The 22 April 1991 Limón, Costa Rica tsunami field survey

Estudio de campo del tsunami de Costa Rica del 22 de abril de 1991

Stuart Nishenko1*, Eduardo Camacho2, Allan Astorga3, Luis D. Morales3 & Jane Preuss4

1Pacific Gas and Electric Company, San Francisco, CA, USA

2University of Panama, Institute of Geosciences, Panama City, Panama

3University of Costa Rica (UCR), San Jose, Costa Rica

4Urban Regional Research, Seattle WA, USA

*Corresponding author: s_nishenko@msn.com

(Recibido: 12/12/2020; aceptado: 29/03/2021)

Revista Geológica de América Central, 65, 2021, doi: 10.15517/rgac.v0i65.47001

ISSN: 0256-7024

abstract: A small tsunami (tsunami magnitude, m ~ 1) was observed following the 22 April 1991 Mw 7.7 Limón, Costa Rica earthquake along the Caribbean coast of Costa Rica and Panama. This report summarizes eyewitness observations from 26 sites along 150 km of affected coast from Rio Matina, Costa Rica to Bocas del Toro, Northwestern Panama, as well as tide gauge recordings from Panama, Puerto Rico and the Virgin Islands. No emergency warnings were issued before the tsunami occurred. While little or no tsunami-related property damage was reported, two or more drownings were attributed to the tsunami. The maximum runup/wave height was reported to be 2 to 3 m, and inundation and coastal flooding extended up to 100 m inland in some locations. Both coseismic coastal uplift and the presence of fringing offshore reefs appear to have provided protection against major runup effects in some coastal areas. Elsewhere, runup effects were amplified at river mouths due to differential compaction of unconsolidated sediments. Widespread tsunami sand deposition near the Costa Rica-Panama border in 1991 is similar to deposition associated with the 7 May 1822 earthquake and tsunami.

Keywords: Caribbean Tsunamis; 1991 Costa Rica Earthquake and Tsunami; Tsunami Questionnaire; Field Records and Observations; Limón; Bocas del Toro.

Resumen: Después del terremoto de Limón, Costa Rica del 22 de abril de 1991 (Mw 7.7) se observó un pequeño tsunami (magnitud, m~ 1) a lo largo de la costa caribeña de Costa Rica y Panamá. Este informe resume las observaciones de testigos oculares de 26 sitios a lo largo de 150 km de costa afectada desde río Matina, Costa Rica hasta Bocas del Toro, noroeste de Panamá, así como mareogramas registrados en Panamá, Puerto Rico y las Islas Vírgenes. No se emitieron advertencias de emergencia antes de que ocurriera el tsunami. Aunque se reportaron pocos o ningún daño a la propiedad relacionada con el tsunami, dos o más ahogamientos fueron atribuidos al tsunami. La altura máxima de las olas fue de 2 a 3 m, y la inundación y las inundaciones costeras se extendieron hasta 200 m tierra adentro en algunos lugares. Tanto la elevación costera cosísmica como la presencia de arrecifes marinos parecen haber proporcionado protección contra los principales efectos del tsunami en algunas zonas costeras. En otros lugares, los efectos de las olas se amplificaron en las desembocaduras de los ríos debido a la compactación diferencial de sedimentos no consolidados. La deposición generalizada de arena de tsunami cerca de la frontera Costa Rica-Panamá en 1991 es similar a la deposición asociada con el terremoto y tsunami del 7 de mayo de 1822.

Palabras clave: tsunamis del Caribe; terremoto y tsunami de Costa Rica de 1991; cuestionario de tsunami; registros y observaciones de campo; Limón; Bocas del Toro.

INTRODUCTION

The 22 April 1991 Mw 7.7 Limón, Costa Rica earthquake is one of the three largest earthquakes to occur in Costa Rica during the last century. In contrast to the 1950 Ms 7.7 and 2012 Ms 7.6 Nicoya Peninsula earthquakes which were located on the seismically active Pacific coast, the 1991 Limón earthquake occurred along the Caribbean coast of southeastern Costa Rica. While the Caribbean coast of Costa Rica and Northwestern Panama is an area of relatively low instrumental seismic activity (Alvarado et al., 2017), it has been affected several times in the past 200 years by large earthquakes associated with back arc thrust faulting (Adamek, Frohlich & Pennington, 1988; Camacho & Víquez,1993; Mendoza and Nishenko, 1989; Plafker & Ward, 1992; Suárez et al., 1995).

The 1991 Limón earthquake occurred on the western part of a southwest dipping fold and thrust belt (North Panama Deformed Belt) that follows the Caribbean coast of Panama and Costa Rica (Plafker & Ward, 1992; Protti & Schwartz, 1994; Silver, Reed, Tagudin & Heil, 1990; Suárez et al., 1995). Field studies by Plafker & Ward (1992) and Denyer, Personius & Arias (1994) indicate a zone of uplift along the inner continental shelf and coast that extended ~ 70 km from north of Limón to the Costa Rica-Panama border. The maximum uplift occurred at Limón (157 cm) and turned sharply westward and inland for an unknown distance. To the southeast, the crest of the uplifted zone is located approximately 3 to 4 km offshore and the amount of uplift gradually decreases towards the Costa Rica-Panama border (Plafker & Ward, 1992).

Researchers from a number of institutions, including the University of Costa Rica, the University of Panama, the Universidad Nacional Autonoma de Mexico (UNAM), the Earthquake Engineering Research Institute (EERI), the University of California Santa Cruz (UCSC) and the U.S. Geological Survey (USGS) conducted field reconnaissance in areas of both Costa Rica and Panama affected by the 1991 Limón earthquake. An area of 15,000 km2 experienced widespread liquefaction and lateral spreading in places characterized by unconsolidated coastal plain and river deposits (EERI, 1991; Camacho & Víquez, 1994; Denyer et al., 1994; Plafker & Ward, 1992). Estimates of damage to infrastructure and lost revenues are at $500 million in Costa Rica (EERI, 1991) and $75 million in Panama (SINAPROC, Panama Civil Defense, written communication, 1992). This report presents tide gauge records and eyewitness accounts of tsunami activity along the Caribbean coast of Costa Rica and Northwestern Panama following the 1991 Limón earthquake.

TSUNAMI OBSERVATIONS

Tide Gauges

A tsunami or series of waves with heights, hmax, of 2 to 3 m were reported by eyewitnesses along the coast from Rio Matina, Costa Rica to Bastimentos Island, Panama following the earthquake (Fig. 1) indicating a tsunami magnitude, m, of ~1 (where m = log2hmax; Table 1; Iida, Cox & Pararas-Carayannis, 1967; Iida, 1970). The Cristobal tide gauge at Coco Solo, Panama (9.35° N, 79.92° W, 320 km southeast of the source zone) recorded a maximum peak to trough amplitude of 7.6 cm (3 in) 1 hour after the earthquake (Fig. 2). Approximately 3½ hours later, bubbler tide gauges in Magueyes Island, Puerto Rico (18.22° N, 67.16° W) and Limetree Bay, St. Croix, Virgin Islands (17.69° N, 64.75° W) both recorded disturbances of about 7 cm peak to trough (Fig. 2). Appendix 1 in the Electronic Supplement contains copies of the three tide gauge records.

Tsunami Questionnaire

In many cases, little if any systematic data collection occurs for tsunamis that have not been accompanied by extensive damage and loss of life, or where the damage has been limited to the immediate area of generation. As a result, a great deal of perishable information related to the local characteristics and behavior of these smaller tsunamis has been lost. To develop a framework for systematic data collection, a questionnaire was designed to facilitate the analysis of smaller events associated with earthquake generated tsunamis or other unusual wave activity related to submarine landslides or volcanic eruptions (Nishenko, 1987). A prototype tsunami questionnaire was field tested in the Monterey Bay, California region following the 1989 Loma Prieta, California (Ms 6.9) earthquake (Preuss & Nishenko, 1992) and was subsequently revised and simplified.

Tsunami questionnaires were also distributed following two later tsunamis in which reported damage was slight - the 1992 Petrolia, California (Ms 7.0) and 1992 Nicaragua (Ms 7.2) earthquakes (Preuss, 1994). The results of these surveys demonstrated the utility of a readily available instrument to disseminate immediately after the event.

The prototype tsunami questionnaire was translated into Spanish (Fig. 3) and distributed in Costa Rica with the help of personnel from the University of Costa Rica approximately one week after the 1991 Limón earthquake. Fourteen eyewitness observations were obtained from 12 sites along 90 km of affected coast, from the mouth of the Matina River, north of Limón to Gandoca, north of the Costa Rica-Panama border. Researchers from the University of Panama Institute of Geosciences recorded observations from 14 additional sites in Panama in the fall of 1991 (Camacho, 1994).

The questions asked during the Costa Rica and Panama field surveys (Fig. 3) were organized into three primary categories:

● Data pertaining to the respondent. Questions 1-5 ask the respondents whether they actually witnessed the tsunami or wave, where they were, and what they were doing at the time.

● Data pertaining to the tsunami. Questions 6-9 ask for information about the water activity (the type of disturbance, its effect, and the direction of wave behavior) and the respondent’s reaction.

● Data pertaining to damage caused by the tsunami. Question 10 asks for information about building and structural damage caused by the waves and the location of these structures relative to the shoreline.

● Additional Information. Questions 11-13 ask for additional information not covered above as well as information related to injuries, fatalities, and their circumstances; whether the respondent received a community tsunami alert or warning; and any response to the alert or warning.

SUMMARY OF EYEWITNESS OBSERVATIONS

Eyewitness accounts of tsunami activity in Costa Rica and Panama following the 1991 Limón earthquake are summarized below. Figure 1 shows the survey site locations and Table 2 lists a summary of the questionnaire responses at those sites. Tsunami runup and coseismic uplift measurements from Plafker & Ward (1992) are also listed in Table 2 as a reference. This report supersedes an earlier, unpublished version (Nishenko et al., 1994) that has been referenced by O’Loughlin & Lander (2003) and the NGDC/WDS Global Historical Tsunami Database (2021). The survey information is also contained in a Google Earth KMZ file located in Appendix 2 of the Electronic Supplement.

All of the observers in Costa Rica and Panama noted that the sea was calm prior to the earthquake (1557 local; 2157 Coordinated Universal Time, UTC) and that no emergency warnings were issued before the tsunami occurred. Depending on location, the sea initially receded 20 to 500 m from the shoreline. The exposure of recognizable features such as reefs and sandbars provided a basis for a more quantitative estimate of the amount of withdrawal in places.

Following the initial retreat, two principal types of runup behavior were observed. The first consisted of a 2 to 3 m high wave which advanced rapidly to the beach. Runup varied according to location and ranged from 10 to 100+ m inland. This behavior was observed at the mouths of major rivers in the area, including Boca Matina (#1) and Bocas del Pantano (#2). Waves were noted to have overtopped 3 m high dikes at the mouth of the Moin River (#5) and penetrated 1 km upstream at the Estrella River (near #7). Widespread liquefaction along the coast and compaction-induced subsidence of poorly consolidated sediments appear to have locally amplified runup effects at the mouths of large rivers (EERI, 1991; Denyer et al., 1994). Large waves were noted along beaches at Puerto Viejo (#8), Punta Uva (SE sector) (#10), Manzanillo (#11), and Gandoca (#12). While no significant damage resulting from the runup was reported by observers, two drownings are reported to have occurred in the canal near Matina (#1) due to the sudden increase in water height. A third drowning has also been documented near Moin (Chacón-Barrantes, personal communication, 2021). Large waves and fast currents were also noted at sites facing the Caribbean in Bocas del Toro, Northwestern Panama. Onshore flooding was noted at Playa de Julio Abrego (#15, Fig. 4), Tiribibi Point (#16, Fig. 5), and Lime Point (#17). Elsewhere, observers described the water activity as a series of strong currents (West Knapp (#18), Ground Creek (#19) and Nancy Key (#23)). Sand deposition by the tsunami was also noted by the respondents at many locations (12 Millas de Moin (#3), Westfalia (#6), Manzanillo (#11), and Gandoca (#12)). At the San-San Natural Refuge (#13, #14), a layer of sand, approximately 1 m thick was deposited on the beach creating dunes and covering turtle nests (Fig. 6).

The second type of runup consisted of a gradual increase in water level without a large wave. In some locations, the sea returned an hour after the initial retreat. This behavior was noted both in areas that experienced coastal uplift (Moin (#5), Limon (Fig. 7), and Cahuita (#7)) and in areas surrounded by fringing reefs (Limón-Moin, Cahuita, Manzanillo, Puerto Viejo in Costa Rica, and Boca del Drago in Panama) which appear to have acted as natural wave barriers or breakwaters (Fig. 1). One observer in Cahuita noted that the water returned to a level that was lower than before the earthquake. The coseismic coastal uplift reported at Cahuita is 38 cm (Plafker & Ward, 1992). The maximum coseismic uplift offshore Cahuita (#7) and Manzanillo (#11) may be comparable to, or larger than, the uplift north and west of Limon (125-140 cm) based on measured tsunami runup heights (Plafker & Ward, 1992).

DISCUSSION

The 1991 Limón tsunami survey demonstrated the utility of having a readily available questionnaire to distribute immediately after the event to capture perishable information related to the local characteristics and behavior of smaller tsunamis. Eyewitness observations from 26 sites along 150 km of affected coast, from Rio Matina, Costa Rica to Bocas del Toro, Northwestern Panama described a variety of wave and runup behavior. The maximum wave height is estimated to be 2 to 3 m, and the inundation and flooding extended between 10 and 150 m inland depending on local conditions. Coseismic uplift of the coast and the presence of fringing reefs minimized the impact of flooding in a number of coastal areas. Elsewhere, tsunami runup effects were amplified at other locations from the differential compaction of unconsolidated sediments at river mouths.

The 1991 earthquake and tsunami demonstrated the vulnerability of the coastal region to strong ground shaking, coseismic uplift and flooding. Local communities and economies were affected by limited coastal access due to significant bridge and roadway damage (Figures 8 and 9; EERI, 1991). While little tsunami-related property damage was reported, two or more drownings occurred at the canal near Matina and near Moin. No emergency warnings were issued before the tsunami occurred. Preparedness programs can educate residents to run to higher elevations when they feel strong ground shaking along the coast – even in the absence of a siren or other warning.

Information about the behavior of recent smaller tsunamis provides a valuable context for interpreting accounts of earlier Caribbean earthquake-related tsunamis. The deposition of sand near the mouth of the San-San River in 1991 is similar to descriptions of tsunami sand deposited on the coast near Gandoca and Monkey Point, near the Costa Rica-Panama border, during the 7 May 1822 (Ms 7.5+) earthquake and tsunami (Roberts & Irving, 1827). Flooding was also noted to have occurred in bays and the mouths of rivers along the Caribbean coast of Costa Rica (González-Víquez, 1910). Farther east, the 7 September 1882 earthquake (Ms 8.3), located in the North Panama Deformed Belt offshore of the San Blas coast of Northeastern Panama, created a tsunami with 2 to 3 m high waves, great destruction, and nearly 100 deaths on the islands of the San Blas archipelago. The French Panama Canal Company tide gauge near Colon recorded a maximum height of 62 cm (Víquez and Toral, 1986; Mendoza and Nishenko, 1989). The April 26, 1916 (M 6.9) Almirante, Panama earthquake also created a small local tsunami that is reported to have covered the southern tip of Carenero Island in the Bocas del Toro Archipelago minutes after the event (Reid, 1917; Duda, Khattri, Purcaru &Schick, 1990).

Hurricane-related flooding and storm surge occurs more frequently than tsunami flooding along the Caribbean coast of Costa Rica and Panama, yet both can produce comparable flooding and coastal impacts. Addressing these common hazards through identification of coastal flooding and tsunami evacuation areas, the marking of evacuation routes, and implementation of coastal building/land use development guidelines can help local communities to become more resilient in the future.

ACKNOWLEDGEMENTS

We thank the United States Geological Survey (USGS) for providing funds to perform the field study in Costa Rica; the Geophysical Commission of the IPGH and CEPREDENAC for providing funds to perform the field work in Bocas del Toro; the Servicio Aereo Nacional (SAN) for logistical support in Bocas del Toro, Panama; J. Espinoza of the Panama Canal Commission and Leonard Hickman of the National Ocean Survey for providing copies of the tide gauge records. Conversations with and help from Silvia Chacón-Barrantes, Serkan Bozkurt, James Dewey, Alvaro Espinosa, Carlos Mendoza, Steve Personius, and George Plafker is gratefully appreciated. Due to the delay in the publication of these survey results, the opinions and conclusions expressed herein may not necessarily represent those of the USGS.

REFERENCES

Adamek, S., Frohlich, C. & Pennington, W.D. (1988). Seismicity of the Caribbean-Nazca Boundary: constraints on microplate tectonics of the Panama region, Journal of Geophysical Research, 93, 2053-2075.

Alvarado, G. E., Benito, B., Staller, A., Climent, Á., Camacho, E., Rojas, W., Marroquín, G., Molina, E., Talavera, J.E., Martínez-Cuevas, S. & Lindholm, C. (2017). The new Central American seismic hazard zonation: Mutual consensus based on up to day seismotectonic framework, Tectonophysics, 721, 462–476. https://doi.org/10.1016/j.tecto.2017.10.013

Camacho, E. (1994). El tsunami del 22 de Abril de 1991 en Bocas del Toro, Panama, Revista Geólogica de América Central, Special Volume: The Limón Earthquake, April 22, 61-64.

Camacho, E. & Víquez, V. (1994). Licuefacción y hundimientos costeros en el noroeste de Panamá durante el Terremoto de Limón. Revista Geólogica de América Central, Special Volume: The Limón Earthquake, April 22, 133-138.

Camacho, E., & Víquez, V. (1993). Historical seismicity of the North Panama deformed belt, Revista Geólogica de América Central, 15, 49-64.

Denyer, P., Personius, S. F. & Arias, O. (1994). Generalidades sobre los effectos geologicos del terremoto de Limón, Revista Geólogica de América Central, Volumen Especial: Terremoto de Limón, 22, 29-38.

Duda, S. K., Khattri, N., Purcaru, G. & Schick, R. (1990). Weichert seismograms of the Gottingen Observatory – microfilming and reinterpretation, Gerlands Beitr. Geophysik, 99, 313-336.

GEBCO Compilation Group 2020 (2020). The GEBCO_2020 Grid - a continuous terrain model of the global oceans and land. British Oceanographic Data Centre, National Oceanography Centre, NERC. doi:10/dtg3

González-Víquez, C. (1910). Temblores, terremotos, inundaciones y erupciones volcánicas en Costa Rica 1608-1910.Tipografia de Avelino Alsina.

Earthquake Engineering Research Institute [EERI] (1991). Costa Rica earthquake of April 22, 1991 Reconnaissance Report, Earthquake Spectra, 7 (Suppl. B), 1-127.

Iida, K. (1970). The generation of tsunamis and the focal mechanism of earthquakes. In W. M. Adams (Ed.), Tsunamis in the Pacific Ocean (pp. 3-18). East-West Center Press.

Iida, K., Cox, D. C. & Pararas-Carayannis, G. (1967). Preliminary catalog of tsunamis occurring in the Pacific Ocean, Data Rep. 5, HIG-67-10. Hawaii Institute of Geophysics, University of Hawaii.

Mendoza, C. & Nishenko, S. (1989). The North Panama earthquake of 7 September 1882: evidence for active underthrusting. Bulletin of the Seismological Society of America, 79, 1264-1269.

National Geophysical Data Center / World Data Service (NGDC/WDS) (2021). Global Historical Tsunami Database. NOAA National Centers for Environmental Information. doi: 10.7289/V5PN93H7

Nishenko, S. P. (1987). A questionnaire for rapid tsunami damage assessment. In E. N. Bernard (Ed.), Proceedings of the International Tsunami Symposium, IUGG, August 18 to 19, 1987 (pp. 36-48). NOAA/PMEL Contribution 1041, U.S. Dept. of Commerce.

Nishenko, S. P., Camacho, E., Astorga, A., Morales, L. D. & Preuss, J. (1994). The 1991 Limon, Costa Rica tsunami. Unpublished manuscript.

O’Loughlin, K. F. & Lander, J. F. (2003). Caribbean Tsunamis: A 500-Year History from 1498-1998. Kluwer Academic Publishers.

Plafker, G. & Ward, S. (1992). Backarc thrust faulting and tectonic uplift along the Caribbean Sea coast during the April 22, 1991 Costa Rica earthquake. Tectonics, 11, 709-718.

Preuss, J. (1994). Field Report: Investigation of the September 2, 1992 Nicaragua tsunami. University of Colorado Boulder, Institute of Behavioral Sciences, Natural Hazards Research and Applications Information Center, Quick Response Research Report 66, Boulder, CO 16 pp.

Preuss, J. & Nishenko, S.P. (1992). The National Earthquake Information Center Tsunami Questionnaire. EOS (Transactions, American Geophysical Union), 73, 267.

Protti, M. & Schwartz, S. Y. (1994). Mechanics of back arc deformation in Costa Rica: Evidence from an aftershock study of the April 22, 1991, Valle de la Estrella, Costa Rica, earthquake (Mw=7.7). Tectonics, 13(5), 1093–1107. https://doi.org/10.1029/94TC01319

Reid, H. F. (1917). Note on the earthquake at Almirante, Republic of Panama, in April 1916. Bulletin of the Seismological Society of America, 7, 27-30.

Roberts, O. W. & Irving, E. (1827). Narratives of voyages and excursions on the east coast and in the interior of Central America; describing a journey up the river San Juan, and passage across the lake of Nicaragua to the city of Leon: pointing out the advantages of a direct commercial intercourse with the natives. Constable & Co.

Silver, E. A., Reed, D. L., Tagudin, J. E. & Heil, D. J. (1990). Implications of the North and South Panama Thrust Belt for the origin of the Panama Orocline, Tectonics, 9, 261-281.

Suárez, G., Pardo, M., Domínguez, J., Ponce, L., Montero, W., Boschini, I. & Rojas, W. (1995). The Limón, Costa Rica earthquake of April 22, 1991: Back arc thrusting and collisional tectonics in a subduction environment. Tectonics, 14, 518-530.

Víquez, V. & Toral, J. (1986). Sismicidad historica sentida en el Istmo de Panama. Geofisica, 27, 135-165.