Post 331: No magic bullets for Maple Avenue traffic

Posted on July 31, 2019

Are there any radical solutions to congestion on Maple Avenue?  Any magic bullets that appear even remotely plausible?

When it comes to Maple Avenue and traffic, people seem to spend an inordinate amount of time discussing solutions that could not plausibly be implemented.  I thought it might be worthwhile just to do a brief writeup of what I believe is probably not feasible on Maple.  Ever.  And here, I am happy to say that if I’ve gotten something wrong, do not hesitate to email me ( so I can correct it.  It seems remarkably hard to track down even basic facts about what might or might not be feasible on Maple.

The gist of this is, there’s really not a whole lot you can do, realistically, to create a vast improvement in Maple Avenue traffic congestion.  If there were, somebody would have implemented it already.

If you already know or have guessed that, you have no need to read this posting.  In this posting, I explain why you’ll probably never see these solutions:

  • Widen the road (feasible, but ugly)
  • Replace lights with traffic circles (not enough room, could not handle peak Maple Avenue volume efficiently)
  • Reversible center lane (dangerous, plausibly only helps with AM rush hour, but not destructive — Maple would look more-or-less the same).
  • Bypass (destructive, requires W+OD crossing, unlikely to carry enough traffic to make a material difference).


Let’s start with what I believe the facts are.

Control of most aspects of Maple Avenue belongs to the Virginia Department of Transportation (VDOT), not to the Town of Vienna.  Literally, VDOT owns the road.  We may call it Maple Avenue, but legally, it’s a section of primary state highway 123.  VDOT allows Vienna to do some things, within reason.  For example, we’re one of a handful of Virginia Towns that maintains its own streets.  And, by law, the Commonwealth pays us to do that.

Significant change in that road is done with the permission of VDOT.  By law, lane markings and stop lights are under VDOT authority.  I.e., the Town could not add or remove a stop light without VDOT’s approval.  I believe that the Town, does, in fact, set the timing of the stoplights.  (I’m not sure whether that requires VDOT cooperation/approval or not.)

Oddly, I think Vienna may control the speed limit on Maple Avenue, at or below its designed limit of (I believe) 35 MPH.  I think this for two reasons.  First, VDOT’s map of speeds on highways that they control specifically omits Vienna.  Second, Commonwealth statute say that if a municipality maintains its roads, then it can set its own speed limits.  (Separately, Fairfax County is one of two Virginia counties that has authority over its own speed limits.)  But, presumably, if we tried to set it to something unreasonable (e.g., 20 MPH), VDOT would step in a prevent that.  You can’t have a single town on a primary highway interfering with the flow of traffic for what is, after all, a regional highway.

Widen the road:  Expensive, destructive, and ugly, but not impossible.

Put aside the aesthetics of this for the time being, and just focus on what it would take to get this done. Can we squeeze any more lanes out of the existing roadway, and if not, what would have to be destroyed to add two more lanes to Maple?

First, how wide do traffic lanes need to be for a low-speed urban arterial highway?  Oddly, there is no agreed-upon standard.  For interstate highways, the standard travel lane is 12′ wide.  For urban roads, anywhere from 9′ to 13′ can be deemed appropriate for a travel (as opposed to a turn) lane.  But conventional wisdom leans toward wider lanes, as described in this article.  Narrow lanes are typically only chosen when there is a traffic-calming goal in mind, or where it is literally not possible to fit wider lanes.

There is a movement suggesting that narrower lanes are both feasible and safer than wider lanes.  That is, a 10′ standard maybe safe and effective.  (Apparently, in some older parts of DC, there are lane widths as narrow as 8′. (See this article for discussion, or this one).   I see little discussion to suggest that anyone would build new urban arterial roads with lanes appreciably narrower than 10′.

In which case, the lanes on Maple are already as narrow as they can be.  For Maple Avenue, in central business district of the Town of Vienna, the asphalt is a fairly uniform 48′ across, with a nine-foot-wide center turn lane, and four travel lanes averaging just under 10′ wide.  (Picture and measurements courtesy of Google Earth.)

Once you get outside of the business district — past the country club, say — lane widths expand to 12′.  The asphalt in one direction is about 24′ wide.

The upshot of all that is that, based on current traffic engineering conventions, you’d be hard-pressed to make the travel lanes on Maple Avenue any narrower in the Vienna business district.   They are already a shade under 10′, which is at the narrow end of what appears to be acceptable.   (I managed to find a couple of examples of short stretches of 9′ lanes in old town Fairfax, but presumably those were squeezed into accommodate the existing old buildings.)  Absent some radical traffic-calming goal, it would be hard to make the case for narrower lanes.

Since we can’t make the lanes any narrower, adding two more lanes to Maple Avenue would require destroying everything within 17′ of the existing asphalt.   That’s 10′ for a new lane, and 7′ for the standard utility strip and concrete ribbon sidewalk.

I believe that Maple was widened to current size around 1961 or so.  Since that time, business have been built around that configuration.  That 17′ path of destruction would mess up a lot of Maple.

On the south side of the street, near as I can tell, the only building that would have to come down is the Historic Mattress District building. I.e., Panera and the row of older shops attached to it.  (The yellow line in the picture below is the measured distance of 17′.)  Otherwise, the Michel Rene building would lose most of its parking, and elsewhere, almost all the street-side parking lot landscaping would be removed.  You’d probably need low brick walls or equivalent to separate the parking lots from the (then) immediately adjacent sidewalk.

On the north side of the street, several of the older buildings would barely scrape by.  E.g., the doors of the Benjamin Moore paint store, Maple Avenue Restaurant, Virginia Tire, Vienna Inn, Pear Tree Cottage would all open directly into the new sidewalk.  But the entire block of old buildings from Vienna Jewelers, Orange Theory Fitness, Potomac River Runners, and the former Starbucks would all have to go.  And, as with the south side, all street-side parking lot landscaping would be destroyed, along with the occasional parking place.

Interestingly, if you were willing to give up the center turn lane for that half-block of Maple (From Panera to Center Street), it looks like you could avoid tearing down any buildings except the former Starbucks.  These buildings fall into the 17′ zone of destruction, but just barely.  With another 4′ or 5′ obtained by eliminating the center turn lane, it looks like you might scrape by.  But the former Starbucks would have to go, because you would still need a left turn lane at that stoplight.

In short, adding two lanes to Maple is not impossible if you are willing to condemn one old building (the former Starbucks), give up the center turn lane for half a block of Maple (to avoid condemning several other old buildings), and lose parking lot landscaping and some parking up and down the length of Maple.  A number of north-side-of-Maple business would have their doors open directly into the sidewalk, as is the case with the former Starbucks now.   And you’d have to put up with a considerably uglier streetscape (no room for street-side parking lot landscaping), as well as move all the utility lines and light poles and such.

Finally, I conclude that the new MAC buildings would not prevent such a widening.  Instead of a 20′ setback from the curb, they’d end up with a 10′ setback from the curb, a standard utility strip and sidewalk, and an extra 3′.  Barely enough to open the shop doors without obstructing the sidewalk.  So it might look awful, but it would not be infeasible.

In short, if you were willing to settle for the look of it, it appears to me that you could add two lanes.  Arguably, by eliminating the center turn lane on one block of Maple, you would only have to destroy only one building (the former Starbucks) do to it.  There would be some loss of store-front parking, certainly at the Michel Rene building, and likely sporadically elsewhere.   Virtually all street-side parking lot landscaping would be gone.  And the resulting road would more-or-less cut the Town in half, as 123 does as it traverses Tysons.

My opinion:  I don’t think the Town would be willing to give up that much of itself for the sake of (possibly) reducing Maple Avenue congestion.  But it’s not flatly impossible.

Roundabouts instead of traffic lights on Maple:  No.

The mini-roundabout at Park and Locust has been, by most accounts, a huge improvement over the previous version of that intersection.  This has led some people to ask whether Maple would work better with roundabouts instead of traffic lights.

I’m fairly sure this one is a “no” for two clear reasons.

One, we don’t have the room to do it.  I looked up minimum dimensions for a low-speed two-lane roundabout and the lowest figure I found was 150′ diameter (from this Federal guide .pdf.).  Here’s what a 75′-radius circle that would look like at Maple and Center, courtesy of CalcMaps.  You (barely) would not have to knock down any buildings, but (e.g.) there’s no longer room for sidewalks between the buildings and the roadway.  This does not even include the widening of the road that might have to occur outside that circle in order to feed roads into the roundabout.  By eye, it looks impossible to fit that into one of our biggest intersections and still have (e.g.) sidewalks.

Two, they would be completely inappropriate/would not work for the very high traffic flows that Maple experiences during rush hour.  For one thing, pedestrians would find it nearly impossible to cross Maple at rush hour due to the continual stream of cars.  But more importantly, Maple peak traffic exceeds what a two-lane 150′ wide roundabout can handle well.  Peak rush hour traffic on Maple is about 2500 vehicles per hour (calculated from “K factors” and total weekday traffic counts in this VDOT publication of Vienna traffic, .pdf.)  Per this Federal guide (pdf.), the absolute maximum capacity of a two-lane urban roundabout doesn’t even go that high, and operation of roundabouts above 85 percent of capacity typically leads to significant delays.  Our peak rush hour volume is literally off the chart for a two-lane urban roundabout 150′ in diameter.

That said, it ain’t like the existing system of stoplights leads to smooth traffic flow during rush hour either.  But, for sure, peak rush-hour flow on Maple, nearly all of which is in one direction, appears to exceeds the design capacity of the two-lane roundabout that we might, barely, be able to squeeze into the existing intersections.

When all is said and done, seems like a non-starter.  Arguably, this is why you don’t see roundabouts used around here in similar situations.  And, given how iffy that is, who in a position of authority would take a gamble on that working?  Nobody is going to start condemning property and displacing utility poles on the off chance that building roundabouts would help Maple Avenue congestion.

Reversible center lane:  Iffy, dangerous, and unhelpful for anything other than rush hour.  But less destructive than alternatives.

Under this proposal, the curb lanes west-bound and east-bound would stay the same, but the other three lanes would change function throughout the day, Monday to Friday, to accommodate rush hour traffic flow.

The center lane of the roadway would become a through lane going inbound toward Tyson’s in the morning, and outbound toward Oakton in the afternoon.   During morning rush, the non-curb west-bound lane would become the left turn lane.  During evening rush, the non-curb east-bound lane would become the left turn lane.

This would require reprogramming all the traffic lights on Maple, modifying some traffic lights, adding lane control signals, re-striping the road, and so on.  This would also require finding a way to bring three lanes of traffic back to two lanes, at either Town of Vienna boundary, in a way that itself did not lead to a traffic jam.

We have non-interstate reversible lanes in a few places in the D.C. area.  (We also have reversible roads, where entire roads become one-way, as with Rock Creek Parkway, but that’s a different issue).  So it’s not as if this is a totally foreign concept.

Reversible lanes can be a significant safety concern, as described in this article about Connecticut Avenue in DC.  The safety concerns would not merely be due to confusion about what direction traffic is flowing in a particular lane.

In addition, in the Town of Vienna, it would mean that left turns would have to cross three lanes of traffic, instead of two as currently.  But this would only be true for then small number of vehicles traveling against rush-hour traffic who wanted to turn left across Maple.

I will also note that, by eye, the AM rush hour really does appear uni-directional (the flow of vehicles west-bound in the AM is slight), but the PM rush hour is not nearly so one-sided.  In my experience, there are plenty of cars trying to travel west-bound (toward Tysons) during the PM rush hour.  So this solution might be useful only for the AM rush, and could not be implemented during the PM rush hour due to the relatively high volume of west-bound traffic at that time.

In its favor, this one requires the least amount of destruction of the existing Maple Avenue.  Maple would look pretty much the same as it does now (plus or minus some of the medians).  But it only would have an impact on rush-hour traffic (e.g., not mid-day or weekend traffic).  And it would require some study to determine how helpful it would be, e.g., would the “choke point” where the road reverts to two lanes end up negating any advantages of three lanes through the middle of town.

My conclusion is that this might be a feasible way to improve traffic flow during AM rush hour.  I don’t think it would help at any other time of day.

Maple Avenue bypass:  A non-starter.

There are at least two routes that could serve as a bypass for Maple Avenue.  Both would require cutting across the W&OD trail and/or putting in a bridge for the W&OD to cross the bypass.  Both would utterly destroy the peace and quiet of the affected neighborhoods.

Return for a moment to this prior post, to understand why the W&OD is one of the main causes of our traffic woes.  The W&OD was here before the road network was developed.  There is a nearly five mile stretch where nothing crosses the W&OD except for Maple Avenue.  The result is that anyone who wants to get from one side of that line to the other ends up on Maple Avenue.

A Town of Vienna bypass would have to breach that line.  After looking at maps for some time, the only route I can see would be to connect Tapawingo and Follin.  This would provide a bypass route for traffic moving between area south of Vienna (e.g., I-66) and Tysons.  (Alternatively, Tapawingo/Electric/Woodford/Old Courthouse would allow traffic to exit further west on Route 123.)

This would require cutting across the W&OD to connect these two streets.

Although much of Tapawingo and Follin are already high-volume through streets, this would certainly burden those neighborhoods further.  Based on 2017 VDOT counts (.pdf), Tapawingo carries about 4000 cars per day where it crosses Cottage, and Follin carries about 8000 cars per day where it intersects Maple.

You would also have to ask whether those streets could carry enough additional traffic to make a dent in Maple Avenue congestion.  And whether drivers would be willing to divert along that route, given the numerous stop signs along Tapawingo.

My guess is, this is a non-starter.  Not enough proven benefit to be willing to trash the quiet neighborhoods at the end of Tapawingo and take public park land to create another W&OD trail crossing.

For a northern bypass, the only route I see would connect Abbotsford with Old Courthouse, destroying significant park land in the process.  You’ll need a map to follow this one, but:  Nutley to Malcom to Lawyers to Abbotsford.  Connect the two pieces of Abbotsford by putting a road through Northside park.  Connect Abbotsford with Old Courthouse.

Another non-starter.  This one destroys some neighborhoods and some parkland, requires a W&OD crossing, and adds almost two miles of narrow suburban roads to the commute, to boot.  I can’t imagine anyone could take this seriously.