Geological Survey of Denmark and Greenland Bulletin
|
|
Nr. 4, Review of Survey activities 2003, pp. 9-12 |
Please use the pdf-file for printing: nr4_p09-12.pdf (~220kb)
PDF-files requires a pdf-reader, Acrobat Reader GSview or similar
9
In an oil reservoir, the geometry of the interface between
water and oil is critical in determining the volume of oil
trapped below the top seal. If the interface is planar and hori-
zontal, the volume calculation is fairly simple, but if the
interface is tilted or undulating, estimation of the volume of
the trapped oil is complex as it depends on the combined
structural and fluid contact geometry. Since accumulation of
the oil may take place over a time span of several million
years, while the reservoir is experiencing burial and com-
paction, the charge history must be studied using dynamic
methods that account for these changes and for flow in both
the oil and water phases. These processes have been studied
quantitatively at the Geological Survey of Denmark and
Greenland (GEUS) in a project that has combined the burial
model with a fluid flow simulator. The modelling study shows
that filling of a chalk reservoir can have a very long and com-
plex history dominated by very low fluid flow rates (cm/year).
water and oil is critical in determining the volume of oil
trapped below the top seal. If the interface is planar and hori-
zontal, the volume calculation is fairly simple, but if the
interface is tilted or undulating, estimation of the volume of
the trapped oil is complex as it depends on the combined
structural and fluid contact geometry. Since accumulation of
the oil may take place over a time span of several million
years, while the reservoir is experiencing burial and com-
paction, the charge history must be studied using dynamic
methods that account for these changes and for flow in both
the oil and water phases. These processes have been studied
quantitatively at the Geological Survey of Denmark and
Greenland (GEUS) in a project that has combined the burial
model with a fluid flow simulator. The modelling study shows
that filling of a chalk reservoir can have a very long and com-
plex history dominated by very low fluid flow rates (cm/year).
The resulting modelled present-day situation exhibits a very
irregular oil distribution and a non-planar geometry of the
fluid contacts, and shows marked similarities to that shown
by the field data.
irregular oil distribution and a non-planar geometry of the
fluid contacts, and shows marked similarities to that shown
by the field data.
Oilwater contact and free water level
The positions of the oilwater contact (OWC), the gasoil
contact (GOC) and the associated free water level (FWL) in
an oil- and gas-field are some of the most important factors
in estimating the in-place hydrocarbon volumes of a given
field. Thus it is important to be able to analyse and predict
tilted or irregular fluid contacts (Dennis et al. 2000; Moss et
al. 2003; Dennis et al. in press; Vejbæk et al. in press).
contact (GOC) and the associated free water level (FWL) in
an oil- and gas-field are some of the most important factors
in estimating the in-place hydrocarbon volumes of a given
field. Thus it is important to be able to analyse and predict
tilted or irregular fluid contacts (Dennis et al. 2000; Moss et
al. 2003; Dennis et al. in press; Vejbæk et al. in press).
The fluid contact can be defined in two radically different
ways: The OWC is defined by setting a threshold for the oil
saturation, whereas the FWL is defined where the pressures
saturation, whereas the FWL is defined where the pressures
The history of hydrocarbon filling of Danish chalk fields
Peter Frykman, Ole V. Vejbæk, Niels Bech and Carsten M. Nielsen
Fig. 1. Map showing top chalk depth
structure for the Danish North Sea area.
Producing chalk fields are shown, with
oil fields green, and gas fields red.
Colour interval is 100 m and contour
interval 50 m. The red line on the
Kraka field shows location of the pro-
file studied (see Fig. 3). Full black
lines are major faults. Dashed black
lines are offshore sector boundaries.
Modified from Vejbæk et al. (in press).
Geological Survey of Denmark and Greenland Bulletin 4, 912 (2004) © GEUS, 2004
10
in the water and the oil phases are equal. In the chalk reser-
voirs in the North Sea, the relationship between the OWC
and the FWL can be described in simple cases by the capil-
lary characteristics of the reservoir rock.
voirs in the North Sea, the relationship between the OWC
and the FWL can be described in simple cases by the capil-
lary characteristics of the reservoir rock.
In the central North Sea (Fig. 1), the fluid contacts in the
Chalk can be naturally tilted by hydrodynamic activity due to
a regional flow of water in the chalk. A regional pressure gra-
dient in the chalk aquifer has been described from available
pressure measurements (Megson 1992), and later refined us-
ing more data (Dennis et al. in press).
a regional flow of water in the chalk. A regional pressure gra-
dient in the chalk aquifer has been described from available
pressure measurements (Megson 1992), and later refined us-
ing more data (Dennis et al. in press).
The regional lateral pressure gradient reflects differential
compaction caused by rapid Neogene deposition with the
highest burial rates in the central Ekofisk area (Japsen 1998).
The water therefore migrates laterally away from this area and
towards the periphery of the North Sea.
highest burial rates in the central Ekofisk area (Japsen 1998).
The water therefore migrates laterally away from this area and
towards the periphery of the North Sea.
Analysis of burial history by backstripping and decom-
paction shows that this pressure was probably caused mainly
by rapid deposition in the time interval from latest Miocene
to Recent times, as the magnitude of the pressure corre-
sponds to the thickness of these deposits (Japsen 1998). This
is consistent with a very low regional permeability of the
chalk (and adjacent sedimentary packages) probably not exceed-
ing 1 mD.
by rapid deposition in the time interval from latest Miocene
to Recent times, as the magnitude of the pressure corre-
sponds to the thickness of these deposits (Japsen 1998). This
is consistent with a very low regional permeability of the
chalk (and adjacent sedimentary packages) probably not exceed-
ing 1 mD.
The flow of water and the accompanying pressure diffe-
rences will influence the position of the FWL (Fig. 2A). If the
oil is also flowing due to either buoyancy equilibration or
active migration, it will affect both the FWL and the GOC
(Fig. 2B, C).
oil is also flowing due to either buoyancy equilibration or
active migration, it will affect both the FWL and the GOC
(Fig. 2B, C).
Factors that modify the position of the FWL include tilt-
ing due to structural movements, and the presence of oil
migrating from the underlying source rocks into the reser-
voir. The reason that these processes influence the present
geometry of the FWL is that both oil and water flow take
place at very low velocities (cm/year), due to the low perme-
ability of the chalk. Even though structural movements are
very slow, the flow is not able to respond quickly enough to
equilibrate the system, even on a scale of millions of years.
migrating from the underlying source rocks into the reser-
voir. The reason that these processes influence the present
geometry of the FWL is that both oil and water flow take
place at very low velocities (cm/year), due to the low perme-
ability of the chalk. Even though structural movements are
very slow, the flow is not able to respond quickly enough to
equilibrate the system, even on a scale of millions of years.
The petrophysical properties of the North Sea chalk reser-
voirs are mainly governed by their high-porosity/low-perme-
ability aspect with porosities usually around 2040% and
average permeability of 1 mD.
ability aspect with porosities usually around 2040% and
average permeability of 1 mD.
Case study
The Kraka field in the southern Danish North Sea (Fig. 1)
has been chosen as the subject of a case study of primary oil
charging and remigration. To study the interaction of the dif-
ferent processes, reservoir fluid flow simulation techniques
have been applied in combination with burial modelling,
including compaction (Vejbæk 2002). The results show that
a time span in the order of 2 Ma is required for the hydro-
carbons to reach the top of the reservoir in an approximately
equilibrium state, if they enter the reservoir section from a
flank position. However, not even dynamic equilibrium can
be fully obtained in this time span if re-perturbation by struc-
tural movements leads to changing water-zone pressure gra-
dients.
has been chosen as the subject of a case study of primary oil
charging and remigration. To study the interaction of the dif-
ferent processes, reservoir fluid flow simulation techniques
have been applied in combination with burial modelling,
including compaction (Vejbæk 2002). The results show that
a time span in the order of 2 Ma is required for the hydro-
carbons to reach the top of the reservoir in an approximately
equilibrium state, if they enter the reservoir section from a
flank position. However, not even dynamic equilibrium can
be fully obtained in this time span if re-perturbation by struc-
tural movements leads to changing water-zone pressure gra-
dients.
The study is focused on a 2D section from the crestal part
through the south-eastern flank of the Kraka field (Figs 1, 3).
Since porosity is the main cause for changes in seismic imped-
ance (Japsen et al. in press), detailed porosity profiles can be
achieved by converting acoustic impedance derived by seis-
mic inversion. These porosity profiles have been modified by
backstripping to reconstruct geometry and porosity. As flow
simulation has only been applied to the Chalk Group layers,
Since porosity is the main cause for changes in seismic imped-
ance (Japsen et al. in press), detailed porosity profiles can be
achieved by converting acoustic impedance derived by seis-
mic inversion. These porosity profiles have been modified by
backstripping to reconstruct geometry and porosity. As flow
simulation has only been applied to the Chalk Group layers,
Fig. 2. Possible dynamic equilibrium situations that may fit a tilted
oilwater contact. Arrows show direction of pressure drop correspond-
ing to flow direction. 1 and 2 represent wells where the pressure depth
plots shown to the left are generated: (A) only the water phase is
dynamic; (B) both oil and water are flowing, but the tilt is maintained
due to a higher lateral pressure gradient in the water phase; (C) the tilt
is maintained only by an oil phase gradient. The situations are physically
distinguishable by the dip of the gasoil contact. Modified from Vejbæk
et al. (in press).
11
detailed porosity profiles have only been constructed for
these layers.
these layers.
The simulation of flow processes in chalk reservoirs is
characterised by the need for end-point scaling and hyste-
resis, in order to account for the marked influence from the
high capillary forces in this low-permeability medium. Since
the dominant process during the filling history is a drainage
process (i.e. oil replacing water), the saturation functions
must also be derived for this type of process. There is a ge-
neral lack of relative permeability analyses for drainage, and
therefore imbibition curves have been the guide for establish-
ing the drainage saturation functions for relative permeabi-
lity.
resis, in order to account for the marked influence from the
high capillary forces in this low-permeability medium. Since
the dominant process during the filling history is a drainage
process (i.e. oil replacing water), the saturation functions
must also be derived for this type of process. There is a ge-
neral lack of relative permeability analyses for drainage, and
therefore imbibition curves have been the guide for establish-
ing the drainage saturation functions for relative permeabi-
lity.
For each rock type (Danian and Maastrichtian), the irre-
ducible water saturation (S
wi
) and capillary entry pressure
(P
ce
) are assumed to depend upon the porosity (
) through
relatively simple relationships. Using these relationships, the
primary drainage capillary pressure is described by means of
the EQR model (Engstrøm 1995).
primary drainage capillary pressure is described by means of
the EQR model (Engstrøm 1995).
The simulation of the filling history uses 8 million years
before present as the starting point, and the entry of hydro-
carbons from an underlying source rock is assumed to occur
on the south-eastern flank (Fig. 3).
carbons from an underlying source rock is assumed to occur
on the south-eastern flank (Fig. 3).
The flow simulation of the filling dynamics of the Kraka
chalk reservoir has a complex geometry due to the high capil-
lary entry pressures in the low-permeability chalks. These
internal barriers re-direct hydrocarbons, such that oil flows in
the Maastrichtian layers for some time before it is able to pen-
etrate upwards into the overlying Danian chalk (Fig. 3A).
lary entry pressures in the low-permeability chalks. These
internal barriers re-direct hydrocarbons, such that oil flows in
the Maastrichtian layers for some time before it is able to pen-
etrate upwards into the overlying Danian chalk (Fig. 3A).
If oil supply is stopped after 1 million years, the oil con-
tinues to move towards the crest, but leaves immobile resi-
dual oil on the migration route. Hydrocarbon charging is
slow and equilibration of hydrocarbons with respect to pres-
sure gradients therefore occurs very slowly.
dual oil on the migration route. Hydrocarbon charging is
slow and equilibration of hydrocarbons with respect to pres-
sure gradients therefore occurs very slowly.
After two million years, the oil is seen to be nearly in
equilibrium even though the FWLs are still slightly inclined
and do not coincide for the two reservoir units (Fig. 3B).
After 4 million years, equilibrium is more obvious (Fig. 3C).
Between 2 million years before present and the present, a
and do not coincide for the two reservoir units (Fig. 3B).
After 4 million years, equilibrium is more obvious (Fig. 3C).
Between 2 million years before present and the present, a
Fig. 3. Modelled oil saturation in the Kraka field profile at different
times during the simulated filling history. Top structures of the Ekofisk
(Maastrichtian) and Tor (Danian) Formations are shown as thin green
and red lines, respectively. Calculated free water levels (FWLs) for
these two reservoir units are shown in thicker green and red lines,
respectively. Charging of the reservoir starts at 8 Ma B.P. by injecting oil
at a very low rate at the flank position shown with an arrow. A: The
situation after 250 000 years, where the injected oil is preferentially
moving in the Maastrichtian reservoir unit. B: Oil distribution after 2 Ma,
where the charging has been sustained over the first 1 Ma, accompa-
nied by equilibration during continued burial. C: Oil is near-equilibrium
at 4 Ma B.P. D: Tilted FWLs resulting from a lateral pressure gradient of
3.5 psi/km (24.1 kPa/km) applied for 1 Ma (from 2 to 1 Ma B.P.) within
the water phase. This causes a water-flow south-eastwards in the
aquifer, and accordingly a tilting of the FWL in that direction, which is
further accentuated until the modelled present-day situation shown
in E. Modified from Vejbæk et al. (in press).
times during the simulated filling history. Top structures of the Ekofisk
(Maastrichtian) and Tor (Danian) Formations are shown as thin green
and red lines, respectively. Calculated free water levels (FWLs) for
these two reservoir units are shown in thicker green and red lines,
respectively. Charging of the reservoir starts at 8 Ma B.P. by injecting oil
at a very low rate at the flank position shown with an arrow. A: The
situation after 250 000 years, where the injected oil is preferentially
moving in the Maastrichtian reservoir unit. B: Oil distribution after 2 Ma,
where the charging has been sustained over the first 1 Ma, accompa-
nied by equilibration during continued burial. C: Oil is near-equilibrium
at 4 Ma B.P. D: Tilted FWLs resulting from a lateral pressure gradient of
3.5 psi/km (24.1 kPa/km) applied for 1 Ma (from 2 to 1 Ma B.P.) within
the water phase. This causes a water-flow south-eastwards in the
aquifer, and accordingly a tilting of the FWL in that direction, which is
further accentuated until the modelled present-day situation shown
in E. Modified from Vejbæk et al. (in press).
12
pressure gradient is imposed in the water zone in order to
allow for the regional pressure distribution during that
period. As a result, the oil is forced south-eastwards towards
the flank (Fig. 3D), which is further accentuated through
time as the water gradient is sustained (Fig. 3E). Again a zone
with residual oil is left behind.
allow for the regional pressure distribution during that
period. As a result, the oil is forced south-eastwards towards
the flank (Fig. 3D), which is further accentuated through
time as the water gradient is sustained (Fig. 3E). Again a zone
with residual oil is left behind.
Conclusions
The modelling reported here demonstrates that oil accumu-
lations in chalk may require several million years to equili-
brate following perturbations resulting from primary migration
or reservoir tilting, if matrix permeability governs fluid flow.
Since naturally occurring disequilibrium oil accumulations
dominate the Danish chalk fields, it must be concluded that
matrix flow dominates fluid dynamics. The modelled filling
scenarios are intended to illustrate the general aspects of geo-
logical timescale oilwater dynamics in chalk reservoirs. The
scenarios are not considered to represent actual filling histo-
ries, as they are constrained by relatively simple model as-
sumptions, but they are geologically plausible. Due to the
long equilibration times, it can be dangerous to interpret
tilted contacts as reflecting only dynamic equilibrium, as they
may be fully dynamic and still actively flowing. This is
revealed locally by non-equilibrium between Danian and
Maastrichtian oil where they are seen to have different FWLs.
It is important to try to understand fluid dynamics during
exploration work, since this strongly affects trap definition
and volumes. The project shows that with simple and geolo-
gically based assumptions, a reasonable filling history can be
modelled quantitatively. A reasonable end-result can be pro-
duced that has many similarities with present-day hydrocar-
bon configurations. With the methods developed in the
project, even a fully dynamic system (with both oil and water
moving), as for example in the DanHalfdan field system,
may be explained.
lations in chalk may require several million years to equili-
brate following perturbations resulting from primary migration
or reservoir tilting, if matrix permeability governs fluid flow.
Since naturally occurring disequilibrium oil accumulations
dominate the Danish chalk fields, it must be concluded that
matrix flow dominates fluid dynamics. The modelled filling
scenarios are intended to illustrate the general aspects of geo-
logical timescale oilwater dynamics in chalk reservoirs. The
scenarios are not considered to represent actual filling histo-
ries, as they are constrained by relatively simple model as-
sumptions, but they are geologically plausible. Due to the
long equilibration times, it can be dangerous to interpret
tilted contacts as reflecting only dynamic equilibrium, as they
may be fully dynamic and still actively flowing. This is
revealed locally by non-equilibrium between Danian and
Maastrichtian oil where they are seen to have different FWLs.
It is important to try to understand fluid dynamics during
exploration work, since this strongly affects trap definition
and volumes. The project shows that with simple and geolo-
gically based assumptions, a reasonable filling history can be
modelled quantitatively. A reasonable end-result can be pro-
duced that has many similarities with present-day hydrocar-
bon configurations. With the methods developed in the
project, even a fully dynamic system (with both oil and water
moving), as for example in the DanHalfdan field system,
may be explained.
Acknowledgement
The work presented in this paper was partly funded by the Danish
Energy Authority (Grant no. 1313/01-0004).
Energy Authority (Grant no. 1313/01-0004).
References
Dennis, H., Baillie, J., Holt, T. & Wessel-Berg, D. 2000: Hydrodynamic
activity and tilted oilwater contacts in the North Sea. In: Ofstad, K.,
Kittilsen, E.-J. & Alexander-Marrack, P. (eds): Improving the explo-
ration process by learning from the past. Norwegian Petroleum
Society (NPF), Special Publications 9, 171185.
Kittilsen, E.-J. & Alexander-Marrack, P. (eds): Improving the explo-
ration process by learning from the past. Norwegian Petroleum
Society (NPF), Special Publications 9, 171185.
Dennis, H., Bergmo, P. & Holt, T. in press: Tilted oilwater contacts
modelling the effects of aquifer heterogeneity. In: Doré A.G. & Vining,
B. (eds): Petroleum geology: North-West Europe and global perspec-
tives. Proceedings of the 6th Petroleum Geology Conference. London:
Geological Society.
B. (eds): Petroleum geology: North-West Europe and global perspec-
tives. Proceedings of the 6th Petroleum Geology Conference. London:
Geological Society.
Engstrøm, F. 1995: A new method to normalize capillary pressure curves.
1995 International Symposium of the Society of Core Analysts, San
Francisco, CA, USA, September 1214, 1995, SCA-9535, 12 pp.
Francisco, CA, USA, September 1214, 1995, SCA-9535, 12 pp.
Japsen, P. 1998: Regional velocity-depth anomalies, North Sea Chalk; a
record of overpressure and Neogene uplift and erosion. American
Association of Petroleum Geologists Bulletin 82(11), 20312074.
Association of Petroleum Geologists Bulletin 82(11), 20312074.
Japsen, P., Bruun, A., Fabricius, I.L., Rasmussen, R., Vejbæk, O.V.,
Pedersen, J.M., Mavko, G. & Mogensen, C. in press: Influence of po-
rosity and pore fluid on acoustic properties of chalk: AVO-response
from oil, South-Arne Field, North Sea. Petroleum Geoscience.
rosity and pore fluid on acoustic properties of chalk: AVO-response
from oil, South-Arne Field, North Sea. Petroleum Geoscience.
Megson, J.B. 1992: The North Sea chalk play: examples from the Danish
Central Graben. In: Hardman, R.F.P. (ed.): Exploration Britain: geo-
logical insights for the next decade. Geological Society (London)
Special Publication 67, 247282.
logical insights for the next decade. Geological Society (London)
Special Publication 67, 247282.
Moss, B., Barson, D., Rakhit, K., Dennis, H. & Swarbrick, R. 2003: For-
mation pore pressures and formation waters. In: Evans, D. et al. (eds):
The Millenium Atlas: petroleum geology of the central and northern
North Sea, 317329. London: Geological Society.
The Millenium Atlas: petroleum geology of the central and northern
North Sea, 317329. London: Geological Society.
Vejbæk, O.V. 2002: A 1, 2 and 3D backstripping procedure with applica-
tion to the Kraka field. Danmarks og Grønlands Geologiske Under-
søgelse Rapport 2002/44, 35 pp.
søgelse Rapport 2002/44, 35 pp.
Vejbæk, O.V., Frykman, P., Bech, N. & Nielsen, C.M. in press: The history
of hydrocarbon filling of chalk fields. In: Doré, A.G. & Vining, B.
(eds): Petroleum geology: North-West Europe and global perspec-
tives. Proceedings of the 6th Petroleum Geology Conference. London:
Geological Society.
(eds): Petroleum geology: North-West Europe and global perspec-
tives. Proceedings of the 6th Petroleum Geology Conference. London:
Geological Society.
Authors' addresses
P.F., O.V.V. & N.B., Geological Survey of Denmark and Greenland, Øster Voldgade 10, DK-1350 Copenhagen K, Denmark. E-mail:
pfr@geus.dk
C.M.N., Danish Energy Authority, Amaliegade 44, DK-1256 Copenhagen K, Denmark.
P.F., O.V.V. & N.B., Geological Survey of Denmark and Greenland, Øster Voldgade 10, DK-1350 Copenhagen K, Denmark. E-mail:
pfr@geus.dk
C.M.N., Danish Energy Authority, Amaliegade 44, DK-1256 Copenhagen K, Denmark.
