Links across the waters
Strait Crossings conference report
Links across the waters Jan 2010
Shani Wallis, TunnelTalk
A tunnel is a tunnel, is a tunnel - right? Well not so fast. A recent symposium in Norway explained why tunnels for strait crossings are special. Deep mined rock tunnels; TBM bored tunnels; immersed tubes; and submerged floating tunnels, all have one thing in common: they are under or through the sea, but each is specific in its own right and, as a collection, different in so many way to tunnels on land. Bridge alternatives and ferries for strait crossings were also considered at the Strait Crossings symposium; one of the rare occasions when bridges, tunnels and ferries are discussed at the same forum. TunnelTalk attended and reports on the proceedings.
Norway is a natural for the discussion of providing links across water divides. With thousands of miles of coastline and communities on different islands and opposite sides of long fjords, Norway is the world's leader in building and operating strait crossings. As well as hundreds of ferries in service, and many bridges of different types and size, Norway has 26 undersea road tunnels with another three in construction and several more in planning and design.
As one of the most interesting, topic-specific gatherings of the year, professionals and engineers gathered in Trondheim to discuss all facets of building, owning, and operating strait crossing bridges, ferries and tunnels. With its specific interest, TunnelTalk concentrated on sessions that discussed tunnels and the post conference tour that included trips through deep mined rock tunnels, over bridges and on new, more environmentally friendly, natural gas-powered ferries for a most memorable journey.

The symposium was the fifth in the series since the first was convened in Norway in 1986. All subsequent events have also been in Norway, the last in Bergen in 2001. Strait crossings is perhaps the fastest moving sector of civil engineering currently, with a wealth of information about technical and project developments from around the world. The concerns and on-going advances of strait crossings as discussed at the symposium illustrated that the next event should not wait six to eight years to be convened, wherever that maybe. In the meantime the organisers and individuals of the 5th in the series at Trondheim last year are to be commended for an excellent and timely convention. Those leading the organisation included SINTEF(1), the Norwegian research organisation based in Trondheim; the NTNU (Norwegian University of Science and Technology(2) Trondheim); Statens vegvensen, the Norwegian Public Roads Administration(3); and NFF, the Norwegian Tunnelling Society(4). The bulk of the worry and the work was carried by the symposium Chairman, Eivind Grøv, and three principal co-ordinators Håvard Østlid, Kaare Flaate and Kåre Senneset.
Submerged floating tunnels
The concept of the submerged floating tunnel (SFT) has been long appreciated but the idea that it is still a far off, futuristic novelty is wrong. Two full sessions and a three-hour workshop at the symposium were devoted to the concentrated work progressing in different countries and by different teams around the world to bring this concept to reality.
Despite a British patent dating back more than 100 years, it is Norwegian engineers who have invested more in developing the technology in recent times. For Norway, the concept provides the best option for providing fixed links across the country's deep fjords, at locations where the divide is either too deep for deep mined rock tunnels or too wide or exposed for long high bridges or floating bridges.

Comparisons of lengths and illustration of an SFT

Set at about 25-30m below the surface, connected to short landfall tunnels at either side, and tethered in some way to counteract buoyancy and the movement of tides, currents and waves, an SFT would:
- be much shorter than a deep mined rock tunnel or a high level suspension bridge;
- be less expensive to build than the deep mined rock tunnel or high level bridge option;
- consume less materials than the alternatives;
- take less time to construct.
Norway came closest to building the world's first floating tube in the 1980s when a two-lane SFT road connection was adopted for a crossing of the Høgsfjord toward the city of Stavanger. Conceptual and preliminary design was complete; planning permission and construction permits had been granted; the financing structure was being finalised; four concepts for design and construction had been prequalified; and the procurement of a construction contract for the selected option was advancing. It was at the eleventh hour that the local county council decided an undersea mined tunnel at a different location would be the fixed link, and not the Høgsfjord SFT as promoted by the country's Public Roads Administration.
Although the opportunity at Høgsfjord was missed so narrowly, it is often forgotten that it was a political decision that defeated the possibility and not the technical feasibility. The four prequalified SFT concepts had drawn together the expertise of tunnellers, bridge builders, materials engineers, university research resources, and the engineering skill of experts advancing the cutting edge design of fixed and floating deep sea platforms needed at the time for Norway's oil and natural gas reserves in the North Sea and the North Atlantic Ocean.
Since then, promoters of the concept in Norway have kept the interest alive in the imagination of the public and against the increasing fire-live demands of operating underground and tunnelled infrastructure. Norway is an active member nation of the Working Group of the International Tunnelling Association (ITA)(5) on immersed and floating tunnels and several Norwegian engineers, most notably Håvard Østlid, have dedicated a professional life-time to promoting the concept world-wide.

Illustration of tethering systems

The technical challenges of building an SFT are not under estimated. The most paramount of these as discussed at the symposium sessions and workshop are listed here in no particular order:
1. The need to prevent resonance vibrations and movements being induced into the hollow tube as well as into the anchoring tethers under different hydraulic and dynamic load conditions
2. The practicalities of fabricating and placing of the tube elements, installing the anchor systems, and realising the landfalls either end.
3. Maintenance and repair of the submerged infrastructure.
4. Addressing the highly improbably, but perhaps possible, event of a submarine or a sinking ship colliding with the suspended tube. Not a small worry in the minds of the public and potential owners.
5. Public perception and operational safety. How to gain the confidence of the public and assure the safety of the infrastructure to clients and the insurance industry? It was suggested that the concept be built first at a theme park like Disneyland to provide a 'soft' introduction to the public.
Pic 2

Typical section of a deep rock subsea road tunnel

All these issues are being considered and addressed by different teams around the world and the locations of possible SFTs continue to be identified. Of several candidates for the first SFT in Norway, the most promising is a crossing of the 4,200m wide, 450m deep Storfjord between the towns of Hareid and Sula near the city of Ålusund. Compared to a deep mined rock tunnel or a long high level bridge, an SFT would be a third the length, consume a third the materials, and offer convincing traffic management advantages.
The first demonstration project may well be the 100m long SFT in Qiandao Lake in China's eastern province of Zhejiang. Advanced by a joint venture created in 1998 between developers in Italy, who have patented their SFT technology as 'Archimedes Bridges', and the Institute of Mechanics of the Chinese Academy of Sciences in Beijing, the Sino-Italian Joint Laboratory of Archimedes' Bridge (SIJLAB), the 4.3m diameter (3.5m i.d.) Qiandao Lake tunnel is large enough for a single lane of traffic. The steel, concrete and aluminium prototype would serve as a full-scale model to help plan and design a much larger project in China for a 3,300m SFT across the Jintang Strait in the Zhoushan archipelago, also in Zhejiang Province. The Qiandao Lake prototype is said to be close to receiving the green light to go into construction from the provincial government.
• Norway has 961 road tunnels for a total of 671,466m. Many are less than 100m long and 40% are less than 500m. Japan has the most road tunnels in the world - 7, 050 - at last count and for an unknown total.
• Bridge or tunnel? Five routes of about 2km each were studied by the Arup/Jacobs JV for the third crossing of the Firth of Forth north of Edinburgh in the UK. The tunnel was a clear favourite according to a local survey but the selected suspension bridge is about 50% cheaper (approx £1.5 billion to about £2.25 billion); faster to build (5.5 years to 7-7.5 years); provides more flexibility; and involve less risk.
• A submerged floating tunnel across Norway's Gardangerfjord would be 19-20% cheaper than a bridge and require a third of the materials
&bull €20 billion is invested each year on auto-transportation research and development by the European Union - perhaps the largest R&D budget in the world.
Internationally the attractiveness of the concept has found fertile ground in many countries. Switzerland considered an SFT as the extension of the Gotthard baseline rail tunnel across Lake Lugano to avoid ruining the tranquillity and beauty of the lake with new rail lines on the surface around the lake.
Engineers in Italy considered an Archimedes Bridge for crossing Lake Lugano on the Italian side of the lake's national divide and a far more ambitious Archimedes Bridge was proposed as an option for the fixed link across the Strait of Messina to Sicily.
Greece has studied SFTs between its many islands; Portugal considered the option for the Rio Tejo crossing in Lisbon before the high bridge was selected; Spain and Morocco studied the possibility for the proposed fixed link across the Strait of Gibraltar, an alternative that would be 15km long rather than 40km long for a mined tunnel under the 300m deep strait; SFTs have been studied for links across the Bosporus to span that Continental divide.
In North America, an SFT is proposed as a fixed link to Vancouver Island on the west coast of Canada across the 26km wide, 365m deep Georgia Strait, and in Washington State in the USA, an SFT is a considered option for fixed links across Lake Washington east of Seattle.
In Asia, Indonesia, Japan and Korea have expressed interest the SFT fixed link concept for quite ambitious projects, but it might well be China that builds the first SFT in the world.
Back in Norway, the brightest brains in the engineering fraternity are investigating the best solution for a crossing of the Sognefjord is 4,000m (12,000ft) wide and some 1,350m (4,000ft) deep. No small challenge at all!
Deep mined rock tunnels
While work continues to bring the first SFT to reality, construction of other tried and tested methods of providing strait crossings continues. In Norway, the preference continues to be for undersea deep mined rock tunnels.
• 1-2%/year of GDP is an average requirement for maintaining national road and highway networks and a modest amount of new build
• The cost of Norwegian road tunnels is about $US6-10,000/m including installations
• Cost comparison rule of thumb - a foot of tunnel in the USA = a metre of tunnel in Norway (three times as much) when salaries for workers in Norway are twice as high as in the US
• 1-2% of the general public have a phobia about using road tunnels
• There are less accidents pro-rata in tunnels than there are on the open road
• There have been no accidents on the 15km long Trans Tokyo Bay bridge/tunnel highway opened in 1998
• 3-7% of the excavation cost of an undersea tunnel in Norway is spent on site investigation
• 4% of the full budget is to be spent on site investigation for the colossal Fehmarnbaelt bridge/immersed tube crossing between Germany and Denmark
• The min. 1,000m2 or max. 1,625m2 Stad ship tunnel is included in Norway's 2009 National Transport Plan
• Currents through the Bosphorus Strait run at 6 knots
Since building its first undersea road tunnel in 1983, to connect the island-town of Vardø to the mainland, Norway has built 25 more and exported the concept to its neighbours Iceland and the Faroe Islands. These add to the many bridges (60 of them longer than 500m) and thousands of road tunnels on land that go together to create continuous highways through the long, narrow mountainous coastal country.
Table 1. List of completed undersea rock tunnels in Norway
No Project Year completed Main rock types Cross section
Total length
Min. rock cover
Max. depth below sea
1 Vardø 1981 Shale, sandstone 53 2.6 28  88
2 Ellingsøy 1987 Gneiss 68 3.5 42 140
3 Valderøy 1987 Gneiss 68 4.2 34 145
4 Kvalsund 1988 Gneiss 43 1.6 23 56
5 Godøy 1989 Gneiss 52 3.8 33 153
6 Hvaler 1989 Gneiss 45 3.8 35 121
7 Flekkerøy 1989 Gneiss 46 2.3 29 101
8 Nappstraumen 1990 Gneiss 55 1.8 27 60
9 Fannefjord 1991 Gneiss 54 2.7 28 100
10 Maursund 1991 Gneiss 43 2.3 20 92
11 Byfjord 1992 Phyllite 70 5.8 34 223
12 Mastrafjord 1992 Gneiss 70 4.4 40 132
13 Freifjord 1992 Gneiss 70 5.2 30 132
14 Hitra 1994 Gneiss 70 5.6 38 264
15 Tromsøysund 1994 Gneiss 60 a) 3.4 45 101
16 Bjorøy 1996 Gneiss 53 2.0 35 85
17 Slöverfjord 1997 Gneiss 55 3.3 40 100
18 North Cape 1999 Shale, sandstone 50 6.8 49 212
19 Oslofjord 2000 Gneiss 79 7.2 32 b) 134
20 Frøya 2000 Gneiss 52 5.2 41 164
21 Ibestad 2000 Micaschist, granite 46 3.4 30 125
22 Bømlafjord 2000 Greenstone, gneiss, phyllite 74 7.9 35 260
23 Skatestraurnen 2002 Gneiss 52 1.9 40 80
24 Eiksundet 2007 Gneiss, gabbro, limestone 71 7.8 50 287
25 Halsnøy 2008 Gneiss 50 4.1 45 135
26 Nordåsstraumen 2008 Gneiss 74 a) 2.6 c) 15 19
27 Finnfast 2009 Gneiss, amphibolite 50 5.7+1.5 44 150
28 Atlanterhavs Tunnel 2009 Gneiss 71 5.7 45 249

a) Two tubes
b) Assumed rock cover from site investigations, proved to be lacking at deepest point
c) Crossing a narrow strait, only 40m length under sea

Stad ship tunnel
Pic 2
As well as new subsea mined tunnels and the prospect of the first SFT, Norway is also developing the Stad shipping tunnel - a tunnel big enough and deep enough to allow ocean-going ships to cut through the isthmus for a short 1,700m long protected all-weather trip through a 36m wide x 49m high (1,625m2) rock tunnel with a 12m deep draft connecting the sea at both ends instead of the more expensive and hazardous journey around the peninsular.
Pic 2
The greatest drawback of undersea tunnels is that the deeper the water the longer the tunnel. To pass under some of the deep fjords in Norway, the connections to the surface either side must extend several kilometres to control the incline and decline gradients. To limit the length of these approaches, the Norwegians have pushed gradients to 10% in some cases, their maximum for vehicular tunnels. Such steep down gradients are hard to control for drivers and create logjams on the steep up gradients. A third lane for slow traffic on the up-grades have been adopted (most road tunnels in Norway have low traffic volume and are therefore single-tube bi-directional facilities) but the safety issue of these steep gradients as well as their obvious consumption of extra fuel energy, has the authorities in Norway considering a limit on the gradients of 5%. Still higher than other international standards but the Norwegians argue their case on the reality of substantially lower traffic volumes compared to criteria against which the standard is set and prefers a case by case approach for its road tunnel needs.
Depth into bedrock below the deepest point of the fjord is another influence on length of these sub-fjord crossings. Some have been as shallow as 15m into bedrock and the story of the Oslofjord subsea tunnel, where the alignment was heading towards intersection with a deep moraine intrusion required a mammoth freezing operation to save the project, is well documented. A minimum 50m depth into bedrock cover is therefore being considered as a requirement. In combination, these potential restrictions do extend considerably the approaches of deep fjord crossings and, with most of the shallower crossings now developed, the next set of undersea tunnels are deeper still and longer.
Norway's longest undersea tunnel is the Bømlafjord road tunnel completed in 2000. At 7.9km long, it is 260m below sea level, has a 35m-bedrock cover and 8.5% maximum gradients. The 7.8km long Eiksund Tunnel, opened in 2007, is the deepest at 287m below sea level, a 50m bedrock cover and a maximum 8% gradients. Today, Norway is set to build the world's longest and deepest undersea road tunnel - the 25km long Rogfast tunnel to dive 350m below sea level under the Boknafjord from Radaberg, north of Stavanger, towards Vestre Bokn Island and with the side tunnel connection to Kvitsøy Island.

Boknafjord undersea rock tunnel

The SFT concept addresses these issues and at Sulafjord, a 4,200m long SFT would replace a 17km long deep undersea rock tunnel at some 630m below sea level to tunnel invert. The Norwegians however are no strangers to long tunnels. The country is home to the longest road tunnel in the world. The Laerdal Tunnel is 24.5km long. The second longest is the 17km long St Gotthard highway tunnel in Switzerland.
But convincing the public to use long road tunnels is a real issue, particularly those that dive under the sea. A paper at the conference (unfortunately not in the proceedings) highlighted the results of a survey by the Norwegian Public Roads Authority that recorded some 20-30% of those asked were 'uncomfortable' driving through long road tunnels. Another 5-7% were 'afraid' and 1-2% of those asked have a phobia of tunnels.
Comparative figures to other forms of transport were not available so it is not known if these results are more or less than those for using high bridges or flying in aeroplanes. Other papers went on to confirm the Norwegian experience that once built the public becomes oblivious to the actuality of the tunnel and find it hard to believe they were once without the convenience.
Global view and preview
The two-and-a-half days of conference were packed with attention demanding presentations to an audience of 200 from 25 nations. Six invited lectures, a keynote address, and 84 presentations from 18 nations covered this very specific topic but from varied angles.
A consultant to the World Bank was invited to address the issue of funding mega-infrastructure projects and highlighted the popularity of PPP and concession arrangements to supplement public funding shortfalls. On the post conference tour, several internationals in the group were aghast at the equivalent $US15 toll on some of Norway's subsea tunnels and a toll not just on the car but per passenger in it, until it was explained that the fare on the ferry alternative was higher. So this was a good deal for the public.
Another invited lecturer spoke of the threat of climate change on coastal and waterfront infrastructure. The basic message was that pre-emptive engineering to anticipate the possible effects of climate change could prove wasteful, that change will be slow enough for protective defences to be engineered and constructed once the consequence of change is evident and the needs more apparent. "There are still too many variable that will influence the rate of climate change and its consequences to be investing in mega projects at this stage," was the principle thought.
TIMBY concept
Pic 2
A new subaqueous fixed link crossing method not included in the conference line up was the TIMBY concept developed by construction conglomerate Bouygues of France and with mechanical and technical design input by Herrenknecht. The concept is a cross between bored and immersed tube tunnelling a system that this also waiting for its first application.
self-build immersed tube - TunnelTalk, Nov 2009
The periods of two and three simultaneous sessions did require some dashing from one auditorium to another but the programme order and times remained pretty much on schedule, which was a credit to the session chairmen. In addition the rooms were close to each other and the presentations were well grouped into specific topics.
Several existing strait crossing tunnels as well as those in construction were highlighted. These included the existing Tokyo Bay tunnel/bridge highway crossing; the Marmaray immersed-tube and bored tunnel railway crossing of the Bosphorus; and construction of Norway's first immersed-tube tunnel, the Bjørvika highway tunnel under Oslo's foreshore.(6)
Trans-Atlantic SFT
Pic 1
The technology for submerged floating tunnels will continue to develop and lead perhaps one day to the possibility of SFTs spanning the world's oceans, connecting continents and providing supersonic travelling systems. The possibilities are already taking shape and the vision of the Trans-Atlantic fixed link is being considered seriously on both sides of the ocean, or 'The Pond' as it is known affectionately. Supersonic trains powered by maglev and travelling in a resistance-free vacuum at 3,000km or more an hour to complete the journey from Boston to Brest within the hour! Now that would be travelling.
Pic 2
International fixed link aspirations included connections between Finland-Estonia (the Helsinki-Talinn railway); between Denmark-Germany across the Femernbaelt; between Hong Kong and Macau; between Korea and Japan; a land bridge between India and Sri Lanka; and the international connection of high profile between Russia and the USA across the Bering Strait (a paper by Russian engineers sadly not presented but included in the proceedings).
Issues discussed that applied to bridge and tunnel crossings included positive and negative environmental and social impacts of fixed links; life-time operating and maintenance costs of different crossing options; back analysis of Norway's experience of sub-sea tunnels and long and floating bridges; the limitations and possibilities for deep-sea site investigation study; and fire-life systems for fixed links.
Although not discussed specifically, there were references to the more important lessons learned as a result of Norway's long-followed policy of low-cost road tunnelling. Economical site investigation studies, limited primary support, early use of shotcrete as a rock support, no lining or only shotcrete with rockbolting as a final lining, early use of grouting for water inflow control, drip sheds for water protection, and limited ventilation and lighting were all part of the low-cost road tunnel model.
Alignment changes and ground freezing to avoid serious geological trouble, high pumping costs to deal with on-going and in some cases increasing water ingress, corrosion of drip sheds, and a series of serious rock falls, fortunately to date with no injuries - the worst being a fall of about 2,000m3 in the Hanekleiv tunnel in December 2006 that blocked the road for some time - have resulted in significant policy revisions.
More detailed geological site investigations using modern, improved techniques; spilling of loose rock instead of relying on relatively thin layers of shotcrete for long term rock support; tightened specifications for rockbolting patterns and lengths; corrosion protection for rockbolts; pvc waterproofing membranes and in-situ concrete for final linings in reaches of poorer rock condition and to control areas of highest ground water ingress; a full canopy of corrosion resistant pvc on corrosion resistant frames in reaches of competent rock to prevent water damage of the road deck and channel water to drains and sumps; and more lighting in tunnels as the traffic levels increase are becoming standard specifications for Norwegian road tunnel.
Wrap up
TunnelTalk was invited to give an overview of the event and there were a few points to highlight.
For development of the SFT concept there was a sense that the engineers engaged are spinning their wheels, that the need to actually build one was needed to cut through the oftenpassionate discourse about laboratory and engineering studies in which the different groups are engaged. Long strings of flummoxing equations were used to argue one point against another and left some, if not most, in the audience on the sideline. It appeared beyond the point where at least the first prototype version of an SFT should be in existence as a facility as well as a real example for study.
Pic 2

The Giertsen waterproofing membrane system erected as the final lining

Several areas of this very specific sector of civil engineering, and its overlap into other sectors of engineering and urban/social planning, were highlighted as requiring more research - traffic behaviour statistics, long-term durability of construction elements, life-cycle costs of operation and maintenance etc - as well as a wider sharing of the experience and knowledge already gathered by the leaders in this field. When Eivind Grøv, Chairman of Norway's SINTEF and Co-Vice President of ITA, asked the delegates at the closing session when should the 6th conference in the series be convened, "soon" was the answer and certainly sooner than the eight years between this event and the last in 2001.
Most countries in the world have already, or have the need, or are planning more strait crossings. The need for sharing the experience about these elements of public infrastructure won't diminish but will grow. In addressing this need the Working Groups of the ITA provide an excellent forum for sharing information about strait-crossing tunnels and long tunnels at great depth(7) and the next in this series of conferences should, and most likely will, attract double or treble the number of delegates. Of course Norway could convene the next strait crossings symposium, and we would all appreciate the opportunity to visit the spectacular country, but perhaps it is time for another host nation to realise the need from its own perspective and bring the experts together.
Pic 2

Photo of the tour group at the Eagle Road view point overlooking the magnificent Geirangerfjord

Following the conference, a group of 40 enjoyed one of the most spectacular three-day post conference tours ever. The weather was perfect and the coach tour, guided by Kaare Flaate and Kåre Senneset who provided excellent commentary as we journeyed, was filled with camaraderie and included stops at all points of interest for the tourist and the engineer in the group. From Trondheim the tour travelled the relatively new route of the E39 and included a ferry ride and several tunnel pass-throughs on the route to Kristiansund. The following day the journey was on the Atlantic Ocean Road to Molde taking in a drive through the route's new 5.7km long sub-sea tunnel which is holed-through and being prepared for official opening in December(8). On driving through, the visitors observed the final section of more than 100,000m2 of the WG T100 waterproofing system from Giertsen Tunnel AS, a system of plastic membrane erected on a light framework attached to the walls of the tunnel to catch and drain any water ingress into the often largely unlined rock tunnels and draining it to the invert drainage systems. Lunch that day was at the famous Vardestua hotel and we continued with a ferry connection and the Trollstigen mountain road of hairpin switchbacks and a stop at the top of Trollveggen where the view was outstanding. After a night in Geiranger, the town at the end of Norway's longest fjord and arrived at via the Eagles Road, the morning was a ferry ride down the majestic Geirangerfjord and a continuation on the road to Ålesund, passing through the world's deepest sub-sea Eiksund road tunnel, stopping to see the location of Norway's second floating bridge and the spot where the country's first SFT might cross the Storfjord between Hareid and Sulasundet.
Copies of the Volume of Proceedings (569pp) are available from by contacting NFF, the Norwegian Tunnelling Society. A copy of the Proceedings is also available at the Library of the ICE in Great George Street London.
3. Norwegian Public Roads Administration
4. Norwegian Tunnelling Society
5. ITA-AITES Working Group 11: Immersed and Floating Tunnels
6. Fire protection for Oslo's immersed tube tunnel - TunnelTalk, Nov 2009
7. ITA-AITES Working Group 17: Long Tunnels at Great Depth
8. Norway's newest undersea road tunnel opens - TunnelTalk, Dec 2009

Add your comment

Thank you for taking the time to share your thoughts and comments. You share in the wider tunnelling community, so please keep your comments smart and civil. Don't attack other readers personally, and keep your language professional.
In case of an error submitting Feedback, copy and send the text to
Name :

Date :

Email :

Phone No :

   Security Image Refresh
Enter the security code :
No spaces, case-sensitive