Earthquakes are unpredictable and can devastate communities, causing large-scale economic damage. Salvatore Iacoletti, Science Specialist at 色多多视频 explains how a scientific exploration of earthquake activity can help risk professionals to be better prepared for these events.
In February, Turkey and Syria were hit by a series of major earthquakes. The first was a magnitude 7.8 earthquake located on the East Anatolian fault. This earthquake triggered a sequence of more than 130 aftershocks with a magnitude above 4.5 (to date). The largest aftershock was a magnitude 7.5 earthquake, which occurred only nine hours after the mainshock around 100 kilometers north.
Before the magnitude 7.8 mainshock, southern Turkey and northern Syria had only experienced three magnitude 6 or larger earthquakes since 1970 (with the largest being a magnitude 6.7 event). Despite the relative seismic quiescence of this region, historical records and geological studies show that this region has experienced significant and damaging earthquakes in the distant past. Aleppo (Syria), for example, was devastated several times historically by large earthquakes (two magnitude 7+ earthquakes in 1138 and 1822).
The 2023 Turkey-Syria earthquakes struck densely populated areas resulting in the tragic loss of life of around 57,000 people. Thousands of buildings collapsed, hospitals were damaged and cold weather conditions hampered rescue efforts. Property damage alone is estimated at over US$100B.
These devastating events underscore the fact that while large earthquakes are rare, they can cause enormous damage, and risk managers should always be prepared.
Unlike other natural disasters, such as hurricanes or volcanos, earthquakes are more unpredictable by nature. This also makes it challenging to model seismic activity and its potential impact. Our science team work to enable us to examine past data to better understand this peril so we can help clients put in place risk mitigation measures.
It’s imperative to quantify past years’ earthquake impacts to ensure that expectations are not artificially lowered and that risk management experts can be as prepared as possible for the potentially disastrous effects of this peril.
Activity in 2022
Worldwide, the year 2022 saw fewer large earthquakes – those with a magnitude greater than or equal to 7.0 – than the 10-year average, while the overall number of earthquakes with a magnitude greater than or equal to 4.5 was broadly similar to the 10-year average.
In 2022, there were 11 earthquakes with a magnitude of 7.0 or greater – the lowest annual number since 2011 – and 125 with a magnitude of 6.0 or higher.
Figure 1 shows the cumulative number of earthquakes by magnitude in 2022 compared with the 2011-2021 average.
Figure 1. Magnitude-frequency distributions for 2022, average and minimum-maximum range
of all years between 2011 and 2021
For risk management, this sounds like it should be positive news. But the size and impact of losses depend on the location as well as the magnitude of the earthquakes. Even a magnitude 6 event can cause considerable losses if it strikes in a densely populated and/or industrial area.
Figure 2 shows a worldwide map of 2022 earthquake locations. The five largest earthquakes were:
- M7.6 (27km depth) Mexico, 2022-09-19 – which caused moderate damage across the region (a in Figure 2)
- M7.6 (116km depth) Papua New Guinea, 22-09-10 – which resulted in some damaged buildings and landslides (b in Figure 2)
- M7.3 (25km depth) Tonga, 2022-11-11 – which occurred offshore, no reported damage (c in Figure 2)
- M7.3 (41km depth) Japan, 2022-03-16 – which caused moderate damage across the region (d in Figure 2)
- M7.2 (236km depth) Peru, 2022-05-26 – which was too deep to cause considerable damage (e in Figure 2)
Overall, 2022 experienced fewer large-magnitude events than the 2011-2021 average, and the largest earthquakes that did occur were in areas of limited exposure and caused a limited amount of damage.
Figure 2. Worldwide map of 2022 earthquakes
Earthquake activity’s impact on losses
While earthquake frequency last year was broadly in line with the 10-year average, and the number of large earthquakes was lower than the 10-year average, to understand the loss impact of earthquakes in 2022 compared with previous years, we would need to conduct detailed portfolio risk calculations.
For the purpose of this article, we use a simplified Risk Index that only takes into account (a) the replacement cost data from the Global Exposure Database provided by the Global Earthquake Model foundation (i.e., not specific to the insurance industry), and (b) the total number of events with a magnitude greater than or equal to 4.5. The Risk Index used here is a crude metric that does not, for example, account for damage susceptibility and vulnerability. But it can give a good sense of how often there’s been a risk of incurring a significant loss at a specific location, per year.
The Risk Index increases – for example – for countries with high Gross Domestic Product (GDP) compared with low-GDP countries, or in areas where there is a large concentration of buildings, or in more seismically active areas. In other words, the Risk Index will be high when there is a considerable loss potential at a specific location or because many earthquakes have been observed in the area.
Figures 3 and 4 show the Risk Index for 2022 and the average Risk Index for 2011-2021, respectively. The Risk Index for 2022 is generally higher in Japan, Southern Europe, Central America, South-East Asia, New Zealand, and the West Coast of North America. The spatial distribution of the Risk Index is broadly consistent with the previous 10-year average.
On the other hand, Figure 5 compares the Risk Index values of 2022 against the average values from 2011 to 2021, showing that the Risk Index values in 2022 are slightly lower than the 10-year average. This confirms that 2022 was an exceptionally mild year in terms of global seismic loss occurrence when considering the previous 10 years.
The Risk Index's spatial distribution in Figure 4 is sparser than in Figure 3. This demonstrates that even though we did not observe major losses in less seismically active areas in 2022, the 10-year average shows that such losses can happen. One striking example is Central USA: Figure 3 shows mostly zero Risk Index values, while non-zero Risk Index values extend as far as Texas in Figure 4.
In less seismically active areas, where people have not experienced large earthquakes in years, it is almost natural to become unprepared for such rare events. But risk managers should always remember that earthquakes can cause significant losses when least expected – the 2010 Canterbury earthquakes in New Zealand being an example of this.
Figure 3. Risk Index for 2022
Figure 4. Mean Risk Index for years between 2011 and 2021
Figure 5. Comparison between the Risk Index for year 2022 and the average Risk Index in 2011 to 2021
What-if analyses
While 2022 was a lower-than-average year in terms of global seismicity and its impact on properties, there are plenty of examples where events could have taken a much worse turn. More earthquakes could have occurred and/or the earthquakes that did happen could have occurred in more exposed areas.
Figure 6 shows a map of the 2022 earthquakes that occurred in North America. The biggest event was the M7.6 earthquake near the coast of Mexico. This event caused moderate damage across the region, mainly because it did not hit a major city. But if this magnitude 7.6 earthquake had occurred only 150 kilometres north-east of its actual epicentre, it would have been as close to Mexico City as the 2017 M7.1 Puebla earthquake, which caused extensive damage and claimed the lives of 370 people.
Figure 6. 2022 earthquakes occurred in North America
Figure 7 shows a summary of the M4.5+ earthquake activity that occurred in 2022 around Taiwan: the right panel displays the map of the earthquakes, the upper-left panel shows the cumulative number of earthquakes (in time), and the lower-left panel shows the magnitude of the events.
Figure 7. 2022 earthquakes around Taiwan. Taipei is highlighted on the map with a red arrow
The two largest earthquakes were the M6.7 in March and the M6.9 in September. After both events, at least two other M5.9+ aftershocks occurred between a few hours to a few months after the respective big shocks. This sequence of events comprising six earthquakes of magnitude 5.9 or greater occurred some way away from a populated zone and so did not garner much media attention.
But the characteristics of this sequence are very similar to the much-remembered 2010 Christchurch and Canterbury events in New Zealand that caused billions of dollars in insured losses. This is confirmation that Canterbury-like sequences happen more often than people think.
As the terrible events of February 2023 remind us, earthquakes can happen at any time causing devastation to families, communities and economies. Rigorous application of building codes, frequent testing of resilience and rescue plans, and heightened risk awareness and understanding must always be top of mind for risk professionals, even when large and damaging earthquakes have not been experienced for some time.
