Sequestration in the Oceans



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 41
Sequestration in the Oceans

OCR for page 41
This page in the original is blank.

OCR for page 41
Direct Injection of Carbon Dioxide into the Oceans PETER G. BREWER Monterey Bay Aquarium Research Institute It is now almost 40 years since we made our first measurements of carbon dioxide (CO2) in seawater, and the science has changed enormously during that time. Last week, I was at sea carrying out a small-scale carbon sequestration experiment. For about the last five years, my laboratory group and our colleagues have been carrying out deep ocean experiments; they are difficult, but fun, and they raise all kinds of important questions. I want to talk with you about some of our results. Instead of dwelling on policy issues, I will focus on the numbers, on the technology, and on the present level of scientific understanding. A 1998 cover story in Environmental Science and Technology, a journal of the American Chemical Society, raised the question of whether we should actively dispose of CO2 in the oceans (Hanisch, 1998). That really begs the question, because we already do. Our current, de facto policy for disposing of carbon dioxide, both in the United States and internationally, is to dispose of it first in the atmosphere. We recognize that the atmosphere then moves across the surface of a large-scale saline “aquifer” containing dissolved carbonate minerals, and we neutralize the CO2 by a reaction with carbonate ion dissolved in seawater, thus converting it to sodium bicarbonate. This aquifer covers 70 percent of the Earth’s surface, and the reaction with the alkalinity of surface ocean waters is the primary modifier of the increase of CO2 in the atmosphere. Ocean circulation then transports these CO2-modified surface waters to water mass conversion regions and subduction zones. By these convective and sinking processes, the fossil-fuel signal is mixed into the abyssal flows. The mean circulation time of oceanic deep waters is about 550 years, and every year about 30 percent of atmospheric fossil fuel CO2

OCR for page 41
emissions are taken up by the ocean. In the very long run, about 85 percent of all CO2 emissions will reside there (IPCC, 1990). We now have more than a hundred years of experience with this “technology,” and the numbers are very large. The volume of the oceanic “aquifer” is about 1021 liters. A very large fraction of the emissions from the early part of the twentieth century are now in deep waters, well along on the exchange path between the upper ocean and the deep ocean; the oceanic fossil-fuel signal has reached a depth of >1,000 meters. The front is moving down at about 1 meter a month (Wallace, 2001). Thus the distinction between ocean “uptake” and ocean “disposal” has become increasingly blurred. Ocean CO2 uptake (in effect surface ocean disposal) is now about 20 to 25 million tons of CO2 per day, of which the U.S. contribution is about 6 million tons of CO2 per day. Like it or not, that is our de facto carbon dioxide policy, and it has been for decades. Every U.S. citizen emits the equivalent of about 120 pounds of CO2 a day, and about a third of that goes rather quickly into the ocean. One significant problem is that we are “disposing” of this CO2 in the surface waters of the ocean where most of the marine life lives and where reef-building corals are. We have already lowered surface ocean pH by about 0.1 pH units, and, if the Intergovernmental Panel on Climate Control “Business as Usual” scenario is followed, by the end of this century, we will have lowered carbonate ion concentrations in surface ocean waters by >50 percent (Brewer, 1997). This will significantly affect the calcification process in coral reefs. Moreover, during the atmospheric residence time of the released CO2, it creates the well known global warming signal. The combined effects of heat and lower pH are causing serious concerns for coral reef systems (Table 1). DIRECT OCEAN SEQUESTRATION Cesare Marchetti (1977) made the first suggestion of direct carbon sequestration in the ocean about 25 years ago. Since then, there have been numerous conferences to study the problem and discuss theoretical analyses. However, only about five or six years ago a number of us decided to initiate small-scale field experiments. We realized that all we had to go on was, in effect, sketches and cartoons of the process—not because contributors to the field were bad or ignorant, but because nobody had any actual experience (e.g., Figure 1). Journal articles were illustrated with sketches and cartoons, leading to all kinds of confusion. The sketch that appears in the 1998 Environmental Science and Technology article shows blocks of dry ice being dropped into the surface ocean—a forbiddingly expensive idea. Another sketch shows the ocean floor with some kind of reactor and a pile of hydrates. These sketches offer intuitive, but possibly confusing, images of how ocean carbon sequestration might work. It was clearly time to carry out real

OCR for page 41
TABLE 1 The Evolving Chemistry of Surface Seawater under “Business as Usual” Time/years pCO2 (µatm) Total CO2 (µmol kg−1) pH HCO3− (µmol kg−1) CO32− (µmol kg−1) H2CO3 (µmol kg−1) 1800 280 2,017 8.191 1,789 217 10.5 1996 360 2,067 8.101 1,869 184 13.5 2020 440 2,105 8.028 1,928 161 16.5 2040 510 2,131 7.972 1,968 144 19.1 2060 600 2,158 7.911 2,008 128 22.5 2080 700 2,182 7.851 2,043 113 26.2 2100 850 2,212 7.775 2,083 97 31.8 Under IPCC “Business as Usual,” the pH of surface seawater drops by 0.4 pH units by 2100. CO3− in surface water drops by 55 percent from preindustrial values. It will be hard to meet even these goals. Fossil fuel CO2 is now a major ion of seawater. Source: IPPC, 1990.

OCR for page 41
FIGURE 1 Options for direct ocean disposal of CO2. Disposal scenarios that are the focus of current research include droplet plume and dense plume dissolution, dry ice and towed pipe dispersion, and isolation as a dense lake of CO2 on the sea floor. Towed pipe and droplet plume scenarios may offer the best approach in the near future. Source: Hanish, 1998. experiments, and we were fortunate to have access to modern, remotely operated vehicles (ROVs) to attempt this. The 1998 President’s Council of Advisors on Science and Technology Energy R&D Panel recommended storing CO2 as a clathrate hydrate on the seafloor. At high pressure and low temperature, CO2 will react with water to form an ice-like solid (CO2.6H2O), which is denser than seawater. This would aid enormously in sinking CO2 to the ocean floor and, it was presumed, greatly extend its time there. CO2 can indeed form a hydrate, and we now have extensive experience of working with this property (Brewer et al., 1999). The nucleation and growth rates can be capricious, but liquid CO2 undergoes a transformation to the solid hydrate form on the seafloor at a depth of 3,600 meters (Brewer et al., 2002).

OCR for page 41
The first question we addressed was the release of CO2 from a pipeline into the mid-depth ocean (e.g., between 400 and 1,500 meters). This scheme formed the basis of a very fine modeling effort carried out by the Bergen group in Norway (Alendal and Drange, 2001; Drange et al., 2001). The hypothesis was that, when CO2 was released in liquid form it would quickly break up into small droplets. Because at these depths liquid CO2 is less dense than seawater, the droplets would float upward and dissolve rather quickly. The Norwegian study showed that, if the location and depth of release were carefully selected, the water masses labeled with this excess dissolved CO2 would be advected to the North Atlantic deep-water formation regions and transported into the abyssal flows. This would ensure sequestration for >250 years before reventilation of the water masses in the Antarctic circumpolar flows. First Experiments Critics of this approach—and I was a bit skeptical earlier—suggested that the dissolution might not be quite that easy. For our study, we took an ROV fitted with a high-definition TV camera and attempted direct imaging of the release, rise, and dissolution sequence. In effect, we had a 7,000-pound vehicle on an almost one-kilometer-long pendulum, subject to continuous, and variable, ocean forces. We requested that the pilots fly to the release point, release a small quantity of liquid CO2, and follow this during upward transit over hundreds of meters, while the ROV takes images of the droplets with a precision of a tenth of a millimeter. This was excruciatingly difficult—painstaking, classical, hard work. Every release required about an hour of intense concentration, as well as hand-eye coordination. But we actually pulled it off, and we were able to track the changing size of droplets in a classical manner (Brewer et al., 2002). We were able to show that the modeling done by the Norwegian group, and also laboratory pressure vessel studies in Japan (Aya et al., 1997), are probably correct. CO2 released in the ocean at a depth of about 800 meters (4.4°C) will dissolve at a rate of about 3 µmol/cm2/sec. This means that for droplets initially about 1 cm in diameter, about 90 percent of the dissolution occurs within 30 minutes and within 200 meters of the release point. That is very close to the modeling result from the Bergen group. When CO2 is injected into the ocean at a relatively shallow depth, both observations and modeling studies show there is a good chance that some of it could return to the atmosphere. Flow in the ocean is primarily along isopycnal (constant density) surfaces, and a key diagnostic tool is where a particular density layer is ventilated, or exposed, to the atmosphere. The North Pacific Ocean has the densest seawater exposed at northern latitudes. Waters deeper and denser are exposed to the atmosphere thousands of miles away, and several hundred years later, in the Antarctic region. Over much of the North Pacific, this isopycnal surface lies at a depth of about 600 meters. Thus, our simple field experiment

OCR for page 41
illustrating dissolution of a plume in the 800- to 600-meter depth zone probably represents the shallowest depth at which effective ocean CO2 sequestration should be considered. The Fate of CO2 Hydrates On the basis of these results, we next asked how important the hydrate skin was in controlling the outcome of the experiment. If we had done the experiment in the warmer deep waters of the Mediterranean, would the results have been different? When we store hydrates on the seafloor, should they be in a stable form? Or would they dissolve? Clearly the deep ocean thermodynamic conditions of temperature and pressure favor hydrate formation, but it is also essential that chemical saturation occur. Deep ocean waters are approximately 500-fold undersaturated with respect to dissolved CO2. We thus decided to do an experiment to measure directly the oceanic dissolution rates of CO2 hydrates themselves—testing the idea of hydrate storage on the ocean floor (Rehder et al., in press). Working with colleagues from the U.S. Geological Survey (USGS) and Lawrence Livermore National Laboratory, we fabricated both CO2 and methane hydrates in the USGS laboratory. These specimens were squeezed at high pressure at Lawrence Livermore National Laboratory into dense solid units about the size of 35-millimeter film cassettes, placed in a specially designed pressure vessel under about 15.5 MPa methane pressure, and packed in ice. This unit was then driven down to our base in Moss Landing, California, and taken out to sea. The specimens were transported by the ROV Ventana to the ocean floor at 1,028 meters (3.6°C) along with a time-lapse camera to record the results. By exceptionally dexterous robotic manipulation, the pressure vessel was opened, and the hydrates were exposed on the seafloor and positioned so that good images could be recorded. Within a few hours, it was clear that both the methane and the carbon dioxide hydrates were dissolving. After appropriate corrections, we found that the effective release rate of CO2 to ocean waters was very similar to the release rate of the liquid droplets mentioned above. The shrinkage rate of the solid diameter was 9×10−2 µm/sec. We were able to observe an apparent correlation between dissolution rate and current velocity. The data were of remarkable quality. The methane hydrate also dissolved, but at a rate about 10.5 times slower. This experiment taught us something valuable about the lifetime of hydrates of all kinds in the ocean and provides a basis for making powerful, simple predictions based on saturated-boundary theory. The ratio of both the solubility and the observed hydrate dissolution rates of CO2 to methane is about 10.5:1. Simply stated, the control on dissolution rate (and the limiting factor of hydrate lifetimes in the ocean) is the existence of a thin,

OCR for page 41
saturated, molecular-boundary layer controlled by diffusion. Higher velocities in the bulk fluid reduce the thickness of this layer and accelerate dissolution. Making and Testing Predictions If we were to go down to a depth of 3,000 meters, we predicted that we would get a reduction in dissolution rate by about a factor of 2, based on the changing solubility (or ease of formation) of the hydrate. We also did that experiment, devising techniques for transporting CO2 to below 2,750 meters, to a depth where the fluid is gravitationally stable. Using one of our newer vehicles, we flew a small amount of CO2 to a depth of 3,600 meters. We punched a small hole in the seafloor so the CO2 wouldn’t roll away and then inserted a pH electrode directly into the mass of liquid. This neither broke the liquid surface, as in pricking a balloon, nor caused simple elastic stretching of the surface. Rather, as the liquid surface deformed microscopic cracks occurred, which were quickly annealed with hydrate, as both water and CO2 flowed into the cracks and combined to renew the skin with remarkable effectiveness. Having thus made a water pocket inside the blob of CO2, we then locked the electrode in place and waited for half an hour. Slowly, the CO2 dissolved into the water; the dissolution rate is given by the observed drop in pH. The result was almost precisely a factor of 2 slower than it was at 1,000 meters—as predicted by thermodynamic-equilibrium and saturated boundary-layer theory. We are now beginning to understand this process quite well at the molecular scale. Our Japanese colleagues conducted a similar experiment earlier in the laboratory and described the hydrate-film rebuilding process (Aya et al., 1997). Thus, in several classic experiments over the course of the last few years, we’ve determined that CO2 in all forms does dissolve at significantly high rates in the ocean. It reacts quickly with water to form carbonic acid and then with carbonate ion to add to the pool of dissolved bicarbonate in ocean waters. The tracer plume that would result from disposal would be detectable by techniques common to recovering the fossil-fuel signal from oceanic observations. Biological Impacts Some obvious questions arise at once. What is the cost? And what are the environmental impacts? We are now beginning to address these questions directly. A colleague of mine, his postdoctoral students, and I are carrying out experiments with CO2-biological interactions right now. We emplace about 20 liters of liquid CO2 in a small corral on the seafloor at 3,600 meters. The corral holds about the same amount of CO2 as an individual U.S. citizen puts into the ocean every day via the atmosphere-ocean gas-exchange process. We then set up a number of experimental enclosures containing a variety of marine animals

OCR for page 41
captured locally at varying distances from the small CO2 source (Barry et al., in press). We measure the pH from a set of recording instruments, and we examine the physiological responses of the marine animals to the lower-pH plume that is formed. Because the velocity signal is dominated by the tidal ellipse, the plume passes over any one site or specimen about twice a day. The work is going very well, but it raises all kinds of interesting issues and technical difficulties. Occasionally, a small amount of CO2 may wash over as a result of ROV thrust—one of the problems of using ROVs near a liquid—or as a result of hydrate formation, thus creating large volume changes. It is very common for deep-sea fish to observe our work closely. They are curious, and they come close up to our experiments, apparently unperturbed, for long periods of time. In a one-month study, we simply used a time-lapse camera to record the fate of a 20-liter corral of CO2 on the seafloor. Although various animals moved closeby, there was no recorded interaction or perturbation in their behavior. The pool of CO2 simply slowly dissolved, with no detectable biological response. SUMMARY These small-scale, very careful experiments are revealing the rate of CO2 dissolution in the ocean, its physicochemical properties, and its environmental impact. They suggest many possibilities for safe and effective oceanic disposal of CO2. We hope we will be able to make some objective evaluations about the feasibility and ethics of this option. Many aspects of this problem have yet to be investigated. As every participant has said during these meetings, a large part of the cost of sequestration is in the initial capture of CO2. A number of people have suggested that one solution to minimize costs is simply to take the CO2-nitrogen mixture resulting from combustion and inject it into the ocean without chemical separation (Saito et al., 2000). At a depth of about 300 meters, the ratio of CO2 solubility to nitrogen solubility changes significantly, with strongly preferential dissolution of CO2. Thus, a bubble stream would quickly evolve into a pure nitrogen gas phase and a dense CO2 rich aqueous phase, which could be piped to great depth. We plan to conduct experiments on this process. Many other issues have been raised, such as the enormous amount of fluids involved, suggesting a very large-scale engineering enterprise. Ken Caldeira and Greg Rau (2000) have examined the use of crushed limestone to provide a carbonate buffer for the CO2-rich fluids, thereby permitting disposal at much shallower ocean depths. Over the next few years, we plan to conduct small-scale experiments to move this science ahead and provide objective data about these extraordinary problems.

OCR for page 41
REFERENCES Alendal, G., and H. Drange. 2001. Two-phase near field modeling of purposefully released CO2 in the ocean. Journal of Geophyical Research 106(C1): 1085–1096. Aya, I., K. Yamane, and H. Nariai. 1997. Solubility of CO2 and density of CO2 hydrate at 30 MPa. Energy 22: 263–271. Barry, J.P., B.A. Seibel, J.C. Drazen, M.N. Tamburri, K.R. Buck, C. Lovera, L. Kuhnz, E.T. Peltzer, K. Osborne, P.J. Whaling, and P.G. Brewer. In press. Direct ocean carbon sequestration: observations of biological impacts during small-scale CO2 releases. Nature. Brewer, P.G. 1997. Ocean chemistry of the fossil fuel CO2 signal: the haline signature of “Business as Usual.” Geophysical Research Letters 24: 1367–1369. Brewer, P.G., G. Friederich, E.T. Peltzer, and F.M. Orr, Jr. 1999. Direct experiments on the ocean disposal of fossil fuel CO2. Science 284: 943–945. Brewer, P.G., E.T. Peltzer, G. Friederich, G. Rehder. 2002. Experimental determination of the fate of rising CO2 droplets in sea water. Environmental Science and Technology 36: 5441–5446. Caldeira, K., and G.H. Rau. 2000. Accelerating carbonate dissolution to sequester carbon dioxide in the ocean: geochemical implications. Geophysical Research Letters 27: 225–228. Drange, H., G. Alendal, and O.M. Johannessen. 2001. Ocean release of fossil fuel CO2: a case study. Geophysical Research Letters 28: 2637–2640. Hanisch, C. 1998. The pros and cons of carbon dioxide dumping. Environmental Science and Technology 32: 20A–24A. IPPC (Intergovernmental Panel on Climate Change). 1990. Climate Change: The IPPC Scientific Assessment , edited by J.T. Houghton et al. Cambridge, U.K.: Cambridge University Press. Marchetti, C. 1977. On geoengineering and the CO2 problem. Climatic Change 1: 59–68. PCAST (President’s Council of Advisors on Science and Technology). 1998. Federal Energy Research and Development Agenda for the Challenges of the Twenty-First Century. Washington, D.C.: U.S. Department of Energy. Rehder, G., S. Kirby, W.B. Durham, L. Stern, E.T. Peltzer, J. Pinkston, and P.G. Brewer. In press. Dissolution rates of pure methane hydrate and carbon dioxide hydrate in undersaturated sea water at 1000m depth. Geochimica et Cosmochimica Acta. Wallace, D.W.R. 2001. Storage and transport of excess CO2 in the oceans: the JGOFS/WOCE global CO2 survey. Pp. 489–521 in Ocean Circulation and Climate. Burlington, Mass.: Academic Press.

OCR for page 41
This page in the original is blank.

OCR for page 41
The Effectiveness and Unintended Consequences of Ocean Fertilization1 KEN CALDEIRA Lawrence Livermore National Laboratory Numerical simulation can shed some fresh light on the idea of ocean fertilization. An analysis of the IS92A Intergovernmental Panel on Climate Change (IPCC) scenario shows that to stabilize climate at 2°C of warming, if climate sensitivity is at the low end of the accepted range, approximately 75 percent of all power production would have to come from sources free of carbon emissions by the end of this century. If climate sensitivity is at the high end of the accepted range, nearly all of our energy would have to come from carbon-emission-free sources. We can perform much the same sort of calculation for a range of climate sensitivities and a range of acceptable levels of warming. For stabilization at 2°C with a midrange climate sensitivity, we would have to add approximately one gigawatt of carbon-free primary power per day somewhere in the world. The magnitude of this problem is enormous, and there is no magic bullet to solve it. As other speakers have suggested, we have to work on reducing energy demand, on sequestration, and on developing nonfossil sources of energy. Speakers today have already discussed geologic storage of carbon dioxide (CO2), ocean storage by direct injection, and land biosphere storage, although this is likely to be limited by land availability. Others have proposed geochemical techniques, such as accelerating silicate or carbonate weathering. 1   This work was supported by the Ocean Carbon Sequestration Research Program of the U.S. Department of Energy (DOE) Office of Biological and Environmental Research. It was performed under the auspices of DOE by the Lawrence Livermore National Laboratory under contract No. W-7405-Eng-48.

OCR for page 41
One problem with putting CO2 into the ocean is that nobody thinks this will be good for the ocean. We may decide to do it, however, if it turns out that the adverse environmental consequences of putting it somewhere else are worse. Ocean carbon sequestration would only make sense if it diminishes the overall adverse consequences of releasing CO2 into the environment. We are already putting two gigatons of carbon into the ocean each year. That works out to about five kilograms per day per U.S. citizen. At present, we are also putting carbon into the atmosphere, which may create significant climate change. But eventually, the ocean will absorb about 80 percent of the carbon released to the atmosphere. The idea of ocean sequestration is to put the CO2 into the ocean deliberately, thus avoiding most of the global warming. The argument is that this could have some adverse impacts on the marine environment, but at least we would avoid most of the climate change. Recognizing that we are already sequestering carbon in the ocean unintentionally is very important, and determining the biological effects of CO2 on organisms in the ocean is one of the most important goals of current research. It is essential that we know the effects of increased oceanic concentrations of CO2, even if we decide to put it into the atmosphere. Several ocean fertilization options have been proposed (e.g., adding chemicals, such as nitrate and phosphates to the oceans). My work is focused on simulations of iron-based ocean fertilization. The basic idea of iron-based ocean fertilization (see Figure 1) is to add iron to the upper ocean to stimulate biological activity and increase photosynthetic activity, and thus generate more organic carbon—removing it from the surface. Some of the organic carbon then sinks into the deep ocean. The goal of fertilization is to remove carbon from the surface ocean, fix the CO2 as organic carbon, and then sink it into the deep ocean mostly by gravitational sinking of the total particles. Because CO2 would come from the surface ocean, the pressure of CO2 in the surface ocean box would be decreased, which would lead to a compensating flux of CO2 from the atmosphere into the ocean and draw more CO2 out of the atmosphere. If this were the end of the cycle, we would have permanent sequestration, and everything would be fine. However, when the organic carbon gets into the deep ocean, it is oxidized back to CO2, which can get mixed back up to the surface ocean and then can escape back into the atmosphere. The time scale of the exchange between the upper ocean and the deep ocean is on the order of several centuries. The upper mixed layer equilibrates with the atmosphere roughly on a time scale of a year or so. Thus, ocean fertilization provides only temporary storage. A number of simulations have been done using general circulation models and schematic ocean models. These simulations suggest that, after fertilizing the southern ocean for a century, it would be possible to store carbon in a range of 100 gigatons to 250 gigatons. I worked on a highly idealized simulation of

OCR for page 41
FIGURE 1 Schematic representation of the concept of ocean fertilization. (1) Adding nutrients to the surface ocean can stimulate marine production of organic carbon. (2) Some of the organic carbon sinks to the ocean interior. (3) CO2 enters the surface ocean from the atmosphere to replace some of the carbon removed from the surface ocean. (4) In the ocean interior, the organic carbon is oxidized to CO2. (5) This CO2 is eventually mixed up to the surface ocean. (6) Once in the surface ocean, the CO2 equilibrates with the atmosphere. fertilization (the Los Alamos Parallel Ocean Program [POP] models) that began with the premise that we could add enough micronutrients to the ocean south of 30 degrees to completely deplete surface macronutrients, such as phosphate. One early discovery with this simulation was that, after only three years, CO2 would already begin to leak back into the atmosphere. If we compared three years, 30 years, and 300 years, we found that previously sequestered carbon was leaking back out over much of the rest of the ocean, and by 300 years, there was significant leakage in the tropics. There are two reasons for leakage: (1) carbon placed in the deep ocean eventually mixes back up to the surface; and (2) along with the organic carbon, we sent nutrients down into the deep ocean, thus increasing the deep-ocean nutrient content at the expense of the surface ocean. Biological productivity in other parts of the ocean then began to diminish. In the POP simulation, approximately 375 additional gigatons of carbon are stored in the ocean over a period of 400 years (see Figure 2). On this time scale,

OCR for page 41
FIGURE 2 Amount of additional carbon stored in the ocean and amount of additional storage per year (net flux to the ocean) as computed at Lawrence Livermore National Laboratory for idealized iron fertilization south of 30 degrees using the POP ocean model. the storage is on the order of about one gigaton per year. The net flux starts out close to eight gigatons. At about 100 years, net additional storage (new storage minus leakage) is about one gigaton per year. At 400 years, net additional storage is about half a gigaton per year. My sense is that these are upper bound numbers because in the real world we would probably not fertilize the entire ocean south of 30 degrees, and the areas that were fertilized would probably not perform up to maximum possibilities. It is important to understand that ocean fertilization, insofar as it works and is environmentally and politically acceptable, might become part of a portfolio of responses. In itself, it won’t solve the problem. As we continue fertilizing, we move phosphate and nitrates away from the upper ocean. Thus, the effectiveness of iron fertilization diminishes over time as the surface ocean runs out of macronutrients. In addition, the ratio of added

OCR for page 41
carbon to the added exports from the surface ocean to the deep ocean decreases with time because previously stored carbon leaks back into the atmosphere. What is the residence time of carbon in the ocean? The ocean transports tracers along surfaces of constant density, controlled largely by temperature. Denser surfaces outcrop at the colder poles. Some surfaces in the deep ocean are not well ventilated, so even though more organic carbon is being transported, more carbon is retained in those areas. In other parts of the ocean, CO2 slips back into the atmosphere much more quickly. To determine how ocean carbon sequestration would change allowable emissions, we can calculate the net benefits as functions of a discount rate and assume a price trajectory. With a zero discount rate, there is no time preference, and there is no point in ocean fertilization because future value would not be discounted. We should look at the discount rate minus the emission cost because, if we have a 3-percent discount rate but the cost of carbon emissions rises at the rate of 3 percent, once again we would gain nothing. Taking the range of discount rates that are typically used in business, we would have to sequester initially three gigatons of carbon, say at 0.33, in order to get one gigaton of carbon’s worth of sequestration value. In other words, we use roughly a factor of three to account for the fact that this is not a permanent sequestration. In one simulation, organic carbon that sank into the deep ocean oxidized, thus consuming ambient dissolved oxygen in the water column. After 300 years, regions formed in the model ocean that had severe oxygen depletion, suggesting potential harm to oxygen-breathing organisms. Green Sea Ventures estimates that the cost of iron fertilization would be $7 to $7.50 per ton. But because it would be a temporary sequestration, we must also consider that it might be necessary to multiply the cost by approximately a factor of three to get the net present value. Macronutrient strategies would be considerably more expensive. Some have also suggested that ships could dribble along some iron to compensate for the flux of CO2 admitted by ships. Models are helpful for clarifying conceptual situations, but a model is only as good as the basic knowledge that goes into it, and most models include many unknowns. We don’t know to what extent adding nutrients to the surface ocean would stimulate marine production of organic carbon or how that would vary from environment to environment. Although we’re making progress, we are still not sure what fraction will sink to the deep ocean when organic carbon production is increased. Of the organic carbon that sinks to the deep ocean, some carbon can mix up from below, and some CO2 can come from the top. A deficit in the surface ocean may also remain. It is not clear how much a flux of CO2 from the atmosphere would compensate for this sinking flux or how deep the CO2 would sink in different environments before it is oxidized. Once it is oxidized, we don’t know how long it would stay down before it cycles back up to the surface. There is also some disagreement, although I think I know the answer, for how we should account for the sort of out-gas seen in de-gasing situations.

OCR for page 41
If all of the CO2 we sequester eventually leaks back into the atmosphere, then all we are really doing is time-shifting emissions. We’re putting CO2 in today. and it’s leaking out 100 years or 200 years from now. How can we put a value on the time-shifting of an emission? This is not simply a question of economics. One advantage might be that it would give us time to invent new, carbon-emission-free energy technologies. It might be worth reducing emissions in the short term in anticipation of new energy technologies coming online in the long term.