To cope with climate change, species must either adapt or move to milder climates. Already, many species are slowly shifting their range polewards, where environments are more tolerable. Mountain species are on the move too, with many migrating to higher altitudes, where the climate is cooler.

Rapidly halting geoengineering would cause land and ocean temperatures to rise at an unprecedented rate - more than double the rate we're experiencing as a result of climate change. Species would need to move over 10 kilometres a year just to keep up. To put that into context, the average migration rate in response to current warming is just 1.7 kilometres a year, about six times slower.

Solar geoengineering involves injecting aerosols into the upper stratosphere to reflect sunlight back into space and so reduce temperatures on Earth. Such measures have never been tested in nature and the impacts of geoengineering on biodiversity had remained largely unknown until now.

"Sudden termination could occur if governments lose the will or means to continue geoengineering," said Chris Trisos from the University of Maryland. To work out what impact this might have, Trisos modelled a world where 5 Teragrams of sulphur dioxide are released into the atmosphere each year, starting suddenly in 2020, and stopping abruptly 50 years later. This amount is equivalent to a quarter of the aerosols released during the volcanic eruption of Pinatubo in 1991. Investigating how solar geoengineering would affect biodiversity during the geoengineering programme and after it ceased revealed that the impacts would be profound.

"The rapid temperature increases from a sudden termination of geoengineering would likely outpace the capacity of many species to migrate to cope with the temperature change, increasing extinction risk," Trisos warned. Similarly, a sudden start to geoengineering would cause rapid cooling and may reverse some of the shifts we've already seen in response to climate change.

If this wasn't enough of a struggle, the rapid changes in climate brought on by stopping geoengineering programmes would cause substantial climate fragmentation - making habitats a mosaic of tolerable and stressful environments. This might generate barriers to dispersal, further reducing the ability of species to move to a cooler climate refuge. Species that fail to track changing climates could go extinct, even if suitable conditions can be found elsewhere.

While the low cost of solar geoengineering relative to other methods of dealing with climate change makes it an attractive prospect, the costs for biodiversity are potentially huge.

"As policymakers begin to discuss geoengineering it is important to understand the potential costs and benefits of the technology," said Trisos. "Our study highlights one potentially large cost, and therefore underlines again how important it is to reduce greenhouse gas emissions to protect biodiversity."

The use of geoengineering is poorly regulated, making the rapid implementation and termination of such technologies a likely prospect. The challenge of maintaining these programmes, and the risks associated with stopping them, present a serious concern for policymakers and scientists alike.