A Framework for Geoengineering

This week's Economist includes coverage of a recent meeting of scientists at Asilomar, in California, to discuss the ground rules for pursuing "geoengineering", the deliberate, large-scale modification of the earth's environment. The purpose of the geoengineering now under consideration is to limit or reverse the effects of climate change, presumably whether man-made or otherwise. This is a notion that provokes great anxiety or outright revulsion on the part of many who feel our only acceptable response to global warming is to return the planet to something approximating its pre-industrial state by eliminating the emissions and land-use changes that have accumulated over the last century or more. However, for those of us who doubt either the efficacy or achievability of such drastic changes in the economy and our lifestyles, geoengineering is at least a legitimate, complementary option along with mitigation, and potentially our last hope of averting a worst-case climate scenario, should one arise.

Anyone who is convinced of the dangers of global warming or climate change, whichever you prefer, implicitly accepts the potential of geoengineering, because anthropogenic climate change (AGW) ultimately amounts to an uncontrolled experiment in geoengineering on a global scale. The kinds of experiments proposed by researchers meeting at Asilomar--the site of other notable, long-view discussions in the past--would operate on a much smaller scale, at least initially, with the goal of either undoing or holding temporarily in abeyance the changes resulting from humanity's emissions of heat-trapping gases in excess of the capacity of the earth's massive natural GHG-recycling facilities to absorb. For that matter, geoengineering might even be useful if it turned out that AGW was only one of several factors combining to shift conditions away from the benevolent state that has supported humanity's rise as the dominant species on the planet.

This is an issue that I've been following for a long time, though I haven't written about it very often here. My interest in geoengineering was piqued in the 1990s by proposals to sequester large quantities of CO2 in the oceans by stimulating plankton growth where there naturally wasn't much. That's only one of many possible approaches that fall into a broad family of carbon-removal strategies constituting one of the two main geoengineering categories The Economist considered. "Solar Radiation Management", the other category, includes strategies for reducing the amount of solar energy the earth receives or retains. That could run to putting large numbers of small particles in the upper atmosphere or orbiting giant mirrors to deflect sunlight off into space. It might even be as simple as painting all rooftops white--a bit of a problem if they're all covered with dark solar panels.

The basic problem seems to be convincing everyone potentially affected--which of course might include everyone on earth, or at least their representatives--to trust researchers to keep the impact of their experiments strictly limited and under tight control. The session at Asilomar apparently endorsed a set of steps called the "Oxford Principles", which describe five key elements for gaining concurrence:

1. Geoengineering to be regulated as a public good.

2. Public participation in geoengineering decision-making.

3. Disclosure of geoengineering research and open publication of results.

4. Independent assessment of impacts.

5. Governance before deployment.

Now, these sound pretty good as a set of basic principles, particularly if your goal as a researcher, or as the institution or nation funding the research, is to get everyone onboard before you start. Among other things that might avoid having someone turn up later to accuse you of making things worse, at least locally. Geoengineering liability is a serious concern at the individual and institutional level, and it could extend to being considered an act of war at the national level, if things turned out really badly. Unfortunately, when I consider how these principles might actually work--including stifling the involvement of for-profit companies in either the funding or actual R&D role--I believe they describe a likely path to doing nothing. Imagine having tried to get the delegates at Copenhagen to agree to let someone put finely-divided salt particles into the atmosphere over, say, the Arctic, to make clouds more reflective. Might as well have tried to sell them the Brooklyn Bridge at the same time.

That's the core of the problem as I see it: If we do end up needing to deploy geoengineering, it's likely to be precisely because we were unable to get every country on earth--or even just the small subset of large emitters--on the same page with regard to climate change, let alone establish a universally-trusted body to oversee their mitigation efforts. If we yoke geoengineering to the same UNFCCC/IPCC process that brought us the Copenhagen Climate Conference and the Kyoto Protocol, then we might as well forget it and try to figure out where to invest in the likely new beachfront property of the 2050s. In any case, as appealing as the Oxford Principles might seem from a stakeholder-engagement perspective for implementing large-scale geoengineering someday in the future, they look too unwieldy to guide the small-scale R&D efforts that would be needed to determine which, if any, of these schemes actually have merit.

One possible alternative would start with the same concept of climate forcing that underpins today's climate models. (And by the way, any serious geoengineering effort is going to require really good, trustworthy global and regional climate models, the inherent limitations of which are one of the main complaints of climate skeptics.) The observed increases in CO2 and other greenhouse gases equate to roughly an extra 2 watts per square meter of heat radiation retained by the earth, out of a total average influx of around 240 w/m2 at the earth's surface. So if 1% more radiation/retention is enough to cause the global warming we have observed, then what is the maximum equivalent level of geoengineering testing we'd be willing to tolerate to see whether any of these techniques might help? 0.01%, or 1/100th of the scale of the problem itself? And what would be the most any one experiment should be allowed to fiddle with? 0.0001%, or one part per million, allowing at least 100 small experiments under the overall limit? (For experiments dealing with carbon-removal, rather than radiation management, this forcing threshold could easily be converted to its tons-per-year of CO2 equivalent.) Whatever the level, the idea would be to keep any individual experiment, and all of them together, below the level at which they could make things noticeably worse by accident--with a healthy margin for error--without preventing any work from being done on this at all.

Some regard geoengineering as yet another outgrowth of our technological hubris and thus unworthy of further research. While I respect anyone's right to that view, I would also question their commitment to the survival of the human race. That's because I'm deeply skeptical that our current approach to climate change can work fast enough and on the necessary scale to avert the worst outcomes scientists suggest we face. We already live in a geoengineered world that couldn't support a fraction of its current population if we returned it all to its natural, pre-industrial state. That's not a license for unlimited tinkering with our environment, and perhaps that's the underlying concern: that the same techniques that might be applied to reduce the impact of climate change might eventually be employed in risky attempts to fine-tune an even more optimal climate than the one we inherited. Science is like that, as demonstrated by nuclear proliferation and questionable medical practices. But while I share those misgivings with respect to the potential misuse of geoengineering, I sure want us to have some of these options in our hip pocket if we ever really need them.