Quote:
Originally Posted by airhero
There's no professional analysis, only my own from around a year ago. I'm a little embarrassed to admit that I did this but I took census population estimates between 2000 and 2010 for each zip code in all three cities (Austin, Denver, and SLC), classified each zip code as low density (which is what I referred to as suburban), medium density, and high density based on the population and area of that zip code. I'm not sure what the cutoff was for each classification but it was the same for all three cities. I grouped together zip codes under the same classifications, added them up for each year, and calculated the percentage of total population and the population growth between 2000 and 2010 for each classification. I then did the same thing for 2010 and 2015, with the 2015 estimates coming from, I think, the American Community Survey. It wasn't a perfect analysis but I'd say pretty good for some guy doing it just for fun.
I still have the document with all the numbers for all the zip codes, but it's on a computer I don't have with me right now. I wanted to do it for more cities but ended up forgetting about it until today. It was also quite a bit of work. It was easy for SLC but there are a ton of zip codes in bigger cities.
|
airhero, thanks for sharing your methodology.
Regarding your first issue of the percentage of a city that lives in a "low density suburban" environment, I think your approach of taking the population of a zip code and dividing it by the area of the zip code to calculate residential density is problematic.
There are commercial, industrial, open space and other areas within a zip code that contribute to the land area total (the denominator) that would consequently dilute the residential density calculation and lead to a mis-characterization of the area's true residential density.
For example, say we have two zip codes each totaling 1000 population and each covering 100 acres. Both zip codes would calculate out to 10 people/acre under your approach. But say Zip Code A's 100 acres consists entirely of single-family homes, and Zip Code B's 100 acres consists of 80 acres of commercial/industrial with the residential packed into the remaining 20 acres as mid-rise towers. Zip Code A's residential density of 10 people/acre is legit (1000/100), whereas Zip Code B's true residential density is 50 people/acre (1000/20)--five times that of Zip Code A. Certainly both zip codes should not be categorized as "low-density suburban" density.
However, I appreciate your effort at taking a quantitative stab at a difficult question. There are alternative ways of getting at what you're trying to accomplish. One idea would be to get the zoning GIS layers for each city, categorize the different residential zone districts by density levels and standardizing those across the cities, and calculate the area of each city that is zoned under each density level. That's a fair amount of work though and requires GIS knowledge.
In 2010, Denver rewrote and remapped its entire zoning code, using a "form-based, context-based" approach. As part of that process, every area was categorized as falling into one of six "neighborhood contexts": Downtown Neighborhood, Urban Center Neighborhood, General Urban Neighborhood, Urban Neighborhood, Urban Edge Neighborhood, and Suburban Neighborhood. I would say the last two (Urban Edge and Suburban) would represent "low-density suburban" type development. I've asked a friend at the city to send me the data so I could calculate the area that falls into each category. Of course, SLC or Austin may not have anything similar.
Regarding your other statement that "
a huge majority of the population growth in these cities has happened in these suburban areas", for Denver, this statement is not accurate, in my opinion.
Denver's borders have been fixed since the late 1980s when the land for Denver International Airport was annexed. The city is largely built out except for two areas where large-scale development is occuring: Green Valley Ranch/Gateway (suburban out by the airport) and the redevelopment of the former Stapleton International Airport, which I would consider semi-suburban.
Gateway/Green Valley Ranch has been master planned for around 30,000 residential units. It's maybe one-third built out. Stapleton has been master planned for around 9,000 residential units, and it's about 80% built out. Combining those, a rough estimate is that about 17,000 residential units in these two large suburban or semi-suburban development areas of Denver proper have been completed since 2000.
Additionally, there are many smaller infill sites scattered throughout the established suburban parts of the city that have been developed. I'm not sure of the total number of these units but would guess perhaps another 5,000-10,000 since 2000. So conservatively, I'd say there's been about 27,000 residential units developed in the suburban/semi-suburban parts of Denver since 2000.
In contrast, there were slightly more than 10,000 residential units completed in the greater Downtown Denver area from 2000-2009, and so far this decade (2010-2017), there have been an additional 10,000 residential units completed in the greater Downtown Denver area, with another 8,000 currently under construction. Therefore, since 2000, there's been (or will be soon) at least 28,000 residential units built in the greater Downtown Denver area--all urban in character--which is at least on par with the number of units developed in the more suburban parts of the city. Add in non-Downtown-but-still-urban neighborhoods like Cherry Creek, and I'd say that
a slight majority of new residential development in Denver proper since 2000 has been higher-density urban, not lower-density suburban.
Another way of getting at this question of urban/suburban development in Denver proper is to look at the apartment market. A local real estate analysis firm, James Real Estate Services, does a nice job of tracking apartment development citywide in Denver. According to their 3Q 2017 Apartment Perspective report (
https://www.jres.com/articles/3rd-qu...t-perspective/) there are about 14,000 apartment units currently under construction in Denver proper. Based on the location of each apartment project, I was able to break it down as follows: 8,152 units (58%) Definitely Urban, 2,648 units (19%) Semi-Urban, and 3,140 units (23%) Suburban.
Sorry for the long post, but I love crunching these types of numbers!
