Structural interpretation - 3-D Seismic Interpretation - page 90

or using an algorithm that did not take proper account of overburden complexities, then
there may be systematic lateral shifts of interpreted features (e.g. faults) from their true
locations. These lateral shifts will be discussed in section 3.3.4 ; for the moment we
assume that they are not a problem. Then all that is needed to convert the reflectors
mapped in two-way time into depth is a knowledge of the seismic velocity in the
subsurface. Sometimes, a quite detailed velocity model will already have been built
by the seismic processors for migration purposes, but as we shall see these are not
necessarily the best velocities to use for depth conversion.
To develop our ideas, it is useful to look at a real seismic section. A display of the
entire section from surface to target level on a workstation is usually too poor in quality
to use for detailed picking, owing to the limited vertical resolution, but is worth making
to plan the strategy for depth conversion (fig. 3.23) . In this example from the UK
Central North Sea, the objective is at or just above the orange horizon, which is the top
of the Ekofisk Formation. Two horizons have been picked in the overburden. They are
levels at which there is substantial discontinuity in the curve of sonic velocity against
depth at a nearby well (fig. 3.24) . The yellow horizon is encountered at a level of about
1300 ms. Above this level, ignoring noise, velocity is nearly constant; at the horizon
there is a slight decrease in velocity, and then velocity increases fairly steadily with
depth to the top of the Sele Formation (green marker in fig. 3.23 ) , at which point there
is an increase in velocity and a more complicated velocity-depth trend which is only
roughly approximated by a linear increase with depth. This suggests that a three-layer
model would be suitable, with constant velocity in the top layer and velocity increasing
with depth in the other two. A similar analysis needs to be carried out at an early stage
of every interpretation, as it determines which overburden layers need to be picked to
carry out the depth conversion. Sometimes, as here, only a few surfaces are needed
to give a reasonable approximation. At other times quite a large number of surfaces
may be needed, if there are large velocity jumps at a number of horizons; this might be
the case if carbonates or evaporites are intercalated within a sand/shale sequence. To
decide whether a given layer is worth including in the model, it is easy to calculate the
error introduced at the well by treating it as part of an adjacent layer. To assess whether
the error is important is harder, and depends on the detailed geometry of the structure
being mapped; critical areas to look at will usually be the culmination and the possible
spillpoint of any structural closure.
If the velocity within a layer is constant, it is obvious how to convert the two-way
time thickness into a thickness in depth. If there is a velocity gradient with depth, we
proceed as follows. Suppose we have a layer which extends from depth z 1 to depth z 2 ,
at which the two-way times are t 1 and t 2 respectively, and that the velocity at any depth
within the layer is given by
v = v 0 + kz .
In such a formula, v is often referred to as an instantaneous velocity ; it describes the
actual seismic velocity at a particular depth (or travel-time) and may be contrasted with