Meteorological data suggest that simultaneous heavy rainfall from Hurricane Hilary and an earthquake in Ojai, California, were coincidences.
By Qiushi Zhai (Caltech Seismological Laboratory, qzhai@caltech.edu), Zhigang Peng (School of Earth and Atmospheric Sciences, Georgia Tech, zpeng@gatech.edu), and Ya-Ju Hsu (Institute of Earth Sciences, Academia Sinica, Taiwan, yaru@earth.sinica.edu.tw)
Citation: Zhai, Q., Peng, Z., Hsu, Y., 2023, Southern California earthquake was unlikely triggered by Hurricane Hilary, Temblor, http://doi.org/10.32858/temblor.322
Southern California is no stranger to earthquakes, but tropical storms like Hilary are rare. It’s even rarer for a magnitude-5.1 earthquake near Ojai, Calif. to strike on the same Sunday afternoon (Aug. 20, 2023) that a tropical storm swept through Southern California. This uncommon confluence of events has sparked a heightened curiosity in the general public, including popular memes and the portmanteau “hurriquake.” But is there a physical connection between these events beyond mere coincidence in space and time? Two recent articles suggest that Hurricane Hilary was unlikely to trigger earthquakes (Turner, 2023; Samuels, 2023). Here, we go a step further and examine available meteorological and seismic data to reveal that the 2023 magnitude-5.1 earthquake near Ojai, Calif. was almost certainly not triggered by Hilary, which had been downgraded to a tropical storm at the time the earthquake struck.
Earthquake triggering and surface processes
Within the scientific community, it’s broadly accepted that energy released during the events like magnitude-5.1 Ojai earthquake originate from long-term tectonic stress loading of active faults — not from extreme weather events like hurricanes. Nevertheless, the timing of such earthquakes could, theoretically, be influenced by significant external stress perturbations that result from either the atmospheric pressure changes or water dropped by such storms.
In the past couple of decades, multiple studies have shown that seasonal variations of snow load (Heki, 2003), barometric pressures (Gao et al., 2000), groundwater storage (Johnson et al., 2017; Hsu et al., 2021), or long-term erosion and unloading (Calais et al., 2010; Steer et al., 2014) are capable of generating stress perturbations on the order of kilopascals at seismogenic depth — the depth at which earthquakes nucleate. (100 kilopascal is equal to about 1 bar, and 1 bar is slightly less than the atmospheric pressure on Earth at sea level; thus, 1 kilopascal is a small fraction of barometric pressure.) Just for reference, a typical earthquake stress drop is 3 megapascals or 30 bar, so a kilopascal is 0.03% of a stress drop — a very small perturbation. Thus, kilopascals of stress change is enough to trigger or modulate seismic activity. However, it is still not clear whether short-term variations (on the order of hours to days), such as extreme weather events (e.g., hurricanes, typhoons or heavy rainfall) can trigger earthquakes. Several recent studies have shown that extreme weather events such as tropical cyclones might trigger additional aftershocks (Meng et al., 2018), change earthquake distribution (Steer et al., 2020), or even shut down an ongoing earthquake swarm (Zhai et al., 2021). But these are only case studies, and none of them can completely rule out that the observed correlations are purely coincidental.
Stress perturbations from hurricanes
Earth consistently experiences lunar tidal stress variations (approximately 2-3 kilopascals; Freed, 2005), and recent studies show that such tidal stresses can trigger or modulate seismic activity (e.g., Tanaka et al., 2002; Ide et al., 2016; Dumont et al., 2023). Therefore, the tidal stress level is a valuable baseline to assess whether the stress perturbation caused by a specific weather event is significant enough to affect seismic activity. Within hours to days during and after a hurricane event, two primary surface processes can induce stress changes in subsurface faults: atmospheric pressure variations near the eye of the hurricane and associated heavy rainfall. In addition, extreme weather events can trigger landslides, and the subsequent removal of landslide debris in the following years can also generate stress perturbations that are significant enough to affect seismic activity (Steer et al., 2020). So, we’d like to know if atmospheric pressure changes or heavy rainfall resulted in stress changes that were greater than tidally driven perturbations at the time the Ojai earthquake occurred.
Atmospheric pressure variations and rainfall
The data recorded by the KOXR weather station at the Oxnard Airport, 23 kilometers (15 miles) away from the epicenter of the magnitude-5.1 Ojai earthquake, reveal a 1.5-kilopascal atmospheric pressure drop during Hurricane Hilary’s passage in Southern California (Figure 1a), slightly below standard tidal stress variations. The resulting stress perturbations would be even smaller if we resolved them along the fault plane at 4 to 5 kilometers (2.5 to 3.1 miles) depth, where the magnitude-5.1 earthquake nucleated. The pressure drop started about one day prior to the earthquake, and the pressure reached its minimum about three hours after the earthquake. This 1.5-kilopascal atmospheric pressure drop from Hilary is much smaller than the 6-kilopascal atmospheric pressure drop from a typhoon that struck Taiwan in 2009 — Typhoon Morakot (Zhai et al., 2021).
Typhoon Morakot was the strongest and wettest typhoon in the past 60 years in Taiwan — a highly seismically active region in East Asia. Hsu et al. (2021) show that the annual average variation of the thickness of the water table in Taiwan is 0.53 ± 0.17 meters, which can result in a Coulomb stress increase of 3 to 5 kilopascal on a 30°-dipping thrust fault at 10 kilometers (6.2 miles) depth. Therefore, the change in water thickness can modulate the subsurface seismicity in Taiwan. In other words, if enough rain permeated the ground and increased thickness of the water table, that could change the stress on active faults and hence trigger earthquakes. So, could heavy rainfall have caused the quake in Ojai?
The accumulated rainfall recorded near the Ojai earthquake epicenter amounted to approximately 0.005 meters (about 0.2 inches) prior to the earthquake (Figure 1b). In comparison, Typhoon Morakot brought approximately 0.3 meters (about 12 inches) of accumulated rainfall to northeastern Taiwan and 3 meters (120 inches) to southwestern Taiwan. Though this 2009 rainfall in Taiwan was orders of magnitude greater than that in Ojai attributable to Hurricane Hilary, Zhai et al. (2021) were unable to observe any clear changes in seismic activity in southwestern Taiwan — the region that saw the heaviest rainfall — shortly following Typhoon Morakot. That’s at least in part because many seismic stations were damaged or out of power. In northeastern Taiwan, Zhai et al. (2021) demonstrated that a 40-day earthquake sequence was shut down right after the passage of typhoon Morakot.
In the Southern California case, the stress induced by a 0.005-meter water column is about 0.05 kilopascal, substantially below tidal stress levels, making it unlikely to be a trigger for earthquakes.
A third possibility for how a hurricane might induce an earthquake is the load on the crust if a massive amount of rain pools, like in a lake, but that would need to be a lot of rain in one place with nowhere to go. Such a scenario is theoretically possible in the inland Salton Sea, as discussed by Turner (2023), but that is certainly not the scenario in Ojai, where runoff reaches the ocean.
A fourth possibility can be the diffusion of rainwater into the subsurface, leading to an increase in pore pressure on active faults. For example, Miller (2008) observed that rain-induced earthquakes primarily occur in karst geological settings, where rainwater rapidly enters underground systems of soluble rock types replete with sinkholes, caves and other cavities. Montgomery-Brown et al. (2019) also observed snowmelt‐triggered earthquake swarms at the margin of Long Valley Caldera, where steeply dipping strata provide high‐permeability pathways for water flow into the ground, toward the faults. With such a rapid funneling of water into the subsurface, earthquakes can potentially result. However, the region in which the Ojai earthquake sequence occurred is within the rugged western Transverse Ranges that comprises mostly Cretaceous and younger clastic sedimentary and metasedimentary rocks near Ojai; Ojai itself sits on Quaternary alluvium.
Could foreshocks be the trigger?
Another piece of the puzzle is that the magnitude-5.1 Ojai mainshock, which occurred on Sunday afternoon local time, was preceded within three days by 19 smaller earthquakes inside a 10-kilometer (6.2 miles) radius. In retrospect, these events are considered foreshocks of the magnitude-5.1 Ojai event. While the underlying physical mechanisms of foreshocks are still debated, these foreshocks clearly occurred before the atmospheric pressure drop and the rainfall (Figure 1). The presence of the foreshocks alone cannot completely rule out the potential influence of external stress perturbation on the time of the mainshock. But it is more reasonable to argue that the occurrence of these foreshocks, rather than the occurrence of Hurricane Hilary, likely played some role in triggering the magnitude-5.1 Ojai mainshock.
Summary
A strong hurricane and its associated rainfall and subsequent sediment removal theoretically has the capability of affecting seismic activity, if we compare the stress perturbation it can cause with tidal stress variations. This is likely the case during the 2009 Typhoon Morakot in Taiwan (Steer et al., 2020; Zhai et al., 2021), and a few other cases (Meng et al., 2018). However, it is highly unlikely that the 2023 magnitude-5.1 Ojai earthquake was triggered by Hurricane Hilary, mainly because this storm and its associated rainfall were insufficient. Whether the foreshocks — which occurred well before the rain — brought the fault closer to failure needs now to be assessed.
Science editor: Dr. Alka Tripathy-Lang, Ph.D.
Reviewers: Dr. Ross S. Stein, Ph.D. and Dr. Wendy Bohon, Ph.D.
References
Calais, E., Freed, A. M., Van Arsdale, R., & Stein, S. (2010). Triggering of New Madrid seismicity by late-Pleistocene erosion. Nature, 466(7306), 608–611. https://doi.org/10.1038/nature09258.
Dumont, S., Custódio, S., Petrosino, S., Thomas, A. M. & Sottili, G. (2023). Chapter 14 – Tides, earthquakes, and volcanic eruptions, Editor(s): M. Green, J. C. Duarte, A Journey Through Tides, Elsevier, 333-364, https://doi.org/10.1016/B978-0-323-90851-1.00008-X.
Freed, A. M. (2005). Earthquake Triggering by Static, Dynamic, and Postseismic Stress Transfer. Annual Review of Earth and Planetary Sciences, 33(1), 335–367. https://doi.org/10.1146/annurev.earth.33.092203.122505
Gao, S., Silver, P., Linde, A., & Sacks, I. S. (2000). Annual modulation of triggered seismicity following the 1992 Landers earthquake in California. Nature 406, 500–504. https://doi.org/10.1038/35020045
Heki, K. (2003). Snow load and seasonal variation of earthquake occurrence in Japan, Earth and Planetary Science Letters, 207(1–4), 159-164, https://doi.org/10.1016/S0012-821X(02)01148-2.
Hsu, Y. J., Kao, H., Burgmann, R., Lee, Y. T., Huang, H. H., Hsu, Y. F., Wu, Y. M., & Zhuang, J. (2021). Synchronized and asynchronous modulation of seismicity by hydrological loading: A case study in Taiwan. Sci Adv, 7(16), eabf7282. https://doi.org/10.1126/sciadv.abf7282
Ide, S., Yabe, S. & Tanaka, Y. Earthquake potential revealed by tidal influence on earthquake size–frequency statistics. Nature Geosci 9, 834–837 (2016). https://doi.org/10.1038/ngeo2796
Meng, X., H. Yang and Z. Peng (2018), Foreshocks, b value map and aftershock triggering following the 2011 Mw5.7 Virginia earthquake, J. Geophys. Res., doi:10.1029/2017JB015136.
Miller, S. A. (2008). Note on rain-triggered earthquakes and their dependence on karst geology. Geophysical Journal International, 173(1), 334–338. https://doi.org/10.1111/j.1365-246X.2008.03735.x
Montgomery‐Brown, E. K., Shelly, D. R., & Hsieh, P. A. (2019). Snowmelt‐triggered earthquake swarms at the margin of Long Valley Caldera, California. Geophysical Research Letters, 46(7), 3698-3705.
Samuels, F.M.D., (2023). Southern California shaken by earthquake during unrelated hurricane, Temblor, http://doi.org/10.32858/temblor.319
Steer, P., Simoes, M., Cattin, R., & Shyu, J. B. (2014). Erosion influences the seismicity of active thrust faults. Nature Communications, 5(1), 5564. https://doi.org/10.1038/ncomms6564
Steer, P., Jeandet, L., Cubas, N., Marc, O., Meunier, P., Simoes, M., Cattin, R., Shyu, J. B. H., Mouyen, M., Liang, W.-T., Theunissen, T., Chiang, S.-H. & Hovius, N. (2020). Earthquake statistics changed by typhoon-driven erosion. Scientific Reports, 10(1), 10899. https://doi.org/10.1038/s41598-020-67865-y
Tanaka, S., Ohtake, M., & Sato, H. (2002). Evidence for tidal triggering of earthquakes as revealed from statistical analysis of global data. Journal of Geophysical Research: Solid Earth, 107(B10), ESE 1-1-ESE 1-11. https://doi.org/10.1029/2001JB001577
Turner, Alice R., 2023, How many Hurricane Hilarys would it take to trigger an earthquake? Probably a lot, Temblor, http://doi.org/10.32858/temblor.318
Zhai, Q., Peng, Z., Chuang, L. Y., Wu, Y.-M., Hsu, Y.-J., & Wdowinski, S. (2021). Investigating the Impacts of a Wet Typhoon on Microseismicity: A Case Study of the 2009 Typhoon Morakot in Taiwan Based on a Template Matching Catalog. Journal of Geophysical Research-Solid Earth, 126(12). https://doi.org/10.1029/2021jb023026
Copyright
Text © 2023 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.
