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


I-66 "spot improvements" underway

A "spot improvement" on westbound I-66 between the George Mason Drive and Sycamore Street is underway.

These projects which are essentially short-distance widenings, have been the subject of great debate for more than a decade. VDOT has long wanted to widen I-66 to three lanes in each direction; Arlington, smart growth and environmental groups have been strongly opposed.

The original agreement to build I-66 included an agreement to not widen the highway beyond two lanes in each direction. VDOT settled on a series of "spot improvements," which widen sections of I-66 to three lanes. (Though most residents of Arlington would disagree that these are "improvements" at all.) This section is called "Spot 1."

In February of 2009, the National Capital Region Transportation Planning Board voted to put these projects on hold, only to reverse itself the very next month as Fairfax County put pressure on Supervisor Cathy Hudgins to break with Arlington. Since then, there has been little to stop this project from moving forward, particularly because it has received earmarked funding from Congress that can be used for nothing else.

I-66 looking westbound from Ohio St.
The contract was awarded in May, and construction has begun.

In the photo at right, you can see the (semi) vegetated median between the left shoulder and the Metrorail tracks. It is being partially removed to make room for the additional roadway width.

You can see the demolition of the median in the top image.

The Custis Trail passes under I-66 along this stretch of highway.  At the point were it crosses underneath there are three bridges: one for each direction of traffic and one for the Metrorail tracks. The westbound bridge is being widened.

To accommodate pedestrian and bicycle traffic during construction, VDOT has constructed a structure for protection (below). There have been occasional detours while work is being done here. The detour is well marked and only slightly longer.

Once the bridge is widened, there will be less daylight at this point. Although there is some relatively inadequate lighting now, it may need to be upgraded as part of the projects.

According to the project's entry in the 2030 Constrained Long-Range Transportation Plan, the main motives for the project are to reduce "recurring congestion, support the economic vitality of the metropolitan area, increase the safety of all users, increase the accessibility of people and freight, and to promote efficient system management and operation."

Furthermore, the project will ostensibly enable the corridor to serve as an efficient emergency evacuation route. But the suggestion that this will improve the road as an emergency evacuation route is just silly. The capacity of I-66 for evacuation will not be increased by adding a lane for a mile and a half. This language is likely a holdover from the arguments used for widening the entire length of I-66 inside the Beltway.

Once the "spot improvements" are all in place, there's a good chance political pressure to widen the relatively short remaining sections will grow very strong, and VDOT will eventually prevail. It will be interesting to see how long that takes.

Steve Offutt has been working at the confluence of business and environment for almost 20 years, with experience in climate change solutions, green building, business-government partnerships, transportation demand management, and more. He lives in Arlington with his wife and two children and is a cyclist, pedestrian, transit rider and driver. 


Add a comment »

Yep. Based on my conversations with the traffic engineers, the evidence that this results in a net improvement for drivers is pretty weak.

The simulatations fail to take account of increases in accidents due to aggressive drivers attempting to move up in the through lane queue by passing other cars in the auxillary lane. Moreover, if we think that accidents might increase, then instead of simply focusing on expected travel time we should also consider changes in its variance. That is, if there are more collisions and collisions have a big negative effect on travel time, then even if the simulations are correct that the average time time decreases -- which is not necessarily the case -- the variability of travel time might be much higher resulting in unpredictable long travel times when a collision occurs.

by Geof on Aug 31, 2010 4:27 pm • linkreport

I realize traffic engineering has a taboo against changing "driver behavior" but in my experience on 66 most of the delays are due to people who don't know how to merge.

by charlie on Aug 31, 2010 4:30 pm • linkreport

Steve --

Thanks for following up on this largely pointless and wasteful VDOT project.

Will there be any improvements to the existing sound walls on I-66 at these points? Some of them aren't really tall enough to dampen the sound very much for the surrounding neighborhoods, and there are also gaps in the walls.

by William on Aug 31, 2010 4:36 pm • linkreport

It's understandable that the agreement with Arlington should be maintained, however it needs to be re-interpreted to what it really meant in the first place ... I.e., I doubt anyone in Arlington at the time the agreement was made actually cared how much traffic was going through there ... only how big of a gash is this thing going to make. And '2 lanes' meant a smaller gash than '3 lanes'. Once the '2 lanes' were built, the area devoted to the highway ... including the median strip AND the side emergency lanes as well as the grassy area on either side defined the actual area used for the road, it defined it's width ... and thus the exect of the 'gash' going through the city. The highway administration should view itself as free to build as many lanes at it can fit in that footprint. It shouldn't feel itself constrained to '2 lanes' simply because that was the 'measurement unit' used ... I wasn't a restriction on the amount of traffic, but simply the width. And as long as they're not exceeding the width, they are in complete accordance with the agreement.

Also, if Arlington can't understand that it's got to think of others and not just its residents, than a simple fix would be removing most of the 'entry ramps' on to I-66 within Arlington. Do like the Dulles road used to do before the toll road got built alongside it. I.e., make it difficult for Arlington residents to actually make use of this road until Arlington cooperates. If you look around DC, there are plenty of streets where we too could have better quality of life within the neigbhorhoods if we hadn't allowed widening. Conn. Ave. down in Dupont comes to mind. Prior to its 50s era (?) widening, its very wide sidewalks would have been a perfect place for big sidewalk cafes.

by Lance on Aug 31, 2010 5:03 pm • linkreport

The push to widen is partially based on the desire to serve areas with afforable homes. Afforable homes are something that is in short supply in Arlington.

The best way to address the root issue is to allow more density in Arlington (or DC).
1) Allow greater building footprints
2) Reduce/eliminate parking requirements
3) relax height restrictions (how can a new Clarendon Condo builing be only 6 stories?)
4) Tax the value of the land and not the land plus structure

I agree with opposing widening 66, but Arlington can't simply point fingers without doing its part. (and GGW needs to remind Arlington about this)

by Madison on Aug 31, 2010 5:07 pm • linkreport

Arlington can't simply point fingers without doing its part.

Arlington's embrace of the best smart growth policies in the region qualify as doing more than their part.

by Vicente Fox on Aug 31, 2010 5:12 pm • linkreport

Could I ask what is the specific argument, besides the agreement, for not widening I-66 inside the Beltway? By specific, I don't mean the general "roads create traffic" argument and the NIMBY argument. Given there is already a lot of development out west, I'm not sure the first general argument applies.

Because currently, there are two or three routes out of DC in the west: I-66, GW Parkway and Rt. 50 which are already congested and also run through Arlington, or (hahahahaha) Metro. We've discussed the capacity crunch and other issues that Metro has on that line already, and it's likely to become worse with the Silver Line.

Is there a specific reason that I-66 shouldn't be widened? It's not like you're putting up sidewalk cafes on the side of the interstate as it is. This seems as good a use to me as any other, with the side benefit of easing congestion which is terrible, even with HOV-2 restrictions.

by varun on Aug 31, 2010 5:12 pm • linkreport

To add a point: an alternative that I'm willing to support is that Metro doubles the number of tracks there and runs trains more frequently. But that doesn't seem like a viable choice, seeing as you'd have to dig a new tunnel. The space is effectively unusable as anything other than a road, unless someone wants to try to run a new Metro line along the median all the way north along I66 to Georgetown as a secondary route to get out of DC.

by varun on Aug 31, 2010 5:17 pm • linkreport

@Madison The best way to address the root issue is to allow more density in Arlington (or DC).
1) Allow greater building footprints
2) Reduce/eliminate parking requirements
3) relax height restrictions (how can a new Clarendon Condo builing be only 6 stories?)
4) Tax the value of the land and not the land plus structure

The first 3 of your 4 suggestions requiring a lowering of living standards (i.e., 'squeeze 'em in', 'make it impossible for them to own and operate their own cars', and 'stack 'em like rabbits in highstanding cages.") Isn't just assuring the access to greater land resources so that people can have both affordable housing AND the freedom of not being 'squeezed' and forced to the 'lowest common denominator' a better option for all involved?

by Lance on Aug 31, 2010 5:18 pm • linkreport

Arlington's got to think of others and not just its residents

Was "thinking of others" the philosophy behind Fairfax County's zoning decisions?

by Marian Berry on Aug 31, 2010 5:34 pm • linkreport

@Varun; the argument against widening is it won't do any good.

Westbound, I have some sympathy. It might help a bit after Ballston. That is where drivers have some real problem understanding how to merge.

Eastbound, into the city, 66 gets clogged up because there is a stoplight. Seriously. And the tunnel.

by charlie on Aug 31, 2010 5:42 pm • linkreport

@Charlie: Yeah, but this entire piece is an argument against westbound widening. Of course it makes no difference eastbound, because I66 is not where the bottleneck is.

But even before Ballston, there is often congestion between the Fort Myer underpass and Glebe Road. There's also often a merge congestion from Spout Run where people abandon GW to come over to I66. There is space for a third land there, and that requires no removal of greenspace. Why not?

Look normally, I'm totally for transit. But I just don't see the argument against widening the westbound road here.

by varun on Aug 31, 2010 5:49 pm • linkreport

I know it isn't cool to say here, but I think between the Dulles Access Road and the Fairfax drive exit in Ballston, one extra lane each way would really help. Waiting to see how the new Metro line affects things is also a prudent first step on this.

by NikolasM on Aug 31, 2010 5:51 pm • linkreport

Third lane* not land.

Or give a shoulder that can be used as a lane during peak outbound time ala I66 between the Beltway and Rt. 50 and the "spot" improvements can serve as the shoulder during those times.

Or run transit there - maybe a dedicated BRT lane. It's not usable as anything but a road, really!

by varun on Aug 31, 2010 5:53 pm • linkreport

@Varun; I was 90% against widening because I saw it as a waste. I am 65% against it now, based more or less on what you say. However, it the problems you mention really are not the roads fault -- they are the drivers fault. Spout ray westbound, for instance, has plenty of room to merge -- it is just drivers can't figure it out.

Also not sure how Custis Trail fits in -- parts of run real close to the highway.

by charlie on Aug 31, 2010 5:56 pm • linkreport

@Charlie: In my experience on the Custis Trail, it's not just separated by a barrier, but often above (grade-separated from) the I66 roadway. I think there's one spot it's in line with the roadway, but that's when Lee is between the interstate and the trail. 24

by varun on Aug 31, 2010 6:05 pm • linkreport

@Vicente - Fair point. I live in (own a home) and love Arlington. It certainly deserves credit for smart growth. I wish more could be done. Getting rid of the density regulations would relax pressure on family budgets for housing and transportation. It would also relax pressure on government budgets for sprawling road and transit infrastructure.
I suppose GGW mentions this in other articles, but when Arlington starts demanding something (even if valid). It is worth GGW reiterating Arlington has obligations as well (and can always do more to address them).

by Madison on Aug 31, 2010 6:09 pm • linkreport

Regarding the first comment made by Geof, typically accident rates decrease, not increase, when auxiliary lanes such as these are added.

by Froggie on Aug 31, 2010 6:52 pm • linkreport

As an Arlington resident, I welcome the current "spot improvement." That is because the backup is large almost all hours of the day approaching Glebe Road through the end of the merge from Fairfax Drive.

As an Arlington resident this will help me because it will move the backup further west.

by Alana on Aug 31, 2010 7:06 pm • linkreport

When the TPB "reversed itself" in March 2009, it only allowed VDOT to construct "Spot 1" prior to completing a full and fair alternatives feasibilty study for the future of the I-66 multimodal corridor. VDOT's representative to the TPB also publicly promised in 2009 that VDOT would not widen I-66 beyond the three proposed "spot improvements".

In June 2010, VDOT's Commonwealth Transportation Board allocated $4 million to conduct a new feasibility study for I-66 inside the Beltway. That 12- to 18-month study should get underway in early 2011 and have a robust public involvement component.

"Spots 1 and 3", which respectively end at the low-volume Sycamore St and Glebe Rd exits, can do little to expand I-66 throughput for commuting or evacuations, because two-lane westbound bottlenecks would remain at Sycamore St and at Glebe Rd.

I-66 inside the Beltway was built as a managed highway with peak-period High Occupancy Vehicle (HOV-4) restrictions, and the best way to move people and minimize traffic congestion in the corridor is to better manage I-66 traffic with 1) expanded HOV hours accompanied with express-bus service(now just 2.5 hours on weekdays in the peak direction only), 2) effect and effectively enforced HOV restrictions (so I-66 is always free-flowing during HOV hours), 3) uncongested busways (in free-flowing HOV lanes and bus-only shoulder use at other times), and 4) ideally, variably priced automated tolling (to let SOV drivers pay to use I-66 during HOV hours).

by Allen Muchnick on Aug 31, 2010 7:12 pm • linkreport

Allen: I'm assuming under your #4 that the proceeds would be used to support the express bus service you promote under #1...ala what MnDOT does with I-35W and I-394 in Minneapolis...

by Froggie on Aug 31, 2010 7:35 pm • linkreport

Yes, *I* would use the toll revenue to subsidize bus and rail transit and ridesharing in the corridor.

by Allen Muchnick on Aug 31, 2010 8:25 pm • linkreport

I never thought that widening 66 was about easing congestions. From the comments I've seen on this board (and others) by some pro-wideners, it seems to be about "Hah! Take THAT Arlington!".

by Max D on Aug 31, 2010 8:30 pm • linkreport

...the Devil's Advocate in me wants to point out that without I-66, it is unlikely Fairfax Dr., Wilson, and Clarendon Boulevards would have reduced lanage and on-street parking. Reducing vehicular traffic in the corridor, as well as the car park barrier has made it feel more comfortable to walk, and that in turn has made strolling to the metro stations along the corridor that much more attractive. Reducing thru-traffic on these roads is a huge positive.

Devil's rant over.

That said, if this added capacity to I-66 were coupled with congestion pricing (equal to or more than the cost of a metro ride), that much the better.

by stevek_fairfax on Aug 31, 2010 8:39 pm • linkreport

I don't know what you're all talking about, this project will accomplish myriad things!! What you ask? Well, let me just list the top five for you:

(1) Move the bottleneck from one section of Arlington (Ballston) - to another (Sycamore Street/East Falls Church).
(2) Allow cyclists and joggers to breathe more car exhaust on Custis Trail when 3 lanes are full of solid traffic commuters trying to get to Fairfax County and Beyond...
(3) Allow more storm runoff to go into Four Mile Run, carrying such rich nutrients as motor oil, antifreeze, and a seasonal dash of rock salt - this way we can make sure that those pesky salamanders have nowhere to live.
(4) Give Orange (and soon Silver) line Metrorail riders more cars to gawk at as they zoom by in the median during rush hour.
(5) Concentrate more CO2 along the corridor, spurring growth of the trees lining the tracks (oh wait, some are being destroyed for the extra lane.) Well, the ones on the other sides of the sound barriers will have all the tasty carbon they can suck up when they have leaves. We all know there isn't enough in the air around here for them already.

by Matt Glazewski on Aug 31, 2010 8:58 pm • linkreport

Your statements regarding density are mistaken. Higher density living does not equate to lower quality of life. The City of Paris is about 40 square miles but houses over 2 million people (53k per square mile). Arlington County is about 26 square miles with about 210,000 people (8k per square mile). Sure living in Paris would be very different than living in Arlington, but many folks would not consider it being stacked like rabbits in cages. Au contraire, many folks would consider it a much higher standard of living.

by FredInRVA on Aug 31, 2010 9:24 pm • linkreport


I understand the intuition behind making it cost more to drive than take Metro, but in practice, can any more people really fit on the Orange Line? High congestion pricing would likely just move vehicular traffic to other roads.

It does seem silly to add extra lanes, rather than an extra track or two, though. If there could be an express train from Vienna straight to East Falls Church, at the very least people could get an idea of the potential gains from wider tunnels/a second river crossing.

by jakeod on Aug 31, 2010 10:16 pm • linkreport

One reason to not widen I-66 is to help the region (or force the region, if you prefer) to analyze much more closely why we have so much traffic. How much of it is truly necessary, and how much of it is just lemmings following lemmings, year after year?

I've become very interested in the impact that Dulles Airport has on all our traffic, even all the way back to I-66 in Arlington.

Technically, Dulles is reachable by mass transit. However, the service is so poorly publicized and the transit schedule is so inadequate that it is minimally used. (In fact, unlike virtually every other city, it is now literally impossible to take an express bus from downtown to Dulles. The venerable non-stop Washington Flyer bus from 16th and K Street was eliminated in 2004.) There is a Flyer bus from the West Falls Church Metro station, but of the 17.8 million passengers who use ground transportation to or from Dulles every year, only about 250,000 use it. Another 500,000-or-so use Metrobus Route 5A from L’Enfant Plaza and Rosslyn. Everyone else – more than 17 million persons – either drives his or her own vehicle, rents a car, gets a car ride from a friend or takes a taxi. Many of them use the Dulles Access Road and I-66.

The problem is the schedule. The Flyer normally operates only every half hour, sometimes less. Bus No. 5A is on a 40-to 50-minute schedule weekdays and hourly on weekends. Complaining to the Flyer management yields the response that there is no demand. “Take a look around you,” I am told. “There are only 10 passengers. We can’t even fill the buses at two per hour, much less if we ran four or six.”

With that kind of faulty logic, a single bus per day would be packed with passengers. In point of fact, it’s obvious that generous scheduling is key to any successful transit service. It is also obvious that there is a fantastic demand to get to Dulles – the airport has a jaw-dropping 40,000 parking spaces (which consume 300 acres). The problem is that parking is plentiful and inexpensive, and transit is scarce -- and that's one of the reasons for the demand to widen I-66.

And it's not just passengers who travel to Dulles. The airport’s workforce is an astonishing 36,000 – for every 12 airline passengers there are an additional nine airline and airport workers driving on the Dulles Access Road and other routes. (These people have it even better. Courtesy the Transportation Security Administration, parking for them is free.)

Maybe it’s not that Americans “love to drive” or “can’t be pried out of their cars.” Maybe they just prefer not to turn down low cost parking and a personalized schedule for expensive and intermittent transit.

Solving this problem is not that difficult. If Dulles increased its parking rates and used the profit to subsidize and jump-start much more frequent bus service, it would begin a "virtuous cycle" of transit use that would reduce pollution *and* reduce traffic on I-66 and other roads.

by Peter Harnik on Aug 31, 2010 10:54 pm • linkreport

I've skipped over the comments & don't know much of the project or the area, but if I'm reading it right: it sounds like a third lane is being added for only a short distance. Unless that lane has a useful endpoint (such as dropping onto a heavily-used off-ramp), then all a short-distance capacity increase would do is induce a new bottleneck where the lane merges back in.

Even in situations where the preceding bottleneck is more restrictive, the human element (that is: motorists don't behave exactly as per fluid dynamics) is such that each individual may increase their separation / headways, expand into the additional capacity, and potentially re-bottleneck at the next capacity reduction.

To go on a bit of a rant... I'd like to think that any traffic engineer with a PE & any recent graduate should be well aware of this -- treating human behavior as separate from fluid dynamics has been pretty solidified in traffic engineering for more than a decade now. I'd be curious to know if the staff-level folk are supportive or against such a project.

...Or it could be I'm just misunderstanding the project, in which case my rant would (hopefully) be moot.

by Bossi on Sep 1, 2010 1:30 am • linkreport

Bossi: the three spot improvements are auxiliary lanes between a given on-ramp and the following off-ramp, if that helps.

by Froggie on Sep 1, 2010 7:11 am • linkreport


Ahh, that does... in that case my comment / rant might go out the window. I'll refrain from chiming in further until I actually know what's really going on, and since it's in Virginia there's a good chance that'll never be so :)

by Bossi on Sep 1, 2010 9:07 am • linkreport

@ Peter Harnik; very interesting data. I've been repeating the one nugget I know -- which MWAA expects 5% of airline passengers to use metrorail -- but your data puts that into some context.

The 5A could clearly be expanded. I've seen it so busy that people can't board.

But in terms of 66, you might be overstating the case for the impact Dulles traffic has on the road.

@Bossi, I tend to agree that all you are doing here is moving the traffic jam a bit further out.

by charlie on Sep 1, 2010 9:25 am • linkreport

I-66 connects, or goes through, the already dense urban corridors of Arlington and Downtown Washington. Adding a single additional lane simply addresses the reality that the road way over capacity as it is. The congestion that exists in the bottleneck (yes, I-66 itself inside the beltway is a bottleneck, in that it goes from 4-5 lanes in Loudon Co. to just two), is dangerous, a waste of time for commuters, visitors, and in-town drivers, plus it's a waste of gas.

The fact that right now there is a traffic jam on I-66 according to Google, at 10:30am on a Wednesday in August, says that something needs to be done, perhaps a third lane as a permanent HOV lane. A two lane road behaves differently than a three lane road. This is what is needed on I-66.

The Washington Metro region is the home of some of the longest commutes in the US, the infrastructure is severely in need of fresh ideas. We need to let go of the old NIMBY arguments for a two-lane I-66 and look at the miserable reality, and the even more troubling inevitable future should we continue this heel-dragging.

by Fabian on Sep 1, 2010 10:31 am • linkreport

Yes, it is September. I was fooled by the weather!

by Fabian on Sep 1, 2010 10:34 am • linkreport

@Fabian; Google is reporting a delay eastbound; WTOP (which I think is more reliable) is not showing any problems there.

66 goes into "Loudon" county? news to me. If 66 is wider in Fairfax, that is because, well, Fairfax is bigger and that is where it meets the Beltway.

Honestly, maybe because I had to deal with Circualtor today instead of the blue bus, and so I am really pissy, but 75% of the traffic problems of 66 inside the beltway are dumb-ass Virginia drivers who can't merge and deal with high volumes.

by charlie on Sep 1, 2010 10:44 am • linkreport

Congestion on 66 is definitely an interesting subject. I have come to the conclusion that part of the reason for the slow driving is simply the short sight lines on the highway (inside the Beltway). It winds a lot and with only two lanes it feels sort of cramped. And with only two lanes, it doesn't take more than a few slowpokes to gum up the whole works. At this point, I'm convinced there are also certain sections where the expectation of a traffic jam causes people to slow down and then - boom - traffic jam.

Interesting point about whether or not VDOT can build a third lane and still live up to the agreement because the "slash" is the same. I hadn't thought of that. Regardless of the legal sufficiency of the argument, I think from a purely logical point of view, it might hold water. On the other hand, there is the policy point of view and that still, I think, argues strongly against widening the freeway because it will just induce more traffic and thus become a bigger mess.

As for the reasonableness of constructing this "auxiliary" lane - sure, it just moves the merge one mile further west, with no exits in that stretch. So it doesn't necessarily make anyone's travel any faster. Perhaps with the longer distance, merges can happen more smoothly, which would speed things up. I don't think we can say until it opens and we find out.

by Josh S on Sep 1, 2010 11:03 am • linkreport

In response to your question re: sound walls. The current project is not going to make any changes--good or bad--to the existing sound walls as far as I'm aware.

Also, because this part of I-66 has the space in the center, it will not affect the Custis Trail in any way (other than the temporary structure I pointed out in the post)

I agree with those of you who argue that it will move the bottleneck to the west. That may not be all bad, because then it will not spill back onto local streets but will be occurring on the highway itself.

@allenmuchnick Thank you for sharing additional details.

@Lance & FredinRVA - A strong indicator of "quality of life is cost of housing. People pay more to live where they perceive their lives will be better. Arlington has high housing prices, so it stands to reason that there is demand to live there--thus, some portion of the population must be perceiving those living conditions as higher quality.

@charlie You are right; I-66 does not go into Loudoun County. It starts (or ends, depending on your POV) in DC and goes west through Arlington, Fairfax, Prince William, Fauquier, Warren counties and ends in Frederick County at the interchange with I-81.

I'll reiterate my opinion that these spot improvements will not make much change in the overall capacity of the road; it will just move things around. This particular spot can be constructed without any impact outside the ROW. However, trying to widen I-66 closer in to to DC becomes more and more difficult, particularly through the Rosslyn tunnel.

by Steve O on Sep 1, 2010 11:28 am • linkreport

@Varun and others who believe that the spot improvements will significantly reduce travel times ...

The mechanism by which Spot Improvements #1 and #3 decrease travel times is by providing more space for drivers to exit and merge onto I-66 and reducing the "noise" -- chaotic speeding and slowing down -- associated with those actions. The problem is the simulations that predict improvements in travel time fail to include some basic human tendencies in the face of slow moving traffic. More specifically, that drivers will move into the auxilliary lane to pass traffic in the through lane when traffic is slow and that will likely lead to more "noise" and collisions.

I discuss it here ...

Anyway, if you're like me and think that government resources should be chasing projects with high returns to the dollar and passing over those with more dubious returns, then this is a bad project.

by Geof on Sep 1, 2010 4:39 pm • linkreport

A few observations from someone who commutes in this corridor daily:

1) As jakeod points out, the Orange Line is nearly at capacity as it is, and the merge of Orange and Blue into the Rosslyn tunnel is the worst bottleneck in the system. It remains to be seen how Metro handles Orange, Silver and Blue trains merging into the Rosslyn tunnel. More eight-car trains would increase capacity, but they're not in the budget anytime soon. So rail transit can't soak up much additional capacity.

2) Even during HOV hours, I-66 slows to single-digit speeds between Lee Hwy/Washington Blvd and Fairfax Drive in the morning and between Fairfax Drive and Sycamore in the evenings. This is a big disincentive to carpooling, and means that any buses routed on I-66 travel much more slowly than rail transit.

3) In the evenings, a lot of traffic enters I-66 on Fairfax Drive and a lot exits at the Dulles Toll Road spur. Since I-66 is already three lanes from Sycamore to the Toll Road, it doesn't seem unreasonable to me that increasing capacity between the large input at Fairfax Drive and the large output at Sycamore/Dulles spur would improve flow.

4) Cars entering the highway at Fairfax Drive are presumably people who work in Arlington but live west of the county. Easing the commute for Arlington workers does seem to be of some benefit to the county.

5) Enforcement of HOV is pretty strong and effective. There aren't a lot of lone rangers, unless you count the hybrid drivers with Clean Special Use plates. But there are quite a few of those. Anecdotally, they seem to be about 40% of the traffic. The state has already ended the hybrid HOV privilege for one class of plates on Shirley Highway HOV lanes. Speaking for myself, I hope these privileges are sunset on I-66 soon. It's very galling to take the trouble of carpooling, only sit in traffic, staring at the bumper of a hulking Highlander Hybrid that gets worse gas mileage than my car, yet has a solo driver thanks to Clean Special Use plates.

6) If I recall correctly, I-66 initially had HOV-3 or HOV-4 restrictions, which was scaled back to HOV-2, which seemed to be the the sweet spot at the time. Shirley Highway is HOV-3. It might be time to consider that for I-66, which would ensure that there is a benefit to carpoolers and buses using the road during HOV hours. But my preference would be to cut back on the hybrid privileges first.

by c5karl on Sep 1, 2010 6:52 pm • linkreport

@Geoff; interesting link, thanks.

But aren't both hypothesis untested: that an increased merge lane will reduce time, and an increased merge lane will encourage "aggressive" driving and cause more backups?

People like me who only encounter 66 occasionally need to be more sympathetic to regular commuters there. But I do have to admit struggling to understand how these latest round of improvement will work better...

by charlie on Sep 1, 2010 7:33 pm • linkreport

@Charlie ...

Yep. You are absolutely correct. I cannot prove my alternative hypothesis. But that is the point. The conclusion that this results in an improvement for westbound drivers is not robust. At least by what I consider scientific standards. And there are intuitive reasons for thinking that it might cause more problems.

Note that the traffic engineer who did the analysis sincerely believed that it would improve matters and seemed quite honest about the limitations regarding the analysis.

by Geof on Sep 2, 2010 9:21 am • linkreport

Oh geez ... I'm terribly sorry for taking up so much cyber-space. I just looked at the bottom and saw that GGW already referenced an old post of mine. Two babies have eaten away at my memory.

... a mind is a terrible thing to lose.

by Geof on Sep 2, 2010 9:28 am • linkreport

That's why I'm not proposing spot improvements - I say, let's add another complete lane westbound to the beltway, i.e.: widen it, not just "spot improve" it. The point I was making is that this article seems to be a blind argument against any road widening, not one that looks at the specifics of this particular road. There is no real alternative use of the space on the side of an interstate. If there is, then by all means, I'm happy to hear it; I even would love to see a dedicated BRT lane or something similar in the median, provided we have buses at least twice as frequently as trains. But otherwise, let's add one complete extra outbound lane and reduce the congestion that exists even during peak HOV hours.

by varun on Sep 2, 2010 6:27 pm • linkreport

It is a myth that Virginia promised not to widen I-66. They promised not to widen the right of way which is completely different. The road can be six lanes in its current footprint. Ultimately it has to be done it simply doesn't make sense why a few residents can cripple NVA. I go to these meetings. The only people that complain live next to the road and think noise will increase dramatically with additional lanes which is simply not the case. To be fair VDOT should be extra vigilant with noise reduction solutions. The region needs more capacity.

by Sivad on Sep 2, 2010 10:39 pm • linkreport

As an Arlington resident and frequent user of the inside-the-beltway portion of I-66, I have to agree that it is broken, as many comments have pointed out. I don't know whether the spot improvements will help, but the highway needs to be fixed. I know the arguments about larger highways producing sprawl, but it seems like there has to be a happy medium with I-66; having a two-lane highway as a major entrance to DC seems doomed to fail.

One comment really picked up on one piece of the problem that I haven't seen discussed. People driving on I-66 inside the beltway drive at slower speeds than on just about any major highway I've seen. It is not uncommon to see two cars driving side-by-side at less than 50 MPH, blocking the only two lanes of traffic, with piles of cars behind them. It is just strange and vexing.

by traffic on Sep 5, 2010 11:33 am • linkreport

To say that I-66 is broken is to imply that there is a possibility that it might be, or was once, whole. But it has never worked. Nor are its problems limited to the Arlington portion. It is more congested outside the Beltway than it is inside the Beltway. It has been widened outside the Beltway several times (for SIX peak hours in each direction it has no shoulder!). These widenings have not helped. It seems clear to me that it should never have been built. Messing with it is simply throwing good money after bad.

by jim on Sep 5, 2010 1:45 pm • linkreport

Why NOT add a new train line? If you have room for two road lanes (one eaxch way) you have room for a train line.

by Nathanael on Sep 8, 2010 3:06 am • linkreport

"There is no real alternative use of the space on the side of an interstate. If there is, then by all means, I'm happy to hear it;"

Express rail.

Yes it requires grade separations, so the median is better for it.

Local rail doesn't work that well in freeway medians (they're poor station locations), but it's great for expresses.

by Nathanael on Sep 8, 2010 3:12 am • linkreport

As an Arlington resident that works West of Arlington, I'm all for ANY improvements that can be made to 66 (improvements meaning any lane expansions). I would've been much happier to see the expansion from glebe all the way to where the third lane starts before the 66/DTR split, but it appears baby steps need to be taken, since people can't agree on it. The NIMBY approach that many Arlington residents have is pathetic imo.

Someone said above that an expansion of lanes outside of the beltway for 66 still results in traffic, and 66 should've never been built?! Wow is all I can say, and I understand why traffic is so bad in the area. Of COURSE the extra lanes help, and 66 itself. All those people taking 66 would have to take other roads, which would be even worse traffic on other roads. Come on people, use your brains.

Also, I don't think the NIMBY/too bad for those that live further out is an intelligent position to have. Tell me other major metropolitan areas that don't have a major freeway (more than 2 lanes) entering or exiting the city from the east/west/north/south? They don't exist, at least nothing in comparison to the size of DC.

I travel opposite the heavy flow of traffic on 66 and still regularly hit substantial traffic (I ride 66 for a whopping ~3 miles each way). That's completely ludicrous. Frankly, I think there should be 3 lanes going each way on 66 inside the beltway. Yes, it doesn't fix some bottlenecks, but it alleviates traffic for everyone that travels between those bottlenecks, which is a BIG improvement. There are plenty of people that work in Tysons/west and live in Rosslyn, which widening 66 would greatly save time for.

As an example of road widenings that help (help defined as moving more people on the roadway in a given period of time) look to 28 that goes through Loudoun and Fairfax county. I remember many years back with lights galore and being bogged down. Now it's a breeze. I think it's sad that a highway built out in the stix is larger/wider than the main Western highway that leads into DC. Maybe DC ought to close the Rosslyn bridge as it's causing too much traffic on Constitution.. *banghead* There's a thought, huh..

by Arlington Resident on Sep 23, 2010 2:27 pm • linkreport

Add a Comment

We are switching over to our new website. Commenting on the old site is now disabled. Thank you for your patience and pardon our dust!

Support Us