PS 12-120
Species-specific tree ring δ15N responses to whole-watershed urea fertilization in four Appalachian hardwoods

Monday, August 10, 2015
Exhibit Hall, Baltimore Convention Center
Mark B. Burnham, Biology, West Virginia University, Morgantown, WV
Mary Beth Adams, Northern Research Station, USDA Forest Service, Morgantown, WV
William T. Peterjohn, Biology, West Virginia University, Morgantown, WV
Background/Question/Methods

The use of tree ring δ15N to study N cycle dynamics through time has become increasingly common. In temperate forest ecosystems, increased N availability due to deposition of reactive N can lead to high rates of nitrification, which fractionates against δ15N. Nitrate (NO3) leaving the system in stream water is thus depleted in the heavy isotope, causing enrichment of the remaining N pool over time. Since a portion of the N acquired by trees from the soil is stored in tree rings, the tree-ring record of δ15N should indicate shifts in N cycle dynamics. However, δ15N trends in the tree rings of different species could vary due to differences in N demand, the primary form of N utilized, mycorrhizal associations, and landscape position. Thus, we studied whether four canopy tree species differed in their ability to record a past major N cycle disturbance in tree ring δ15N. In 1971, a one-time urea fertilization occurred in a small watershed in the Fernow Experimental Forest. We cored 4 Fagus grandifolia and Quercus rubra trees and 5 Prunus serotina and Liriodendron tulipifera trees, and measured the δ15N in individual rings for 1967-1980, and in 5-year composite samples for 1980-2000.

Results/Conclusions

Urea fertilization in 1971 caused a ~350 μM spike in stream water NO3 concentration, and the tree ring δ15N of three species (F. grandifolia, Q. rubra, and P. serotina) increased >1‰ from 1967-70 to 1972-76. In contrast, L. tulipifera tree ring δ15N did not change immediately following fertilization, but did increase ~1.5‰ from 1980 to 1985, a time period that corresponds with increased stream water nitrate due to long-term deposition. Although stream water NO3 declined rapidly in the 2 years post-fertilization, tree ring δ15N was not as responsive; F. grandifolia and P. serotina tree ring δ15N remained elevated through 2000, while that of Q. rubra declined ~0.5-0.75‰ from 1980-2000. Compared to F. grandifolia and Q. rubra, P. serotina more closely reflected the changes in stream water chemistry, although isotopic shifts in this species were delayed by ~2 years. In conclusion, we found that the record of tree ring δ15N for 3 species accurately reflected the onset of enhanced stream water nitrate concentrations, but species varied in their ability to record extent and the timing of the recovery.  We also found the isotopic record in some species may be sensitive to the nature of the change in N cycle dynamics.