NATIONAL CAPITAL:Internet Boom Bogs Traffic

LA is thinking big on truck toll lanes, HOT, new toll roads, and some major freeway widenings if the Southern Calif Assoc Govt’s plans are any indication. And it has given up squandering money on wasteful rail transit. (see TRnl#23 Jan 98) Washington by contrast has had timid leadership and is planning rather little of promise. Its highest transp priority is reconstruction and widening of the Wilson bridge, the Beltway’s southern Potomac River crossing. Of 6-lanes only it serves local traffic as well as the 8-lanes of the Beltway on either side, so is a major bottleneck, as well as in structurally poor shape.

Plans are to replace it with a 12-lane design but this project is being massively mishandled – a vintage Washington showcase for the world of how NOT to go about a major infrastructure rebuild. The project team showed contempt for established law and procedures for assessing historic and environmental impacts, and having thrown out the rulebook, has been forced by a judge to restart the long permitting process.

A more wasteful, inflexible roadway format (3/3/3/3) would be difficult to imagine. 6m (20') of bridge deck is required for each 3.65m (12') travel lane in order to provide double breakdown shoulders on each of four separate roadways (2-local, 2-through traffic plus HOV). Fixed barrier separating the four roadways will ensure many lanes are used under capacity, while others are overloaded. There are no plans to draw on the successes of variable pricing to balance traffic flows or prevent overloading and breakdown into stop and start driving – in short this is an unimaginative 1960s design guaranteed to provide minimal benefit to motorists at maximum cost and maximum impact on the city of Alexandria through which it passes. The proposed 74m (244') width of pavement is equivalent to about two complete blocks through the old town. 40% will lay unused waiting on a smash, or someone out of gas.

Cost is currently estimated at $1,900m for the new bridge including the reconstruction of two closeby interchanges on either end. Never before has a bridge of this scale been built without toll financing. Like the George Washington Bridge, or the Tappan Zee bridges in New York, or the McHenry and Harbor tunnels in Baltimore, the Wilson bridge carries 80 to 90% commuter and local business traffic, and an important 10 to 20% interstate or inter-regional through traffic. As on these Hudson River and I-95 crossings, tolls are the only way to sensibly charge a diverse collection of users for the bridge, and to provide a revenue stream for the operations and upkeep of the Wilson bridge. Yet in a break with the longstanding tradition of using toll revenue bonds for large crossings, the states of Virginia and Maryland are attempting to get the bulk of the funds ($1,500m) out of the US Government, and propose getting the remaining $400m out of their own taxpayers.

The Congress has committed $900m in TEA21, but the VA and MD govts are lobbying congress for another $600m. Tolls are still a possibility if Congress refuses to indulge the Washington DC area with special handouts that it would never provide to New York, Baltimore or San Francisco.

Many of those familiar with the examples of 91-Express and I-15 in San Diego see the Wilson Bridge as a perfect example to use variable tolls manage traffic on a larger scale. The new Wilson bridge is likely to carry over 200k veh/day when it opens putting it right up there in volume with the largest crossings in the country.

Compromise: toll-XLs

A workable compromise could be to impose variable tolls on the inner roadways, operating them as premium service toll express lanes, while using the federal handouts to fund the outer roadways, where motorists would have to take their chances on the vagaries of flow in unpriced lanes. Then at least a portion of the traffic flow could be managed to provide a higher level of service and lower emissions.

Maryland DOT has a set of studies underway of priced toll express lanes (XLs) on its portion of the Beltway (I-495), as well as several major radials leading into the Beltway - MD-210 whose interchange with the Beltway is right next to the Wilson Bridge, and US-50, I-95 and I-270 - so the pricing of the inner lanes of the Wilson bridge would fit with that. There’s the genesis of a HOT lanes/toll express network for the area.

There might be support from environmentalists who are influential. Michael Replogle of the Environmental Defense Fund (EDF) has said he favors taking the Beltway format from its present 4/4 format to 3/2/2/3, the inner 4-lanes being HOT lanes, and the outer-6 unpriced. Virginia is presently set on taking its portion of the Beltway to 12-lanes with the 3/3/3/3 format of the Wilson bridge being the most likely, though 4/2/2/4 and 2/4/4/2 are still possibilities. Virginia DOT has shown little interest so far in the potential of variable tolls to manage traffic flows and still talks pure HOV lanes for the Beltway and the big radial motorways off it (I-95, I-66, Dulles).

A recently released draft Northern Virginia 2020 Transp Plan (NoVA-2020 plan) allocates $2,000m or over half of the total highway improvements of $3,802m to the Beltway widening. The extra 96 lane-km (60 lane-mi) will cost $21m/lane-km ($33m/lane-mi). Much of the money is in reconstruction of 7 interchanges where expensive right of way will have to be acquired, though between ICs the reserve is wide enough to go to 12 lanes. A separate project already under construction is the $500m rebuild of the area’s most heavily trafficked IC (I-95/I-495/I-395) in Springfield. This IC will provide duplicate ramps for all movements for the inner roadways, be they HOV, toll express or through traffic.

I-66 going west from the Beltway is presently 3/3 and 2/2, though in rush hours the use of shoulders as travel lanes allows it to run an extra lane peak direction. Plans are to rebuild 28km (18mi) to a 4/2/4 format, the center 2-lanes being reversible HOV, at a cost of $535m. By our estimate it is another 113 lane-km (71 lane-mi) at $4.7m/lane-km ($7.5m/lane-mi).

Plans are already underway to widen the reversible barrier HOVL facility on I-95 and I-395 (its continuation inside the Beltway) from its present 2-lanes to 3-lanes for 34km (21mi) going as far north as the Pentagon.

In the northwest no roadway expansion is proposed for the Dulles Toll Road (which has room for 14-lanes plus transit), though part of the Dulles Greenway is to go to 6-lanes with 2 HOVLs being added. Major expansion of capacity envisaged here is the progressive upgrade of some 30km (19mi) of Leesburg Pike, presently a 2x2 and 2x3 lane signalized arterial to motorway standard through construction of some 8 interchanges. Since this will be directly competitive with the Dulles Toll Road and Dulles Greenway it would be logical to fund this work with tolls to have the two routes competing on an equal footing. Leesburg Pike would then be a real pike once more!

Another proposed enhancement of a signalized arterial to motorway standard is VA-28, a north-south route that runs from the Manassas-Centerville area south of I-66 right past the eastern entry to Dulles airport and up as far as VA-7 (aiming right for Gaithersburg MD!). It was built to its present 2x3 lanes with a special tax on adjoining property owners who were legislated to be members of a special tax district. Ten ICs would be built to replace intersections to progressively upgrade this to full motorway standard.

The northern stretch of VA-28 would combine with VA-7 to provide a second motorway route in the Dulles-Leesburg corridor and an alternative to the Dulles Greenway toll road. That could be a prospect for tolling too, or otherwise the Greenway would lose traffic to a competitive free road of comparable quality. ICs are also to replace some intersections on the Fairfax County Parkway, FC-7100, a kind of second beltway which was built earlier this decade with fairly good access control but without grade separation. Fairfax County official Katherine Ichter said after the opening of the road that the county got much criticism for the backups at the traffic signals: “Everyone asks now why we didn’t made it a proper parkway with interchanges. At the planning meetings beforehand the people who showed up wanted grade intersections and signals, so that’s what we did. Now we are going to have to go back and progressively install interchanges.”

Big Story

Almost every day this summer the WASHINGTON POST has managed to make a story out of traffic congestion in Northern Virginia (NoVA). The VA subregion (1.8m pop) is less that four-tenths of the greater Washington DC metro area (pop 4.8m) or only a quarter of the 7.5m if you follow the Census Bureau finding that there is now sufficient cross-commuting and other interconnections to form a Washington-Baltimore area metropolis.

But congestion problems are growing faster on the west bank of the Potomac, because that is where the growth of business and population is fastest. NoVA is the Internet Capital of the World, the place where URLs are allocated and administered, home of Newtork Solutions and Internic, home of AOL the world’s largest ISP, home to UUNet and MCI Worldcom and a whole heap of internet venture capital companies. Cofounder of Netscape, Marc Andreesen was recently featured on the cover of WASHINGTON BUSINESS for buying a large house in McLean VA for himself and his bulldog. Direct air service Dulles-San Jose is heavily traveled as NoVA develops deeper relationships with Silicon Valley. NoVA unfortunately doesn’t have Silicon Valley’s extensive network of freeways and the high quality arterials that it calls expressways. Business leaders including Steve Case of AOLand a bunch of smaller entrepreneurs are calling for action to ease congestion. It makes a good news story because there is a chorus of calls for higher taxes dedicated to new transport projects pitted against a Republican Governor (James Gilmore) who got into office with a campaign against taxes. Gilmore calls for something “more creative,” but is as yet vague about what that might be.

Klinge Commission

Gilmore late August came up with an interim list of projects he’d support from general funds but any structural change seems likely to wait on the Klinge Commission on transport policy, a group of about 20 NoVA officials and business leaders who are working to a Dec 2000 deadline for the governor.

We asked Kenneth Klinge, a former USDOT official, political consultant, and longtime friend of the governor whether his commission was going to define a policy on tolls and road pricing. He said: “The governor has asked us to look at everything, and we plan to do that, so certainly we’ll be looking at tolls.”

Klinge who lives in Alexandria volunteered: “Northern Virginia has one extremely successful toll road, the Dulles Toll Road. It is possibly one of the most successful toll roads in the country. It has been a winner from the day it opened. The Dulles Greenway got off to a weak start, but it is doing well now and it will be very successful before long also.”

He told us that when he was first told about the idea of a toll road in the Dulles corridor in about 1980 his reaction had been: “People here won’t pay tolls. It won’t fly.” He quickly added: “That shows how much I know.” The Dulles toll road idea, he said, was pushed by Til Hazen, a major NoVA developer whose genius was in thinking up the idea of running a toll road alongside the Dulles Airport Access road, without encroaching on the airport road. (The whole road now has a 4/2/2/4 configuration with the two inner roadways being the original Airport Access Road and the outer 4-lane roadways forming the toll road.) All the pressure until that time had been to open the existing Airport Access road to local traffic, which the airport adamantly resisted.

Hazen had broken the political logjam with the separate toll road idea and it was “accepted in about five minutes.”

Klinge said he is a strong supporter of a third major Potomac River crossing to improve connections between Maryland and Virginia. He agreed it could well be a toll project, but said financing is not really an issue at this point because of the reluctance of Maryland to agree to any highway across the rural strip on its side of the river: “We’d have it (a crossing) built in no time if it was up to us.”

Grow

Transp expert at the Greater Washington Board of Trade (GWBoT), Robert Grow, is another who thinks extra Potomac River crossings are essential to the future of the capital area. His organization sponsored traffic modeling that suggested the need for two new crossings upstream (north) of the Beltway and one downstream. The NoVA-2020 plan has large vague arrows suggesting one Northern Potomac Crossing and one Southern Potomac Crossing, but has them slated for post-2020. Many think they have a much higher priority, especially the Northern.

The Potomac Conference, a group of local businessmen and officials led by Steve Case, CEO of AOL is working on a proposal it calls TechWay, a concept to provide high quality transport between the hightech areas of Fairfax and Loudoun counties NoVA and Montgomery and Howard counties in Maryland across the river. Grow says TechWay will be multimodal and therefore more than just a highway, but clearly it will involve some kind of road link. The rationale for the project is that the heavily biotech and health research of Maryland and the internet/hightech of northern Virginia can both benefit from better accessibility to one another. And NoVA would benefit by having improved access to the Baltimore-Washington airport area, just as Mongomery and Howard counties in MD would benefit from an alternative to the Beltway to get to Dulles airport. Herndon VA to Gaithersburg MD, from the heart of telecom/internet to the heart of biotech is 43km (27mi) taking Dulles Toll Road/Beltway/I-270, a U-plan trip that is upwards of an hour in the extended peak periods. A direct connection via the proposed Northern Potomac Crossing would be about 26km (16mi) and about 20mins, for a 17km (11mi) distance savings. If managed for free flow it could provide 30min to 40min time saving.

Wayne Lemmon, a veteran Washington consultant says the idea is “very logical” but says that Maryland officials fear the competition of a more business-oriented wheeler-dealer culture in VA. They think a new crossing would encourage Maryland residents and workers to use Dulles rather than BWI airport, and encourage them to shop in VA which has a 1c lower sales tax. Lemmon says MD politicians see the long unbridged section of the Potomac as a protective moat. Others oppose the N Potomac Crossing because they think it would encourage development in the rural western part of Montgomery county through which it would pass.

Bob Grow of the GWBoT says both fears can be addressed. He notes that already job growth and retailing is growing much faster in NoVA than in MD. 120k northern Marylanders each day go to jobs in NoVA whereas only 40k Virginians go to jobs in northern MD.

“It is already happening,” he says of unbalanced development and he suggests they need to address directly any sources of uncompetitiveness. An enthusiastic supporter of the Maryland Inter-County Connector (ICC), Grow says this planned toll road is essential to tie northern MD together better and to link the I-270 corridor of Montgomery county to Howard and Prince George’s counties and to the BWI airport area. He thinks some kind of new link between VA-7/VA-28 Dulles VA and the I-270 Corridor will come to be accepted by both sides as beneficial. He says such a TechWay (as it has been dubbed) need not have any interchanges or station stops between the Potomac River and I-270 in order to preserve the rural character of the east bank of the river. Zonings and limited right of way for future ICs should address Montgomery co preservation concerns.

The ICC has been recommended as a 2x2-lane variably priced, managed flow toll road (TRnl#36 Feb 99 p1) by a Transportation Solutions Group set up by the Maryland Governor Paris Glendening with former TRB head Tom Deen as its chair. Montgomery and PG county executives strongly support the new road as essential to the economic health and quality of life of their two counties which have 1.5m pop in total. Environmental impact assessment was halted 18 months ago after the USEPA region issued inflammatory comments about the road, which had indeed not been designed to minimize impacts on a number of wild creek valleys through which it passes. Failure to tackle upfront how to take the road through natural reserves without harming flora and fauna was a serious mistake by the MDOT.

Advocates for the ICC, a local pro-road group says the concerns of anti-road groups can be met with “top-down” construction of slim and high bridging through the wild areas. And they point out that on its original master plan alignment the road is in a developed area and indeed entirely within a priority zone approved by the state planning office for infrastructure support under the so-called ‘smart growth’ program. So it cannot validly be opposed as a ‘sprawl’ maker.

Nevertheless as we go to press (9/22) Gov Glendening says he only wants to build deadend spurs of motorway standard in from I-270 eastward and from I-95 westward leaving a 12.4km (8mi) gap (MD-97 to US-29) in the middle of the 27.7km (16.9mi) route! (The Gov is angling for a position in an Al Gore admin, they say.) An intriguing thought is that Montgomery and Prince George’s counties, which strongly support the full road, and in whose territory it is located, could sponsor it as a bi-county toll road and request private sector proposals.

In northern Virginia lipservice is given as regularly as anywhere to the politically correct homily that “we cannot build our way out of congestion.” It is a kind of incantation to the environmentalist gods, a ritual affirmation of respect for the ecofaith, and once gotten done with, discussion immediately turns to what needs to be built in order to relieve congestion. For as one reporter wrote: “They say that if you build it, they will come. But in northern Virginia, they’ve already come - booming businesses and new residents by the hundreds of thousands...” (WASH POST Sept 5, 99 pC2) Almost everyone wants to build in NoVA. The arguments are over what to build and how to fund it.

Trains Still Reign in Ole Virginny

Rails have a powerful grip on the imagination in these parts, at least among those who show up at “public meetings” and among those who plan. They had much “public outreach,” a citizens advisory committee, surveys, and four public meetings for NOVA-2020. The draft report says the “public” wanted the new plan to “be more than a highway plan” and that they thought “more emphasis should be given to public transp” and “serious consideration given to rail.” Representing the 1.8m citizens of NoVA were a grand total of 102 people who attended those meetings (p36), each public meeting attender apparently representing 17,647 citizens, an attendance rate of 0.0057 percent. But the planners listened to the people’s pleas for rail.

In addition to $3802m of highway projects by 2020, the planners propose (see table) seven rail projects totaling 135km (84mi) of new rail costing $5,754m, almost half again what is to be spent on the roads which cater to the overwhelming majority of trips in the region. Moreover of transit expenditures proposed only $86m goes to bus or 1.5% of the total. Over 98% of the outlays go for the rail mode though it caters to a minority of transit riders in NoVA.

The 2020 plan adds 504 lane-km or 6.6% to the base 7666 lane-km highway system for $2,952m at $5.9m/lane-km ($9.4/lane-mi) and builds or rebuilds 44 ICs for $850m or nearly $20m ea. ICs usually double per-lane throughput, so these will benefit about 500 lane-km so a rough guess would be that road capacity is being increased about 10%. A Long Range Plan already on the books adds about the same so plans are for an overall increase in capacity of around 20%. That compares with a projected growth of total commute trips of 64% (p83). Worse gridlock looms!

But trains are being mobilized to come to the rescue!

The rail in NoVA-2020 is 135km (82mi) of route, 73km of metro route and 61km of trolley route, at a cost of $27m/track-km for the metro rail and $15m/track-km for trolley at grade. There is presently 47km of metro rail in NoVA, so the plan increases that by180% or 2.8 fold. VA has no rail-lite, so the addition of 61km of trolley weighted at half the capacity of metro rail makes the combination 145km or a 3.5 fold increase in capacity. The area has 130km of commuter rail (VRE) - not proposed for expansion though several extra stations within the existing mileage are proposed.

CRITIQUE: Such is the faithbased belief in rail that the proposed new lines have been adopted in the plan even though few of them have been specifically studied for ridership, their costs compared with benefits. Neither has rail been studied seriously for cost-effectiveness versus other modes.

The rail projects are quite duplicative of one another which will cause them to poach one another’s clientele. The Orange line extension to Centreville is close to the existing Manassas VRE commuter rail, just as the extension of the Blue line to Lorton runs all the way alongside the VRE commuter rail from Fredericksburg. Two of the three new light rail lines closely duplicate metro rail.

The proposed Purple Beltway metro line is a circumferential route that has never been found in studies or in trials to be likely to attract substantial ridership. A 15-min frequency rush hours/ otherwise hourly Tysons VA to Bethesda MD trial express bus service on the Beltway currently carries just under one thousand passengers/day. And this is considered the most promising Beltway leg for transit. A Beltway rail line to the Bethesda area would involve 8km extra in MD on top of the 11kmwithin VA to the state line for a total cost of $1275m on the NoVA-2020 cost numbers. For 4k or 5k/day riders?

As for the $1947m Dulles rail, this was studied (KPMG Peat). The report said that express bus (TRnl#34 Dec 98 p1) could operate “at a fraction” of the cost of rail. Combining express bus/van/cab into HOT lanes would cost even less. The idea that any substantial number of airport passengers will use a metro rail line with 19 station stops to the center of Washington and bring luggage up and down the world’s longest escalators is fanciful. One study (a Parsons Brinckerhoff work with arithmetical errors and some high growth assumptions) cited in support of Dulles rail suggested that without mode shift to rail up to 22-lanes would be needed compared to the present 12. But that MIS failed to consider (1) the now planned upgrade of parallel VA-7 to mwy standard (2) the potential of a North Potomac Crossing in relieving the Dulles Corridor and the Beltway (3) the use of movable barrier to match capacity to directional flow (4) the use of variable pricing to manage traffic.

NoVA-2020 justifies this bonanza of new rail starts by producing a matrix of trips, projecting higher densities along corridors, projecting car travel times and assuming transit travel times, then running this through a standard gravity model that distributes trips by relative travel times. The model predicts major mode shifts in favor of transit (rail plus bus) for commute trips – from 8.4% to 13.1% (p116). An assumption fed into the modeling that seems highly questionable is the speed assumed for metro rail – 60mph. This compares with an assumed 30mph for HOV and 45mph for express bus in the modeling. The HOV and express bus speeds nominated are to account for rideshare/bus boarding time and waiting times, according to notes (p61). The Washington area metro in fact rarely reaches a maximum 60mph and operates currently at 30 to 36mph from door-close to door-open, the lower being the rush hour speed applicable to commute trips. Wait times at the station will add about 5mins/trip and home-station and station-work times bring the overall trip speed down to around 20mph. Typically in America rail transit trips are slower door to door than private car trips driven even in congested conditions, which is one explanation for why the mass of American commuters endure congested rush hour travel even when there is a train. So the unreal 60mph assumption skews the model toward rail.

A further bias is that the NoVA-2020 model revolves entirely around trips to work, which are most susceptible to mode shift. The latest Nat Personal Transp Survey confirmed earlier findings that a minority now of rush hour trips are trips to work. The majority are other trips – people in work vehicles, freight, people on personal business, shopping etc. These trips are far harder to move to rail transit from cars than the commute trip.

This plan for the NoVA subregion of the greater Washington area is at odds with the Metro Washington Council of Govts in assuming a big mode shift to rail from private cars. The National Capital Region Congestion and Mobility Task Force recently reported (p9) the share of transit is most likely to continue to decline slightly. They pointed out that it is especially costly and difficult for transit to serve suburb to suburb trips which are the ones growing fastest. And it is more difficult for rail transit to serve these trips than road-based transit, which makes the Beltway metro and the VA-28 trolley especially dubious projects.

The NOVA-2020 plan projects an additional 81k commute trips/day attracted by the seven new rail lines and a number of far more economical bus projects, all within the capital program of $5,860m. If that ridership were to be achieved then each extra daily work trip on transit will cost $73,000 in capital investment and a per trip subsidy of about $2 on the increase in operating costs of $131m/yr.

By contrast the road improvements costing $3,802m and extra annual operating costs of $5m/yr will serve an extra 640k daily non-transit commute trips for about $6,000 each in capital investments. And there will be a per trip subsidy of 1.3c. (calculated from #s p83)

Even if rail achieves its big gain at the expense of road in NoVA, and if transit goes from 8.4% to 13.1% of commutes, transit is costing 12-times the capital per extra trip that providing an extra road trip road costs ($73k vs $6k) and 150-times the annual operating subsidies ($2.00 vs 1.3c). If rail gets only half its projected ridership – a common occurrence – then the gap is doubled to well over 24-times the per trip capital cost and more than 300-times the per trip subsidy of road.

Bob Grow of the Board of Trade says the plan is “too heavy on rail, too light on road.” He puts it too mildly. The rail projects will burden the Virginia economy whereas the road projects are at least an order of magnitude cheaper to implement per trip served. Some of the road projects may also be self-financing with tolls, whereas there is no attempt to make rail pay its way. VA-7, VA-28 and a Northern Potomac Crossing look readymade for full tolling with perhaps some express lanes on I-66, the Beltway and the Wilson bridge?

Governor Gilmore who has resisted the shrill demands for taxes to fund this rail-heavy program should cast a critical eye on these rail projects, and get some traffic and revenue studies done on how much road expansion can be funded by toll revenue bonds. (Contacts Kenneth Klinge 703 683 3279, Steve Read VDOT 703 383 2216, Bob Grow, Bd of Trade 202 857 5935, Wayne Lemmon 202 298 6712)