New data research sees the forest from the trees

Two new research papers outline how new mathematical models can more accurately predict how climate change could affect the world's forests.

Microsoft and several academics published two research papers on Thursday outlining how improved data analysis and new mathematical models can more accurately predict how climate change could affect the world's forests.

The research could prove to be a damper for climate change critics, who argue that human activity -- which causes the release of large amounts of greenhouse gases such as carbon dioxide -- is not responsible for changes in the environment, said Drew Purves, an ecologist who works in Microsoft's computational science research branch in Cambridge, England.

In as few as five years, scientists could use these more accurate mathematical models to gain a much broader understanding of how forests will react to environmental changes, he said. While other aspects of the environment, such as the oceans, have been studied more in-depth, the impact of forests on the environment has not been probed so well.

Both papers were published in Science magazine on Thursday. One of the papers, authored by Purves and Stephen Pacala at Princeton University in the US deals with "dynamic global vegetation models," which can simulate how forests react to various climate changes.

The models are essentially sets of mathematical equations that attempt to accurately describe how the forests will behave. Other forestry researchers have been able to do this well with individual trees or groups of trees in an isolated locale, but the models haven't scaled well to look at forests from a worldwide perspective, Purves said.

"We think we've managed to crack that problem," Purves said. "We understand how the growth and death of individual trees leads to the long-term future of forests."

The new models are more realistic now, due to new math models and better computational data analysis, Purves said. They're drawing on forestry records from as long as 200 years ago, many of which are very detailed studies. The information comprises metrics such as the species of tree, growth over a 10-year period and when the tree dies.

That information is crunched in an equation. The math models aren't perfect yet since they're attempting to describe facets of the earth's environment that may not be completely understood. When those uncertainties are highlighted, "it is very easy for skeptics to attack the whole exercise," Purves said. There's bound to be a learning curve, he said.

The second paper focuses on how tree seeds have been dispersed in around 90,000 plots of land in Spain. The paper concludes that seeds eaten and distributed by the animals have a greater chance of becoming trees than seeds distributed by wind. That's because wind-distributed seeds are more likely to end up in less conducive places for tree growth.

The paper was written by Purves and Daniel Montoya of the Universidad de Alcala in Madrid, with Miguel A. Rodriguez of the Universidad de Alcala and Miguel A. Zavala of Centro de Investigacion Forestal, Instituto Nacional de Investigacion y Tecnologia Agraria y Alimentaria in Madrid.

Links to both of the papers are on Microsoft's Research Web site.

Join the newsletter!

Error: Please check your email address.

More about Microsoft

Show Comments

Market Place