Sugarcane - a principal crop for biofuel - reduces the local air temperature compared to pasturelands or fields growing soybeans or maize, according to a new study from researchers at Stanford University and the Carnegie Institution for Science. But sugarcane's effect on temperature is a "double-edged machete," as it increases ambient temperatures compared with natural vegetation.
These small local changes should be taken into consideration in studies of global climate change, the researchers said.
The researchers looked at changes in vegetation in the Brazilian Cerrado - a vast tropical savanna lying south of the Amazon basin - large areas of which have been converted from natural vegetation to agriculture in recent decades.
Increasingly, these existing agricultural areas are now being converted to sugarcane for use in biofuel production. Brazil is now second only to the United States in ethanol production, much of which is used domestically.
What the effect on global climate would be if sugarcane farming were to expand significantly is not yet clear, said David Lobell, an assistant professor in environmental Earth system science at Stanford and center fellow at the Program on Food Security and the Environment.
"The temperature changes are happening locally, where the land-use change is happening," Lobell said. "It does not seem to spill over into other countries, for example, at least as far as we can tell right now."
But Lobell said sugarcane growing in the Cerrado is definitely expanding and given that the region encompasses approximately 1.9 million square kilometers (733,000 square miles) - an area larger than Alaska - the potential exists for a globally significant effect.
Using maps and data from hundreds of satellite images, the researchers calculated the temperature, the amount of water given off and how much light was reflected rather than absorbed for each of the different types of vegetation. They found that compared to land cultivated with other annual crops, sugarcane reduced the local air temperature by an average of 0.93 degrees Celsius (1.67 F).
But compared to the natural vegetation of the Cerrado - mainly grass and shrubs - the sugarcane fields warmed the ambient air by 1.55 C (2.79 F).
Lobell said the bulk of the temperature difference is due to evapotranspiration - the moisture released to the air through the leaves of the plants and the soil. Most of the land put into sugarcane had previously been converted from natural vegetation to pastureland, said Scott Loarie, a postdoctoral researcher at Carnegie. "If someone has a farm that once was natural vegetation, that transition to pasture and annual crops caused local warming," he said. "So now as the farm is going to sugarcane, by comparison it is cooling temperatures locally."
Their research, Direct Impacts on Local Climate of Sugarcane Expansion in Brazil, is described in the current issue of Nature Climate Change.
This local cooling does not necessarily mean that the global climate is cooling as a result. It depends in part on what happens with the agriculture that was displaced by the sugarcane, Loarie said. For example, if cattle used to graze on a tract of land and some Amazon forest is cut down to provide new pasture for them, net carbon emissions will actually increase.
"You might not make any difference as far as cooling the world globally at all; in fact, you might make the world marginally warmer," he said.
"The global implications of these local effects were not a part of this study, and any discussion of mitigating global climate should consider the potential for these land use cascades."
One of the important aspects of the study, Lobell said, is that it demonstrates how satellite data can be used in real time to understand the effects of environmental changes. Most research studying the impact of biofuel use on climate has been done with computer modeling.
"I think the coolest thing about this study is you actually can see these temperature effects happening already," Lobell said. "In terms of the more general point about bio energy, I think it is another good example of why looking only at greenhouse gases is not the full picture."
Another takeaway from the study, Loarie said, is that the temperature findings support the existing rule of thumb that biofuel crops are best located on land that is already used for agriculture. That general guideline stems from the fact that there is less carbon released to the atmosphere by converting land where the existing vegetation contains low amounts of carbon, such as pasture or crops, than by cutting down the dense, carbon-rich forests in the Amazon.
Loarie said that while the study clearly showed that planting sugarcane moves the temperature closer to what it would have been if the natural vegetation had not been removed from the land, that doesn't mean the land is any closer to its natural state in other respects.
"Converting pasture to sugarcane is definitely not ecological restoration," said Chris Field, a professor of biology and of environmental Earth system science, who was involved in the research.
"Still, the direct effect on climate is potentially important enough to play a role in future decisions about land use and land management in large parts of the tropics," he said.
The study was funded by the Stanford University Global Climate and Energy Project.
Greg Asner, a professor, by courtesy, of environmental Earth system science, is a coauthor of the paper. Lobell is also a center fellow at both the Freeman Spogli Institute for International Studies and the Woods Institute for the Environment. Field is also a senior fellow at the Precourt Institute for Energy and at the Woods Institute, and director of the Department of Global Ecology at the Carnegie Institution.
FSI researchers examine the role of energy sources from regulatory, economic and societal angles. The Program on Energy and Sustainable Development (PESD) investigates how the production and consumption of energy affect human welfare and environmental quality. Professors assess 
