Geology Reference
In-Depth Information
seismic principles, automatic imaging programs produce subsurface veloc-
ity models without human intervention and should be treated with healthy
scepticism. Mis-picked arrivals will be treated as real by the computer, but
would usually be recognised in manual interpretation. Moreover, if the ac-
tual velocity change at a boundary were to be greater than the maximum
velocity change allowed by the software, the interface would be spread
out, with overestimation of the velocities on the low-velocity side and
underestimation of velocities on the high-velocity side. This could lead
to an erroneous choice of ripping equipment (see Figure 11.2). Even the
basic intercept-time method (see Section 13.2.1) would give a better result
for shallow dipping layers in this case. It is always prudent to check the
results of one type of modelling against another, to increase confidence in
the interpretation, but imaging techniques have, generally speaking, signifi-
cantly improved the interpretation of refraction data in complex geological
settings and have radically reduced interpretation time and effort.
13.3 Limitations of the Refraction Method
First-arrival refraction work uses only a small proportion of the information
contained in the seismic traces, and it is not surprising that interpretation is
subject to severe limitations. These are especially important in engineering
work; in low-velocity-layer studies only a time delay estimate is sought and
short shots alone are often sufficient.
13.3.1 Direct waves
The
ground roll
consists of a complex of P and S body waves and Love and
Rayleigh surface waves travelling with different but generally slow veloc-
ities. There is often some doubt as to which component actually produces
the first break, since conventional geophones respond only poorly to the
horizontal ground motions of direct P-waves. Close to the source, enough
energy is associated with the P-waves for the response to be measurable,
but at greater distances the first breaks may record the arrival of S-waves,
surface waves or even the air wave.
The complex character of the direct wave may be amongst the reasons for
the commonly observed failure of best-fit arrival lines to pass through the
origin. Delays in the timing circuits may also play a part but can be measured
directly using a detonator or a light hammer blow close to a geophone. A
more important reason may be that the amplifier gains at geophones close to
the shot-point are often set so low that the true first-arrivals are overlooked
(Figure 13.11). Full digital storage of the incoming signals should allow the
traces to be examined individually over a range of amplifications, but if this
is not possible, then the most reliable velocity estimates may be those that
do not treat the origin as a point on the line.
