We use performance cookies to collect information about how you use our website (for instance which pages you visit most often). Cookies help us to improve your online experience with Aspen. Find out more here

close

Aspen Opinion

Seismic Sequences - Unfinished Business?

January 25, 2017

Kirsty Styles PhD, Aspen Re's Earthquake Risk Scientist, considers the recent earthquake sequence in Italy and the predictive capabilities of seismic and satellite data.

The occurrence of the Norcia M6.5 event in the 2016 sequence was not entirely unexpected, but more analysis is needed to understand the relationship between the 2016 sequence and those in 2009 and 1997. Meanwhile, as 2017 begins with four significant quakes in the region, the suggestion that events in 2016 were unfinished business seems prescient.

2016 Shallow seismic sequence

The African Plate continues to collide with the Eurasian Plate as it has done for millions of years and in 2016 Italy once again bore the brunt of the impact. A shallow seismic sequence started shaking central Italy on 24 August with an earthquake of moment magnitude M6.0 striking Amatrice, a small town in the central Apennines region. A M5.4 quake, 15km to the north-northwest (NNW), followed an hour later. Since then, there have been over 40,500 (to 24 December 2016) seismic events along a 60km stretch of the Apennines. Approximately 880 of these events have registered a magnitude of between M3 and M4 and a further 50 have been between M4 and M5. Five events, shown in Figure 1, recorded a magnitude greater than M5 with the largest quake of the sequence striking the town of Norcia on 30 October as a M6.5 event. 

Figure 1: Seismic sequence between 24 August and 24 December 2016. Blue circles denote events post 24 August; red, orange and yellow circles denote events within 72 hour period of 14.00h December 24, 2016

Source: INGVterremoti.com (January16, 2017)

Tectonic complexity

This region of Italy is extremely complex tectonically. Approximately one million years ago, Italy started splitting along the Apennines as a consequence of the collision between the African and Eurasian plates. The section to the east, now part of the Adria microplate, is moving away from the part to the west, which is part of Eurasia (Stein and Sella, 2005). The movement is slow (one to two millimetres a year), but still enough for faults to accumulate strain and therefore generate earthquakes.

In contrast, more than two million years ago, the Adria microplate was moving in the opposite direction forming a subduction zone as it sank beneath Italy from east to west. The convergence of these plates pushed up the Apennines mountain range. Most seismologists believe this subduction activity has now ceased (Anderson and Jackson, 1987), which is evidenced in the regional geology transitioning from convergence to extension. In other words, the Apennines are home to old faults that are typical of compression, or convergence, and also active faulting and earthquake mechanisms that show extensional behaviour as Adria moves away from Eurasia. All the quakes in the 2016 seismic sequence exhibited east-west extensional behaviour.

Geodesy and Geophysics: Analysis of satellite and seismic data

Scientists analysed the 24 August and 26 October quakes using data from Sentinel-1, a constellation of two polar-orbiting satellites from the European Space Agency mission. These satellites can measure ground motion that occurs along a fault by comparing pre and post event images. The analysis showed that the section of crust between the 24 August and 26 October quakes had been stressed or brought closer to failure. The 30 October M6.5 quake filled the gap that had been created by the 24 August and 26 October quakes.[1]  

Scientists also analysed the seismic data collected from the 24 August – 26 October sequence with a particular emphasis on the 24 August M6.0 Amatrice event and calculated the Coulomb stress transferred from the Amatrice event to neighbouring faults.[2] Fault geometry, slip from an earthquake and the coefficient of friction in a region can all be employed to identify Coulomb stress changes on a fault plane and highlight any areas of heightened stress. Findings, shown in Figure 2, identified that the Vetorre-Bove and Gorzano faults have been brought slightly closer to failure by the August quake. Indeed, most of the seismicity since 24 August has been attributed to the activation of this fault which also ruptured during the 30 October M6.5 quake. The fault had been active in the past but, more recently, was considered to be ‘silent’ and so represented a possible seismic gap (Galadini and Galli, 2000). The presence of a gap is not proven to be predictive of future events but simply a gap in the rupture zones of past earthquakes. 

In this instance, satellite and seismic data plus the historical seismic record were combined to suggest where another quake might occur. It is difficult to predict a precise timeframe, but narrowing the location is still helpful.

Figure 2: The Coulomb Failure Function (CFF) on faults surrounding the rupture of 24 August M6.2 Amatrice event. Reduced stress is indicated on the CFF scale by the colour blue with red for increased stress.

Source: INGVterremoti.com (January16, 2017)

Past seismic sequences

This region in Italy is renowned for its seismic sequences and there have been more than 10 damaging earthquake sequences with magnitudes above 4.7 since the 13th century.

In September 1997, a M6.0 earthquake killed 11 people, injured over 100 and destroyed approximately 80,000 homes. It occurred 25 km northwest of the 24 August M6.2 Amatrice event and was part of the Umbria – Marche seismic sequence, which included eight events of M ≥ 5.0 in a two-month period between September and November of that year (USGS).

In April 2009, a M6.3 earthquake occurred 72 km to the southeast of the 26 October M6.1 event near the town of L’Aquila. The quake killed 308, injured over 1,000 and left 55,000 homeless and was also followed by a vigorous aftershock sequence, including five other events of M ≥ 5.0 (USGS).

Linking the sequences

The 1997, 2009 and 2016 sequences are linked but exactly how is still not certain. Seismologists are excited by such sequences of related quakes of medium magnitude that are ongoing over many months. They prompt questions concerning cause and effect of event triggers and distinction between aftershocks and foreshocks. They are also far less common than long stretches of seismic calm punctuated by large quakes and the following smaller and less frequent aftershocks.

Did the 1997 sequence trigger the sequence in 2009? Studies suggest that the 2009 sequence probably triggered the 2016 sequence. Serpelloni et al. (2012) calculated the Coulomb stress transferred to the region by the 2009 L’Aquila quake. This brought the fault that subsequently ruptured during the 24 August M6.0 Amatrice quake closer to failure - albeit only by a small amount (about 0.05 to 0.10 bar). As a general guide, stress increases of at least 0.1 bar are associated with aftershocks and an increased probability of future main-shocks, so the stress change was small but positive.

The fault systems in the central Apennines are relatively young at less than one million years old, short and structurally complex, and contrast with those that have simple continuous features. Therefore, earthquakes in the region are not large by global standards (often M<6.8), but are often shallow and occur in sequences or clusters as stress is transferred from fault to fault.

Filling the gap?

The location of the 2016 earthquake sequence lies predominantly in a gap between the aftershock sequences of the September 1997 M6.0 Umbria – Marche event and the April 2009 M6.3 L’Aquila event (Figure 3). However, this is such a tectonically complex region that it is difficult to know if the seismic gap has been filled by the recent seismic sequence.

The gap between the 1997 and 2009 shocks has a history of moderate quakes so it is not aseismic. Galli, Galadini and Calzoni (2005) found evidence of repeated surface ruptures in the past 20,000 years along the region’s Norcia fault zone. It is still not known whether the 30 October M6.5 quake was the largest and final quake of the 2016 sequence. Collection of additional geophysical and geodetic data from the recent events should reveal whether stress has been transferred elsewhere, and historical data can help determine whether seismic gaps might remain. However, forecasting where and when another M6 quake will strike remains uncertain.

What we do know is that with such uncertainty, we can turn to probability and the most recent probabilistic seismic hazard map of Italy performed excellently. The recent quakes have all struck in the most seismically active part of Italy and the observed shaking was at roughly the maximum level forecast. These expectations are plotted on a probabilistic seismic hazard map that shows the potential hazard from earthquakes within a region based on the combined views of geologists and seismologists. It is probabilistic in the sense that it considers uncertainties in the size and location of earthquakes and the resulting ground motions that can shake a particular site.

Figure 3: Seismicity of the Central Apennines since 1985.

Source:INGVterremoti.com (January16, 2017)

Unfinished business?

Seismic activity in the Apennine region is a complex sequence of related earthquakes, occurring on more than one fault segment. Aftershocks will continue for months and the possibility of similar sized or larger events cannot be ruled out, although the probability of a larger event is low. Analysis of satellite data suggests that deeper parts of the Vettore-Bove fault remain locked (i.e. could still rupture) and so seismic risk remains. Analysis of the seismic data suggests that the Gorzano fault has been stressed by the 2016 events. It was also stressed by the L’Aquila quake in 2009.

References

[1] Apennines earthquakes aftershocks Italy (January 16, 2017)

Bibliography

Anderson H.A., Jackson J.A. The deep seismicity of the Tyrrhenian Sea. Geophys J R Astron Soc 1987; 91:613-637

Galadini F., Galli P., 2000. Active tectonics in the Central Apennines (Italy) - Input data for seismic hazard Assessment. Natural Hazards, 22, 225-270.

Galli P., Galadini F., Calzoni F., 2005. Surface faulting in Norcia (central Italy): a ‘paleoseismological perspective’. Tectonophysics, 403, 117-130. 

Serpelloni E., Anderlini L., and Belardinelli, M.E., (2012) Fault geometry, coseismic-slip distribution and Coulomb stress change associated with the 2009 April 6, Mw 6.3, L’Aquila earthquake from inversion of GPS displacements, Geophys. J. Int., 188, 473–489 doi: 10.1111/j.1365-246X.2011.05279.x

Stein, S. and G. Sella, Pleistocene change from convergence to extension in the Apennines as a consequence of Adria microplate motion, in The Adria Microplate: GPS Geodesy, Tectonics and Hazards, Nato Science Series, 21-34, edited by Pinter, N., G. Grenerczy, J. Weber, S. Stein, and D. Medak, Springer, 2005

Click here for a pdf version of the article.

Back to articles

The above article/opinion reflects the opinion of the author and does not necessarily represent Aspen's views. The article reflects the opinion of the author at the time it was written taking into account market, regulatory and other conditions at the time of writing which may change over time. Aspen does not undertake a duty to update these articles.