Nutrient cycling in streams involves some downstream transport before the cycle is completed. Thus, the path traveled by a nutrient atom in passing through the cycle can be visualized as a spiral. As an index of the spiralling process, we introduce spiralling length, defined as the average distance associated with one complete cycle of a nutrient atom. This index provides a measure of the utilization of nutrients relative to the available supply from upstream. Using 32P as a tracer, we estimated a spiralling length of 193 m for phosphorus in a small woodland stream.Key words: downstream transport, nutrient cycling, phosphorus, spiralling, stream

We conducted two experiments to determine the relative effects of herbivory and nutrients on an algal community in Walker Branch, a stream having effectively two trophic levels: primary producers and herbivorous snails. The first study (1989), performed in streamside channels, tested the effects of three factors: (l) stream water nitrogen (N), (2) phosphorus (P), and (3) snail grazing, on periphyton biomass, productivity, and community composition. The second study ( 1990), conducted in situ, tested the effects of snail grazing and nutrients (N + P). In the 1989 study, nutrients had positive effects, and herbivores had negative effects, on algal biomass (chlorophyll a, ash-free dry mass, total algal biovolume) and primary productivity (areaand chlorophyll-specific). Likewise, both nutrients and snail grazing exerted effects ( + and -, respectively) on biomass measured in the 1990 study (chlorophyll a, algal biovolume). Grazed communities were dominated by chlorophytes and cyanophytes, which were overgrown by diatoms when herbivores were removed. Algal species that were reduced most by herbivores were increased most by nutrient addition, and vice versa, suggesting a trade-off between resistance to herbivory and nutrientsaturated growth rates. Increases in algal biomass and productivity were slight with the addition of either N or P compared to responses observed when both nutrients were added together, suggesting that both nutrients were at growth-limiting levels. The greatest changes in periphyton structure or function were observed when both nutrients were added and simultaneously, grazers were removed, in contrast to lesser effects when nutrients were added under grazed conditions or grazers were removed at low nutrient levels, indicating dual control by both factors. Nutrient addition also positively affected snail growth in both experiments, indicating tight coupling between herbivore and algal growth (top-down effects) and that bottom-up factors that directly affected plant growth could also indirectly affect consumers belonging to higher trophic levels. Indices quantifying the direct effects of top-down factors relative to bottom-up factors (top-down index, TDI) and the importance of interactions between these factors (interaction coefficient, IC) were computed. These indices showed that the relative strength of top-down and bottom-up factors varied among biomass and productivity parameters and that top-down and bottom-up effects, alone, were less important than their combined effects.

The limiting role of phosphorus on leaf decomposition and primary producers was investigated in a second—order woodland stream in Tennessee by experimentally enriching, for 95 d, adjacent reaches with an average of 60 and 450 mg PO4—P/L, respectively, over upstream control levels of °4 mg/L. Red oak (Quercus rubra) leaf packs in the enriched sections lost mass 24% faster than control packs (P < .05). Nitrogen content of the enriched packs increased 60% more, and P content increased 83% more than the respective increases in the control packs (P < .05). Differences in mass loss and N and P levels between the low and high enrichments were not significant (P > .05). Respiration rates of subsampled leaf discs were significantly higher than control rates only at the high level of enrichment. The increased respiration rates in the low and high enrichments accounted for 10 and 34% of the increased mass loss in the respective enriched sections, suggesting that the enrichment also produced increases in mechanical breakdown through faster microbial conditioning, increases in macroinvertebrate feeding, or both. Effects of the enrichment on aufwuchs initially consisted of increased chlorophyll a levels, followed by increased aufwuchs biomass levels. Dense growth of filamentous algae, including some Oscillatoria, which may be a nitrogen fixer, developed immediately downstream of P inputs. In addition, Nostoc, a known nitrogen—fixing blue—green alga, sampled after the enrichment, was significantly more abundant in the enriched sections than the control (P < .05). Densities of the snail, Goniobasis clavaeformis, a grazer—shredder sampled after the enrichment, also were significantly greater in the enriched reaches, suggesting that the lack of a sustained response of chlorophyll a to the enrichment may have been a result of increased grazing on algal biomass. These findings indicate that nutrient limitation of detrital processing is a significant factor in natural streams. The apparent increases in densities of benthic macroinvertebrates in the enriched sections, along with reported relationships between detrital food richness and macroinvertebrate growth and survivorship, suggest that nutrient limitation in streams also has ramifications on higher tropic levels.

The term spiralling refers to the interdependent processes of cycling and downstream transport of nutrients in a stream ecosystem. To describe spiralling in Walker Branch, a first—order woodland stream in Tennessee, we released 32PO4 to the stream water and measured its uptake from the water and then followed its dynamics in coarse particulate organic matter (CPOM), fine particulate organic matter (FPOM), aufwuchs, grazers, shredders, collectors, net—spinning filter feeders, and predators over a 6—wk period. Rates of transfer among compartments and rates of downstream transport were estimated by fitting a partial differential equation model of the ecosystem to the data. With the resulting coefficients, the model was run to steady state to estimate standing stocks and fluxes of exchangeable phosphorus. Phosphorus moved downstream at an average velocity of 10.4 m/d, cycling once every 18.4 d. The average downstream distance associated with one cycle, defined as the spiralling length, was therefore 190 m (10.4 m/d ° d). Spiralling length, at steady state, is approximately the ratio of the total downstream flux of phosphorus per unit width of stream (720 mg°d—1°m—1) to the rate of P uptake from the water (3.90 mg°m—2.d—1). CPOM accounted for 60% of the uptake, FPOM for 35%, and aufwuchs for 5%. Turnover times of P in particulates ranged from 5.6 to 6.7, except for FPOM, which showed a second, slower turnover time of 99 d. Of the P uptake from water by particulates, 2.8% was transferred to consumers, while the remainder returned directly to the water. About 30% of the consumer uptake, in turn, was transferred to predators. The spiralling length was partitioned into: (1) an uptake length associated with transport in the water column (165 m), (2) a particulate turnover length associated with transport in FPOM and CPOM (25 m), and (3) a consumer turnover length associated with animal drift (0.05 m). FPOM transport accounted for 99% of the particulate turnover length. The small consumer turnover length reflected low consumer uptake of P from particulates and slow downstream drift velocity (0.013 m/d). In spite of the low rate of phosphorus uptake, the combined consumer—and—predator community accounted for 25% of the standing stock of exchangeable P in the stream. The retentiveness of this community is attributable both to the low drift rate and to a long turnover time (152 d) for P within the community.

No abstract