Crazy Transit Pitches

[ritchiew]

Are there streets around Boston that are wide enough to support elevated rail,

Revere Beach Parkway is plenty wide and could work as part of a ring (though that's not typically the chosen path).

How wide a street do you think you'd need? Just as a jumping off point, the new Lechmere is on a relatively nice, quiet viaduct. Cambridge Crossing was willing to put up buildings about 15ft from the trains. Lechmere Station itself if about 60ft wide. So very roughly, you might consider rights-of-way more than 90ft wide. Looking past an urban ring, you can find streets that wide. Blue Hill Ave, Warren St, Columbia Road, Fellsway, Broadway (Somerville), Washington where the Orange el used to run (though I agree with @TheRatmeister , a lot of Washington is pretty narrow. The trains through Egleston must have been loud.)

 

[bigeman312]

Are there streets around Boston that are wide enough to support elevated rail … I'm thinking either Miami People Mover style, if it's a short distance …

Some candidates that come to mind:

• Logan Airport, connecting the airport terminals, Airport Station (Blue Line), and the rental car center.
• Between North and South Station, either over:
  • some combo of Purchase St, Cross St, Atlantic Ave, and Surface Rd, ideally steering clear of the actual Greenway footprint for the most part
  • Congress (to either North Washington and Beverly or Merrimack and Causeway)
• Extending the North-South People Mover past North Station to Sullivan via one of the following:
  • Rutherford Ave (widest)
  • Main St (most centrally located)
  • Bunker Hill St (most underserved currently)
• Extending the North-South People Mover past South Station to Seaport via:
  • Seaport Blvd
  • Summer St, hitting BCEC on the way, possibly continuing on to City Point
• Looping the North-South People Mover via Causeway, Staniford (connecting to Charles/MGH), Charles (connecting to Public Garden and Boston Common), Boylston (connecting to Boylston Chinatown Stations), and Essex St.

… Or Vancouver Skytrain style for medium distance …

Many, many candidates that I want to come back to this.

 

[JeffDowntown]

Revere Beach Parkway is plenty wide and could work as part of a ring (though that's not typically the chosen path).

How wide a street do you think you'd need? Just as a jumping off point, the new Lechmere is on a relatively nice, quiet viaduct. Cambridge Crossing was willing to put up buildings about 15ft from the trains. Lechmere Station itself if about 60ft wide. So very roughly, you might consider rights-of-way more than 90ft wide. Looking past an urban ring, you can find streets that wide. Blue Hill Ave, Warren St, Columbia Road, Fellsway, Broadway (Somerville), Washington where the Orange el used to run (though I agree with @TheRatmeister , a lot of Washington is pretty narrow. The trains through Egleston must have been loud.)

Agree with a lot of this sentiment.

You are going to have a very hard time convincing residents in Charlestown or Roxbury, who have spent the last 4 decades helping their neighborhoods recover from the blight of the old Orange Line ELs through their narrow streets, that they want ELs again.

 

[TheRatmeister]

though I agree with @TheRatmeister , a lot of Washington is pretty narrow. The trains through Egleston must have been loud.)

I believe this is a photo of Egleston station, even with the black and white photo you can really see just how dark Washington St was, especially near stations.

For any elevated rail you should really consider the area under it to be lost space, in Paris and Berlin for example it's often totally inaccessible. On wider streets though it could be clawed back by using it for street parking.

Blue Hill Ave, Warren St, Columbia Road, Fellsway, Broadway (Somerville), Washington where the Orange el used to run (though I agree with @TheRatmeister , a lot of Washington is pretty narrow.

Of this list I'd strike Warren St, Columbia Rd, and Washington St. Even at Dover on upper Washington St where the street is about 80ft wide the station basically blocked out the sun. I think 100-110 is a more reasonable minimum width.

And so that leaves us with Melnea Cass, Revere Beach Pkwy, Fellsway/McGrath, Rutherford Ave, Broadway, and BHA. Here's my evaluation on those:

Melnea Cass: Would be great for connecting Andrew/Widdett Circle and Nubian.

Revere Beach Pkwy: The section from Broadway to Bell Circle could be useful, but it bypasses the densest parts of Revere. The rest could maybe be an orbital connection between Everett and Revere but again it bypasses the important parts of both and there also just isn't tons of demand for such a connection.

Fellsway/McGrath: Connects nicely to the Grand Junction but doesn't go anywhere particularly useful otherwise.

Rutherford + Broadway: Could be a good corridor between Medford or Everett and Downtown, but a tunneled continuation would be needed on both sides to make it useful, and if we're building a new subway line downtown I think sending it to Chelsea is a better choice, although Rutherford is well positioned to get to Everett.

BHA: What BHA doesn't really need is a fast line to downtown, that already exists and it's called the Fairmount Line. What BHA needs is a faster local service, capable of branching often and running many different services. (Aka, buses and/or trams)

 

[kdmc]

@Teban54 's green line reconfig map suggests that the some northside Green Line trains would use Park St outer tracks to continue up through the Pleasant St incline (near Bay Village) and continue to an el from South Bay, over 93, then over Melnea Cass, to Nubian. The key idea here is that Nubian, as the system's largest bus hub, deserves a rapid connection to downtown that the Silver Line nor a South End surface trolley would be able to provide.

If you look at the dotted lines, there's also suggestion for how that same el could simultaneously serve as part of an urban ring.

 

[ritchiew]

Agree with a lot of this sentiment.

You are going to have a very hard time convincing residents in Charlestown or Roxbury, who have spent the last 4 decades helping their neighborhoods recover from the blight of the old Orange Line ELs through their narrow streets, that they want ELs again.

For sure. I've lived near some unpleasant els, and I'm still kind of skeptical it's a good idea to build more. But there are good els, and they've gotten so much better compared to the old Orange line. They have benefits over subways. If they seem reasonable, it could be worth trying to convince people.

For any elevated rail you should really consider the area under it to be lost space, in Paris and Berlin for example it's often totally inaccessible.

If I thought the space under an el was completely lost, then I'd just say we shouldn't build any. It wouldn't be worth it. It's that simple. But I don't think that has to be the case. They can be pleasantly integrated into a city. They can be nice. Even in Chicago, under the ancient els in the Loop, the streets are busy.

I think 100-110 is a more reasonable minimum width.

Yeah, that could right. I was picking 90 as a bare minimum, in part just so there were a few more streets to consider. I also don't think there's a hard and fast number to pick. It'll depend on the height of the el, the heights of adjacent buildings, setbacks, existing street configurations, and on and on.

BHA: What BHA doesn't really need is a fast line to downtown, that already exists and it's called the Fairmount Line. What BHA needs is a faster local service, capable of branching often and running many different services. (Aka, buses and/or trams)

Yeah, Fairmount will be the transit solution there for the foreseeable future. But I don't think it's absurd to imagine rapid transit down BHA. I don't think I've ever seen even distant plans to have Fairmount run more than every 10 minutes. That would be a massive improvement, but crummy rapid transit. Stop spacing is on the long side, and the stops aren't actually at lot of people's destinations, which are businesses and big apartment buildings along along BHA. Rapid transit on BHA would be such a low priority, but not absurd for a crazy transit pitch, I think.

 

[TheRatmeister]

The key idea here is that Nubian, as the system's largest bus hub, deserves a rapid connection to downtown that the Silver Line nor a South End surface trolley would be able to provide.

I do need to interject with my 'thing' about this proposal, which is that the time savings is somewhere between 4-5 minutes. It would need to be pretty cheap to justify the added expense. If saving time is important, a subway going the full length from Nubian to Boylston would probably be a better use of money even if it costs more. It serves more people and destinations while also shaving off even more time due to the direct routing.

 

[TheRatmeister]

and the stops aren't actually at lot of people's destinations, which are businesses and big apartment buildings along along BHA.

This gets less and less true the further away from Downtown you get. Around Franklin Park the density starts shifting away from BHA more towards Washington St and the 22/23 bus routes at places like Codman Square. Once you get past Talbot Ave the Fairmount Line is pretty much right alongside BHA, and obviously BHA station itself is right there.

I've thought about a rapid transit corridor using the outer tracks and a Washington Subway, but extending it past Grove Hall or Franklin Park has never really made sense to me. A more logical continuation (to me) would be to join the Fairmount Line for the rest of the trip to Readville, although this comes with several other problems too. Maybe this (now well beyond crazy) line could follow the 22 to Ashmont, then terminate at Morrissey?

 

[The EGE]

I do need to interject with my 'thing' about this proposal, which is that the time savings is somewhere between 4-5 minutes. It would need to be pretty cheap to justify the added expense. If saving time is important, a subway going the full length from Nubian to Boylston would probably be a better use of money even if it costs more. It serves more people and destinations while also shaving off even more time due to the direct routing.

The higher reliability of a grade-separated alignment versus surface running, as well as the ability to support higher frequency (ie multiple through-running branches), are also points in favor of a subway or elevated to Nubian. I've posted previous in the Green Line Reconfiguration thread that I believe you need a grade-separated route to Nubian if you want it to serve as a primary bus terminal.

 

[TheRatmeister]

as well as the ability to support higher frequency (ie multiple through-running branches),

As with anything Green Line though, the capacity of the outer portions isn't the problem, it's the trunk. GLRC already puts 2 branches (Riverside/Needham) through the outer tracks, if we want two more branches we're going to need to sacrifice frequency, and if half the people need to wait longer than they save by using the faster route then why bother at that point?

 

[Teban54]

As with anything Green Line though, the capacity of the outer portions isn't the problem, it's the trunk. GLRC already puts 2 branches (Riverside/Needham) through the outer tracks, if we want two more branches we're going to need to sacrifice frequency, and if half the people need to wait longer than they save by using the faster route then why bother at that point?

Except that Nubian routes aren't competing with Riverside and Needham for frequency.

Let's be handwavey and say any GL trunk can support 4 branches. Riverside and Needham only take up 2 of them, and chances are they won't both go to Tremont St (you probably want to send one of them to Seaport). Even if you include a Heath St / Hyde Park / Arborway branch to Tremont, you still only have 2 branches there.

This means 2 additional branches can be sent to Nubian. If you want to maximize capacity on Tremont St, you have to send 2 to Nubian.

 

[TheRatmeister]

If you want to maximize capacity on Tremont St

But do we actually want to do that? Maximum capacity looks like the GL trunk of today, slow and prone to delays. Something like 7 minutes each for Riverside and Needham and 5 minutes for Nubian only works out to just over 2 minutes between trains, headways shorter than that is really just asking for trouble in my opinion. Off peak could be 10/10/7 for more like 3 minute headways, which would mean a fast, smooth ride all the way from Bay Village to North Station and not the stop-starty mess of the current subway.

 

[kdmc]

But do we actually want to do that? Maximum capacity looks like the GL trunk of today, slow and prone to delays. Something like 7 minutes each for Riverside and Needham and 5 minutes for Nubian only works out to just over 2 minutes between trains, headways shorter than that is really just asking for trouble in my opinion. Off peak could be 10/10/7 for more like 3 minute headways, which would mean a fast, smooth ride all the way from Bay Village to North Station and not the stop-starty mess of the current subway.

From the GLR thread I gathered that the flat junction at Copley is a big issue for GL trunk frequency. I assume that Bay Village would be a flying junction.

 

[TheRatmeister]

From the GLR thread I gathered that the flat junction at Copley is a big issue for GL trunk frequency. I assume that Bay Village would be a flying junction.

Of course, but on any double track line headways will be limited to dwell times+emergency braking time. So more and more TPH means distance between trains is decreased, so speeds need to go down to reduce the time it takes to come to a stop. That puts the theoretical limit for headways at about 90 seconds between trains, plus 30s for a bit of slop, just in case.

 

[ritchiew]

Since transit to Logan has come up in other threads lately, here's (I think) the best possible solution. Assuming a world with NSRL, Logan should be connected to the commuter rail. And it should look like this:

That blue line is a TBM'd tunnel, a little longer than three miles, that would connect directly to the NSRL tunnels (without a connection to surface North Station). The airport station would be directly under central parking.

Here's my thinking:

People here have pointed out that the airport largely serves two different groups of people. The first is locals using the airport, which skews heavily suburban. The suburban group is a big sign that a rail connection most naturally fits on the commuter rail network, not rapid transit. I don't think a rapid transit connection is a bad idea, but it's not the best fit. This proposal gives lots of people in southern suburbs and even Rhode Island a one-seat-ride directly to the airport. For those with a two-seat-ride, the transfers will be really easy. For many it would be a cross platform transfer, and for others it would be a simple up and over without having to leave the underground NSRL stations.

The other group of travelers is visitors to Boston, who are mainly headed downtown. This might be the best Airport-to-Downtown rail link of any city in the world. This takes people directly to major destinations and hotel clusters at North Station, Aquarium, South Station, and Back Bay. And it would hit each stop extremely fast. The curve I drew has a half-mile radius. I can't figure out exactly what is required, but that'd roughly let the trains run 80mph, I think. That would be Logan to North Station in 3-4 minutes. Airport to South Station in maybe 8. For any downtown station, this would be faster than driving, even if there is little car traffic. Under common conditions, the train could get to Back Bay before a car could get from Terminal B to the Ted Williams Tunnel.

Also, for people living in the city trying to get to the airport, this connects to every rapid transit line.

Is this a waste to have such a short commuter rail line? No, I don't think so, for two reasons. First, Logan is a major trip generator. There are 17,000 employees who could benefit from this. TSA regularly screens more than 70,000 passengers a day, and I'll assume the inbound is about symmetrical, so 140,000 people coming and going. If this line captures just 15% of those people, that'd be in the range of the busiest commuter line today. Second, branching off the north side of the NSRL tunnel takes advantage of slight issue with NSRL. Namely, there are more branches and trains coming from the south side than the north. Adding this new north side branch gives a new place for trains to go. In a Regional Rail world, the 7 south side branches could easily send a total of 8 trains an hour to the airport, meaning a train every 7.5 minutes.

Cost: expensive. But I don't know, I'd be curious to hear how expensive people think. The commonly proposed transit-to-Logan is extending the Silver Line tunnel through Seaport, then a new cross harbor tunnel. That is also super expensive, and I don't think I've ever seen a version that put a station in so useful a location. For this commuter rail proposal, this is all TBM, so the setup is expensive but then the cost per mile is relatively cheap. Also, in a project like this it's the stations that get expensive, and here there is only one station. Building under the parking garage could get expensive. Demolishing a small part of central parking and putting a station in its place would be much, much cheaper. And the station wouldn't have to be deep. Even if the tunnel is 100ft deep under the Mystic River, there's over a mile straightaway under East Boston to gently bring the track up to surface level at the airport

One far distant benefit is this could also by used by Amtrak. Having the Northeast Regional start/stop there could make Logan the best airport for some people into Connecticut. The East-West Rail project could do the same for people in Springfield, or even as far as Albany. (I lived in the Berkshires for a time and would fly out of Hartford or Albany, but sometimes it was easier to skip a flight connection by just driving to Boston.) In some future with real high speed rail in the US, a station at Logan could be used to eliminate some shorter connecting flights.

One more thing that is maybe wishful thinking, but perhaps something like this gets NSRL to actually happen. Most people simply don't care about NSRL, or transit generally. But people often do care about transit to the airport. Some argue governments tend to overvalue connections to the airport. So something like this might get the public and politicians to care about NSRL. We can say "Build NSRL, and then Boston can get the best Airport-to-Downtown train in the world." It might help.

 

[TheRatmeister]

Since transit to Logan has come up in other threads lately, here's (I think) the best possible solution. Assuming a world with NSRL, Logan should be connected to the commuter rail. And it should look like this:

View attachment 56420

That blue line is a TBM'd tunnel, a little longer than three miles, that would connect directly to the NSRL tunnels (without a connection to surface North Station). The airport station would be directly under central parking.

Here's my thinking:

People here have pointed out that the airport largely serves two different groups of people. The first is locals using the airport, which skews heavily suburban. The suburban group is a big sign that a rail connection most naturally fits on the commuter rail network, not rapid transit. I don't think a rapid transit connection is a bad idea, but it's not the best fit. This proposal gives lots of people in southern suburbs and even Rhode Island a one-seat-ride directly to the airport. For those with a two-seat-ride, the transfers will be really easy. For many it would be a cross platform transfer, and for others it would be a simple up and over without having to leave the underground NSRL stations.

The other group of travelers is visitors to Boston, who are mainly headed downtown. This might be the best Airport-to-Downtown rail link of any city in the world. This takes people directly to major destinations and hotel clusters at North Station, Aquarium, South Station, and Back Bay. And it would hit each stop extremely fast. The curve I drew has a half-mile radius. I can't figure out exactly what is required, but that'd roughly let the trains run 80mph, I think. That would be Logan to North Station in 3-4 minutes. Airport to South Station in maybe 8. For any downtown station, this would be faster than driving, even if there is little car traffic. Under common conditions, the train could get to Back Bay before a car could get from Terminal B to the Ted Williams Tunnel.

Also, for people living in the city trying to get to the airport, this connects to every rapid transit line.

Is this a waste to have such a short commuter rail line? No, I don't think so, for two reasons. First, Logan is a major trip generator. There are 17,000 employees who could benefit from this. TSA regularly screens more than 70,000 passengers a day, and I'll assume the inbound is about symmetrical, so 140,000 people coming and going. If this line captures just 15% of those people, that'd be in the range of the busiest commuter line today. Second, branching off the north side of the NSRL tunnel takes advantage of slight issue with NSRL. Namely, there are more branches and trains coming from the south side than the north. Adding this new north side branch gives a new place for trains to go. In a Regional Rail world, the 7 south side branches could easily send a total of 8 trains an hour to the airport, meaning a train every 7.5 minutes.

Cost: expensive. But I don't know, I'd be curious to hear how expensive people think. The commonly proposed transit-to-Logan is extending the Silver Line tunnel through Seaport, then a new cross harbor tunnel. That is also super expensive, and I don't think I've ever seen a version that put a station in so useful a location. For this commuter rail proposal, this is all TBM, so the setup is expensive but then the cost per mile is relatively cheap. Also, in a project like this it's the stations that get expensive, and here there is only one station. Building under the parking garage could get expensive. Demolishing a small part of central parking and putting a station in its place would be much, much cheaper. And the station wouldn't have to be deep. Even if the tunnel is 100ft deep under the Mystic River, there's over a mile straightaway under East Boston to gently bring the track up to surface level at the airport

One far distant benefit is this could also by used by Amtrak. Having the Northeast Regional start/stop there could make Logan the best airport for some people into Connecticut. The East-West Rail project could do the same for people in Springfield, or even as far as Albany. (I lived in the Berkshires for a time and would fly out of Hartford or Albany, but sometimes it was easier to skip a flight connection by just driving to Boston.) In some future with real high speed rail in the US, a station at Logan could be used to eliminate some shorter connecting flights.

One more thing that is maybe wishful thinking, but perhaps something like this gets NSRL to actually happen. Most people simply don't care about NSRL, or transit generally. But people often do care about transit to the airport. Some argue governments tend to overvalue connections to the airport. So something like this might get the public and politicians to care about NSRL. We can say "Build NSRL, and then Boston can get the best Airport-to-Downtown train in the world." It might help.

Is North Station really so important that it justifies adding over a mile of tunnel? I'm somewhat doubtful.

 

[ritchiew]

Is North Station really so important that it justifies adding over a mile of tunnel? I'm somewhat doubtful.

Yes, it is. Most people on Logan-bound trains will still just be people traveling from south and west of the city to somewhere downtown Boston. That includes the North Station area. There are lots of jobs in that area, even if it isn't as dense as other neighborhoods. There's a ton of residential density around it. And it's still growing. The skyline around there is unrecognizable compared to even 10-15 years ago, with more buildings planned. That growth should be encouraged and planned for, and that means trains should still go through North Station. We also shouldn't be diverting trains before they get to the trip generator of the Garden. And maybe most importantly is all the possible connections that could be lost with the Orange and Green Lines. It would be a huge missed opportunity if trains from Logan miss the Green Line. Also, if North Station is skipped, then trains from Logan headed for Old Colony lines never connect to the Orange line. So yes, North Station is really, really important as a destination and as an interchange.

As for the extra mile of track, again, that typically isn't where the costs come from with a TBM project. Mining the connection into the NSRL tunnel will be expensive, and same for the TBM launch pit and the station. But once the TBM gets going, the cost of added tunnel length is (usually) marginal. A mile of TBM'ed tunnel is not where costs get big.

 

[ulrichomega]

I guess my question is: Is the cost of a TBM for that length worth it when we can probably get away with reusing the existing ROW and then a much shorter (probably doable without a TBM) tunnel segment under Chelsea Creek and 1A. It's a longer trip, but it seems like it would save a lot of money (that map has the track loop around the terminals to serve each one, but obviously you can do as many or few stations as you want).

 

[TheRatmeister]

View attachment 56423
I guess my question is: Is the cost of a TBM for that length worth it when we can probably get away with reusing the existing ROW and then a much shorter (probably doable without a TBM) tunnel segment under Chelsea Creek and 1A. It's a longer trip, but it seems like it would save a lot of money (that map has the track loop around the terminals to serve each one, but obviously you can do as many or few stations as you want).

Yeah if we're digging a deep bore tunnel then we may as well just start from Chelsea. No need to go nearly as deep like you would need to for connecting to NSRL if you do that either.

 

[737900er]

View attachment 56423
I guess my question is: Is the cost of a TBM for that length worth it when we can probably get away with reusing the existing ROW and then a much shorter (probably doable without a TBM) tunnel segment under Chelsea Creek and 1A. It's a longer trip, but it seems like it would save a lot of money (that map has the track loop around the terminals to serve each one, but obviously you can do as many or few stations as you want).

It's also worth considering a future with A-B Connector, A-E Connector, and Terminal E Expansion Phase 2 where Terminal E will connect directly to the Blue Line and every other terminal will be connected to Terminal E airside.

I guess what I'm wondering is whether a 1-station line to Logan really the best use of limited capacity in the NSRL?