Satellite images could be used 'to predict animal extinction threat from climate change'

Satellite images could be used by scientists to help predict which species are under the most severe threat of extinction due to global warming.

Louise Gray, The Telegraph 5 Apr 09;

The photographs from space show the amount of vegetation available in an area for animals like giraffe, buffalo or wildebeest to eat.

Already the information is being used to show the impact of climate change on great wilderness areas.

Now scientists at the Zoological Society of London (ZSL) want to use the information to predict which species may be worse affected if the temperature rises and the grassland becomes a desert.

Animals that rely on the vegetation include elephants, bushbuck and antelopes and, indirectly, carnivores like lions and cheetahs.

The study, due to appear in The American Naturalist, examined patterns for 13 herbivore species in 77 African national parks.

By comparing the images known as a "normalized difference vegetation index" or NDVI with traditional aerial and on-the-ground surveys of animal numbers, researchers were able to prove a correlation between healthy grasslands and wildlife.

It means there could be potential to predict which species will die out first if climatic conditions change, giving conservationists a chance to focus on those species before it is too late.

Dr Nathalie Pettorelli, ZSL researcher and lead author of the study, said images from space could show scientists which animals may need to be protected and even moved to a different area.

"Several climatic models can be used to predict changes in NDVI, allowing scientists to forecast how climate change might affect vegetation. The correlation discovered means that the effects of climate change on wildlife could also ultimately be predicted quantitatively," she said.

"This is a really important step forward in helping to determine conservation priorities in a changing climate."

Dr Pettorelli said not every species of animal studied showed a strong relationship between NDVI and abundance. However she was confident these differences were caused by additional factors including poaching, predation or competition rather than a problem with the data.

"Even though we were unable to correct for several factors such as poaching intensity, predator density or soil nutrient status, we were still able to report a relationship between satellite indices and wildlife abundance," she said.

"This suggests that the underlying relationship between satellite data and abundance might be even stronger than is apparent from our initial analysis."