Modelling needs for subsurface CO 2 storage UKCCSRC 2017 - Work Package B3 Mid-term Review September 4-5, Edinburgh, United Kingdom Sam Krevor Department of Earth Science & Engineering, Imperial College London Storage site Regional storage resource Pore Rock core
Low predictability in subsurface flow modelling This is widely known in the petroleum industry However standards and expectations for CO 2 storage may be different Sleipner, Norway In Salah, Algeria Frio, USA Ringrose et al. 2009. Williams et al. 2018. First Break , 27 p 85 –89. DOI: Haszeldine and Cavanagh (2014) Cowton et al., (2018) DOI: Kampman et al. 2014. 10.1016/j.ijggc.2017.11.0 10.1016/j.ijggc.2013.11.017 10.1016/j.epsl.2018.03.038 DOI: 10.1016/j.chemgeo.2013.11.012 10 BGS Univ Edinburgh Univ Cambridge
Regional scale and simplified modelling approaches are needed IPCC techno-economic pathways achieving 1.5 o and 2 o C are more sensitive to the availability of CO 2 storage than anything else Geographic and pressure limitations to CO 2 storage need to be incorporated in energy systems models Multi site systems management will become increasingly important in planning and development of regional resources Illinois Basin (USA) Basal Aquifer (Canada) Bunter Sandstone (UK) Huang et al., 2014 Noy et al., 2012 Birkholzer and Zhou, IJGGC 2009
Commercial simulators are optimised for water flooding In contrast, with CO 2 storage we are interested in • Major impacts of small scale heterogeneity • Gravity override • Trapping mechanisms • For regional and global estimates very large spatial scales are required ~10 2 -10 3 km
Important physical processes at a range of scales At each scale the key questions are (1) does it matter and (2) what can we do about it? Storage site Pore Rock core Regional resource Captain sandstone Bunter and Captain Bunter sandstone Sandstones Noy et al., 2012 10 3 10 0 10 6 m 10 -6
Lab to field Do small scale heterogeneities matter for large scale flow? Data and core availability allow us to study this Planned injection site for (discontinued) Peterhead • CCS project, aim to store ≈ 20Mt CO 2. Storage unit - Captain D, lower Cretaceous • Sandstone, 100m thick. Sample of 48 rock core plugs from depth 2950m – 3050m Typical North Sea Sandstone: • Poorly consolidated • Marshall et al. 2017. DOI: 10.1144/PGC8.18 Shell U.K., Peterhead High permeability • CCS project. Document # PCCS-05-PT-ZR-3323- Thin mudstone layers • 00002 Example rock plug sample used for characterisation
Exhaustive sample characterisation > 40 rock cores characterised to develop a “ground truth” for modelling the Captain Sandstone Porosity-depth Permeability-depth Capillary pressure characteristics
Benchmark against industry measurements Consistency with industry measurements of single phase flow properties – porosity and permeability – provides confidence in our measured dataset
What scale of heterogeneity matters? Start at the laboratory core scale – correlation lengths are larger than the core samples 20 cm
What scale of heterogeneity matters? At the well log scale, heterogeneities persist with length scales 1-4 m
What scale of heterogeneity matters? Generate statistical realisations of storage reservoirs constrained by the measurements
Synthetic field – 1m correlation
Synthetic field – 4m correlation
Synthetic field – 6m correlation
Unlayered heterogeneity spreads and slows the plume This is the typical conception of the impact of capillary pressure characteristics in large scale modeling Heterogeneous P e Homogenous P e 50m
Centimetre-scale layered heterogeneity significantly increases plume migration rate The effect is only present if heterogeneity in the multiphase flow properties – capillary pressure characteristics, are taken into account Homogenous P e Heterogeneous P e 50m
Small scale heterogeneity impacts large scale flow Upscaling – How can we represent these Characterisation – How can we measure impacts at the large scales required to properties on enough samples with model fields? sufficient resolution? 25 cm 50m
Characterisation Evaluation of digital rock modelling for characterising heterogeneities shows that these approaches can capture key properties without the need for experimental calibration Zahasky et al., 2019
Upscaling Accurate upscaling approaches have been identified that permit the incorporation of the impacts of small scale heterogeneities into field scale simulations Jackson et al., 2019
Simplified modelling Simplified models can also be used to incorporate Field scale simplified models allow us to place constraints on maximum migration pressure limited, dynamic, regional, storage distance, pressurisation, and allow for rapid capacity estimates into energy systems analysis screening Cowton et al., (2018) DOI: 10.1016/j.epsl.2018.03.038 De Simone et al., 2019
Summary CO 2 storage requires advances in modelling the upscaled impacts of small • scale heterogeneity, the effects of gravity, across large scales, for rapid screening Digital rock characterisation opens the door to rapid characterisation of small • scale heterogeneity in multiphase flow properties Upscaling techniques have been developed that allow for the incorporation of • the impacts of heterogeneity, gravity, but require incorporation into commercial numerical simulation packages Illinois Basin (US) (Birkholzer and Zhou, IJGGC 2009) Simplified modelling approaches have been developed to inform energy • systems models, for multi site resource management, and for reservoir screening Storage site Rock core Pore Regional resource
References Published De Simone, S., Jackson, S. J., & Krevor, S. (2019a). The Error in Using Superposition to Estimate Pressure During Multisite Subsurface CO 2 Storage. Geophysical Research Letters , 46 (12), 6525-6533. To be submitted De Simone, S., Krevor, S., (2019b) A methodology for the assessment and optimization of the CO 2 storage capacity in saline aquifers Jackson, S., Krevor, S. (2019) The emergent impacts of small-scale heterogeneities on field scale subsurface flow Jackson, S., Lin, Q., Krevor, S., (2019b) A multi scale study of representative elementary volume, hysteresis, and heterogeneity in multiphase flow Zahasky, C., Jackson, S.J., Lin, Q., Krevor, S. (2019) Pore network model predictions of Darcy-scale multiphase flow heterogeneity validated by observations Conference Presentations De Simone, S., Jackson, S. J., Zimmerman, R. W., & Krevor, S. C. (2018, December). On the applicability of simplified models to assess pressure evolution during CO2 injection into multiple site systems. In AGU Fall Meeting Abstracts . Jackson, S. J., & Krevor, S. C. (2018, December). The Emergent Impacts of Small Scale Capillary Heterogeneity on Field Scale CO2 Flow and Trapping. In AGU Fall Meeting Abstracts .
Recommend
More recommend
