CO2 Migration Research Articles

A deep learning-based direct forecasting of CO2 plume migration

Accurate and timely forecasts of CO2 plume evolution in geological reservoirs are crucial for CO2 migration detection, leakage risk assessment, and operation decision support. Conventional forecasting usually adopts a two-step strategy, first calibrating reservoir model parameters against observations using iterative inverse modeling (or history matching) and then applying the calibrated model for predictions. This method impedes real-time forecasts due to the heavy computational demand in inverse modeling and may suffer from poor prediction accuracy because of the limited observation data. In this work, we propose a deep learning-based latent space mapping framework to forecast CO2 plume migration directly by avoiding the inverse modeling. We first use the convolutional autoencoder to map the high-dimensional complex plume extents onto low-dimensional latent space. Next, we use neural networks to learn the relationship between the observation variables and the prediction latent variables. And then for given observation data, we infer the prediction values directly. This one-step direct forecasting is computationally efficient which requires a few number of parallelizable reservoir simulations and it can provide accurate predictions with limited observations by learning the observation-prediction relationship in the reduced dimension. Therefore, our proposed method enables an in-time forecast of dynamic CO2 plume distributions. We demonstrate the effectiveness and accuracy of our method in predicting the CO2 plume migration using four metrics such as plume area, centroid movement distance, and plume spreading in the primary and secondary directions. And the spatio-temporal evolution patterns of plume migration under diverse geological complexities are also accurately quantified.

Deep-Learning-Based Flow Prediction for CO2 Storage in Shale–Sandstone Formations

Carbon capture and storage (CCS) is an essential technology for achieving carbon neutrality. Depositional environments with sandstone and interbedded shale layers are promising for CO2 storage because they can retain CO2 beneath continuous and discontinuous shale layers. However, conventional numerical simulation of shale–sandstone systems is computationally challenging due to the large contrast in properties between the shale and sandstone layers and significant impact of thin shale layers on CO2 migration. Extending recent advancements in Fourier neural operators (FNOs), we propose a new deep learning architecture, the RU-FNO, to predict CO2 migration in complex shale–sandstone reservoirs under various reservoir conditions, injection designs, and rock properties. The gas saturation plume and pressure buildup predictions of the RU-FNO model are 8000-times faster than traditional numerical models and exhibit remarkable accuracy. We utilize the model’s fast prediction to investigate the impact of shale layer characteristics on plume migration and pressure buildup. These case studies show that shale–sandstone reservoirs with moderate heterogeneity and spatial continuity can minimize the plume footprint and maximize storage efficiency.

Numerical Simulation of CO2 Migration and Geochemical Reactions in Shihezi Formation Caprock, China

CO2 geological storage, which is an effective way to reduce CO2 emissions, is of great significance to mitigate the current greenhouse effect. In long-term CO2 storage, although the chemical reaction rate of CO2–brine–rock is slow, it can significantly change the mineral composition of rock and the structure and properties of pores and joints, and then change the transport process and distribution state of CO2 in porous media. Therefore, a simplified 2D geological model is established based on the geological data of the Shihezi Formation in the Ordos Basin, China. The mechanism of the CO2–brine–rock reaction and its effect on mineral transformation and pore permeability are studied. In the early stage of CO2 geological sequestration, the rate of CO2 intrusion into the caprock is fast, the CO2–brine–rock reaction in the early stage is mainly a dissolution reaction, and the porosity and permeability of the caprock show an increasing trend. During the period from 100 to 1000 years of CO2 sequestration, the vertical distance of CO2 intrusion into the caprock does not change much. During this period, the type of CO2–brine–rock reaction is mainly a precipitation reaction, which reduces the porosity and permeability of the caprock and increases the sealing ability of the caprock to a certain extent. Our results can not only provide theoretical support for the site selection and risk assessment of CO2 geological sequestration, but also provide a theoretical basis and practical guidance for large-scale commercial storage in the future.

Multi‐Layered Systems for Permanent Geologic Storage of CO2 at the Gigatonne Scale

AbstractThe effectiveness of Carbon Capture and Storage (CCS) as an imperative decarbonization technology relies on the sealing capacity of a fine‐grained caprock to permanently store CO2 deep underground. Uncertainties in assessing the caprock sealing capacity increase with the spatial and temporal scales and may delay CCS deployment at the gigatonne scale. We have developed a computationally efficient transport model to capture the dynamics of basin‐wide upward CO2 migration in a multi‐layered setting over geological time scales. We find that massive capillary breakthrough and viscous flow of CO2, even through pervasively fractured caprocks, are unlikely to occur and compromise the storage security. Potential leakage from the injection reservoir is hampered by repetitive layering of overlying caprocks. This finding agrees with geologic intuition and should be understandable by the public, contributing to the development of climate policies around this technology with increased confidence that CO2 will be indefinitely contained in the subsurface.

Deep learning for characterizing CO2 migration in time-lapse seismic images

Time-lapse (or 4-D) seismic data play an important role in monitoring the spatial CO2 distribution during and after the injection period. However, traditional interpretation or prediction of CO2 distribution is time-consuming and might be sensitive to the quality of 4D seismic data. To solve these problems, we propose a deep-learning-based method to efficiently and accurately characterize CO2 plumes in time-lapse seismic data. We first introduce a workflow to build 3-D realistic impedance models containing CO2 plumes with various shapes, sizes, and locations. From the impedance models, we then simulate synthetic seismic datasets and automatically obtain the corresponding CO2 label volumes. We extract real noise from field seismic datasets and add the noise to the synthetic ones to make them more realistic. We further construct a diverse and realistic training dataset with the combination of synthetic data containing CO2 plumes and real data without CO2 plumes that are randomly cropped from field seismic data before CO2 injection. We finally utilize the training datasets without any human labeling to train a 3-D deep U-shape convolutional neural network for detecting CO2 plumes in the Sleipner time-lapse seismic images. Compared to traditional interpretation methods that take several days or even weeks, our method takes only 29 s using one graphics processing unit (GPU) to predict CO2 plumes in a 512*512*256 seismic volume. Besides, our CO2 prediction can achieve 95.8% accuracy (compared to the manual interpretation) and could distinguish reflections of CO2 plumes from the ones of pre-existing fluids, thin layers, and noise. To more accurately characterize the CO2 plumes migration, we use dynamic image warping to compute relative shifts that register the time-lapse seismic volumes before and after CO2 injection and then apply the same shifts to the predicted CO2 plumes. By doing this, we are able to reduce the inconsistencies that may be introduced by acquisition, processing, push-down effect (velocity decrease by injected CO2), and pull-up effect (wavelet distortion), which is helpful to more accurately characterize the CO2 plumes migration.

Study on vertical migration of CO2 in a karst cave system

Study on vertical migration of CO₂ in a karst cave system

Construction of urchin-like multiple core-shelled Co/CoS2@NC@MoS2 composites for effective microwave absorption

Designing nanocomposite with multi-level interfaces is an effective approach to improving the microwave absorbing capability of materials. In this study, the urchin-like multiple core-shelled Co/CoS2@NC@MoS2 composites (CCM) are designed and fabricated by high-temperature pyrolysis and solvothermal process, where the structure and phase composition are affected by the migration of Co. Abundant interfaces in the multiple core-shelled structures favor the polarization relaxation loss, defects of MoS2 and nitrogen doping carbon (NC) matrix induce the dipole orientation polarization loss. The CCM composite exhibits excellent microwave absorption with the strongest reflection loss (RL) of − 43.9 dB. Meanwhile, the broadest effective absorption bandwidth (EAB, RL<−10 dB) is 4.4 GHz with a thickness of 1.5 mm. Besides, with the thickness of 1.0–3.5 mm, all peaks of RLmin< −20 dB (99 % absorptivity). It is believed that this work can provide an effective strategy for the design of tunable multi-interface nanostructures to regulate microwave absorption.

Joint monitoring of CO2-ECBM based on multiple geophysical methods: A case study of Shizhuang Town, Shanxi Province, China

Among CCUS methods, carbon dioxide-enhanced coalbed methane (CO2-ECBM) is one of the most important applications. The CO2 migration range in the reservoir and impact on the fractures are the keys to safe storage. Multiple geophysical methods of time-lapse DAS-VSP, passive microseismic, and InSAR have been taken to monitor the CO2-ECBM project in Shizhuang Town, Shanxi Province, China. The time-lapse DAS-VSP technology captures the changes of seismic amplitude due to CO2 injection. Three phases of passive microseismic monitoring detected 217 microseismic events and 48 earthquakes. Based on the distribution of the microseismic events, the CO2 migration direction was inferred to be northeast of the injection well. A quantitative calculation method for the impact of natural earthquakes on the storage site was proposed. Studies on the relationship between microseismicity and earthquakes suggest that natural earthquakes do not significantly affect CO2 storage in the study area. InSAR combined unmanned aerial vehicle (UAV) photography measurement technology obtained the surface deformation results with a maximum displacement of 21 mm. The surface deformation also shows CO2 migration in a northeast direction, consistent with microseismic results. Finally, in order to verify the monitoring results of the above methods, the gas detection in the production well around the injection site was analyzed. The analysis results show that the primary direction of CO2 migration in the study area is the northeast direction, followed by the east direction. These geophysical techniques validate the accuracy of the monitoring results and provide new technical support for future CCUS projects.

Application of machine learning in carbon capture and storage: An in-depth insight from the perspective of geoscience

Greenhouse gas emissions cause serious global climate change, and it is urgent to curb CO2 emissions. As the last-guaranteed technology to reduce carbon emissions, carbon capture and storage (CCS) is emerging and has the potential to become an important choice to mitigate the greenhouse effect in the future. Forward-looking project deployment and accurate forecasting techniques occupy an indispensable position in CCS. Machine learning (ML), one of the fastest developing intelligent technology fields at present era, is considered as a substantial means to realize forecast demand relying on computer science and data statistics. This work provided a comprehensive review of ML applications in CCS, based on classical ML methods and mainstream research directions in CCS. The study shown that ML algorithms such as artificial neural network (ANN) and convolutional neural network (CNN) were widely used, mainly for predicting physical properties, evaluating mechanical stability, and monitoring CO2 plume migration and leakage during CO2 storage. Support vector machine (SVM) was generally combined with other ML methods for the prediction of petrophysical properties and sensitivity analysis of influencing factors. Deep learning (DL) algorithms, represented by generative adversarial network (GAN) and long short-term memory (LSTM), had shown good results in real-time monitoring of CO2 migration and leakage. Decision tree (DT) and random forest (RF) were mainly used to establish risk assessment and decision analysis framework, and estimate the success probability of CCS. This review summarized the applications of ML algorithms in CCS and presented the challenges and future prospects, from the geoscience perspective. The findings of this work can help for better understanding of the key role played by ML in CCS, as well as guiding the selection of ML algorithms and the development of new models in CCS research.

Quantitative interpretation of time‐lapse seismic data at Farnsworth field unit: Rock physics modeling, and calibration of simulated time‐lapse velocity responses

AbstractThis study investigates the contribution of fluid saturation variation to the time‐lapse velocity response by performing fluid substitution modeling. The methodology is exemplified by the time‐lapse seismic monitoring of carbon dioxide at Farnsworth field unit (FWU). In order to evaluate the fluid distribution in a matured oil reservoir, the Southwest Regional Partnership (SWP) acquired multiple vertical seismic profile (VSP) surveys at different times during the CO2–water alternatinggas (WAG) injection period. In this work, we present a thorough methodology for computing the elastic response of the saturated rock for different fluid saturations using a site‐specific petro‐elastic model (PEM). The output from the PEM was combined with results from a fluid compositional model to compute the seismic velocities at times corresponding to each VSP survey. To produce a calibrated simulated response, the measured time‐lapse seismic velocities were integrated into the numerical simulation model. The mismatches between the predicted and measured time‐lapse velocities were minimized through an iterative calibration process using a trained artificial neural network proxy (ANN) coupled with a particle swarm optimizer (PSO). Our study indicates that the hybrid optimization workflow can effectively perform the history matching. With an accurate prediction of the hydrodynamic properties, the migration of CO2 within the subsurface was modeled by predicting the spatial velocity distribution for a radius of 305 m around the injection well. The technology demonstrated and the expertise gained from this study can guide similar CO2‐WAG projects. © 2022 Society of Chemical Industry and John Wiley &amp; Sons, Ltd.

A screening assessment of the impact of sedimentological heterogeneity on CO2 migration and stratigraphic-baffling potential: Johansen and Cook formations, Northern Lights project, offshore Norway

• Johansen-Cook storage unit consists predominantly of wave-dominated deltaic sandstones. • Novel method used to screen impact of sedimentological heterogeneity on CO 2 migration and stratigraphic-baffling potential. • Screening method combines experimental design, sketch-based reservoir modelling and single-phase flow diagnostics. • Multiple types of sedimentological heterogeneity have significant effects. We use a method combining experimental design, sketch-based reservoir modelling, and single-phase flow diagnostics to rapidly screen the impact of sedimentological heterogeneities that constitute baffles and barriers to CO 2 migration in the Johansen and Cook formations at the Northern Lights CO 2 storage site. The types and spatial organisation of sedimentological heterogeneities in the wave-dominated deltaic sandstones of the Johansen-Cook storage unit are constrained using core data from the 31/5-7 (Eos) well, previous interpretations of seismic data and regional well-log correlations, and outcrop and subsurface analogues. Delta planform geometry, clinoform dip, and facies-association interfingering extent along clinoforms control: (1) the distribution and connectivity of high-permeability medial and proximal delta-front sandstones, (2) effective horizontal and vertical permeability characteristics of the storage unit, and (3) pore volumes injected at breakthrough time (which approximates the efficiency of stratigraphic baffling). In addition, the lateral continuity of carbonate-cemented concretionary layers along transgressive surfaces impacts effective vertical permeability, and bioturbation intensity impacts effective horizontal and vertical permeability. The combined effects of these and other heterogeneities are also influential. Our results suggest that the baffling effect on CO 2 migration and retention of sedimentological heterogeneity is an important precursor for later capillary, dissolution and mineral trapping.

Investigation of solubility trapping mechanism during geologic CO2 sequestration in Deccan Volcanic Provinces, Saurashtra, Gujarat, India

Solubility of CO2 in connate water plays a crucial role in facilitating the safe disposal of CO2 in geological formations, and this phenomenon in CO2 geological sequestration is termed solubility trapping. A multiphase-reactive transport code that solely considers CO2 dissolution and its accompanying pH fluctuation are used to investigate the influence of the geological and sequestration parameters on a solubility trapping mechanism. A 3-dimensional synthetic geological domain is modelled based on the geological data of Deccan volcanic provinces in Saurashtra, Gujarat, India. The stairsteps kind of morphology of geological formation obtained from literature was considered. Simulation results have shown that parameters such as caprock morphology, petrophysical properties, and injection location significantly affect solubility trapping. Solubility trapping can be observed by examining the pH variation, pressure distribution, and CO2 migration patterns. Analysis of such domains can give foresight about the possible implementation of CO2 sequestration and its influence on the solubility trapping mechanism.

Method for upscaling of CO2 migration in 3D heterogeneous geological models

Numerical reservoir simulation of CO2 storage in geological formations is computationally expensive due to the complex, and often coupled, multiphase flow phenomena, and the large spatial and temporal scales. Upscaling is a family of techniques used to decrease petrophysical degrees of freedom from a fine-scale model to a coarser one such that the computational costs can be substantially reduced while maintaining key fine-scale flow mechanisms, at least approximately. This research aims at developing methods suitable for upscaling CO2 migration in 3D heterogeneous geological systems. Two percolation-based methods are proposed to upscale the multiphase flow functions of the fine-scale models with large and zero correlation lengths, respectively. To confirm the applicability of the proposed upscaling methods, a case study is carried out. The combination of the two methods is adopted to upscale a representative of eolian sandstones with mixed correlation lengths. These are achieved thanks to the consideration of the dominant CO2 migration direction in aquifers. The computational errors between fine- and coarse-scale models, in terms of averaged CO2 saturation, were less than 10%, thereby demonstrating the robustness of the upscaling methods. The advantages of the proposed method are also highlighted by comparing it with other commonly used upscaling approaches. Overall, this research sheds light on the understanding of fine-scale CO2 migration in heterogeneous geological systems and illustrates the benefits of properly incorporating small-scale capillary barriers effects into coarse-grained scale models.

A monitoring system for CO2-EOR and storage based on reactive transport simulation of CO2 migration in groundwater

A monitoring system for CO2-EOR and storage based on reactive transport simulation of CO2 migration in groundwater

Biomimetic Tremelliform Ultrathin MnO2/CuO Nanosheets on Kaolinite Driving Superior Catalytic Oxidation: An Example of CO.

Highly efficient three-dimensional (3D) kaolinite/MnO2-CuO (KM@CuO-NO3) catalysts were synthesized by a mild biomimetic strategy. Kaolinite flakes were uniformly wrapped by ultrathin tremelliform MnO2 nanosheets with thicknesses of around 1.0-1.5 nm. Si-O and Al-O groups in kaolinite hosted MnO2 nanosheets to generate a robust composite structure. The ultrathin MnO2 lamellar structure exhibited excellent stability even after calcination above 350 °C. Kaolinite/MnO2 exhibited abundant edges, sharp corners, and interconnected diffusion channels, which are superior to the common stacked structure. Open channels guaranteed fast transportation and migration of CO and O2 during CO oxidation. The synthesized KM@CuO-NO3 achieved a 90% CO conversion efficiency at a relatively low temperature (110 °C). Furthermore, the abundant oxygen vacancies on KM@CuO-NO3 assisted the adsorption and activation of oxygen species and thus enhanced the oxygen mobility and reactivity in the catalytic process. The mechanism results suggest that CuO introduced to the catalyst not only acted as CO active sites but also weakened the Mn-O bond, subsequently improved the mobilities of the oxygen species, which was found to contribute to its high activity for CO oxidation. This study provides new conceptual insights into rationally regulating structural assembly between transition metal oxides and natural minerals for high-performance catalysis reactions.

Residual Trapping of CO2 and Enhanced Oil Recovery in Oil-Wet Sandstone Core – A Three-Phase Pore-Scale Analysis Using NMR

HypothesisCombining Carbon Geo-sequestration (CGS) with Enhanced Oil Recovery (EOR) in three-phase flow oil reservoirs is an effective and economically attractive technology to mitigate global warming caused by emission of anthropologic CO2 from fossil fuel combustion, as it offsets some of the costs of CO2 extraction. The targeted oil reservoirs display oil-wet characteristics, primarily because of the strong affinity of polar molecules present in oil; a mixture of a large number of chemical compounds to the rock's surface; and oil-wet surfaces that drastically reduce storage capacity of geological trapping mechanisms and increase the vertical migration of CO2. Thus, it is essential to characterise the effect of wettability of rock on pore-scale physics in a pore network that dominate three-phase flow, which in turn control overall reservoir‐scale fluid dynamics and; thus, determine residual CO2 trapping and the success of CO2‐EOR projects. ExperimentalTo fully examine this, we present the first in situ NMR T1-T2 2D images and transverse relaxation time (T2) to quantify residual CO2 (capillary) trapping for an oil-wet core at reservoir conditions, representative of virgin oil reservoirs at around 1000-meter depth. We also systematically measure how much additional oil can be produced, where we compared the results with measurements performed on an analogous water-wet core. ResultsThe results show a significant residual CO2 saturation (12%) can be stored, which is similar to the values found in analogue two-phase flow for oil-wet sandstone samples, although some details are different and displacements are significantly complex, and a substantial amount of oil (51%) was recovered. In contrast, results of similar tests conducted on an identical analogous water-wet core show a higher CO2 trapping (20%), where oil production was substantially higher (77%). We have also presented the oil displacement efficiency (η) versus CO2 injection (PV), showing the percentage oil displacement efficiency of water-wet (0.6%) to be significantly higher than that of analogue oil-wet core (0.4%). Interpretation of the findingThese results significantly improve geo-storage integrity; i.e. capacities and containment security, over reservoir-scale implementations, in addition to having a substantial impact on outlining CO2-EOR projects in term of budgets and storage capacities for CO2-EOR and CO2 geo-sequestration.

Porosity estimation by neural networks for CO2 storage in Otway site

Dynamic simulation of CO2 migration requires a variety of modeling parameters fed by geomechanical models. The confidence of these parameters of material groups such as porosity and permeability is crucial in achieving successful simulations. Based on the geomechanical and geophysical parameters, we estimated porosity distributions on the Paaratte Formation in the Otway site, one of the CO2 storage project in Australia. Considering the nonlinear relations between porosity logs and seismic data, we applied the neural network scheme that addresses the porosity value across a whole domain. With only one monitoring well and two injection wells at the site, seismic data are used to restore the spatial absence in porosity. The technique of the neural network was conducted based on the integration of the well logs to the seismic volume and the inversion of acoustic impedance. The results indicated that a correlation value of the well and the seismic tie is 75% and the value between the recorded and the estimated porosity is 87% on average. Further, the time slice maps of porosity at a depth of the injection interval demonstrated a CO2 plume developed in the Paaratte formation of the Otway site.

Modulating electronic structure of lattice O-modified orange polymeric carbon nitrogen to promote photocatalytic CO2 conversion

Adsorption and activation of CO2, and high carriers transform efficiency are of great significance for improving the CO2 photoreduction performance. However, electronic structure regulation plays an important role in the dynamical behavior of carriers. In this study, O atoms were successfully introduced into polymeric carbon nitrogen to synthesize an orange O-modified polymeric carbon nitrogen and achieve the rearrangement of electron density distribution, which promotes the adsorption and activation of CO2 and efficient separation and migration of carriers. DFT calculation demonstrates that the O atoms perform as the dominant active sites and reduce the free energy of determining step. Moreover, the introduction of O atoms broadens the light absorption range by changing the electronic layer structure of polymeric carbon nitrogen. Therefore, the fabricated O-modified polymeric carbon nitrogen exhibits significant performance. Under visible light irradiation, the CO generation efficiency achieved a 10-fold boost. We synthesize theory and experiment to highlight the important role of modulating electronic structure in designing photocatalysts.

High-Resolution Pore Space Imaging, Mineralogical Characterization, and Sealing Capacity Estimates of Confining Units at a Geologic Carbon Storage Demonstration: The Illinois Basin–Decatur Project, USA

At the Illinois Basin–Decatur Project, a large-scale CO2 capture and geologic storage demonstration project in the saline Mount Simon Sandstone in central Illinois, three overlying and laterally continuous shale formations (the Eau Claire, Maquoketa, and New Albany) are considered confining units overlying the sandstone reservoir. The Mount Simon reservoir contains internal mudstone baffles that will influence CO2 migration pathways and future interaction with seals. Understanding the sealing properties of rock units deemed seals or confining units is vital to commercialization of geologic carbon storage. For this article, nanoscale-resolution focused ion beam scanning electron microscopy (SEM), mercury injection capillary pressure (MICP), X-ray diffraction (XRD) spectroscopy, and quantitative evaluation of minerals by SEM were used to characterize the controls on the sealing integrity of these seals and reservoir confining units. Results show that porosity and pore size generally decrease with depth, except in the carbonate-rich Maquoketa Shale. The Maquoketa contains the highest pore volume owing to abundant dolomite in the mineral matrix compared with the other mudstone and shale intervals, which are clay rich. The shallowest seal sample, the organic-rich New Albany Shale, contains the highest frequency of the smallest pore throat size and is most comparable, with respect to pore sizes and entry pressures, to the deepest black shale and primary Mount Simon reservoir seal, the Eau Claire. Point-specific MICP threshold pressure results, theoretical calculations based on a range of permeabilities, and column height calculations indicate that the internal Mount Simon mudstone and Eau Claire Shale are effective seals of CO2 in the Mount Simon reservoir.

A numerical study on CO migration after blasting in high-altitude tunnel by inclined shaft

On the western plateau of China, ventilation problems brought on by low atmospheric pressure must be overcome. And CO migration after blasting in high-altitude tunnel by inclined shaft has become a significant scientific issue. In this study, the Computational Fluid Dynamics (CFD) method was used to analyze the flow field characteristics at the junction of the inclined shaft and tunnel. In addition, the effects of different fan opening modes and different initial CO concentration distributions on the ventilation were discussed. The simulation results showed that the main difference in the ventilation wind field was reflected in the position of the vortex region due to the different fan opening modes. Meanwhile, various initial CO concentration distributions showed different migration when there was no air volume difference between the left and right tunnels. Eliminating vortex zones and fully using high velocity airflow could improve relative ventilation efficiency by at least 18%. CO would accumulate in the opposite direction of the tunnel if only one of the fans was turned on. Therefore, a two-stage ventilation scheme was proposed, and the energy consumption was reduced by at least 33%. This research can provide guidance on high-altitude tunnel construction with multiple working faces to improve ventilation efficiency and reduce energy consumption.