Rethinking safe: the post-Canterbury view of risk

A 20-storey rockfall tsunami depicted in Norwegian film The Wave could happen in Milford Sound.

A 20-storey rockfall tsunami depicted in Norwegian film The Wave could happen in Milford Sound.

Rockfall tsunamis in Milford Sound, landslide dams forming and bursting on the Waimakariri River. Who really wants to think about the worst that could happen following an earthquake like the Alpine Fault? JOHN McCRONE talks risk with the experts.

If the Canterbury earthquakes have taught us anything, it is that we need to redefine what we mean by safe, experts say.

Created by the twisting upthrust of two opposing continental plates, New Zealand has some of the most unstable landscape on the planet. And almost as importantly, says Davies, we have lived here too short a time for past experience to provide much of a guide to its hazards.

So where other older countries might be more cautious, we have heedlessly built up to the edge of every cliff-top, shore-line and river-edge, chasing today's lovely view rather than worrying too much about tomorrow's possible disaster.

Canterbury University geohazard expert Professor Tim Davies is relaxed in his office. Brushing his floppy fringe out of his eyes, he swings a foot on a chair and tilts back to gaze at the ceiling. We are on to his favourite topic now. Risk at almost a philosophical level.

Canterbury University geohazard expert Tim Davies says risk depends on your point of view. JIM MCCRONE / FAIRFAX NZ

Right, an example, he says. One that his former PhD student Dr Jesse Dykstra has drawn attention to. What are we going to do about Milford Sound? The sound attracts up to a thousand tourists a day, yet it will be a death trap when the magnitude 8 Alpine Fault eventually goes, Davies says.

There is a one in five chance of that happening in the next 20 years. A 50 per cent chance of it going in the next 50. And then almost certainly great slabs of rock will peel off the sound's 2-kilometre high slopes – among the world's highest sea cliffs – crashing into the water to create a local tsunami wave.

Such landslips have made displacement waves 20 storeys high in narrow fjords elsewhere. Norway's Tafjorden disaster of 1934, which killed 41 people, has just been turned into a movie. So the danger to travellers in Milford Sound has to be considered off-the-scale unacceptable, says Davies.

"Anyone who's been to Milford won't have any trouble conjuring up the horrific scenario. You'll get about two minutes warning, and very likely the earthquake will go on for that long anyway. You'd be trying to evacuate yourself while the ground was still shaking itself to pieces."

Ad Feedback

Davies says the policy-maker's calculation of hazard is "risk equals probability times consequence" – death being a major consequence. Generally we tolerate only a 1 in 10,000 annualised chance of death from everyday dangers like a car crash, an epidemic, or drowning. We are quite conservative when it comes down to it.

But say a Milford Sound tsunami is bound to kill 400 unwary tourists during the next 1000 years. The quake must happen, there can be no escape. That works out to a 1 in 40 scale risk. "It way exceeds any acceptable figure. It's 4000 times the normal limit."

Davies says just imagine the international law suits , the scale of the condemnation which could follow, after it is realised New Zealand allowed such a situation to persist. "It's a huge societal and political risk, because you kill 400 tourists all at once and the repercussions will be enormous."

Yet the tricky thing, says Davies, is there is then the equally valid individual traveller's point of view. How people assess the risk when deciding to drive to Milford Sound for the day.

"If you're there as a tourist for about four hours, your individual risk to life is likely much less than 1 in a million. Which of course seems perfectly acceptable." In that context, any talk about having to close down an iconic tourist destination would seem obvious nanny state overkill.

Up to a thousand tourists a day visit Milford Sound.

As it happens, says Davies, government agencies are considering the tsunami hazard in Milford Sound with some urgency. Since the Canterbury earthquakes, New Zealand's shaky landscape is being taken a lot more seriously.

The Milford example illustrates that risk assessment is complicated. In this instance, if you only take the high level view on hazard probabilities, you could in fact be over-reacting because this may fail to take sufficient account of what risk looks like from the individual or even local community point of view.

Davies rubs his chin and pushes his hair back up his brow again. Yes, it is quite an intellectual puzzle, he says. But exactly the kind of thing about which New Zealand – reacting to its most recent experience, and the realisation it may have been pushing the envelope – is having to do some speedy catch-up thinking right now.


Conversations with hazard experts have a way of quickly turning alarming. Did I realise the mighty Waimakariri River used to flow through Christchurch only about 1000 years ago, asks Canterbury University geospatial researcher Dr Matthew Hughes a little too airily?

We are chatting after a presentation on the new "multi-hazard scenario approach" which will underpin the hazard mapping in post-quake Christchurch though processes like the regional-level Natural Environment Recovery Programme ( Nerp ) and city-level Replacement District Plan.

This is what can happen after a big earthquake, says Hughes. Your major rivers start going walkabout. High up in the hills the rocks are loosened, erosion increases. Stones and silt wash downstream. Where they are deposited, they lift the river right out of its existing channel and cause it to switch course.

Canterbury's braided rivers are exceptionally mobile. The Waimakariri River has a hinge point up around the Springfield gorge, says Hughes. Right now it is confined by "1 in 10,000" year stopbanks to keep it up beyond Kaiapoi, its most northern natural extent. It helps that the river's gravel beds are being continually mined for road-building and earthquake recovery construction too.

But if the Alpine Fault goes, or even a lesser Canterbury Hills fault like Porters Pass, that is just the sort of perturbation that has seen the Waimakariri divert south to pass through Halswell and Christchurch. Or even a few thousand years earlier, head right round the back of the Port Hills to exit through Te Waihora/Lake Ellesmere.

Christchurch still learning how the Avon River is responding. Flooding in St Alban's after 2014 rainstorms. STACY SQUIRES/ FAIRFAX NZ

Yikes, the forces of nature, I say. And how to factor that into the belief that out west in Christchurch – on the very shingle beds left by previous Waimak wanderings – is now the best place to build post-quake?

Hughes replies it does take maybe 50 years for the debris to collect, for a river to actually shift. The danger is not exactly imminent. But it illustrates the way that earthquakes inject energy into the landscape and the resulting response can roll on for decades if not centuries.

Dr Deirdre Hart, a lecturer in coastal studies at Canterbury University, says this is why Christchurch is trying to look at its natural hazards in a holistic fashion with its Nerp and District Plan processes.

Hart says it has become realised that the city's story is not just about an earthquake, but an earthquake in interaction with a shoreline. As a flat development on reclaimed swampland, Christchurch's biggest long-term concern now is how it is going to manage its drainage plan.

Everyone knows that most of the central and eastern city has subsided, about 10 per cent of it by over 50 centimetres, she says. How the effects of that will play out are still in the process of being discovered by property owners.

Hart says an example is the surprise flooding of the Flockton Basin and other parts of the city in last year's heavy rains. Subtle new tilts in the local topography can combine to create dramatic ponding .

"A lot of the plain down around the Avon River has changed so we are possibly more vulnerable there than we realise," Hart says. People are being lulled because the post-quake Avon appears to be draining quite normally after moderate rain.

"If you live five houses back from the Avon, you might think I'm not in danger because I'm well inland. I'm not really near the river at all."

However Hart says new modelling suggests that in future rainstorms, the Avon could suddenly cross a threshold and behave in ways no one has ever seen before.

UC's Dr Deirdre Hart warns of Christchurch's change water table levels. DEAN KOZANIC / FAIRFAX NZ

Then there are the longer term landscape dynamics that Christchurch's hazard planning will have to consider. Hughes, who is studying the city's estuary, says it is still rocking with changes.

The Heathcote River end has been raised, making the Heathcote harder to drain. The Avon end has slumped, killing the marsh grass and increasing erosion to Southshore spit. Sediment is choking the estuary mouth, leading to a 15 per cent reduction in the tidal flow.

Hughes says the estuary could go back to "normal" in time. "The Heathcote could re-erode to cut a deeper channel." 

However how do you predict its behaviour over the next 50 years in a way that you can decide where a city's sub-divisions , electricity sub-stations, and other assets can safely go?

Opening the lens wider, Hart says Christchurch has to consider sea-level rise caused by climate change and its interaction with the city's new water table.

Controversially, the Christchurch City Council wants a land drainage plan that takes into account an expected 1 metre rise in sea level over the next century. There is talk of managed retreat from the coastline – restriction on further development which will hit people's property values and hazard notices that may affect the insurability of homes.

The level of local protest has grown so much that in September, the debate was shelved for a while. But Hart says as with the Avon River, the problems are not just for those who are absolute waterfront but may extend much further inland than is generally realised.

She says it is the whole city that does not drain as well because its slope has changed. The land has dropped down towards the existing water table, and over coming decades, sea-level rise is also going to push this entire water table up from below. That means even swampier soils to keep drained and a creeping spread of the suburbs which will be liquefiable when the Alpine Fault next does its thing.

Hart says it is all about how far you want to step back from the immediate here and now in making what can be costly choices about settlement patterns.

The whole planet is having to take a "two shorelines" view of sea level rise these days. There is the shoreline as it presently is, and then the shoreline as it will be once the ice melts and the world returns to its last inter-glacial high-water mark seen only 6000 years ago – which for Christchurch was when Riccarton was the sandy beachfront.

Psychologically, says Hart, it is about making the paradigm shift from seeing a landscape that never changes – allowing you to invest right up to its rugged edges – to seeing a landscape which pulses with rhythms that might make you want to be rather more cautious.


Canterbury's experience is driving natural hazard thinking both nationally and even internationally.

Back in his office, Davies says a big lesson of the Canterbury earthquake sequence was that the improbable is indeed what often happens in life. And again it reflects the difference between a high level and low level view of risk.

Around the world, earthquake planning is still mostly based on simple event statistics, says Davies. This has a way of over-focusing attention on a few large well-known hazards – like in New Zealand, the Alpine Fault – while rare events, such as the 1 in 10,000 year probability Canterbury earthquakes, are almost automatically filtered out of the thinking.

"As soon as you start assigning probability numbers to these events, you immediately begin to take the big but infrequent ones out of your consideration." This feels rational at the national scale, he says. Yet from a local community point of view, the disaster that actually hits it is quite likely to be one of the many lurking rare surprises.

It is just like the Milford Sound issue. As you narrow the view to that of the individual, almost everything bad – a heart attack, a coach crash, fatal food poisoning or a quake tsunami – starts to look like it strikes at life out of the same empty blue sky.

For this reason, says Davies, New Zealand's geohazard planners are starting to shift away from pure probability forecasting to a more "scenarios" based approach. This means really involving communities in the decision making rather than simply imposing what seems the outside expert's perspective.

Franz Josef is a case in point. The West Coast township is built right on top of the Alpine Fault. Hotels and petrol stations straddle the fracture. And since the Canterbury earthquakes, the Westland District Council has felt it has no choice but to push through a hazard plan which effectively red-zones properties 100m to either side of the line.

Franz Josef's petrol station straddles the Alpine Fault. SARAH-JANE O'CONNOR / FAIRFAX NZ

Davies says when you get down to ground level, all sorts of other considerations begin to gain weight. Science can't be so certain the Alpine Fault will go in exactly the same place every time. So you could force the hotels and petrol stations elsewhere – only to find that is exactly where it next breaks.

Or you could push property owners nearer to other local hazards, like the steep mountainsides that might crumble or the local rivers that might break their stopbanks.

Davies says these other risks may each be counted as rare in comparison to the general inevitability of the fault going. But when making a decision that is good for the community – especially when there is a heavy cost to making a change – the many other ways that Franz Josef could be catastrophically disrupted need to be weighed in the overall balance.

Scenarios thinking brings in the wider picture. Davies says it is the way we also need to think about big events like the magnitude 8 Alpine Fault itself.

The disaster statistics approach focuses attention tightly on the fact of whatever particular day it will finally happen. The number of tourists washed away that day in Milford Sound. The fate of Franz Josef's hotels and petrol station.

A more rounded approach is to take the fault as a general example of the extreme things which could occur in New Zealand and then focus on the huge range of knock-on consequences that might follow.

For instance, the Alpine Fault would cause landslides all through the Southern Alps, cutting off the entire West Coast, says Davies. Lewis Pass is liable to be out for six weeks, Arthur's Pass for six months.

"If it's high summer and there's 100,000 tourists over on the coast, it's going to be a difficult situation. Who's going to look after them? There's only ever a few days food and fuel stored over there."

That realisation has made it clear money needs to be spent on the one safeguardable route from Nelson through Reefton, Davies says. Its bridges should be strengthened, its rockfaces secured.

It also shows that as you pan back from the immediacy of the quake itself, the numbers who might die or suffer for other reasons can start to shape the priorities.

Davies says another thing to worry about after the Alpine Fault goes – or any other reasonable-sized mountain quake – is that slips and erosion will block river valleys to create landslide dams.

So the Waimakariri River changing course over 50 years would be one thing. It is true that after any "big one", rivers would be sent snaking about the countryside in ways that threaten towns and farms all over the place. Yet if you had to, you could always build new stopbanks and diversions.

Davies says more of a bother is the idea of the Waimakariri gorge getting plugged, swelling monstrously over a few weeks, then bursting to send a terrestrial tsunami in the direction of Christchurch.

I tell Davies this is not comforting me much. Again, it seems the more you know, the more there is to be worried about. It gets to the point you just want to shut off.

The advantage, Davies says, is it means the Alpine Fault does get put in perspective. Our national reaction does not get too caught up in a single event on a single day, but is instead shaped by a general feel for the odds of a highly dynamic landscape.

The threat of a Waimakariri dam break becomes another small reason why you might be cautious about investing in assets in its path. West Coast communities likewise will have in the back of their minds they might one day be sharing a post-disaster environment with a rather large number of trapped tourists.

Davies says New Zealand needs to think like this because its own experience of what could happen is so recent, its environment so rugged and mobile. But the whole world is now moving fast with population growth and climate change, entering an era when its past history is not that much of a guide.

So really Canterbury does have some interesting lessons to offer everyone, Davies concludes, as he once more leans back in his chair, clasps his hands behind his head and contemplates his office ceiling.

Post-quake preparedness a priority for town
Alpine Fault spreads across South Island, researchers say
Alpine Fault earthquake could isolate West Coast for six week

 - Stuff


Ad Feedback
special offers
Ad Feedback