Short-term canopy and understory nitrogen addition differ in their effects on seedlings of dominant woody species in a subtropical evergreen broadleaved forest

As atmospheric nitrogen (N) deposition continues to increase, information on how N deposition affects seedling performance of tree species is critical for predicting forest regeneration. In this study, we examined the effects of canopy N addition at 25 and 50 kg ha−1 year−1 of N (CAN25 and CAN50), and understory N addition at 25 and 50 kg ha−1 year−1 of N (UAN25 and UAN50) on the survival, physiology, and growth of seedlings of two dominant woody species (Schima superba and Ardisia quinquegona) that were transplanted into a subtropical evergreen broadleaved forest. At two years after transplanting, understory N additon greatly decreased seedling height and biomass production for both species, especially at 50 kg ha−1 year−1 of N. N-addition treatments, however, had only minor effects on seedling survival and biomass allocation. UAN50 decreased leaf light-saturated photosynthesis (Asat) in the dry season and increased stem N concentration; UAN25 increased intrinsic water-use efficiency (WUEi); and CAN50 increased concentrations of N and P in the roots and stems of S. superba seedlings. Both UAN25 and UAN50 generally increased stem and leaf N concentrations and the leaf N/P ratio in A. quinquegona seedlings. UAN50 decreased whole-plant N-use efficiency of both species, and CAN50 reduced whole-plant N- and P-use efficiencies of S. superba seedlings. Overall, S. superba seedlings performed better than A. quinquegona seedlings in terms of height, biomass, Asat, and WUEi. Our results indicate that N deposition can profoundly influence the seedling growth of woody species, and that seedling height, basal diameter and biomass are more sensitive to the understory addition of N than to the canopy addition of N, i.e., the traditional use of understory addition of N to simulate atmospheric N deposition may overestimate the effects of N deposition on seedling performance.