|(Photo by Carlye Calvin.)|
"Human uses of land, especially clearing of forest for agriculture and reforestation of abandoned farmland, are an important cause of regional climate change," concludes Bonan. The cooling is the result of the changeover of the region to crops, which reflect more sunlight back into space than forests.
The impact of land-use changes on climate is currently one of the most uncertain factors contributing to climate warming, according to the recent third assessment report from the Intergovernmental Panel on Climate Change. Most of the work to date on this subject has been with computer models and has focused on deforestation in the tropics in areas such as the Amazon. Bonan's is one of the first observational studies on the effect of temperate-forest changes on regional climate. His own earlier model results hinted at this cooling effect in the U.S. Midwest, but, explains Bonan, "I wanted to see if this pattern really exists in nature."
Since accurate temperature and land-use records do not exist for the Midwest 150 years ago, when agricultural clearing began to deforest the region, Bonan relied on a direct modern-day comparison between temperatures in predominantly forested areas and those in cropland areas to see if the different types of land cover were associated with different temperatures.
He used temperatures from 65 U.S. weather reporting stations from 1986 to 1995, where the surrounding land cover was either crops or forests and there were no nearby cities or water bodies, which can have their own distinct effects on temperatures. The cropland sites were predominantly in the Midwest, where 80% of the land is now under cultivation; the forested stations were in the Northeast, where just 20% of the land is now agricultural.
The data showed that the daily temperature range was lower in the Midwest than in the forested Northeast. This was because the daytime heating of agricultural stations across the Midwest was consistently lower than that of the forested northeastern stations. The result was a surprise, because previous regional climate studies showed that the Midwest should have a larger daily temperature range than the Northeast, due to the moderating influence of clouds on daytime heating. The eastern United States is generally cloudier than the Midwest, and more clouds reflect more solar energy back into space. Bonan's study found that temperatures in the Midwest did not rise as much during the day as they did in the Northeast, contrary to what was expected from these regional differences in cloud cover.
The seasonal character of the differences in temperatures between the agricultural Midwest and forested Northeast suggested a strong influence by the land cover. Bonan found that the cooling was most prominent in the Midwest in late spring and summer, just when crops reached their full growth. The temperature difference diminished in the fall, after harvest time.
In 1850 croplands were on their way to being the dominant land cover in the Northeast, but forests and grasslands still dominated the Midwest, with only 5% of the land under cultivation. Just 30 years later, when northeastern croplands reached their highest level, the Midwest had caught up to match the Northeast, with both at 50% of land under cultivation. As cropland then steadily declined in the Northeast, forests returned; the spread of midwestern agriculture continued for the next 100 years, peaking at about 80% in the 1980s.
To make sure that the results he was seeing were not happening in only one decade of the temperature and land-use measurements, Bonan also analyzed a 100-year record of U.S. temperatures. Before 1940, when the two regions had more similar amounts of cropland, the difference in regional daily highs was much smaller than it is today. Since 1940, as agriculture continued to spread across the Midwest and northeastern farm lands returned to forests, the temperature difference steadily increased. The Northeast became warmer in the spring and summer as forests returned.
Bonan is currently using a computerized model of the climate to further investigate the impact of historical deforestation on the eastern United States. This project is funded by NASA's Earth Observing System.
Edited by Bob Henson,
Prepared for the Web by Jacque Marshall
Last revised: Wed Aug 8 17:05:07 MDT 2001