SGA, Storia Geofisica Ambiente, Bologna, Italy

The need to understand the activity of the main seismogenetic structures, to calculate the recurrence periods of major earthquakes and to identify their main epicentral areas, requires wide-ranging research in the field of historical seismology. The present research was conducted in the framework of the projects of the Istituto Nazionale di Geofisica (1991-1995). The presence of different populations and ruling élites, and hence of languages and cultures, has in many cases confused the historical context of the medieval Sicilian sources. However, by going beyond the local sources, and analysing the events in a wider European and Mediterranean context, taking due account of the Byzantine, Latin and Arab sources, it has been possible to identify five seismic events that do not exist in the Italian catalogues or whose dates are very difficult to establish. Six spurious events have been deleted thanks to the revision (659, 785, 796, 963, 1069, 1259); the date of an event has been corrected (1140 into 1125) and five unknown events have been discovered: 853, 1172, 1203-1204, 1255-1256 and 1295-1296. The data on which these findings are based are in many cases dispersed and unused within the many specialised sectors of historical research; in some cases they are unknown even to historians. An example of textual analysis of the sources is also presented to show by what a roundabout route the chronological parameter for the 1125 earthquake was reached. The new events confirm the high seismicity of the eastern area of Sicily; nevertheless, the newly identified series does not seem to show any seismic event comparable to the one that struck the Messina Strait, in 1908 (I0 = XI MCS, M = 7.2): this historical element can provide information regarding the return times of such great event; a hypothesis is formulated by the authors: it will have to be carefully examinated by archaeology research. The criteria used in this research are explained, as well as the problems tackled in accordance with the method for the revision of the historical earthquakes adopted by ING-SGA.

Within the SGA research on the historical seismicity of the Crimean Peninsula (SGA Report, 1990), interest has been focused on the case of the earthquake of 63 B.C. According to regional seismic catalogues as well as to historic and archaeological literature, two late Roman sources. Dio Cassius and Paulus Orosius, allegedly give evidence of an earthquake which happened in the Crimea in this year; the event was linked to the death of Mithridates V1 Eupator, eventually the king of Pontos. Local archaeologists claimed to have found evidence of this event in the excavations of Panticapaeunl (present-day Ker?). In fact. this is the result of a restricted analysis of the written sources. Thence stems a sort of iivulgatan. currently accepted by scholarship, yet not really supported by the evidence. A re-examination of the whole question, including an analysis of all sources avalaible on earthquakes in the Eastern Mediterranean. showed that in that period no seismic event took place in the Crimea. Dio's and Orosius' accounts are instead concerned with another earthquake, already known for Syria from other sources. This historical case gives a proper methodological example of the problems concerned with the analysis of the evidence in historical seismology. not only of Antiquity, but of almost any premodern period.

The present contribution describes the method of work, the types of source materia] used, and the historio- graphical and historico-eismic tradition of Armenia. The catalogue' s territorial frame of reference is that of socalled historical Armenia (which included part of present Eastern Turkey, and part of present Azerbaijan). The sources belong to different languages and cultures: Armenian, Syriac, Greek, Arab, Persian and Georgian. A comparison of the local sources with those belonging to other cultures enab]es the historical and seismological I"adition of the Mediterl'anean to be "linked" with that of the Iranian p]ateau, traditionally considered as two separate areas. We analyzed historical events listed in the most recent catalogues of earthquakes in the Armenian area compiled by Kondorskaya and Shebalin (1982) and Karapetian (1991). Important and valuable though these catalogues are, they are in need of revision. We found evidence for six hitherto unrecorded seismic events. Numerous errors of dating and location have been corrected, and several new localities and seismic effects have been evidenced. Each modification of the previous catalogues has been documented on the hasis of the historiographical and literary sources and the data from the written sources have been linked with those concerning the history of Armenian cities and architecture (monasteries, churches, episcopal complexes). On the whole. the revised earthquakes seem underestimated in the previous catalogues. The aim of this catalogue is to make a contribution to the knowledge of historical seismicity in Armenia, and at the same time to underline the specific nature of the Armenian case, thus avoiding a procedure which has generally tended to place this area in a marginal position, within the wider field of other research on historical earthquakes.