California HSR should use Tejon Pass

The current California HSR plan has significant flaws; a major mistake was selecting a detour via Palmdale instead of the direct route via I-5 across Tejon pass to connect Bakersfield to Los Angeles. The cost of the Palmdale route -- about $32 billion in current dollars -- is more than twice that of the direct link, and nobody wants to pay for it. The HSR project will be restricted to the Central Valley until the less expensive Tejon route is embraced.

Saturday, September 8, 2012

Why California's HSR project is failing

Note: This posting was published in the Sacramento News & Review (Sept. 13, 2012) as a letter to the editor in response to an article in a recent issue of that journal.

In his article on the California high speed rail project ("The right track," 9/6/12), Max Pringle hints at why this effort is in deep trouble and unlikely to get built in its current guise: it is simply the wrong rail project for this state. The studies that led to the passage of Proposition 1A in 2008 were rigged to satisfy specific parochial concerns, foremost an alignment across Pacheco pass to serve San Jose and then up the Peninsula to terminate in San Francisco. The resulting system is at least twice as expensive and less useful than a sensible, more direct alignment utilizing Altamont pass in the north and Tejon pass (I-5) in the south. The cost of an initial system linking the Bay Area to the Los Angeles Basin should have been capped at $30 billion, which the current plan cannot deliver.

The current effort is almost certain to collapse within a couple of years: the initial segment between Bakersfield and Fresno has become politically toxic at the national level ("a railroad to nowhere"), and Congress is unlikely to follow through on appropriating federal funding for it. Perhaps this eventuality will allow a less expensive alternative to be pursued. In particular, the California HSR system should be built from south to north, starting with a link across Tejon pass from the L.A. Basin to Bakersfield. This segment alone generates over half of the full "Bay to Basin" ridership, at a cost of less than $12 billion. Moreover, it provides the missing rail link joining northern and southern California. No other 100-mile initial segment comes anywhere close to this performance.

Tuesday, July 24, 2012

Building an HSR spine for California

This post appears in the RailPAC Weekly E-Newsletter, July 30, 2012.

California can complete the North-South spine of its HSR project by 2030 without voter approval of further State bonds, but only if it completely revamps the system currently being pursued by the CHSR Authority. The Authority has chosen an extremely bad route, which is longer, slower, and vastly more expensive than the direct route via Tejon and Altamont passes.

Proceeding beyond the first segment from Madera to Bakersfield, the next step is to complete an initial operating segment (IOS) by building the second segment southward, from Bakersfield to Los Angeles via Tejon pass. The cost of this segment, based on the 2005 EIR, is around $10 billion in addition to the $6 billion being spent on the first segment. With 50% federal support, the State would have to draw an additional $5 billion from Prop 1A bonds. Ideally, the IOS should be extended to Anaheim as inexpensively as possible, using dedicated tracks in a rebuilt (and fully grade separated) Metrolink right of way.

Building the IOS would use up nearly all of the Prop 1A bonds, and other sources of support are needed to continue construction northward. Fortunately, the IOS (and successive extensions) are expected to generate significant net income from fares, sufficient to support well over $5 billion in revenue bonds. Together with federal grants, another $10 to $15 billion in additional construction is possible. This will cover the cost of the section from Madera to Oakland via Altamont pass, estimated to be $10 billion in the 2005 EIR. This will complete the spine of the HSR project, running from Oakland to Anaheim. The total cost is well under $30 billion (current dollars).

Once the spine is operational, branches to San Jose, Sacramento, and Riverside, and an extension to downtown San Francisco can be built when justified and financing is available. But it is premature to do detailed planning for these extensions, since their construction cannot possibly occur until after 2030. Furthermore, most of the extensions can be built at standards far below the 220 mph design speed for the spine, and may even involve track sharing with high performance local rail service.

Comparisons of proposed spine (including an extension to S.F.) with the CHSR Authority's current plan:

Distance (LA-SF): 420 miles vs. 450 miles
Run time (LA-SF): 2:17 vs. 2:38
Capital cost (LA-SF): $32 B vs. $62 B (!!!)

References:

Final Program EIR/EIS Report: Volume 3: Appendix 4-C, "Capital Cost: High Speed Train Alternative" (CHSR Authority, 2005)

Revised 2012 Business Plan Source Documents: "Cost Changes from 2009 Report to 2012 Business Plan" (CHSR Authority, 2012)

Comment by Noel Braymer, Newsletter editor:
I don't think that the current plan by the California High Speed Rail Authority is written in stone and I expect more changes will happen to it over the years. RailPAC's priority now is to push for a new rail connection between Palmdale and Bakersfield and get passenger rail service between Southern California, the San Joaquin Valley and the Bay Area in the next few years. It is unlikely that the Tejon Pass will be used for rail service in the near future. ACE is planning to expand service between Merced and San Jose in the next 5 years over the Altamont Pass to connect with the San Joaquins. Few details are available of what route or how frequent the service will be for the new Altamont pass ACE service.

Saturday, July 7, 2012

California HSR should use Tejon pass

Before the CA state senate vote on July 6th I thought that the federal stimulus grant for HSR would be better spent in the Los Angeles Basin, but after the vote I've changed my mind. It's better to have a somewhat over-built segment from Madera to Bakersfield that can run at speeds close to 200 mph than to have a segment centered on LAUS that may be limited to 150 mph or possibly only 125. This sort of practical limit on speeds in urban areas was acknowledged by the CHSR Authority in its recent business plan, which recommends operating high speed trains both in the L.A. Basin and in the Bay Area initially on the same tracks as conventional passenger trains ("blended" operations). The additional $1 billion-plus authorized by the state legislature for blended operations in L.A. will cover a large portion of the costs of this project, particularly if it is matched by local (Metrolink) money.

For this assignment of federal funds to the San Joaquin Valley to be effective, the Authority must adopt the Merced to San Fernando scenario for Phase 1 of HSR, giving priority to the Tehachapi crossing (per RailPAC's adopted position). But there is a serious flaw in Prop 1A, which mandates a detour through Palmdale and across Tehachapi pass. The cost of this alternative has now gone through the roof: about $16 billion for Bakersfield-Palmdale-San Fernando compared to a $7 billion estimate in the 2005 EIR! Presumably, the cost of the I-5 (Tejon pass) alternative has not seen so dramatic an escalation, but the Authority is apparently forbidden by Prop 1A to revisit this option. The legislature needs to clarify Prop 1A to allow Tejon pass to be considered as an alternative to the now prohibitively expensive Palmdale routing.

This small change to Prop 1A is absolutely essential in order to move forward with a relatively inexpensive Phase 1 using Prop 1A bonds. The cost of the I-5 alternative presented in the 2005 EIR was about $8 billion (current dollars), which is sufficiently small to legitimately ask for half-share federal funding in a single grant for a project of this importance. With improvements to the Metrolink/Surfliner corridor in the L.A. Basin, Phase 1 of HSR construction would effectively provide an initial operating segment for HSR from Merced to Anaheim, a distance of over 300 miles!

References:

Final Program EIR/EIS Report: Volume 3: Appendix 4-C, "Capital Cost: High Speed Train Alternative" (CHSR Authority, 2005).

Revised 2012 Business Plan Source Documents: "Cost Changes from 2009 Report to 2012 Business Plan" (CHSR Authority, 2012)

Monday, July 2, 2012

High Speed Rail should start in the L.A. Basin

Portions of this post appear in RailPAC Weekly E-Newsletter, July 9, 2012.

Most of the federal rail stimulus money awarded to California ($3.3 billion) should be shifted from the San Joaquin Valley to the urban areas -- chiefly the Los Angeles basin. NARP opposes this shift, calling construction of HSR in urban areas mere "enhancements of conventional infrastructure" (Hotline #765 -- June 29, 2012), but this position ignores the evolution of how to implement about HSR in California's urban areas. The original California proposal for federal funding included a request for a $2.2 billion grant for HSR between L.A. and Anaheim, matched by another $2.2 B of Prop. 1-A bonds. The CHSR Authority now plans to substantially upgrade existing urban rail corridors, so that (for example) high speed trains in the L.A. basin can share tracks with Pacific Surfliner and Metrolink services. The revised proposal would implement this "blended" system, at less than half the cost of full HSR. This plan would provide immediate benefits to the most important rail corridor in the U.S. outside the Northeast Corridor, and prepare the way for an HSR system operating across the Tehachapis to Bakersfield (preferably via Tejon pass).

The combination of a new rail crossing of the Tehachapis with rail improvements in the L.A. Basin is the single most important rail project in the entire U.S., generating more new intercity rail ridership than is current carried on the Northeast Corridor (about 15 million per year, compared to 12 million). The total cost of this projec, extending from Bakersfield to Anaheim (or Irvine), is around $15 billion, well within the funding capability of Prop. 1-A bonds together with matching federal grants. And it far outperforms any other option for HSR in California at a comparable cost. The bottom line is that the Tehachapi crossing is the single indispensable segment in a viable north-south rail line through the San Joaquin Valley, and details of extensions in the Vally should be deferred until the initial segment is in place. In the meantime, existing rail services at both ends of the HSR segment will serve as feeder lines with the possibility of offering run-through services such as Sacramento-San Diego.

One reason for initially limiting HSR in California to a relatively short system is that there are a large number of other rail projects that must also be carried forward, such as modernization of the Caltrain and ACE corridors in the north. Rail planners now think that "Super ACE" is the most important passenger rail project in the north, even if built to accommodate diesel operations at a maximum speed of 110 mph. Moreover, Super ACE can potentially provide San Joaquin service from San Francisco to Bakersfield at a run time of around 4 hours (average speed of 75 mph). But building HSR in northern California in advance of the Tehachapi crossing is almost a complete waste of money.

More broadly, the California HSR project as defined in Prop. 1-A is effectively a dead letter because of ballooning costs. The total cost for the entire 800-mile project now stands at some $150 billion and the $9 billion authorized by Prop. 1-A cannot realistically leverage more than a minimal initial segment. But the planning effort has been useful in some regards: we now have a more solid appreciation of the costs, ridership, and benefits of improved passenger rail in California, and the realization that it is impossible to effectively plan and finance this large of a project in the scattershot fashion allowed (and even encouraged) by mandates written into Prop. 1-A.

Building an anchor in the L.A. Basin is clearly the best place to start the California HSR system, since it provides immediate benefits (improved Pacific Surfliner and Metrolink services) that a stranded HSR line in the San Joaquin Valley cannot come close to matching. By starting in the L.A. Basin, the state can legitimately pursue federal funding for the Tehachapi crossing, using what is left of Prop. 1-A bonds for the local match. This is probably all California can afford to spend on HSR in the next 20 years, but coupled with greatly improved San Joaquin service (and possibly run-through trains to Los Angeles) it provides a clear goal that is much more likely to succeed than the dim prospects for the current, unfocused efforts.

Comment by Noel Braymer, Newsletter editor:
Since we can't take taxpayers money for granted we need to get as much out of what we get and serve as many people with it as possible. Each time we have a winner for rail service the more likely we will have support to build more improvements.

Thursday, May 17, 2012

Amtrak and FRA Plans for Improvements to the Surfliner/Coastal Corridor

The Amtrak Pacific Surfliner is currently the premier intercity train service in California, and will continue to play a major role even if high speed rail (HSR) is implemented through the Central Valley. Much of the trackage in this corridor is under the control of two regional rail agencies (Metrolink and Northern [San Diego] County Transit District), which are coordinating with Caltrans on infrastructure improvements that will benefit both Surfliner and regional rail services. Caltrans is also planning to reinstate Coast Daylight service between Los Angeles and San Francisco, aiming to provide two round trips per day by 2025. This memo will review planning efforts in the California coastal corridor over the past 15 years, concluding with a recommendation that implementation of Coast Daylight service be pursued much more aggressively.

Amtrak and LOSSAN Plans (2001, 2007)

A dozen years ago Amtrak sponsored a study to identify possible improvements to the California costal corridor and their effect on run times, level of service, and potential ridership. With an investment of $6.61 billion, this study forecast significant reduction in run times for the existing Surfliner service and limited resumption of Coast Daylight service (Los Angeles to San Francisco) at two round trips per day. Much of the investment was concentrated between San Diego and Los Angeles, targeting a run time of under two hours. Ridership was estimated at over six million passengers per year. See Table 1 for comparisons with other plans.

More recently, the LOSSAN Rail Corridor Agency (Caltrans) updated this plan for the LOSSAN corridor from San Diego to San Luis Obispo, with very little modification to project costs and service levels. No new estimates of run times and ridership were provided. SANDAG (presumably in coordination with the North County Transit District) has also developed its own priorities for rail improvements in San Diego County, and these are largely consistent with LOSSAN's priorities.

FRA Study (1997)

Four years earlier, the Federal Railroad Administration published evaluated the prospects for better passenger rail in a number of promising corridors in the United States. This study considered a range of options, from modest improvements to existing services (90 mph maximum speed) to full-blown HSR (around 200 mph) and even Maglev. It proposed using tilt-train technology to reduce capital investments, and recognized the need for electrification to achieve speeds above 125 mph. This study considered improving the California coastal corridor (San Diego to San Francisco) to allow maximum speeds of 90 mph and 110 mph, and higher speed options for the parallel Central Valley route.

Capital costs and ridership in the FRA study were estimated using generic models, which gave a much more optimistic view of the prospects for intensive service between Los Angeles and San Francisco (up to 30 RTs per day!) than the Amtrak study seemed to consider possible. Another difference between them is that the FRA study proposed running more trains at lighter loads to achieve essentially the same ridership.

Discussion

San Diego to Irvine. Mid-term improvements include almost complete double tracking in San Diego County except along the Del Mar bluff and only minor improvements in Orange County south of Irvine, at a cost of around $1 billion. Long term ("vision") projects include tunnels in four sections -- Miramar, Del Mar, San Clemente bypass, and San Juan Capistrano -- at a total cost of over $2 billion. These tunnels are needed to achieve full double tracking of the San Diego to Irvine section. In the mid-term rail operations are going to be impacted by a number of single-track segments which will affect frequency and reliability more than run times. There are alternatives -- for instance, double tracking along San Clemente beach is technically feasible but nearly impossible due to local opposition. San Juan Capistrano quite reasonably wants to reduce the number of trains operating through its historical district, and adding a single-track bypass tunnel under I-5 might be an acceptable alternative.

Irvine to Burbank Junction. This segment is eligible for HSR (Prop. 1-A) bonds, and major improvements are likely in the near future if the HSR Authority goes ahead with its current plan for a "blended" system in urban areas. High priority projects include run-through tracks at L.A. Union Station (LAUS) at a cost of $430 million and almost complete separation from freight operations between Fullerton and LAUS.

Burbank Junction to San Luis Obispo. Metrolink bought right-of-way from Burbank to Moorpark from Southern Pacific twenty-five years ago, and plans to double track most of this section by 2025. Further north, Caltrans plans include adding some passing sidings and realigning the track in places for a cost of $1.6 billion. No new tunnels are contemplated.

San Luis Obispo to San Francisco. Caltrans plans to spend an additonal $1.2 B in this section, with double tracking the Cuesta grade the largest single project. Caltrain is expected to make major improvements to the section on the San Francisco peninsula, including electrification, at an additional cost of perhaps $2 B.

Conclusions

1. In the southern section of the California coastal corridor (between San Diego and Los Angeles), it should be possible to nearly double the frequency of the Surfliners to 20 RTs per day and reduce run time from San Diego to LAUS to an even two hours, even without expensive tunneling.

2. Ridership on the northern section of the Surfliner corridor is about half that on the southern, so 10 RTs per day are possible, at least to Santa Barbara. The Amtrak study suggests that two hour run times for both L.A.-Santa Barbara and Santa Barbara-S.L.O. are feasible.

3. The FRA study is probably correct that there is a significant latent market for passenger rail between Los Angeles and San Francisco (two to three million riders per year), even at a run time as high as seven hours. Even so, the volume of trains proposed in the this study (up to 30 RTs per day) is probably excessive. The Amtrak study, on the other hand, recommends making track improvements to reduce the run time for the Daylight to 8 hours and starting with a modest frequency of just 2 RTs per day. This level of service will provide an accurate estimate of the demand for the Daylight, and service can be expanded as needed.

4. Ridership on the coastal route would be drastically reduced if HSR is implemented via the Central Valley at a run time of five hours or less, but the coast provides a quick and relatively inexpensive alternative in the meantime.

5. The utility of tilt train technology as an inexpensive alternative to costly capital improvements needs to be taken seriously in future rail planning for the California coastal corridor. Fifteen years ago, the FRA study advocated using tilt trains on this route, projecting a run-time of about 7:20 for tilt trains operating at speeds up to 90 mph and under six hours for speeds up to 110 mph.

6. Electrification can reduce run times by up to 10%, and would benefit the intercity trains even if implemented only in the urban areas provided they use hybrid (dual power) locomotives.

7. Base cost of needed improvements to the coastal corridor is around $5 billion, with tunnels in the southern section adding another $2 B. Much of this cost will be borne by Metrolink and Coaster agencies, but Caltrans is solely responsible for improvements north of Oxnard.

8. The Union Pacific owns the tracks between Moorpark and San Jose, and its active cooperation is needed for improved passenger rail service in this section. Freight operations are currently pretty minimal (four trains per day), and this is not likely to change very much in the future. As long as Caltrans pays for the improvements it needs to run passenger trains on a reliable schedule, U.P. has little reason to object to increased frequency of passenger trains since it should get a fair compensation for hosting the trains. The relatively small number of freight trains should be easily accommodated.

References

Table 1.
Comparision of Amtrak and FRA Plans for Surfliner/Coastal Corridor

2011 Actual1 Amtrak Plan
2020 Vision
LOSSAN Plan
2025 Vision
FRA Study
90 mph Case
FRA Study
110 mph Case
Project Costs (Billions of 2010 Dollars)
San Diego -
Los Angeles
---$3.892$4.47$0.693$1.003
Los Angeles - San Francisco ---$2.722,4$2.825$1.283$3.403
Number of Daily Round Trips
San Diego -
Los Angeles
111616256266
Los Angeles - Santa Barbara 577------
Santa Barbara - San Luis Obispo 224------
Coastal (L.A. - San Francisco) 022297317
Run Times
San Diego -
Los Angeles
2:451:57---2:001:48
Los Angeles - Santa Barbara 2:422:04---------
Santa Barbara - San Luis Obispo 2:312:11---------
Coastal (L.A. - San Francisco) 11:0788:11---7:1795:469
Annual Ridership (Millions)
San Diego -
Los Angeles
---------2.93.0
Coastal (San Diego-S.F. 2.796.18---6.37.9
  1. From Amtrak National Timetable, Spring-Summer 2011.
  2. Inflation factor of 1.267 (2000-2010) applied to original data.
  3. Inflation factor of 1.509 (1993-2010) applied to original data.
  4. $1.55 billion for Los Angeles - San Luis Obispo; $1.17 billion for San Luis Obispo - San Francisco.
  5. Investments in Surfliner (Los Angeles - San Luis Obispo) section only.
  6. No service north of Los Angeles
  7. All trains assumed to operate from San Diego to San Francisco.
  8. Northbound Coast Starlight, Los Angeles - Oakland.
  9. Factor of 0.9 applied to run time given in source (Irvine to San Francisco).

Thursday, May 3, 2012

Can we use the Coast Line as an alternative

to the High Speed Rail Project?

To Noel Braymer, 4/22/12

Included in RailPAC Weekly E-Newsletter, April 30, 2012

The HSR Authority probably realizes that it isn't likely to get additional federal grants for at least ten years. That's probably the main reason it's moving ahead with "blended" operations in the urban areas, so it can use local funds to match Prop. 1-A bonds. This is fine with me -- a modern, electric rail line from Lancaster to Laguna Niguel (outer limits) is a much better project to demonstrate modern rail service project than a stranded Fresno-to-Bakerfield segment -- but I would still prefer to see some progress in closing the Bakersfield-Palmdale gap.

I also think that the HSR Authority is evading a couple of fundamental questions -- if the rest of the Prop. 1-A bonds are chewed up by construction in the urban areas, what happens to the overall HSR project? And don't the restrictive conditions written into Prop. 1-A -- construction to 220 mph standards, alignments through Pachecho and Tehachapi passes, 2:40 running time between SF and LA -- effectively disappear once the Prop. 1-A bonds are exhausted? It seems to me that the project will then have to be completely re-evaluated, and again that isn't a bad outcome. But the state will arrive at this point with three isolated HSR segments, and unable to connect them short of another public vote to authorize more state bonds.

Since this point is not likely to be reached for another 15 years, what other, less expensive intercity rail options should be pursued in the meantime? I've felt for some time that the Daylight option on the Coast is a good, relatively inexpensive alternative to the Central Valley route for SF-LA service in the short-term. With a relatively modest investment -- perhaps $5 billion, but that's peanuts compared to the cost of the Valley alignment -- the SF-LA running time along the Coast could be reduced to as little as 6 hours. And this service could run electric in the urban areas (using dual-power locs), which automatically helps to improve train performance. Through-running from San Diego (which I understand you'd like to see) would also be automatic. But it won't help me a lot (Sac-LA), which will still require a bus bridge (Bakersfield-Newhall) for a long time into the future.

150 mph is fast

To Noel Braymer, 4/8/12

Included in RailPAC Weekly E-Newsletter, April 16, 2012

You've probably seen the op-ed piece in today's (4/8) Sac Bee by Daniel Carson discussing the dismal prospects of funding the California HSR project. The piece is interesting because it is not overtly anti-HSR, but sticks to pointing out the financial fantasies concocted to support rapid completion of the IOS and the "Bay to Basin" plan. Carson's views are, in fact, very close to my own.

To be fair, there are some positive aspects to the latest version of the HSR plan: designating the Tehachapi crossing as the next segment to be built, and undertaking to improve existing passenger rail lines with a view towards implementing "blended" HSR services where appropriate. But funding of these improvements is even hazier than funding HSR itself.

What is missing in the debate about HSR is an understanding that modernization of existing rail services at some reasonable speed is much less expensive than the full-blown HSR option built to 220-mph standards. Moderization would include full grade separtation (from roads) and almost complete separation from freight operations, to allow running of diesel passenger trains up to 125 mph and electric trains up to 150 mph. The target should be a running time of under 5 hours from the Bay Area to Los Angeles, and a half-hour less for Sacramento-L.A.

The 220-mph system should be deferred to the far future, and the state should concentrate on improving existing services plus the Tehachapi crossing. The current federal funding of about $3.3 billion, plus an equal match from Prop. 1-A bonds, gives the state $6.6 billion to start with. The state should renegotiate its federal contract to build the Fresno-Bakersfield segment as rapidly as possible (to a 150-mph standard), using up the federal grant to meet the 2017 deadline. The state bonds would then be devoted to building the Tehachapi crossing, starting with the Bakersfield-Tehachapi segment and perhaps even reaching the L.A. county line to link up with Metrolink.

Concurrently, the existing Metrolink Antelope Valley and the Orange County lines would be moderized, so that electric service could be operated from Irvine to Fresno (300 miles) by 2020. This is a practical alternative to the currently proposed IOS, and should cost no more than $10 billion.

Many thanks for the notes you publish every week -- I look forward to reading them.

Letter to Sacramento Bee, 3/26/12

[Not published.]

Re: Dan Walters' op-ed piece, 3/23/12

The plan for high speed rail (HSR) in California should be drastically scaled back to a precursor system ("HSR lite") that can be built in a reasonable time frame for a reasonable cost. The target of under three-hour service between San Francisco and Los Angeles has already been effectively abandoned, and with it the need for 220 mph trains anywhere in between. HSR lite, with a maximum speed of around 125 mph, is a much better fit for the immediate passenger rail needs of the state. This alternative would be designed to provide greatly improved service in both the northern and southern urban areas, with a connection between them through the Central Valley via Tehachapi and Altamont passes. And it is affordable, with a probable cost of under $25 billion, compared to an estimated cost of well over $100 billion for the current plan for HSR.