| Special Topic: Machine Learning and Data Analytics for CCUS |
Bayesian Optimization for Field Scale Geological Carbon Sequestration
Mary Wheeler, Xueying Lu University of Texas at Austin; Kirk Jordan, Edward O. Pyzer-Knapp, Matthew Benatan, IBM Research We present a framework of the application of Bayesian Optimization (BO) to well management for geological carbon sequestration. The coupled compositional flow and poroelasticity simulator, IPARS, is utilized to accurately capture the underlying physical processes during CO₂ sequestration. We further couple the IBM Bayesian Optimization (IBO) with IPARS for parallel optimization of well injection strategies during field-scale CO₂ sequestration. Bayesian optimization builds a surrogate for the objective and quantifies the uncertainty in that surrogate using a Bayesian machine learning technique, Gaussian process regression, and then uses an acquisition function defined from this surrogate to decide where to sample. IBO addresses the three weak points of standard BO in that it supports parallel (batch) execution, scales better for higher-dimensional problems, and is more robust to initializations. An application to optimize the CO₂ injection schedule in the Cranfield site using field data benchmarked with genetic algorithm (GA) and covariance matrix adaptation evolution strategy (CMA-ES) shows that IBO achieves competitive objective function value with significantly less number of forward model evaluations. Furthermore, the Bayesian framework that IBO builds upon allows uncertainty quantification and naturally extends to optimization under uncertainty. |
Deep Learning Enhanced Joint Framework for Monitoring CO₂ Storage
Yanyan Hu, Xuqing Wu, Jiefu Chen, University of Houston Monitoring carbon dioxide (CO₂) storage is crucial to tracking CO₂ movement, evaluating storage integrity and early detection of CO₂ leakage. Using joint inversion of cross well electromagnetic data and seismic data to reveal underlying geology has drawn considerable research attention due to their characteristics of the complementary resolutions. We propose a deep learning enhanced joint inversion framework to simultaneously reconstruct the resistivity and velocity models to index the CO₂ plume, by fusing different types of geophysical data.
A key issue of joint inversion is to develop effective strategies to link different geophysical data in a unified mathematical framework. In our work, we enforce the constraint of structural similarity by a deep neural network (DNN) during the learning process. The framework is designed to combine the DNN and the traditional separate inversion workflow together and improve the joint inversion results iteratively. The network can be easily extended to incorporate multi-physics without structural changes. In addition, this learning-based framework demonstrates excellent flexibilities when the sensing configuration changes or different discretization is used for different models. Numerical experiments show that our deep learning enhanced joint inversion framework can reconstruct more accurate both physical property values and structures of the CO₂ plume than separate inversions and traditional cross-gradient based joint inversion. |
Machine Learning-guided Sand-volume Mapping for CO₂ Storage Site Location: A GOM Example
Hongliu Zeng, Mariana Olariu, Ramón Treviño, BEG, University of Texas at Austin In recent years, machine learning (ML) has been proven a useful method in many geoscience applications, including facies analysis and reservoir prediction for CO₂ storage purpose. We have completed a pilot study in TexaLa Merge survey area, in which 400 km2 3D seismic data and 5 wells were analyzed with a ML work flow for sand-volume estimation in Miocene. In the past, this kind of data condition only allows for amplitude mapping or conventional seismic inversion, generating low-grade, low resolution targets. Yet a sparse well-based ML learning is typically biased for a complex depositional sequence. Our new workflow started with building a large acoustic impedance model with the sparse well data as truth model using geostatistical method (sequential Gaussian Simulation), then converting the impedance model to synthetic seismic training model as truth data. We trained the model using random-forest algorithm, and applied it to the field seismic data volume for predicted impedance, which in turn was used to calculate the volume of shale and thickness of sandstone at high-frequency sequence level. Major conclusions are: (1) ML inversion has higher resolution than calculated seismic resolution. In the pilot project, the resolution is 5 m, or 10-30% higher than conventional poststack inversion. (2) Results are of higher accuracy than seismic inversion. In model testing, R2 score ranges 0.7-0.9, higher than conventional inversion (around 0.5-0.6), which means fewer dry wells in real world. (3) The workflow can be expanded to a larger survey (e.g., 10,000 km2) with current computing power. These results, in our view, show significant improvement over current approaches without any new requirement on data quality and availability, which would be meaningful to future CCS implementation. |
Predicting CO₂ Gravity-Driven Drainage Saturation using Machine Learning
Hailun Ni, Sahar Bakhshian, T. A. Meckel, Bureau of Economic Geology, Jackson School of Geosciences, The University of Texas at Austin; Anthony Zuniga, The Department of Computer Science and Engineering, University of North Texas To ensure the long-term security of injected CO₂, reservoir simulations are conducted to model CO₂ plume migration and trapping. Research has shown that cm-to-m scale heterogeneity can have a significant influence on the amount of CO₂ residual trapping and hence impacting the plume migration speed and extent. Typically, computationally-intense fluid flow simulations need to be conducted on fine-grid heterogeneous geologic domains to obtain the effective CO₂ trapping amount to be integrated into field-scale simulation models. In this study, we present a much faster machine learning model that can predict the post-injection CO₂ saturation at percolation for any heterogeneous domain based on primary geologic parameters.
Initial data generation is done by conducting CO₂ gravity-driven drainage simulations using modified invasion percolation techniques (PermediaTM Static Migration module). A planar CO₂ source at the bottom of the domain is allowed to migrate under gravity, and simulation terminates when the CO₂ plume percolates at the top of the domain. A dataset is generated for a diverse set of high-resolution, small-scale models (59 sedimentary bedform geometries, each with 40 matrix-laminae grain size contrast cases and 50 realizations) containing 118,000 data points of CO₂ saturation for heterogeneous domains. The average of the 50 realizations yields an average saturation for each bedform model and grain size contrast (2360 data points). Next, we extract predictors as inputs into the machine learning models. Extracted predictors include numeric ones such as the matrix-laminae grain size contrast and the geological entropy, as well as categorical ones that are the geologic descriptors of the bedform architecture. Machine learning models trained include linear models, K-nearest-neighbor models, tree-based models, and neural networks. According to the cross-validation metric associated with the training set, the gradient-boosted tree model is found to have the best performance. The corresponding test set regression performance on 8 bedforms for all 40 cases of grain size contrast is above R2=0.9. We have found that the grain size contrast predictor is the most important parameter for determining CO₂ saturation at percolation, with bedform architecture playing a minor role. |
Upscaling relative permeability from pore scale to field scale using conditional deep convolutional GANs (Conditional DCGAN)
Zihan Ren, Sanjay Srinivasan, The Pennsylvania State University Relative permeability (kr) is an important parameter that controls CO₂ displacement during geologic sequestration. However, kr is mostly measured from cores (cm scale) or from high-resolution micro-CT images (μm scale). A novel technique that can model relative permeability at field scale (km scale) combining pore scale simulation with machine learning-based pore technique has to be developed together with application of scaling analysis for considering CO₂ flowing behavior at field scale.
Pore network model can be extracted from representative micro-CT image to perform multiphase flow simulation to get relative permeability curves. Deep convolutional generative adversarial network (DCGAN) is widely used for image generation, including 3D micro-CT image reconstruction, which is made without conditioning to 2D slices. The regeneration process of DCGAN can be controlled through adding extra conditional layer to the input, which can be customized with geological properties such as porosity, permeability and pore size distribution etc. By doing so, the reconstruction of heterogeneous representative 3D micro-CT volume can be conditioned to reservoir specific data such as from well logs. Different relative permeability curves can be simulated on these reconstructed porous media to characterize heterogenous flowing behavior of CO₂ in subsurface.
Results indicate that 3D micro-CT images can be reconstructed to reproduce hydraulic properties such as porosity and absolute permeability. Therefore, given multiple spatial realizations of porosity, permeability and pore size distribution near wellbore, the generator extracted from trained conditional DCGAN will be able to generate multiple realizations of kr curves in each grid block, which characterizes the flowing behavior of CO₂ in high resolution while considering both representative pore network model and field scale heterogeneity.
The novelty of this work arises from (1) physics assisted data driven approach that can reconstruct porous media while combining both pore scale features and field scale data; (2) new workflow to successfully predict relative permeability given only static hydraulic features such as porosity and permeability etc.; and (3) extrapolate small and sparse expensive data (for example relative permeability extracted from micro-CT images.) to a large volume conditioned to field scale data in the form of well log derived porosity and pore size distribution etc., and (4) more accurate characterization of flow behavior in subsurface to support future reservoir simulation development. |
| Theme 1: Subsurface Storage |
3-D Model-Based Inversion for Onshore CO₂ Storage in STACK play, Anadarko Basin
Victor Fakeye, Camelia C. Knapp, Oklahoma State University In a bid to attain an atmospheric carbon-neutral or -negative future energy economy, society will need to apply all possible methods at its disposal. One method at the disposal of many regions worldwide is to sequester carbon dioxide (CO₂) and other greenhouse gases into depleted hydrocarbon or deep saline reservoirs. This research focuses on the characterization of potential storage reservoirs in Anadarko Basin centered in the western part of the state of Oklahoma and the Texas Panhandle, and extending into southwestern Kansas and southeastern Colorado. To provide a detailed evaluation of the porosity and permeability – essential properties for reservoir performance, a model-based inversion of seismic data is necessary to obtain the Acoustic Impedance (AI) values for the potential reservoir and seal intervals. The AI reflectivity is used to predict porosity from post-stack seismic inversion. The inversion procedure involves well to seismic calibration, wavelet extraction, estimation of low-frequency model and model-based inversion for available seismic dataset. In more detail, the initial model of impedance is generated by using the P-impedance logs calculated from the sonic and density logs from the well log with a low-pass filter. Cross plotting is an effective method to link the acoustic impedance with porosities. Results will show the reservoirs suitable for CO₂ storage in the study area." |
A Micro-CT Investigation into the Pore Network and CO₂ Storage in Stevens Sandstone of California, USA
Liaosha Song, Elizabeth Duginski, Tabitha Guadian, California State University Bakersfield The Miocene Stevens Sandstone in San Joaquin Basin of California is a promising target for CO₂ geological storage and sequestration. Significant heterogeneity in reservoir properties have been noticed in this formation due to the complicated geology in Miocene sedimentation in San Joaquin Basin. A thorough understanding of the porous system in the rock will benefit both CO₂ storage and enhanced oil recovery (EOR).
In this research, Stevens Sandstone core samples were collected from a well in San Joaquin Basin. A three-dimensional characterization on the pore networks of these samples was conducted. Samples were analyzed at micrometer scale. High resolution X-ray micro computed tomography (micro-CT) scan, a non-destructive analytical method, was utilized to investigate the pore network. The pore networks and quantitative results were extracted using image analysis, including porosity, effective porosity, pore size distribution, etc. Absolute permeability simulation was then performed, and CO₂ flow was modeled. |
An Approach for Sealing Depleted Oil and Gas Reservoirs while Sequestering Carbon Dioxide
Daniel J Soeder, Daryl-Lynn Roberts, Hunaid Nulwala, Carbon Blade Corporation; Gokce K. Ustunisik, Bret N. Lingwall, South Dakota School of Mines & Tech Carbon dioxide dissolved in water creates carbonic acid (H2CO3). Substituting a different cation for hydrogen will form carbonates that can be precipitated as minerals. Examples include calcite (CaCO3), magnesite (MgCO3), siderite (FeCO3) and nahcolite (NaHCO3). Recent field experiments in Iceland discovered that CO₂-saturated water injected into basalt reacted with plagioclase feldspar minerals in the rock, releasing Ca ions that precipitated substantial quantities of calcite in as little as two years. Many marine organisms from clams to corals precipitate carbonate out of seawater for shells through the process of biomineralization. Multidisciplinary research at the South Dakota School of Mines & Technology has been investigating the potential for genetically engineered microbes to biomineralize carbonate from CO₂ solutions much more rapidly than inorganic processes and in a variety of different subsurface environments. One such environment of interest is depleted oil and gas fields containing orphaned or abandoned wells. Many of these old wells have not been properly plugged and produce fugitive emissions of methane. A number of U.S. states are concerned about methane leakage from orphaned wells, and several have set aside substantial budgets to properly P&A these. The microbes under investigation are extremophiles, capable of surviving the pressure, temperature, salinity, and pH regimes likely to be encountered in conventional oil and gas reservoirs at depths between 2,500 feet (762 m) and 10,000 feet (3,045 m). Oilfield brines obtained from depleted production wells in an inverse five-spot pattern can be saturated with CO₂, inoculated with engineered microbes and returned to the reservoir via a central injection well. The original trap will contain the CO₂ until it is biomineralized. There may be enough cations (Na, Mg, and Ca) from salt in the brines for the microbes to precipitate carbonate minerals directly, or the injection well can be filled with crushed basalt to release Ca ions into the reservoir. Carbonate precipitation into subsurface pore space will occlude it and block fluid flow, and also eventually plug the abandoned wells, ending fugitive emissions of methane. The CO₂ source is a distributed, direct air capture (DAC) technology consisting of a self-contained system mounted on camper-trailer sized units. A small wind turbine supplies power and acts as an air contact surface for capturing CO₂. The carbon capture technology is a sodium hydroxide-bicarbonate-sulfate acid/base reaction that collects one to two tons of CO₂ per day using an electrodialysis bipolar membrane (EDBM) to regenerate the acid and sodium hydroxide. The stand-alone nature of the DAC units allows them to be deployed nearly anywhere, including remote, depleted oil fields that may lack electrical power and pipeline infrastructure. Renewable wind power and the low energy demands of the EDBM produce low cost CO₂. Applying this technology in depleted oilfields offers carbon storage that is both permanent in the form of solid minerals, and substantial, given the large volumes of subsurface pore space. The DAC technology requires bench testing and optimization, followed by a field test of the biomineralization. Once developed, carbon offsets and the 45Q tax credit could provide favorable economics. |
Application of XRF-Based Elemental Data to Optimize Core and Cuttings Characterization for CCUS Projects: Ready Adaptation of Workflows Developed for Oil and Gas Operations
Michael Charles Dix, Tim Prather, Austin Morrell, Giovanni Zanoni, Nicholas Nelson, Harry Rowe, Premier Oilfield Group; Julie M. Bloxsom, Hannah Chambers, Stephen F. Austin State University Because of their rapid acquisition and versatility, elemental data obtained from core and cuttings by portable X-ray fluorescence (XRF) instruments are now routinely used in many oil and gas drilling programs. While these XRF-based workflows are straightforwardly adapted to the characterization of CO₂-injection zones and seals, they may be unfamiliar to some geologists and engineers working on CCUS projects. The XRF-based CCUS workflow described below is equally applicable to sandstone, carbonate, and mudstone lithologies. Examples from two analog injection-zone sandstones in the East Texas Basin will be presented (Jurassic Cotton Valley and Cretaceous Blossom).
A typical workflow for a new core begins with acquisition of high-resolution XRF data from the core face just after slabbing (typically at a spacing of 0.1 to 0.2 feet). The XRF data (usually 25-30 elements) is processed through a mineral model, and then classified into elemental lithotypes (chemofacies) using hierarchical cluster analysis. The detailed compositional profile constructed from the XRF data is combined with log responses to optimize sample selection for routine core analysis, special core analysis, petrology, and rock mechanics. Petrology analyses (thin section, SEM, and XRD) are specifically focused on quantifying the amount and distribution of potentially reactive pore-facing mineral phases. Lastly, the XRF data is used in conjunction with core-description sedimentary facies for chemostratigraphic correlation, improving the understanding of overall depositional architecture. Relevant analytical parameters from all of these results are parsed or integrated to provide input for reactive transport modeling.
Following integration and interpretation, the workflow is inverted. Mathematical relationships are established between measured rock parameters and XRF elemental data. The most robust of these relationships (typically including mineralogy and rock mechanics parameters) can be applied to new XRF data acquired from cuttings as additional injection wells are drilled. The new XRF results can recognize lateral changes in injection zone composition, and potentially provide input for updating reactive transport models. |
De-risking Danish large-scale CO₂ injection into depleted chalk reservoirs
Tobias Orlander, Walter Wheeler, Leonardo T. P. Meireles, Ida L. Fabricius, Technical University of Denmark; Anders Nermoen, Roman Berenblyum, Thomas G. Petersen, Norwegian Research Centre AS; Frederik P. Ditlevsen, Geo, Denmark On the Danish continental shelf, large infrastructure investments have been done in pipelines, wells, and platforms for fossil fuel extraction from chalk fields. These chalk fields are now depleting.
Large savings could be made if the same infrastructure were reused and given a 2nd-life, serving as transport and storage sites of CO₂. Reservoir insights and experience of the Danish oil and gas industry could be used and developed – as a fossil industry in reverse. Chalk is made up of compacted carbonate ooze, so concerns for storage safety have been raised – can large-scale CO₂ injection weaken chalk reservoir rocks? This means that the risk of compaction must be identified and critically evaluated.
We take a fundamental approach by evaluating to which extend chemical rock-fluid interaction would mechanically soften and weaken Danish chalk reservoirs. As such, if reservoir rocks maintain elastic stiffness and plastic strengths upon CO₂ injection, or if the risks of the softening and weakening is manageable, then significant cost cuts in the transport and storage of captured CO₂ are anticipated.
From basic scientific viewpoints, there are unresolved questions on how the CO₂, when mixed with formation waters, affects chalk. Mixing CO₂ and water forms a weak acid that dissolves calcite minerals until a new buffered equilibrium is met. Dissolution of the chalk framework could thus reduce its stiffness and strength. Whether chemical dissolution plays a significant mechanical role depends upon how chalk grains are held together (cementation of grain contacts as expressed in Biot’s coefficient), and how much calcite would dissolve in a given injection scenario – especially whether calcite cement in contacts would diminish or even grow. The latter process would strengthen the frame, although it could potentially also clog pore throats.
The overall importance of dissolution must be evaluated to determine if this is a real risk. Further, these chalks possess a large porosity, but low permeability due to the small mineral grains and corresponding high specific surface area. Thus, surface-effects are much more important in chalks than in, e.g., sandstones.
The calcite mineral has charged surface sites, where differently charged ions can adsorb from the water phase, causing a net electrostatic potential. When two charged calcite crystals are closely together, these electrostatic potentials overlap so that a disjoining pressure develops. A range of chalk and calcite experiments have shown how these overlapping electrostatic layers may soften and weaken the rock, but other experiments have shown that when CO₂ is injected with saline water during continuous creep experiments, the creep-rate is reduced – implying a strengthening of the chalk.
Tests are therefore planned where finite amounts of CO₂ is injected into reservoir rock samples, with controlled pore fluid composition and temperature. Fluid will be sampled to evaluate the amount of dissolution during injection, while mechanical stiffness is estimated from sonic velocities before, during and after CO₂ exposure. We estimated strength properties from stress-strain curves recorded during triaxial testing. |
De-risking long-term CO₂ containment on the Upper Gulf Coast
Alexander Bump, Gillian Apps, Frank Peel, Madeleine Laidlaw, Bureau of Economic Geology Against the backdrop of increasing public concern about climate change, there is rapidly growing interest in Carbon Capture and Storage (CCS) as a flexible solution for decarbonizing hard-to-abate industries such as dispatchable power, natural gas processing and the manufacture of cement, steel and petrochemicals. With over 50 years of experience injecting CO₂ into subsurface reservoirs, researchers and industry pioneers have proven that CCS is safe and can be deployed at industrial scale. However, developing new sites and projects still requires careful geologic characterization to find the capacity needed and demonstrate reliable containment. While historic oil and gas experience is valuable the need for ongoing assurance of containment under increasing pressure puts new focus on seal performance and the risks posed by existing breaches (faults and legacy wells).
The US Gulf Coast offers a timely example and a natural place to explore ways to de-risk containment. Texas and Louisiana lead the nation in point-source CO₂ emissions and the region is a hub of hydrocarbon production, with proven subsurface reservoirs and seals, plus abundant infrastructure, data and expertise. Similar storage opportunities exist both onshore and offshore but offshore has the added attractions of little to no fresh water, fewer wells, fewer competing uses and a single landowner.
Detailed mapping of the coastal region shows an abundance of potential storage sites. Most have been explored for oil and gas and seals on some are proven by the presence of hydrocarbons that have been retained on geologic timescales. The exploration “dry holes” raise a question, however—why did those prospects fail? The possibility of a problem with the seal casts doubt on their viability for carbon storage. The work presented here combines geologic mapping, sequence stratigraphy, column height data, seismic interpretation and flow modeling to investigate the capacity and extent of the main regional seals. We show that many of the dry holes failed not because of top seal but because of defocused charge. This work helps to de-risk CO₂ containment on the Gulf Coast but it also suggests methods for doing so elsewhere. Demonstration of robust containment will be critical if CCS to grow to the scale needed to mitigate climate change. |
Developing a Regional Framework to Define and Assess CO₂ Storage Systems in the Midwestern to Northeastern United States
Amber Conner, Mark Kelley, Autumn Haagsma, Priya Ravi Ganesh, Neeraj Gupta, Battelle Memorial Institute; Sallie Greenburg, Hannes Leetaru, Illinois State Geological Survey; Steve Greb, Kentucky Geological Survey; Jessica Moore, West Virginia Geological and Economic Survey; Kristen Carter; Pennsylvania Geological Survey; William Harrison, Michigan Geological Repository for Research and Education, Western Michigan University Geologic providences can encompass multi-state regions and include multiple potential storage systems suitable for carbon capture, utilization, and storage (CCUS) projects. Identification and characterization of these systems requires multi-organization collaboration, along with the development of reservoir and caprock properties to determine storage capacity, sealing potential, and structural integrity of a storage site. Deployment of a CCUS project presents numerous technical challenges concerning geologic storage potential such as capacity, containment, low injectability, and public acceptance. The Midwest Regional Carbon Initiative (MRCI) project was established as a regional collaboration to address these challenges that prevent the establishment of CCUS. The collaboration is assessing 20 states in the Midwest and Northeastern United States and are examining several major Carbon Systems within several basins, including the Michigan, Illinois, and Appalachian basins. As part of the MRCI initiative the collaboration compiled geologic property data including, porosity, permeability, structure maps, seismic, pressure and temperature, and core analyses. These data will be used to expand characterization of potential single and stacked storage complexes, reduce uncertainties in reservoir and caprock properties, characterize geologic structure including basement faulting and stress state, and reduce uncertainty storage integrity risks. The initial effort will focus on the analysis of nearby sources, assessing storage certainty (storage resource to storage reserves) based on site geology, and assessing CO₂ containment. The storage complex definition for each location will account for location, reservoir and caprock types and descriptions, reservoir properties, production histories of oil and gas fields, modeling parameters where available, state of stress based on regional data, and fluid properties. The collaboration produced a first-of-a-kind regional database, and the results will be used for identifying prime sites and evaluating site readiness.
Project is funded through U.S. DOE DE-FE0031836 |
Pre-Feasibility Study for On Site CO₂ Subsurface Storage at an Ethanol/Biofuel Plant Located in the Northern San Joaquin Basin, CA
Jamar Bynum, David Katz, Ehsaan Nasir, Santi Randazzo, Suat Bagci, Mehagan Hopkins, Aemetis Carbon capture and storage (CCS) holds potential for dramatic reduction in anthropogenic greenhouse gas emissions. Stated carbon reduction targets of governments and industry require cost effective and timely solutions. CCS projects may be implemented as large-scale industrial “hubs” with substantial pipeline infrastructure to transit CO₂ to established storage reservoirs. However, hubs require considerable capital investment, have counterparty risk, and due to scale and complexity possess long development timelines. An alternative development model pursued by early adopters are ad hoc local storage networks. These storage networks are anchored by a sufficiently large singular CO₂ source which could also gather from a number of smaller proximal sources. Local networks possess advantages for accelerating the timetable for CCS due to lower CAPEX and less counterparty dependency. These advantages can be outweighed by the difficulty of proving sufficient storage capacity or integrity with limited data availability. This study pertains to a pre-feasibility investigation of a local storage network located in north central California within the San Joaquin Basin. The aim of this study was to evaluate storage potential of the site and determine whether geologic criteria exist in the subsurface for pursuing a Class VI well permit from the EPA for onsite CO₂ injection.
The project goal was to construct a geologic box model with the capacity to run numerical simulations for testing potential injection rates and plume migration scenarios. The potential injection reservoirs identified are Late Cretaceous strata, and record a series of interbedded marine sandstones and mudstones. The depositional system is characterized by progradation of submarine fans, slope, and shelf deposits with sediment sourcing from the Sierra Nevada uplift to the north and east. Regional correlations were performed on well logs and tied to a 2D seismic line for identifying basement structure, significant stratigraphic surfaces, geometry of depositional packages, and injection reservoir candidates with potential sealing intervals. Facies analysis was conducted to distinguish reservoir facies (sandstone) vs confining facies (mudstone), and are defined principally by the estimated volume of shale (VShale) calculated through the basic well log suites acquired from hydrocarbon exploration efforts in the 1960’s (spontaneous potential, gamma ray, and resistivity). While these data sets were critical for establishing the architecture of the geologic model it posed challenges for establishing the detailed reservoir parameters needed for numerical simulation exercises. Reservoir parameters typically acquired through modern well log suites, full seismic surveys, and core analysis were not available. Reservoir parameters (porosity, permeability, etc.) were constrained based on regional reports of correlatable strata.
Taking a multi-disciplinary approach during pre-feasibility investigations poses modeling challenges due to limited data availability, but it also provides key environmental and economic reference points for project viability. Successively, this workflow allowed investment decisions to be made on allocating capital to fund a stratigraphic test well. The intent of this investment is to improve the next phase of subsurface modeling efforts for fulfillment of the EPA site characterization guidelines for a Class VI CO₂ injection well permit. |
Quantifying Rock Characteristics in the San Andres Formation that Promote CO₂ Sequestration, Permian Basin, USA
Mitchell Schneider, Colorado School of Mines, Zane Jobe, Jonathan Knapp, Bruker Nano Analytics The San Andres Formation is a conventional carbonate reservoir on the Central Basin Platform, Permian Basin. The San Andres has been a prolific producer of oil and gas, but vertical and lateral heterogeneity within and between fields make reservoir characterization and thus recovery difficult. Carbon dioxide (CO₂) flooding has long been used for enhanced oil recovery operations within the San Andres Formation, and some fields unintentionally sequester large volumes of CO₂. However, the rock characteristics that allow effective CO₂ sequestration (e.g., lateral geological heterogeneity, diagenetic evolution, pore-network dynamics, fluid-rock interactions) are still uncertain. There is a desire to transition these reservoirs into permanent CO₂ sequestration sites due to existing infrastructure and the history of CO₂ injection. Using thin-section data from several fields on the Central Basin Platform, we quantify heterogeneity within pore and pore throat networks using to provide a ranking for transitioning these reservoirs into permanent CO₂ sequestration sites.
We analyzed thin sections using field-emission scanning electron microscopy to document the mineralogy and porosity network at the micron scale. The resulting data documents pore dimensions, pore-lining minerals, and pore-network heterogeneity between different facies and stratigraphic intervals of the San Andres Formation. We integrate this thin section data with existing core-plug porosimetry and field-wide production data to quantify the CO₂ trapping capability of the San Andres Formation, and thus the viability for a particular field to be converted to CO₂ sequestration. This newly collected data can help make informed economic decisions on the future utilization of depleted carbonate petroleum reservoirs, not only in the Permian Basin, but globally. |
The Effects of Wettability and Relative Permeability Variations on the CO₂ Injectivity, Storage Capacity, and Trap Mechanism–A Laboratory Study for WY CarbonSAFE III Project
Ying Yu, Davin Bagdonas, Charles Nye, Zunsheng Jiao, Matthew B Johnson, Jonathan Fred McLaughlin, Scott Quillinan, Center for Economic Geology Research, University of Wyoming Achieving carbon neutrality is a top global initiative. Wyoming CarbonSAFE (WY C-SAFE) at the Dry Fork Station in Campbell County, Wyoming, is considered one of the flagship carbon capture and storage projects (CCUS) in the United States and is led by the Center for Economic Geology Research (CEGR) at the University of Wyoming. The project is focused on geologic characterization and permitting for the injection of 50 MT of CO₂. One stratigraphic test well has been drilled, logged, and pre-feasibility and feasibility work has been performed for this project. Evaluating wettability and relative permeability in the CO₂-water fluid flow system is essential for injectivity, storage capacity, and contentment assessment for any commercial scale CCUS project. This paper aims to focus on the wettability and petrophysical properties characterization of the target reservoir formations (the Lower Cretaceous Lakota Sandstone, Jurassic Hulett Sandstone and Pennsylvinian Minnelusa Formation) near the Dry Fork Station and the impact of these properties on the CO₂ storage injectivity and capacity of the reservoir.
In this study, rock samples from the target storage formations Lakota, Hulett, and Minnelusa were selected for analysis, based on the representative lithology, permeability, and porosity of the respective formations. The rock samples are all fine-grained sandstone with sedimentation, cement, and banding variance. The porosity and permeability vary with the range of 9.1–14.3% and 0.1–14.2 mD, respectively. These rock samples are then adopted and cropped in two forms, square pieces and core plugs, for wettability, relative permeability, and CO₂ drainage and storage capacity evaluation. The quantitative variance of the wettability, relative permeability, and the corresponding CO₂ storage capacity results among the target storage formation samples are determined and discussed. Additionally, the wettability and lab capillary pressure correction for sealing formations, Fuson (directly above Lakota and Hulett) and Opeche (directly above Minnelusa), are also included in the study. Finally, the best candidate reservoir is recommended while the wettability and petrophysical characteristics that could suggest the best candidates for CO₂ storage are also concluded. The systematic evaluation presented in this study provides a countable standpoint for choosing the best storage formation candidates and a time-efficient workflow that benefits CO₂ storage projects. |
Transforming the Assessment of Risk and Volumetrics for Carbon Capture and Sequestration through the Application of Basin Modeling Technology
Robert Tscherny, ConocoPhillips The recent oil and gas industry focus on Carbon Capture and Storage (CCS) projects require two integrated assessments; (1) the volume of carbon dioxide (CO₂) that can safely be stored in the subsurface, and (2) the containment risk and leakage rate over the subsequent decades or centuries. The presented proof-of-concept study illustrates how utilizing basin modeling techniques can impact selecting the best storage sites, outline expectations around monitoring, and most importantly, ensure containment & storage safety.
The starting point for this study is a basin screening followed by an in-depth analysis of potential storage sites. The proposed screening workflow is straightforward. Use a basin model's ambient pressure and temperature, and perform a ray-tracing / flow path analysis of a geological reservoir-seal coupling by injecting and migrating CO₂. This analysis calculates the maximum storage volume and phase using trap characteristics (fill and spill), dynamic constraints of capillary seal capacity, and ambient CO₂ density. Integrating these results with surface constraints, such as distance to an emission source, pipeline, etc., helps high-grade potential storage sites for further analyses with high-resolution asset scale 3D geocellular models.
These asset scale 3D models use the PetroMod's "Nested Model" workflow coupled with the percolation migration method. The Nested Model" workflow is similar to sector modeling in the reservoir engineering world and was collaboratiively developed and realized by Schlumberger and ConocoPhillips. At the core is the ability to "nest" volume-of-interest (VOI) or a sector model into a regional basin model. It is an independent model that allows modeling at genuine prospect and well scales, maintaining the boundary conditions of a regional model without the need to "carry" tens of millions of unnecessary cells.
This novel workflow results in an interactive and fast-paced evaluation of the sequestered CO₂ volume (free and in water phase), the containment risk, and the injection and monitoring wells placement. The models in this study have 2 to 5 million flow cells and run fully coupled within minutes to hours while capturing all fundamental physical and chemical processes influencing storage efficiency and containment within a saline aquifer. These include the injection and migration of CO₂ in saline aquifers, dynamic leakage through top- and lateral seals during the plume migration, dissolution (and diffusion) within the water phase, the impact of the injection pressure on the seal integrity, and mineral trapping of CO₂ as carbonate minerals.
In summary, examples from the study illustrate how the basin modeling technology and techniques improve CO₂ storage site assessment. The workflow is by design a complementary precursor to asset-specific reservoir models. It highlights the strengths and weaknesses of the storage sites. It further helps to understand the optionality from a subsurface point of view, ultimately aiding in sound business decisions and communicating challenges. |
| Theme 3: Monitoring |
DNA Diagnostic enabled Total Fluid Monitoring Tool applied as a Non-Invasive, Cost-Effective, Low Carbon Footprint Surveillance Technology for CCUS/CCS Complexes
Caroline Burke, Matthew Haggerty, Mathias Schlecht,Thomas Ishoey, Biota Technology, Inc. Carbon Capture and Storage is now accepted as the only technologically viable method to mitigate carbon emissions and control climate change, while keeping net-zero targets within reach. Federal and State incentives have made CCS/CCUS economically viable and the focus of much of the growing decarbonization effort. While significant progress has been made in CCS technology in the last decade, there are still many challenges to commercial deployment.
Biota presents an independent tool to help high-grade, de-risk and monitor subsurface carbon storage opportunities, inform management strategies and provide long-term monitoring to improve efficiency and meet regulatory requirements.
The application of DNA markers to inform CCUS applications is a parallel to Biota’s current commercial applications of DNA monitoring in Oil & Gas which have facilitated insights into Drained Rock Volumes (DRV’s), fracture barrier integrity, total fluids production allocation, enhanced recovery with pressure management across a pad and improved sweep efficiency with water flooding.
Subsurface DNA diagnostics provides a spatially and temporally scalable, non-invasive measurement for tracking fluid movement in the subsurface. DNA enabled total fluid monitoring can inform both cap-rock integrity over time and flag changes in aquifers and rock formations that should be safely above the cap-rock or beyond the modelled limits of the CO₂ plume. This is a high resolution, cost-effective and low carbon foot-print monitoring technology which can be carried out over the lifetime of a project. The ease of sampling and rapid turnaround time will allow for active field development management ensuring maximum CO₂ storage. The technology provides leading indicators which can be used to trigger more cost and effort-intensive field monitoring or development technologies.
The applications discussed in this paper will include high resolution, time-lapse monitoring of total fluid from the storage complex and overburden enabling detection of any perturbations in the microbial system, CO₂ plume migration monitoring, the use of leading indicators for supercritical-CO₂ front detection as well as potential microbial remediation upon CO₂ dissolution.
DNA diagnostic enabled total fluid monitoring must be part of an integrated, multi-disciplinary approach for effective CCS/CCUS implementation and surveillance. The DNA marker stratigraphy is also a valuable new dataset to integrate with other subsurface data to enable improved, high resolution reservoir models leading to more robust predictions of CO₂ migration.
This technology promises to be an integral tool to ensure safe storage of CO₂ and in doing so play an important role in the journey to net zero. |
Low-Environmental-Impact Seismic CO₂ Monitoring in a North Dakota Carbon Capture and Storage Project Integrated with Ethanol Production
César Barajas-Olalde, Trevor L. Richards, Donald C. Adams, Kris MacLennan, Justin T. Kovacevich, Amanda J. Livers-Douglas, Kerryanne M. Leroux, John A. Hamling, Energy & Environmental Research Center University of North Dakota; Dustin Willett, Red Trail Energy, LLC; Ziqiu Xue, Research Institute of Innovative Technology for the Earth; Barry M. Freifeld, Class VI Solutions, Inc.; Julia Correa, Lawrence Berkeley National Laboratory Red Trail Energy (RTE), an ethanol production plant in Richardton, North Dakota, seeks to make ethanol more valuable by integrating carbon capture and storage (CCS) to reduce CO₂ emissions from ethanol production. The goal of the RTE CCS project, a multiphase research and development effort, is to create the first integrated CCS system in North Dakota. Led by the University of North Dakota Energy & Environmental Research Center (EERC), with support from RTE and the U.S. Department of Energy, technical partners in this research include the Plains CO₂ Reduction Partnership Initiative, the Research Institute of Innovative Technology for the Earth, and Class VI Solutions, Inc.
In 2016, the EERC started research for the RTE CCS project. Initial research demonstrated the preliminary technical and economic feasibility of CCS technology at the site. More in-depth assessments have been conducted in the subsequent research phases, such as detailed designs and plans for capturing, transporting, and storing CO₂. The research investigated the site’s geology and gathered information needed to comply with state regulations for injection and permanent storage.
Results of a 3D seismic survey performed in March 2019 showed promising geology under the ethanol plant east of Richardton. Seismic data helped identify two possible injection zones: the Broom Creek Formation at 6400 feet deep, with an average thickness of 295 feet, and the Inyan Kara Formation at 4800 feet deep, with an average thickness of 410 feet. Building on those results, two boreholes were drilled to the potential storage zones to collect rock samples, fluids, and other geologic data to enhance the geologic models and improve the accuracy and precision of the CO₂ injection simulations. Drilling and sample collection were completed in spring and fall 2020. Analyzing the data and evaluating the geology/refined models lasted the remainder of 2020. Recent activities include equipment contracting, public outreach, and development of the first North Dakota CO₂ storage facility permit application, submitted in February 2021 and approved on October 19, 2021.
In late 2021, the EERC will begin 3–5-year research project around the RTE CCS site, exploring a lower-cost, less invasive seismic technique that could replace the large-scale 3D seismic surveys by using four stationary seismic sources called surface orbital vibrators (SOVs) and a small number of seismic sensors deployed sparsely within the same 8-square-mile area investigated with the 3D seismic survey conducted in 2019. In the project’s first phase, an active and passive noise test will be conducted. Furthermore, drones will collect land-use images to determine possible sensor locations for the second phase.
The project’s second phase will last about 3 years. Sensors will be placed sparsely and in minimally disruptive locations throughout the study area to record seismic reflection before (baseline) and after the start of CO₂ injection (monitoring) from daily/weekly SOV operation. InSAR measurements will be assessed as a CO₂ monitoring tool in the study area in the second phase of the project.
In this talk, we present the results of Phase 1 and the preparation for Phase 2 of this project. |
Quantifying Natural Groundwater Variability as a Function of Time for Establishing a Geochemical Baseline in Pre-Injection Monitoring for CO₂ Sequestration
Anna Ahlstrom Littlefield, Alexis Navarre-Sitchler,Colorado School of Mines, Joel Moore, Towson University Geologic sequestration of CO₂ in deep saline reservoirs is currently the most practical approach for mitigating the release of anthropogenic CO₂ into Earth’s atmosphere. Establishing effective containment of CO₂ in the subsurface is a fundamental criterion for obtaining a Class VI permit for CO₂ injection. Leakage of CO₂ or brine from the injection formation into overlying intervals presents a risk of groundwater contamination. To address this risk, the EPA’s Class VI permitting guidance requires applicants to implement a comprehensive testing and monitoring plan that includes groundwater quality monitoring. The protocols outlined by the EPA are designed to ensure the preservation of groundwater quality in an active CO₂ injection site, by monitoring any deviation from the natural baseline of geochemical parameters such as pH, alkalinity, major anions and cations, and TDS. Operators are required to establish a baseline for these parameters, capturing natural variability prior to injection operations.
Previous work has established that these geochemical parameters have varying degrees of sensitivity to CO₂. Here we address the importance of time in establishing a geochemical baseline for groundwater monitoring. By analyzing existing groundwater datasets from the USGS, EPA, as well as state and local agencies, we will quantify geochemical variability at progressive timesteps and identify statistically significant changes in variance through time. By using temporal data from multiple regions, we will determine the duration of sampling required to capture baseline natural variability for each parameter.
This effort will provide a data driven evaluation of the length of time needed for baseline monitoring to accurately capture natural variability across different groundwater systems. These findings will be applicable not only to the delineation of monitoring timelines but also to CO₂ leakage modelling. These data and analyses will aid in the identification of significant exceedance of natural variability as it relates to a containment breech of a CO₂ sequestration project. |
Utilizing Distributed Acoustic Sensing (DAS) for CO₂ Containment Monitoring
Carson Laing, Martin Karrenbac, Steve Cole, Andres Chavarria, Todd Bown, OptaSense The Oil and Gas industry has seen a substantial increase in utilizing and monitoring on fiber-optic cables with Distributed Acoustic and Temperature Sensing (DAS/DTS) technologies to characterize subsurface geology, monitor well completions and operations, and demonstrate containment and conformance of injected fluids into the reservoir. Over their last several years CCUS applications have focused mostly on pilot studies and small commercial CCS projects in order to understand the monitoring technologies and their limitations. For subsurface monitoring typically a combination of DAS sensing, standard 3C seismic sensors, DTS, strain and pressure measurements are employed. Active source and passive seismic analysis techniques have been used, tested and developed in order to verify plume location, migration paths and subsurface property changes due to CO₂ replacing previously existing pore fluids. In this presentation we describe the background of monitoring and focus on a well-studied and published example, Shell’s Quest CCS Facility in Canada. We highlight the relevant conclusions, best practices and also rely on our own experience in acquiring and analyzing distributed acoustic sensing (DAS) data for the Quest project. We will also speak to considerations for future CCUS projects in terms of various fiber-optic layouts and the integration of several distributed fiber sensing technologies to ensure the conformance and containment of the injected CO₂. |
| Theme 4: Risk Assessment |
Deploying Petroleum Play-Based Exploration Methodologies for Carbon Storage Site Characterization
Katya Casey, Marel Sanchez, Actus Veritas Geoscience; Kevin Schofield, U3 Explore The EPA application process for the injection project permit is an iterative process, requiring the completion of an extensive questionnaire that covers a detailed description of the subsurface, addressing regional and local aspects of the storage container. We propose that the Play-based assessment methodology developed and successfully used in Petroleum exploration during the last two decades could be deployed for the characterization of the risks and uncertainties of the Area of Review (AoR) as defined by Environmental Protection Agency (EPA) for Class 6 well (CCUS) permit approval process.
The first application for this permit took six years to get approved, largely because of “recycling” as the requirements were properly understood. Recently, data show that Class 6 well permit approvals now take about two years. The utilization of play-based assessment methods offers opportunities to further reduce the cycle time of the application and improve the quality of the subsurface characterization.
The Common Critical Risk Segment methodology (CCRS) used in play-based assessment defines the risks and manages the uncertainties in the critical elements used to characterize a hydrocarbon accumulation (Source, Reservoir, and Seal). Well, seismic, and potential fields data are used to reduce the uncertainty in the ranges of the parameters used in the characterization. The same data sources and methodology can be applied to understanding the range in critical parameters required for developing an appropriate injection site…the absence of petroleum fluids, the presence of saline aquifer with sufficient porosity and permeability, and an overlying seal of known capacity.
Reducing uncertainty in subsurface characterization in the permit application process is possible utilizing the consistent approach to data quality and reliability assessment and statistically valid data analysis embedded within petroleum business processes that were developed to deliver a better understanding of the subsurface variability.
CCRS maps may be used to define a sweet spot for injection well positioning determined by the configuration of the most favorable parameters of each critical element. |
Evaluating Fault and Fracture Risks for CCS in the San Juan Basin Using Rock Volatiles Stratigraphy of Cuttings and Cores from Old and New Wells
Michael P Smith, Christopher Smith, Patrick Gordon, Timothy Smith, Advanced Hydrocarbon Stratigraphy; William Ampomah, Luke Martin, New Mexico Institute of Mining and Technology AHS is a key technology partner on the multi-disciplinary NMT-DOE CCS project providing a novel geochemical tool aimed at mitigating seismic risks associated with CCS in the San Juan Basin (SJB).
AHS’s patented gentle multi-pressure-extraction cryogenic-separation mass spectrometry system provides unique insights for fluid migration conduits, seals, and reservoirs.
Cuttings from the CCS target Jurassic section and adjoining strata from the 1961 Kirtland #1 well in the SJB and the 1982 Stephenson #1 well on the Hogback Monocline are analyzed. The Kirtland and Stephenson wells are in near each other but drilled on opposites sides of the very large displacement, multiple 1000’s of feet, Hogback Fault. Core and cuttings from the soon to drill nearby SJB CarbonSAFE#1 well will also be analyzed and included in our presentation.
The Kirtland CO₂ cuttings log shows higher CO₂ below 7200’, and lower CO₂ shallower, suggesting a pressure seal at 7200’ within the Saltwash Member of the Morrison Formation. This finding reduces the risk we assign to vertical loss of CO₂ through the Jurassic section at Kirtland #1..
There is no Jurassic oil or gas production in the NW of the petroliferous SJB, but oil occurs in cuttings throughout the Jurassic section in Kirtland #1 as a series of oil spikes. Previous work by AHS on samples in different basins shows oil spikes in our data indicate oil filled fractures.
Oil migrates through the Jurassic section along these fractures above and below the vertical seal to CO₂ at 7200’. Oil spikes occur in all Jurassic formations in Kirtland-1, including both reservoirs and seals. Vertical oil migration from the base to the top of the Jurassic at Kirtland #1 is unlikely. The horizontal fractures probably tap into a vertical oil migration conduit near Kirtland #1 well, probably the Hogback fault.
The oil filled fractures could increase CO₂ loss risk if the fractures are repurposed for CO₂ migration to the Hogback fault.
This work together with other analysis conducted as part of the SJB CarbonSAFE and the SJB fault characterization projects will assist in siting future CO₂ injection wells within the Basin.
This material is based upon work conducted as part of the San Juan Basin Fault Characterization project, supported by the Department of Energy under Award Number DE-F0032064. Additional support was received from the San Juan Basin CarbonSAFE Phase III project, supported by the Department of Energy under Award Number DE-FE0031890. |
| Theme 5: Case Studies |
Characterization Evolution Based on a CO₂ Huff and Puff of the Siliciclastic Carper Sandstone Greenfield ROZ in the Illinois Basin
Nate Grigsby, Scott Frailey, Nathan Webb, Fang Yang, Illinois State Geological Survey The Carper Sandstone (lower Mississippian) is a siliciclastic greenfield residual oil zone (ROZ) in the Illinois Basin. Conventional Carper oil production occurs on anticlines in the northern portion of the basin, but drilling records indicate widespread oil shows in off-structure areas where the Carper is not productive. This presentation details how new data from a single well CO₂ huff and puff (HnP) and related tests advanced our understanding of the characteristics (rock and fluid properties) of the Carper ROZ.
The test well is in Cumberland County, IL, in an off-structure greenfield ROZ at least 15 miles from the nearest Carper oil production. This well was drilled as a wildcat test of conventional oil production from the Carper, but no oil was produced despite observed oil saturation in core. Well log derived oil saturation was at or below residual oil saturation (~25%) in this and other wells in the greenfield. The well was hydraulically fractured, natural fractures were also suspected based on high water production during the initial oil production attempt that could not be explained by the core derived permeability (0.2 mD).
A HnP and related tests were designed to demonstrate efficacy of CO₂-EOR and to better understand the properties of the ROZ. Pressure response falloff behavior from pre-HnP water injection pressure transient and baseline production tests supported the existence of a natural fracture network. The HnP test injected 1,000 tons of CO₂ and produced 50 bbls of oil and 360 tons of CO₂, storing the remaining 640 tons in the reservoir. Both oil recovery and CO₂ storage exceeded reservoir simulation forecasts based on the pretest assumptions (natural fracture network and low viscosity oil only within the tight matrix), which showed that CO₂ would remain in fractures and have limited contact with matrix bound oil.
Oil samples acquired throughout the test were consistently denser (median 29° API) and more viscous (median 30 cP) compared to published conventional Carper oil density and viscosity data of 36°-39° API and 4-11 cP, respectively, due to attenuation of the light hydrocarbon fraction. Core flood experiments using the 30 cP oil resulted in a residual oil saturation higher than observed in log analyses (~65% compared to ~25%). The core flood also found that CO₂ would mobilize more of the viscous oil than expected from such tight rock.
The HnP and related testing led to previously unknown Carper characteristics that changed the geologic and geocellular models and improved flow modeling results to match HnP observations. The oil composition in the model was revised to match that of the produced oil. Because of the revised oil composition, initial oil saturation was added to the natural fractures and matrix oil saturation was increased. The natural fracture properties were also tuned to observed pressure responses. The resulting model better represents the greenfield Carper with greater utility for expanding simulations to field-scale development. The evolved characterization has implications for improving the regional understanding of the Carper Sandstone greenfield ROZ for CO₂-EOR and associated storage. |
CO₂ Upstream Emission Control: A Proposed Case Study Cooper Basin Australia.
Shane Arick Eiring, Shane S. Namie, Moones Alamooti Australia’s CO₂ emissions yearly report shows 39.6 MMt of CO₂ produced from fossil fuel extraction as of May 2021 (Hanna, 2021). Under the Paris agreement, Australia has committed to reducing greenhouse gas emissions by 26-28% by 2030 (Hanna, 2021). Additionally, as of October 2021, the government has committed $180 MM to develop commercial-scale carbon capture, use, and (CCUS) projects (Lepic, 2021). Therefore, emissions control is critical in the success of developing any future fossil fuel production in Australia. These emissions are categorized into three groups or “Scopes” by the most widely-used international accounting tool, the Greenhouse Gas (GHG) Protocol. These include Scope 1 direct emissions from owned or controlled sources; Scope 2 emissions, the use of outside sources of energy; and Scope 3 emissions, all other including indirect emissions in a company’s value chain which are not included in this study. This research covers an in-depth process for assessment, including the implementation of practices for mitigating emissions throughout the upstream development. The oil and gas field case study consists of a four-way dip closure east of Moomba’s facility by 150 km and 10 km south of the Southwest Queensland Pipeline. The oil and gas production consists of multiple-net pay zones starting at the top of the Toolachee Formation (1,200 m).
The Scope 1 and 2 emissions for the current location assessment through drilling operations on a per well basis is about 529 Tonnes of CO₂. Additionally, all activities correlated to the period of field production over a 15-year life, the case study indicates that CO₂ emissions would be about 9.9 Mt. Therefore, emissions collection for this project considered the use of integrating a carbon capture unit developed by Compact Carbon Capture (Baker Hughes, 2021). Due to the geological subsurface characteristics of the Cooper Basin, the operational requirements for the carbon capture units are viable and economical for optimal operation. Furthermore, the high geothermal gradient in the basin is a strategic advantage to implementing these units in the Cooper Basin (Habermehl, 1982).
The proposed case study location is ideal for developing a working template for future applications related to upstream emission control from well implementation to CCUS processes. Additionally, as energy demand increases, so will petroleum production, which will strain Australia’s CO₂ emissions goals under the Paris Agreement. This case study introduces methods for emissions accounting for all aspects of activities in Exploration & Production (E&P) that include Scope 1 and Scope 2. Once adequate emissions accounting is achieved, mitigating efforts for project-specific activities are introduced and applied. This case study will give a much-needed working model for future exploration and development into achieving near net-zero CO₂ production status in the E&P industry for Australia. |
First Wave of Incentive-Driven CCS Projects in North Dakota
John A. Hamling, James A. Sorensen, Wesley D. Peck, Kevin C. Connors, Bethany A. Kurz, Energy & Environmental Research Center University of North Dakota North Dakota Industrial Commission orders granting the first geologic CO₂ storage facility permits have been issued though the North Dakota State Class VI Primacy Program. Through these orders, the geologic storage of carbon dioxide is addressed, along with the amalgamation of storage reservoir pore space and the determination of financial responsibility. These projects are at the forefront of the first wave of incentive-based, dedicated geologic CO₂ projects to be implemented in the United States.
The North Dakota Administrative Code provides a comprehensive set of statutes overlaid on an Underground Injection Control Class VI Primacy Program that creates a regulatory environment enabling commercial geologic storage of carbon dioxide. Several of the key overlays include amalgamation of pore space, postinjection title transfer and release of liability, establishment of long-term trust funds, and a statutorily defined timeline for issuing a final decision on a storage facility permit application.
These first landmark permits coupled with the prudent regulatory environment and excellent geology and stacked storage potential of North Dakota have resulted in a broad range of commercial carbon capture and storage (CCS) projects being advanced in the region. The dozens of CCS projects lining up include capture from coal-fired power generation, ethanol production, industrial processes and natural gas compression, processing, and generation. Each application is driven by project-specific business cases and varying combinations of incentive programs. Each of the commercial CCS projects being advanced boosts investment in the region and provides a means of significantly reducing the carbon intensity of the energy or products produced.
Despite the range of project types, each follows a similar development arc and timetable that consists of key stages and an integrated adaptive management approach. This staged adaptive approach provides a mechanism that manages and balances a project’s risk tolerance against a financial investment strategy. The implementation strategies being developed for projects in North Dakota, including integration with an array of incentive program compliance requirements, are already being adapted to enable CCS projects throughout the United States.
While the central interior of North America offers a tremendous storage resource, optimization will be key to meaningfully reducing the more than 2.6 billion tons of CO₂ emitted annually from large point sources in the United States. These initial projects tend to have the most favorable economics and are often able to take advantage of the best geology, with little to no competition for resource. This initial wave of projects is informing investment in and creating the opportunity for innovation that will enable the next wave of incentive-based CCS projects. Such innovation is critical for reducing costs, enhancing the design and operability of the next wave of CCS projects, and realizing more of the 250 billion tons of storage resource in North Dakota. |
The Quest Carbon Capture and Storage Operation - Decarbonizing the Oil Sands
Sarah Kassam, Anne Halladay, Simon O'Brien, Robert Liston, Nial Smith, Stephen Harvey, Nathan Robinson Shell The Quest carbon capture and storage (CCS) facility, near Edmonton, Alberta, has been operating for over seven years, demonstrating that that large-scale CO₂ capture is a safe and effective measure to reduce CO₂ emissions from industrial sources in the Canadian Oil Sands.
The purpose of Quest is to deploy technology to capture CO₂ produced at the Scotford Upgrader and to compress, transport, and inject the CO₂ for permanent storage in a saline formation near Thorhild, Alberta. Approximately 1.2 Mt/a of CO₂ is being captured, representing greater than 35% of the CO₂ produced from the Scotford upgrader.
Since opening in late 2015, the facility has capture and stored over 7 million tonnes of CO₂ safely underground in a world-class saline aquifer.
Reservoir performance and injectivity assessments thus far indicate that the project will be capable of sustaining adequate injectivity for the duration of the project life; therefore, no further well development should be required. MMV activities are focused on operational monitoring and optimization and MMV data indicate that no CO₂ has migrated outside of the Basal Cambrian Sands(BCS) injection reservoir to date.
Highlights from these seven years of operations include:
-Sustained, safe, and reliable operations.
-Execution of first turn-around at the Quest capture unit in 2021
-Strong integrated project reliability performance with operational availability at 98.8%.
-First successful halite remediation of a CCS injection well
-Continued cost and operational optimization of MMV activities since start-up
This presentation will deliver a comprehensive overview of the seven years of operations at the Quest CCS facility, including some key lessons learned in maintenance and integrated operations. |
| Theme 6: Economics |
A Probabilistic Framework for Integrated Design and Economic Modeling of CO₂ Hubs and CCUS Projects
Mohammad Evazi, ESGWAY Corporation Technical and economic uncertainty is inherent in the design, evaluation and economic modeling of large scale Carbon Capture, Use and Storage (CCUS) projects. Understanding and quantifying the uncertainty of project cost, revenue and storage performance is critical for optimum design and success of these capital-intensive projects. This is of utmost importance in large scale CCUS projects, such as CO₂ hubs, to ensure continuous success of the technology from technical, impact and economic perspective to gain confidence of general public, finance, government etc.
In this work, an integrated economic modeling and forecasting framework is developed for design, evaluation and optimization of CCUS projects. This framework allows deterministic and probabilistic modeling of cost of various components of a CCUS project, performance of storage reservoir system and revenue from carbon storage incentives and additional recovery. Sensitivity analysis workflow is developed to identify key performance indicators and leverage in taking actions for uncertainty reduction and lowering project risk.
Comprehensive cost modules are developed for modeling of operating & capital cost of CO₂ capture, compression, transport and injection for storage and enhanced recovery. As project revenue is driven by incentives from CO₂ storage and sale of additional hydrocarbon recovery, various tax credit and incentives (such as 45Q, LCFS etc.) and additional recovery is probabilistically modeled. Physics-based simulation of CO₂ injection performance is used to quantify variability in CO₂ injection (storage) rate. Dimensionless curve approach is incorporated to generate CO₂ EOR performance and quantify project performance.
We use the developed framework for design and evaluation of carbon capture and storage in a gas-condensate reservoir. We demonstrate the impact of various uncertainties on the economics of the project and identify key performance indicators and input variables that control the project performance. This workflow provides a systematic approach to guide in identifying the additional evaluations and data gathering that will assist in reducing project uncertainties.
In addition, we use the developed framework to design and optimize a hypothetical CO₂ infrastructure (hub). Multiple CO₂ emission sources, compression, pipeline and storage reservoirs are defined and incorporated in the project. The developed framework is used to design multiple project scenarios and investigate project performance under various uncertainty scenarios. We demonstrate that project performance is controlled by the project input distributions and pinpoint that inaccurate quantification of input data range and distribution could drastically impact the success of project. This project simulation further demonstrates that project design optimization under uncertainties is critical for successful early-stage decision making and ensuring sustained and impactful deployment of CCS projects. |
| Theme 7: Non-Technical Considerations |
Environmental Justice Considerations for CCS projects
Ramon A. Gil-Egui, Susan D. Hovorka, University of Texas at Austin It has long been recognized that public engagement in obtaining license to operate is essential for a successful Carbon Capture and Storage (CCS). Recently this task has added a specific focus on Environmental Justice (EJ) as defined by EPA’s environmental justice screening tool’s guidance (EJScreen tool). Three components of projects may impact different or partly overlapping groups of residents: capture at a point source, transportation via pipeline, and residents over and nearby the storage site or Area of Review (AOR), defined by the extension of the CO₂ plume(s) and the elevated pressure area.
Assessment of two hypothetical projects in the Gulf Coast (onshore and offshore) is being conducted to learn about the application of the EPA EJ tool with a focus on storage sites. Storage sites are located where pore space rights can be leased or easements acquired; the cost of negotiating access drives projects to areas with few owners, which are predominantly in rural areas or marine settings. The EJ tool is based on census block groups and, in our cases are biased toward population centers that are likely only impacted indirectly by the project; therefore, we see the need to carry out more specific and granular evaluations of the site to identify population, the potential positive and negative impacts of the storage project on them, and determine optimal ways to communicate. In addition, it is possible that a significant proportion of human connections may be to workers active in rural areas or even fishermen. The EJScreen tool combines the environmental hazards exposure (Environmental Indicator) with a Demographic Index (socio-economic average indicators) as a measure of community vulnerability. The weighted combination of both results in an EJ Index that allows comparing selected community’s EJ situation with larger geographic areas (state, region, or nation) to prioritize actions.
EPA’s underground injection control (UIC) Class VI rules seem to have good potential to harmonize with EJ needs built into the rules both in terms of requiring protection from health hazards and groundwater protection and via a requirement for consultancy. Preliminary results show that the communities within the test AORs have relatively low exposure to environmental hazards but high vulnerability (high percentage of minorities and low-income among others), revealing an EJ community presence and requiring targeted actions. |
Hybrid Approaches for High-Performance Scientific Teams
Jorge Luis Barrios, Batteelle Accelerated changes in the geopolitical and market environment in the energy sector, which are due to different kinds of factors, lead organizations to develop accelerated transformation processes, in order to be more adaptive and flexible in order to respond with assertiveness, generating mechanisms that increase organizational efficiency. Faced with this reality, companies in the energy sector and governments around the world are inserting themselves in the trends of change to adopt new energy approaches in a more environmentally friendly way, as key factors in their value chains, with the intention of being directed to achieve efficiency, productivity, and permanence over time.
The complexity of managing projects within the Triple constraints (normal or inverted) has been increasing day by day. Quick browsing through the changes over the past decade tells us how fast the “Complexity” of managing projects has been changing. Various factors contributed and will continue for the next decade, specifically CCUS projects face a dilemma in their geoscientific teams, which is the best and most effective approach to approach the development of their projects.
When we talk about complexity, it is important to have theoretical abstractions embedded in complexity, which for this presentation are supported by the proposals of authors such as Morín (1986), Nonaka and Takeuchi (1995), Martínez (1997), Maldonado (2003), Godet ( 2007), Lobo (2011), Zooback (2014), Bertalanffy (1968), Bautista (2016) among others. Methodologically we are framed in the complex paradigm.
The objective of this methodology was to propose the orientations and strategic actions based on the competencies of the companies based on the scenarios of its general and competitive environment.
As many scientists know, a research project often changes from the original proposal, (because you don't know at the beginning if your theory is right or how many trials you will attend to get the success) and it is generally messier than following a series of sequential tasks. This presents a problem: the waterfall methodology is often not flexible enough to be applied to academic research.
By contrast, the philosophy underpinning agile project management prioritizes flexibility.
From the analysis of the perceptions received, it is concluded that the organization must be open to the emergence of the new approach to handle projects related to the characterization of geological models and Carbon Capture Utilization and Storage (CCUS) operations, conceived from a complex approach, which allows broadening the vision of a way to provide flexibility and the ability to act in anticipation in a highly changing environment. |
U.S. EPA Class VI Injection Permit- Understanding Regulatory Requirements for Rapid Approval
Tiraz Birdie, TBirdie Consulting, Inc; Eugene Holubnyak, Jennifer Raney, Lynn Watney, Kansas Geological Survey The U.S. EPA Class VI Carbon-dioxide geologic sequestration permit application process requires significant investment of human and financial resources. Salient features of the permit application contained in the online Geologic Sequestration Data Tool are presented and lessons learned from previous permitting experience shared with examples to expedite the permit review process.
Key challenges include detailed regional and local site characterization, delineating the CO₂- plume and pressure-based Areas of Review, addressing seismic risk due to injection, developing robust but cost-effective testing/monitoring and emergency/remedial response plans, reducing the default post-injection site care period of 50 years, and lowering the multi million-dollar financial assurance obligation.
Briefly discussed will be technologies acceptable to EPA for periodic monitoring of CO₂ plume and pressures in the injection zone, methods for estimating potential for fault slippage, establishing a seismic monitoring network that ties into a customized Seismic Action Plan for ensuring safe injection, methods for mechanical integrity testing of the injection well, developing a per-operational testing plan involving coring, logging, and formation testing to fulfill data deficiencies, EPA’s quality assurance and surveillance plan for QA/QC, and development of an operating plan for safe and efficient injection. Also briefly discussed are monitoring and reporting requirements to avail federal sequestration credits or Direct-Pay. |
| Theme 8: Infrastructure |
Accounting for Sensitive Areas When Siting CCUS Infrastructure
Jared Hawkins, Battelle Memorial Institute Sensitive areas not only present technical obstacles to CCUS development and operations but can also cause public opposition. This analysis considers Sensitive Areas as tracts of land protected by statute or that pose a significant issue for public acceptance of a project. Three strategies for sensitive land designations were developed: Avoid, Limit, and Informational. The analysis of whether the legal or public acceptance issues preclude (avoid), impede (limit), or affect (informational) a project will be applied to the Midwest Regional Carbon Initiative (MRCI) study area using mapping and publicly available data. The interplay between our analysis and other considerations, such as Pipeline and Hazardous Materials Safety Administration (PHMSA) high consequence areas (HCAs) for pipeline operations, will also be discussed. |
Core and Cuttings Repository Networks for Initiating CCUS Projects
Tim Prather, Michael Nieto, Sean Arrington, Harry Rowe, Premier Oilfield Group As oil and gas operators continue to purge their rock collections, through the dumpster or through donation, we should pause to understand how simple the decision to save core and cuttings should be. These materials were expensive to collect and analyze, they provide tremendous insight, and they (and their associated data/interpretations) should be treated in a fashion akin to how NASA treats Moon rocks.
At this stage in our understanding of how CCUS techniques will be deployed, it seems clear that CO₂ will be injected into highly characterized formations, with highly characterized caprocks, in highly characterized reservoirs that, in large part, were initially targets for oil and gas exploitation. In essence, much of the initial work has been or can be accomplished with existing materials, as long as they are properly curated.
This presentation will focus on outlining 1) what makes a modern core repository, 2) how best a core repository can serve its CCUS clientele, and 3) what future developments are required to optimize repository longevity, connectivity, and usability. |