Nicholas A. Azzolina
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Nicholas A. Azzolina has a total of 82 co-authors. Here are the top co-authors by number of shared publications:
Charles D. Gorecki
Univ. of North Dakota, Grand Forks, ND (United States)
4 shared publicationsJames A. Sorensen
4 shared publicationsAmanda J. Livers-Douglas
Univ. of North Dakota, Grand Forks, ND (United States)
3 shared publications- Bethany Kurz
Univ. of North Dakota, Grand Forks, ND (United States). Energy & Environmental Research Center
3 shared publications David V. Nakles
Univ. of North Dakota, Grand Forks, ND (United States)
3 shared publications- Trevor L. Richards
Univ. of North Dakota, Grand Forks, ND (United States)
3 shared publications
Latest Publications
- 30 January 2026
- Published by Office of Scientific and Technical Information (OSTI)
Williston basin associated co2 storage field laboratory
The Williston Basin Associated CO2 Storage Field Laboratory (WBCFL), led by the Energy & Environmental Research Center (EERC) in partnership with Denbury (now ExxonMobil), was established to advance understanding of CO2 enhanced oil recovery (EOR) and associated geologic CO2 storage in the Williston Basin. The specific objective of the WBCFL was to conduct field and laboratory activities to inform decisionmakers regarding the 1) efficacy of associated storage in stacked geologic reservoirs during EOR operations; 2) CO2 injectivity, oil production, and storage in residual oil zones (ROZs); 3) use of a novel CO2 storage monitoring, verification, and accounting (MVA) technology (active seismoelectric [ASE] monitoring), and 4) life cycle and techno-economic factors for EOR development scenarios targeting ROZs. The project location was the Pennel Field in southeast Montana, which is part of the Cedar Creek Anticline (CCA) geologic structure within the Williston Basin. The field was selected because it was being developed by Denbury for tertiary-phase CO2 EOR targeting the Red River Formation, and the availability of CO2, which was being supplied by the Greencore Pipeline, offered opportunities to evaluate the EOR potential of the other oil-bearing formations within the structure. The focus of this project was a vertically stacked reservoir system, including the Interlake Formation, a complex carbonate interval with both main pay zones (MPZs) and underlying ROZs. ROZs are oil-bearing zones that have been naturally waterflooded and, while they often contain moderate oil saturations, they cannot be economically produced using conventional primary or secondary recovery. However, CO2 EOR of ROZs in the Permian Basin have been successful, as the injected CO2 can mobilize oil trapped in the reservoir by swelling the oil, reducing oil viscosity, and achieving miscibility under reservoir conditions. The initial project activities focused on data collection and preparation of the field demonstration within the Pennel Field. Denbury drilled, logged, and cored a new well, 23-18B, as a CO2 injector paired with a nearby production well, 13-18B. Samples were collected from the field location, including rock, brine, and oil samples, for subsequent analysis at the EERC and to support the laboratory testing, geomodelling, and reservoir simulation components of the project. Comprehensive laboratory analyses were performed to characterize produced oil and brine, evaluate CO2–oil miscibility, and measure interfacial tension and contact angles at reservoir temperature and pressure. Results showed that Interlake oils achieve miscibility with CO2 at approximately 2600–2800 psi, indicating potential suitability for CO2 EOR. Brine and carbonatedbrine testing highlighted modest but measurable changes in interfacial properties, while contact angle measurements indicated predominantly water-wet rock—an important condition that can strongly influence CO2 storage and oil displacement behavior. These laboratory results were integrated into a geologic model of the Pennel Field, constructed from well logs, core data, structural horizons, and petrophysical properties. The Interlake Formation model was subdivided into six depositional cycles that capture heterogeneity between the MPZ and ROZ intervals. A 1-sq-mi dynamic reservoir simulation model was vii developed from this geologic framework, incorporating five producing wells and the drilled injection well. Historical well performance, including more than five decades of production, was used to history-match reservoir behavior and calibrate the simulation model. Using the calibrated simulation model, the EERC evaluated three CO2 injection strategies: continuous CO2 injection, a 1:1 water alternating gas (WAG) scheme, and a tapered WAG schedule designed to reduce early gas breakthrough. Across all strategies, CO2 injection increased oil production relative to primary depletion, including meaningful contributions from the ROZ. The tapered WAG strategy produced the highest incremental oil recovery and the most efficient CO2 utilization, aided by improved reservoir pressure support and reduced early gas migration toward producers. Notably, Well 22-18B, located farther from the injector and historically more productive, significantly outperformed the closer 13-18B, both in projected incremental oil and in CO2 storage amount. Simulations suggested that developing CO2 EOR in the 1-sq-mi area of interest in the Interlake MPZ/ROZ system could yield 34,000–54,000 barrels of incremental oil at 13-18B and 181,000–217,000 barrels at 22-18B over the 5- and 6-year modeled injection periods. Associated CO2 storage ranged from 80,000 to over 190,000 tonnes, depending on the case, with WAG strategies storing a larger fraction of the injected CO2. These high storage rates reflect the strong retention characteristics of carbonate ROZs and their compatibility with miscible CO2 flooding. Dimensionless curve analysis was used to extend these simulation results to 15 years, enabling long-term projections for economic and life cycle analysis (LCA). A screening-level LCA and techno-economic analysis evaluated the environmental and economic impacts of each EOR scenario, integrating simulation-based forecasts of oil production, recycled CO2, and long-term storage. Results emphasized that CO2 EOR—particularly tapered WAG—can provide substantial net storage while increasing oil recovery, producing favorable CO₂ storage per barrel, and strengthening alignment with federal incentives such as the Section 45Q tax credit. The stacked-storage development concept, targeting both MPZ and ROZ horizons together, offers a pathway for maximizing combined resource recovery and geologic CO2 containment at legacy oil fields. To explore the feasibility of cost-effective monitoring strategies, the project deployed a baseline active seismoelectric survey—a method...
- 30 January 2026
- Published by Office of Scientific and Technical Information (OSTI)
Williston basin associated co2 storage field laboratory
The Williston Basin Associated CO2 Storage Field Laboratory (WBCFL), led by the Energy & Environmental Research Center (EERC) in partnership with Denbury (now ExxonMobil), was established to advance understanding of CO2 enhanced oil recovery (EOR) and associated geologic CO2 storage in the Williston Basin. The specific objective of the WBCFL was to conduct field and laboratory activities to inform decisionmakers regarding the 1) efficacy of associated storage in stacked geologic reservoirs during EOR operations; 2) CO2 injectivity, oil production, and storage in residual oil zones (ROZs); 3) use of a novel CO2 storage monitoring, verification, and accounting (MVA) technology (active seismoelectric [ASE] monitoring), and 4) life cycle and techno-economic factors for EOR development scenarios targeting ROZs. The project location was the Pennel Field in southeast Montana, which is part of the Cedar Creek Anticline (CCA) geologic structure within the Williston Basin. The field was selected because it was being developed by Denbury for tertiary-phase CO2 EOR targeting the Red River Formation, and the availability of CO2, which was being supplied by the Greencore Pipeline, offered opportunities to evaluate the EOR potential of the other oil-bearing formations within the structure. The focus of this project was a vertically stacked reservoir system, including the Interlake Formation, a complex carbonate interval with both main pay zones (MPZs) and underlying ROZs. ROZs are oil-bearing zones that have been naturally waterflooded and, while they often contain moderate oil saturations, they cannot be economically produced using conventional primary or secondary recovery. However, CO2 EOR of ROZs in the Permian Basin have been successful, as the injected CO2 can mobilize oil trapped in the reservoir by swelling the oil, reducing oil viscosity, and achieving miscibility under reservoir conditions. The initial project activities focused on data collection and preparation of the field demonstration within the Pennel Field. Denbury drilled, logged, and cored a new well, 23-18B, as a CO2 injector paired with a nearby production well, 13-18B. Samples were collected from the field location, including rock, brine, and oil samples, for subsequent analysis at the EERC and to support the laboratory testing, geomodelling, and reservoir simulation components of the project. Comprehensive laboratory analyses were performed to characterize produced oil and brine, evaluate CO2–oil miscibility, and measure interfacial tension and contact angles at reservoir temperature and pressure. Results showed that Interlake oils achieve miscibility with CO2 at approximately 2600–2800 psi, indicating potential suitability for CO2 EOR. Brine and carbonatedbrine testing highlighted modest but measurable changes in interfacial properties, while contact angle measurements indicated predominantly water-wet rock—an important condition that can strongly influence CO2 storage and oil displacement behavior. These laboratory results were integrated into a geologic model of the Pennel Field, constructed from well logs, core data, structural horizons, and petrophysical properties. The Interlake Formation model was subdivided into six depositional cycles that capture heterogeneity between the MPZ and ROZ intervals. A 1-sq-mi dynamic reservoir simulation model was vii developed from this geologic framework, incorporating five producing wells and the drilled injection well. Historical well performance, including more than five decades of production, was used to history-match reservoir behavior and calibrate the simulation model. Using the calibrated simulation model, the EERC evaluated three CO2 injection strategies: continuous CO2 injection, a 1:1 water alternating gas (WAG) scheme, and a tapered WAG schedule designed to reduce early gas breakthrough. Across all strategies, CO2 injection increased oil production relative to primary depletion, including meaningful contributions from the ROZ. The tapered WAG strategy produced the highest incremental oil recovery and the most efficient CO2 utilization, aided by improved reservoir pressure support and reduced early gas migration toward producers. Notably, Well 22-18B, located farther from the injector and historically more productive, significantly outperformed the closer 13-18B, both in projected incremental oil and in CO2 storage amount. Simulations suggested that developing CO2 EOR in the 1-sq-mi area of interest in the Interlake MPZ/ROZ system could yield 34,000–54,000 barrels of incremental oil at 13-18B and 181,000–217,000 barrels at 22-18B over the 5- and 6-year modeled injection periods. Associated CO2 storage ranged from 80,000 to over 190,000 tonnes, depending on the case, with WAG strategies storing a larger fraction of the injected CO2. These high storage rates reflect the strong retention characteristics of carbonate ROZs and their compatibility with miscible CO2 flooding. Dimensionless curve analysis was used to extend these simulation results to 15 years, enabling long-term projections for economic and life cycle analysis (LCA). A screening-level LCA and techno-economic analysis evaluated the environmental and economic impacts of each EOR scenario, integrating simulation-based forecasts of oil production, recycled CO2, and long-term storage. Results emphasized that CO2 EOR—particularly tapered WAG—can provide substantial net storage while increasing oil recovery, producing favorable CO₂ storage per barrel, and strengthening alignment with federal incentives such as the Section 45Q tax credit. The stacked-storage development concept, targeting both MPZ and ROZ horizons together, offers a pathway for maximizing combined resource recovery and geologic CO2 containment at legacy oil fields. To explore the feasibility of cost-effective monitoring strategies, the project deployed a baseline active seismoelectric survey—a method...
- Open Access
- 26 November 2025
- Published by Office of Scientific and Technical Information (OSTI)
Findings on subtask 1.6 – basin electric carbon storage research project: novel monitoring techniques
The Energy & Environmental Research Center (EERC) led a study to validate novel and emerging technologies as commercial monitoring techniques for application in carbon dioxide (CO2) injection operations. This applied research was conducted at Basin Electric Power Cooperative’s (Basin Electric’s) active CO2-injection operations in Beulah, North Dakota, in two phases. The EERC previously completed a set of baseline (preinjection) activities in Phase 1, which included 1) design of an automated, integrated, modular (AIM) monitoring station; 2) time-lapse electromagnetic (EM) field surveys; 3) drone-based surveillance studies; 4) time-lapse monitoring with seismic methods; and 5) advanced wellbore-monitoring methods. This report focuses on key accomplishments and major findings for the Phase 2 follow-on operational monitoring activities, including 6) deployment of an AIM monitoring network, 7) EM monitoring of CO2 with real-time data processing, 8) continued seasonal drone-based surveillance studies, 9) passive and active seismic surveys, and 10) wellbore monitoring with nuclear magnetic resonance (NMR) for near-surface characterization. All CO2-injection operations, including enhanced oil recovery, acid gas, or dedicated storage, must be permitted under the Underground Injection Control (UIC) program and are subject to monitoring requirements to ensure containment and that underground sources of drinking water are protected. Current state-of-the-art monitoring techniques often require many labor-intensive hours dedicated to the acquisition of data and typically rely on commercial facilities to process raw data from the field. The next generation of monitoring techniques should be cost-effective, have a lower impact, generate more continuous and autonomous data streams, and perform faster in terms of time to acquire and/or process the data. Monitoring data collected during Phase 2 have demonstrated the commercial readiness of multiple novel monitoring techniques that comply with the most rigorous UIC requirements for application across all kinds of CO2-injection projects, and learnings facilitated by this research have directly benefited other CO2-injection projects. Major accomplishments in Phase 2 include the following: • Three AIM stations were deployed, demonstrating a low-impact method for acquiring monitoring data efficiently. In addition, an automated soil gas-sampling system and real-time dashboard for processing and visualizing soil gas results and attributing source through a decision-based framework were successfully tested at significant cost savings relative to traditional manual soil gas-sampling methods. • A 3D tTEM survey and a traditional 2D ERT survey to characterize near-surface conditions were collected for comparison. tTEM resulted in much lower impact and faster acquisition and yielded high-quality results in three dimensions at significant cost savings. The interpretation of resistivity structure in the shallow subsurface was supported with well-based NMR and lithology logs. • A time-lapse CSEM survey was acquired using metal-cased groundwater wells to transmit current deep into the earth and image CO2 injection into the reservoir. A spCSEM response was observed and evaluated in time-lapse, which uniquely captured the interplay of the reservoir’s connectivity and response to pressure and flow. This was the first time this method used multiple stages of toggling CO2 injection between two wells to understand well interaction. • Three drone missions were repeated with photogrammetry, lidar, thermal, and multispectral to support geophysical survey work and characterize seasonal changes of vegetation to monitor surface conditions. In addition, a novel workflow using multispectral imagery was developed for optimizing survey acquisitions to ensure repeatability between surveys by accounting for seasonal effects. • A passive seismic survey was repeated and imaged the reservoir and Precambrian basement. • Two repeat active seismic surveys with low-energy, low-impact sourcing with AWD were acquired and showed a time-lapse response for imaging the injected CO2 in the reservoir at significant cost savings relative to traditional seismic acquisition methods. Additionally, compressive sensing decimation and reconstruction was successfully implemented to image the CO2 injection. The EERC has successfully completed all activities and achieved all milestones within the performance period of Subtask 1.6. Of the techniques tested, the AIM concept, spCSEM, tTEM, and AWD are noted for their novelty, acquisition efficiency, cost-effectiveness, low impact, and compliance with UIC requirements. This subtask was funded through the EERC–U.S. Department of Energy Joint Program on Research and Development for Fossil Energy-Related Resources Cooperative Agreement No. DE-FE0024233. The EERC also recognizes its project partners—Basin Electric Power Cooperative, Dakota Gasification Company, Carbon Vault Holdings LLC, and Coteau Properties Company—for their significant contributions to the project, including but not limited to providing access to the site, sharing commercial data for comparison to research results, and supporting fieldwork with their staff.
- 26 November 2025
- Published by Office of Scientific and Technical Information (OSTI)
Open AccessFindings on subtask 1.6 – basin electric carbon storage research project: novel monitoring techniques
The Energy & Environmental Research Center (EERC) led a study to validate novel and emerging technologies as commercial monitoring techniques for application in carbon dioxide (CO2) injection operations. This applied research was conducted at Basin Electric Power Cooperative’s (Basin Electric’s) active CO2-injection operations in Beulah, North Dakota, in two phases. The EERC previously completed a set of baseline (preinjection) activities in Phase 1, which included 1) design of an automated, integrated, modular (AIM) monitoring station; 2) time-lapse electromagnetic (EM) field surveys; 3) drone-based surveillance studies; 4) time-lapse monitoring with seismic methods; and 5) advanced wellbore-monitoring methods. This report focuses on key accomplishments and major findings for the Phase 2 follow-on operational monitoring activities, including 6) deployment of an AIM monitoring network, 7) EM monitoring of CO2 with real-time data processing, 8) continued seasonal drone-based surveillance studies, 9) passive and active seismic surveys, and 10) wellbore monitoring with nuclear magnetic resonance (NMR) for near-surface characterization. All CO2-injection operations, including enhanced oil recovery, acid gas, or dedicated storage, must be permitted under the Underground Injection Control (UIC) program and are subject to monitoring requirements to ensure containment and that underground sources of drinking water are protected. Current state-of-the-art monitoring techniques often require many labor-intensive hours dedicated to the acquisition of data and typically rely on commercial facilities to process raw data from the field. The next generation of monitoring techniques should be cost-effective, have a lower impact, generate more continuous and autonomous data streams, and perform faster in terms of time to acquire and/or process the data. Monitoring data collected during Phase 2 have demonstrated the commercial readiness of multiple novel monitoring techniques that comply with the most rigorous UIC requirements for application across all kinds of CO2-injection projects, and learnings facilitated by this research have directly benefited other CO2-injection projects. Major accomplishments in Phase 2 include the following: • Three AIM stations were deployed, demonstrating a low-impact method for acquiring monitoring data efficiently. In addition, an automated soil gas-sampling system and real-time dashboard for processing and visualizing soil gas results and attributing source through a decision-based framework were successfully tested at significant cost savings relative to traditional manual soil gas-sampling methods. • A 3D tTEM survey and a traditional 2D ERT survey to characterize near-surface conditions were collected for comparison. tTEM resulted in much lower impact and faster acquisition and yielded high-quality results in three dimensions at significant cost savings. The interpretation of resistivity structure in the shallow subsurface was supported with well-based NMR and lithology logs. • A time-lapse CSEM survey was acquired using metal-cased groundwater wells to transmit current deep into the earth and image CO2 injection into the reservoir. A spCSEM response was observed and evaluated in time-lapse, which uniquely captured the interplay of the reservoir’s connectivity and response to pressure and flow. This was the first time this method used multiple stages of toggling CO2 injection between two wells to understand well interaction. • Three drone missions were repeated with photogrammetry, lidar, thermal, and multispectral to support geophysical survey work and characterize seasonal changes of vegetation to monitor surface conditions. In addition, a novel workflow using multispectral imagery was developed for optimizing survey acquisitions to ensure repeatability between surveys by accounting for seasonal effects. • A passive seismic survey was repeated and imaged the reservoir and Precambrian basement. • Two repeat active seismic surveys with low-energy, low-impact sourcing with AWD were acquired and showed a time-lapse response for imaging the injected CO2 in the reservoir at significant cost savings relative to traditional seismic acquisition methods. Additionally, compressive sensing decimation and reconstruction was successfully implemented to image the CO2 injection. The EERC has successfully completed all activities and achieved all milestones within the performance period of Subtask 1.6. Of the techniques tested, the AIM concept, spCSEM, tTEM, and AWD are noted for their novelty, acquisition efficiency, cost-effectiveness, low impact, and compliance with UIC requirements. This subtask was funded through the EERC–U.S. Department of Energy Joint Program on Research and Development for Fossil Energy-Related Resources Cooperative Agreement No. DE-FE0024233. The EERC also recognizes its project partners—Basin Electric Power Cooperative, Dakota Gasification Company, Carbon Vault Holdings LLC, and Coteau Properties Company—for their significant contributions to the project, including but not limited to providing access to the site, sharing commercial data for comparison to research results, and supporting fieldwork with their staff.
- Open Access
- 22 August 2025
- Published by Elsevier in International Journal of Greenhouse Gas Control
- Vol. 146
Quantifying CO2 plume stabilization at carbon storage projects, North Dakota, USA
- 22 August 2025
- Published by Elsevier in International Journal of Greenhouse Gas Control
- Vol. 146
Open AccessQuantifying CO2 plume stabilization at carbon storage projects, North Dakota, USA
- Open Access
- 1 October 2023
- Published by KEAI Publishing in Petroleum Science
- Vol. 20 (5) , 2921-2938
Optimization of operational strategies for rich gas enhanced oil recovery based on a pilot test in the Bakken tight oil reservoir
- 1 October 2023
- Published by KEAI Publishing in Petroleum Science
- Vol. 20 (5) , 2921-2938
Open AccessOptimization of operational strategies for rich gas enhanced oil recovery based on a pilot test in the Bakken tight oil reservoir
- Open Access
- 4 June 2023
- Published by Office of Scientific and Technical Information (OSTI)
Screening Assessment of Prospective CO2 Storage Resource Potential in Saline Carbonate Systems, Central North Dakota Region
Conference presentation at 2023 Rocky Mountain Section-American Association of Petroleum Geologists (RMS-AAPG) Annual Meeting, Bismarck, ND, June 4–6, 2023. As part of the U.S. Department of Energy-funded North Dakota CarbonSAFE Phase III effort, a screening-level assessment was conducted over central North Dakota, including McLean, Mercer, and Morton Counties, to determine if viable CO2 storage resource potential exists within carbonate saline reservoirs. Published literature and data from 173 wells were used in the study. The study focused on identifying carbonate saline reservoirs and suitable confining zones for CO2 storage, with emphasis on three key questions: 1) are the carbonate saline reservoirs viable for CO2 storage? 2) what is their range of storage resource potential? and 3) what are the main elements driving uncertainty and risk of potential capacity and containment?
- 4 June 2023
- Published by Office of Scientific and Technical Information (OSTI)
Open AccessScreening Assessment of Prospective CO2 Storage Resource Potential in Saline Carbonate Systems, Central North Dakota Region
Conference presentation at 2023 Rocky Mountain Section-American Association of Petroleum Geologists (RMS-AAPG) Annual Meeting, Bismarck, ND, June 4–6, 2023. As part of the U.S. Department of Energy-funded North Dakota CarbonSAFE Phase III effort, a screening-level assessment was conducted over central North Dakota, including McLean, Mercer, and Morton Counties, to determine if viable CO2 storage resource potential exists within carbonate saline reservoirs. Published literature and data from 173 wells were used in the study. The study focused on identifying carbonate saline reservoirs and suitable confining zones for CO2 storage, with emphasis on three key questions: 1) are the carbonate saline reservoirs viable for CO2 storage? 2) what is their range of storage resource potential? and 3) what are the main elements driving uncertainty and risk of potential capacity and containment?
- 25 April 2023
- Published by Office of Scientific and Technical Information (OSTI)
Risk Management for CO2 Storage Projects
- 25 April 2023
- Published by Office of Scientific and Technical Information (OSTI)
Risk Management for CO2 Storage Projects
- 28 March 2022
- Published by Office of Scientific and Technical Information (OSTI)
Balancing CO2 Pipeline Infrastructure Challenges
- 28 March 2022
- Published by Office of Scientific and Technical Information (OSTI)
Balancing CO2 Pipeline Infrastructure Challenges
- Open Access
Deep Learning-Accelerated 3D Carbon Storage Reservoir Pressure Forecasting Based on Data Assimilation Using Surface Displacement from InSAR
Fast forecasting of reservoir pressure distribution in geologic carbon storage (GCS) by assimilating monitoring data is a challenging problem. Due to high drilling cost, GCS projects usually have spatially sparse measurements from wells, leading to high uncertainties in reservoir pressure prediction. To address this challenge, we propose to use low-cost Interferometric Synthetic-Aperture Radar (InSAR) data as monitoring data to infer reservoir pressure build up. We develop a deep learning-accelerated workflow to assimilate surface displacement maps interpreted from InSAR and to forecast dynamic reservoir pressure. Employing an Ensemble Smoother Multiple Data Assimilation (ES-MDA) framework, the workflow updates three-dimensional (3D) geologic properties and predicts reservoir pressure with quantified uncertainties. We use a synthetic commercial-scale GCS model with bimodally distributed permeability and porosity to demonstrate the efficacy of the workflow. A two-step CNN-PCA approach is employed to parameterize the bimodal fields. The computational efficiency of the workflow is boosted by two residual U-Net based surrogate models for surface displacement and reservoir pressure predictions, respectively. The workflow can complete data assimilation and reservoir pressure forecasting in half an hour on a personal computer.
Open AccessDeep Learning-Accelerated 3D Carbon Storage Reservoir Pressure Forecasting Based on Data Assimilation Using Surface Displacement from InSAR
Fast forecasting of reservoir pressure distribution in geologic carbon storage (GCS) by assimilating monitoring data is a challenging problem. Due to high drilling cost, GCS projects usually have spatially sparse measurements from wells, leading to high uncertainties in reservoir pressure prediction. To address this challenge, we propose to use low-cost Interferometric Synthetic-Aperture Radar (InSAR) data as monitoring data to infer reservoir pressure build up. We develop a deep learning-accelerated workflow to assimilate surface displacement maps interpreted from InSAR and to forecast dynamic reservoir pressure. Employing an Ensemble Smoother Multiple Data Assimilation (ES-MDA) framework, the workflow updates three-dimensional (3D) geologic properties and predicts reservoir pressure with quantified uncertainties. We use a synthetic commercial-scale GCS model with bimodally distributed permeability and porosity to demonstrate the efficacy of the workflow. A two-step CNN-PCA approach is employed to parameterize the bimodal fields. The computational efficiency of the workflow is boosted by two residual U-Net based surrogate models for surface displacement and reservoir pressure predictions, respectively. The workflow can complete data assimilation and reservoir pressure forecasting in half an hour on a personal computer.