Fifth CO2 Geological Storage Workshop

CO2 storage needs economic business cases through cost-effective exploration and production and needs license-to-operate through public support. Re-interpretation and reprocessing of vintage geophysical data is a means to achieve cost-effective exploration whereas de-risking and conformance control of storage operations is a means to obtain public support. Seismic exploration should identify risk elements for CO2 storage such as the risk of leakage, risk of pressure build-ups or drops, unexpected increase or decrease of storage capacity and spill points to name a few. These risks elements are often caused by hidden features such as a failing overburden seal, closed or open faults in either reservoir or seal and high- or low-permeability streaks in the reservoir. We have investigated a seismic reprocessing workflow for imaging and de-risking CO2 storage reservoirs and seals. The workflow includes statics, demultiple, velocity modeling, Prestack Time Migration, high resolution sparse spike deconvolution and Non Local Means filtering. Non Local Means filtering increases signal to noise ratio while preserving edges and the sparse spike deconvolution produces results with superior vertical and lateral resolution. This workflow manages at low cost to considerably de-risk the CO2 storage reservoirs and seals by identifying previously hidden faults, seal-reservoir contacts and thin reservoir streaks.

In many applications, the wettability of the rock surface is assumed to be constant in time and uniform in space. However, many fluids are capable to alter the wettability of rock surfaces permanently and dynamically in time. We simulate the dynamic system using a bundle-of-tubes (BoT) approach, where an empirical model for contact angle change is introduced at the pore scale. The resulting capillary pressure curves are then used to correlate the time-dependent term to the upscaled version of the wettability model. This study shows the importance of time-dependent wettability for determining capillary pressure over timescales of weeks and months. The impact of wettability has implications for experimental methodology as well as macroscale simulation of wettability-altering fluids.

In July 2018, the Ministry of Petroleum and Energy (MPE) offered an area in the northern Stord Basin for applications to exploit a subsea reservoir for injection and storage of CO2. This will be the first licence regulated by the Norwegian regulation for CO2 Storage (2014). By offering the area south of the Troll Field for an exploitation licence, it was decided to continue with the qualification of the Johansen-Cook aquifer for the Northern Light project.

In this abstract, we develop simulation models to study and show the potential for field-scale application of microbially induced calcite precipitation (MICP) as a leakage mitigate solution in CO2 sequestration. Based on laboratory experiments, field-scale cases, and numerical studies from the literature, two injection strategies for efficient MICP are developed: (I) injection of pre-stimulated microorganisms and urea into the subsurface, resulting in calcite precipitation around the body of the microbes; and (II) the classic approach of injecting microorganisms together with chemicals to stimulate growth of biofilm, and subsequent calcite precipitation from the biofilm. To enable field-scale simulations of (I) and (II) at low computational cost, we simplify the processes that have little contribution to the flow, while keeping input parameters and assumptions as realistic as possible. The injection strategies were simulated on field-scale, synthetic 2D radial models. The simulation results showed that both injection strategies produce significant porosity/permeability decrease at targeted locations away from the injection well. Moreover, it was seen that injection strategy (II) produced significantly more porosity/permeability decrease compared to (I).

We study the enhanced mass transfer of CO2 in water for a CO2 saturated layer on top of a water saturated porous medium, experimentally and theoretically. A relatively large experimental set-up with a length of 0.5 m and a diameter of 0.15 m is used in pressure decay experiments to minimize the error of pressure measurement due to temperature fluctuations and small leakages. The experimental results were compared to the theoretical result in terms of onset time of natural convection and rate of mass transfer of CO2 in the convection dominated process. In addition, a non-isothermal multicomponent flow model in porous media, is solved numerically to study the effect of the heat of dissolution of CO2 in water on the rate of mass transfer of CO2. The effect of the capillary transition zone on the rate of mass transfer of CO2 is also studied theoretically. The simulation results including the effect of the capillary transition zone show a better agreement with experimental results compared to the simulation result without considering a capillary transition zone. The simulation results also show that the effect of heat of dissolution on the rate of mass transfer is negligible

Sorption- and stress-induced coal permeability alteration may occur considering injection of carbon dioxide in coal seams for CCS. To take into account properly these phenomena, a microscale model was developed for the modelling of injection experiments carried out in laboratory. This work presents this model and first experimental results obtained from an injection test.

To evaluate subsurface reservoirs for CO2 sequestration, the grain scale properties and role of diagenesis is important for the injectivity and the subsequent mobilization. This study focuses on Johansen Formation of Jurassic age in the vicinity of Troll field within the northern North Sea. Johansen Formation is a saline aquifer and no hydrocarbon discovery has been reported in this reservoir so far. We analysed 24 wells using petrophysics and rock physics techniques to obtain net reservoir, net to gross ratio, effective porosity, volume of shale and level of cementation, and attempted to relate these parameters with the factors influencing them. The reservoir properties were found to be optimal approximately around depths shallower than 2000m (below sea floor, BSF). Even the shallowest sandstones exhibited cementation indicating calcite precipitation while the sediments deposited. Presence of shale however found to inhibit the quartz cementation possibly preserving the porosity. These findings will help understanding the complexity of the Johansen Sandstone as storage reservoir and the influence of heterogeneity on CO2 migration.

Equinor with Shell and Total, evaluate the feasibility for full-scale CO2 capture and storage project in Norwegian continental shelf. One of the challenges for CO2 storage sites is to assure containment and to assess possible leakage paths to the surface. Understanding the overburden’s geological setting is crucial for this assessment. As part of the overburden risk assessment, we investigate the potential for quantifying porosity from seismic inversion data. Knowledge of the porosity distributions, may enable us to employ porosity-permeability models in the future to assess leakage pathways into the overburden.

Different injection methods have been already proposed by different researchers to improve the solubility of CO2 in the formation brine. In this study an injection technique is presented to cool down (liquefy) the supercritical CO2 in the wellbore by the use of a downhole cooler equipment. CO2 with a higher temperature enters the cooling equipment and exits the equipment with a lower temperature at the down-stream in a same injection pressure. The colder (liquid) CO2 has a higher solubility in brine, higher density and viscosity which increases the security of CO2 storage. With this method the supercritical CO2 is cooled down to a liquid phase to increase the solubility at the wellbore and thus it eliminated the risk of phase change or pressure and rate fluctuation in liquid CO2 injection from the surface. To simulate this technique two cases have been considered by changing the relative permeability curves. The results show that using the combination of CO2STORE and THERMAL options shows a higher dissolution compared with only inserting the relative permeability curves corresponding the injection condition.

Injection of CO2 into a reservoir increases the pressure above initial values, resulting in overpressure of a hydrostatically charged formation. Without careful monitoring and management, excessive pressure can lead to a number of serious complications for a CO2 storage operations. Using numerical simulations with four distinct porosity/permeability distributions to represent reservoirs with random and structured heterogeneity. We initially consider the impact heterogeneity has on pressure propagation from a CO2 injection well; in particular the effect of channels on the lateral extent of the region of increased pressure. Subsequently, we investigate how heterogeneity influences the efficacy of water production as a pressure management tool and the optimisation of well positioning. For a channelized reservoir the most effective production well, which reduces the area of high pressure by up to 88%. Even in a randomised reservoir with no structured distribution of porosity and permeability, water production can still reduce the high pressure footprint by 60–88%. The location of the production well relative to the heterogeneity has been shown have a significant effect. The most effective production well location may not always be close to the target, but should be connected to the target by relatively high permeability pathways.

In 2016, a study identified the Smeaheia area located 30km off the western coast of Norway, as a suitable storage site for CO2. A concept selection study requested by the Gassnova public enterprise was subsequently performed by the Northern Lights subsurface team, a group comprised of personnel from Equinor and partners. The study revealed challenges with the various geological structures planned for CO2 storage, as well as the importance of understanding the pressure connectivity with the neighbouring hydrocarbon producing Troll field. Due to these challenges Smeaheia was not found mature enough for concept selection at this stage

Large scale CCS is crucial to reduce the cost associated with minimizing climate change. Energy system models should thus include CCS at regional or global scale with a proper evaluation of pressure limitations and injectivity, which are currently ignored. To this aim, the use of simplified analytical solutions is highly useful because they provide fast evaluation of pressure and plume evolution without the computational costs of the numerical models. Application of these solutions to assess storage capacity has been extended to cases of multiple well injection. In these cases, the pressure build-up is evaluated as the superposition of the analytical solutions for pressure associated with each individual well. In this study we investigate the validity of the superposition procedure, given the non-linearity of the multiphase flow. We quantify the error associated with the application of superposition to estimate reservoir pressurisation in different scenarios of.multi-site CO2 injection in a large regional aquifer. We find that the error associated with the adoption of this procedure increases with time and with the number of wells in proportion to the area invaded by CO2 in the reservoir.

In reservoir engineering, the predictive analyses of CO2 sequestration in subsurface formations commonly employ numerical models of subsurface formations. A significant number of work have utilised numerical modelling techniques to predict the impact of the reservoir’s boundary conditions and interlayer communication on CO2 storage capacity in aquifers. To the best of our knowledge, no study on the impact of boundary conditions on CO2 storage efficiency has focused on the combined effect of this factor in the reservoir and saturation functions in the caprock. To this end, this study examined the effect of integrating both processes on pressure evolution in the caprock during the numerical simulation of CO2 injection into a deep saline aquifer. Utilising the Sleipner benchmark model, we also showed how varying saturation functions in the caprock can affect the storage efficiency in the reservoir formation.

Re-using depleted fields (and platforms and wells) offers advantages over developing storage projects in saline formations. However, with reservoir pressures after production sometimes below 20 bar, there can be a large pressure difference between the reservoir and the transport pipeline at the surface, which will be typically at pressures in the range of 80 – 120 bar. This pressure difference must be carefully managed to ensure that the temperature of the CO2, the surface installations and the well, remain within materials specifications and within proper operating boundaries. Pressure drops of the CO2 result in potentially large decrease in temperature, due to its high Joule-Thomson coefficient; in addition, the temperatures and pressures that occur in a typical CO2 transport and storage system are such that two-phase flow is likely to occur. Pipeline pressure and temperature management can easily be done in a single source- single sink scenario as the pipeline pressure is a free parameter. However, if the pipeline must act as a backbone for multiple wells at different reservoir pressure, pressure and flow management must be balanced carefully. In this paper, the differences between a pipeline as transport and a pipeline as backbone will be discussed in detail.

A revised analysis of seismic data at Sleipner has revealed large-scale, roughly north-trending, channels at a range of levels in the Utsira Sand. The seismic data also reveal localised chimneys within the reservoir and overburden, some of which show evidence of having provided vertical conduits for earlier natural gas flow. Reservoir flow models were set up with flow properties constrained by the observed levels of CO2 accumulation in the reservoir and the arrival time of CO2 at the reservoir top just prior to the first repeat survey in 1999. The initial model with laterally homogeneous sand units separated by thin semi-permeable mudstones achieved a moderate match to the observed time-lapse seismics. Subsequent flow models, progressively incorporating higher permeability vertical chimneys through the mudstones and large-scale channelling within the reservoir sands, yielded a progressive and marked improvement in the history-match of key CO2 layers within the plume. The preferred plume simulation flow model was converted into a seismic property model using Gassmann fluid substitution with an empirical Brie mixing law. Synthetic seismograms generated from this show a striking resemblance to the observed time-lapse data, both in terms of plume layer reflectivity and also of time-shifts within and beneath the CO2 plume.

We present the results of reservoir simulations and feasibility study of surface seismic monitoring applied to the CO2 sequestration at the CaMI Field Research Station (FRS). We first test the influence of injection parameters, as reservoir temperature, maximum bottom-hole pressure and of the ratio vertical permeability over horizontal permeability on the amount of CO2 you can inject and on the gas plume shape. We demonstrate that if the reservoir temperature has a very small influence on the injectivity, the maximum bottom-hole pressure and the ratio of permeabilities play a key role on the gas injection. The next step is fluid substitution, necessitated to estimate the variation in elastic parameters induced by the gas injection. We test different methods to compute the bulk modulus of the fluid (Reuss, Voigt, HRV and Brie) and compare their results. We finally use a 3D finite difference modeling to simulate the seismic response in the elastic models generated for the baseline, for 1 year of injection and for 5 years of injection.

The latest CO2 foam technology reviews were conducted to understand recent research trends in CO2 enhanced oil recovery (EOR). In general, it is expected to improve CO2 sweep efficiency resulting in better oil recovery and prevention of early breakthrough. From CCUS point of views, the delay of gas breakthrough has a significant advantage in underground storage of industry-originated CO2. The reviews highlighted that various types of nano-additives have been investigated to develop further advanced foam technology. Key points to be focused on are how achieving more robust foam stability. Even a conventional CO2 foam generated with surfactant agents might be deteriorated in short period, those additives can extend foam half-life time. As additives, the recent researches have paid attention to nano-particles, polymer, viscoelastic surfactant, etc. The investigation measured half-life, viscosity, and differential pressure in core flood as key performance indicators. In addition, “high temperature (HT)” and “high salinity (HS)” are keywords in their researches. Namely, screening criteria of experimental conditions are aiming to more harsh conditions. However, the reviewed reports have not covered up to our target conditions in typical Middle East region. Thus, we have been concentrating to develop nano-additive enhancing CO2 foam technology in HTHS.

A core flooding experiment was carried out to simulate an Enhanced Gas Recovery (EGR) process to inject supercritical Carbon Dioxide (SCO2) into a core sample saturated with methane (CH4). This was done to investigate the flow behaviour of the injected SCO2 at the flow conditions when the injection orientation was switched from horizontal to vertical during the CH4 displacement. From the results, it was found that gravity has significant effects on the flow behaviour of SCO2 at lower flowrates; more pronounced is the seemingly lower permeability in the horizontal orientation compared with the vertical orientation. So the choice of the injection pattern or direction during EGR by SCO2 injection for the purpose of additional recovery of CH4 and subsequent sequestration of the injected CO2 should be made in conjunction with the determination of optimum injection rate for efficient injectivity.

In this work the physics of a fluid CO2 – crude oil mixture are explained and correlated to the evaluation of the best performance of a CO2 EOR project. The impact of different factors on the miscibility of the two fluids is described. Based on this knowledge some methods for the determination of the minimum miscibility pressure (MMP) are introduced and their pros and cons are discussed. Additionally, the concept of using miscibility enhancing additives to improve the oil recovery for successful CCUS projects is introduced. At the end a good understanding of the complex CO2 – oil mixture and its influencing parameters is developed. The reasons for good or poor miscibility are understood. An approach to make reservoirs applicable for CO2 EOR which were naturally not is shown by the application of the miscibility enhancing additives in order to improve the economics and to provide a proper justification for CCUS.

Impurities in CO2 streams influence the chemical reactivity in and mineral alterations of CO2 storage formations. Fluid-rock interactions have been investigated by means of reactive transport simulations using TOUGHREACT V3.0-OMG. A novel method has been established through which co-injection of SO2, NO2, O2 and H2 with temporally varying concentrations can be implemented in reactive transport model scenarios. The paper presents (i) model testing and validation against simulation results obtained by Xu et al. (2007) , and (ii) results acquired from 1D-radial multiphase reactive transport simulations investigating two generic Bunter Sandstone reservoir formations. Results gained applying the novel hybrid approach show that modelling-based inaccuracies have largely been eliminated and inconsistencies are minimized. For the investigated generic Bunter Sandstone reservoir formations, two major geochemical processes are apparent. While the acidifying impurities SO2 and NO2 trigger carbonate dissolution coupled to anhydrite precipitation, presence of O2 leads to dissolution of iron-rich chlorite and subsequent goethite precipitation. Absolute changes of porosity for the two generic Bunter Sandstone formations are below 1%. The total quantitative impact of SO2, NO2, O2 and H2 on mineral reactions is rather limited and their impacts on the petrophysical properties of the two investigated generic Bunter Sandstone formations are geotechnically negligible.

Leakage of reservoir fluids from injection site, e.g. through faults, is one of the key risks associated with long-term CO2 geological storage. Leakage monitoring technologies applied at different levels: in-situ, groundwater and surface, are necessary to ensure safe CO2 storage. Development and testing of the monitoring technologies is an objective of the ENOS project. In this paper, in-situ leakage detection from analysis of well bottom hole pressure is discussed. Modern CO2 injection wells are usually equipped with Permanent Downhole Gauges (PDGs), providing pressure measurements during the whole well life-span including injection and shut-in periods. A practical way to apply Pressure Transient Analysis (PTA) to such measurements for leakage detection is in the focus. A simulated well test of near-fault water injection into saline aquifer was employed to evaluate capabilities of PTA in detecting leakage through the fault. These mechanistic reservoir simulations were followed by similar simulations on an actual geological setting. A reservoir segment of the potential LBr-1 injection site containing a fault was used to demonstrate PTA-based leakage detection under actual geological conditions. Both simulation studies have confirmed that the PTA-based detection may be a useful component of the multi-level leakage monitoring technologies relying on readily available facilities (PDGs).

Geophysical monitoring techniques are commonly used to image the subsurface and potential changes. These monitoring techniques are important for CO2 storage projects to ensure a safe operation. A detailed image of the subsurface can be achieved from borehole seismic where mostly transmitted and reflected waves are investigated. However, these measurements are time consuming and costly as receivers and sources need to be moved within the well during the acquisition. We investigate the monitoring potential of tube waves, which propagate along the interface between the well and geological formation. An experiment is conducted where the signal from a rotating metal pipe in a borehole is recorded in a nearby observation well. The tube wave velocity can be measured with a high precision, around ± 1.2 m/s, during the experiment, which is an important measure to evaluate the potential of the method. Therefore, it might be possible to use noise sources like CO2 injection phases to monitor changes of the formation surrounding the well. This would reduce the time and cost needed for borehole seismic as only receivers at a constant position are required. Further field test are needed to investigate the feasibility at larger scales and for real injection cases.

CO2-WAG injection has been applied in offshore Brazilian carbonate reservoirs aiming to improve oil recovery and promote a safe destination to CO2 naturally being produced alongside with hydrocarbon gas. A gas re-utilisation strategy can potentially lead to multiple benefits: residual oil saturation reduction, maintenance of reservoir pressure, avoidance of gas flaring and development of the infrastructure and expertise necessary to make CO2 storage more accessible once oil production is complete, paving the path for a low carbon future, whereas mature basins can be a potential hub for Carbon Capture, Utilisation and Storage (CCUS). This study aims to develop a methodology to design CO2-WAG projects that not only achieve a high Net Present Value (NPV) but also maximizes the capacity and safety of geological CO2 storage.

Gassnova is working to establish what could become Europe’s first industrial CCS project. The project will demonstrate that carbon capture, transport and storage (abbreviated to CCS) is possible and safe to implement. A full-scale CCS project can provide lessons and experiences that new CCS projects can take advantage of. In this abstract we present some experiences from the maturation of storage sites on the NCS.

Carbon Capture and Storage (CCS) is a key climate mitigation technology required to meet the Paris Agreement goal of limiting global warming. Commercial-scale demonstration projects such as the Quest project in Alberta, Canada, or the Illinois Basin-Decatur project in Illinois, USA, have shown that the technology is feasible and safe. These projects demonstrate that existing technologies are sufficient for the successful implementation of CCS at the mega-tonne/year scale. However, scaling these technologies to meet the future need for giga-tonne/year storage remains a shared industry challenge. Responding to it demands addressing the low-probability, high-impact storage risks that cannot always be avoided within a large and diverse portfolio of CO2 storage projects. These include the risk of induced seismicity and fault reactivation, pressure management to improve storage security, exposure to legacy wells, and lowering the cost of large-scale containment monitoring. We propose four technology development pathways to address these giga-ton/year challenges, highlighting key focus areas.

Operatonal update on the storage facility operations at Quest CCS.

To improve the imaging and the subsequent quantification of the injected CO2 at the Ketzin pilot site, as well as to improve knowledge about CO2 storage at Ketzin, we combine different geophysical techniques (e.g., seismics and geoelectrics) using joint inversion. The resulting 3D models of geophysical parameters, and their changes over time, can then be jointly interpreted to obtain reservoir parameters (e.g., pressure and saturation), using rock physics inversion. To accomplish this, a new joint inversion method using structural constraints was developed combining seismic full waveform inversion (FWI) and electrical resistivity tomography (ERT). The joint inversion combines the strength of the different techniques (e.g., high spatial resolution of seismics and the sensitivity of geoelectrics in terms of CO2 saturation), and results in models that are consistent with each other, with all data sets, and any a priori information. This new method is tested using realistic synthetic-, and real data from the Ketzin pilot site in Germany, and the results are presented.

We present an integrated methodology for quantitative CO2 monitoring using Bayesian formulation. A first step consists in full-waveform inversion and CSEM inversion solved with gradient-based inverse methods. Uncertainty assessment is then carried out using a posteriori covariance matrix analysis to derive velocity and resistivity maps with uncertainty. Then, rock physics inversion is done with semi-global optimisation methodology and uncertainty is propagated with Bayesian formulation to quantify the reliability of the final CO2 saturation estimates.