Why did the Port Hills faultline rupture?
The Port Hills Fault that caused so much death and destruction in Christchurch on February 22 last year is a tiny feature compared with other faults in the central South Island. Science reporter PAUL GORMAN investigates why it broke when it did in disastrous fashion.
Titanic forces between the crustal plates surrounding and pushing into New Zealand chose a spot under Christchurch's Port Hills to break out on February 22, causing the country's most destructive earthquake in 80 years.
Set against other active faults across Canterbury and the central South Island, the Port Hills Fault, as it later became known, is a diminutive split in basement rock a bit over 10km in length that runs almost west-southwest/east-northeast from near Cashmere underneath the Avon-Heathcote Estuary and a few kilometres out to sea.
The small fault was a previously unknown feature which first hinted at its existence a few mornings after the September 4, 2010, magnitude-7.1 earthquake.
But small does not mean it was weak, even though the magnitude-6.3 aftershock it generated began at its weakest point. In fact, scientists say it was a very strong fault, a consequence of which was the severity of the shaking it produced when it ruptured.
Often touted before September 4, 2010 as a region with low seismicity, Canterbury has had more than its share of large earthquakes during the past 150 years.
The Southern Alps are riddled with faults, including the Alpine Fault running along the western edge of the Main Divide and its subsidiaries, the longest of which fan out and cut through North Canterbury and Marlborough.
The North Canterbury foothills are home to a number of faults, while faults are also in evidence offshore under Pegasus Bay and hidden beneath the Canterbury Plains and Banks Peninsula, as we now know only too well.
The Alpine Fault marks the boundary between the Pacific Plate, which Canterbury sits on, and the Australian Plate.
The Pacific Plate is being driven obliquely into the Australian Plate at the rate of between 35mm and 38mm a year.
About 25mm a year of this convergence in central South Island is taken up by intermittent slip on the Alpine Fault during large earthquakes.
The remainder causes distortion of the crust, stressing the network of faults throughout the South Island until they break, generating earthquakes.
Calculations show that the Alpine Fault accommodates about 70 per cent of that plate boundary motion, releasing it in great earthquakes of around magnitude-8.0 on average once every 300 to 400 years.
Most of the rest of that motion is soaked up on the larger regional faults and only a small amount of strain, perhaps less than five per cent, is harboured on the recently discovered faults around Christchurch.
Local scientists have warned for some years of the likelihood of hidden, "blind thrust", faults below the gravels and sands of the Canterbury Plains.
The September quake, a complicated event consisting of at least three or four quakes generated almost simultaneously by as many separate faults, revealed where some of those concealed structures are.
There were other unknown hidden faults too below the volcanic rocks of the Port Hills and Banks Peninsula the February 22 fault among them which sparked into life after September.
GNS Science seismologist Martin Reyners said these faults were formed about 12 million years ago, when the Lyttelton volcano punched up through the basement greywacke rocks and forced cracks into them.
Retired Canterbury University active tectonics expert Jocelyn Campbell said many of Canterbury's hidden faults were "basically all pointing at one place - Christchurch".
"It was always a matter of if, not when, these would start to arrive in our neck of the woods," she said.
Since the first big quake, the regional stress has slowly moved east, triggering the February 22 quake, then the June 13 double whammy and most recently the December 23 aftershocks.
Campbell has a theory that Canterbury's current earthquake sequence may have had its genesis in a magnitude-6.7 quake 10km from Arthur's Pass in June 1994.
The quake was felt strongly across the region and in Christchurch, and was followed by a 6.1-magnitude shake and seven others of more than magnitude-5.0 within weeks.
Aftershocks from that event have, over the years, gradually drifted south from near the Harper River across Lake Coleridge and down the Rakaia River before turning east around the Malvern Hills towards Hororata and coming almost within sight of the western end of the Greendale Fault.
Since September 17 months ago, that stress has transferred further east across Canterbury, heading below the Port Hills and offshore near Christchurch.
So if the Port Hills Fault is such a small player in the regional scheme of things, why was the February 22 quake so bad?
The rupture began just before 12.51pm and 43 seconds about 7km to 10km below the Port Hills and tore upwards towards the city at about 3km a second, with a maximum slip between the two sides of the fault estimated at between 2.4m and 3.6m after 2.5 seconds, beneath the estuary.
The rupture energy arrived at the surface after about four seconds.
As the waves raced across the city, huge peak ground accelerations were recorded in southern and eastern parts, with vertical accelerations of more than twice that of gravity (2.2g) measured at Heathcote Primary School, the strongest recorded in any New Zealand quake.
GNS Science seismologist Bill Fry said it was a rare combination of factors that caused such intense shaking.
The fault was unusually strong, which meant when it broke it released more energy, like snapping a piece of plywood in half rather than a sheet of polystyrene.
The strength also meant the speed of the underground rupture was faster than for most quakes of the same magnitude.
The ground shaking was "punchy" because the waves from the rupture were travelling in the same direction as the rupture itself, making it more energetic, he said.
"It's like bulldozing snow. You are pushing snow along that you've already collected, but all the time you are adding more."
The dip of the Port Hills Fault, sloping down towards the southeast, meant it was effectively pointing back directly at Christchurch like a loaded gun. When it ruptured, the energy shot northwest straight across the city.
Layers of sediment below Christchurch also separated under the intense vertical acceleration, an example of the recently discovered "trampoline effect".
Weaker upper layers bounced up further than stronger layers lower down, then slapped back down against the lower layers as they came up, producing very high impacts, Fry said.
"Think of jumping on a trampoline. If you change the rate at which you jump and get out of synch with the trampoline, when you come down and land as the trampoline is still heading upwards, you get quite a jolt to your knees."
Scientists also found that the Port Hills Fault had not needed a lot more "encouragement" to break and may only have been decades away from that.
Using Coulomb failure stress (CFS), a measure of a fault's state and how close it might be to rupturing, they found stress on the Port Hills Fault increased by only a "modest" amount, less than 0.1 of a megapascal, after the September 4 quake.
Fry said that suggested the fault was already highly stressed and close to failure before September.
Unfortunately the prospect cannot be ruled out that more hidden faults below Canterbury will one day make themselves known. However, scientists think the Greendale and Port Hills faults have very long recurrence intervals of tens, possibly even hundreds, of thousands of years.
Those faults are therefore unlikely to rupture again in such dangerous fashion for many, many years.