THE BUSINESS OF GLOBAL WARMING: OPPORTUNITIES FOR THE OIL AND GAS INDUSTRY IN GREENHOUSE GAS MITIGATION

Transcription

1 THE BUSINESS OF GLOBAL WARMING: OPPORTUNITIES FOR THE OIL AND GAS INDUSTRY IN GREENHOUSE GAS MITIGATION by Steven Bryant Associate Professor, Dept. of Petroleum and Geosystems Engineering and Director, Geological CO 2 Storage Joint Industry Project Center for Petroleum and Geosystems Engineering The University of Texas at Austin Austin, TX steven_bryant@mail.utexas.edu ABSTRACT Mitigating the effect of anthropogenic emissions of carbon dioxide will require multiple technologies, including fuel switching and storage of carbon dioxide in subsurface geologic formations. Using natural gas instead of coal for power generation will become more attractive in a carbon-constrained world. An important advantage of geologic storage is that know-how and hardware needed for large-scale storage are available offthe-shelf in the oil & gas industry. But the economics of byproduct disposal differ from the economics of resource extraction. Moreover geologic storage involves intrinsic technical uncertainties. Thus geologic storage will require new modes of interaction between technology, risk assessment and regulation. Managing these interactions will demand a notable, perhaps unprecedented, degree of collaboration and communication between engineers, geoscientists and lawyers. Indeed, the greatest challenge for mitigating carbon dioxide emissions will be developing a regulatory environment that helps, rather than hinders, the cause. This paper briefly reviews the reasons that fuel switching and geologic storage must contribute of any serious effort to mitigate greenhouse gas emissions. The role of fuel switching has significant implications for supply and demand of natural gas. Evaluating the size of a geologic storage industry is crucial, because the scale of that industry is the single most important factor driving policy, regulation and technology. A non-technical description of four technical aspects of geologic storage follows, setting the context for a discussion of the cross-disciplinary cooperation needed to enable a storage industry. INTRODUCTION The relationship between carbon dioxide and climate change has become an important and complicated issue for the energy industry. At one extreme, environmental factions call for cessation of fossil fuel consumption. At the other, business-as-usual factions 1

2 dismiss global warming as a hoax. Selective presentation of data is common on both sides. Debate is often conducted in moral, rather than scientific, terms, and often descends into vitriol. This is no accident; for the more vocal sides the debate is about a world view, not about global warming per se. The business-as-usual faction fears any action that would limit economic growth. The earth-first faction regards industry as the root of many ills afflicting humanity and the biosphere. Global warming is merely the latest bone of contention between these factions. The fear of the business-as-usual faction is well-founded. The solution to global warming commonly proposed by the earth-first faction may be summed up as leave the coal in the ground. (Read coal to include all fossil fuels.) This strategy is not practical on the time scale of interest (the next few decades), and if imposed on that time scale, it would certainly lead to an economic downturn, as discussed in the next section. But the fears of the earth-first faction are also well-founded. Declaring that human activity cannot perturb the climate is a high-risk policy, given the evidence (also discussed in the next section). Because the time scale for the climate to respond to a perturbation can be long (decades), it could be far too late to take preventive action, such as GHG mitigation, by the time a human influence became clear enough to be unarguable. Amidst the debate, and regardless of whether scientific or political consensus exists, large investment decisions are turning on the perceived impact of projects on CO 2 emissions. Permits for a coal-fired plant for electricity generation are becoming more difficult to obtain in the US and can lead dramatic restructuring, as in the case of TXU (now Luminant). The re-insurance industry and the credit ratings industry now gauge a company s possible exposure in the event of a carbon-constrained world. Companies include in their annual reports to shareholders a carbon inventory for their operations. The high profile of carbon within many companies and industries suggests a way around the impasse on global warming. Paying for waste disposal is now part of doing business in the modern world. This was inevitable as the scale of human activity grew after the Industrial Revolution. Industry after industry has changed its practices to reduce the volume or concentration of byproducts from its operations. Lead was removed from motor gasoline; oxides of sulfur and of nitrogen were removed from coal combustion flue gas; fluorocarbons were phased out of routine use. None of these actions caused the demise of industry. By setting a price on carbon dioxide, the last major byproduct of energy consumption would be brought into the cost of doing business. Once a price is established, a market can emerge. Companies will figure out how to profit from myriad forms of GHG mitigation. Overall emissions will be avoided or reduced without draconian reductions in economic activity; in fact, the opportunities for activity will increase. These include major opportunities for the oil and gas industry. This view of a market-driven response to GHG emissions may be optimistic, as nothing of this scale has been tried before. The cap-and-trade system for sulfur dioxide emissions worked well, but carbon dioxide emissions are ten to one hundred times larger. Because substantive efforts to mitigate greenhouse gas (GHG) emissions will have tremendous effect on the energy industry and the global economy, it is worthwhile to review the case 2

3 for action. It is then instructive to consider the options for action and the role that the oil and gas industry can play in those options. The Case for Action The argument for mitigating GHG emissions begins with two uncontroversial facts. The carbon dioxide molecule absorbs radiation in the infrared spectrum, and the atmospheric concentration of carbon dioxide has been increasing in century or so. The rate of increase became especially dramatic in the last fifty years, as shown in Fig. 1. Fossil fuel consumption has also grown rapidly in the last century, Fig. 2. The common, but by no means unchallenged assumption is that human activity, primarily in the form of burning fossil fuel, is responsible for the increase in carbon dioxide concentration. The corollary is that human activity is responsible for any temperature increase associated with the larger carbon dioxide concentrations. Unfortunately, it is very difficult to predict the time scale and the extent of warming driven by an increase in carbon dioxide concentrations. The Earth s climate is an extraordinarily complicated collection of feedback loops, and carbon dioxide is but one of many contributors to the forcing functions that drive climate change. The problem of prediction is the crux of the debate about whether GHG mitigation is necessary. The geologic record, the last four hundred thousand years of which is shown in Fig. 3, illustrates both sides of the argument about human influence on the climate. On one hand, the correlation between atmospheric carbon dioxide concentrations and average temperature in the biosphere is clear. On the other, there was no anthropogenic contribution to changes occurring in the geologic past. It is also clear that the climate is self-regulating. The variations in carbon dioxide concentration are bounded within a relatively narrow range, from about 180 parts per million (ppm) to 280 ppm. Perturbations from any source do not grow without bound. Yet the climate does exhibit large swings over relatively short periods. This behavior is consistent with a dynamic system containing many feedback loops. A small perturbation need not cause runaway behavior, but it can easily lead to quite different operating conditions in a short period of time. If we then regard the climate as being in a state of dynamic equilibrium, the perturbation of atmospheric carbon dioxide concentration in the last century is highly significant. The concentration exceeded 300 ppm just under one hundred years ago, and now exceeds 380 ppm. This range of values is unprecedented in the last four hundred thousand years, Fig. 3. As discussed above, it is not possible to predict accurately the response of the climate to this perturbation, but it would be surprising indeed if there were no response. It is also useful to consider the magnitude of human contributions to Earth s carbon cycle. Table 1 summarizes the fluxes between several compartments that exchange carbon in various forms. Land use changes and fossil fuel consumption are anthropogenic; the other fluxes are natural. The contributions of these mechanisms are small compared to the naturally occurring fluxes. However, to understand the behavior of a dynamic system, it is not just the magnitude of the fluxes but the net flux between compartments that matters. 3

4 The very large exchanges between atmosphere and ocean and between atmosphere and vegetation and soil are almost in balance. The last column of the table shows that the contribution of human activities to the net flux into the atmosphere is significant. Indeed, fossil fuel consumption makes the single largest net contribution of carbon into the atmosphere. In the last century, humans have become sufficiently numerous and sufficiently advanced technologically to play a significant role in the carbon cycle on this planet. Table 1. Selected fluxes in the Earth s carbon cycle. A negative (positive) value of net flux indicates carbon leaving (entering) the atmosphere. Data from National Petroleum Council Facing the Hard Truths About Energy, Figure 5-1, p downloaded 31 May 2008 from Ch5-CarbonMgmt.pdf Source Sink Flux, Net flux, Mechanism Compartment Compartment Gt C/y Gt C/y CO Surface ocean Atmosphere 2, 90 hydrocarbon -2 exsolution and Atmosphere Surface Ocean absorption 92 Vegetation, soil, detritus Atmosphere Vegetation, soil, detritus Atmosphere Global net Atmosphere 60 primary Vegetation, soil, production and detritus respiration 61.2 Atmosphere 1.6 Land use Vegetation, soil, changes 0.5 detritus Fossil Fuel Atmosphere Combustion 7 +7 Much of the debate about global warming centers on the choice of proxy for temperature. It is inevitably unsatisfactory to choose a single number to represent the state of the climate. In this context it is useful to consider a simple, broad based, model-independent observation. Records compiled over the last century or so indicate that a large fraction of flora and fauna are moving upward, if they live in mountainous regions, or poleward if they live far from mountains (Parmesan, 2006; Rosenzweig et al., 2008). An even larger fraction of species are blooming, breeding or migrating earlier in the year. These changes are consistent with a response to warmer temperatures indicated by other physical changes in the same habitat range. The changes cannot be explained by a species evolving a tolerance to a broader range of conditions, and thereby expanding its habitat. As these changes are widely distributed across the globe, they constitute very strong evidence for warming. 4

5 Options for Action Business and governments alike are calling for action to avoid global warming. It seems likely that a carbon-constrained world will come to pass, without waiting for consensus on the human contribution to climate change. What actions are appropriate? Many global models have been run to evaluate stabilization scenarios. The goal is to identify a suite of responses that would result in a plateau of atmospheric CO 2 concentration. The time scale for these scenarios is typically one century, possibly two. Two examples from the IPCC are reproduced in Fig. 4 (IPCC, 2006). The difference between total carbon dioxide production and total emissions to the atmosphere is achieved by implementing a variety of technologies. These examples make different assumptions about economic activity, population growth, government policies etc. The point here is that two of the major contributors to GHG mitigation are fuel switching and carbon capture and storage (CCS). This is typical of most stabilization scenarios; the differences lie in timing (later in this century in the MESSAGE model, earlier in the MiniCAM model) and amount. The other essential point in Fig. 4 is the scale. Current (2008) emissions are 30,000 Mt (million metric tons) of CO 2 per year, or 30 Gt/y. A meaningful emissions reduction must be a few Gt CO 2 /y. The role anticipated for CCS in the MiniCAM model grows to 10 Gt CO 2 /y by the middle of the 21 st century, while the MESSAGE model anticipates 1 to 2 Gt CO 2 /y by the same time. To put these values in perspective, consider that the density of CO 2 under typical geologic storage conditions will be about 600 kg/m 3. These CCS rates thus correspond to 1.7 to 17 Gm 3 /y, or 29 million to 290 million barrels per day. In 2007, the world produced about 85 million barrels of oil per day. Thus a meaningful CCS industry would have to be of size comparable to the current oil and gas industry by the middle of this century. This is an extraordinary challenge by any measure. Fuel Switching About a quarter of anthropogenic carbon dioxide is emitted by burning coal. Most of that combustion is done to produce electricity. A 1000 MW coal fired power plant will produce about 7.6 Mt (million metric tons) CO 2 per year. In contrast a 1000 MW gasfired power plant will produce 3.9 Mt CO 2 /y. This simple difference is anticipated to be the chief driver for fuel switching for electricity generation in a carbon-constrained world. The economics of fuel switching for power generation are already complex. A carbon price will only make the matter more complicated. Increased demand for gas for new generating capacity will drive up its price. And burning methane still produces CO 2, so a gas-fired plant will still be at a disadvantage compared to nuclear or hydroelectric. Meanwhile existing coal plants may see lower coal prices but be faced with the need to retrofit carbon capture and storage technologies. The capture process requires a significant energy input, about a third of plant output for the monoethanolamine (MEA) absorption process. New technologies may reduce the energy consumption for capture by a factor of two, but the power required will still be significant. Gas turbines may become the technology of choice so that existing plants can still produce at nameplate capacity. The choice of fuel for new generating capacity will be the most difficult to forecast. But an increase in demand for gas appears inevitable in a carbon-constrained economy. 5

6 While gas is much heralded as a clean fuel, switching to it should not be regarded as a quick solution for GHG emissions. This is because of the sheer scale of current fossil fuel consumption. Installed electricity generating capacity worldwide in 2007 is 4000 GW. To reduce carbon dioxide emissions by 5 Gt CO 2 /y a fraction of the amount needed to stabilize atmospheric concentrations of carbon dioxide industry would have to construct 1350 gas fired plants each of 1000 MW capacity, instead of constructing 1350 coal fired plants of the same capacity. This corresponds to one third of total installed capacity today. On the other hand, the same amount of GHG mitigation would be achieved if industry built 730 coal-fired plants of 1000 MW each with capture and storage of the CO 2 (assuming the power to run the capture process was produced without fossil fuels.) Geologic Carbon Storage Storing CO 2 produced by fossil fuel combustion must play a significant role in mitigating GHG emissions, because strong demand for these fuels will continue for the foreseeable future. This is because energy consumption is strongly correlated to quality of life, and because energy in the modern world is predominantly supplied by fossil fuel. These simple realities, combined with the vast scale of the modern global economy, are why leaving the coal in the ground is an unrealistic strategy. Geologic formations depleted hydrocarbon reservoirs and deep, saline aquifers are one of the very few places on Earth with enough volume to accommodate substantial amounts of CO 2. Moreover the know-how for transporting and injecting CO 2 into deep formations already exists within the oil industry. In effect, an off the shelf service could be provided. Hence geologic storage is generally regarded as an essential technology for any serious effort at GHG mitigation. The obvious first choice for geologic storage is a mature oil reservoir from which additional oil can be recovered if CO 2 is injected. The revenue from the incremental oil will be extremely important to the economics of early sequestration efforts. However, operating a simultaneous EOR/sequestration project is not the same as a traditional EOR project. The recycling of produced CO 2 and the alternate injection of water and CO 2 slugs are important for the reservoir engineering of this type of EOR. Both practices would work against a sequestration project, however. Water would occupy pore space that could be used for sequestration. Recycling CO 2 reduces the amount of new CO 2 that could be sequestered, and would greatly complicate the accounting for emission credits. Beyond these practical difficulties is a more fundamental limitation. The cumulative amount of CO 2 to be stored is measured in the hundreds of gigatons. If geologic storage were to handle, say, 100 Gt CO 2 of the total, then a pore volume of about one trillion barrels would be needed. This is the same as the volume of original conventional oil in place on Earth. Adding in the fact that many fixed sources of CO 2 are not near oil fields, it is clear that other sedimentary rocks will be needed. This is the reason that deep saline aquifers command so much attention in the CCS community. 6

7 A carbon-constrained world thus presents a remarkable opportunity for the oil and gas industry. On one hand demand for oil and gas is unlikely to decrease, even if a carbon price or carbon tax makes those fuels more expensive. The transition to other fuels for the world s economy will take decades. At the same time, a carbon storage industry will be needed to redress the consequences of using fossil fuels. That industry will have to grow to the size of the modern oil and gas industry if mitigation is carried out at meaningful levels. ENGINEERING ASPECTS OF GEOLOGIC STORAGE IN AQUIFERS Several key aspects of the process of injecting CO 2 into an aquifer will strongly influence the operation and regulation of geologic storage. CO 2 will be buoyant; geologic formations are not always leakproof; certain rock/fluid properties promote immobilization of CO 2 ; and injecting CO 2 will necessarily displace brine. In the prototypical geologic storage project, CO 2 is captured from a flue gas stream, compressed and injected into a deep saline aquifer. Under typical conditions in the storage formation, the density of the CO 2 phase will be about 60% the density of the brine initially occupying the aquifer. This means the CO 2 will be buoyant. Given the opportunity, the CO 2 will rise. Consequently the single most urgent question for the design and regulation of geologic storage is whether the CO 2 will leak out of the storage formation. If the CO 2 migrates during or after injection and encounters an escape route a conductive fault or an old wellbore or if the integrity of the impermeable formation above the CO 2 is compromised, then buoyancy guarantees that some CO 2 will escape. The challenge is to assess the probability of a CO 2 plume encountering an escape route, and if it does, to estimate the resulting flux. Many formations within the Earth s crust have trapped buoyant hydrocarbons for very long times, far longer than the storage time needed for CO 2. The existence of the oil and gas industry is proof that subsurface formations can be reliable storage containers. Unfortunately, it is also true that conductive paths are widespread in the Earth s crust. In fact the oil and gas industry provides examples of the poor containers as well as the reliable ones. The best exploration strategy for the first seventy five years of the oil business was to find surface seeps of oil or natural gas, then drill them. Much of the hydrocarbon ever produced at depth simply rose under buoyancy until it reached the surface, where it dissipated into the atmosphere or surface water. Besides naturally occurring leakage paths, great numbers of wells have been drilled into the Earth s crust, each creating a potential breach of sealing formations. The formation and exploitation of hydrocarbon reservoirs teaches a clear lesson. Some, but not all, of the Earth s crust is a very secure place to store CO 2. The problem for the geologic storage industry will be to distinguish the secure places from the rest. Establishing the security of sealing structures adds cost to the storage enterprise, and there will be considerable pressure to limit costs. 7

8 After CO 2 injection ends, the CO 2 plume will continue to migrate because of its buoyancy. Migration will continue until an impervious boundary is encountered. If the CO 2 is injected at the top of an anticline, no migration will occur after injection unless the seal is compromised. On the other hand, migration is inevitable if the injection takes place downdip on a structure, or if the injection is deliberately limited to the lower section of a thick interval. Somewhat counterintuitively, two processes occur during migration that increase the security of storage. CO 2 is soluble in brine (about 5 weight % under typical storage conditions). Thus when the leading edge of the migrating plume encounters brine not yet saturated with CO 2, dissolution into the newly contacted brine occurs. This increases the density of the brine slightly. The density increase means the brine saturated with CO 2 cannot escape upwards from the storage formation. The second process occurs as the leading edge of the plume rises. Because the fluids are relatively incompressible, brine must fill in at the trailing edge to replace the migrating CO 2. This invasion is known as imbibition, and as the water/co 2 interface advances at the level of individual pores in the rock, it causes the CO 2 phase to break up into tiny droplets. The droplets are of different sizes, occupying individual pores to as many as a few hundred pores within the rock. Whatever their size, once disconnected the droplets are permanently held in place by capillary forces. This phenomenon is applies generally to any nonwetting phase (oil, natural gas, CO 2 ) being displaced by a wetting phase (brine). These processes are known as dissolution trapping and residual phase trapping. Once held in these forms, CO 2 has negligible probability of escaping the host formation. Hence any process that transfer CO 2 from the connected, mobile phase into either of these forms increases the long-term security of storage. Thus, some migration is a good thing, as long as it does not enable the CO 2 plume to encounter a leakage path. Because some of the moving CO 2 is left behind as residual, it follows that migration of a finite plume cannot continue indefinitely. Eventually the saturation of still mobile CO 2 will approach residual, and the migration will halt. The inject low and let rise strategy for maximizing security of storage takes advantage of this phenomenon. An obvious fact with significant consequences is that all pore space in the deep subsurface is already host to one (or more) fluid phases. If CO 2 is injected into an aquifer then the brine initially residing there must be displaced. Simple calculations indicate that the brine can be displaced significant distances (hundreds of meters) from an initial location tens of kilometers from the injection well. Thus displacement of brine at a local or regional scale where many storage projects are underway may not substantial. The movement of brine may prove to be a bigger issue than the risk of CO 2 leakage. The probability of the CO 2 plume reaching underground sources of drinking water (USDW) is small for a prototypical storage scheme. This is because of the long vertical distance the CO 2 would have to travel to reach USDW and the large attenuation (dissolution and residual trapping) that will accompany that travel. But the probability could be 8

9 considerably larger that the same USDW be contaminated by brackish or saline water displaced by the CO 2 plume. When considering a geologic storage project, the US EPA is likely to be concerned with the contamination of USDW, whether directly by CO 2 or indirectly by displaced brine rather than buoyant migration of CO 2 per se. Furthermore the regulatory authority for groundwater already exists, while regulations for CO 2 are still pending. RISK ASSESSMENT AND RISK MINIMIZATION Fuel switching and sequestration of carbon dioxide in geologic formations must be implemented at a very large scale to mitigate emissions of greenhouse gases. Thousands of sequestration projects will be needed, so a prerequisite to implementation is a project certification framework that is simple to apply. The most challenging aspect of such a framework is the assessment of risk associated with CO 2 migration after closure. The physics of the CO 2 /brine/rock intereaction is distinctive and strongly influences the balance between migration and immobilization of CO 2. Regulations informed by these physical realities are likely to be much more cost-effective. This is an important consideration, as both cost and security of storage will be critical factors in public acceptance. However regulations that allow for concepts such as effective trapping entail greater flexibility and uncertainty than is customary in waste disposal operations. Detailed numerical simulations of several CO 2 storage schemes in deep saline aquifers were conducted. The simulations cover two periods: the injection phase (typically thirty years) followed by one thousand years during which the injected CO 2 can move by buoyancy. Three schemes are compared: the standard approach (injection of CO 2 into the full thickness of an aquifer), the inject low and let rise approach (injection of CO 2 only into the lower part of an aquifer), and a surface dissolution approach (captured CO 2 is dissolved into brine in surface facilities, and the CO 2 -saturated brine is injected into the aquifer). For each case, the primary mode of CO 2 immobilization (brine dissolution, residual phase trapping, hydrodynamic trapping, stratigraphic trapping) is identified and the period of time needed to achieve that immobilization is estimated. Each mode of trapping presents unique requirements for monitoring and verification. The standard scheme of CO 2 injection will have the least operating and capital costs. The inject-low-and-let-rise scheme will have somewhat larger costs, as it may require more wells to be constructed. Because it requires many brine extraction wells, the surface dissolution approach would have about 50% greater operating costs and 100% greater capital costs than the standard scheme (Burton and Bryant, 2007). The long-term security of the CO 2 increases as the cost of the scheme increases. For the standard scheme, much of the CO 2 is held beneath a sealing stratum (a situation analogous to a hydrocarbon reservoir), Fig. 5. The CO 2 will remain indefinitely (diffusion and dissolution into underlying brine would occur over tens to hundreds of millenia) as long as the integrity of the seal is intact. If the seal is breached, the CO 2 will escape. This scheme therefore imposes a long-term obligation to monitor the integrity of the seal. The 9

10 key difficulty with this approach is that the Earth s crust is intrinsically leaky. Convincing regulators and the public that a particular seal will remain intact for very long periods of time will be difficult. The inject-low-and-let-rise scheme, Fig. 6, relies on the buoyancy of CO 2 relative to brine to cause controlled migration of the plume after injection (Kumar et al., 2005). The migration necessarily increases residual phase trapping within the plume and increases dissolution trapping at the leading edge of the plume. The migration leads to nearly complete immobilization of the injected CO 2 over a few centuries to a few millennia. The immobilization is independent of the integrity of any seal. This scheme requires monitoring to ensure that the migration is proceeding as planned. The surface dissolution scheme, Fig. 7, completely eliminates the chance of buoyancydriven migration of the CO 2 plume (Burton and Bryant, 2007). It therefore requires no monitoring and in principle could be regulated as wastewater disposal is regulated. No additional time is required for immobilization after injection ends. The footprint of the CO 2 plume is much larger than the CO 2 plume for the other schemes. However, the footprint of brine displacement is 30% smaller than the standard approach. The benefit depends on whether groundwater or bulk phase CO 2 is the subject of regulation. The inject-low-and-let-rise scheme is based on physical principles (gravity, capillary pressure) that are guaranteed to apply. In essence, it states that controlled migration after injection ends increases the degree of immobilization that requires no further monitoring. If regulators were to impose a strict no-migration requirement, analogous to the Underground Injection Code for wastewater, operators would not be able to implement this scheme, despite its advantage relative to the standard scheme. The issue here is not that one scheme is better than the other, but that regulation could inadvertently favor one over another. Similarly, were regulators to prescribe monitoring measurements intended to ascertain mobile CO 2 saturation, an exemption should be made explicitly for the surface dissolution scheme. The self-limiting nature of certain types of CO 2 plume migration leads to the concept of effective trapping. The term connotes migration of the CO 2 from the original storage volume within a formation that does not cause harm, Fig. 8. The impact to be considered would primarily be in other regions of the subsurface such as hydrocarbon reservoirs, mines, underground sources of drinking water, extending to the near subsurface and then to the atmosphere (Oldenburg et al., 2007, 2008). This concept is deliberately more flexible than the familiar containment or no-migration approaches. To regulate storage with this concept would demand a new level of understanding and transparency. The notion of effective trapping is appealing, but it does demand faith on the part of stakeholders that the process will continue just as the operators claim. The question of time scale for achieving a certain level of immobilization is likely to emerge. Here lies another challenge of communication between the technical and legal communities. For a typical aquifer, buoyancy driven flow is slow. As a result, it will take centuries, perhaps 10

11 millennia for the mobile phase saturation to become immobilized, Fig. 9. Most human institutions do not last that long. CONCLUSIONS Large-scale reduction of emissions of carbon dioxide will require large-scale geologic storage, raising new issues for environmental law and regulation, but simultaneously creating a world-scale opportunity for the oil and gas industry. This is because deep sedimentary rocks are one of the few places with enough volume to hold decades of greenhouse gas emissions. Moreover the technology and the know-how to transport fluids around the globe at large scale and inject them into the Earth s crust already exist in the oil industry. The industry can thus provide an almost off-the-shelf CO 2 sequestration service starting as soon as a regulatory framework is in place, and a price for carbon exists. GHG mitigation will also add substantial impetus to the natural gas industry, simply because gas produces a given amount of heat with less carbon. Continued growth in electricity demand will present a difficult optimization problem in a carbonconstrained world: new coal-fired capacity with capture and storage vs new gas-fired capacity without capture, while retrofitting existing coal plants with capture and storage technology., Carbon dioxide is produced in far greater quantities than other byproducts of fuel consumption, so existing regulations for air quality are not good models for regulating carbon dioxide. Carbon dioxide is benign compared to high level radioactive waste, and the approach taken to permitting and regulating the latter will not be effective for CO 2. Subsurface storage of carbon dioxide involves qualitatively different physical phenomena than wastewater disposal, so the Underground Injection Code for regulating the latter is not a good model for regulating the former. A new approach to regulation is needed. In particular it should acknowledge the engineering and physical realities of geologic storage and allow for geologic uncertainties. It must also encourage technical innovation and allow a profitable sequestration industry to emerge. The advent of a carbon tax, a cap-and-trade system for carbon dioxide emissions, or some other mandate to reduce emissions should not be automatically regarded as a threat to the oil and gas industry. On the contrary, while the cost of producing and using fossil fuels would rise, demand is unlikely to fall rapidly because alternatives will not be available at the needed scale for decades. Thus continuing to use oil and gas (and coal) while mitigating the effect of combustion is very likely to be the path followed. This should be a win-win for the oil and gas industry, as it would continue to extract fluid hydrocarbons while expanding to handle the return of carbon, now in the form of CO 2, to the earth s crust. REFERENCES Burton, M. and Bryant, S. Eliminating Buoyant Migration of Sequestered CO 2 through Surface Dissolution: Implementation Costs and Technical Challenges, SPE , 11

12 Proceedings of the 2007 SPE Annual Technical Conference and Exhibition Anaheim, California, U.S.A., November 2007 Intergovernmental Panel on Climate Change (IPCC) Special Report on Carbon dioxide Capture and Storage, pp. 45, Downloaded April 2007 from Kumar, A., Ozah, R., Noh, M., Pope, G.A., Bryant, S., Sepehrnoori, K., and Lake, L.W., "Reservoir Simulation of CO 2 Storage in Deep Saline Aquifers," Society of Petroleum Engineers Journal, Vol. 10, No. 3, pp , September Oldenburg, C. and Bryant, S. Certification Framework for Geologic CO 2 Storage, National Energy Technology Laboratory 6 th Annual Conference on Carbon Capture and Sequestration, Pittsburgh, PA, 7-10 May, Oldenburg, C., Nicot, J.-P., and Bryant, S. "Case studies of the application of the Certification Framework to two geologic carbon sequestration sites." Proceedings of Greenhouse Gas Technologies 9th Conference, Washington DC, Nov , Ozah, R. C., Lakshminarasimhan, S., Pope, G. A., Sepehrnoori, K., and Bryant, S. L., Numerical Simulation of the Storage of Pure CO 2 and CO 2 -H 2 S Gas Mixtures in Deep Saline Aquifers, SPE 97255, Proceedings of the 2005 SPE Annual Technical Conference and Exhibition, Dallas, Texas, October 9-12, Parmesan, C. Ecological and Evolutionary Responses to Recent Climate Change, Annu. Rev. Ecol. Evol. Syst , pp Rosenzweig, C., D. Karoly, M. Vicarelli, P. Neofotis, Q. Wu, G. Casassa, A. Menzel, T. L. Root, N. Estrella, B. Seguin, P. Tryjanowski, C. Liu, S. Rawlins, and A. Imeson, Attributing physical and biological impacts to anthropogenic climate change, Nature, May 15, 2008, 453, pp

13 Atmospheric CO 2 Concentration, ppmv Mauna Loa, annual avg Siple Year Figure 1 Atmospheric concentrations of carbon dioxide from direct measurements at Mauna Loa, Hawaii (annual average of monthly data) and from ice cores at the Siple Station in Switzerland (Data downloaded 11 October 2008 from and from Concentration rose gradually during the 19th century, then more rapidly in the first half of the 20th century, and much more rapidly in the latter half of the 20th century Emission from Fossil Fuel, Mt C/y Year Figure 2 Fossil fuel consumption started increasing after the Industrial Revolution. In the last half century, global consumption has increased five-fold. The corresponding emission of carbon dioxide into the atmosphere is now approaching eight billion tons of carbon per year (30 Gt CO 2 /y). This is larger than the net natural fluxes of carbon between atmosphere and land surface (vegetation and soil) or between atmosphere and surface ocean. Data downloaded from 11 October

14 Figure 3 Atmospheric concentrations of carbon dioxide (blue line) inferred from ice cores from Antarctica varied between 180 parts per million (ppm) and 280 ppm during the last four hundred thousand years. The corresponding deviation of average surface temperature from the long-term mean (red line) tracks the carbon dioxide concentration. (Data downloaded 26 May 2008 from Only in the very recent past, geologically speaking, has the carbon dioxide concentration moved outside its long-term limits. This is widely believed to be the consequence of fossil fuel combustion, Fig

15 Figure 4 Greenhouse gas stabilization scenarios for the 21 st century (examples shown here are the MESSAGE model and the MiniCAM model; cf. IPCC, 2006) vary in their assumptions about development, but nearly all of them envision significant contributions from fuel switching (replacing coal fired electricity generation with gas fired) and carbon dioxide capture and storage (CCS). Mitigating greenhouse gas emissions thus represents very substantial opportunities for the oil and gas industry. 15

16 shale CO 2, S g > S g,r CO 2 cannot move Brine cannot displace CO 2 Will not form residual saturation Mass transfer only at CO 2 /brine interface aquifer Figure 5 The standard approach to storing CO 2 in a geologic formation involves injecting bulk phase CO 2. The CO 2 collects at the top of the structure. The CO 2 exists as a mobile phase, so if the overlying seal ever loses its integrity, the CO 2 will escape. This mode of storage will require monitoring indefinitely. shale CO 2, S g ~ S g,r aquifer CO 2 moves finite distance Contacts more rock - More dissolution - More residual saturation Self-limiting movement Figure 6 The inject low and let rise approach to CO 2 storage takes advantage of the buoyancy of CO 2 relative to brine in typical deep saline aquifer conditions. A finite volume of CO 2 is injected downdip and/or in the lower part of an interval. At the end of injection, the buoyant CO 2 rises. As the leading edge of the plume moves, it encounters fresh brine, into which it dissolves. Meanwhile at the trailing edge, brine imbibes into the plume, trapping CO 2 as a residual phase. 16

17 Coal-fired Power Plant CO Pressurized Mixing 2 Brine Dense CO2 Saturated Brine ft Brine Aquifer Figure 7 Schematic of the brine surface dissolution strategy (Burton and Bryant, 2007) includes pumps for the brine extraction, brine injection, and compression of the captured CO2 stream. Brine is lifted from the target aquifer and pumped to an adequate mixing pressure. The two fluids are mixed until the CO2 dissolves, and then the saturated brine is re-injected. Figure 8 (a) Generic geologic cross section of potential GCS site showing reservoir and sealing formations, faults, wells, USDW, and near-surface and surface environments. (b) Generic cross section with Certification Framework (Oldenburg et al., 2008) source and compartments overlaid. Compartments key is as follows: ECA = Emission Credits and Atmosphere; HS = Health and Safety; NSE = Near-Surface Environment; USDW = Underground Source of Drinking Water; and HMR = Hydrocarbon and Mineral Resources. 17

18 CO 2 Stored in Various Forms Total CO 2 Sequestered Percentage CO 2 as Free Gas Percentage CO 2 as Residual Gas Percentage CO 2 in Aqueous Phase ,000 10,000 Time, years Figure 9 As reported by Ozah et al. (2005), the inject low and let rise strategy ensures that after injection of CO 2 ends, the fraction of the stored CO 2 that is mobile (exists as a free gas in the formation) decreases monotonically. The fraction that is immobilized, either as a discrete small droplets of residual phase, or dissolved into brine which thereby becomes slightly dense than the original brine Total CO 2 Stored, million metric tons 18