This project aimed to assess the potential for brine production through dedicated wells in target Carbon Dioxide (CO2) storage formations to increase CO2 storage capacity and reduce overall cost of storage - as well as any other potential benefits for CO2 store operators associated with brine production. Brine production is proposed as a method to manage pressure in storage sites, as a corollary to water injection during hydrocarbon extraction. In the case of CO2 storage, the concept is that the production of water creates voidage to increase storage capacity and reduce the extent of pressure increase due to CO2 injection. This in turn reduces the risk of caprock failure, fault reactivation and induced seismicity. Additionally, brine production reduces the energy available to drive fluids through legacy well paths and other potential seep features. Spatial reduction in the extent of the pressure plume cuts down the area of potential drilling interference, the number of impacted legacy wells, and the area of investigation for monitoring where brine movement is a concern. This report presents findings from the entire project, and references other project reports where appropriate.

The Pembina Cardium CO2 Monitoring Pilot was used as a test site to determine the relative roles of trapping mechanisms. Two methods to assess this distribution are presented. A geochemical approach using empirical data from the site was used to determine the phase distribution of CO2 at a number of production wells that were sampled monthly during a two-year CO2 injection pilot. In addition, a simplified reservoir simulation was performed. Results indicate that significant amounts of CO2 are stored in the oil phase thus reducing the amount of CO2 available as a buoyant free phase and hence increasing storage security.

Notional partitioning and long term security of storage of CO2 in geological formations (after IPCC Special Report, 2005) Introduction Solubility trapping in formation fluids is one of the mechanisms whereby CO2 can be sequestered in geological formations. Unlike saline aquifers, potential storage in hydrocarbon reservoirs, where CO2 may be utilized to enhance oil recovery (EOR), offers the prospect of solubility trapping in both the aqueous and hydrocarbon phases. Here, CO2-EOR may provide a greater a quantity of securely stored CO2 than a purely non-EOR storage operation.

This report shows that accelerating deployment of CCS can enable CO2- EOR in the UKCS. Part of the CO2 that would otherwise need to go directly to dedicated storage in CCS projects can be used to drive CO2-EOR. That gives significant benefits to the wider UK economy - extending the producing life of the North Sea, reducing imports of oil, maintaining employment, developing new capability to drive exports, and additional direct and indirect taxation revenues. At a national level this synergy between CCS and CO2-EOR could provide the overall most cost effective way to accelerate this energy transition between 2018 and 2030, to meet Committee on Climate Change de- carbonisation pathways. This CO2-EOR route also achieves two desirable UK objectives. A business demand is created, which drives sequential construction of CO2 capture, which develops learning and reduces costs of CO2 supply, which enables cheaper low-carbon electricity. CCS by this route, with secure CO2 storage already proven, develops more rapidly to protect the onshore UK economy and industry from increasing carbon prices.

Many diverse challenges – political, economic, legal and technical – face the continued development and deployment of geological storage of anthropogenic CO 2 . Among the technical challenges will be the satisfactory proof of storage site security and efficacy. Evidence from many past geotechnical projects has shown the investigations and analyses that are required to demonstrate safe and satisfactory performance will be site specific. This will hold for the geomechanical assessment of saline aquifer storage site integrity where, compared to depleted hydrocarbon fields, there will be no previous pressure response history or rock property characterization data available. The work presented was carried out as part of a project investigating the improvement in levels of confidence in all aspects of saline aquifer site selection and characterization that could be expected with increasing data availability and in-depth analysis. Attention focused on the geomechanical modelling and the rock mechanics data used to populate models of two storage sites in geological settings analogous to those where CO 2 storage might be considered. Coupled geomechanical models were developed from reservoir simulation models initially incorporating generic rock mechanical properties and then laboratory-derived site-specific properties. The models were run in various configurations to investigate the effect of changing the rock mechanical properties on the geomechanical response of the storage systems. Modelling results showed that the pressure response at one site due to low injectivity caused significant potential for fault reactivation. Increasing the number of injection wells, thereby reducing the individual rates needed to deliver the target capacity, reduced the injection pressures and ameliorated, but did not eliminate, this adverse response.

Abstract When CO2 is injected into a saline aquifer for storage purposes, it rises through the brine to form a thin plume underneath a caprock (or under an intermediate shale layer). This is challenging to model, because a fine grid is necessary to resolve the plume and, since aquifers are extensive, many grid cells are required. On the other hand, simulations may be speeded up by using vertical equilibrium (VE) which assumes instantaneous separation of fluids due to gravity. The focus of this work is to compare VE models with conventional black oil (BOS) and compositional simulations (CS) to determine conditions under which the VE approximation is valid. A simple 2D homogeneous pilot model was tested first. The base case model had 81×1×81 cells, and the refined models had 243×1×243 and 729×1×729 cells. Subsequently, a realistic 3D model representing the Bunter formation in the southern North Sea was used. This type of formation is expected to be used for CO2 storage in the future. The results from the pilot model showed that the VE model could adequately represent the plume with a coarser grid than the BOS or CS models. For a homogeneous model, the VE approximation is therefore very useful. However, when applying VE to a heterogeneous 3D model, the results must be treated with caution. For the CS model, the areal distribution of the CO2 at the top of the aquifer showed that CO2 had migrated only through the high-permeability cells while the VE model showed evidence of CO2 migration through the high and moderate permeability cells. The VE model saved processor time and gave reasonable results at low grid resolution. Even though the results may be less accurate for heterogeneous aquifers, the use of VE is important for the extended reservoir simulations needed to generate multiple realizations to capture the uncertainties and their effects on the aquifer-description parameters.

Carbon capture and storage (CCS) brings new entrants to subsurface exploration and reservoir engineering who require very high levels of confidence in the technology, in the geological analysis and in understanding the risks before committing large sums of capital to high-cost drilling operations. Many of the subsurface techniques used for hydrocarbon exploration are capable of translation to CCS activities. Unfamiliarity may, however, lead new entrants to openly question their applicability in order to transform their current understanding to a level where large capital investment can be organisationally justified. For example, some may make the erroneous assumption that a good CO2 subsurface store should resemble the pressure vessel type of containment that is prevalent with surface installations. Basic concepts such as utilising the rock structure and mineralogy to control fluid flow and securing the CO2 by residual trapping (between the rock grains) or by dissolution, as a superior storage mechanism, are counter intuitive and challenging to communicate effectively. To achieve success and reliable operation in CO2 emission reduction for coal- and gas-burning electricity power generation, all elements of the CCS chain have to function. In 2008 the CO2 Aquifer Storage Site Evaluation and Monitoring project (CASSEM) was one of the first UK based projects to attempt integration and full-chain connectivity from, capture and transport to injection, storage and monitoring. Its research is aimed at development of workflows that describe a CCS entry path for a target audience of potential new entrants, i.e. power utilities, engineering sector and government. In contrast to other studies, the CASSEM project has applied the specification of the full CCS chain, using two exemplar sites (coal-fired power plants) with contrasting geological conditions in the subsurface, to tailor storage site selection and analysis. Centred on the Ferrybridge Power Station in Yorkshire (Figure 1.1), a 'simple' site underlain by a thick, uniform sandstone with diverse legacy information available was sought onshore in the English Midlands. The offshore extension of this (Bunter) sandstone has been highlighted as a large potential aquifer store for CO2 captured from power plants in eastern and South East England. A 'complex' site was sought offshore of eastern Scotland, centred on the Longannet Power Station on the Firth of Forth near Edinburgh (Figure 1.2). This site was intended to confront the difficulties of investigating subsea structures with sparse legacy and incomplete information from hydrocarbon investigations. The selected site is a faulted and folded geological structure and the issues of seismic reflection surveys, detection of faults and fractures, and quality of the target reservoir, are similar to those which challenge offshore hydrocarbon exploration beneath the North Sea.

One of the many challenges facing carbon capture and storage will be to provide convincing evidence of the geomechanical integrity of any proposed geological storage site. Contrary to storage in depleted hydrocarbon fields, storage in saline aquifer presents many more unknowns in this respect because there will probably be no known previous pressure response history or rock property characterisation. The work presented here was carried out as part of a project investigating the improvement in levels of confidence in all aspects of site selection and characterisation that could be expected with increasing data availability for saline aquifers. Attention here was focused on geomechanical modelling and the rock mechanics data used to populate these models. The models initially used generic geomechanical property data and the potential for shear failure of the intact rock and (fault) reactivation of fractured rock investigated. The models were then updated with laboratory measured rock mechanical properties for actual rock from the proposed storage system locality. The modelled results were changed marginally but did not identify any significant issues of criticality because of the relative geomechanical “benignness” of the storage site.

Modern geomechanical simulation techniques – based on the finite element method, and using an advanced poro-plastic material description – produce numerical outcomes that are in good agreement with experimental models of faulting processes, and that also agree with outcrop observations of natural fault damage zones. The correspondence between these independent investigation methods enables us to suggest that the geomechanical numerical simulation approach can serve as a useful proxy for reality, even though there is no way to prove that the simulation results are correct. In compensation, however, the geomechanical simulation results provide spatial and temporal information that cannot be obtained from physical models or outcrops. Specifically, the simulation outcomes contain a complete characterisation of the mechanical state at all points of a model as it evolves. Here, we show how the progressive deformed states of simulation models can be used as input to create flow simulations that enable us to examine the flow consequences of faulting. To transform the mechanical state to petrophysical properties, we develop algorithms based on the state of strain – since it is the texture of the rock that is altered by strain, and it is the same texture that controls the flow properties. Our algorithm has a primary dependence on the volumetric strain, since this determines whether permeability is increased or decreased; additional dependencies can be introduced to account for the directionality introduced by distortional strains. Single-phase flow simulations reveal how variations in the rock mechanics behaviours are ultimately expressed in terms of flow system differences. We illustrate the development of fault seals and reservoir flow barriers, and the impact of faulting on caprock seals.

Gary Couples, Jingsheng Ma, Helen Lewis, Peter Olden, Juan Quijano, Tomi Fasae, и Rebecca Maguire, из Heriot-Watt Institute of Petroleum Engineering описывают некоторые методы, в которых степень деформации пород может сказываться на изображении и интерпретации разрывных нарушений. Сейсмические исследования представляют один из главных методов характеристики пространственной изменчивости резервуаров. Карты могут отображать несколько регионов с различными сейсмическими характеристиками внутри целевого интервала; они могут способствовать выявлению литологических и диагенетических особенностей резервуара и созданию ограниченных геомоделей межскважинных участков. Могут быть закартированы конфигурации резервуаров и отрисованы более крупные разрывные нарушения в них в виде пересечения горизонталей и их смещений. Сейсмика, однако, не очень эффективна при определении структурных деталей в пределах разломной зоны (рис. 1). Должны быть предусмотрены другие методы, чтобы обеспечить оценки воздействия разломов на движение флюидов. Наблюдения на дневной поверхности некоторых хорошо известных разломов позволило исследовать местность, чтобы разработать типичную модель пространственных структур, которые обычно имеют место в разломных зонах, воздействующих на разрез кремнеобломочных пород. Мы можем назвать такую пространственную структуру зоной разломного нарушения. Идеальная зона разломного нарушения (FDZ) состоит из узкой центральной зоны, которая заключает в себе интенсивно расколотые породы (которые именуются как тектоническая брекчия), окруженной областями менее деформированных пород. Центральная зона разлома и зона деформаций более мощные, когда увеличивается смещение по разлому, наводя на мысль о прогрессирующем развитии разломного нарушения.

The integrity of the top seal of a reservoir may be compromised when that sealing layer is deformed, either as a consequence of compaction of the underlying materials, or by a tectonic event. Here we assess, using a geomechanical simulator, the stress state that can develop in a simple fold-fault system where the seal and its surrounding materials are folded, or flexed, as a consequence of incremental displacement on an underlying fault, into a gentle monoclinal shape. We have conducted an extensive set of numerical simulations to consider the effects on the resulting deformation of variations in fault displacement, of material-property contrasts between layers, and the specification of boundary conditions. The stress states within the seal and its surrounding layers are dominated by the bending process, even at very small fault displacements. All our simulations create along-seal variations in minimum principal stress (σ 3 ) that are a consequence of bending. These σ 3 variations can exceed 10 MPa where the seal and its enclosing shales have a strong strength contrast, but still reach 1-2 MPa when the seal and its surrounding rocks have identical properties. However, the details of the variations are strongly dependent on the specifics of the model, so quantification requires that a model be related to a particular set of properties and constraints. If our results are applicable to natural events, leak-off test (LOT) values (as a proxy for the minimum stress) that are measured in seals above reservoirs can be expected to indicate significant, deformation-induced variations over short lateral distances. By considering the timing of deformation with respect to initial hydrocarbon retention and/or overpressure, together with alteration of the seal's petrophysical properties during deformation, our results can be used to make general predictions about the sealing capacity.