Although Alan Robock's "20 Reasons Why Geoengineering May Be a Bad Idea" raises legitimate questions, it seems to argue against implementation rather than against studying the underlying science. Few people are actively advocating for immediate, full-scale implementation of geoengineering techniques as a means of addressing climate change. But many people are suggesting that we learn more about the efficacy of such techniques--including Alan, who was recently awarded a National Science Foundation grant to study the effectiveness and possible consequences of injecting aerosol particles into the stratosphere to reduce incoming solar radiation.
In terms of geoengineering concerns, it's helpful to group them into three categories:
Efficacy. Clearly, any geoengineering technique first needs to achieve the intended effect for a reasonable cost--whether the goal is buying time for more sustainable solutions by reducing incoming solar radiation or addressing the root cause of warming by removing carbon dioxide from the atmosphere. It's critical that scientists be allowed to study efficacy through experimentation and modeling without being stigmatized by the assumption that their work will cause a rush to full implementation.
Impact. The environmental impacts of the technique must either be minimal or acceptable relative to the benefits of action and the consequences of inaction. Martin Bunzl makes this point clearly in "An Ethical Assessment of Geoengineering," an accompanying essay on p. 18 of Alan's article. In the case of ocean iron fertilization, 12 small, open-ocean experiments have already been conducted by oceanographers to improve understanding of both efficacy and impacts. (See "Mesoscale Iron Enrichment Experiments 1993-2005: Synthesis and Future Directions".) From the start, it was clear that these experiments weren't a danger to the environment and that their effects wouldn't last long. Scientists should be encouraged to study impacts through experimentation and modeling as long as it can be reasonably presumed that their impacts are short-lived.
Implementation. If a technique is both effective and sensitive to the environment, the following implementation questions become important: Who implements it? Who regulates it? And how do we incorporate these activities into existing regulatory and legal frameworks and treaties? These questions are difficult but not intractable, as many carefully negotiated international agreements already demonstrate, including the International Maritime Organization's London Convention on ocean dumping (signed by 80 countries in 1972, including most of the developed world) and the U.N. Law of the Sea treaty (signed and ratified by most countries except the United States).
More broadly, the provocative title of Alan's article and the quick treatment of individual concerns obscure the complexity behind these subject areas--as Martin and Ken Caldeira have addressed. We expected a summary of research results suggesting a priori that geoengineering is a bad idea, but didn't find one. Also, we found it distracting that many of Alan's concerns (i.e., ozone depletion, acid deposition, effects on cirrus clouds and plants) are specific to one technique--aerosol seeding--but offered as reasons why geoengineering in general is a bad idea. Another of Alan's examples presumes that "humans [adopt] geoengineering as a solution to global warming, with no restriction on continued carbon emissions." No one is suggesting that geoengineering replace emissions reduction.
Most surprising is Alan's conclusion that global warming is a not a difficult technical--but rather purely a political--problem, and therefore, geoengineering isn't required to solve it. We disagree. The road ahead is paved with difficult technical challenges in addition to the considerable political ones. Many new, clean technologies that promise incremental improvements in efficiency also require substantial scientific achievements--such as genetic modification of organisms to make novel substances (i.e., enzymes that process various feedstocks for cellulosic ethanol) or revolutionary advances in materials and process sciences (i.e., new thin-film technologies for solar power). Emission reductions don't simply follow from mandates; we must innovate alternatives to fossil fuels.
That we need to contemplate geoengineering to buy us time for that innovation is unfortunate. That we have the scientific, technical, and human potential to do so responsibly is not.