The Washington, DC region is great >> and it can be greater.


Which Metro stations are physically "walkable"?

Anyone who's seen the area around a variety of Metro stations knows that some are very walkable and some are not. Is there a scientific measure of that? Metro planners crunched the numbers to find out.

Metro rider surveys have shown that most people are willing to walk up to about a half mile to get to a Metro station. Research in other cities also has settled on the half-mile zone.

But the land within a half mile of a station is not the same all across the system. You can't walk in any direction; there are things in the way, whether buildings, rivers, or highways. Where there is a good grid of streets near the station, it's possible to reach a lot by walking up to half a mile. Elsewhere, most of that half-mile radius circle actually requires a longer walk.

Landover, for instance, is right next to a highway. There is only one road leading to the station's parking lot, and no connection over the highway to the nearest residential neighborhood. At Takoma, on the other hand, the street grid lets riders reach many commercial streets and neighborhoods with a half-mile walk.

Metro planners calculated the percentage of land within a half mile you can reach by walking a half mile. It's little surprise that the worst stations are mostly in Fairfax and Prince George's, two jurisdictions that did not try to locate their stations in walkable areas or, during Metro's first few decades, work very hard to plan transit-oriented development around them.

Images from WMATA.

Which stations and jurisdictions fare best and worst?

The worst stations in DC appear to be Fort Totten, a station in the middle of a federal park, and Rhode Island Avenue, a station hemmed in by strip mall development and lacking a good street grid on most sides. (The pedestrian and bicycle bridge over the railroad tracks to the Metropolitan Branch Trail may improve that station's score once it opens.)

In Arlington, it's National Airport (no surprise there; you can't walk on most of an airport) and East Falls Church (but the county has a plan for that area). The Rosslyn-Ballston corridor in Arlington, most of DC (especially in the L'Enfant city) and Montgomery County (particularly inside the Beltway) fare well.

Alexandria is very mixed, with two stations hemmed in by the Beltway and in areas with weaker grids. Prince George's stations are generally more unwalkable than walkable, with College Park the biggest exception. In Fairfax, only Huntington gets anywhere close to a good score. It will be interesting to see how the Tysons stations rank once they open, now and in the future.

The planners also found that the walkability rank correlates very strongly with a station's morning peak ridership. This makes sense, because at the vast majority of stations, even when there is parking there is not that much compared to all the capacity of the trains that pass through. The stations which get a lot of use are those with many people living or working nearby.

There's more to walkability

It's important to note that this is one of several measures of walkability. This analysis computes the size of a station's "walk shed," or how far you can physically get by walking. That is a necessary first step to making a place walkable.

While the Metro planners excluded highways, this analysis still treats roads the same, even though some have no sidewalks, or are multi-lane high-speed roads that are intimidating and unsafe to walk on. But since most of the time good street grids go hand in hand with safer streets to walk on, that shouldn't affect the results much.

More significantly, when people talk about walkable neighborhoods, they are generally thinking beyond just the literal ability to walk. Walkability also includes whether there are amenities such as stores, parks, and more that you can reach by walking. The WalkScore tool computes these in its scores for an area.

Some Metro stations are in places which are physically walkable, but where there isn't much to walk to except for the houses immediately nearby. Glenmont or Forest Glen might be good examples. On the other end of the scale, Prince George's Plaza has a terrible walk shed, but there are lots of stores right near the station.

Regardless, this analysis says something important, and something that's most directly under government planners' control. If jurisdictions want their Metro stations to thrive, a critical first step is making sure people can get to them from the immediate area without having to drive and take up a scarce (and expensive) parking spot.

David Alpert is the founder of Greater Greater Washington and its board president. He worked as a Product Manager for Google for six years and has lived in the Boston, San Francisco, and New York metro areas in addition to Washington, DC. He now lives with his wife and two children in Dupont Circle. 


Add a comment »

I'd love to see a land ownership/zoning analysis (and maybe physical site assessment if that's relevant) for all the parcels within a mile of a metro station. Looks like there is maybe 20 acres of buildable land near Landover metro. Is there a reason it's empty e.g. no market/unmotivated owner/environmental limitations?

by BTA on Jun 17, 2014 1:16 pm • linkreport

Why is the National Cemetery considered green but the airport red? Seems kinda dumb to me to knock the airport because its an airport and you can't roam around on the taxiways and stuff. Airport station doesn't need to be in a walkable area. It needs to be in the airport (which by definition is pretty dang walkable...because its an airport terminal).

by Another Nick on Jun 17, 2014 1:22 pm • linkreport

2. wrt Another Nick's comment, he's right that it's not always good to use the same analytical framework for every place, when some places, like airports, are outliers as it relates to these particular principles. (Although I did once walk from National Airport to Capitol Hill at night, after getting arriving at the airport after the Metro was closed and being unwilling to wait in line.)

by Richard Layman on Jun 17, 2014 1:25 pm • linkreport

Very interesting. Slightly surprising that every station in Montgomery County is 46% or higher (and is the only jurisdiction with that distinction), including Shady Grove and Glenmont.

The results from Prince George's and Fairfax otoh aren't really surprising. The Silver Line stations wouldn't fare any than better Ffx's current stations. In fact even well into the future, the Silver Line's location in the median of the Dulles Toll Road will severely hamper it's "walkability" (according to these metrics), just like it has the Orange Line in the center of I-66.

by King Terrapin on Jun 17, 2014 1:28 pm • linkreport

A pretty good argument for more in-fill stations.

by charlie on Jun 17, 2014 1:29 pm • linkreport

Fort Totten sticks out like a sore thumb. Even without touching the single-family homes, there's so much DC could do to make this neighborhood more vibrant.

by Tom Veil on Jun 17, 2014 1:38 pm • linkreport

Note strongly, what DA pointed out - this is about the percentage of blocks within a half mile distance that are actually a half mile on a legal walking path. It does not reflect the amount of retail activity, say, within that area, or other things impacting walkability. Its a measure of the permeability of the street grid, not urbanism. Thats how Huntington is more "walkable" than Dunn Loring or Vienna. In fact, someone looking to live near a metro and have a walkable lifestyle, is more likely to choose the above orange line stations than Huntington.

As for the new silver line stations, while the existing weak street network will lower their scores on this metric (though that will improve over time) the level of activity in the area will certainly mean it's a more desirable place for someone seeking a WUP than Huntington (or I think most of PG) is.


costs about a a hundred million bucks for an infill station. Now much retrofitting of the street grid can you do for 100 million?

by AWalkerInTheCity on Jun 17, 2014 1:43 pm • linkreport

@AWalkerInTheCity; the point is that on the outer lines the stations are far more than a mile apart. Get you out of the 1/2 mile radius.

by charlie on Jun 17, 2014 1:52 pm • linkreport

East Falls Church surprises me. Expand the definition from 0.5 miles to 1 mile, and I suspect the results would be drastically different. There are lots of single family homes right by the Metro in the Arlington County portion, but once you get into Falls Church City itself, there’s a decent amount of development right along Broad Street (with more planned). The whole area is easily walkable/bikeable with sidewalks, relatively narrow streets, and low speed limits. Broad Street is only a 15 minute walk from the Metro, with plenty of bus service to Falls Church City. When I have some more time, I want to look closer at their methodology for ranking stations.

by Jason on Jun 17, 2014 1:53 pm • linkreport

Why is Dupont Circle only in the 61-75% category? For that matter I'm surprised Woodley Park, Cleveland Park and Van Ness aren't higher as well.

by WestEgg on Jun 17, 2014 1:55 pm • linkreport

Know much retrofitting of the street grid can you do for 100 million?

Less than you'd think, especially if utility relocation is required! That said, it usually opens up land for construction, and you can get developers to pay for it.

by Neil Flanagan on Jun 17, 2014 2:07 pm • linkreport

Is this from a report that is downloadable? Can you share a link?

by Dan on Jun 17, 2014 2:17 pm • linkreport

I'd like to echo above comments that anything that finds Arlington Cemetery has a better 'walkshed' than East Falls Church methodologically flawed.

by Kolohe on Jun 17, 2014 2:42 pm • linkreport

Heck, finding NoMa to be far less 'walkable' than Union Station also indicates a huge methodological flaw.

by Kolohe on Jun 17, 2014 2:45 pm • linkreport

Why is Dupont Circle only in the 61-75% category? For that matter I'm surprised Woodley Park, Cleveland Park and Van Ness aren't higher as well.

Just a guess: inaccessible Rock Creek Park-land. Unless you feel like basejumping off of the Taft & Duke Ellington Bridges (in the case of Woodley Park).

by Dizzy on Jun 17, 2014 2:48 pm • linkreport

I suspect Dupont and other Red Line stations in NW are losing points for connectivity issues around Rock Creek Park that dont seem terribly valid. I wouldn't get too bogged with methodology since this is clearly a blunt instrument but those ranking in the bottom two categories clearly need some work.

by BTA on Jun 17, 2014 2:49 pm • linkreport

NoMa is surrounded by North Capital, Florida, and NY Ave. It's not my favorite place to walk outside of the 5 blocks or so that are nice.

by BTA on Jun 17, 2014 2:52 pm • linkreport

the half mile metric is flawed. Surveys won't capture real life use: I prefer a half mile metro I'll use it at a mile. Picture an apartment building in that landover map just outside the circle along the freeway with a walkway right to the station, everyone would use it. its why meridian hill park is considered as inaccessible by metro as the Gateway costco.
also flawed because:
1) does not take into account walking distance (as the crow flies half mile
2) does not measure from exits - some exits are over a quarter mile from the 'center' of the stop meaning actual walking cutoff is far short of a quarter mile which is ridiculous.

by markus on Jun 17, 2014 3:01 pm • linkreport

It's a good metric to measure streets access to land around the metro station.

It's a terrible metric to measure the actual livability of a place, as it doesnt take into account the land use.

As people have pointed out, Arlington Cemetery isnt exactly a great place for the living to spend the night.

PG plaza probably has more apartments and people living within a 1/2 mile than college park. In fact why college park ranks so highly, when so much of it is cut off by a river, a highway, a CSX line, and an airport I don't know.

This kind of metric, combined with the population and other numbers can be useful, but alone it doesnt provide much insight.

by Richard on Jun 17, 2014 3:21 pm • linkreport

What people are missing is that this establishes a walk shed definition that is far more useful than drawing a half mile radius circle around the station. With that done, you can look at population within the more realistic walk shed, other activity, the quality of the sidewalks, etc.

While they were doing that it was easy to compare the refined walk shed to the crow flies walkshed and get a metric. As Richard says, thats a metric of street access, not of all things impacting walkability or livability.

by AWalkerInTheCity on Jun 17, 2014 3:31 pm • linkreport

+1 Richard "As people have pointed out, Arlington Cemetery isnt exactly a great place for the living to spend the night."

by JDC on Jun 17, 2014 3:36 pm • linkreport

As if to reiterate why walking (and avoiding the roads as much as possible is good):

by JDC on Jun 17, 2014 3:38 pm • linkreport

Even beyond that I'd say it establishes a metric that you could use to argue that some pedestrian improvements could significantly expand the usefulness of certain stations. In the case of Landover, they've cut off over half their walk shed simply by having no west side access to the station. West Falls Church for example is in a similar situation though arguable it's a function of having the yard there.

by BTA on Jun 17, 2014 3:43 pm • linkreport

The metric is what it is, but it creates some weird outcomes. For instance, the Cemetery gets a high score, despite the fact the river and several roads really restrict where you can go.

On the other hand, Van Dorn gets a real crappy score, because there is no way to correct the fact that it is stuck against the Beltway, severely hindering any movement across the Beltway. I guess that's the point, but it just happens to be where the rail tracks were.

by Jasper on Jun 17, 2014 3:51 pm • linkreport

The other problem at Van Dorn is access across the railroad tracks and backlick run, and there is a proposal for a bike/ped bridge specifically to address that.

by AWalkerInTheCity on Jun 17, 2014 3:57 pm • linkreport

Growing up in PG county during the building of the line to New Carrollton I seem to remember the Landover station as planned having a pedestrian bridge over John Hanson highway from the station to the community to the north. The bridge also would have linked the station to drop off area along Parkwoood street. Community opposition or a cut in the budget apparently got the bridge removed from the plan.

by Erich on Jun 17, 2014 5:23 pm • linkreport

Neal Flanagan -- not really, in retrofit situations where they don't have land control, developers expect TIF funding at the very least to pay for "new" road networks.

by Richard Layman on Jun 17, 2014 7:20 pm • linkreport

I've been fascinated by this phenomenon for some time now, that mass transit can, ironically, be so pedestrian-unfriendly at times.

Fort Totten Station was originally planned to be right in the interchange of two freeways - the North Central Freeway along the B&O tracks and the I-95 extension from there to the Beltway.

by Frank IBC on Jun 17, 2014 8:11 pm • linkreport

Its nice to see WFC is actually becoming a better walkable hub than other areas (and redevelopment at the surface parking would really expand that) but I find something to be off with the EFC numbers. Anyone who has used EFC I think would say there are a lot more things around that metro than WFC... is it that most of 7 corners/Falls Church City is just outside of the 1/2 mile? Also, roads around EFC/Westover all have much better bike and pedestrian infrastructure than WFC vicinity.

by Navid Roshan on Jun 17, 2014 8:35 pm • linkreport

Building the Metro lines along railroad lines was penny-wise and dollar-foolish, but luckily Washington's Metro did this a lot less than Baltimore's Metro.

Once again, my turn-back-the-clock, contrary-to-fact fantasy:
-Line straight up Wisconsin Avenue from Georgetown
-Line straight up 7th St. and Georgia Ave. out to Silver Spring, then to Glenmont.
-Line from Union Station, out North Capitol then out either Rhode Island or Michigan Ave. to Greenbelt
-Line from Union Station, out H St. NE to Benning Rd. to Central Ave.

And in Virginia - build the line through Old Town under Washington St.

by Frank IBC on Jun 17, 2014 8:37 pm • linkreport

@ WestEgg

"Why is Dupont Circle only in the 61-75% category? For that matter I'm surprised Woodley Park, Cleveland Park and Van Ness aren't higher as well."

For Dupont Circle, I suspect that having Rock Creek Park on the west side may be the factor.

For Van Ness, the area is dominated by the UDC campus and the embassy complex on the west side of Connecticut, and and then there's another superblock on the east side of Connecticut, with the Howard University School of Law campus immediately behind that. And it's also bounded on the north and south by Soapstone Valley Park and Hazen Park.

Cleveland Park doesn't have any modern superblocks but it is also bounded by several parks and institutions - Hazen Park, Klingle Park, the Zoo, and Washington International School.

by Frank IBC on Jun 17, 2014 8:52 pm • linkreport

I wouldn't mind learning when they're going to build that pedestrian tunnel for Forest Glen as well as the added left turn lane for the Hospital and other long-needed road improvements for safety.
Nobody wants to get killed or see other people getting killed crossing that intersection.

Glenmont suffers from being on the Red Line and getting half the train service of even Silver Spring.

by asffa on Jun 17, 2014 9:36 pm • linkreport

And Woodley Park is also next to Rock Creek Park, the Zoo and two large hotel campuses (Wardman Park and Shoreham).

by Frank IBC on Jun 17, 2014 9:41 pm • linkreport

And in Virginia - build the line through Old Town under Washington St.

Where would you have put the stations? Old Town was already an historic district and Washington St is the spine of it, more than King St is. I can't imagine the fight it would have been to pick which old building is getting knocked down and which has to have a Metro station pit next to it.

Turning to the west would have meant a lot more tunnel too. I'm also suspecting the water table under Old Town might not have been too conducive to tunnels. Putting one of the stations next to the train station was also a pretty good idea.

by Another Nick on Jun 18, 2014 12:26 am • linkreport

You're right about the water table in Old Town, Another Nick.

by selxic on Jun 18, 2014 7:39 am • linkreport

Frank IBC -- the early plans for a subway network in Baltimore City and the metropolitan area were much more extensive and not as tied to rail lines.

The light rail was an add on in the face of dashed subway expansion plans, that specifically "took advantage" of the availability of the determined unneeded Northern Central Railway line, which by that time, mostly had an industrial function, serving areas that didn't have a lot of potential for residential and commercial use.

(I happened to see a map of the NCR track system from the 1870s, and back then it actually went almost as far as Lake Ontario in New York State, and of course into the 1950s it had passenger service, although by then not much further than York.)

by Richard Layman on Jun 18, 2014 12:42 pm • linkreport

Thanks everyone for their close look at the rail map diagram. Your eagle eyes alerted me to a few errors on that diagram that resulted in under-counting the walkable area around several stations, including Dupont Circle and the western end of the Orange Line. I corrected and replaced the map today - new version posted over on PlanItMetro!

Keep in mind measuring the size of the walkable area is only the first step in measuring walkability and Metrorail. Stay tuned for updates on where we're heading next: accounting for un-developable land like parks, improving the walking network, tallying the development in the walk shed, and then looking for a link to walk ridership.

by Justin on Jun 18, 2014 2:59 pm • linkreport

Noma is so high when a large portion of the area can not be walked to easily. All of New York Ave east of Florida Ave is horrible as is every intersection that connects with it bridges over it. It doesnt even have sidewalks on both sides continually and forces pedestrians to go out of their way. Rhode Island Ave is far more walkable

West Falls Church needs to be completely redone. Long ass path to get you out of the stations property takes 10-15 minutes to walk. Needs a connection to Idlywood Rd and needs sidewalks there are places where the sidewalk ends and leads you straight to the Parking lot with no type of path to take. Park Ride and Falls Church Drive need sidewalks

This seems skewed also cause Minnesota Ave is very walkable excluding the Pepco Plant

@ Frank IBC

How did Washington's Metro build less of the system next to railroads compared to Baltimore ? If you add up the distance between New Carrolton & Minnesota Ave, Union Station & Forest Glen, Twinbrook & Shady Grove, Greenbelt & about 1/2 mile after College Park, Potomac Yard to Franconia Springfield, Yellow Line Bridge over Potomac all of that would be about the same distance as the entire Baltimore Metro !

If you add in areas with highways its worst.

The only portion of the metro that was done right was between Ballston & Stadium Armory (thought I think they ould have spaced the stations a bit further apart) Benning Road & Addison Road, Anacostia to Georgia Ave, Union Station to Bethesda

by kk on Jun 18, 2014 10:12 pm • linkreport

Kk -

These are the parts of the Baltimore metro which were NOT built along a railroad (or highway) right-of-way:

--Subway from Mondawmin to Johns Hopkins Hospital

Light rail line from MD Institute of Art to Camden Station

That's quite a bit shorter than the combined length of the Red Line from Union Station to Twinbrook, the Green Line from Naylor Road to Fort Totten, and the Orange Line from Ballston to Stadium-Armory, as I see it.

Another item for my counterfactual wish list - put the Orange line along the W&OD right-of-way between East Falls Church and Dunn Loring - this would be better than along I-66.

by Frank IBC on Jun 19, 2014 3:26 pm • linkreport

@ Frank IBC

Why are you mentioning the Lightrail you said Baltimore Metro not Baltimore Lightrail ?

"Building the Metro lines along railroad lines was penny-wise and dollar-foolish, but luckily Washington's Metro did this a lot less than Baltimore's Metro."

This is what you said

So your statement

"That's quite a bit shorter than the combined length of the Red Line from Union Station to Twinbrook, the Green Line from Naylor Road to Fort Totten, and the Orange Line from Ballston to Stadium-Armory, as I see it."

Does not make sense as they (Washington Metro) did not do it alot less

by kk on Jun 19, 2014 5:47 pm • linkreport

I see that WMATA must have modified their walkable area map. Now the station colors for the Fairfax Orange line stations are shown as 46-75% walkable area on the map on the WMATA site. There are still shown as 20-45% on the GGW map. See the note below the map on the WMATA site.

by Bruce Wright on Jun 22, 2014 3:55 pm • linkreport

Add a Comment

Name: (will be displayed on the comments page)

Email: (must be your real address, but will be kept private)

URL: (optional, will be displayed)

You can use some HTML, like <blockquote>quoting another comment</blockquote>, <i>italics</i>, and <a href="http://url_here">hyperlinks</a>. More here.

Your comment:

By submitting a comment, you agree to abide by our comment policy.
Notify me of followup comments via email. (You can also subscribe without commenting.)
Save my name and email address on this computer so I don't have to enter it next time, and so I don't have to answer the anti-spam map challenge question in the future.


Support Us