Page 272 - Petroleum Geology

247

(Smith, 1980; and I know of a field with similar evidence). This suggests
that there is indeed some critical vertical dimension to the oil accumulation,
but that it is relatively large - 500-1000 m - for these faults.
Now, zero effective permeability to oil or gas in the fault plane does not
mean zero permeability to water in those parts of the fault plane where aqui-
fers are juxtaposed (but it does mean zero effective permeability to water in
the accumulation because the water saturation in the accumulation will be at
irreducible minimum). We conclude from this that no large water-pressure
discontinuity can be generated in juxtaposed aquifers, so that the effective
stress in the aquifer on one side of the fault is not significantly different from
that on the other side.
Migration of petroleum (or indeed water) in a fault plane, up or down, can
only take place in significant quantities under certain conditions:
(a) The fault plane must have permeability.
(b) There must be a potential gradient up or down the fault plane and in
the fault plane.
(c) The beds on either side of the fault plane must have very low perme-
ability, so that the product KAAh/l within the fault plane is large compared
to that in the enclosing beds in a direction parallel to the fault. Note carefully
the factor of area, because the area of a fault normal to the direction of flow
within the fault is very small compared to the area of sedimentary rocks on
either side of the fault.
Evidently these criteria require a tendency for the fault blocks to separate
in beds of very low permeability; and that migration of water or petroleum in
the fault plane, up or down, will terminate at the first permeable bed in which
the pore fluids have less energy than the migrating fluids.
A normal fault is the commonest fault in which there could be a tendency
for the blocks to separate, and the stress field around a normal fault under
flat topography is such that the greatest principal stress is vertical, and the
intermediate and least principal stresses are horizontal. Hubbert (1951, p. 367)
showed that the strength of normal sediments is such that the least principal
stress is compressive below a few hundred metres at most, and the least com-
pressive stress exceeds the cohesive strength of most sedimentary rocks. Thus
it is quite conceivable that faults act as conduits for fluids at shallow depth.
Indeed, seepages along faults are common. The question arises, are there any
conditions that can lead to important fluid migration in fault planes at the
depths of most oil and gas fields?
Secor (1965) concluded that there are. Taking a composite failure envelope
(Fig. 11-8) that satisfies both theory and experiment reasonably well, he de-
rived the following expression for the maximum depth at which open frac-
tures could exist in regions where the greatest principal stress is vertical:
z,, = gOo/Pb&? (1 -1) (11.1)
where uo is the tensile strength of the material, p b is the mean bulk mass den-