When a magnitude 6.3 earthquake scored an almost direct hit on Christchurch in February, I discussed the possibility that rather than being a simple aftershock of example of the earlier quake triggering activity on an already loaded adjacent fault. However, although I thought that this explanation was more consistent with the wider tectonic picture, it was a hypothesis – an informed guess – rather than a definitive statement. Without much better knowledge of the structures involved, and how much stress they had accumulated prior to the Darfield earthquake, it is difficult to say anything more definitive. So when Twitter pal shortstack81 pointed me to a recently published paper in Nature’s online Open Access journal, Scientific Reports, which concludes “that the Darfield earthquake contributed to promote the rupture of the Christchurch fault [their name for the fault that ruptured in February – NZ geoscientists seem to commonly refer to it as the Port Hills Fault]”, I had to take a closer look. Unfortunately, I’m not sure this paper provides any more information than we had already about the linkage between the two earthquakes.
Italian seismologist Salvatore Stramondo and his colleagues mapped the deformation of the landscape around Christchurch in response to the Darfield earthquake in September 2010 by comparing before and after radar images, and calculated how stress was redistributed in the crust as a result of this deformation. Their calculations indicate that some of the largest stress changes occurred close to the rupture point of the Port Hills earthquake, and that these stress changes would have pushed the fault that generated it closer to failure. Whilst this is an interesting observation, apart from the technical details of the model used, this is not exactly a new result. Very similar stress calculations had already been made even before the Port Hills earthquake, and I included a figure that showed a stress increase near the rupture (from here) in my original post about the Port Hills earthquake in February.
In order to really nail causes-and-effects, though, we need more than this. Essentially, there are three possible hypotheses about the link between the Darfield earthquake and the Port Hills earthquake five months later:
- There is no link: the Port Hills earthquake would have happened as it did regardless of whether or not the Darfield earthquake had happened.
- There was triggering: the Port Hills earthquake would have occurred at some point, but the Darfield earthquake altered its timing
- The Port Hills earthquake was an aftershock: it would not have occurred at all without the stress change imparted by the Darfield earthquake.
The fact that the Darfield earthquake did change the stress in the crust around the Port Hills rupture probably rules out hypothesis 1, but as I just said, we knew that already. The more important problem is distinguishing between hypotheses 2 and 3 – and this new paper does not really try to do that. Their modelling assumes there was a previously loaded fault just south of Christchurch and then calculates that the stress changes caused by the Darfield earthquake would have pushed it closer to failure. At best, this confirms that triggering is a plausible hypothesis, but arguably we knew that already as well: the idea that stress changes due to deformation following an earthquake can significantly increase (or decrease) the likelihood of earthquakes on adjacent faults is pretty well established. But it being purely an aftershock is also (just about) plausible, based on the statistical expectations of aftershock behaviour after a large earthquake.
It seems to me that what we really need is modelling that shows that the stress changes resulting from the Darfield earthquake were insufficient to produce a magnitude 6.3 aftershock. But sadly we are lacking a few important bits of knowledge before we could run such a test: the detailed subsurface structure of the faults near Christchurch, an idea of how much strain was loaded across them prior to the Darfield earthquake… and, oh yes, an accurate physical model of how faults fail in earthquakes. Given such formidable gaps in our knowledge, you can hardly blame the authors of this paper for not being able to prove triggering or not; but the fact remains that despite its publication, we’re left pretty much where we were in March.
For more informed commentary on this paper, New Zealand’s Science Media Centre has collected a number of responses from Kiwi geoscientists.
Stramondo, S., Kyriakopoulos, C., Bignami, C., Chini, M., Melini, D., Moro, M., Picchiani, M., Saroli, M., & Boschi, E. (2011). Did the September 2010 (Darfield) earthquake trigger the February 2011 (Christchurch) event? Scientific Reports, 1 DOI: 10.1038/srep00098