R.C. Harkness, PhD Urban Systems Planning

Feb. 7, 2001

Summary

The process of addressing Puget Sound’s transportation problems has been faulty for many years and Sound Transit’s light rail plan is simply one faulty result of that faulty process, an unfortunate end result as it were. Today’s core issue is whether to implement the results of that faulty process or whether to place rail on hold, amend the process and honestly investigate other alternatives.

The reasons the light rail plan is faulty are numerous but the primary one is that it achieves very little for its enormous cost. There is evidence that it actually has a negative cost benefit. Although marketed as an answer to traffic congestion light rail will have miniscule impact. When marketed as an alternative to driving or a way to remove cars from the road it seems anything but cost effective. Strategically, committing this much resource and political attention to a non-solution for the regions transportation problems may preclude the Puget Sound region from being able to implement more cost effective alternatives for many years.

The reasons the planning process is faulty are also numerous. The first key problem is that a clear objective (such as reducing regional travel delay) has never been consistently applied as the primary measure against which any and all alternatives are judged or compared. The second key problem is that a full range of alternatives has never been investigated, while those that were investigated were not investigated honestly and/or thoroughly. Of primary concern here is the fact that the current light rail plan has NEVER been evaluated against other alternatives. The most recent treatment of light rail in an alternatives analysis was when the Regional Transit Project (RTP, the predecessor of the Regional Transit Authority that renamed itself Sound Transit) considered light rail proposals from various citizen groups in the ’92 timeframe and dismissed light rail as inadequate (the RTP favored modified heavy rail at the time).

Nevertheless there is indication that rail in one form or another was always the inside favorite throughout this multiyear process, while public hearings, alternatives analysis and outside criticism were all viewed as obstacles to be ignored or overcome. If heavy rail wouldn’t sell, our transit agencies would swallow their pride and sell light rail.

Since Sound Transit has never done proper alternatives analysis to show the relative merits of their light rail plan, the plan should be put on hold while the proper studies are completed. Federally mandated transit planning processes may have been violated thus voiding the FFGA. This paper concludes with some specific recommendations.

The history of alternative analyses relevant to Link light rail has been summarized in Sound Transit’s short white paper called "Transit Technology Review" dated 2/2/99. This author’s critique of these so-called alternative analyses comprises six main points:

As noted in the above summary report light rail as a technology was considered along with other transit technologies in PBKE’s August 1992 study. However the 21 mile light rail system Sound Transit hopes to build was never compared to other systems that could accomplish roughly the same objectives.

In the last major alternatives analysis, documented in RTP’s 1993 FEIS, light rail was not even one of the alternatives studied. What RTP studied (as their "rail/TSM alternative") was a rapid rail system running 100% on a grade separated right of way, capable of speeds up to 70 MPH, averaging 35 to 40 MPH, operating on as little as 90 second headways, and with a capacity of 22,400 persons per hour per direction. (See RTP’s 1993 FEIS pages xxxii, xli, 2-24, 2-58) RTP emphasizes throughout their report that this is not a light rail system and says "the rail system that is proposed would fall into the definition of heavy rail" (p xxxii).

The essence of that 1993 study was to compare this heavy rail system with a system of buses on HOV lanes called "TSM", and with a third alternative of buses partly on exclusive busways called "Transitway/TSM". This study was done approximately right in so far as one total system with specific routes, stations, etc. was compared in detail with two alternate systems. Since this study compared bus alternatives against a 125 mile heavy rail system it is clearly not relevant to Link light rail and in no way would seem to satisfy any statuary requirements for project approval contingent on having completed an alternatives analysis.

Table xxxiii in that ’93 FEIS shows that the heavy rail alternative was the most costly in dollars per rider, yet somehow it became the preferred alternative submitted to voters in 1995. It was defeated. Subsequently Sound Transit was formed, morphed the heavy rail into light rail, scaled the system back into what was supposed to be a $1.67 Billion dollar project, and obtained voter approval in 1996.

At no time since the 1993 RTP study has a light rail system with its reduced performance relative to heavy rail ever been re-evaluated against the bus/HOV alternative or the bus/transitway alternative. This is the crux of where the process went astray.

As a result neither the FTA, nor the citizens of Seattle, nor the elected officials have any data to show that the Link light rail is better than bus alternatives, or better than other transit and non-transit alternatives.

The central transportation problem in Puget Sound is traffic congestion. There are many demand side alternatives for reducing transportation demand that might help alleviate this problem, many ways to use the existing systems more efficiently, and other supply side alternatives beside mass transit. To the authors knowledge there has been no comprehensive overview study which considered this broad range of alternatives so the merit of improving mass transit vis-a-vie other alternatives is unknown. Conducting such a study is beyond Sound Transit’s scope or expertise, but it does mean that (lacking any broader study showing the merits of mass transit) there is no assurance that spending money on light rail will continue to receive public support in the future if other alternatives gain popularity while Link is mired in construction problems for the next ten or 15 years. Indeed the political and popular support for Link is based on myth and could evaporate quickly.

It is not clear why the rail alternative was placed on the ballot in 1995 when it was the least cost-effective alternative in the 1993 FEIS. The logic for that transition in thinking is not documented in the FEIS, although there are indications in the FEIS that the RTP thought the supposedly higher capacity of rail in the DSTT was a distinct advantage. (See page 2-58 for instance). Could this have been the key reason for picking rail and ruling out bus solutions? If heavy rail was picked over bus in 1993 because it had more capacity it does not follow that rail would still win if it were light rail that were being compared against bus. In short the rational for favoring rail over bus -in the last mass transit alternatives analysis ever conducted in this region- can not be assumed as being true or relevant for the light rail system now being proposed.

To be more specific, the RTP calculated heavy rail capacity through the DSTT as 22,400 pph in one direction (p.1-10). On that same page they cited max bus capacity as 13,400. They never say they flat out rejected the bus alternatives because of capacity considerations, but they so imply on page 2-58 by tating there is a peak demand for 15,000 pph. In additon, on page xviii and elsewhere they make quite a point out of capacity issues as though capacity is a paramount requirement. Therefore one can suspect, if not prove, that relative capacity in the DSTT was the main reason (or most convenient excuse) used to dismiss the bus alternatives. If so, it appears unfair, unwarranted, and possibly a breach of proper process.

If the IG determines that bus alternatives were dropped because of capacity concerns the IG must determine if the calculated capacities were technically correct and if some specific level of capacity in the DSTT is actually a hard requirement. Here are some considerations.

1. It might not be necessary to have as high a level of capacity in the DSTT as RTP specified because: 1) transit demand may have been overestimated and in any case is likely to be less for light rail than for heavy rail, 2) not all bus traffic need transit the DSTT, indeed thru traffic should probably stay on the freeway, 3) there are parallel NS routes through the city on the streets and on Alaska way that a bus system could divide its capacity among. In short capacity in the DSTT should not be the killer criteria used to rule out bus, monorail or other non-rail solutions.
2. The bus tunnel is apparently running at only a fraction of capacity today. 1) Perhaps bus operations could be better managed so that all the buses that transit the tunnel are actually full (a personal survey this morning showed they are far from it)*. 2) Perhaps buses could be platooned to increase capacity beyond prior assumptions. 3) If the tunnel is only running at half capacity today, as has been reported, is it reasonable that destinations downtown will double in the foreseeable future? What else would double bus origins/destinations in the DSTT?
3. It is highly unlikely that voters will approve an extension of light rail over I-90 thus light rail in the DSTT may never reach the volumes that would help justify the need for rail versus bus.
4. Light rail may not have near the DSTT capacity that the RTP felt needed, and may have even less than bus. Per the ’93 FEIS page 2-50 light rail can’t realistically run at less than six minute headways in mixed traffic, as will be the case on the south line through Rainer valley. At 533 passengers per light rail train at crush loading this means the south line of link light rail will have a permanent limitation of 5330 pph. Can such a low capacity be acceptable for a so-called regional solution serving the whole south end of this urban area? If the trains just carry on through the DSTT to serve the north line as would seem sensible then the capacity of the north line is also limited to 5330 pph in which case the whole system would have a capacity far less than the 13,000 pph capacity of buses. An alternate scenario is that loads to/from the north end are far heavier than from the south and thus two trains running from the U District could interleaf on 2-minute headways with one train running all the way from the U district to the airport. This would create a 22,000 pph capacity thru the DSTT and along the entire north line. But it requires a way to reverse the two north route trains somewhere south of the CBD. Is this the scenario that mandates 22,000 pph capacities through the DSTT? Still another possibility is merging two trains from the eastside, but this scenario seems highly unrealistic and should not become the linchpin rationale for trying to justifying rail over bus.
5. All Sound Transits headway assumptions need verification. The RTP assumed 90 seconds for heavy rail. It is not clear what ST is assuming in the DSTT and in Rainer valley.
6. Many bus trips can be made to downtown without transfers as is quite evident from the many routes converging on the DSTT. With rail many of these riders will need to transfer from bus to rail. Transfers are disliked. Has ST been honest in accounting for this in their patronage modeling and mode split process?
7. About the same effect as light rail could probably be achieved by running platoons of buses on the HOV lanes and thru the DSTT. They could run as frequently as the trains, have at least the same capacity thru the DSTT, and perhaps run as fast along an approximately parallel route on I-5. That option should be modeled. It would be far less costly to implement.

4) Prior alternatives analyses did not consider monorail and group rapid transit alternatives adequately

Monorail, automated guideway transit and group rapid transit were mentioned and rather casually dismissed in the ’92 and ’93 studies. Granted they have some disadvantages, but so does rail. However they have a key advantage that could outweigh all. They have the potential to be far less expensive if they can follow Puget Sound’s terrain and existing corridors rather than require a hugely expensive tunnel. If they can be built at a lower per mile cost we could afford to build more route miles thus greatly increasing ridership. This potential advantage is great enough so that ST should not dismiss these alternatives but rather fund a serious route cost analysis. If indeed the routes are significantly less costly then the other disadvantages mentioned previously may fade by comparison or prove vacuous.

As noted above, capacity may have been one criterion used against these more modern technologies. The IG must ensure this criterion has not and is not being used unfairly. One consideration is that an ideal system is more like a network and less like a single line. Perhaps two north south tending lines through Seattle would make more sense than a single line. The whole system should not be constrained by a single bottleneck if there are workarounds.

To summarize, it is not an adequate alternatives analysis to just to look at the technology in a general way as RTP did in 1993, one must do a systems level comparison which includes laying out and costing specific alignments and looking at one complete systems solution against another. These alternate technologies should have been developed into full-scale alternatives like the TSM, Transitway/TSM, and rail/TSM alternatives were in ’93. Each should have been costed in detail. Each should have been modeled and its patronage calculated. Last but not least the evaluation should be done objectively.

Almost 10 years have elapsed since monorail and other non-rail technologies were judged unproven, etc. Perhaps things have changed enough to answer some of the uncertainties cited in those early studies. Clearly an alternatives analysis completed nearly ten years ago can’t be thought adequate to satisfy Federal requirements.

RTA’s 1993 FEIS did not compare apples to apples in the sense that both the costs and the benefits were variables. Either each alternative should have been proposed for implementation to a degree that would have made their benefits equal, or their costs equal. We are left with questions like how many transit riders would have been achieved if the same amount budgeted for rail had been spent on bus?

The last time light rail was considered in an alternatives analysis was in RTP’s 1993 FEIS. Light rail was not proposed by RTP but rather by two citizen groups to which RTP was compelled to respond. Here is what Sound Transit’s predecessor the RTP had to say about light rail at the time:

"They typically have an average speed of 5 to 20 mph and capacities between 4000 and 12,000 persons per hour" (FEIS p. 2-49)

"Conventional transit practice…suggests that when train frequencies are under 6 minutes, cross traffic on arterials will affected…and grade separation is necessary. Between 6 and 16 minute headways, traffic levels, levels of service on cross streets, and the importance of cross streets to community and emergency services become important criteria for assessing operational feasibility. These constraints limit the capacity of surface LRT systems, as compared to grade separated systems" "… these considerations will reduce system speeds, schedule reliability, or both" (FEIS p. 2-50)

"Surface LRT options were analyzed to the point that it became clear that these (light rail) options did not adequately serve the goals and objectives of the Regional Transit Project." (FEIS p. 2-61)

Since surface light rail is now a key part of Sound Transit’s plan one can only say that Sound Transit is now acting contrary to the conclusions of their own alternatives analysis. This is one specific example of how Sound Transit acts without apparent principle, morphing and finessing at every occasion to achieve their single objective of building some type of rail system regardless of its merit.

Sound Transits alternative analysis is severely flawed. The current light rail now approved for FFGA has never been compared against alternatives. Sound Transits process for alternatives analysis, or lack thereof, has produced a result that appears far from optimal, and it has failed to provide hard evidence that light rail is any better than other alternatives. Either the FTA sanctioned process, which guides behavior of agencies like Sound Transit, should be rectified to improve the alternative analyses process and more clearly spell out the influence that analysis should have on final decisions, or Sound Transit should be sanctioned for not having complied with that process. In any event the current plan should be put on hold until a proper alternative analysis is conducted and the results widely reviewed. Other recommendations are as follows:

1. FTA should encourage the State of Washington or Puget Sound governments to undertake a broad study of alternate ways to reduce traffic congestion and travel delay. The study should consider a wide range of supply side and demand side options.
2. Sound Transit should be required to fund a full-blown analysis of other transit alternatives to their current light rail plan. However, because of an inherent conflict of interest, Sound Transit should have no say in who does these studies or how they are managed.
3. Each alternative put forth for consideration, such as monorail, AGT, various bus based alternatives, and vanpools should be designed by its strongest advocate in order to assure it will not be shortchanged. Each promising alternative should be developed to the level of a full system description with at least as much detail as used to evaluate TSM, and other alternatives in the 1993 RTP EIS. However the cost and performance of each alternative should be evaluated by an objective third party.
4. All alternatives should be evaluated against a common set of criteria using "least cost planning" techniques. Those criteria should specifically include impact on traffic congestion and impact on travel delay.
5. All alternatives should be designed so either their costs are equal or their primary benefits (like degree of delay reduction) are equal. This is necessary for an apples to apples comparison.
6. A public vote on the alternatives should be encouraged to narrow the field. This should be followed by another public vote to actually fund or endorse the selected alternative or combination of alternatives.

Whether or not to build Link Light Rail is the biggest public works decision in the central Puget Sound's history. We need to get it right.

________________________________________________________________________

* The author counted buses at the Westlake station from 8:32 am to 8:52 am today and estimated passenger loading on the 18 northbound and 13 southbound buses that passed during that 20 minutes. The results were that the northbound lane was carrying roughly 1560 pph while the southbound lane carried about 1530 pph. Most buses were lightly loaded. Several were empty. Only 7 were heavily loaded with standees. It is hard to imagine why transit demand through the DSTT would grow into the 15,000 pph range needed to rule out bus solutions. This is of course only a casual observation, but it does suggest forecast demands through the DSTT might be exaggerated.

Return to the Public Interest Transportation Forum home page.

Last modified: October 21, 2008