I've been talking to people lately about the upcoming University Ave. LRT, and whether or not it's going to have tunnels (or bridges) at two key intersections: Washington Avenue at the U of MN, and at Snelling Avenue in Saint Paul.
The problem doesn't have an easy answer, but it points toward something I call the Transit Dilemma. Good transit is most expensive where it's most needed. Those places that are in dire dire need of a transit option are those places where the best option is an underground tunnel. (The short list is Lyndale/Hennepin, and the U of MN.) And of course that's god-damned expensive.
It's because of density, which is the most basic prerequisite for effective transit. In places like the Stadium Village stretch of Washington Avenue, you've got by far the most heavily clogged, pedestrian-laden place in the Twin Cities. This is a place where cars stream in at 45 mph from downtown only to be met with one lane of stoplight-encrusted road with busses galore, parking, and the most people per square foot for 500 miles in any direction. It's a real buzz kill. (Here's the strib story on the topic.)
To even think about putting LRT at street grade here would make the U of MN congestion ten times worse, and that's why this whole business about "Tunnel or No Tunnel?" is mere window dressing. The powers that be at the U of MN would never condone it.
Less so but still a problem is the Snelling Avenue interchange. This is the #2 traffic intersection in Saint Paul, and is only going to become more heavily trafficked as the Bus Barn site gets built up and the SuperTarget (and SuperWalMart?) go in at this site. Putting a LRT down the middle might tie this area in a Gordian knot, and maybe an unsightly Lake Street-style overpass (or a tunnel) would be a good idea.
But I don't think it'll happen here. Transit planners are probably thinking that University Avenue auto traffic will decrease somewhat with the LRT going in. (Though the overall traffic will be more than compensated by the train numbers.) And they're really trying to keep costs down. But, that being said, this will be the probable flashpoint of any anti-LRT aggression. Particularly since the damned Midway Books guy is going to be out protesting every time a train goes by.
These debates reveal why TOD is such a cost-effective way of building a transit system, but only in the long term. (In the short term it doesn't make any sense because you're building a rail corridor where there isn't any demand.) But in the long term, as infrastructure follows the capitol investment, you've got a great transit surrounded by density. And you've done it much cheaper than if you'd built the transitway after the density.
(Not to mention the whole chicken-and-egg conundrum, about whether density follows transit or vice versa . . .)