THE VALUE OF ARCHAEOLOGICAL EVIDENCE IN PALAEOSEISMOLOGY
 

‘Palaeoseismology’ can be defined as the ‘identification and study of Prehistoric earthquakes’ (Sieh, 1978).
Pre-instrumental seismicity can only be inferred by geological, archaeological, and historical investigation.
Geologists can read signs of past tremors imbedded in sedimentary sections (faults and fractures in the bedrock structures) and sometimes present as geomorphic features (marine landslides, etc.).
Archaeological evidence instead is built on a more conjectural pedestal based on the effects of strong earthquakes on buildings and soil.
 
Generally, historical seismicity is based on descriptions of destructive earthquakes and can stretch back to 2000 years or more. When a written record is available, archaeological seismicity, which supplements the historical data, paints a very time and location specific picture, quite unlike the geological record .

The main contribution that seismic archaeology can offer to active tectonics is the localisation, dating and evaluation of seismic events.
The relationship between qualitative data (the description of effects) and quantitative data (magnitude, depth) is still approximate.
Archaeologists need to interpret and place material and written clues into a wider temporal context in the appropriate socio-economic framework of the sampled region. In other words, to establish the extent and magnitude of a certain destructive episode, i.e. working within a rigorous scientific outline, the archaeological study has to look beyond the often scant written record of the event. Local patterns of destruction and damage, filtered through historical parameters, can point to a more precise description of an ancient earthquake.
 

Identification of palaeosismic phenomena

First we have to ask if an earthquake is at all possible within the present geophysical knowledge of a certain area.
 

It is also crucial that a good stratigraphical check is carried out so that the observed deformations are not the result of later earthquakes.

Second we need to establish that human factors are not at the origin of these seismic signs.
Wars, fires, building over older structures, partial or complete demolition, restructuring, reconstruction and repair can all be confused with tectonic dynamics.

Finally we need to correlate the potential seismic event with the known historical setting of the site. We also have to establish the locality of the event, i.e. if neighbouring areas were also affected.

Criteria for the identification of earthquakes from archaeological data:
 

·Displacement drums of dry masonry columns. Seismic oscillation of a free-standing monolithic column produce strong extensional stresses at the base of the columns. At this point columns can fracture.
Seismic oscillations of a multi-block column causes translation of the single drums if not reinforced by metal or wood. Toppled columns display a characteristic domino-style arrangement of its drums.

· Opened vertical joints and horizontally slided parts of walls in dry masonry walls.

· Diagonal cracks in rigid walls. Horizontal seismic acceleration tend to deform a rectangle to parallelograms. Fissures are often near openings and in corners and again they tend to arrange themselves into diagonal fractures in brick-and-mortar fillings.

· Triangular missing parts in corners of masonry buildings.

· Cracks at the base or top of masonry columns and piers. Because of the geometry and distribution of stress fields in vertical and solid constructions, cracks appear on the most rigid part of the building (the base) and the most oscillating part (the top).

· Inclined or sub-vertical cracks in the upper parts of rigid arches, vaults and domes, or their partial collapse along these cracks. An arch (vault or dome) under the influence of a seismic force is stretched and as a result, joints open in a dry and rigid masonry structure. As a consequence a total or partial collapse occurs.

· Down-slided keystones in dry masonry arches and vaults. Keystone (or even the uppermost voussoirs) may slide down and the entire structure collapse.

· Several parallel fallen columns. This exclude the natural and gradual fall of individual columns through the ages.

· Constructions deformed as by horizontal forces (rectangular transformed to parallelograms). Stress and strain applied by a ground motion of oscillatory nature can stretch and deform manmade objects such as pave-stones.
 

Well-dated destruction of buildings correlating with historical (including epigraphic) evidence of earthquakes. The partial collapse of the retaining wall of Acropolis in late 17th and 18th century is well documented. Historical evidence suggests that this destruction is associated with an earthquake dated 1705. Along the river Jordan near Jericho there are written records of about 30 earthquakes which have occurred in the last 2000 years. Ben Shean was destroyed by an earthquake under Roman occupation in AD363. The damage extended 300 km from Petra in the South to Panias in the North suggesting an unusually large magnitude, estimated at around 7. This event also destroyed dozens of synagogues in Galilee. From the direction in which these columns fell it is possible to infer the ground motion during this earthquake. In most of the  investigated sites which are West of the Dead Sea fault, columns fell north-westerly, whereas in sites East of the fault, columns fell south-westerly. Assuming that these more or less free-standing columns fell in the direction opposite to the initial horizontal strong ground motion, the directions may indicate the future direction of the fault dynamic.

Even the Bible can be a source for palaeoseismology; Zechariah’s prophecy describes a large earthquake which occurred during the reign of King Uzziah around 760 BC. The earthquake happened somewhere East of Jerusalem, most likely along the Jericho fault. Apparently, the offset of the rocks across it was great enough to reveal the northward slip of the eastern side relative to the southward slip of the western side. This motion is remarkably similar to the motion observed in the 1927 Jericho earthquake, and is consistent with the N-S movement of the plates in this area. Many more examples includes new evidence of earthquake destruction in late Minoan Crete, in the south-western Peloponnese, the disappearance of Dioscura and Sebastopolis.

Other likely candidates are the Temple of Zeus Olympus in Sicily, the 6th city of Troy, and the destruction of the Rodi colossus in 224 a.C.

Palaeoseismology does not need to based only on destructive episodes. There is evidence that the uplifting of the ancient harbour of Aigeira in the Corinthian Gulf was also due to a well know seismic event. Carbon dating applied on a sample of fossil Dendropoma resulted in an estimated uplift centred around AD 1000-1200.

Conclusion

Archaeological evidence in palaeoseismology is a relatively new source of investigation. It was put into a more scientific framework only in 1928 by Sir Arthur Evans. Based on evidence of a destruction layer, he established the tradition of regarding earthquake horizons as benchmarks in archaeological stratigraphy and chronology. Some archaeologist have gone as far as blaming major earthquakes for the destruction of several major constructions, even putting the blame for the collapse and disappearance of ancient civilisations.

For the Earth scientist archaeological evidence, when can be irrefutably considered such, can supply another piece of evidence when trying to construct a geological profile of a given area.

Ancient earthquakes can also warn geologists about a presently tectonically quiet region and a potential future return of activity.
 

References
 
Ambraseys, N. 1971. Value of historical records of earthquakes. Nature. 232 375-379.
Guidoboni, E. 1996. Archaeology and Historical Seismology: the need for collaboration in the Mediterranean Area. Fitch Laboratory, British School at Athens. Athens. Greece.
Lee. W. H., H. Meyers & K. SHimazaki 1988. Historical seismograms and earthquakes of the world. Academic Press, Inc., San Diego, California, USA.
Karz, I. & Kafri, U. 1978. Evaluation of supposed archaeoseismic damage in Israel. Archaeological Science. 5 237-253.
Kirikov, B. 1992. Earthquake resistance of structures: from antiquity to our times. Mir, Moscow. Russia.
Pavlides, S. B. 1996. Palaeoseismology: a branch of Neotectonics linking Geological, Seismological and Archaeological data. Fitch Laboratory, British School at Athens. Athens
Pavlides, S &  D. Mountrakis. 1986. Neotectonics: an introduction to recent geological Structures. University Studio Press. Thessaloniki. Greece.
Rapp, G. 1986. Assessing archaeological evidence for seismic catastrophes. Geoarchaeology 1 365-379.
Sieh. K. E. 1978. Prehistoric large earthquakes produced by slip on the San Andreas fault at Pallet Creek, California. J. Geophys. Res. 83  3907-3939.
Sinopoli, A. 1991. Dynamic analysis of a stone column excited by a sine wave motion. Applied Mechanics Revue. 44/2 S246-S255.
Sinopoli, A. 1989a. Kinematic approach in the impact problem of rigid bodies. Applied Mechanics Revue. 44/2 S233-S244.
Stiros, S. C. 1996. Identification of Earthquakes from Archaeological Data. Fitch Laboratory, British School at Athens. Athens. Greece.