predicted from depositional facies and provenance
studies which define the starting composition for
burial diagenesis. Both observations and theoretical
arguments suggest that advective transport can not
significantly change the rock compostion below the
reach of meteoric water flow. Short distance transport
by diffusion may nevertheless be important.
Quartz cementation was often interpreted to occur
as events of relatively short duration (approximately
<10 million years) that could start and stop late in a
sandstone’s burial history (Robinson and Gluyas
1992). This would imply that the quartz cementation
was controlled by advective transport of silica in solution and the source of silica from other reactions. Fluid
inclusion data, however, shows that quartz cementation occurs throughout the temperature range
corresponding to the burial history.
Modelling of quartz precipitation (Walderhaug
1996, Olkers et al. 2000, Walderhaug et al. 2001) is
based on the assumption that this is a continuous
process controlled by the kinetics and therefore by
temperature. This is the basis for the practical models
widely adopted by the petroleum industry (e.g. Exemplar and Touchstone).
The assumption that the quartz precipitation is the
rate limiting factor may not always be strictly true in
clean quartz arenites where silica sources (stylolites)
are widely spaced, resulting in decreasing quartz
cementation away from the stylolites (Walderhaug
and Bjørkum 2003). The modelling does, however,
require that the burial curve and the temperature as a
function of geological time are known. It is also very
sensitive to changes in the primary sediment composition, which strongly influence both the mechanical
compaction and the chemical reactions. The porosity
loss and increased sediment density resulting from
chemical compaction as a function of temperature
cause basin subsidence (Bjørkum et al. 1998, 2001;
Bjørkum and Nadeau 1998). Temperature-driven
chemical compaction results in a volume reduction
(shrinkage) and the strain is then independent of effective stress. As a result, differential stresses in siliceous
rock will be relaxed by the compaction processes
during basin subsidence as long at the temperature
exceeds about 80
C (Bjørlykke 2006).
4.16 Porosity/Depth Trends in
Sedimentary Basins
Data from wells in sedimentary basins which have
undergone almost continuous subsidence can be
regarded as records of a natural compaction experiment. We can use the log porosity and in the cored
intervals we have data from core plugs.
At depths shallower than about 2 km (80–100
C)
we can compare the log porosities with experimental
compaction of similar sands in the laboratory. There
will then be a marked effect of overpressure reducing
the effective stress. Poorly sorted sands will lose most
of their porosity at relatively shallow depth but well
sorted sand may have 30–35% porosity at 2–3 km
depth, which corresponds to experimental compaction
at about 20–30 MPa effective stress. This suggests that
there is little creep over long geologic time.
Data from deep wells will always show a trend
towards lower porosities with depth but there may
also be intervals where this trend is reversed. This
does not mean that net porosity has been created by
diagenetic processes. Because of the very low solubility of silica and even more so of aluminium in
porewater it is very difficult to explain how minerals
in several metre thick sandstones can be dissolved
without precipitation of other minerals in the same
sandstones. Each lithology will have a characteristic
porosity depth trend (Fig. 4.16) and increases in porosity reflect variations in the primary composition.
As discussed above the rate of quartz cementation
can be modelled if the surface area available for quartz
cementation and the time-temperature history during
burial are known.
At about 4.0 km burial depth (120–140
C) the
amount of quartz cement may be 10–15% so the
remaining porosity may be only 10–15%. We do however find good reservoir quality (>20% porosity) at
greater depths and higher temperatures but this is due
to grain coatings. Prediction of porosity at great depth
therefore requires that the occurrence of coating of
chlorite, illite, haematite or microquartz can be
predicted. Such prediction of the primary sediment
composition must again be linked to facies and
provenance.
142
K. Bjørlykke and J. Jahren
studies which define the starting composition for
burial diagenesis. Both observations and theoretical
arguments suggest that advective transport can not
significantly change the rock compostion below the
reach of meteoric water flow. Short distance transport
by diffusion may nevertheless be important.
Quartz cementation was often interpreted to occur
as events of relatively short duration (approximately
<10 million years) that could start and stop late in a
sandstone’s burial history (Robinson and Gluyas
1992). This would imply that the quartz cementation
was controlled by advective transport of silica in solution and the source of silica from other reactions. Fluid
inclusion data, however, shows that quartz cementation occurs throughout the temperature range
corresponding to the burial history.
Modelling of quartz precipitation (Walderhaug
1996, Olkers et al. 2000, Walderhaug et al. 2001) is
based on the assumption that this is a continuous
process controlled by the kinetics and therefore by
temperature. This is the basis for the practical models
widely adopted by the petroleum industry (e.g. Exemplar and Touchstone).
The assumption that the quartz precipitation is the
rate limiting factor may not always be strictly true in
clean quartz arenites where silica sources (stylolites)
are widely spaced, resulting in decreasing quartz
cementation away from the stylolites (Walderhaug
and Bjørkum 2003). The modelling does, however,
require that the burial curve and the temperature as a
function of geological time are known. It is also very
sensitive to changes in the primary sediment composition, which strongly influence both the mechanical
compaction and the chemical reactions. The porosity
loss and increased sediment density resulting from
chemical compaction as a function of temperature
cause basin subsidence (Bjørkum et al. 1998, 2001;
Bjørkum and Nadeau 1998). Temperature-driven
chemical compaction results in a volume reduction
(shrinkage) and the strain is then independent of effective stress. As a result, differential stresses in siliceous
rock will be relaxed by the compaction processes
during basin subsidence as long at the temperature
exceeds about 80
C (Bjørlykke 2006).
4.16 Porosity/Depth Trends in
Sedimentary Basins
Data from wells in sedimentary basins which have
undergone almost continuous subsidence can be
regarded as records of a natural compaction experiment. We can use the log porosity and in the cored
intervals we have data from core plugs.
At depths shallower than about 2 km (80–100
C)
we can compare the log porosities with experimental
compaction of similar sands in the laboratory. There
will then be a marked effect of overpressure reducing
the effective stress. Poorly sorted sands will lose most
of their porosity at relatively shallow depth but well
sorted sand may have 30–35% porosity at 2–3 km
depth, which corresponds to experimental compaction
at about 20–30 MPa effective stress. This suggests that
there is little creep over long geologic time.
Data from deep wells will always show a trend
towards lower porosities with depth but there may
also be intervals where this trend is reversed. This
does not mean that net porosity has been created by
diagenetic processes. Because of the very low solubility of silica and even more so of aluminium in
porewater it is very difficult to explain how minerals
in several metre thick sandstones can be dissolved
without precipitation of other minerals in the same
sandstones. Each lithology will have a characteristic
porosity depth trend (Fig. 4.16) and increases in porosity reflect variations in the primary composition.
As discussed above the rate of quartz cementation
can be modelled if the surface area available for quartz
cementation and the time-temperature history during
burial are known.
At about 4.0 km burial depth (120–140
C) the
amount of quartz cement may be 10–15% so the
remaining porosity may be only 10–15%. We do however find good reservoir quality (>20% porosity) at
greater depths and higher temperatures but this is due
to grain coatings. Prediction of porosity at great depth
therefore requires that the occurrence of coating of
chlorite, illite, haematite or microquartz can be
predicted. Such prediction of the primary sediment
composition must again be linked to facies and
provenance.
142
K. Bjørlykke and J. Jahren