Blog Post < Previous | Next >
SMART
Is SMART smart? Richard Hall and Frank Leahy debate the issues
The following resulted from the publication of Richard Hall’s Assessing the True Cannibalization Cost of SMART, on the Marin Post. Frank Leahy of Mill Valley and Richard Hall of San Rafeal debate the facts and issues, with Bob Silvestri moderating and adding occasional comments.
FL
We just spent 5 weeks in Europe this summer, and the amount and convenience of public transportation is absolutely staggering when compared to the US. Take Gothenberg, Sweden where we were for a week. I'd bet that > 95% of the trips taken were on public transpiration. Stops were plentiful. I never waited more than 5 minutes day or night for a tram or bus. It made SF and the Bay Area look like a poor second cousin. Shameful actually how far behind we are.
So, my question for you Richard Hall is -- what is your alternative to the car? From what I read above you basically want to continue pumping money into car based projects (CalTrans, Novato Narrows, Greenbrae Interchange) and put nothing into public transit. Do I have that right? My personal opinion is that we underfund projects like SmartTrain. What I'd like to see is a SF across the GG bridge tram system with branches into MV, Fairfax corridor, San Rafael, etc. so that people could get out of their cars. And then maybe we could attack the housing problem with a good transportation system in place.
RS
I think Richard's point is not about being for or against public transportation. I think it's about telling voters one thing and then having what we funding turn out to be something else. SMART is costing almost 4 times what it was advertised to cost, and will only serve 1/20th of the riders they originally stated (they fudged the figures). I'll let Richard speak for himself, but I think it's about spending and getting bang for the buck.
It's also about comparing apples to apples. Gothenburg, Sweden, is the second largest city in the county, with a population of about 549,000 people in the urban core and almost one million in its metropolitan area. Marin, in total, has about 250,000 people and we're very spread out. It's comparing urban to suburban / rural. In addition, although I think there's much to be admired about Sweden (schools, healthcare, etc.), people in Sweden pay enormous taxes to support all those things, including subsidizing all their public transportation. We just don't have the tax structure or public funds to do that (note the recent Marin IJ article about ridership of buses decreasing because the agency is so poorly managed).
Lastly, SMART is unfortunately not just a 19th century solution to a 21st century problem, but it's not addressing our most pressing public transportation need, which is all the traffic generated within one mile of highway 101, going to and from services and schools in the corridor and back into the hills and canyons on all sides.
Perhaps an electric, app driven shuttle system would be a better use of our precious funds?
FL
Traffic is just people trying to get somewhere. Give them a single option -- cars -- and that's what they'll take. Give them an option that works (e.g. ferry and bus service at commute times) and they'll use it. Don't give them an option, or give them an option that doesn't work (e.g. most bus routes on 30 or 60 minute schedules) and it's guaranteed to fail.
RH
But does that mean you should spend $1bn on an option that's unlikely to get used, or cost ineffective?
The issue with the train vs the car is the person has to decide between:
- travel time (the train may not leave at the precise time they want to go or return);
- comfort (sitting in a chair adjusted to you, listening to your favorite music with a/c set just right); or perhaps the comfort of not being stressed by driving in traffic;
- convenience (door to door);
- connections: each connection adds wasted time and a risk of missing connections.
Then there's reliability. In Marin IJ comments on bus ridership declining one commenter mentions that after getting to work late / returning home late after repeated experiences with bus unreliability she considered switching to driving - while this involved traffic, she considered driving to be more reliable overall than the bus.
Remember this train:
- is single track, with limited passing points, any delays will ripple up and down the line;
- uses totally new DMUs that are not time proven;
- is in a flood plain;
- service intervals are every half hour, miss a train or if a train is cancelled and riders will wait a LONG time.
FL
I agree there are almost unlimited reasons to think smart will fail. And if smart guys like the folks on this thread keep putting it down instead of figuring out to make it work within a ten or twenty year framework then it's going to be that much harder to succeed.
But I came back from Amsterdam and Switzerland and Germany and Denmark and Sweden agog at the incredible public transportation infrastructure, and wonder why we can't at least start thinking about how to do even maybe just a little bit better, rather than throwing up our hands and saying it's not fundable, it'll never work, the sky is falling.
Someone had the balls to get the train running again...more power to them. Can we riff on that and use it to advantage instead of tearing it down?
RS
The great line from the film "Chinatown" is that "politicians, old buildings, and whores all become respectable if they stay around long enough" is apt here. Yes, someday people will say SMART was a great old train idea, and praise its promoters, long after the costs that burdened the public are written off and forgotten. Unfortunately, some ideas are just bad ones, and a waste of precious taxpayer funds. Still, what's done is done. So the real question is since everyone agrees that public transportation is a great thing, so long as it addresses the requirements Richard has outlined (which are correct: I grew up in NYC, where owning a car was only for going to Long Island), how do we fund it? And who would run it?
It's not a question of the sky falling. It's just fiscal reality. So that's why even Charles McGlashan proposed a pod based, monorail system in that very precious train right of way, running high above traffic intersections, instead of going back 100 years and building crossings we worked so hard to get rid of for the last 50 years, because they are very very dangerous. So now we have no right of way to use for 21st century, and much cheaper, and much greener technologies, and we have debt. So financing and wasting opportunities matters.
So there are many of us already working on alternative solutions, many of which have been published and discussed. But so long as large, top down planning agencies control the game, we will never even remotely get to what Germany, Sweden or Denmark take for granted.
It's easy to just believe something can be done if we just want it to, but it just doesn't work out the way the decision making and funding system is set up. I'd suggest that that's the problem we have to fix to fix the others.
RH
I am a former European who likes trains, but I would not make such an inappropriate comparison as to compare the mature and extensive train network of a country like Sweden to SMART, and use this to justify SMART.
The 5 minute waits you experience are a far cry from SMART - a single track diesel with limited passing places. If you look at the draft SMART timetable you will see that trains are 30 minutes apart, and that is only during commute hours, you will likely have to wait 15 minutes on average - and you will likely need to connect and wait even longer to get to most genuine destinations.
Draft SMART timetable:
http://www2.sonomamarintrain.o...
Re: CAR ALTERNATIVE
You ask "what is your alternative to the car?" This is a leading question - clearly you presume that the car is not the answer. Maybe you think that the train is and you are trying to back into that supposition?
What is just around the corner are autonomous cars offered as shared ride services operated by giants like Google, Apple, Uber and Lyft. They will go door to door, unlike trains. If a route proves economically unviable then $1 billion+ has not been wasted as with the train.
You state "What I'd like to see is a SF across the GG bridge tram system with branches into MV, Fairfax corridor, San Rafael, etc. so that people could get out of their cars."
Sure if money were no object I would agree with you, but money IS AN OBJECT. Back in 1956 BART considered extending to Marin - by far the majority of Marin's residents - 87% according to a survey - wanted BART to come to Marin. However BART foresaw that Marin did not have the population density and tax base to support just that one rail artery through the county and pulled out.
What you describe would cost likely $10 billion plus. But you presume it will get used. If we look at the figures for transit use in Marin we will see transit use has steadily declined. Why? We are a relatively wealthy population that values privacy, convenience and comfort. We vote by driving instead of taking transit.
So in addition to $10 billion plus there would have to be a wholesale "attitude reversal". Tell me, why do you believe such an attitude *reversal* would take place?
The $10 billion+ would add an immense tax burden making Marin even less affordable. Already our taxes are unreasonably high and we face a crisis with unfunded pensions.
Finally your radical proposal, based on massive expenditures and a presumption of a reversal of attitudes would happen in parallel to the introduction of autonomous car fleets that chain together, reducing congestion, and render rail systems obsolete.
RS
Richard, I would have to add that the whole "cars are bad" argument needs to be re-evaluated also from the scientific, environmental impact side. The cars are bad mantra is from the early 1970s. But what has changed and is changing more rapidly than even car makers can keep up production for are hybrid cars, electric cars and trucks and buses, and now hydrogen cars, which Toyota and Honda are now selling in the US (in Japan they are becoming ubiquitous because they are building refueling stations). These cars do not pollute or barely pollute. Your gas lawn mower does more environmental damage, as does leaving all your electronic devices and computers "on" all the time.
So this leaves us with an entirely separate problem, which is traffic congestion. And solutions to that are varied, and yes, to include self-driving vehicles, car sharing, Uber and Lyft, Zip car, and perhaps more importantly getting rid of bogus deductions in the IRS code that allow people to write off vehicles over 6,000 pounds as "business vehicles" even if they're used to go to Safeway or camping.
RH
You're quite correct. There seems to be an unhealthy popularly held myth that cars are all bad - they're dirty. I would almost go so far as to say that in certain circles it's trendy to bash cars.
The reality is that transit, unless it can be filled with riders, emits much more than cars. In suburban and rural locations such as Marin attaining even a modicum of ridership is a significant challenge. As you mention cars like the Toyota Mirai are available today that emit...water.
https://ssl.toyota.com/mirai/f...|google||
FL
Gothenburg is smaller than SF, and it makes SF transportation look like a (very) poor second cousin. I bike to the Sausalito ferry in the mornings, and the first day I got back, I looked at the traffic on the Embarcadero and saw four lanes of traffic going both ways. In Gothenburg those four lanes would have been 1 lane for bikes, 1 lane for cars, and two lanes for buses and trams. And almost no cars on the road.
And help me understand how self-driving cars are going to reduce the trips taken in the mornings and evenings? Same people sitting in a self-driving car is the same traffic that's there today. What's different about it?
RH
And help me understand how self-driving cars are going to reduce the trips taken in the mornings and evenings? Think Uber Pool + autonomous cars.
Door to door service. Far fewer single occupant vehicles. Then there's "chaining" which reduces the gap between cars. Cars use WiFi or an equivalent so their computers are linked together. When the front car brakes it signals all the cars behind it to brake.
Then with computer driven cars there's the reduction in lane switching, and unnecessary acceleration and deceleration.
FL
Check out: http://motherboard.vice.com/re...
Over the last 10 years the big problems in Marin and Mill Valley are (were) - 101S in the morning, 101 north at night, Mill Valley morning commute (Blithedale to 101, Tam Junction to 101, and in-town near Tam High and MVMS), and Mill Valley weekend commute (drivers going to Muir Woods and Stinson).
101 south in the morning. Queueing theory says that when the service rate is greater than the arrival rate, then all is good...and all was good after the bridge went to all electronic tolls.
I remember when 101N backed up into the rainbow tunnel regularly, but after they opened up the SR 3rd lane it got better. But it's now back to its old ways with backups caused by the single lane exit trying to get to the Richmond San Rafael Bridge. The only way to solve this is to increase the flow from 101 to RSR the bridge (the proposed re-stripings are not going to make a whit of difference). That means at least two lanes so that there's no backup onto 101.
Tam Junction to 101 in the morning was better for a while after they changed from a stop sign at Manzanita to stop lights. The reason is that traffic flow increased substantially because 50% of the traffic heads to SF, and the light at Manzanita is almost always green towards southbound 101. Traffic went back to its old ways when they added the light at Tennessee Valley Road because of a very straightforward queuing theory problem -- because there's a light at TVR, drivers now have a known time delay, whereas prior to the light the delay was unknown, and either people were polite and let people turn, or they weren't and you sat there forever.
Because of the unknown time delay many more drivers elected to not turn there and went past Walgreens and turned at Walgreens where it didn't impact either northbound or southbound traffic on Shoreline. Sigh. The only way to fix this is to remove the light at TVR for anything except foot traffic crossing Shoreline, so that drivers cannot turn there with a known time wait.
The weekend traffic entering Tam Junction on a nice day is a real problem, and I regularly have to work around it by either going up to Blithedale or behind the police station. Buses to Muir Beach and Stinson could help here, but they would need to be super convenient, price effective, and there would need to be a big parking lot somewhere. Two lanes past the Dipsea with the right lane designated "local use, no left turn onto Rt 1 to Stinson Beach from this lane" could help, but I don't see such being built. I guess this is what we get by living so close to something like Muir and Stinson beaches.
In town traffic is very solvable, with a combination of school buses, better GG commute options to SF, introduction of separate bike lanes, support for electric bikes, and introduction of round-abouts at key junctions such as Tam Junction, Joes Taco Lounge, and at the Blithedale 101 overpass.
Just my 2c...
RH
CITATIONS & DATA – A regular freeway lane has a capacity of 2,200 cars per hour. A study by Bose and Ioannou assessed the increase in capacity based on different market penetration rates of autonomous vehicles:
- 0% penetration: 2,200
- 50% penetration: 2,685 (22% increase)
- 80% penetration: 50% increase in capacity
- 100% penetration: 80% increase in capacity
SOURCE:
https://www.semanticscholar.or...
FL
At least 10 and more likely 20 years before there will be a measurable impact of using autonomous cars. Lots we can and should do before then.
RH
But not a billion dollar train that doesn't have a hope of generating a critical mass of riders for 20 years. Yes we may have it, but we need to stem the poor forward- thinking and blatant misrepresentations of these types of transit projects.
For instance there is even greater madness waiting in the wings - there is a desire to put trolleys into the San Anselmo corridor that would serve to reduce the capacity of this area that already has acute traffic congestion. Trolleys are long and travel slowly creating an obstacle that must be overtaken. The EIR for the Anaheim trolley, if properly analyzed, showed that it reduce the capacity of the streets that it served.
FL
Quick question to set my expectations here...are there any public transit options that would serve Marin that you would support?
RH
Yes. Many. I've long advocated that the best solution is to much more significantly subsidize bus fares - creating a proven incentive that will increase ridership. I've been very consistent on this point.
Buses are much better suited to the narrow ( lower density population) routes of our county. We also need to subsidize school buses. A lot of the issues in Mill Valley are people doing school runs. I also favor bike paths - just not ones across bridges or through tunnels costing tens of millions but used by a couple of dozen regular riders or less.
It's important to maintain objectivity. I love trains - I drag my family onto them when we vacation - but I don't let my preferences color objectivity.
FL
Fare subsidies won't do it, foremost it must be convenient and next it must be fast. You could pay me to take a bus that came only once or twice an hour but I still wouldn't use it. Can't be more than a 10 min wait but preferably 5 min at peak times. And it must go late. I hate that the last Sausalito ferry is 8 pm. Makes it impossible to stay late without a car (and don't make me laugh by mentioning the late bus… what a joke).
RH
You say "fare subsidies won't do it" but there's a strong body of evidence that:
- reducing fares increases transit ridership;
- implementing rail causes an upward pressure on all transit fares (including buses) causing an overall decline in transit ridership system wide.
The details are published on my website - but this article itself was authored by Thomas Rubin, the CFO of Southern California Transit during the time when fares went up and down due to rail (with inversely proportional transit ridership):
http://planningforreality.org/...
As a mathematician you should find the cause and effect interesting. I also agree with you that convenience and speed are major factors - like you say long wait times are an issue, so are connections which add risk of more extended delays.
If people try something and it's not reliable they quickly return to their former more reliable methods.
FL
No question that raising fares reduces ridership, I'll completely agree with that. What I was responding to was the statement that fare subsidies will **increase** ridership. Sure, if you have an existing system that goes someplace that people already use, then fare subsidies will likely increase ridership at the margin.
But, if you have a system, like Marin's, that doesn't support the transit habits of the vast bulk of the populace (> 95%?), then transit subsidies will have essentially zero impact. (As I said before, you could try and ***pay*** me to take the bus to San Francisco during off peak commute hours, but I still wouldn't take it because it's slow, it isn't convenient, and it doesn't go anywhere I want to go.)
If one really wants to attack our traffic problem with public transit, then the transit system needs to provide fast, convenient last mile service to within say 400-800 yards of the homes of 90% of the populace. The Smart train could be the cornerstone of this system, covering the primary north south corridor transit, with east west travel corridors that go out the major arteries in Novato, San Rafael, Larkspur, San Anselmo, Fairfax, Mill Valley, Tiburon and Sausalito.
You mentioned that someone is talking trolleys out Sir Francis Drake Blvd. Cool! But instead of slow old-fashioned "trolleys", what about the fast Porsche trams they're using in Prague (see https://en.wikipedia.org/wiki/.... We took these a couple of Christmas' ago, and they were incredible, smooth and fast. Imagine SFD Blvd with a dedicated tram line where one could travel from Fairfax to the Larkspur Ferry in under 15 minutes (it is only 6 miles after all) every 15 minutes all day and night long. Combining that with ferries that traveled more regularly into SF and beyond (Oakland, Alameda, Palo Alto) would be transformative, and would likely reduce the need to do anything about the 101S interchange.
Is there the political will to pull off something like this? I don't know, but my experience in Gothenberg showed how transformative an amazing mass transit can be.
Speaking of traffic, does anyone know if there's an overview document that describes what steps are being proposed over the next 10 years for traffic mitigation in Marin? I'm thinking of something that talks at both a high and low level. From how to fix the Larkspur interchange to fix the afternoon 101 North problem, to how to implement rotaries in Mill Valley to fix the morning school commute problem.
As an engineer (electrical, computer science) and mathematician (operations research, queuing theory) by trainin, I shake my head every time I go by the Tennessee Valley traffic light because it's so obvious what the problem is and how to solve it (from a queueing theory perspective).
RS
Doesn't exist. TAM, our own planning agency, is hopelessly mismanaged. In more than a decade, they have been unable to even address the 101 / Sir Francis Drake congestion in any meaningful way. MTC (the major source of regional transportation funding) is focused now on funding more and more high density development, not new or improved transportation systems (this is what the battle for planning control between ABAG and MTC has been about). MTC has stated that they are even intentionally not fixing roads to make driving more difficult. How this is going to work out without funding massive improvements in public transportation remains a mystery.
RH
Startlingly Dianne Steinhauser states that SMART is the number one project to mitigate traffic congestion in Marin. In contrast SMART's Farhad Mansourian presents that SMART is an "alternative". In other words it doesn't mitigate traffic at all.
This is a major contradiction and disconnect with the #1 concern of most Marin residents. Really money should be prioritized to add capacity to the Novato Narrows, Greenbrae Interchange and Richmond Bridge.
In 10+ years autonomous cars may start to help - and these again will use roads.
FL
I would love to be part of a working group if there were such a beast. And if not maybe we could start one? I keep thinking about creating a ppt to explain and highlight the issues in Mill Valley and possible solutions but need an impetus to get it started and a forum to think and discuss.