Print PageTree survival is an important process affecting forest communities. Identification of the factors that influence survival of tree seedlings, a critical life history stage, should increase our understanding of forest dynamics in general. The goal of this study was to assess the relationships between species-specific seedling survival and a range of functional traits for seven tree species common to the eastern US. Planted seedlings were sampled from eight plots, with open or closed canopies, within five stands, each occurring in one of two different regions in Michigan’s Lower Peninsula. Height, growth, and status (dead or living) for each of 7683 seedlings were monitored over the period 2009-2010. We explored the factors underlying differential survival across species and sites, and hypothesized that differences could be partly explained by variation in species-specific functional traits. We sampled 96 seedlings, and measured specific leaf area (SLA), wood density, biomass, and height for each individual. Light response curves were measured for Acer saccharum and Quercus velutina to quantify maximum photosynthetic rate, dark respiration rate, and light-use efficiency. Multiple regression analyses were used to determine if functional traits could help explain seedling survival for the 2009 growing season and the 2009-2010 winter season.
Results/Conclusions
Survival ranged from 10% (Carya ovata, winter) to 100% (multiple species). In the regression analysis, functional traits significantly (p < 0.05) explained 34% of the variation in winter survival but did not explain growing season survival. The best predictor of winter survival was wood density (p = 0.002) such that survival was higher for species with higher wood density. Though the model for growing season survival was not significant, possibly due to sample size, species associated with higher aboveground biomass tended to have higher survival (p = 0.020). When species and plot were incorporated, these effects explained an additional 51% and 38% of the variation in growing season and winter survival, respectively; however, the regressions were not significant (p > 0.05), likely due to a small sample size compared to the number of species and plots sampled. However, this does suggest that identification of other species-level traits and environmental factors not considered in this analysis should improve our ability to predict seedling survival in these species. In general, this work highlights the potential for using functional traits to predict seedling survival, indicating that functional traits may be important for understanding broader-scale forest dynamics.
