Are the 2021 and 2010 Haiti earthquakes part of a progressive sequence?

Haiti’s 2010 earthquake likely brought the fault that ruptured in the Aug. 14, 2021 quake closer to failure. Some elements suggest a westward earthquake progression, but highly stressed sections of the fault system to the east remain.
 

By Ross Stein, Ph.D., Temblor, Inc., California (@rstein357), Shinji Toda, Ph.D., IRIDeS, Tohoku University, Japan, Jian Lin, Ph.D., Southern University of Science and Technology, China, and Woods Hole Oceanographic Institution, Massachusetts, and Volkan Sevilgen, M.Sc., Temblor, Inc., California (@volkansevilgen)
 

Translations of this article are available in Japanese, French, Chinese and Spanish.
 

Citation: Stein, R.S., Toda, S., Lin, J., Sevilgen, V., 2021, Are the 2021 and 2010 Haiti earthquakes part of a progressive sequence?, Temblor, http://doi.org/10.32858/temblor.197
 

On Jan. 12, 2010, a magnitude-7.0 earthquake struck Léogâne, Haiti, just outside the capital of Port-au-Prince. The quake killed some 300,000 people, according to the Haitian government, and displaced hundreds of thousands more people. Though the quake was initially thought to have struck on the Enriquillo-Plantain Garden Fault, where the Caribbean Plate is separated from the Gonâve microplate, further investigation eventually indicated the quake occurred on a blind thrust fault now known as the Léogâne Fault. (Blind thrust faults are those that don’t reach Earth’s surface and result from compression.) On Aug. 14, 2021, a magnitude-7.2 quake struck along the same fault system, to the west of the epicenter of the 2010 quake. The death toll is already above 1,400 and thousands more are displaced. Those numbers are expected to rise, especially as a tropical storm barrels down on the island.

Our analyses suggest that the disastrous 2010 earthquake likely brought the fault that ruptured in the Aug. 14, 2021 quake closer to failure. Some elements suggest a westward earthquake progression, but highly stressed sections of the fault system well to the east remain.
 

Stress transferred by the 2010 quake to the site of the 2021 event

Two weeks after the 2010 earthquake, we published a Coulomb stress analysis to gain insight as to what could happen next (Lin et al., 2010). We identified sections of the Enriquillo-Plantain Garden Fault (Mann et al., 2002) to the east and west of the 2010 rupture zone with significantly increased stress and hazard, as shown in Figure 1. The 2021 magnitude-7.2 Nippes, Haiti, earthquake struck on a patch that was brought 0.1 bar closer to failure. While 0.1 bar has been shown in many studies to be large enough to trigger earthquakes, it is nevertheless much smaller than the stress increases closer to both ends of the 2010 rupture, which we calculated to be 5-10 times larger. Symithe et al. (2013) obtained similar and more extensive results, providing independent confirmation of our calculations.
 

Figure 1. This is Figure 1 from Lin et al. (2010), annotated with the epicenter of the Aug. 14, 2021, magnitude-7.2 Nippes mainshock, which likely struck on a patch of the Enriquillo-Plantain Garden Fault System. Credit: Authors, after Lin et al. (2010)
Figure 1. This is Figure 1 from Lin et al. (2010), annotated with the epicenter of the Aug. 14, 2021, magnitude-7.2 Nippes mainshock, which likely struck on a patch of the Enriquillo-Plantain Garden Fault System. Credit: Authors, after Lin et al. (2010)

 

The fault map in Figure 1 might be oversimplified, as there are likely two adjacent faults running along the peninsula— the Enriquillo, as well as a blind thrust fault (Calais et al., 2010; Hayes et al, 2010; Prentice et al., 2010; Hashimoto et al., 2012; Douilly et al., 2013). Here, we refer to both faults as a fault system. But for both fault geometries, the 2010 earthquake stressed the site of the 2021 event.

Why the 11-year wait for the next shoe to fall, and why the unruptured fault section between the two earthquakes (labeled “jump?” in Figure 2)? In our calculations, the impact of the stress change on seismicity fades with time, just as do the rate of aftershocks. So, while an immediate trigger is more likely, long delays are still possible. The jump or gap between the 2010 and 2021 ruptures could be explained by a strong fault patch that requires still more stress to rupture, it could mark a bend or break in the fault (Saint Fleur et al., 2020), or this section could have ruptured prehistorically, lowering the stress relative to adjacent patches. Irrespective of those speculations, we cannot rule out the possibility that the fault could soon rupture through that gap in a magnitude~6.5 aftershock.
 

Figure 2. Progressive westward rupture of the 2010 and 2021 earthquakes. It appears that there is a 15-kilometer-long jump or gap between them, one candidate among several for a future large earthquake. Credit: Temblor Inc.
Figure 2. Progressive westward rupture of the 2010 and 2021 earthquakes. It appears that there is a 15-kilometer-long jump or gap between them, one candidate among several for a future large earthquake. Credit: Temblor Inc.

 

Stress transferred by the 2021 earthquake

How has the fault stress been altered by the magnitude-7.2 event, which is about twice as large as its 2010 forerunner? In Figure 3, we calculate this in several different ways. Regardless of which approach we take, one can see that sections of the fault system to the east and west of the magnitude-7.2 rupture were brought significantly closer to failure. This is easiest to visualize in Figure 3a. But Figure 3b shows something else of importance: The faults that ruptured in the 2010 aftershocks have been re-stressed by the 2021 event, and so we could see a reawakening of the 2010 aftershock zone.
 

Figure 3. Stress transferred by the Aug. 14, 2021, magnitude-7.2 mainshock to surrounding faults. In (a), we calculate stress on faults with the same geometry as the mainshock, based on a preliminary finite fault model (USGS, 2021). Lobes of stress increase (red) extend in four directions. (b) Here we use the focal mechanisms of past earthquakes to infer the geometry of surrounding faults, which indicates that the faults that ruptured in 2010 aftershocks were re-stressed by the 2021 earthquake (red beachballs). (c) Because the density of focal mechanisms is very sparse, here we interpolate between focal mechanisms for a smooth grid. Credit: Temblor Inc.
Figure 3. Stress transferred by the Aug. 14, 2021, magnitude-7.2 mainshock to surrounding faults. In (a), we calculate stress on faults with the same geometry as the mainshock, based on a preliminary finite fault model (USGS, 2021). Lobes of stress increase (red) extend in four directions. (b) Here we use the focal mechanisms of past earthquakes to infer the geometry of surrounding faults, which indicates that the faults that ruptured in 2010 aftershocks were re-stressed by the 2021 earthquake (red beachballs). (c) Because the density of focal mechanisms is very sparse, here we interpolate between focal mechanisms for a smooth grid. Credit: Temblor Inc.

 

The 2021 earthquake had a predecessor in 1770

Saint Fleur et al. (2020) excavated a trench along the Enriquillo–Plantain Garden fault near Clonard (shown in Figure 2), which turns out to be within the rupture of the 2021 mainshock. They found evidence for an earthquake in about 1770, with a range of uncertainty of 40 years, which they suggest corresponds to the June 3, 1770, magnitude~7.5 earthquake. The Enriquillo Fault has a long-term slip rate of about 9 millimeters/year at this location, so in the 250 years between the two events, about 2.0-2.5 meters of potential slip would have accumulated. According to the USGS (2021) finite fault model, the mean slip in the 2021 earthquake was about 1.5 meters, and so these inferences are roughly consistent: Stress released in the 1770 earthquake had rebuilt by about 2010, when a small amount of additional stress was transferred to the site, further ratcheting the fault toward failure.
 

Earthquake forecast for the next 12 months

We make a probabilistic forecast (Figure 4) of the next year by using the stress transferred from the 2010 and 2021 mainshocks, our smoothed grid of focal mechanisms (shown in Figure 3c), and the Global Earthquake Activity Rate (GEAR) model of Bird et al. (2015) (Figure 4b). The method we adopt (Toda and Stein, 2020) assumes that the impact of a large earthquake decays with time. The absence of a strong national seismic network limits our ability to test the forecast in Haiti, but the same approach has performed well in Japan, California and Chile.

In the forecast (Figure 4a), we calculate an elevated hazard extending from Jeremie (population 122,000) near the west end of the peninsula, all the way to Port-au-Prince in the east, a surprisingly long extent of 200 kilometers. But in support of this calculation, Figure 2 shows numerous earthquakes both east of Port-au-Prince, and in the region around the Aug. 14, 2021, magnitude-7.2 rupture zone, more than the 11 years before the magnitude-7.2 event struck (the grey quakes). These quakes probably indicate that the hazard was high along this section of the Enriquillo-Plaintain Garden fault system since the 2010 magnitude-7.0 event struck.
 

Figure 4. Probabilistic forecast for 12-month periods. (a) This forecast considers the decaying impact of the stress transferred by the 2010 magnitude-7.0 and the 2021 magnitude-7.2 earthquakes on surrounding faults, following the approach of Toda and Stein (2020). Except for the 40 kilometers centered on the Aug. 14, 2021 rupture, a 220-kilometer-long section of the Enriquillo-Plantain Garden Fault System has a higher likelihood of hosting magnitude-5.0 or bigger quakes than during an average 12-month period, as shown in (b). Credit: Temblor Inc.
Figure 4. Probabilistic forecast for 12-month periods. (a) This forecast considers the decaying impact of the stress transferred by the 2010 magnitude-7.0 and the 2021 magnitude-7.2 earthquakes on surrounding faults, following the approach of Toda and Stein (2020). Except for the 40 kilometers centered on the Aug. 14, 2021 rupture, a 220-kilometer-long section of the Enriquillo-Plantain Garden Fault System has a higher likelihood of hosting magnitude-5.0 or bigger quakes than during an average 12-month period, as shown in (b). Credit: Temblor Inc.

 

Is the earthquake sequence headed west?

One might be tempted to infer that the magnitude-7 earthquake sequence is progressing westward, in which case any future large shocks would strike less-populated areas along Haiti’s southern peninsula. The argument for this view is that the aftershocks of the 2010 and 2021 earthquakes are separated by about 15 kilometers, and in addition, the 2010 and 2021 events ruptured largely to the west, where most of their aftershocks lie.

But our forecast tells a different story: The eastern sections of the fault are also calculated to have an elevated probability of large shocks, and unfortunately, these areas are more populated. One might ask if the 11 years since the 2010 quake without a large shock to the east means that we can exclude this possibility. But there have been about 10 shocks to the east of the 2010 rupture in the past decade (Figure 2), and so, in our judgment, a larger earthquake there remains a possibility.
 

References

Bird, P., D.D. Jackson, Y.Y. Kagan, C. Kreemer, and R.S. Stein (2015), GEAR1: A global earthquake activity rate model constructed from geodetic strain rates and smoothed seismicity, Bull. Seismol. Soc. Amer., 105, 2538–2554.

Calais, E., A. Freed, G. Mattioli, F. Amelung, S. Jónsson, P. Jansma, S.-H. Hong, T. Dixon, C. Prépetit, and R. Momplaisir (2010), Transpressional rupture of an unmapped fault during the 2010 Haiti earthquake. Nature Geosci., 3, 794-799, doi:10.1038/ngeo992.

Douilly, R., et al. (2013), Crustal structure and fault geometry of the 2010 Haiti earthquake from temporary seismometer deployments, Bull. Seismol. Soc. Am., 103, 2305–2325, doi: 10.1785/0120120303.

Hashimoto, M., Y. Fukushima, and Y. Fukahata (2011), Fan-delta uplift and mountain subsidence during the Haiti 2010 earthquake, Nature Geosci., 4, 255–259, doi:10.1038/ngeo1115.

Hayes, G. P., R. W. Briggs, A. Sladen, E. J. Fielding, C. Prentice, K. Hudnut, P. Mann, F. W. Taylor, A. J. Crone, R. Gold, T. Ito, and M. Simons (2010), Complex rupture during the 12 January 2010 Haiti earthquake, Nature Geosci., 3, 800–805, doi:10.1038/ngeo977.

Lin, J., Stein, R. S., Sevilgen, V., and Toda, S. (2010), USGS-WHOI-DPRI Coulomb stress-transfer model for the January 12, 2010, MW=7.0 Haiti earthquake: U.S. Geological Survey Open-File Report 2010-1019, 7 p.

Mann, P., E. Calais, J.-C. Ruegg, C. DeMets, P. E. Jansma, and G. S. Mattioli (2002), Oblique collision in the northeastern Caribbean from GPS measurements and geological observations, Tectonics, 21, 1057, doi:10.1029/2001TC001304.

Prentice, C. S., P. Mann, A. J. Crone, R. D. Gold, K. W. Hudnut, R. W. Briggs, R. D. Koehler, and P. Jean (2010), Seismic hazard of the Enriquillo Plantain Garden fault in Haiti inferred from palaeoseismology, Nature Geosci., 3, 789-793, doi:10.1038/ngeo991.

Saint Fleur, N., Y. Klinger, N. Feuillet (2020), Detailed map, displacement, paleoseismology, and segmentation of the Enriquillo-Plantain Garden Fault in Haiti, Tectonophysics, 778, 228368.

Symithe, S. J., E. Calais, J. S. Haase, A. M. Freed, and R. Douilly (2013), Coseismic slip distribution of the 2010 M 7.0 Haiti earthquake and resulting stress changes on regional faults, Bull. Seismol. Soc. Am., 103, 2326–2343.

Toda, S., and Stein, R. S. (2020), Long‐ and short‐term stress interaction of the 2019 Ridgecrest sequence and Coulomb‐based earthquake forecasts. Bull. Seismol. Soc. Amer. 110, 1765-1780.

USGS (2021), Finite Fault model for 14 August 2021 M 7.2 Nippes, Haiti, Earthquake
https://earthquake.usgs.gov/earthquakes/eventpage/us6000f65h/finite-fault