OFFICE OF THE PRIME MINISTER’S SCIENCE ADVISORY COMMITTEE
The Canterbury Earthquakes:
Scientific answers to critical questions
The Canterbury region has had six months of unexpected and extremely difficult challenges as a result of a sequence of damaging and deadly earthquakes and the associated aftershocks. The result is significant uncertainty within the public about why the February 22 aftershock was so damaging and deadly compared to the larger magnitude event on September 4, 2010. In order to provide information on these and other questions, the Royal Society of New Zealand, together with the Office of the Prime Minister’s Science Advisory Committee, has convened a group of earthquake science experts including those from the Natural Hazards Research Platform, a collaborative research consortia hosted by GNS Science, to provide answers to critical questions based on current best scientific information and knowledge.
An enormous scientific effort is underway but it is difficult to define all of the critical attributes of an earthquake quickly after the event. As a result, our understanding of these earthquakes will improve and much more information will become available to be used in further scientific analyses and advice in the weeks and months ahead. The information below is provided for the purpose of improved understanding of the science relating to the Canterbury earthquakes.
Why didn’t scientists know about the faults that caused the two earthquakes?
Prior to September 4th, there were no surface signs of the Greendale Fault or the fault that generated the Lyttelton aftershock and there was no evidence for seismicity on these faults (i.e. ‘foreshocks’). Seismic surveys have located some ‘hidden’ faults across parts of the Canterbury Plains, but these particular regions had not been surveyed for this purpose. An oil-gas seismic survey had been carried out but did not reveal any convincing evidence for the presence of the Greendale Fault. Following September 4th, there was significant aftershock activity in the area of the Lyttelton Fault and around many faults in the region but there was no clear indication that a larger earthquake was imminent there.
It was predicted that aftershocks from the September 2010 earthquake might reach magnitude 6, and some smaller aftershocks had already occurred under Christchurch city.
Why wasn’t some warning given about the possibility of a big and damaging aftershock under the city?
Warnings were given over the risks from large aftershocks. The prediction of aftershocks of approximately magnitude 6 is based on statistical analysis of historical earthquakes (Bath’s Law) which states “the average difference in magnitude between a mainshock and its largest aftershock is 1.2, regardless of the mainshock magnitude”. A quick survey of some of New Zealand’s largest historical earthquakes conforms to this average, although there is significant variability. The 6.3 aftershock is not outside the average range. The isolated and smaller aftershocks that occurred under the city CBD do not necessitate the presence of a larger fault capable of generating larger earthquakes, however this is possible. Seismic and aerial surveys are targeting this area in the near future to provide constraints on the geometry, extent, and magnitude potential of a fault under the city.
Why did the February aftershock occur so long after the September 2010 earthquake ?
It is not unusual to have a 6 month gap between a magnitude 7.1 earthquake and a 6.3 aftershock.
How predictable was the 22 February quake - and what was the influence of the moon and the tides; do more earthquakes happen at night?
Despite substantial scientific effort, the specific timing, location, or magnitude of earthquakes cannot be predicted. Although it is not possible to reliably predict individual earthquakes, it is possible and routine to identify areas and times of higher or lower earthquake activity based on models of crustal stresses and faulting. In addition, on the basis of observations of past earthquakes, it is possible to expect a series of aftershocks after an earthquake which follows a general and decreasing pattern, but there is always some level of unpredictability as to their timing, location and severity. For this reason, it is not possible to give specific predictions about aftershocks in terms of severity, location and timing.
Recent popularized ‘predictions’ regarding the Canterbury earthquakes and their relationship to phases of the moon are not scientifically correct. Although there is some evidence that the moon can sometimes influence very small earth tremors, there is no credible evidence linking the moon to medium or large earthquakes.
There is no relationship between the time of day and the frequency of earthquakes. More earthquakes will be felt at night only because people are in bed or at rest and thus have more contact with the ground, making them more sensitive to feeling the ground movement during earthquakes.
Are the recent natural disasters in Queensland, Japan and New Zealand linked in any way?
No. The Canterbury earthquakes are in no way related to the flooding events in Queensland or the massive earthquake in Japan.
Figure 2: Christchurch Earthquake Population Exposure showing the extent of earthquake ground shaking (represented in colour) overlain on population density (represented as height of vertical bars) at a grid size of 1 km2. The colour key is based on the Mercalli scale, which is a measure of shaking, and explains how sever the ground shaking is for each colour. (Source: USGS)
Figure 3a and 3b: Comparisons of the ground accelerations in the Christchurch CBD between the 2011 Christchurch and 2010 Darfield earthquake. Left to right, the recordings are arranged roughly in increasing distance from the epicenter of the earthquake on 22 February 2011
Was the Christchurch magnitude 6.3 earthquake an aftershock?
Yes. The term ‘aftershock’ refers to an earthquake that is smaller in magnitude than the preceding main shock, part of the sequence of earthquakes that closely follows the main shock in time, and in the region influenced by changes in crustal stress levels due to the main shock. The M 6.3 earthquake fits all of these criteria and is therefore considered an aftershock.
What is liquefaction and what causes it?
Liquefaction is the term used to describe when some soils behave more like a liquid than a solid during the shaking from an earthquake. During an earthquake the soil particles are rearranged and compacted, forcing out water onto the surface, creating sand volcanoes or sand boils, water fountains and surface cracking (See Figure 3). Along with liquefaction, lateral spread can also occur, which is when the liquefied soil flows in to lower areas such as river channels, under the force of gravity.
Figure 4: A representation of the liquefaction process (source: Environment Canterbury)
After liquefaction, does the ground ever stabilise again?
The risks from liquefaction will remain. Strength-depth profiles under some parts of Christchurch indicate up to 9m of ‘liquefiable’ material. Immediately following some of the largest aftershocks from the 7.1 earthquake, liquefaction reappeared in the same areas. During the 6.3 earthquake, liquefaction was widespread and vents continued to surge during the aftershocks immediately following this event. Although some ground settlement may occur, the large reservoir of liquefiable material and these examples suggest that similar characteristics of ground shaking are likely to result in similar amounts of liquefaction in the future.
Have new faults appeared under Christchurch after these earthquakes?
It is not yet known. All aftershocks are earthquakes and most earthquakes occur on faults. There has been some aftershock activity under Christchurch, however this activity has occurred on small faults that may not be connected. In order to better understand the extent of faults under the city, further seismic studies are required.
Are the faults around Christchurch a southern extension of the Marlborough Fault Zone?
No. The Marlborough Fault Zone is a system of four major and many minor faults that link the offshore plate boundary located east of the North Island, to the Alpine Fault along the West Coast, South Island.
Will there be another damaging earthquake in Christchurch in the future, linked to these quakes?
The pattern of aftershocks so far (which is fairly typical) suggests that a magnitude 5 aftershock or greater is still likely during the time that the aftershocks continue to decline in intensity and frequency. The exact timing and location cannot be predicted. What damage such an aftershock might cause depends critically on location and depth of the earthquake, just as it did for the February event. There are some indications that the aftershock locations are moving eastwards away from Christchurch.
How long will the current sequence of aftershocks go on for?
The frequency of aftershocks will generally decrease over time, and earthquakes that can be felt will become less common. After 1 year from now the rate of earthquakes that people can feel may be on the order of one per month.
Is the next one likely to happen at sea, and could there be a tsunami?
A major tsunami is very unlikely. Calculations of the possible tsunami that may occur if the magnitude 7.1 Sept 4th earthquake were to occur offshore in the future indicate the hazard is mostly restricted to beach areas and estuaries. The best way to be prepared for any tsunami that may follow such an earthquake is to move away from the beach or other low lying areas as soon as the earthquake shaking is over, which might last for 20-30 seconds.
A number of aftershocks associated with the September 2010 earthquake and the 22 February 2011 earthquake have occurred offshore to the east of Christchurch, and further aftershock activity of a generally declining nature in that region is expected. We know that to generate a significant tsunami generally requires earthquakes larger than magnitude 7.5, and a substantial vertical displacement of the sea floor.
Will these earthquakes trigger an eruption of the Banks Peninsula volcano, and are other volcanic areas like Timaru, Oamaru and Dunedin likely to get earthquakes like this?
No. The last time the Banks Peninsula volcanoes erupted was ~5.8 million years ago. There is no evidence of magma remaining beneath Banks Peninsula. Hot springs are common in non-volcanic regions throughout the South Island but this does not indicate renewed volcanic activity. There is no link between these geologically ancient events and the current patterns of earthquakes in Canterbury.
Are the recent earthquakes in Wellington caused by the Christchurch ones?
No. In 2009, New Zealand experienced over 500 earthquakes larger than magnitude 4.0. In a typical year, magnitude 4.0-4.5 earthquakes occur, on average, at least once per day3 . The magnitude 4.5 and 4.7 earthquakes that occurred near the Kapiti coast and beneath Upper Hutt are typical of previous earthquakes in the area. Earthquakes in the Wellington area are part of a different segment of the plate boundary through New Zealand. Thus the current Canterbury earthquakes are in no way related to the faults of the Wellington region and in no way increase the likelihood of earthquakes there, or anywhere else other than Canterbury.
Do these earthquakes mean that New Zealand is in a more active phase of seismic activity and there will be more earthquakes in other places in New Zealand?
There is no evidence that the rate of earthquake activity in New Zealand or anywhere else in the world is increasing. More than 100 earthquakes of equal or larger size to the Christchurch event occur each year around the globe. The rate of this activity has not changed over decades.
Is the big earthquake predicted for Wellington now more likely to happen in the near future?
No. The effects of the Canterbury earthquakes in changing the stresses in the New Zealand crust do not extend as far as Wellington – the changes are contained within the Canterbury region. Thus the likelihood of a major earthquake in the Wellington region has neither increased nor decreased as a result of these events.
Which city is safest to live in to avoid another earthquake – Christchurch, Wellington, Hamilton or Auckland? What about other risks in those cities – volcanoes, cyclones, tsunamis?
All regions of New Zealand are vulnerable to natural hazards. For some locations it is earthquakes, or volcanic activity; for other locations it may be flooding, or tsunami. Although some locations may have lesser hazard to local earthquakes, a large earthquake along the plate boundary faults in either the North or South Islands would cause strong ground shaking over large areas including all of the major cities in New Zealand.
Does every year that the Alpine Fault does not rupture make the risk in a subsequent year higher and does every year it does not happen increase the severity of the earthquake when it happens?
The Alpine Fault is significantly longer than any of the faults in the Canterbury region, and thus it can produce substantially larger earthquakes (see Figure 5). Current estimates are that an earthquake of magnitude 8.0 or larger would be possible, if the entire Alpine Fault were to rupture in a single earthquake. Given its distance from Christchurch, its impact is unlikely to be greater than the M7.1 earthquake. The recurrence rate for major Alpine Fault earthquakes is between 150-500 years, with the last major event inferred to be in 17174, and thus an event in the next 40 years is considered to have a high probability.5 This probability does not systematically increase each year and the severity of the earthquake is related to the length of the fault rather than the duration between quakes.