Wow. What's really interesting about that document is the routing for the LRT variant, which morphed into "surface subway." Down the middle of Don Mills, through the old Leaside Spur, along the railway through the Don Valley, across into Broadview and back, along Bayview to Queen and then a tunnel under King.. It does the same thing as the long subway version, but (+) costs less and (-) serves communities less directly. I think it would have to use more of a light rail technology for narrow ROWs, curves and grades, but in longer configurations.
I'd definitely agree that a sleeker, quieter, narrower-bodied vehicle would be an optimal choice for a surface subway instead of the hulking TR. Something made to order, like Siemen's Inspiro. Having said that, I do have some serious queries (qualms?) about this surface subway and how it'd get from Broadview Stn to King. I've looked at this section numerous times, and the only way I think a non-tunneled solution could work as they describe would involve closing the section of Bayview Ave between River and Front. A level crossing is out of the question considering the frequencies (and that this is a subway train). I think a better idea would be to have the line elevated as it enters through a portal just north of Gerrard. Another thing I don't get is why it would take longer to get from Broadview to downtown (9.4min) than from Pape (8.8min) - which would have two more stations and an added km. Nor why on p.33 it says Broadview to
Union when everywhere else it says it will stop at St Andrew.
I also have doubts about how realistic it'd be to have a subway in Don Mills' median, and along the Leaside Spur. 2-car LRVs is one thing, or something cut/covered and/or infrequent. But 7-car subways at 3.5min frequencies running on the surface is a tall order. Also, it claims that this line would have seven stations. But what's listed is only six (Don Mills, Eglinton, Broadview, Sherbourne, King, St Andrew). Would the seventh be at Thorncliffe, or Lawrence, or Dundas/Gerrard area? If stopping at Thorncliffe or Regent, it would really boost its Social Inclusion & Equity score. But aside from that, this is an interesting, peculiar proposal. Highest relief, highest benefit-cost ratio, lowest per rider cost, etc. Should lead to interesting discussions.