11562. the impact of geomechanics on predicted seismic...

1st Sustainable Earth Sciences Conference & Exhibition – Technologies for Sustainable Use of the Deep Sub-surface

8-11 November 2011, Valencia, Spain

11562. The Impact of Geomechanics on Predicted Seismic Attributes for CO2 Injection and Storage

T. Lynch (University of Leeds), D. Angus* (University of Leeds), Q.J. Fisher (University of Leeds) & P. Lorinczi (University of Leeds) Main objectives In this study, we explore seismic attribute predictions due to the depletion stage of depleted hydrocarbon reservoirs for injection CO2 programmes. Current work is focussing on compare results using seismic predictions based solely on fluid-flow simulation to seismic predictions based on coupled fluid–flow and geomechanical simulation. In the former case, only changes in saturation and pressure influence the seismic velocities of the reservoir. In the latter case, the influence of geomechanical deformation is also included with changes in fluid saturation and pore pressure.

New aspects covered

coupled fluid-flow and geomechanical simulation, seismic prediction, rock physics

Summary

Seismic methods have been used as a tool to monitor the flow and storage of injected CO2 for various carbon capture and storage (CCS) pilot studies. Surface seismic and vertical seismic profile (VSP) time-lapse surveys have demonstrated the capability of temporally and spatially tracking the storage of CO2 within the subsurface geological formation. Microseismic monitoring has been used to monitor the injection of CO2 and the integrity of the cap rock. Although these techniques have demonstrated capability to track the flow and containment of CO2, quantitatively linking changes in seismic attributes (e.g., amplitude) to fluid-flow properties (e.g., volume and saturation) is often difficult and uncertain. In this study, we explore seismic attribute predictions due to the depletion stage of depleted hydrocarbon reservoirs for injection CO2 programmes. Current work is focussing on compare results using seismic predictions based solely on fluid-flow simulation to seismic predictions based on coupled fluid–flow and geomechanical simulation. In the former case, only changes in saturation and pressure influence the seismic velocities of the reservoir. In the latter case, the influence of geomechanical deformation is also included with changes in fluid saturation and pore pressure.

Topic(s)

Reservoir modeling (CO2 Storage) Monitoring (CO2 Storage) Exploration & storage assessment (CO2 Storage)



Introduction

Seismic methods have been used as a tool to monitor the flow and storage of injected CO2 for various carbon capture and storage (CCS) pilot studies. Surface seismic and vertical seismic profile (VSP) time-lapse surveys have demonstrated the capability of temporally and spatially tracking the storage of CO2 within the subsurface geological formation (e.g., Daley et al., 2008, at Frio; Chadwick et al., 2010, at Sleipner; White, 2010, at Weyburn). Microseismic monitoring has been used to monitor the injection of CO2 and the integrity of the cap rock (e.g., Urbancic et al., 2009, at Ostego; Verdon et al., 2010, at Weyburn). Although these techniques have demonstrated capability to track the flow and containment of CO2, quantitatively linking changes in seismic attributes (e.g., reflection amplitude) to fluid-flow properties (e.g., volume and saturation) is often difficult and uncertain. The coupling of fluid-flow simulators to geomechanical modelling software allows consideration of the impact of geomechanical deformation on the reservoir, which may have a significant impact on the reservoir pressures, and mechanical behaviour of the reservoir. Furthermore, coupled approaches enable the assessment of reservoir deformation in the overburden and hence quantify the security of containment. Verdon et al. (2011) have shown the benefits of integrating coupled flow-geomechanical simulation with seismic observations using a poroelastic model. In this study, we use a poroelastoplastic constitutive material model and a model geometry representing a simple faulted graben style sandstone reservoir to investigate the seismic and geomechanical attributes of a CO2 injection scenario, using both coupled and non-coupled fluid flow simulations. The aim of the study is to generate comparisons of predictions of P and S wave velocities and reflection amplitudes using both simulation techniques.

Method

Coupled fluid-flow and geomechanical simulation: Although formulations exist for fully coupled fluid flow and geomechanical simulation, their computational expense in terms of simulation time make iteratively or loosely coupled schemes highly attractive approaches (Dean et al., 2003). Furthermore, since there are various commercial fluid–flow and geomechanical simulation packages already available, there has been an interest in developing coupling techniques and workflows to link these various industry software packages for improved “in–house” reservoir modeling. We use an algorithm that links the TEMPESTTM (ROXAR) reservoir flow simulators with the geomechanical simulator ELFENTM (Rockfield Software Ltd.). Segura et al. (in press) provide an extensive study on the influence of reservoir geometry and material properties on stress path during production using coupled fluid-flow and geomechanical simulation. Rock physics modelling: To predict the seismic response based on the results of the reservoir simulator as well as the coupled fluid-flow/geomechanical simulator, rock physics models are required that map changes in fluid saturation, pore pressure, and effective stresses into seismic stiffness. The workflow used to build the dynamic elastic model is based on constructing an aggregate elasticity (Angus et al., 2007) starting from the micro-scale (e.g., intrinsic anisotropy) and working up to the macro-scale (field-scale fractures). Such a model should account for intrinsic rock properties and microstructural fabrics (e.g., Kendall et al., 2007), stress-dependent seismic velocities (e.g., Verdon et al., 2008), and fluid substitution effects either at low frequency (Gassmann, 1951) or including dispersive effects induced by squirt-flow (Chapman, 2003). The influence of coherent fracture sets is modelled using Hudson et al. (1996) and Schoenberg & Sayers (1995). In this study, we use Gassmann’s equation to predict the influence of fluid saturation on seismic velocity and for the coupled fluid-flow and geomechanical simulation we also map the influence of stress perturbations onto seismic velocity changes. To do this we adopted the micro-structural model of Verdon et al. (2008) because it represents a conceptually attractive approach to describe the stress dependence of seismic velocity and anisotropy. This model assumes that a fraction of the total



porosity of the rock can be considered compliant. Although negligible in volume, the compliance of these features, sometimes referred to as microcracks, dominates the nonlinear stiffness response. As stresses increase, these features are forced closed, increasing the seismic velocity and reducing the stress-sensitivity, matching the nonlinear response that is empirically observed.

Model

The model we used to study the geomechanical impact of CO2 injection on predicted seismic attributes has been used in previous studies. Angus et al., (2008) presented initial results from the coupled flow-geomechanical simulations to show the effect of fault transmissibility due to production on various seismic attributes. For the high transmissibility example, travel-time anomalies for both the overburden and reservoir are observed over the lateral extent of the reservoir and indicate that the two normal faults may act as a stress guide. As fault transmissibility is reduced, the travel-time anomalies become more localized. Shear–wave splitting shows similar patterns. Angus et al. (2010) predict production induced microseismicity and show, that for this reservoir geometry, fault movement as well as fluid extraction can influence the spatial, temporal and scalar moment of microseismicity. For non-sealing faults, failure occurs within and surrounding all reservoir compartments and significant distribution located near the surface of the overburden. Movement of faults leads to increase in shear-enhanced compaction events within the reservoir and shear events located within the side-burden adjacent to the fault. The moment magnitude distributions of shear events show low values near the surface, moderate values near the faults and high values along the reservoir boundary. Overall, the results from the study indicate that it may be possible to identify compartment boundaries based on the results of microseismic monitoring.

Figure 1: Geometry of the graben sandstone reservoir model shown (a) in the unstructured finite-element mess used in the geomechanical simulation, (b) plane and vertical sections, and (c) detailed view of the fault geometry (Angus et al., 2010). The model consists of a sandstone reservoir with two normal faults subdividing the reservoir into three compartments (see Figure 1). The surrounding shale rock and reservoir sandstone are initially seismically isotropic. The bounding model is a rectangular volume having lateral dimensions of 18.6 km x 9.3 km, and depth 3.72 km. The reservoir has lateral dimensions of 7 km x 3.5 km, thickness 76 m and is located at a depth of 3.048 km. Four coupled flow–geomechanical simulations are performed; two varying the fault transmissibility multiplier from high (0.98) to zero (0.00001), and



two varying the fault plane coefficient of friction from a high value (µ =0.750) to a low value (µ=0.375). For all simulations, CO2 is injected at a maximum rate controlled by the bottom hole pressure (60 MPa for this model).

Results

Initial results of this study are shown in Figure 2 for the depletion stage of the injection modelling, showing decrease in velocity in the overburden and increase in velocity in the side-burden and within the reservoir. The results also show the development of seismic anisotropy, which is indicative of non-hydrostatic stress development that can be problematic for top-seal integrity. It should be noted that the nonlinear rock physics model appears to be overly stress sensitive near the surface and this is because the input parameters where calibrated using rock core at reservoir depths (Angus et al., 2009). These results along with those of Angus et al. (2010) indicate that the reservoir stress path asymmetry is important and should be better understood for potential depleted hydrocarbon CO2 storage sites, as the asymmetry could influence the effective storage capacity (i.e. negatively impact storage reserves).

Figure 2: Initial results from production stage of depleted reservoir CO0 injection programme. (a) Initial P-wave velocity, (b) P-wave velocity after 10 years of production, (c) developed stress-induced P-wave anisotropy, and (ε), (γ) and (δ∗) are the Thomsen’s anisotropy parameters after 10 years of



production.

Conclusions

The integrated workflow has been used to assess the controls on reservoir stress path, use of microseismicity and 4D seismic for detecting reservoir compartmentalization as well as top seal integrity, and the need for coupled fluid-flow and geomechanical production simulation modelling. In terms of fully understanding geomechanical impacts of CO2 injection, there are still significant knowledge caps that need to be improved to reduce the uncertainty of full scale CCS initiatives. In this study, we explore seismic attribute predictions due to the depletion stage of depleted hydrocarbon reservoirs for injection CO2 programmes. Current work is focussing on compare results using seismic predictions based solely on fluid-flow simulation to seismic predictions based on coupled fluid–flow and geomechanical simulation. In the former case, only changes in saturation and pressure influence the seismic velocities of the reservoir. In the latter case, the influence of geomechanical deformation is also included with changes in fluid saturation and pore pressure.

Acknowledgments

We thank ROXAR for use of TEMPESTTM, Rockfield Software for use of ELFENTM, and ITF and the sponsors of the IPEGG project, BG, BP, Statoil and ENI.

References

Angus, D.A., J-M. Kendall, Q.J. Fisher, M. Dutko, A.J.L. Crook, S.A. Hall and J.P. Verdon (2007) Linking coupled fluid-flow and geomechanical models with rock physics derived elastic models for reservoir seismic forward modelling applications, EAGE 69th Conference & Exhibition, London.

Angus, DAC; Kendall, J; Fisher, QJ; Segura, JM; Skachkov, S; Crook, AJL; Dutko, M (2010) Modelling microseismicity of a producing reservoir from coupled fluid-flow and geomechanical simulation , Geophysical Prospecting, 58, pp.901-914.

Angus, D.A., J.P. Verdon, Q.J. Fisher and J-M. Kendall (2009) Exploring trends in microcrack properties of sedimentary rocks: An audit of dry-core velocity-stress measurements: Geophysics, 74, E193-E203.

Chadwick, A., G. Williams, N. Delepine, V. Clochard, K. Labat, S. Sturton, M-L. Buddensiek, M. Dillen, M. Nickel, A. L. Lima, R. Arts, F. Neele and G. Rossi (2010) Quantitative analysis of time-lapse seismic monitoring data at the Sleipner CO2 storage operation. The Leading Edge (2010) vol. 29 (2) pp. 170.

Chapman, M. (2003) Frequency-dependent anisotropy due to meso-scale fractures in the presence of equant porosity, Geophysical Prospecting, 51(5), 369-379.

Daley, T.M., L.R. Myer, J.E. Peterson, E.L. Majer and G.M. Hoversten (2008) Time-lapse crosswell seismic and VSP monitoring of injected CO2 in a brine aquifer. Environmental Geology (2008) vol. 54 (8) pp. 1657-1665.

Dean, R., Gai, X., Stone, C. and Minkoff, S. (2003) A comparison of techniques for coupling porous flow and geomechanics. In: Proceedings of the 17th Reservoir Simulation SymposiumSPE 79709.

Gassmann, F. (1951) Uber die elastizit at poroser medien: Vierteljahrsschrift det Naturforschenden Gesellschaft in Zurich, 96, 1-23.

Hudson, J.A., E. Liu and S. Crampin (1996) The mechanical properties of materials with interconnected cracks and pores, Geophysical Journal International, 124(1),105-112.

Kendall, J-M., Q. J. Fisher, S. Covey Crump, J. Maddock, A. Carter, S.A. Hall, J. Wookey, S.L.A. Valcke, M. Casey, G. Lloyd and W. Ben Ismail (2007) Seismic anisotropy as an indicator of reservoir quality in siliciclastic rocks, Structurally complex reservoirs, eds S.J. Jolley, D. Barr, J.J. Walsh and R.J. Knipe, Geological Society Special Publication, London, 123-136.

Schoenberg, M. and C.M. Sayers (1995) Seismic anisotropy of fractured rock, Geophysics, 60(1), 204-211. Segura, J.M., Q.J. Fisher, A.J.L. Crook, M. Dutko, J. Yu, S. Skachkov, D.A. Angus, J. Verdon and J-M. Kendall, (in press)

Reservoir stress path characterisation and its implications for fluid-flow production simulations, Petroleum Geosciences. Urbancic, T.I., J. Daugherty, S. Bowman and M. Prince (2009) Microseismic Monitoring of a Carbon Sequestration Field

Test, EAGE Passive Seismic. Verdon, J.P., J-M. Kendall, D.J. White, D.A. Angus, Q.J. Fisher and T.I. Urbancic (2010) Passive seismic monitoring of

carbon dioxide storage at Weyburn. The Leading Edge, vol. 29 (2) pp. 1-7 Verdon, J.P., Kendall, J-M., White, D.J. & Angus, D.A. (2011) Linking microseismic event observations with

geomechanical models to minimise the risks of storing CO2 in geological formations, Earth and Planetary Science Letters, 305(1-2), 143-152.

White, D.J. (2010) Geophysical monitoring of the Weyburn CO2 flood: Results during 10 years of injection. Energy Procedia, pp. 1-8

11562. the impact of geomechanics on predicted seismic...

Documents