January 28, 2010 | A new study led by NCAR scientist Peter Lawrence has found that impacts to Earth's hydrological cycle—not changes in the reflection or absorption of solar radiation—are the most important driving force in how human-caused land use changes affect climate. The findings contradict previous research and suggest that land cleared of forest for agriculture could have a greater warming impact on regional temperatures than previously thought.
The goal of Lawrence's study was to examine a discrepancy between model simulations and real-world observations. Scientists have widely accepted that the main factor in how land use affects climate is surface albedo. Models generally show that deforestation has a cooling effect, as darker leaves are replaced by lighter-colored plants such as grasses and crops. Observations indicate, however, that regional climate has warmed substantially in mid-latitude and tropical regions where forests have been replaced with grasses and crops.
By running experiments with the NCAR-based Community Climate System Model (CCSM), Lawrence found that the reason for the discrepancy was rooted in how many models simulate land surface hydrology. The CCSM showed that replacing forests with grasslands and crops resulted in increased land surface evapotranspiration, contrary to observations. After Lawrence corrected the CCSM's land component to more closely reflect observed surface hydrology, model results matched real-world observations in showing that clearing land for agriculture and other uses leads to reduced evapotranspiration. This reduction in turn weakens evaporative cooling, with feedbacks that lessen rainfall and cloud cover.
The study's conclusion that deforestation in mid-latitudes and tropics may substantially enhance regional warming on top of projected global warming has implications for biofuels production, which could require widespread landscape conversion to agriculture. It also underscores the need for more research aimed at fine-tuning how climate models represent the hydrological cycle.