Earthquakes - Extreme Events: A Physical Reconstruction and Risk Assessment - page 175

blows and mud deformation features, landslides, anomalous sediment layers in
lakes and destruction of ancient human built features provide us with an accu-
rate long-term history of the frequency of these past events. Their magnitude
can also be ascertained from the prehistoric evidence using the extent of surface
displacement features, both along a fault and the vertical displacement of the
ground, and the degree of offset of features such as aqueducts. The written and
oral records of ancient civilisations also frequently record earthquakes because
of the devastating impact they had on these societies.
The infrequent occurrence of earthquakes at a given location makes it dif-
ficult to obtain long-term records of sufficiently fine resolution to determine
whether they have varied significantly in their frequency over time. The long-
term records, however, can provide information on the average time interval
between these extreme events and have often shown that the historical record
is a poor reflection of the nature of the earthquake regime for a region. For exam-
ple, parts of the Dead Sea Fault in Syria had been thought to be inactive and
not likely to produce a severe earthquake in the near future. But the Meghraoui
et al .(2003)analysis of the long-term record showed that an earthquake of greater
than magnitude 7 is imminent and possibly overdue in this region. Once again,
and like so many long-term records of extreme natural events, it is clear that we
cannot rely upon short historical records to make reasonable predictions about
the future impacts of earthquakes or, indeed, any natural hazard.