This report considers climate intervention strategies for deliberately modifying the energy budget of Earth to produce a cooling designed to compensate for some of the effects of warming associated with greenhouse gas increases. The physical principles for modifying the energy budget to cool the planet are discussed more thoroughly below, but they also appear to all of us in our everyday lives. For example, in the temperate and polar regions, winter temperatures are generally colder than summer temperatures, because those regions receive less sunlight in the winter. The energy principles controlling temperature on a hot day or cool night result from and influence weather on a day-to-day local scale and also operate on climate at seasonal through millennial timescales over the globe. For example, in 1784, Benjamin Franklin speculated that “a constant fog over Europe” arising from volcanic eruptions near Iceland diminished the heating effect of the rays of the sun, and that it was responsible for the abnormally cold winter of 1783-1784 in Europe (Franklin, 1789). Since that time, the connection between cooler temperatures and volcanic eruptions (which release particles into the atmosphere that scatter sunlight back to space) has been well established.
These principles operate everywhere in nature; as understanding of Earth’s physical system has increased, some scientists have begun to consider deliberately making use of these physical principles to counter global warming. Budyko (1974) was the first to suggest that global warming might be countered by burning sulfur on airplane flights high in the atmosphere to make small particles (called aerosols) that, like volcanic emissions, would reflect sunlight. Since that time, a variety of suggestions have been made regarding ways to reduce the amount of sunlight absorbed at the planet’s surface.
Climate intervention ideas have been explored in a variety of ways: (1) through basic theoretical considerations, (2) through the study of climate-relevant features that occur today and have occurred in the past that serve as approximate analogues relevant to the methods being suggested for engineering the climate, and (3) through
computer models. Climate models, known to be only an approximation of the real world, suggest that it might be possible to intervene in the climate system to counter some of the effects of global warming, but they also point to negative consequences and new issues of concern from these proposed techniques. Models provide an incomplete and imperfect picture of the world, and one must be cautious in interpreting their results. Nevertheless, these results indicate to some scientists that it would be worthwhile to continue to do research to better evaluate and understand the possibility of deliberately modifying the climate. The need to carefully evaluate and understand these proposals is highlighted by the limited success of previous attempts to deliberately control weather and climate, discussed in Box 2.1.
In the remainder of this chapter we introduce the major themes that are explored at length in subsequent chapters. The principal terminology used throughout this report is summarized in Box 2.2, together with alternate terminology used at places in the existing literature to refer to similar concepts.
BOX 2.1 HISTORICAL CONTEXT FROM PREVIOUS ATTEMPTS TO CONTROL WEATHER
Humans have inadvertently affected regional and global weather in different ways. History has demonstrated the human capability to deploy technologies that affect climate at global scale. As agriculture spread across the continents, land use changes meant that in many areas dark forests were replaced by lighter colored croplands, and in high latitudes this caused a regional cooling (IPCC, 2013a). Sulfate aerosols, largely from coal-fired power plants with inadequate pollution controls, have a global cooling influence, but the effect is most pronounced over large parts of the Northern Hemisphere. Of course, our fossil fuel emissions are affecting climate the world over (IPCC, 2013b). At first, people were not aware that such activities would affect climate and thus unknowingly undertook climate modification. Although humans have never undertaken actions with the express intent of altering regional or global climate on a large scale for a sustained period of time, there have been efforts to affect local weather and proposals to alter regional or global climate (see below).
Visionary proposals for weather and climate control have a long history (see Byers, 1974; Fleming, 2010b, 2012; Huschke, 1963). The National Science Foundation produced a report, Weather and Climate Modification, in 1966 (NSF, 1966) and the National Research Council followed this up with an update in 1973, titled Weather and Climate Modification: Problems and Progress (NRC, 1973).
Many early weather modification proposals did not move beyond the discussion stage, and the ones that did mostly did not produce the desired effects on the physical environment. In many cases, these proposals gave rise to complicated political, social, and economic issues. As we look forward at proposals for intentionally modifying Earth’s climate, society can learn important lessons from previous weather modification proposals.
In 1841, James Espy, the first U.S. national meteorologist, proposed a massive rainmaking scheme based on the convective updrafts theory, the best science of his day. Inspired by volcano dynamics, he proposed burning woodlots each week along the Appalachian Mountains to enhance convection and provide regular rains to the east coast. Espy claimed this would keep the rivers navigable, break up cold snaps and heat waves, and also provide a health benefit by clearing the air of miasmas (Espy, 1841). The immediate result was public criticism, and even ridicule, for Espy (Fleming, 2010a). This is one example of a common theme through history: proposals to modify weather have tended to produce strong public opposition.
A century later, in 1946, Nobel Laureate Irving Langmuir believed he and his team at the General Electric Corporation had discovered a means of controlling the weather with cloud-seeding agents such as dry ice and silver iodide. The following year, in conjunction with the U.S. military, they sought to deflect a hurricane from its path, but planned publicity for the experiment went awry. After seeding, the hurricane veered suddenly, due to what were later determined to be natural steering currents (rather than the seeding), and smashed ashore on Savannah, Georgia (Fleming, 2010a). An important lesson is that those who conduct experiments that substantively alter weather—regardless of whether the interventions had any actual effect—can potentially be held legally liable for damage caused by the altered weather. (See further discussion in Appendix C, including descriptions of cloud-seeding activities that are ongoing today.)
Prospects for larger-scale, even planetary, intervention in the climate system arrived after World War II with the dawn of several transformative technologies. Proposed weather modification projects included ideas such as cloud-seeding techniques, weakening hurricanes with biodegradable oil slicks, and breaking up polar ice with nuclear weapons, often as part of the Cold War quest to militarize the atmosphere (Fleming, 2010b; Hoffman, 2002, 2004). These previous attempts highlight both societal and scientific difficulties in attempting to exert deliberate control over nature, in particular the challenge of demonstrating the efficacy of the modification against a background of natural variability.
A 2003 NRC study, Critical Issues in Weather Modification Research (NRC, 2003), concluded that there was “no convincing proof” that cloud seeding is effective at increasing precipitation. However, peer-reviewed studies have indicated some modest increases in precipitation resulting from cloud seeding in some cases (Breed et al., 2014; California Department of Water Resources, 2005; Morrison et al., 2009).
History teaches us that things change—often in surprising or unanticipated ways—and that a certain amount of clarity can be gained by looking backward as we inevitably rush forward. Although there have been proposals aimed at attempted control of weather and climate that have had some success, there have also been many that have fallen well short of their goals. The potential for public opposition, the potential liability for any negative consequences, and the complex nature of the weather-climate system all point to the need to approach any future proposals for modifying Earth’s climate with caution. A further discussion of previous attempts at planned weather modification is found in Appendix C.
BOX 2.2 SUMMARY OF TERMINOLOGY USED IN THIS REPORT
Albedo modification: Intentional efforts to increase the amount of sunlight that is scattered or reflected back to space, thereby reducing the amount of sunlight absorbed by the Earth, including injecting aerosols into the stratosphere, marine cloud brightening, and other efforts to enhance surface reflectivity. This set of approaches is often referred to by the acronym SRM, standing most often for the term “solar radiation management” but sometimes also “sunlight reflection methods” (Caldeira et al., 2013; The Royal Society, 2009). The committee prefers the term “albedo modification” because it is a more straightforward and neutral description of the physical process involved, and it is free of the connotations of a precise, routine, and orderly process carried by the term “management.”
Stratospheric aerosol injection: A proposed method of albedo modification that involves increasing the amount of small reflecting particles (aerosols) in the stratosphere. The stratosphere is a layer in the upper regions of the atmosphere (starting at approximately 18 km altitude in the tropics) above the more turbulent troposphere layer where rainfall and most conventional “weather” occurs. The aerosol increase is generally not accomplished by injecting aerosols themselves, but by injecting chemical precursors such as sulfur dioxide (SO2), which transform into aerosols via subsequent processes.
Marine cloud brightening: A proposed method of albedo modification that involves injecting substances near the surface of Earth that increase the reflectivity of low cloud layers. The emphasis is generally on clouds over the ocean (which has a low albedo), because these present the best opportunities for increasing reflectivity.
It has been known since the work of Fourier in the early 1800s that the temperature of Earth is determined by the requirement that, in steady state, the rate at which energy is lost to space in the form of outgoing infrared radiation balances the rate at which energy in the form of incoming solar radiation is absorbed by Earth. A mismatch in this balance would cause Earth to warm or cool. The rate at which infrared radiation is emitted increases as the temperature of the surface and atmosphere increases, so the planet can come into equilibrium by warming up or cooling down until balance is achieved. Convection and other vertical mixing processes tightly couple most of the atmosphere to the surface temperature, and for that reason the surface temperature can largely be determined by the top-of-atmosphere energy balance without explicit reference to the details of how energy is transferred between the surface and the atmosphere (Pierrehumbert, 2010).
The climate system can be compared to a heating system with two knobs, either of which can be used to set the global mean temperature. The first knob is the concentration of greenhouse gases such as CO2 in the atmosphere that affects the infrared side of the energy balance; increases in concentration of these gases reduce the rate at which infrared radiation is emitted to space for any given surface temperature (Figure 2.1). As more greenhouse gases are added to the atmosphere, the system (if otherwise undisturbed) will warm up until outgoing infrared radiation increases sufficiently to restore Earth’s energy balance. The other knob is the reflectance of the planet, which controls the amount of sunlight that the Earth absorbs. Sunlight is reflected or scattered by clouds and particles in the atmosphere, and by the surface. One could instead attempt to restore the balance at the original temperature by increasing the proportion of sunlight that Earth’s surface and atmosphere reflect back to space, reducing energy reaching Earth’s surface (Figure 2.1). The technical term for this proportion of reflected incoming sunlight is “albedo,” which comes from the Latin
FIGURE 2.1 Schematic illustration of the energy balance of the preindustrial climate (left panel) and a modified high-CO2 climate following a climate intervention by albedo modification (right panel). In the albedo-modified high-CO2 climate, the infrared cooling to space (red arrow) is reduced relative to the preindustrial climate, but the effect on the energy budget is offset by a corresponding reduction in the amount of solar energy absorbed. The solar absorption is reduced by increasing the albedo so as to reflect more sunlight back to space.
root meaning “whiteness.” For example, adding tiny particles to the upper atmosphere scatters light and brightens the sky, increasing the planet’s albedo. However, these two knobs do more than affect global mean temperature. In differing ways, they also influence regional temperatures, the global hydrological cycle, land plants, and other components of the Earth system. So, turning up one knob and turning down the other might be able to restore Earth’s global mean temperature but could nevertheless produce substantial changes to Earth’s environment (see Chapter 3 for further discussions).
By way of analogy, consider a home heated in winter by passive solar heating, where sunlight entering the windows maintains a comfortable interior temperature. If insulation is added to the roof and walls, the rate at which heat is lost to the outside would decrease, and the temperature inside the house would increase until a balance is restored with the amount of solar energy streaming through the windows. As a result, the house could become uncomfortably hot. One could address this problem by pulling down the window shades a bit, reducing the amount of sunlight entering the house.
There are a number of means by which the amount of sunlight absorbed by Earth could be altered. Objects such as mirrors, lens arrays, or orbiting clouds of reflecting particles could be placed in outer space, diverting some sunlight before it can encounter Earth (Early, 1989). Small particles (aerosols) or substances that lead to their formation could be injected into the stratosphere and renewed as needed (Budyko, 1974). Substances can be injected near the surface of Earth that either directly reflect sunlight or cause low-level clouds to become more reflective (Latham, 1990). See Figure 2.2 for an illustration. Finally, the land surface reflectivity can be directly modified, for example, by adding white roofs and parking lots or by planting light-colored vegetation to cover or replace darker surfaces (Irvine et al., 2011). All of these ideas have been proposed as possible mechanisms to modify Earth’s albedo on a large scale, and some of these proposed strategies are discussed in more detail in Chapter 3.
Climate change is driven by an imbalance in Earth’s energy budget. The magnitude of this imbalance (after accounting for some adjustment processes) is radiative forcing (see Box 2.3), typically quoted in units of watts per square meter (W/m2) of Earth’s surface. The radiative forcing caused by doubling of the preindustrial CO2 concentration is approximately 4 W/m2. Sunlight is absorbed by Earth at a rate of about 240 W/m2, so reflecting back to space approximately 2 percent of the currently absorbed sunlight would offset the top-of-atmosphere radiative imbalance caused by a doubling of atmospheric CO2 content (Govindasamy and Caldeira, 2000; Kravitz et al., 2013a). Because aerosols are very effective reflectors of sunlight (see Chapter 3), the required
FIGURE 2.2 Illustration of the two proposed approaches for increasing albedo that are discussed in this report: increasing the concentration of reflecting particles in the upper atmosphere (specifically, the stratosphere) and increasing the reflectivity of low clouds.
change in albedo can in theory be accomplished by maintaining a small mass of aerosols in the atmosphere; this is the chief appeal of climate intervention by aerosol injections.
To put the required increase in albedo into perspective, the 1991 Pinatubo eruption, which is estimated to have been the largest eruption since Krakatau in 1883, led to a radiative forcing of approximately −3 W/m2 within a month following the eruption, decreasing to nearly zero over the subsequent 2 years (IPCC, 2007b, Fig. 2.18) and causing the average surface air temperature to cool an estimated 0.3°C over a period of 3 years.
BOX 2.3 RADIATIVE FORCING AND ALBEDO
Radiative forcing provides a measure of the amount by which a change in some given characteristic of the Earth system (e.g., atmospheric CO2 concentration) alters Earth’s energy budget, all other things being held constant. The larger the radiative forcing, the more the surface and atmospheric temperature must change in order to restore balance. The forcing is referred to as “radiative” because essentially all energy enters or leaves the Earth system in the form of electromagnetic radiation—largely infrared or visible light. Radiative forcing is measured in units of watts per square meter (W/m2), corresponding to the change in amount of energy per unit time per unit of Earth’s surface area entering or leaving the top of the atmosphere. The change in energy is referenced to a baseline period, typically in recent preindustrial times.
Radiative forcing can be divided into long- and short-wavelength components. The long-wavelength component refers to changes in the amount of infrared radiation emitted by Earth to space and is controlled primarily by changes in the greenhouse gas content of the atmosphere. The short-wavelength component refers to changes in the amount of solar energy absorbed by Earth and is controlled primarily by the proportion of sunlight reflected back to space by the atmosphere and the surface. This proportion is known as the albedo. Albedo is commonly quoted as a percentage; an albedo of 100 percent would mean that all of the incident sunlight is reflected back to space and none is absorbed, whereas an albedo of 0 percent would mean that none of the incident sunlight is reflected and all of it is absorbed. The best current estimates of Earth’s albedo put the value between 29 percent and 30 percent for the past decade (Stephens et al., 2012).
For a more precise technical definition of radiative forcing, see IPCC (2013a, Box 8.1).
Reviews by the NRC (2011b) and the Intergovernmental Panel on Climate Change (IPCC, 1991, 1997, 2003, 2007a, 2013b) have concluded that the anthropogenic climate change has the potential to cause substantial harm. IPCC Working Group I projects that the RCP8.5 “business-as-usual” scenario will result in 4°C (7.2°F) warming by year 2100 (IPCC, 2013b). At this level of warming, IPCC Working Group II projects “severe and widespread impacts on unique and threatened systems, substantial species extinction, large risks to global and regional food security, and the combination of high temperature and humidity compromising normal human activities, including growing food or working outdoors in some areas for parts of the year.” For example, under a business-as-usual scenario, climate models project that by the end of the twenty-first century most summers in the tropics will be hotter than the hottest summer experienced in the twentieth century, which could potentially threaten tropical crop productivity (Battisti and Naylor, 2009).
The IPCC (2013b) estimates anthropogenic releases of aerosols to the atmosphere are currently offsetting about 30 percent of the radiative forcing from anthropogenic greenhouse gases, primarily by affecting planetary albedo. The IPCC (2013b) further estimates that albedo change due to land use change offsets about 5 percent of the radiative forcing from anthropogenic greenhouse gases. Crutzen (2006) raised the question of whether humanity might want to develop the capability to intentionally modify Earth’s albedo to a greater degree and offset a larger amount of forcing. Unfortunately, today’s aerosols emissions create large health and environmental problems. Thus, it is important for society to know whether it is possible to alter Earth’s albedo by much greater amounts while being sure that the effort will do a large amount of good and only a small amount of harm.
Should it ever become important for society to cool Earth rapidly, albedo modification approaches (in particular stratospheric aerosol injection and possibly marine cloud brightening) are the only ways that have been suggested by which humans could potentially cool Earth within years after deployment. Over the past 15 years, stratospheric aerosol injection and marine cloud-brightening ideas were tested in modern climate models, and results for an idealized set of scenarios across a broad spectrum of models (Kravitz et al., 2013a) yielded consistent results on the direct cooling effects of such approaches and some indirect processes. These models indicate that decreasing the amount of sunlight absorbed by Earth can offset most of the global mean warming caused by elevated greenhouse gas levels (Kravitz et al., 2013a). Changes in the hydrological cycle are more complex and harder to summarize; these are discussed in Chapter 3. Although these model results are consistent with one another, the remaining unknowns with respect to the overall effects of increasing Earth’s albedo raise the risks if they are not well understood before embarking on any deployment.
Nonetheless, climate models, observations of volcanic effects, and basic physical theory indicate that it would be possible for humans to cool Earth within a few years after deployment by reflecting more sunlight to space. Some assessments have been made on the feasibility of deploying albedo modification methods (see Chapter 3 below). Engineering analysis suggests that at least some of the proposed methods to achieve substantial cooling may be within the realm of technological feasibility and would have relatively modest direct costs, not including, however, the costs of the necessary control and monitoring infrastructure. The accuracy with which a targeted degree of cooling can be achieved is unclear, and indirect costs of potential damages have not yet been quantified and could be substantial. For these reasons, there has been interest in learning more about albedo modification proposals.
There are a number of hypothetical but plausible scenarios in which deployment of albedo modification might be considered. One scenario is a response to sudden and severe climate change, which is sometimes referred to as a “climate emergency.” If, for example, global warming resulted in massive crop failures throughout the tropics (e.g., Battisti and Naylor, 2009), there could be intense pressure to temporarily reduce temperatures to provide additional time for adaptation.1 In such circumstances, there could be demands for immediate deployment of albedo modification, even in the absence of a rigorous assessment of the implications or an adequate monitoring system.
It has also been suggested that albedo modification with strictly limited magnitude might be initiated without waiting for a climate emergency to occur (Burns, 2011; Keith, 2013; Wigley, 2006). For example, the international community might agree to a gradual phase-in of albedo modification to a level that is expected to create a verifiable modification of Earth’s climate (e.g., 1 W/m2) as a large-scale field trial aimed at gaining experience with albedo modification in case it needs to be scaled up in response to a later climate emergency. A limited deployment of albedo modification might also be considered as part of a portfolio of actions to reduce the risks of climate change.
Finally, as a matter of physical and economic capability, a single nation, a large corporation, or a group of individuals with sufficient means could potentially deploy albedo modification in the absence of an international consensus or coordination (Bodansky, 2011; Victor et al., 2009). Such attempts might begin at small scales (e.g., a few small ships for modification of low clouds) or as an attempt to modify regional climate (e.g., an attempt to restore a failed Indian monsoon or to ameliorate a severe European heat wave). However, in practice, unilateral capability is likely to be limited to those states with significant political and economic power and world stature, such that it would be difficult or costly for others to make them stop an unsanctioned albedo modification program through the threat or act of military attacks against deployment devices and associated infrastructure (Parson and Ernst, 2013). There is also the possibility, however, that similar countermeasures could be used by a sufficiently powerful dissenter against a sanctioned deployment by other nations.
As described in the next section, such scenarios bring with them a wide range of concerns and a likelihood of unintended consequences (also see Robock, 2014). It is these risks and concerns that form the chasm between what may be technically feasible and what might constitute wise and prudent action. Substantial research would be required and understanding developed before this gap could be bridged, and such
1 Albedo modification would not be an effective response to some types of climate emergencies, such as a rapid collapse of the West Antarctic ice sheet, which are not driven by surface air temperatures (Barrett et al., 2014).
research should be done before albedo modification is seriously considered. The unilateral and uncoordinated actor scenario raises questions of how we could detect albedo modification activities and attribute changes in climate to such activities. Arguments to oppose such unilateral action would be bolstered by better understanding of the underlying science of the albedo modification, its detection, and its unintended consequences. The state of knowledge on these techniques and future research directions are discussed in Chapter 3.
The increase in greenhouse gas concentrations from anthropogenic emissions introduces many risks to the planet. Deploying albedo modification could produce a generally cooler climate, but it would introduce risks of a different type. Compensation by albedo modification is only approximate, and some manifestations of high CO2 concentrations are not addressed at all. This imprecise compensation implies that there could be regional disparities in the distribution of benefits and risks (Kravitz et al., 2014; Moreno-Cruz et al., 2012), and a means would need to be found to agree on the right mix of albedo modification in the portfolio of responses, if it were ever to be deployed (Ricke et al., 2013). Any of these decisions, however they are made, would benefit from a more informed understanding of the nature of the climate response. The bulk of this report is devoted to reviewing the extent to which the response is understood currently and the research agenda needed to address questions that remain open.
Earth’s albedo is governed by cloud, water vapor, aerosols, land surface, and sea ice processes that link dynamically to all other aspects of the climate system, all of which are affected by both addition of anthropogenic greenhouse gases to the atmosphere and actions aimed at increasing the albedo. The uncertainties in modeling of both climate change and the consequences of albedo modification make it impossible today to provide reliable, quantitative statements about relative risks, consequences, and benefits of albedo modification to the Earth system as a whole, let alone benefits and risks to specific regions of the planet. To provide such statements, scientists would need to understand the influence of various possible activities on both clouds and aerosols, which are among the most difficult components of the climate system to model and monitor.
Albedo modification can in principle reduce the annually averaged global mean temperature to a given target level, but the resulting climate will be different in a number of important ways from the low-CO2 climate with natural albedo. There is potential for substantial consequences to other aspects of the climate system, including precipitation; regional temperature; atmospheric and oceanic circulation patterns; stratospheric temperature, chemistry, and dynamics; and the amount and characteristics of sunlight reaching the surface (see sections in Chapter 3 on modeling and environmental consequences).
The geographical and seasonal distribution of radiative forcing due to albedo modification is substantially different from that arising from a decrease of CO2. The atmosphere and ocean respond to radiative forcing by redistributing the heat in a way that alleviates the mismatch, but this requires changes in circulation patterns and also can leave regional climate anomalies uncompensated to one extent or another. Additionally, increasing albedo alters the surface energy budget by reflecting sunlight that would otherwise sustain evaporation (and hence precipitation); this can have effects on precipitation patterns. The ratio of change in precipitation to change in temperature is greater for a change in albedo than it is for a change in carbon dioxide content. Furthermore, albedo modification does not address the ocean acidification problem (Matthews et al., 2009), which, in the absence of ocean alkalinization (see Box 2.4), is an
BOX 2.4 OCEAN ACIDIFICATION
Albedo modification techniques could address some, but not all, of the consequences of rising atmospheric carbon dioxide that extend well beyond alterations in the radiative balance of the planet and climate change. Of particular importance, the ocean uptake of excess atmospheric carbon dioxide—the excess above preindustrial levels driven by human emissions—causes well-understood and substantial changes in seawater chemistry that can negatively affect many marine organisms and ecosystems (Doney et al., 2009; Gattuso and Hansson, 2011).
The additional carbon dioxide causes direct changes in seawater acid-base and inorganic carbon chemistry in a process often termed ocean acidification. Long-term ocean acidification trends are clearly evident over the past several decades in open-ocean time-series and hydro-graphic survey data, and the trends are consistent with the growth rate of atmospheric carbon dioxide (see Figure) (Doney, 2013; Doney et al., 2014; Dore et al., 2009).
The biological impacts of ocean acidification arise both directly—via effects of elevated carbon dioxide, lower pH, and lower carbonate ion concentrations on individual organisms—and indirectly—via changes to the ecosystems on which they depend for food and habitat (Doney et al., 2009, 2012). Ocean acidification leads to a decrease in the saturation levels of calcium carbonate (CaCO3), a hard mineral used by many marine microbes, plants, and animals to form shells and skeletons. The potential biological consequences due to acidification are slowly becoming clearer at the level of individual species, but substantial uncertainties remain, particularly at the
ecosystem level (Doney, 2013; Gattuso and Hansson, 2011). Ocean acidification acts as a stress on marine ecosystems and will likely also exacerbate other human perturbations such as climate change, overfishing, habitat destruction, pollution, excess nutrients, and invasive species.
The magnitude of ocean acidification and biological impacts is related to the concentration and growth rate of excess atmospheric carbon dioxide. Thus, approaches for mitigating future ocean acidification impacts require curbing human carbon dioxide emissions to the atmosphere and/or developing atmospheric carbon dioxide removal and sequestration methods. Proposed strategies for limiting the potential negative impacts of ocean acidification also include a combination of targeted adaptation strategies and evolving coastal management practices (Washington State Blue Ribbon Panel on Ocean Acidification, 2012).
FIGURE Top: Time series showing the increase in dissolved CO2 in the ocean and increasing ocean acidity (decreasing pH) over the past several decades. Partial pressure of CO2 in seawater calculated from dissolved inorganic carbon (DIC) and total alkalinity (TA) (blue symbols) and in water-saturated air at in situ seawater temperature (red symbols). Bottom: Time series of mean carbonic acid system measurements within selected depth layers at Station ALOHA, 1988-2007. In situ pH, based on direct measurements (green symbols) or as calculated from DIC and TA (orange symbols), in the surface layer and within layers centered at 250 and 1000 m. SOURCE: Dore et al., 2009.
inevitable consequence of the uptake of CO2 emissions by the oceans. (For the same reason, albedo modification does retain the benefits of CO2 fertilization of land plants [Govindasamy et al., 2002].) These considerations apply to all albedo modification schemes and are discussed in detail in Chapter 3.
Additional considerations apply specifically to albedo modification techniques that involve stratospheric aerosols. Stratospheric aerosols heat the stratosphere at the same time they cool the surface, which can have important implications for the climate of both the stratosphere and the surface, as well as for stratospheric chemistry (see further discussion in Box 3.2 and stratospheric aerosol modeling sections in Chapter 3).
Intervening in the climate system through albedo modification therefore does not constitute an “undoing” of the effects of increased CO2 but rather a potential means of damage reduction that entails novel and partly unknown risks and outcomes. Approaches that limit or reduce levels of CO2 in the atmosphere address the major cause of human-induced climate change, whereas albedo modification attempts to counter some effects of high greenhouse gas concentrations without addressing the causes. This nonequivalence of climate states has a bearing on any decision that will ultimately be made regarding the proper place of albedo modification in the portfolio of responses to the problems caused by greenhouse gas emissions arising from human activities. Along the continuum of hypothetical climate futures—ranging from those with comparatively low CO2 and little or no albedo modification (because greater reliance has been placed on mitigation and carbon dioxide removal [CDR]), extending to scenarios with unrestrained emissions and very high CO2 and a correspondingly high degree of albedo modification—the risk increases as one moves toward higher CO2 because the climate system is forced further outside the range in which it has known, historically established behavior. As one example of such a consequence, consider that if CDR were ramped up to very high levels to compensate very high levels of CO2, one would expect the diurnal cycle of temperature to be reduced significantly with the potential for significant impacts on ecosystems.
The less CO2 that humans release to the atmosphere, the lower the environmental risk from the associated climate change and the lower the risk from any albedo modification that might be deployed as part of the strategy for addressing climate change. It is widely recognized that the possibility of intervening in climate by albedo modification does not reduce the importance of efforts to reduce CO2 emissions. Notably, an assessment by The Royal Society (Shepherd et al., 2009) concluded that “[g]eoengineering methods are not a substitute for climate change mitigation, and should only be considered as part of a wider package of options for addressing climate change.” The findings of this committee, summarized in Chapter 5, support this conclusion.
Another important difference between an albedo-modified high-CO2 state and the preindustrial state arises from the mismatch in timescales between the high rate of dissipation of substances introduced into the atmosphere, for the purposes of modifying albedo, and the very low rate of removal of CO2 from the atmosphere by natural processes. Marine cloud brightening dissipates in a matter of days to weeks after the cessation of active climate intervention, and stratospheric aerosols dissipate within 1 to 2 years (as evidenced by the lifetime of volcanic forcing). In contrast, the climate forcing due to CO2 persists for millennia even if emissions cease (Archer et al., 2009; NRC, 2011a; Solomon et al., 2009).
If CO2 emissions into the atmosphere were not reduced and instead albedo modification was relied on as the primary means to avoid CO2-induced warming, the amount of albedo modification required would continue to escalate as atmospheric CO2 concentrations increased. This scenario of increasing reliance on albedo modification coupled with increasing atmospheric CO2 concentrations is a scenario of profoundly increasing risk. As the albedo modification system was ramped up, negative consequences would likely amplify because at higher CO2 levels imperfections and nonlinearities in the attempted climate change cancellation would become more pronounced (Bala et al., 2003). Furthermore, as the amount of CO2 in the atmosphere and the scale of offsetting albedo modification effort increases, termination, whether it be gradual or sudden, becomes more problematic and risky. If albedo modification activities are ceased abruptly, rapid warming of potentially large magnitude will ensue (the magnitude rising with the level of CO2 being dealt with).
The committee refers to the set of potential challenges that may confront such long-term maintenance of albedo modification in this class of deployments as the problem of millennial dependence risk. These issues are discussed at length in Chapter 3.
Rather than discuss every potential means of modifying Earth’s albedo that has been proposed, this report focuses on the two strategies that have received the most attention and which may most feasibly have a substantial climate impact: stratospheric aerosol injection and marine cloud brightening. The stratospheric aerosol and marine boundary layer cloud schemes are the ones that have been most extensively studied so far, and they are also the ones that are the closest to being deployable in the lim-
ited sense of technical ability to inject sufficient material into the atmosphere to cause a significant (if not necessarily well-controlled) modification to Earth’s albedo. The physical basis of these techniques, their technical feasibility, the nature of the climates produced when they are used to partly offset the effects of high CO2, and the physical risks involved are discussed in detail in Chapter 3.
Other proposed albedo modification techniques include placing large arrays of reflecting satellites in space or altering the reflectivity of the land or ocean surface. As described in the Chapter 3 section “Other Methods,” these other proposed techniques are generally either prohibitively expensive or difficult to scale to the point where they could offset a substantial amount of CO2 radiative forcing. Proposals to modify cirrus clouds, which are not formally an albedo modification method but use another means to modify the planet’s energy balance, have received less attention thus far and are also discussed briefly in this section.
One of the charges of this committee is to assess the technical feasibility of albedo modification techniques. Although it might be possible to deploy albedo modification procedures rapidly and at modest expense (in comparison with the cost of rapidly decarbonizing the world economy), doing so would entail substantial risk and uncertainty. The risk of inadvertent and possibly harmful side effects is increased in the absence of adequate monitoring needed to determine what climate forcing was actually achieved by a given intervention. Some preliminary work based on control theory analysis (MacMartin et al., 2014) suggests that it may be possible to design intervention strategies that rely on temperature measurements alone, but it is unclear at present whether such strategies can actually be implemented by known ways of affecting albedo. The infrastructure needed to accurately monitor albedo and aerosols involves developing capabilities to model the albedo modification caused by a particular injection protocol, to observe the resulting change in aerosol content and albedo of the atmosphere to determine what modification was actually achieved, and to detect the response of climate to the modification. There is considerable uncertainty about whether it would be possible to create an observational infrastructure that would greatly reduce unnecessary risk. If it were possible, the amount of time and resources it would take to develop such an infrastructure is also at present unsettled. This is a crosscutting issue that applies to all albedo modification techniques, and therefore it forms a key part of our feasibility assessment in Chapter 3.
Sociopolitical issues raised by the prospect of climate intervention by albedo modification are taken up in Chapter 4, including a discussion of governance that might be required in order to regulate experiments on albedo modification that involve controlled emissions. Many of the risks associated with albedo modification are socio-
political in nature. These are among the hardest risks to assess, and the expertise to perform such an assessment is for the most part beyond the capabilities of the committee. Though the chief recommended actions in this report are to move forward with research but not with deployment, expansion of research in albedo modification is not without risk, and most of the risks are sociopolitical in nature; on the other hand, ignorance (through failure to carry out research) of consequences of albedo modification deployment also entails considerable risk. In Chapter 5, the committee suggests a way forward toward appropriate research on albedo modification, synthesizing findings from the present report with insights derived from the committee’s report on CDR technologies.