• Greece - 74 tunnels on new Egnatia Motorway
• UK Parliament - Tunnel Vision - Full Report
• India - Proper Use of Tunnels...

The Gateway team are deliberately comparing apples and oranges:

They claim the route through Annieville compared to to a tunnel is $188 vs $820 million respectively.

I believe the cost of the tunnel could be around $300M. My preliminary estimate is shown below. Multiply my costs by about 3 to get the government rate for a project like this that will require "due diligence" or input from a myriad of "consulting engineers" who will ask more questions than propose solutions and $300 million will be pretty close. Now the asking of questions is relevant to any large-scale project such as this one, since there are many unknowns that need to be assessed to ensure safety and operating reliability of the project well into the future. For example, one question that could be asked about the proposed Annieville two-tier highway would be: "how will the structure stand-up to possible flooding of the Fraser River to a height that would impact on the road structure?" With the recent U.K. Global Warming report issued on October 30th, the economists are predicting a catastrophic impact by 2030 on coastlines around the world due to temperature increases that cause sea levels to rise by several metres.

A similar "due diligence" question on the tunnel option would be: "how will the tunnel stand-up to a major earthquake seeing as it will be located in glacial till material rather than hard-rock?"

Many of these types of questions are important and their answers cost money to study and the solutions will cost money to implement. The impact of such studies on both options is likely to scale-up the project costs by as much as two times. So the proposed two-tiered highway proposal cost will grow to ~$380 million, and the tunnel costs might double to $600 million - as questions are posed and answered and required solutions implemented. In every case, implementation will demand a risk analysis to determine if the expense of installation is justified for either option.

So by just comparing these two routes, one certainly gets the impression that the tunnel cost will be about double that of the highway option and the Gateway numbers bring this ratio to 4:1.

However, if you extend the comparison to include the East Ladner bypass option and the Hoover-Nass options with their respective Annieville route (highway vs. tunnel), a different conclusion emerges.

                                            H&N                SFPR
                                               $                      $
East Ladner                       100-200            200-400
Tunnel                                 300-600                -
Tiered Annieville Highway                          188- 376
---------------------------------------------------------------------
Combined Costs               400-800            388-776
---------------------------------------------------------------------

The difference in these two ranges is small (i.e., ~3%).

Now, there will be an annual operating and maintenance cost associated with the tunnel that needs to be "NPVed" and added to this estimate, but the improved flow of traffic under your plan vs. the Gateway plan will have a significant incremental NPV that will reduce the added H&N/tunnel operating costs to below that of Gateway's current proposal. In addition, with the severe slope stability difficulties along the Annieville route, one can anticipate annual maintenance costs that have not been included by the Gateway proposal that will significantly exceed that of most normal roads.

So their analysis aims to elevate the costs for a tunnel to close to a billion dollars, which they then conveniently compare to the Annieville two-tiered highway costs without accounting for the cheaper cost of the H&N route vs the East Ladner bypass and ignoring the escalations that will be necessary to avoid significant geotechnical problems and high long-term maintenance requirements.
 

Tunnel Safety and other Route Safety Issues:

Good ideas on tunnel safety measures and the cost savings from urban tunnels are described in this presentation on a new Handbook on Tunnel Safety Methods. The presentation suggests that recent changes in Europe have reduced accident rates in tunnels close to that of regular open highways. We need to obtain a copy of this Handbook:
http://www.astra.admin.ch/media/norway/elvik.pdf

This presentation on tunnel safety is a good one as well:
http://www.astra.admin.ch/media/norway/zulauf.pdf

Countries such as Switzerland and Norway have tens of tunnels of the length or greater than what we are talking about here. The data show accident rates in tunnels have declined by an order of magnitude from 1995 to 2005. All presentations can be linked to from here:
http://www.astra.admin.ch/html/en/news/medien/presse.html

This one shows that the Swiss intentionally burned up a whole bunch of cars in a tunnel in order to develop procedures to reduce impact of a fire in a tunnel and to fight such fires - simply amazing!  http://www.astra.admin.ch/media/norway/weatherill.pdf

and this one shows the incredible ventilation work done in Norway on really long tunnels:
http://www.astra.admin.ch/media/norway/lotsberg.pdf

A Rough Estimate of Tunnel Costs

Now regarding the tunnel - a 4 km (or even a 5 km) tunnel that is 4-lane (two rectangles about 12-13 m in width and 6 m high), cannot possibly cost 820 million dollars unless it becomes a "cash-cow" project for every engineering consulting firm in the region - this is simply a ludicrous estimate with the clear aim of blowing the tunnel idea out of the water.

The people who I talked to at the October 21st Gateway presentation at the Delta River Inn confirmed my suggestion that the Ministry of Transportation has a serious bias against tunnels. Safety issues are the number one reason for their fear, and together with their argument on costs which have been escalated to a ridiculous level in order to make you and the Annieville folks look silly. TYour idea of a tunnel all the way down 120th Street is being used by Gateway to point to the "craziness" of your ideas. So no other options for locating the tunnel should be put forward or acknowledged by your group - the longer the tunnel, the more it becomes unattractive and the more they will dump on you for presenting "nonsense". Instead you should be grateful that their suggested location reduces the tunnel length compared to your suggested location by over a full kilometre.

Now I am not a Civil Engineer, but I do know a few things about driving tunnels underground in a mine. While a tunnel boring machine can be used to speed up the advance and schedule the work in a predictable way for civil projects, tunnels in soft-rock (such as the glacial till under Annieville) can instead be rapidly driven using the cheaper method of drilling and blasting together with mucking the material by loader and trucks. This is how we mine underground and have been doing so for many, many years. After blasting, the roof and sidewalls are rock-bolted and shotcreted to provide a safe working environment for the miners as they muck out the blasted rock. Installation of the required pre-cast concrete for structural support can be sequenced behind. To speed up construction, the tunnel can be driven towards the centre from each end simultaneously. By using these conventional mining methods, the amount of material to be excavated can be reduced by 50%. This is a major cost savings over using a tunnel borer in this situation.  Instead of achieving a circular tunnel, it will have a rectangular cross-section significantly reducing the quantity of rock to be mined.

Here is my crack at ball-parking the cost of a tunnel using drilling and blasting:

Assume a 4 km tunnel with 4 lanes - each lane = 3.5 m with a 2 m access way on each side and a pre-support installation height of 6 m, here are some numbers:

This is very rough and I have left stuff out, but I think the total estimate is likely fairly close to the possible real cost. For government work, of course, you have to multiply everything by about 3 times to get the true total cost with all the studies and consultants involved.

* Ventilation Shaft Details:
   Number = 4
   Cross-Sectional Area of each = 4 m²
   Est. Length of each = 100 m
   Concrete in each = 400 m³ (total = 1,600 m³)  
   Raise-boring = 400 m

I had a look at the tunnel design by Hatch, Mott, MacDonald in the back of Technical Memorandum 2. My design involves two tunnels each of which are rectangular as opposed to circular. I believe my design would require removing only about half the material that their design does to provide the same cross-sectional area for the 4 lanes. So, this partially explains why Gateway engineers are claiming such a higher capital cost to build the tunnel.

I am not an expert on building public traffic tunnels and so, I don't know if my design is practical and would meet all the stringent requirements of civil construction projects. There is no question, however, that it is cheaper to drill and blast the tunnel in a single pass for all 4 lanes rather than the two tunnel requirements for the tunnel boring machine. Loaders could muck the blasted rock onto a conveyor or onto trucks in much the same way that the boring machine would discharge continuously onto a conveyor.
 

From the geotechnical section of Gateway's application:

The 7.9 km long section from 500 m west of Elevator Road in Delta to 100 m east of King Road/140 Street in Surrey has undulating vertical alignment placed over existing fills which are underlain by peat and soft organic silts for most of the alignment. A test hole [TB1]82-3 at Portage Road (75m from the railway and about 300m northwest from SFPR alignment) indicates that drill casing sank under its own weight (Becker hammer blow counts=0) in sand and silty clay from 2.7m to 4.3m. Hammer blow counts increased to 10 below this level, with further increases to 20 at 12m depth where more compact sands were encountered. This test hole is 300 m away from the SFPR alignment, however it points out the fact that we must be prepared to encounter such unfavourable conditions and incorporate this possibility into the design and in the estimates. As noted in Subsection 4.7 above, highly variable stratigraphy, perpendicular to the SFPR alignment, will require attention in the geotechnical design. A photograph of a small sinkhole on the shoulder of King Road east of 138 Street (Fig 11) shows high water level beneath the pavement. Use of a geotextile barrier under the pavement structure could prevent sinkhole formation in this high groundwater area.

Clearly the geotechnical problems with the two-tiered highway through Annieville are still a mystery. Risk therefore is high and the possibility of added costs is real. 

 Review of the East Ladner Route vs the Gateway Proposal:

"Route Alternatives in SW Delta 䴋 Technical Memorandum 1"

Responses to the Conclusions and Executive Summary of Technical Memorandum 1.

1. The H & N proposal is a dedicated truck corridor for east-west movements in Southwest Delta. As total traffic on Highway 17 would only be reduced by about 10% with the H & N proposal, issues pertaining to congestion, safety and incident management on River Road and Highway 17 would not be as effectively mitigated when compared with the South Fraser Perimeter Road alignment.

A reduction of 10% of traffic is a misleading statistic since this traffic reduction is totally trucks. Each truck on highway 17 and highway 99 through the tunnel represents 3 private vehicles because of the length of the vehicles and their relative acceleration abilities. As such, reduction of truck traffic will actually impact driving delays along highway 17 by 30% - which is a considerable improvement. Furthermore by diverting all trucks to highway 91 north to the Alex Fraser Bridge, the impact on tunnel traffic by the H&N proposed route will be much more significant than the SFPR route which does nothing to  discourage trucks from taking the tunnel into Richmond and Vancouver.

Diversion of trucks to the east along 99 at a period of time when traffic flows in that direction are low and then, north along an improved highway 91 will reduce the cost of upgrading the route north of 99 and will still provide quality access to the Tilbury/Sunbury industrial park from the north as well as connecting to whatever route is planned along the river north of the Alex Fraser Bridge. The number of trucks entering the Tilbury/Sunbury industrial park is a relatively low percentage of total truck traffic 䴋 sending these trucks south into the park will be more benign than sending all trucks north along the SFPR option.

2. Analysis indicates that under the H & N proposal travel time for trips between BC Ferries/Delta Port and Tilbury/Sunbury for commercial vehicle operators would increase, due to the circuitous nature, by more than one and half times. This is inconsistent with a key SFPR objective to reduce travel times.

Since the total truck traffic going to the Tilbury/Sunbury industrial park is a relatively low percentage, this is a rather specious argument. The majority of truck traffic from Delta Port goes east to the US and into Surrey. The BC Ferry traffic does favour the tunnel into Richmond, the Vancouver Airport, and Vancouver itself, but these vehicles can still follow that route along highway 17 and north onto 99 into the tunnel. So there is no impact on travel times for BC Ferry traffic going to any destination 䴋 if the truck traffic from the Ferries can access the H&N bypass, this in fact , would provide a shorter route for east-bound truck traffic. The small increase in travel time for trucks going to Tilbury/Sunbury is more than made up for by the reduced travel times for vehicles going elsewhere which are in the majority. Furthermore since truck traffic is on a dedicated route that few private vehicles will take, the ability to drive faster along the H&N route will limit the increase in travel time to about a 25% increase, not 50%. By separating trucks from private vehicles a much safer traffic system results 䴋 this is often done in the mining industry where large truck fleets may have two different size trucks that travel at different speeds. By separating these two types into dedicate service routes, overall congestion and waiting times in queues is significantly decreased. This can be proven and quantified using Queuing Theory simulation tools.
 

3. The proposal does not improve access to key commercial and industrial areas along the Fraser River; such as Tilbury/Sunbury.

This is quite simply untrue 䴋 access from the north using H&N will be improved to the same overall degree as the southerly entrance using the SFPR option.

4. The proposal increases traffic demand on the Alex Fraser Bridge, Hwy 91 and other arterials in west Surrey and Richmond. The traffic data also indicates that the H & N proposal moves truck traffic away from their intended destination along routes that will exacerbate congestion in other areas (Hwy. 91 & Hwy. 99 Interchange and Highway 91 & Nordel Way Interchange).

This is true. However congestion at these interchanges, if they are properly upgraded, will not increase anywhere near to what will take place at the tunnel when daily truck traffic increases from 2,000 today to 5,000 over the next few years. The Alex Fraser Bridge has 3 lanes compared to 2 (and the counterflow) lanes through the tunnel 䴋 the bottleneck at the tunnel is a much greater problem than that at Alex Fraser. The Nordel Way & Highway 91 interchange definitely requires upgrading, so traffic can flow onto the bridge without being controlled by traffic lights 䴋 this must be done regardless of whether the SFPR or the H&N option is chosen.

Truck traffic on H&N that does wish to access the tunnel may still do so at the 99 and highway 10 interchange and in fact, this will give access for such traffic to the counterflow lane at rush hour which is not as congested as the regular two lanes. Thus, both congestion relief and access are achieved.

Review of the East Ladner Route vs the Gateway Proposal:

"Route Alternatives in SW Delta & Technical Memorandum 1"

Responses to the Conclusions and Executive Summary of Technical Memorandum 1.

1. The H & N proposal is a dedicated truck corridor for east-west movements in Southwest Delta. As total traffic on Highway 17 would only be reduced by about 10% with the H & N proposal, issues pertaining to congestion, safety and incident management on River Road and Highway 17 would not be as effectively mitigated when compared with the South Fraser Perimeter Road alignment.

A reduction of 10% of traffic is a misleading statistic since this traffic reduction is totally trucks. Each truck on highway 17 and highway 99 through the tunnel represents 3 private vehicles because of the length of the vehicles and their relative acceleration abilities. As such, reduction of truck traffic will actually impact driving delays along highway 17 by 30% - which is a considerable improvement. Furthermore by diverting all trucks to highway 91 north to the Alex Fraser Bridge, the impact on tunnel traffic by the H&N proposed route will be much more significant than the SFPR route which does nothing to  discourage trucks from taking the tunnel into Richmond and Vancouver.

Diversion of trucks to the east along 99 at a period of time when traffic flows in that direction are low and then, north along an improved highway 91 will reduce the cost of upgrading the route north of 99 and will still provide quality access to the Tilbury/Sunbury industrial park from the north as well as connecting to whatever route is planned along the river north of the Alex Fraser Bridge. The number of trucks entering the Tilbury/Sunbury industrial park is a relatively low percentage of total truck traffic 䴋 sending these trucks south into the park will be more benign than sending all trucks north along the SFPR option.

2. Analysis indicates that under the H & N proposal travel time for trips between BC Ferries/Delta Port and Tilbury/Sunbury for commercial vehicle operators would increase, due to the circuitous nature, by more than one and half times. This is inconsistent with a key SFPR objective to reduce travel times.

Since the total truck traffic going to the Tilbury/Sunbury industrial park is a relatively low percentage, this is a rather specious argument. The majority of truck traffic from Delta Port goes east to the US and into Surrey. The BC Ferry traffic does favour the tunnel into Richmond, the Vancouver Airport, and Vancouver itself, but these vehicles can still follow that route along highway 17 and north onto 99 into the tunnel. So there is no impact on travel times for BC Ferry traffic going to any destination 䴋 if the truck traffic from the Ferries can access the H&N bypass, this in fact , would provide a shorter route for east-bound truck traffic. The small increase in travel time for trucks going to Tilbury/Sunbury is more than made up for by the reduced travel times for vehicles going elsewhere which are in the majority. Furthermore since truck traffic is on a dedicated route that few private vehicles will take, the ability to drive faster along the H&N route will limit the increase in travel time to about a 25% increase, not 50%. By separating trucks from private vehicles a much safer traffic system results 䴋 this is often done in the mining industry where large truck fleets may have two different size trucks that travel at different speeds. By separating these two types into dedicate service routes, overall congestion and waiting times in queues is significantly decreased. This can be proven and quantified using Queuing Theory simulation tools.
 

3. The proposal does not improve access to key commercial and industrial areas along the Fraser River; such as Tilbury/Sunbury.

This is quite simply untrue: access from the north using H&N will be improved to the same overall degree as the southerly entrance using the SFPR option.

4. The proposal increases traffic demand on the Alex Fraser Bridge, Hwy 91 and other arterials in west Surrey and Richmond. The traffic data also indicates that the H & N proposal moves truck traffic away from their intended destination along routes that will exacerbate congestion in other areas (Hwy. 91 & Hwy. 99 Interchange and Highway 91 & Nordel Way Interchange).

This is true. However congestion at these interchanges, if they are properly upgraded, will not increase anywhere near to what will take place at the tunnel when daily truck traffic increases from 2,000 today to 5,000 over the next few years. The Alex Fraser Bridge has 3 lanes compared to 2 (and the counterflow) lanes through the tunnel, the bottleneck at the tunnel is a much greater problem than that at Alex Fraser. The Nordel Way & Highway 91 interchange definitely requires upgrading, so traffic can flow onto the bridge without being controlled by traffic lights, this must be done regardless of whether the SFPR or the H&N option is chosen.

Truck traffic on H&N that does wish to access the tunnel may still do so at the 99 and highway 10 interchange and in fact, this will give access for such traffic to the counterflow lane at rush hour which is not as congested as the regular two lanes. Thus, both congestion relief and access are achieved.