Carbon Sequestration in the Southwestern United States:

Using the ‘String of Pearls’ Model for Cost and Source-to-Sink

Assessments

Peter H. Kobos1, Len A. Malczynski

1, David J. Borns

1 and Brian J. McPherson

2

1Sandia National Laboratories 2University of Utah & NM Institute of Mining and Technology

27th USAEE/IAEE North American Conference, Houston, TX, September 16 − 19, 2007

This paper describes an Integrated Assessment analytical model used by the Southwest

Regional Partnership on Carbon Sequestration (SWP) to assess up to hundreds of CO2 source

and geological sink sequestration projects in the Southwestern United States. The model was

first developed as a central presentation tool, later into an integrated assessment source-to-sink

matching tool (based on the ‘String of Pearls’ framework), and continues to evolve to give

regional summaries for large-scale carbon sequestration potential cost and performance metrics.

The model’s development is part of the larger Southwest Regional Partnership on Carbon

Sequestration consisting of over 20 organizations, including geological surveys, national labs,

federal and state agencies, and is one of seven such partnerships organized by the National

Energy Technology Laboratory.

The Integrated Assessment (IA) model development team at Sandia National

Laboratories continues to expand and refine the model as the larger project continues to include

additional source and sink data, technology assessments, and looks to employ increasingly

standardized cost metrics. Additionally, the model is able to help decision makers (e.g., policy

analysts and interested energy companies) determine where a CO2 source (e.g., power plant)

could be built given a set of planning constraints based on current power plant locations, sink

availability, and existing pipeline infrastructure right-of-ways.

The working results indicate that the cost of capturing carbon dioxide is by far the

majority of a potential project’s overall capital cost. For example, the capture, transportation

and storage-associated cost breakdown of a plant in northern New Mexico may be 95%, 2% and

3% of the initial cost estimate, respectively. The IA also develops overarching results such as

regional CO2 sequestration totals, relative cost issues, and sink lifetimes across an initial fifty-

year time horizon. The initial results indicate the region may support anywhere from several

decades to over ten thousand years’ worth of CO2 sequestration potential capacity depending

upon the assumptions regarding CO2 source and sink resources. To sequester this much CO2,

however, may have substantial parasitic energy penalties ranging from 15 to 40% associated

with capturing the CO2 from, for example, power plants. The IA team, as part of the larger SWP,

continues to include additional and more refined capture cost data from a larger working group

across all of the regional partnerships in an effort to develop a more ‘seamless story’ for carbon

sequestration at a high level. These standardization efforts also allow for common technological

progress to be included in the partnership’s modeling efforts to address the cost and energy

penalty issues. Ultimately, however, many of the costs will be highly site-specific for power

plant-scale carbon sequestration projects.

Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United

States Department of Energy’s National Nuclear Security Administration under contract DE-AC04-94AL85000.

Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United

States Department of Energy’s National Nuclear Security Administration under contract DE-AC04-94AL85000.SAND2007-5153C

INTRODUCTION The U.S. Department of Energy developed seven regional partnerships throughout the

United States to assess the geological and economic potential for large-scale CO2 sequestration and the associated changes to the existing infrastructure. The Southwest Regional Partnership (SWP) for carbon sequestration developed several thematic committees over the last several years to address a potential CO2 sequestration network.

i The Integrated Assessment (IA) thematic committee developed an interactive computer model to initially serve as a central presentation tool, then as a general cost and CO2 flow model, and continues to develop as an overarching systems view framework to integrate additional information coming from the potential pilot projects in the SWP region.

The original Integrated Assessment’s approach to the larger SWP began by presenting a causal loop diagram and a brief description of system dynamics as a conceptual approach to form the backbone of the integrated assessment work. The main goals of these activities were to develop scenarios for regulators and policy makers, determine what information private industry was interested in, and to develop outreach sessions where the regulators, industry participants and public participants could comment on and understand the complex interactions of policy, engineering and economics.

The primary deliverable was a workshop-based, iterative mediated group modeling-oriented series development process. These sessions included interaction through face-to-face and web-enabled meetings plus explicit directions/input from the DOE sponsors to include other features in the model. The project developed such that a test case model was developed in 2004, an initial minimal spanning tree algorithm then became the ‘String of Pearls’ network in 2005, a larger model was built to include all of the sinks and sources in 2006, and in 2007 a power plant aging chain and replacement module was included to further refine the timing of potential carbon capture installation given the existing power plant fleet.

THE INTEGRATED ASSESSMENT MODEL: AN OVERVIEW OF THE

DEVELOPMENT PROCESS

The Integrated Assessment working group developed a dynamic simulation computer

model used by the Southwest Regional Partnership on Carbon Sequestration to assess CO2 source (power plants) and geological CO2 sink combinations at a high level. The model is a tool that highlights the amounts of CO2 a sequestration project might capture at a power plant, calculates the required CO2 pipeline capacity for the project, and addresses the time scales involved with such a project (e.g., the number of years it may take until a geological sink is filled with CO2 from a power plant of a certain type) along with the illustrative economics

The Integrated Assessment Team largely served as a committee in the first phase of the project to help present the information generated from the select thematic committees in a generic framework. Figure 1 illustrates the Integrated Assessment model’s underlying framework.

Figure 1. The Integrated Assessment model’s underlying cost and CO2 model structure.

Figure 2 illustrates the general structure of the SWP, and highlights the role of the Integrated Assessment within the overall project.

Figure 2. Thematic Committees & Partner Integration

(adapted from McPherson, 2005a). The Integrated Assessment team used the simulation model on an initial select set of

sources and sinks in New Mexico, then moved on to the entire state, and finally included the entire region covered within the region for the Southwest Regional Partnership. This ‘bottom up’ modeling approach allowed the SWP to continuously improve the model in a consistent and transparent manner.

The initial test case model calculated all of the illustrative cost and CO2 flow combinations between four power plants in New Mexico, and seven geological sequestration sites. The model then ranked the cost combinations from lowest to highest.

The original test case’s CO2 separation and capture calculations were based directly on the Integrated Environmental Control Model (IECM) developed at Carnegie Mellon University (CMU, 2004). ii The model also employs the CO2 emissions data for power plants based on the 2002 version of the Emissions and Generation Resources Integrated Database (eGRID) (EPA, 2005). The CO2 transport and injection equations for CO2 storage calculations were based on Ogden (2002), Williams (2002) and through periodic industry participation by SWP including

Source Capture Transportation Storage Metering

Power Plants

Cost Modeling

Pipelines

CO2 Sinks

Ongoing

Integrated Assessment

Information,

GIS / Database

Information,

GIS / Database

CO2 Sinks &

Distributed

Sources

CO2 Sinks &

Distributed

Sources

Public

Education &

Involvement

Public

Education &

Involvement

Infrastructure,

Separation,

Capture &

Point Sources

Infrastructure,

Separation,

Capture &

Point Sources

Regulatory

Compliance

Regulatory

Compliance

Data-Sharing StatesData-Sharing States WGA and Other Critical PartnersWGA and Other Critical Partners

Industry Advisory Panel

Mike Hirl (Hirl, 2004).iii The CO2 sinks database was developed by the geological sinks thematic committee of the SWP, and managed by the Utah AGRC.iv Additional data used in the model on power plants was obtained from the Electricity Supply and Demand Database and Software (NERC, 2006).

USER OPTIONS AND DEVELOPING THE INTERFACE FOR THE INTEGRATED

ASSESSMENT MODEL

The Integrated Assessment team expanded the model to include the full suite of power

plants and geological sinks within New Mexico. This model version was the first to employ the ‘String of Pearls’ model module (Kobos et al., 2006). The ‘String of Pearls’ is a metaphor for the string of new or existing pipelines connecting a collection of potential CO2 sources and geological sinks (the ‘pearls’). The ‘String of Pearls’ module calculates the transportation distances to move CO2 based on a great circle distance algorithm from the source of the CO2 (e.g., the power plant) to the closest sink (e.g., the geological reservoir) (Stephens, 1998). The ‘String of Pearls’ module, therefore, builds on hypothetical (potential) pilot project pipelines and cost-saving connections with the existing pipeline infrastructure in the region. Figure 3 illustrates the pipelines and select CO2 sources and geological sinks in New Mexico considered in the initial ‘String of Pearls’ prototype model.

Figure 3. Map of the CO2 sources (power plants) and geological sinks considered in the

‘String of Pearls’ Integrated Assessment New Mexico model prototype. The CO2 sinks are assigned their place in the ‘String of Pearls’ so that as one fills to

capacity, the pipeline network extends to the next sink, fills it, and continues to move down an existing pipeline or build a new one to reach the next viable sink. All of the links between the

multitudes of source-sink combinations are the potential pipeline routes. This technique serves as a linear proxy for new pipeline routes, and lays the groundwork for future, additional analysis throughout both the SWP area in the United States and beyond. With this technique, the model can address additional metrics (e.g., largest sink volume, acceptable distance between sinks, etc.) for a deeper level of systems insight. The full ‘String of Pearls’ version of the Integrated Assessment model includes 218 geological sinks (plus an additional 107 points along existing CO2 pipelines), and 83 power plants across the five core states in the SWP. The full ‘String of Pearls’ model determines the source to sink distances and associated economics for any of the source sink combinations between the power plants, geological sinks and select pipeline nodes (with more refinements to come as the SWP continues to progress). The model includes high-level measurement, monitoring and verification (MMV) costs of approximately $0.16 to $0.31 per tonne of CO2 and can illustrate up to the ten closest sinks to any ten power plants (Benson et al., 2004). It can also perform the calculations for all of the sink assessments relative to any of the selected sources.

The cost metrics are based on previous work using the IECM model results and a regression equation proxy to approximate the cost to capture CO2 as a working model component as the other regional partnerships come to a more standardized method to calculate the costs, or, the model allows for custom cost metrics if more information is known.v Additionally, the carbon capture community is looking to reduce uncertainty surrounding the overall carbon capture systems’ costs and performance parameters (e.g., using an amine (MEA)-based system holds substantial potential to reduce the cost to half of their current costs according to recent work by Rao and Rubin, 2002, and Rao et al., 2006).

One of the core innovations of the ‘String of Pearls’ model is the ability to include existing CO2 pipeline infrastructure and down select the specific classes of geological sinks (e.g., oil/gas, coal bed methane, saline aquifers) in the network algorithm. The model includes portions of the Bravo Dome, Cortez, and Sheep Mountain pipelines to allow for ‘piggybacking’ into current CO2 pipelines’ corridors across the southwest region. This gives planners the ability to consider the pipeline as a sink or a politically viable corridor to other sinks because of its established CO2 pipeline transportation right-of-way designation and the ability to move CO2 between basins.

Figures 4 illustrates the main ‘System Results’ front page from the model. The front page illustrates the top 10 closest single source to multiple sink cost results according to the respective components, the number of years each sink will last along the hypothetical pipeline network, the assumed percentage capture of CO2 and corresponding capture costs, pipeline and well cost multipliers, and finally the MMV costs within a slider to allow for a type of overall systems cost sensitivity analysis.

Figure 4. The San Juan Power Plant CO2 Source to Multiple Geological Sinks using

the full ‘String of Pearls’ Pipeline Network Algorithm.

The model user can also query the sinks information based on the geonode of the sink,

the type of sink (e.g., coal bed methane, oil/gas, saline aquifer), the maximum distance between the source and sinks for the algorithm to consider it (e.g., to minimize the distance of the pipeline network to be developed), and the minimum desired capacity of the sinks (e.g., million metric tonnes of CO2 capacity). The specific sinks can also be selected per state, which serves as a legend for the sinks’ names. For example, figure 4 shows sink number 14 as the first to meet the criteria, which is the Barker Dome-Hermosa Group (oil/gas) formation. It has 10 million metric tonnes of CO2 storage capacity. The next sink selected was the San Juan and Paradox Basin 1-Miss Leadville Limest (saline aquifer) formation with 1,000 million metric tonnes of capacity. Figure 5 illustrates the San Juan power plant as a potential source of CO2, but includes only the oil and gas reservoirs in the algorithm for the entire region. This demonstrates how the pipeline network results adjust to the new constraint.

Figure 5. The Full ‘String of Pearls’ model only considering oil and gas formations that

are at least 500 million metric tonnes in size as geological sinks in the region. (This

approximates to 60+ years’ worth of capacity from the San Juan Power Plan).

The user can also select a specific location for a power plant by entering its latitude and

longitude coordinates. Lastly, the model user has the ability to select specific geological sinks throughout the region. This allows one to assess a specific power plant’s CO2 and cost profile relative to a specific geological sink to determine the high-level metrics a pilot project might entail (e.g., costs, length of the pipeline, sink size relative to the plant’s CO2 output).

THE POWER PLANT REPLACEMENT MODULE

The Integrated Assessment model has been designed to allow for additional data to be taken into account as it becomes available for additional/existing CO2 sources, sinks and other salient components. The model also includes a power plant construction cycle based on the work of Ford (2001) and Malczynski et al. (2007). The model includes changes in the plant’s capacity due to the parasitic energy losses associated with carbon capture technologies. Anticipating the capacity gap, the model then calculates construction and licensing time delays which can result in some oscillatory behavior to build capacity over the coming decades. Figures 6 & 7 show a summary screen of the model which demonstrates the existing utility coal and gas plants being replaced over time, and various regional totals for the carbon dioxide captured at the plants.

Figure 6. The prototype interface presenting electricity generation capacity due to carbon

capture penalty and new capacity growth in the U.S. up to 2050.

Figure 7. The ‘Replacing Plants’ interface. (Note: Most of the utilities’ coal and gas

powered electricity generating capacity could be replaced by 2050.)

Given a modest 3% increase in MWh demand per year the working results indicate the majority of existing plants will be replaced by the year 2050. When total projected demand exceeds the power generated from existing plants and their replacement plants, the model builds new plants of a given size. The replacement cycle provides an opportunity to build more efficient plants that are carbon capture ready. For perspective, the SW region includes a relatively small proportion of the nation’s CO2 emissions (Figure 8).

2,611

22036

0

500

1,000

1,500

2,000

2,500

3,000

NM SWP US

Million m

etric tonnes of

CO2 / year

Figure 8. Carbon Dioxide Emissions for New Mexico, the Southwest Regional Partnership

(SWP) states, and the United States (Note: ‘SWP’ is the sum of the emissions from

Arizona, Colorado, New Mexico, Oklahoma and Utah, (EPA, 2005; 2000 data)).

Exports of electricity across state lines will raise policy challenges when the debate comes as to where, when and how electricity producers and users will pay for the added costs associated with CO2 sequestration technologies. For additional context to the scale of cross-state electricity flows, Arizona, Colorado, New Mexico, Oklahoma and Utah, for example, export roughly 63, 11, 73, 32 and 65 percent of their electricity (EIA, 2004) to regions beyond their state’s boundaries, respectively.vi Figures 9 and 10 illustrate the installed megawatts and the CO2 emissions associated within the selected states.

0

5,000

10,000

15,000

20,000

Arizona Colorado New Mexico Oklahoma Utah

Utility M

egawatts

Coal Gas Hydro Nuclear

Figure 9. Utility-based Installed Megawatts for States in the Southwest U.S. in 2000

(Note: Oil-based & other fuels represented 2% or less of the total installed megawatts)

(EPA, 2005).

0

20

40

60

Arizona Colorado New Mexico Oklahoma Utah

Million M

etric Tonnes of

Carbon Dioxide

Coal Gas

Figure 10. Million Tonnes of Carbon Dioxide Emissions from Utilities in the core states

within the Southwest Regional Partnership on Carbon Sequestration. (Note: Oil-based &

other fuels represented 2% or less of the total CO2 emissions. EPA, 2005; 2000 data).

The sources database will continue to develop as different sources are considered for the overarching region(s) associated with the Integrated Assessment model. The sinks database will also continue to develop as different data is considered for the overall regional assessment throughout the SWP, and beyond. Figure 11 illustrates the geological sinks data in the full ‘String of Pearls’ version of the Integrated Assessment as of July 2007. The SWP continues to collect CO2 sinks data for Texas, other states, and potentially other pilot projects as the overall partnership looks to address several key issues including what types of sinks should be included in the analysis (e.g., those of a certain size, location, depth, within certain regulatory constraints, etc.).

Figure 11. Carbon Dioxide Storage Capacity (million metric tonnes) in the core states of the

SWP in the working database (Note: Coal Bed Methane capacity in UT is 57 million metric

tonnes in the working database and some Saline Aquifer data was unavailable as of July

2007).

0

5999

0 0

1431 1126

8440

577

0

2,500

5,000

7,500

10,000

AZ CO NM OK UT

Million Metric Tonnes of

Carbon Dioxide

Oil & Gas

Saline Aquifer

27,072 2,055,010

THE INTEGRATED ASSESSMENT AND MODELING SOFTWARE ISSUES

The uniqueness of the Integrated Assessment model, and in particular the ‘String of

Pearls’ framework, lies within the model’s structure and flexibility. The model has been built in Powersim Studio, a system dynamics software often used in forecasting.vii The ‘String of Pearls’ (SOP) application focuses primarily on a core analysis capability associated with source-to-sink matching and optimization-oriented elements to help ‘tell the story’ of where, and how much CO2 sequestration may cost from regional power plants. The primary reason to develop this capability is to address a range of uncertainty derived from the set of stakeholders in the SWP. Although using system dynamics modeling does not immediately imply the construction of a computer based model, the demands of stakeholders driven by their accumulated experience with general modeling and simulation-oriented thinking of the modeling software demands that system dynamics applications include increasingly non-system dynamics features (Sterman, 2000). Over time, as the model became increasingly intricate, the corresponding application (the core analytical model code plus the custom user interface) in turn became increasingly more intricate. The Integrated Assessment Team is looking to develop a website to allow for additional user and public feedback regarding the ‘String of Pearls’ model. This website can help increase potential partnership feedback while decreasing the costs (time, $) to receive this feedback across the larger partnership. Figure 12 illustrates a prototype screen as part of the website materials under development.

Figure 12. A Prototype Interface Screen for the Integrated Assessment’s ‘String of Pearls’ Source-Sink Model Materials.

The partnership continually made decisions about what and how to include model features into the application that fall outside the system dynamics framework. The goal of these decisions, of course, was to have a useful and meaningful application. Powersim Studio 2005 has the standard system dynamics methodological tools required to develop an interface capability with the necessary modeling elements. The modeling team made use of the ‘VBFUNCTION()’ tool that permits a modeler to build their own functions if expressed in Visual Basic Script (VBScript).

The SOP model includes a substantial amount of externally accessed data in the form of standard spreadsheet files. Much of this data is easier to organize outside of Studio and then manipulated in Studio for the purposes of interface development, data aggregation, or relationship

calculation(s). VBScript provides a simple solution to these challenges. Studio, like most system dynamics software, has a large set of built in functions. It is not, however, a general purpose programming language. The algorithm employed to estimate the great circle distance between sources and sinks must iterate to an approximate solution. This could be done within the traditional stock and flow paradigm, but with additional manipulation beyond the scope of the SOP at this time. This capability was therefore added via employing the VBScript. Finally, the minimal spanning tree algorithm was obtained in the form of pseudo-code from the research community. It was then constructed in VBScript to operationalize it within the Studio environment.

DISCUSSION

The overarching goals of this project have evolved from developing a presentation tool to

refining a more widely-applicable analytical tool available to help potentially site a carbon sequestration project in the Southwestern United States. The central goals as submitted to the Department of Energy have been met despite the substantial evolution in the model’s scope. The shift was due to a combination of changing requirements/suggestions over time from partnership members, and due to the complex data needs beyond the original scope of the work efforts. The model requires detailed information on power plants, geology, and various types of technology. The modeling efforts during Phase I (2004 – 2005) focused mainly on oil and gas reservoirs within New Mexico, and select parts of the southwestern United States. The central insight gained from these initial efforts is the cost to capture carbon dioxide at the point sources represents the majority of the overall system’s costs to the order of 90% or more. Thus, in a carbon constrained world, carbon sequestration projects might focus their efforts on cost effective technologies to capture carbon dioxide in concert with geological assessments of sinks with sufficient capacity to provide a useful sink lifetime for the project.

Phase II (2006 and beyond) efforts have included even more detailed, region-wide source-sink matching and model components to eventually include pilot projects. Additionally, the model includes a power plant aging and replacement cycle that accounts for the parasitic power losses which may occur in the face of wide-scale adoption of carbon capture and sequestration technologies to the existing energy infrastructure.

Potential additions and modeling capabilities may include developing time-to-build constraints for the infrastructure, developing a capital budgeting-oriented sub-framework (e.g., if given a limited budget, which project might develop), and looking to determine how the regulatory environment might drive (or constrain) the wider adoption of carbon sequestration technologies. Additional modeling granularity (e.g., costs, sinks’ capacity, etc.) will be incorporated as it develops from the other thematic committees participating in the SWP.

Construction of the ‘String of Pearls’ model and interface posed some challenges. Using a more general approach toward developing the software interface on top of the underlying quantitative model has proved to be extremely useful when demonstrating the model to both general and specialized audiences, including to members of other regional partnerships.

With the combined insight provided by the ‘String of Pearls’ Integrated Assessment model as well as the power plant replacement module, planners can assess the technologies, economics and associated issues using an integrated, high-level view when deciding where to develop future carbon sequestration projects and understanding the overall potential carbon sequestration future in the U.S. Much of the overall cost per carbon sequestration project, however, will be highly site-specific for power plant-scale projects.

ACKNOWLEGEMENTS

The authors wish to thank the National Energy Technology Laboratory (NETL), Barry Biediger (Utah Automated Geographic Reference Center (AGRC)), collaborators from Los Alamos National Laboratory (LANL), and other colleagues within the Southwest Regional Partnership on Carbon Sequestration (SWP) for their assistance in these ongoing research efforts.

REFERENCES

Benson, S.M., Hoversten, M., Gasperikova, E. and M. Haines. (2004). Monitoring Protocols and

Life-Cycle Costs for Geologic Storage of Carbon Dioxide. GHGT7, Vancouver, Canada. Biediger, Barry. (2007). The Southwest Regional Partnership on Carbon Sequestration,

The Sinks Database, Personal Communication, Images reproduced with the permission of Biediger of the Utah Automated Geographic Reference Center (AGRC), July 26, 2007.

Carnegie Mellon University (CMU). (2004). The Integrated Environmental Control Model

(IECM). IECM, Update version 4.0.4.1, September. Drennen, T.E., Baker, A.B., Jones, S.A., Kamery, W.J., Lutz, A.E., Rosthal, J.E. (2004). The

Hydrogen Futures Simulation Model (H2Sim) Technical Description. SAND2004-4937. October.

Environmental Protection Agency (EPA). (2005). Emissions & Generation Resource Integrated

Database (eGRID), eGRID2002, Version 2.01. Year 2000 Data, 27, June 2005. http://www.epa.gov/cleanenergy/egrid/index.htm

Energy Information Administration (EIA). (2004). Individual State Data. 19, March, 2004.

http://www.eia.doe.gov/emeu/states/_states.html

Ford, Andrew. (2001). Waiting for the boom: a simulation study of power plant construction in California. Energy Policy, 29 (2001) 847-869.

Hirl, M. (2004). CO2 Sequestration Integrated Model Workshop, Personal Communication,

April 20. Kobos, P.H., D.J. Borns, L.A. Malczynski and O.H. Paananen. (2005a). Southwest Regional

Partnership on Carbon Sequestration: A Test Case Model in the San Juan Basin. Poster Presentation at the Fourth Annual Carbon Capture and Sequestration Conference, Alexandria, VA, May 2 – 5.

Kobos, P.H., D.J. Borns and L.A. Malczynski. (2005b). The Southwest Regional Partnership on

Carbon Sequestration: Employing the Integrated Assessment Model for Systems Insight. Proceedings of the 25th Annual N. American Conference of the USAEE/IAEE, September 19 – 21, 2005, Denver, CO.

Kobos, P.H., L.A. Malczynski and D.J. Borns. (2006). Southwest Regional Partnership on

Carbon Sequestration: The ‘String of Pearls’ Integrated Assessment Model. Proceedings of the 5th Annual Carbon Capture and Sequestration Conference, Alexandria, VA, May 8 – 11.

Malczynski, L.A., P.H. Kobos and D.J. Borns. (2007). Understanding Carbon Sequestration: A Systems Analysis Tool. International Conference of the System Dynamics Society, Boston, MA, July 29 – August 2, SAND2007-4650 C.

McPherson, B. (2005a). Phase I Project Overview, DE-PS26-O3NT41983, October 12, NETL

Regional Carbon, Sequestration Partnerships Review Meeting Pittsburgh, Pennsylvania. McPherson, B. (2005b). U.S. Department of Energy Solicitation DE-PS26-05NT42255,

Technical Narrative March 15. Meyer, H. (2006). Personal Communication, utilized The Integrated Environmental Control

Model (IECM), As of February 2. Ogden, J.M. (2002). Modeling Infrastructure For a Fossil Hydrogen Energy System with CO2

Sequestration. Sixth Greenhouse Gas Control Technologies Conference, Kyoto, Japan, September 30 – October 4.

Paananen, O. (2004). A Framework for Assessing Carbon Storage in the Southwestern United

States. Presentation at the Third Annual Conference on Carbon Sequestration, Alexandria, VA, May 3 – 6.

Rao, A.B. and E. Rubin. (2002). A Technical, Economic, and Environmental Assessment of Amine-Based CO2 Capture Technology for Power Pant Greenhouse Gas Control. Environ. Sci. Technol. 36, pp. 4467-4475.

Rao, A.B., Rubin, E.S., Keith, D.W. and M.G. Morgan. (2006). Evaluation of potential cost

reductions from improved amine-based CO2 capture systems. Energy Policy, Vol. 43, Issue 18, pp. 3765-3772.

Stephens, R. (1998). Ready-To-Run Visual Basic Algorithms, Second Edition. John Wiley &

Sons. Sterman, J.D. (2000). Business Dynamics, Systems Thinking and Modeling for a Complex

World. Irwin-McGraw Hill. Williams, R.H. (2002). Toward Zero Emissions for Transportation Using Fossil Fuels. In VII

Biennial Asilomar Conference on Transportation, Energy, and Environmental Policy: Managing Transitions, K.S. Kurani and D. Sperling (eds.), Transportation Research Board: Washington, D.C. - forthcoming, and Prepared for Managing Transitions in the Transportation Sector: How fast and how far? VII Biannual Conference on Transportation, Energy & Environmental Policy, Monterey, CA. 11-14, September, 2001, final draft revised 7.15.2002.

Zagonel, A.A and T.F. Corbet. (2006). Levels of Confidence in System Dynamics Modeling: A

Pragmatic Approach to Assessment of Dynamic Models. 2006 International Conference of the System Dynamics Society.

i This paper builds on several works including those of Paananen (2004), Kobos et al., (2005a, 2005b, 2006), the work of Malczynski et al., 2007 and is the next iteration of this ongoing work effort.

ii The Gas Technology Institute (GTI) used the IECM to calculate the CO2 capture costs for several power plants in Phase I of the SWP. iii Based on the Ogden (2002) model of costs for CO2 disposal and sequestration, Williams (2002) develops the general framework where the cost of CO2 disposal is a function of the cost of the pipeline transmission (CPT) + the cost of disposal wells (CDW) + the cost of surface piping near the disposal wells (CSP); where CPT($/tCO2) = CPT0*(Quantityn/Quantity0)^-0.53 * (length of pipelinen/length of pipeline0)1.24; Cost per well ($/well)=$1.0 million + ($1.25 million/km)*[depth(km)]; CSP=0.138*(Quantity-104.17)^0.253. The calculations developed for the IA also draw on the work of Drennen et al. (2004) and the work of Kobos et al. (2005a, 2005b). iv Barry Biediger (2007) of the Utah AGRC has been instrumental in the SWP by managing and maintaining the core SWP sinks data. The majority of the working sinks database has been incorporated where possible into the Integrated Assessment model unless missing data prevented further analysis (e.g., sink’s depth from the surface) or size constraints limited their usefulness (those less than 10 million metric tonnes in size, which equates to approximately one year’s worth of storage (or less) for a large coal-fired power plant). v The Integrated Assessment Model began using direct results from the IECM model based on GTI work for hire (Meyer, 2006). As of the summer of 2007, the model employs the following working regression equations for coal and natural gas-fired power plants, respectively; Capture cost (coal plant) = 48.9683 + (0.0003 * Power plant MW) + (-0.2030 *% CO2 captured); Capture cost (natural gas plant) = 117.6814 + (0.0409* Power plant MW) + (-0.6665* % CO2 captured); R

2=0.9553 and 0.9574, respectively . Additionally, the model allows users to specify custom capture costs as more detailed information becomes available. vi Colorado imported roughly 4% of its electricity in 2000 (EIA, 2005). vii Zagonel and Corbet 2006.