The Physics of Energy, page 155
At mid-latitudes the increased availability of CO2 may lead to increased plant growth and food production. As mentioned above, as of 2015 some 30% of human carbon emissions are absorbed in increasing terrestrial biomass. It is not well understood how far this trend will continue or whether it may reverse beyond a certain point.
In some dry and tropical regions increased desertification from increased evaporation will place additional pressure on water supplies, with impact both on human and natural ecosystems.
The Arctic ecosystem will likely experience the greatest impact; substantial habitat loss due to changing conditions may destroy parts of this ecosystem completely.
Because the pulse of warming from anthropogenic carbon emissions is projected to persist over a period of at most several thousand years, species and ecosystems will not have sufficient time to adapt substantially to changes. Many species whose range and habitat is already limited may be pushed to extinction by the short-term change in their environment. According to IPCC projections [246], if global temperature rise is between 1.5°C and 2.5°C, 20% to 30% of species worldwide will be at an increased risk of extinction. With temperature rise of 3.5°C or greater, over 40% of species may go extinct. Note that projections of this kind are highly uncertain, and this risk should be interpreted in a context in which many species are already threatened with extinction by habitat loss and other human impact. The populations of a large fraction of the non-domestic animals weighing more than a few kilograms are already severely impacted by human activities such as land use and fishing. Even in the absence of global warming, it is likely that many large land and marine animal species will be driven to extinction by human activities in the twenty-first century. Global warming will hasten the extinction of those species already living at the limit of their resiliency.
35.3.6 Ocean Acidification
As described above, some of the excess CO2 produced by human activity is absorbed by the world’s ocean waters, at a level corresponding to roughly 2.3 Gt of carbon per year. As the CO2 comes into equilibrium in the ocean, reactions with water produce carbonic acid (H2CO3), which in turn can release hydrogen ions (H+) and bicarbonate ions (HCO3–). The acidity of a solution is measured by pH, which is roughly proportional (precisely in the original definition) to the (negative of the base-10) logarithm of the density of H+ ions. Over the last several hundred years, the pH of ocean surface waters has decreased from roughly 8.25 to 8.14, corresponding to an increase by roughly 30% in hydrogen ions. As the ocean becomes more acidic, carbonate ions (CO3–) combine with free hydrogen ions to form bicarbonate ions. As discussed in §35.1.4, carbonate ions are used in production of shells and skeletal materials by many marine organisms, including in particular corals as well as mussels, clams, oysters, snails, sea urchins, and calcereous phytoplankton. Most of these organisms live at levels in the ocean that are oversaturated with carbonate ions. As the ocean pH drops, these areas become undersaturated and it is more difficult for the organisms to build shells. In recent years, many coral and shellfish populations have suffered extreme degradation from the decrease in pH as well as other human impacts. With CO2 levels approaching or exceeding 560 ppm, an even more dramatic decrease in pH, by 0.2 or 0.3, may be expected over the next century. While, as discussed previously, atmospheric CO2 levels were higher many millions of years ago, the rate of change of ocean pH from the current anthropogenic CO2 increase is several orders of magnitude faster. Although the populations of marine organisms affected could presumably migrate over periods of many thousands of years or more to depths with an appropriate carbonate ion level, such migration is much more difficult over a period of a few decades. Thus, many marine scientists have great concerns regarding the impact of rising atmospheric CO2 levels on marine ecosystems.
35.3.7 Nonlinear Climate Effects
Most modeling and climate analysis is based on the assumption that for given external conditions there is a unique equilibrium climate configuration, and that small fluctuations in the external conditions will lead to a linear climate response. The analysis of climate feedbacks in §34.5, for example, is based on this assumption of linearity. The complete set of equations governing atmosphere and ocean circulation, ice melt, deep water formation, interaction with biological systems, and all the other complexities of climate systems, however, are highly nonlinear. In general, nonlinear dynamical systems like this can have many different local equilibria, even for fixed external conditions. As external parameters shift, such a system can undergo a rapid shift between regimes where solutions have very different qualitative features. Over the last several million years there have been periods of rapid warming at the beginning of interglacials following long glaciated periods. Some climatologists believe that this rapid shift is evidence for a bistable system in which small perturbations can cause the climate system to rapidly move between one locally stable equilibrium and another. It is possible that anthropogenic CO2 emissions and the associated warming may move the climate system far enough from its current equilibrium for some nonlinear effects to kick in that could take the system into a new equilibrium. Such events are difficult to describe accurately in computer models due to a limited understanding of the precise mechanisms for such processes. There is currently no consensus among climate scientists regarding the likelihood of events of this type.
One scenario of this type that has been discussed extensively is the possibility that fresh water melting into the North Atlantic from Greenland and other sources may reduce the salinity of the water sufficiently to reduce the rate of deep water formation and slow the rate of northward heat transport in the Atlantic Ocean, affecting global circulation patterns. Such a slowdown of the meridional overturning current may have been responsible for North Atlantic cooling during the Younger Dryas 12 000 years ago, and oscillations in temperature during earlier glacial periods measured from Greenland and Antarctic ice cores may also have been caused by this mechanism. While recent analysis suggests that this possibility is not likely to play a large role in climate changes over the next few hundred years, research in this area is ongoing.
Another nonlinear effect that could lead to a rapid climate shift is the sudden collapse of the West Antarctic ice sheet, as discussed above, which would lead to a significant contribution to sea level rise and increased albedo feedback in the southern polar regions.
One other contribution to climate change with some nonlinear features may come from methane clathrates (§33.3) under frozen permafrost and in ocean sediments that could be released extensively after a certain temperature threshold is passed. It is believed that this mechanism may have contributed to the increase in methane levels following warming at the end of glacial periods identified in the geological record. Because methane is a highly potent greenhouse gas (see §34.4), the release of significant amounts of methane could provide a substantial positive feedback for warming. On the other hand, because of the long time scale of warming surface ice and ocean sediments, this process may not occur particularly suddenly or quickly; furthermore, methane’s short lifetime in the atmosphere would mitigate this effect.
While the particular effects just mentioned are not believed likely to lead to dramatic climatic shifts, these and many aspects of climate systems are not well understood. From a statistical point of view this indicates that the distribution of possible outcomes of the climate “experiment” may be broader than a typical normal distribution. Practically speaking, this means that significant deviations in either direction outside the estimated error bars around the average predicted values cannot be ruled out with high certainty. From a policy point of view, it is important to incorporate the uncertainty in climate predictions as well as the mean of the distribution. In particular, it may be prudent for society to take significant precautions to preclude the occurrence of events at the tail of the probability distribution if those events would have catastrophic consequences, even if the estimated probability of such events is relatively small.
35.4Mitigation and Adaptation
Given the fact of ever-increasing anthropogenic CO emissions and our scientific understanding of the role of CO in increasing radiative forcing, and notwithstanding our less-than-complete understanding of how the carbon cycle and global climate will react in the coming decades, it seems inevitable that Earth will experience a warming climate over the next century, though the precise extent of this warming is still unclear. In this section we discuss some of the scientific aspects of options that humans may adopt in responding to these circumstances.
35.4.1 Options for the Future
Faced with rising atmospheric CO levels from fossil fuel usage, there are four logical possibilities for how humans may react:
1. Move to non-carbon based power sources and/or reduce energy use substantially.
2. Capture carbon, from the atmosphere and/or at the point of release, and sequester it away somewhere.
3. Reduce radiative forcing directly by geoengineering.
4. Do nothing and deal with whatever consequences arise.
Of course, these options are not mutually exclusive; more than one will almost certainly come into play in the next century. Much of this book has been dedicated to describing the physical principles of non-carbon based power sources. In this section we briefly discuss the other three options.
35.4.2 Carbon Capture and Sequestration
One way to mitigate the excess carbon dioxide emissions caused by fossil fuel usage is to capture the CO and remove it from the global carbon cycle by placing it in some kind of long-term storage, a process known as carbon capture and sequestration (CCS). The easiest way to capture the CO is at the point of emission; large point sources such as coal plants currently account for roughly 40% of human CO emissions. It has also been proposed that carbon can be removed directly from the atmosphere, a process known as direct air capture (DAC).
Figure 35.24 The first large-scale CO sequestration facility in operation: Sleipner Vest natural gas field, Norway. Beginning in 1996, this Statoil facility sequestered roughly 1 Mt of CO per year. (Credit: Bloomberg/Bloomberg/Getty Images)
Several options have been suggested for sequestration. One possibility is to compress the CO and pump it into geological repositories such as oil or gas fields from which original fossil fuels have already been extracted. It has also been suggested that CO can be stored in the deep ocean, either by dissolving it into the ocean at a depth of several thousand meters, or injecting it below 3000 m depth, where carbon dioxide becomes more dense than water and will pool at the bottom for a long time before re-entering the carbon cycle. Methods have also been suggested for fixing the carbon into carbonates that can be stored in compact solid form.
There are many technical challenges to the capture and sequestration of carbon that go beyond the purview of this book. Here we consider some aspects of CCS from point sources where the CO is relatively concentrated. DAC is explored in Problems 35.11 and 35.12. A useful lower bound on the effort entailed in CCS can be obtained by considering the entropy change that must occur when CO is separated from other gases with which it had been mixed. If molecules of CO are originally mixed with other gases at a fractional concentration of c, then when the CO is separated out, the entropy of the system is reduced by the entropy of mixing as described in Example 8.1,
(35.5)
But the second law of thermodynamics requires that the entropy of the universe must always increase; thus at least this entropy must be released to the environment. If the ambient temperature is T then the separation of the gases requires that energy
(35.6)
be expelled into the environment. An equivalent way to think about this calculation is through Gibbs free energy (8.68); eq. (35.6) is just the statement that the Gibbs free energy increases when the separation is performed.
One currently favored approach to long-term sequestration of CO is storage in compressed form in geological repositories. If captured CO is stored at high pressure, additional work must be done to compress the gas. A small calculation (Example 35.2) shows that capturing the CO from the burning of coal in air and compressing it to 100 atm reduces the power available from the coal-fired power plant by at least 17%. This calculation only gives the theoretical minimum energy needed; actual energy costs would be significantly greater.
Example 35.2 Energy Cost of CO Capture and Sequestration
A coal power plant with carbon capture and sequestration captures CO from coal that is burned in air, compresses it to100 atm, and stores it in an underground repository. We estimate a lower limit percentage of the power plant’s output required to separate the CO from flue gases and to compress it (isothermally) to 100 atm at an ambient temperature of 300 K. We assume that the coal is burned with the minimal amount of air required for complete combustion and that the efficiency of the power plant (without considering the energy required for CCS) is 30%.
First consider the energy required to satisfy the second law of thermodynamics. Every molecule of O in the combustion air is converted to a molecule of CO. Thus the concentration of CO in the flue gases is , the same as the concentration of O in air. Equation 35.5 enables us to compute the entropy reduction by separation (per mole of flue gas), and eq. (35.6) relates this to the minimum energy that must be expelled to the environment by the separation process. Burning 1 kg(C) produces 1/0.012 = 83.3 mol(CO), which in turn, generates 83.3/0.21 = 397 moles of flue gas. The minimum separation energy per mole of flue gas at 300 K is
or per kilogram of carbon, kJ/kg(C). Compressing the resulting 83.3 moles of CO to 100 atm isothermally requires 958 kJ/kg(C). Thus the total energy required for CCS is kJ/kg(C). Burning a kilogram of coal produces MJ of thermal energy, of which 30% or 8.8 MJ is available for useful work. Thus CCS requires at least of the power plant’s useful output.
Since the concentration of CO in the effluent is the principal variable in CCS, we show the minimum percentage of power plant output required for sequestration as a function of CO concentration in the figure above. The case considered above and the case of direct air capture are marked on the figure. Note that for direct air capture of a comparable quantity of CO after emission from the plant, over 30% of the plant’s output would be used at least. Other difficulties with DAC are explored in Problem 35.12.
Some pilot plants have been developed with post-combustion CO capture. Estimates of actual energy cost for post-combustion CO capture are in the range of 25%–40% of plant output. The corresponding increase in fuel required for a given usable energy output in turn increases the energy used for mining and transporting the coal fuel to the plant, which in turn produces additional CO, reducing the effective offset from the carbon capture in the plant. By incorporating pre-combustion carbon capture into an IGCC plant (see §33), carbon emissions can in principle be reduced at lower energy cost, because the CO is present at higher concentrations and higher pressure in pre-combustion syngas than in post-combustion emissions. While the chemical engineering needed to capture carbon at high pressure and temperature is challenging, it has been estimated that pre-combustion carbon capture may be possible with energy requirements in the range 11–22% of plant output.
In addition to capture from point sources and direct air capture by conventional chemical processes (see Problems 35.11 and 35.12), a number of alternative approaches to direct air capture of CO have been considered. These include using biological systems – terrestrial biomass, for example – to concentrate carbon, and sequestering the mass thus produced. It has been suggested that seeding the ocean with iron or other nutrients could enhance the rate of growth of phytoplankton over large areas, increasing the rate at which the biological pump transfers the carbon from atmospheric CO to the deep ocean; while experiments have confirmed that added iron can stimulate plankton blooms, the efficacy of this approach for carbon sequestration is still under debate. Concerns about the widespread disruption of marine ecosystems, among other things, have diminished interest in this approach [274]. More speculative approaches include ideas like bioengineering organisms to optimize conversion of atmospheric CO to carbonates that can be compactly stored over long periods with small degradation rate.
35.4.3 Counteracting Radiative Forcing through Geoengineering
If fossil fuels continue to be used in large quantities, and rising atmospheric CO levels lead to warming with undesirable consequences, and if removal of CO directly from the atmosphere is too difficult, another option that has been raised is the direct manipulation of radiative forcing using large-scale geoengineering – option 3 on the list above. This approach may be less expensive in the short term than large-scale carbon capture or investment in low-carbon energy sources. It has several disadvantages, however. Any geoengineering effort that only affects the radiative forcing will not change the forcing due to CO directly, only offset it using another mechanism. Thus, the geoengineering technique must be maintained as long as atmospheric CO levels remain elevated. If CO levels continue to rise, then the geoengineering project must be increased commensurate to the increase in radiative forcing (which is, however, logarithmic in the CO level, as discussed previously). Finally, geoengineering to offset radiative forcing would not address other aspects of elevated CO levels, including in particular ocean acidification.
We briefly summarize a few of the methods that have been suggested for geoengineering to modify radiative forcing.
Reflection of incoming radiation One possibility that has been suggested is the emplacement of large systems of mirrors or reflecting material to effectively increase Earth’s albedo. It has been suggested that these mirrors be placed in space, though this would be prohibitively costly. A more economical approach would be to cover large tracts of land, perhaps in desert areas, with inexpensive reflecting material. For example, to offset a radiative forcing of 3.7 W/m as expected for a doubling of CO, a simple estimate of the land coverage needed assuming additional reflection of 30% of an incoming daily average of 300 W/m would be % of the planet’s surface area. This is roughly the area ( M km) of all the world’s deserts. For comparison, the rough estimate in §23.6 of the fraction of desert that would need to be covered by solar systems to generate all world energy needs is roughly 10% of desert area or 2 M km.
