The response of tree ring δ15N to whole-watershed urea fertilization at the Fernow Experimental Forest, WV

Abstract

Plant tissue δ15N is frequently used as a proxy for N availability and N cycle dynamics, and the δ15N signature of tree rings could potentially be used to reconstruct past changes in the N cycle due to forest disturbance or anthropogenic N deposition. However, there are substantial uncertainties regarding how effectively tree ring δ15N records N cycle dynamics. We used increment tree cores from a forested watershed that received a one-time application of urea, along with the long-term stream water chemistry record from that watershed and a nearby reference watershed, to determine the effectiveness of tree ring δ15N in recording a change in N availability, and whether its effectiveness differed by species or mycorrhizal type. Tree ring δ15N of three species increased rapidly (within ~1 to 3 years) following fertilization (Quercus rubra, Fagus grandifolia, and Prunus serotina), while that of Liriodendron tulipifera did not respond to fertilization but increased ~10 years later. Tree ring δ15N tended to remain elevated throughout the measured time period (1967–2000), well past the pulsed fertilization response in stream water. This extended δ15N response may be partially caused by chronic atmospheric N deposition in the region, which also contributed to greater losses of nitrate in stream water by ~1980. Additionally, local recycling of N compounds, and retranslocation of N within the trees, may account for the persistence of elevated tree ring δ15N levels beyond the direct fertilization effects. Collectively, these results confirm that tree ring δ15N from some species can document the onset of historical changes in the N cycle. We suggest that studies utilizing tree ring δ15N as a proxy for long-term N cycle dynamics should look for a consistent pattern of change among several species rather than relying on the record from a single species.