Offsetting Climate Change by Engineering Air Pollution to Brighten Clouds

Research shows that the judicious targeting of clouds and selection of the size and composition of particle emissions can produce substantial cooling effects.

Natural, industrial, and residential combustion produces both aerosols that cool the Earth and CO2 that warms it, and the amount of combustion worldwide has increased substantially since the invention of the steam engine as well as with the increase in populations relying on wood and char burning. Natural and early man-made combustion processes emitted aerosols and CO2 roughly proportionally, although the ratios of emission types were dependent on burning conditions. In the wake of smog-induced respiratory-health-related deaths in London in 1952, and ensuing legislation in favor of limiting emissions in the United States and Europe, “air quality engineering” was developed to reduce combustion-related aerosol emissions. But the reductions—without corresponding reductions in CO2 emissions—led to more warming (with some offset for reductions in absorbing aerosol emissions). One approach to “climate engineering” is to undo these reductions in aerosol emissions in a way that avoids the health and visibility impacts of pollution but still allows for particles to cool the Earth both by reflecting sunlight directly and by brightening clouds (which magnify the scattering of light with water). The engineering challenge with this approach is that clouds are the least understood component of the climate system, and current models cannot reliably predict their formation and properties. Recent research in the Eastern Pacific Emitted Aerosol Cloud Experiment (E-PEACE) 2011 illustrates that judicious selection of the meteorological regime and the size and composition of particle emissions can achieve substantial cooling effects. However, socioeconomic questions about climate engineering remain—such as the possibility that, if implemented, sudden cessation of enhanced particle emissions could exacerbate the climate effects on ecosystems and might interfere with oceanic and terrestrial ecosystem processes—thus requiring cautious and comprehensive research.


The fundamental physics that control the global mean surface temperature are well understood: about one-third of the incoming solar radiation is reflected back to space by the Earth’s albedo and the remaining two-thirds is absorbed at the surface, then emitted as longwave energy. In this way the incoming and outgoing energy at the top of the atmosphere largely balances the energy leaving, after partially trapping some of the energy by the greenhouse effect of atmos-pheric water vapor and clouds as well as greenhouse gases (IPCC 2007). These interactions constitute the Earth’s radiative energy balance, as illustrated in Figure 1. Higher surface mean temperatures (Tsurf) are due to the greenhouse effect, caused by the man-made release of CO2 and other greenhouse gases. Increasing albedo (a) can offset CO2-enhanced greenhouse warming by increasing the shortwave reflection of clouds. And clouds can be brightened to increase their reflectance by adding aerosol particles, which increase the number and decrease the mean size of cloud droplets. An example of such brightening is provided by the “ship tracks” created by the emissions of cargo ships crossing the Pacific Ocean, as shown in Figure 2.

FIgure 1

  Figure 2

Keeping in mind that maintaining global mean surface temperature does not imply that regional temperatures or precipitation patterns are kept constant, engineering the global mean surface temperature to reduce changes from present-day conditions could be sufficient to alleviate some of the most severe effects of global warming. Adding aerosol is straightforward, since particle production is a side effect of most combustion processes as well as a result of vaporization of liquids in condensable conditions. The real challenge in engineering aerosol particles to offset climate change by brightening clouds is predicting how the Earth system, and in particular its clouds, will affect the albedo response to increased particles.

Recent Model Simulations of Cloud Brightening

Model simulations have established the climate impacts of distributing enough particles to modify enough clouds to offset sufficient global warming to delay or lessen some of the effects expected in the Earth’s changing climate (Latham 1990, 2002; Latham et al. 2008). Some schemes focus on a perceived need for engineering and development of new technology, such as Flettner rotors and high-efficiency seawater atomization (Salter et al. 2008). Other studies use detailed global modeling investigations to show what fraction of clouds are brightened, with more aggressive increases in brightening resulting in exacerbation of climate in some regions even as others are improved (Rasch et al. 2009). Global simulations have also shown that where clouds are targeted is important, as some choices result in exacerbation of drought conditions in some regions (Korhonen et al. 2010; Rasch et al. 2009). In addition, recent studies have investigated the complexities of aerosol cloud interactions, including the damping of cloud brightening by reductions in cloud supersaturation (Korhonen et al. 2010) and by overlapping plumes1 of particles (Wang et al. 2011).

However, aerosol-cloud-radiation interactions are widely held to be the largest single source of uncertainty in projections of climate change due to increasing anthropogenic emissions. The underlying causes of this uncertainty in modeled predictions are the gaps in fundamental understanding of cloud processes (IPCC 2007). Although there has been significant progress with both observations and models, and the qualitative aspects of the indirect effects of aerosols on clouds are well known, the quantitative representation of these processes is nontrivial and limits the ability to represent them in global climate models. Current global models lack (1) accurate aerosol particle activation, with associated implications for the profiles of supersaturation, vertical velocity, liquid water content, and drop distribution; (2) realistic microphysical growth and precipitation processes that control the formation and impacts of drizzle on cloud structure, lifetime, and particle concentration; and (3) eddy-based transport processes that control the effects of entrainment on cloud thickness and lifetime as well as the dispersion of aerosol plumes. These basic scientific issues have not been addressed by climate models or by climate engineering proposals that involve perturbing marine stratocumulus; the following section describes work by our multi-institution collaboration to address them.

New Experimental Evidence of Cloud Brightening

To learn more about the cloud physical processes that affect aerosol-cloud-radiation interactions, we designed the Eastern Pacific Emitted Aerosol Cloud Experiment (E-PEACE) 2011 as a targeted aircraft campaign with embedded modeling studies, using the Center for Interdisciplinary Remotely Piloted Aircraft Studies (CIRPAS) Twin Otter aircraft and the R/V Point Sur in July 2011 off the coast of Monterey, California, with a full payload of instruments to measure particle and cloud number, mass, composition, and water uptake distributions (Russell et al. 2013; Shingler et al. 2012). Three central aspects of the collaborative E-PEACE design are described below, followed by highlights of the findings.

Figure 3

  1. Controlled particle sources were used to separate particle-induced feedback effects from natural variability. We have investigated three types of sources of different particle sizes and compositions to characterize specific aspects of aerosol-cloud interactions: (1) ship-emitted particles at rates of 1016 to 1018 s−1 with dry diameters between 50 and 100 nm (Coggon et al. 2012), (2) shipboard smoke generator particles at rates of 1011 to 1013 s−1 with dry diameters between 50 nm and 1 µm, and (3) aircraft-based milled, coated salt particles at rates of 109 s−1 with dry diameters between 3 and 5 µm. The shipboard smoke generators are shown in Figure 3.
  2. Satellite observations showed that not all ship tracks cause cloud brightening (Chen et al. 2012), indicating a variety of cloud feedback responses to increased particle concentrations. These observations were compared to the features predicted by large eddy simulations and aerosol-cloud parcel modeling of the impacts of turbulence, precipitation, and other cloud processes on the number concentration and size distribution of cloud drops (Lu and Seinfeld 2005, 2006; Russell et al. 1999).
  3. The track from the controlled emission of smoke-generated particles demonstrated efficient cooling of clouds at very low warming cost, using existing technology and minimal resources. We noted that cooling outweighed warming by a factor of 50 on the day that a track was observed (Russell et al. 2013). This cooling effect exceeds that of commercial shipping, for which track-making ships induce twice as much cooling as warming (on days when tracks formed).

One of the most interesting results of E-PEACE was the activation into cloud droplets of smoke particles composed almost entirely of organic constituents (Shingler et al. 2012). This result was surprising because many organic components are hydrophobic and do not serve as effective cloud nuclei at supersaturations below 0.2%. The large diameter of smoke particles makes it possible for them to activate with fewer soluble constituents.

A second finding is the formation of tracks from the smoke particles in cloud-covered marine boundary layers. The organic smoke particles not only activated to cloud droplets but also did so in sufficient numbers to form a track with a detectable increase in brightness. However, there was a range of brightening observed for the many different tracks formed by particle emissions from fairly similar cargo ships (Chen et al. 2012), indicating that cloud feedback processes play an important role in determining cloud brightening.

The third important finding of E-PEACE was the frequency of low clouds with multiple layers, which reduce the impact of particles on clouds. Tracks did not form every day because of either the absence or structure of clouds near the ship. To produce a track with a significant albedo effect, the cloud layer needed to be uniform and single-layered. In addition, for rapid mixing of particles into the cloud layer, the layer needed to be below approximately 500 m. During the 12 days of the E-PEACE cruise, multiple cloud layers of 100 to 1000 m were present on more than half. Since particles emitted by ships on the ocean surface are usually transported only to the lowest cloud layer, their modification of droplet distributions does not appreciably change the albedo seen from above the top cloud layer. In such cases, particles have little effect on the radiation balance. The presence of low cloud layers overlying the layers affected by the smoke particles resulted in a low frequency of track formation. This finding is significant because it shows the need for representing small-scale cloud structure in global climate models in order to improve predictions of aerosol-induced cloud albedo changes.

Implications for Climate Engineering

The E-PEACE results provide a proof of concept that cloud brightening to reduce global mean warming is possible, with existing, decades-old technology, for some cloud conditions (but it will not reduce drought or ocean acidification). Track formation requires sufficient particle production to increase droplet number by 100 to 300 cm−3 over well-mixed boundary layers 100 m to 600 m high and spanning track widths of several kilometers. Cargo ships and portable smoke generators can both easily emit 1016 to 1018 s−1, which is sufficient at wind speeds of up to 10 m s−1 to make tracks in unpolluted marine air. The advantage of smoke generators for climate engineering is that the lower fuel consumption by the much smaller ship has a substantially lower CO2 cost, making cooling more efficient.

However, the radiative effects are not the only ones to be considered before deploying on a large scale (Russell et al. 2012). In particular, careful research is needed to assess the impacts of particle deposition on ocean and downwind terrestrial ecosystems; sustained changes in particle deposition could have deleterious impacts on ocean and land biota. Furthermore, shifts in precipitation patterns and direct radiation at the surface, if substantial, could affect crop production. And implementation of cloud brightening in regions near susceptible human populations could affect health.


Although the technology for particle emission and distribution exists, the engineering required for cloud brightening is hardly trivial. The most critical challenges to engineering the design of large-scale cloud brightening are (1) cloud feedback processes that affect the cloud response to aerosol enhancements and reduce the expected brightening, (2) multilayered clouds that mask changes in underlying clouds, and (3) ecosystem impacts of particle deposition (Russell et al. 2012). These issues require region-specific observations and small-scale, short-duration testing to determine realistic constraints for modeling. In addition, although particle production is feasible with existing technology, there are ample opportunities for optimizing the efficiency of particle emission processes and for minimizing their ecosystem impacts.

Knowledge of aerosol-cloud interactions remains sufficiently uncertain that consideration of their use for climate engineering is premature. Substantial advances are needed in understanding of aerosol and cloud physics to quantify their role in climate change, and such advances require experimental as well as modeling studies. If such studies demonstrate the effectiveness of particles for cloud brightening, it may be possible to use this method to offset some of the warming to the global mean surface temperature caused by greenhouse gases.

Going Forward

The seriousness of the consequences of global warming merits research into the possibility of using cloud brightening for climate engineering. However, while cloud brightening will target atmospheric emissions outside of national boundaries (since offshore marine stratocumulus have some of the largest impact on albedo) in areas that largely lack environmental regulations, any large-scale implementation should involve multinational agreement and cooperation, as well as compensation for unexpected and harmful consequences. Furthermore, as with any solar reflection method that does not also reduce greenhouse gases, once initiated the cessation of cooling would likely cause accelerated warming as the system returns to the nonmasked warming (Russell et al. 2012).

In summary, while cloud brightening could be appropriate to prevent tipping points (such as massive sea ice loss, which some predict may occur as early as 20152), implementation of cloud brightening to offset climate warming should be considered as an option only after sufficient research is devoted to better constraining aerosol-cloud-radiation interactions.


E-PEACE was supported by the National Science Foundation under grant AGS-1013423. Thanks to Kurt Nielsen of the Monterey Naval Postgraduate School for compiling the satellite image in Figure 2.


Chen Y-C, Christensen MW, Xue L, Sorooshian A, Stephens GL, Rasmussen RM, Seinfeld JH. 2012. Occurrence of lower cloud albedo in ship tracks. Atmospheric Chemistry and Physics 12:8223–8235.

Coggon MM, Sorooshian A, Wang Z, Metcalf AR, Frossard AA, Lin JJ, Craven JS, Nenes A, Jonsson HH, Russell LM, Flagan RC, Seinfeld JH. 2012. Ship impacts on the marine atmosphere: Insights into the contribution of shipping emissions to the properties of marine aerosol and clouds. Atmospheric Chemistry and Physics 12:8439–8458.

IPCC [Intergovernmental Panel on Climate Change]. 2007. Climate change 2007: The physical science basis. Contribution of Working Group I to the fourth assessment report. In IPCC (Intergovernmental Panel on Climate Change), ed. Solomon S, Qin D, Manning M, Chen Z, Marquis M, Avery KB, Tignor M, Miller HL. Cambridge UK: Cambridge University Press.

Korhonen H, Carslaw KS, Romakkaniemi S. 2010. Enhancement of marine cloud albedo via controlled sea spray injections: A global model study of the influence of emission rates, microphysics and transport. Atmospheric Chemistry and Physics 10:4133–4143.

Latham J. 1990. Control of global warming. Nature 347: 339–340.

Latham J. 2002. Amelioration of global warming by controlled enhancement of the albedo and longevity of low-level maritime clouds. Atmospheric Science Letters 3:52–58.

Latham J, Rasch P, Chen CC, Kettles L, Gadian A, Gettelman A, Morrison H, Bower K, Choularton T. 2008. Global temperature stabilization via controlled albedo enhancement of low-level maritime clouds. Philosophical Transactions of the Royal Society A–Mathematical Physical and Engineering Sciences 366:3969–3987.

Lu ML, Seinfeld JH. 2005. Study of the aerosol indirect effect by large-eddy simulation of marine stratocumulus. Journal of the Atmospheric Sciences 62(11):3909.

Lu ML, Seinfeld JH. 2006. Effect of aerosol number concentration on cloud droplet dispersion: A large-eddy simulation study and implications for aerosol indirect forcing. Journal of Geophysical Research Atmospheres 111:D02207.

Rasch PJ, Latham J, Chen CC. 2009. Geoengineering by cloud seeding: Influence on sea ice and climate system. Environmental Research Letters 4:045112–045119.

Russell LM, Seinfeld JH, Flagan RC, Ferek RJ, Hegg DA, Hobbs PV, Wobrock W, Flossmann AI, O’Dowd CD, Nielsen KE, Durkee PA. 1999. Aerosol dynamics in ship tracks. Journal of Geophysical Research Atmospheres 104(D24):31077.

Russell LM, Rasch PJ, Mace GM, Jackson RB, Shepherd J, Liss P, Leinen M, Schimel D, Vaughan NE, Janetos AC, Boyd PW, Norby RJ, Caldeira K, Merikanto J, Artaxo P, Melillo J, Morgan MG. 2012. Ecosystem impacts of geoengineering: A review for developing a science plan. Ambio 41:350–369.

Russell LM, Sorooshian A, Seinfeld JH, Albrecht BA, Nenes A, Ahlm L, Chen Y-C, Coggon M, Craven JS, Flagan RC, Frossard AA, Jonsson H, Jung E, Lin JJ, Metcalf AR, Modini R, Mülmenstädt J, Roberts GC, Shingler T, Song S, Wang Z, Wonaschütz A. 2013. Eastern Pacific Emitted Aerosol Cloud Experiment (E-PEACE). Bulletin of the American Meteorological Society (in press).

Salter S, Sortino G, Latham J. 2008. Sea-going hardware for the cloud albedo method of reversing global warming. Philosophical Transactions of the Royal Society A-Mathematical Physical and Engineering Sciences 366:3989–4006.

Shingler T, Dey S, Sorooshian A, Brechtel FJ, Wang Z, Metcalf AR, Coggon M, Mülmenstädt J, Russell LM, Jonsson HH, Seinfeld JH. 2012. Characterisation and airborne deployment of a new counterflow virtual impactor inlet. Atmospheric Measurement Techniques 5:1259–1269.

Wang H, Rasch PJ, Feingold G. 2011. Manipulating marine stratocumulus cloud amount and albedo: A process-modelling study of aerosol-cloud-precipitation interactions in response to injection of cloud condensation nuclei. Atmospheric Chemistry and Physics 11:4237–4249.


1 Cloud brightening is nonlinear, so two plumes of particles that overlap each other do not typically produce twice as much brightening.

 2 See for example an article in the September 14, 2012, issue of The Guardian, “Arctic expert predicts final collapse of sea ice within four years”; available online at (accessed November 9, 2012).

About the Author: Lynn M. Russell is a professor of atmospheric chemistry at the Scripps Institution of Oceanography of the University of California, San Diego.