Crazy Transit Pitches

I'm confused. That study @Tallguy links to is wild (and quite fascinating -- longer response later) -- but while it helpfully gives us some numbers to do some guesstimating around tunnel costs, the very same study also concludes that a cross-harbor tunnel is dubious at best. While it's true that the study was comparing NSRL vs Airport tunnel, it still lays out a whole bunch of fundamental problems with the Airport alignment.

If you want a cross-harbor tunnel to support 24tph service to the North Shore, the Airport, and Downtown, we already have that, and it runs between Aquarium and Maverick.
So....BLX to....Beverly? Newburyport?
 
Alright, so let's take your costs from this '94 study, and correct for inflation (which is closer to 75%, not 50%), and also include the costs to get from SS to the harbor tunnel, and from the tunnel to Airport Station, Airport Station itself, and Airport Station to the existing ROW (since the all-in matters and you've gotta do all of that, the tunnel doesn't exist in a vacuum). That's $675M in 1994 dollars, or $1.2B today, omitting right-of-way, vents and exits, tracks, signals, electrification, environmental remediation, engineering and slushfund I mean, management and contingencies. The study observes that the latter three typically are allotted an additional 20-25% in budget ($220-275M). The study also observes that vents and exits for the Logan tunnel were expected to cost at least $80M over the CA/T alignment, for a total of $170M. That includes the Chelsea Creek tunnel, which we wouldn't be using, so cut it in half (although the Harbor crossing almost certainly was a bigger share of cost than Chelsea Creek). $85M, translated to 2020 dollars gets you $150M. Due to lack of information in the study, I've omitted any right-of-way costs in Eastie, track, signal, electrification, and environmental remediation costs. These costs will exist, I'm just not a civil engineer so any estimate I make would be a shot in the dark.

So our price tag is a floor of $1.7B for the cross-harbor tunnel. Again, we know it will be more than that ultimately, but let's stick with that. BLX + Son of Ted, in comparison, are between $1.8B and $2B all-in depending on the exact assumptions made. That's not a large difference, and due to several of the assumptions I made the floor is probably a little bit higher for the cross-harbor, and a little bit lower for BLX and 3rd Ted.


Understood.


I don't see why we should base our costs of a Ted bore off of a Draft Study from 1994 when we have the actual, real-world, cost of the Ted tunnels that have already been built. I'd be interested in the estimates from a 2020 study, however.


I'm happy to evaluate the options based on whatever dataset you choose, as long as it's consistent. The T sandbags all the time, but it's at least an expected sandbagging. I'll compare T to T, or Harvard to Harvard, or we can agree on a reasonable reduction factor to reduce the T figures to something we can compare to Harvard's study. But bear in mind that it makes the existing CA/T and BLX proposals cheaper if we assume that the T is sandbagging everything.


Combination done above, with proper inflation factor applied. I only factored in the costs as listed under the "Logan Airport to South Station" header, anything north of the Airport was omitted.


No data to support that TBM would be cheaper. So sticking with the data you've presented.


Of that, you are correct (primarily because I based this off the originally presented graphic). However, several of your assumptions are incomplete at best, as I've illustrated above.



Considering the fact that WOR-NS access via the GJ comes up fairly regularly, it's a fair bet that WOR would rather one-seat access to NS than the Airport. It also would mean that NSEWRL would not be easing the downtown crush-loads at all by keeping WOR riders making that transfer at BB as they currently do.


SL1 already exists. It gives direct terminal access. You're solving a problem that largely doesn't exist, at least from the southside.


I'll look later, but I've definitely seen long-range GL extensions proposed on this very forum to Chelsea and possibly Logan without needing to displace the current CR.

Leaving Malden Center on a branchline isn't exactly ideal either if you want to shoot Orange out there, but F-Line can address that better than I can).


Considering how close NS and Haymarket are, ER riders already have that access functionally. Just walk, or take one of the distended Lynn buses that will still be making the run all the way to Haymarket given that the Lynn routes will still be screwy. Not everyone needs a one-seat to every single CBD station.


You can make this work, but my contention is that there are easier, cheaper, and universally more productive ways to make these improvements. And I don't appear to be alone on that stance.
So re:inflation, I was actually assuming 100% to 200%, although I also understand that tube costs might not be linear, as there may be some fixed costs that dont scale.
So assuming linearity(275m for the tubes, as I remeasured and its actually 65% as long) with a 75% inflation factor, thats more like $485ishM. The Logan study also assumed an SSU on land whereas more recent proposals call for a Channel placement. I went with that assumption and expect that tubes would connect directly to the station.
Let's also use the Logan estimates(as you rightly say, for consistency, for two bored tunnels of 3600ft I thought that 18k per ft, which is between the lower and upper tunnelling estimates, would do. That results in a figure of $112M, with your 75% inflation factor. $597M so far. The study assumed an 1100ft station but lets keep that, just in case Amtrak sticks its beak in, at $52Mish. $649M so far? 1100ft of trench from there to the old ROW trench $30Mish, adjusted. Did I miss anything? Looks like $679M through in your mitigation, engineering etc $150M?
Regarding tracks and ventilation, I ve assumed a wash between CA/T and Harbor, but okay, $979M. I also never mentioned the operational advantages in not having two tunnel feeds in Charlestown.
And as for comparing the Ted, those tunnels are huge compared to a two track train tunnel. That alone makes them apples and oranges
And if youve ever ridden the SL1, you would know why Massport planned for a $1B people mover to Airport. Not unusual at all to take 20+min in the Ted alone.
 
Sorry, Riverside, forgot to address Chelsea/Malden.
Two main points.
The first, is that substituting RT for RR would be lightyears easier and less expensive than placing RT NEXT to CR.
Secondly, yes I know branching at Sullivan has its downside regarding Malden, but present operations are 6 min frequencies and if you went to 3-3.5 on the trunk, which you might have to because of additional Revere and Chelsea pax(also provides additional capacity for Worc to OL types), I think that both branches, which are almost exactly the same length, would be adequate. And with a GL to West, so OL traffic would be diverted at Sully
And regarding Worc-NS, I think that GJ GL would satisfy many that would want NS service. Possibly not.
 
So re:inflation, I was actually assuming 100% to 200%, although I also understand that tube costs might not be linear, as there may be some fixed costs that dont scale.
So assuming linearity(275m for the tubes, as I remeasured and its actually 65% as long) with a 75% inflation factor, thats more like $485ishM. The Logan study also assumed an SSU on land whereas more recent proposals call for a Channel placement. I went with that assumption and expect that tubes would connect directly to the station.
I'm confused as to where you're getting a 65% reduction in length. Are you measuring from the channel location to the Jeffries Point shoreline? Because if so, you've got to now bore under Jeffries Point to roughly reach the BL Airport station (assuming we're doing a superstation here) rather than doing a cut-and-cover as as the 94 study calls for, and I'm not seeing that accounted for here. Honestly, probably a wash with the deleted tunnel from SSU to the downtown shoreline but still needs accounting for, we can't just not add it into your recalculations.

I also think you may have shorted the tunnel measurements somewhat in that case, but honestly I'm not sure it's by enough for it to matter.

Let's also use the Logan estimates(as you rightly say, for consistency, for two bored tunnels of 3600ft I thought that 18k per ft, which is between the lower and upper tunnelling estimates, would do. That results in a figure of $112M, with your 75% inflation factor. $597M so far.
I wouldn't do that. The study is pretty clear about the cost difference in the two project areas. The Chelsea portion (which we're discarding since we aren't trying to bend the NEC back to the Western Route in this exercise) has lower costs that don't match the costs of tunneling on Logan property. I wouldn't be arbitrarily second-guessing their costing now that we've decided that they're a reliable cost source. Minor in the grand scheme, though.

Regarding tracks and ventilation, I ve assumed a wash between CA/T and Harbor, but okay, $979M.
Why would you assume a wash between CA/T and Harbor? The study very explicitly notes the difference in ventilation. Specifically, that you can't just throw up ventilation buildings and egress shafts in the middle of the harbor, so both air and people have to be moved laterally to a safe location to emerge.

The more I keep looking at the numbers in this study, the more I doubt them. Even with the T sandbagging, the cost differences are just too staggering between the 94 CA/T costs and current 2-track CA/T costs. It's pretty obvious to me that this study is missing several somethings. The pricing is literally a factor of 10 less than current cost estimates, even adjusting for inflation. If the T were sandbagging that hard, I would hope that we all knew it. Commonwealth Mag thinks the T is off by a factor of 2, so... I guess I have to wonder, what did they miss back in 1994, and is this discussion even remotely worthwhile if the cost estimate is junk?

I also never mentioned the operational advantages in not having two tunnel feeds in Charlestown.
I mean, you're still going to have a spaghetti junction at SSU since it's the one shared point, as opposed to being able to sort traffic at SSU going north, have it hit NSU, and then off to wherever needed, and the reverse coming south. Would that not be par at best?

And as for comparing the Ted, those tunnels are huge compared to a two track train tunnel. That alone makes them apples and oranges
The study you linked disagrees, interestingly enough. It makes a direct comparison to the Ted, stating that "this 7000' segment would be roughly the diameter of one of the highway barrels in the Third Harbor Crossing". So no, it's pretty apples to apples.

And if youve ever ridden the SL1, you would know why Massport planned for a $1B people mover to Airport. Not unusual at all to take 20+min in the Ted alone.
The traffic in the TWT has precisely zero to do with Massport's APM project. They're worried about Logan internal traffic (Massport shuttles, Uber/Lyft dropoffs, hotel shuttles, etc). While the APM would connect to your Airport superstation, it's anything but a prerequisite for the build. Logan has circulatory issues wholly aside from the nature of the transit tie-in.

I've ridden SL1 many a time. Which is why I'd prefer there be a dedicated RT tube instead of sitting in traffic. But the APM doesn't change Ted traffic for good or for bad.


Sorry, Riverside, forgot to address Chelsea/Malden.
Two main points.
The first, is that substituting RT for RR would be lightyears easier and less expensive than placing RT NEXT to CR.
It was me. But placing RT next to CR is still cheaper than building a cross-harbor tunnel just to relocate the CR.

Secondly, yes I know branching at Sullivan has its downside regarding Malden, but present operations are 6 min frequencies and if you went to 3-3.5 on the trunk, which you might have to because of additional Revere and Chelsea pax(also provides additional capacity for Worc to OL types), I think that both branches, which are almost exactly the same length, would be adequate. And with a GL to West, so OL traffic would be diverted at Sully
Orange could use additional frequencies now to be perfectly honest. Branching would preclude improving that situation at MC by default.

And regarding Worc-NS, I think that GJ GL would satisfy many that would want NS service. Possibly not.
It's not much of an improvement (if any), and it's still a transfer (though admittedly, at least one that occurs outside of the core). I'd be pretty sore if I lived in Worcester and the state spent billions on a NSRL that could have given me direct access to North Station, but instead still has me transferring 5 days a week to get to work, while giving me a one seat ride for that handful of times a year that I fly without the family. Especially with Worcester's own Tim Murray previously pushing that plan (RR running over the GJ, not LRT), there's definitely demand for a one-seat ride to North Station.
 
Last edited:
Two tunnels would probably not be at theoretical maximum 48tph on either route.
Regardless of theoretical max, my point really was that a tunnel that only serves the Eastern Route will run fewer trains than one that serves the entire North Side. All else being equal, a tunnel that connects to every line, is of far greater usefulness than one that does not.
 
You are convinced that access to the backwater which is Revere Beach and Orient Heights is so desirable that the expense is worth it and it seems no amount of research or analysis will dissuade you from your "faith".
Sometimes a fresh analysis of a problem results in a solution that others had not noticed before. Happens all the time. Now, perhaps there are technical issues that none of us is aware of that would drive up the costs of this routing. But what we do know is that this routing is more central to the population, jobs, retail and entertainment of the area. I have seen no data to counter mine. Do the advantages outweigh the disadvantages of this NSRL routing? That is a legitimate debate. There is no arguing against blind faith.

Nice try...no dice. You have to cite some research results first before levying that charge. Where does the cited study that was never advanced past draft flow to conclude mutual exclusivity in project selection?

Still waiting for the empirical underpinning of that "I think..." prerequisite despite these thread waters getting awfully thick with chum.
 
Nice try...no dice. You have to cite some research results first before levying that charge. Where does the cited study that was never advanced past draft flow to conclude mutual exclusivity in project selection?

Still waiting for the empirical underpinning of that "I think..." prerequisite despite these thread waters getting awfully thick with chum.
Not sure why I need to produce more data when no one has made a case FOR BLX based on something more than intuition.

For instance...the Gloucester Draw is being replaced right now. It is a 200ft single track replacement bridge and it is costing $100M.
The drawbridge necessary at River Works is 500ft, 2 tracks, and goes through Rumney Marsh OF CRITICAL ENVIRONMENTAL CONCERN! What could possibly go wrong? My wallet is breaking out in a cold sweat. Plus the other bridge, at the other end of a 2 mile marsh. The bridges alone could run north of $250-300M.

See? Facts. Data.

Maybe the Corps of Engineers will want 4 miles of viaduct like at SCR. What might that run?
 
How about BLX to Kenmore, continue down Brookline Ave, jog under the Muddy River to South Huntington and Heath, turn south to Columbus, Seaver, and Blue Hill Avenue to Mattapan Square. Bonus points for consuming the trolley, as we're dealing with small Blue Line cars, and continuing to Ashmont.

Stations:
  • Charles/MGH
  • Esplanade
  • Mass Ave (maybe)
  • Kenmore
  • Fenway (entrances at Kilmarnock and Park Drive)
  • Longwood Ave
  • (maybe) Riverway (South Huntington/Huntington)
  • Heath Street/VA
  • Jackson Square
  • Egleston Square
  • Humboldt Ave
  • Franklin Park
  • Talbot Ave
  • Morton Street
  • Blue Hill Avenue (CR)
  • Mattapan Square
 
I'm confused as to where you're getting a 65% reduction in length. Are you measuring from the channel location to the Jeffries Point shoreline? Because if so, you've got to now bore under Jeffries Point to roughly reach the BL Airport station (assuming we're doing a superstation here) rather than doing a cut-and-cover as as the 94 study calls for, and I'm not seeing that accounted for here. Honestly, probably a wash with the deleted tunnel from SSU to the downtown shoreline but still needs accounting for, we can't just not add it into your recalculations.

I also think you may have shorted the tunnel measurements somewhat in that case, but honestly I'm not sure it's by enough for it to matter.


I wouldn't do that. The study is pretty clear about the cost difference in the two project areas. The Chelsea portion (which we're discarding since we aren't trying to bend the NEC back to the Western Route in this exercise) has lower costs that don't match the costs of tunneling on Logan property. I wouldn't be arbitrarily second-guessing their costing now that we've decided that they're a reliable cost source. Minor in the grand scheme, though.


Why would you assume a wash between CA/T and Harbor? The study very explicitly notes the difference in ventilation. Specifically, that you can't just throw up ventilation buildings and egress shafts in the middle of the harbor, so both air and people have to be moved laterally to a safe location to emerge.

The more I keep looking at the numbers in this study, the more I doubt them. Even with the T sandbagging, the cost differences are just too staggering between the 94 CA/T costs and current 2-track CA/T costs. It's pretty obvious to me that this study is missing several somethings. The pricing is literally a factor of 10 less than current cost estimates, even adjusting for inflation. If the T were sandbagging that hard, I would hope that we all knew it. Commonwealth Mag thinks the T is off by a factor of 2, so... I guess I have to wonder, what did they miss back in 1994, and is this discussion even remotely worthwhile if the cost estimate is junk?


I mean, you're still going to have a spaghetti junction at SSU since it's the one shared point, as opposed to being able to sort traffic at SSU going north, have it hit NSU, and then off to wherever needed, and the reverse coming south. Would that not be par at best?


The study you linked disagrees, interestingly enough. It makes a direct comparison to the Ted, stating that "this 7000' segment would be roughly the diameter of one of the highway barrels in the Third Harbor Crossing". So no, it's pretty apples to apples.


The traffic in the TWT has precisely zero to do with Massport's APM project. They're worried about Logan internal traffic (Massport shuttles, Uber/Lyft dropoffs, hotel shuttles, etc). While the APM would connect to your Airport superstation, it's anything but a prerequisite for the build. Logan has circulatory issues wholly aside from the nature of the transit tie-in.

I've ridden SL1 many a time. Which is why I'd prefer there be a dedicated RT tube instead of sitting in traffic. But the APM doesn't change Ted traffic for good or for bad.



It was me. But placing RT next to CR is still cheaper than building a cross-harbor tunnel just to relocate the CR.


Orange could use additional frequencies now to be perfectly honest. Branching would preclude improving that situation at MC by default.


It's not much of an improvement (if any), and it's still a transfer (though admittedly, at least one that occurs outside of the core). I'd be pretty sore if I lived in Worcester and the state spent billions on a NSRL that could have given me direct access to North Station, but instead still has me transferring 5 days a week to get to work, while giving me a one seat ride for that handful of times a year that I fly without the family. Especially with Worcester's own Tim Murray previously pushing that plan (RR running over the GJ, not LRT), there's definitely demand for a one-seat ride to North Station.
This MBTA chart (albeit 5yrs old) shows ridership patterns on the OL. MC is well before peak ridership: full trains start at Sullivan. So, Sullivan is fed by 12 bus routes. Three go by Sweetser Cir, where there would assumedly be a station. So, most of that ridership is now on Chelsea OL. Three go Downtown. Four of the remaining cross the new GLX. One comes within two blocks. One would not be affected. Wellington feeds seven routes, five of which also go to Sweetser. And ridership is not equally distributed: four of the most traveled go through Sweetser. Three min frequency would double capacity to 16000 ph. So capacity to OG would remain the same as now, as would frequency. Even 4 min frequency would probably provide sufficient capacity, as ridership north of Sullivan will drop, but frequency would drop to eight min. If anything, with time, the Chelsea branch might exceed the OG.
Also, if GL to Sullivan were implemented, a significant number of pax would detrain from OL
OrangeLineFlows.png
 
I'm confused as to where you're getting a 65% reduction in length. Are you measuring from the channel location to the Jeffries Point shoreline? Because if so, you've got to now bore under Jeffries Point to roughly reach the BL Airport station (assuming we're doing a superstation here) rather than doing a cut-and-cover as as the 94 study calls for, and I'm not seeing that accounted for here. Honestly, probably a wash with the deleted tunnel from SSU to the downtown shoreline but still needs accounting for, we can't just not add it into your recalculations.

I also think you may have shorted the tunnel measurements somewhat in that case, but honestly I'm not sure it's by enough for it to matter.


I wouldn't do that. The study is pretty clear about the cost difference in the two project areas. The Chelsea portion (which we're discarding since we aren't trying to bend the NEC back to the Western Route in this exercise) has lower costs that don't match the costs of tunneling on Logan property. I wouldn't be arbitrarily second-guessing their costing now that we've decided that they're a reliable cost source. Minor in the grand scheme, though.


Why would you assume a wash between CA/T and Harbor? The study very explicitly notes the difference in ventilation. Specifically, that you can't just throw up ventilation buildings and egress shafts in the middle of the harbor, so both air and people have to be moved laterally to a safe location to emerge.

The more I keep looking at the numbers in this study, the more I doubt them. Even with the T sandbagging, the cost differences are just too staggering between the 94 CA/T costs and current 2-track CA/T costs. It's pretty obvious to me that this study is missing several somethings. The pricing is literally a factor of 10 less than current cost estimates, even adjusting for inflation. If the T were sandbagging that hard, I would hope that we all knew it. Commonwealth Mag thinks the T is off by a factor of 2, so... I guess I have to wonder, what did they miss back in 1994, and is this discussion even remotely worthwhile if the cost estimate is junk?


I mean, you're still going to have a spaghetti junction at SSU since it's the one shared point, as opposed to being able to sort traffic at SSU going north, have it hit NSU, and then off to wherever needed, and the reverse coming south. Would that not be par at best?


The study you linked disagrees, interestingly enough. It makes a direct comparison to the Ted, stating that "this 7000' segment would be roughly the diameter of one of the highway barrels in the Third Harbor Crossing". So no, it's pretty apples to apples.


The traffic in the TWT has precisely zero to do with Massport's APM project. They're worried about Logan internal traffic (Massport shuttles, Uber/Lyft dropoffs, hotel shuttles, etc). While the APM would connect to your Airport superstation, it's anything but a prerequisite for the build. Logan has circulatory issues wholly aside from the nature of the transit tie-in.

I've ridden SL1 many a time. Which is why I'd prefer there be a dedicated RT tube instead of sitting in traffic. But the APM doesn't change Ted traffic for good or for bad.



It was me. But placing RT next to CR is still cheaper than building a cross-harbor tunnel just to relocate the CR.


Orange could use additional frequencies now to be perfectly honest. Branching would preclude improving that situation at MC by default.


It's not much of an improvement (if any), and it's still a transfer (though admittedly, at least one that occurs outside of the core). I'd be pretty sore if I lived in Worcester and the state spent billions on a NSRL that could have given me direct access to North Station, but instead still has me transferring 5 days a week to get to work, while giving me a one seat ride for that handful of times a year that I fly without the family. Especially with Worcester's own Tim Murray previously pushing that plan (RR running over the GJ, not LRT), there's definitely demand for a one-seat ride to North Station.
"Two bored tunnels of 3600ft" I used the 18k figure which was with 1ishK pft which was the high.
And it is disingenuous to pick out one reason amoungst several and say "x is not worth b" when I argued "x+y+z is worth b"
 
Not sure why I need to produce more data when no one has made a case FOR BLX based on something more than intuition.

Oh, the turf...the turf! There's nothing left of it but goalpost holes!

Now we have flipped to this no longer being a matter of arbitrarily-chosen competing projects "🍎" and "👞" are thrown together by an "I think..." into mutual exclusion by data progression from "NO U" ad hominems. And installed a *new* arbitrary prerequisite where "🍎" must self-defeat itself on scoring so that a vacuum exists where only "👞" can waltz into the void, no pesky head-to-head scoring now required remedying the fact that "👞" never lived long enough as a study candidate to be benchmarked. Because "👞" is better than "🍎" when "🍎" has been removed from consideration by magic beforehand. Except "🍎" having 70 years of study basis and multiple endorsed maximum recs in that time doesn't matter because all those studies were published by liars, says I. So really the most important thing is that only sucky liars like "🍎"'s, so I'm right.

You ask again..."What does "🍎" have to do with "👞"?" Because I've read a lot of data in my time, and I don't need to show the whatnow about how they're related because your intuitions...which will remain unmentioned by all...are just, unnngh, *sigh*.


For instance...the Gloucester Draw is being replaced right now. It is a 200ft single track replacement bridge and it is costing $100M.

I know! For a double-track bridge, that's clearly way too expensive for a single-track bridge! We should totally scrap "🍎" that's 24 miles south of there to build "👞" 28 miles south of there, because goshdarn LOOK WITH YOUR OWN EYES at those numbers and tell me they pass your nose's smell test. I've seen data for the marshes. That Gloucester Harbor...just...won't...hunt...for a bridge over nowheresville Revere Beach when we're talking about the granite rocks under the Inner Harbor. FACT!



So...we on the same wavelength yet? Or is there some truthiness still not accounted for yet by intuition? 🤡
 
How about BLX to Kenmore, continue down Brookline Ave, jog under the Muddy River to South Huntington and Heath, turn south to Columbus, Seaver, and Blue Hill Avenue to Mattapan Square. Bonus points for consuming the trolley, as we're dealing with small Blue Line cars, and continuing to Ashmont.

Stations:
  • Charles/MGH
  • Esplanade
  • Mass Ave (maybe)
  • Kenmore
  • Fenway (entrances at Kilmarnock and Park Drive)
  • Longwood Ave
  • (maybe) Riverway (South Huntington/Huntington)
  • Heath Street/VA
  • Jackson Square
  • Egleston Square
  • Humboldt Ave
  • Franklin Park
  • Talbot Ave
  • Morton Street
  • Blue Hill Avenue (CR)
  • Mattapan Square

Awful God-modey with the street widths and any subway paralleling of the same Muddy River floodplain that imperils the D portal. Consider that the Urban Ring Phase III MOAR TUNNEL complete-circuit grade separated rail tries to make hay on a lot of these same street corridors...and didn't get very far along on the conceptualization before aborting the mission and punting it off to re-study in some well later era where somebody's got the foggest clue what realistic pathways make for a testable-theory design/build to benchmark. And no, Egleston to exclusion of Nubian isn't going to cut it here. Nubian is the single largest bus hub on the system. Egleston transit suffers as an accessory to Nubian transit suffering...but Roxbury doesn't rehabilitate treating in the reverse order. Egleston's too dependent on a Nubian source. Which means you have to figure out how to build Urban Ring SW between Kenmore and Numbian before there's any additional pieces hitting Kenmore-Egleston but not Nubian like a Blue trunk. Which means double-double tunnel trouble!

Nothing's impossible. But that SW quadrant is such an unstitchable mess I think you've got to retreat to *much* more basic questions like how you're going to get plausibly between Fenway/Longwood and any part of Roxbury on any kind of separable rail mode before the Blue Hill Ave. corridor is ever a glint in anyone's eye. Which basically means...solve for the same Urban Ring SW sweepstakes first, then worry about what color line gets sent through the intermediary deep out to the desired endpoint.
 
I'm confused as to where you're getting a 65% reduction in length. Are you measuring from the channel location to the Jeffries Point shoreline? Because if so, you've got to now bore under Jeffries Point to roughly reach the BL Airport station (assuming we're doing a superstation here) rather than doing a cut-and-cover as as the 94 study calls for, and I'm not seeing that accounted for here. Honestly, probably a wash with the deleted tunnel from SSU to the downtown shoreline but still needs accounting for, we can't just not add it into your recalculations.

I also think you may have shorted the tunnel measurements somewhat in that case, but honestly I'm not sure it's by enough for it to matter.


I wouldn't do that. The study is pretty clear about the cost difference in the two project areas. The Chelsea portion (which we're discarding since we aren't trying to bend the NEC back to the Western Route in this exercise) has lower costs that don't match the costs of tunneling on Logan property. I wouldn't be arbitrarily second-guessing their costing now that we've decided that they're a reliable cost source. Minor in the grand scheme, though.


Why would you assume a wash between CA/T and Harbor? The study very explicitly notes the difference in ventilation. Specifically, that you can't just throw up ventilation buildings and egress shafts in the middle of the harbor, so both air and people have to be moved laterally to a safe location to emerge.

The more I keep looking at the numbers in this study, the more I doubt them. Even with the T sandbagging, the cost differences are just too staggering between the 94 CA/T costs and current 2-track CA/T costs. It's pretty obvious to me that this study is missing several somethings. The pricing is literally a factor of 10 less than current cost estimates, even adjusting for inflation. If the T were sandbagging that hard, I would hope that we all knew it. Commonwealth Mag thinks the T is off by a factor of 2, so... I guess I have to wonder, what did they miss back in 1994, and is this discussion even remotely worthwhile if the cost estimate is junk?


I mean, you're still going to have a spaghetti junction at SSU since it's the one shared point, as opposed to being able to sort traffic at SSU going north, have it hit NSU, and then off to wherever needed, and the reverse coming south. Would that not be par at best?


The study you linked disagrees, interestingly enough. It makes a direct comparison to the Ted, stating that "this 7000' segment would be roughly the diameter of one of the highway barrels in the Third Harbor Crossing". So no, it's pretty apples to apples.


The traffic in the TWT has precisely zero to do with Massport's APM project. They're worried about Logan internal traffic (Massport shuttles, Uber/Lyft dropoffs, hotel shuttles, etc). While the APM would connect to your Airport superstation, it's anything but a prerequisite for the build. Logan has circulatory issues wholly aside from the nature of the transit tie-in.

I've ridden SL1 many a time. Which is why I'd prefer there be a dedicated RT tube instead of sitting in traffic. But the APM doesn't change Ted traffic for good or for bad.



It was me. But placing RT next to CR is still cheaper than building a cross-harbor tunnel just to relocate the CR.


Orange could use additional frequencies now to be perfectly honest. Branching would preclude improving that situation at MC by default.


It's not much of an improvement (if any), and it's still a transfer (though admittedly, at least one that occurs outside of the core). I'd be pretty sore if I lived in Worcester and the state spent billions on a NSRL that could have given me direct access to North Station, but instead still has me transferring 5 days a week to get to work, while giving me a one seat ride for that handful of times a year that I fly without the family. Especially with Worcester's own Tim Murray previously pushing that plan (RR running over the GJ, not LRT), there's definitely demand for a one-seat ride to North Station.
This MBTA chart (albeit 5yrs old) shows ridership patterns on the OL
View attachment 8622
 
This MBTA chart (albeit 5yrs old) shows ridership patterns on the OL
View attachment 8622
Would it kill you to consolidate your replies to me into one post? You're killing me with the three posts to ping back and forth to read and respond to lol.

So, by your own chart, MC already matches almost the entire southern end of Orange, and that's 5 year old data. I'd love to have more up to date data, but anecdotally, ridership levels weren't exactly dropping prior to COVID. Add to that the eventual extension of Orange to Reading in an NSRL (or NSEWRL) world due to the increasing difficulty of running that as a CR branch (adding more ridership on what's currently your Oak Grove branch), and things start getting dicey. Presumably the state would find a way to shiv in a PnR stop where the line crosses 128, netting additional ridership.

You've nicely demonstrated a compelling need for Rapid Transit heading East from Sullivan (hence the Urban Ring). I agree with that need. You have not demonstrated a need for that to be the heavy rail OL vs the Light Rail GL, imo.

As far as cherry-picking the cross-harbor costing, that's rich considering you decided to nitpick my cost assumptions to make your plan look more cost-reasonable. It's possible for us both to be wrong. I do, however, want to know why we should take the 94 study's estimate as a more accurate cost basis than the T's 2018 reassessment, given that the 2018 reassessment is a factor of 10 more expensive on the comparable routing even when adjusting for inflation. Are we arguing that the T is sandbagging just that much? Because if we are... Why are we arguing, clearly we can build literally everything we want within the T's 10x inflated NSRL budget? BLX would literally disappear within it, you could do BLX to Salem (let's hit the home run, why not, the T is overpricing by 10x right?), your harbor crossing, the Red-Blue connector (giving Lynn and Salem direct access to MGH and one-stop outside of the crush loaded core access to Cambridge) AND a 2-track CA/T if the 94 costs are actually correct and 2018 is just Price Is Right numbers.

But of course, neither of us actually believes that. So I'm asking honestly, how do we reach a number that's tethered to reality without being clearly overestimated as the T tends to do? Because right now we're engaging on false terms. I'm evaluating your cross-harbor tube on 94 cost estimates, but additional Ted bores are based on as-built costs, and BLX has been studied to death over the past 80 years (there should actually be a DEIR study out there that addresses your faux-concern over Rumney Marsh from 2008 or 2009), and there's another study that was supposed to be completed this April (I assume delayed by COVID) regarding the feasibility of a BLX, which is being pushed by Lynn politicians. Prior to that, it was studied in 2005 which is still a decade more current. In hindsight, the whole comparison is flawed. There's no point in continuing the cost discussion unless we can accurately compare.

Oh, and before I forget - you're criticizing F-Line and myself for harping on inter-neighborhood connections, but you're pushing a one-seater to the Airport with no proven constituency, and the study that you yourself dug up took the premise as giving Amtrak direct access to Logan, the CR connection was seen as nothing but a bonus. It literally calls Amtrak access to Logan the "raison d'être" of the Logan routing. In your scenario where Amtrak isn't using the tunnel, you're removing the study's primary goal. So again, what exactly is the constituency for this?
 
Last edited:
Sorry about the fragmentation. I usually do this on my phone and I can barely read what I am writing:)

So, you're right, both GL and OL to Chelsea/Revere have plusses and minuses.
Green offers potential one seat to West,but man, I feel sorry that poor viaduct, trying to bring Union Sq, Medford and whatever percentage of UR trains people will want to push through to GC. What is frustrating is how much four track infrastructure remains.
Reading? 1200 pax from 7-830, of which 400 ride the crunch time of 8-830. A p&r, combined with more frequent service, could easily double that, if not more.
The key question is how much ridership gets diverted to GLX. That 8000 hourly limit should rise to 12000 with the new fleet at max. Assuming three min frequency is the max for a while, we could get that up to 16000, but thats the last trick up anyones sleeves.
Splitting BL at Airport is an option, as ridership falls off dramatically further out, but getting over or under Chelsea River is going to be pricey.
I know tha FLine and others favor the PoPs route, but I just dont see an easy way of getting from one side of Rt 1a to the other. Also, I know the ROW is technically clear, I am reminded of the Blues Brothers, where Elwood has El trains shaking by continuously.
But you are correct in saying the estimates for BLX are sketchy as well.
As far as the constituency for harbor routing? There is no one overriding raison d'etre. It that it does several different things, not necessarily as well as other stand alone projects. It is the only way to so dramatically reduce time to SS. If we assume that Chelsea is joined by a Sweetser and a Sullivan CR Station, then even with EMUs and full high platforms, then each stop is a 75 sec penalty. That's four minutes. The route is 4 miles shorter. Thats 4 min, at 60mph. Its 6 min at 45mph. It would not surprise me if the Airport routing was 10 min quicker. But it also obviates the need for a second bridge across the Mystic, whichever mode. Fitting four tracks wont necessarily lead to the cost bloat of GLX, but there is a significant cost savings, probably upwards of $200M. And Im not saying that there NO travel between Lynn and Revere Beach
In actuality, no one route would have to bear the brunt of missing upper CBD. You could spread the joy. Lets say the Worcester train runs 4x hr as does the Providence. Each could have, in one case only have the :15 and :45 go to ER and the other the :00 and the :30. But remember, as you said, you cant offer everyone a one seat to everywhere. If all Worcester runs went east, you could change at BB to either Prov/Stoughton, running every 7.5 min, or at SS and take RL or switch to any of the other trains coming through every 2.5 min.
And when it comes down to it, without more data, neither of us has a handle on the cost differential between the two options.I have no horse in this race regarding NS access; I live to the west and dont usually fly.
 
Oh, the turf...the turf! There's nothing left of it but goalpost holes!

Now we have flipped to this no longer being a matter of arbitrarily-chosen competing projects "🍎" and "👞" are thrown together by an "I think..." into mutual exclusion by data progression from "NO U" ad hominems. And installed a *new* arbitrary prerequisite where "🍎" must self-defeat itself on scoring so that a vacuum exists where only "👞" can waltz into the void, no pesky head-to-head scoring now required remedying the fact that "👞" never lived long enough as a study candidate to be benchmarked. Because "👞" is better than "🍎" when "🍎" has been removed from consideration by magic beforehand. Except "🍎" having 70 years of study basis and multiple endorsed maximum recs in that time doesn't matter because all those studies were published by liars, says I. So really the most important thing is that only sucky liars like "🍎"'s, so I'm right.

You ask again..."What does "🍎" have to do with "👞"?" Because I've read a lot of data in my time, and I don't need to show the whatnow about how they're related because your intuitions...which will remain unmentioned by all...are just, unnngh, *sigh*.




I know! For a double-track bridge, that's clearly way too expensive for a single-track bridge! We should totally scrap "🍎" that's 24 miles south of there to build "👞" 28 miles south of there, because goshdarn LOOK WITH YOUR OWN EYES at those numbers and tell me they pass your nose's smell test. I've seen data for the marshes. That Gloucester Harbor...just...won't...hunt...for a bridge over nowheresville Revere Beach when we're talking about the granite rocks under the Inner Harbor. FACT!



So...we on the same wavelength yet? Or is there some truthiness still not accounted for yet by intuition? 🤡
Thank you for your valuable input. I will give it every bit the consideration it is due.
 
Yo I'm just gonna say this -- Regional Rail to Lynn vs BLX is, you know, one topic (well worth discussing). Building a cross-harbor tunnel is totally and completely different.

We've already established that the travel time difference between a Chelsea surface and an Eastie tunnel alignment is on the order of 5 minutes, maaaaybe 10. We're not talking about straightening the NEC's S-curve through NYC here.

And while the Eastie tunnel grants you better access to Airport Station (but still need to transfer to get to Terminals), you lose out on access to a wider swath of Downtown and on improved connectivity to the subway network. The Chelsea alignment results in direct transfers to the Green and Orange Lines at North Station/Haymarket, Red Line at South Station and maybe Orange (and future Green) at Sullivan. The Eastie alignment dumps nearly everyone out at South Station, where they need to squeeze onto the Red Line to complete their journey to Green/Orange (along a stretch where Red is already maxed out).

Regional Rail to Lynn at 8-12 tph is definitely worthwhile, I don't think anyone disagrees on that. (The disagreement is whether it replaces or supplements BLX, but that's not the point here.) But Regional Rail to Lynn absolutely does not entail an Airport-South Station tunnel.
Yo I'm just gonna say this -- Regional Rail to Lynn vs BLX is, you know, one topic (well worth discussing). Building a cross-harbor tunnel is totally and completely different.

We've already established that the travel time difference between a Chelsea surface and an Eastie tunnel alignment is on the order of 5 minutes, maaaaybe 10. We're not talking about straightening the NEC's S-curve through NYC here.

And while the Eastie tunnel grants you better access to Airport Station (but still need to transfer to get to Terminals), you lose out on access to a wider swath of Downtown and on improved connectivity to the subway network. The Chelsea alignment results in direct transfers to the Green and Orange Lines at North Station/Haymarket, Red Line at South Station and maybe Orange (and future Green) at Sullivan. The Eastie alignment dumps nearly everyone out at South Station, where they need to squeeze onto the Red Line to complete their journey to Green/Orange (along a stretch where Red is already maxed out).

Regional Rail to Lynn at 8-12 tph is definitely worthwhile, I don't think anyone disagrees on that. (The disagreement is whether it replaces or supplements BLX, but that's not the point here.) But Regional Rail to Lynn absolutely does not entail an Airport-South Station tunnel.
I think you are discounting the utility of BL in putting people in the same areas that GL and OL would. Also, for the stretch from State to Haymarket, SSU transfer to NSRL CA/T remains an option, taking some heat off of RL.
And though all this(and Im not pointing a finger at anyone in particular) I think some people are not taking into account that around 4 million pax pass through Logan per month. It is not unusual for 175K a day to use it. Logan itself employs 17000 people, plus many others at the many hotels and other businesses in the area. Even 40% utilization would result in 80ishK pax per day. I dont think that's anything to sneeze at.
The one seat from ER to NSU or CSU or whatever we call it is a concern, but only for those people who walk to work from there. For many others, transfers at SSU, BB, Lansdowne and West get the job down. (And dont get me started on E Line to Transitway transfers at BB;) )
 
Sorry about the fragmentation. I usually do this on my phone and I can barely read what I am writing:)

So, you're right, both GL and OL to Chelsea/Revere have plusses and minuses.
Green offers potential one seat to West,but man, I feel sorry that poor viaduct, trying to bring Union Sq, Medford and whatever percentage of UR trains people will want to push through to GC. What is frustrating is how much four track infrastructure remains.
Reading? 1200 pax from 7-830, of which 400 ride the crunch time of 8-830. A p&r, combined with more frequent service, could easily double that, if not more.
The key question is how much ridership gets diverted to GLX. That 8000 hourly limit should rise to 12000 with the new fleet at max. Assuming three min frequency is the max for a while, we could get that up to 16000, but thats the last trick up anyones sleeves.
I agree with most of this! But again, I think you're missing potential inter-neighborhood travel (and general demand stimulation from people who could, but don't presently, commute via the CR line). Right now, your only real local transit option on Reading north of Oak Grove is the 136/137 out of Malden Center. The CR frequencies aren't anywhere near enough to be useful for much of anything other than heading straight to the CBD, and of course the bus frequencies aren't ever going to even sniff Orange-peak frequencies, whether those are 3 or 6 minutes. Having 6 min frequencies is likely to stimulate transit demand along that corridor, even if it's not beating down the doors Day 1. It's likely to be an extension compelled by operational needs more than ridership, but that doesn't mean the ridership won't end up being substantial.

Splitting BL at Airport is an option, as ridership falls off dramatically further out, but getting over or under Chelsea River is going to be pricey.
Honestly I'd love a Blue split at Airport or Maverick to head to Chelsea, Everett, and Saugus. I just don't really know how you manage to get over there at a reasonable price. It looks a little weird on a map, but add UR to allow cross-town connections without going into the core and I think it's a winner if we can figure out how the hell to build it and figure out the branching dispatching, since Blue is unbalanced to begin with (having the CBD at a terminal instead of the midpoint). There's no shortage of demand over there, it's a matter of how to serve it. I think a pre-req for a Blue split is an extension to ??? on the other side of the CBD. So, definitely Crazy, possibly God mode, but a boy can dream.

I know tha FLine and others favor the PoPs route, but I just dont see an easy way of getting from one side of Rt 1a to the other. Also, I know the ROW is technically clear, I am reminded of the Blues Brothers, where Elwood has El trains shaking by continuously.
But you are correct in saying the estimates for BLX are sketchy as well.
There's a half-and-half option, cutting across at Oak Island instead of going whole-hog down the PoP routing. I think that's probably the best option to avoid NIMBYs from smothering it in the crib. Cutting across at Oak Island also puts you neatly on alignment to go side-by-side on the 4-track ROW with CR, and pop that Lynnport (nee River Works) station in there, as well as potential space for a larger yard.

As far as the constituency for harbor routing? There is no one overriding raison d'etre. It that it does several different things, not necessarily as well as other stand alone projects. It is the only way to so dramatically reduce time to SS. If we assume that Chelsea is joined by a Sweetser and a Sullivan CR Station, then even with EMUs and full high platforms, then each stop is a 75 sec penalty. That's four minutes. The route is 4 miles shorter. Thats 4 min, at 60mph. Its 6 min at 45mph. It would not surprise me if the Airport routing was 10 min quicker. But it also obviates the need for a second bridge across the Mystic, whichever mode. Fitting four tracks wont necessarily lead to the cost bloat of GLX, but there is a significant cost savings, probably upwards of $200M. And Im not saying that there NO travel between Lynn and Revere Beach
In actuality, no one route would have to bear the brunt of missing upper CBD. You could spread the joy. Lets say the Worcester train runs 4x hr as does the Providence. Each could have, in one case only have the :15 and :45 go to ER and the other the :00 and the :30. But remember, as you said, you cant offer everyone a one seat to everywhere. If all Worcester runs went east, you could change at BB to either Prov/Stoughton, running every 7.5 min, or at SS and take RL or switch to any of the other trains coming through every 2.5 min.
And when it comes down to it, without more data, neither of us has a handle on the cost differential between the two options.I have no horse in this race regarding NS access; I live to the west and dont usually fly.
The fact that it does several different things, but not as well as the targeted projects, is my entire beef with the idea. I'm not sold that 10 min faster direct access to South Station is worth eliminating the one-seat to North Station in the opinion of people living in the target area (Lynn and points Northeast). Remember, right now these folks have no direct access to South Station. So in the minds of the public who needs to support this spending, you're not arguing "hey, you get to South Station 10 minutes faster", you're arguing "hey, you get to go to the Airport and South Station, but if you commute to the Haymarket/North Station area, I've just forced you onto an additional transfer". You're also not addressing any demand from the northeast to Cambridge and environs. Those passengers now only have one transfer, but it's a transfer onto the most painfully overloaded segment of Red. BLX doesn't directly solve this, of course - but you can build the Red-Blue connector to do so, similar to your proposal to branch Orange to take advantage of the vacated CR ROW through Everett and Chelsea.

I'm also firmly in the camp of not planning transit around extremely infrequent trips. For the vast majority of people, heading to Logan is infrequent (I believe that Logan employees have access to some sort of program subsidized by Massport to take the Logan Express as well) - going to work, however, is by definition 5 days a week under normal circumstances. The natural constituency for direct access to both CBD stations is flat-out higher than direct Airport access. That doesn't mean Airport access isn't valuable, but it does mean it is less valuable than giving everyone (or at least, a greater fraction of everyone) access to the entire CBD. The thru-running to give southern suburbs access to Woburn and Waltham office parks just happens to be a cherry on top.

As far as spreading the joy on missing out on upper CBD access, that sounds like a recipe to dampen support for NSRL entirely. Direct access to both CBD stations is the big hook for everyone who isn't as invested in transit as we are. Telling the public that we're going to spend 5+ years and $X billion digging out these tunnels beneath Boston and the Harbor, for what amounts to slightly faster Airport access and reduced North Station access for people who currently have direct access isn't exactly a winning strategy. Now, if we were doing 6 -track NSEWRL and some reverse-branching on the Eastern, and some ran via Chelsea and some via Airport, that's an easy sell on the merits, but the price tag would make the Legislature absolutely blush. It's not that I don't like the idea of a cross-harbor tunnel. I do! But there are a number of issues inherent to the design that make it less desirable compared to the base 4-track CA/T NSRL that most area transit advocates are pushing for. Cost is only one of them.
 
I agree with most of this! But again, I think you're missing potential inter-neighborhood travel (and general demand stimulation from people who could, but don't presently, commute via the CR line). Right now, your only real local transit option on Reading north of Oak Grove is the 136/137 out of Malden Center. The CR frequencies aren't anywhere near enough to be useful for much of anything other than heading straight to the CBD, and of course the bus frequencies aren't ever going to even sniff Orange-peak frequencies, whether those are 3 or 6 minutes. Having 6 min frequencies is likely to stimulate transit demand along that corridor, even if it's not beating down the doors Day 1. It's likely to be an extension compelled by operational needs more than ridership, but that doesn't mean the ridership won't end up being substantial.


Honestly I'd love a Blue split at Airport or Maverick to head to Chelsea, Everett, and Saugus. I just don't really know how you manage to get over there at a reasonable price. It looks a little weird on a map, but add UR to allow cross-town connections without going into the core and I think it's a winner if we can figure out how the hell to build it and figure out the branching dispatching, since Blue is unbalanced to begin with (having the CBD at a terminal instead of the midpoint). There's no shortage of demand over there, it's a matter of how to serve it. I think a pre-req for a Blue split is an extension to ??? on the other side of the CBD. So, definitely Crazy, possibly God mode, but a boy can dream.


There's a half-and-half option, cutting across at Oak Island instead of going whole-hog down the PoP routing. I think that's probably the best option to avoid NIMBYs from smothering it in the crib. Cutting across at Oak Island also puts you neatly on alignment to go side-by-side on the 4-track ROW with CR, and pop that Lynnport (nee River Works) station in there, as well as potential space for a larger yard.


The fact that it does several different things, but not as well as the targeted projects, is my entire beef with the idea. I'm not sold that 10 min faster direct access to South Station is worth eliminating the one-seat to North Station in the opinion of people living in the target area (Lynn and points Northeast). Remember, right now these folks have no direct access to South Station. So in the minds of the public who needs to support this spending, you're not arguing "hey, you get to South Station 10 minutes faster", you're arguing "hey, you get to go to the Airport and South Station, but if you commute to the Haymarket/North Station area, I've just forced you onto an additional transfer". You're also not addressing any demand from the northeast to Cambridge and environs. Those passengers now only have one transfer, but it's a transfer onto the most painfully overloaded segment of Red. BLX doesn't directly solve this, of course - but you can build the Red-Blue connector to do so, similar to your proposal to branch Orange to take advantage of the vacated CR ROW through Everett and Chelsea.

I'm also firmly in the camp of not planning transit around extremely infrequent trips. For the vast majority of people, heading to Logan is infrequent (I believe that Logan employees have access to some sort of program subsidized by Massport to take the Logan Express as well) - going to work, however, is by definition 5 days a week under normal circumstances. The natural constituency for direct access to both CBD stations is flat-out higher than direct Airport access. That doesn't mean Airport access isn't valuable, but it does mean it is less valuable than giving everyone (or at least, a greater fraction of everyone) access to the entire CBD. The thru-running to give southern suburbs access to Woburn and Waltham office parks just happens to be a cherry on top.

As far as spreading the joy on missing out on upper CBD access, that sounds like a recipe to dampen support for NSRL entirely. Direct access to both CBD stations is the big hook for everyone who isn't as invested in transit as we are. Telling the public that we're going to spend 5+ years and $X billion digging out these tunnels beneath Boston and the Harbor, for what amounts to slightly faster Airport access and reduced North Station access for people who currently have direct access isn't exactly a winning strategy. Now, if we were doing 6 -track NSEWRL and some reverse-branching on the Eastern, and some ran via Chelsea and some via Airport, that's an easy sell on the merits, but the price tag would make the Legislature absolutely blush. It's not that I don't like the idea of a cross-harbor tunnel. I do! But there are a number of issues inherent to the design that make it less desirable compared to the base 4-track CA/T NSRL that most area transit advocates are pushing for. Cost is only one of them.
So, we are trying to get data on NS destinations and modes now. We know that a significant percentage take OL,GL and shuttles to Kendall and MGH. That is a key question. If say, 80percent (I know, unlikely, but exaggerating for emphasis)then go to GC or State St, Downtown X, Park St, MGH or Tufts(or even SS;), then my point is made, if its 5 percent then yours is proven. Of course, station placement(and how many) also affect utility.
As far as Cambridge and points northeast, I think that your point is has some validity regarding Kendall and Central Sq, but as far as Central and Southern MIT and Harvard Sq, then West to UR GL should fit the bill. And I am also assuming that NSRL would give the RL the relief it needs to carry any ER pax.
And, yes I have presupposed RBC.
Also, while this doesn't meet every need as well as several individual projects, it does most of it at less cost than all the projects together. And it does a few things that the other projects can't. I will appraise you of our data when it comes in.
 

Back
Top