Rellie Goddard


On the 25th of April 2015, a Mw1 7.8 earthquake, known as Gorkha, ripped through the Main Himalayan Thrust (MHT) just 77 km north-west of the country’s capital, Kathmandu1. This event propagated a colossal 160 km to the east-south-east, killing 9000 people and resulting in 10 billion US dollars’ worth of damage.  As the dust continues to settle, the question that’s always tempting to ask is “should we have known?” As a structural geologist the question on my lips is more “what’s left to come…?” Here, we take a quick look at the extent to which the Nepalese earthquake was predictable and the likelihood of another event.

First a few things about faults

Figure 1: Map showing plate boundaries and the relative movement of each plate.

The Earth’s surface is divided up into tectonic plates (Figure 1), the movements of which, though influenced by a number of mechanisms, are thought to mainly be driven by the subducting plate dragging the remaining plate in the direction of subduction (a theory known as slab pull). The boundaries between different plates are made up of faults. Faults are approximately planar features in the Earth’s crust that deform via rapid, localised movement. Stress resulting from the plate movements builds on these faults until friction is overcome and the fault fails. An earthquake’s magnitude is proportional to the area of fault that slips, and therefore can be assumed to be proportional to the fault length2. Which makes sense – longer faults, larger earthquakes. India’s subduction under the Tibetan Plateau has led to a long connected fault surface known as the Main Himalayan Thrust (MHT). This fault surface is shown schematically in Figure 3 and continues 2200 km along the Himalaya mountain belt. It was on this fault that the Gorkha earthquake occurred.

‘Hindsight is a wonderful thing’
A huge amount of work has been devoted to investigating the amount of stress that was released during and after the event.  Opinion is generally divided on whether warning signs were apparent preceding the event. Some scientists identified an increase in earthquake activity 3-4 days before Gorkha3. Others claim that no predicable precursory signals were recorded1. Irrespectively, even if the region was more active than normal, one could be forgiven for a lack of concern over any small increase in earthquake activity in one of the most rapidly deforming regions on Earth.

The fact that the rupture area was a known ‘seismic gap’, an area lacking any significant earthquakes, may have made the event more predictable.  Specifically, the area examined here had not had any significant earthquakes in the last 300 years4.  Models can be used to quantify the relationship between an earthquake’s magnitude and its frequency5, therefore allowing one to predict the time interval between events.  As such, it was known that as an earthquake in this area was long overdue.

Inevitable, yes, but predictable? Any earthquake, regardless of its size, has the ability to trigger another earthquake6. As such, magnitude-frequency relationships rarely apply to individual areas under continuous deformation. With a stress system so interconnected and complex, predictions are fairly limited. What can be said is the aftershocks tend to occur on gaps in the fault where movement hasn’t occurred in the initial rupture1. It can be assumed that where there are clear gaps in the fault rupture area, the occurrence and rough magnitude of an aftershock could be predicted.

Figure 2: Cross section showing the portion of the fault that slipped in the 2015 events. The red and blue lines show the areas of the fault that moved during and after Gorkha respectively.

Looking to the future
The majority of the motion during (Figure 2, red line) and following (Figure 2, blue line) the Gorkha event occurred down-dip of the original earthquake nucleation site4. Therefore, the shallow region of the fault, which remains in the frictional breakable section of the crust, has experienced movement neither during or after the event3. Earthquake triggering results when stress is redistributed in the area surrounding an earthquake7. It is sensible then to consider that Gorkha may have loaded the un-ruptured areas of the Main Himalayan Thrust fault, pushing them closer to failure. Likewise in the west, where there has been no rupture during this event and also no earthquake larger than an Mw 7.5 since 1505, there is also an area of increased risk.

The issue remains that earthquakes are fairly unpredictable; an earthquake could occur in Nepal today, tomorrow, and it would be no surprise. Likewise, the stress could continue to accumulate for years or decades to come.  All we can know is that a large amount of stress remains stored in the fault, and that any future earthquakes are likely to occur on the shallower regions. If the next earthquake breaks the surface, and causes a total ‘unzipping’ of the Main Himalayan Thrust (and therefore movement of a large fault surface) the result could be even more disastrous than what has just occurred3.

1. Mw  is a new magnitude scale based on the seismic moment. The seismic moment is the measure of the size of an earthquake based on the average amount of slip and the shear strength of the faulted rock.

 

References

  1. Gualandi, A. et al. Pre- and post-seismic deformation related to the 2015, Mw7.8 Gorkha earthquake, Nepal. Tectonophysics (2016).
  2. Cooke, M. L. & Murphy, S. Assessing the work budget and efficiency of fault systems using mechanical models. J. Geophys. Res. B Solid Earth 109, 1–13 (2004).
  3. Hayes, G. P. & Briggs, R. W. Introduction to the special issue on the 25 April 2015 Mw 7.8 Gorkha (Nepal) earthquake. Tectonophysics 2015–2017 (2016).
  4. Li, Y., Song, X., Shan, X., Qu, C. & Wang, Z. Locking degree and slip rate deficit distribution on MHT fault before 2015 Nepal Mw 7.9 earthquake. J. Asian Earth Sci. 119, 78–86 (2016).
  5. Davidsen, J., Gu, C. & Baiesi, M. Generalized Omori-Utsu law for aftershock sequences in southern California. Geophys. J. Int. 201, 965–978 (2015).
  6. Gu, C., Schumann, A. Y., Baiesi, M. & Davidsen, J. Triggering cascades and statistical properties of aftershocks. J. Geophys. Res. Solid Earth 118, 4278–4295 (2013).
  7. Freed, A. M. Earthquake Triggering By Static, Dynamic, and Postseismic Stress Transfer. Annu. Rev. Earth Planet. Sci. 33, 335–367 (2005).

 “I am a first year PhD student in the Earth Sciences Department researching the partition of stress in multi-mineral rocks. Other than my PhD topic, my interests lie in large scale structural geology, particularly in large magnitude earthquakes and their corresponding faults.”

Advertisements