2 February 2015

Not enough time for geoengineering to work?

Piers Forster

Piers Forster

Piers Forster is a professor of physical climate change at the School of Earth and Environment at the University of Leeds, and the principal investigator of the ...


There is consensus among scientists and global leaders on what has to happen to arrest climate change: The latest Intergovernmental Panel on Climate Change report, endorsed by 120 country delegations in November 2014, says that without an immediate and large reduction in carbon dioxide emissions, global temperatures will sail into dangerous territory within the next 30 years, triggering effects such as major sea-level rise and increases in heavy rainfall.

Despite being in accord, though, when the same country delegations met for UN climate negotiations in Lima in December, they couldn’t reach agreement on what to do. This lack of action on mitigating emissions, despite what we know, is making more people look towards geoengineering—large-scale, technology-driven ideas that seek to either remove (or “capture”) massive amounts of carbon dioxide from the atmosphere or reduce the amount of solar radiation absorbed by the Earth.

It’s a tempting solution. Climate change is like a steam train heading towards a ravine where the bridge is out. The train is speeding up as we add more and more coal to the fire. Reducing the amount of coal being added will take a long time to slow the train, just as, even if humans were to start heavily mitigating carbon dioxide emissions, it would take centuries for atmospheric levels to begin to come down. Using the train analogy, geoengineering could provide faster solutions: carbon capture would remove burning coals from under the boiler, and solar techniques would slam on the brakes.

So could geoengineering be a quick fix? We have been researching the feasibility of such technologies as part of Britain’s Integrated Assessment of Geoengineering Proposals project, which spans engineering and the physical and social sciences. We examined two carbon capture technologies and six solar technologies in as much detail as possible, and identified two main stumbling blocks. The first issue has to do with the deployment time necessary to introduce a technology at scale, and the second with how long we would need to commit to geoengineering to make a difference.

Rocky rollouts. None of the proposed technologies really exist on anything other than paper. This raises serious concerns as to whether any of them could be developed and deployed at scale within the next few decades. We investigated one technology first proposed around 15 years ago in which sea-salt particles would be sprayed into clouds from ocean-going ships to increase the low clouds’ reflectivity. We chose this technique as a test case because some of the proposed engineering details are in the public domain, so we could use them to improve the realism of our simulations.

Our simulations found three issues that reduced the efficacy of the spraying mechanism: only certain clouds were susceptible to spraying at certain times of day; many of the sea-salt particles coagulated and rained out before they reached the cloud; and the particle plume generated by the moving ship had a tendency to sink rather than rise to cloud level (due to the evaporation of water from the generation of sea-salt). No doubt many of these obstacles would be surmountable, but development and testing take time.

International governance and legal obstacles will also slow any attempts at implementation. Even a carbon-capture technology like tree planting, which already exists and is benign on a small scale, becomes problematic when deployed on a large scale, requiring that competition for land and resources be taken into account. And all of the solar technologies we simulated led to side effects, particularly in the form of changing rainfall patterns. The side effects were uncertain, crossed national borders, and often occurred on the other side of the Earth from the deployment location. The possibility of a rogue state conducting unilateral geoengineering aside, people and governments would have to develop international legal protocols to manage the process of deployment before any technologies could be put in place.

Watching and waiting. We also found that with current observation capabilities and the inherent variability of the climate system, it would take at least a decade of careful observations to determine the impact and side effects of geoengineering. Climate is defined as the “average weather.” Floods and droughts happen even without any manmade interference, so weather needs to be averaged over at least a decade to determine the climate. A similar 10-year average would be needed to see what effect geoengineering was having on weather statistics.

There would be further complications. The effectiveness of geoengineering deployment could be affected by the weather. For example, we studied injecting sulfate aerosols into the stratosphere to reflect sunlight before it hits the Earth. But the winds in the stratosphere may blow differently than expected and produce abnormal distributions of sulphate aerosol in the stratosphere. Similarly, the technology that would have ships injecting sea-salt particles into the air to brighten marine clouds would need a strategy to cope with varying patterns of cloudiness. Careful collection of statistics to measure the effects these technologies were having on climate, and their effectiveness, would require time. Unfortunately, time may not be on the side of the would-be geoengineers: Imagine that soon after injecting particles into the stratosphere, a country experienced unprecedented flooding. The engineers would be unable to say whether the technology was to blame or not. This uncertainty could easily lead to paranoia as to what effects geoengineering was having, even if it was blameless.

It is hard to predict how much geoengineering could cool the climate over a given time frame due to a lack of sufficient information on the proposed technologies. However, we were able to roughly gauge the maximum potential of several radiation management technologies. We found that marine cloud brightening and cirrus cloud thinning may not be able to cool the planet by much more than 1 degree Celsius globally. A slightly greater cooling may be achieved by injecting sulfur dioxide into the stratosphere. Other schemes we investigated had very large local effects on either temperature or rainfall, making them less attractive as global cooling mechanisms. Moreover, any possible cooling needs to be put in context with the expected 1 degree Celsius of additional warming over the next 20 to 30 years from continued emissions of greenhouse gases. Unless we reduce greenhouse gases in the atmosphere, rising temperatures could at best only be delayed for a short while.

We need to remember, too, that even if geoengineering appeared to be effective in all the ways we hoped, it wouldn’t be possible to simply switch it off. For it to continue to be effective in a world of rising emissions, the scale of its deployment would need to grow commensurately. Suddenly stopping would then become a problem, as very rapid warming would result.  Rapid warming is damaging to many biological ecosystems which do not have time to adapt. By setting off down the geoengineering path, we would be committing future generations to the technology.

Just as slamming on the brakes can slow down a train headed for a ravine, geoengineering can cool the climate. But it won’t happen without a lot of bruising, and it’s not a quick fix. Any hero worth his or her salt should try to stop the train falling into the ravine by every means possible, but we need to design, build and fit the braking system first. We therefore need to continue to urgently research these technologies.