-DEMOGRAPHIC DATA
The MIS1 population growth projection is outdated and its projections of population growth, and especially home based work trips are exaggerated. They are based on the estimate from MWCOG round 5.1, which in turn reflects population growth rates that occurred during the 1970´s and 1980´s.
Census data for the 1990-1995 period estimate that the US population grew at less than 1% per year. For Virginia, during this period the population growth was 0.9% per year. For the Northern Virginia area the population growth was less than 1.3% per year. And for Prince William County the growth averaged less than 3% per year.2 In all cases the growth rate declined significantly during the period. For Prince William County the decline is particularly striking. The estimated annual growth rate in 1988 was 6.19%, an historical high, but had declined to 2.7% in 1995. By the end of 1996 it is expected to have fallen to as low as 2%; a total decline of 68%. These trends will continue. The 1980 growth rates that peaked in 1989 and declined thereafter are well known to be a result of the "Baby Boom," a highly transitory event.
Based on reasonable extrapolations of the National, State and Local trends, it is probable that the growth rate will decline over the new decade to a value near 1%. This would be consistent with the present birth rate and other data. The MIS traffic forecasts assume a growth rate of 2.7% through the year 2020.
Reasonable projections population increase in the study area in the year 2020 (compared to 1995) ranges from a low of 20% to a high of 35%. The comparable value used in the MIS for the period is 87%, a large overestimate. The erroneous data was pointed out in a letter from EPA to VDOT in Dec. 1995.
Even using the more recent growth rate data may overstate the case because it does not include the effect of Federal Government downsizing, the ripple effect leading to corporate down-sizing, the growth in tele-commuting and other recognized and predictable traffic limiting effects.
-NORTH-SOUTH TRAFFIC NEEDS
According to the MIS, 91% of the traffic in the study area is either east-west or radial (for example, from the north-west zone to the central-central zone), and 14% of the 91% is all the way to DC. The average number of daily trips on this new road in 2020, assuming the optimum alignment, is 37,000 per day, based on the outdated demographic data used in the MIS. The more likely value for total north-south trips in the three western zones in 2020 is perhaps 20,000 trips per day.
Of these trips, the great majority (93%) are short trips. Fewer than 2700 trips per day would use a large fraction of the road (Leesburg/Sterling/Dulles area to I-95/Fredrickesburg trips.) Using more realistic demographics, the number of long distance trips is likely to be less than 1500 per day.
The data in the MIS do not support the need for a major north-south highway, particularly in view of the large east west capacity provided by the Prince William Parkway (which is the most important function of the southern segments of the WTC) and the large north-south capacity to be provided by the soon to be completed 234 By-Pass and widening of the portion of Va 234 south of the 234 By-Pass. Completion of the long planned Va 28 By-Pass would provide substantial additional north-south capacity in an area that could make optimal use of it.
-TRAFFIC MODEL FLAWS
The MIS states that there is a net decrease in I95 traffic between the WTC and the Prince William Parkway (PWP) of about 13,000 vehicles per day. But traffic on the PWP, VA-234, and Dale Boulevard are unchanged3. This is a substantial discrepancy and casts doubt on the validity of the model and any projections make by it.
For several intersections, particularly on roads between Prince William and Fairfax Counties, the model calibration runs generate several thousand vehicles, while in reality, traffic counts give a "handful" of vehicles. On numerous other roads the model generates traffic that is a factor of two higher than actual traffic counts. These large errors and major discrepancies such as the I95 traffic evaporation result cited above seriously challenges the validity of the model projections used to support the need for the WTC.
According to VDOT personnel4 involved in using this model (It is the "famous" Dulles Corridor Model) the Manager of the Project responsible for Model development for the I66 MIS found so many flaws and impossible results that he proposed to VDOT that the Model be abandoned as unworkable. Nevertheless this model, that cannot be calibrated, and that gives highly inaccurate trip table results that vary from actual traffic counts by 200% to as much as 800% was used to justify the need for the WTC. Now that some of the sources of the problems in the model have been identified5, it is clear that much of the traffic forecast to exist on the WTC is entirely the result of the flaws in the model. This Model simply does not function correctly for 80% of the traffic that it must account for - all of the non-work trips.
It is worth noting that this same model with its error producing methodology was used to justify the Dulles Greenway. As is well known the model overstated the potential traffic and failed to adequately account for the impact of the toll fee on the traffic volume.
-TRAFFIC CONGESTION RELIEF
Traffic reduction on major highways is inconsequential. Quoting from the MIS "There is a small reduction of daily vehicles on I-66 just west of Route 28 of up to 5000 daily vehicles." This is, in fact, an inconsequential reduction of 3.5%, and is effective for less than 6 miles.
Again quoting from the MIS, "The amount of traffic reduction on I-95 north of the Prince William Parkway, is fairly minor with no more than 8,500 daily trips or about 5% of the total traffic on I-95."
Even these small traffic reduction figures cannot be fully achieved. For example, the alignment that produces the I-66 reduction of 3.5%, results in less than 4% reduction on I-95. The inverse is true also.
There is no meaningful reduction in traffic on any other roads in the study area, including Route 234, or US 17, or US 15. The reason is simple: in 2020 this new road will carry only 5% of the traffic originating or terminating in the three western zones. (Including the substantial traffic within a zone raises the total from the 91% cited above to 95%.) (There is a moderately large percentage decrease in traffic on US 15, but this is a reduction in an already relatively small amount of traffic - 15,000 trips/day in 2020)
As is obvious to most real commuters, the problem is east-west or radial traffic, and for this problem a north-south road does not help.
The published Draft MIS provides no traffic projections for the Prince William Parkway Va 234 and the 234 By-Pass. The only significant traffic reduction occurs for a short distance on I-95 between the WTC and the Prince William Parkway (21,000 trips per day; that is, an additional decrease of 13,000 trips/day for that segment.) This implies that much of the traffic on the WTC is the result of travelers choosing the WTC rather than the PWP or VA 234 (no one wants to drive even a few miles on I-95). Of the average of 36,000 daily trips projected for the WTC, it appears that about 13,000 trips result from the diversion of trips from VA Route 234 or the PWC. Consequently there should be a matching traffic decrease on the PWC and the 234 By-Pass. Examination of the complete traffic study documentation indicates that there is no change in traffic on these two roads for any alternative. This is a serious discrepancy and casts doubt on the validity of any of the data in the MIS. Furthermore the MIS explicitly states that there is a matching reduction in the traffic on the PWC and 234, a statement that is demonstrably false.
-COMMUTING TIME REDUCTION
The MIS establishes that significant commuting time savings are non-existent. For 95% of the projected trips, time savings never exceed 8 minutes even in peak hours. (In off-peak hours time savings are 3-4 minutes.) Quoting from the MIS "Because of the design of the new facility alternative, only the long distance trips tend to save measurable time...", but the MIS projects that there will be an insignificant number of such long distance trips, probably less than 1000 trips/day during peak hours where time savings up to 24 minutes could occur. (Off peak hour time savings are less than 15 minutes.)
If other reasonable options for improving transportation (for example the Route 28 By-Pass, whose right of way is already public property) are chosen, they provide smaller time savings for a larger number of travelers, making the comparative time savings even smaller. This road has been endorsed by numerous citizens groups as well as the City of Manassas.
-MISLEADING RESULTS FROM ARBITRARY ZONE BOUNDARIES
In the MIS the boundary between the NW zone and the NC zone is the Loudon-Fairfax County Line. This has the effect of putting Dulles in the NW zone, and exaggerating the need for the new road by converting east-west or radial traffic into north-south traffic. If the boundary were further west, a substantial fraction of the traffic would become radial or east-west rather than north-south.
Traffic on the 28 by-pass and Segment 13 combined (Tri-County Parkway) north of I-66 is virtually the same as the new western corridor, dramatically illustrating that the real traffic needs are in the developed areas and not in the western part of the study area.
Similarly, the so-called South-West zone extends to the Potomac River and includes the Quantico Marine Base. These zone boundaries have the effect of causing a substantial amount of East-West traffic to appear to be North-South traffic. This is a mis-leading approach and does not engender confidence in the objectivity of the contractors or VDOT.
-OTHER MIS SHORTCOMINGS
The MIS did not include the anticipated substantial impact of the transportation improvements that could result from the I-66 corridor MIS, the Rt. 29 Corridor Development Study and the Fredrickesburg By-Pass.
The impact of a mass transit component has not been studied, particularly a Metro rail line to Dulles, which has been projected to have a major impact on traffic on roads serving Dulles.
The MIS does not address the crucial issue of compatibility of the proposed roads with land use plans and particularly does not address the major issue of the impacts of increased development pressure on farmlands, woodlands and wetlands.
The only conclusion that can be derived from the MIS is that any WTC alternative (new facility or linkage/upgrade) simply does not meet any real needs, would have large negative impacts on numerous communities in the study area, would create substantial pressure for development that would violate existing long range land use plans, and would divert a very substantial amount of funds from transportation improvements that would meet the real needs of people living in developed areas.
-DULLES VEHICLE TRAFFIC
A prominent WTC proponent has claimed that in 2020 there will be 120,000 vehicles per day into Dulles, a 427% increase. If it is assumed 45% of these trips are passenger trips (the MIS Purpose and Need, Section 6 page 7) and that half the vehicles pick up One traveler and the other half deposit One traveler per vehicle, then it follows that to accommodate these 54,000 travelers there must be at least one 200 passenger Boeing-757 passenger planes either landing or taking off every 2.2 minutes continuously for 10 hours. Of course the actual number of passengers per vehicle is greater than one, and hence the number of aircraft must be increased accordingly. This surely is an exaggeration of the potential increase in passenger traffic at Dulles, must imply a similar exaggeration in the service traffic as well.
It is claimed that less than half of the above 120,000 vehicles are from the "traditional eastern direction". This statement implies that more than half will come from the western direction. Such an implication is demonstrably false, particularly as it pertains to justifying the Western Corridor. The projections of traffic on the new WTC in the region of Dulles Airport range from 30,000 to 40,000 trips per day. Of these, a fraction will exit to Dulles using Rt. 50, the Greenway or the (presently mythical) Dulles back entrance. There are no data in the MIS that allows this fraction to be calculated. It can be crudely inferred from differences in traffic south of Rt. 50, south of the Greenway and South of Rt. 7. I infer about 15,000 trips per day to Dulles. But this estimate ought to be reduced if more realistic population growth rates are used. About 10,000 vehicles per day is a reasonably generous estimate for trips from the west in the year 2020.
Summarizing: Trips to Dulles originating west of Dulles most probably will not exceed 10,000 per day in 2020; The total of 120,000 trips per day is an overestimate because it uses both exaggerated demographic projections and optimistic projections of growth in Dulles passenger service. Regardless of population growth projections, traffic into Dulles, originating west of Dulles in the three western zones (excepting the eastern end of Loudon) cannot exceed 12% of the total, and this small fraction will be easily accommodated by I-66, VA 28 (to be enhanced by grade separations at all intersections) and US 15.
The remainder of the traffic (88%) originates from the Northern sector and Eastward in a semicircle terminating to the South including Manassas. Much of this traffic would be abundantly well served by the VA28 Bypass. If the Tri-County Connector segment from the 28 Bypass to the Greenway or US 50 is constructed this traffic would be over-served.
-TRUCK TRAFFIC ON THE WTC
Truck traffic into Dulles Airport originating from I-95 that would use the WTC is minimal. According to the MIS the total long distance traffic is expected to be less than 3000 (one-way) trips per day. Using the standard ratio of passenger cars to trucks, the number of trucks would be less than 150 per day from the I-95, Route 17, Fredricksburg area to the Dulles, Sterling, Leesburg area (less than 10 per hour). If half of these trips terminate at Dulles, then there will be about 75 long distance truck trips per day into Dulles.
If, as has been claimed by Dulles Airport spokesmen, the WTC is needed to support Dulles Freight traffic, and if freight traffic grows by 6% per year, truck traffic will increase 574% by 2020. In 1990 freight tonnage was 211,061 tons per year. (The MIS erroneously says 211,061 tons per day.) In 2020 it is projected to be 1,212,227 tons per year or 3,321 tons per day. If large trucks typically carry approximately 30 tons of cargo, there will be 110 trucks per day into Dulles from all directions. Perhaps one quarter to one half of these trucks would use the WTC. Only a fraction of the trucks using the WTC would make use of its full length, with a majority using only the segment north of I66, and these have clearly defined alternatives such as Va 28 or the Tri-county Connector. Combining these figures the number of truck using the WTC would be no more than 50 per day and probably no more than about 25 per day in 2020. As few as 25 trucks per day seems to be meager justification for building a major highway.
On to Part 2: Revised Cost Estimate
Return to Report Summary Page
Return to The List
2. These data are from the US Census Bureau, the Virginia Population Tracking organization and Prince William County Planning Department.
3. Technical Memorandum from Robert W. Wilson, Director of Public Works for Prince William County, to Lawrence D. Huges, Acting County Executive; November 21, 1996. This memorandum documents that the traffic analysis shows no difference in traffic volumes on the PWP, VA 234 or Dale Boulevard whether or not the WTC is built, despite the fact that in all of the build options, there is a reduction in traffic on I95 of at least 13,000 vehicles per day for the segment between the WTC and the PWP. For this situation to be true, these 13,000 vehicles would have to exit onto the WTC, take some route that would put them onto the PWP whence they would return to I95.
4. I-66 CORRIDOR MIS, Technical Advisory Committee Meeting, March 20, 1997. Statement of Phil Hopkins of VDOT referring to a conversation with Jeff Bruggeman of KPMG, the technical manager of the Model Refinement project, that occurred in January of 1997. A major revision in philosophy occurred following this conversation and the recognition that the model probably could not be calibrated in a manner that would result in reasonably accurate trip tables for trips other than home based work trips.
5. The Dulles Corridor Model was intended to be a much more sophisticated and , in principal, more realistic approach to traffic forecasting. It is based on the "composite impedance" approach for distributing trips on the road network. The composite impedance is a procedure in which many factors, such a time, distance, cost, etc are combined to produce a single factor (the impedance) which determines the likelihood that a particular trip will occur. For home based work trips, this concept appears to work well, probably because there are few options for the traveler; for example, the possibility of going to an alternate destination is negligible, the cost of parking appears to be a minor factor, and most other factors are largely inconsequential. For non-work based trips, the factors that seem to apply are largely unaccounted for in the model, and those that are used seem to be not very significant to the traveler. Furthermore, there seem to be a fatal flaw in the concept, because small changes in the weights applied to the various factors and constants cause the traffic estimates to vary substantially, and there is no means of deciding which possible result is the correct one if both are equally consistent with observations.
1. All references to the Western Transportation Corridor Major Investment Study are to the Draft Documents published October 3, 1996. In particular this analysis makes use of data published in one or the other of the Technical Memoranda: Purpose and Need or Traffic and Transportation, prepared by Parsons Brinckerhoff Quade & Douglas, Inc. Quotations are from Technical Memorandum: Traffic and Transportation.