What makes a city walkable?

July 23

New Google feature: pound the pavement

By Seth Budick
For PlanPhilly

Sometimes it seems like every few months a new list appears claiming to rank America’s best walking cities (see two recent examples here and here).  Last Thursday, the newest walkability rankings went up on the Walk Score website and as usual, Philadelphia places near the top with a #5 rating, behind San Francisco, New York, Boston, and Chicago.  Though these types of rankings are remarkably good at attracting media attention, it’s not always clear what they mean for the average pedestrian.

Fortunately, Walk Score is particularly transparent about what its ratings try to capture; the availability of nearby businesses and amenities like parks and schools.  This fits in well with a rapidly expanding body of literature in the fields of planning and public health which is taking a more scientific approach to linking a place’s walkability to quantifiable aspects of its built environment.  As interest in walkability grows, driven in part by rising gas prices and booming downtown populations, Philadelphia is well-positioned both to serve as a model to other cities, and to learn from the lessons of other walkable places.

What makes a walkable place?

So what is it that makes a place walkable?  Or to put it another way, what are the key features of an environment, whether it’s a street, neighborhood, or entire city, that make it conducive to walking?  Common sense is probably a reasonable guide in answering this question, according to Nando Micale, Principal at Wallace Roberts & Todd.  “When it comes to walkability, this is not rocket science,” he said.  But while any pedestrian is intuitively aware that some streets are more walkable than others, there has been a growing interest in providing a more quantitative answer to this apparently simple question. 

In recent years especially, transportation planners and urban designers working in this area have been joined by many in the field of public health who share an interest in the effects of the built environment on physical activity.  As a result, it’s become common to see assessments of the impact of low density, single-use development on walking and other forms of physical activity.  The predominance of suburban sprawl clearly adds to the relevance of this work, a goal of which is to develop a scientific approach to walkability assessment and a set of recommendations for creating more walkable, and thus potentially healthier places.

According to Dan Burden, the Director of Walkable Communities and a consultant on walkability, suburbanization and automobile oriented planning have forced us to relearn many of the basic lessons for creating environments intended for people instead of cars.  As he explained, “There was a time, of course, in history, when there was no such need to assess walkability.  Every neighborhood, every street was walkable.  There was just no other way to design.”

So how can one determine what makes a place walkable; what factors affect whether or not people actually use the street?  It seems obvious that this might depend on a combination of things – everything from the speed of traffic to the presence of a grocery store – making it difficult to imagine how to measure the effects of those individual features.  There are a variety of methods for dealing with this complexity of the real world, but the solution for many researchers working in the area has been to use statistical tools that can tease apart the impact of multiple variables.

In many cases, these factors are initially separated into those that can be measured from maps or databases – things like block length and population density – and those small scale properties of the environment whose measurement requires observers to actually visit a street – such as the number of benches and the condition of the sidewalk.  Researchers then gather data on the length and number of walking trips made by people in the area, often by using surveys, in hopes of finding correlations between walking and some features of the environment.

By now, a growing number of studies have suggested that many of the large-scale, neighborhood-level variables do have a clear impact on the number of walking trips.  For instance, in a review of the recent literature, Reid Ewing & Robert Cervero considered the relative importance of what they call the 3Ds – density, diversity (mixture of land use), and design – on the fraction of trips that people make by car or by alternate modes of transportation.  In evaluating dozens of studies, they found that in areas of more diverse land use and higher density (both population density at the trip origin and employment density at the destination) a much greater proportion of trips were made on foot or by bicycle.

A good example of this type of result comes from a recent study by a group led by Christine Hoehner at St. Louis University.  She and her colleagues looked at the effects on physical activity of everything from the presence of sidewalks to the availability of public transit.  After quantifying a huge range of variables that characterized the street, they found that the number of destinations within walking distance of residents’ homes (for example supermarkets, dry cleaners, etc.) was most strongly correlated with activity (including walking). 

The idea that people will walk more if they have places to go is certainly intuitive, and this metric has been popularized by the Walk Score website.  Walk Score gives any location a point value based on its proximity to retail and amenities like parks and libraries.  To Matt Lerner, Chief Technology Officer of Front Seat, the software company that developed Walk Score, a walkable place is one where a person has “the ability to get by without a car” and this is precisely what the website tries to capture.

The new rankings include maps of America’s 40 most walkable cities, color-coded based on the Walk Score metric.  These scores clearly help to distinguish generally more walkable cities from those that are relatively car-centric; Charlotte, Nashville and Jacksonville round out the bottom of the top 40 for instance.  At the citywide scale though, average scores should probably be interpreted cautiously since they can lump together highly walkable downtowns with more suburban neighborhoods, like some sections of Northeast Philadelphia.

The neighborhood boundaries used by Walk Score come from Zillow, which means that residents of University City, for example, may be surprised to discover that they wake up in Cobbs Creek.  Nevertheless, of Walk Score’s 138 most walkable neighborhoods in America, Philadelphia is home to 7.  Those include Center City East and Center City West, as well as a part of town designated “Riverfront” which extends roughly from Front St. to the Delaware River between Reed and Poplar Streets and which has a population of only 942. 

Other high scoring neighborhoods include Wharton-Hawthorne-Bella Vista, Poplar-Ludlow-Yorktowne, Pennsport-Whitman-Queen, and Fairmount-Spring Garden.   All of these fit Walk Score’s definition of a “Walker’s Paradise” meaning that most everyday tasks can be accomplished on foot and that many people get by without owning a car.

One obvious complication with using only the proximity to retail and amenities is the difficulty in capturing differences in the quality of those destinations.  According to Lerner, the website gets a lot of questions about why it doesn’t differentiate between McDonald’s and any other restaurant in town (it’s treated no differently from Le Bec Fin in Philadelphia’s case).  But quality is in the eye of the beholder of course, and as Lerner put it, “On the one hand, McDonald’s is one of the only places you can spend a dollar or two and sit inside somewhere warm, but on the other hand, it’s not very healthy food.”  Instead, the developers of Walk Score have decided to stay impartial and not make any judgments about the quality of businesses.

This raises the question of whether an automated algorithm will ever be able to capture all of the facets that make a place walkable.  For instance, Nando Micale agreed that Center City Philadelphia has great walkability, in large part because it’s possible to walk downtown and access anything you need within a few blocks.  At the same time, “you still need to have a qualitative and subjective analysis portion of the work.”  While the quality and the depth of the data may someday be sufficient to automate this sort of analysis, he believes that “ultimately, you need to have human judgment.”

That is not to say that efforts at quantifying walkability stop at the level of retail availability.  Another large-scale variable that is frequently examined in analyses of walkability is the form of an area’s street network.  “The overall factor that drives a lot of it [walkability] is the network of streets with sidewalks” according to Micale.  For instance, a grid-like layout, with its large number of four-way intersections, is highly connected, allowing a pedestrian to move efficiently from place to place along a choice of direct routes. 

Further, having many street choices steers traffic off of a small number of arterials, creating a more pedestrian friendly environment according to Frank Jaskiewicz, a Transportation Planner at JzTI Transport Planning.  “Instead of having a few streets that are overburdened with traffic requirements, you have a lot that are all carrying just a little bit of traffic.  You have that flexibility in designing something that’s really truly balanced between traffic circulation and pedestrians.”

Similarly, networks characterized by short blocks are considered more pedestrian friendly according to Reid Ewing, Research Professor at the National Center for Smart Growth.  By this criterion, it’s interesting to note that despite its enormous numbers of pedestrians compared to most US cities, many of Manhattan’s East-West blocks are actually quite pedestrian unfriendly due to their length, in Ewing’s opinion. 

In their review of the recent literature, Ewing and Cervero found some disparity in the results of studies on the effects of the street network.  At least some of those studies, however, suggested that the amount of travel by car is reduced in areas with short blocks and well connected street networks.  According to Ewing, street connectivity is important, but probably a little less important than land use mix and density in shaping travel choices.

At the other end of the spectrum from these large-scale factors, researchers have been exploring the impact of small-scale features of the street environment on the number of people who walk.  These sorts of studies generally require observers to fan out into neighborhoods and score them on everything from sidewalk condition to the presence of street trees, while also surveying residents about their walking. 

Perhaps not surprisingly though, it may be difficult to assess the impact of very local physical features on the number of pedestrians using a street.  For example, one recent paper by a research group led by Mariela Alfonzo at Virginia Polytechnic University found that the amount of walking by residents in several California cities did not depend on variables such as the number of street trees or the presence of sidewalks.

Since the effect of any one of these types of features is likely to be small, many researchers instead combine a variety of related characteristics, such as the presence of graffiti and vacant buildings, into general categories like “safety” in their analyses.  In some cases, this has revealed the impact of a several related features, as in the study by Alfonzo where an index of “safety” did significantly impact the amount of walking. 

This result is consistent with the experience of Paul Levy, President & CEO of the Center City District, who was unambiguous when asked about the most important things that a city can do to encourage pedestrians: “Clean and safe streets are absolutely fundamental.”  This assertion was backed up by Dan Burden who says that a sense of security is the most important determinant of walkability.  “It always has been and will always be.  People will walk where they feel secure and they’ll choose not to walk where they don’t feel secure; it’s the most basic human instinct.” 

Rather than directly assessing the effects of a street’s physical characteristics, a somewhat different approach, according to Ewing, is to think of walkability as deriving from peoples’ perceptions of the street environment.  “Urban designers talk in terms of physical characteristics, but also in terms of these kind of more basic perceptual qualities,” Ewing said. 

These qualities include things like a sense of enclosure – the perception of being in an outdoor room – as well as the feeling that the street is designed at a human scale.  While these perceptions arise from a combination of individual physical properties, it may make sense to interpret peoples’ reactions to the environment in terms of their response to these more general qualities, according to Ewing.

One other key urban design quality, according to this school of thought, is complexity – a place’s visual richness.  Looking at Philadelphia through this lens, it’s easy to see how the city’s streets might qualify as some of the most walkable in America.  As Dan Burden put it, “If I’m in Philadelphia and I go down South Street, there are things that I will see on the 1000th walk down that street that are brand new to me, it’s so complex, rich, vibrant.  It’s not that they weren’t there before, it’s just that there are so many things to take in.”

Philadelphia’s walkability

So how does Philadelphia stack up when it comes to the variety of factors that shape walkability?  Unfortunately, the city has not been the subject of many recent academic studies to specifically focus on these sorts of questions.  But in the opinion of many who know the city well, Philadelphia is clearly a relative pedestrian paradise.  Center City at least “rates as ones of the best American cities, easily, in terms of its walkability,” according to Frank Jaskiewicz.

When it comes to connectivity for instance, much of Philadelphia benefits from a grid of relatively short blocks.  Philadelphia’s Center City blocks tend to average around 400 to 500 feet, giving pedestrians a substantial degree of freedom in their choice of routes between two points.  Dan Burden and Reid Ewing agree that blocks over 500 feet start to push the boundaries of walkability.  As Ewing put it, “Generally shorter is better.  Four hundred feet probably is O.K., 600 feet is getting a little long, 800 feet is too long because you’re requiring people to walk four or five hundred feet to get to an intersection.”

Focusing on block length also highlights the potential disparity in walkability across the city.  Blocks in parts of Northeast and Northwest Philadelphia can be much longer than those in Center City, not to mention that they may lack sidewalks entirely.  As Deborah Schaaf, Senior Planner at the Philadelphia City Planning Commission and Chair of the Bicycle/Pedestrian Advisory Task Force put it, “There are suburban parts of the city.”

Another mark in Philadelphia’s favor, according to Schaaf, is that traffic tends to move slowly on the city’s narrow streets.  “Even though there’s a lot of traffic, the traffic can’t go very fast, so that’s good for pedestrians.”  In addition to the speed of traffic, having a buffer between pedestrians and traffic, such as a parking lane, a landscaped strip, or at minimum even a bike lane, can make a big difference for pedestrian comfort according to Jaskiewicz.  For that reason, he thinks that many of Philadelphia’s most walkable streets are those that have parallel parking on both sides and are therefore outside of Center City.

As a transportation planner Schaaf has an especially interesting take on what makes Philadelphia walkable.  In addition to short blocks and narrow streets, she thinks that the brief signal cycles at crosswalks are integral.  In much of the city, signal cycles don’t exceed 60 seconds meaning that you don’t have to wait more than half a minute to cross the street.  “You don’t have to wait long to cross the street, and when you do cross, it’s a short crossing,” she said.

Similarly, sidewalk width, and the ratio of the width of the sidewalk to that of the street, make a noticeable impact on walkability according to Nando Micale, particularly on wide streets like Broad and Market.  Market Street, for instance, may be five or six lanes wide in some locations, but that doesn’t make it feel unwalkable, he said, because of the generous width of the sidewalks.

As Jaskiewicz sees it, it’s not just street widths that are important, but the design of intersections, since those can be “major conflict points between traffic and pedestrians.”  Intersections are generally engineered to maximize vehicle flow, he explained.  This means adding turning lanes and making intersections wide enough to allow large vehicles to make turns at relatively high speeds.  This negatively impacts pedestrians by lengthening the crossing and “allowing cars to really fly around those curves which makes everything a bit uncomfortable.”

Fortunately, this is one area where Philadelphia benefits from its age.  Center City streets are generally simply too narrow to add turning lanes, according to Jaskiewicz.  This may be a mixed blessing though, since those narrow streets often aren’t wide enough for parallel parking on both sides of the street, eliminating that important pedestrian buffer and increasing traffic speeds.

The inclusion of parallel parking is one reason why, when asked to name his ideal Philadelphia street for walking, Jaskiewicz did not choose Walnut or Chestnut Streets in Center City.  While emphasizing the abundance of great examples, he cited Chester Ave., as it goes through University City, as an extremely walkable street.  Even he admits though that it’s hard to pinpoint what makes it special.  “It has all the main elements; parallel parking, relatively tame traffic; but it also has a lot of amenities; it has a lot of trees, it has nice architecture to look at, it has a nice urban feel to it.“

A good example of a how a street can become more walkable, according to Jaskiewicz, is the change that occurred recently along Walnut Street in University City.  The University of Pennsylvania’s new campus plan aimed to create a strong pedestrian emphasis on that street, Jaskiewicz said.  As a result, one of the two curbside lanes was converted from a peak period travel lane into a parallel parking lane.  In his opinion, this led to a much more comfortable pedestrian experience on both sides of the street at all times of day.

While Philadelphia’s density, mixture of uses, and connectivity clearly are an enormous asset for making it such a walkable place, recently, there haven’t really been any recent comprehensive guidelines for creating pedestrian friendly streets in the city, according to Schaaf.  Meanwhile, many cities, including places like Los Angeles that probably don’t spring to mind when one thinks of pedestrian utopias, are currently generating new sets of guidelines and recommendations for walkable streets.

In Philadelphia, a new Pedestrian and Bicycle Coordinator is being hired, and the Planning Commission is starting a Pedestrian Plan for the city.  According to Deborah Schaaf, the steering committee that’s working on the plan will be considering policy issues for the entire city and addressing physical improvements to sidewalks in about a third to a half of the geographic area of the city.  This will include South Philadelphia, Center City, North Philadelphia and most of Northwest Philadelphia with the emphasis being on connectivity.  Specifically, the committee will be addressing gaps in the bicycle and pedestrian network.

Beyond the basic question of sidewalk availability, the Planning Commission will focus on issues like sidewalk width, especially in some of the more congested parts of the city like Center City and Manayunk.  Schaaf points out that it’s not just sidewalk width that’s important, but the sidewalk’s clear width, or actual walking width, which takes into account factors like steps, cafes, signs, bus shelters, newspaper boxes, parking meters, bike racks, trees, etc.  While there are existing standards, they are not really enforced adequately she said, and whether those standards are adequate needs to be addressed.  The outcome of this study, which might be expected within 18 months, will be a list of specific physical improvement recommendations prioritized for both bicycles and pedestrians including a set of policy and programmatic recommendations.

While making the necessary adjustments to create a more walkable city, Philadelphia also has to follow PennDOT’s traffic engineering guidelines.  Guidelines for important parameters like sidewalk width arise largely from Federal recommendations that are then adopted by the state, according to David Bachmann, Bicycle & Pedestrian Program Manager at PennDOT.  Nevertheless, the state agency tries to accommodate specific local requirements.  For instance, Bachmann says, when it comes to sidewalks, “there are some local ordinances, for example, requiring sidewalks or requiring certain kinds of things being installed, and we usually work with the local folks on that.”

Walkability on the central Delaware waterfront

One obvious canvas on which these lessons for walkability are being applied is the PennPraxis Civic Vision for the Central Delaware.  Indeed, that framework explicitly includes many of these features, notably, the extension of the city’s existing street grid to the waterfront and the creation of new, pedestrian-scale, mixed use development.

In addition to the connectivity benefits of short blocks and four-way intersections, extending Philadelphia’s street grid to the waterfront makes sense for a variety of reasons, according to Nando Micale.  The grid is ideal since it both serves as an appropriate response to the historical and cultural context of Philadelphia and because it can accommodate a wide variety of land uses.  “The plan for the central Delaware is rooted in the idea that Philadelphia, as a place, has this great history of a street network that’s been flexible over 200 to 300 years,” Micale said.

The extension of the grid to the central Delaware, as described in the Civic Vision, would result in an average block size of about 400 by 500 feet.  This represents something close to an average Center City block, according to Micale, and would yield relatively short blocks that are still large enough to support a variety of forms of development.  For instance, it could allow for the insertion of secondary streets, similar to Sansom St. in Center City, that can further subdivide the blocks into units that might be appropriate for a particular scale of development.  For example, “residential land use has a smaller increment of development, as opposed to something that might be a retail and entertainment center,” Micale said.

In order to get to this new grid of streets, of course, pedestrians will first have to deal with the highly unwalkable reality of I-95.  Asked about the problem of getting pedestrians across highways, Reid Ewing was clear: “It’s a huge impediment.”  While the Civic Vision suggests that rebuilding the Center City portion of I-95 has the greatest potential to impact the future of the central Delaware, it is also realistic about the highway being likely to remain a significant barrier in the immediate future.  Some of its suggestions for mitigation therefore include the installation of public art, high quality lighting and landscaping beneath the highway.  It also proposes incorporating recreational areas beneath the elevated structure, along with linear parks that could connect the waterfront with neighborhoods west of the interstate.

Michael Southworth, Professor of City & Regional Planning and Landscape Architecture & Environmental Planning at the University of California, Berkeley, agrees that creating inviting linkages between the waterfront and the rest of the city’s grid is essential.  “I know they’re dreary zones that are extremely difficult.  I have given students a project in Oakland similar to this and they come up with crazy ideas like bowling alleys, disco clubs; activities that make a lot of noise and are compatible.”

In addition to I-95 itself, Delaware Ave. could serve as a formidable barrier to the waterfront, according to Frank Jaskiewicz.  As that street has become more automobile oriented in the last 20 years or so, many of the intersections have been designed to maximize traffic flow, resulting, in many places, in wide intersections with turn lanes that are hostile to pedestrians.  “In coordination with extending the grid, we also have to be very careful about how we design those intersections,” he said. 

Ultimately, connectivity and accessibility are only part of the walkability equation, according to Southworth; people will still need a good reason to cross the highway.  As he put it, “Obviously, you need to have places people want to go or connectivity doesn’t do much of anything.  You not only need connectivity, hopefully with fairly frequent intersections, and choice of pathways, but you need land use patterns and activities that support human life, things that people want to do.”

That “attraction package” as Micale calls it, will hopefully include commercial activity spilling over onto the waterfront from Center City, as well as the potential for whole new neighborhoods further to the North and South.  To fully amenitize those neighborhoods, they’ll need to have dispersed parkland, he said, which should offer a variety of reasons to want to reach the waterfront.  “I think that in different locations on the waterfront, there’s a different mix of uses, parks being one of those uses, that make you want to get there, to go to the edge, to be part of that community.”

And as that community develops, the waterfront may be the most conspicuous opportunity for the city to demonstrate its commitment to maintaining and building on its legacy as a highly walkable place. Though in many places that walkability has been slowly eroding, according to Frank Jaskiewicz, he and others hope that the upcoming changes to the zoning code will mark a return to a focus on pedestrian friendly streets.  As he put it, “hopefully, we’ll draw a line in the sand, that we can’t really eat into our urban, walkable assets anymore.”

Reader feedback:  What streets do you consider to be the most or least walkable in Philadelphia?  Please feel free to let us know about your favorite (or least favorite) streets.

Contact the reporter at sbudick@gmail.com

View Http-planphilly-com-sites-planphilly-com-files-bv_newsletter_fall2008-pdf PDF