Seismic Reflection Technique
General Seismic Reflection Techniques
Introduction to Seismic Reflection
The
physical process of reflection is illustrated in FIGURE
1 where raypaths
for successive layers are shown. There are commonly several layers
beneath the earth's surface which contribute reflections to a single
seismogram.
Thus, seismic reflection data are more complex than refraction data because it is these later arrivals that yield information about the deeper layers. More noise is present at later times on the record making reflections difficult to extract from the unprocessed record.
|
FIGURE 2 indicates the paths of arrivals that would be recorded on a multichannel seismograph. Note that the figure indicates that the subsurface coverage is exactly one-half of the surface distance across the geophone spread. The subsurface sampling interval is one half of the distance between geophones on the surface.
Another important feature of modern reflection data acquisition is illustrated below. If multiple shots, S1 and S2 are recorded by multiple receivers, R1 and R2, and the geometry is as shown in the figure, the reflector point for both rays is the same. However, the raypaths are not the same length, thus the reflection will occur at different times on the two traces. This time delay is called "normal-moveout". With an appropriate time shift, called the normal-moveout correction, the two traces (S1 to R2 and S2 to R1) can be summed, greatly enhancing the reflected energy and canceling spurious noise.
In FIGURE 3, this method is variously called the common reflection point, common midpoint, or common depth point (CDP) method. If all receiver locations are used as shot points, the multiplicity of data on one subsurface point (call "CDP fold") is equal to one-half of the number of recording channels. Thus a 24-channel seismograph will record 12-fold data if a shot corresponding to ever receiver position is shot into a full spread of geophones. Thus for 12-fold data every subsurface point will have 12 separate traces summed to represent that point after appropriate time shifts of each trace.
Arrivals
on a seismic reflection record can be seen in FIGURE
4. The receivers
are arranged to one side of a shot which is 15 m (50 ft) from the
first geophone. The geophone spacing is 3 meters (10ft). Various arrivals
are identified on the figure. Note that the gain is increased down
the trace to maintain the signals at about the same size by a process
known as automatic gain control(AGC). One side of the traces is shaded
to enhance the continuity between traces.
The ultimate product of a seismic reflection survey is a corrected cross-section with reflection events in their true subsurface positions. Though more than 70 years of development have gone into the seismic reflection method in the search for petroleum, the use of reflection for the shallow subsurface (less than 50 m) remains an art. This discussion cannot give every detail of the acquisition and processing of shallow seismic reflection data. The difference between deep petroleum-oriented reflection and shallow reflection work suitable for engineering and environmental applications are principally cost and frequency bandwidth.
One measure of the nominal frequency content of a pulse is the inverse of the time between successive peaks. In the shallow subsurface, the exploration objectives are often at depths of 15 to 45 m (45-150ft). At 450 m/s (1500 ft/s) a pulse with 10 milliseconds peak to peak (nominal frequency of 100 Hz) is 45 meters (150 feet) long. To detect(much less differentiate between) shallow, closely spaced layers, pulses with nominal frequencies at or above 200 Hz may be required. A value of 1,600 m/s (5000 ft/s) is used as a representative velocity corresponding to saturated, unconsolidated materials because without saturated sediments, both attenuation and lateral variability make reflection generally difficult.
The daily crew cost for a petroleum exploration crew ranges upwards from five figures.
Seismic Reflection Field
Techniques
A shallow seismic reflection crew consists of three to six persons.
The equipment used allows two to three times the number of active geophones
to be distributed along the line. A switch (called a "roll-along switch")
allows the seismograph operator to select the particular set of geophones
required for a particular shot from a much larger set of geophones
that have been previously laid out. The operator can then switch the
active array down the line as the position of the shot progresses.
Often the time for a repeat cycle of the source and the archiving time
of the seismograph are the determining factors in the production rates.
With sufficient equipment, one or two persons
can be continually shifting equipment forward on the line while a shooter
and an observer are sequencing through the available in-place equipment.
If the requirements for relative and absolute surveying are taken care
of at a separate time, excellent production rates, in terms of number
of shot points per day, can be achieved. Rates of one shot point per minute
or 3-400 shot points/normal field day can be achieved. Note that the spacing
of these shot points may be only 0.6-1.2 m (1-2 ft), so the linear progress
may be only 300 m (1000 ft) of line for very shallow surveys. Also note
that the amount of data acquired is enormous. A 24-channel record sampled
every 1/8 millisecond that is 200 milliseconds long consist of nearly
60,00 thirty-two bit numbers or upwards of 240 KB/record. Three hundred
records might represent more that 75 MB of data for one day of shooting.
Field data acquisition parameters are highly site specific. Up to
a full day of testing with a knowledgeable consultant experienced in
shallow seismic work may be required. The objective of these tests is
demonstrable reflections on the raw records. If arrivals consistent
with reflections from the zone of interest cannot be seen, the chances
that processing will recover useful data are slim.
One useful testing technique is the walkaway noise test. A closely
spaced set of receivers is set out with a geophone interval equal to
1 or two percent of the depth of interest-often as little 30 or 60 cm
(1-2 ft) for engineering applications. By firing shots at different
distances from this spread, a well-sampled long-offset spread can be
simulated. Variables can include geophone arrays, shot patterns, high
and low-cut filters and AGC windows, among others. Software by INTERPEX,
LTD. of Golden, Colorado is used by MicroGeophysics Corporation
to evaluate field tests.
Processing is typically done by professionals using special purpose
computers. These techniques are expensive but technically robust and
excellent results can be achieved. Exposition of all the processing
variables is well beyond the scope of this discussion. However a close
association of the geophysicist, the processor, and the consumer is
absolutely essential if the results are to be useful. Well logs, known
depths to marker beds, results from ancillary methods, and the expected
results should be furnished to the processor. At least one iteration
of the results should be used to ensure that the final outcome is successful.
One important conclusion of the processing is a true depth section.
The production of depth sections requires conversion of the times of
the reflections to depths by derivation of a velocity profile. Well
logs and check-shot surveys in holes are often necessary to confirm
the accuracy of this conversion.
These warnings are important because the powerful processing algorithms
can produce very appealing but erroneous results. Most seismic data
processors are oriented to petroleum exploration and volume production.
The effort and cooperation required by both the geophysicist and the
processor for a shallow reflection survey are beyond that normal in
petroleum scenarios.
Conclusions-Seismic Reflection
1. It is possible to obtain seismic reflections from very shallow depths,
perhaps as shallow as 3 to 5 m (10-16 ft).
2. Variations in field technique are required
depending on site and depth parameters.
3. Success is greatly increased if shots and geophones are near or
in the saturated zone.
4. Severe low-cut filters and arrays of a small (1-5) number of geophones
are required.
5. If possible, reflections should be visible on the field records
after all recording parameters are optimized.
6. Data processing should be guided by the appearance of the field
records and extreme care should be used not to stack refractions or
other unwanted artifacts as reflections.
Okay, where do I start?
Additional considerations you should contemplate are given under the
logistical portion of the methods section. If seismic reflection seems
to be the weapon of choice, get together a suspected stratigraphic section
with rock types and thicknesses and call MicroGeophysics Corporation.
We will review your problem and generate a synthetic seismogram for
different input frequencies. This step, based on the inferred geology,
is vital to forecasting the probability of success and the cost of your
survey. We look forward to solving your geologic problem with a geophysical
solution.
.
MicroGeophysics Corporation (MGC) will be happy to respond
to your questions!!
Phone (1-800-GEOPHYSics) fax (303-424-0807) or E-mail
(microgeo@aol.com).
