Climate change is rapidly transforming forests over much of the globe in ways that are not anticipated by current science. Large scale forest diebacks, apparently linked to interactions involving drought, warm winters, and other species, are becoming alarmingly frequent. Models of biodiversity and climate have not provided guidance on if, where or when such responses will occur. Instead models tend to provide potential numbers of extinctions, but such forecasts are not linked in any mechanistic way to the processes that could cause them. Both modeling and field studies rely on aggregate measures of species presence or absence, or their relative abundance at regional scales. However, climate acts on individuals. Aggregating data on individual trees to the level of a whole species hides or may even change predictions of climate effects. This study aims to link individual scale tree processes to regional species level responses by sampling and analyzing data about individuals across their entire range and corresponding range in climate conditions. It will use data from existing research sites, plus the platform of sites that form the core of the new National Ecological Observatory Network. These data will be collectively synthesized and used to develop computer models that can help determine when and where predicting climate impacts on biodiversity is a plausible goal. The models will also reveal where surprises are likely to occur and can provide feedback to expectations of individual tree health and vulnerability to environmental changes.
This study will provide the first forecasts of the vulnerability of forest biodiversity to changes in climate that are directly linked to the biological processes that are most sensitive. The goal is to provide forecasts of the distribution, growth, reproduction and risks of mortality for tree species making up the nation's forests. These predictions will help scientists, forest managers and policy makers anticipate the combined risks of increasing drought and longer growing seasons. Methods and results developed during this project will be disseminated through workshops for training resource managers, as well as graduate students and postdoctoral associates at a number of universities.
