produce fluid flow that is often referred to as seismic
pumping. This mechanism may work in metamorphic
and well-cemented sedimentary rocks where the rock
strength is high enough to keep fractures from being
closed by tectonic stress. In these types of brittle rocks
most of the water is present in the fractures and little in
the rock matrix. Softer sediments are by contrast more
ductile in their response to tectonic stress and fractures
will normally not stay sufficiently open to transmit
fluid rapidly. In compacting (normally consolidated)
sediments most of the water is in the sediment matrix
and even if fractures should remain open the ratelimiting step is the flow of porewater from the matrix,
which often consists of low permeability mudstones
and shales.
If the low permeability seal overlying or
surrounding an overpressured part of the basin is broken, i.e. through faulting, a rapid pulse of porewater
flow upwards may follow. However due to the low
compressibility of water (4:3 Â 10
À10 Pa
À1 ) the
upward flow of porewater necessary to reduce the
pressure is relatively small. For an overpressure of
10
7 Pa (potentiometric surface 1 km above sea level),
the average expansion of the water would be
4.3Â10
À3 (Leonard 1993). In a vertical column
through an overpressured sequence this would result
in an average upwards flow of 4.3 m for each km of
sequence, if all the overpressure was released at one
time. If the flow was focused through a smaller crosssection, this figure would increase proportionally.
It is very unlikely that large volumes of shales
would reduce their overpressure over a very short
time, given their low permeabilities, so the potential
for episodes of rapid flow of compaction water is
limited. If the water is saturated with respect to gases
like carbon dioxide or methane the compressibility of
the pore fluid will increase significantly. A reduction
in the pressure then will cause gas to come out of
solution and form a separate phase, which has a high
compressibility. The degree of overpressure is reduced
by porewater flow through a leaking seal, but only by a
small amount. The pressure will not drop much below
fracture pressure before the fine fractures close. In the
relatively shallow section this small increase in effective stress may result in some mechanical compaction.
Chemical compaction (>100
C), however, occurs at a
very slow rate which is mostly a function of temperature and time and is relatively independent of the
pressure changes related to fracturing. Compaction
will therefore slowly build up the pore pressure again
unless there is continued flow from the overpressured
section.
As we saw from the calculations, the flow resulting
directly from the pressure release is rather limited.
The main expulsion of porewater is due to sediment
compaction, which is an indirect consequence of the
reduction of overpressure and increase in effective
stress (effective stress ¼ overburden stress minus
pore pressure). Compaction and expulsion of
porewater resulting from increased effective stress is
a gradual and rather slow process and this strongly
influences the rate of water supply to the faults from
mudstones. The permeability of the surrounding
mudstones further limits the rate of flow into the
fault zone. If a fault plane does not extend up to the
surface (or seafloor), the upwards-moving porewater
will have to be accommodated in shallower strata.
Large volumes of porewater can not suddenly be
injected into shallower sandstones even if the latter
are normally pressured.
Flow into shallow aquifers of limited extent will
result in a temporary pressure build-up before the
water can be displaced, and this will reduce the flow
rate. The highest flow rates can be expected when the
fault extends all the way to the surface so that the
porewater can escape into the water column or onto
the land surface. In many sedimentary basins like the
North Sea basin, most of the faults extend only into the
Cretaceous or lower Tertiary section. These faults were
therefore not very active during the early Tertiary, and
certainly not during the later Tertiary and Quaternary.
The timing of faulting must be taken into account when
faults are called upon to explain fluid flow and diagenetic reactions in sedimentary basins. In a sequence
with a high clay/sand ratio, faults may be sealing
between two sandstones due to the clay smear on the
fault plane.
10.11 Episodic Flow
When the source of the fluids is hydrothermal the flow
may be episodic, at least when considering individual
fractures or a limited area, because new fractures
develop and close. Igneous intrusions (sills and
dykes) may cause boiling and episodic flow. For a
296
K. Bjørlykke
pumping. This mechanism may work in metamorphic
and well-cemented sedimentary rocks where the rock
strength is high enough to keep fractures from being
closed by tectonic stress. In these types of brittle rocks
most of the water is present in the fractures and little in
the rock matrix. Softer sediments are by contrast more
ductile in their response to tectonic stress and fractures
will normally not stay sufficiently open to transmit
fluid rapidly. In compacting (normally consolidated)
sediments most of the water is in the sediment matrix
and even if fractures should remain open the ratelimiting step is the flow of porewater from the matrix,
which often consists of low permeability mudstones
and shales.
If the low permeability seal overlying or
surrounding an overpressured part of the basin is broken, i.e. through faulting, a rapid pulse of porewater
flow upwards may follow. However due to the low
compressibility of water (4:3 Â 10
À10 Pa
À1 ) the
upward flow of porewater necessary to reduce the
pressure is relatively small. For an overpressure of
10
7 Pa (potentiometric surface 1 km above sea level),
the average expansion of the water would be
4.3Â10
À3 (Leonard 1993). In a vertical column
through an overpressured sequence this would result
in an average upwards flow of 4.3 m for each km of
sequence, if all the overpressure was released at one
time. If the flow was focused through a smaller crosssection, this figure would increase proportionally.
It is very unlikely that large volumes of shales
would reduce their overpressure over a very short
time, given their low permeabilities, so the potential
for episodes of rapid flow of compaction water is
limited. If the water is saturated with respect to gases
like carbon dioxide or methane the compressibility of
the pore fluid will increase significantly. A reduction
in the pressure then will cause gas to come out of
solution and form a separate phase, which has a high
compressibility. The degree of overpressure is reduced
by porewater flow through a leaking seal, but only by a
small amount. The pressure will not drop much below
fracture pressure before the fine fractures close. In the
relatively shallow section this small increase in effective stress may result in some mechanical compaction.
Chemical compaction (>100
C), however, occurs at a
very slow rate which is mostly a function of temperature and time and is relatively independent of the
pressure changes related to fracturing. Compaction
will therefore slowly build up the pore pressure again
unless there is continued flow from the overpressured
section.
As we saw from the calculations, the flow resulting
directly from the pressure release is rather limited.
The main expulsion of porewater is due to sediment
compaction, which is an indirect consequence of the
reduction of overpressure and increase in effective
stress (effective stress ¼ overburden stress minus
pore pressure). Compaction and expulsion of
porewater resulting from increased effective stress is
a gradual and rather slow process and this strongly
influences the rate of water supply to the faults from
mudstones. The permeability of the surrounding
mudstones further limits the rate of flow into the
fault zone. If a fault plane does not extend up to the
surface (or seafloor), the upwards-moving porewater
will have to be accommodated in shallower strata.
Large volumes of porewater can not suddenly be
injected into shallower sandstones even if the latter
are normally pressured.
Flow into shallow aquifers of limited extent will
result in a temporary pressure build-up before the
water can be displaced, and this will reduce the flow
rate. The highest flow rates can be expected when the
fault extends all the way to the surface so that the
porewater can escape into the water column or onto
the land surface. In many sedimentary basins like the
North Sea basin, most of the faults extend only into the
Cretaceous or lower Tertiary section. These faults were
therefore not very active during the early Tertiary, and
certainly not during the later Tertiary and Quaternary.
The timing of faulting must be taken into account when
faults are called upon to explain fluid flow and diagenetic reactions in sedimentary basins. In a sequence
with a high clay/sand ratio, faults may be sealing
between two sandstones due to the clay smear on the
fault plane.
10.11 Episodic Flow
When the source of the fluids is hydrothermal the flow
may be episodic, at least when considering individual
fractures or a limited area, because new fractures
develop and close. Igneous intrusions (sills and
dykes) may cause boiling and episodic flow. For a
296
K. Bjørlykke