![The detectability of free-phase migrating CO2:
A rock physics and seismic modelling feasibility study
Rami Eid 1, Anton Ziolkowski 1, Mark Naylor 1, Gillian Pickup 2
1 University of Edinburgh, Edinburgh, EH9 3JW, UK
2 Heriot-Watt University, Edinburgh, EH14 4AS, UK
Introduction Methodology
Results
Discussion
To investigate the range of surface seismic responses due to CO2 migration, we estimate the
change in P-wave velocity (Vp) over a theoretical, clean, homogeneous sandstone reservoir (Fig 2)
through the application of a three-stage model-driven workflow (Fig 3).
Patchy -vs- Uniform
saturation
Vp as a function of CO2 (Fig 4)
- Uniform saturation predicts a
rapid change in Vp as CO2
saturations increase to 20%, with
minimal to no changes
thereafter.
- Patchy and Modified-patchy
saturation predicts a quasi-linear
relationship, showing gradual
change in Vp with increasing
saturation.
Great implications for
- Interpretation of CO2 trapping
mechanisms
- Detection of free-phase
migrating plume front
- Depleted hydrocarbon fields
Results from the petrophysical modelling highlight two main points:
1) The importance of determining and understanding the fluid distribution model, as well as
the relationship between fluid saturation and seismic characteristics,
2) The importance of relative-permeability and capillary pressure curves and their influence
on the detectability of migrating free-phase CO2.
The velocity is directly related to the volume of CO2 occupying the pore-space of a migrating
plume front which in turn is controlled by the relative-permeability and capillary-pressure
curves input into the simulation. This highlights the importance of understanding the role and
impact of relative-permeability and capillary-pressure curves used to predict the seismic
response of the CO2 as this is the main contributor controlling the detectability of free-phase
migrating CO2 within the reservoir.
- Generate synthetic geophysical time-lapse responses by
simulating the acquisition of geophysical seismic surveys,
- Detailed analysis of the role of relative-permeability and
capillary-pressure curves, with the idea of generating a
detectability threshold for storage reservoir end-
members, representing best- and worst-case scenarios
for the detection of a migrating plume front.
Apply methodology to two complex models:
- Bunter model, a heterogeneous saline reservoir (Fig 8),
- ROAD model, a heterogeneous depleted gas field (Fig 9).
Future work
Fig 5: Difference in P-wave velocity with the baseline for the
simulation (a), uniform saturation (b), patchy saturation (c), and
the modified-patchy saturation (d). Migrating plume front has
been highlighted and enlarged.
Fig 3: Detectability workflow used to assess the detectability of free-
phase migrating CO2. The workflow consists of three stages: 1) Fluid-
flow, 2) Petrophysical and 3) Geophysical modelling
Rami Eid
PhD Student
Grant Institute of Earth Sciences
School of GeoSciences
University of Edinburgh
uk.linkedin.com/in/eidrami
Rami.Eid@ed.ac.uk
0
2
4
6
8
10
12
14
16
18
20
0%
2%
4%
6%
8%
10%
12%
14%
Frequency
CO2 Saturation (%)
Fig 7: Two-phase rel-perm curves (modified from
Bennion & Bachu 2006). Solid lines represent drainage,
dashed lines represent imbibition.
Fig 6: Saturation histogram showing the frequency
of saturations within the secondary reservoir
First instance of a breach (Fig 5)
- Detectability directly related to
simulated CO2 pore saturation
- Saturations range from 0-7%
(Fig 6)
- Challenging to differentiate between
background velocity
- Uniform: change of ∼ -200 m/s
- Mod-patchy: change of ∼ -50m/s
- Disturbing nature of injection means
saturation is ‘expected’ to follow
patchy distribution. Implications on
detectability?
Control on saturation of plume front
- Directly related to relative-
permeability (rel-perm) and capillary-
pressure curves applied (Fig 7)
- Site-specific issue
- Significant impact on detectability
Fig 4: P-wave velocity (Vp) as a function of CO2 saturation
calculated using Gassmann-Reuss uniform saturation (Vpuni),
Gassmann-Hill patchy saturation (Vppatchy) and the modified
Gassmann-Hill patchy saturation model (Vpmod).
To capture the range of possible responses which
could be encountered, three fluid saturation
distribution models were assessed:
1) Gassmann-Reuss, uniform saturation,
2) Gassmann-Hill. patchy saturation,
3) Modified Gassmann-Hill. Patchy saturation .
Two key monitoring stages, critical when monitoring
the containment of CO2, were assessed:
1) first instance of a breach,
2) initial contact with the secondary seal.
The monitorability of CO2 storage sites requires the demonstration of containment within the
intended formation, and importantly, the identification and quantification of any movement of
CO2 from the primary storage reservoir.
It is favorable that CO2 injection occurs within a storage complex with two reservoir-seal pairs,
representative of a primary and secondary storage site (Fig 1). Should a loss of containment
occur, movement of CO2 from the primary to secondary reservoir occurs, termed migration.
The ability for seismic techniques to monitor
structurally trapped CO2 has been successfully
demonstrated due to the changes in the acoustic
properties of the reservoir produced by the
displacement of brine by less dense and more
compressible CO2. However, the ability for seismic
methods to detect free-phase CO2 migration is still
moderately understood. The ability to detect an
initial loss of containment could provide operators
with an early warning system allowing for
remediation activities to be undertaken such that
probability of a leak outside of the storage complex
is negligible.
The process of CO2 injection disrupts the
equilibrium of the reservoir, resulting in multi-
phase fluid distributions of different
compressibilities within a pore space. As the
seismic response depends on both the fluid type
and distribution, end-member fluid distribution
models are required to predict the possible range
of velocities impacting on the seismic responses.Fig 1: A schematic diagram of the subsurface
highlighting the storage complex, storage site and
respectable reservoir-seal pairs.
d) Vp difference- Modified-
patchy saturation
b) Vp difference- Uniform Saturationa) CO2 Saturation
c) Vp difference- Patchy
saturation
CO2 Saturation Vp difference [m/s]
Vp difference [m/s]Vp difference [m/s]
Primary seal
Secondary reservoir
Secondary seal
Overburden
Primary reservoir
Storage complex
Reservoir-seal pair
Storage site
Free-phase migration
2700
2750
2800
2850
2900
2950
3000
3050
3100
3150
3200
0 0.2 0.4 0.6 0.8 1
Vp(m/s)
CO2 Saturation (/)
Ksat.patch
Ksat.mod
Ksat.uni
Vppatch
y
Vpmod
Vpuni
0
0.2
0.4
0.6
0.8
1
0 0.2 0.4 0.6 0.8 1
Kr
CO2 Saturation (/)
Kr_CO2
Kr_W
Kr_CO2_Imbib
Kr_W_Imbib
Migrating
plume front
Maximum
saturation
Uni: 2776
Mod: 3051
Uni: 2880
Mod: 3117
Uni: 2728
Mod: 2728
Residual
saturation
Fig 2: Theoretical model used to represent a clean
homogeneous sandstone reservoir separated by an
intraformational seal with a zone of weakness.
Impermeable seal
1000 m
400 m
400 m
300 m
50 m
1000 m
2150 m
Intraformational seal Reservoir
Injection Well
500 m
Reservoir Depth
500 m
Fig 8: Bunter sandstone model, with each individual
dome labelled (modified from Williams et al., 2013).
Fig 9: 3D view of the top reservoir surface of the P-
18 fields. Faults are shown in grey (Arts et al., 2012).
ARTS, R. J., VANDEWEIJER, V. P., HOFSTEE, C., PLUYMAEKERS, M. P. D., LOEVE, D., KOPP, A. & PLUG, W. J. 2012. The feasibility of CO2 storage in the depleted P18-4 gas field offshore the Netherlands (the ROAD project). International Journal of Greenhouse Gas Control, 11.
BENNION, D. & BACHU, S. Year. Dependence on temperature, pressure, and salinity of the IFT and relative permeability displacement characteristics of CO2 injected in deep saline aquifers. In: SPE Annual Technical Conference and Exhibition, 2006.
WILLIAMS, J. D. O., BENTHAM, M., JIN, M., PICKUP, G., MACKAY, E., GAMMER, D. & GREEN, A. 2013. The effect of geological structure and heterogeneity on CO2 storage in simple 4-way dip structures; a modeling study from the UK Southern North Sea. Energy Procedia.](https://image.slidesharecdn.com/5efc11ef-bf64-4d11-80b9-fe3acdb55756-150625142730-lva1-app6891/85/The-detectability-of-free-phase-migrating-CO2-1-320.jpg)
The detectability of free-phase migrating CO2
- 1. The detectability of free-phase migrating CO2: A rock physics and seismic modelling feasibility study Rami Eid 1, Anton Ziolkowski 1, Mark Naylor 1, Gillian Pickup 2 1 University of Edinburgh, Edinburgh, EH9 3JW, UK 2 Heriot-Watt University, Edinburgh, EH14 4AS, UK Introduction Methodology Results Discussion To investigate the range of surface seismic responses due to CO2 migration, we estimate the change in P-wave velocity (Vp) over a theoretical, clean, homogeneous sandstone reservoir (Fig 2) through the application of a three-stage model-driven workflow (Fig 3). Patchy -vs- Uniform saturation Vp as a function of CO2 (Fig 4) - Uniform saturation predicts a rapid change in Vp as CO2 saturations increase to 20%, with minimal to no changes thereafter. - Patchy and Modified-patchy saturation predicts a quasi-linear relationship, showing gradual change in Vp with increasing saturation. Great implications for - Interpretation of CO2 trapping mechanisms - Detection of free-phase migrating plume front - Depleted hydrocarbon fields Results from the petrophysical modelling highlight two main points: 1) The importance of determining and understanding the fluid distribution model, as well as the relationship between fluid saturation and seismic characteristics, 2) The importance of relative-permeability and capillary pressure curves and their influence on the detectability of migrating free-phase CO2. The velocity is directly related to the volume of CO2 occupying the pore-space of a migrating plume front which in turn is controlled by the relative-permeability and capillary-pressure curves input into the simulation. This highlights the importance of understanding the role and impact of relative-permeability and capillary-pressure curves used to predict the seismic response of the CO2 as this is the main contributor controlling the detectability of free-phase migrating CO2 within the reservoir. - Generate synthetic geophysical time-lapse responses by simulating the acquisition of geophysical seismic surveys, - Detailed analysis of the role of relative-permeability and capillary-pressure curves, with the idea of generating a detectability threshold for storage reservoir end- members, representing best- and worst-case scenarios for the detection of a migrating plume front. Apply methodology to two complex models: - Bunter model, a heterogeneous saline reservoir (Fig 8), - ROAD model, a heterogeneous depleted gas field (Fig 9). Future work Fig 5: Difference in P-wave velocity with the baseline for the simulation (a), uniform saturation (b), patchy saturation (c), and the modified-patchy saturation (d). Migrating plume front has been highlighted and enlarged. Fig 3: Detectability workflow used to assess the detectability of free- phase migrating CO2. The workflow consists of three stages: 1) Fluid- flow, 2) Petrophysical and 3) Geophysical modelling Rami Eid PhD Student Grant Institute of Earth Sciences School of GeoSciences University of Edinburgh uk.linkedin.com/in/eidrami Rami.Eid@ed.ac.uk 0 2 4 6 8 10 12 14 16 18 20 0% 2% 4% 6% 8% 10% 12% 14% Frequency CO2 Saturation (%) Fig 7: Two-phase rel-perm curves (modified from Bennion & Bachu 2006). Solid lines represent drainage, dashed lines represent imbibition. Fig 6: Saturation histogram showing the frequency of saturations within the secondary reservoir First instance of a breach (Fig 5) - Detectability directly related to simulated CO2 pore saturation - Saturations range from 0-7% (Fig 6) - Challenging to differentiate between background velocity - Uniform: change of ∼ -200 m/s - Mod-patchy: change of ∼ -50m/s - Disturbing nature of injection means saturation is ‘expected’ to follow patchy distribution. Implications on detectability? Control on saturation of plume front - Directly related to relative- permeability (rel-perm) and capillary- pressure curves applied (Fig 7) - Site-specific issue - Significant impact on detectability Fig 4: P-wave velocity (Vp) as a function of CO2 saturation calculated using Gassmann-Reuss uniform saturation (Vpuni), Gassmann-Hill patchy saturation (Vppatchy) and the modified Gassmann-Hill patchy saturation model (Vpmod). To capture the range of possible responses which could be encountered, three fluid saturation distribution models were assessed: 1) Gassmann-Reuss, uniform saturation, 2) Gassmann-Hill. patchy saturation, 3) Modified Gassmann-Hill. Patchy saturation . Two key monitoring stages, critical when monitoring the containment of CO2, were assessed: 1) first instance of a breach, 2) initial contact with the secondary seal. The monitorability of CO2 storage sites requires the demonstration of containment within the intended formation, and importantly, the identification and quantification of any movement of CO2 from the primary storage reservoir. It is favorable that CO2 injection occurs within a storage complex with two reservoir-seal pairs, representative of a primary and secondary storage site (Fig 1). Should a loss of containment occur, movement of CO2 from the primary to secondary reservoir occurs, termed migration. The ability for seismic techniques to monitor structurally trapped CO2 has been successfully demonstrated due to the changes in the acoustic properties of the reservoir produced by the displacement of brine by less dense and more compressible CO2. However, the ability for seismic methods to detect free-phase CO2 migration is still moderately understood. The ability to detect an initial loss of containment could provide operators with an early warning system allowing for remediation activities to be undertaken such that probability of a leak outside of the storage complex is negligible. The process of CO2 injection disrupts the equilibrium of the reservoir, resulting in multi- phase fluid distributions of different compressibilities within a pore space. As the seismic response depends on both the fluid type and distribution, end-member fluid distribution models are required to predict the possible range of velocities impacting on the seismic responses.Fig 1: A schematic diagram of the subsurface highlighting the storage complex, storage site and respectable reservoir-seal pairs. d) Vp difference- Modified- patchy saturation b) Vp difference- Uniform Saturationa) CO2 Saturation c) Vp difference- Patchy saturation CO2 Saturation Vp difference [m/s] Vp difference [m/s]Vp difference [m/s] Primary seal Secondary reservoir Secondary seal Overburden Primary reservoir Storage complex Reservoir-seal pair Storage site Free-phase migration 2700 2750 2800 2850 2900 2950 3000 3050 3100 3150 3200 0 0.2 0.4 0.6 0.8 1 Vp(m/s) CO2 Saturation (/) Ksat.patch Ksat.mod Ksat.uni Vppatch y Vpmod Vpuni 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 Kr CO2 Saturation (/) Kr_CO2 Kr_W Kr_CO2_Imbib Kr_W_Imbib Migrating plume front Maximum saturation Uni: 2776 Mod: 3051 Uni: 2880 Mod: 3117 Uni: 2728 Mod: 2728 Residual saturation Fig 2: Theoretical model used to represent a clean homogeneous sandstone reservoir separated by an intraformational seal with a zone of weakness. Impermeable seal 1000 m 400 m 400 m 300 m 50 m 1000 m 2150 m Intraformational seal Reservoir Injection Well 500 m Reservoir Depth 500 m Fig 8: Bunter sandstone model, with each individual dome labelled (modified from Williams et al., 2013). Fig 9: 3D view of the top reservoir surface of the P- 18 fields. Faults are shown in grey (Arts et al., 2012). ARTS, R. J., VANDEWEIJER, V. P., HOFSTEE, C., PLUYMAEKERS, M. P. D., LOEVE, D., KOPP, A. & PLUG, W. J. 2012. The feasibility of CO2 storage in the depleted P18-4 gas field offshore the Netherlands (the ROAD project). International Journal of Greenhouse Gas Control, 11. BENNION, D. & BACHU, S. Year. Dependence on temperature, pressure, and salinity of the IFT and relative permeability displacement characteristics of CO2 injected in deep saline aquifers. In: SPE Annual Technical Conference and Exhibition, 2006. WILLIAMS, J. D. O., BENTHAM, M., JIN, M., PICKUP, G., MACKAY, E., GAMMER, D. & GREEN, A. 2013. The effect of geological structure and heterogeneity on CO2 storage in simple 4-way dip structures; a modeling study from the UK Southern North Sea. Energy Procedia.