Friday, March 15, 2013

How Much Parkland is Enough?

When the subject turns to public parks, there is a truism in land-use entitlement.  If a proposed development site has been used by the public for casual recreation, someone will argue that the property should instead become a park.

It doesn’t matter if the property has been owned by a family for years and they now hope to retire off the proceeds of selling it or if a developer just put down much of his net worth and a healthy bank loan to acquire the property, someone will propose that the owner give the land to the city for the public good.

It doesn’t matter if the city has no interest in owning the land because the city is already in debt and can’t afford the additional maintenance, someone will claim that the public good requires the land to be given to the city.

I’ve spent a surprisingly large chunk of my professional career listening to Planning Directors, Hearing Officers, and Planning Commission Chairs patiently explain that we live under a system that permits the ownership and fair use of private property and that nobody has the authority to take land without compensation.

In the past week, I’ve heard three different stories of entitlement processes in which someone has proposed that the land instead be given to the city.  Perhaps it will always be thus.

As an urbanist, I have a complex relationship with parks.  I love parks that are loved by the public, are often filled with happy people, and fill a community need.  But I abhor ill-conceived parks that often sit empty and, by their size and location, encourage people to get into cars to drive to destinations beyond the parks.

This leads us to two key questions.  What is the correct amount of parkland for a city to own?  And how do we ensure that the land is configured such that it finds public affection?  I’ll tackle the first question today and defer the second to a later date.

Actually, saying that I’ll “tackle” the first question is a misnomer.  The correct amount of parkland is a question that quickly becomes mired in subjectivity.  Not every acre of parkland is equal.  Adding an acre to a beloved and heavily-used park in the center of town would have far greater value than adding an acre to a regional park that receives fewer and more dispersed users.

Furthermore, there are often sizable, largely undeveloped parks/nature preserves on the edges of city.  Whether those parks are inside or outside of the city limits is often a matter of political circumstances, not civic commitment to parks.

But trying to parse the park data to account for all of these possibilities would require a doctoral thesis in demographics.   Instead, I’ll do the best I can in this shorter format.

New York City and its Central Park are often suggested as the gold standard for city parks.  From the city’s website, New York City has 29,000 acres of park.  (Ironically, Central Park is only the fourth largest park.  The top three spots are held by nature preserve parks.)  With its 2011 population of 8,244,900, New York City parks provide 3.5 acres of park per 1,000 residents.

Looking to a pair of North Bay examples, Napa has 77,867 residents and a park area of 800 acres, which is equal to 10.3 acres of park per 1,000 residents.  Petaluma has 492 acres of park and a population of 58,453, which equates to 8.4 park acres per 1,000 residents.

(A note on the Petaluma calculation: The city webpage includes both Helen Putnam Park and Tolay Lake Regional Park in its park summary, with a note that both lay outside the city limits.  To try to keep the comparison parallel to other cities, I included the Putnam park acreage in the total, but excluded Tolay.  My thinking is that Putnam immediately adjoins the city, functions much like an undeveloped city park, and would likely be inside the city limits except for the California focus on urban growth boundaries.)

And then there is this City Parks Blog webpage that offers a summary of the top twenty large cities as measured by percent of parkland relative to entire city area.  (It’s an odd way to measure parks.  It’s as if the other lands of the city need parks.  Population-based measures seem far more reasonable.  Unfortunately however, the webpage on which the underlying data was provided is no longer active, so developing an alternative list isn’t an option.)

From this data, park acreage per 1,000 residents ranges from 4.6 to 1,794.  The upper end of the range, which is Anchorage, is affected by having a large chunk of a National Forest within the city limits.  The second highest calculation is 128 park acres per 1,000 residents in Jacksonville.

To further illustrate the subjectivity of the data, please note that City Parks Blog gives New York City credit for 10,000 more acres of parkland than New York City claims for itself.

I’m not familiar with all of the cities on this list.  But of the ones that I do know, many are bounded by natural geographic features that likely result in large native preserves on the urban fringes. 

The deeper we dig, the fuzzier the data becomes.  But it seems reasonable to assert that neither Napa nor Petaluma, although not at top of the list for park acres per resident, is particularly deficient in parkland.  Any calls for park dedication in place of entitlement are likely misplaced.

If there’s a perceived need for more parks, it’s more likely the result of park design not meeting community needs.  And that’s a whole different subject, which we’ll discuss another time.

As always, your questions or comments will be appreciated.  Please comment below or email me.  And thanks for reading. - Dave Alden (


  1. When we statistically lump all kinds of parkland together in a single measure, we get these kinds of nonsensical answers. Garbage in, garbage out.

    In reality, there are at least three different kinds of parks.

    Nature preserves are just that--large, and largely unimproved and unprogrammed (except for hiking trails). Passive. Critter habitat, not people habitat (unless you're a hermit). Keep in mind, however, that nature preserves often have neighborhood parks carved into their edges.

    Neighborhood parks are the exact opposite--small, and massively improved. Usually they have a lawn, some play equipment, sitting space, some shade trees, and some aesthetic device or another (sculptures, fountains, etc.) When they're properly placed, they become the neighborhood's center. Such parks only need to be about a square block (on average) in size.

    In between the two is the regional park. Parts of regional parks function as neighborhood parks, but the core of the park is a structured experience of nature (as well as providing space for space-intensive pursuits).

    While squares and neighborhood parks come out of the tradition of the plaza, and nature preserves hunting preserves, regional parks actually come out of the tradition of manor grounds--particularly the English garden style of Capability Brown and his American successor Frederick Law Olmsted. Principal regional parks seem to function best between two and three thousand acres, while secondary ones (where the city can afford them) tend to be one to two thousand acres.

    Because regional parks are thus essentially manor grounds environments set aside for public use, they are expensive to maintain and can only really be supported at between 500k-1 million people in high-density environments. New York is able to maintain one regional park (or collection of parks that amount to one regional park) for each of its counties: Central Park, Prospect Park, Bronx Park, Corona Park/Flushing Meadows/Forest Park/Kissena Park, FDR Boardwalk Park). Boston only has one regional park system: the Emerald Necklace. Philadelphia maintains two major regional parks. Frisco only has one--Golden Gate Park (assuming the Presidio functions as a nature preserve)--and so on.

    The point I'm making is that lumping all parks into "green space" acreage is useless. Many different kinds of parks exist, three with particular importance for city dwellers. And since each are different, each needs to be accounted for differently.

    For neighborhood parks:
    -Does each neighborhood have a quality focal park?
    I.e. one neighborhood park, no more than about 100 acres in size (most less than 50) per neighborhood.

    For nature preserves:
    -Has x kind of habitat been preserved in the region? If not, where can we preserve it?
    Obviously, this means one nature preserve per preexisting biome. More are nice, but generally not necessary.

    For regional parks:
    -Does our (region of the) city have space for more space-intensive uses?
    -Note that nature preserves often fulfill active parkland needs regional parks do.
    -If not, is there suitable space to establish such a park?
    -Regional parks are measured as large park access per capita: that is, one regional park per section of the city of ~500k-1 million citizens.

    1. Steve, thanks for writing. I appreciated the thoughts on the origins of the three scales of park. Your comments seem reasonable, but I'd never before heard it expressed that way.

      The only point on which I disagree with you is whether there is any value in the bulk acres per 1,000 residents. At the upper end, I agree it's meaningless. It only measures when the city limits happen to include nature preserves or national forests.

      But at the lower end, I think there is some meaning. We can safely assume a city with 8 acres per 1,000 residents is better meeting the needs of its residents than one with 2 acres per 1,000 residents.

      It's a crude measure, but fairly easily calculated. As long as we avoid over-reliance, it can tell us something.

      Thanks again for writing.