Changes in the growth rates of trees in Eastern North America accounting for the fitness – suitability hypothesis

Manuel Bernal-Escobar, Department of Biology, University of Miami, Miami, FL, Kenneth J. Feeley, Department of Biology, University of Miami, Coral Gables, FL and Daniel F. Zuleta, ForestGEO and NGEE–Tropics, Smithsonian National Museum of Natural History, Washington, DC

Background/Question/Methods
There is a growing interest in quantifying the long-term effects of climate change on tree growth and forests around the world. It is hypothesized that trees will respond to increasing atmospheric CO2 concentrations with accelerated growth, creating a large net carbon sink that could help to offset anthropogenic greenhouse gas emissions and thereby slow climate change. Support for this hypothesis is mixed; some studies show increasing tree growth rates through time while others show decreasing tree growth rates. A possible explanation for these conflicting results is that the response of trees to climate change is highly idiosyncratic and can depend on factors that have typically not been accounted for in these previous studies. More specifically, according to the fitness-suitability hypothesis, trees that are growing in areas where conditions are becoming more favorable through time should increase in growth while trees growing in areas where conditions are becoming less favorable through time should decrease in growth despite increasing CO2. To account for this possibility, it is important to standardize the climatic conditions experienced by different individuals or populations through time. In this study, we calculated tree growth rates for thousands of trees of 37 species at 558 locations throughout Eastern North America based on tree-ring data available in the International Tree Ring Data Bank (ITRDB). For each species, we created species distribution models (SDMs) based on available occurrence locations in the Botanical Information and Ecology Network (BIEN) and the Global Biodiversity Information Facility (GBIF). We then used these SDMs to estimate changes in climatic suitability for each tree population from 1915 to 1995 using the CRU TS3.10 historic climatic databases. Lastly, we assessed the relationships between tree growth (correcting for tree size and ontogeny) and climatic suitability using linear mixed models accounting for species and site-level effects.
Results/Conclusions
We found that climate suitability has a positive effect on tree growth rates but that tree growth rates are tending to decrease through time regardless of changes in suitability, although this relationship was highly variable across populations (i.e. each species x site combination). These results indicate that changes in climate suitability are important for driving patterns of tree growth, but that changes in other climatic and environmental factors that are not included in our study may be having negative impacts on tree growth, outweighing any potential benefit of increasing CO2.