Protecting the Arctic

Further written evidence from Professor Stephen Salter, Emeritus Professor of Engineering Design, Edinburgh University

I have been working on the design of seagoing hardware to implement John Latham’s proposal to reverse global warming by increasing the reflectivity of marine stratocumulus clouds. I would like to comment on some points made by Professor Lenton in his evidence to your committee of 21 February 2012.

In reply to the question about costs from the Chair (Q 40) he talks about meddling with Arctic cloud cover. It is important to distinguish between two very different techniques for reducing incoming solar radiation. The first method injects sulphur high in the stratosphere. From studies of volcanic eruptions we know that this does produce a general world-wide cooling which lasts about two years. The second method injects sea salt particles into the low troposphere, sometimes only a few hundred metres above the sea surface, to increase the number of drops in a cloud by providing extra condensation nuclei which are scarce in mid ocean. For the same amount of water, a cloud containing a large number of small drops reflects more than a small number of large ones. The lifetime of the salt residues is short, only a few days, and so spray has to be done continuously.

The long life of the stratospheric injection means that the aerosol will get everywhere but the short life of the tropospheric injection means that we have control of where and when we do the spraying. I agree with Professor Lenton where he speaks about changing gradients (Q 40). But if there are gradients which are having adverse effects we can plan the spray patterns to make them more acceptable. We can observe the results carefully and respond quickly in the same way that a driver can pass bends in an unfamiliar road. The final effect will be net cooling but we are in control of the gradients in the initial stages. Professor Lenton could point to gradients in sea surface temperature of which he does not approve and the spray vessels would come to the rescue.

The figure below is from a paper by the Hadley Centre using a very high resolution climate model that shows the changes to incoming solar radiation resulting from the injection of stratospheric sulphur. The results do not seem gradient free. This model predicts that injecting enough sulphur to produce a global cooling of 1.1 watt per square metre will work in the wrong direction in the Arctic, so making ice loss and methane release very much worse by increasing shortwave solar radiation by as much as10 watts per square metre in some places. This may be because the sun’s rays, which are coming in at a low angle of incidence and might have missed the earth, are scattered from high in the stratosphere at 90 degrees from the direction of the incoming beam.

Figure: [1] Geoengineering by stratospheric SO2 injection: results from the Met Office HadGEM2 climate model and comparison with the Goddard Institute for Space Studies Model E", from Atmospheric Chemistry & Physics 10 2010, Figure 2: Annual mean change in incident surface radiation due to 5 million tonnes of SO2 as predicted by HadGEM2.

But there is a second effect in play. Work by Kristjansson at the University of Oslo has shown that there is also a warming effect in winter because reflecting particles cannot tell up from down. They act like a blanket and will send back long-wave radiation that would otherwise have gone out to dark cold space. The advantage of the short life of tropospheric injections is that we can be sure than we never let any of the sprayed material get near the Arctic.

If we can accept the gross engineering simplification that the climate system is a machine which moves heat from the hot tropics to the cold poles then we can see that heat which is reflected en route will not reach its previous destination. Using cloud albedo control anywhere will tend to cool the pole of the hemisphere in which it is released no matter where it is done. This has been confirmed by Rasch at Pacific North Western with a far more sophisticated analysis than mine.

Professor Lenton also repeats a point made by many critics of geoengineering that once you start geoengineering you have to continue. I must disagree. You have to continue only until emissions have fallen sufficiently or CO2 removal methods have proved effective or there is a collective world view that abrupt global warming is a good thing after all. No action by the geoengineering community is impeding these. Indeed everyone working in the field hopes that geoengineering will never be needed but fears that it might be needed with the greatest urgency. This is like the view of people who hope and pray that houses will not catch fire and cars will not crash but still want emergency services to be well trained and well equipped with ambulances and fires engines.

The urgency of the need for geoengineering will increase if the PIOMAS model for Arctic ice volume turns out to be accurate because the release of methane is irreversible. I draw the attention of the Committee to the comparison between PIOMAS and direct observation from US submarines from 1975 to 2005 in the figure below. If anything PIOMAS is over-estimating the thinner ice measurements. I fully share the anxieties of Arctic Methane Emergency Group.

Figure: [2] Uncertainty in Modelled Arctic Sea Ice Volume", Axel Schweiger, Ron Lindsay, Jinlun Zhang, Mike Steele and Harry Stern, February 2011, Figure 2: Comparison of PIOMAS ice thickness estimates with observations from US submarines.

Professor Lenton also says that 40% of anthropogenic emissions can be eliminated at zero cost. I hope that he is correct but the remorseless increase in the slope of the Keeling curve of atmospheric concentration of CO2 shows that this has yet to happen. At present a methane concentration of about 1.8 parts per million is widely thought to contribute about one third of the warming of CO2 at nearly 400 parts per million. If a small fraction of stored methane was released it could easily take over from CO2 as the main driver of climate change.

Two urgent programmes should be put in place.

The first is more comprehensive observations of methane concentration, ice thickness and energy fluxes over the Arctic region with results linked to many different computer climate models.

The second is the design, construction and testing of engineering hardware which could reverse Arctic warming so that reliable equipment is ready to be deployed immediately that the political decision to do so has been taken. I very much hope that this will not be necessary but, if it is needed, the need may be desperate and the time very short.

5 March 2012

[1] Not printed ( )


[2] Not printed. (

Prepared 20th March 2012