Professor KEVIN P FURLONG, a geoscientist, asks: What's the deal with all these aftershocks?
OPINION: Cantabrians have become experts in detecting the subtle differences among earthquakes. But such expertise does not bring with it much comfort in light of continuing events.
Since the September 4 earthquake, we have experienced hundreds of felt earthquakes within a total of thousands of aftershocks following the main event.
Throughout this entire episode, the earthquake-science community continues to reassure the public that there is nothing strange or out of the ordinary about the aftershocks. But for many, aftershocks remain a real source of personal concern.
Perhaps it is now useful to step back and explain why aftershocks occur, what controls their locations and sizes, and why the current spate of events is not a cause for alarm, but rather a reminder of the value in maintaining vigilance and preparedness.
Aftershocks are just earthquakes; they obey the same laws of physics as other earthquakes.
We use the term aftershocks to differentiate them, since we consider them to have been caused or encouraged by the initial main- shock earthquake.
In other words, if the main shock hadn't occurred we wouldn't be getting the aftershocks.
But not all aftershocks are the same. In general there are two main categories of aftershocks - ones that occur on or very near the fault plane of the main earthquake, and others that occur in areas where earthquake-caused crustal stress changes help trigger these small earthquakes.
The main magnitude 7.1 earthquake occurred primarily on a nearly-vertical fault plane extending about 10 to 15 kilometres downward into the crust and perhaps as much as 50km along its east-west length.
About 30km of this fault length ruptured all of the way to the surface, producing the impressive fault we see extending from west of Greendale to the western outskirts of Rolleston.
That 30km-long surface fault is only part of the story, and geophysical evidence suggests that the fault continues in bedrock beyond the ends of the surface rupture.
This continuation of the fault at depth is typical of many earthquake ruptures; in fact in the Haiti earthquake there was little or no surface rupture.
During our main earthquake, there was slip along the fault that averaged about 2.5 to 3 metres, but the slip during the earthquake was different at different places along the fault, so there were patches along the fault that were not in equilibrium with nearby fault patches.
Many of the aftershocks, particularly those occurring during the first few weeks after the main earthquake and that were situated on or near the main fault, result from the processes by which these slip differences are reduced or smoothed.
Their occurrence near the mapped fault trace and in regions such as the swath from Rolleston to Lincoln record these fault equilibration processes at work.
The size or magnitude of an earthquake is proportional to the product of the fault area multiplied by the distance it slipped.
The main magnitude 7.1 earthquake with a fault area of about 500 square km (50km length x 10km depth extent) and an average slip of 2.5m to 3m dwarfs a magnitude 5 aftershock with slip of about 10cm on a fault patch about 3km by 3km, or a magnitude 4 aftershock rupturing a 1 square km patch with slip of only 5cm.
Most of the aftershocks this past month record the second category of earthquakes triggered by stress changes.
These earthquakes, including the notorious Boxing Day event and the 5.1 on January 20 did not occur on the September 4 fault; rather they are within the crust of the surrounding region where stresses changed enough to cause earthquakes on other small faults.
The spatial pattern of these triggered aftershocks is not random.
Models of the stress changes caused by the main earthquake can be used to identify where changes are sufficient to either favour or inhibit aftershocks.
These stress models show that the occurrence of aftershocks in the region of Christchurch city and east of the CBD, north of the main fault trace near Darfield and further north, and west of Horarata are all in places where stress changes favour increased aftershocks.
These triggered aftershocks occur on existing structures or zones of weakness in the crust, and are often aligned along linear trends.
Maps of Canterbury earthquakes since September 4 show several such trends. In this way the aftershocks help scientists to map the fabric of the crust and identify trends and locations that merit further study.
The why, where, and when of aftershocks since the September 4, 2010, earthquake is consistent with the expectations and experiences of earthquake scientists.
Most fall neatly into the two likely causes - either adjustments on and near the main fault rupture or small triggered events in regions where stresses increased slightly as a result of the main earthquake.
Therefore when a small cluster of events occurs in any particular location, we understand their origin in relation to the stress resulting from the main event and they are unlikely to indicate that another large event on the Greendale Fault is pending.
The patterns of these aftershock earthquakes do, however, help to improve our understanding of fault rupture during the main event, and to identify other active structures within the Canterbury crust.
Perhaps most importantly, they remind us that preparations and continued vigilance are an important responsibility of living in earthquake country.
* Kevin P Furlong is a Visiting Erskine Fellow at the University of Canterbury and Professor of Geosciences at Pennsylvania State University in the United States.
- The Press