Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
6 ATMOSPHERIC CHEMICAL FEEDBACKS SUMMARY Understanding atmospheric chemistry feedbacks are crucial not only for future climate projections but also to connect measured concentrations of greenhouse gases with their emissions and for formulating control strategies and policies. Both gas phase and aerosol chemistry are influenced by temperature, humidity, cloudiness, and precipitation and have the potential to feed back on climate change. Both tropospheric and stratospheric chemical processes interact with temperature, humidity, circulation, and air composition changes. The oxidizing capacity of the atmosphere and the processes that convert effluents into pollutants may be altered by climate change. Current knowledge of aerosol composition, surface characteristics, and their spatial and temporal variations are inadequate. More research on atmospheric processes is required, with the goal of representing them more comprehensively in global climate models to elucidate the feedbacks. The physical and chemical processing of aerosols and trace gases in the atmosphere, the dependence of these processes on climate, and the influence of climate-chemical interactions on the optical properties of aerosols are the key areas that require better understanding and quantification. The recent improvements in the instrumentation for in situ aerosol characterization (e.g., chemical composition of aerosol particles on individual particle basis), optical extinction, and scattering measurements will allow rapid progress in this area. The ability to monitor aerosols from satellites and LIDAR (aircraft and ground-based) will allow large-scale characterization of aerosol climatologies and properties. Deployment of these instruments in clear and cloudy conditions in well-planned field studies augmented by laboratory and modeling studies is needed. Development of aerosol climatologies along with other variables such as emission inventories is essential and feasible. 76
ATMOSPHERIC CHEMICAL FEEDBACKS 77 Greenhouse gases (e.g., water vapor, carbon dioxide, methane, nitrous oxide, ozone, chlorofluorocarbons) and aerosols in the atmosphere interact with solar and terrestrial radiation and thus alter the radiative balance of Earth's climate system. Human activities have clearly altered the atmospheric abundance of many greenhouse agents since the pre-industrial era (circa 1750~. Humans alter some by direct emissions of the agents themselves (e.g., the gases CO2, CH4, N2O, and CFCs, and the aerosol soot), some through the emissions of precursors that through atmospheric chemistry impact the greenhouse agents (e.g., emissions of SO2 are oxidized to form sulfate aerosol, or emissions of NO make O3 and destroy CH4), and some through changes in temperature and other related factors (e.g., water vapor abundance). The largest individual greenhouse gas contributions to the overall human-driven rise in radiative forcing since 1750 is 1.46 W m~2 from CO2, 0.48 W m~2 from CH4, 0.35 W m~2 from 03, 0.17 W m~2 from CFC-12, and 0.15 W m~2 from N2O (IPCC, 2001a). Atmospheric chemical feedbacks arise when alterations in the surface temperature, precipitation, and other changes in climate interact with air chemistry to alter the abundance or properties of greenhouse gases or aerosols, which then produce an additional climate change. It is also important to note that interactions between climate and air chemistry can produce regional changes in air quality that may be a very important aspect of climate change. Water vapor, an important greenhouse gas, is predominantly determined in the troposphere by physical and dynamical processes in the natural climate system. Only in the stratosphere are chemical processes central in the determination of the water vapor concentration. To a first approximation, atmospheric chemical processes do not affect the abundance of water vapor and carbon dioxide in the troposphere. In the stratosphere, however, approximately 50 percent of water vapor (Kley et al., 2000) is generated by methane oxidation and is therefore influenced by chemical changes in the troposphere. Changes in the abundance of water vapor greatly affect chemical changes in the atmosphere. These processes affect such species as methane, ozone, and aerosols. For example, an increase in atmospheric water vapor due to increases in sea surface temperature will increase the production of the OH radical, the agent that cleanses the troposphere and controls the abundances of such greenhouse gases as methane and tropospheric ozone; changes in cloud abundance and cloud water content alters the rates of heterogeneous and multiphase chemical reactions. Changes in cloud coverage could also alter the actinic radiation that drives the photochemistry in the troposphere. Understanding how chemical processes are altered by changes in temperature and water abundance in the
78 UNDERSTANDING CLIA~1TE CHANGE FEEDBACKS atmosphere, and how chemical processes alter the concentrations of greenhouse gases and aerosols, constitute an important area of climate research. Climate-air chemistry feedback is obviously essential for projecting the consequences of the current emissions and for making policy decisions regarding regulating, decreasing, and trading emissions. While the long-lived greenhouse gases (e.g., CO2) are well mixed in the troposphere, other short-lived gases and aerosols are not well mixed and hence extremely variable in space and time. One of the keys to understanding the geographic distribution of future radiative forcing is the ability to project the abundance of greenhouse gases and aerosols and their spatial and temporal variations. Ozone in particular is an important species because it is produced in the atmosphere by photochemical processes. Aerosols are also often generated in the atmosphere. The distributions of both are controlled by their rates of production and destruction, as well as their atmospheric transport. There are a few other factors related to chemical processes that are of central importance to climate and its variation. Rapid, nonlinear changes in greenhouse gases, such as the release of methane from clathrates, can lead to catastrophic changes. AN EXAMPLE OF THE MULTIPLE DIMENSIONS OF CLIMATE- AIR CHEMISTRY FEEDBACK Methane is the greenhouse gas whose increase since the pre-industrial era provides a climate forcing that is second only to carbon dioxide. Methane is also a chemically active species that affects the abundance of the OH radical, the most important tropospheric oxidizer. Therefore, if methane is emitted into the atmosphere, it can decrease the abundance of OH radicals, which in turn will make methane degrade more slowly. This is a purely chemical feedback. Climate enters the feedback process if the rate of methane formation or destruction is affected by a climate variable such as temperature or water vapor abundance. This example also shows the possibility that the climate response to methane emission can also change the lifetime of methane in the atmosphere. To a crude approximation, the atmospheric lifetime of methane, · · . tCH4, IS given oy: 4 [k1 (T) x tOH]]
~ TMOSPHE~C CHEMICAL FEEDBACKS 79 where kitty is the rate coefficient for the reaction of OH with methane in the atmosphere at the temperatures in the atmosphere (DeMore et al., 1997; Vaghjiani and Ravishankara, 1991~. The product of kit and [OH] is averaged over the entire region where methane is degraded. Because the abundance of OH can be decreased by an increased emission of methane, clearly the lifetime of methane will increase. The change in temperature will alter the rate coefficient kit and hence alter the atmospheric lifetime and abundance of methane. Changes in water vapor will also greatly affect OH since the primary process that produces OH in the atmosphere involves a competition between water vapor and nitrogen (oxygen) for the removal of Of D) produced by the photolysis of ozone. O3 v > 0~1D'+O2 O(1 D) + N2 /O2 ~ GAP) + N2 /O2 o(1D)+H2O~2 OH This example also shows how changes in other atmospheric constituents feed back on greenhouse gases through alterations in their lifetime. Ozone levels affect the abundance of other radiative gases and alter the abundance of ozone itself. Therefore, it is very important to note the feedbacks that involve ozone, an important anthropogenic greenhouse gas. The way ozone affects the abundance of chemically active radiative gases is through the alteration of the capacity of the troposphere to oxidize such species. For example, an increase in ozone abundance in the troposphere will lead to an increase in the production of OH, which in turn affects the tropospheric lifetimes of species. An increased oxidative capacity, when coupled with emissions of hydrocarbons and nitrogen oxides, leads to further production of ozone itself (i.e., an increase in hydrocarbons and nitrogen oxides leads to more production of ozone [Seinfeld and Panis, 19983~. Clearly the above example is not simple. An increase in gas phase hydroxyl radical can also enhance the abundance of gas phase hydrogen peroxide (;~:I2O2~. H2O2 oxidizes SO2 in liquid droplets. Thus, even though an increase in OH will increase the sulfate production rate (either through gas phase or through liquid phase reactions), the consequences to the atmosphere could be different. Gas phase production of H2SO4 can lead to a burst of new particles when there are few existing particles, while liquid phase oxidation will only grow existing particles (Seinfeld and Panis, 1998~. The consequences of a larger number of particles are different from the same
80 UNDERSTANDING CLIMATE CHANGE FEEDBACKS mass of larger particles. For example, the former may lead to brighter clouds than the latter given the same amount of water vapor (Twomey, 1991~. H2O in, / Light Nitrogen ~ Radicals oxide ~ / - Hydrocarbon CO FIGURE 6.1 A simplified illustration of feedbacks between ozone and the hydroxyl radical. Unlike other greenhouse gases, ozone is photochemically produced from other emissions (Figure 6.1~. This unique nature of ozone makes its abundance highly susceptible to changes in other atmospheric abundances and conditions. In addition to its direct role as a greenhouse gas, ozone is also the precursor to the OH radicals during the daytime and NO3 radicals at night, which initiates the degradation of most atmospheric species. Ozone itself is a gas phase oxidant for various olefin organic species and a liquid phase oxidant for many sulfur species, particularly SO2 that is converted into sulfuric acid (Seinfeld and Panis, 1998~. Hence, changes in atmospheric variables can change ozone abundance and drive feedbacks through ozone. Hydrocarbons, nitrogen oxides, and sunlight dictate the production of ozone in the troposphere, the OH radical is the initiator of its production. The abundances of all these species will affect ozone production. The impact of climatological variables on OH abundance was discussed earlier. The emission of hydrocarbons is controlled by anthropogenic sources and the biosphere. The abundance of nitrogen oxides is controlled by emission (anthropogenic and natural) and production by lightning. Thus, connections of atmospheric ozone to biosphere, hydrological cycles, clouds, and temperature are evident. This is an example of how atmospheric variables and their changes can lead to a feedback in the chemical system.
ATMOSPHERIC CHEMICAL FEEDBACKS OVERVIEW OF FACTORS CONTROLLING CLIMATE, CHEMISTRY, AND AEROSOL INTERACTIONS 81 An increase in the oxidative capacity of the troposphere will lead to changes in the rates of production and destruction of species that lead to the generation and growth of aerosols. Clearly a large fraction of the aerosol in the troposphere is produced from gas phase emissions that lead to condensable chemicals. Water and these chemicals, either by themselves or with other condensables, lead to the production of aerosol. For example, sulfate aerosol is generated by the oxidation of SO2 to gas phase sulfuric acid, which condenses (some times with other species such as ammonia) in the presence of water vapor to produce sulfate aerosol. Therefore, an increase in OH would lead to more H2SO4 and hence more sulfate. Thus, there is a direct coupling between the abundances of reactive species and aerosol. One of the key factors in the formation, sustenance, and composition of the clouds is the cloud condensation nuclei that are needed. It is known through various studies that cloud condensation nuclei can be generated by aerosol, which is processed in the atmosphere. The atmospheric processing can render those unique properties needed to be a cloud condensation nucleus to an aerosol particle. Oxidants in the troposphere can alter the surface (and even the bulk) of the aerosol. For example, a hydrophobic organic aerosol (or an aerosol coated with an organic layer) can be oxidized to produce chemical functional groups that convert a hydrophobic to a hydrophilic aerosol. Hydrophilic aerosols are a key to many processes in the atmosphere, and especially for the formation of cloud condensation nuclei. Thus, changes in gas phase composition can alter the hygroscopicity of aerosols and hence the ability of aerosols to induce nucleation of droplets (i.e., cloud formation). In addition, the presence of aerosols in the atmosphere greatly alters the composition of the atmosphere because of heterogeneous and multiphase reactions that occur on or in the aerosol. The consequences of heterogeneous and multiphase reactions on the composition can be dramatic as in the case of the Antarctic ozone hole. They can also be less dramatic but extremely significant for global budgets. For example, aerosols convert active nitrogen oxides to nitric acid and hence reduce the ability of the atmosphere to photochemically generate ozone. Thus, chemical processing, coupled with the generation of aerosols in the atmosphere, couples atmospheric chemical processes with the important cloud feedback mechanism. The biosphere interacts with the atmosphere, and these interactions have a significant impact on the climate system. A large number of the chemicals in the atmosphere originate in the biosphere. These include such gases as
82 UNDERSTANDING CLIAl 4 TE CHANGE FEEDBA CKS methane, hydrocarbons that alter the chemistry of the troposphere, and precursors to aerosols from the terrestrial regions and dimethylsulfide (DMS), methyl bromide, organics, and others from oceans. Of particular note is the emission of DMS from the oceans, such emissions have been proposed as a possible feedback on the climate system (Charlson et al., 1987; Shaw, 1983~. This well-publicized feedback links the production of aerosol from dimethylsulfide oxidation and the subsequent change in the cloudiness of the planet to surface temperature and insolation, which then affect the emission of DMS from the oceans. The increased emission of DMS to the atmosphere leads to an increased sulfate aerosol production. In this hypothesis the increased sulfate production is expected to alter the properties of the clouds, which decrease the incoming solar radiation and thus cool the surface and decrease insolation. Whether such a change at the surface would increase or decrease DMS emissions was left open by Charlson et al. (1987~. Such a feedback system clearly connects the temperature and insolation changes with oceanic emissions, to gas phase processes, to cloud processes, to radiation changes. Some links in the feedback have some support, such as the seasonal correlation of cloudiness, non-sea salt sulfate and sea-to-air DMS fluxes in the Southern Hemisphere (Ayers et al., 1991; Boers et al., 1994~. There is also some tentative support for a positive correlation between the sea-to-air DMS flux and surface solar radiation, suggesting that the feedback may be negative (see Chapter 8~. In contrast Bates and Quinn (1997) found the DMS output in the equatorial waters to be invariant and concluded that the connection between DMS emission and atmospheric and oceanic variables remains "elusive." In general the mechanistic understanding is far from complete. For example, the reaction pathways of DMS beyond its original reaction with OH are poorly known (Davis et al., 1999), as is the relationship between cloud droplet number and cloud condensation nuclei (Lohman et al., 1999~. As discussed in Chapter 8, the processing of DMS by marine planktonic ecosystems is just beginning to be elucidated. Therefore, this hypothesis of a strong climate feedback process involving DMS is in an uncertain state. As discussed above, the connection between gas phase oxidation to new particle formation and its coupling to cloud condensation nuclei formation is an example of the coupling between purely chemical processes and other atmospheric feedbacks.
TMOSPHENC CHEMICAL FEEDBA CKS Chemistry and Transport Interactions 83 The change in the abundance of the upper tropospheric reactive species because of changes in the transport processes or the increase in water vapor is another example of coupling between chemical and other atmospheric feedback processes. As the climate warms, the lapse rate and the abundance of water vapor in the upper troposphere will change (see Chapter 3~. These changes will influence both the transport and mixing of chemicals to the upper troposphere and their chemical transformations in the upper troposphere. A change in the vertical transport of reactive species that act as precursors for OH for example acetone, methyl hydroperoxide, formaldehyde, or acetaldehyde will enhance the chemical reactivity of the upper troposphere and increase the production of ozone in this radiatively crucial region of the atmosphere. This is especially efficient if nitrogen oxides are transported along with the other active ingredients into the upper troposphere. Stratospheric Chemistry-Surface Climate Interactions Changes in the stratosphere and their impact on the troposphere are also of major interest. Increases of greenhouse gases or the release of chemicals that destroy ozone in the stratosphere can produce large dynamical changes in the stratosphere that influence the surface climate. Within the stratosphere, chemical processes, radiative processes, and dynamical processes are all strongly coupled. Changes in stratospheric ozone abundances and their future levels can be altered by the changes in stratospheric temperature and water vapor, which are driven by greenhouse gases. Greenhouse warming at the surface and cooling of the stratosphere by increased carbon dioxide, methane, and water vapor may delay the expected recovery of the ozone layer, the ozone hole may persist longer, and Arctic ozone depletions may continue beyond the time currently estimated. These changes will impact the ultraviolet (UV) radiation available in the troposphere. Furthermore, the dynamical and transport consequences of the ozone changes on the troposphere can also be significant (Hartmann et al., 2000; Shindell et al., 2001; Thompson and Solomon, 2002~.
84 UNDERSTANDING CLIMES TE CHANGE FEEDBACKS Regional Climat - Air Chemistry Interactions An impact of global climate change will be regional changes in chemical composition, UV levels, deposition rates, emission rates, precipitation rates, and other variables. Therefore, regional and urban air quality will be impacted by global climate changes superimposed on regional and local conditions. For example, changes in water vapor due to climate change will impact local and regional ozone production and the rate at which chemicals are removed from the location of emission. The global- scale changes will also alter the transport of species in and out of a given region of interest. Therefore, requirements for attainment of an air quality standard in a given region or location will be affected by climate change. Factors That Control Chemical Feedbacks Processes that are affected by changes in temperature, water vapor abundance, and other climate variables by means of atmospheric chemical processes are 1. gas phase oxidation processes; 2. heterogeneous and multiphase chemical processes; 3. photolytic processes; 4. transport and mixing of atmospheric constituents; 5. emissions from the terrestrial and oceanic sources that control the flux of species into the atmosphere; and 6. deposition of atmospheric degradation products and constituents that remove the chemical constituents from the atmosphere. These processes are also affected by factors such as UV radiation, flux into the atmosphere, and flux out of the atmosphere. The representation of these processes in models is the key to the recognition and quantification of the role of feedbacks. DEVELOPING A SCIENTIFIC STRATEGY As discussed above, there are many chemical feedbacks in the atmosphere; most of them have been qualitatively identified and some of them have been assessed to a limited extent (i.e., the sign of the feedback is known and in some cases the magnitude is known roughly). The effects of
A TMOSPHERIC CHEMICAL FEEDBACKS 85 other feedbacks discussed in this document on atmospheric chemistry feedbacks can be large and greatly alter the impact on crucial areas such as regional and urban air quality. Quantitative assessments of the effect of the feedbacks outlined in this chapter on the overall radiative balance, the surface temperature change, or some other "impact proxy" are lacking. Many of the atmospheric chemistry feedbacks are identified in principle and have been semi-quantitatively evaluated by their proponents, but reliable quantification awaits a consensus from the community that includes a more accurate treatment of the key processes and their integration into comprehensive models. The first step in dealing with climate-air chemistry feedbacks is to better understand the atmospheric processes responsible for the formation and destruction of the chemical species of interest. These processes must be understood at a sufficient level to quantitatively evaluate them. Such quantification requires detailed knowledge of the chemical processes (e.g., their rates, products, their variation with atmospheric conditions) and an accurate knowledge of the composition of the current atmosphere. Because of the spatio-temporal complexity of the climate system, the role of transport and mixing processes, and the local nature of many of the feedback processes, it is necessary to incorporate these processes in a global climate model to test their global significance and assess their local consequences. Because many of the species of interest, especially the aerosols, are highly variable in space and time, the resolution of the models has to be sufficient (e.g., 1° x 1°, unless processes such as convection are being explicitly simulated, which require higher resolution) to capture the nonlinearities in the processes. The representation of processes in models must be sufficiently faithful representations of nature to deal with nonlinearities in processes and their coupling to other Earth system processes. Such an advance is essential before the contribution of a feedback can be calculated. When a sufficient fundamental understanding of the basic processes that couple air chemistry and climate is achieved, these processes should be incorporated in regional and global atmospheric climate models. These models are essential to integrate and hence quantify the key climate- chemistry feedbacks. Regional models are needed to evaluate detailed emissions-chemistry-climate interactions, and global models are needed to evaluate interactions with the atmospheric general circulation and broader Earth system. Aerosols and their studies also deserve special attention. Although the paradigm for studying gas phase processes appears to be reasonably well established and has been reasonably successful, studies of and on aerosols are at a very early stage. Currently knowledge of the composition, surface
86 UNDERSTANDING CLIMATE CHANGE FEEDBACKS characteristics, and their spatial and temporal variations is at best rudimentary. The processes that lead to the production of aerosols (the nucleation processes) are incapable of explaining alone the current observations of aerosol distributions. Therefore, the physical and chemical processing of aerosols, the dependence of these processes on climate, and the influence of climate-chemical interactions on the optical properties of aerosols must be elucidated. They can be done in steps (e.g., observation and understanding of how aerosols change with conditions, connecting the atmospheric conditions to climate variables, and measuring the optical properties under different conditions of temperature, humidity, and composition) The research needs can be summarized as follows: · A complete understanding of the emissions, atmospheric burden, and final sinks for carbonaceous aerosols needs to be developed. This class of aerosols includes a wide range of different species that are often simply characterized as organic and elemental or soot. They act as greenhouse agents, can either warm (soot) or cool (organic) the climate, and alter clouds and the hydrological cycle. To understand the role of anthropogenic activities in changing the atmospheric burden of the carbonaceous aerosols, speciated measurements of the aerosols at the emission source and in the atmosphere need to be made with the same techniques so that atmospheric burdens can be attributed to specific sources. In addition, absolutely calibrated emissions inventories need to be developed for each species of carbonaceous aerosols so that the atmospheric measurements provide a true test of the global models. This should include airborne, satellite-based, and ground-based observations. Airborne and satellite measurements have become more feasible because of improvements in instrumentation, data reduction algorithms, and input data. Reduction of existing satellite data to retrieve aerosol optical depth is being investigated. · The processes and the global range of conditions under which carbonaceous and other aerosols can interact with the cloud and hydrological cycle need to be defined. The key atmospheric processes that influence the radiative, cloud condensation, and ice condensation properties of aerosols need to be characterized. Intensive regional measurement campaigns (on the ground, airborne, by satellite) should be mounted that are designed specifically to improve global aerosol models so that the improved knowledge of the processes can be directly applied in the predictive models that are used to assess future climate change scenarios. Better use needs to be made of the recent development of instrumentation to measure the chemical composition of aerosols, ability to measure in-situ extinction and
ATMOSPHERIC CHEMICAL FEEDBACKS 87 scattering, advances in the microphysical modeling, and availability of other ancillary input data. These characterizations should be done in concert with other field, lab, and modeling studies to improve the needed database for process understanding. · The key processes that control the abundance of tropospheric ozone need to be quantified, including but not limited to stratospheric influx; natural and anthropogenic emissions of precursor species such as NOx, CO, and VOC; the net export of ozone produced in biomass burning and urban plumes; and the loss of ozone at the surface. Improved characterization is required of the type and magnitude of chemistry-climate feedbacks that would lead to alteration of these processes with future climate change. · The chemical feedbacks that can lead to changes in the atmospheric lifetime of CH4 need to be identified and quantified (This could be defined equivalently as a measure of the global mean OH abundance) by careful integration of models and measurements; there is no clear method for deriving these feedbacks from measurements alone. These feedbacks on tropospheric OH include stratospheric ozone depletion, increasing temperatures and water vapor in a future climate, changing emissions of NOx and CO from both natural ecosystems and anthropogenic activities, alterations in lightning production of NOx, and of course the increasing abundance of CH4.
