The Darfield Earthquake struck at 4:36 am on Saturday 4 September 2010. It was magnitude 7.1, and since then GeoNet, the natural hazards surveillance arm of GNS Science, has recorded more than 1200 aftershocks. As anticipated, they are slowly diminishing in frequency and size with time.
I neither felt the earthquake nor was woken by it here in Ngaio, Wellington. I do remember hearing the phone ring at about 7:00am but chose to ignore it and dozed off to sleep again. Some hours later, I determined that the call was from The Kim Hill Show at Radio NZ. In due course I responded and suggested to the producer that they had best contact my colleague Warwick Smith at GeoNet, which they did.
Through the day I listened to the Radio NZ coverage of the event and its aftermath as the nature of the damage became known. The days have since stretched to weeks.
In the week after the Darfield Earthquake, I was in Auckland attending the 11th International Engineering Geology Association congress at Sky City. There were 700 delegates including, as luck would have it, many of the world’s leading earthquake engineers. They were champing at the bit to visit Christchurch and see things for themselves.
I have no doubts that some people in Wellington were woken by the earthquake. I have friends in Paraparaumu who were woken by a large family dog that leapt on to their bed at exactly 4:36am. My mother was woken in Dunedin as was my brother and his family. Not to mention my many relatives and friends living in Canterbury!
So what happened?
Research by seismologists at GNS Science thus far indicates a complex event involving both compressional (thrust) and shear (strike-slip) motion. The strong motion records for the earthquake show two major pulses of strong shaking separated by the order of 5-10 seconds.
A 29km long, roughly east-striking fault break was produced by the earthquake, where active faults have previously not been mapped. The surface expression of the new fault has been named the Greendale Fault. The northern side of the fault was moved about 4 metres to the east relative to the southern side.
The main earthquake epicentre (the point on the earth’s surface projected vertically above the hypocentre) was located some 40 kilometres west of Christchurch near Darfield, hence the name of this event. The hypocentre (the actual point source of the earthquake at depth within the earth, where the earthquake rupture process began) was located some 10 kilometres below the surface. The Greendale Fault ruptured between Greendale and Rolleston, offsetting numerous linear features in the rural, flat, farmed landscape of the Canterbury Plains, including streams, irrigation channels, ditches, drains, hedges, fences, tracks and roads.
The numerous aftershocks relate to adjustment and “settling down” of the fault, and extend beyond the observable surface rupture. Some small events on other faults in the general region may have been triggered by the Darfield event.
While the magnitude and rare occurrence of the Darfield earthquake was already anticipated in both nationally-based and Canterbury-based seismic hazard calculations for the region, the exact timing and geometry of the event and causative fault was not. The occurrence of the event will focus our efforts on trying to improve seismic hazard models to better forecast events for (e.g.) the time span of a human lifetime.
Given that the Darfield earthquake ruptured the ancient (16,000 year old) surface of the Canterbury Plains for the first time in the area, the recurrence interval of the causative Greendale Fault must be measurable in the tens of thousands of years.
In response to the Darfield Earthquake, scientists from GNS Science and various universities have been mobilised for a variety of research activities and in particular the recording and mapping of surface deformation phenomena. Many instruments (mainly seismometers and GPS instruments) and methodologies have been deployed and in time a great deal of new data and research will result in a better understanding of what happened. It is early days as yet. We shall have to wait and see what satellite radar interferometry studies have to offer in terms of how the surface of Canterbury has changed shape, along with detailed seismology and GPS studies.
My GNS Science colleagues have traversed the total length of the Greendale Fault rupture on foot, measuring offset linear features and en echelon geomorphic features developed along its length. These include many broad welts, ridges and fissures along the trace of the fault. They are substantive, producing a relief involving uplift and subsidence of at least 3 metres and deformation of a surface (dominated by river and glacial outwash gravels) that has been subject to minimal deformation and/or erosion processes for 16,000 years. Such geomorphic features have not been recognised previously in this part of the Canterbury Plains. They are fragile features, and will be largely gone from the landscape with ongoing land use in the area (e.g. ploughing).
In terms of plate motion and plate collision, which is the underlying ‘awesome force’ responsible for the Darfield Earthquake, this event was ‘business as usual’. It was the result of pent-up accumulated strain within the earth’s crust as a result of the collision between the westward moving Pacific Plate and the northward moving Australian Plate. The actual plate boundary, the Alpine Fault, lies some 90 kilometres west of Christchurch, so Christchurch and most of Canterbury are located on the Pacific Plate.
The rate of collision is well-established from direct measurement and is 4 to 5 centimetres per year. Research suggests that some 80% of this motion is accommodated by movement on the actual plate boundary, and that the remaining 20% of deformation is taken up on faults distributed within a broad zone of Canterbury up to 100 kilometres east of the Alpine Fault. The Darfield Earthquake may therefore be interpreted as part and parcel of this 20% ‘relief effort’.
The east-west orientation of the Greendale Fault rupture came as a surprise to most earth scientists. This is because the main ‘structural grain’ of New Zealand is NE-SW, which is the sum vector orientation one might expect between two entities (the plates) moving in different directions (one to the west, the other to the north). Furthermore, most active faults associated with plate collision are oriented NE-SW.
However, if we consider Canterbury as just an on-land western end of the Chatham Rise, we get a clearer picture. If we ‘remove’ Banks Peninsula (an intra-plate basalt volcanic complex that erupted between 10 and 6 million years ago), and the young gravels (less than 2 million years old) of the Canterbury Plains, what we would see is geology compatible with that of the Chatham Rise, including major east-west oriented faults.
Oil company exploration of the Chatham Rise in the 1970s produced a great deal of very useful data that were subsequently summarised and published by GNS Science in the early 1990s. Our knowledge of the geology of the Chatham Rise is largely based on this work but it is also well-supported by our knowledge of the geology of the Chatham Islands. Indeed, the east-west prolongation of northern Chatham Island is controlled by major east-west trending normal faults. ‘Normal’ faults involve elongation of the earth’s surface as a function of stretching, whereby one side of the fault slides down the other. This is the opposite of a thrust or ‘reverse’ fault which is associated with a net shortening of the earth’s surface.
These old normal faults relate to stretching of the crust when Zealandia was rifted away from Gondwanaland. This ‘stretching’ phase of our geological history lasted for about 100 million years but has been reversed i.e. subject to compression or collision, for the past 23 million years. It is thought that the Greendale Fault is most probably one of these old Zealandian normal faults that have been reactivated. However, it may not be!
A common question about large earthquakes is this: how do you know that it isn’t just a precursor to something even bigger? The simple answer is that we don’t, at least not initially. However, things improve greatly as time passes. Some statistical studies give probabilities of a larger earthquake occurring soon after the original large earthquake, and in a close-by location. If nothing happens though, we can be more confident that things are settling down. This knowledge is based on almost 100 years of sustained global research effort on patterns of seismicity associated with large earthquakes.
Christchurch and Canterbury had good planning and good luck on their side. It is nothing short of miraculous that no one was killed. To put things in stark perspective, the magnitude 6.5 Boxing Day 2003 Bam Earthquake in southern Iran involved rupture on a blind fault, not unlike the Greendale Fault, and 50,000 people were killed. The more similar 7.0 Haiti Earthquake on 12 January this year killed 200,000 people.
Everyone in Canterbury, if not the whole of the South Island, but especially those in Christchurch, will have their own story to tell of the Darfield Earthquake. I came across two very different versions last weekend which are worthy of mention.
I was on my way to an event at the Te Hikoi Southern Journey Heritage Museum in Riverton which involved driving from Dunedin. I picked up a hitch-hiker just south of Gore. (It was snowing at the time). He told me that he was ‘fleeing from Christchurch’ and that he was ‘terrified by earthquakes’. He had lost all his possessions which he had had in storage in a friend’s brick garage. The brick walls had collapsed and crushed his possessions and then if that were not enough, everything was ruined by water damage resulting from liquefaction-related effects. He was on his way to his mother in Invercargill and was clearly traumatised. He had been with his sister in Methven since the main earthquake.
I eventually arrived in Riverton after dropping the hitch-hiker off at his mother’s house and was met by a lady at Reception at Te Hikoi. She was simply raving about the Darfield Earthquake. To her it had been the most wonderful experience and she felt very privileged to have been in Leeston at the time to feel it all. All care and no responsibility perhaps but oh what joy she expressed! Each to their own…
The Darfield Earthquake is the most damaging earthquake to strike New Zealand since the 1931 Hawkes Bay Earthquake that so badly affected Napier, Hastings and Wairoa with more than 250 deaths.
The Darfield Earthquake was notable in two other respects: as measured by strong motion instruments, it generated the highest recorded ground accelerations (1.25 times that of gravity) ever recorded from New Zealand, and it also generated a great deal of liquefaction.
In many respects the Darfield Earthquake was similar to the Inangahua Earthquake of 1968: it too measured 7.1, the hypocentre was 10 kilometres deep, there was very strong ground acceleration and considerable liquefaction; furthermore it was complex, perhaps involving movement on more than one fault plane. Interesting!
Previous earthquakes that have adversely affected Christchurch include the following:
1869, June 5: Magnitude less then 6 "Christchurch" earthquake
1870, August 31: less than 5 "Lake Ellesmere" earthquake
1888, September 1: 7.1 "North Canterbury” (Hope Fault)
1929, March 9: 7.1 "Arthur's Pass“
…and more recently:
1987, March 9: 5.2, epicentre located in Pegasus Bay at a depth of about 30 km
2003, September 29: 4.8, also located in Pegasus Bay at a similar depth
2004, September 5: 4.2, south of Amberley at a depth of 37 km.
The Darfield Earthquake will go down in history as a reminder, short-lived no doubt, of the phenomenal amount of energy that can be unleashed within the Earth’s crust in a major earthquake. The aftershocks have been especially sobering and distressing for those living proximal to the Greendale Fault.
So, what part of New Zealand is next in line for a major earthquake?
We cannot say. There is no particular pattern in the ongoing seismic activity that New Zealand is subject to. GeoNet records between 16,000 and 18,000 earthquakes a year, of which about 200 are felt, and only six or so are greater than magnitude 6. Yet despite this large ‘data set’ there is no trend that enables a specific forecast to be made. The entire plate boundary running NE-SW through New Zealand is ‘alive’. This means that we ALL have to be prepared for a major earthquake at any time, especially those of us living within 100 kilometres of the plate boundary.
As geologist at Te Papa, I have often been asked about the Shake House experience in ‘Awesome Forces’ and in particular what size earthquake does it represent and how does it relate to the Darfield Earthquake? What is experienced is a relatively gentle but long wave-length motion that approximates with what the 7.1 magnitude Darfield Earthquake felt like in Dunedin more than 300 kilometres away. So, the Shake House approximates to a large, shallow earthquake some hundreds of kilometres away, say a magnitude 6 at 10 kilometres depth some 150-200 kilometres away. What it feels like all depends on how close you are to the source.
Another common question relates to the energy involved. How much energy was expended in the 7.1 magnitude Darfield Earthquake? This is difficult to express in terms of meaningful everyday human experience. However, I am reliably informed that the largest nuclear device every exploded generated seismic wave records commensurate with a magnitude 7 earthquake. The energy can be calculated in various ways. If it is compared with how much energy flows through the New Zealand electricity grid each year, the Darfield Earthquake probably amounts to about 10 years worth. But now I am guessing! Time to stop.
Acknowledgements
This account has benefited from discussions with a number of my GNS Science colleagues including: John Beavan, Dick Beetham, John Begg, Russ van Dissen, Bill Fry, Caroline Holden, Richard Jongens, Martin Reyners, Warwick Smith, Mark Stirling, Pat Suggate. As expressed above, a great deal of new science will undoubtedly emerge from this event and what is written here is merely an informed interpretation of what happened, a preliminary report. At the time of the earthquake, 37 seismic recording instruments were operating within a 50 kilometre radius of the epicentre. More have been added since. In world terms, this must rate as one of the ‘best’ recorded earthquakes ever!
