Nitrogen uptake and turnover in riparian woody vegetation.

The nutrient balance of streams and adjacent riparian ecosystems may be modified by the elimination of anadromous fish runs and perhaps by forest fertilization. To better understand nitrogen (N) dynamics within stream and riparian ecosystems we fertilized two streams and their adjacent riparian corridors in central Idaho. On each stream two nitrogen doses were applied to a swathe approximately 35 m wide centered on the stream. The fertilizer N was enriched in 15N to 18 per thousand. This enrichment is light relative to many previous labeling studies, yet sufficient to yield a traceable signal in riparian and stream biota. This paper reports pre-treatment differences in delta15N and the first-year N response to fertilizer within the riparian woody plant community. Future papers will describe the transfer of allochthonous litter N to the stream and its subsequent processing by stream biota. Pre-treatment delta15N differed between the two creeks (P=0.0002), possibly due to residual salmon nitrogen in one of the creeks. Pre-treatment delta15N of current-year needles was enriched compared to leaf litter, which was in turn enriched compared to needles aged 4 years and older. We conclude that fractionation due to retranslocation occurs in at least two phases. The first phase, which optimizes allocation of N in younger needle age classes, is distinctly different from the second, which conserves N prior to abscission. The delta15N difference between creeks was eliminated by the fertilization (P=0.42). In the two dominant conifer species, Abies lasiocarpa and Picea engelmannii, most fertilizer N was found in the current-year foliage; little was found in older needles and none was detected in litter (P=0.53). The only N-fixing shrub species, Alnus incana, took up only a small amount of fertilizer N [mean percent N derived from fertilizer (%Ndff) 5.0+/-1.6% (SE)]. Far more fertilizer N was taken up by other deciduous shrubs (mean %Ndff=33.9+/-4.5%). Fertilizer N made up 25% (+/-4.2%) of the N in deciduous shrub litter. These results demonstrate the feasibility of light labeling with 15N and the potential influence of riparian plant species composition on stream nutrient dynamics via allochthonous leaf litter inputs.