8/22/2016

1

Midwest Regional Carbon Sequestration PartnershipDOE/NETL Cooperative Agreement # DE-FC26-0NT42589

Neeraj Gupta, Battelle (gupta@battelle.org)

Mastering the Subsurface Through Technology Innovation and Collaboration: Carbon Storage and Oil and Natural Gas Technologies Review Meeting

August 16-18, 2016

Outline

• Project Overview

• Technical Status− Injection Test

− Modeling

− Monitoring

− Regional Assessment

− Outreach

• Accomplishments

• Synergy Opportunities

• Project Summary

2

This presentation will provide a technical summary

8/22/2016

2

Benefit to the Program

3

MRCSP supports DOE Program Goals

DOE Program Goal MRCSP Approach/Benefit

Predict CO2 storage capacity in geologic formations to within ±30%

Correlate geologic characterization and reservoir models with monitoring and regional mapping.

Demonstrate that 99% ofCO2 remains in the injection zones

Account for CO2 during EOR operations

Assess monitoring options for tracking and imaging CO2 plume, storage and retention

Improve reservoir storage efficiency while ensuring containment effectiveness

Test in EOR fields in various life cycle stages and examine strategies for utilizing the pore space created by the oil and water production

Development of Best Practices Manuals (BPMs)

Contribute to BPMs through large-scale test and regional analysis across MRCSP

RCSP Goal MRCSP Approach and Success Criteria

Goal 1 – Prove Adequate Injectivity and Available Capacity

• Injecting 1 million metric tons of CO2 in CO2-EOR fields within permitted reservoir pressures

• Pressure analysis and modeling used to evaluate capacity

Goal 2 – Prove Storage Permanence

• Site selection to include impermeable caprock, geologic structure

• Seismic and well data used to evaluate storage mechanisms and containment

• Monitoring wells used to measure containment over time within the reef and immediate caprock

Goal 3 – Determine Aerial Extent of Plume and Potential Leakage Pathways

• Monitoring portfolio employed to understand migration • Using monitoring data to compare to and validate plume

models

Project Overview

4

Michigan Basin Large-Scale Test Goals and Objectives

4

8/22/2016

3

55

RCSP Goal MRCSP Approach and Success Criteria

Goal 4 – Develop Risk Assessment Strategies

• Risk assessment for events, pathways, and mitigation planning

• Comparing predicted to actual field experience for all stages of the project

Goal 5 – Develop Best Practices

• Phase III builds on Phase II best practices in siting, risk management, modeling, monitoring, etc.

• Key emphasis is on operation and monitoring and scale-up to commercial-scale

Goal 6 – Engage in Public Outreach and Education

• Appropriate outreach efforts for both Phase II and Phase III sites as well as technology transfer and information sharing with stakeholders

Project OverviewMichigan Basin Large-Scale Test Goals and Objectives

Project Overview

6

MRCSP scope of work is structured around six tasks

Task 1Regional Characterization: Develop a detailed actionable picture of the region’s geologic sequestration resource base

Task 2Outreach: Raise awareness of regional sequestration opportunities and provide stakeholders with information about CO2 storage

Task 3Field Laboratory Using Depleted EOR Field: Pressurize a late-stage EOR field with CO2 injection to test monitoring technologies and demonstrate storage potential

Task 4CO2 Storage Potential in Active EOR Fields: Monitor CO2 Injection and recycling in active EOR operations with different scenarios

Task 5CO2 Injection in New EOR Field(s): Monitor CO2 injection into an oil field that has not undergone any CO2 EOR to test monitoring technologies and demonstrate storage potential

Task 6 Program Management

8/22/2016

4

Technical Status

1. Injection Test

2. Modeling

3. Monitoring

4. Regional Characterization

5. Outreach

7

Technical updates grouped into five categories

Injection Test Status – EOR Life-Cycle

8

Large-scale test site leverages industrial EOR operations

8

Natural gas processing is the source of the CO2

Central Processing Facility

Late-stage

Active

Main Test Bed

Active

Pre EOR

Active

Active (new)

Active (new)

Active

Active

Active

8/22/2016

5

Variations in reef characteristics

• # of compartments, compartmentalization

• Lithology – dolomite vs limestone, Anhydrite

• Availability of core, seismic, well log data

• Presence of salt plugging

9

Injection Test Status – Accounting for CO2

10

269K MT

0.17K MT

269K MT15K MT42K MT

57K MT

-18K MT

289K MT

271K MT

139K MT

139K MT0K MT

74K MT

243K MT

317K MT

38K MT124K MT162K MT

67K MT21K MT88K MT

Net CO2 in Reef

CO2

ProducedCO2 Injected

-87K MT

154K MT

67K MT

• Nine reefs in Northern Michigan [Otsego County]

• All in various stages of EOR

• ~570K MT net injection in nine reefs during monitoring period (Feb. ‘13 – July ‘16)

• EOR still ongoing, with a new reef (CC-16) being added

Monitoring PeriodFebruary 2013 – June 2016

54K MT0K MT54K MT

8/22/2016

6

Injection Test Status – Cumulative StorageAccounting – Incidental storage over EOR lifetime

139,000357,000

597,000

974,000

1,456,000

1,647,000

0

500,000

1,000,000

1,500,000

2,000,000

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

Net in

 Reef CO2 (MT)

Net in Reef CO2 (MT)

Total EOR Net In Reef CO2 (MT)

• >1.6 million metric tons of CO2 stored over the 20 Years lifetime of EOR operations with 77% since new operator took over in 2005

Injection Test – Performance Metrics

12

A Dashboard to Manage Storage and EOR

• 9-panel dashboard of key metrics used to track performance

• Understanding of CO2-flood maturity, EOR and storage performance trends, and storage capacity

• Example: Beyond 80% by HCPV of CO2 injection, incremental oil recovery began plateauing and reef entered the late stages of EOR

Dashboard Panel for late stage reef

8/22/2016

7

Injection Test – Performance Metrics

13

Comparison Across Fields

• 4-panel dashboard used to compare storage and recovery performance across all reefs

• Normalized to %HCPV (hydrocarbon pore volume) injected

• After CO2-EOR, around ~45% of oil still remains unrecovered in the reservoirs

• D-35 is the best performing reef by oil recovery performance, and likely will have most incidental CO2-storage at the end of CO2-EOR

Example of Inter-Reef Dashboard

Injection Test – Storage Capacity Limits

14

Pressure response in Late-Stage during injection

8/22/2016

8

Geologic Modeling - Static Earth Models

15

Characterization of Diverse Michigan Niagaran Reefs

• Niagaran reefs effectively used for EOR

• Diverse geology of reefs makes characterization and SEMs challenging

• Key issues include:

Limestone vs dolomite

Salt plugging

Multi-pods

Diagenesis

Data availability

Geologic heterogeneity0.5

0.5

0.5

0.5

0.5

0.5

0.50.5

0.5

0.5

0.5

1

1

1

1

1

606000 608000 610000 612000 614000 616000 618000 620000 622000

606000 608000 610000 612000 614000 616000 618000 620000 622000

4880

0049

0000

4920

0049

4000

4960

0049

8000

5000

0050

2000

5040

0050

6000

488000490000

492000494000

496000498000

500000502000

504000506000

0.100.200.300.400.500.600.700.800.901.00

% Dolomite1

2

3

4

5

6

78

9

Geologic Modeling - Static Earth Models

16

Development of Efficient Workflow

• Developed approach to integrate data and to simplify SEMs

Formations and faciesDefine zonesCalculate petrophysicalpropertiesAnalyze whole core Depositional model

Organize log data and correlate to formations and faciesDetermine componentsUse descriptive statisticsApply geologic conceptsDefine modeling rules

Geometry and structureBuild segmentsProperty modelCalculate volumetrics

Geologic Interpretation

Geostatistics

Static Earth Model • Workflows are

repeatable and efficient

• Collaboration with WMU and Core Energy, LLC

8/22/2016

9

Geologic Modeling - Static Earth Models

17

New Insights into Michigan Niagaran Reefs

• Salt plugging can be extensive and traceable

• Definition of reef geometry with 3D seismic is critical

• Geostatistics can assist with modeling decisions and be used to predict electrofacies

• Increased dolomitization often leads to better quality reservoirs

Dynamic Modeling - Injection Response Validation

1-335594251603

Injection Schedule

19

Variable Initial Saturation Model

Variable Initial Saturation Model

Equivalent Homogeneous Compositional Reservoir Model

• Reasonable match except near the end of injection

• Further model calibration is in progress

8/22/2016

10

20

• Tighter Rock:

Lower rock compressibility

Tighter (i.e. less permeable flanks)

• Smaller pore volume for HC fluids-in-place and CO2:

Lower model pore volume

Higher water saturation outside the core reservoir in the flanks

Amount of CO2 present in the system

Dynamic Modeling - Injection Phase Pressure BuildupScenarios to explain pressure response

Dynamic Modeling – Fundamental Approaches

21

Using Synthetic Models

• Use numerical model representing typical depleted reef reservoir with simulated primary production followed by CO2 injection (but no production)

• Create synthetic datasets for analyzing pressure fall-off response and injectivity at injection well:

Pressure falloff data Horner analysis to estimate reservoir properties and identify boundaries

Injectivity index (injection rate normalized by pressure buildup) commonly-used oil-field metric of well performance

• [Q] What to expect in a multiphase environment?

8/22/2016

11

Dynamic Modeling - Synthetic ModelsFall-off Pressure Response

[Q] Does CO2 injection into an oil-gas system create a multi-bank (composite) system with different near-field and far-field characteristics?

• Pressurization in closed reservoir evident in falloffs after injection periods 4-6

• Upward shift in time-lapse Horner plots confirms evidence of boundary effects

Dynamic Modeling - Synthetic ModelsPressure Derivative Analysis

Inner zone responseInner and outer zone responses

with boundary effect

(k/)t (kkrg/g)front

(k/)t inner zone (kkrg/g)front

(k/)t outer zone (kkro/o)undisturbed

8/22/2016

12

Dynamic Modeling - Synthetic ModelInjectivity Index for Well Performance:

• [J] = Ratio of injection rate [q] to pressure buildup [Pwf-Pi]

• Useful metric for comparing well performance over timeor comparing formations

• Transient period plot of [J] versus [time] shows stabilization

• Pseudo-steady-state period plot of [1/J] versus [Q/q]: Intercept 1/J

Slope 1/(Vpct)

Dynamic Modeling - Injectivity IndexMRCSP and other field and model data show correlation of injectivity index with transmissivity

8/22/2016

13

Modeling Status: Synthetic modelingKey Points

• Inner zone total mobility (permeability divided by viscosity) related to gas-phase mobility in the vicinity of CO2 front

• Outer zone total mobility related to oil-phase mobility in the undisturbed reservoir

• Cannot determine absolute permeability from mobility, due to unknown multiphase viscosity

• Injectivity index behavior during transient and boundary dominated periods different

• Empirical correlation found between injectivity and permeability-thickness product (helpful for screening analysis and quick-look estimation of absolute permeability)

Monitoring Status – Late Stage Reef

27

Currently in After Injection Monitoring Stage

ActivityBefore

InjectionEarly

InjectionMid

InjectionLate

InjectionAfter

Injection

CO2 flow accounting X X X X

Pressure and temperature

X X X X

PNC logging X X Aug 2016

Borehole gravity X Aug 2016

Fluid sampling X X X

Vertical seismic profile X Sep 2016

Microseismic X X

InSAR (Satellite radar) X X X Complete

Characterization WellDrilling

Sep 2016

8/22/2016

14

28

Microseismic performed during final injection at Late-Stage Reef above discovery pressure

Monitoring – Microseismic

Reef Outline

Well with MS Array

InjectionWell

Well Map

Monitoring performed by Paulsson Geophysical, Inc.

29

String shots in off-set well used to “calibrate” microseismic Monitoring Status

• 5 of 6 string shots located with “good” accuracy

8/22/2016

15

30

Two types of microseismic events detectedMonitoring Status

“FOCUSED” “DISTRIBUTED?”

31

Injection Rate vs Detected EventsMonitoring Status

• 28 day injection following 6-day installation and baseline monitoring

• Step increase in injection

• ~7,000 events during initial installation and Zero-offset OVSP

• Steady frequency during initial injection

• Increased frequency after booster pump started

100

200

300

Preliminary data – do not Quote or cite

8/22/2016

16

Pulsed Neutron Capture (PNC)Monitoring Status

• New triangulation model developed to provide better resolution and better estimations of gas, water, and oil saturations

WaterOilGasDepth

Sig

ma

RATO13

Sigma vs. RATO13

Porosity

Sig

ma

Sigma vs. Porosity

RA

TO

13

Porosity

RATO13 vs. Porosity

WaterOilMethaneCO2

Saturation Estimation AnalysisMonitoring Status 0 1

Water

1 0

Oil

1 0

Gas

0 0.2

XPHIA

0 65

Sigma

• Water, oil, and gas saturations are of interest for CO2 EOR and CO2 storage

Better estimations of saturations using triangulation method

Baseline and repeat logging show changes in saturations

38%20%

42%

25%

21%54%

WaterGasOil

Repeat A1 Carb Saturations

Baseline A1 Carb Saturations

8/22/2016

17

35

Vegetation and snow are challenging for radar, but there were a reasonable number of natural reflectors

Artificial reflectors augmented the data for injection monitoring

BHP (psi)

Displacement (mm)

Time series displacement data show no correlation to injection

No Meaningful Displacement Observed Monitoring Status – INSAR Results

Geomechanical modeling to validate INSAR results

Monitoring Status

Average Dynamic Young Modulus and Poisson Ratio transferred to D-33

8/22/2016

18

37

Monitoring StatusGeomechanical modeling, multi-phase fluid flow

Com

plex

Mod

elS

impl

e M

odel

Reservoir Section

Simple model including overburden

• Bulk volume, pore volume, porosity, and permeability for each sector same as complicated model: Reservoir Flank, Aquifer, A1 Carb., Core Reservoir.

38

Geomechanical modeling agreed with INSAR resultsMonitoring Status

• Predicted surface displacement is less than 1 mm - insignificant

• Agreement among model and field observations

Top Vertical Displacement from Geomechanics (mm)

Time (Date)

8/22/2016

19

39

NESW

Kickoff 4350 ft MDPlanned Characterization• Logging• Coring/core analyses• Hydrologic and

geomechanical testing• Pressure/fluid sampling

Existing Injector

B Salt

A 2 CarbonateA 2 Evaporite

A1 Carbonate

Brown Niagaran

TD 6185MD TVD 5824 These data will improve static

and dynamic reservoir models

A new well in late-stage reef will provide key dataWhat’s Next – Characterization Well

Regional Assessment Status

40

• Population growth has not been accompanied by an increase in emissions from power plants.

• Declining market-share of coal.

• Increased availability of cheaper gas has led to more power plants switching out of coal into natural gas.

Emissions from Power plants in the MRCSP region

8/22/2016

20

41

MRCSP 10-State team conducting regional studiesRegional Assessment Status

Cambro-Ordovician Storage Potential

Led by Indiana

East Coast Offshore and Onshore Storage Targets

Led by Rutgers

Silurian Pinnacle Reef Reservoirs

Led by W. Michigan University

CCUS Opportunities in Appalachian BasinLed by Pennsylvania

Storage and Enhanced Gas Recovery for Organic Shale

Led by Kentucky

Reservoirs for CO2-EOR, EGR, and other Commercial Uses

Led by West Virginia

Use of Multiple Datasets for Regional Analysis

43 Co-Funded by

y = 0.0097e0.5776x

R² = 0.744

0.0001

0.001

0.01

0.1

1

10

100

1000

10000

100000

0 5 10 15 20 25 30

Per

mea

bilit

y (m

D)

Porosity (%)

Rose Run Porosity Permeability Transform

• Dataset varies throughout the region, formation by formation:

Basic and advanced wireline data synthesis

Core analysis for geomechanical, petrophysical, and porosity-permeability assessment

Facies analysis using petrophysicaland statistical methods

Depth, structure, isopach, thickness, and porosity map generation by formation

Regional formation assessment for storage potential

Regional Assessment – Upper Ohio Valley

8/22/2016

21

44

Formation

Mt CO2 /km3 Pore Volume

Total Prospective CO2 Storage Resource (Mt) ESaline Depositional Environment (CO2-

SCREEN; IEAGHG, 2009)P10 P50 P90 P10 P50 P90

Theoretical Max.

Esaline P50 (avg.)

Beekmantown 5 18 43 652 2,137 5,227 97,207 2.20% Dolomite: Unspecified

Rose Run 5 20 61 188 757 2,305 30,320 2.50% Clastics: Peritidal

Upper Copper Ridge 5 18 42 436 1,462 3,498 66,236 2.21% Dolomite: Unspecified

Copper Ridge B 5 18 42 205 674 1,634 30,776 2.19% Dolomite: Unspecified

Lower Copper Ridge 5 17 42 1,090 3,561 8,637 163,846 2.17% Dolomite: Unspecified

Kerbel Sandstone 6 22 63 134 505 1,464 18,610 2.71% Clastics: Delta

Conasauga 5 17 42 393 1,321 3,194 29,480 4.48% Dolomite: Unspecified

Rome 5 18 42 1,639 5,556 13,281 250,824 2.22% Dolomite: Unspecified

Basal Sandstone 6 24 70 990 3,904 11,348 130,915 2.98% Clastics: Shallow Shelf

Calculation of Prospective Stacked CO2 Storage ResourceRegional Assessment Status

Preliminary data – do not Quote or cite

45

Geologic Modeling – Upper Ohio ValleyMultiple Scales

• Regional structural model based on geologic data, regional maps, and available seismic data

• Local scale assessments at sites of interest

• Dynamic modeling of CO2

scenarios at local scale

• Analysis of image and acoustic log data with core data for analysis of mechanical properties

• Static and dynamic modeling of geomechanical caprockbehavior

• Fracture analysis and modeling of behavior

8/22/2016

22

Outreach Status

• Convening/participating in the Outreach Working Group

• Communicating results to a broad audience via site visits, fact sheets, conference and meetings, and the website

• Topical highlights:

CO2 accounting in closed reservoirs

Performance Measures

Numerical Modeling

Monitoring-Modeling Loop

Regional Storage Opportunities

• MRCSP website moving to a mobile friendly platform (transitioning in August)

46

Technology transfer is a growing focus

www.mrcsp.org

Accomplishments

47

MRCSP positioned for developing its storage potential

• ~575,000 metric tons injected across all reefs (ongoing)

• Completed injection at main test bed

Performed microseismic monitoring in final injection stage

Post-injection PNC, microgravity, and VSP underway

• Developed performance metrics to assess storage capacity

• Advancements in static and numeric modeling processes

• Collaborative team for regional assessments across ten states

• Technology transfer is focus of outreach

8/22/2016

23

Synergy Opportunities

• Knowledge share with Plains CO2 Partnership on closed reservoirs modeling and monitoring

• Knowledge share with other RCSPs on monitoring technologies and depleted oilfield modeling

• Testing NRAP models and CO2Screen tools

• Collaboration with international projects on modeling and CO2

EOR to Storage transitions

• IEAGHG monitoring/Modeling Network

• Input to DOE Best Practices Manuals

48

Research is complementary to the RCSP projects

Project Summary

• MRCSP Large-Scale Test ~60% completed with diverse EOR field setting and variety of monitoring options

• Multiple monitoring options are being tested

• Both monitoring and modeling are essential for understanding performance – imperative to be able to do much with limited data

• Regional characterization helping identify new storage zones and estimate storage resources – setting stage for commercial scale CCUS

• Results will contribute to developing standards and best practices, NRAP tools, CO2 capacity estimate tools

49

Key Findings and Lessons Learned

8/22/2016

24

Project Summary

• Completing post-injection monitoring and data analysis for late-stage field

• Drilling a new characterization well in late-stage field post-injection, applying results to validate/improve models

• Implementing metering improvements for MVA

• Applying methodologies and lessons to new EOR reefs

• Extending findings to the entire Michigan reef trend

• Expanded technical outreach

50

Future Plans

Acknowledgements

Battelle’s MRCSP Current Contributors – Mark Kelley, Srikanta Mishra, Matt Place, Lydia Cumming, Sanjay Mawalkar, Charlotte Sullivan, Priya Ravi Ganesh, Autumn Haagsma, Samin Raziperchikolaee, Amber Conner, Glen Larsen, Caitlin McNeil, Joel Main, Jacob Markiewicz, Isis Fukai, Ashwin Pasumarti, Jackie Gerst, Rod Osborne, and several others

DOE/NETL – Agreement # DE-FC26-0NT42589, Andrea McNemar (PM)

Core Energy, LLC – Bob Mannes, Rick Pardini, Allen Modroo, Bob Tipsword, and others

Ohio Development Services Agency’s Ohio Coal Development Office

MRCSP’s technical partners, sponsors, and host sites

The MRCSP Region’s State Geology Surveys and Universities

51

8/22/2016

25

Contributions From Partners Have Helped Make MRCSP Successful

52

Questions?

53

Please visit www.mrcsp.org

8/22/2016

26

Midwest Regional Carbon Sequestration Partnership

54

BACK UP SLIDES

Organization Chart

55

Charlotte SullivanAlain Bonneville

CharacterizationMonitoring Support

Prime ContractorNeeraj Gupta, PI/PM

M. Kelley, Characterization/MonitoringL. Cumming, Outreach & Regional Geology

A. Haagsma and A. Conner, GeologyS. Mishra and P. Ravi Ganesh, ModelingC. McNeil, PM support and coordinationM. Place, J. Markiewicz, J. Holly, Fieldwork

DOE/NETLAndrea McNemar

MRCSP Program Manager

Robert Mannes Rick Pardini

Large‐Scale Test Host

Kris Carter, PA Geo SurveyJohn Rupp, IN Geo Survey

Regional Characterization Task Coordinators

Sarah Wade

Outreach Working Group Coordinator

David Cole

Geochemical Monitoring

8/22/2016

27

MRCSP Task Schedule

56

MRCSP Phase III Schedule Year FY20

No. Task Quarter 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1

1.0 Regional Characterization

2.0 Outreach

3.0 Reservoir Studies in Depleted Niagaran Reefs

NEPA EQ and Site Workplan

Advanced Geological Characterization

Reservoir Modeling and Analysis

CO2 Injection

Monitoring and Analysis

Site Transfer

4.0 Reservoir Studies in Active Niagaran Reefs

NEPA EQ and Site Workplan

Reservoir Modeling and Analysis

CO2 Injection and Mass Flows

Monitoring and Analysis

5.0 Reservoir Studies New Niagaran Reefs

Site Characterization Plan

Advanced Geological Characterization

Reservoir Modeling and Analysis

CO2 Injection

Monitoring and Analysis

Site Transfer

6.0 Project Management

7.0 Deep Saline Formation Activities

Document and Close St. Peter SS Well

Approval of workplan before field work.Approval of basline geologic report before injectionIndustry Review at MRCSP Annual Meeting Task ReportsPost-transfer monitoring

FY12 FY13 FY14 FY15 FY16 FY17 FY18 FY19

60% Complete

60% Complete

60% Complete

20% Complete

10% Complete

70% Complete

90% Complete

Bibliography

• Barclay, T. and Mishra, S., 2016, New correlations for CO2-Oil solubility and viscosity reduction for light oils. J Petrol Explor Prod Technol, DOI 10.1007/s13202-016-0233-y

• Gerst, J., Cumming, L., Miller, J., Larsen, G., Gupta, N., Modroo, A., 2014, Using baseline monitoring data to strengthen the geological characterization of a Niagaran Pinnacle Reef. Energy Procedia, v. 63, p. 3923-3934.

• Hawkins, J., Mishra, S., Stowe R., Makwana, K., Main J., in press. CO2 storage capacity and potential CO2-EOR in oilfields of Ohio. Environmental Geosciences.

• Kelley, M., Abbaszadeh, M., Mishra, S, Mawalkar, S., Place, M., Gupta, N, Pardini, R., 2014, Reservoir Characterization from Pressure Monitoring during CO2 Injection into a Depleted Pinnacle Reef – MRCSP Commercial-scale CCS Demonstration Project. Energy Procedia, v. 63, p. 4937-4964, ISSN 1876-6102

• Ravi Ganesh, P., Mishra, S., Mawalkar, S., Gupta, N and Pardini, R., 2014, Assessment of CO2 injectivity and storage capacity in a depleted pinnacle reef oil field in northern Michigan. Energy Procedia, v. 63, p. 2969-2976.

• Miller, J. , Sullivan, C., Larsen, G., Kelley, M., Rike, W., Gerst, J., Gupta, N., Paul, D., Pardini, R., and Modroo, A., 2014, Alternative conceptual geologic models for CO2 injection in a Niagaran pinnacle reef oil field, Northern Michigan, USA. Energy Procedia, v. 63, p. 3685–3701.

• Gupta, N., Paul, D., Cumming, L., et al., 2014, Testing for Large-scale CO2-EOR and Geologic Storage in the Midwestern USA, Energy Procedia, v. 63, p. 6393-6403.

• Gupta, N., L. Cumming, et al., 2013. Monitoring and Modeling CO2 Behavior in Multiple Oil Bearing Carbonate Reefs for a Large Scale Demonstration in Northern Lower Michigan. Energy Procedia, v. 37, p. 6800-6807.

• Oruganti, Y.D., and S. Mishra, 2013, An improved simplified analytical model for CO2 plume movement and pressure buildup in deep saline formations. International Journal of Greenhouse Gas Control, v. 14, p. 49–59

• Swickrath, M., Mishra, S., and Ravi Ganesh, P., 2015, An Evaluation of Sharp Interface Models for CO2-Brine Displacement in Aquifers, Groundwater, DOI 10.1111/gwat.12366

57