Crops vs. Forest? Land Use Models Predict Future Climate Change

“The land use choices we make can cause significant differences in our climate,” Gordon Bonan of the National Center for Atmospheric Research said to a crowded room in the Nicholas School, Thursday.

Depending on deforestation and agricultural expansion in the coming decades, the Earth’s climate will be impacted in various ways because different types of land cover interact with the environment differently.

Forests, in general, absorb large amounts of CO2 and solar radiation, and cool the air around them through evapo- transpiration. Cropland, on the other hand, is not as effective at capturing carbon.

However, cropland reflects sunlight better than some types of forest, raising an interesting question: is it more important to capture CO2, or to reflect sunlight and keep surface temperatures cooler? In other words, should forests with low reflectiveness be chopped down to make way for more-reflective fields? Some models say that higher reflectiveness and cooling that results from deforestation is more significant than warming caused by the increase in CO2. But a great deal of data is missing, and the debate is far from over.

Some things about climate change are clear: tropical rainforests, for example, are our “planetary salvation,” according to Bonan. They absorb a tremendous amount of carbon, but also cool the atmosphere and reflect heat. To mitigate climate change, major rainforest conservation and reforestation efforts must be undertaken.

Bonan and his colleagues at the National Center for Atmospheric Research work with computer models to predict future climate change. But Bonan’s talk, if anything, emphasized how uncertain climate models can be. The more factors that are included, such as land cover change, irrigation, and the carbon cycle, the more uncertainty there is bound to be. According to Bonan, there’s no clear way to model climate change that would result in any sort of consensus among the international scientific community.

However, consensus or not, Bonan and his colleagues are attempting to make their models as accurate as possible. Right now, their land use models “don’t account for diversity of crop types at all,” and assume a generic crop type for all cropland. Bonan hopes that a future model might be able to evaluate based on specific crop systems, such as soybeans, wheat, and corn.

Bonan’s team continually tests their model-generated hypotheses against observed data, collected at 15 sites spanning the globe. (Duke Forest is one of them.) “Experiments identify the shortcomings of a model,” Bonan said.