Icelandic epic – tunnelling methods
Chemical grouting prevented an Icelandic tunnel project becoming a total wipe out. Michael Forrest reports
The north coast of Iceland is home to arctic gales, significant snow fall, high mountains, deep fjords and some of the youngest rocks on the planet. The weather and topography isolates the settlements of Olafsfjordur and Siglufjordur for significant periods of the year requiring long circuitous drives to the south or a sailing trip. To alleviate these problems, Vegagerdin, the Icelandic Road Administration, decided to develop two tunnels to link the towns.
Olafsfjordur and Siglufjordur are on the central part of the north coast, to the west of the fissure swarm that marks the mid-Atlantic ridge. Like most of the country, the geology is dominated by the volcanic outpourings that have occurred over the past 16 million years. The geology of this part of the north coast is olivine basalt, a hard competent rock occurring in flows 10-15m thick. However, the rock has been affected by a series of thrusts, often with interbeds of sediments between 0.1m and two metres thick. In addition, many sub-vertical dykes and faults are common throughout the succession. This feature has a big impact on the hydrogeology, which in turn has a significant geothermal component. Both settlements have district heating based on nearby hot wells. The overburden (depth below surface) of the tunnels varies from 80m to over 800m with associated rock stress levels. The area is also seismic.
The tunnelling method was drill and blast using two Axera Tamrock T11-315 rigs. They were fitted with a semi-automatic technology computer aided design, enabling accurate positioning of the holes into which explosives are loaded. Although the geology is relatively well known, part of the drill/blast cycle encompassed probe drilling to 25-33m in up to four holes 51mm in diameter. The purpose was to confirm rock strength, and to test water inflow.
To create the tunnel face, around 110-120 blast holes per round were drilled at a depth of 2.5-5.3m. For the most part, the chosen explosive is ANFO (ammonium nitrate/fuel oil) emulsion using non-electric detonators and 25g boosters. Further drilling was undertaken for inserting rock bolts to support the roof, but this was not necessary in the walls during construction. Finally, ‘umbrella’ holes were drilled for pre-grouting to ensure stability and deter water ingress. Bolting was then completed and shotcreting of the tunnels before and after waterproofing.
Advancing the tunnelling from Siglufjordur was rapid at 200m/month with little in the way of water ingress as the tunnel inclined upwards for 1.6km. In the Olafsfjordur tunnel, difficult conditions reduced the rate to 180/month.
For the most part, the drilling and blasting was only constrained by the drill bit penetration rate and the cyclical blasting and mucking. Although the dominant rock type in both tunnels is hard basalt, the occasional inter-flow sediments caused some problems, particularly if they sub-cropped in the roof of the tunnel. Removal was the most common method of dealing with the problem.
The water’s lovely
Ermín Stehlík, Project Director, at Metrostav a.s. Prague, says, ‘Another major concern was the possible interference of the water table that might diminish the geothermal wells that supplied district heating to both Siglufjordur and Olafsfjordur, as both supplying wells were near tunnels portals’. Fortunately no disruption has been seen.
More serious, however, were the vertical dykes and faults that often connected from the tunnel roof to the surface and their effect on the hydrogeology. The number and scale of these dykes, although easily mapped at the barren rock surface, required probe drilling up to 30m ahead of excavation. In the Siglu tunnels, water at between 10°C and 20°C was encountered, often at considerable pressure. Stopping the flow required pre-grouting holes, usually with cement, to seal the rock jointing. With up to 800m of overlying rock, these dykes and fractures effectively provided a high pressure conduit of water to the tunnel waiting to be tapped by the probing drill.
In the Olafsfjordur tunnel, the problem was exacerbated by greater overburden due to the higher mountain, resulting in water pressures in the drill hole of up to 32bar. Furthermore the water temperature, was between two and three degrees Celsius making for extremely tough working conditions, and crucially preventing the setting of cement. The contractor turned to polyurethane (PU) resin to seal the flow.
‘The scale of use was unprecedented,’ says Stehlík. ‘We used nearly 30,000 cannisters of PU, equivalent to 630 tons to seal the inflow. Often in highly fractured sections the rock face resembled an amber deposit. In one probe hole, the pressure only reduced from 32bar to 19bar after 10 days with no measurable decline in flow of 50-60l/s. Hydro-pneumatic packers finally sealed the flow.’
The final finish of the tunnel requires shotcreting over a waterproof membrane of 45mm thick
polyethylene (PE) foam. This also acts as frost protection to the tunnel walls and is held in place by 0.3m bolts of 20mm diameter. ‘The total number of holes drilled to secure the membrane exceeded 46,000,’ says Stehlík, ‘Our demand for small-diameter drill bits cleared the entire European market’.
Grand day out
Support for the tunnel both from the town populations and the Government was strong and informal. Breakthrough celebrations frequently had to be repeated and locals took to driving into the tunnel at weekends to see progress. This became a safety issue and large machinery had to be placed in the portal to prevent further incursion. The tunnels are due to open later this year.