Geology Reference
In-Depth Information
4.12 4D seismic surveys
Once an oilfield is in production, the oil and/or gas is
extracted and its place in the pore spaces of the reservoir
rock is taken by inflowing groundwater. Since the pore
fluids are changing, the seismic response of the forma-
tions also changes. Even in an extensively developed
field with many wells, there are large intervals between
the wells, of the order of 1 km. It is impossible from mon-
itoring the well-flow to be sure how much of the hydro-
carbon is being extracted from any particular part of the
reservoir. Often oil reservoirs are cut by numerous faults
and some of them may isolate a volume of the reservoir
so that the hydrocarbons cannot flow to the nearby
wells. If the location of such isolated 'pools' can
be found, additional wells can be drilled to extract these
pools and hence increase the overall hydrocarbon recov-
ery from the reservoir.
It is apparent that if the location of such features as the
oil-water contact and gas accumulations can be mapped
with a seismic survey, then repeated surveys at time
intervals during the production of the field offer the
prospect of monitoring the extraction of hydrocarbons,
and contribute to the management of the production
phase of the field operation. This is the rationale for 4D
seismic surveys, which essentially consist of the repeated
shooting of 3D (and often 3C) surveys over a producing
field at regular intervals. The fourth dimension is, of
course, time.
The practical implementation of 4D surveying is far
from simple ( Jack 1997). The essential measurements
made by a seismic survey are the values of amplitudes of
seismic waves at specific locations and times after a seis-
mic source has been fired. Any factor which affects the
location, amplitude or timing of seismic waves must be
allowed for when comparing two sets of data recorded in
different surveys. Obvious effects would be different
geophones in different locations, for each survey. Other
effects are much more subtle. The seasonal change in
level of the water table may be enough to affect the trav-
el time of seismic waves in the near-surface such that all
deep reflections will be systematically mistimed between
two surveys in different seasons. As an oilfield develops,
the increased plant (pumps, drill-rigs, vehicles) changes
(and increases) the background seismic noise with time.
In the processing of the raw data to make the final seismic
sections for comparison many different mathematical
operations change the amplitudes of the data. Each of
these must be rigorously checked and identical process-
ing must be carried out for each separate dataset.
Fig. 4.40
A three-component geophone.
Knowledge of the behaviour of both body waves pro-
vides important additional information. In a lithified
rock formation, such as an oil reservoir, the P-wave is
transmitted through both the rock matrix and the fluids
in the pore spaces. The behaviour of the P-wave
is thus determined by the average of the rock matrix
and pore fluid properties, weighted with respect to the
porosity of the rock.
The S-wave on the other hand is only transmitted
through the rock matrix, since the shear wave cannot
propagate through a fluid. Comparison of the P-wave
and S-wave velocities of the same formation thus can
give information about the porosity of the formation
and the nature of the fluids filling the pore spaces. The
relationships can be complex, but the presence of hy-
drocarbons, especially if accompanied by gas, can be
identified directly from the seismic data in favourable
circumstances. Derivation of measures which reliably
predict the presence of hydrocarbons,
direct hydrocarbon
indicators
(DHIs), is an important part of modern seismic
processing (Yilmaz 1987, 2001), though the details of
this are beyond the scope of this topic.
The ability to detect these features is an enormous
advantage to the hydrocarbon industry and has had a
marked effect on the success rate of exploration bore-
holes in locating oil or gas reservoirs. Since the cost of
drilling a borehole can often reach or exceed $10 m, the
additional effort in seismic data acquisition and process-
ing is very cost-effective.
