GLOBAL warming predictions could be wildly inaccurate because climate models fail to take changes in land use into account.
Turning more land over to farming, introducing irrigation and reforestation could have a greater effect on global warming than changes in industrial carbon dioxide emissions, says Roger Pielke, an atmospheric scientist at Colorado State University.
Pielke鈥檚 team reached their conclusion after modifying a standard climate change model to take account of changes in land use since the beginning of the 18th century. Among the factors they included were estimated changes in forest sizes, areas of farmland and irrigated land.
Advertisement
When they ran the modified model, they found that these changes had a significant influence on the climate over the past 300 years. They raised midwinter temperatures in western Europe by about 3 掳C, and speeded up the rapid jet streams high above the Atlantic and Pacific by about 6 metres per second.
The model shows forests help to cool the atmosphere as water evaporating from leaves absorbs heat. Turning forest into farmland removes this cooling effect, although irrigation made the temperature rise less severe.
Climate experts will need to rethink how they model climate change, says Pielke. 鈥淭he human influence on climate is more complex and diverse than we have appreciated. We are going to have to start thinking more holistically,鈥� he says.
Pielke also found that certain proposals in the Kyoto Protocol to reduce global warming could actually have the opposite effect. For example, planting forests to absorb carbon dioxide can actually lead to warming if they are planted where snow falls, because the land will then reflect less sunlight.
Richard Betts, a climate expert at the Hadley Centre for Climate Prediction and Research in Bracknell, Berkshire, says Pielke鈥檚 findings should be taken seriously. 鈥淭his work suggests that there are shortcomings in the protocol that have to be addressed,鈥� he says.