11Montero.qxd 477 ANNALS OF GEOPHYSICS, VOL. 47, N. 2/3, April/June 2004 Key words Central America – historical earth- quakes – catalogues 1. Introduction Central America has an important seismic and volcanic activity due to the interaction of several plates (fig. 1). That activity has conditioned its economic and social development. In order to de- fine the seismic hazard of this region, a necessary step is to collect seismicity data by means of «earthquake catalogues». In order to have a better understanding of the different earthquake sources and its activity rates, historical earthquake infor- mation must also be included. For this reason, during the last two decades important research ef- forts have been conducted in historical seismicity (pre-instrumental earthquakes) and on developing a digital catalogue for this region. Before 1980 there were some valuable re- gional and national catalogues, although the in- formation they contain was scattered in differ- ent reference sources. Also, modern techniques were not applied to integrate the regional pre- 1900 earthquake information in a systematic and methodical way. However, since 1980 there has been a remarkable increase in the historical seismicity investigation at regional and local levels, and its results have been partially incor- porated in a digital catalogue. In the next para- graphs we review the previous work with espe- cial emphasis in the historical seismicity re- search topic (pre-1900 earthquake studies). 2. Pre-1980 descriptive earthquake catalogues Research and compilations of the Central American destructive earthquakes which oc- curred since the Spanish Conquest in the 16th century up to 1900 are found in different works. Current knowledge on the Central America historical seismicity: an analysis of recent catalogues Walter Montero and Giovanni Peraldo Escuela Centroamericana de Geología, Universidad de Costa Rica, San Pedro, Costa Rica Abstract Historical seismicity studies in Central America have increased since 1980. Currently, there is a well-organized regional digital catalogue database structured according to modern techniques. It contains more than 17 000 earthquakes which occurred between 1520 and 1992. A regional tsunami catalogue is also available. The histor- ical seismicity research included a more complete earthquake documentation obtained from foreign and Central American depositories. New intensity data point estimates and isoseismal maps allows among others, macro- seismic magnitude calculations and seismogenic source determinations, for many pre-1900 earthquakes. Future trends in Central America historical seismicity research are discussed. Mailing address: Dr. Walter Montero, Escuela Cen- troamericana de Geología, Universidad de Costa Rica, Apartado 214-2060 UCR, San Pedro, Costa Rica; e-mail: wmontero@geologia.ucr.ac.cr 478 Walter Montero and Giovanni Peraldo At the Central American level, the principal compilations came from Montessus de Ballore (1888), Sapper (1925), Díaz (1933) and Grases (1974). At national level we mention González (1910) for Costa Rica; Lardé (1960) and Mar- tinez and Maximiliano (1978) for El Salvador and Kirkpatrik (1920) for Panamá. All the pre- vious works were descriptive catalogues com- piling historical earthquake information, but they do not include intensities or isosesismal estimates or define focal parameters obtained from macroseismic information. 3. Current earthquake and tsunami catalogues Currently, descriptive and parametric cata- logues, and a mixture of both are found in Central America. Leeds (1974) compiled for Nicaragua, the first published parametric catalogue including pre-1900 earthquake information for a Central American country. After this publication, other parametric catalogues were compiled. Sutch (1981) published one for Honduras; White (1984) produced one for the Polochic-Motagua fault sys- tem of Guatemala; Miyamura (1980) and Mon- tero (1989) published catalogues for Costa Rica and Víquez and Toral (1987) published one for Panamá. The previous catalogues also included instrumental seismicity. Rojas (1993) and Rojas et al. (1993) carried out the main seismological digital parametric catalogue currently existing at the regional level. They included information of around 17 000 earthquakes which occurred between 1505 and 1992. This catalogue gathered seismological in- formation from the previous catalogues and oth- er diverse reference sources, although no archival or any specific earthquake investigation was conducted. The catalogue merged the in- strumental and macroseismic information apply- ing the Standard Nordic Format, which includes three files: 1) The Main Catalogue file includes the main focal parameters of the event, as well as the possibility to include several entrances for any event. It includes instrumental and macro- seismic event information. 2) The Catalogue file is a bridge that connects files 1 and 3. 3) Exten- sional files include additional information of the event, such as macroseismic data and references, among others. For some earthquakes intensity data points are included. Besides the parametric catalogues, national and regional descriptive or mixed catalogues were compiled. Feldman (1985, 1986 and 1993) gath- ered from diverse archives a large number of new primary documents about destructive earthquakes which occurred in Guatemala and in Central America between 1500 and 1899. He compiled descriptive catalogues, not including intensity da- ta or isoseismal maps. Peraldo and Montero (1999) published a mixed earthquake catalogue including informa- tion obtained from archives and other different reference sources. They studied Central Ameri- ca destructive earthquakes which occurred from 1469 to 1899 and obtained new seismological interpretations such as intensity data points, isoseismal maps and macroseismic focal pa- rameters. Figure 1 shows the epicentre loca- tions and the suggested seismogenic sources of the historical earthquakes studied by Peraldo and Montero (1999). In relation to 20th century Central America seismicity studies, we briefly review two impor- tant recent works, among other valuable specific or national studies. We mentioned the work of Ambraseys and Adams (2001) who gathered in- strumental and macroseismic information on Central American earthquakes which occurred during the period 1898-2000. The study includes a regional descriptive and parametric catalogue, and maps with macroseismic locations but with- out intensities estimations. They also obtained new magnitude calculations for a considerable number of events. On the other hand, White and Harlow (1993) studied the upper crust Central American earthquakes which occurred since 1900. They obtained magnitudes for several of the shocks and compiled isoseismal maps for some damaging earthquakes. An important number of earthquake specific studies are also found in the literature but they will not be reviewed in this ar- ticle. For a general overview see for instance Per- aldo and Montero (1999) for pre-instrumental seismicity and Ambraseys and Adams for 20th century seismicity. The occurrence of tsunamis is another kind of earthquake hazard related phenomenon in 479 Current knowledge on the Central America historical seismicity: an analysis of recent catalogues Central America. A recent tsunami catalogue was compiled for this region (Molina, 1997; Fernández et al., 2000). 4. Scope, professional background and financial research agencies since 1980 As a result of the research carried out in his- torical seismicity of Central America since 1980, now there are improved and more complete cata- logues. They constitute the fundamental tool to obtain better seismic hazard estimates. Seismolo- gists and historians, working in separate or in in- terdisciplinary ways carried out the investigation. However, according to our experience, a previous work of interdisciplinary investigation, including the analysis of the historical context where the seismic event took place, favors better seismolog- ical interpretations. Fig. 1. Spatial-temporal historical seismicity along the different Cocos-Caribbean plate boundary convergent segments (blank circle events). Also shown are the epicentre locations related to other regional seismogenic sources. The large number of earthquakes shown for Northern Central America is explained by its early colo- nization and its relatively larger development during the Colonial period. The scarce seismic activity of the Caribbean side of Central America is also shown (modified after Peraldo and Montero, 1999). 480 Walter Montero and Giovanni Peraldo Most of the investigation performed by his- torians has been focused toward finding new primary information in diverse archival sources. We here highlight the work carried out by the historian Feldman (1993) at the regional level, investigation that received financial support of the Agency for International Development (AID) and of the United States Geological Sur- vey (USGS). Also, the historians Hilje and Al- varez (1994) carried out a similar research in Costa Rica, work financed by the Pan-Ameri- can Institute of Geography and History (PIGH). In 1992, the geologist Peraldo and the histo- rian Claudia Quirós gathered primary historical Central American seismological information from different archives. It was used together with other documents and references in the seismological interpretations carried out by Peraldo and Montero (1999). The historical context of the seismic events was considered in their interpretations. The previous investigators received financial support from the PIGH through a project financed by the International Development Research Centre of Canada (IDRC). Also, they obtained local support from the University of Costa Rica (UCR). The seis- mological analysis carried out by these authors is somehow extensive and focused on studying most of the pre-instrumental destructive region- al events. The digital regional parametric catalogue merged by the seismologists Rojas (1993) and Ro- jas et. al. (1993) was supported by the Norwegian Agency for International Cooperation (NORAD). The 20th century seismicity study carried out by the seismologists Ambraseys and Adams (2001) was financed by the National Scientific Coopera- tion Programme of Brussels (CEC) and the Inter- national Seismological Centre of England (ISC). Table I summarizes the characteristics of the recent regional catalogues. At the national level, Peraldo and Montero (1994) using primary and secondary sources studied the Colonial seismicity of Costa Rica. This work, that includes seismological interpre- tations, is considered quite intensive, because the archival sources were intensely reviewed, although only few Costa Rican colonial time Table I. Main characteristics of Central America seismological catalogues. Authors Year Type Period Catalogue type Format IDP or intensity map Intensity Scale Financial Agency Rojas 1993 Parametric 1505- 1992 Compiled seismological information D Partial- ly MM NORAD, UCR Feldman 1993 Descriptive 1500- 1899 Primary Historical information P No AID, USGS Peraldo and Montero 1999 Descriptive and parametric 1469- 1898 Mixed P Both MM PIGH, IDRC, UCR Ambraseys and Adams 2001 Descriptive and parametric 1898- 1995 Summary of seismological information P No CEC, ISC D = digital, P = printed; IDP = Intensity Data Point; MM = Modified Mercalli. 481 Current knowledge on the Central America historical seismicity: an analysis of recent catalogues earthquakes were documented. The University of Costa Rica supported this work. In Panamá, Camacho and Víquez (1993) studied the histor- ical seismicity of the North Panamá Deformed Belt, a seismic source area located on the Caribbean side of Panamá and Costa Rica (fig. 1). The Pan-American Institute of Geography and History and the University of Panamá sup- ported this research. 5. Primary information sources The documents found in archives and li- braries including seismic phenomena descrip- tions from the Spanish time up to 1821 (1903 for Panamá as an independent republic), are chronicles, reports and correspondence elabo- rated by conquerors, governors, ecclesiastical authorities, among others. After 1821, Munic- ipal and Government archive documents, ec- clesiastical reports, newspapers reports, and naturalist and travellers’ chronicles, especially by foreign people, also record damaging earth- quakes that affected the Central American re- publics. That historical data baggage consti- tutes the basic primary information with which the diverse seismological Central American catalogues have been elaborated since the end of the 19th century. The primary source reports have been obtained from archives and libraries of Europe and America. They include, among others, the Archivo Gen- eral de Indias and of Simancas, Museo Naval of Madrid and Biblioteca Nacional (Spain), Archivo General de la Nación (México), Archivo General de Centro América (Guatemala), Archivo Nacional and Archivo Eclesiástico of León (Nicaragua), Archivo Nacional and Archivo Eclesiástico of the Cu- ria Metropolitana of San José (Costa Rica). Let us briefly review the depositories re- searched at the regional level. Feldman (1993) carried out a profuse investigation in different archives and libraries, including several of those mentioned previously. Peraldo and Mon- tero (1999) carried out their investigation based on documents obtained from archives of Spain and Central America, and compiled other pri- mary and secondary documentary sources. Am- braseys and Adams (2001) gathered historical and seismological information published by di- verse authors. At the national level, the depositories re- searched by Hilje and Alvarez (1994) and by Peraldo and Montero (1994) were the Costa Ri- can archives. Camacho and Víquez (1993) re- searched the Canal of Panamá Library, the Archivo Nacional and the Biblioteca Nacional of the Republic of Panamá, the Biblioteca Simón Bolivar of the University of Panamá and the Archivo Nacional and the Biblioteca Na- cional of the Republic of Colombia. 6. Intensity scales used The scale used to estimate the levels of in- tensity in the diverse studies carried out on his- torical seismicity in Central America has been the Modified Mercalli intensity scale (MM). Exceptionally, the intensity scale MSK (Medvedev-Sponheuer-Kárník) or its later modifications was applied. Rojas (1993) gath- ered some historical earthquake MM intensity data in digital format. Peraldo and Montero (1999) assigned MM intensity levels to Central American earthquakes and drew isoseismal maps when enough estimates of intensity exist- ed. Otherwise, they only showed maps with in- tensity data points, as is the case of most of the earthquakes of the Costa Rican colonial period studied in Peraldo and Montero (1994). The MM intensity scale was also used by Camacho and Víquez (1993) and by White and Harlow (1993), among others. 7. Catalogue completeness According to Rojas et al. (1993), the Cen- tral American catalogue is reasonably complete for earthquakes of Ms ≥ 5.5 from 1900 and of Ms ≥ 7.0 from around 1820. Ambraseys and Adams (2001) consider their catalogue com- plete for earthquakes of magnitudes Ms > 5.0 starting from 1920 and of Ms > 6.0 starting from 1910. Table II shows the earthquake num- ber per century included in different catalogues. Figure 2 shows the historical accumulative 482 Walter Montero and Giovanni Peraldo earthquake number found in Peraldo and Mon- tero (1999). A remarkable slope increment in the seismicity reported is shown around the be- ginning of the 18th century, probably related to the consolidation of the Spanish colonial period and accordingly, to a better documentation of the historical earthquakes. 8. Difficulties and solutions in historical research Several difficulties were found during the historical seismicity research and among these we emphasize: i) Manuscript reading difficul- ties related to the grammatical and calligraph- ic language evolution, especially from the old to the modern Spanish. This problem has been partially solved with the participation of histo- rians trained in paleography or in the case of professionals coming from other fields, by learning paleographic techniques; ii) Physical deterioration of the documents caused by im- proper conservation, humidity damage, ink scattering, writing fading and paper destruc- tion caused by fungi and diverse animals; iii) Document losses due to such diverse causes as fires and robberies; iv) Document misadminis- tration including inadequate classification and improper organization of the historical materi- al and inappropriate loan practices; v) Scat- tered archival information, in some cases oc- curring in different document series in the same archive or between distinct depositories, many of them difficult to access; vi) Changes in the nomenclature of towns and localities or wrong geographical locations; vii) Physical disappearance or seat change of populations; viii) Earthquake dating errors included in pri- mary and secondary reference sources. An ex- ample of an error in the event occurrence date is found in González (1910), who based on a municipal record, mentioned an earthquake which occurred on 3 April 1827, in NW Costa Rica. However, a primary source located by Peraldo (1993) in the Government File of the Costa Rican National Archive referred the oc- currence of an earthquake on 3 April 1826, in the same region. This leads Peraldo (1993) to propose that the earthquake occurred in 1826 instead of 1827, something sustained by the new investigations carried out by Montero and Peraldo (work in progress). For instance, a municipal record dated 4 April 1826, from Alajuela town, located east of the epicentral region, explicitly mentions the occurrence the day before of a strong earthquake that caused some damage. Table II. Number of earthquakes included in the seismic catalogues. Authors Type Period 16th c. 17th c. 18th c. 19th c. 20th c. Total Rojas (1993) Parametric 1505-1992 92 300 16 698 17 188 Feldman (1993) Descriptive 1500-1899 37 45 66 217 365 Peraldo and Montero (1999) Descriptive and parametric 1500-1899 16 22 65 70 173* Ambraseys and Adams (2001) Descriptive and parametric 1898-1995 – – – 6 1823 1829 (*) This study investigated the earthquakes with intensities MM larger than VI. 98 483 Current knowledge on the Central America historical seismicity: an analysis of recent catalogues An important aspect that must be analyzed carefully in any historical earthquake research is the historical conditions prevailing at the time the documents were written, including the spe- cific political and social situation. Examples are the August and September 1717 and the July 1773 seismic sequences, that affected the city of Antigua, Guatemala (Peraldo and Montero, 1996). In both cases, a series of documents sug- gest widespread city destruction allowing some authorities to propose its relocation. On the oth- er hand, another group of documents refers ma- jor damage although with remarkable excep- tions and in this case other authorities argue that the city should remain in the same location. In this case two opposed political factions existed, with different concerns about the necessity or not of the city relocation, partially because they had different property damage level assess- ments. Both theses also reflected a power fight between Clergyman and the «Real» authorities. The identification and opposition of the differ- ent document discourses and their historical ex- planation allowed a better estimate of the dam- age level produced by those earthquakes. 9. Current knowledge status and future work The seismic history before the Spanish ar- rival in Central America shows an information gap because only few documentary sources have been identified. The principal ones are the Popol Vuh, the Memorial de los Cackquiqueles, and the interpretation of indigenous narratives, although they do not necessarily correspond to the pre-Hispanic period (Peraldo and Montero, 1994; 1999). A possible solution to expand the historical seismicity record to the pre-Spanish epoch would be to increase paleoseismological investigations with the objective to identify and date paleoearthquakes. Another way would be to begin research in seismic archaeology, be- cause it is possible that some earthquakes af- fected the physical and social structure of some Central American indigenous cultures, espe- cially those from the Mayan period. In relation to the seismicity which occurred between 1500 and 1900, a first investigation phase in future research would be to evaluate the information contained in the catalogue of Fig. 2. Temporal variation of the accumulative number of events according to Peraldo and Montero (1999). The graph incorporates only well documented seismic events. The absence of appropriately documented damaging earthquakes is evident for the period between 1500 and 1700. Some periods of apparent seismic quiescence are identified at regional level between 1540 and 1565, 1594 and 1607 and 1625 and 1648. Starting from the 17th century the slope increases, indicating a better documentation due to the consolidation of the Spanish coloniza- tion process. Also, the periods of apparent seismic quiescence shorten. 484 Walter Montero and Giovanni Peraldo Rojas (1993). A necessary step will be to up- date that catalogue with the new historical seis- mological information that has been generated since 1993 at regional and national levels. By doing so, the seismic parameters information incorporated in Peraldo and Montero (1994, 1999), Ambraseys and Adams (2001) and other newly published works including some inten- sive studies of earthquakes, can be incorporated in that catalogue. A second phase would consist in carrying out a systematic and careful evaluation of the historical records gathered in the diverse archives, libraries and reference sources ex- plored. A good approach to proceed can be found in Guidoboni and Stucchi (1993), where very useful guidelines are found to review the historical documentation. With regard to the primary documentation of earthquakes from the conquest up to the end of the 20th century, it is considered that the investigation carried out until now by the diverse authors has been in general extensive and not intensive. The next objective will be the identification of the most important destructive events that still have not been studied intensively, in order to proceed to its investigation in promissory lo- cal, national and foreign little explored archives. One important step of any future historical intensive investigation should be to increase the completeness of the Central America seismic catalogue, including the search for new events and a better documentation of those not well known. Finally, the procedures followed in esti- mating earthquake parameters from macroseis- mic data must be reviewed and some homoge- neous and robust criteria must be identified to obtain well-supported macroseismic earth- quake parameters. Finally, seismicity studies for the 20th cen- tury would benefit from macroseismic studies, such as defining intensities and isoseismals for specific damaging earthquakes, because this important research tool can better constrain the earthquake focal parameters. This is especially true for events which occurred during the first six decades of that century, because some dam- aging shocks before 1960 had a magnitude lev- el below the world seismographic network de- tection capacity. For that reason, their instru- mental magnitudes were poorly or not deter- mined. Another problem is that the instrumen- tal epicentres were poorly constrained due to the scarce coverage of the world seismograph- ic network before 1960. Acknowledgements We are grateful to Virginia García Acosta, Massimiliano Stucchi and Paola Albini for inviting us to write this paper. REFERENCES AMBRASEYS, N.N. and R.D. ADAMS (2001): The Seismicity of Central of America: a Descriptive Catalogue 1898- 1995 (Imperial College Press, London), pp. 309. CAMACHO, E. and V. VÍQUEZ (1993): Historical seismicity of the North Panamá Deformed Belt, Rev. Geol. Am. Centr., 15, 49-64. DÍAZ, V. (1933): Conmociones Terrestres de América Cen- tral, Editorial José Ibarra, Guatemala, pp. 267. FELDMAN, L. (1985): Disasters, Natural and Otherwise, and Their Effects upon Population Centers in the Reino de Guatemala, in Estudios del Reino de Guatemala, Dun- can Kinkead (Escuela de Estudios Hispano-Ameri- canos, Seville), 49-60. FELDMAN, L. (1986): A Master List of Historic (Pre-1840) Earthquakes and Volcanic Eruptions in Central Ameri- ca, in Inv. Nat. Haz. Lat. Am. History, edited by ROBERT H. CLAXTON (Ed. Studies in the Social Sciences, West Georgia College), vol. 25, 66-106. FELDMAN, L. (1993): Mountains of fire, lands that shake. Earthquakes and Volcanic Eruptions in the Historic Past of Central America (Labyrinthos, California), pp. 295. FERNÁNDEZ, M., E. MOLINA, J. HAVSKOV and K. ATAKAN (2000): Tsunamis and Tsunami Hazards in Central America, Natural Hazards, 22, 91-116. GONZÁLEZ, C. (1910): Temblores, Terremotos, Inunda- ciones y Erupciones Volcánicas en Costa Rica 1608- 1910 (Alsina, San José), pp. 200. GRASES, J. (1974): Sismicidad de la Región Centroameri- cana Asociada a la Cadena Volcánica del Cuaternario (Instituto de Geociencias-Organización de Estados Americanos, Caracas), vol. II, pp. 253. GUIDOBONI, E. and M. STUCCHI (1993): The contribution of historical records of earthquakes to the evaluation of seismic hazard, Ann. Geofis., XXXVI (3-4), 201- 215. HILJE, B. and F. ALVAREZ (1994): Historia de los Sismos y Desastres Naturales en Costa Rica, (Observatorio Vol- canológico y Sismológico de Costa Rica, Heredia), pp. 292. KIRKPATRIK, R. (1920): Earthquakes in Panamá up to Janu- ary 1, 1920, Bull. Seismol. Soc. Am., 10, 120-128. 485 Current knowledge on the Central America historical seismicity: an analysis of recent catalogues LARDÉ, J. (1960): Obras Completas, (Publicaciones del Ministerio de Cultura, San Salvador), pp. 597. LEEDS, D. (1974): Catalogue of Nicaraguan earthquakes, Bull. Seismol. Soc. Am., 64, 1135-1158. MARTINEZ, H. and A. MAXIMILIANO (1978): Cronología Sís- mica y Eruptiva de la República de El Salvador a Par- tir de 1520 (Ministerio de Obras Públicas, Centro de Investigaciones Geotécnicas), pp. 40. MIYAMURA, S. (1980): Sismicidad de Costa Rica (Editorial Universidad de Costa Rica, San José), pp. 190. MOLINA, E. (1997): Tsunami Catalogue for Central Ameri- ca 1539-1996, Technical Report No. II 1-04, Reduction of Natural Disasters in Central America (University of Bergen). MONTERO, W. (1989): Sismicidad histórica de Costa Rica, período 1610-1910, Rev. Geofis. Int., 28, 531-559. MONTESSUS DE BALLORE, F. (1888): Sur les Tremblements de terre et les éruptions volcaniques dans le Centre Amerique: Depuis la Conquête Espagnole Jusqu’ a nos Jours, Société des Sciences Naturelles de Saone-et- Loire (Ed. Jobard, Dijon), pp. 293. PERALDO, G. (1993): El geólogo ante el discurso del docu- mento histórico. Rev. Geol. Am. Centr., 15, 92-97. PERALDO, G. and W. MONTERO (1994): Los temblores del Período Colonial de Costa Rica (Editorial Tecnológica, Cartago), pp. 162. PERALDO, G. and W. MONTERO (1996): La secuencia sísmi- ca de agosto a octubre de 1717 en Guatemala, Efectos y Respuestas Sociales, Historia y Desastres en Améri- ca Latina, coordinated by V. GARCÍA ACOSTA, I, 295- 324. PERALDO, G. and W. MONTERO (1999): Sismología Histórica de América Central, Instituto Panamericano de Geografía e Historia (IPGH), México, Publ. No. 513, pp. 342. ROJAS, W. (1993): Catálogo de sismicidad histórica y re- ciente en América Central, desarrollo y análisis, Tesis de Licenciatura en Geología (Universidad de Costa Ri- ca, San José), pp. 91. ROJAS, W., H. BUNGUM and C. LINDHOLM (1993): Historical and recent earthquakes in Central America, Rev. Geol. Am. Centr., 16, 5-22. SAPPER, K. (1925): Los Volcanes de la América Central (Max Niemeyer, Halle), pp. 116. SUTCH, P. (1981): Estimated intensities and probable tec- tonic source of historic (pre-1898). Honduran earth- quakes, Bull. Seismol. Soc. Am., 71, 865-81. VÍQUEZ, V. and J. TORAL (1987): Sismicidad histórica senti- da en el istmo de Panamá. Instituto Panamericano de Geografía e Historia, Rev. Geofís., 27, 135-166. WHITE, R. (1984): Catalogue of historic seismicity in the vicinity of the Chixoy-Polochic and Motagua faults, Guatemala, U.S. Geol. Surv. Open File Rep. 84-88, pp. 24. WHITE, R. and D. HARLOW (1993): Destructive Upper- Crustal Earthquakes of Central America since 1900, Bull. Seismol. Soc. Am., 83, 1115-1142.