Geoscience Reference
In-Depth Information
are discussed in
chapter 8
.
Appendix 1
contains a brief summary of the hardware and
software issues involved in managing interpretation workstations in practice, and finally
Appendix 2
contains a glossary of technical terms. This is not intended to be exhaustive;
1.6
Conventions: seismic display, units
There are two topics to mention here that may cause confusion to the unwary reader:
display polarity and systems of units. Display polarity is the more important of these;
arguing about polarity wastes large amounts of interpreter time. The problem is this: if
we have an interface at which impedance increases downwards, when we make a wiggle
display of a seismic trace with the time axis vertical, does such an interface give rise to a
deflection to the left (a trough) or to the right (a peak)? Classically on paper sections, the
peaks were shaded to produce a display in which the peaks were black and the troughs
appeared white, and a similar convention is often used for workstation displays so that
peaks are black or blue and troughs are red or white. In principle, polarity is fixedatthe
time of recording the data, and preserved throughout the processing sequence. Many
modern datasets are transformed to zero-phase so that a single interface is represented
by a single loop with some lower-amplitude wiggles on either side of it. Processors
often describe the polarity of the final data as 'SEG normal' or 'SEG reverse'. This
refers to a convention promulgated by the Society of Exploration Geophysicists (SEG),
according to which SEG normal would correspond to an increase in impedance down-
wards being represented by a peak. The reverse convention (SEG reverse) is commonly
employed in some hydrocarbon provinces, e.g. the UK North Sea. Unfortunately, it is
quite possible for mistakes in acquisition or processing to result in final displays with
polarity opposite to the processors' stated convention. The interpreter needs to check
for himself what the polarity of a given dataset really is. It might be thought that this
can easily be done by comparing the data with well synthetics
(section 3.1)
. However,
many seismic sections show long intervals of wiggles of about the same amplitude and
frequency, and over such an interval it may be easy to establish plausible ties using either
polarity convention, if bulk shifts are allowed; such shifts are almost always present in
real data owing to various limitations in processing. A good check is to find an isolated
interface with a large and sharp impedance change across it, which will give rise to a
strong and isolated seismic loop; inspection of this will reveal the polarity of the data.
In the marine case, it is tempting to use the seabed for this purpose, but care is needed
because the true seabed may have been removed by the application of a trace mute
during processing. In view of the potential for confusion, it is good practice to state
polarity explicitly in discussions of seismic data, with some phrase such as: 'increase of
impedance downwards gives rise to a red (or blue, or black, etc.) loop'. This is a cum-
bersome convention, but is at least clear. In some cases, where conversion to zero-phase
