• Who We Are

    City Parks Blog is a joint effort of the Center for City Park Excellence at the Trust for Public Land and the City Parks Alliance to chronicle the news and issues of the urban park movement. Read more about us.
  • Urban Park Issues

  • Enter your email address to receive notifications of new City Parks Blog posts by email.

  • Archives

  • Urban Green Cover Ad

What is Your City’s ParkScore?

How many people in your city live within walking distance of the nearest park? In what neighborhoods should park improvements or additions be targeted to maximize impact? How well is your city’s park system serving the needs of its residents? Are there disparities between the inner-city core and the lower-density urban fringe, or between different demographic groups?

Today, with the launch of The Trust for Public Land’s ParkScore, it became easier to answer these questions – and more importantly, to begin to develop solutions to park shortages. ParkScore is the most comprehensive park rating system ever developed, combining advanced GIS analysis and data collected by the Center for City Park Excellence.

San Francisco came out on top of the ParkScore rankings, edging out Sacramento, Boston, and New York. Credit: Flickr user Phillie Casablanca.

The Trust for Public Land analyzed the park systems of the nation’s forty most populous cities, and ranked them according to three categories:

Acreage: a city’s acreage score is based equally on two data points – median park size and the percentage of the city’s area covered by parkland.

Access: a city’s access score is based on the percentage of the city’s population that lives within a half-mile walk of the nearest park, taking into consideration the layout of the road network and barriers to access such as railroads, freeways, and fences.

Service & Investment: a city’s service & investment  score is based equally on two data points – total spending per resident and playgrounds per 10,000 residents.

Park access in Dallas, which placed 21st overall in ParkScore. Areas without walkable park access are either red (very high need) or orange (high need), depending on three demographic factors: density, income, and presence of children. Interactive maps for all cities are available at the ParkScore website.

Combined, these factors provide a fair and comprehensive basis for comparison within cities, between cities, and over time. ParkScore is designed to help city residents quantify their need for more and better parks, and for city governments to craft effective and efficient plans to create excellent park systems.

There is a wealth of information in ParkScore that we will delve into in much greater detail in the coming months. For now, here’s an overview of the best urban park systems and those most in need of improvement. Visit the ParkScore website for all the in-depth rankings, maps, and information.

9 Responses

  1. Really?!?! Minneapolis was not even included in the rankings. Also, as of 9:20 AM CST, your links do not work.

  2. Thanks for your comment, Curt. ParkScore included the 40 largest cities by population. While the Minneapolis-St.Paul-Bloomington metro area, as defined by the census, is the 16th most populous in the U.S., the city of Minneapolis is actually the 49th largest by population.

    Other large metro areas whose primary city is not in the 40 largest by population include: Tampa, St. Louis, Pittsburgh, Orlando, Cincinnati, and Cleveland.

    We certainly plan on expanding the scope of ParkScore in the near future, and Minneapolis would be included. Additionally, we have been collecting data on Minneapolis for years in our City Park Facts report, available at tpl.org/cityparkfacts. The 2012 report will be out this summer.

    The Trust for Public Land is heavily invested in the Twin Cities. Check out our Parks for People – Twin Cities projects at: http://www.tpl.org/what-we-do/where-we-work/minnesota/parks-for-people-twin-cities.html

  3. I think the study is somewhat flawed when you are looking at cities with urban/county governments. For instance, Louisville is judged on its massive city/county boundary (400 sq. mi), not the city itself, which has a pretty respectable park system for a city with 300,000 people.

    Other cities on your list that do a similar consolidated gov’t are Indianapolis, Nashville, and Jacksonville just off the top of my head.

  4. City boundaries are arbitrary when it comes to population. For example, Louisville, Columbus and Indianapolis all have artificially high population numbers because their city boundaries include the entire county in which they exist. Meanwhile, cities like Cincinnati, Minneapolis, Atlanta, Pittsburgh and Cleveland are much larger cities than what their official city population would indicate because of their politically constrained boundaries.

    The fact that El Paso (680k), Fort Worth (740k), Fresno (510k), Memphis (1.3 million), Mesa (440k) and Virginia Beach (438k), but Cincinnati (2.2 million), Cleveland (2 million), Minneapolis (3.3 million), Pittsburgh (2.3 million), St. Louis (2.8 million) and Tampa (2.5 million) are not is a bit silly.

  5. Hi, Randy. Thanks for your input. That’s definitely an interesting and valid point. The issue of where a “city” begins and ends is a complicated one with all sorts of implications. I would argue that if the city boundary is too restrictive in some cases (i.e., it doesn’t truly reflect the urban form of places like Cincinnati and Minneapolis), the metropolitan areas that you cite aren’t restrictive enough.

    Take Minneapolis, for which the official metropolitan area (population 3.3 million) extends all the way to St. Croix County, Wisconsin. According to the 2010 census, the population density of St. Croix County is 0.2 people per acre. Oklahoma City, the least-dense city included in ParkScore, is more than seven times as dense, at 1.5 people per acre.

    To rank “city park systems” using areas with less than 1 person per acre would be extremely misleading.

    Unfortunately, there is no consistent, universal way to define a city in such a way that all “urban” areas are included and all “suburban” areas are excluded. We think that the city boundary is most appropriate. Fortunately, we plan on expanding ParkScore, and many of the cities you mentioned will certainly be included.

  6. To clarify a point made by Randy A. Simes. The city boundaries of Columbus DO NOT include the entire county. Suburbs are separate municipalities, as are unincorporated township areas.

  7. Perhaps you could use urban areas as defined by the Census. Urban areas are more tightly defined then metropolitan areas.


  8. This is really interesting. But somewhat flawed. It only seems to be interested in a City’s parks (apparently as defined by the City’s parks program) but leaves out many other open spaces that function as much as parks as the official City ones. I live in a very dense area of Austin with a huge open space (sports fields and running trails) only a few blocks away and used by anyone in the neighborhood, but as it’s owned by the University it doesn’t even show up. I really like the spirit of the study though.

  9. […] Earlier this year, TPL launched its ParkScore project, which ranks the park systems of the nation’s 40 largest cities.  Using computerized mapping technology plus some data from City Park Facts, TPL gave highest rankings (4.5 “Park Benches” out of 5) to San Francisco and Sacramento.  (To see the full rankings, click here: https://cityparksblog.org/2012/05/23/what-is-your-citys-parkscore) […]

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s