Abstract Forest ecosystem responses to nitrogen (N) deposition depend on N fate, yet most simulated N deposition experiments focus on understory addition, overlooking canopy N processes. Here we compared canopy (CAN) and understory (UAN) N addition in a secondary forest using a 15 N tracer approach, adding 15 N as 15 NO 3 - or 15 NH 4 + . Total 15 N recovery was higher under UAN than CAN initially, but recoveries converged by day 365. CAN increased long-term N retention in woody biomass, with stems becoming the dominant sink, accounting for up to 34.9% of 15 N recovered after one year. UAN promoted short-term N retention in understory layer and long-term N retention in soil. 15 N recovery increased with soil depth under CAN but decreased under UAN from 120 to 365 days. Trees absorbed more 15 NO 3 - than 15 NH 4 + , whereas soil retained the two forms similarly. These findings indicate that N manipulation methods affect N retention magnitude and spatial distribution.
Yang et al. (Thu,) studied this question.