MBTA Image

Bruce recently wrote about the “cheaper, brighter future of American passenger rail” thanks largely to the relaxation of rules that will allow lighter rolling stock on conventional tracks. While the rule change doesn’t come close to solving all of the problems of U.S. passenger rail, allowing lighter trains does go a long way to reducing costs and improving service on legacy tracks. Last Thursday, the Massachusetts Department of Transportation released an exciting 5-year capital plan that would invest $252 million in Diesel Multiple Unit service.

The MBTA commuter rail’s 12 lines serve a ridership of roughly 140,000 each weekday. Even though there is already mid-day service on all lines, they are not nearly as useful as they could be. With the exception of a few stations such as Porter Square, the commuter rail interfaces awkwardly with the MBTA’s rapid transit lines, and the system has very few useful infill stations. Crosstown connectivity is notoriously terrible citywide, and many inner neighborhoods are neglected, including Allston-Brighton, Dorchester, Chelsea, and Lynn.

Enter Boston’s 2024 vision: in addition to implementing DMUs on upgrades already underway – such as the Fairmount Line in Dorchester, a shuttle train from Back Bay to the Convention Center, and the new Boston Landing station planned for New Balance’s headquarters in Brighton – the plan calls for creating Boston’s version of the London Overground: DMU service to Chelsea and Lynn on the Newburyport/Rockport line, and to Medford, Winchester, and Woburn on the Lowell line.

Even more ambitious are proposed new interfaces with subway lines, with a branch from Auburndale to the D-Line at Riverside, and a new line between North Station and Boston University via MIT (though this would require expensive upgrades through Cambridge to be useful). Meanwhile, traditional heavy commuter rail would continue at the outer stations but would run express through the new DMU lines.

This would create a really sensible service pattern, where the outer stations get what they need most (reduced travel times) and the inner stations get what they need most (higher frequency). And all this at what seems to be a really reasonable price tag: $252 million.

MBTA 2024

Stay tuned for a post from Bruce on applying DMUs in our local context.

57 Replies to “Boston’s Exciting DMU Plans”

  1. PERFECT SOLUTION for many of North Sounders inept blunders to date.
    Today: NO mention the entire line is shut down on ST’s home page. (you have to really dig for it)
    Last Year: Pathetic ridership on the line is answered with buying more free parking in Edmonds.
    Previous Years: 3,000 daily riders has never been met, and the cost per rider is in the stratosphere.
    Negotiations w/ BNSF: Purchasing 4 time slots for 1/4 Bil with no escape clause was INSANE.
    Planning leading up to N. Sounder: Many from SnoCo in transportation planning saw this train wreck coming.
    Buy one DMU and run or don’t run the 4 time slots daily to preserve ST’s operating slots in perpetuity.
    It’s time to cut their losses and do something different.

      1. YES. With multiple buttons to click on, you have to select the right one to find out an entire line is shut down.
        OH, did I mention the entire South Sounder is shut down this morning?
        BTW, I love the 1/4 page ad for Jack Frost and the really cute yellow boots on the home page.
        (priorities people!)

    1. Sounder will be kept alive because of who rides it, not how many. The typical passenger isn’t coming to downtown and going to their job at Claire’s selling barrettes and scrunchies. The people on those trains are the who’s who of ST, The Port, and King County government, and as long as the region’s movers and shakers ride those trains, Sounder won’t be going away any time soon, no matter how dismal the ridership or expensive the service.

    2. I can’t imagine why anyone who lives in Edmonds (or north Kitsap, for that matter) and works in downtown Seattle would want to fight the I-5 rush-hour traffic when the Sounder gets there in only 29 minutes. The problem, of course, is that the population in south SnoCo isn’t in Edmonds–it’s centered around Lynnwood.

      On the other hand, it would seem ST’s missed a golden opportunity by not adding a stop in Ballard.

  2. Sounder North is a glaring example of the problems created by a combination of sub area equity, and uniform taxation. Without it, there wouldn’t be enough money spent for Snohomish county, and so the tax rate there would have to be lower. But because of uniform taxation, this means everyone’s tax rate has to be lower, which in turn would mean that there’d be less money to do things elsewhere.

    The bottom line is that everyone’s ability to fund projects is limited by the least ambitious subarea. The problem is that the political requirement of sub-area equity is causing a misprioritization of projects.

  3. An imaginative and resourceful vision from the T.

    Unfortunately, our existing rail lines are not as well positioned to make a difference as those in Boston. In particular, we don’t have anything with remotely the potential usefulness of that underused little freight track through Cambridge. For a comparison, imagine if we had an east-west surface line from Ballard to Sand Point via Ravenna that was used for just a few freight trains a day.

    1. How about Ballard to Duvall via the U-District and Lake Forest Park?
      Or SLU to Ballard via Freemont?
      Or Renton to Snohomish, Woodinville to Issaquah and Monroe to North Bend?

      Oops, we ripped all those up.

      1. The Burke-Gilman corridor is — and was always — far too winding and indirect to provide efficient passenger serving in the modern era, and is not particularly analogous here.

        The Grand Junction connector, as David says, shoots arrow-straight across Cambridge’s original water line (and the arguable geographical center of the Boston-Cambridge urbanized area), providing cross-connections to points that can otherwise require an out-of-direction slog on train or bus.

        Unfortunately, the success of this particular aspect of the plan will require far greater capital expense than any of the others. Whereas 100% of expenses on the radial “Overground” network (Lynn, Woburn, Allston/Newton, and Dorchester/Fairmount) lay in purchasing equipment, building infill stations, upgrading signals, and figuring out fare integration, the Grand Junction connector would require a total rebuild to be useful, involving significantly expanded double trackage and full grade separation at Mass Ave and ideally at Main and Broadway.

        Mass Ave separation is non-negotiable — even today’s very rare locomotive movement can cause a rush-hour clusterfuck on this north-south crosstown bus and general-traffic corridor of paramount importance. Main/Broadway separation would no doubt become a political sticking point, and would be made exorbitant by the existence of this and by the Red Line immediately below the surface of Main.

        If MassDOT and the MBTA try to cheap out on this one, it will fail as hard as Nathanael’s Sand Point hypothesizing.

      2. Yeah, I just can’t see the Grand Junction part working out.

        Frankly, even though it would be really expensive, I’m still holding out for a real North-South Rail Link, where every north-end commuter rail route is through-routed with a south-end route. You’d have a two-seat ride between pretty much any two points in the Boston metro area, and even to other places in the Northeast Corridor (assuming that the tracks underneath North Station and Central Station were electrified). I think that would be a much better use of money than however much it would end up costing to completely rebuild the Grand Junction Line.

      3. A Grand Junction done on the cheap would be a disaster.

        A connection between North and South Stations would be nice but it doesn’t seem like a high priority – particularly since folks coming in at North Station can connect to Amtrak trains at Back Bay Station via the Orange Line. I’m not sure what either the current demand or potential ridership would be and if that increase would justify the associated cost.

      4. The only way to grade-separate the Grand Junction is road-over-rail. Which is going to be disliked extensively by the locals. Seems very unlikely, unfortunately. N-S Rail Link would be more popular.

      5. Yeah, road-over-rail can be really bad for pedestrians, and the Main/Broadway area is becoming more pedestrianized by the day.

        Rail-under-road would be quite feasible at Mass Ave.

      6. A North-South Rail Link has a whole pile of benefits. First, let me sketch out the original plan that was sent for federal approval:

        – Build a pair of tunnels, with a total of four tracks, that go between North and South Station (and further south).
        – Build new underground platforms at North and South Station for these new tracks.
        – All service running through the tunnels would be electric-only.
        – Build a new underground Central Station, near where Aquarium Station is today.
        – Extend the Northeast Corridor (electrified) north to Woburn.
        – Gradually electrify MBTA service. As it’s electrified, shift it to the underground tracks, and through-route north-end and south-end service.

        The benefits include:

        – In-city transfers: There would be a direct connection between every electrified commuter train, and every subway line. You could easily transfer to the Green/Orange Lines (North Station), the Red/Silver Lines (South Station), or the Blue Line (Central Station). This avoids a lot of “friction” in the current system.

        – Suburban transfers: Right now, if you want to go between the northern suburbs and the southern ones, you don’t have many good options. (Nevermind the fact that many commuter rail stops are within the city of Boston!) For example, if you want to get from Chelsea to Fairmount — a trip between two urban places — you have to take a train to North Station, then two subways to South Station, then a train to Fairmount. The Rail Link would reduce this from four to two.

        – Operating efficiency: The underground North/Central/South Stations would be through stations, not terminals. Electrified trains from the north would continue through as electrified trains from the south. This makes it possible to serve a higher number of passengers with a smaller footprint. It also minimizes dwell time, which means that you can get more service out of the same amount of platform hours.

      7. @Alex-

        I do think the connection between North and South Stations is a good idea for the reasons you mentioned. I don’t think there’s much will behind it and I don’t see it as a priority for the folks at the T unless there’s a big check coming in from the feds.

      8. @Alecks,

        Whoa up, sir. The path south from North Station has already been claimed by the Big Dig. Just like Seattle, each tunnel has been placed where it was easiest at the time, with little or no thought for future expansions.

        Remember that heavy rail commuter trains — even DMU’s — have much more rigorous grade and curvature constraints than do subway or light rail trains.

        I doubt Boston could do what you’re proposing for less than $20 billion. Nobody is going to stump up that much money for through-running DMU’s.

      9. The platinum-plated, diamond-encrusted, Cadillac-mounted-on-a-Cadillac plan for the Big Dig would have built the North-South Rail Link on its own level, directly below the I-93 roadway, for an extra couple of billion.

        Obviously, this did not happen. But I can’t remember whether they went through with the lesser plan of sinking deeper highway support beams to allow a lower level to be hollowed out in the future, with minimal disruption and no need need for a separate, “new” tunnel.

        If the engineers did this, then the North-South Rail Link remains at least a hypothetical possibility, if it can ever be mathematically justified. If this was not done, then what Aleks describes will never, ever come to pass.

      10. I don’t see how you could possibly fix up the Grand Junction line for less than a billion dollars. Most of the line is one-track, including the bridge. And it runs through some very narrow rights-of-way that are politically and/or technically impossible to rebuild. Grade-separating the entire thing could end up being a huge undertaking, even if you just do cut-and-cover. Not to mention the fact that much of the *underground* right of way is already taken up by MIT.

        And if you spend this billion dollars, what you’re left with is pretty much one interesting station (Boston University). And the best part is, the trip from North Station to BU is already a one-seat ride on the Green Line!

        The North-South Rail Link would be a megaproject, on the order of the Big Dig. But the benefits would be much broader. The estimates I’ve seen suggest that the cost would be somewhere on the order of $3-10 billion, not the $20 billion that Anandakos claims. And it would simplify hundreds, if not thousands, of different trip pairs.

        I’d rather spend $10 billion dollars on a useful project than $1 billion on a very dubious one.

      11. “Rail-under-road would be quite feasible at Mass Ave.”

        Water table? Building foundations? Basements? Under-street utilities like water and sewer? Getting back up to the surface in time to cross over the RED LINE at Main Street?

        Lowering the Grand Junction line at Mass Ave enough to go under the existing road grade is not very practical. You might possibly be able to do half-and-half (lower the tracks some, raise the road some).

        Main Street is an even bigger grade separation problem, by the way. The North-South Rail Link is probably cheaper than trying to grade-separate the whole of the Grand Junction.

      12. “But I can’t remember whether they went through with the lesser plan of sinking deeper highway support beams to allow a lower level to be hollowed out in the future, with minimal disruption and no need need for a separate, “new” tunnel.”

        I believe they did. The messy part is the approaches and the underground stations for North and South.

      13. It occurs to me that there’s already a tunnel that vaguely connects the Northeast Corridor to North Station. I’m curious whether anyone ever considered expanding or repurposing the Orange Line tunnels, as an alternate way of closing the gap. This obviously isn’t a slam dunk — you still have to build something, and you potentially have to reroute the Orange Line, which could be temporarily or permanently disruptive, and you probably need a major expansion/rebuild of both North Station and Back Bay, and this isn’t a solution for the Fairmount and Old Colony lines. Still, this seems like the kind of thing I would expect to find in an EIS, if only to explain why it’s a bad idea.

      14. Nathanael,

        The three of us agree that a “low-budget” upgrade of the Grand Junction is not as simple as its boosters might make it out to be. But please try not to get carried away.

        1) Memorial Drive dips under Mass Ave, below the river level, without trouble. Dozens of adjacent MIT buildings have full basements. There is no “water table” impediment to a Grand Junction undercrossing that couldn’t easily be solved with the cheapest available form of retaining wall.

        2) Mass Ave and Main Street are 1/3 mile apart. Yes, there’s plenty of room to underpass Mass Ave and return to grade. Or, for perhaps tens of millions, you could pass under the Red Line, which is immediately below the street, without digging especially deeply.

        I’d like to see the rail “layer” of the Big Dig resurrected as a long-term possibility, but we’re still looking at a couple of billion dollars to make that happen. Meanwhile, I don’t think an at-grade cheap-out on the Grand Junction is a good idea.

        But “the North-South Rail Link is probably cheaper than trying to grade-separate the whole of the Grand Junction” is most densely-compacted pile of bullshit you’ve ever written. It would still be an order of magnitude cheaper to do the Grand Junction right than it would be to dig out the space beneath I-93.


        It’s an interesting idea, but it doesn’t seem remotely feasible. The Washington Street Orange Line tunnel dates to 1908, and follows such a skinny ROW that about half of it is already stacked. There’d be no digging below it without destroying the present line, interrupting the Orange Line for years, and destroying most of the street above it in the process.

        You’d be looking at billions more than the billions needed to dig the under-93 space.

      15. I agree with you that a grade-separated Grand Junction would be an order of magnitude cheaper than the real deal. I just don’t think it’s worth the money. Let’s say it costs a billion dollars. What do you get for your money? Maybe one or two unique stations (somewhere in Cambridge and somewhere near BU). Maybe, if you really wanted, you could run some trips from Worcester or Riverside to North Station. Or maybe you could reroute the Lake Shore Limited to free up some space at South Station.

        These are kind of nice things, but they aren’t worth anywhere close to a billion dollars, especially when you consider that pretty much all of these trips are already possible today, and that some of them (like North Station to BU) are already one-seat rides!

        I’d rather spend $10 billion on a game-changer than $1 billion on a dud. The whole point of the DMU plan is to create a meaningful service improvement for a tiny amount of money; the Grand Junction part of that plan just doesn’t belong.

      16. Funny… I was all set to agree with you that it wasn’t worth the money. And then you suggested a Riverside-Brighton-Cambridge-North Station line, which is actually a pretty good idea. ;-)

        If I were to ballpark it, I’d guess that it would cost only $100-$200 million to get the Grand Junction right. That’s a lot more than MassDOT seems to be pledging to it today, but it’s a lot less than $1 billion. Meanwhile, if Nathanael is right that the Big Dig left excavate-able space (w/supports) beneath itself, you’d be looking at $2-$3 billion for that project, which is fortunately far less than $10 billion.

        Is a $100 million Grand Junction worth it? Maybe. Maybe not. If only there were still such a thing as a feasibility/cost-benefit study without a forgone conclusion baked into it!

        Either way, I appreciate your parsing as always. Nathanael’s proclamation that a Grand Junction rebuild (still mostly at grade, with a handful of 90% open-trenched grade separations) would cost more than the RER-ing of downtown Boston was infuriating.*

        *(And sadly typical of his pronouncements about places he hasn’t spent much time.**)

        **(Which is most places. Because he lives in Ithaca and literally will not travel where or when a train can’t take him.***)

        ***(Which is sad, because he’s smart and a little broader experience might tone down some of his most dubious rhetoric.)

    2. In my younger, *far stupider* years while I was a grad student at BU, I was told by a friend that those tracks were abandoned. So I thought I’d walk across the rail bridge in the middle of winter for fun, over the frozen Charles River. And of course, a repositioning Amtrak locomotive came within probably 50′ of hitting me. I had just enough time to jog over to the Cambridge side. If I’d had to jump, I’d probably never have been heard from again.

  4. Just FYI, for those unfamiliar with Eastern Massachusetts geography: the new map is super extra not-to-scale.

    1. Indeed. Most of the Green Line extension to College Avenue runs along the Lowell Line right of way (if cuts over from Lechmere and joins the Lowell Line around Washington Street. So along that one right of way you’d have Green Line Service to College Avenue, then DMU service to Anderson, and traditional Commuter Rail to Lowell (and maybe into New Hampshire someday)?

      Map of Green Line Extension:


    1. Okay, I saved the most important reply for last…

      This is actually why I warned of the not-to-scaleness of the map above, and why David L. lamented the poor positioning of Seattle’s existing rail infrastructure, and why it is probably a good idea to see what cheap and locally-applicable DMU corollaries Bruce comes up with before we speculate into a void about the transposability of this heavier-lifting proposal.

      For someone not familiar with Boston geography, the map might lead you to believe that the MBTA intends to flood the inner half of its commuter-rail network with high-frequency DMUs. This couldn’t be further from the truth. While the outer commuter spindles stretch as far as 63 miles from downtown Boston, the DMU proposal resides entirely within the 12-mile radius that is the Boston area’s fully-urbanized agglomeration of contiguous cities and inner suburbs.

      That this proposal is intended to solve problems of missing connectivity within and across the interconnected urbanized area cannot be overstated, because those interconnections are where the true multi-directional and all-day-frequency demand comes from. On the most important lines (Fairmount and Allston/Newton), infill stations will yield a final stop spacing of 3/5 mile and 1 mile, respectively (again, impossible to tell from the map). This isn’t “interurban” — this is “intra-urban” to its core.

      It’s not that Renton-Seattle doesn’t have all-day demand of its own. (It clearly does, though not at high-capacity-transit levels.) It’s that the demand it has would not be well served by a Boeing/SoDo corridor that misses literally every intermediate destination and 95% of the demand. There’s simply no urbanity and no corridor where the tracks currently go.

      Mistaking the Boston plan as a distance-enabler, when it is thoroughly urban in approach, may lead to analogies that are severely lacking.

      1. But why wouldn’t a multi-modal station that would allow buses, light rail and DMU service make sense? It seems a pretty natural focal point to me. In this case, it shouldn’t matter that the particular location is not in a dense residential zone but it does serve a large industrial zone. Bus services that currently come to downtown from South King County could be truncated at the B.A.R. station and reduce congestion downtown and on the freeways approaching downtown.

      2. Because as a general rule, high-capacity proposals that serve zero places directly, and that aim to garner 100% of their ridership through feeder applications and forced transfers, do not provide enough aggregate benefit to the region to justify the transfer penalties and 3-seat rides they induce.

        I’m heartily in favor of high-capacity trunks with low-pain transfers. Indeed, my experience as a Bostonian tells me that this is the only way public transit can perform the heavy lifting required to garner major metropolitan modeshare. But your trunks absolutely must go to and through the real places that people are trying to reach. People don’t respond well to being sent on wild goose chases just because those geese inhabit existing train track.

      3. (p.s. “Large industrial zone” is synonymous with “no places directly served”. It’s not that industrial zones aren’t important — they obviously are — but that by their very nature they are sprawling places, with low “employment density” and little chance of improvement in the forseeable future.)

  5. I love the idea of using DMUs to provide higher-frequency service to the inner suburbs, especially the ones that aren’t well served by the T right now. I think the lines to Chelsea/Lynn and to Woburn (Amtrak) are great, and would be very popular. I particularly line that many of the DMUs are running just as far as the nearest Amtrak station; for example, if you want to get from Portland (Maine) to Medford, you can transfer at Anderson/Woburn and take DMU service the rest of the way.

    In practical terms, there are a few things I’m surprised/disappointed to see.

    The biggest thing is that the MBTA should be running *electric* service on the Providence Line (either with EMUs or with electric locomotives). The whole line is already electrified; why not take advantage of it?

    Long-term, I would also like to see the following associated changes:

    – Electrify the Needham Line. Or, even better, convert it into an Orange Line extension.
    – Reroute the Franklin and Stoughton Lines to use the Fairmount Line to get to Boston, rather than using the Northeast Corridor. (This might require upgrading the Fairmount Line to 3 or 4 tracks.)
    – Figure out a way to allow the Franklin and Stoughton Lines to reach Route 128 Station, without putting diesel trains on NEC tracks. That would allow Route 128 to be a convenient transfer point between virtually all south-side service, which is really important if Franklin and Stoughton trains are no longer stopping at Back Bay. And it would also provide a high-frequency DMU connection between Route 128 and downtown.

    Together, these changes would remove all diesel service from the Northeast Corridor in Massachusetts (possibly excepting freight trains?). They would reduce congestion (and pollution) on this important corridor, and they would enable higher frequencies and better scheduling.

    The Grand Junction Line is a mess, and turning it into anything practical would be an enormous undertaking. The real solution is the North-South Rail Link (though that’s obviously quite a bit more expensive).

    I’m surprised that they aren’t proposing DMU service to Brandeis. I think it would be very popular.

    It seems like a shame that there isn’t an Amtrak station somewhere in Newton or Wellesley. Framingham is a bit too far to need the frequency that you would get from DMUs. But I think an Amtrak station that was easily accessible from Newton would get a ton of use, especially with the additional branch to Riverside. Maybe the “West Station” on the map would become an Amtrak station, too.

    1. The MBTA’s allergy to electrification seems to be worse than GO Transit’s, and no, I don’t understand it either.

    2. Ordering a homogeneous fleet gives a lot better operational flexibility, may reduce the cost of maintenence, and might even bring down the per unit cost of the order. Of course EMUs would be nice to have. It would almost certainly be more environmentally friendly to operate EMUs. And the higher availability and greater acceleration of EMUs would be welcome. But it really looks like they are trying to do an awful lot with a comparitively small amount of money, and the numbers really might work out best the way they have it.

      1. The thing is, they *don’t* have a homogeneous fleet. They have both single-deckers and double-deckers (the latter only run on certain lines from South Station). They also run a handful of different engines, including some that they appear to borrow from Amtrak. I think the Old Colony Line also uses special equipment that has electric doors,

        And, of course, on the rest of the MBTA, each of the lines — Green, Red, Orange, Blue, Mattapan — has completely separate equipment and signaling. This is understandable for the Green Line (light rail), and the Blue Line (switches from third rail to catenary), but there’s no reason the Red and Orange Lines couldn’t share equipment if you wanted to.

        And there’s also the *four* trackless trolley routes that the MBTA runs, out of 183 bus routes. If homogeneity was the goal, these would have been converted to diesel operation decades ago.

        In this case, electrification would be a major service improvement, both in terms of frequency, reliability, and environmental effect. Given that the Providence Line has the highest ridership of any MBTA commuter rail line, the cost of heterogeneity seems to be worth it.

      2. Kevin, it seems to come and go. When I was a kind, I remember taking double-deckers on the Newburyport Line. Then they seemed to completely disappear. Then I moved to Providence, and the double-deckers were back. Maybe they’re running them into North Station again.

      3. I’m from Winchester and have consistently seen them going through town when I’m back there (as recently as a month ago).

      4. Just because they currently manage a heterogeneous fleet doesn’t mean it makes sense to add both a new DMU class and a new EMU class instead of a single DMU class. My understanding is that at least some of the lines that would be served by the new MUs are not currently electrified, and I would be surprised if electrification of these lines didn’t bust the budget, badly.

      5. Oh, DMUs would be quite reasonable (if the FRA problems are ever resolved), but they’re no substitute for electrification.

        It would make sense for MBTA to slowly start electrifying the south side lines, given that one of them is *already* electrified, and the others all branch off from it at one point or another. Long-term planning from the state government is going to require electrification of New Haven-Hartford-Springfield and eventually Springfield-Worcester-Boston for intercity rail anyway.

        Unless N-S Rail Link is funded, electrifying the north side lines makes less sense.

        I know MBTA swaps equipment around between the north and south sides, but really they do it rarely enough that it would make perfect sense to start building up an electrified fleet on the south side.

      6. there’s no reason the Red and Orange Lines couldn’t share equipment if you wanted to.

        The Red Line is like three feet wider than the Orange Line. And longer. And the ceilings seem taller. And my hunch is that the platforms are higher.

        The Cambridge Subway was the original very-high-capacity rapid transit experiment.

      7. The Red Line is like three feet wider than the Orange Line. And longer. And the ceilings seem taller. And my hunch is that the platforms are higher.

        Right. What I mean is, fundamentally, there isn’t some inherent technical difference between the systems that makes it impossible to share equipment. If you wanted to, you could widen the tunnels a bit. (New York did this on several lines, when converting from one of their incompatible subway systems to another.)

        Anyway, my point isn’t that this should happen or that it needs to; it’s that the MBTA hasn’t historically placed a high value on homogeneity of equipment, and I think it would be a mistake to accept inferior (non-electrified) service in exchange for maintaining that homogeneity.

    3. Re: Needham:

      The Needham Heights end of the line is barely two miles from Newton Highlands, along partially-intact trackage. There has been some level of official discussion of severing the commuter rail at Needham Junction and attaching Needham Center to the Green Line.

      Of course, this runs up against the Green Line’s capacity problems, but with the DMU branch picking up much of Riverside’s commuter traffic, the splitting of service at the most suburban end of the D Line at peak would be justified by the vast improvement in connectivity to Needham. (Also, Needham trains could be turned around at Kenmore if need be.)

      Re: Fairmount:

      This is the single most vital application of the new DMU program — a staggeringly populous, transit-dependent corridor that was the victim of decades of environmental racism on the part of MBTA administrations past. This is the one that is most important to get right. And that means predictable trains with 100% level rolling stock coming, eventually, every 10 minutes all day. It means complete fare integration with the rest of the rapid transit network, ideally someday including a dedicated platform at South Station within the fare-paid area and with its own ramp to the Red and Silver Lines.

      Running a handful of the more peak-focused outer lines over this branch would be a distraction from the line’s upgraded purpose. Furthermore, a large enough percentage of Franklin and Stoughton commuters are headed to Ruggles and Back Bay that it makes little sense to re-route their commutes for nebulous reasons.

      1. I agree with you about the importance of getting Fairmount right. My assumption was that part of my proposed change would be upgrading the Fairmount line to 3-track or 4-track service. The express trains to the suburbs — some of which *already* use the Fairmount line — would be completely separate from the “Indigo Line” service.

        The “nebulous reason” is that the Northeast Corridor is chronically overscheduled. That’s why some Franklin Trains use the Fairmount Line. That’s also why the Providence Line has a dismal on-time performance record.

        FWIW, the Providence Line has the highest ridership of any MBTA commuter rail line. The Framingham/Worcester Line (also a shared line with Amtrak) is second, and nothing else even comes close. The high and growing ridership of the Providence Line suggests that the line could easily accommodate even more passengers, if the service were more reliable and more frequent. The only way to do this is to reduce contention or increase flexibility. This means running fewer non-NEC trains on the *existing* NEC tracks.

        You could build more track on the NEC, but that involves coordinating with Amtrak, and anyway, the corridor is so space-constrained that I highly doubt it’s even possible to build more track without elevating or tunneling. In contrast, building parallel express tracks along the Dorchester (Fairmount) Line seems much more feasible, and it’s something that the MBTA could do completely on their own. And it also lets you eliminate diesel locomotives from the NEC, which improves scheduling as well.

        I don’t dispute that it would be unfortunate to lose the direct connection to Ruggles and Back Bay. The mitigation is that you would be able to transfer at Route 128 to a very-high-frequency train that would take you to either of those two stations.

        My point about the Needham Line is that it’s a tiny line — shorter than the Red Line to Braintree — and it just so happens that all of the line’s unique stops immediately follow the terminus of Orange Line. It seems like an Orange Line extension would be relatively cheap — you’ve already got the track and the stations, so all you need is some third rail, fare gates, and possibly higher platforms. That seems cheaper than acquiring new ROW to connect Needham with the D Line, and also better, since you don’t lose any of the existing stations. But maybe I’m missing something?

      2. When the Orange Line was moved into the Northeast Corridor in the ’80s, the intention was to extend it beyond Forest Hills. I don’t remember if it was supposed to continue toward Needham or toward Hyde Park. Regardless, the extension was cancelled because, at least at the time, demand was too low to support anything resembling full subway service — West Roxbury and Roslindale were solidly “car culture” at the time.

        Demand along the existing Needham Line ROW remains low — so low that weekend service was withdrawn, though who knows how much of that is the chicken/egg problem inherent in lousy service. With Roslindale newly acting as a release valve for Jamaica Plain gentrification, extension of the Orange Line should probably be revisited, but Hyde Park/the Northeast Corridor probably provides a better demand fit than West Roxbury and Needham.

        Anyway, here’s what you’re missing about the Green Line idea. The ROW already exists, and the T already seems to own it.

      3. (The fact that you probably didn’t realize those two places were so close is precisely what this connection would aim to rectify.)

      4. The same could be said for Auburndale and Riverside, while are only ~1/2 a mile apart but almost no one knows that.

      5. Precisely. And on top of the added urban/inner-suburban connectivity, Riverside commuters and the baseball throngs would see their trip times cut nearly in half.

        That little ½ mile is a remarkable win-win for both commuters and all-day users.

      6. I knew that Needham was pretty close, but I didn’t realize that the necessary track was already in place!

        Well, that makes sense for Needham Junction and the stops to its north, though it would be unfortunate for folks boarding to the west of Newton Highlands. But like you say, with commuter rail service at Riverside, that might not be a huge loss.

        I still think it would make more sense to provide real rapid transit to West Roxbury and Roslindale. You could imagine branching the Orange Line, and having one branch head west to Needham Junction, and the other head south to Hyde Park and Readville. As of 2009 (can’t find a newer source), West Roxbury and Roslindale Village both had double the daily ridership of any station on the Fairmount Line. In other words, I think both of them are examples of places that would have much higher transit usage if they had transit service worth using.

        Of course, this still wouldn’t address my single biggest complaint about the Green Line, which is that it commingles reliable grade-separated service with completely unreliable at-grade service. I have no idea how to solve that problem without spending a giant pile of money, or without making some people very unhappy (e.g. by terminating all B/C trains at Kenmore, and all E trains at Copley). But in the meantime, if I lived in West Roxbury, I can say for sure that I would definitely prefer an Orange Line extension over a Green Line one.

      7. In 2009, the Fairmount Line only had two urban stops and ran about six times a day, so even that isn’t a particularly fair comparison.

        Your counterpoints are valid, but I think we need to be realistic about how much all-hour demand drops off past the 7- or 8-mile radius before we talk about building two new branches of multi-mile, double-tracked, turnstile-controlled Orange Line subway.

        Fun fact: Did you know that when they first moved the Green Line underground at Kenmore (1932), the intention was to continue tunneling west under Commonwealth, eventually turning the B Line into a full-scale subway? That’s why the Kenmore turnback exists: so C Line streetcars could reverse direction after dropping passengers for a cross-platform transfer to the heavy-rail trunk. (The D and E lines didn’t yet exist.)

  6. Repurposing rail lines for DMU service can be cost-effective if the situation presents itself (like it does in a few places in the Boston area). Doesn’t MBTA control most of the track for these proposed new DMU services? In our potential corridors in Seattle, wouldn’t restoration costs be really high (as compared to these corridors in the Boston area)?

  7. MBTA also controls how big their stations are too – some concrete, a shelter, and not much else.
    Where’s the huge ass mezzanine, glass and steel art-deco treatments, under and overpasses and full length shelters in the photo?
    “That’s no way to run a railroad”

    1. As I’ve said before, the consensus that (relatively) high-performing transit with a high degree of scalability is required for the Boston area to function* has forced a minimum standard of competence from the ever-imperfect (and occasionally malignant) bureaucrats that have run the MBTA in my lifetime.

      That said, the T has not been immune to some of the sillier trends in American public transit in recent decades. Here, for example, is the mezzanine of a BRT tunnel station completed in 2004. The main fare gates are behind the camera, as is the street access. There is no exit of any sort at the other end. To date, no human has ever crossed this expanse, except to take luminously purple photographs of the expanse itself.

      Did I mention that surface buses on Summer Street, a block away, run faster than buses in the tunnel below? Oy!

      When home in November, I did observe that Yawkey station (the first to be reconstructed on the Auburndale line; it can be found between Back Bay and a new “West Station” on the map above) has been transformed from this into this. So it’s not as if they’re suddenly taking minimalism to an extreme. On the other hand, there’s nothing here remotely as extravagant as the purple money pit above. And once the DMUs are running, Yawkey may become a major urban transfer point, and these platforms may host tens of thousands of baseball enthusiasts on game days.

      Hopefully the scope of the DMU proposal, and its low pricetag, reflect a renewed commitment to right-size every segment and every station in pursuit of effectiveness over showiness.

      *(by contrast with Seattle, where a significant portion of the body politic continues to see transit as optional or, even worse, as a whimsical new civic showpiece that must look better than it works)

Comments are closed.