Aerosols and Climate
- Bjørn H. SamsetBjørn H. SamsetCICERO Center for Climate and Environmental Research
Among the factors that affect the climate, few are as diverse and challenging to understand as aerosols. Minute particles suspended in the atmosphere, aerosols are emitted through a wide range of natural and industrial processes, and are transported around the globe by winds and weather. Once airborne, they affect the climate both directly, through scattering and absorption of solar radiation, and indirectly, through their impact on cloud properties. Combining all their effects, anthropogenic changes to aerosol concentrations are estimated to have had a climate impact over the industrial era that is second only to CO2. Their atmospheric lifetime of only a few days, however, makes their climate effects substantially different from those of well-mixed greenhouse gases.
Major aerosol types include sea salt, dust, sulfate compounds, and black carbon—or soot—from incomplete combustion. Of these, most scatter incoming sunlight back to space, and thus mainly cool the climate. Black carbon, however, absorbs sunlight, and therefore acts as a heating agent much like a greenhouse gas. Furthermore, aerosols can act as cloud condensation nuclei, causing clouds to become whiter—and thus more reflecting—further cooling the surface. Black carbon is again a special case, acting to change the stability of the atmosphere through local heating of the upper air, and also changing the albedo of the surface when it is deposited on snow and ice, for example.
The wide range of climate interactions that aerosols have, and the fact that their distribution depends on the weather at the time and location of emission, lead to large uncertainties in the scientific assessment of their impact. This in turn leads to uncertainties in our present understanding of the climate sensitivity, because while aerosols have predominantly acted to oppose 20th-century global warming by greenhouse gases, the magnitude of aerosol effects on climate is highly uncertain.
Finally, aerosols are important for large-scale climate events such as major volcanoes, or the threat of nuclear winter. The relative ease with which they can be produced and distributed has led to suggestions for using targeted aerosol emissions to counteract global warming—so-called climate engineering.
On June 15, 1991, Mt. Pinatubo in the Philippines erupted, ejecting a massive mushroom cloud of volcanic ash and dust. Billions of tons of fine-grained magma were emitted to the atmosphere, along with 20 million tons of sulfur compounds. The dust cloud gained altitudes of over 20 kilometers, entering the stratosphere, and in the following months spread out to cover the entire Northern Hemisphere.
At the same time, global mean temperature dropped by several tenths of a degree Kelvin, and stayed low for around two years. The Mt. Pinatubo eruption and its influence on global temperature made for a dramatic example of the impact that aerosols—minute particles suspended in the atmosphere—can have on our climate. It also marked a turning point in the scientific study of aerosols as an important and complex factor in climate evolution. A U.S. research group, led by climate scientist James Hansen, used a numerical climate model to correctly predict the broad features of the climate’s response to the eruption (Hansen, Lacis, Ruedy, & Sato, 1992).
Volcanoes are a natural, and unpredictable, source of atmospheric aerosols. Their effects are potentially very strong but always transient. Over the industrial era, however, anthropogenic activities have also added sources of aerosol emissions. Major examples are soot from incomplete combustion in diesel engines, cook stoves, and large-scale agricultural burning, and sulfate compounds resulting from SO2 emissions, mainly from fossil fuel usage. Like volcanic ash, these emissions also have an impact on our climate and need to be studied in detail if we are to understand the observed ongoing changes to global temperature and precipitation patterns.
In the early 21st century, aerosol physics, chemistry, and climate connections are among the most active subfields of climate science. It is generally recognized that changing aerosol emissions, both natural and anthropogenic, have played a key part in the climate evolution of the 20th century, obscuring the impact of the strengthened greenhouse effect caused by emission of CO2 and other long-lived greenhouse gases.
Due to the many ways in which aerosols may impact the climate, their actual role is however far from certain. In the summaries of anthropogenic radiative forcing regularly produced by the UN Intergovernmental Panel on Climate Change (IPCC) (Myhre et al., 2013b), increased CO2 concentration can be seen to dominate, but the main uncertainty on our net impact on the global energy balance comes from the wide range of aerosol–climate interactions.
This review first gives an overview of what atmospheric aerosols are and where they come from: emissions and secondary production, and anthropogenic and natural processes. Then, the various ways in which aerosols can affect the climate are discussed, before tackling their precise contributions to ongoing climate change. Global concerns about aerosols are covered, such as past and future major volcanic eruptions, the specter of nuclear winter, and the possibility of using them for large-scale geoengineering. Finally, some historical notes on the evolution of the scientific study of aerosols in the climate context are given, and some still-unanswered scientific questions are listed.
What Aerosols Are and Where They Come From
Aerosols, or particulate matter, are airborne particles suspended in the Earth’s atmosphere. The particles can be liquid, solid, or of mixed phase, and their diameters typically range from 0.01μm to 100μm (Boucher et al., 2013).
Aerosols can be either emitted directly to the atmosphere, as is the case for dust or sea spray swept up by winds and soot from incomplete combustion, or produced as a result of atmospheric chemical reactions from so-called precursor gases, as when SO2 emissions react with water vapor to produce sulfate compounds.
Once airborne, aerosols may undergo further chemical reactions or merge through coagulation (Isaksen et al., 2009). Some may change their size through hygroscopic growth. They may get washed back out of the atmosphere by encountering precipitation, or simply fall out through gravitational settling. The average lifetimes of aerosols in the atmosphere range from days to weeks, depend strongly on their size, emission location, and chemical properties, and are hard to measure.
Major Aerosol Types and Sources
Aerosols present in the atmosphere today are commonly categorized as either primary (emitted) or secondary (produced), and natural (resulting from a natural process) or anthropogenic (the product of industrial or other human activities). This overview uses assessments of lifetime and abundances from Boucher et al. (2013).
Natural Aerosol Sources
The most common natural aerosols are salt particles from sea spray droplets, mineral dust, and carbonaceous compounds from wildfires.
Sea spray aerosols are emitted to the air through the bursting of bubbles when waves break, and their production is therefore dependent on the strength of surface winds. It is estimated that 2,000–7,000 Tg (teragrams) of sea spray is emitted annually, mainly from the mid-latitude oceanic regions in both hemispheres.
Mineral dust is produced by disintegration of larger rocks, and swept up by winds from arid land regions. Major sources are desert regions of Northern Africa and the Middle East, Australia, and parts of Southern Asia. Estimates of annual dust emissions range from 1,000 to 4,000 Tg.
Emissions of carbonaceous aerosols, or soot, from natural wildfires are less well constrained, and also vary more from year to year. Wildfires occur in regions that both support significant growth of vegetation and are vulnerable to extended dry seasons. Examples are South America, boreal forests, and many regions of Central Africa. Aerosol emissions from such wildfires are commonly termed biomass burning aerosols.
In addition to these primary sources, natural emissions of aerosol precursors such as dimethyl sulfides, monoterpenes, isoprene, and biogenic volatile organic compounds are estimated to induce formation of hundreds to thousands of Tg of aerosols annually.
Anthropogenic Aerosol Sources
Major types of anthropogenic aerosols are sulfate and nitrate compounds, organic aerosols, and black carbon. Of these, sulfate and black carbon have received the most attention in the context of anthropogenic climate change.
Anthropogenic sulfate aerosols result from emissions of sulfur dioxide (SO2), which, once airborne, react with water vapor to produce sulfate compounds. The main source is combustion of fossil fuels and biomass. Global, annual emissions are estimated to be around 50 Tg of primary sulfur, concentrated around the largest industrial regions, and along major shipping lanes. Sulfate aerosols are highly hygroscopic; that is, they can undergo significant growth if they are emitted or transported into humid air.
Anthropogenic black carbon (BC), or soot, is a collective term for the products of incomplete combustion of fossil fuels, biofuels, and biomass. BC comes in a wide variety of shapes and sizes and can undergo significant changes after emission that alters its radiative and chemical properties (so-called BC aging). Around 5 Tg is estimated to be emitted annually—primarily from industrial and major crop-burning regions—though there is at present considerable uncertainty surrounding this number. A key feature of BC, distinguishing it from organic carbon and most other species of aerosol, is its ability to absorb solar radiation, which gives it a climate effect similar to a short-lived greenhouse gas.
Nitrate aerosols are produced from the oxidation of NOX emissions, primarily from combustion of nitrogen-bearing fuels. As for sulfate, they originate mainly from industrial regions and along shipping lanes, though reductions of nitrogen content in ship fuels from the late 2000s has reduced the latter source.
Organic aerosols (OA) are carbonaceous compounds that do not absorb solar radiation. They come from primary emission sources such as combustion of fossil fuels, biofuels, and biomass (known as primary organic aerosols, or POA), or from organic emissions, for example, from vegetation (known as secondary organic aerosols, or SOA). Sometimes the term organic carbon is also used. Recent research indicates that some OA may be weakly absorbing, particularly in the blue–to–ultraviolet part of the solar spectrum, placing them somewhere between the established categories OA and BC (Liu et al., 2015). Such aerosols are becoming known as brown carbon because of the spectral variation in absorption.
Observing and Modeling the Global Aerosol Abundance
Knowing the main primary and secondary sources of aerosols is not sufficient to determine their concentration and composition in the atmosphere at any given location and time. After emission or production, aerosol transport and evolution will be subject to the prevailing weather, ambient temperature, and humidity. The particles may be washed out by rain more or less immediately, or they may get transported to high altitudes and across vast distances. Global wind patterns are such that aerosols rarely cross from one hemisphere to the other, but transport of aerosols from one continent to another is not uncommon.
To determine global abundances and composition of aerosols, a combination of observations and detailed modeling is used. Observations of aerosol properties are done from satellites, permanent ground-based stations, and mobile stations on ships and aircraft. Such observations are mainly able to discern the amount of sunlight either scattered or absorbed by the total amount of aerosols in the atmospheric column, and to some extent the size distribution of particulate matter. They are rarely able to divide aerosols into source categories, as done here in the section Major Aerosol Types and Sources.
The most commonly used observable quantity to study aerosol abundances is the aerosol optical depth (AOD), which quantifies the amount of incoming sunlight removed by aerosols when traversing down to the surface through scattering and absorption. By measuring AOD at several wavelengths, information on aerosol size distributions may be inferred.
Numerical models, on the other hand, use explicit emission inventories for each source category, and attempt to treat the complex physical and chemical processes that transport and alter aerosols after emission. They are, however, limited in resolution and hampered by incomplete knowledge of emissions, the physical and optical properties of the particles, and their treatment of weather and clouds.
Here, an optical depth of 1 means that the solar radiation reaching the surface is reduced to a fraction of its top-of-atmosphere value, 1/e. The map indicates the present-day major sources of aerosols: biomass burning emissions from some regions of Africa and South America, dust from arid regions around the equator, and anthropogenic emissions in the industrial regions of North America, Europe and—most notably—Southeastern Asia. The mid-latitude sea spray regions can also be seen.
Overlaid are pie charts showing decompositions into individual aerosol categories, as calculated by a numerical model (the OsloCTM, a detailed chemical transport model). They show how the local composition varies from being dominated by anthropogenic species such as sulfate and BC close to industrial regions, to dust and sea salt in remote areas. This difference in composition with region matters significantly, for both the local and global climate impact of aerosols, as will be discussed in the section Aerosols in the Context of Anthropogenic Climate Change. In addition, there is significant variability of species with altitude, which influences the climate impact of absorbing aerosols such as BC.
In summary, while aerosols are present across the globe, throughout the troposphere, and even in the stratosphere, their abundances and composition vary greatly with region, season, and altitude. All of these factors are of importance for the net climate impact of aerosols—both in general, and as part of the ongoing anthropogenic climate changes.
Broadly stated, the boundary conditions of the climate are determined by the balance and flow of energy in the ocean/atmosphere system. The Earth is hit by solar radiation, some of which is instantly reflected back into space by clouds and high-albedo land surfaces. A significant fraction is however absorbed by the surface, notably the oceans, and later emitted back as longwave heat radiation. Some of this radiation is kept in the atmosphere for a while, by absorption and re-emission by greenhouse gases; that is, the greenhouse effect. Over time, the surface temperature of the planet will adapt to this radiative balance and reach an equilibrium value, with some interannual variations in temperature and precipitation largely due to internal rearrangement of energy in the climate system. Since the climate of the Earth is broadly quite stable, having until recently seen less than a degree Kelvin of change over the last 10k years (the present interglacial hot period, the Holocene), we know that the balance of incoming and outgoing energy from the Earth system must be remarkably good.
For aerosols to impact the climate system, they must in some way affect the incoming shortwave energy, the outgoing longwave energy, or the internal flow of energy. The example in the introduction of the drop in global temperatures after the Mt. Pinatubo eruption shows that such aerosol–climate interactions exist and can be significant. Since the 1980s, aerosol climate science has uncovered a wide range of climate interconnections, and determined that some of them have been significant enough to have had impacts on the climate, almost on par with those from increased greenhouse gas concentrations.
Direct Interaction With Incoming Sunlight
The most obvious aerosol–climate interactions are the ability of aerosols to scatter or absorb incoming sunlight.
The first of these, scattering, is what was in play after the Mt. Pinatubo eruption. The volcano emitted large amounts of dust, and of SO2, which subsequently transformed into sulfate compounds in the atmosphere. A significant fraction of the sulfate ended up in the stratosphere, where there is no precipitation to wash the aerosol back out and atmospheric dynamics acts to keep the aerosol airborne.
Over the course of a few months the sulfate spread around most of the globe. This layer of particles scattered back so much additional sunlight to space that surface temperatures dropped in response. The global energy balance had, albeit temporarily, been offset.
Scattering of Incoming Sunlight by Aerosols Can Cool the Climate
All species of aerosol are, to various degrees, able to scatter sunlight. To determine their precise scattering power, however, we need to know their sizes and optical properties. Optical properties vary between species, and are usually determined through lab experiments. Aerosol size distributions are more challenging, as they are strongly modified by hygroscopic growth. For example, sulfate is highly hygroscopic, meaning that it very readily grows if the ambient humidity is high; this process strongly modifies its scattering properties. Nitrate is also hygroscopic, but to a lesser degree than sulfate. Black carbon is not hygroscopic when it is emitted, but can become so if it gets coated with sulfate or other hygroscopic compounds.
Optical processes, size distributions, growth, and aging factors are at present included in global climate models (Mann et al., 2014). Based on Mie theory, numerical radiative transfer libraries solve the differential equations describing the interactions of incoming shortwave sunlight with both aerosols and other gases in the atmosphere. Of all the aerosol–climate interactions, this scattering effect—part of what is commonly termed the aerosol direct radiative effect—is the one best understood.
Aerosol Absorption of Incoming Sunlight Can Warm the Climate
Some aerosol species, notably black carbon, can also absorb incoming shortwave radiation (Bond et al., 2013). This heats the aerosol particles, which in turn emit longwave heat radiation to the ambient air. Increasing concentrations of absorbing aerosols may therefore warm the climate, similar to the effect of greenhouse gases.
As was the case for scattering, the ability of an aerosol particle to absorb sunlight depends on size, optical properties, and morphology. In the early 21st century, “black carbon” is understood to include a wide range of products of incomplete combustion, from small, single spheres to long chains with greatly varying shapes. Understanding the time evolution of black carbon optical properties after emission is presently an active field of research.
Location also matters, as does altitude. When located above a highly reflective surface, such as ice or clouds, absorbing aerosols may interact both with downwelling and reflected radiation. As a result, averaged aerosol absorption efficiency has been found to increase dramatically with altitude, up to the tropopause, above which there are no more clouds. This means that the impact of black carbon on the global energy balance gets stronger the higher up in the atmosphere it is transported. It is also elevated above ice sheets such as at the poles or over Greenland. Hence, thorough knowledge of transport mechanisms, removal rates, and atmospheric lifetime is crucial to determine the total perturbation to the global energy balance from absorbing aerosols.
Deposition of Black Carbon on Snow
Black carbon, which, as the name implies, has a dark color, has an additional climate interaction after it is removed from the atmosphere. If the surface it lands on originally has high albedo, black carbon will darken it, causing absorption of sunlight and hence heating of the surface.
This process is especially important on snow and ice. In the Arctic, long-range-transported black carbon deposition is thought to contribute both to an increased melting rate of sea ice and to the observed heightened rate of Arctic warming relative to the global mean (Arctic amplification) (Pithan & Mauritsen, 2014). Deposition on glaciers is also thought to contribute to the significant reduction in volume of Himalayan glaciers.
Aerosols Affecting Clouds
In addition to direct interaction with sunlight, aerosols may also affect the formation, albedo, and lifetime of clouds. These processes have been termed the aerosol indirect and semidirect effects (see the subsections Cloud Albedo Effect and Cloud Lifetime Effect for alternate names).
Cloud Albedo Effect
Clouds form when supersaturated water vapor condenses into droplets. To get started, this process needs physical objects for the first water molecules to attach to—so-called cloud condensation nuclei (CCN). Many aerosol species are highly efficient as CCN, meaning that a change in aerosol concentration will alter the rate of cloud droplet formation.
Natural aerosols, for example, dust or sea spray, are present almost everywhere throughout the troposphere. Hence, a shortage of CCN is usually not an inhibiting factor for cloud formation. However, the concentration of CCN may still determine the number of droplets formed for a given amount of atmospheric water.
Due to multiple scattering, a cloud consisting of many small droplets will appear whiter than one made up of fewer, larger drops containing the same amount of water. Whiter clouds reflect more sunlight. Hence, an increase in the concentration of aerosols in a cloud-forming layer will lead to increased reflection of incoming radiation—that is, a cooling effect on the Earth’s atmosphere.
The idea that anthropogenic aerosol emissions could alter cloud properties in this way was first introduced by Sean Twomey in a series of papers in 1977 (Twomey, 1977). When satellite imagery of ocean regions became readily available in the late 1980s, the concept was shown to be highly relevant as ship exhaust was clearly seen to modify clouds above and downwind of the emission tracks. See Figure 3 for a recent example.
The cloud albedo effect is sometimes termed the Twomey effect in the literature, or the 1st indirect effect.
Cloud Lifetime Effect
Next, altering the number concentration and size distribution of cloud droplets will also affect the atmospheric lifetime of a cloud (Albrecht, 1989). Clouds dissipate through evaporation of droplets, or through droplet growth via collisions, which eventually leads to precipitation. A larger concentration of CCN and subsequent change in droplet concentration and size, will act to reduce the rate of precipitation from a given cloud, but may also enhance evaporation, depending on cloud type and local thermodynamic conditions.
Whether the cloud lifetime effect from an increase in aerosols will heat or cool the climate depends both on the altitude of the modified cloud and on the sign of the lifetime change. Low clouds normally cool the surface by blocking incoming sunlight. Conversely, high-altitude clouds heat the surface by reflecting back outgoing longwave radiation. Hence, the net magnitude and even the sign of the cloud lifetime effect from present anthropogenic aerosol emissions are poorly known.
The cloud lifetime effect is sometimes alternately termed the Albrecht effect in the literature, or the 2nd indirect effect.
Finally, the presence of aerosols may affect the convective stability of the atmosphere and the ambient temperature within a cloud. Both of these effects may indirectly alter cloud formation rates and hence influence the climate. Collectively, these effects are termed the aerosol semidirect effects.
Convective cloud formation begins with updraft from the heated planetary surface. The strength of this updraft is dependent on the change in temperature with altitude, or lapse rate, through the troposphere. If an absorbing aerosol, like black carbon, is inserted at a high altitude, it will heat the ambient air and cool the surface, reducing the lapse rate. Hence, cloud formation below the aerosol layer may be reduced. At the altitude of the aerosol, however, the increased temperature may inhibit cloud formation. Further, absorbing aerosols embedded in droplets within an already existing cloud may contribute to its evaporation, so-called cloud burnoff.
The semidirect aerosol effect is at present poorly understood, although it is an active field of research. Improvements are needed, through both observations and the representation of the semidirect effect in models.
In summary, atmospheric aerosols have a range of direct and indirect interactions with the climate. They can both directly scatter and absorb incoming, or reflected, shortwave radiation, and affect emitted longwave heat radiation through absorption and by affecting clouds. By acting as CCN, they alter both the formation and evaporation rates of clouds, and hence also their lifetimes and precipitation rates. By altering atmospheric stability, they also affect convective cloud formation. All of these interactions are potentially significant enough to have global impacts on the Earth’s climate, should the long-term concentration of aerosols change.
Aerosols in the Context of Anthropogenic Climate Change
Over the industrial era, anthropogenic activities have introduced sources of atmospheric aerosols in addition to their natural formation processes. While aerosols stay in the atmosphere only for some days, the anthropogenic sources are still large enough to alter their global mean concentrations. In the early 21st century, aerosols are counted as one of the major factors that have altered the global energy balance since preindustrial times. Their role is, however, much harder to pin down than that of the well-mixed greenhouse gases. One reason for this is the short atmospheric residence time of aerosols, which means that to get a handle on their total radiative effects, we would really need to measure them everywhere and at all times—which we obviously cannot do. Another is that because their composition in the atmosphere is so variable, both spatially and over time, their radiative impacts at the surface and at the top of the atmosphere can be largely independent, while for greenhouse gases, one may be inferred from the other. These and other effects turn aerosols into a major source of uncertainty in projections of future climate.
In this section, we review present estimates of the radiative forcing exerted by changes in aerosol abundances, through the climate interactions discussed in the section Aerosol–Climate Interactions. Radiative forcing (RF) is defined as the perturbation to the global energy balance due to a change in the climate system, before any response in temperature or other climate processes has occurred. Forcing estimates are primarily taken from the IPCC 5th Assessment Report (AR5). We also discuss other topics relevant to anthropogenic climate change, such as the link between aerosols and precipitation, and how they hamper our ability to determine the transient climate sensitivity from historical data.
Direct Radiative Forcing From Anthropogenic Emissions
Radiative forcing due to the direct interaction of atmospheric aerosols with radiation, colloquially termed direct radiative forcing, has been estimated both from models and from observational studies. Since there are multiple aerosol species, with varying scattering and absorption strength, even the net direct aerosol RF over the industrial era (here taken as the years 1750–2010) is quite uncertain.
Hatched boxes show the result from a recent multi-model assessment, while the solid boxes combine these results with independent observational constraints and expert assessment.
Sulfate can be seen to be the dominant scattering component, having had a cooling effect on the climate over the industrial era. BC is found to be the major absorbing component, with a positive RF. Primary organic aerosol (POA) and nitrates are also scattering, with RFs likely to be negative. For biomass burning aerosol, secondary organic aerosol (SOA), and mineral dust due to industrial activities, estimates vary even in sign.
The AR5 assessed the net RF of the direct radiative effect from aerosols to be –0.27 Wm–2, with a 9–95% confidence range of [–0.77,0.23] Wm–2; that is, it is highly likely to have been negative, but with a small possibility for a positive value.
Aerosol indirect effects are harder to assess. Fewer observational constraints exist, meaning we must rely on model information. Unfortunately, cloud microphysics is among the hardest processes to treat in global climate models, as computations quickly become prohibitively expensive. Hence model results on the indirect effect are limited in precision both by their internal representation of clouds, and by their parameterizations of the various aerosol indirect interactions.
Sulfate is reckoned as the major contributor to the cloud albedo and lifetime effects, due to its high number concentration. Black carbon, as the only major absorbing aerosol, is counted as the major species for the semidirect effect.
Figure 5 shows the IPCC AR5 summary of radiative forcing over the industrial era. The estimate of –0.27 Wm−2 from the direct effect can be seen in the column labeled “Aerosols and precursor.” Below this, the AR5 best estimate for the combined indirect effects, or cloud modifications due to aerosols, is listed as –0.55 (–1.33, –0.02) Wm−2.
As Figure 5 shows, the IPCC AR5 estimate for industrial era RF from CO2 is 1.68 Wm–2, with a 90% uncertainty range of 1.33–2.03 Wm−2.
While the total aerosol contribution is assessed to be much smaller in magnitude, its uncertainty range is relatively larger. Hence, while anthropogenic emissions of greenhouse gases are the main drivers of the present climate energy imbalance, one main component in the remaining uncertainty comes from aerosol processes. Specifically, the indirect aerosol effect is, at the time of writing, the climate interaction that has the largest uncertainty relative to its magnitude. Significant effort is being put into reducing this uncertainty, but the due to lack of computational power, good observations, and even present understanding of the physical mechanisms involved, it is hard to overcome.
Historical Evolution of Emissions, and Global Dimming
These estimates for aerosol RF are taken over the whole industrial era, that is, the conditions in 2010 relative to those estimated to have prevailed in 1750. Between these times, aerosol emissions and concentrations have naturally had a nonlinear, spatially heterogeneous evolution. This evolution broadly follows the progress of industrialization.
In Europe and North America, emissions of black carbon from closed combustion sources is estimated to have peaked around 1920, and then to have rapidly declined as combustion engines improved and air quality concerns came to the fore. In the rest of the world, black carbon emissions started to increase gradually around 1950, and are estimated to have peaked around 2010. Sulfate emissions have followed similar trends, with European and U.S. emissions peaking in the early 20th century, then declining with various U.S. and European clean air laws and directives. Southeast Asia, on the other hand, has seen a strong increase in sulfate loading from SO2 emissions since the 1960s.
This shift from European and American sources to East Asian sources, combined with the relatively short atmospheric lifetime of aerosols, implies a shift in local radiative forcing. This shift has been implicated in the observed behavior of regional surface temperature, which shows a peak around 1950 in Europe and the United States, but not in Asia (Stocker et al., 2013). The simplified picture is that aerosol scattering of incoming sunlight, from sulfate in particular, held global warming in check over parts of the world for some decades. The phenomenon itself, known as global dimming, is well established (Wild et al., 2009). Its connection to the temperature evolution is, however, more tenuous, as volcanoes and natural variability of the sun and oceans may also have played significant parts.
Aerosols and Climate Sensitivity
The perhaps most hotly burning question in all of climate science is the sensitivity of the global temperature per unit of radiative forcing—or the global climate sensitivity. While both observations and model estimates have improved greatly since the 1980s, the range of estimates of the climate sensitivity has stubbornly refused to converge. In fact, the range from the IPCC AR5 is virtually identical to that from the 1st Assessment Report, which came out in 1990.
One challenge we face when trying to determine the climate sensitivity is that there have been several processes at work at the same time, altering the global energy balance in different ways. While the well-mixed greenhouse gases have added energy to the system, aerosols, as we have seen, have most likely acted to cool the Earth. The situation is much like a tug-of-war, with greenhouses gases and aerosols pulling in opposite directions. What we observe is the resulting change in global mean temperature, and from this we must try to infer the two opposing forces.
For climate sensitivity, it has been shown that the uncertainty interval from aerosol radiative forcing matters strongly. While an overestimation of aerosol radiative forcing would not greatly affect estimates of the climate sensitivity, an underestimation could mean that the sensitivity is quite high (Myhre, Myhre, Samset, & Storelvmo, 2013a). While recent research indicates that climate sensitivities above 4K for a doubling of CO2 are highly unlikely, the remaining aerosol uncertainty is one reason why this troubling possibility can still not be completely ruled out.
Aerosols and Precipitation
Overall, increased aerosol concentrations are thought to reduce average precipitation.
As shown in the section Aerosol–Climate Interactions, aerosols can perturb both cloud microphysics and atmospheric stability, both of which may lead to changes in precipitation. In addition, since they can act to cool or heat the climate, they also contribute to the changes in precipitation expected as part of global warming. Further, it has been suggested that the recent shift in aerosol loading from Europe and the United States, over to Southeast Asia, could be linked to changes in the tropical rain belt and to the Pacific Decadal Oscillation (Allen, Norris, & Kovilakam, 2014).
Basic physics suggests that a change in aerosol concentration will affect precipitation on two time scales: one set of fast effects, connected to the rapid response of clouds and atmospheric stability, and another set connected to the long-term change in surface temperature—and hence evaporation. Recent modeling studies indicate that an isolated increase in black carbon would, at first, cause a decrease in precipitation due to reduced convection (Samset & Myhre, 2015). Over time, precipitation will again increase, as the BC absorption acts to heat the surface. Overall, however, the precipitation change due to BC is estimated to be negative. For sulfate, on the other hand, there is little initial precipitation change, but a strong decrease over time as the aerosol cools the surface.
These general relations, however, apply when looking at the planet as a whole. As we have seen, however, aerosols and their perturbations to clouds and atmospheric absorption have strong regional variations. The impact of a given aerosol distribution on precipitation will further depend on local humidity and convection, surface albedo, and other conditions. Extreme precipitation, which is an area where the impacts of a changing climate are quickly felt by society, may also follow other trends than global mean precipitation. The precise links between observed and future changes in aerosol concentrations and precipitation trends are presently under very active investigation.
Aerosols as Climate Mitigation
As aerosol emissions have significant climate impacts, and their atmospheric lifetimes are short, both scientists and policy makers have argued for emission reductions as climate mitigation strategies. It is commonly argued that, due to the link between aerosols, air quality, and respiratory disease, such reductions would be win-win scenarios.
Recently, however, several studies have noted that the climate impact that can be achieved through aerosol emission mitigation is relatively modest, compared to the long-term effects of reduced greenhouse gas emissions. Aerosols are also rarely emitted in isolation, but are co-emitted from the same sources. For example, a coal fire will emit both black carbon and SO2, meaning that a reduction of such fires would remove both a source of atmospheric heating and cooling.
There is no doubt that reductions in aerosol emissions would give health benefits. Their climate mitigation potential is, however, presently a hot and controversial topic among scientists and policy makers.
Expected Trends in Anthropogenic Aerosol Emissions
Anthropogenic aerosol emissions have changed drastically over the 20th century in magnitude, composition, and geographical pattern. While it is hard to predict future trends, a common expectation is that in the West, reductions in emissions will continue as technology improves, the use of combustion engines in road vehicles decreases, and the composition of ship fuel changes. In the East, where the majority of anthropogenic emissions can presently be found, concerns about air quality and rapid technological development are both already thought to be reducing emissions. Hence, it has been stated by experts that “the age of aerosols may soon be over” (Smith & Bond, 2014).
This view is reflected in the emission pathways considered for the IPCC 5th Assessment Report. All pathways assume drastic reductions relative to year 2000 values, first in Europe and the United States, then, after 2020, also in Asia. While the details differ, all pathways project a reduction by half of anthropogenic aerosol emissions by 2100. Globally, the climate impacts of such a reduction are expected to be minor, due to the increasing dominance of radiative forcing from long-lived greenhouse gases. Locally, however, the reductions in aerosol burden may have significant impacts.
Large-Scale Climate Impacts From Aerosols
The section “Expected Trends in Anthropogenic Aerosol Emissions” discussed the climate impact of anthropogenic activities, that is, emissions from industry, road transport, shipping, and large-scale agricultural burning and land use change. There are, however, several other areas—both manmade and natural—where aerosols may have even larger impacts. This section discusses three of these: volcanic eruptions, nuclear winter, and climate engineering by sulfate emissions.
Volcanic eruptions are among the most dramatic of natural events, both in visual splendor and in their consequences for society. The introduction to this review showed how the 1991 eruption of Mt. Pinatubo noticeably lowered global temperatures for several years. Larger eruptions, such as Mt. Tambora in 1815 and Krakatoa in 1883, are well documented in history books, scientific records, and even art. The global dimming and cooling following the Mt. Tambora eruption led to 1816 being labeled “the year without summer.”
The large-scale climate impact of volcanoes comes from aerosols. When large volcanoes erupt, they can emit copious amounts of ash and SO2. If the eruption is of sufficient explosive strength, this dust may reach tens of kilometers into the atmosphere and end up in the stratosphere. There, well above the precipitation processes that normally reduce aerosol concentrations, the fine-grained sulfuric acid compounds formed from the emitted SO2 can remain suspended for a number of years. Due to aerosol transport with prevailing winds, the precise impacts of a given eruption will depend both on its geographical location and the time of year, in addition to the explosive power and volume of mass ejected.
Historically, the radiative forcing exerted by volcanic aerosols is the strongest short-term perturbation to the climate system. In the forcing time series included in the IPCC AR5, volcanic eruptions can be seen as spikes with global mean values in excess of –4 Wm–2. The report further observes that volcanic eruptions are random in nature, meaning that their size and location cannot easily be predicted. Hence, they constitute a major source of uncertainty in future climate predictions. Should a major volcanic eruption occur in the near future, it may well offset global warming due to long-lived greenhouse gases for a short duration, or even cause the opposite situation: global cooling and precipitation changes sufficient to impact food production for a number of years.
Presently, the duration of the climate perturbation after a major eruption is being discussed in scientific literature. An example is the study of the so-called “Little Ice Age,” a period around the year 1500 where many temperature reconstructions show a colder climate, most notably in Europe and North America. While several factors are thought to have been important for this cooler period, including the Maunder grand solar minimum and changes to ocean circulation, several authors have linked the effect to periods of heightened volcanic activity. This suggests that in some cases, volcanic aerosols may have climate impacts that last beyond the 2–3 years estimated from the most recent major eruptions.
While an armed conflict involving the use of nuclear weapons would be horrific in itself, scientists in the 1980s started warning about an additional hazard. Were a large number of nuclear explosions to take place over major cities, an additional result would be large, widespread, and long-lasting fires. The smoke from these fires, that is, a layer of aerosols, would spread across the globe and cause significant global dimming. The resulting hypothetical cooling was termed nuclear winter, and became part of the public debate on the nuclear arms race.
At the time, numerical climate models had many shortcomings. In recent years, the topic has however been thoroughly investigated using state-of-the-art modeling tools and updated estimates for emissions from major city fires.
The conclusions remain the same: a major, international nuclear war would have, as yet another consequence, a global cooling of up to tens of degrees Celsius, lasting for several years. Such an event would have devastating impacts on global agriculture, vegetation, and wildlife, and hence for society. Recent work also suggests that even a limited, regional conflict between two neighboring nuclear powers could have subsequent climate effects exceeding those of even very strong volcanic eruptions (Robock & Toon, 2012). Such consequences would be global, even if the conflict that sparked them was not.
Climate Engineering via Aerosols
As the consequences of the ongoing climate changes start being felt, and the difficulties of achieving global agreements on emission mitigation or widespread adaptation become increasingly apparent, many have suggested that there may be a third option: could we, through technological means, engineer the climate back to the state we have been used to? (Shepherd, 2012)
One of the first suggestions to be made for such a large-scale climate engineering scheme is to deliberately emit sulfates or other scattering aerosols. The idea is that their dimming of incoming sunlight can balance the extra energy stored due to the enhanced greenhouse effect. In other words, we would mimic the climate impact of volcanoes, using aerosols emitted from balloons or airplanes in the stratosphere.
Another suggestion involving aerosols has been to increase the flux of sea spray droplets in regions where marine stratocumulus clouds form. Here, we would utilize the aerosol indirect effect to make the clouds whiter, again reflecting out more sunlight. A third aerosol-based option is to seed high-altitude cirrus clouds, increasing their evaporation rate. This would reduce their heating of the surface due to re-emission of longwave radiation.
While such options may sound attractive in principle, they are all still very far from being ready for deployment. Climate engineering in general is a contentious issue, both for practical and for ethical reasons. Research into the topic is presently conducted primarily with climate models, and even here, under idealized conditions, results show that the climate impacts of large-scale climate engineering via aerosols would be highly unpredictable. At present, the technology to efficiently disperse sulfate or sea spray aerosol at the right altitudes is also not available.
Experts on climate engineering presently hold a wide range of views on whether or not such a strategy could ever be implemented. The main bulk of research, however, indicates that while the global energy budget could indeed be brought back into balance through increased aerosol scattering, an exact cancellation of all the other impacts of a heightened greenhouse effect on the distribution of energy in the atmosphere/ocean system is not possible. Hence, we would see changes to precipitation and large-scale atmospheric circulation in any case. Further issues include the question of what would happen if the climate engineering measures were ever to be terminated, and the inability of aerosol climate engineering to counter ocean acidification.
The Scientific Study of Aerosols
While the notion that dust and other minute particles may influence the climate goes back several centuries, the study of aerosol–climate interactions only became a fully-fledged scientific subfield after 1990. Due to the multiple possible interactions, the large number of aerosol types, and their short atmospheric lifetimes, it was only with the advent of supercomputers that scientists were able to make firm statements about the net role that aerosols in general—and anthropogenic aerosols in particular—have had on the climate over the industrial era. Presently, aerosol research is one of the most active parts of climate science, with regular publications appearing in the most prestigious academic journals.
Early Ideas Linking Aerosols to Climate Evolution
When Krakatoa exploded in the East Indies in 1883, ash and dust reduced incoming sunlight around the world for months. Scientists at the time had no way to measure changes to global temperature, but it was obvious that such major emissions did indeed cause at least regional cooling. As one of nature’s most dramatic spectacles, volcanoes were in the early 1900s hypothesized to be the causes both of the ice ages and of more recent deviations found in early climate records. In the 1950s, when the first reports of a gradually rising global temperature started showing up, many initially attributed this to the long preceding period without major known eruptions.
At the same time, air pollution and urban smog were becoming major topics of concern. Measurements from ships far out at sea showed that industrial byproducts were being transported by air to the furthest reaches of the globe. Few, however, thought that these observations could have any implications for weather and climate.
The link between local pollution, global climate, and the basic physics and chemistry of aerosols was only made in the 1960s, when the study of atmospheric aerosols gradually became a full-fledged scientific field. Early pioneers such as Robert McCormic, John Ludwig, Reid Bryson, Murray Mitchell, and Hubert Lamb debated, in increasingly technical terms, whether aerosols had so far had a climate impact. At the same time, they urged the relevant communities to form stronger ties to make a more focused attempt at understanding the full role of aerosols.
In a pioneering study in 1971, Ichitaque Rasool and Stephen Schneider showed that it had become possible to perform numerical studies of aerosols that were detailed enough to give realistic answers. They discussed the distinction between aerosols that scatter sunlight and thus could cool the atmosphere, and absorbing aerosols that could lead to warming. Their early, admittedly primitive results indicated that, dependent on composition and amount, aerosols could on relatively short timescales cause climate perturbations of several degrees Celsius.
The 1970s saw a debate among scientists as to whether aerosols were then primarily cooling or heating, and whether we should worry about a possible global cooling catastrophe from industrial pollution. At the same time, numerical models were being continually improved and new links between climate and aerosols were being described scientifically. For example, the first indirect aerosol effect was described by Sean Twomey in 1977.
Through the 1980s, the major players in the problem of anthropogenic aerosols emerged: soot, sulfate, and their interactions with radiation and clouds. In 1982, Mother Nature gave a helping hand through the eruption of El Chichon in Mexico. Measurements of the amount of sulfur emitted to the atmosphere and the resulting regional cooling could be compared to rudimentary climate models, allowing for tuning of parameters that made the calculations far more realistic.
Aerosol Science in the Age of Supercomputers
In spite of two decades of advances, the first IPCC report in 1990 noted that, regarding anthropogenic aerosol emissions, “at this stage, neither the sign nor magnitude of the proposed climatic feedback can be quantitatively estimated.”
The real turning point came with the event described in the introduction to this review: the 1991 massive eruption of Mt. Pinatubo. Following the event, the group of James Hansen at NASA quickly published a model based prediction of the ensuing global cooling—a prediction that was, broadly, borne out by observations over the subsequent years. Their results, among others, led to the realization that previous calculations of past and future temperature change due to greenhouse gas emissions may have been significantly off since they did not include the cooling effect of sulfates.
A revised calculation from the U.K. Hadley Center, published in 1995, showed that including sulfates indeed gave a significant improvement in the predictive skill of their numerical model. It became clear that sulfate emissions from Western industry, from 1940 onwards, had temporarily held global warming in check. This insight was one key aspect that allowed the IPCC in 1995 to conclude that anthropogenic climate change was by then discernible.
Recent Developments and Outstanding Issues
Since the mid-1990s aerosol science has attracted ever more attention. The aerosol representation in the complex earth system models used to predict future climate is becoming ever more complex and realistic. Observations of aerosol distribution, transport, and loading are available from the surface, from satellites and from instruments mounted on aircraft.
As shown in the section Aerosols in the Context of Anthropogenic Climate Change, however, the scientific uncertainty on the net climate impact of aerosol is still relatively large. In the 2013 Working Group I part of the IPCC 5th Assessment Report, a whole chapter was devoted to recent developments related to aerosols and their climate impacts, together with the influence of clouds on climate and of aerosols on clouds. This combination of topics presently feels natural, as it has become clear that the total aerosol climate impact cannot be estimated without a thorough knowledge of cloud microphysics, and of the dynamic and thermodynamic behavior of clouds in general.
A wide range of topics are presently being discussed in the aerosol climate field. While it is believed that most major interactions between particulate matter and climate evolution are identified, the validation of computer modeling of these effects by observational data is a key issue. Issues remain both with the numerical representations and with the remote sensing and in situ observations available. A major setback to the field came with the loss of the GLORY satellite on launch in 2011. GLORY was, among other things, to have observed in detail the spatial and temporal distributions and physical and optical properties of aerosols with unprecedented detail. It is presently unclear if and when a replacement satellite will be launched.
Burning scientific aerosol–climate issues at the time of writing include the following:
What are the global annual aerosol emissions from natural and anthropogenic processes?
How far and how high are aerosols transported after emission?
How significant is the warming effect of black carbon, and can mitigation of black carbon emissions be used as an effective short-term strategy to counter global warming?
How can we sufficiently model the microphysical interactions of aerosols with cloud droplets in global earth system models?
Do the cyclic variations in the flux of galactic cosmic rays, as modulated by the change in the solar magnetic field over a sunspot cycle, have a noticeable impact on aerosol formation and hence on clouds?
To what degree have scattering aerosols masked global warming from long-lived greenhouse gases over the 20th century?
What, if any, is the link between aerosol emissions and extreme precipitation?
Presently, atmospheric aerosols are part of a wide range of climate discussions. As an example of this, the World Climate Research Programme (WCRP, a research promoting organization sponsored by the World Meteorological Organization [WMO], the International Council for Science [ICSU], and the Intergovernmental Oceanographic Commission [IOC] of UNESCO), recently defined a set of five “grand challenges” that the international climate research community faces. Of these five, improved knowledge about aerosols is crucial for four of them: “Clouds, circulation and climate sensitivity,” “Melting ice and global consequences,” “Climate extremes,” and “Water availability.”
While our understanding of aerosols, their basic physics and chemistry, formation, transport, and various climate interactions has improved tremendously there is still significant work to be done. Aerosols will remain a topic at the forefront of climate research for the foreseeable future.
- Boucher, O. (2015). Atmospheric aerosols: Properties and climate impacts. Dordrecht, The Netherlands: Springer Verlag.
- Fleming, J. S. (2010). Fixing the sky: The checkered history of weather and climate control. New York: Columbia University Press, 2010.
- Weart, S. (2008). The discovery of global warming (rev. ed.). Cambridge, MA: Harvard University Press.
- Albrecht, B. A. (1989). Aerosols, cloud microphysics, and fractional cloudiness. Science, 245(4923), 1227–1230.
- Allen, R. J., Norris, J. R., & Kovilakam, M. (2014). Influence of anthropogenic aerosols and the Pacific Decadal Oscillation on tropical belt width. Nature Geoscience, 7(4), 270–274.
- Bond, T. C., Doherty, S. J., Fahey, D. W., Forster, P. M., Berntsen, T., . . . Zender, C. S. (2013). Bounding the role of black carbon in the climate system: A scientific assessment. Journal of Geophysical Research: Atmospheres, 118(11), 5380–5552.
- Boucher, O., Randall, D., Artaxo, P., Bretherton, C., Feingold, G., Forster, P., . . . Wyant, M. (2013). Clouds and aerosols. In Stocker, T. F., Qin, D., Plattner, G.‑K., Tignor, M., Allen, S. K., Boschung, J., . . . Midgley, P. M. (Eds.), Climate change 2013: The physical science basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, U.K.: Cambridge University Press.
- Hansen, J., Lacis, A. Ruedy, R., & Sato, M. (1992). Potential climate impact of Mount Pinatubo eruption. Geophysical Research Letters, 19(2), 215–218.
- Isaksen, I. S. A., Granier, C., Myhre, G., Berntsen, T. K., Dalsøren, S. B., Gauss, M., . . . Wuebbles, D. (2009). Atmospheric composition change: Climate-chemistry interactions. Atmospheric Environment, 43(33), 5138–5192.
- Levy, R. (2015). MODIS Atmosphere L2 Aerosol Product. Edited by G. S. F. C. NASA MODIS Adaptive Processing System, USA.
- Liu, S., Aiken, A. C., Gorkowski, K., Dubey, M. K., Cappa, C. D., Williams, L. R., . . . Prévôt, A. S. H. (2015). Enhanced light absorption by mixed source black and brown carbon particles in UK winter. Nature Communications, 6.
- Mann, G. W., Carslaw, K. S., Reddington, C. L., Pringle, K. J., Schulz, M. Asmi, A., . . . Henzing, J. S. (2014). Intercomparison and evaluation of global aerosol microphysical properties among AeroCom models of a range of complexity. Atmospheric Chemistry and Physics, 14(9), 4679–4713.
- Myhre, G., Myhre, C., Samset, B. H., & Storelvmo, T. (2013a). Aerosols and their relation to global climate and climate sensitivity. Nature Knowledge Project, 4(5).
- Myhre, G., Shindell, D., Bréon, F.-M., Collins, W., Fuglestvedt, J., Huang, J., . . . Vernier, J.‑P. (2013b). Anthropogenic and natural radiative forcing. In T. F. Stocker, D. Qin, G.‑K. Plattner, M. Tignor, S. K. Allen, J. Boschung, . . . P. M. Midgley (Eds.), Climate change 2013: The physical science basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (pp. 659–740). Cambridge, U.K.: Cambridge University Press.
- Pithan, F., & Mauritsen, T. (2014). Arctic amplification dominated by temperature feedbacks in contemporary climate models. Nature Geoscience, 7(3), 181–184.
- Robock, A., & Toon, O. B. (2012). Self-assured destruction: The climate impacts of nuclear war. Bulletin of the Atomic Scientists, 68(5), 66–74.
- Samset, B. H., & Myhre, G. (2015). Climate response to externally mixed black carbon as a function of altitude. Journal of Geophysical Research: Atmospheres, 120(7), 2913–2927.
- Shepherd, J. G. (2012). Geoengineering the climate: an overview and update. Philosophical Transactions of the Royal Society A, 370(1974), 4166–4175.
- Smith, S. J., & Bond, T. C. (2014). Two hundred fifty years of aerosols and climate: The end of the age of aerosols. Atmospheric Chemistry and Physics, 14(2), 537–549.
- Stocker, T. F., Qin, D., Plattner, G.-K., Tignor, M., Allen, S. K., Boschung, J., . . . Midgley, P. M. (Eds.) (2013). Climate change 2013: The physical science basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, U.K.: Cambridge University Press.
- Twomey, S. (1977). The influence of pollution on the shortwave albedo of clouds. Journal of the Atmospheric Sciences, 34(7), 1149–1152.
- Wild, M., Trüssel, B., Ohmura, A., Long, C. N., König-Langlo, G., Dutton, E. G., & Tsvetkov, A. (2009). Global dimming and brightening: An update beyond 2000. Journal of Geophysical Research: Atmospheres, 114(D10).