Boston, Mass. – Linking vegetation models to climate models when approximating the Earth's past and future climates may make climate predictions more accurate and could provide a better picture of the effects of global warming on the Earth, according to Penn State researchers.
"Recent studies show that if accurate vegetation is not included in global climate models, anomalies of up to 4 degrees Fahrenheit and a third of an inch of rain per day can occur," says Persaram O. Batra, Penn State graduate student in geosciences.
The way that vegetation is incorporated into a climate model is important. The worst case is of course when vegetation is completely ignored. Assigning uniform fixed vegetation, i.e. grass land or mixed forest, to the land masses does get vegetation into the model, but not accurately. A better choice would be to assign fixed accurate vegetation to land masses, putting grassland where there was grassland and coniferous forest where there were coniferous forests. While this is a good option, it is still a static one. The best incorporation of vegetation is accurate, interactively modeled vegetation data that can influence and be influenced by the climate model.
Atmospheric global climate models often do not, in themselves, include vegetation data, but can be linked to separate vegetation models. The climate model values of such variables as temperature and precipitation are fed into the vegetation model, which produces a vegetation cover for the Earth deciding where tundra, savannahs, temperate and tropical forests would occur. This data, and the effects on climate, including variables like temperature changes and reflectivity are fed back into the climate model that is adjusted and the process is repeated numerous times.
Batra, David Pollard, research associate, and Eric Barron, professor of geosciences and director, Penn State College of Earth and Mineral Sciences' Environment Institute, looked at four different vegetation models and linked them to the GENESIS atmospheric global climate model.
"We were looking at three time periods in the past, that of the Miocene, 20 million years ago, oxygen isotope stage three between 30 and 42 thousand years ago, and the last glacial maximum 21,000 years ago," Batra told attendees at the spring meeting of the American Geophysical Union today (May 29) in Boston.
Batra compared the results of the four vegetation models to what is known about actual vegetation during those time periods on Earth. Information on vegetation during the last glacial maximum is fairly complete, but what is known about the Miocene is less complete. "We want to see how robust the various vegetation models are at different time periods," says Batra. "Then we can use the best models to see how climate change would affect vegetation patterns in the future."
Vegetation can have a substantial impact on climate. If an area is covered with tundra type vegetation, the high reflectivity of the snow, when it falls, will cool that area of the planet. However, if that same area is covered with coniferous forests, the snow would fall to the ground and the dark surface of the treetops would absorb more of the sun's energy and warm, rather than cool, the area. The researchers found that none of the climate models tested were perfect and that some differences occurred between the models. Some models did better in modeling tropical vegetation while others were better at temperate vegetation. Also, some of the vegetation models consider the effects of carbon dioxide on plants, while others ignore carbon dioxide.
Because plant growth and type is dependent on the levels of carbon dioxide and carbon dioxide serves as a greenhouse gas, the researchers believe that its inclusion in the vegetation models may be important.
Atmospheric global climate models are large, complex computer programs that are only as accurate as the data they have and the variables they cover. Adding vegetation into the mix, provides a better picture of the interconnected changes that occur as climate changes.
Materials provided by Penn State. Note: Content may be edited for style and length.
Cite This Page: