Quantification of the turning point saturation for cross bedded CO2 storage reservoirs

https://doi.org/10.1016/j.ijggc.2020.103185Get rights and content

Highlights

  • Cross bedded sedimentary structures are prominent features of CO2 storage reservoirs.

  • Turning point saturation is important to determine the amount of capillary trapping.

  • Range of turning point saturation for rock properties in cross bedding is determined.

Abstract

Cross bedding is a prominent sedimentary structure commonly found in fluvial and coastal to shallow-marine sedimentary sequences, which are often considered for the geological CO2 storage. A millimeter-thick silt- or mudstone lamina draped over a lens of sandstone is characteristic for cross bedding. This type of sedimentary structure gives rise to a greater capillary heterogeneity trapping capacity compared to other sedimentary structures such as planar bedding. The estimation of capillary trapping depends on turning point CO2 saturation. However, turning point saturations for a range of rock type combinations in cross bedded structures have not been quantified. This study uses field data to build 120 multiphase flow realizations to quantify the range of turning point saturation. The results show that maximum turning point saturation can vary between 40–95 % of end-point CO2 saturation depending on rock type properties. Statistical analysis is used to quantify the impact of different rock properties and their associated heterogeneity on turning point saturation. Results show that heterogeneity in entry pressure contributes about 69 % of the total variability in the maximum CO2 saturation under drainage. The results are used to derive multivariable linear regression models for the direct estimation of turning point CO2 saturations using intrinsic rock properties.

Introduction

Geological carbon storage is deemed to be an effective technology mitigating global warming (Benson and Orr, 2008; Friedmann, 2007; Metz et al., 2005). As part of this scheme, CO2 is injected into a geological reservoir where it is permanently stored via one or more of the four different trapping mechanisms: stratigraphic, structural, residual and mineral trapping (Friedmann, 2007; Metz et al., 2005). The storage capacity of the four trapping mechanisms has been well documented for different reservoirs (Bickle et al., 2007; Bradshaw et al., 2000; Metz et al., 2005; Pires et al., 2011; Steeneveldt et al., 2006). Most of these studies have focused on relatively homogeneous reservoirs in terms of lithology (Baklid et al., 1996; Ennis-King and Paterson, 2002; Korbøl and Kaddour, 1995; Pruess et al., 2003; Van der Meer, 1996; Xu et al., 2004a).

However, CO2 storage reservoirs are often lithologically heterogeneous (Doughty et al., 2001; Flett et al., 2005; Law and Bachu, 1996; Lindeberg, 1997). Reservoir heterogeneity can be characterized based on properties and scale of interest. In this study, we address heterogeneity in terms of flow and petrophysical properties like porosity, permeability and capillary pressure, at mm- to cm-meter scale. At such scales, lithological heterogeneity exits as intraformational baffles (Gibson-Poole et al., 2009; Yu et al., 2017), which are rock units with a higher capillary entry pressure due to their finer grain size compared to adjacent rock units. The baffles are usually interbedded within high-porosity and -permeability reservoir rocks to a varying extent (Flett et al., 2007). Due to their low porosity and permeability and high capillary entry pressure, the intraformational baffles control fluid flow pathways and act as flow barriers (Frykman et al., 2009; Green and Ennis-King, 2010a). This retards the migration of supercritical CO2 to shallow depths rising under buoyancy and leads to capillary heterogeneity trapping in addition to residual trapping by snap-off mechanism within the pore space (Gershenzon et al., 2017; Krevor et al., 2011; Li and Benson, 2015; Saadatpoor et al., 2010b; Trevisan et al., 2017). Lithologically heterogeneous reservoirs with higher volumetric proportions of intraformational baffles might trap a significant additional volume of CO2 compared to the homogeneous reservoirs by local capillary forces (Flett et al., 2007; Saadatpoor et al., 2010b).

In a heterogeneous siliciclastic reservoir, intraformational baffles exist as a part of different sedimentary structures such cross bedding, flaser bedding and lenticular bedding (Miall, 2016). Out of these, cross bedding is one of the most prominent sedimentary structures characteristic of reservoir rocks, for example in the coastal to shallow marine deposits of the Paaratte Formation, Otway Basin (Australia) (Mishra et al., 2020). The cross bedding usually forms by the downstream migration of sediments along the flow direction within a distributary channel depositional facies environment. The sediment migration creates alternate dipping layers of coarse grained beds known as foresets and fine grained sediments known as lamina. The lamina layers have a maximum thickness of 1 cm while the foresets are usually a few centimetres thick (Miall, 2016). In a cross bedded strata, lamina act as intraformational baffle while foreset beds act as reservoir rocks.

The cross bedded strata might significantly affect the migration of supercritical CO2 and enhance capillary heterogeneity trapping (Corbett et al., 1992; Honarpour and Saad, 1994; Huang et al., 1996; Ringrose et al., 1993; Saad et al., 1995). This is mainly because of the geometry of intraformational baffles comprising cross bedding. The foreset beds typically have a lens like shape and are completely enveloped by mm- to cm-thick lamina layers (Miall, 2016; Reineck and Singh, 1975) (Fig. 1a). The capillary heterogeneity trapping of supercritical CO2 within the foreset beds is expected to be enhanced because of this specific geometry of cross beddings compared to baffles with other geometries such as planar laminations. The enveloping shape of lamina implies that under low fluid flow velocities, the CO2 flow paths within the foreset are limited essentially to vertical direction as the lateral migration of CO2 will be restricted by the presence of dipping laminae. On the contrary, the planar bedded baffles are characterised by a limited lateral dip of laminae resulting in lateral fluid flow pathways allowing CO2 to migrate around the baffle (Green and Ennis-King, 2010a, b; Green and Ennis-King, 2013; Li and Benson, 2015; Saadatpoor et al., 2009, 2010a; Saadatpoor et al., 2010b, 2013). Capillary trapping in such planar bedded reservoir rocks is higher than in homogeneous reservoirs (Ni et al., 2019), but could be significantly lower than in a cross bedded reservoir rock.

Under a capillary limited flow regime, CO2 injected below a cross bedded strata rises under buoyancy during drainage and reaches the interface between the foreset rock and the overlying lamina. Due to higher capillary entry pressure characteristic of the lamina, the up-dip migration of CO2 is impeded and it starts accumulating underneath the lamina. This increases CO2 saturation below the foreset-lamina interface (Fig. 1b). Additionally, the presence of dipping lamina layers around the foreset bed restricts lateral CO2 migration. CO2 saturation increases over time within the foreset bed resulting in an increase of buoyancy pressure. Once the buoyancy pressure of CO2 exceeds the entry pressure characteristic of the lamina, CO2 migrates into the lamina (Bech and Frykman, 2018; Naylor et al., 2011). At this stage, CO2 saturation below the foreset-lamina interface reaches its maximum, or turning point, value (Fig. 1b) but is typically smaller than the end point CO2 saturation value at irreducible water content post drainage. After the turning point saturation is reached, CO2 saturation within the foreset layer decreases during imbibition.

Given the high volume proportion of cross beddings in sedimentary reservoir rocks and their potential for enhanced capillary heterogeneity trapping compared to other baffle geometries, it is important to explore the range of CO2 turning point saturation within the foreset beds and the governing factors. The turning point saturations vary between a maximum value below the lithological interface and a minimum value at the bottom of the foreset layer (Fig. 1b) (Bech and Frykman, 2018). A characterization of maximum turning point saturation in cross bedding is important as it provides a key parameter for the determination of the maximum expected amount of capillary trapping (Carlson, 1981; Jerauld, 1997; Kleppe et al., 1997; Land, 1968; Spiteri et al., 2008; Suzanne et al., 2003).

For heterogeneous reservoirs, several studies have focused on understanding fluid flow migration (Green and Ennis-King, 2010a, b; Li and Benson, 2015; Saadatpoor et al., 2009), the impact of hysteresis of relative permeability (Ataie-Ashtiani et al., 2002; Juanes et al., 2006; Honarpour et al., 1995) and capillary pressure curves (Bech and Frykman, 2018; Pini and Benson, 2017) on trapping and a quantification of capillary heterogeneity trapping (Green and Ennis-King, 2013; Krevor et al., 2011; Ni et al., 2019; Saadatpoor et al., 2013). Many of these studies are based on heterogeneous domains where intraformational baffles are planar and have limited lateral dip and extent (Green and Ennis-King, 2010a, b; Li and Benson, 2015; Saadatpoor et al., 2009). Hence, turning point saturation determined from these studies cannot be applied to reservoirs with cross bedding due to differences in the flow dynamics caused by the baffle geometry. Certain studies have considered cases with laminations perpendicular to flow (Bech and Frykman, 2018; Burnside and Naylor, 2014; Ni et al., 2019), thereby serving as approximations to cross bedding. However, a range of possible maximum turning point saturations expected for different rock types in cross bedded strata has not been determined (Burnside and Naylor, 2014). The rock types are expected to significantly impact the magnitude of maximum turning point saturation relative to CO2 saturation corresponding to irreducible water saturation. Additionally, previous studies have shown that residual trapping is dependent on a range of rock properties (Chatzis and Morrow, 1984; Geistlinger et al., 2014; Holtz, 2009, 2002; Iglauer et al., 2011; Krevor et al., 2015; Ruspini et al., 2017) and have quantified the relative dependence of trapping (Ni et al., 2019) and plume speed (Li and Benson, 2015) on various rock properties and their associated heterogeneity. However, the impact of common rock properties and their heterogeneity on turning point saturation has not been quantified in absolute terms for a cross bedded strata. Also, the range of rock properties are usually implemented in a limited number of model domains (Bech and Frykman, 2018; Li and Benson, 2015; Ni et al., 2019). A limited range of simulations may mean the derived turning point saturation is not well constrained and introduces a larger uncertainty.

This study addresses the following three objectives in order to better understand and predict turning point saturation in cross-bedded sediments. Firstly, the range of maximum CO2 saturation within the foreset layer during drainage is modelled for a time before CO2 invades the lamina. The range is determined for different rock types and flow conditions typically characteristic of cross beddings. Secondly, the dependence of maximum turning point CO2 saturation on different rock properties and their associated heterogeneity is quantified in absolute terms. Thirdly, regression models are derived determining the maximum turning point saturation directly from intrinsic rock properties.

Section snippets

Geostatic modeling

The dimensions of cross bedding foresets and laminae have been determined using core sample data from the Paaratte Formation, Otway Basin (Australia), which is a heterogeneous siliciclastic sequence located in the state of Victoria, Australia (Dance, 2018). The cross bedded strata constitute about 16 % of the sequence (Mishra et al., 2020). The samples comprise coarse grained, lens shaped foresets with a thickness 2–3 cm enveloped by fine grained laminae with a thickness of 0.5–1.0 cm (Fig. 2a)

Simulation results from simplified geometry models

CO2 injected at the base of the models rose under buoyancy through the foreset layer and accumulated underneath the lamina. The saturation built up until CO2 entered the baffle. Fig. 5 shows the evolution of CO2 saturation profiles for one representative realization from each of the four scenarios: realization numbers 1 (Fig. 5a), 31 (Fig. 5b), 61 (Fig. 5c) and 91 (Fig. 5d). The initial CO2 saturation distribution for the four realizations was similar. However, significant differences in the

The importance of rock types in cross bedding controlling the maximum CO2 saturation

The present study quantifies the variation of maximum CO2 saturation during drainage within the foreset beds of cross bedding sedimentary structures. The results show that the maximum CO2 saturation varies between 40–95 % of the end point saturation values at irreducible water content depending on the rock types of the foreset and the lamina (Fig. 8a). This is a large range and can have important implications on the amount of capillary trapping within reservoir rocks with cross bedding.

Ideally,

Conclusions

The present study quantifies the range of maximum turning point CO2 saturation during drainage for cross bedding sedimentary structures in CO2 storage reservoirs. The maximum CO2 saturation below the foreset-lamina interface at the time of CO2 entry into the lamina is derived for 120 scenarios and the results are used to develop insights into rock types and depositional environments favorable for heterogeneity enhanced capillary trapping. The results show that maximum turning point saturation

CRediT authorship contribution statement

Achyut Mishra: Conceptualization, Data curation, Formal analysis, Methodology, Writing - original draft. Ralf R. Haese: Funding acquisition, Supervision, Writing - review & editing.

Declaration of Competing Interest

The authors declare no conflicts of interest.

Acknowledgements

This study is a contribution to GeoCquest, a BHP-supported collaborative project by The University of Melbourne (Australia), the University of Cambridge (UK) and Stanford University (USA), and with CO2CRC Limited as a data provider. The GeoCquest project aims to develop a better understanding of small-scale heterogeneity and its influence on CO2 migration and trapping mechanisms. The authors thank BHP for providing project funding. The authors also acknowledge the support from Melbourne

Funding

This work was supported by BHP.

References (82)

  • J. Hermanson et al.

    Representation of geological heterogeneities and their effects on mineral trapping during CO2 storage using numerical modeling

    Procedia Earth Planet. Sci.

    (2013)
  • R. Korbøl et al.

    Sleipner vest CO2 disposal-injection of removed CO2 into the Utsira formation

    Energy Convers. Manage.

    (1995)
  • S. Krevor et al.

    Capillary trapping for geologic carbon dioxide storage–From pore scale physics to field scale implications

    Int. J. Greenh. Gas Control.

    (2015)
  • D.H.S. Law et al.

    Hydrogeological and numerical analysis of CO2 disposal in deep aquifers in the Alberta sedimentary basin

    Energy Convers. Manage.

    (1996)
  • B. Li et al.

    Influence of small-scale heterogeneity on upward CO2 plume migration in storage aquifers

    Adv. Water Resour.

    (2015)
  • B. Li et al.

    Influence of capillary-pressure models on CO2 solubility trapping

    Adv. Water Resour.

    (2013)
  • E. Lindeberg

    Escape of CO2 from aquifers

    Energy Convers. Manage.

    (1997)
  • A. Mishra et al.

    Composite rock types as part of a workflow for the integration of mm-to cm-scale lithological heterogeneity in static reservoir models

    Mar. Pet. Geol.

    (2020)
  • M. Naylor et al.

    Calculation of CO2 column heights in depleted gas fields from known pre-production gas column heights

    Mar. Pet. Geol.

    (2011)
  • H. Ni et al.

    Predicting CO2 residual trapping ability based on experimental petrophysical properties for different sandstone types

    Int. J. Greenh. Gas Control.

    (2019)
  • R. Pini et al.

    Capillary pressure heterogeneity and hysteresis for the supercritical CO2/water system in a sandstone

    Adv. Water Resour.

    (2017)
  • J.C.M. Pires et al.

    Recent developments on carbon capture and storage: an overview

    Chem. Eng. Res. Des.

    (2011)
  • S. Puranik et al.

    A novel machine learning approach for bug prediction

    Procedia Comput. Sci.

    (2016)
  • B. Ren et al.

    Maximizing local capillary trapping during CO2 injection

    Energy Procedia

    (2014)
  • P.S. Ringrose et al.

    Immiscible flow behaviour in laminated and cross-bedded sandstones

    J. Pet. Sci. Eng.

    (1993)
  • C. Ruprecht et al.

    Hysteretic trapping and relative permeability of CO2 in sandstone at reservoir conditions

    Int. J. Greenh. Gas Control.

    (2014)
  • L.C. Ruspini et al.

    Pore-scale modeling of capillary trapping in water-wet porous media: a new cooperative pore-body filling model

    Adv. Water Resour.

    (2017)
  • E. Saadatpoor et al.

    Effect of capillary heterogeneity on buoyant plumes: a new local trapping mechanism

    Energy Procedia

    (2009)
  • E. Saadatpoor et al.

    Estimation of local capillary trapping capacity from geologic models

    Energy Procedia

    (2013)
  • R. Steeneveldt et al.

    CO2 capture and storage: closing the knowing–doing gap

    Chem. Eng. Res. Des.

    (2006)
  • L. Trevisan et al.

    Impact of 3D capillary heterogeneity and bedform architecture at the sub-meter scale on CO2 saturation for buoyant flow in clastic aquifers

    Int. J. Greenh. Gas Control.

    (2017)
  • L.G.H. Van der Meer

    Computer modelling of underground CO2 storage

    Energy Convers. Manage.

    (1996)
  • T. Xu et al.

    Numerical simulation of CO2 disposal by mineral trapping in deep aquifers

    Appl. Geochem.

    (2004)
  • N. Zhou et al.

    Pore-scale visualization of gas trapping in porous media by X-ray CT scanning

    Flow Meas. Instrum.

    (2010)
  • A. Alcott et al.

    Using petrasim to create, execute, and post-process TOUGH2 models

    Proceedings of the TOUGH Symposium

    (2006)
  • M.A. Babyak

    What you see may not be what you get: a brief, nontechnical introduction to overfitting in regression-type models

    Psychosom. Med.

    (2004)
  • A. Baklid et al.

    Sleipner vest CO2 disposal, CO2 injection into a shallow underground aquifer, SPE annual technical conference and exhibition

    Old Spe J.

    (1996)
  • S.M. Benson et al.

    Carbon dioxide capture and storage

    MRS Bull.

    (2008)
  • J. Bradshaw et al.

    GEODISC project 1- regional analysis stage 2 basins- Oerth Basin, Western Australia

    Australian Petroleum CRC.

    (2000)
  • R. Brooks et al.

    HYDRAU uc properties of porous media

    Hydrology Papers

    (1964)
  • F.M. Carlson

    Simulation of relative permeability hysteresis to the nonwetting phase

    SPE Annual Technical Conference and Exhibition

    (1981)
  • Cited by (0)

    View full text