Detailed Analysis of Passenger Rail Transit vs Bus Rapid Transit
by Mark Mesiti‐Miller, P.E.
Board Chair, Santa Cruz County Friends of the Rail and Trail
This detailed analysis is focused on answering the question of why the County of Santa Cruz should move forward to implement passenger rail transit rather than bus rapid transit along the existing rail corridor. A number of issues are explored and analyzed. Most of the data used in the analysis is taken directly from the Unified Corridor Investment Study (UCS). Other data sources are referenced when used.
Passenger Rail Transit on the rail corridor is easy to understand and implement as it simply consists of upgrading the existing railroad tracks, adding boarding platforms and other appurtenances and running some type of new passenger rail vehicle along the tracks between Watsonville and Santa Cruz. There are many modern passenger rail vehicles to choose from and a final selection should be made after studying the specific needs of this county.
Bus Rapid Transit (BRT) on the rail corridor is a bit more challenging to understand as the proposed BRT system only uses about 40% of the rail corridor (8.5 mi of 20.5 mi). The rest of the time, BRT will be running on local surface streets or along Highway 1. Even in the short 8.5 mile section of the rail corridor proposed for BRT use, more than 70% of this length (6.1 mi of 8.5 mi) requires running BRT on local surface streets near the rail corridor. Below is an illustration of the proposed BRT system taken from the UCS, Table 12. Take a moment to study the “BRT on the rail corridor” proposal ‐ it is important to understand what BRT on the rail corridor really means. Regarding vehicles, there are basically two types of buses from which to choose for BRT purposes: a standard bus or an articulated bus (sometimes referred to as a ‘bendy’).
As proposed, BRT on the corridor will consist of two‐way traffic (2.4 miles) and one‐way traffic (6.1 miles) with reverse direction on parallel local streets for a total of 8.5 miles. There are four places along this 8.5 mile stretch where two‐way BRT traffic will occur over one‐lane segments under signal control (buses will be held until lane is clear of bus traffic in opposing direction).
Two types of BRT service levels were identified in the UCS. BRT with 20 stops and BRT Express with 8 stops between Watsonville and Santa Cruz. Where data is provided for both service levels, only the data for BRT Express service is used to compare “apples to apples”.
The UCS is silent regarding what types of site appurtenances (boarding areas, shelters, intermodal transfer facilities, ticketing machines, bike parking, auto parking, etc) and other corridor improvements (signal controls at surface street intersections, traffic barriers between busway and trail, etc) would be included in the implementation of BRT on the rail corridor. Yet, costs for these items appear to be included in the UCS estimated costs for rail transit. Thus, cost comparisons between BRT and rail transit are not accurate.
BRT on the corridor only uses about 40% of rail corridor (8.5 mi of 20.5 mi) between SC and Watsonville. This fact alone raises important questions that must be addressed including:
- what happens to the other 12 miles of the rail corridor between Watsonville and Aptos?
- how much of the remaining 12 miles exists as easements only and would be subject to expensive and lengthy litigation resulting from presumed abandonment of the rail line?
- how much delay in building the rail trail would occur while easement issues are resolved and the rail trail master plan and environmental documents are redone?
- Assuming an average 3% annual rate of inflation and a delay of 10 years, means the cost of the Rail Trail currently estimated to cost $283M would go up $97M raising the total cost to $380M. This delay is an unnecessary waste of time and money.
- If we add the cost of trail delay to the current estimated cost of BRT on the rail corridor, the estimated cost of BRT on the corridor rises to $362M, well beyond the $325M estimated cost of passenger rail transit.
- could we lose portions of the rail corridor all together ending the opportunity to have a continuous rail trail – what potential upside public benefit justifies taking such a risk?
As proposed BRT on the rail corridor fails to address the ‘Aptos Strangler’ identified as the obvious single biggest transportation bottleneck by internationally acclaimed transportation expert, Jarrett Walker, as BRT on the rail corridor starts/ends at State Park Drive. The ‘Aptos Strangler’ is a substantial source of social inequity and denies south county residents from enjoying the same access to education and employment opportunities as north county residents. The ‘Aptos Strangler’ denies south county residents the freedom north county residents have now and denies them the improved freedom north county residents would enjoy with a more robust transit system afforded by BRT on the rail corridor. Accordingly, BRT on the corridor appears to be discriminatory in that it fails to provide the same social equity and economic opportunity as rail transit would to all citizens.
The above figure illustrates how convoluted BRT on the corridor will be. Less than 30% of the rail corridor proposed for BRT allows unrestricted two-way bus traffic. Accordingly, the proposal doesn’t meet the basic definition of a BRT system. If one simply studies the above illustration, it is clear that BRT on the rail corridor is what you would do if you had no other options. In that sense, BRT on the rail corridor appears to fail the common sense test because an efficient option exists, namely passenger rail transit.
What about CapEx: According to the UCS, BRT on the corridor is estimated to cost $265M. Dividing $265M by the 8.5 miles on the corridor = $31.2M/mile. Rail transit is estimated to cost $325M. Dividing $325M by the 20.5 miles on the corridor = $15.9M/mile. Obviously, rail transit provides a far more economical utilization of the rail corridor AND successfully addresses the ‘Aptos Strangler’ problem.
Question: How much more would it cost for BRT on the corridor to address the ‘Aptos Strangler’ problem? To address the ‘Aptos Strangler’, the next location available for buses using surface streets to access the rail corridor would require diverting buses from Highway 1 at San Andreas Road then proceeding down San Andreas to Seascape Blvd, then down Seascape Blvd and connecting to the rail corridor near the intersection of Seascape Blvd and Sumner Ave (near the entrance to Seascape Resort). This would add about 2.9 miles to the length of BRT on the corridor. BRT CapEx is about $28.6M/mi excluding the cost of bus vehicles ($22M incl contingency). Thus, the cost to extend BRT on the rail corridor to solve the ‘Aptos Strangler’ problem would be $83M ($28.6/mi x 2.9mi). Adding $83M to the original UCS estimate for BRT on the rail corridor of $265M results in a total estimated CapEx of $348M for BRT on the rail corridor. Add to this figure the $97M cost of delaying the Rail Trail for ten years and the total CapEx for BRT on the corridor rises to $445M. Whether the grand total for BRT is $348M or $445M or some figure in between, the total cost to implement BRT on the rail corridor will be well beyond the $325M needed to implement passenger rail transit between Watsonville to Santa Cruz. For a transportation service that is inferior to passenger rail transit in so many ways and puts the rail corridor at risk, BRT simply does not make sense for the taxpayers of the County.
What about OpEx: The OpEx figures provided in the UCS stand in stark contrast to the figures found in the National Transit Database published by the Federal Transit Administration. According to Exhibit 6 of the NTD, the operating cost of bus rapid transit is $1.07 per passenger mile while the operating cost of commuter rail is $0.51 per passenger mile (52% less than BRT). The operating cost of light rail is $0.79 per passenger mile (26% less than BRT). Since we don’t know what type of rail transit service will be implemented, it is reasonable to say passenger rail transit will probably be 30‐40% less per passenger mile than BRT when a true apples to apples comparison is provided. Since the UCS presents relative OpEx cost estimates so different than the NTD, I offer the following as possible explanations.
- Based on the figures provided in Appendix B‐12, new Bus Rapid Transit (BRT) service with 15 min frequency on the rail corridor will only require an additional 29 hours of increased bus service on weekdays to support the new BRT operation. Yet according to figures in Appendix B‐10, new passenger rail transit service on the rail corridor with 30 min frequency, will require 265.5 hours of increased weekday and weekend bus service to support new rail transit service (almost 10 times more). Assigning 10 times more cost to support rail transit than BRT is internally inconsistent. Either BRT on the corridor is receiving insufficient support or Rail Transit is receiving excessive support. Common sense suggests the level of supporting bus service necessary to serve both of these new faster and exclusively ‘main line’ type transit options should be essentially the same.
- Bus vehicles must be replaced more frequently than rail vehicles. Accordingly, total vehicle life cycle costs must be included in any comparative OpEx analysis. Per the Federal Transit Administration the expected useful service life of a bus is 14 years while the expected useful service life of a commuter rail self‐propelled passenger car (such as considered in the UCS) is 39 years. According to the UCS, BRT buses are estimated to cost $22M (including contingency) but if they must be replaced 2.78 times as often (39yrs/14yrs), the real present day value of bus vehicles is $61M (22 x 2.78). The $39M additional cost ($61M ‐ $22M) for more frequent bus vehicle replacements should either be added to CapEx or realized as an added OpEx cost. Given the 39 year life of the comparable rail vehicle, the math is easy and $1M should be added to the annual BRT OpEx figure.
- The UCS does not appear to include costs of maintaining the one‐lane, two‐way traffic signal systems, nor costs of maintaining signal systems at local street crossings of which there are many.
Summarizing, once true cost adjustments are made to the estimated OpEx for BRT, OpEx for passenger rail transit is likely to be substantially less than OpEx for BRT on the corridor.
What about Ridership: The UCS predicts daily ridership of 7,396 for rail transit and 3,949 for BRT. The 3,447 difference in ridership expressed as a percentage improvement, means passenger rail transit will carry almost 87% more passengers every day than BRT. The difference in predicted ridership is explained in the UCS as follows: “The main factors that reduced ridership is that BRT between Watsonville and Santa Cruz is not a dedicated facility for the entire length in both directions and that the travel time for BRT is longer than for rail.” This monstrous difference in ridership alone justifies abandoning any further consideration of BRT on the rail corridor.
What about Transit Times: The UCS predicts transit travel times between Watsonville and Santa Cruz will be 41 minutes for rail transit whether travelling during AM or PM peaks. The UCS predicts the average transit travel time will be 58 minutes for a BRT service equivalent to rail service (i.e. BRT Express Service with 8 intermediate stops). A little math tells us BRT will be 17 minutes longer or result in a 41% increase in travel time. The difference becomes even more pronounced when one considers the UCS predicts AM travel time for BRT Express will be 63 minutes, 22 minutes longer or a 54% increase in travel time. Given the much poorer travel time performance for BRT, it is easy to understand why predicted ridership for BRT on the rail corridor is so much lower than for rail transit.
Rail transit suppors and will result in more bike ridership than BRT. Buses do not provide the same bike carrying capacity nor ease of use for cyclists. For buses, bicycles must be lifted up and onto a rack located at the front or back of a bus, a task many cyclists are unable to complete. On the other hand, bicycles can be simply rolled onto a passenger rail vehicle, a task any cyclist can perform. Accordingly, passenger rail transit will encourage greater use of active transportation than a bus. (Note: When SMART system opened in August of 2017, each two car train set had the capacity to carry 22 bicycles. Four months after starting operation, SMART added a third car to meet the unexpected demand for bicycles.)
Rail transit travel times are reliable and BRT travel times are variable. Because BRT substantially relies on surface streets and therefore relies on current AM/PM peak directional traffic patterns which are likely to change as many current commuters will testify, travel time performance estimates of BRT will be more variable and are likely to deteriorate over time as traffic congestion increases on the highway and local surface streets. Because passenger rail transit operates in a dedicated corridor, travel times will be unaffected by changes in traffic patterns. Travel time reliability is valuable.
BRT does not provide significant peak travel time relief. Because BRT transit times are so much longer than that of passenger rail transit, BRT doesn’t appear to provide much, if any, peak travel time advantage to south county commuters particularly during the morning commute (it should be noted: one third of the entire county population lives south of La Selva Beach, commonly known as south county.)
BRT in the rail corridor will create substantial amounts of new impervious paved areas (8.5 miles at avg 18’ width = 18 acres or if extended to Seascape Resort, 11.4 miles at 18’ width = 25 acres), a significant environmental issue requiring mitigation. Because railroads are constructed with gap graded ballast, railroads are considered pervious surfaces and therefore require no storm water runoff mitigation.
Because buses are not constrained by tracks and will wander, running buses on the corridor will require a substantially heavier crash protection system separating the busway from the adjacent multiuse trail instead of the visually transparent cable rail fence proposed for use with rail transit. Besides visual impact, BRT crash barriers may present challenges for wildlife crossings.
BRT has limited capacity (BRT will require about 5 times more buses to carry the same number of commuters as a typical passenger rail vehicle – even more if rail transit is scaled up) if residents were worried about 60 trains a day, imagine their reaction when they hear 300 buses a day are coming their way. Rail transit moves more people using less money and energy – per the UCS:
1 rail car OpEx is $547 hr but moves 300+ people
1 bus OpEx is $200/hr but it takes 5 buses to move as many people at cost of $1,000
BRT on the rail corridor has limited scalability (limited bus vehicle capacity combined with the functional throughput limitations where two way traffic occurs on signal controlled one lane sections severely hampers maximum bus passenger throughput.)
Rail transit is far more scalable – adding one car to a two car train set adds 50% more capacity at nominal added cost. Doing the same with buses would require 3 more buses at $200/hour each.
Passenger rail transit provides the 80,000 south county residents access to fast, modern, quiet, comfortable rail transit and does not leave them stuck them with slower, inferior bus transit. Passenger rail transit improves the balance of social and economic equity.
UCS indicates mode share for passenger rail transit would be 25% higher than for BRT even though UCS considered BRT would run on 15 min frequency vs passenger rail transit would run on 30 min frequency.
Without substantial expense to provide both tracks and pavement, implementing BRT on the corridor will eliminate / break the continuous railroad tracks between Watsonville and Santa Cruz. Doing so:
- denies Roaring Camp (a thriving local business) access to nationwide rail network for replacement rail vehicles and other needs
- denies freight rail service to customers north of Watsonville. Every freight car replaces 4 highway trucks (saving space on our highways, saving fuel, reducing GHG emissions, and making roadways safer). Moving freight by rail can also give local businesses a competitive advantage in the market.
Rail transit leads to more compact development patterns – reducing sprawl and leading to more efficient use of our most valuable resource, developable land area. Bus routes can be changed without notice and are less dependable over the long term. Transit Oriented Development thrives under rail transit systems that are essentially permanent. Housing providers and other developers like permanent.
Passenger Rail Transit on the corridor
- attracts more riders and more active transportation users
- offers significantly faster travel times
- offers unparalleled reliability and scalability
- protects the existing rail easements and avoids costly time consuming litigation
- allows the rail trail to be completed ASAP
- is eligible for substantial outside funding
- will require less local money to build and operate
- will give our community the best public transportation system possible
Accordingly, the RTC should focus resources on developing passenger rail transit on the rail corridor combined with an optimized METRO bus system with the goal of transforming the public transportation system serving our community ASAP.