THE $200M Manapouri project, which took 10 years to complete, represents New Zealand’s largest energy-efficient undertaking. The tunnel 10km long and 10m in diameter has allowed Manapouri to generate up to 730MW, enough additional power to supply the needs of 64,000 homes.

Manapouri power station is located in the heart of Fiordland National Park and the United Nations World Heritage Area, a scenic but isolated section in the remote South Island of New Zealand. Power at the station is generated by diverting water from Lake Manapouri to an underground power house from where it is discharged through a 10km long tailrace tunnel to Deep Cove in Doubtful Sound.

Built in the 1960s, the station had been unable to generate more than 590MW, far short of its original design expectations of 700MW, due to insufficient capacity in the tailrace tunnel. In the early 1990s, Electricity Corporation of New Zealand (ECNZ) developed a concept for increasing the output from the station by constructing a second tailrace tunnel. URS was engaged to investigate the technical and economic feasibility of the plan, and to develop the detailed design for the Second Manapoui Tailrace Tunnel.

The project presented a wide range of challenges, from technical and engineering issues to the demand for a cost-effective and environmentally sensitive solution. Its geographical isolation and pristine environmental conditions, including rock material and terrain, needed to be preserved. The paramount requirement was to protect the World Heritage Area, a site of international interest that attracts thousands of tourists each year.

A strict environmental specification was required to ensure that no unacceptable pollutants were brought in and designated waste was removed from the park on completion. Potential pollutants, such as dirty tunnel water discharges, diesel fuel and oil, were managed carefully and met all wastewater clarity standards. The carefully engineered waste rock dumps were contoured and replanted with locally sourced native plants. All construction waste and debris that could cause environmental problems, except for the tunnel spoil, was transported out of the national park. A diving programme was implemented to monitor rare black coral to ensure that sediment released during channel excavation did not harm underwater plants or marine life.

With the Deep Cove portal located below sea level, a significant sheet pile cofferdam and deep well pumping system were required to provide construction access to the tunnel groundwater inflows had been a major problem during construction of the original tunnel. Design studies indicated that the second tunnel could be positioned within a groundwater ‘pressure shadow’ around the existing tunnel to reduce total inflows. Its positioning was also calculated to prevent the risk of tapping leakage flows from the existing tunnel.

Geological problems

The potential for differing site condition claims is inherent in underground construction, and nothing can eliminate the risk of encountering unexpected conditions. In the past, tender documents for tunnel construction projects provided factual data gathered from nearby construction and subsurface exploration programs but left it to the tenderers to evaluate and cope with the risk that variable conditions might be encountered. For the tailrace tunnel, an evaluation was provided in document called a Geotechnical Baseline Report (GBR). It was the first time such a document was used for a major project in New Zealand. The purpose of the GBR was:

• To set forth the designer’s interpretation of the available subsurface data and previous construction records, which were detailed in the accompanying Geotechnical Data Report.

• To describe how the anticipated subsurface conditions had influenced the construction plans and specifications.

• To provide clear baseline statements of the conditions likely to be encountered during construction which would be used as a basis for contractors in the development of their tenders and as a key tool in assessing the merits of differing site condition claims if they should arise during the work.

The baseline statements in the GBR did not represent a guarantee that those conditions would actually be met; they represented a set of contractual assumptions to be applied to all tenderers equally and to assist in the administration of the differing site conditions clause in the contract. Some aspects of the project particularly those that resulted from the contractor’s work methods were not assigned baseline parameters.

Specific geological problems were anticipated to be met along the way, including poor ground conditions. To counter those conditions, the design incorporated specific ground support systems for various ground classes and provided for probe drilling to investigate and pre-drain the rock and formation grouting if required. Also, a specific type of support system was designed to protect against the possibility of rock spalling.

Tunnel design and construction

A new tunnel boring machine (TBM) was designed to meet the specific project requirements and ground conditions described in the factual Geotechnical Data Report and the interpretive GBR. An open-face hard rock type TBM was specified for the project because it offered technical advantages, including compatibility with the anticipated rock conditions and proposed stabilisation systems; the ability to install stabilisation requirements; and the ability to achieve higher advance rates in good quality rock. It also provided more flexibility than a closed machine if difficult fault conditions were encountered.

The tunnel design recognised three main ground types, each with different support requirements. Type T1 ground was high quality rock supported only by pattern rock bolts in the crown of the tunnel. In this ground class, elimination of the concrete lining allowed faster construction and a larger diameter of the finished tunnel. Coupled with a TBM for excavation, which ensured a smooth walled tunnel in good rock, an unlined tunnel guaranteed better hydraulic efficiencies and resulted in a greater head gain than expected because the use of good rock made it possible to leave a higher percentage of the tunnel unlined.

During construction, it was necessary to manage large groundwater inflows and to dewater the tunnel for inspection. A tunnel dewatering system was designed consisting of six 560kW pumps, each capable of moving 800l/sec. All six were controlled through two 1000V motor control systems with soft starters controlled by PLCs.

Safety was a critical issue throughout the entire project. Construction of the original station and first tunnel resulted in the death of 16 workers. With careful attention paid to construction methods, the second tunnel was completed with no serious injuries.

Technical action teams

A key part of the design process was the use of a technical action team that met very early on in the design process to discuss the design and confirm the overall design concept. Team members included internationally recognised tunnel designers, owners’ representatives, and independent technical specialists. The design process resulted in the decision to construct the tunnel using a tunnel boring machine rather than traditional drill and blast methods. As a further check, the owner, Meridian Energy appointed an independent board of review to oversee the development of the project right through construction.

The Manapouri project was completed with minimal disruption to the vital power generation of the station. Since its completion, it has exceeded all projected estimates of additional output.