Introduction
Congress is considering policies to reduce U.S. emissions of greenhouse gases. Prominent among these policies are those promoting the capture and direct sequestration of carbon dioxide (CO2) from manmade sources such as electric power plants and manufacturing facilities. Carbon capture and sequestration is of great interest because potentially large amounts of CO2 produced by the industrial burning of fossil fuels could be sequestered. Although they are still under development, carbon capture technologies may be able to remove up to 95% of CO2 emitted from an electric power plant or other industrial source.
Carbon capture and sequestration (CCS) is a three-part process involving a CO2 source facility, a long-term CO2 sequestration site, and an intermediate mode of CO2 transportation—typically pipelines. Some studies have been optimistic about pipeline requirements for CO2 sequestration. They conclude that the pipeline technology is mature, and that most major CO2 sources in the United States are, or will be, located near likely sequestration sites, so that large investments in CO2 pipeline infrastructure will probably not be needed.1 Other studies express greater uncertainty about the required size and configuration of CCS pipeline networks.2 A handful of regionally-focused studies have concluded that CO2 pipeline requirements for CO2 sources could be substantial, and thus present a greater challenge for CCS than is commonly presumed, at least in parts of the United States.3
Divergent views on CO2 pipeline requirements introduce significant uncertainty into overall CCS cost estimates and may complicate the federal role, if any, in CO2 pipeline regulation. They are also a concern because uncertainty about CO2 pipeline requirements may impede near-term capital investment in electricity generation, with important implications for power plant owners seeking to reduce their CO2 emissions.
In the 110th Congress, there has been considerable debate on the capture and sequestration aspects of carbon sequestration, while there has been relatively less focus on transportation. Nonetheless, there is an increasing perception in Congress that a national CCS program could require the construction of a substantial network of interstate CO2 pipelines. The Carbon Dioxide Pipeline Study Act of 2007 (S. 2144), introduced by Senator Norm Coleman and nine cosponsors on October 4, 2007, would require the Secretary of Energy to study the feasibility of constructing and operating such a network of CO2 pipelines. The America's Climate Security Act of 2007 (S. 2191), introduced by Senator Joseph Lieberman and nine cosponsors on October 18, 2007, and reported out of the Senate Environment and Public Works Committee in amended form on December 5, 2007, contains similar provisions (Sec. 8003). The Carbon Capture and Storage Technology Act of 2007 (S. 2323), introduced by Senator John Kerry and one cosponsor on November 7, 2007, would require carbon sequestration projects authorized by the act to evaluate the most cost-efficient ways to integrate CO2 sequestration, capture, and transportation (Sec. 3(b)(5)). The Energy Independence and Security Act of 2007 (P.L. 110-140), signed by President Bush, as amended, on December 19, 2007, requires the Secretary of the Interior to recommend legislation to clarify the appropriate framework for issuing CO2 pipeline rights-of-way on public land (Sec. 714(7)).
This report examines key uncertainties in CO2 pipeline requirements for CCS by contrasting hypothetical pipeline scenarios in one region of the United States. The report summarizes the key factors influencing CO2 pipeline configuration for major power plants in the region, and illustrates how the viability of different sequestration sites may lead to enormous differences in pipeline costs. Power plants, particularly coal-fired plants, are the most likely initial candidates for CCS because they are predominantly large, single-point sources, and they contribute approximately one-third of U.S. CO2 emissions from fossil fuels. The report discusses the implications of uncertain CO2 pipeline requirements for CCS as they relate to evolving federal policies for carbon control.
Scenarios for CO2 Pipeline Development
Under a national CCS policy, a key question is how to establish a CO2 pipeline network at the lowest social and economic cost given the current locations of existing CO2 source facilities and the locations of future sequestration sites. On its face, this may appear to be a straightforward analytic problem of the type regularly addressed in other network industries. The oil and gas industry, among others, employs myriad analytic techniques to identify and optimize potential routes for new fuel pipelines.4 In the context of CCS, however, predicting pipeline routes is more challenging because there is considerable uncertainty about the suitability of geological formations to sequester captured CO2 and the proximity of suitable formations to specific sources of CO2. One recent analysis, for example, concluded that 77% of the total annual CO2 captured from the major North American sources could be stored in reservoirs directly underlying these sources, and that an additional 18% could be stored within 100 miles of the original sources.5 Other analysts suggest that captured CO2 may need to be sequestered, at least initially, in more centralized reservoirs to reduce potential risks associated with CO2 leaks.6 They suggest that, given current uncertainty about the suitability of various on-site geological formations for long-term CO2 sequestration, certain specific types of formations (e.g., saline aquifers) may be preferred as CO2 repositories because they have adequate capacity and are most likely to retain sequestered CO2 indefinitely.
The Department of Energy estimates that the United States has enough capacity to store CO2 for tens to hundreds of years.7 However, the large-scale CO2 experiments needed to acquire detailed data about potential sequestration reservoirs have only just begun. Given current uncertainty about potential sequestration sites, policy discussions about CCS envision various possible scenarios for the development of a CO2 pipeline network. If CO2 can be sequestered near where it is produced then CO2 pipelines might evolve in a decentralized way, with individual facilities developing direct pipeline connections to nearby sequestration sites largely independent of other companies' pipelines. The resulting network might then consist of many relatively short and unconnected pipelines with a small number of longer pipelines for facilities with no sequestration sites nearby. Alternatively, if only very large, centralized sequestration sites are permitted, the result might be a network of interconnected long distance pipelines, perhaps including high-capacity trunk lines serving a multitude of feeder pipelines from individual facilities. A third scenario envisions CO2 sequestration, at least initially, at active oil fields where injection of CO2 may be profitably employed for enhanced oil recovery (EOR). Indeed, a CO2 pipeline network already exists for EOR purposes in the southwestern United States, although it is limited in geographic reach. Whether CCS policies ultimately lead to one or more of these scenarios remains to be seen; however, the configuration of the resulting CO2 pipeline network, and its associated costs, may have a significant bearing on which CCS policies best serve the public interest.
Hypothetical CO2 Pipelines in the Midwest
Infrastructure requirements and policy implications related to CO2 pipelines become clearer when considering what actual pipeline projects might look like. This section outlines contrasting scenarios for hypothetical CO2 pipeline development in the region covered by the Midwest Regional Carbon Sequestration Partnership (MRCSP). The MRCSP is one of seven regional partnerships of state agencies, universities, private companies, and non-governmental organizations established by the Department of Energy to assess CCS approaches. The MRCSP serves as a good illustration of CO2 pipeline issues because it has a varied mix of CO2 sources and potential geologic sequestration sites, and because geologists have completed a number of focused studies relevant to CCS in this region.
Sequestration in the Rose Run Formation
The MRCSP has identified key CO2 sources and geologic formations potentially suitable for carbon sequestration within its seven-state region encompassing northeast Indiana, Kentucky, Maryland, Michigan, Ohio, Pennsylvania, and West Virginia. Figure 1 shows the locations of 11 of the largest CO2 sources located in the MRSCP region—all coal-fired electric power plants emitting over 9 million metric tons of CO2 annually.8 There are numerous other CO2 sources in this region, including many other power plants and large industrial facilities, but the 11 power plants in this analysis include the very largest in terms of annual CO2 emissions.
|
Figure 1. Major Power Plants and the Rose Run Formation |
|
Source: MRCSP. Geologic data for NY are not provided by the MRCSP. |
Figure 1 also shows the locations of the Rose Run sandstone, a deep saline formation identified by the MRCSP as a potential carbon sequestration site.9 As the figure shows, the plants all lie above or near to this formation, so suitable CO2 injection sites presumably could be located very near to each of these plants. If the Rose Run formation proves to be viable for large-scale CO2 sequestration, then some plants may be able to inject CO2 directly below their facilities, and CCS pipeline requirements for some of the other 11 power plants could be small. If this were the case, then the CCS CO2 pipeline network for the 11 plants might appear as shown in Figure 2.
|
Figure 2. Hypothetical CO2 Pipelines to the Rose Run Formation |
|
Source: MRCSP, CRS. Geologic data for NY are not provided by the MRCSP |
The hypothetical pipeline layout in Figure 2 assumes that a 25-mile diameter, non-overlapping reserve area is needed for each plant's sequestration site and that any location within the Rose Run formation is viable for sequestration. Figure 2 also assumes that each power plant is either located at or is connected to the center of its respective sequestration field by a large trunk pipeline built along existing rights of way and capable of carrying its peak CO2 output. Smaller pipelines branching from the centrally-located plant or from the trunk line distribute the CO2 to multiple injection wells in the sequestration site. These smaller pipelines are not considered in detail in this report.
Figure 2 shows that the longest trunk pipeline for CO2 transportation is 32 miles long, and the average pipeline is approximately 11 miles long. According to models developed at Carnegie Mellon University (CMU), the capital costs to construct an 11-mile pipeline in the Midwestern United States with a capacity of 10 million tons of CO2 annually would be approximately $6 million. The levelized cost would be approximately $0.10 per ton of transported CO2, including costs for operation (e.g., compression) and maintenance.10
Potential Barriers to Rose Run Sequestration
Although the Rose Run formation is identified by the MRCSP as a major potential sequestration site, it has characteristics which may ultimately limit its viability for large-scale CO2 sequestration. The most important of these is overall sequestration capacity. Because the Rose Run formation has low to moderate permeability and thickness, geologic models show that it is unlikely all of the CO2 emitted in the Rose Run region can be efficiently sequestered in the Rose Run formation.11 The Rose Run formation is also relatively fractured.12 Geologists have concluded that injecting pressurized CO2 into the Rose Run formation potentially could induce minor earthquakes along certain preexisting (but undetected) faults in otherwise seismically stable areas.13 Faults and fractures can, in some cases, provide additional sequestration capacity and be beneficial for sequestration. But faults or fractures can also be permeable conduits for leakage and "can be a significant pathway for the loss of sequestered CO2."14 While studies are not yet available to establish the validity of any of these concerns, future research may conclude that significant parts of the Rose Run formation would be unsuitable for large scale, permanent CO2 sequestration.15
Alternatives to CO2 Sequestration in Rose Run
The CO2 sequestration capacity of the Rose Run formation may turn out to be too limited because of its of overall size or integrity. If the policy goal is to sequester CO2 from all major sources in the region, then at least some of the largest power plants in the MRCSP will need to sequester their carbon emissions elsewhere. The alternative sites for potential CO2 sequestration nearest to Rose Run are unmineable coal beds, oil and natural gas fields, and another large saline formation—the Mount Simon sandstone.
Unmineable Coal Beds
The MRCSP region contains unmineable coal beds underlying the same general geographic footprint as the Rose Run formation, but located at different depths underground. Studies suggest that such coal beds may be suitable for sequestration. In some cases injected CO2 could replace methane trapped in the coal seam, increasing natural gas available for extraction wells in a process similar to EOR known as enhanced coal-bed methane recovery. However, the potential capacity for storing CO2 in regional coal beds is only about 5% compared to the Rose Run sandstone, and the practicability of storing CO2 in coal seams is virtually untested.16 In addition, removing groundwater from coal seams prior to CO2 injection may create environmental problems related to water disposal, and some studies indicate that coal swelling associated with CO2 injection may curtail the permeability of the coal seam, limiting its overall capacity to store CO2.17
Oil and Natural Gas Fields
The MRCSP region includes a number of oil and natural gas fields which may offer opportunities for CO2 sequestration. The region also includes a number of natural gas storage reservoirs, both natural and manmade, which suggest that CO2 could be similarly stored. However, according to the MRCSP, the ten largest oil and gas fields in the region have an average CO2 sequestration potential of only 251 million tons.18 By comparison, the 30-year CO2 output of the 11 plants in this analysis would range from 270 to 491 million tons at current emission levels. The oil and gas fields in the MRCSP region, therefore, even if they could achieve their stated sequestration potential, may not individually have sufficient capacity to sequester CO2 from one of the 11 power plants in this analysis operating with current emissions over a 30-year period. Multiple fields possibly could be used by individual power plants to achieve adequate long-term sequestration, but this would require multiple pipeline networks and, consequently, could increase CO2 transportation costs and complexity.
Oil and gas production fields also present CO2 sequestration challenges due to numerous boreholes from historical well-drilling activity. Geologists are concerned that old oil and gas wells may be inadequately sealed and that their locations may be uncertain.19 Increased leakage risks from old wells, as well as associated mitigation and monitoring costs, may reduce the economic CCS sequestration potential in oil or gas fields. Although revenues from CO2 sales for EOR projects could offset CO2 transportation and sequestration costs for some source facilities, long-term CO2 emissions in the MRCSP region would far exceed CO2 requirements for EOR. It is possible, therefore, that because of their limited sequestration capacity and wellbore leakage concerns, oil and natural gas fields in the MRCSP region may not be viable sequestration sites for the largest CO2 sources either.
Mt. Simon Formation
If neither the Rose Run formation nor regional coal, oil, or gas fields can provide adequate CO2 sequestration for the major power plants in the MRCSP region, the next best potential CO2 sequestration site is the Mt. Simon formation. This formation is a deep saline aquifer like the Rose Run formation, but it is over four times larger in terms of sequestration capacity and is less fractured.20
|
Figure 3. Hypothetical CO2 Pipelines to the Mt. Simon Formation |
|
Source: MRCSP, CRS. Geologic data for western Indiana are not provided by the MRCSP. |
Figure 3 shows hypothetical CO2 pipelines which might be required if any of the major power plants in this analysis were required to transport CO2 to the Mount Simon formation. As in the Rose Run case, Figure 3 assumes pipelines use existing rights of way and that a 25-mile diameter, non-overlapping reserve area is needed for each plant's sequestration site. However, consistent with the Rose Run limitations, the Mt. Simon scenario assumes that the thinnest parts of the formation (the easternmost contours on the contour map) are unsuitable sequestration sites. As the figure shows, the pipelines required in such a scenario could be substantial, ranging in length from 130 to 294 miles, and averaging 234 miles. According to estimates from CMU, the approximate capital costs for these pipelines would range from $70 million to $180 million, and would average $150 million. The average levelized cost would be approximately $2.00 per ton of transported CO2.21
Although Figure 3 shows a pipeline route for all the 11 power plants in question, how many of these pipelines might be needed depends upon which plants may be able to sequester their CO2 emissions closer to home. Furthermore, there are potential scale economies for large, integrated CO2 pipeline networks that link many sources together rather than single, dedicated pipelines between individual sources and sequestration reservoirs.22 The individual pipelines required in Figure 3 may be so large on their own that combining multiple CO2 flows from multiple plants through shared trunk lines may be limited.23
While the Mt. Simon scenario in Figure 3 is far less favorable in terms of cost and siting requirements than the Rose Run scenario in Figure 2, it is not necessarily the "worst" case in terms of overall pipeline requirements. Future work on sequestration capacity may conclude that the Mt. Simon sequestration sites should be located in thicker parts of the formation (in central Indiana and Michigan) to absorb the tremendous volumes of CO2 generated by these power plants. Such a westward shift would require even longer pipelines than those illustrated here.
Policy Implications
The MRCSP pipeline scenarios, while only illustrative, nonetheless highlight several important policy considerations which may warrant congressional attention. These include concerns about CO2 pipeline costs, siting challenges, pipeline and sequestration site relationships, and differences in sequestration potential among regions.
Variability of CO2 Pipeline Costs
The cost of CO2 transportation is a function of pipeline length (among other factors), which in turn is determined by the location of sequestration sites relative to CO2 sources. The scenarios in this report illustrate how different assumptions about sequestration site viability in the MRCSP region can lead to a 20-fold difference in CO2 pipeline lengths and, therefore, similarly large differences in capital costs. (In this regard, CO2 pipeline costs may present the cost component in integrated CCS schemes with the greatest potential variability.) At the international and national policy levels, some studies have recognized this potential variability. For example, an MIT analysis states that the costs of CO2 pipelines are highly variable due to "physical ... and political considerations."24 The IPCC report likewise estimates total costs of CO2 mitigation of $31- $71 per ton of CO2 avoided for a new pulverized coal power plant, assuming CO2 pipeline transportation costs, including operations and maintenance costs, of $0 to $5 per ton.25 Recent increases in the global price of steel used to make line pipe could push CO2 pipeline costs above this range.26 At $5 per ton of transported CO2, pipeline costs account for a modest share of aggregate carbon control costs—between 7% and 16% based on the IPCC estimates. Nonetheless, if CCS technology were deployed on a national scale, overall CO2 pipeline costs could be in the billions of dollars. Minimizing these costs while achieving environmental objectives may therefore be an important public policy objective.
From the perspective of individual power plants, or other CO2 sources, highly variable costs for CO2 pipelines may have more immediate ramifications. If CO2 pipeline costs for specific regions reach hundreds, or even tens, of millions of dollars per plant, then power companies may have difficulty securing the capital financing or regulatory approval needed to construct or retrofit fossil fuel-powered plants in these regions. For example, in August 2007, the Minnesota Public Utilities Commission rejected a developer's proposal to construct a new coal-fired power plant in the state, in large part because the associated costs of a 450-mile CO2 pipeline to an EOR site in Alberta, over $635 million, were not viewed to be in the public interest.27 To the extent that other, lower-cost power plant options are available, the failure of a costly project like the Minnesota plant may not be a problem. However, if other generation sources are constrained (e.g., nuclear, renewable), then the inability to construct a new fossil-fueled power plant may negatively impact the regional balance of electricity supply and demand. Higher electricity prices or reliability concerns might ensue.
Some analysts believe that CO2 pipeline costs will be moderated in the future because generating companies will construct new power plants geographically near sequestration sites. Recent network cost models suggest otherwise. On a mile-for-mile basis, these models show that electricity transmission costs (including capital, operations, maintenance, and electric line losses) generally outweigh CO2 pipeline costs in new construction. Accordingly, the least costly site for a new power plant tends to be nearer the electricity consumers (cities) rather than nearer the sequestration sites if the two are geographically separated.28 Analysts have therefore concluded that "a power system with significant amounts of CCS requires a very large CO2 pipeline infrastructure."29
CO2 Pipeline Siting Challenges
Any company seeking to construct a CO2 pipeline must secure siting approval from the relevant regulatory authorities and must subsequently secure rights of way from landowners. There is no federal authority over CO2 pipeline siting, so it is regulated to varying degrees by the states (as is the case for oil pipelines). The state-by-state siting approval process for CO2 pipelines may be complex and protracted, and may face public opposition, especially in populated or environmentally sensitive areas.30 Securing rights of way along existing easements for other infrastructure (e.g., gas pipelines), as the scenarios in this report assume, may be one way to facilitate the siting of new CO2 pipelines. However, questions arise as to the right of easement holders to install CO2 pipelines, compensation for use of such easements, and whether existing easements can be sold or leased to CO2 pipeline companies.31 Although these siting issues may arise for any CO2 pipeline, they become more challenging as pipeline systems become larger and more interconnected, and cross state lines. If a widespread, interstate CO2 pipeline network is required to support CCS, the ability to site these pipelines may become an issue requiring new federal initiatives.32
Pipeline and Sequestration Site Relationships
Due to potential CO2 transportation costs, individual generating plants have a strong interest in the selection of specific sequestration sites under future CCS policies. Since transporting CO2 to distant locations can impose significant additional costs to a facility's carbon control infrastructure, facility owners may seek regulatory approval for as many sequestration sites as possible and near to as many facilities as possible. Furthermore, capacity limitations at favorably located sequestration sites (like the Rose Run formation) may lead to competition among large CO2 source facilities seeking to secure the best local sequestration sites before others do. How the development of sequestration sites will be prioritized and how competition for such sites may evolve have yet to be explored, but they may create new and significant economic differences among facilities.
Because CO2 pipeline requirements in a CCS scheme are driven by the relative locations of CO2 sources and sequestration sites, identification and validation of such sites must explicitly account for CO2 pipeline costs if the economics of those sites are to be fully understood. Proposals such as S. 2323, which would require an integrated evaluation of CO2 capture, sequestration, and transportation (Sec. 3(b)(5)), appear to promote such an approach, although the details of future sequestration site selection have yet to be established. If CCS moves from pilot projects to widespread implementation, government agencies and private companies may face challenges in identifying, permitting, developing, and monitoring the large number of localized sequestration reservoirs that may be proposed.
Advantaged and Disadvantaged Regions
Geologists have long recognized that some regions in the United States have high potential for carbon sequestration and others do not. For example, a 2007 study at Duke University concluded that "geologic sequestration is not economically or technically feasible within North Carolina," but "may be viable if the captured CO2 is piped out of North Carolina and stored elsewhere."33 Likewise, states in the Northeast, Minnesota, Wisconsin, and possibly parts of other states appear to lack geological formations with potential for large-scale sequestration of the volumes of CO2 they produce. If national CCS policies are implemented, power plants and other CO2-producing facilities in these states may face more extensive, and more costly, pipeline requirements than other states if they are to sequester their CO2. States such as North Carolina, with limited sequestration potential and a relatively high proportion of coal or natural gas in their electric generation fuel mix, may face particular challenges in this regard. The Duke study, for example, estimated it would cost $5 billion to construct an interstate pipeline network for transporting CO2 from North Carolina's electric utilities to sequestration sites in other states.34
One particular concern among some stakeholders is that high CO2 transportation costs could increase electricity prices in "sequestration-poor" regions relative to regions able to sequester CO2 more locally. For states like Massachusetts, for example, which has some of the highest electricity prices in the country and may have little sequestration potential, CO2 transportation costs could raise electricity prices even higher above the national average. Moving beyond this illustrative example to evaluate comprehensively the distribution of CO2 transportation costs across the United States is beyond the scope of this report. Nonetheless, these kinds of regional price impacts, and their implications for regional economies, may become an issue for Congress.
Conclusion
The socially and economically efficient development of the nation's public infrastructure is an important consideration for policymakers. In the context of a national program for CCS, CO2 pipelines may be a major addition to this infrastructure. Yet there are many uncertainties about the cost and configuration of CO2 pipelines that would be needed to meet environmental goals within an emerging regulatory framework. Exactly who will pay for CO2 pipelines, and how, is beyond the scope of this report, but understanding ways to minimize the cost and environmental impact of this infrastructure may be of benefit to all.
In addition to specific questions about CO2 pipeline requirements, the scenarios in this report raise larger questions about the ultimate development and allocation of sequestration capacity under a national CCS policy. How much individual companies may have to spend to transport their CO2 depends upon where it has to go. However, even as viable sequestration reservoirs are being identified, it is unclear which CO2 source facilities will have access to them, under what time frame, and under what conditions. While Congress is beginning to turn its attention to these questions, it will likely require sustained attention and the input of many stakeholders to refine and address them. Given the potential size of a national CO2 pipeline network, many billions of dollars of capital investment may be affected by policy decisions made today.