Feature

Slip Sliding Away

The landslides causing concern in the Cromwell Gorge are not simple little slips, but huge ancient masses of broken rock that pose problems for both geologists and Clyde Dam engineers.

By Vicki Hyde

The announcement of the vast sums of money required to "fix" the landslides threatening the Clyde Dam project have prompted a flurry of helpful suggestions. These have included building retaining walls, planting trees on the slopes and carting away the loose rock.

The nature of the suggestions reveals the misperceptions that abound over the issue. The 10 landslides in and around the dam area are ancient formations dating back 50,000 years, consisting of folds of schist rock and thin clay seams. The landslides are deep, measuring 50 to 150 metres thick, and cover areas of several square kilometres. The largest, at Nine Mile Creek, runs to a billion cubic metres of broken schist, gravel, silt and debris.

They aren't sliding actively, but are creeping along at around 50 mm per year. Despite the thousands of years since their formation, � the landslides have yet to encroach upon the gravel terraces running alongside the Clutha River. Several large earthquakes that have rocked the region over the last 18,000 years have not shifted the landslides by much.

Early Concerns

In the 1970s, the preliminary studies suggested that the slides were � stable, and a small contingency budget was set aside for any unforeseen geological problems. Nevertheless, even in the early days of the Clyde Dam project severe reservations were noted concerning the geology of the area. Files show constant complaints of a lack of sufficient personnel or finances to do a thorough geotechnical assessment.

Years of careful monitoring were required to confirm some of the early concerns. It was not until over 10 years after the project got its go-ahead that geologists began to find evidence that the landslides were more active and more widespread than first thought.

It was suggested that the slow slippage be stopped and stabilised by dumping huge quantities of fill material at the bottom of the slope. This toe buttressing would also provide support for the base of the slide when it was drowned by the rising water of the filling dam.

Trapped Water

To complicate matters further, a high pressure system of trapped groundwater was discovered under the No. 5 Creek landslide. Further drilling revealed more trapped water at other sites. The water reduces the effective weight of the overlying debris and encourages slippage, and also places great pressure on the slope. Complex drainage measures have been developed to remove the water and provide some stability.

Initial estimates for remedial work were set at less than $10 million. The discovery of the need for drainage work and major buttressing has raised this to over $330 million.

The geological problems didn't stop with landslides along the edge of the proposed lake. A seismotectonic hazard study looked at the active and secondary fault lines in the area. The River Channel Fault, discovered to be running across the dam site, meant that 270,000 cubic metres of material had to be excavated to produce a solid foundation. The original estimate called for only 14,000 cubic metres to be removed.

Design Changes

Eventually the design of the dam itself was changed to make allowances for newly identified geological hazards. The active Dunstan Fault meant that geological forces could push one half of the dam up by as much as 60 centimetres. To counter this, a slip joint running down the centre of the dam face was added to react flexibly to any major earth movements.

It is imperative that local geology is considered when designing dams. In 1962, in Vaiont, Italy, a hillside took 45 seconds to plunge into a new reservoir. The resultant wave and flooding downstream killed more than 2,000 people and generated air pressure changes strong enough to pull people through windows.

The Ministerial Review of the Clyde Project clearly identifies the concerns of the geologists, the Geological Society, the DSIR and other associated experts concerning the potential hazards of the site. Perhaps the most telling comment comes from a 1983 paper by four DSIR geologists: "The present site was finally selected ahead of alternative sites because of factors which were not directly related to engineering or geology."

Those factors have cost a great deal more than a well-funded geological assessment team could ever have hoped to spend.

Vicki Hyde is the editor of New Zealand Science Monthly.