I imagine that in a D-E connector world (where a connection is built between Riverway and Brookline Village), the 20 tph currently consumed by the D and E would get divvied up between Riverside, Needham, and Heath, probably not evenly. The simplest way to envision this would be to reroute half of E trains to Needham:
- 12-min freq to Needham
- 6-min freq to Riverside
- 12-min freq to Heath
(Worth noting that, on the one hand, the VA Medical Center is roughly a six minute walk from Riverway, which normally would be considered comfortably within the walkshed; however, given the more vulnerable population served by the VA Medical Center, having some level of front-door service remains important.)
6 min probably slightly overserves Riverside, and 12 min probably slightly underserves Heath, so probably you’d want to tweak these a bit — perhaps swap 1 or 2 Riverside trains per hour out through Heath instead. Those dropped Riverside trains could be supplemented by short-turns at Kenmore, which I think could be feasible.
(Needham is jumping from one train per hour to sub-15 min freqs, so I think 12 min isn’t an unreasonable starting point. The three Needham stops currently see a little bit more ridership combined than uses Riverside station alone, so I also don’t think throughput would be a problem on Day 1.)
But also, keep in mind that a surface D-E connector could easily serve as an Aldgate Junction (actually a dual Aldgate Junction), meaning that 12-min Park St <> Heath frequencies could be supplemented by trains coming from Kenmore and/or Reservoir. (Kenmore would require a somewhat fancier junction at Brookline Village, but a Reservoir <> Heath service would simply require a third leg to the junction at the intersection of Huntington + South Huntington.)
Finally, it’s worth remembering that 5 branches is not actually a huge problem. Assuming one branch turns at Park St, we know the Green Line has historically handled 40 trains per hour. 40 divided by 4 branches yields 6-min freqs; 40 divided by 5 branches yields 7.5-min freqs, which is still comfortably within turn-up-and-go, and I suspect would not raise the ire of the community. Especially after this last year on the T… reliable 7.5 min frequencies probably sound pretty good:
So, yeah. Particularly if the fifth branch is to Needham, I’m not super convinced that adding a fifth branch creates an immediate capacity concern. The T would definitely still need to keep their eye on the ball and continue work on a long-term plan to fix Copley Junction (e.g. a subway to Pleasant St via Back Bay), but I don’t think it needs to be a show-stopper.
(Also, bleh those headways on June 14 [chosen at random] at absolutely dreadful, especially on the E. What even is this?)
And from a throughput perspective, assuming single car T10’s with a capacity of 400 passengers, one hour’s worth of inbound Green Line trains at 10-min frequencies (6 tph) comes out to 2400 passengers.
Which means that one hour’s worth of trains would easily be able to mop up the 1400 inbound riders from all four stations in Needham (according to the 2018 survey). The higher numbers in the Bluebook — 2900 — would take a bit more than 7 trains’ worth, and could be buzzed through in 1h10m or 1h20m.
In reality, the morning rush gets smeared out over about 2 hours (departing Needham between 6am and 8am), during which there will be capacity for 4800 passengers leaving Needham during that time. So there should be plenty of capacity and room to grow. (And if they run double-car T10 super trains, there’s no contest.)