Aims and methods in territorial archaeology: possible clues to a strong fourth-century AD earthquake in the Straits of Messina (southern Italy)

Abstract

This research was stimulated by the need to extend in time the record of Italy's largest earthquakes, which commonly have repeat times of the same order as the length of the available historical record. As a test case we used the 1908 Straits of Messina earthquake, a large event that geologists assume to recur at intervals of roughly a millennium but whose predecessors are as yet unknown. The 1908 earthquake caused enormous territorial upheaval and left signs in the settlements that are still largely recognizable today. We hypothesized that the Straits of Messina, which were densely populated even in ancient times, may similarly retain evidence of one or more much older 'upheavals' of the settlement network, and that this evidence may be recognized through a careful analysis of archaeological observations. We found evidence that the settled area around the Straits of Messina contracted substantially around the middle of the fourth century AD, when many sites were abandoned or relocated. This contraction can hardly be justified by the then current economic and military setting. Specific archaeological findings within the cities of Messina and Reggio Calabria also suggest a serious decline of the region during the same period. The archaeological hypothesis is in good agreement with the available historical and palaeoseismological evidence and suggests that a large earthquake, perhaps similar to the 1908 event, took place in the area surrounding the Straits of Messina around the middle of the fourth century AD.One of the most important findings of earthquake studies in Italy over the past decade is that the repeat time of individual large earthquakes is of the order of one or more millennia. Similar results had previously been obtained for many seismic areas in the world, but the frequency of large historical earthquakes in Italy had brought seismologists to the conclusion that Italian earthquakes repeated themselves every few centuries or so. Some Italian cities indeed appear to have been struck several times in their history, but results supplied by palaeoseismology (a recently developed discipline aimed at uncovering and dating any direct surface evidence of large earthquakes of the past) suggest that this is due to multiplicity of destructive seismogenic sources rather than to the frequency with which individual sources generate large earthquakes (see, e.g. Boschi et al. 1994).

Archaeoseismology: Methodological issues and procedure

“…In this light, some works have used the architectural history of the buildings (usually churches) in order to improve the knowledge of the effects of earthquakes sparsely known (e.g., ENEL 1986; Stiros et al 2006). In case of the 1117 earthquake in northern Italy, an evident clustering of interventions to romanesque churches has been identified in the zones which, based on the sparse historical information, should represent the most damaged areas (ENEL 1986;Guidoboni et al 2000;.…”

“…This term has been used in the methodological and research works by Guidoboni (2000) and Guidoboni et al (2000) to indicate the correlation of archaeoseismic effects throughout a region in order to define the extension of the mesoseismal area. The extension is crucial because the main difference between earthquake damage and damage related to other natural causes is the distribution of the effects.…”

“…For these reasons, the identification of buildings which have been simultaneously damaged in several locations in periods of political stability (line 11 of Table 1) or reviewing published or unpublished archaeological material on this aspect (line 1, column 2 of Table 1) may be a logical method to test the archaeoseismic hypothesis (e.g., Galadini and Galli 2004;Guidoboni et al 2000;Nur and Cline 2000;Jones and Stiros 2000;Stiros 2001). …”

“…Also in case of poor information, an historical event is usually defined in terms of the year of occurrence or of a time span of few years compared to decades in archaeology. This implies that relating observed destruction to an historical event, especially in regions of frequent destructive earthquakes, cannot be considered reliable (Guidoboni 2000;Guidoboni et al 2000). Since the list of historical earthquakes cannot be complete, archaeoseismic evidence due to different events may be attributed only to the known historical earthquake.…”

“…Since the list of historical earthquakes cannot be complete, archaeoseismic evidence due to different events may be attributed only to the known historical earthquake. This procedure leads to the enlargement of the perceived destruction and consequently to an overestimation of the event size (Guidoboni 2000;Guidoboni et al 2000;Ambraseys 2005).…”