I was trying to think of other examples where there's rail built 35ish feet above the highway with road bridges passing perpendicularly underneath. Do they have that in LA (I'm just not that familiar with their system)? Would love to see other examples of similar solutions.
Seattle is building many miles of LRT viaducts along major freeways, which dip down to the surface when possible, and cross under roads when they can. The Boston metro area could certainly use some miles of this type of elevated rail structure, which is orders of magnitudes cheaper than tunneling.
Vancouver's SkyTrain is another system that does a great job with elevated tracks. Apparently modern els in general are notably cheaper than tunneling?
A one-minute, not-serious proposal that's partially inspired by my comment on expanding rapid transit boundaries:
Extend the GL D branch from Riverside to Wellesley College via Wellesley Farms, Wellesley Hills and Wellesley Square. This allows Regional Rail to Framingham/Worcester to skip these stations and speed up their trips.
Or is a future Blue Line extension (via tunnel beneath Worcester Line to also serve the Newton stations) a better idea?
But then I looked at the current schedules, which seem to suggest a full thirty minutes of travel time between Wellesley Farms and Back Bay, which seems... very high? (Where skipping the Newtons only shaves off 4 min)
So, I mean, the Green Line would still be slower. But to your overall point about rapid transit poking out beyond 128 here, I wholeheartedly agree it's worth looking at. Norwood, Brockton, and Framingham all set at the ends of higher-density "arms" that stretch out from Metro Boston:
The arms themselves aren't shockingly high density, but are comparable to parts of Newton and Arlington/Lexington, which we pretty regularly consider candidates for rapid transit.
So, yes, I'd say there's some future, or alternate-history, version of the T where the Red Line runs to Brockton, the Orange Line to Norwood, and the Blue Line to Wellesley or Framingham.
Slightly less God-mode-y, but still fun to think about would be a short extension of the D Line to Newton Lower Falls, ending in what could be a cute little TOD village:
But then I looked at the current schedules, which seem to suggest a full thirty minutes of travel time between Wellesley Farms and Back Bay, which seems... very high? (Where skipping the Newtons only shaves off 4 min)
Landsdowne -> Back Bay is a big culprit here. Anyone who rides the Worcester Line regularly can attest to the fact that it absolutely crawls over that stretch.
It's scheduled for 7-10 minutes, which I'm sure some of that has to do with buffers for Keolis to have a better OTP and for extended dwells at peak times, but it doesn't help how slowly it moves when it's moving over that section.
For perspective, Kenmore -> Copley was 4 minutes prior to the slow zones earlier this year, so half the time of the scheduled Commuter Rail.
Transitmatters claims Back Bay and Landsdowne should be two minutes apart and Wellesley Farms -> Back Bay should be 19 minutes on local trains, but that's very ambitious.
Landsdowne -> Back Bay is a big culprit here. Anyone who rides the Worcester Line regularly can attest to the fact that it absolutely crawls over that stretch.
It's scheduled for 7-10 minutes, which I'm sure some of that has to do with buffers for Keolis to have a better OTP and for extended dwells at peak times, but it doesn't help how slowly it moves when it's moving over that section.
For perspective, Kenmore -> Copley was 4 minutes prior to the slow zones earlier this year, so half the time of the scheduled Commuter Rail.
Transitmatters claims Back Bay and Landsdowne should be two minutes apart and Wellesley Farms -> Back Bay should be 19 minutes on local trains, but that's very ambitious.
For what it's worth, the weekend outbound schedule claims a travel time of 3 minutes from Back Bay to Lansdowne, which really illustrates the effect of padding and dwell times.
Bumping this question to add another one: I'm also curious about data on cost of cut-and-cover subways with utility relocations (especially undocumented utilities) vs deep bored tunnels.
I know @JeffDowntown has always been of the belief that C&C with utilities may not even be cheaper than TBMs, so I'd love to see data on that, especially given that many of our roads are built on landfill. If costs of C&C are underestimated, that significantly changes the mentality of transit planning.
Edit: Soon after I said this, I came across yet another discussion on Reddit with many inputs on C&C vs TBM. Quoting part of one such comment:
But people involved with the transitcosts.com project have said that the difference between full cut-and-cover and the mentioned world standard [bored tunnels, C&C stations] is not that big.
Much of the discussion was also in the context of the BART San Jose extension, where people pointed out political opposition from residents and commercial districts that not only stopped the whole route from using C&C, but stopped even the station from using C&C (which is standard practice for bored subways).
Since you're going to West either way, another option might be a more traditional El with LRT over the B&A and at-grade intersections with TSP at the 14 or so bridges from Boston Landing to Riverside. Newton Corner would obviously be tricky. Tunneling here, when there's no negotiation about the highway going away, just seems nuts.
In all seriousness though, I think 15 minutes is about the max of what you could do inside 128 limited to 2-tracks on the B&A Mainline will still maintaining slots for services west of Worcester at useful running times. Newton seems to want to prioritize upzoning along the B&A rather than the Highland Branch, and the whole stretch in Boston from Newton Corner to Beacon Park/West is ripe for increased density if there were better transit options.
Suppose a Grand Junction rapid transit route is built. How feasible and costly would it be to add a Red Line infill at where it crosses the Grand Junction?
Intuition behind it: The transfer between Red Line and Grand Junction absolutely sucks, which doesn't help people living on the Red Line who want to transfer to a ring route. While southbound transfers have a low-cost solution of a Harvard - West Station - BU branch, northside transfers, especially to northside OL*, don't have an equivalent option. I expect the Red Line infill to be expensive, but the question is how does it compare to a whole new deep bored subway north from any of Kendall station, Central or Harvard.
* GLX to Porter actually helps somewhat, but it doesn't reach OL until North Station.
I think a Tech Square infill is somewhere between "reasonable" and "crazy", but probably not god mode. I think the station itself would be a good idea. Centered on Albany Street, you're looking at about 1,700 feet spacing, and the density would certainly justify it. Relief of platform crowding at Kendall would reduce some of the increased running time from adding a station.
It would be difficult but doable to construct it without closing the Red Line entirely, though you'd certainly have night/weekend closures. Main Street would probably either need closed entirely or temporarily made one-way, but it's an area that should be trying to reduce the number of vehicles anyway. Hopefully you could avoid dealing much with nearby foundations except for integrating headhouses. It's a 100% cut-and-cover job on a shallow tunnel, which helps. With Boston transit prices you're probably talking in the $100-150M range.
Suppose a Grand Junction rapid transit route is built. How feasible and costly would it be to add a Red Line infill at where it crosses the Grand Junction?
Intuition behind it: The transfer between Red Line and Grand Junction absolutely sucks, which doesn't help people living on the Red Line who want to transfer to a ring route. While southbound transfers have a low-cost solution of a Harvard - West Station - BU branch, northside transfers, especially to northside OL*, don't have an equivalent option. I expect the Red Line infill to be expensive, but the question is how does it compare to a whole new deep bored subway north from any of Kendall station, Central or Harvard.
* GLX to Porter actually helps somewhat, but it doesn't reach OL until North Station.
That's not necessarily a reason not to do it. Park and DTX are also super close, but I don't think anyone is arguing that it would be good to have all RL transfers use the platforms at Park St. They're crowded enough as it is.
I imagine the platform to be placed just west of Grand Junction, and end nicely at the edge of the green space in Tech Square at exactly 430 ft, which is about the length of most RL platforms. This means its distance from Kendall (center to center) is 1572 ft or 0.300 miles:
Comparing this 0.300 miles to other stops (in miles, source is this MBTA dataset):
Park St to Downtown Crossing: 0.127
Downtown Crossing to South Station: 0.336
Fields Corner to Shawmut: 0.586
Haymarket to State: 0.286
Ruggles to Roxbury Crossing: 0.519
The rest of OL South average about 0.5
Aquarium to State: 0.312
Government Center to Bowdoin: 0.176
Beachmont to Suffolk Downs: 0.528
Arlington to Copley: 0.354
Fenway to Longwood: 0.356
Medford/Tufts to Ball Square: 0.597
(Major edit: After writing this, I found that there are some inconsistencies with the MBTA data. Park-DTX looks like edge-to-edge distance, but GC-Bowdoin seems like center-to-center, and NS-Haymarket is inbetween the two measurements. The edge-to-edge distance of Kendall-Tech Square would be 0.216 miles. Though this is still similar to Chinatown-TMC and NS-Haymarket, and the point about RL being able to afford more stops still stands.)
So 0.3 miles is completely reasonable by downtown standards, and only slightly closer than average for stations further away from downtown. It certainly seems close on the Red Line, but that's only because it's the line with the longest stop spacing by far (Alewife to Kendall average about 1 mile per station except Davis-Porter, and South Station to Savin Hill is also longer than other HRT lines). And considering how much of a major employment hub Kendall is, how spread out it is and how much its ridership is peak-oriented (which means greater peak ridership and longer dwell times, unlike Harvard), having another station to increase coverage and relieve the load at Kendall seems reasonable.
That's not necessary a reason not to do it. Park and DTX are also super close, but I don't think anyone is arguing that it would be good to have all RL transfers use the platforms at Park St. They're crowded enough as it is.
Funny enough, yesterday I was just revisiting this blog post by Ari Ofsevit explaining why combining Park St and Downtown Crossing makes no sense, despite the fact that it was somehow mentioned by the MBTA's Fiscal and Management Control Board. (It's also a very informative post that explains the Red Line's other headway bottlenecks, like Harvard Curve and Alewife.)
--------------------------------------------
I also checked the walksheds. First, here's the 10-min walksheds of Kendall and Central:
There's a visible gap between the walksheds. It contains a bunch of apartments north of Main St, but also the triangular area between Main St and Mass Ave, which contains Novartis headquarters, Pfizer and many other offices. They already contribute to commuting ridership on the 1 bus, and I suspect are the reason for the (surprisingly high) commuting ridership at Central. There aren't even any buses running on Main St (although BNRD's T70 takes care of that).
Now add our Tech Square station into the mix:
Even though there's a lot of overlap with Kendall, the non-overlapping regions of Tech Square nicely fill the gap between Kendall and Central, and even extend further north. The station is also much closer to MIT's core campus (to the point where I'd consider renaming the stations Tech Square/MIT and Kendall), and many offices, apartments and other developments at Tech Square itself and further up the Grand Junction corridor.
While the only trigger for such an infill to be considered is a Grand Junction rapid transit route, I do think there are other benefits to the station beyond just the transfer.
There are a few other ways you can place the 430' platform:
(Yellow is the one I presented earlier. I didn't intentionally make 7 alignments for a rainbow, there just happened to be that many.)
This is an area with many tall buildings (some of which go above the Grand Junction, which I find cool but presents problems for transit), so the exact alignment will likely depend on building foundations and how far the buildings are from the street. Each one has pros and cons: moving towards purple is easier but closer to Kendall, while moving towards red improves walkshed but puts your platform in a narrower section between two closer buildings next to Portland St.
Outbound headhouses will almost certainly take the green spaces, either in the middle of Tech Square or at the park next to Galileo (with stone carvings that say "Grand Junction"), integrated with the GJ platform. Inbound can be a bit tricky, but they can take the small triangular space west of Albany St before the new building. The cyan/blue/magenta options can also take some space on MIT campus where the Collier Memorial is.
FWIW, even a station centered at Portland St doesn't improve walkshed coverage by much.
Suppose a Grand Junction rapid transit route is built. How feasible and costly would it be to add a Red Line infill at where it crosses the Grand Junction?
Intuition behind it: The transfer between Red Line and Grand Junction absolutely sucks, which doesn't help people living on the Red Line who want to transfer to a ring route. While southbound transfers have a low-cost solution of a Harvard - West Station - BU branch, northside transfers, especially to northside OL*, don't have an equivalent option. I expect the Red Line infill to be expensive, but the question is how does it compare to a whole new deep bored subway north from any of Kendall station, Central or Harvard.
* GLX to Porter actually helps somewhat, but it doesn't reach OL until North Station.
I've read that when the Red Line was built there was supposed to be 5 stations in Cambridge, but people around Harvard wanted to get to Boston faster, so it was limited to to just Kendall, Central, and Harvard. Those stops are really too far apart, I think. I've been wondering if there were ever plans to fill in the "missing" two stations in Cambridge. So as pointed out, a new station at GJ would fill in part of that gap (even if it's close-ish to Kendall). Another infill station would be between Central and Harvard, maybe centered on Dana St. That would be big improvement, I think. That's something I wish was done a century ago, but hard to say it would be worth the $100's of millions today.
I've read that when the Red Line was built there was supposed to be 5 stations in Cambridge, but people around Harvard wanted to get to Boston faster, so it was limited to to just Kendall, Central, and Harvard. Those stops are really too far apart, I think. I've been wondering if there were ever plans to fill in the "missing" two stations in Cambridge. So as pointed out, a new station at GJ would fill in part of that gap (even if it's close-ish to Kendall). Another infill station would be between Central and Harvard, maybe centered on Dana St. That would be big improvement, I think. That's something I wish was done a century ago, but hard to say it would be worth the $100's of millions today.
A Dana St infill station seems questionable. If it were super cheap then maybe, but it's probably not. "Tech Square" is (IMO) way more justifiable, arguably more of a Crazy Transit Pitch than a godmode one.
A Dana St infill station seems questionable. If it were super cheap then maybe, but it's probably not. "Tech Square" is (IMO) way more justifiable, arguably more of a Crazy Transit Pitch than a godmode one.
Agreed. Given the existence of the 1/T1 bus, a much more cost-effective (and perhaps even better) solution would be to improve its reliability and frequency. Especially in the world of many crayonists that call for LRT conversion of the 1 bus, it basically fulfills the role of a Red Line infill. (Whether that should be done is another question.)
Another infill station would be between Central and Harvard, maybe centered on Dana St. That would be big improvement, I think. That's something I wish was done a century ago, but hard to say it would be worth the $100's of millions today.
Growing up in Cambridge, my mother and I used to walk from Harvard Square to Putnam Ave to Kinnaird St all the time, and it seemed like a short walk. I never thought of the Harvard and Central Square stations as being very far apart.
Growing up in Cambridge, my mother and I used to walk from Harvard Square to Putnam Ave to Kinnaird St all the time, and it seemed like a short walk. I never thought of the Harvard and Central Square stations as being very far apart.
Sure, but there are other factors to consider for stop spacing, like residential density, job density, platform crowding, etc. An infill between Central and Harvard would shave off a bit of walking time, but for a ton of people. I don't know how to factor all those in, but the 1 mile spacing in Cambridge seems like it's at the far end of what's reasonable. It makes public transit less usable for lots of people because of the added walk and crowded platforms. Just for comparison, the Orange Line south from Jackson Square has stops about every half mile, despite much lower job and residential density.
