts at judging some level of sprawl or urbanity myself - but generally I only complete small areas in the south. Especially in more sprawling cities such as Atlanta, where it is nearly impossible to decipher what the urban core of the city is - very few block groups of over 10k per square mile surrounding. Whereas in vast suburban swaths of NYC, 10k is common place. I have found it difficult to judge each city on the same scale, but also have found it innacurate to use a relative scale.

Here is another idea about 'how large my city would be if it had xxx city limits' - use census block points & select them based on another city's limits, having copied the municipal polygon & snapped it to a center point for the other city. Not very accurate - but could be interesting.

Some time ago I had made a somewhat similar ranking of city populations based on spatial distance. Unfortunately, I had the idea of comparing population growth based on ESRI's pop estimates from 2004 to 2000. A few northeastern forummers, particularly Philadelphia chewed me up for that. Because obviously - less dense city centers will most likely (or should) have a greater amount of population growth than built out city centers. But it was an exercise worth trying I thought.

Anyways - very interesting.

[teshadoh]

This is just a fun goof off thing I did - so don't take it fully face value. But it does indeed indicate how big a difference there is between even Houston & Dallas and Atlanta.

If Houston & Dallas had the same corporate limit size as Atlanta did - I mean the exact city limits, this is what their population would be:

2000 Census

Houston

583668

Dallas

518216

Compared to Atlanta's

Atlanta

416559