Nitrogen-15 natural abundance in a montane cloud forest canopy as an indicator of nitrogen cycling and epiphyte nutrition.

Nutrients obtained by epiphytes may either be of atmospheric origin or from organic matter in the canopy, which decomposes to form canopy soil on large branches. We hypothesised that the N supply for epiphytes on small branches was lower, and a larger proportion provided by rainwater, than for epiphytes rooting in canopy soil. We tested this by measuring the N concentration and isotopic composition in terrestrial and canopy soil and in various canopy compartments of a Costa Rican cloud forest. In general, epiphytes on small branches without canopy soil had lower N foliar concentrations and δN signals than plants rooted in canopy soil, suggesting that the former receive a higher proportion of N directly from the rain. Epiphytes on small branches also had less negative δC values, indicating more frequent water stress. Epiphytes had lower δN values (-3.9±2.3‰) than ground-rooted trees (-1.1±1.6‰), and canopy soil had lower values (0.7±1.2‰) than terrestrial soil (3.8±0.7‰). Assuming that the isotopic effect of terrestrial and canopy soil organic matter formation is similar, our findings support earlier results showing that canopy soil is derived mainly from epiphytes, with only minor inputs from host tree matter. Thus, the epiphyte N cycle appears to be largely detached from the tree-soil cycle. Epiphylls on leaves of understorey shrubs had higher δN signals than cryptogams in the upper canopy, as a result of either N accumulation in throughfall or increased N fixation. The correlation between epiphyll and understorey host leaf δN suggests some exchange of N between epiphylls and host leaves. Differences between epiphyte groups also appear to be related to uptake of N through mycorrhizas or N fixation. Thus, the source and quantity of N supply is highly variable, depending on the systematic group and canopy position.