COLLAPSE OF a section of Australian motorway tunnel during construction last month has cast further doubts about the use of the New Austrian Tunnelling Method (NATM).
The collapse occurred at the intersection of a ventilation adit and a running tunnel during construction of the privately financed Lane Cove tunnel, 75km north of Sydney.
NATM work on the project involved excavating up to 2.5m of tunnel at a time, using road headers to form a 7m high, 8.1m wide bore.
Contractors then installed a row of rockbolts pre-tensioned to 50kN, and applied a 75mm thick layer of C28/35 shotcrete. Rockbolts were then grouted with final waterproofing and another layer of shotcrete applied later.
The cave-in caused the partial collapse of a three-storey block of flats above the intersection, although nobody was injured.
Ground conditions at the intersection include 'highly fractured' low strength weathered sandstone, high strength shale and laminated layers.
It is understood that engineers were rockbolting or excavating the ventilation adit when the collapse occurred, creating a 25m deep hole.
Residents in the flats above the collapse were immediately evacuated and contractor Thiess John Holland pumped more than 1400m3 of C40/50 concrete to stabilise the hole. This was completed by 7am the same morning.
Four days later, a structural assessment revealed that 12 units were safe to inhabit, but nine required further strengthening. The whole structure has been temporarily shored.
A report published by the local Lane Cove council reveals that the intersection was originally located under a park, 65m east of the block of flats. The council said a redesign of the ventilation scheme in December 2002 led to the decision to shift the junction between the ventilation shaft and running tunnel under the flats. This is believed to have contributed to the severity of the collapse.
'As the location of the air tunnel seems to be a crucial element in the tunnel collapse, all the circumstances relating to the changes to the ventilation system, including what level of assessment and consultation was undertaken, need to be investigated, ' says the report.
The council also claimed that design and build contractor Thiess John Holland Joint Venture had not consulted residents adequately over the redesign. In a statement, Thiess John Holland claimed the revised ventilation scheme had been reviewed by an 'independent verifier' and that the community liaison group had been briefed in November 2004 over the changes. It added that the tunnel had also been constructed in the correct location.
Thiess John Holland has appointed rock mechanics expert Professor Ted Brown to carry out an independent review of the incident.
A 'bad luck' event
Imperial College senior rock mechanics lecturer John Harrison said the collapse could have taken place while breaking out the ventilation tunnel to connect with the off-ramp tunnel.
'This is always awkward, because it releases big blocks of rock where the ground is already fractured, ' he said. He added that since the collapse was so localised, it could have been 'just bad luck that it happened under a building'.
'It seems that they must have intercepted a low-strength feature in the rock, ' said Harrison, who drew comparisons with the Channel Tunnel Rail Link collapse at Stratford in 2002, where a tunnelboring machine hit a natural well.
NATM was originally developed for tunnelling in soft, less predictable rock strata. It is an open face method, with stability provided by rockbolting and shotcreting.
There have been a number of high-pro'le failures over the last 20 years, the most notable in the UK being the Heathrow Express tunnel collapse of October 1994.
A month before, another NATM collapse in Munich swallowed a bus, killing four people.
In 1999 the Health & Safety Executive estimated there had been at least 39 cases of collapse or serious deformation on NATM projects between 1973 and 1979.