Where Would a Don Mills Subway Go?

There has been a lot of discussion here about potential alignments for the eastern leg of a downtown relief line.  On occasion I have mentioned a route rather different from the commonly discussed one via Pape, the Leaside Bridge and Overlea, and I am sure this has caused some confusion.

One advantage of having been at this transit advocacy business for a long time is that I have a long memory and archives to match.  For your delectation, here is a proposed route from Don Mills and Eglinton to downtown.  It is a TTC Subway Construction Department drawing dated December 12, 1973.


A few things worth noting about this drawing:

The route north from Danforth is via Donlands, not Pape.  This provides access to Greenwood yard via the connection shown.  It also aligns the route further east to simplify the valley crossing north of O’Connor.

The route passes through the middle of Thorncliffe Park and proceeds north to Eglinton.  This is more or less the sort of alignment I have been talking about for the east leg of a DRL (or, for that matter, for the Don Mills LRT if it came south of Eglinton).

Two alternative alignments from the CNR line to Queen are shown.  One goes straight south while the other runs along the rail corridor.  Going west along Queen brings its own problems, and these were discussed in an earlier, 1968 report that I will present in a separate post.  (Please don’t clutter up the comments thread here with questions about that part of the alignment.  You will get your chance.)

I present this information mainly so that people can see that the idea of a subway to Eglinton and Don Mills is hardly new, and it’s not even mine — I simply resurrected an old TTC concept.  When we discuss transit plans, it is useful to know some of the history.

20 thoughts on “Where Would a Don Mills Subway Go?

  1. Hmm,

    I always had the Pape terminus in my head, as my first encounter with the idea of a DRL was the Network 2011 plan, which as I recall was introduced under Premier Peterson in the late ’80s.

    However, if we are to retain the essence of the DRL idea, but reopen the routing…

    I’m not a fan of the Donlands option. What works in favour of the Pape option is that it is a major bus terminal (for Don Mills) and that that it is in the heart of Greektown a busy tourism and retail area.

    Donlands by contrast is in a quiet (some might say near dead) area of the Danforth, with no significant bus traffic), and no major trip generators.

    IF..(Big if) we were to rethink where a line might come down from Don Mills……

    Here’s my thought on a radical idea. Don Mills, on a map, aligns almost flawlessly with Coxwell Avenue.

    Take Don Mills Road out of the valley. (the only purpose it serves there is the DVP interchange, which while used, is not all that busy for a highway interchange. (the interchange removal has been previously considered by the City)

    Route Don Mills Road directly south on a Bridge, across the valley to Coxwell Avenue.

    Run the subway underneath the new bridge on a deck.

    This provides a direct route for the subway to Coxwell Station, which would allow it to serve 2 major trip generators, East York Collegiate (one of the largest High Schools in the City); and East General Hospital.

    Further, under this new alignment, Don Mills buses (in reduced frequency, post-subway) could go directly south via the new bridge to Coxwell station, a much faster route.

    The only knocks on this idea, that I’m aware of are the extraordinary cost of a new bridge, and that this might result in under servicing the Thorncliffe Community.

    However, in the case of the bridge, the cost is only so large, assuming that we are comparing it with a crossing utilizing the existing Millwood Bridge. If we require a new bridge for a subway crossing anyway, the added cost of decking said bridge for cars is marginal, and partially off-set by the reduced future expense from removing the valley portion of Don Mills Road.

    As to servicing Thorncliffe. Everyone there would be withing walking distance of a subway station at Don Mills and Overlea in this scheme. But, further, we could reroute the Thorncliffe bus to the new subway route and the travel time to the new station would be negligible.


    All that said. I’d be quite happy if they just built a DRL to Pape!

    But if we get to rewrite the maps, I think the Coxwell corridor is a viable and interesting choice.

    Steve: There are problems with every corridor be it Pape, Donlands or Coxwell. In a future post, I will show a detailed map of the alignment south from Donlands Station. It goes through a residential neighbourhood and under Eastern High School of Commerce. Tricky, that.


  2. Don Mills and Eglinton? Interesting.

    Would it try to capture Don Valley traffic? There’s nothing there other than the Ontario Science Centre and future Aga Khan Ismaili Centre and Museum otherwise.

    I would think that a better terminus for capturing bus / LRT traffic going downtown would include the Leslie buses. (I suppose they could be turned east instead of west as currently done, though.)

    Steve: It would intercept the Eglinton LRT and be fed by the Don Mills LRT whose ridership would be higher than the existing bus simply because it went to a subway line at Eglinton. The main reason for taking the subway up to Eglinton is that it has to be undrground south of the valley anyhow, and it makes more sense to have the transfer point at a major junction. That intersection is also ripe for development.


  3. I think that a DRL line should rather follow the hydro corridor east past Overlea so as to provide one or two stops within the Flemingdon Park neighbourhood, and then up the valley (possibly beside the train tracks, I don’t know if this is feasible) past Eglinton to Wynford Heights and Concorde Gate, another high-density node with some commercial and industrial jobs already there. The Don Mills LRT could terminate at Don Mills and Overlea, connecting with the subway still, or join the alignment and head downtown.


  4. Why not have the DRL using an underground LRT connecting the Don Mills LRT and Jane LRT. It would help utilize the infrastructure that will be built first (hopefully) and people from the outer edges of Toronto all the way up to Steeles (and past?) would have access to a semi-rapid route to downtown.


  5. Shouldn’t the value of the corridor be more important than yard proximity?

    Steve: Yes, but you have to store the trains somewhere. In a future post, I will include a detailed map of the area between the Don River, East York and Greenwood to show the difficulty of threading any line through these neighbourhoods. The days of wholesale demolition of buildings in the name of transit progress is long past.


  6. Well it’d go down Don Mills, to Pape (or donlands), then down to the Rail line and then… Thats where it gets tricky. It could follow the rail line, or, go down to Queen. Then the two meet once more, where again, it could follow the rail line, or follow queen.

    I still favour an LRT solution to the problem. My favoured path is to take the line down Pape (or donlands for that matter) to the rail line, and then follow the tracks (above or below ground – whatever works) to Queen and the rail line. Since one LRT trackset will not be able to handle the demand, we will build two, and express, and a local. From here the express line would head to Union, and the local line would head downtown, under Queen and/or King.

    If, however, we are going with a full-fledged DRL subway, we should make the route as short as possible, and also put it under Queen or King and not head it to Union. The reality is that King and Bay is close enough to Union station that people can walk.


  7. That Coxwell bridge proposal would be shot down in a heartbeat.


    Costs are way too high! The bridge itself would cost hundreds of millions of dollars.

    The EA would state extraordinary and unnecessary harm to the environment would occur due to the nature of building extremely high pillars to support the foundation of the bridge within the valley. It could cause major disruption to wildlife along the Taylor Creek Park in that area.

    In theory this bridge proposal is very good for planning & linking important roads in the network but in practice this should have been done 50 years ago, not now. I think the only reason that Don Mills runs between Overlea and O’Connor is just to connect it to the DVP and unload traffic there rather than send everyone from Thorncliffe & Flemingdon Park northeast to Eglinton just to access the highway.

    This Don Mills section is also used by the 25 quite extensively (and bypasses Thorncliffe so running time is equal for the route since it doesn’t have to make stops in Thorncliffe or wait to turn left at Overlea as that is cancelled with the left turn at O’Connor & Don Mills) as a detour IF there is a blockage at the end of the Millwood bridge and the buses can’t access Overlea. It’s also used by the 81 which reroutes into Thorncliffe via that method, and the 100 which just runs down the same way as the 25 bypassing Thorncliffe, although I am not quite sure what happens with the 56… the 56 might have to run up Don Mills, Overlea, Beth Neilson and Wicksteed in this case.

    That last bit was FYI to all readers!


  8. Steve,

    I must have mentioned to you a few times … when you look at the LRT/Transit City maps for WWLRT… on Queen’s Quay, East of Bay where the 509/510/WWLRT would go north to Union … there is a dashed white/grey which either from you or someone else told me it would be a seperate transit initiative … then there was that meeting for the redeveloping of the Cherry Street area (bottom of the Don River) which maps show a 2 or 3 transit routes….

    Steve: The “separate” initiative you may be talking about near Union Station is the Bremner line which comes in south of Skydome, swings northeast at Lower Simcoe, and connects into the existing tunnel at Bay.

    The plans shown for the Don Lands include alternative routes not all of which would actually be built. I will blog on this issue sometime in the new year.

    So … what would happen if you “combine” the DRL with the Don Mills LRT and make an EWLRT (East Waterfront LRT):

    Bring the DMLRT south past Danforth via Pape then Carlaw (Pape breaks up for a part due to tracks and Riverdale Shopping Centre, it also only goes to Eastern).

    Carlaw goes to Commissioners. the DMLRT continues south to a loop on let’s say commissioner and carlaw where it could meet the EWLRT, maybe even making the WWLRT and EWLRT one line.

    The DRL would leave Pape station to Queen then turn west and go it’s merry way.

    Steve: The line will have to be underground south from Danforth as the streets involved are only four lanes wide. The projected peak demand on the DRL south of Danforth is over 10K/hour and this could not be provided by surface operation. I am not sure that we need to bring any more demand into Union Station from the east, and we need the capacity of the eastern waterfront LRT for demand in that area, not for through trips from Don Mills.


  9. James,

    The problem with your proposed alignment is that Coxwell Avenue becomes Coxwell Crescent north of O’Connor. It’s a narrow side street with single famly homes (some quite expensive) that extend up to the valley, where there’s a small park. While building subways under side streets is nothing new (Strathmore, just 2km south, comes to mind), I suspect the residents in that pocket would put of a voracious fight, citing concerns with noise, vibrations, and other negative effects to their properties. Having said that, I think the alignment is a good idea, and much better than Donlands. But realistically, the Pape corridor has the highest density in the area, and should be the first choice. Coxwell would make a reasonable second option, but I doubt that would ever happen.

    Steve: Just to clear things up: I am not wedded to either a Pape or Donlands alignment, but wanted to show what sort of things had been discussed in the past. If we wind up needed a new bridge across the Don to reach Thorncliffe Park, then we need to be open to a route that does not slavishly follow the existing street pattern.


  10. The idea of a Don Mills/Coxwell bridge is not too far-fetched and could be made affordable through the use of, dare I say it, road tolls. I am against tolling existing roads because that only shifts congestion to roads where it messes up transit, but to build a new connection (a “missing link” I believe is the term used by some) that is tolled is a whole other issue. Effectively, we would get a subway bridge that could be paid for by the road on top of it.

    That said, the points brought up by Leo Gonzalez about residents in the area are by far the impediment to this rather than cost. Though, one block west of Coxwell, at Lankin Blvd, there is nothing north of O’Connor, though the jog would be odd – unless we took out the gas station at Coxwell and O’Connor.

    One possible solution would be to have a bridge that crosses the valley at a lower level (perhaps only half way up – low enough could make a connecting station with the Richmond Hill GO line practical) that effectively enters a portal to a deep tunnel on the south side, gradually rising as it moves south to Danforth. Such a deep tunnel, hopefully, would be less of a noise problem and certainly less of a construction inconvenience to local residents. This depth would also likely make it easier to align the route with whatever road is decided upon from Coxwell to Pape. On the north side of the valley, there are more options over entering a portal in a steep side of the valley, so deep tunneling may or may not be necessary there.

    Steve: Please let’s not confuse transit projects with road building. The last thing we need is to turn Coxwell Avenue into a continuation of Don Mills Road.


  11. In the back of my mind I always had the DRL going along the CN to the west side of Greenwood and now I know where that idea came from. I like the alignment along the CN. If a DRL is to bleed passengers away from Yonge and Bloor Danforth then it should provide a faster trip downtown.

    The problem with the existing subway is that it is trying to do two jobs, be a rapid transit line in from the suburbs and also be a HRT line with semi local service. It can’t do both jobs well. I believe that this line should have stops 2 – 3 km apart except where they meet a major transit line. All stations should be major interchanges and none should be built with only walk in riders.

    I am conflicted about the alignment west of Queen and De Grassi. Part of me wants it under Queen to help with downtown distribution and part of me wants it to stay on the CN to cut costs and speed up the line. I fear that running it to Union would make the future problems at Bloor Yonge seem mild. I shall await your missive on the west end.


  12. It should be mentioned that a DRL would require turning Queen and Osgoode (or King and St Andrew or mega-Union) into major interchanges. It would require new stairs, new platforms, perhaps widened platforms, perhaps relocated platforms. This is not necessarily cheaper than the beefing up of Bloor-Yonge.


  13. How would the presence of the DRL affect plans in the works for the new waterfront LRT lines?
    What about the plan for a Kingston Road LRT that would go along Danforth to Victoria Park Station? If there is a DRL would you ignore that option and have it continue along Kingston Road and Queen to meet up with the DRL?

    Steve: I have never thought that the scheme of having a continuous line to Union Station from the far reaches of Kingston Road via the eastern waterfront made any sense at all. It’s worse than the Western Waterfront scheme on several counts, not the least that we need the capacity of the eastern Queen’s Quay line to get local traffic into Union.

    If a DRL does come down from the east, it will probably not go far enough south to pick up much of the waterfront traffic. The new residential areas are south of the rail and expressway corridors, and in the case of the proposed housing in the Port Lands, a long way to the south.

    A lot more can be done to improve bus service in southern Scarborough before it needs a streetcar.


  14. I think that any routing which bypasses Thorncliffe Park would be a mistake given the existing population (30,000 according to an unsourced Wikipedia article) and current redevelopment potential along Overlea Blvd.

    It’s also unreasonable to expect people to walk to Don Mills & Overlea, as it’s 800 metres alone to get there from Overlea & Beth Nealson/Thorncliffe Park, and most of the apartments are some distance away from that point. You would need to maintain pretty heavy bus service to that area if it were bypassed by any new line.

    Another key point to keep in mind is the relatively high (for East York) population density at Pape & Cosburn, which happens to be just about the right distance away from Danforth (1.2km).

    Finally, Gerrard Square & Riverdale shopping centre are 1.3km from the Danforth, and Queen is about 1.2km from Gerrard & Pape. Just from looking at the major trip generators and intersecting routes it looks like a no-brainer – of course I realize it’s not that simple!


  15. The sewer under Coxwell crosses under the Don River meaning it is really really deep under the street. I don’t think the subway needs to go that deep.


  16. Regarding the radical suggestion by James to run the subway up Coxwell: “The only knocks on this idea, that I’m aware of are the extraordinary cost of a new bridge, and that this might result in under servicing the Thorncliffe Community.”

    In regards to the latter – one solution is simply to build the Don Mills LRT along Don Mills, Overlea, Leaside and Pape – in addition to the new subway line. This would allow the new subway line to simply have stops at Eglinton, Overlea, somewhere north of Danforth and Danforth. The LRT could be extended south from Pape Station to Pape/Gerrard (or Pape/Queen) and intersect the new subway there. The LRT would then provide service for Thorncliffe, either towards Don Mills Road, or towards Danforth.

    As far as the bridge – I’m sure that a new road bridge couldn’t be sold. But a subway may be different. If the subway was to get quite deep at O’Connor (which shouldn’t be an issue if the only station is at Cosburn or Mortimer), then the bridge structure may be only in the valley, and wouldn’t raise the ire of residents. An potentially could be underground the entire way – though I’d think it would be a challenge to climb enough to not make an Overlea station very deep.

    A Don Mills subway – to Eglinton at least, doesn’t have to preclude the LRT coming south of Eglinton.


  17. To Nicholas Fitzpatrick:

    Assuming that the Pape section of LRT must run underground, it would be a challenge to get funding for both the subway tunnel under Coxwell and the Pape LRT tunnel.


  18. “Assuming that the Pape section of LRT must run underground, it would be a challenge to get funding for both the subway tunnel under Coxwell and the Pape LRT tunnel.”

    Very true – though they do seem to be trying to move away from putting the Jane and Don Mills LRTs underground at all (perhaps the wider 23-27 metre ROW on Donlands comes into play here).

    I’ve always figured this was the way to go, but I figured the LRT on Pape would come 10-15 years before there was any serious DRL discussion – and if that was built bare-bones just for streetcars, then a second tunnel would be a no-brainer. Now that there are some signs DRL may come sooner, then harder to justify. But operationally I think that it would work superbly, with enough local traffic to justify both – except perhaps between Overlea and Eglinton.


  19. The spacing between stops will be further apart if the DRL is implemented. So my question is weather the TTC will keep the Queen Streetcar in operation once the subway line is completed. This will help adress local needs.

    Steve: Considering that the DRL probably will not run along Queen Street, that’s a reasonble assumption. Looking at where the demands are, only the east leg of the DRL is likely to be built within the next decade at best, and its catchment area will be quite different from that of the 501/2/3. In the west end, the DRL may serve Liberty Village, but it won’t replace the Queen car.


Comments are closed.