The demography of tree species response to climate: sapling and canopy tree growth

Despite the clear need to predict the effects of climate change on the distribution and abundance of temperate tree species, there is still only a rudimentary understanding of how climate influences key demographic processes that determine the current distribution and abundance of tree species. We use data from the U.S. Forest Service Forest Inventory and Analysis (FIA) program to quantify the relationships between two key climate variables-mean annual temperature and effective growing season precipitation-and rates of sapling and canopy tree growth for the 50 most common tree species in the eastern United States. Our models include the effects of tree size, competition, and anthropogenic nitrogen (N) deposition, both to avoid confounding effects and to provide context for the importance of variation in climate relative to other factors known to influence tree growth. The 50 species show a broad range of relationships between size and growth, in contrast to predictions of metabolic theory. The 50 species differ widely in shade tolerance, and both saplings and canopy trees show a wide range of competitive responses to total stand basal area. The competitive responses of canopy trees were more sensitive than were saplings to the size of an individual relative to the median size of trees in the stand. As has been shown in other studies with FIA data, species responses to N deposition also varied widely and were related to the type of mycorrhizal association of the tree species. Relationships between the two climate variables and tree growth were surprisingly modest, and bore little obvious relationship to the distributions of the species along climate gradients. For over a quarter of the species, there was no statistical support for a relationship between 5-yr average growing season precipitation and 5-yr average growth, and for most of the remaining species, the relationship was effectively flat over a wide range of precipitation. Responses to regional variation in mean annual temperature were stronger, but again showed little obvious correlation with the distribution of abundance of most species along the temperature gradient.