U. ILLINOIS—The composition of some U.S. forests might be quite different 200 to 400 years from today, a new study suggests.
Researchers from the University of Illinois report that temperature and photosynthetic active radiation are the two most important variables in predicting what forest landscapes may look like in the future. After the year 2200, uncertainties become very high.
Approximately 100,000 acres of forested area west of Lake Superior which make up the Boundary Waters Canoe Area was used for the study. Using computer models PnET-II and LANDIS-II, the researchers were able to simulate 209 possible scenarios, including 13 tree species and 27 possible climate profiles to predict how the landscape will look over time.
“The tools that we developed and we’re using for the research project can be applied to any discipline dealing with risk and uncertainty in decision making,” says researcher George Gertner. “We were dealing with the uncertainties in global change predictions using the projections established by the United Nations Intergovernmental Panel of Climate Change. These projections were based on different CO 2 reduction scenarios and global circulation models. ”
The study found that the most important source of uncertainty in the forest composition prediction is from the uncertainty in temperature predictions. The second most important source is photosynthetic active radiation, the third is carbon dioxide, and the fourth is precipitation. Findings were published in Global Change Biology 2009.
“The Boundary Waters Area is significant because it’s a transitional area between boreal forests—like those in Canada, Russia, Sweden, and Norway—and temporal forests,” Gertner says. “So, if there are changes in the climate you’ll see the changes—if it gets warmer, the temporal forests will move north. Because of its proximity to Lake Superior, rainfall is not so critical there. It’s very moist. So, if you were to do a similar sort of study, say, in Illinois, temperature may not cause so much uncertainty; rainfall might.”
The team drew from the disciplines of statistics and ecology to interpret the data.
Gertner explains that in traditional uncertainty analysis, the variables are considered to be independent of one another. “But in reality, they are all interrelated. We try to account for the actual correlation of these inputs—these relationships. And that’s where the methodology is new, because of that.”
The relationships of the variables are more complicated than just raising the temperature and lowering the amount of rainfall. “One scenario might be if we establish a policy to reduce CO 2 greenhouse gas emissions by a certain level,” Gertner explains. “If we have agencies around the world who adopt these policies to make these reductions, over time the scenarios predict what will happen, but with uncertainty.”
The question is what to do about it? How to adapt? How to manage the forest for global change?
“The bottom line is that we have to have very robust systems that can handle this variability. It can’t be rigid. If we have robust systems, whatever happens, it can handle it. Sustainability comes into play in the robustness. You try to manage those areas by having more diversity, not monocultures.”
Gertner says that management can be easier with agricultural systems. “Over short intervals you can adapt very quickly. You can make big changes very quickly, but with a forest, the lifespan is 100, 200 years, so once you do something it’s longer term. We need to be making policies now that will affect our forests hundreds of years from now.”
Funding was provided by U.S. Department of Agriculture McIntire-Stennis funds and U.S. Army Corps of Engineers Construction Engineering Research Laboratory funds.
University of Illinois: www.aces.uiuc.edu/news