04Bautista.qxd 379 ANNALS OF GEOPHYSICS, VOL. 47, N. 2/3, April/June 2004 Key words historical earthquake catalog – Philip- pine earthquakes – seismic hazard assessment – archival documents – historical accounts 1. Introduction Reliable seismic hazard studies depend on having a robust earthquake catalog. The longer the extent of the catalog and the more reliable the parameters are, the better it is for those do- ing seismic hazard analysis. There are two kinds of earthquake catalog: one is the instru- mental or recent catalog and the other is the his- torical catalog. In this study, instrumental or re- cent catalog refers to the time when seismic monitoring existed while historical refers to the pre-instrumental period. In the Philippines, in- strumentally-derived parameters are available for events from 1892 onwards. The year 1892 was selected because the first earthquake with instrumentally derived parameters was deter- mined by Abe (1994) for an earthquake in 1892. The historical period for the case of the Philip- pines could be considered as covering the peri- od before 1892. Unlike other countries with long earthquake histories and despite the The Philippine historical earthquake catalog: its development, current state and future directions Maria Leonila P. Bautista and Bartolome C. Bautista Philippine Institute of Volcanology and Seismology (PHIVOLCS), Department of Science and Technology (DOST), Quezon City, Philippines Abstract This report will trace the development of the historical earthquake catalog of the Philippines, assess its present state and recommend future research directions. The current Philippine historical earthquake catalog is culled from var- ious catalogs, both global and local, that were developed since the first Philippine catalog by Perrey was published in 1860. While early global catalogs gave simple mention of earthquakes in the Philippines, more focused earth- quake catalogs about the Philippines gave more explicit descriptions of earthquake accounts and adopted descrip- tions by local historians. Over the years, various historians and seismologists continued to compile their catalogs whose contents depended on the author’s perspectives and purposes. These works varied from simple listings to others including detailed descriptions. It was only recently that an attempt made to parameterize the magnitudes and epicenters of Philippine historical earthquakes using magnitude-felt area relations was done. A more detailed catalog, however, is now underway that will show details of intensity distribution for each significant historical earthquake. By comparing the historical catalog with the recent catalog and assuming that the recent catalog is complete, we find that there are still a substantial amount of historical earthquakes that needs to be reviewed and located. Possible sources of new information are local libraries, museums and archives in the Philippines, Spain and other Southeast Asian countries to which the country was in contact with during historical times. Mailing address: Dr. Maria Leonila P. Bautista, Philip- pine Institute of Volcanology and Seismology (PHIVOLCS), Department of Science and Technology (DOST), C.P. Gar- cia Avenue, UP Campus, Diliman 1100, Quezon City, Philippines; e-mail: leyo@phyvolcs.dost.gov.ph 380 Maria Leonila P. Bautista and Bartolome C. Bautista Philippines being seismically active, the Philip- pine historical catalog is not that long compared to other countries. The historical accounts only started during the time when the country was conquered by Spain in the late 15th century. This study will discuss the development of the historical earthquake catalog in the Philippines, assess its present state and recommend future directions. 2. Existing catalogs The current Philippine historical earthquake catalog is culled from various catalogs, both global and local, that were developed since the first Philippine catalog by Perrey was published in 1860. Early global catalogs made simple mention of earthquakes in the Philippines. Meanwhile, catalogs that mainly focused on the Philippines gave more explicit descriptions of earthquake accounts and adopted descriptions by local historians. Over the years, various historians and seis- mologists kept on building their own catalogs whose contents varied depending on each au- thor’s perspective and purpose. For example, some authors would make simple listings of earthquakes while some would include detailed descriptions. Some authors would only list tsunamigenic events. Others gave their own in- tensity estimates as well as drew their own iso- seismal maps. A recent work by Bautista and Oike (2000) has parameterized the magnitudes and epicenters of well-described Philippine his- torical earthquakes using magnitude-felt area relations. 2.1. Global earthquake catalogs It was only when colonization of the Philip- pines started in the late 15th Century that earth- quakes affecting the country started to be no- ticed and documented. These earthquake cita- tions may be grouped into those cited in global catalogs while the other group belongs to the country-specific catalogs. Global earthquake catalogs such as those made by von Hoff (1841), Mallet (1853, 1854, 1855) and Milne (1912) mentioned earthquakes about the Philippines. The work of von Hoff (1841), written in Ger- man, listed three earthquakes and three volcanic eruptions in the Philippines. Mallet’s works were in English and among his references were the works of De Guignes (1808), Garnier (1839) and articles from the Singapore Chronicle (1824, 1829). Meanwhile, Milne (1912) listed 66 Philippine historical earthquakes from 1589 to 1909 in his catalog. His source is mainly the work of Maso (1895). More recently, Dunbar et al. (1992) listed epicenters, magnitudes and damages attributable to Philippine earthquakes from 1599 to 1990. Meanwhile, some authors listed tsunamigenic earthquakes only (Rudolf, 1887; Heck, 1947; Iida et al., 1967; Berning- hausen, 1969; Cox, 1970; Nakamura, 1978). Re- viewing global data since 1616, Iida et al. (1967) listed 26 tsunamigenic earthquakes in the Philip- pines. They also listed the areas affected and es- timated heights based on descriptions. Berning- hausen (1969), meanwhile, listed tsunamis in Southeast Asian region which included 29 events in the Philippines. His database covered the period 416 to 1965. The various works of Abe (1981, 1984, 1994) and Abe and Noguchi (1983) especially their redetermination of the magnitudes by Gutenberg and Richter covering the early 20th century period helped to improve the earthquake data in the Philippines including those during the 1892 to 1900 period. 2.2. Country-focused historical earthquake catalogs: Philippines While the earliest earthquake reported for the Philippines was the one which happened in 1589, the first known country-focused earth- quake catalog for the Philippines is the work of Perrey (1860) which was in French. This cata- log described both earthquakes and volcanic eruptions in the Philippines from the 15th to 18th centuries starting with an earthquake in 1601. His sources included various European travellers in Asia who wrote books, chronicles or accounts of their travels. Despite exaggera- tions and misquoted names of places, Perrey’s work became one of the most referred to earth- quake catalog for the Philippines. 381 The Philippine historical earthquake catalog: its development, current state and future directions The creation of the Manila Observatory in 1865 paved the way for the systematic collection of earthquake reports in the country. Maso, a re- searcher at the Manila Observatory, published an earthquake catalog in 1895 written in Spanish. His work could be considered the first attempt to document the historical accounts made by Span- ish priests about earthquakes affecting their local- ities. His work is more comprehensive than that of Perrey, although the main difference is that it dealt exclusively with earthquakes and excluded vol- canic activities. His work also included the obser- vations of the Manila Observatory and its various stations in the provinces. In addition, Maso also drew isoseismal maps of selected large earth- quakes using the Manila Observatory Intensity Scale of I to VI (table I). Later, culling much from the work of Maso, Algue (1900), who was also from the Manila Observatory, published his own catalog covering the period 1870 to 1897. In 1901, after the Spanish conquerors had left the country, the new American government reorganized the Manila Observatory and creat- ed a Philippine Weather Bureau. Aside from weather monitoring and forecasting, the Bureau also conducted seismic monitoring. Maso con- tinued on his compilation and updating of earthquake events and published several cata- logs (Maso, 1909; 1927, Maso and Smith, 1919). The 1895 and 1927 Maso catalogs have basically the same contents except that the 1927 version only contained the damaging events. In 1928, Fr. William Repetti joined the Manila Observatory and the Philippine Weather Bu- reau. He was placed in charge of all seismolog- ical stations in the Philippines. He was Chief of the Section of Seismology and Terrestrial Mag- netism of the Philippine Weather Bureau for more than a decade. In 1946, Repetti published the most comprehensive historical catalog of the Philippines covering the period 1589 to 1899. Repetti translated various accounts writ- ten in various languages into English. Repetti’s work included the results of Perrey and Maso and at the same time included other accounts culled from history books, bulletins, letters and newspaper articles. After World War II, the Weather Bureau was reorganized into the Philippine Atmospheric, Geophysical, Astronomical and Services Ad- ministration (PAGASA). In 1982, the Office of Foreign Disaster Assistance (OFDA) of the United States Agency for International Devel- opment (USAID) undertook a project called «Seismic Data Gathering, Southeast Asia». This project was undertaken under the auspices of the Southeast Asia Association on Seismology and Earthquake Engineering or SEASEE. Among the projects of SEASEE is an earthquake catalog published in 1985 that compiled Repetti’s histori- cal data, instrumental data from the Manila Ob- servatory and PAGASA (1976) as well as reports of investigations of earthquakes from 1950 to 1983 (SEASEE, 1985). Aside from these, the SEASEE catalog also included earthquake data determined for the Philippines by Gutenberg and Richter (1954), Gutenberg (1956), by the Interna- tional Seismological Centre (ISC) and the U.S. Geological Survey (USGS). The SEASEE catalog was divided into eight chapters: i) Catalogue of Philippine Earth- quakes 1589-1864; ii) Catalogue of Philippine Earthquakes 1865-1899; iii) Catalogue of Philippine Earthquakes 1901-1942; iv) Cata- logue of Philippine Earthquakes 1948-1983; v) Catalogue of Destructive Earthquakes 1589-1983; vi) Assessment of Seismic Intensi- ty of Philippine Historical Earthquakes; vii) Seismic Source Zones of the Philippines, and viii) Seismotectonics of the Philippines. The result is an 843-page compilation of earth- quakes from 1589 to 1983. Although some of these previous works, in- cluding the SEASEE catalog, estimated epicen- Table I. Manila Observatory Intensity Scale (after SEASEE, 1985). Intensity Description I Perceptible II Light III Regular IV Strong V Violent VI Destructive 382 Maria Leonila P. Bautista and Bartolome C. Bautista ters, maximum intensities and even contained iso- seismal maps, there was no attempt to make a sys- tematic parameterization of these historical earth- quakes to determine their magnitudes and epicen- ters. The first attempt was made by Bautista (1993) who made use of magnitude-intensity at epicenter (Io) relationships derived from Philip- pine earthquake data to infer the sizes of Philip- pine historical earthquakes. Later on, Bautista (1996) revised her empirical equations through the addition of more recent events and developed new empirical equations relating magnitude to felt area for various intensity levels. The same study estimated the epicenters and magnitudes of 74 historical earthquakes. A more extensive work by Bautista and Oike (2000) re- viewed the historical accounts of more than 3000 earthquake events from 1589 to 1895. From these events, they were able to estimate the magnitudes and epicenters of 485 historical earthquakes in the Philippines. While there were 3000 earthquake events, most of these were ei- ther felt only in one area or had no damage re- ported. Only these 485 events were sufficiently described or could be assessed for parameter de- termination. Meanwhile, as guides to future users of the catalog, their study also gave a qual- ity assessment for each earthquake. This quality assessment is based on the number of places where accounts are available. To have a higher quality, an earthquake should have many ac- counts for many different places. Having numer- ous accounts but limited to one place would still mean a lower quality of reporting. Out of the 485 events, quality «A» was given to earthquakes with 10 or more reporting places, «B» to those with five to nine reporting places and «C» to those with four or less number of reporting places. Using these criteria, they found out that only 52 of the total 485 events may be consid- ered very reliable (quality A) and 80 events as quite reliable (quality B). The remaining events were given quality C and their determined pa- rameters are not considered so reliable. 3. Issues and concerns Inaccuracies, exaggerations and biases sometimes creep into interpretation of histori- cal accounts. Thus, it is always better to go back to the primary source of information. The search for new sources of information started from the original list of references by previous authors. The original list has since more than doubled and potential new sources of information are now being considered. The on- ly issue is how to find or locate these references for historical earthquake evaluation. Meanwhile, aside from other factors such as change of names and jurisdiction, it is also im- portant to consider knowledge of local history, geography and existing socio-political condi- tions when these earthquakes occur when re- viewing historical earthquakes. Samples of such errors were cited in Bautista and Oike (2000). One example is after an earthquake in 1645, when only the Spaniards who perished during the earthquake were counted, while per- sons of «no account» were disregarded. Since the number of casualties are, at times, used to gauge how damaging an earthquake was, the re- sults from this procedure of counting could give quite misleading results. Meanwhile, knowledge of local geogra- phy is also important. Errors in naming places can be discerned if one is familiar with local geography. For example, the name of a place called «Cagayan» was changed by Per- rey (1860) into «Camarines» for an earth- quake in 1627, when in fact the two places are 400 km apart. To solve such kinds of issues for various cases, it was always important to find out which place was actually affected by getting the infor- mation from the original source, if available, or by relating existing accounts with one another. 4. Status of the Philippine historical catalog After Bautista and Oike (2000) have deter- mined the epicenters of 485 historical events, they also evaluated the completeness of the re- sulting historical catalog. They plotted the typical yearly number of earthquakes with M 6.5 and above for the Philip- pines for both the historical and recent periods. They found out by comparing the historical cata- log with the recent catalog and assuming that the 383 The Philippine historical earthquake catalog: its development, current state and future directions recent catalog is complete, that there are still a substantial number of earthquakes that need to be reviewed and located especially during the period from 1600 to 1850. On the other hand, they also found out that earthquakes that occurred from the period 1850 to 1900 are almost completely re- ported. Instead, what is missing in the catalog for the same time period are events with magnitudes lower than M 6.5 (fig. 1). 5. Future directions The continued search for new information to cover the 1600 to 1850 period for all magnitude ranges and from 1850 to 1900 for magnitudes less than 6.5 will be actively pursued. Among the pos- sible sources are local libraries, museum and archives, libraries in Spain and in Southeast Asia. Spanish-era old churches in various towns are also promising sources as some of them keep records of when they were damaged by natural phenomena while some of them main- tain a museum or archive. Specific books on town histories are another potential source of earthquake histories. The Internet offers information, although it should still be verified, about histories of churches, dates of their constructions, effects of calamities, architectural styles and many other details. The Mindanao area in southern Philip- pines where the Muslim mosques are found is also a fresh avenue of research considering that previous research concentrated mainly on ac- counts from a Christian perspective. The geoscientific institutions, archives and libraries around the Philippines are also anoth- er source of possible new information that must be searched on further. The National Records and Archives Office of the Philippine National Library is another promising source of data. As it is, archival Fig. 1. Number of earthquakes per 50-year period in the Philippine earthquake catalog. By assuming that the recent period from 1900-1997 is complete, the historical period 1850-1900 may be considered well reported and almost complete, too. Meanwhile, there are still many earthquakes that need to be discovered during the earlier period from 1600 to 1850 (table after Bautista and Oike, 2000). Earthquake Count Year Magnitude Range 384 Maria Leonila P. Bautista and Bartolome C. Bautista documents on some historical earthquakes and volcanic eruptions which were not seen by previous authors including Maso and Repetti had already been found. They are all handwritten and in Spanish. They were writ- ten by governors, military commandants, priests, mayors and other government offi- cials and may be considered as the official damage reports for these earthquakes. How- ever, they are difficult to read and translate because they were handwritten in very com- plicated hand characteristic of old Spanish documents. But their scientific value cannot be overemphasized because they are able to add significant information to these earth- quakes. An initial project to have these trans- lated properly is a possible future action. Another future activity envisioned is the eventual production of a catalog that will show the detailed intensity distribution of most of these well-described historical earthquake events. Aside from the inclusion of the isoseis- mal maps, this Philippine catalog will include the tabulation of the individual earthquake ac- counts for the period 1589 to 1899. These tabu- lated accounts will be further categorized for easy usage into structural damage, effects to people, effects to the environment, human and economic loss estimates, references, and their assigned intensities. Acknowledgements We would like to thank two anonymous re- viewers whose insights and comments greatly improved this manuscipt. We would also like to thank Dr. Paola Albini of the Istituto Nazionale di Geofisica e Vulcanologica, Sezione di Mi- lano, for her thoughtful and constructive com- ments throughout the various stages of the edit- ing process. REFERENCES ABE, K. (1981): Magnitudes of large shallow earthquakes from 1904-1980, Phys. Earth Planet. Int., 27, 72-92. ABE, K. (1984): Complements to «Magnitudes of large shallow earthquakes from 1904 to 1980», Phys. Earth Planet. Int, 34, 17-23. ABE, K. (1994): Instrumental magnitudes of historical earthquakes, Bull. Seismol. Soc. Am., 84 (2), 415-425. ABE, K. and S. NOGUCHI (1983): Revision of large shallow earthquakes, 1897-1912, Phys. Earth Planet. Int., 22, 1-11. ALGUE, J. (1900): Focos Sismicos, in El Archipelago Fil- ipino, Chapter X, 268-377. BAUTISTA, M.L.P. (1993): Historical earthquakes in the Philippines, in Proc. 1993 Joint Conference of Seis- mology in East Asia, October 29-November 2, 1993, Tottori, Japan, p. 53. BAUTISTA, M.L.P. (1996): Estimation of the magnitudes and epicenters of Philippine historical earthquakes, Un- publ. M.Sc. Thesis (Graduate School of Science, Kyoto University). BAUTISTA, M.L.P. and K. OIKE (2000): Estimation of the epicenters and magnitudes of Philippine historical earthquakes, Tectonophysics, 317, 137-169. BERNINGHAUSEN, W.H. (1969): Tsunamis and seismic se- iches of Southeast Asia, Bull. Seismol. Soc. Am., 59, 289-297. COX, D.C. (1970): Discussion of «Tsunamis and seismic se- iches of Southeast Asia» by William H. Berninghausen, Bull. Seismol. Soc. Am., 60, 281-287. DE GUIGNES, M. (1808): Voyages a Peking, Manille et l’île de France faits dans l’intervalle des annees 1784 a 1801, de l’imprimerie imperiale (Paris). DUNBAR, P.K., P.A. LOCKRIDGE and L.S. WHITESIDE (1992): Catalog of significant earthquakes 2150 B.C.-1991 A.D., incl. quantitative casualties and damage (World Center A for Solid Earth Geophysics. National Geo- physical Data Center, Boulder, Colorado, U.S.A.), Re- port SE-49, pp. 320. GARNIER, J.G. (1839): Traite de meteorologie du Physique du Globe (Librairie Scientifique-Industrielle, Paris), pp. 480. GUTENBERG, B. (1956): Great earthquakes 1896-1903, Trans. Am. Geophys. Un., 37, 608-614. GUTENBERG, B. and C.F. RICHTER (1954): Seismicity of the Earth (Princeton University Press), 2nd edition, pp. 310. HECK, N.H. (1947): List of seismic sea waves, Bull. Seis- mol. Soc. Am., 37, 269-279. HOFF, K.E.A. VON (1841): Chronik der Erdbeben und Vulcan Ausbrüche, II. Theil, (Justus Perthes, Gotha), pp. 406. IIDA, K, D.C. COX and G. PARARAS-CARAYANNIS (1967): Preliminary catalog of earthquakes occurring in the Pa- cific Ocean, International Tsunami Information Center and Office of Naval Research, Report no. 5. MALLET, R. (1853): Catalogue of recorded earthquakes from 1606 B.C. to A.D. 1850, Brit. Ass. Rep., 1853, 1-176. MALLET, R. (1854): Catalogue of recorded earthquakes from 1606 B.C. to A.D. 1850, Brit. Ass. Rep., 1854, 118-213. MALLET, R. (1855): Catalogue of recorded earthquakes from 1606 B.C. to A.D. 1850, Brit. Ass. Rep., 1855, 2-326. MASO, S.M. (1895): La Sismologia en Filipinas (Es- tablicimiento Tipo Letograpico de Ranuirez Y Compa- nia, Manila), pp. 121 (+ figures). MASO, S.M. (1909): Catalogue of Violent and Destructive Earthquakes in the Philippines, 1599-1909 (Weather 385 The Philippine historical earthquake catalog: its development, current state and future directions Bureau, Manila Central Observatory). MASO, S.M. (1927): Philippine Earthquakes, Weather Bu- reau Seismological Bulletin for 1926, 67-103. MASO, S.M. and W. SMITH (1919): The relation of seismic disturbances in the Philippines to the geologic struc- ture, Phil. J. Sc., 8, 199-233. MILNE, J. (1912): A Catalogue of Destructive Earthquakes, A.D. 7 to A.D. 1899, BAAS (London), pp. 92. NAKAMURA, S. (1978): On statistical tsunami risks of the Philippines, South East Asian Studies, 15, 581-590. PAGASA (1976): Catalogue of Philippine earthquakes, 1585- 1976, in Earthquake Catalog (Philippine Atmospheric, Geophysical, Atmospheric and Services Administra- tion, Quezon City, Philippines). PERREY, A. (1860): Documents sur les tremblements de terre et les eruptions volcaniques dans l’archipel des Philippines, Mem. Ac. Sc., Arts et Belles-Lettres de Di- jon, VII, 85-194. REPETTI, W.C. (1946): Catalogue of Philippine Earthquakes, 1589-1899, Bull. Seismol. Soc. Am., 36, 133-319. RUDOLF, E. (1887): Über submarine Erdbeben und Eruptionen, Gerl. Beitr. Z. Geophysik, Stuttgart, I, 360-365. SEASEE (1985): Philippines, edited by E.P. ARNOLD, Southeast Asian Association of Seismology and Earthquake Engineering, Series on Seismology, IV, pp. 843. SINGAPORE CHRONICLE (1824), November 25: article about the October 26, 1824 earthquake in Manila. SINGAPORE CHRONICLE (1829), January 1: article about the December 7, 1828 earthquake in Manila.