How to mend a shattered city from the ground up

The future cityscape of Christchurch is unlikely to feature any multi-storey buildings such as the Catholic Basilica Cathedral. Photo / Dean Purcell

On the day of the big quake, engineering experts from Auckland were in Christchurch preparing to give a seminar. The subject: making older brick buildings more earthquake-resistant so walls and parapets might not fall down and kill people.

The three academics were in a conference room in the Holiday Inn, next to the high-rise Grand Chancellor Hotel in Cashel St, when the shaking started.

Associate Professor Charles Clifton of the University of Auckland's civil engineering department, soon realised he was okay. So he moved towards the windows.

"I had always wanted, if I was ever in an earthquake, to try to determine accurately how much buildings move. Once I worked out I was safe, I lined up the Westpac building against the window frame and watched it move 1m."

The intensity of shaking, says Clifton, was more than twice that of the September quake near Darfield, and unprecedented in New Zealand - and probably worldwide - for its Richter magnitude.

But the 10-storey Westpac tower performed as it should, withstanding a "maximum considered event" in earthquake design level-speak.

Clifton, with colleagues Professor Mick Pender and Associate Professor Jason Ingham, then stepped outside to see "massive disintegration" of one corner of the Grand Chancellor.

It is suspected a wall supporting the cantilevered eastern side of the 28-storey building gave way, possibly due to foundation failure beneath.

Understanding why some buildings failed while others survived will be pivotal to long-term decisions about the future of Christchurch. If there is to be rebuilding (and in some areas that may be in doubt), the design and standard of building; whether seismic systems and building codes should change.

Some are suggesting that multi-storey buildings should be ruled out in CBD reconstruction, others that suburbs should be abandoned with rebuilding elsewhere, on land less prone to liquefaction.

These are understandable emotional responses but they have, as yet, no scientific basis, says Kelvin Berryman, natural hazards manager at GNS Science.

The Grand Chancellor is likely to have to come down. Most other casualties in the Christchurch CBD are considerably older, unreinforced masonry buildings built to a lower earthquake resistance standard on shallow piles or slabs. Last week, 775 buildings were "red stickered" as unsafe and they may be deemed incapable of being repaired to a safe standard.

Christchurch in the aftermath of the earthquake is a tale of two cities - the CBD and the suburbs. In the CBD the story is one of loss of life, injury, collapsed buildings, devastating damage in close proximity and business brought to a halt. In the sprawling suburbs the damage sprawls too - widespread across a liquefaction zone that has wreaked havoc on housing and infrastructure, breaking roads, water pipes and sewers.

"The general public just see destruction," says Professor Kevin Furlong, an American geoscientist.

"But the causes - liquefaction and shaking - are different so the responses have to be different," says Furlong, who spent six months in Canterbury and is now at Waikato University on sabbatical.

"Constructing buildings for levels of ground shaking is probably more advanced in engineering terms than for liquefaction."

Beca's technical director of earthquake engineering, Dr Richard Sharpe, who was in Christchurch last week, says there is another narrative in the CBD that has gone almost unnoticed - that, almost without exception, modern high rise buildings have performed extremely well.

"It really is of concern that people are saying they would never go back into any multi-storey building. I would have felt safer in a building like the PricewaterhouseCoopers building than I would be perhaps in the [low-rise] accommodation I was in."

People need to understand that in terms of engineering buildings for earthquake resilience, "modern" actually means in the last 20 or 30 years. It doesn't apply to the Pyne Gould Corporation building, built around 1963, nor the Canterbury Television building, designed around 1975.

"That date is quite crucial because '75 was before the real stringency or knowledge in reinforced concrete was hitting mainstream practice," says Sharpe.

Scientists and engineers have a talent for making the impossible sound plausible. Amid utter chaos and ruin, where the focus is on extracting bodies from the rubble, it's early days to be thinking about stitching Christchurch back together.

But there is a wealth of knowledge on what needs to happen. In some cases it's a matter of going back on work underway after the first quake. city from the ground up

First, the experts need to understand what happened: how houses, buildings and infrastructure responded to the forces applied. The starting point is increased knowledge about what is going on under Canterbury - seismic mapping.

"If we listen to the earth it will tell us certain things," says Furlong. Seismic mapping and trend analysis will help to identify any other potential faults and inform decisions about where and whether to rebuild, about ground rehabilitation and restoring infrastructure including roads and underground services.

Sharpe says in terms of mitigation preparedness and building evaluation, New Zealand's response has been as good as any in the world. "The impression is that things we worked 40 years for have worked very well."

Nearly 80 per cent of houses have water supply. But restoring roads, sewage and stormwater pipes so people can go home will be a massive undertaking. Some residents were still using Port-a-loos five months after Darfield.

University of Auckland senior lecturer in construction management Dr Suzanne Wilkinson is a member of Resilient Organisations (ResOrgs), a multi-disciplinary team researching disaster recovery. She says New Zealand has done a lot of preparation for the earthquake and response work has been very well done. Wilkinson says the critical part now is how we manage the long-term recovery and reconstruction.

"Our sense is the Government should take the lead on this one but it needs to do it in partnership with industry.

"My thoughts are that you leave the city where it is, but that has consequences in terms of timing, planning, codes and regulations.

"From our understanding people generally want to go home - they want to feel safe in their home so if you relocate the city that sense of home is lost."

Without power, toilets and water, many people can't wait.

An estimated 70,000 - a fifth of Christchurch's population - have evacuated. After five months of aftershocks from the Darfield quake, some say they won't be back. There have been calls to abandon some areas.

But liquefaction is not thought to have killed anyone. After the September quake, engineers deemed most houses still habitable and in need of relatively inexpensive repairs. The Government signed up to plans to remediate land which would make it more resistant to liquefaction and lateral movement in a future quake.

Devised by Tonkin and Taylor and a bevy of advisers, the plan involved installing perimeter walls in the ground along "free edges" such as rivers, streams and the coast to contain soils and prevent lateral spread.

Other ways to reduce the effects of liquefaction include compaction of surface soils - even drilling down and injecting cement to bind sand and gravels. Such fixes are costly but economies of scale are possible when large swathes need to be rehabilitated.

First, studies need to identify not just the aerial extent but the depth of liquefaction, says Associate Professor Misko Cubrinovski, an expert in liquefaction and earthquake engineering at University od Canterbury. Then costs must be quantified.

Cheap land supply campaigner Hugh Pavletich has renewed his call to fast-track subdivision west and south of the city on land less prone to liquefaction, rather than undertake costly land remediation which could be ruined in a future quake.

Pavletich, a Christchurch-based developer, says this would allow residential builders to get cracking and "those who wished to do so could sell their existing homes, allowing others with destroyed or damaged homes to relocate to them".

He also suggests high-rise reconstruction should be avoided.

"It seems likely people will not be interested in returning to what they consider suspect buildings - those of older construction and above two levels. Going forward, Christchurch is going to be, in construction terms, a 'low and light city'."

Peter Marshall, director of architects Warren and Mahoney, also worries fear will dictate future building form.

"I think in Christchurch in our lifetime there will not be any multi-storey buildings built," he told the Herald midweek.

The council would also have to take the hard decision not to allow rebuilding in areas with the worst liquefaction, he said.

"We are going to have to relocate one or two suburbs."

Such predictions are speculative and premature, say many experts.

Cubrinovski says it's premature and speculative to talk about abandoning suburbs or imposing building restrictions before people understand the causes and effects of what happened.

"It might be the eventual conclusion but we don't have any evidence at this point to decide. And we'd need to consider what that would mean, not only in a technical sense but in economic and social costs. We need to consider everything.

"But I cannot see any solution that is cheap - this is going to be a costly exercise whatever we do. It is really widespread, it is affecting significant parts of Christchurch."

Suzanne Wilkinson, from Auckland University, says research on the recovery following the bush fires in Australia provides some useful guidelines.

The Government appointed a lead engineering company which put a system in place and got the clean-up done very quickly. But when everything was cleaned away, huge areas of empty plots of land were left.

"What was lacking was the next phase of actually intervening in the resource allocation for the rebuilds."

A year and half after the bushfires things looked little different to how they were after six months .

"A few public buildings had gone back, but you had very few houses being rebuilt.

" The communities don't want to go back if there is no economic activity, but people don't want to start up their businesses if there is no community."

Path to recovery: issues that will form the basis of expert consideration on the future of Christchurch

Seismic gaps
A GNS Science image shows the fautline from the September earthquake running through the Canterbury countryside.The Greendale fault, which caused the September 4 Darfield quake, was unknown to earthquake geologists. Officially, GNS Science won't even name the new fault which ruptured in the Port Hills because it has no surface expression. Some experts, including Canterbury University geologist Mark Quigley, say Christchurch could be vulnerable to further unknown faults and extensive and expensive seismic mapping should take place before rebuilding is contemplated.

Others say while data-gathering should inform long-term planning, it need not delay work to get Christchurch back on its feet. Better seismic data will help longer-term decision making about building design, building codes, standards of infrastructure construction and the location of future development.

GNS Science natural hazards manager Kelvin Berryman says wide-ranging studies need to be commissioned, including mapping to identify other potential faults. Historical patterns will help to gauge the likelihood of future events. He says it's also important to understand the reasons for building failures.

"[But] we can move on without knowing everything that needs to be known."

Berryman says aftershock patterns ease concerns that there could be another large aftershock in the gap between the two faults.

"It's all crystal ball-gazing but the aftershock activity tends to make us think that a lot of the energy has been released in there. It tends to make us think that there is not a big through-going structure that's still waiting to produce a really big earthquake."

Lifelines
Compared with the first quake, roads and underground pipes were much more badly damaged by land movement and liquefaction this time around. So while houses in many suburbs may have stood up well, the lack of sewerage, water and electricity has forced residents to abandon them.

Liquefaction can make underground pipes and manholes buoyant and sever connections.

"If the ground that was supporting a pipe turns into a porridge, it tries to float the pipe upwards," says Richard Sharpe. In some instances pipes will break, but the other effect is that the gradual gradients of sewer system are disrupted. "Suddenly it's got a section that's uphill." Pumps are being used to deal with changes in pipe gradients.

But replacement infrastructure can be made more earthquake resistant - for instance, by using flexible joints and more ductile materials such as PVC in place of ceramic pipes. Professor Mick Pender, an expert in geotechnical engineering, says there's plenty of international knowledge on replacing underground services in a more robust manner.

Misko Cubrinoski says an area-wide approach should be taken to restore lifelines. "We need a solution which is robust enough to operate as a system rather than looking to individual structures and properties."

Land remediation
Liquefaction
"When you get an earthquake that has got a decent strong motion in it, uniform sands or silts [beneath topsoils] compact a bit and the water pressure in between them rises," says Richard Sharpe of Beca. "It turns into a thin porridge and the pressure tries to escape."

What happens next is that the porridge slips or slumps towards lower ground - what's known as lateral spreading. The worst of the residential and roading damage near rivers has been caused by lateral spreading because the layer of liquefied soil has lubricated the sliding surface. On slopes, the effect is worse.

If the pressure can't be released sideways it spurts upwards through any crack in the ground it can find.

"The public could expect that regional authorities will be more cognisant of this potential when approving building consents in the future," says Sharpe.

Some Christchurch homeowners have in recent years spent extra money on putting piles under their houses because they recognised the potential for liquefaction. The additional cost, says Sharpe, is surprisingly low.

After the September quake, lateral spreading was evident up to 400m away from "free edges" such as river banks. The Government signed up to Tonkin and Taylor recommendations to contain the ground at the free edge with methods such as perimeter walls, using columns or groins, in the ground. This would make subsoils more resistent to liquefaction in future quakes.

The solution was based on evidence that very few buildings actually collapsed and that most houses could continue to be occupied and repaired cheaply.

After the new quake, liquefaction is much more widespread, particularly in the eastern and northern suburbs, and the damage to infrastructure and housing is much more severe. The extent is thought to closely match a map developed in 2004 by Environment Canterbury of potential liquefaction if an earthquake occurred when the water table was high - expected in winter. That it is so widespread when the water table is low is thought due to the intensity of the shaking.

But geotechnical experts say the "confinement" approach to minimising lateral spreading remains valid. Other options may be needed in specific areas. These range from shallow compaction - to create a non-liquefiable raft at the surface - to "densification" of subsoils using heavy vibration equipment and using boring machines to inject cement to strengthen and bind sands and fine gravels beneath the topsoil.

University of Canterbury Associate Professor Misko Cubrinovski says the same measures used to combat lateral spreading can be used to deal with liquefaction.

Slips
A distinct problem is stabilising landslips threatening houses in eastern hillside suburbs including Redcliffs and parts of Sumner. Some properties may be too damaged to repair and rebuilding may be deemed unsafe.

Building design/types
Houses
Old-fashioned timber-framed houses on concrete-block perimeter foundations appeared to withstand the first quake better than some recent homes' concrete slab foundations. Misko Cubrinovski says even brand-new homes built on unreinforced slabs have been damaged. American geoscientist Kevin Furlong says timber frame homes on piles and concrete perimeter foundations may have distorted but didn't break. This raises questions about why solid wall homes on unreinforced slabs were allowed in areas known to be liquefaction-prone. It's early days, but more robust pile foundations may be required on new houses in liquefaction zones. Floating "waffle" slabs are another option.

Houses damaged by the second quake will need to be examined case-by-case. Where foundations have cracked due to sideways ground movement, houses have cracked as well. Another cause of damage is slumping due to liquefaction.

But as in the first quake, many houses stood up well. "They did what they were supposed to do - they didn't collapse," says Pender.

Commercial buildings
As in the September quake, older, unreinforced masonry buildings tended to suffer most damage. Engineers say all that's known about how buildings should be built comes after earthquakes. Having the same city struck twice in five months, by forces at 90 degrees to each other and at different intensities, provides a unique case study, says University of Auckland's Charles Clifton.

Clifton and colleagues Jason Ingham and Mick Pender are part of a joint project to promote earthquake strengthening for unreinforced masonry buildings - options such as securing parapets and strengthening walls.

Larger, modern, multi-storey structures in the CBD should be less vulnerable to liquefaction because of their deep pile foundations. However, some land remediation and compacting will be needed.

"You need to consider the type of structure sitting on top," says Cubrinovski. "You use different methods for a high rise building than for a one or two-storey building - the high rises need much deeper improvement of soils. Most [multi-storey] structures with robust pile foundations have performed well."

Beca's Richard Sharpe says there is no such thing as an earthquake-proof building. But the multi-storey buildings in the CBD built to the latest code have stood up well, with shaking that exceeded design parameters, he says. The earthquake standard has minimum loading levels to deal with a "very low probability big event" such as the latest quake.

"We build toughness into the building so that it flexes and graunches at all the joints, but does not actually pancake down."

The exception is the Grand Chancellor building, but even that, in not falling down has met the minimum resilience requirement.

The failure of the Pyne Gould and CTV buildings is less understandable. Yes, they were built before more stringent standards came in. But engineers have known about designing seismic "toughness" into buildings for some time.

Furlong says another puzzle is the collapse of staircases in some high-rise buildings. "They are supposed to be the part left standing if other parts fail."