Earthquakes - Extreme Events: A Physical Reconstruction and Risk Assessment - page 154

When uncertainty margins from dated sand blows across a region overlap, those
events cannot technically be distinguished and the possibility that the same
earthquake generated a suite of sand blows must be considered. The only way
to determine whether the sand blows occurred independently of each other is
to use dating techniques of sufficiently high resolution so as to obtain small
uncertainty margins. Dating liquefaction events can also be made more diffi-
cult if sand-bearing water has intruded the overlying sediment and formed sand
dikes and sills without venting to the ground surface. Where this is the case,
theliquefaction feature can only be dated based on the age of the uppermost
intruded stratigraphic unit (Tuttle and Schweig, 1996).
There are several other problems associated with using liquefaction features
to date prehistoric earthquakes. One of the most problematic is differentiating
between earthquake-generated structures and liquefaction features formed by
other geological processes that mimic seismic liquefaction events (Sims and
Garvin, 1995). Non-seismic sand boils form when water seeps beneath levees dur-
ing floods; these boils can mimic sand blows developed during seismic events (Li
et al ., 1996). Misinterpretations, where flood-induced sand boils are mistaken for
earthquake-induced sand blows, can lead to a hypothesis of shorter earthquake
recurrence intervals and thus an exaggeration of the seismic hazard. This can
also result in an exaggeration of the magnitude of prehistoric earthquakes as
the spatial extent of these events can be overestimated.
Li et al .(1996)use six criteria to distinguish between earthquake-induced
liquefaction and flood-induced sand boils. These are:

(1) earthquake-induced liquefaction deposits are broadly distributed along

an epicentral area, whereas flood-induced sand blows are limited to a
small band along river levees;

(2) the conduits of most earthquake-induced sand blows are planar dikes,

whereas the conduits of flood-induced sand blows are most commonly
tubular;

(3) depression of the pre-earthquake ground is usual for sand blows, but

not for sand boils;

(4) flood-induced sand boils tend to be composed of better sorted and much

finer-grained sediments than sand blow deposits;

(5) source beds for earthquake-induced deposits occur at a wide range of

depths, whereas the source bed for sand boils is always near the surface;
and

(6) materials removed from the walls surrounding the vent of a sand blow

are seen inside the sand blows. In general, flood-induced sand boils
are interpreted to represent a less energetic genesis than earthquake-
induced liquefaction.