Reflecting Sunlight to Cool Earth: The NAS Weighs Controversial Measures in New Report

February 10, 2015
Peter Frumhoff
Director of Science and Policy and Chief Climate Scientist

The president’s science advisor John P. Holdren has often observed that humanity has three basic options for dealing with climate change: Mitigation (reducing heat-trapping emissions), adaptation (coping with unavoidable impacts of climate change), and suffering.  The more swiftly we both mitigate and adapt, the less suffering we endure and impose on future generations.

Suppose, however, that we falter and temperatures continue to rise to dangerous levels. In a climate emergency, facing high risks of major and otherwise unavoidable impacts, should the U.S. or other governments consider forced cooling of Earth by injecting reflecting aerosol particles into the stratosphere?

Today, the U.S. National Academy of Sciences (NAS) weighs in with a pair of major reports examining the scientific basis for considering this and other possible “climate interventions” — deliberate, potentially large-scale actions to reflect sunlight away from Earth or remove carbon dioxide from the atmosphere should mitigation and adaptation prove insufficient to limit the risks of dangerous climate warming.

Kudos to the National Research Council (NRC) panel, chaired by Marcia McNutt, editor-in-chief of the journal Science and former director of the U.S. Geological Survey, for tackling this set of challenging and controversial issues. It is one of a growing number of scientific and related policy assessments on a suite of potential and problematic climate responses most commonly referred to as “geoengineering.”

Reflecting sunlight to cool Earth

Here’s a synopsis of key findings from the NRC report on Climate Intervention: Reflecting Sunlight to Cool Earth. (In a related post, my colleague Brenda Ekwurzel looks at their report on Climate Intervention: Carbon Dioxide Removal and Reliable Sequestration):


“Technologies that prevent sunlight from reaching Earth’s surface could reduce average global temperatures within a few years, similar to the effects of large volcanic eruptions. While many albedo-modification [i.e. solar energy reflecting] techniques have been proposed… two strategies that could potentially have a significant impact are injection of aerosols into the stratosphere and marine cloud brightening. [T]hese methods would not require major technological innovation to be implemented and are relatively inexpensive…

However, albedo modification would only temporarily mask the warming effect of greenhouse gases and would not address atmospheric concentrations of CO2 or related impacts such as ocean acidification. In the absence of CO2 reductions, albedo-modification activities would need to be sustained indefinitely and at increasingly large scales to offset warming, with severe negative consequences if they were to be terminated. In addition, albedo modification introduces secondary effects on the ozone layer, precipitation patterns, terrestrial and marine ecosystems, and human health, with unknown social, political, and economic outcomes.

Many of the processes most relevant to albedo modification — such as those that control the formation of clouds and aerosols — are among the most difficult components of the climate system to model and monitor. Present-day observational capabilities lack sufficient capacity to monitor the environmental effects of an albedo-modification deployment. Improvements in the capacity to monitor direct and indirect changes on weather, climate, or larger Earth systems and to detect unilateral or uncoordinated deployment could help further understanding of albedo modification and climate science generally.

 [I]t would be “irrational and irresponsible” to implement sustained albedo modification without also pursuing emissions mitigation, carbon dioxide removal, or both. [The Committee] oppose[s] deployment of albedo-modification techniques, but recommend[s] further research, particularly “multiple-benefit” research that simultaneously advances basic understanding of the climate system and quantifies the technologies’ potential costs, intended and unintended consequences, and risks.

 Albedo-modification research will have legal, ethical, social, political, and economic ramifications. The committee recommend[s] the initiation of a serious deliberative process to examine what international research governance structures may be needed beyond those that already exist, and what types of research would require such governance. The degree and nature of governance should vary by activity and the associated risks, and should involve civil society in decision-making through a transparent and open process.”

No substitute for dramatic reductions in heat-trapping emissions

In other words: Proposed strategies to alter the amount of sunlight hitting the Earth’s surface by (for example) deliberately injecting millions of tons of sulfate aerosols into the stratosphere pose enormous risks and uncertainties and don ‘t address the underlying causes of global warming or other major risks from rising concentrations of carbon dioxide, such as ocean acidification. They should not be deployed today and we should do everything possible to avoid their being deployed in the future. As the NRC report emphasizes, there is no substitute for dramatic reductions in heat-trapping emissions. Preventative medicine is far more attractive than getting treated in the emergency room.

But the Committee also recommends the U.S. government invest in an “albedo-modification research program” focused on improving understanding of the intended and unintended impacts of these technologies on climate, people, and ecosystems. They consider — and firmly reject — the “moral hazard” argument that such research would somehow distract from efforts to reduce emissions, concluding that “as a society we have reached a point where the severity of the potential risks from climate change…outweigh[s] the potential [moral hazard] risks associated with a suitably designed and governed research program.”

I strongly agree. We need to better understand these technologies and their risks, even if we are determined to never deploy them. They are relatively low-cost, and if deployed unilaterally by others, would have global consequences. In the U.S. and internationally, societal debate over their use would be well served by better understanding their risks and consequences. A fuller understanding of their risks, informed by science, might well reinforce our collective determination to never use them and motivate greater commitment to mitigation and adaptation. And, should we falter in that effort, we would be well-served to better understand the impacts of such emergency-room measures.

Needed: a transparent, participatory process to guide research on impacts and risks

That said, the question of who decides what research is appropriate is tricky. To date, studies have largely been confined to computer modeling. The NRC notes that “small-scale field experiments with controlled emissions [e.g. releasing reflecting aerosols into the atmosphere] may….be helpful.” Some scientists are eager to initiate field research. In my view, the NRC Committee has it exactly right when they call for any planning of such research to be subject to a “serious deliberative process” to weigh options for its governance. Such a process, they argue, should be fully transparent and informed by the active participation of civil society.

That process should begin now and subsequent guidance on the governance of albedo-modification research established before the U.S. supports any scale-up of albedo-modification research.

About the author

More from Peter

Peter C. Frumhoff is director of science and policy and chief climate scientist at UCS. He guides organization-wide strategies to bring robust scientific expertise to bear on strengthening public understanding and public policies; guides innovative science and policy initiatives and post-doctoral fellowships; and serves as senior liaison with the scientific community, policymakers and the media.