Shrinking Cities, going for "less is more" quality over quantity.

[TheArtist]

  http://www.newsdaily.com/stories/tre56j3cr-us-usa-states-cities/


Seems like more and more cities are facing the scenario of large parts of their city, or neighborhoods, emptying out. What I find interesting is that both, cities that are growing and shrinking, are finding that more dense "Urban Village" nodes, often surrounded by green-space, parks, etc. are desirable.

This goes back to my theory of the Ideal City model where there is a central core often having higher density and big civic attractions like arenas, stadiums, museums, central gov. etc., but around that are nodes of Main Streets - Urban Villages. These mini cities/nodes seem to operate best with populations up to around 100-200 thousand. (Think Tulsa back in its heyday when it was the "most beautiful city in America") They can be pedestrian friendly, built around mass transit, very manageable (better crime control), cost effective (less streets, less need of a car, etc.), have a great sense of community, school spirit, etc. and be connected to each other and the central core via mass transit like rail. They can be both urban and suburban, containing a mixture of some of the best qualities of both.

Lot of this is related to good land use and zoning. Whether a growing city or a shrinking one, the pattern seems to be more and more examples of trending back to older models, a more European model, of development.

I have seen examples in Europe where they have green spaces that they have decided are not going to be much developed. Stays mostly countryside and is very expensive if you want to have your little "mansion or villa in the countryside".  Yet there are spots scattered around that start out as a little main street with houses around them. They are set up in such a way that the development can still be fairly quick and inexpensive suburban, you don't lose that economic factor, but will be able to evolve into more dense Urban Village nodes, small pedestrian friendly, transit oriented, cities over time.  Lot of this is simply the mindset, the habits and traditions, which are also codified into their zoning laws, building regulations and such. We just seem to embrace a; devil may care, mish mash, what ever happens happens, approach to things lol.  Sometimes it works out ok, often its a mess.

Even small steps like more half mile streets and a pedestrian friendly "main streets"  with retail up to the sidewalks, seem to offer a good growth model. Cherry Streets and Utica are both half mile streets which allow for traffic to be diverted and filtered out more, so that the main corridors dont have to be widened as much and trips, walking or driving, dont have to be as long. Cherry Street and Brookside were the suburban sprawl of their day, but were set up, perhaps even without thinking or knowing, it was just the way things were done, so that they could evolve more easily into denser, pedestrian friendly, nodes and communities. 

Notice too how Cherry Street, and now Brookside is starting to, had the "main street" then behind that quite a bit of medium density living in the form of small apartment buildings, and then behind that your quiet neighborhoods. The classic form starts in this stepped down, mixed use pattern. Urban mixed use street with living or offices above, to medium density living, stepping down to less dense more suburban neighborhood. You can almost have all your lifestyle choices and stages within half a mile, from an urban apartment on or near a bustling street that may appeal to a young person, down to quiet suburban style, single family homes near a park (Swan Lake).  These areas can be as high a density or low a density as you allow them to be. The potential and ease with which these types of areas can reinvent themselves, can evolve and revitalize over time, seems much greater than areas we tend to create today.

[swake]

While I appreciate the message you are trying to convey, Tulsa is NOT an emptying city. Tulsa was hit pretty hard with the recession at the start of the decade, but did recover and has to this point not been hit hard by the current recession. The city of Tulsa's population fell from 2000 to 2005 but has been growing again since and never fell very much. Tulsa does not have large numbers of vacant housing like is described in the cities in the article you link to. These are cities that were hit very hard, in the early to middle part of the 20th century, and that never really recovered.

Here are the stats on the cities in the article, of them and Tulsa, which one doesn't belong?

Flint, MI, peak population 196,940 (1960), current estimated population, 112,900 (down 43% from peak 49 years ago, down 10% this decade)

Youngstown, OH, peak population 170,002 (1930), current estimated population, 73,818 (down 57% from peak 79 years ago, down 10% this decade)

Cleveland, OH, peak population 914,808 (1950), current estimated population, 438.042 (down 52% from peak 59 years ago, down 8% this decade)

Highland Park, MI, peak population 52959 (1930), current estimated population, 14,709 (down 72% from peak 79 years ago, down 12% this decade)

Philadelphia, PA, peak population 2,071,605 (1950), current estimated population, 1,449,634 (down 30% from peak 59 years ago, down 5% this decade)

Detroit, MI, peak population 1,849,568 (1950), current estimated population, 937,506 (down 51% from peak 59 years ago, down 4% this decade)

Syracuse, NY, peak population 220,583 (1950), current estimated population, 139,079 (down 37% from peak 59 years ago, down 6% this decade)



Tulsa, Ok, peak population 393,049 (2000), current estimated population, 385,635 (down less than 2% from peak, which was at the start of this decade)

[TheArtist]

My apologies. I was not trying to imply that Tulsa was one of the shrinking cities. I was throwing out yet again the notion of clusters of Urban Village type developments.  There is a tendency on many forums and discussions to have the conversation be about urban vrs suburban. I think there is a very nice middle way with the Urban Village/ small city node way of doing things. Also not saying that all development has to be that way, but would like add another form into the debate and have it become a more prevalent presence in our usual development patterns.

I think this could be useful when thinking about parts of our city that have emptied out to some degree like in North Tulsa. How should we redevelop those areas? Would there have been any benefits if the grocery store on Pine and some of the other development near it, had been of the Urban Village type versus the suburban model? How about areas in east Tulsa that do not yet have any development? Should we develop those and how? Or should we leave them as green space and redevelop areas like 21st and Mingo-Garnett as the Main Street or Urban Village model? How about areas that are still developing and infilling near Woodland Hills? Is there a way to encourage some half mile or through streets with pedestrian friendly, mixed use areas? Some way to shift the refill and infill now, so that over time this area evolves even more density and have it be smoothly functioning, mass transit friendly density?   

For even some cities that are growing in population are clearing out some blighted parts of their city while other parts of the city are becoming a little more dense and pedestrian/mass transit friendly. Rather than continuous, homogeneous sprawl, there are more concentrated clusters radiating outward from a central core.

We could more than double our population and take up less than half our land area, and still be only moderately dense. We once had around 6,000 people per square mile. We are now at about 2,000 people per square mile. Los Angeles, the epitome of sprawl is I believe around 7,000 people per square mile.

Once people have an expectation for the way things are or should be, it tends to happen. The rules and habits follow. Again, just wanting to keep throwing one more possibility into the "how we do things" mix.  A possibility that imo we dont think about as much as we should in these parts. As a man thinketh... As a city thinketh...

[Red Arrow]

We could more than double our population and take up less than half our land area, and still be only moderately dense. We once had around 6,000 people per square mile. We are now at about 2,000 people per square mile. Los Angeles, the epitome of sprawl is I believe around 7,000 people per square mile. 

2000 people/sq mi is about 3 people/acre. (I have a better mental picture of people per acre than per sq mi.)  There are certainly areas like that in Tulsa. There are also areas with about 1/4 acre lots and single family housing. If there were 3 people per house that would be 7600 people/sq mi.  I'm reasonably sure that density maps for Tulsa exist.  It would be interesting to see the density of mid-town areas compared to other developed areas of the city. When I fly over the metro area I see mostly 2 types of newly developed areas.  There are bucks-up places with huge houses on small to medium lots and houses for the rest of us on small lots with barely enough (in some cases, not enough) room for a driveway to the back yard. Those developed densities have to be well over 2000 people/sq mi.  The commercial zoned areas along the arterials bring down the average for an arbitrarily defined square mile.  In the more suburban areas, usually a quarter section or less is developed at a time.  Using that square mile with only 1/4 of it developed will give artificially low densities.   The densities of the developed housing areas is what I would like to know.

[TheArtist]

I think we might be suprised at some of the densities in areas around Woodland Hills. And with more and more apartment complexes going in within a mile or two radius of it, that density is still increasing. However, the mall, the parking and such itself leaves a big gap.

But it does show that population density isnt everything, especially when your having a conversation concerned about mass transit, whether our current bus system or any future rail possibilities.  Though we can find pockets of density in the area, perhaps even whole square miles of pretty good density, walking or biking to some place you may want to go may be very difficult. Streets may be windy and within large neighborhoods, developments may not connect so you have to go around, streets are very wide, parking lots which themselves may not connect or can have barriers between them, etc. But still you would be better off in that area than many in Tulsa because there is so much near you and your trips will be shorter and more convenient.  A few tweaks here and there and that area could indeed be much more connected and convenient. I can indeed see it evolving into the most dense and urban area of our city. If it isnt already. Allow for more mixed use development, connect areas better with roads running the length of the front of shopping centers, add sidewalks and trails for biking and walking while removing barriers like berms, hedges, fencing, etc. between shopping areas and residential areas and shopping areas,,,. 

If you look at the "Walk Score" of living in the area, your not bad off. But what its actually calculating is "whats" nearby, not how easy it is in real life to get to it lol. My house near the Promenade gets a respectable 85. The best I could find around Utica Square or Cherry Street was an 87. Though we might categorize those areas as more "walkable" than mine.

http://www.tulsanow.org/forum/index.php?action=post;topic=13914.0;num_replies=3

I think there is also a site where you can find the population within a chosen radius of a spot or within a zipcode.

But really I dont think we need statistics or numbers as much as just taking a look around. Which areas are set up such that a lot of people can; quickly, conveniently, and pleasantly get to the most things. Moderately dense, mixed use, pedestrian friendly areas, with pleasing streetscapes sound wonderful to me. We dont need NYC density. Not always about quantity, but more about quality imo. Though more people in an area means its more likely they can support a wider variety of shops, restaurants, businesses etc. And the more pedestrian friendly an area is, obviously means your closer to stuff. Both of those together can really help the situation if you want to ever have good mass transit.

[cannon_fodder]

I've been looking for a map showing population density by zip code, I have failed.  At least at finding a free one.

Oklahoma GIS project has tons of information, maybe I am just incompetent at accessing it. 
http://geo.ou.edu/MapsFrame.htm

Found a wiki map that is by zip code, but not enough detail to make it work for our purposes. 


And I'm not competent enough to get the Census data I need, though it is the source for the above map.
http://factfinder.census.gov/home/saff/main.html?_lang=en

However, it is safe to say that there are areas of Tulsa that are suitably dense.  And other areas that are not even close.

[SXSW]

One of the denser parts of Tulsa is in Riverview where there are many apartments and apartment buildings and very little parking mixed in with an already dense single family neighborhood.  I think this area, especially the part from Boulder to Boston south of 15th and north of 18th could see even higher densities in the future.

[dsjeffries]

I took ZIP Code data from City-data.com (http://www.city-data.com/zipmaps/Tulsa-Oklahoma.html) and created my own population density maps for the City and County.

It's not as accurate as doing it by census tract, but until I can get my hands on that data and a map to go along with it, it's the best I can offer:

[cannon_fodder]

dsjeffries:

AWESOME.  Thanks for putting in the effort.  Now someone overlay a road map so I can get a good feel for it.

It really helps show our sprawl.  Generally the newer the part of the city, the less dense it is (post 1960's anyway).   Is that because they haven't filled in all the way or just because it is more spread out?  Seems to be that it is because it is more spread out.

Which creates inefficiencies in police and fire,  sewer, water, power, not to mention roads and other common maintenance issues.  Doesn't make that much sense, 74114 is the richest zip code in Oklahoma (statistically, I help drag that average down) but maintains a level of density.    Oh well, lets hope we regain some density, particularly in that 74119 zip.

[tshane250]

I have access to GIS so I created a map for you guys.  It's pretty crudely constructed, but hopefully it paints the picture you all want to see.  I mapped the density down to the Census Block Group level, so it's detailed.  The highways aren't that great and there is some labeling that covers up some stuff, particularly the Jones Riverside Airport label.  Anyway, hope you like it.  (Source: U.S. Bureau of the Census, SF3 2000, Block Group level geography & Center for Spatial Analysis at the University of Oklahoma & ESRI, GIS Online mapping services)

[dsjeffries]

Tshane: Thanks for putting my hours-long project to shame!
Is there a way you could remove the streets and pull density numbers for the tracts?

[tshane250]

Okay, let's see if this works.  I removed the highways/labels and am now showing population density by Census Tract.  Each Census Tract has an ID that corresponds with the ID field in the tables included below the map.  The tables identify the Census Tract name (number), total 2000 population, total square miles, and population/square mile for virtually all the Census Tracts on the map.  I limited the data to just Tulsa County and those tracts in Osage and Wagoner Counties that are within Tulsa corporate limits.  I also included the portions of Broken Arrow in Wagoner County.
 

[dsjeffries]

A million thanks, tshane! That's some very compelling data--I can't wait to pour over all the details!
I've already found my tract (21) and am very pleased that it has a density of more than 8,000/sq. mi!


And, Artist, I just did a little Wikisearch and Los Angeles's density is actually more than we previously thought. It's 8,205/sq mi. Tulsa's density is 2,152/sq mi., which means that Los Angeles, the epitome of sprawl, is 3.81 times more dense than Tulsa.

For more comparison, here are some U.S. cities and their densities:
NYC: 27,440/sq mi
Portland, Oregon: 4,288.38/sq mi
Seattle: 7,179.4/sq mi
Dallas: 3,697.44/sq mi
Houston: 3,828/sq mi
San Francisco: 17,323/sq mi
Atlanta: 4,018/sq mi
Chicago: 12,649/sq mi
Boston: 12,561/sq mi

I think it's rather telling that LA is more dense than Portland, which is hailed by urbanists. I think it is an example of our perceptions not matching reality.

[SXSW]

A million thanks, tshane! That's some very compelling data--I can't wait to pour over all the details!
I've already found my tract (21) and am very pleased that it has a density of more than 8,000/sq. mi!


And, Artist, I just did a little Wikisearch and Los Angeles's density is actually more than we previously thought. It's 8,205/sq mi. Tulsa's density is 2,152/sq mi., which means that Los Angeles, the epitome of sprawl, is 3.81 times more dense than Tulsa.

For more comparison, here are some U.S. cities and their densities:
NYC: 27,440/sq mi
Portland, Oregon: 4,288.38/sq mi
Seattle: 7,179.4/sq mi
Dallas: 3,697.44/sq mi
Houston: 3,828/sq mi
San Francisco: 17,323/sq mi
Atlanta: 4,018/sq mi
Chicago: 12,649/sq mi
Boston: 12,561/sq mi

I think it's rather telling that LA is more dense than Portland, which is hailed by urbanists. I think it is an example of our perceptions not matching reality.

If you've ever been to LA you realize it is a very large and dense city.  There are less dense, suburban sprawl type areas but they are in the far-flung southern and eastern areas.  You need a car to get around most of LA because of its size but most areas within the city are pretty pedestrian-friendly i.e. most retail fronts sidewalks with parallel parking, most buildings have underground parking, lots of mixed-use development with condos/offices above retail, etc.  Basically like having a more dense Brookside or Cherry Street all over the city...

[Red Arrow]

I have access to GIS so I created a map for you guys.  It's pretty crudely constructed, but hopefully it paints the picture you all want to see.  I mapped the density down to the Census Block Group level, so it's detailed.  The highways aren't that great and there is some labeling that covers up some stuff, particularly the Jones Riverside Airport label.  Anyway, hope you like it.  (Source: U.S. Bureau of the Census, SF3 2000, Block Group level geography & Center for Spatial Analysis at the University of Oklahoma & ESRI, GIS Online mapping services)

This (Census Block Group Level) is the type map I had in mind.  On the later map (Census Tract), block 137 at the very north edge of Bixby between 101st and 121st east of Memorial to Mingo is a good example of average density in arbitrary square mile(s).  Look at that same area in Google maps satellite view and you will see everything from a tree farm to almost 6 houses per acre.  Call it 5 houses per acre to allow for roads and at 2 people per house that would be 6400 people/sq mi for that housing development.  Some of that area has been developed since 2000 (if I remember correctly) so that may skew the numbers lower a bit than the present Google map would indicate.