Large, distant earthquakes can trigger microseismicity in Costa Rica

For the first time in Costa Rica, scientists document that earthquakes can communicate with faults, even at a distance of more than a thousand kilometers.
 

By Sonia Hajaji and Esteban J. Chaves, Ph.D., Volcanological and Seismological Observatory of Costa Rica, OVSICORI, Universidad Nacional
 

Citation: Hajaji, S., and Chaves, Esteban J., 2024, Large, distant earthquakes can trigger microseismicity in Costa Rica, Temblor, http://doi.org/10.32858/temblor.343
 

When large earthquakes strike, smaller magnitude events called aftershocks occur — but why? Aftershocks are a response to an increase in static stresses induced by sudden slip during a mainshock. These smaller events generally illuminate the rupture area of the main event, and thus tend to occur within a range of about one to two fault lengths. For instance, if the rupture length is 100 kilometers, then aftershocks most often occur within a radius of 100 to 200 kilometers. Such aftershock behavior has been very well characterized by seismologists, with examples from recent events in Türkiye and Japan.

At greater distances — hundreds to thousands of kilometers away from the mainshock — triggering of smaller shocks is rare (Gomberg and Johnson, 2005). Nevertheless, for cases in which a relationship can be shown between mainshock and distant earthquakes, seismologists refer to this process as “dynamic triggering.” In this scenario, surface waves generated by a distant (or teleseismic) event can perturb the strain state along faults, speeding up their loading cycle (or the time during which elastic energy, to be released as a future earthquake, accrues on the fault) and bringing distant faults to failure (Brodsky and van der Elst, 2014). The first — and quite remarkable — documented example of dynamic triggering was produced by the 1992 magnitude 7.3 earthquake in Landers, California. This event triggered seismicity in the eastern United States just minutes after its surface waves passed (Hill et al., 1993).

Although the underlying physics that governs dynamic triggering of seismicity is not well understood, results from numerical simulations and laboratory experiments have highlighted several processes (e.g., permeability enhancement, granular flow, viscous creep, rate and state friction and others) as plausible mechanisms driving the triggering process. In one proposed mechanism, fault heterogeneities (also called asperities) may focus transient energy from surface waves, thus inducing local stress perturbations (Langer et al., 2015). In other words, as energy from surface waves passes by a bump on a fault, the local stress can be changed.

Scientist have observed and documented dynamic triggering in multiple regions all over the world, from Southern California (e.g., Aiken and Peng, 2014; Fan et al., 2020) to Northern China (Peng et al., 2010), and even Antarctica (Peng et al., 2014). Overall, data show that surface waves are responsible for triggering seismicity at great distances. Among the surface waves, Rayleigh waves are more efficient at triggering than Love waves. Regions experiencing extension, like geothermal fields, are more prone to dynamic triggering than regions experiencing compression, like subduction zones (Brodsky and van der Elst, 2014).

During the passing of surface waves, locally triggered earthquakes generally occur instantly or within a few minutes after their arrival. The subsequent cascade of microseismicity (small magnitude seismicity) may last for hours or even days. In Costa Rica, the dynamic triggering of earthquakes by distant events has not been documented — until now.
 

Evidence of dynamic triggering in Costa Rica

For 25 earthquake events, each of which occurred hundreds to thousands of kilometers away from Costa Rica, we investigated whether the passing of surface waves subsequently triggered local microseismicity. We focused on mainshock events with magnitudes greater than or equal to 7.5 (Figure 1). We presented our study at the 2023 American Geophysical Union Annual Meeting in San Francisco, California.

In our investigation, we evaluated the seismic records of all stations operated by the Volcanological and Seismological Observatory of Costa Rica at Universidad Nacional (OVSICORI-UNA) between 2010 (when the digital era of the observatory began) and 2023.
 

 

Among the 25 analyzed events, we noticed an increase in the local microseismicity following events 14 and 24. Event 14 corresponds to a magnitude 7.5 earthquake that struck in the Swan Islands of Honduras on Jan. 10, 2018 (Figure 1). Event 24 corresponds to a magnitude 7.8 earthquake that occurred along the border of Türkiye and Syria on Feb. 6, 2023 (Figure 1). Although some events were larger or closer to Costa Rica, seismic data (both waveforms and earthquake catalogs) from stations available during these events showed no evidence of instantaneous or delayed triggering of microseismicity. Therefore, the question of why larger and closer teleseismic events did not trigger seismicity in Costa Rica is an avenue for future research.

The Honduras event triggered microseismicity that was focused along the Tenorio Volcano of northern Costa Rica, generating events with magnitudes between 0 and 3.0. The Türkiye-Syria earthquake last year resulted in microseismicity with magnitudes between 0 and 4.4 along the subduction zone in southern Costa Rica immediately after the event (Figure 2).

We quantified the sudden increase in local earthquakes following three approaches: 1) visual inspection in the time and frequency domain, 2) computing the  statistic, as done in previous studies (e.g., Fan et al., 2020) and 3) evaluating the probabilistic power spectral densities, or PPSDs (MacNamara and Bulland, 2004) for all stations available in Costa Rica hours before and after the passing of the regional and teleseismic surface waves.
 

 

The visual inspection

In the data from Costa Rica’s seismic network, signals from nearby dynamically triggered events are generally hidden in large amplitude waves produced by the large magnitude teleseismic event. The magnitude of the subsequently triggered microseismicity oscillates between 1.0 and 4.0. Thus, these small earthquakes are rich in high frequencies but low amplitudes. These signals contrast with teleseismic events’ long period, high amplitude signals. Indeed, the high frequency signals produced by large mainshocks rapidly attenuate, or fade away, and cannot be measured hundreds to thousands of kilometers from the source. Earth essentially works as a low pass filter.

Thus, for all events analyzed in this study (Figure 1), we filtered the signals. We kept only high frequencies generated by local seismicity, which let us look for the presence of nearby earthquakes that correlate temporally with the arrival times of the surface waves. Figure 3 shows an example of this analysis for the Türkiye-Syria earthquake. We focused on the vertical components recorded by three different stations (COVE, PEZE and DUNO shown in Figure 2). The triggering is most evident in the frequency domain, as highlighted by the “sparks” or bright vertical lines in panels c, g and k (Figure 3).
 

 

Computing the β statistic

Another way to identify dynamic triggering is to robustly characterize seismicity rate changes (how many earthquakes occur in a given amount of time), which can be accomplished by various statistical approaches. Of these approaches, the β-statistic is the most widely used and measures the difference between the observed versus expected number of earthquakes relative to the expected variability of number of earthquakes during that period (e.g., Aiken et al., 2018).

Thus, β is positive when there is an increase in seismicity rate compared to background levels, and it is negative when there is a decrease. β greater than or equal to two is typically considered an indicator of a seismicity increase at the 95% significance level (e.g., Aiken et al., 2018).

By computing β, we made two important observations. First, β exceeds values of 25 and 50, correlated temporally with the occurrence of the Honduras earthquake and the Türkiye-Syria earthquake, respectively (Figure 4). This lends further support that Costa Rican microseismicity was dynamically triggered by these large magnitude events located at regional and teleseismic distances. Second, a β of two is a very low value for a region with a high background seismicity rate, like Costa Rica. This is particularly evident in Figure 4b (green dashed line at β=2).
 

 

Probabilistic Power Spectral Density

The Probabilistic Power Spectral Density (PPSD) is a concept used in signal processing, seismology, and other fields. It extends the traditional Power Spectral Density (PSD) analysis by incorporating the probabilistic distribution of spectral amplitudes. The PPSD shows how the power of a signal is distributed over different frequencies and offers a more detailed view by accounting for the variability of the signal’s power over time.

We used this concept of PPSD to analyze the seismic records from all OVSICORI’s seismic stations that were available hours before and after the occurrence of the 2018 magnitude 7.5 Honduras earthquake and the 2023 magnitude 7.8 Türkiye earthquake. Specifically, during the 60 minutes before the P-wave arrival (red curves in Figure 5), we computed PPSD every 900 seconds. We followed the same procedure during the 60 minutes after the arrival of surface waves (blue curves in Figure 5).
 

 

What can we learn?

For both events analyzed here, the implications are different. In the case of the 2018 Honduras earthquake, dynamically triggered seismicity occurred in the Tenorio Volcano and surrounding volcanic field. The sequence of events lasted several days before dying out.

However, an alternative scenario could have occurred: the cascade of events could have led to a change in the internal dynamics of the volcano, releasing pressure and culminating in an eruption that could have affected lives and public infrastructure. For instance, the geothermal field next to the volcano produces energy for the country. In Costa Rica, where the 99% of the energy produced is green, this scenario would be a game changer if energy production for the country were reduced. This would affect public and private infrastructure, and consequently, the growing economy.

For the second example, the 2023 Türkiye-Syria earthquake triggered seismicity along the subduction zone in southern Costa Rica. Similarly, the sequence lasted several days, and the largest triggered event was a magnitude 4.4 earthquake located on the subduction interface.

That seismicity was triggered in a subduction zone — a compressive system — may be counterintuitive (e.g., Brodsky and van der Elst, 2014). However, in this region, subducting sediments could be lubricating the fault, introducing different material properties and rheologies that might be susceptible to minor strain changes compared to the central and northern subduction zones (DeShon et al., 2006; Bangs et al., 2016; Edwards et al., 2018).

Of concern, in this region, the Osa Peninsula last experienced a magnitude 7.4 earthquake in 1983, which produced significant damage (Tajima and Kikuchi, 1995). With an estimated recurrence time of about 40 years, an event of similar magnitude may be impending. Recent seismological observations have shown how big ruptures like the 2011 magnitude 9.1, Tōhoku, Japan earthquake or the 2014 magnitude 8.2 Iqueque, Chile earthquake were preceded by distinct spatial and temporal patterns of microseismicity (Lay et al., 2014). However, we do not yet understand if a cascade of dynamically triggered microseismicity can also lead to a larger earthquake rupture in a critically stressed region. But it is a concept that should be considered, both in risk assessment and in future research.
 

Science editor: Dr. Alka Tripathy-Lang, Ph.D.
Reviewers: Dr. Chastity Aiken, Ph.D. and Dr. Ross Stein, Ph.D.
 

References

Aiken, C., & Peng, Z. (2014). Dynamic triggering of microearthquakes in three geothermal/volcanic regions of California. Journal of Geophysical Research: Solid Earth, 119(9), 6992-7009. https://doi.org/10.1002/2014JB011218.

Aiken, C., Meng, X., & Hardebeck, J. (2018). Testing for the ‘predictability’of dynamically triggered earthquakes in The Geysers geothermal field. Earth and Planetary Science Letters, 486, 129-140. https://doi.org/10.1016/j.epsl.2018.01.015.

Brodsky, E. E., & van der Elst, N. J. (2014). The uses of dynamic earthquake triggering. Annual Review of Earth and Planetary Sciences, 42, 317-339.https://doi.org/10.1146/annurev-earth-060313-054648.

Fan, W., Barbour, A. J., Cochran, E. S., & Lin, G. (2021). Characteristics of frequent dynamic triggering of microearthquakes in Southern California. Journal of Geophysical Research: Solid Earth, 126(1), e2020JB020820. https://doi. org/10.1029/2020JB020820.

Gomberg, J., & Johnson, P. (2005). Dynamic triggering of earthquakes. Nature, 437(7060), 830-830.https://doi.org/10.1038/437830a.

Hill, D. P., Reasenberg, P. A., Michael, A., Arabaz, W. J., Beroza, G., Brumbaugh, D., … & Zollweg, J. 1. (1993). Seismicity remotely triggered by the magnitude 7.3 Landers, California, earthquake. Science, 260(5114), 1617-1623. https://doi.org/10.1126/science.260.5114.1617.

Langer, S., Finzi, Y., & Olsen-Kettle, L. M. (2015). Dynamic triggering of earthquakes is promoted by crustal heterogeneities and bimaterial faults. Physics of the Earth and Planetary Interiors, 238, 34-41. https://doi.org/10.1016/j.pepi.2014.10.012.

Lay, T., Yue, H., Brodsky, E. E., & An, C. (2014). The 1 April 2014 Iquique, Chile, mw 8.1 earthquake rupture sequence. Geophysical Research Letters, 41(11), 3818-3825. doi:10.1002/2014GL060238.

McNamara, D. E., & Buland, R. P. (2004). Ambient noise levels in the continental United States. Bulletin of the seismological society of America, 94(4), 1517-1527. doi: https://doi.org/10.1785/012003001.

Peng, Z., Walter, J. I., Aster, R. C., Nyblade, A., Wiens, D. A., & Anandakrishnan, S. (2014). Antarctic icequakes triggered by the 2010 Maule earthquake in Chile, Nat. https://doi.org/10.1038/ngeo2212.

Peng, Z., Wang, W., Chen, Q. F., & Jiang, T. (2010). Remotely triggered seismicity in north China following the 2008 M w 7.9 Wenchuan earthquake. Earth, planets and space, 62, 893-898. https://doi.org/10.5047/eps.2009.03.006.

Tajima, F., and Kikuchi, M. (1995.) Tectonic implications of the seismic ruptures associated with the 1983 and 1991 Costa Rica earthquakes, in Mann, P., ed., Geologic and tectonic development of the Caribbean plate boundary in southern Central America: Geological Society of America Special Paper 295, 327–340.

Edwards, J. H., Kluesner, J. W., Silver, E. A., Brodsky, E. E., Brothers, D. S., Bangs, N. L., … & Okamoto, K. (2018). Corrugated megathrust revealed offshore from Costa Rica. Nature Geoscience, 11(3), 197-202.https://doi.org/10.1038/s41561-018-0061-4.

DeShon, H. R., Schwartz, S. Y., Newman, A. V., González, V., Protti, M., Dorman, L. M., … & Flueh, E. R. (2006). Seismogenic zone structure beneath the Nicoya Peninsula, Costa Rica, from three-dimensional local earthquake P-and S-wave tomography. Geophysical Journal International, 164(1), 109-124. https://doi.org/10.1111/j.1365-246X.2005.02809.x

Bangs, N. L., McIntosh, K. D., Silver, E. A., Kluesner, J. W., & Ranero, C. R. (2016). A recent phase of accretion along the southern Costa Rican subduction zone. Earth and Planetary Science Letters, 443, 204-215.https://doi.org/10.1016/j.epsl.2016.03.008.
 
 

Copyright

Text © 2024 Temblor. CC BY-NC-ND 4.0

We publish our work — articles and maps made by Temblor — under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) license.

For more information, please see our Republishing Guidelines or reach out to news@temblor.net with any questions.