As a bit of a follow up to my previous reply on the difficulties on using a TBM during rock-soil interface conditions, I thought I would examine in a bit more detail why going above ground for certain portions of the line might be a logical choice. In the last comment, I was focused on the northern interface, and here I'd like to focus on the south. Keep in mind this might not be why ML chose to do this, it is just an examination of some of the risks that could be associated with the underground alignment, particularly in the Lower Don crossing area. All of this is just my opinion and is a
hypothetical risk assessment based on some readings I've done.
If we take a look at the current Don River, we can see it is channelized. It is also worth noting the concentration of highrise buildings on the left, the very high concentration of rail tracks in the area, and the proximity of the DVP to the the river. Also very much worth noting the old eastern ave bridge "[...] is used by Enbridge Gas to carry a major gas main." (Source:
Wikipedia)
I'm pointing these out because these present some very high levels of
risk to any underground project that would take place near them.
(Source: Google Maps)
And of course, it should be noted that the Don River was not always channelized. The historic alignment of the Don River may have looked something more similar to this:
(Source: Bateman C., blogTO,
That time the Don River was straightened)
I'm sure you can see the large sinuous curves. These sorts of curves would carve out bedrock over millions of years and several glacial periods, creating a subsurface profile that could conceivably look similar to the one shown below:
(Source: Andres A. et al.,
Impacts of Quaternary History on Critical Zone Structure and Processes: Examples and a Conceptual Model From the Intensively Managed Landscapes Critical Zone Observatory)
Note: This is
not the Don River.
However, it is a river that formed through similar soils (glacial till) during similar glacial-interglacial periods. Thus we can extrapolate that the area of the Don we're looking at would have a similar profile. You can see in the river valleys that the soil is not flat or horizontal, but are quite undulous. This presents similar problems of volume loss, but there are more severe issues than slight surface settlement.
In this case, the transition zone could see a switch between soil and rock as bedrock maps show it being relatively shallow here. If this occurs back and forth, you'd essentially have a very rough transition zone across the
entire stretch of river. And remember, the "river" is not just the 40m or so channelized one that exists today, but includes all the historic alignments and curves which leave an impression in the bedrock. These possible difficult geological conditions would exist under water pressure. We might expect to see similar issues to other TBM projects under rivers:
To balance the external soil and groundwater pressures, the TBM had to be operated at high chamber pressures. TBM excavation chamber pressures up to 6bar occurred during tunnel excavation where the 55cm full hydrostatic pressure was realized
The TBM saw considerably lower advance rates in the more challenging geologic conditions.
(Source: Fekete S. et al.,
Tunnelling under the Fraser River at 6 bar)
The Fraser River project included a specially design TBM and, while a much longer tunnel under a much wider river, shows some of the issues that can occur. Would you want to spend the extra money to specially design a TBM for what could amount to a 40m problem zone? What if it ends up being 100m? 200m? Are you going to spend the extra money drilling a borehole every 3m across the entire river, on the active DVP, and through the brand new Corktown Commons to ensure there's no risk?
Another example:
Most of the tunnelling for the North East Line in Singapore was carried out using earth pressure balance (EPB) shields. Settlements were generally well controlled; however, there were occasional large settlements, exceeding the design estimates, sinkholes or losses of ground.
(Source: Shirlaw N. et al.,
Local settlements and sinkholes due to EPB tunnelling)
I think we can all agree that having sinkholes in a public park, beneath rail lines, under an active highway, or affecting a major gas main on a bridge are all very, very bad outcomes. (Additional readings on risk in urban areas: Kovari K. and Ramoni M.,
Urban Tunneling in soft ground using TBM's)
It also doesn't seem feasible to close any or
all of these while the TBM's would be progressing beneath these areas. We can expect that the advance rate of a TBM would be around 1-4m/hr in the best conditions (Source: Mohammadzamani D. et al.,
Evaluation of required thrust force based on advance rates in shielded TBMs under squeezing conditions) which would be considered very generous, it might take as long as 13 days to pass beneath an area from Broadview Ave to Bayview Ave if this is where the historic river channel lies. Imagine suggesting closing the DVP-Gardiner interchange
both ways for 13+ days! Along with the GO lines, gas mains, nearby roads, and parks.
I don't want to say the choice to go above ground is politically motivated or not, taking a stab at those who didn't 'Vote Doug', but there are other things to consider in an engineering project. This isn't Sim City or Cities Skylines, where you can slap a tunnel under anything with no issue. These are incredibly complex issues with high amounts of risk associated with them, especially when in such a dense urban environment. It's entirely possible that the levels of risk associated with tunneling under the Don, which we haven't done very often, is considered too great this time. Especially for such a high profile project, in comparison to something like the Coxwell Bypass, which runs parallel to the river and is 50m deep (almost certainly in shale, judging by the large grey piles visible from the Bloor St bridge). It's also possible that Doug is just trying to F-over everyone who didn't vote for him.