Koeberg jitters after Cape quakes and new find

Two earthquakes felt in Cape Town last weekend unearthed simmering concerns about the long-term safety of the Koeberg nuclear power station, on the coastline 30km north of the city.

Now there's a new potential worry: scientists have discovered what appears to be a previously unknown geological fault zone on the bed of Table Bay, even closer to the 36-year-old power station than the epicentre of the 1969 Tulbagh earthquake that killed 12 people.

Koeberg's proximity to fault areas and a metro were two major concerns when the plant was built in the face of major opposition. But the government says the power station was designed to withstand earthquakes far worse than those recorded to date in the relatively "stable" Cape.

The Council for Geoscience this week confirmed the new find. There is no evidence the Table Bay fault is active or poses a seismic threat, it said, but the discovery will be incorporated into future hazard mitigation planning. "This is not a new fault - that is, newly formed - but rather an ancient structure that has been newly mapped. This fault has been present for at least the last 100-million, if not 500-million, years," said council spokesperson David Khoza.

Scientists stumbled on the feature while conducting high-resolution mapping. "This work is assisting in identifying previously undefined geological features that may have a wide range of implications," said Khoza.

The strength of the past two major South African earthquakes, in Milnerton in 1809 and Tulbagh in 1969

—  6.3 on the Richter scale

The possibility of a Table Bay fault was described in November last year in a paper in the Journal of African Sciences. It revealed a series of fault indicators, such as rock type and orientation, as well as folds in the rock and the presence of fractures.

But lead author Michael MacHutchon said there is no conclusive evidence, such as a recorded tremor or noticeable displacement along the 10km length of the suspected fault.

"It could also be the case that it is indeed a major fault but that it has not moved since Neoproterozoic times (approximately 600-million years ago, when it would have formed)," MacHutchon told the Sunday Times this week.

Rock characteristics of the inferred fault area match two faults near Cape Town, including the Wellington-Piketberg fault that caused the Tulbagh quake.

Strength of the September 26 quake

—  6.2 on the Richter scale

"Because it has the same orientation and movement characteristics as the other two and it occurs in the same rock types as these, we postulated that it is related," MacHutchon said.

"Of course, if there is a major fault zone running through Table Bay it would increase the risk profile of the city, but with no recorded events along its axis we simply cannot conclusively say it is active or predict future magnitude events."

Cape Town is an earthquake hotspot compared with most of SA, according to the Council for Geoscience seismic risk map. The document is partly based on historical records, which show at least two major quakes since the 1600s, but the map notes that "no earthquakes have been located in the Cape Town area since instrumental recording began in 1972".

News of the new possible fault zone comes at an awkward time for Eskom, which has applied to extend Koeberg's life by 20 years. The plant, the only nuclear power station in Africa, came online in 1984 and was due to be decommissioned in 2024.

It has built-in earthquake safety features, which operate as giant shock absorbers made of neoprene. However, one disaster management source this week claimed these absorbers have a finite warranty and will need to be replaced.

Eskom did not respond to queries about Koeberg's safety features or its seismic hazard profile.

The first of six replacement steam generators arrived at Koeberg this week, but if the extension is approved it will be counter to the advice of Koeberg critics, among them the anti-nuclear Koeberg Alert Alliance, which opposes the R20bn refurbishment.

Alliance director Peter Becker said Eskom had refused to release its seismic hazard report. "The original seismic study was deeply criticised," Becker told the Sunday Times. "They agreed that it needs to be redone. We know it has been done as they submitted it to the nuclear regulator as part of the new nuclear site licence application, but they are refusing to release it for public scrutiny."

Last weekend's quakes occurred hours apart, the first, on Saturday night, measuring 6.2 on the Richter scale in the Southern Ocean, and the second, much smaller tremor, 9km north of Malmesbury, measuring 2.9.