Head of Team : Prof. Nanang T. Puspito

Team Members : Dr. David P. Sahara and Dr. Andri Dian Nugraha

Introduction

The phenomenon of earthquake interaction has been studied using static stress transfer techniques [1] The study of stress triggering of earthquakes has shown to be important in the estimation of seismic hazard on major faults throughout the world, e.g. San Andreas, North Anatolian, western Turkey and Greece [2, 3, 4, 5]. In such studies, one of remaining issues is whether a small change in static stress (about hundreds of KPa) caused by an earthquake can really increase or decrease the potential of the next earthquake occurrence [6].

Here we explore how events in the Great Sumatran Fault (GSF) interact, cluster, and propagate each other. We first present a theory of the modeling of Coulomb stress transfer associated with blind and surface cutting arbitrary faults. We then show how these models explain the observed following mainshock distributions along GSF and its surrounding. This is then tested whether this model can be used as an indication of the next mainshock occurrence outside the major GSF segments.

Stress transfer modeling

Earthquake interaction might be explained by static Coulomb stress transfer [2]. The deformation resulting from the earthquake slip will cause stress redistribution around the fault plane. The illustration on how earthquake slip producing stress perturbation is presented in Fig. 1.

If the stress change is increasing, then the area is brought closer to failure; or in other words, the probability of aftershock occurrence is higher on that area. On the other hand, if the stress change is decreasing, then it will inhibit failure [2, 4].

FIGURE 1. Illustration of dislocation theory in a finite homogeneous elastic material (modified from [9] and [10]).

Aceh 2016 earthquake

According to the BMKG press release on Wednesday, Dec 12th 2016, there was a big earthquake with magnitude 6.5 occurring in the Pidie Jaya, Aceh, Indonesia. The hypocenter of the earthquake was located at a depth of around 10 km. Due to its shallow source and close to the big cities, this earthquake caused a great damage to the society; more than 100 and 700 people were killed and wounded, respectively.

We modeled the stress transfer of the previous slip on the segments of GSF within 200 km from the Pidie Jaya. In the past 30 years, there were 26 focal mechanism data of events with magnitude greater than 5.5 available in the Harvard and Global Centroid Moment Tensor (GCMT) catalog. Those events were distributed on four segments of GSF, i.e. Aceh, Seutimeum, Anu Batee and Blangkejeren segments.

Figure 2 shows the modeled stress transfer distribution due in the Pidie Jaya and its surroundings. It can be seen that there is an increasing stress of around 2 Bar at the location of Pidie Jaya 2016 earthquake. According to the previous studies, e.g. Harris [2] and King [11], this is twice of the threshold of the minimum stress increase to trigger next earthquake. This is a strong indication of the increasing seismic occurrence potential in the Pidie Jaya. However, it has to be kept in mind that this information could not be used to infer the occurrence time and magnitude of the triggered event.

FIGURE 2. Static stress transfer modeling due to previous major slips in the GSF segments. 26 events with magnitude greater than 5.5 are included in this modeling. The center coordinate (0, 0) is located at 5.3°N and 96.17°E. The Feb 7th 2013 event is denoted as fault plane number 8 centered at around -15 km X and 35 km Y.

Conclusions

In this paper we introduce the modeling of stress triggering in analyzing the possible triggered event in the surrounding areas of the Great Sumatran Fault. Based on our modeling of stress transfer using a database of focal mechanisms 30 years prior to the Aceh 2016 earthquake, we showed that the February 7th, 2013 event withmagnitude 6.1 caused a significant stress increase in the Pidie Jaya area. The stress increase was twice higher than the triggering threshold. Therefore, we concluded that the 2013 event was responsible for the occurrence of the Pidie Jaya 2016 earthquake.

The finding of the triggered event outside the major segments of GSF due to previous slips at GSF could be applied to analyze the occurrence of other devastated earthquake offset from the GSF segment. In 2017, there were four devastated earthquakes occurring outside the major segment of GSF. Furthermore, the procedure of stress transfer modeling and its application in inferring the probability of the next earthquake occurrence could be applied also to predict and anticipate the occurrence of the major event within and outside the GSF segments.

References

1. L. Hardebeck, J. J. Nazareth and E. Hauksson, J. Geophys. Res. Solid Earth 103, 24427–24437 (1998).
2. A. Harris, Curr. Sci. 79, 1215–1225 (2000).
3. Karakostas, E. Papadimitriou and D. Gospodinov, Geophys. J. Int. 197, 525–541 (2014).
4. G. C. P. King, R. S. Stein and J. Lin, Bull. Seismol. Soc. Am. 84, 935–953 (1994).
5. Lin and R. S. Stein, J. Geophys. Res. Solid Earth 109, B02303 (2004).
6. A. Harris, J. Geophys. Res. Solid Earth 103, 24347–24358 (1998).
7. Kanamori, Annu. Rev. Earth Planet Sci. 22, 207–237 (1994).
8. J.K. Gardner and L. Knopoff, Bull. Seismol. Soc. Am. 64, 1363–1367 (1974).
9. Okada, Bull. Seismol. Soc. Am. 82, 1018–1040 (1992).
10. Okada, Bull. Seismol. Soc. Am. 75, 1135–1154 (1985).
11. S. King, Int. J. Rock Mech. Min. Sci. Geomech. Abstr. 20, 237–241 (1983).
12. W. Lin, E.-H. Yeh, J.-H. Hung, B. Haimson and T. Hirono, Tectonophysics 482, 82–91 (2010).