Crazy Transit Pitches

[ritchiew]

So You Want To Build A Route 128 Transit Line

A circumferential transit line using 128 is one of those ideas that pops up every so often. There's currently almost no circumferential service beyond ~5 miles from downtown Boston, even in the densest areas. There used to be - streetcar lines and later bus lines along the section between Needham and Salem - but those gradually died out during the 20th century. Last night, my insomnia wondered what it would take to have useful transit at a ~10 mile radius from Boston.

Three things are clear immediately:

• It's 128 or nothing. Except in Peabody, there aren't any circumferential rail rights-of-way. The local roads that connect the 128-belt towns are largely older two-lane roads that couldn't support much in the way of BRT-style enhancement. You could string together useful circumferential services from rail lines in 495-land, but not 128-land.
• Anything other than stuck-in-traffic buses is pretty instantly a megaproject. At peak hours, almost the entire 128/95 concurrency is slowed in one or both directions. The minimum viable service would be to convert the innermost lanes to bus-only; even then, you have to build some pretty substantial stations just to be reachable from street level. Some of the most valuable stations - transfers with existing radial lines - are the most complex.
• It would cater to relatively short trips. In most cases, electrified regional rail via NSRL will beat a circumferential line for trips along a large portion of the arc. The value of the ring service would largely be in connecting those radial lines to points a few miles away.

With that, here's my sketch of a full build-out, assuming automated light metro. Some combination of stealing lanes, running alongside 128, and viaduct over the highway.
View attachment 60391

You can hit every commuter rail line but Braintree and three of the four subway lines. Three potential spurs to Lincoln Labs/Hanscom, Anderson RTC, and Danvers.

View attachment 60392
Salem is the obvious northern endpoint, though going through Peabody would either require a tunnel or viaduct through the downtown grade crossings.

View attachment 60393
I went back and forth on whether to have separate stations for the Framingham/Worcester Line and Green Line transfers. This is the combined station, which lets you stay along the 128 mainline and gives a CR/GL transfer, but would need a new park-and-ride built at the interchange. (The existing Riverside station would become a through station with less parking.)

View attachment 60394
This assumes that the Needham Line has already been replaced by GL to Needham Junction and OL to VFW Parkway. If you build the circumferential line, I think that justifies extending the OL to Needham with a new transfer.

View attachment 60395
Braintree can be reached either by a viaduct above the rail lines or by the Route 3 alignment. You could plausibly go for Quincy Center instead, especially if you could share tracks with the Red Line.

This is a neat idea, and I keep thinking about the value of circle lines. But I don't really see anything like this working very well. Most of the stops are in low density areas with bad pedestrian access. In a lot of cases that probably won't be fixed because it's in a highway, which is not a good place to try to wedge in pedestrian-friendly areas. The stations are mostly not natural origin or destinations points for anyone's trips. That would mean this offers virtually zero one-seat-rides, maybe some two seat rides, and realistically nothing too much easier than a three-seat-ride (maybe bus to circle line to a subway). As bad as driving might be, it could still be preferable.

But again, I like this idea in principle. Are there some travel patterns you think this would be especially useful for? Or some stations you see as especially useful (or ripe for development)?

 

[The EGE]

Yeah, this is not intended to be a particularly serious proposal, though there definitely is a need for additional suburban and circumferential services. Simply getting local bus service between suburbs would be a great start, and perhaps a good use case for the sort of smaller automated buses that have been popping up. The fact that you can't currently take any sort of public transit from Stoneham to Woburn or Wakefield, or Needham to Wellesley, is ridiculous.

Other than Salem-Peabody-Danvers (which would be easier as a surface light rail or regional rail line), the North Waltham to Needham Crossing segment is the most viable. For 8 miles of construction you can hit two commuter rail lines, two Green Line branches, and two major employment nodes. I suspect that segment has a lot of shorter-distance circumferential trips that could reasonably be diverted to transit.

 

[jbray]

Yeah, this is not intended to be a particularly serious proposal, though there definitely is a need for additional suburban and circumferential services. Simply getting local bus service between suburbs would be a great start, and perhaps a good use case for the sort of smaller automated buses that have been popping up. The fact that you can't currently take any sort of public transit from Stoneham to Woburn or Wakefield, or Needham to Wellesley, is ridiculous.

Other than Salem-Peabody-Danvers (which would be easier as a surface light rail or regional rail line), the North Waltham to Needham Crossing segment is the most viable. For 8 miles of construction you can hit two commuter rail lines, two Green Line branches, and two major employment nodes. I suspect that segment has a lot of shorter-distance circumferential trips that could reasonably be diverted to transit.

Why don't we try and kill 2 birds with one stone. OLX to VFW/GLX to Needham and the Ring. Is it impossible to create a Wye junction with Newton Upper falls and Eliot? Here's my pitch:

Phase 1
-VFW Parkway (OLX Connection) to Worcester line via Needham Junction and Riverside. Is it a new station on Worcester line? Is is Auburndale? I prefer Auburndale, personally.

Phase 2 or 3 North
- North via Lexington St and a Moody Street line (subway? Streetcar?) to Waltham Center
- North on Lexington st to Totten Pond Road (So far this has recreated the streetcar line that used to exist.
- TPR to the High power electrical ROW which is right behind all of the labs there and can be followed all the way to Hanscom and Hartwell Ave where the Lexington Line will eventually cross.

Phase 2 or 3 South
- VFW Parkway South to Dedham Corporate or 128 (preferred)

Phase 4 or 5 North
- Hartwell to Burlington Mall
- You can use the highway or Burlington Mall Road or even land taking through the various parking lots to get across.
- Highway ROW to Main street in Woburn turning onto the old ROW to Anderson/Woburn.
- If OLX north goes to Anderson/Woburn we head back south to the highway ROW via the power ROW, if it only goes to Reading, then we go there although it would be better for the orange line to meet there.

Phase 4 or 5 South
- With south coast rail, Going to Canton is mostly out of the question So I would duplicate what you have presented.

Phase 6 North
- I would also use your ROW to Salem via the Peabody line although maybe there's a justification for going north to Lynnfield and using that ROW past the Northshore Mall to get to Peabody but that all looks low density to me.

 

[Riverside]

Phase 1
-VFW Parkway (OLX Connection) to Worcester line via Needham Junction and Riverside. Is it a new station on Worcester line? Is is Auburndale? I prefer Auburndale, personally.

Wait a second, yeah, this is really interesting. (I like the other Phases too, some overlap with ideas I’ve been playing around with myself.) The main question would be: is it faster to the get to the TV Place et al jobs from Auburndale via a light rail Aldgate Junction at Cook St, or a bus shuttle? Because the downside of the light rail is that the office parks are still somewhat spread out, and a shuttle would offer front door service.

But it is true, the way the radial ROWs are shaped there, you do kinda get a decent parallel to 128.

 

[TheRatmeister]

- VFW Parkway South to Dedham Corporate or 128 (preferred)

What about Readville? There's the stadium at Dedham HS to get around or under, but apart from that there's a clear ROW along VFW Parkway and then following the old railway alignment, which is mostly free from encroachment here. You could then continue to Routet 128/University Park if desired.

 

[Riverside]

Prompted by this discussion, I made a Google Map. I'll try to post more detail tomorrow, though I'm guessing a fair amount of what I'll say will be self-evident in the map.

 

[Riverside]

Alright, so let me explain some of my thinking and observations here.

First, we should articulate two (separate) goals of 128 service.

The "Outer Ring" problem: This is a generic problem that impacts all systems of this size. If you live in Braintree and work in Auburndale, it sucks to take the train all the way into Boston and then all the way back out again. All the more so if you live in Braintree but work at University Park. In my observation, systems often first build a circumferential "Inner Ring" service about 3 mi (5 km) from the core of the system. See, for example, the Urban Ring proposal, or the G Line in NYC. Most rapid transit systems in North America (again, in my observation) stretch out about 10 miles (15-16 km) from the core; when built, the "Outer Rings" of these systems are usually between 6 and 9 miles (10-15 km) away from the core. The TriBoro Rx proposal in New York and the Purple Line in Maryland fit this description, as would a 128 service.

Outer Rings create an interesting mathematical problem. As @The EGE notes, particularly with speedy radial Regional Rail services, there are many 128 <> 128 destination pairs where it would make more sense to ride through the core: for example, Woburn <> Westwood, or even Braintree <> Waltham. As it turns out, there's a geometrically elegant quantification of this: assuming equal speeds on all services, if your destination is more than (about) one-third of the way around the Ring, you're better off going through the core.

The circumference of a circle is the radius x 2 x pi -- or, written differently, the radius x 6.28. For example, treating 128 as being consistently 10 miles from downtown, we find that the Outer Ring would be 62.8 miles (101 km), minus the ~15 miles (25 km) taken up by the ocean. Any journey through the core will always be 20 miles, which is a little less than one-third of the length of the Outer Ring. For reference, Braintree (I-93 @ Route 3) to Waltham is just over 21 miles (34 km).

The math can get more complicated if your destination is somewhere between the core and the Outer Ring. For example, Braintree <> Forest Hills is roughly equidistant whether via Readville or donwtown. Likewise, Watertown Square <> Mishawum is also roughly equal whether via 128 or downtown. But the bar will always be the same: is the journey less than 20 miles? In theory, this would mean that an L-shaped journey like Watertown <> Mishawum could only be competitive if the circumferential leg is less than 10 miles; in practice, it won't quite work out this way because there are a finite number of radial lines, creating distorting gaps. (Consider the northwest corner, where there's a long gap between the Fitchburg Line and the Lowell Line.) But, in general, an Outer Ring circumferential service will be useful to travel along for 10 miles for 2SRs and 20 miles for 1SRs.

For reference, here are the mile markers for potential areas of interest and transfer points along an 128 Outer Ring service (starting at Quincy Adams and ending in Salem via an as-the-crow-flies hypothetical route from Boston Children's Peabody):

Point	Miles	Miles (rounded)	Kilometers (rounded*)
Quincy Adams (Red Line)	0	0	9
Route 128/University Park (NEC)	8	8	13
Dedham Corp Center (Franklin Line)	9.8	10	16
Cutler Park (Needham Line)	14.2	14	23
TV Place (Needham GLX)	16.6	17	27
Woodland (Newton-Wellesley Hospital)	18.8	19	30
Mass Pike	19.7	20	32
Route 20 (Fitchburg Line)	21.9	22	35
Totten Pond Road (North Waltham office parks)	23.9	24	38.5
Route 2	26.1	26	42
Lexington Branch	29.1	29	47
Middlesex Turnpike (Burlington Mall)	31	31	50
Mishawum (Lowell Line)	35.4	35	57
Lake Quannapowit (Reading Line)	38.3	38	62
South Lynnfield (Route 1)	43.5	43.5	70
Downtown Peabody	47.4	47	76
Salem (Eastern Route)	49.1	49	79

( * kilometers rounded from original miles)

All of which is to say... Outer Ring services are hard. Edge cities notwithstanding, you're out in the suburbs, so you're already fighting lower density. You're following a circle, meaning, definitionally, you are taking the "long way around" for any pair of destinations (though this is less true in practice for 128, where there are legitimate straightaways of substantive length). And you're almost certainly going to be competing with higher-speed, higher-capacity radial services that will usually require equal or fewer transfers.

I'll also note that ~50 miles one-way is long for a rapid transit service. If light rail, it would slightly edge out Los Angeles's A Line as the longest light rail line in the world. (And, from what I can tell, would itself be a contender for longest non-branched non-interlined rapid transit line in the world.)

So this brings us to the second goal for an Outer Ring 128 service: serving (specific) supersized office parks.

(Which I'm going to put in a follow-up post, which may or may not be written today).

 

[Riverside]

The second goal of an Outer Ring 128 service would be serving (specific) supersized office parks.

I would argue that this is actually a better primary goal, and a better starting point for planning.

The "Route 128 Corridor" has long been noted for being home to a range of technology companies (once described as Masaschusetts' "Silicon Valley"). The Corridor has grown in recent decades to also feature a wider array of businesses, as well as large shopping plazas and other developments. I wouldn't quite call any of these "edge cities," though Burlington Mall probably comes closest. This growth has also merged into existing corridors, such as Route 1 in Dedham/Norwood.

For the most part, recent developments have not been particularly "transit-oriented," aside from University Park (though even there, the development is still very car-centric). Indeed, despite being relatively centrally located within the bounds of its surrounding office parks, Mishawum station has always struggled for ridership -- unsurprising, given the lower frequencies and inflexible schedules historically provided by the commuter rail, combined with it being "not too bad" to commute by car, and by having commuters dispersed well-beyond the service area of the Lowell Line.

(Dedham Corporate Center gets decent ridership, but from what I can tell it is overwhelmingly oriented toward Boston, probably mostly driven by PnR traffic, and commuters from adjacent TOD housing.)

For these office parks to attract meaningful transit ridership, I think an increase to proper SUAG frequencies would be needed, in conjunction with "hop-on-hop-off" shuttle services to bring commuters closer to the front door of their office. The "convenience margin" is very thin in this scenario. Sitting in traffic sucks, but you have (the perception of) more control over when you travel, more flexibility to do errands on the way home, and often more comfort than on a train. Add a 15-minute walk in 85° humid heat, or pouring rain, or windy freezing twilight hours, and the inconvenience of sitting in traffic will almost always seem preferable by comparison (for those who can afford it). Rightly or wrongly, I think the need to cater to a "comfortable last mile" will be particularly strong (at least, in the current configuration of these parks; new housing, or more pedestrian-oriented streetscapes, could tip the scales).

As depicted on my map, I think there are seven key areas of interest for transit service along 128:

1. "Mishawum"
   • Stretching ~3 miles from Salem St to Oxbow Dr, hugging 93 and the Lowell Line
2. Burlington Mall
   • the length of Burlington Mall Road, plus Middlesex Turnpike up to Mitre St
3. Hanscom
   • Somewhat ill-defined, but essentially everything on Hartwell Ave and Barksdale St
4. North Waltham
   • Running parallel to 128 between Route 2 and Waltham's Main St, with an arm extending west to Waltham Woods
5. Needham Crossing
   • Much smaller area, hemmed in by the Charles River and 128 on opposite sides
6. Dedham Corporate Center
   • Largely well-defined borders, arbitrarily truncated at the Star Market on my map
7. University Park
   • Pretty much everything that hugs University Road, but I also note that there is a nearby park along Royall St

I agree that there are other places that would benefit from a 128 Outer Ring, such as the South Shore Plaza. My choices above are somewhat arbitrary, but are meant to focus on developments which would not be there if not for Route 128. South Shore Plaza and Crown Colony sit on the edge of high-density Quincy, and don't seem to have the same "vibe" IMO. But, again, my choices are arbitrary, though I think they are defensible in terms of being high priorities.

So, I would start my Outer Ring planning by identifying the best options for serving each of those areas, and then building from there.

Current Conditions

One interesting feature shared by several of these business parks is the wide boulevards favored by modern road designs. University Ave, Burlington Mall Road, and Commerce Way all have center medians, and most of the other parks have 4-6 lanes along their primary axis. The irony here is that these highway-sized widths were obviously designed for cars (and lots of them), but arguably could be partially reclaimed for transit use.

Another coincidental (?) common feature is the presence of/proximity to power line ROWs. Indeed, many of North Waltham's parking lots sit underneath power lines.

The combination of boulevards and power line ROWs actually creates a remarkably cohesive path from Mishawum to Anderson/Woburn to Burlington Mall to the Minuteman Branch; the power lines continue south from there, and the roads to Hanscom could probably be widened to include a median (or could have tracks laid alongside). Power line ROWs marked here:

That's... actually pretty remarkable. As I note in the Google Map, the Anderson/Woburn <> Burlington ROW looks roundabout, but it actually is almost exactly the same distance as it would be via the Lowell Line + 128 ROWs. And in addition to Burlington Mall Road itself, there is a consistent line of parking lots, between 128 and the buildings, which could serve quite nicely as a dedicated ROW (and likely get closer to the Lahey buildings than a median service on Mall Road could):

Heading south, the power line ROW continues past the intersection with the Minuteman Branch, but it traverses some complicated terrain, crossing the Cambridge Reservoir and 128 in close succession, which would be much more expensive to cross.

Once past that, however, the power lines run above the aforementioned parking lots, along a narrow linear corridor of office parks -- the kind of geography that lends itself well to fixed transit; this corridor is marred only by a dicey junction at the intersection with the Mass Central, where it would be more difficult to build a connection to a new Weston/128 PnR:

More details on the current conditions at Needham Crossing, Dedham Corp Ctr, and University Park can be found in the Google Map, but there's nothing (IMO) quite as impressive as the string of boulevards and power lines to the north.

So what would I propose?

 

[Riverside]

Building transit to 128's Office Parks

Phase -1: Maximize bidirectional all-day frequencies to Anderson/Woburn, Dedham Corp Ctr, and Route 128. I don't think any of this has a prayer of working without 20-min frequencies at the very least (and I think it's possible it'd still be DOA until/unless we achieve 13-min SUAG headways). I would also look at extending the Orange Line to Reading and then to Anderson/Woburn, partially by way of -- you guessed it -- a power line ROW that vaguely runs between the two mainline ROWs. The Reading Line corridor is well within normal commuting distance to Burlington Mall, to say nothing of Mishawum, so the Orange Line connection is a significant benefit. (Optionally, also complete Needham GLX and Weston/128.)

Phase 0: Implement high-frequency low/medium-capacity shuttle services between stations and offices, in "Mishawum" (idk an actual name for this area), Burlington Mall, North Waltham (if Weston/128 is built), Needham Crossing (if GLX) and University Park. I think Dedham may not need a shuttle, although if the parcels toward the north end up getting larger businesses, that might drive enough demand; for the moment, Legacy Place is a 6-minute walk from the platforms. I think Hanscom is too far from potential transfer points.

Phase 1: Using the power line ROW and the line of parking lots, build a light rail line from Anderson/Woburn to Burlington Mall, with 10-12 minute headways. Running non-stop between Anderson/Woburn and the edge of Burlington Mall Road, it should be able to run at a pretty decent clip. Stops within the Burlington Mall area TBD, but I would truncate the line before it crosses Middlesex Turnpike. In principle, this entire segment could be in a dedicated ROW (albeit with some grade crossings).

Phase 2: Create an Aldgate junction between the Riverside and Needham branches, and extend tracks from Riverside to either Auburndale or Norumbega/128; run a light rail shuttle service that connects the Worcester Line to Needham Crossing (and Newton Wellesley Hospital and downtown Needham). I don't expect this to be an amazingly high-yield extension, but in principle it would be relatively simple to do.

Phase 3: Extend the Burlington Mall Line. At this point, you have some choices:

3a: Blue Sky Center: looking strictly at job counts, turning north along/parallel to Middlesex Turnpike probably has the highest payoff

3b: Hanscom: following the power line ROW gets you to the Minuteman Branch

One option from there would be to run in or along the streets to Hanscom Air Force Base. For the most part, buildings are set back from the road, so it could be possible to create a median, or dedicated ROW alongside.

3c: Red Line connection

The other option would be to turn southeast and use the Minuteman Branch to reach a Red Line transfer station in Lexington, Arlington, or at Alewife. This would significantly increase the number of 2SRs and 3SRs to the Burlington Mall

3d: some combination of the above

Perhaps there is a pair of services: Anderson/Woburn <> Blue Sky Center, and Arlington Heights <> Burlington Woods, with interlining within Burlington Mall proper. Or perhaps it's Anderson/Woburn <> Hanscom, Arlington Heights <> Blue Sky Center, with interlining along the power line ROW and a good transfer point somewhere near the AMC cinema. At this point, Burlington Mall will have had its Anderson/Woburn light rail service for a little while, so it will be clearer where demand would be.

Phase 4: Build a light rail line through North Waltham (maybe). On the one hand, this is a pretty narrow linear corridor (the arm off to Waltham Woods notwithstanding) that could do well with transit. On the other hand, it's a bit cramped and there's less space for a small maintenance facility than there is to the north; the highway itself also pretty cleanly divides the area in two, and it would be hard for a single line to serve both sides. If Route 2 and the Cambridge Reservoir could be crossed, in theory you could have a line running all the way from the Fitchburg Line to the Lowell Line. But, for the reasons I went through previously, I'm skeptical of the viability and need for something like that. On the balance, I think this corridor will probably be a better fit for frequent shuttles for a while.

Phase 5: Cross the Charles River (maybe maybe maybe). Brandeis and Auburndale are just over 1 mile (just under 2 km) apart. It seems like it should be easy to build a line between them. But as soon as you start looking more closely, it gets gnarly fast. Rivers, islands, residential neighborhoods... absent the aforementioned subway under Moody St, or stealing lanes from 128, or building a viaduct over 128 (all of which seem like tall orders), I don't think there's really a good way to do this.

Phase Triangle (by which I mean, done independently of the above phases): implement an open BRT system on 128. In an open BRT system, buses running on non-BRT infrastructure (e.g. mixed traffic) move into dedicated BRT infrastructure for some amount of time before leaving again. (Somewhat similar to streetcar branches entering a subway and exiting at the other end.) Open BRT is useful when there's a high concentration of overlapping routes, but with a disparate range of destinations. Adding BRT infrastructure along 128 would allow for a layer cake of routes that can originate within town centers (not just constrained to being near 128 itself), speed along 128, and exit at various office parks where they can provide door-to-door services for different offices.

Despite BRT's reputation as the "low-cost alternative" to everything, I'd argue that proper open BRT on 128 would actually be a significant investment, since it would probably require significant stretches of grade separation where there's no room to widen 128 -- meaning, viaducts. The viaducts would also need dedicated access points -- potentially new ramps, although I've always wondered whether an open BRT system that has at-grade intersections with overpasses might actually be viable; there would be long stretches without intersections, so high speeds would be achievable, and the at-grade intersection would solve the access point problem. And since the overall number of vehicles would be lower, it might not be a problem to have the hard right angles of the intersections.

In closing

Like I said, I think 128 is hard. I think the particular location and layout of the Burlington Mall (and Middlesex Turnpike) area means that one or two services centered on it might be unusually viable. But, as much as anything, that's because there's an unusually wide gulf between the Lowell Line and Fitchburg Line. So, perhaps I'd say that I'm mildly optimistic about a "Burlington Mall transit expansion", and modestly pessimistic about a "128 Outer Ring".

(That all being said, to be clear -- I think @The EGE's map is really cool in its own right. My comments are meant to complement his, not try to rebut.)

EDIT: I want to more explicitly acknowledge @jbray here — he laid out a lot of what I presented here already (including making the IMO vital observation about the parking lots in Burlington Mall). My suggested phasing differs from his, but many of the underlying ideas are his!

 

[The EGE]

This is fantastic analysis! I hadn't considered those power line ROWs, but they're an interesting idea, especially for shorter-distance lines where speed isn't as critical. They do have some steep grades, and certainly wouldn't be suitable for mainline rail, but light rail might be doable.

 

[ritchiew]

This is fantastic analysis! I hadn't considered those power line ROWs, but they're an interesting idea, especially for shorter-distance lines where speed isn't as critical. They do have some steep grades, and certainly wouldn't be suitable for mainline rail, but light rail might be doable.

Yeah, this is a neat idea, but the grades along the power lines just don't work for rail. From a quick look, this hill that would be part of the route is far too steep. For about 500 feet, the average slope is over 13%. This hill is over 10%. There are bunch of other smaller quick drop-offs. Even light rail can't handle this terrain. To follow these ROWs would mean subways or massive viaducts. That's expensive for what is really a low density route. Also, if you really wanted to just build subways, there's less reason to need to follow the power lines.

 

[Riverside]

This is fantastic analysis! I hadn't considered those power line ROWs, but they're an interesting idea, especially for shorter-distance lines where speed isn't as critical. They do have some steep grades, and certainly wouldn't be suitable for mainline rail, but light rail might be doable.

Yeah, this is a neat idea, but the grades along the power lines just don't work for rail. From a quick look, this hill that would be part of the route is far too steep. For about 500 feet, the average slope is over 13%. This hill is over 10%. There are bunch of other smaller quick drop-offs. Even light rail can't handle this terrain. To follow these ROWs would mean subways or massive viaducts. That's expensive for what is really a low density route. Also, if you really wanted to just build subways, there's less reason to need to follow the power lines.

[alt text: screen capture from Star Trek II: The Wrath of Khan, where Kirk is looking to Spock, who is subtitled as saying "His pattern indicates two-dimensional thinking."]

(I say self-deprecatingly.)

There are other ways to handle steep grades, including open cuts, but agreed that that increases cost and complexity on an already-challenging proposal.

I'd be willing to look more closely at taking space along 128 itself for the ~3 miles between Mishawum station and Burlington Mall; I do think there's a valid need for transit along here, but it's hard. For reference, OnTheMap visualizations below. (Boy, North Waltham looks really tempting.)

 

[ritchiew]

[alt text: screen capture from Star Trek II: The Wrath of Khan, where Kirk is looking to Spock, who is subtitled as saying "His pattern indicates two-dimensional thinking."]

(I say self-deprecatingly.)

I can be quick to point out your mistake only because I've done it so many times myself.

 

[Riverside]

For funsies, here's the elevation profile of the Woburn <> Burlington power line ROW according to Google Earth Pro:

Some problem points:

Looking at the most painful of these:

^ This one is probably the "easiest" to address, in that it would be about 1000 feet, and the cut wouldn't need to be that deep.

This one is hard, as you basically have no choice but to rise 50 feet. In theory, I think you could do that along ~725 feet of distance. It also looks like there are no road crossings here, so I wonder if you could use earthworks at less expense.

^ Similar story here, though the elevation change is ~75 feet, which would need 1100 feet for a <7% grade. Also, this one would need to cross some water.

^ This one is the gnarliest. Would need to be a tunnel or open cut (or I guess a viaduct, but that seems even harder to imagine). It's about 2000 feet; with 6.9% grade, you could rise and then fall ~70 feet along that distance, which means the peak of 7% grade tracks would be at about 170', compared to the peak of the existing contour, 264'. That would be a 100' deep cut, which seems... daunting.

In summary, I think the back of the napkin suggests:

• 3000 feet of cut or tunnel (at least 2000 of which probably would just need to be a tunnel)
• 1800 feet of viaducts or earthworks

If that last segment can't be a tunnel or cut, I think you'd be looking at about 3000 feet of viaduct to handle that (maybe a little less if you can do a more modest cut at the top), for a total of 4800 feet (1.46 km) in total.

The overall path for just the power line ROW (excluding anything within Burlington Mall) is 3.8 miles (6.1 km), 0.9 miles (1.45 km) of which would be elevated. Assuming ~$250M per mile at surface level and ~$500M per mile for elevated, the overall cost for this segment might be estimated at (250 x 2.9) + (500 x 0.9) = 725 + 450 = ~$1.2B. Adding the Burlington Mall segment would probably bring it to around $1.7B.

For comparison, a path along 128 from Mishawum station would be just under 2.8 miles (4.5 km); at the naive estimate of $500M per mile, that would put the non-Burlington Mall segment at ~$1.4B. (But I'm guessing that the expense would be significantly higher than $500M, since you'd be building over a busy highway.)

And finally... just for comparison. If we imagined that the power line ROW posed no grade problems, its 3.8 miles times $250M per mile would be ~$950M.

All of which is to say, navigating the grades might not actually blow out the costs much more than numerous other aspects of the project. If it were somehow to make sense to build at $950M, it seems like it would probably make sense to build at $1.2B too.

(Why do I get pulled down these rabbit holes? Who can say.)

(Also, like, the power line ROWs look convenient, but, you know, idk how wild any contractor would be about building elevated viaducts under high-power electrical lines.)

 

[Stlin]

I think one of the issues of a I95 circumferential service is how it also basically represents the extreme edge of the MBTA bus service area - and one of the weaker aspects of our transit system is the weak connections between the regional RTAs, which would likely be better suited to providing that last-mile connectivity into the individual office parks and providing access and connectivity from the "outside" of the outer loop to include the closer Gateway cities like Lowell & Lawrence. In terms of that last mile, I do think that while Lexingtons Lexpress & the Beverly Shuttle are good examples, they also represent a incredibly fragmentary system such that its really quite difficult to do route planning.

That said, I am intrigued by the possibilities in an almost pure play park and ride scenario - the pitch to commuters could be as simple as "instead of sitting in I95 traffic, only drive as far as your local office park for a nearly seamless shuttle to work, no matter which office park your job is actually in" - that has possibilities to me in reducing VMT and improving job mobility and seems really marketable to office working folks. In theory, given the roughly balanced traffic loads seen on I95, the parking load should be balanced across the corridor, so no office park should get overloaded - if a majority join in, and if there's local demand? The RTAs now have a ready base to provide connectivity for. The 128 Business Council network of office parks, extended into the areas highlighted above, seem like it would be a good basis for expansion as a pilot, if you allow them to run bus-on-shoulder.

 

[jbray]

EDIT: I want to more explicitly acknowledge @jbray here — he laid out a lot of what I presented here already (including making the IMO vital observation about the parking lots in Burlington Mall). My suggested phasing differs from his, but many of the underlying ideas are his!

Thanks!

The second goal of an Outer Ring 128 service would be serving (specific) supersized office parks.

I would argue that this is actually a better primary goal, and a better starting point for planning.

The "Route 128 Corridor" has long been noted for being home to a range of technology companies (once described as Masaschusetts' "Silicon Valley"). The Corridor has grown in recent decades to also feature a wider array of businesses, as well as large shopping plazas and other developments. I wouldn't quite call any of these "edge cities," though Burlington Mall probably comes closest. This growth has also merged into existing corridors, such as Route 1 in Dedham/Norwood.

For the most part, recent developments have not been particularly "transit-oriented," aside from University Park (though even there, the development is still very car-centric). Indeed, despite being relatively centrally located within the bounds of its surrounding office parks, Mishawum station has always struggled for ridership -- unsurprising, given the lower frequencies and inflexible schedules historically provided by the commuter rail, combined with it being "not too bad" to commute by car, and by having commuters dispersed well-beyond the service area of the Lowell Line.

(Dedham Corporate Center gets decent ridership, but from what I can tell it is overwhelmingly oriented toward Boston, probably mostly driven by PnR traffic, and commuters from adjacent TOD housing.)

For these office parks to attract meaningful transit ridership, I think an increase to proper SUAG frequencies would be needed, in conjunction with "hop-on-hop-off" shuttle services to bring commuters closer to the front door of their office. The "convenience margin" is very thin in this scenario. Sitting in traffic sucks, but you have (the perception of) more control over when you travel, more flexibility to do errands on the way home, and often more comfort than on a train. Add a 15-minute walk in 85° humid heat, or pouring rain, or windy freezing twilight hours, and the inconvenience of sitting in traffic will almost always seem preferable by comparison (for those who can afford it). Rightly or wrongly, I think the need to cater to a "comfortable last mile" will be particularly strong (at least, in the current configuration of these parks; new housing, or more pedestrian-oriented streetscapes, could tip the scales).

I agree with this completely and I've bolded and sized up what I think is the most important thought. People are usually arguing about whether an outer ring is valuable for connecting trips or lines, but the real issue is that a significant amount of jobs are being loosely concentrated along all of 128 and if we don't come up with a plan to make those spaces transit oriented, then they will continue to sprawl and make traffic worse. When I envision the outer ring, it's not about trips from Braintree to Reading via a ring, it's about connecting all of the density to the jobs along the corridor while also allowing regional rail to have an approximate two-seat ride to most jobs on 128. So for example, to answer Rat:

What about Readville? There's the stadium at Dedham HS to get around or under, but apart from that there's a clear ROW along VFW Parkway and then following the old railway alignment, which is mostly free from encroachment here. You could then continue to Routet 128/University Park if desired.

It's probably easier on paper to build it to Readville, it connects to two lines at once instead of two separate stations; however, Readville is a trainyard with industrial and there is a need to turn it into a maintenance facility and expand it's footprint. Additionally, I didn't want to take space out of the main line ROW to go further South and West. Going to Dedham Corporate and 128 via Legacy place allows us to connect people to job centers and people to other people. If Norwood becomes a secondary Airport, we have an easy way to attach a line. I wasn't trying to find the quickest route but instead the one that touches the most things without being too squiggly (there is certainly some squiggle). Which is why, if I'm going to spend money on viaducts and tunnels to get over the water barriers between Newton and Waltham, I would always advocate to put the train down Moody street instead of going to the Route 20 superstation; it's like the Salem tunnel and we don't question whether that was valuable based upon Salem's density which is comparable to Waltham (although not as long as what I'm potentially proposing). This was all historically connected by rail. All I've done is swapped Lake street in Waltham for Totten Pond Road because that gets us closer to the Labs sooner without getting too much in the way of the hills there. Put a station there at the junction of the power row and TPR and feed buses from Border Road and Winter Street.

This type of line does not lend itself to high speed travel for someone going long distances, but is very comparable to people taking the OL or RL from near end to end like Quincy Center to Harvard.

@Riverside, I specifically avoided that Woburn power ROW because it's terrible especially compared to using the highway with potential stations at Winn st, Main street, then using the norther part of the Woburn loop (gotta move that food pantry) and connecting to Anderson; it's already flat.

 

[TheRatmeister]

This is more applicable to the Waltham/Burlington area where the office parks are most heavily concentrated, but what about some kind of busway, similar to the Metrobus in Istanbul? Running in the highway median, you could have some key stations at denser areas, and then move off the highway for short stops in the various office parks. Private entities could run their own shuttle services if desired, something that's really not possible with rail, and you could branch to your hearts content for lots of 1SRs.

 

[Riverside]

This is more applicable to the Waltham/Burlington area where the office parks are most heavily concentrated, but what about some kind of busway, similar to the Metrobus in Istanbul? Running in the highway median, you could have some key stations at denser areas, and then move off the highway for short stops in the various office parks. Private entities could run their own shuttle services if desired, something that's really not possible with rail, and you could branch to your hearts content for lots of 1SRs.

Yeah, this is my suggestion for an open BRT network. There's a lot of flexibility opened up by this. (And, I mean, it does kinda make sense: office parks were designed to be reached by large numbers of vehicles, with roughly equal ratio of vehicles to destination pairs, that are funneled through high volume dedicated corridors; an open BRT system with lots of routes still uses the same formula, just with adjusted variables -- fewer vehicles, more than one destination pair per vehicle, and BRT lanes rather than highways.)

... the regional RTAs, which would likely be better suited to providing that last-mile connectivity into the individual office parks and providing access and connectivity from the "outside" of the outer loop to include the closer Gateway cities like Lowell & Lawrence. In terms of that last mile, I do think that while Lexingtons Lexpress & the Beverly Shuttle are good examples, they also represent a incredibly fragmentary system such that its really quite difficult to do route planning.

That said, I am intrigued by the possibilities in an almost pure play park and ride scenario - the pitch to commuters could be as simple as "instead of sitting in I95 traffic, only drive as far as your local office park for a nearly seamless shuttle to work, no matter which office park your job is actually in" - that has possibilities to me in reducing VMT and improving job mobility and seems really marketable to office working folks. In theory, given the roughly balanced traffic loads seen on I95, the parking load should be balanced across the corridor, so no office park should get overloaded - if a majority join in, and if there's local demand? The RTAs now have a ready base to provide connectivity for. The 128 Business Council network of office parks, extended into the areas highlighted above, seem like it would be a good basis for expansion as a pilot, if you allow them to run bus-on-shoulder.

This is really interesting. Is your thought that every office park along 128 would essentially become a park-n-ride? Or do you also envision this system extending beyond that further into the suburbs?

I wonder how long it would take a bus to "deviate" from 128 to serve each office park along the way....

@Riverside, I specifically avoided that Woburn power ROW because it's terrible especially compared to using the highway with potential stations at Winn st, Main street, then using the norther part of the Woburn loop (gotta move that food pantry) and connecting to Anderson; it's already flat.

It's almost as if there's a reason no one has suggested it before me!

I basically agree with you, although it is true that we're trading difficult grades for difficult construction/disruption along the highway, with more close abutters. On the balance, it's probably still a better payoff to go via 128, but like I said I think the math potentially looks broadly similar for both.

I've thought about intermediate stops at Winn St and Main St; it's seemed dicey because a) relatively low density around, b) Winn St doesn't have un-/underdeveloped lots to upzone, and c) the highway creates an disincentive in the walkshed. (That's a word salad clause there, but, you know, "highway, station, walking, bad".) But Main St does have the shopping plazas and several large parking lots, so that could be an option.

The northern Woburn Loop is pretty obscured, especially north of Dexter Ave. I would probably focus on building a transfer station at the current Mishawum station, and then potentially extending further to Anderson/Woburn via Commerce Way (either the existing median or some sort of rework expanding the road's footprint).

 

[jbray]

This is more applicable to the Waltham/Burlington area where the office parks are most heavily concentrated, but what about some kind of busway, similar to the Metrobus in Istanbul? Running in the highway median, you could have some key stations at denser areas, and then move off the highway for short stops in the various office parks. Private entities could run their own shuttle services if desired, something that's really not possible with rail, and you could branch to your hearts content for lots of 1SRs.

I don't even think you need it in the median unless you're trying to run express service between the Fitchburg line and the Lowell line to Hartwell Ave/Hanscom and Burlington. Waltham just needs the 128/20 station and then you can run buses up Bear Hill Road/2nd Ave/Winter St on one side and Border Rd/5th/Prospect/3rd/Wyman St/Smith/Spring where it stops at Shire and then heads south down Hayden. You could send it to Alewife for now, but in a perfect world it would connect to the red line at Lexington Center.

An arcing rail line allows faster service over the whole are with better bus transfers than the current ones at Alewife. The bus equivalent (assuming it doesn't leave the median or equivalent row) would be stops for transfers at 128/20, Winter st/Totten Pond Rd, Trapelo Rd, Marret Rd, Bedford st, Middlesex, South Bedford (Cambridge Street also?), Winn, Main Street (Woburn), Mishawum, (depart highway) Somewhere on Commerce way, Anderson/Woburn. Not much of that actually gets you to your destination so we're talking three seat rides at a minimum. If the have to get on and off there's a lot of logistical problems of getting buses to and from a station to the highway that would hurt the rapidity of the service.

It's almost as if there's a reason no one has suggested it before me!

I basically agree with you, although it is true that we're trading difficult grades for difficult construction/disruption along the highway, with more close abutters. On the balance, it's probably still a better payoff to go via 128, but like I said I think the math potentially looks broadly similar for both.

I've thought about intermediate stops at Winn St and Main St; it's seemed dicey because a) relatively low density around, b) Winn St doesn't have un-/underdeveloped lots to upzone, and c) the highway creates an disincentive in the walkshed. (That's a word salad clause there, but, you know, "highway, station, walking, bad".) But Main St does have the shopping plazas and several large parking lots, so that could be an option.

The northern Woburn Loop is pretty obscured, especially north of Dexter Ave. I would probably focus on building a transfer station at the current Mishawum station, and then potentially extending further to Anderson/Woburn via Commerce Way (either the existing median or some sort of rework expanding the road's footprint).

Ah, but Winn St has Beacon Village and Arboretum way and is, in fact, the densest residential in Woburn. You could just use Main and send a bus down from Winn, or you could just build the two stations and have buses from further afoot on Winn got to the Winn station which is my preference.

the loop shows up well in the terrain feature but it also never went to Anderson, I'm just taking advantage of that being approximately level and formerly industrial to make the connection which I prefer to Mishawum because its a regional hub with Amtrak connections and I think the Orange line should end up there too. Proper prior planning prevents poor performance.

 

[Stlin]

This is really interesting. Is your thought that every office park along 128 would essentially become a park-n-ride? Or do you also envision this system extending beyond that further into the suburbs?

I wonder how long it would take a bus to "deviate" from 128 to serve each office park along the way....

Basically! It'd likely have really limited utility for not-office park / mall based folks, but the sheer concentration of those types of workplaces in finite pockets along the 128 Corridor, from Danvers to Quincy makes it interesting to me. I'm basically imagining the 128 Alewife shuttles, just PnR oriented. The diversions would possibly be a problem, but with some sort of "traffic skip," especially in your open BRT concept, they'd probably be able to make up that time whenever there's meaningful traffic, absent local congestion.

Generally, I think complexity is a deterrent to transit ridership, so I probably wouldn't extend it along the radial highways, at least not as a first step. Connections seem dicey to me, and I don't think you usually see too much traffic on those radial highways until I95. Although there's similar office park density on I90 / the I495 ring, a lot of office parks are off corridor. I think it would all depend on whether there'd be demand for that out further where there isn't as much traffic.