A very useful paper (abstract|pdf|discussion space) comes out today in Atmospheric Chemistry and Physics by Tim Lenton and his student Naomi Vaughan. Tim told me when I was reporting the Andy Ridgwell paper on leaf albedo (Nature story|blog entry) that he’d become pretty interested in evaluating geoengineering schemes, and was setting up a group at the University of East Anglia to assess them. This paper presumably represents the first fruits of that interest, providing a ranking of most of the geoengineering schemes proposed in the literature in terms of the amount of radiative forcing they can provide.
Radiative forcing is, more or less, the difference in terms of energy per square metre that’s associated with any given action that changes the climate; it’s a pretty routine way of expressing things in IPCC-land. The IPCC puts the radiative forcing associated with the greenhouse gas industrial and industrialising societies pumped into the atmosphere from 1800 to 2005 at about 1.6W/m², and the forcing for a doubling of CO2 at about 3.7W/m².
Lenton and Vaughan first divide geoengineering proposals into two sorts: shortwave and longwave. Shortwave schemes seek to reduce the amount of energy that gets into the earth system by reflecting away incoming sunlight. Longwave schemes seek to increase the amount of energy leaving the earth system by making the atmosphere more transparent to outgoing infrared radiation — that is, by reducing the greenhouse effect. Then they assess the two with some very simple modelling (well, for the longwave there are some wrinkles, but it’s all in principle pretty simple). They don’t claim that the figures they come up with are the best available in any particular case, just that they are all derived the same way, and so allow fairly straightforward comparisons. By standardising the techniques they also show up a few errors in previous analyses: for example, if you increase the total amount of light reflected back into space by clouds, you reduce the amount reflected by the surface, simply because less light gets there in the first place.
The first and most striking conclusion is that if you want to have a big effect, go shortwave. Sulphate aerosols in the stratosphere (which were the main topic of this piece and these Climate Feedback posts) and mirrors/refractors in space (also in that piece, and in this paper by Roger Angel) both have the potential to provide as much by way of negative forcing as a doubling of CO2 provides by way of positive forcing. Not surprising; if you’re not constrained by money or by concerns about environmental side effects, you can put mirrors in the sky and particles in the stratosphere until it’s darkness at noon.
When you leave these global technologies behind, the other shortwave interventions rank, unsurprisingly, more or less according to the area they affect. Increasing the brightness of marine stratocumulus clouds, as proposed by John Latham, would affect about 17% of the earth’s surface, and the Lenton-Vaughan analysis suggests that the whitening effect would have to be considerably more marked than previous work has assumed; but if that brightening could be achieved then a negative forcing that averages more than 3W/m² should be possible. Covering non-sandy deserts with aluminium and polyethylene (not an idea I had come across before, and a pretty silly one as far as I can see: more here if you want it) makes 2% of the surface a lot brighter, and gets you an average 1.7W/m² of negative forcing, obviously very unevenly spread. Increasing the brightness of the planet’s grassland as Robert Hamwey has discussed (pdf) gets you 0.64W/m², and the Ridgwell et al idea of planting brighter crops gets you 0.44W/m² at best, croplands being smaller than grasslands. Lightening everywhere that people actually live (another idea from the Hamwey paper) gets you 0.19W/m²; increasing the area of plankton blooms that seed the creation of clouds in parts of the southern ocean gives you just 0.016W/m² (and that may be an overestimate) and restricting yourself to just creating shinier cities gives you no more than 0.01W/m².
What of the longwave?
In principle, capturing carbon dioxide from the air (pdf of the Keith et al paper) and burying it in the ground could give you whatever radiative forcing you wanted; the limits to such a scheme are entirely economic, rather than being imposed on the earth system. All the other schemes, though, which involve making changes in the natural carbon cycle, are quite constrained, with none able to counter a doubling of carbon dioxide, even given the most extreme assumptions.
The biggest effect comes from really aggressive planting of forests, as described in an essay (pdf) by Peter Read on his global gardening plans. This involves growing enough plant material in the next 50 years to more than completely make up for all the arbon dioxide lost through deforestation and land use change over the past few centuries, which is really remarkably ambitious, especially if people are still going to have some space to grow food. By 2050 this strategy gets you an effective 0.49W/m² of negative forcing thanks to 88 gigatonnes of carbon dioxide being stored away. A variant of the idea in which you grow the biomass and burn it in power stations fitted out for carbon capture and storage does even better: 0.69W/m² by 2050 and almost 2W/m² by 2100 (For the longwave calculations, the radiative forcing depends on how long the programme has been going on. It also depends on what assumptions you make about how effective carbon-emissions control is; Lenton and Vaughan calculate all the forcings in terms of what extra relief the carbon-dioxide drawdown provides in a world that is already making serious cuts in emissions).
A lower tech idea that Read is fond of, as for that matter am I, is turning biomass into biochar and ploughing it into the ground. Jim Lovelock, Lenton’s mentor and friend, was extolling this as a possible way of making things better in New Scientist last week, speaking to the in-this-case-aptly-named Gaia Vince. This may make sense for all sorts of reasons, and the fact that making the charcoal also provides you with fuel (see Johannes Lehmann’s commentary in Nature a few years ago) is obviously a plus, but even a really aggressive campaign along these lines gives ou a negative forcing of only 0.40W/m² by 2100.
After that come a bunch of ocean fertilization schemes, using phosphorous, nitrogen and iron, all of which offer something in the region of 0.1-0.2W/m². A system of pumping nutrient-rich water up to the ocean surface sketched out by Lovelock and Chris Rapley (earlier blog entry) delivers a truly meagre 0.003W/m² by 2100.
None of this, as Lenton and Vaughan are at pains to make clear, counts as an endorsement; all the schemes have side effects and risks, as well as in some cases (ahoy there, vast fleet of space parasols) quite remarkable costs. But looking at the options this way does allow a sense of what might be possible, and a way of seeing what might be done in a mix and match sort of way. And the fact that the paper is published in the discussion section of ACPD means that the various researchers whose work is discussed will have a chance to answer back, correct any poor assumptions, and carry the debate forward.
Image from flickr user gianky, under a creative commons licence.