Page 170 - The Petroleum System From Source to Trap

162 Downey

examination of the distribution and composition of
hydrocarbon shows and production.

F A ULTS AS SEALS OR LEAKS
An enormous body of dogma is quotable from
technical literature providing opinions that particular
faults "must have" leaked or "must have" sealed.
�ithout arguing a ainst anyone's particular prejudices,
!?
It seems worthwhile to provide some guidance as to
when faulting can aid entrapment and when faulting
c a uses leakage from traps.
Considerable efforts are commonly made by inter­
�
p reters to id ntify and map faults; this is proper and
Important. It IS, however, only the first step toward eval­
uating hydrocarbon entrapment. On an interpretative
map, the black line identifying a fault trace does not seal Figure 8.1. Leakage of hydrocarbons along the fault plane
hydrocarbons (or cause them to leak, for that matter). of tensional faults is likely at shallow depths and common
The fault trace is itself only a representation of the in geopressures.
narrow discontinuity where capillary properties, fluid
properties, and structural dips of the co-joined rock
layers have a high probability of being altered. . There � e a few excep�ons to this statement. Faulting
The fault plane itself can behave as an open fracture in m areas WI�h abundant thick layers of plastic clays (such
three particular circumstances: ( 1 ) a fault plane will as the Tertiary of the Gulf of Mexico and Niger Delta)
generally behave as a transmissive open fracture in can cause plastering of clay smears across permeable
teriSional settings at shallow depths; (2) a fault plane will horizons along considerable portions of a fault surface
often transmit fluids in teriSional settings within geopres­ (Smith, 1966, 1980; Weber et al., 1978; Bouvier et al.,
sures; and (3) the f a ult plane can be expected to transmit 1989). Also, considerable granulation along the fault
hydrocarbons during fault movement (Downey, 1990). surface can locally alter lateral transmissibility in juxta­
�
In each of these cases, the fault may behave as an open po ed . � servoirs. F�lly, fluid movement through reser­
fracture and provide passage of hydrocarboriS along the vOirs JOined by faulting can induce pore-filling diagen­
plane of the teriSional fault (Figure 8.1). It is sometimes esis in the reservoirs near a fault (Foster et al., 1987). Such
thought that leakage from such hairline-width openings phenomena are real and create important impediments
is negligible; nothing could be farther from the truth. that profoundly affect reservoir production performance.
:rrariSport capa�ty (permeability) along planar openings It is doubtful, however, that they create absolute lateral
IS enormous; a smgle planar opening 0.001 in. wide in the seals to hydrocarbon movement over a geologic time
seal overlying a 500-ft oil column could cause leakage at scale.
a rate exceeding 150 million bbl per 1000 yr (see figure 7, A simple calculation can serve to indicate the
Downey, 1984, for details of calculation). magnitude of the difficulties in relying on an absolute in
To summarize, (1) it is unlikely that a pure fault trap situ seal to be created along a fault plane. Assume that
for �ydrocarbon wou_ld exist at very shallow depths; (2) the area of the fault-joined reservoirs is 1 mi by 100 ft.
�
a high level of nsk eXIsts for teriSional fault seals within What degree of perfection of clay smearing is necessary
overpressured sectioriS; and (3) leakage can occur during to ensure that the juxtaposed reservoirs will not allow
fault movements. In active compressional settings, the leakage through the fault plane? Let's take the example
�
fa lt lane is rarely a pathway for hydrocarbon of a Gulf of Mexic !ertiary sandstone of 30% porosity,
�
J?
�
migration, as the fault plane itself is almost never an 1000-md perm abihty, and 20-psi differential driving
�
open fracture. pressur . I f a smgle square foot of the entire fault-joined
surface IS left unaltered and the 20-psi driving pressure is
dissipated over a distance of 1 ft, then leakage at a rate of
When a Fault Is a Seal about 10 billion bbl/ m.y. can occur.
A fault plane surface is often invoked by interpreters
as a seal to laterally migrating hydrocarbons; this rarely Maps of Fault Planes
h ppens. Technical interpretations often stop after identi­
�
fying the fault plane in the mistaken belief that the fault Faults are most commonly seen and studied in the
plane itself creates a lateral seal. A fault plane is not a tw � dimensions of a seismic section. Mapping of an
magical membrane, an impervious seal iriStantly created entire fault plane is necessary if the interpreter wants to
�
�
J
by the fault. In the overwhelming predominance of cases, t ?derstand l teral eal risk on a fault-sealed prospect. A
hydrocarbons pass readily from a permeable horizon on smgle two-dimensiOnal cross section at right angles to
�
one side of the fault plane to a juxtaposed permeable th fault can serve to describe the fault seal only at those
horizon on the other side of the fault plane. pomts.