Alpine Fault shake in Canterbury could be of 'higher intensity' than previously thought

Saturday, 15 April 2017

Milford Sound will be ruined by an earthquake on the Alpine Fault.

A massive earthquake on the South Island's Alpine Fault is predicted to shake Canterbury at a 'much higher intensity' than previously thought, new research shows.

The findings come a month after it was reported that urgent national planning for a magnitude-8.2 rupture of the Alpine Fault is underway, with a formal plan to be released next year.

Misko Cubrinovski says using simulation to predict shaking means 'you are quite specific in what you are predicting'.

Shaking near the epicentre, not far from Haast, would be unlike any felt in modern New Zealand history.

The new research, completed by Canterbury University researchers, provides the first 'high-fidelity modelling' of the severity of shaking from future major earthquakes on the Alpine Fault.

Simulations of how much the ground will move during a major Alpine Fault earthquake forecast a much higher intensity of ground movement in places such as Canterbury, according to new research.

* Preparing for the big one on the Alpine Fault

* New Zealand's large, fragile mountains face twin threats**

Simulations of how the ground would move during a large earthquake on the Alpine Fault predicted a 'much higher intensity' of ground movement, particularly in Canterbury and northern parts of the South Island, than previous models.

In Canterbury, the simulation of an earthquake centred at the southern tip of the Alpine Fault, near Milford Sound, would produce peak ground velocity amplitudes that were 1.5 to 3 times greater than the empirical prediction.

According to the research, empirical predictions showed there would be peak ground velocity of zero to 10 centimetres per second.

'As illustrated . . . the ground motion estimates from the presented simulation models suggest greater amplitudes in several notable areas (namely Canterbury and the northern half of the South Island), compared to the previously utilised empirical predictions.'

The new research had been handed over to the Project AF8 programme, which proposes an 8.2 earthquake triggering near Haast.

That project involves emergency management officials preparing for what to expect, and how to coordinate a response.

Canterbury University professor of civil and natural resources engineering Misko Cubrinovski said there were two principal ways to predict ground shaking.

One was the empirical approach, based off records from previous earthquakes.

'They're telling you roughly what is the relationship between the size of the earthquake, measured by magnitude, and the distance from the fault.

'On that basis, based on empirical evidence, you can predict what kind of shaking is going to occur at Christchurch for example,' Cubrinovski said.

However, simulation, used in the latest research, was a more 'rigorous' approach.

'So you are simulating the rupture process.

'That generates some seismic waves that travel through [the] particular topography and geometry of the Canterbury Plains, and other features of the South Island, and is reaching particular locations of interest,' Cubrinovski said.

'So you are quite specific in what you are predicting . . . there is a qualitative upgrade in terms of the prediction.'

Magnitude-8.0 or greater earthquakes were expected on the Alpine Fault approximately every 300 to 500 years.

According to the research, the last major rupture of the Alpine Fault occurred 300 years ago, in 1717.

Asked exactly what a magnitude-8.2 earthquake would feel like in Canterbury, Cubrinovski said this was dependant on a number of variables.

'Because you're talking about a rupture which is hundreds of kilometres, so it's completely different if it goes in the Northland direction or in the Southland direction, or if it splits and goes both ways.

'All those features are obviously important, but the key is that with that kind of [simulation] model now you can check all those scenarios.

'There is no simple answer . . . but what you can see is that you can really quantify and say 'well it is highly likely that we are getting maybe higher responses that what we anticipated from the empirical models,' Cubrinovski said.

There was an estimated 30 per cent chance of a major Alpine Fault rupture over the next 50 years.

*Comments are closed on this article