Seismicity induced by tropical storms Eta and Iota in the karstic region of northern Guatemala
Sismicidad inducida por las tormentas tropicales Eta e Iota en la región kárstica del norte de Guatemala
Robin Yani-Quiyuch1,*
, Diego Castro1, Erwin Quiroa1, Fredy Monterroso1
1 Instituto Nacional de Sismología, Vulcanología, Meteorología e Hidrología (INSIVUMEH), 7th. Av. 14-57 zone 13, Guatemala City, Guatemala 01013.
* Corresponding author: (R. Yani-Quiyuch) This email address is being protected from spambots. You need JavaScript enabled to view it.
How to cite this article:
Yani-Quiyuch, R., Castro, D., Quiroa, E., & Monterroso, F. (2026). Seismicity induced by tropical storms Eta and Iota in the karstic region of northern Guatemala. Boletín de la Sociedad Geológica Mexicana, 78(2), A161225. https://doi.org/10.18268/BSGM2026v78n2A161225
Manuscript received: July 30, 2025. Corrected manuscript received: November 20, 2025. Manuscript accepted: November 30, 2025.
ABSTRACT
We present a case of induced seismicity in the departments of Alta Verapaz and Quiché, in northern Guatemala, and its relationship with severe flooding caused by tropical storms Eta and Iota, in a subsurface already saturated by an unprecedented Atlantic hurricane season in 2020. The epicentral region was located in an area with predominance of carbonate rocks, where a complex karst system has developed. The mechanisms that triggered this seismicity may be mainly associated with the direct hydraulic connection of the flash floods with the seismic sources and diffusion processes in the pores, suggested by the temporal relationship between the rainfall and the seismicity episodes, according to previous studies of induced seismicity in karstic regions. The identification of earthquake families through correlation analysis in the frequency domain of seismic records suggests that several cracks or faults were activated, in a tectonic setting predisposed to nucleate earthquakes, which represents a significant seismic hazard. This is the first time that this type of induced seismicity caused by extreme meteorological phenomena in karstic environments has been instrumentally recorded and analyzed in Guatemala.
Keywords: induced seismicity, Eta and Iota, karst, floods, repeating earthquakes.
RESUMEN
Presentamos un caso de sismicidad inducida en los departamentos de Alta Verapaz y Quiché, en el norte de Guatemala, y su relación con las inundaciones severas provocadas por las tormentas tropicales Eta e Iota, en un subsuelo ya saturado por una temporada de huracanes del Atlántico sin precedentes en el año 2020. La región epicentral se localizó en una región con predominancia de rocas carbonáticas, en la cual se ha desarrollado un complejo sistema kárstico. Los mecanismos que dispararon dicha sismicidad pueden estar asociados principalmente con la conexión hidráulica directa de las inundaciones súbitas con las fuentes sísmicas y procesos de difusión en los poros, sugeridos por la relación temporal entre las lluvias y los episodios de sismicidad, de acuerdo con estudios previos de sismicidad inducida en regiones kársticas. La identificación de familias de sismos mediante análisis de correlación en el dominio de la frecuencia de los registros sísmicos, sugiere que varias grietas o fallamientos fueron activados, en un entorno tectónico con predisposición a nuclear terremotos, lo que representa una amenaza sísmica importante. Esta es la primera vez que se registra y analiza instrumentalmente este tipo de sismicidad inducida por fenómenos meteorológicos extremos en ambientes kársticos en Guatemala.
Palabras clave: sismicidad inducida, Eta e Iota, karst, inundaciones, sismos repetitivos.
1. Introduction
The 2020 Atlantic hurricane season was the most active on record and the fifth consecutive season to exceed average activity, with 30 tropical storms, 13 of which became hurricanes. Hurricanes Eta and Iota caused disasters in Nicaragua, Honduras, and Guatemala (Bello and Peralta, 2021).
Although their paths did not reach Guatemalan territory (Figure 1a), and during their closest approach both had downgraded to tropical storms (Pasch et al., 2021; Stewart, 2021), their areas of influence caused intense rainfall that affected the population between November 3 and 17, resulting in floods, dozens of landslides, and mudflows. These events impacted 16 of the country’s 22 departments, including Alta Verapaz and Quiché (Figure 1a), whose predominant geology is composed of carbonate rocks (Figure 1b), where an extensive karst system has developed.
According to the Coordinadora Nacional para la Reducción de Desastres (CONRED), 60 people were reported dead, 30 injured, at least 100 missing, around 60,000 homes with moderate to severe damage, and a large amount of national and community infrastructure affected (Bello and Peralta, 2021).
During the development of these tropical storms, the Red Sismológica Nacional (RSN), operated by the Instituto Nacional de Sismología, Vulcanología, Meteorología e Hidrología (INSIVUMEH, 1976), recorded unusual seismic activity in the departments of Alta Verapaz and Quiché (Figure 1b). Several of these events were felt by the population, causing alarm. Combined with the flooding and the restrictions imposed by COVID-19, this particular context resulted in a significant humanitarian crisis in the region.
In this study, we analyze the relationship between the occurrence of this seismicity and the extensive flooding of the karst system, which was documented through fieldwork. The seismic sequence developed in different stages, forming at least three main clusters, whose temporal occurrence may provide clues about the mechanisms involved in nucleating this seismicity.
We also explore the possible activation of different pre-existing faults in a tectonic environment susceptible to generating earthquakes.
2. Geological description
Dengo (1969) divided the region that includes Guatemala into two tectonic blocks known as the Maya Block and the Chortis Block. The Maya Block constitutes the continental part of the southern North American Plate and includes the Yucatán Peninsula, Belize, northern Guatemala, and southeastern Mexico. The Chortis Block forms the eastern part of the Caribbean Plate, comprising southern Guatemala, Honduras, El Salvador, and part of northern Nicaragua. These two blocks are separated by the Motagua Suture Zone. However, some authors, such as Fourcade et al. (1994), define the Polochic Fault as the boundary between the Maya and Chortis blocks (Figure 1a).
Northern Guatemala is part of the Maya Block, where the oldest rocks of the cratonic basement are igneous and metamorphic, unconformably overlain by Upper Paleozoic metasedimentary rocks (Finch and Dengo, 1990). The thick carbonate deposits are responsible for the karstic topography in northern Guatemala (Figure 1b).
A karstic area is a terrain with distinctive hydrology and landforms that arise from a combination of high rock solubility and well-developed secondary porosity (Ford and Williams, 2007). Karst primarily develops in carbonate rocks such as limestone, dolomite, and marble, which possess high solubility due to their calcium carbonate (CaCO₃) composition. Karst can also form in evaporites composed of minerals such as anhydrite or gypsum, although the dissolution rates and conditions in these rocks differ from those occurring in carbonates (Klimchouk, 1996; Klimchouk and Andrejchuk, 2002; Lu et al., 2013).
Water is the triggering element for dissolution in carbonates (van Beynen, 2011), since fractures are the pathways through which precipitation enters the system; both conditions are necessary for karst to develop. However, to effectively dissolve carbonate minerals, the water must be acidic. Karstification begins in the atmosphere, where water absorbs small amounts of carbon dioxide (CO₂), and the acidity of the water increases significantly as it absorbs more CO₂ stored in the soil, originating from the decomposition of organic matter. This is a continuous process and thus a constantly evolving system.
Dissimilar flow patterns that occur in variable zones within the karst system appear to influence the characteristics and sequences of karstic dissolution forms (Hernández-Flores et al., 2021). Karstification is a relatively long-term process that modifies the surface and subsurface, creating complex forms. The evolution of the unique features of karst landscapes and their aquifers arises mainly from exogenous agents that give rise to resulting structures such as subsidences and collapse features. However, from a broader perspective, this evolution is related to a more complex interaction of various factors, such as hydrological, geological, climatic, and biological (Stokes et al., 2011).
Sapper (1989) mapped two carbonate units in the department of Alta Verapaz (see classification Ksd in Figure 1b), naming them (without defining type sections) “Cobankalke” and “Rudistenkalke”. Vinson (1962) designated a type section for the Cobán Formation and defined the Campur Formation. The division by Sapper and Vinson of the Cretaceous section of thick dolomite overlain by thin limestone has prevailed in eastern and northern Guatemala and Belize. However, in western Guatemala, lithological distinction is commonly more difficult. For this reason, Termer (1932) introduced the name Ixcoy Formation for a carbonate section in western Guatemala whose relationship with Sapper’s two units was uncertain. Anderson et al. (1973) continued using Termer’s classification in western Guatemala. The Cobán Formation is a massive dolomitic limestone that is highly recrystallized, as seen in outcrops along the southern margin of the Maya Block. Intraformational breccias are common; these are predominantly dolomitic but contain siliceous limestones and sparse shales and siltstones. Thin limestone sections tens of meters thick with little dolomite occur near the top of the Cobán. Evaporitic solution breccias are common (Blount and Moore, 1969), and several occurrences of secondary gypsum are found near outcrops of this unit.
3. Flooding caused by Eta and Iota
Eta formed on October 31 as a tropical depression, intensified into a tropical storm on November 1, and became a hurricane by November 2, rapidly strengthening to reach Category 4 intensity on November 3, when it made landfall on the coast of Nicaragua. As it moved slowly across northern Nicaragua, it continuously weakened, becoming a tropical storm on November 4 and a tropical depression on November 5 as it crossed Honduran territory. After nearly dissipating, it emerged over the Gulf of Honduras shortly before November 6, heading toward the Caribbean. Its evolution continued until it dissipated around November 13 along the northeastern coast of the United States (Pasch et al., 2021).
Iota, in turn, formed in the Caribbean Sea on November 13 as a tropical storm, gradually strengthening. On November 14, Iota began a period of rapid intensification, eventually turning westward. Between November 14 and 16, Iota continued to strengthen and moved toward Central America as a hurricane. It later weakened as it passed over the cooler waters left by Eta two weeks earlier, making landfall around November 17 on the coast of Nicaragua. As it passed through the country, it weakened into a tropical storm and, upon crossing southeastern Honduras on November 18, it became a tropical depression. Later that same day, Iota dissipated over Salvadoran territory (Stewart, 2021).
Due to the pre-existing soil saturation resulting from the intense 2020 Atlantic hurricane season, the development of Eta and Iota and their associated rainfall (see section 5.1) exceeded the recharge and discharge capacity of the karst system in northern Guatemala, causing flooding in many communities. One outstanding example was the community of Campur (Figure 2b), where floodwaters reached heights of approximately 22 meters. In early November 2020, unusual seismic activity occurred near the affected regions (Figures 1b and 2a), suggesting a relationship between the flooding and the triggering of this seismicity.
During the dry season of 2024, a field campaign was carried out to verify flood heights based on the water surface level of currently accessible sources. In other cases, the reference point was the ground level near the flood height measurement site. The flood levels were determined from natural marks identified with the assistance of residents from the visited communities. The information collected is shown in Figure 2a and in Tables 1 and 2. The sites visited were selected based on accessibility and proximity to the epicenters recorded by the RSN, and the campaign did not aim to document all the flooding events in the region.
4. Rainfall-triggered seismicity
Induced seismicity associated with artificial or anthropogenic activities is a widely studied topic, such as that generated by fluid injection into the subsurface (e.g., Keranen et al., 2013; Diehl et al., 2017; Alghannam and Juanes, 2020; Schultz et al., 2023; Silva et al., 2024) or that triggered by changes in water level in reservoirs (e.g., Talwani, 1997; Gupta, 2002; Kangi and Heidari, 2008). The generation of this type of seismicity is commonly associated with changes in pore pressure in the vicinity of the investigated sites or in more distant regions due to diffusion processes.
A less explored branch of induced seismicity involves meteorological phenomena, including seasonal variations (Hsu et al., 2021), ice melt (van Loon et al., 2016), or seismicity triggered by tidal changes at ocean ridges (Scholz et al., 2019). Within this category, a particular case of rainfall-triggered seismicity in karst environments has been identified, with three well-documented cases in Germany, Switzerland, and France (Hainzl et al., 2006; Husen et al., 2007; Rigo et al., 2008; Miller, 2008), where the triggering mechanism is considered more complex than simple pore pressure alteration and diffusion processes.
Rigo et al. (2008) propose that an abrupt vertical load on the land surface in a well-developed karst system could generate sufficient overpressure to trigger seismic activity in the upper kilometers of the crust. However, analyses carried out by Perrochet et al. (2023) of the three aforementioned cases estimate that vertical stress changes, as well as poroelastic deformation due to increasing water load, can hardly be considered as the triggering process for rainfall-induced seismicity. Instead, they propose four categories to explain the triggering mechanisms: surface loading, poroelastic changes in pore pressure associated with that load, pore pressure diffusion, and direct hydraulic connection. A key difference lies in the order of magnitude of the associated stress variations: while traditional pore pressure and diffusion analyses are in the order of kilopascals (kPa), direct hydraulic connection can reach the order of megapascals (MPa).
The geological conditions of the karst environment, particularly the connection between faults and the karst system—especially at depth—suggest which mechanism plays the most important role in triggering seismicity. According to Perrochet et al. (2023), pore pressure increases caused by surface loading and poroelastic deformation are much lower than the increase in pore pressure generated by a direct hydraulic connection and its diffusion at shallow depths. This also determines the timing of seismicity onset after the saturation of the karst environment by intense rainfall: if the seismicity begins immediately or with only a short delay of a few hours, it is likely due to a direct hydraulic connection, while if the delay is longer—ranging from hours to a few days—a coupled solution can be assumed, i.e., a rapid increase in pore pressure from a direct hydraulic connection followed by a diffusion process toward focal depths. On the other hand, Rigo et al. (2008) propose longer delays for the onset of seismicity after intense rainfall, especially when related to fluid diffusion processes in pores, which may take weeks or months to begin.
5. Seismicity triggered by Eta and Iota
In addition to the complex karst system in the departments of Alta Verapaz and Quiché, located north of the Chixoy-Polochic fault system, the region is crossed by a series of active seismic faults, some of which are well identified, such as the Ixcán and Las Conchas faults (Guzmán-Speziale, 2010; Styron et al., 2020; Guzmán-Speziale and Molina, 2022) (Figure 1b), and have been associated with some historical earthquakes (White, 1984). Moreover, the topography reveals numerous lineaments with similar east-west orientations to those of the Ixcán and Las Conchas faults (see Figures 1b and 2a), although no significant historical seismic events have been recorded along them. On the other hand, similar orientations of surface karst features (sinkholes or dolines) in the department of Alta Verapaz (Ramos et al., 2004) suggest that their distribution is closely related to these local faults and that both systems may be hydraulically connected.
Although this region is seismically active (Figure 3b), it is important to consider that historically there was no sufficiently dense seismological network to characterize this seismicity in detail. In 2020, the RSN operated with only a few seismic stations near the study area (Figure 3a), limiting the accurate detection of low-magnitude events and resulting in relatively high errors in the location of larger earthquakes.
Another challenge for hypocenter location is the lack of a specific velocity model for northern Guatemala, one that also accounts for the geological conditions of karst systems. Despite these limitations, it was possible to detect an increase in seismic activity and to characterize earthquakes with local magnitudes (Ml) between 2.6 and 4.7 that coincided with the rainfall and impacts caused by Eta and Iota at the end of 2020 (Figure 3b) in the study area.
We carried out the relocation of the hypocenters following the strategy of Franco et al. (2009), who used the velocity model proposed by Molina and Tenorio (2000) and a Vp/Vs ratio of 1.73 in a nearby study area (blue box in Figure 3a). We found that the relocated hypocenters using this velocity model were mostly located at depths of less than 10 km (Figure 3c), which are values more consistent with the depths reported by Franco et al. (2009) for earthquakes located north of the Polochic Fault using a considerably dense temporary seismic network (see Figure 4 in Franco et al., 2009).
Considering the location errors for the spatial coordinates (represented by error ellipses in Figure 1b), we have obtained a good indication of the seismic activity area. Additionally, we approximately identified three clusters of epicenters: in the central region of the seismic activity (cluster A), to the west of it (cluster B), and to the northeast (cluster C) (Figure 1b). As expected, the depth error estimates were larger (Figure 3c); nevertheless, important observations were obtained regarding the development of the seismicity triggered by Eta and Iota.
5.1. SPATIOTEMPORAL EVOLUTION OF THE SEISMICITY
The seismic station closest to most of the epicenters was the RSN short-period station AVCB, located in Cobán, Alta Verapaz (Figure 3a). In the same location, there is also a meteorological station from the INSIVUMEH National Meteorological Network, whose rain gauge recorded the rainfall during the period addressed in this study. This dual data acquisition proved advantageous for properly analyzing the spatial and temporal evolution of both the rainfall and the recorded earthquakes. We temporally classified the recorded seismicity into five episodes (Figure 4), as described below.
As shown in Figure 4a, no earthquakes were recorded in the region during October, while rainfall at the Cobán station was intermittent throughout the month, almost never exceeding 40 mm per day. At the end of October and the beginning of November, rainfall began to exceed 50 mm per day, coinciding with the first two earthquakes recorded by the RSN (blue circles in Figure 4b). This occurred simultaneously with the formation of Eta in northeastern Nicaragua. In the following days, as Eta moved through Central America, rainfall increased, reaching its peak on November 5—the day of Eta’s closest approach to Guatemalan territory (Figure 1a) (Pasch et al., 2021)—with a daily total of 173 mm and a cumulative rainfall of 535 mm. Over the next two days, rainfall decreased, and by November 8 it had ceased completely at the AVCB meteorological station. Around November 10, a notable increase in seismicity was observed, with most epicenters located in cluster A (green circles in Figure 4b). The estimated depths for the majority of these earthquakes ranged from 3 to 16 km (Figure 3c).
Rainfall resumed on November 15 at the AVCB station, two days after Iota formed in the Caribbean Sea. On November 17 and 18, Iota crossed parts of Nicaragua and Honduras before dissipating in El Salvador (Figure 1a) (Stewart, 2021). Between November 17 and 22, rainfall exceeded 60 mm per day, accumulating 375 mm (Figure 4a). Although this cumulative rainfall was lower than that recorded during Eta at this station, rainfall in other regions of the country was abundant. For example, at the Las Vegas station in the department of Izabal (near station IZAB in Figure 3a), a total of 510 mm was reported.
During the rainfall days associated with Iota, three scattered earthquakes were recorded (yellow circles in Figure 4b), and on November 22, another increase in seismicity was observed, continuing through the end of November (Figure 4a). Several epicenters continued to be located in cluster A, with events ranging in depth from a few hundred meters to 10 km, but a new sequence also began, forming cluster B in the department of Quiché (orange circles in Figure 4b), where most epicenters were relatively shallow (Figure 3c). Some earthquakes also occurred to the northeast, in cluster C.
After these episodes triggered by the approach of Eta and Iota to Guatemala, rainfall significantly decreased. However, seismicity continued through December (Figure 4a), with epicenters forming cluster C (red circles in Figure 4b). Earthquakes were also recorded south of cluster B, possibly associated with the Chixoy-Polochic Fault rather than with rainfall-triggered seismicity. In cluster C, no earthquakes occurred at depths as shallow as in the previous two clusters; estimated depths ranged from approximately 3 to 6 km (Figure 3c).
5.2. REPEATING EARTHQUAKES
Due to limitations in instrumental coverage, it was not possible to record lower-magnitude earthquakes or focal mechanisms that could provide evidence for the activation of specific geological faults, as has been demonstrated in areas of artificially induced seismicity monitored with dense seismic networks (e.g., Silva et al., 2024; Yin et al., 2025).
To explore the possibility that the seismicity within the proposed clusters could originate from defined seismic sources, we conducted a search for repeating earthquakes using the REDPy software (Hotovec-Ellis, 2016). This tool groups repeating earthquakes into families based on waveform similarity via cross-correlation, using data from multiple stations or from repeated recordings at the same station.
Since REDPy does not require prior information about repeating seismicity, we only provided the P-wave arrival times of the manually recorded earthquakes at four broadband seismic stations (APG, PETF, CHIE, and FG14) and one short-period station (AVCB) (Figure 3a). Additionally, we specified the frequency range with the highest content for the earthquakes recorded at those stations. Although most of the stations are considerably distant from the epicentral area (except AVCB), the relatively high earthquake magnitudes allowed for suitable waveform recordings for this analysis (configuration values for each station used in REDPy can be found in the supplementary material).
After analyzing the continuous records from the five seismic stations individually, we identified 10 families (see Figure 5a) with correlation coefficients above 60% (example shown in Figure 5c). In cluster A, 4 families were identified, each with only 2 members. In cluster B, 3 families were found: two with 3 members and one with 2. Cluster C showed two families: one with 2 members and another with 4, the latter being the largest family of all (see inset in Figure 5a and Figure 5b). One additional family with 2 members was identified north of the main clusters.
As shown in Figure 5a, the largest family in cluster C was independently identified by three seismic stations, and a similar case occurred with the family detected in the northern region. The significant spatial separation observed between some members may be due to epicentral location errors. Twenty-eight earthquakes were not associated with any family and were therefore classified as “orphans,” represented by black circles in the figure. Given the RSN’s detection capabilities for this sequence, it is possible that more unlocated members could be added to the identified families, or that other families associated with the orphan earthquakes may exist.
It is worth noting that the APG station experienced operational issues during the sequence and was out of service from November 20 to 27 (coinciding with the period of highest seismicity) due to problems caused by intense rainfall. As a result, only the largest family in cluster C was identified at this station, most of which occurred toward the end of the entire sequence in December.
6. Discussion and analysis
According to the distribution of epicenters, the sequence took place in a region dominated by carbonate rocks (Ksd, Figure 1b), where different geological formations form a complex karst system. Its geomorphological structures—dominated by sinkholes and underground rivers— form part of the environment inhabited by communities in the departments of Alta Verapaz and Quiché. Based on the orientation (approximately E–W) of numerous surface features observed in satellite images (e.g., Ramos et al., 2004), it is proposed that they are influenced by faulting in the region and may be hydraulically connected. The widespread flooding, which reached up to 25 meters above the base water level, clearly demonstrates that the storage and discharge capacity of the karst system was exceeded during the rainfall episodes caused by Eta and Iota.
The earthquakes that occurred in the analyzed area were scattered throughout the year (grey circles in Figure 4b) before the heavy rainfall and flooding episodes, indicating that the area is predisposed to earthquake nucleation. The stress field in the crust of the study area is not well defined. To the northwest, strike-slip focal mechanisms associated with the Ixcán fault have been documented, along with inverse mechanisms near its northwestern tip, approaching the border with Mexico (Guzmán-Speziale, 2010), suggesting a compressive or transpressional regime (Álvarez-Gómez et al., 2019). In addition, according to the World Stress Map, to the north of the Polochic fault, the maximum horizontal stress orientations (approximately NE–SW) appear consistent with these focal mechanisms (Cáceres et al., 2005; Heidbach et al., 2018). A single normal focal mechanism has been documented in the same region (Guzmán-Speziale and Molina, 2022).
Following cases of artificially induced seismicity, it has been observed that various fractures or faults with complex geometries can be activated with different rupture mechanisms (e.g., Martínez-Garzón et al., 2017; Silva et al., 2024; Yin et al., 2024). A similar behavior can be seen in the case of the earthquakes generated by flooding in Mt. Hochstaufen, Germany, where focal mechanisms with different rupture styles were found, making it difficult to define the tectonic processes associated with the seismicity (Kraft et al., 2006).
Due to the scarcity of near-field data that would allow a more precise definition of fault activation and rupture types, we focused on correlating the seismicity with rainfall timing, as well as identifying waveform similarities to discuss indications of activation of various fractures or faults in the area.
As shown in Figure 4a, and following the mechanisms proposed by Rigo et al. (2008) and Perrochet et al. (2023), the temporal relationship between torrential rainfall and increased seismicity suggests the presence of random earthquakes triggered during days of intense rainfall, which could be related to a direct hydraulic connection. Significant seismic sequences immediately following each tropical storm’s rainfall episode may suggest the combined influence of direct hydraulic connection and short-range diffusion processes. A sequence that began roughly a week after the end of rainfall from Iota could be mainly attributed to fluid diffusion in the medium. It is not ruled out that lower-magnitude earthquakes may have occurred, and that the start or end times of the different episodes may vary slightly.
A clear geographical trend in the distribution of the three seismic activity increases can also be observed (Figure 4b), consistent with the three proposed clusters in Figure 1b. The identification of families also provides insight into the occurrence of this seismicity: one family was linked to the north of the sequence, four to cluster A, three to cluster B, and two to cluster C. In cluster C, the family with the largest number of members was identified; these were also the earthquakes with the smallest hypocentral errors (Figures 1b and 3c), and they were possibly caused by fault activation due to diffusion processes.
Finally, it is worth noting that the seismicity generated in the sequence analyzed in this study involved relatively large local magnitudes (Ml 2.6 to 4.7) compared with similar sequences reported in Switzerland (1.0 to 2.4), Germany (–0.2 to 2.4), and France (1.7 to 2.8), although the depths were similar (between the surface and 5 km in all three cases), at least on average.
7. Conclusions
We associate the unusual seismicity recorded in November and December 2020 in the karstic region of northern Guatemala with the widespread flooding caused by the approach of tropical storms Eta and Iota to the country, and we explore the possible mechanisms that triggered this seismicity based on its temporal occurrence: direct hydraulic connection to the seismic sources, a coupled effect of direct hydraulic connection with short-range pore fluid diffusion, and a more delayed diffusion process.
Through waveform correlation analysis from five seismic stations, ten families were identified that may be associated with the activation of fractures or seismic faults in a tectonic environment predisposed to earthquake nucleation.
The feasibility of this karstic environment to nucleate significant-magnitude earthquakes in response to sudden and prolonged flooding clearly represents an important seismic hazard for local populations. This should be further analyzed given the potential for future scenarios of intense rainfall, exacerbated by climate change.
Supplementary data
Seismic catalog, description of REDPy, available at Zenodo bit.ly/3SWr5Zv
Contributions of authors
(1) Conceptualization: RY; (2) Data analysis or acquisition: RY, DC, EQ, FM; (3) Methodological/technical development: RY, DC, EQ, FM; (4) Original manuscript writing: RY, EQ, DC; (5) Revised and edited manuscript writing: RY, DC, EQ; (6) Graphic design: FM, DC, EQ; (7) Fieldwork: EQ, FM; (8) Interpretation: RY, DC, EQ.
Acknowledgments
Seismic records from the auxiliary seismic station APG (AS-037), part of the International Monitoring System (IMS) of the Comprehensive Nuclear-Test-Ban Treaty Organization (CTBTO), were used for the analyses.
Conflicts of interest
The authors declare no conflict of interest.
Handling editor
Carlos Alberto Vargas Jiménez.
References
Alghannam, M., & Juanes, R. (2020). Understanding rate effects in injection-induced earthquakes. Nature Communications, 11(1), 3290. https://doi.org/10.1038/s41467-020-16860-y
Álvarez-Gómez, J. A., Staller Vázquez, A., Martínez-Díaz, J. J., Canora, C., Alonso-Henar, J., Insúa-Arevalo, J. M., & Béjar-Pizarro, M. (2019). Push-pull driving of the Central America Forearc in the context of the Cocos–Caribbean–North America triple junction. Scientific Reports, 9(1), 10839. https://doi.org/10.1038/s41598-019-47617-3
Anderson, T. H., Burkart, B., Clemons, R.E., Bohnenberger, O. H., & Blount, D. N. (1973). Geology of the western Altos Cuchumatanes, northwestern Guatemala. GSA Bulletin, 84(3), 805–826. https://doi.org/10.1130/0016-7606(1973)84<805:gotwa c>2.0.co;2
Bello, O., & Peralta, L. (Eds.). (2021). Evaluación de los efectos e impactos de las depresiones tropicales Eta y Iota en Guatemala (LC/TS.2021/21). Comisión Económica para América Latina y el Caribe (CEPAL). https://repositorio.cepal.org/server/api/core/bitstreams/ff46f3cd-48cd-4c3c-b525-27c98190b7c7/content
Blount, D. N., & Moore, C. H., Jr. (1969). Depositional and non-depositional carbonate breccias, Chiantla Quadrangle, Guatemala. GSA Bulletin, 80(3), 429–442. https://doi.org/10.1130/0016-7606(1969)80[429:DAN CBC]2.0.CO;2
Bonis, S., Bohnenberger, O. H., & Dengo, G. (1970). Mapa geológico de la República de Guatemala [Map] 1:500,000. Instituto Geográfico Nacional.
Cáceres, D., Monterroso, D., & Tavakoli, B. (2005). Crustal deformation in northern Central America. Tectonophysics, 404, 119–131. https://doi.org/10.1016/j.tecto.2005.05.008
Dengo, G. (1969). Problems of tectonic relations between Central America and the Caribbean. Transactions of the Gulf Coast Association of Geological Societies, 19, 311–320. https://archives.datapages.com/data/gcags/data/019/019001/0311.htm
Diehl, T., Kraft, T., Kissling, E., & Wiemer, S. (2017). The induced earthquake sequence related to the St. Gallen deep geothermal project (Switzerland): Fault reactivation and fluid interactions imaged by microseismicity. Journal of Geophysical Research: Solid Earth, 122(11), 9037-9055. https://doi.org/10.1002/2017jb014473
Finch, R. C., & Dengo, G. (1990). NOAM–CARIB plate boundary in Guatemala: A Cretaceous suture zone reactivated as a Neogene-transform fault. Dallas Geological Society. https://archives.datapages.com/data/dgs/019/019001/1_dgs0190001.htm
Fourcade, E., Mendez, J., Azéma, J., Bellier, J.-P., Cros, P., Michaud, F., Carballo, M., & Villagrán, J.-C. (1994). Dating of the Settling and Drowning of the Carbonate Platform, and of the Overthrusting of the Ophiolites on the Maya Block during the Mesozoic (Guatemala). Newsletters on Stratigraphy, 30(1), 33–43. https://doi.org/10.1127/nos/30/1994/33
Ford, D. and Williams, P. (2007). Karst Hydrogeology. In Karst Hydrogeology and Geomorphology. Wiley https://doi.org/10.1002/9781118684986.ch5
Franco, A., Molina, E., Lyon-Caen, H., Vergne, J., Monfret, T., Nercessian, A., Cortez, S., Flores, O., Monterosso, D., & Requena, J. (2009). Seismicity and crustal structure of the Polochic–Motagua fault system area (Guatemala). Seismological Research Letters, 80(6), 977–984. https://doi.org/10.1785/gssrl.80.6.977
Gupta, H. K. (2002). A review of recent studies of triggered earthquakes by artificial water reservoirs with special emphasis on earthquakes in Koyna, India. Earth-Science Reviews, 58(3-4), 279-311. https://doi.org/10.1016/s0012-8252(02)00063-6
Guzmán-Speziale, M. (2010). Beyond the Motagua and Polochic faults: Active strike-slip faulting along the western North America–Caribbean plate boundary zone. Tectonophysics, 496(1–4), 17–27. https://doi.org/10.1016/j.tecto.2010.10.002
Guzmán-Speziale, M., & Molina, E. (2022). Seismicity and seismically active faulting of Guatemala: A review. Journal of South American Earth Sciences, 115, 103740. https://doi.org/10.1016/j.jsames.2022.103740
Hainzl, S., Kraft, T., Wassermann, J., Igel, H., & Schmedes, E. (2006). Evidence for rainfall-triggered earthquake activity. Geophysical Research Letters, 33(19), L19303. https://doi.org/10.1029/2006GL027642
Heidbach, O., Rajabi, M., Cui, X., Fuchs, K., Müller, B., Reinecker, J., Reiter, K., Tingay, M., Wenzel, F., Xie, F., Ziegler, M., Zoback, M., & Zoback, M. (2018). The World Stress Map database release 2016: Crustal stress pattern across scales. Tectonophysics, 744, 484–498. https://doi.org/10.1016/j.tecto.2018.07.007
Hernández-Flores, G., Gutiérrez-Aguirre, M. A., Cervantes-Martínez, A., & Marín-Celestino, A. E. (2021). Historical analysis of a karst aquifer: Recharge, water extraction, and consumption dynamics on a tourist island (Cozumel, Mexico). Annales de Limnologie – International Journal of Limnology, 57(16), 1–17. https://doi.org/10.1051/limn/2021013
Hotovec-Ellis, A. J., & Jeffries, C. (2016, April). Near real-time detection, clustering, and analysis of repeating earthquakes: Application to Mount St. Helens and Redoubt volcanoes [Conference]. Seismological Society of America Annual Meeting, Reno, Nevada.
Hsu, Y.–J., Kao, H., Bürgmann, 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. Science Advances, 7(16), eabf7282. https://doi.org/10.1126/sciadv.abf7282
Husen, S., Bachmann, C., & Giardini, D. (2007). Locally triggered seismicity in the central Swiss Alps following the large rainfall event of August 2005. Geophysical Journal International, 171(3), 1126–1134. https://doi.org/10.1111/j.1365-246X.2007.03561.x
Instituto Nacional de Sismología, Vulcanología, Meteorología e Hidrología (INSIVUMEH). (1976). GI: Red Sismológica Nacional [Data set]. International Federation of Digital Seismograph Networks. https://doi.org/10.7914/SN/GI
Kangi, A., & Heidari, N. (2008). Reservoir-induced seismicity in Karun III dam (Southwestern Iran). Journal of Seismolog y, 12(4), 519–527. https://doi.org/10.1007/s10950-008-9104-4
Keranen, K.M., Savage, H.M., Abers, G.A., & Cochran, E.S. (2013). Potentially induced earthquakes in Oklahoma, USA: Links between wastewater injection and the 2011 Mw 5.7 earthquake sequence. Geolog y, 41(6), 699-702. https://doi.org/10.1130/G34045.1
Klimchouk, A. (1996). Speleogenesis in gypsum. International Journal of Speleology, 25(3), 61–82. https://doi.org/10.5038/1827-806X.25.3.5
Klimchouk, A., & Andrejchuk, V. (2002). Karst breakdown mechanisms from observations in the gypsum caves of the Western Ukraine: Implications for subsidence hazard assessment. International Journal of Speleology, 31, 55–88. https://doi.org/10.5038/1827-806X.31.1.4
Kraft, T., Wassermann, J., & Igel, H. (2006). High-precision relocation and focal mechanism of the 2002 rain-triggered earthquake swarms at Mt Hochstaufen, SE Germany. Geophysical JournalInternational,167(3),1513–1528.https://doi.org/10.1111/j.1365-246X.2006.03171.x
Lu, Y., Liu, Q., & Zhang, F. (2013). Environmental characteristics of karst in China and their effect on engineering. Carbonates and Evaporites, 28(2), 251–258. https://doi.org/10.1007/s13146-013-0158-1
Martínez-Garzón, P., Kwiatek, G., Bohnhoff, M., & Dresen, G. (2017). Volumetric components in the earthquake source related to fluid injection and stress state. Geophysical Research Letters, 44(2), 800–809. https://doi.org/10.1002/2016GL071963
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
Molina, E., & Tenorio, C. (2000). An approach to a regional crustal velocity model for Central America. University of Bergen.
Pasch, R. J., Reinhart, B. J., Berg, R., & Roberts, D. P. (2021). Tropical Cyclone Report: Hurricane Eta (AL292020). National Hurricane Center, NOAA. https://www.nhc.noaa.gov/data/tcr/AL292020_Eta.pdf
Perrochet, L., Preisig, G., & Valley, B. (2023). Quantifying the mechanisms of rain-triggered seismicity in karstic regions. Frontiers in Earth Science, 11, 1234856. https://doi.org/10.3389/feart.2023.1234856
Ramos, R., Machorro, R., & Granados, P. (2004, May 25-29). Ecohydrology of a karstic terrane in northern Alta Verapaz [Conference]. BALWOIS 2004, Ohrid, FY Republic of Macedonia.
Rigo, A., Béthoux, N., Masson, F., & Ritz, J. (2008). Seismicity rate and wave-velocity variations as consequences of rainfall: The case of the catastrophic storm of September 2002 in the Nîmes Fault region (Gard, France). Geophysical Journal International, 173(2), 473–482. https://doi.org/10.1111/j.1365-246X.2008.03718.x
Sapper, K. T. (1899). Über den Gebirgsbau und Boden des nördlichen Mittelamerika. Gotha, J. Perthes. https://archive.org/details/flowersandtheir01morlgoog
Scholz, C. H., Tan, Y. J., & Albino, F. (2019). The mechanism of tidal triggering of earthquakes at mid-ocean ridges. Nature Communications, 10, Article 2526. https://doi.org/10.1038/s41467-019-10605-2
Schultz, R., Park, Y., Aguilar, A. L., Ellsworth, W. L., & Beroza, G. (2023). En echelon faults reactivated by wastewater disposal near Musreau Lake, Alberta. Geophysical Journal International, 235(1), 417–429. https://doi.org/10.1093/gji/ggad226
Silva, J. A., Khosravi, M., Yoon, H., Fehler, M., Frailey, S., & Juanes, R. (2024). Mechanisms for microseismicity occurrence due to CO₂ injection at Decatur, Illinois: A coupled multiphase flow and geomechanics perspective. Seismological Research Letters, 114(5), 2424–2445. https://doi.org/10.1785/0120230160
Stewart, S. R. (2021). Tropical Cyclone Report: Hurricane Iota (AL312020). National Hurricane Center, NOAA. https://www.nhc.noaa.gov/data/tcr/AL312020_Iota.pdf
Stokes, T. R., Ramsey, C., & Griffiths, P. (2011). Constraints for karst landscape evolution on Vancouver Island, British Columbia, Canada: Concepts, research plans and outcomes. Australasian Cave and Karst Management Association, 1–17.
Styron, R., García-Pelaez, J., & Pagani, M. (2020). CCAF-DB: The Caribbean and Central American active fault database. Natural Hazards and Earth System Sciences, 20(3), 831–857. https://doi.org/10.5194/nhess-20-831-2020
Talwani, P. (1997). On the nature of reservoir-induced seismicity. Pure and Applied Geophysics, 150(3-4), 473–492. https://doi.org/10.1007/s000240050089
Termer, F. (1932). Geologie von Nordwest-Guatemala. VEB Geogr.-kartogr. Anst.
Van Beynen, P. E. (2011). Karst management. Springer. https://doi.org/10.1007/978-94-007-1207-2
van Loon, A. J., Pisarska-Jamroży, M., Nartišs, M., Krievāns, M., & Soms, J. (2016). Seismites resulting from high-frequency, high-magnitude earthquakes in Latvia caused by Late Glacial glacio-isostatic uplift. Journal of Palaeogeography, 5(4), 363–380. https://doi.org/10.1016/j.jop.2016.05.002
Vinson,G.L.(1962).UpperCretaceousandTertiary stratigraphy of Guatemala. AAPG Bulletin, 46(4), 425–456. https://doi.org/10.1306/BC743835-16BE-11D7-8645000102C1865D
White, R. A. (1984). Catalog of historic seismicity in the vicinity of the Chixoy-Polochic and Motagua faults, Guatemala (Open-File Report No. 84-88). U.S. Geological Survey. https://doi.org/10.3133/ofr8488
Yin, X., Jiang, C., Zhai, H., Yin, F., Zheng, Y., Zhang, Y., Jiang, C., & Li, J. (2025). Seismic activity reveals the coexistence of multiple mechanisms of fault reactivation induced by hydraulic fracturing in the Gonghe EGS project in Qinghai, China. Seismological Research Letters, 96(3), 1573–1589. https://doi.org/10.1785/0220240244
Peer Reviewing under the responsibility of Universidad Nacional Autónoma de México.
This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)