I was a bit skeptical until I saw this user map of the Northeast Corridor:
https://www.google.com/maps/d/u/0/viewer?mid=1dH5vJWx6lwBlBv...
I was a bit skeptical until I saw this user map of the Northeast Corridor:
https://www.google.com/maps/d/u/0/viewer?mid=1dH5vJWx6lwBlBv...
I don't really know this area. Say you wanted to build an entirely new and mostly straight Stamford - Waterbury - Hartford - Providence alignment. Some of that would need to be in tunnel. But how much of that could you do on the surface without needing to demolish too many houses?
The route that the existing track takes is insanely suboptimal and seems so hard to fix considering the existing urban sprawl in the relevant areas.
High speed rail goes well beyond that speed and needs much larger curves, at which point you're deviating enough from main roads you may as well just take a different path entirely.
The above isn't a perfect answer to all situations, but it helps a lot.
The rail solution to that would probably be a tunnel. A tunnel is a cheaper affair with just two tracks than if it has to be wide enough for two tracks and six to eight highway lanes.
In any case when you start building HSR it becomes progressively harder to eke out savings and the highest come from urban area speedups. Going from 50 to 100MPH cuts a lot more time than going from 150MPH to 200MPH. And the former simply doesn't work in the median of a highway.
But then the bizarre/historic track geometry ruins the fast service...
Rerouting so far has been a headache; in New Haven-Boston where the tracks are so curvy new tracks have to be built, the FRA proposed going through entirely different population centers (the geography of the existing coast line is very challenging). This has resulted in an interesting fight where people want the train in their town, but also do not want the impact of newly constructed tracks through their town.
They really can't.