This dissertation explores the relationship between the physical environment and the spatial pattern of the concentration of nitrogen in tree leaves, or foliar N. Maps of foliar N are of great interest due to relationships among foliar N, forest production, and N cycling. I propose that the spatial variability of foliar N is the outcome of a localized process involving species functional traits, the environment experienced by the species, and humans’ effect on that environment. I obtained species-level and whole-canopy foliar N data from a diverse set of seventy-five forest plots within the Adirondack Park, New York. These plots contained a wide range of variation in foliar N and six factors hypothesized to control its spatial distribution (i.e. spatial controls): species composition, atmospheric N deposition, disturbance history, temperature, moisture availability, and bedrock geology. I also used a subset of the field data and imagery from the Hyperion hyperspectral sensor to predict whole-canopy foliar N with thirty-meter spatial resolution across two 185x7.7 km satellite images. Analysis of the field and satellite foliar N datasets against descriptions of spatial controls obtained from field surveys and a geographic information system (GIS) revealed that species composition was the primary control on the spatial pattern of whole-canopy foliar N. In fact, inter-specific variation in the functional trait of mean foliar N accounted for 93% of the variation in the field whole-canopy foliar N data. The remaining intraspecific variability was related to all six hypothesized spatial controls, and especially to the anthropogenic controls of N deposition and disturbance history. Interestingly, the marked species differences in foliar N response to a single spatial control of N deposition, or to multiple spatial controls were strongly related to species’ functional traits of leaf mass per area (LMA) and shade tolerance. In sum, I found the inter- and intra-specific sources of variability to explain 97% of the spatial variability in my field whole-canopy foliar N dataset. Given this evidence, I strongly suggest that maps of foliar N can be further developed as tools to discover fundamental ecological principles and indicate ecosystem response to environmental impacts.
|