Long-term experiments can reveal insights that were never anticipated when they were planned. We analyzed results of a three-decade-old N addition and herbivory experiment at the Cedar Creek Ecosystem Science Reserve (e001:1982-2015) to assess the relative importance of the nitrogen addition and fencing experimental treatments along with climatic variability and neighborhood effects. We report 9 years of unpublished data on the effects of a fencing treatment crossed with nitrogen (N) treatments on plant community composition and use the full 31-years of data to evaluate the impacts of two non-treatment factors – climatic variation and neighborhood effects. We wondered if non-treatment effects might be as important as the experimental treatments.
Design: From 1982-2004, 54 fenced plots had two control treatments and 7 levels of added nitrogen with 6 replicates each. In 2005, half the replicates for each treatment were randomly assigned to be unfenced, crossing fencing with nitrogen.
Results/Conclusions
We find that while N remained the most important driver, deer florivory, neighborhood effects and climate were all determinants of the plant community.
When fences were removed deer (Odocoileus virginianus) presence decreased the abundance of: Symphyotrichum oolentangiense, Oligoneuron rigidum, Lathyrus venosus, and Euphorbia corollata. The mechanism appears to be the removal of inflorescences, that is deer florivory. In contrast, in unfenced plots Artemisia ludoviciana resisted florivory reaching >50% relative abundance in some years.
Species abundances and species richness in a plot positively related to those of neighboring plots. A neighborhood with high species richness appears to have buffered the negative effects of nitrogen addition on species richness.
Growing season rainfall and temperature were also significant drivers of productivity that depended on the level of N. Productivity increased in hot and wet years and decreased under drought, but these effects were strongest with added N >9.52 gm-2 where an invasive C3 grass Elymus repens dominates.
A lesson from long-term research is that what ecologists think may matter when an experiment is designed can grow richer and more interesting as time progresses. Here we report additional factors beyond N, neighborhood effects, deer florivory and climate that all changed plant community composition.