Suddenly stopping the technique, because of technology failure or a breakdown in international co-operation, would cause temperatures to rise rapidly. The speed of this rise depends mainly on the concentration of greenhouse gases in the atmosphere, a team from Canada and the US found, as well as how long the geoengineering system had been in place. Climate sensitivity was only a secondary factor.

"Should solar radiation management be used and then terminated, decadal surface temperature trends nearly everywhere on land would be extremely large compared to trends that have occurred historically," Kelly McCusker of the University of Victoria, Canada, told environmentalresearchweb. "This would have major implications for the ability of living organisms to adapt or migrate, and for the maintenance of a satisfactory level of food production."

McCusker and colleagues found that abruptly stopping injection of sulphate aerosols would bring sudden and sustained warming over a large area of land. "Surface warming trends in particular following solar radiation management cessation are much spatially broader and greater in magnitude than anything one would find just due to natural processes," said McCusker.

In an absolute sense, the temperature changes would be largest in winter over high latitude land, the team discovered. But compared to historical fluctuations, the tropics in summertime – an area where there is usually very little variation – would see the largest changes.

"According to our simulations, tropical regions like South Asia and sub-Saharan Africa are hit particularly hard, the very same regions that are home to many of the world's most food insecure populations," McCusker explained.

The team paid special attention to summer warming over land as this is crucial for many crops. Previous studies, in contrast, have used global or long-term averaging of temperature trends. After geoengineering cut-off, the land-averaged five-year temperature trend was 3.3 °C per decade, while in some places warming was as fast as 15 °C per decade.

Global mean precipitation would also increase rapidly after such a geoengineering scheme stopped, but the trend would be less uniform than for temperature, with almost half of land experiencing drying. Warming and drying trends together could severely affect food production.

Some have suggested combining solar radiation management with cutting greenhouse gas emissions to combat ocean acidification as well as limit global warming. Now it seems that cutting greenhouse gas emissions is also essential to avoid the risk of rapid warming if geoengineering, once implemented, has to stop.

"Reduction of greenhouse gas emissions is still the answer," said McCusker. "Even if solar radiation management is deployed in the event that the impacts of global warming are deemed to have become too severe, reduction of greenhouse gases must occur concurrently to avoid the risk of rapid and spatially extensive climate changes."

The geoengineering technology would need to be maintained for thousands of years in order to allow carbon dioxide concentrations to return to natural levels.

To come up with their results, the team used the Community Climate System Model version 4 (CCSM4) under the RCP8.5 scenario, which gives a radiative forcing of about 8.5 W per square metre above preindustrial levels by 2100. The team introduced geoengineering via stratospheric sulphate aerosols in the year 2045, increasing the sulphate loading from zero to 8 Tg of sulphur equivalent over three years. This brought temperatures down roughly 1 °C back to their level at the start of the 21st century.

Over time the team increased the sulphate concentration in the model, to stabilize global climate at this 2000 level as greenhouse gas concentrations rose. Once geoengineering had been in place for 25 years, in the year 2060, they removed the sulphate to simulate a sudden shutdown.

As a result of the stoppage, global average surface air temperature in the model increased by nearly 4 °C within 30 years. Without geoengineering and its removal, the temperature would have risen less than 2 °C over the same timeframe. For the first twenty years of the shutoff, temperatures rose at slightly over 1 °C per decade, roughly six times faster than the global warming trend since 1975.

Geoengineering with stratospheric aerosols could also cause problems such as stratospheric ozone depletion, or changes in ocean and atmospheric circulation that destabilize the West Antarctic ice sheet, the researchers say.

So what's next for the team? "There are many more aspects of geoengineering with solar radiation management that are not well known, especially regarding how it would affect the ocean," said McCusker. "We will continue to study how the climate responds to solar radiation management and focus on areas that are not well understood."

McCusker and colleagues reported their work in Environmental Research Letters (ERL).

Related links

Related stories