Climate, light, earthworms, nutrients and deer herbivory as predictors of seedling growth at the prairie-forest ecotone in western Minnesota

Background/Question/Methods Forests in the eastern United States face a warming climate, but how important might anticipated warming be in comparison to other forces shaping tree population dynamics? To address this question, in early 2012 we transplanted over 4000 seedlings to areas south of their seed source along a 5°C transect at the prairie-forest ecotone in western Minnesota.  Study species include three native trees—Acer saccharum (sugar maple), Quercus macrocarpa (bur oak), and Tilia americana (basswood)—and the invasive Rhamnus cathartica (European buckthorn).  Here we report on growth of surviving seedlings after three growing seasons as a function of climate, deer density, light, soil nutrients, and invasive earthworm counts and biomass. At each site, seedlings grow in 12 plots, arranged into four clusters, with a high, medium, and low seedling density plot in each cluster.  Each plot is divided into two subplots, one of which is protected from deer herbivory by a small exclosure (2.5 x 1.5 m).  Temperature and moisture are monitored via sensors with dataloggers.  We estimated light using hemispherical photos, deer density using fecal pellet counts, and earthworm biomass using hot mustard extraction.  Soil properties analyzed include nitrogen, phosphorous, organic carbon, and pH.

Results/Conclusions All four tree species exhibited substantially greater height growth when protected from deer herbivory.  T. americana seedlings have grown the most both inside deer exclosures (+301%) and in adjacent control plots (+121%), and T. americana has the largest gap in height growth for protected versus unprotected seedlings.  At the other extreme, Q. macrocarpa in our control plots have not, on average, grown at all.  To examine the effects on growth of climate versus other stressors, we performed an analysis of deviance and compared predictors and suites of predictors to a base model.  Climate (temperature and moisture), soil nutrients (considered together), light, earthworms (count and biomass), and deer density all substantially improved growth predictions over the base model, though not equally so.  For example, climate reduced overall deviance by 63% more than did earthworms and 25% more than the suite of soil nutrients, but light was more important than climate by 19%.  For trees growing in our control plots, deer density was 94% more important for predicting growth than was climate.  Thus we conclude that while climate change of the magnitude predicted for the coming century will impact tree growth, other factors are of similar or even greater importance.