Solutions for Today | Options for Tomorrow
U.S. DOE’s National Risk Assessment Partnership:Assessing Carbon Storage Risk Performance to Support Decision Making Amidst Uncertainty February 16, 2016
2
National Risk Assessment Partnership
NRAP leverages DOE’s capabilities to quantify storage risks amidst system uncertainties, to help remove barriers to full-scale CO2 storage deployment.
Stakeholder Group
Wade, LLC
Objective: Building tools and improving the science base to address key questions related to environmental impacts from potential release of CO2 or brine from the storage reservoir, and potential ground-motion impacts due to injection of CO2
Technical Team
3
NRAP’s approach to quantifying performance relies on reduced-order models to probe uncertainty in the system.
NR
AP In
tegr
ated
Ass
essm
ent
(Sys
tem
) Mod
els
Storage Reservoir
Release and Transport
Potential Receptors or
Impacted MediaData
IAM
E. Exercise whole system model to explore risk performance
A. Divide system intodiscrete components
B. Develop detailed component models that are validated against lab/field data
C. Develop reduced-order models (ROMs) that rapidly reproduce component model predictions
D. Link ROMs via integrated assessment models (IAMs) to predict system performance
4
NRAP Phase I Accomplishments (2011-2016)Assessing environmental risk and quantifying uncertainties
• Pioneered the movement into quantitative risk assessment, uncertainty quantification, and reduced order modeling for carbon storage
• Developed insights into key technical issues around storage-security relationships• Transferring technology to CCUS stakeholders
IJGGC Virtual Specia Issue (August, 2016)
NRAP Tools Workshop and Webinar Series
Final release, https://edx.netl.doe.gov/nrap
5
• NRAP Outreach• Tools: 10
• https://edx.netl.doe.gov/nrap/• External Beta Tool Testers: 112
• Publications: 389• Manuscripts (submitted/under development): 46
• Technical Reports Series documents (TRSs)• 56 published; 4 in final review
• Presentations at conferences/meetings: 490• International Journal of Greenhouse Gas Control Special Issue – 54 articles• Report: National Risk Assessment Partnership (NRAP) Phase I Accomplishments (2011-16)
NRAP Phase I
6
Storage Reservoir ResponseNRAP Phase I Accomplishments
Wel
lbor
es
Faul
ts &
ISSize of CO2 Plume Size of Pressure Plume Pressure at at a Location
Bromhal et al., 2014
Develop reduced-order models (ROMs) that rapidly reproduce component model predictions
Identify Critical Reservoir Storage/Risk Relationships
7
System Integration and Strategic MonitoringNRAP Phase I Accomplishments
Prototype Design Approaches for Strategic Monitoring
Wel
lbor
es
Faul
ts &
IS
Probabilistic assessment of whole-system containment and leakage risk
NOTE: Hypothetical cases for demonstration purposes only
8
Potential Induced SeismicityNRAP Phase I Accomplishments
Wel
lbor
es
Faul
ts &
IS
Experiments to reduce uncertainty in fault/fracture slip-induced permeability changes
Estimating Ground Motion Response from potential Induced Seismicity
Measured
Forecast
Even
ts /
Day
Time, hours
Forecasting short-term, injection-related induced seismicity
9
Areas of Research Focus:• Containment Assurance• Induced Seismicity Risk• Strategic Monitoring for Uncertainty Reduction• Synthetic and Field Data collection, use, and dissemination
NRAP Phase II Focus
Managing environmental risk and reducing uncertainties for CO2 storage sites
10
Use case example: conformance evaluation
Monitoring data
Compare simulations to
monitoring data
Updated risk assessmentInitial risk assessment
Containment Assurance researchEarly Progress
• Drafting formal software design basis document, clearly defining Phase II IAM capabilities and the QA/QC process.
• Collecting intended IAM ‘use cases’ from project participants (complete) and from stakeholders
• Designing next generation reduced-order model of leaky wellbores that consider geomechanical and geochemical processes and permeability evolution over time
11
• Real-time Hazard Forecasting• Focus: Improve Short-Term Seismic Forecasting (STSF) tool by testing new forecasting methods and improving tool
usability.
• Active Seismicity Management• Focus: Study effectiveness of different techniques (e.g. pressure control) for managing seismicity at problematic sites.
• Probabilistic Seismic Risk Assessment• Focus: Transition NRAP workflow to a practical industrial workflow by partnering with stakeholders in the seismic risk
consulting world.
• Fault Leakage• Focus: Targeted monitoring and active mitigation of fault leakage (through, e.g., hydraulic barriers).
• Seismicity Management Protocol• Focus: Best-practices protocol for CO2 seismicity management, supported by a suite of tools to help stakeholders
implement a practical workflow.
Induced Seismicity Research
12
Multiphysics models
Input for geophysical modeling
Inversion output
IAM-CS
Monitoring data
Phase I
Phase II
STRATEGIC MONITORING FOR UNCERTAINTY REDUCTION
Identified interactions between modeling, monitoring, and data processing
• Development of Methods to Model Monitoring Techniques
• Risk-Based Monitoring Network Design Tool
13
• Illinois Basin – Decatur Project• Demonstrating applicability of AIM groundwater tool to IDBP site
• Sensitive hydraulic parameters have not been measured at the site• Non-adjustable ROM parameters, such as the initial pH and TDS and the pH no-impact
threshold, are significantly different at the site
• Midwest Regional Carbon Sequestration Partnership (MRCSP)• Estimating risk using DREAM monitoring design tool and Wellbore Leakage
Analysis Tool (WLAT)• Incorporate substantial database of abandoned wellbore properties
Field Applications
13
Average Total Well Integrity
Indicator (per 1 km2 Area)
1210864
• FutureGen• Gathered data originally collected during license application process
and put in usable format for potential application of several NRAP models
• Investigating risk-based area of review
• CarbonSafe and BEST• Working with new DOE field projects to identify opportunities for
collaboration and tool validation
14
• Tool Development• Synthetic datasets generated during development have been or will be archived on EDX• Tool developers have recently been surveyed to assess what kinds of synthetic datasets
would be helpful to future tool development
• Outreach to CCS Community• Work with International Data Sharing Group• Prepare synthetic data sets for use by others
Synthetic/Community Datasets
14
16
NRAP Phase I CO2 Storage Risk Assessment Toolset
Integrated Assessment Model – Carbon Storage (NRAP-IAM-CS) - Simulates long-term full system leakage and containment behavior (reservoir to aquifer/atmosphere)Reservoir Evaluation and Visualization (REV) Tool - Generates pressure and CO2 plumes sizes over timeReservoir ROM Generator (RROMGEN) – Converts reservoir simulation results for input to NRAP-IAM-CSWellbore Leakage Analysis Tool (WLAT) – Evaluates existing well leakage potentialNatural Seal ROM (NSealR) - Estimates flux through a fractured or perforated seal Aquifer Impact Model (AIM) - Estimation aquifer volume impacted by a leak (for pH, TDS, select metals and organics)Design for Risk Evaluation and Management (DREAM) -Selects optimal monitoring design for minimum time to detectionShort Term Seismic Forecasting (STSF) - Forecasts seismic event frequency during injection, over hours/daysGround Motion Prediction application for potential Induced Seismicity (GMPIS) - Predicts ground motion response from potential induced earthquakesMultiple Source Leakage ROM (MSLR) – Characterizes atmospheric dispersion of leaked CO2
17
Wells and Seals Leakage PerformanceNRAP Phase I Accomplishments
Well Leakage Scenarios in WLAT
Brine leakage through Fractured Cement
Cemented Wellbore with Thief Zone
Well Permeability (m2)Num
ber o
f wel
ls
Open Wellbore
(Jordan et al., 2015; Harp, et al., 2016)
(Huerta, et al., 2016)
Fractured Caprock
(Lindner, 2016; Namhata et al., in review)
18
Rapid exploration of trends in potential well leakage
Well Permeability (m2)
Num
ber o
f wel
ls
0
1
2
3
4
0 500 1000 1500 2000 2500 3000Brea
kthr
ough
Tim
e (y
ears
)
Distance from Injection Site (m)
Well Age, Completion Quality & Distance from Injection Site vs. Breakthrough Time
2000's1980's1950'sPoor Well Completion
Carey, 2014 • What is the relative role of individual well parameters?
• Can we use additional data to rank wells and develop monitoring and mitigation strategies?
NRAP Phase I Accomplishments
19
Rapid Prediction of Receptor ResponseNRAP Phase I Accomplishments
Wel
lbor
es
Faul
ts &
IS
Rapid estimation of atmospheric dispersion
Predicting groundwater impacts from potential leakage
Hydraulic ROM
Geochemistry Scaling
Function
Coupled Groundwate
r ROM
20
• Identifying critical leakage conditions for wells
• Measuring relative permeability for fractured wells
Containment Assurance researchEarly Progress
21
Oklahoma Field Application
• Objective: Use observation data and seismic simulations to characterize poroelastic parameters of the Arbuckle group.
22
Oklahoma Field Application
23
Background seismic velocity model using wireline log data from the initial stratigraphic borehole
Modeling of seismic monitoring for the FutureGen 2.0 CO2 storage site
24
Leakage Detection Probability Maps
(a) pH
Tim
e (y
ear)
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0.2
0.2
0.5
0.8
1
1
1
1
1
1
1
1
0.1
0.1
0.2
0.4
0.7
0.8
0.8
0.8
0.8
0.8
0.9
0.8
0
0
0.1
0.2
0.3
0.4
0.5
0.5
0.5
0.5
0.7
0.5
0
0
0.1
0.1
0.2
0.3
0.4
0.4
0.4
0.4
0.5
0.4
0
0
0
0.1
0.1
0.2
0.2
0.3
0.3
0.3
0.4
0.3
0
0
0
0
0.1
0.1
0.2
0.2
0.2
0.2
0.3
0.2
0
0
0
0
0.1
0.1
0.1
0.2
0.2
0.2
0.3
0.2
0
0
0
0
0.1
0.1
0.1
0.1
0.1
0.1
0.2
0.1
0
0
0
0
0.1
0.1
0.1
0.1
0.1
0.1
0.2
0.1
0
0
0
0
0
0
0.1
0.1
0.1
0.1
0.1
0.1
0
0
0
0
0
0
0
0
0
0
0.1
0
0
0
0
0
0
0
0
0
0
0
0
0
(fixed sensor, tolerence interval thresthod)
20 100 300 500 700 900 1500
1
2
3
4
5
10
20
30
40
50
100
200
(b) TDS
0
0
0
0
0
1
1
1
0
0
0
0
0
0
0
0
0
1
1
1
0
0
0
0
0
0
0
0
0
0.2
0.2
0.2
0
0
0
0
0
0
0
0
0
0.1
0.1
0.1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
(fixed sensor, tolerence inter
20 100 300 500 700 900 1500
1
2
3
4
5
10
20
30
40
50
100
200
(c) Benzene
Monitoring grid distance (m)
Tim
e (y
ear)
0
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
0.2
0.2
0
0
0
0
0
0
0
0
0
0
0.1
0.1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
(groundwater sampling, detection limit thresthod, CV=0.01)
20 100 300 500 700 900 1500
1
2
3
4
5
10
20
30
40
50
100
200
(d) Combined detection pr
Monitoring grid distance (m)
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0.2
0.2
0.5
0.8
1
1
1
1
1
1
1
1
0.1
0.1
0.2
0.4
0.7
0.8
0.8
0.8
0.8
0.8
0.9
0.8
0
0
0.1
0.2
0.4
0.5
0.5
0.5
0.5
0.5
0.7
0.5
0
0
0.1
0.1
0.2
0.3
0.4
0.4
0.4
0.4
0.5
0.4
0
0
0
0.1
0.1
0.2
0.2
0.3
0.3
0.3
0.4
0.3
0
0
0
0
0.1
0.1
0.2
0.2
0.2
0.2
0.3
0.2
0
0
0
0
0.1
0.1
0.1
0.2
0.2
0.2
0.3
0.2
0
0
0
0
0.1
0.1
0.1
0.1
0.1
0.1
0.2
0.1
0
0
0
0
0.1
0.1
0.1
0.1
0.1
0.1
0.2
0.1
0
0
0
0
0
0
0.1
0.1
0.1
0.1
0.1
0.1
0
0
0
0
0
0
0
0
0
0
0.1
0
0
0
0
0
0
0
0
0
0
0
0
0
20 100 300 500 700 900 1500
1
2
3
4
5
10
20
30
40
50
100
200
Spatial and Temporal Detectability
Apply Monte-Carlo simulations for leakage simulations and monitoring modeling to explore risk scenarios given uncertainty
Use statistical methods to combine multiple monitoring techniques and provide decision support for evaluating proposed monitoring plan
Total probability of leakage detection (PD) given a monitoring network =
∑𝒍𝒍=𝟏𝟏𝑳𝑳 probability of leakage detection of leakage pathway(l)∗prior probability of leakage pathway (l)
Risk-based Monitoring Design
25
Containment and Monitoring Institute (CaMI) Field Research Station
Containment Tools and Methodologies Field Demonstration: Leakage Analog Site
25
CO2 injectors
Dept
h
CO2
Observationwell(s)
Wel
l-bas
ed m
onito
ring
tech
nolo
gies
surface monitoringtechnologies
CO2 300 m
500 m
BGP
• Undertake controlled CO2 release at 300 m (phase I) & 500 m depth (phase II); Injection of ~1000 t/yr will start in June/July.• Determine CO2 detection thresholds • Develop improved monitoring technologies.• Monitor fugitive gas emissions.
Phase I layout
Injection wellMonitoring well