Temporal variation of nutrient uptake in streams may be large because nutrient uptake is driven by many factors that vary substantially over time. Although many studies have compared nutrient uptake among streams, the range and variation of nutrient uptake within streams is known only for a few streams and a few nutrients. We investigated the monthly variation of NH₄ , NO₃⁻, and PO₄³⁻ uptake in 2 New Zealand streams over 1 y. To measure uptake, each nutrient was added individually along with a conservative tracer (Cl⁻) into each stream on 3 successive days in each month. Ambient nutrient concentrations were low and nutrients were efficiently removed from the water column, with maximum uptake velocities (v_f) of 71, 12, and 11 mm/min for NH₄ , NO₃⁻, and PO₄³⁻, respectively. Nutrient uptake varied considerably during the year (CV = 37–109%), with shortest nutrient uptake lengths (S_w) and highest v_f generally in spring and summer months. The range of v_f occurring within the streams spanned 25 to 89% of the range of v_f among other streams. The range of uptake rates (U) within the streams was lower, accounting for 2 to 40% of the range among other streams. Variation in S_w was largely explained by changes in velocity and effective depth. Physical factors (temperature, transient storage) and chlorophyll a were generally poor predictors of v_f and U. There was little correlation in uptake among nutrients, suggesting different factors were responsible for uptake of each nutrient. Our results show that the range and variation of nutrient uptake within some streams can be large. Within-stream variation should be considered when comparing among streams and may be useful for understanding what factors drive nutrient uptake in streams.

Riparian zones can strongly influence the exchange of nutrients between streams and their watersheds. Most riparian studies have been done in mesic watersheds, which differ significantly from arid-land watersheds hydrologically. The goals of our work were to determine the strength and direction of hydrologic linkages between stream and riparian zone, and to estimate the extent of uptake of streamwater N by riparian trees in Sycamore Creek, a Sonoran Desert stream. Br⁻ and ¹⁵NH₄ were added simultaneously to the surface stream to trace water and N from stream to riparian zone. Br⁻ concentrations in riparian wells installed downstream of the release point increased during the addition, demonstrating a strong hydrologic linkage from stream to riparian zone. Percentage stream water in wells increased in a downstream direction, suggesting little or no input of water laterally from uplands or vertically from deep groundwater. Leaf and wood samples collected from willow trees downstream of the addition point became significantly labeled with ¹⁵N during the addition, indicating uptake of streamwater N. Other tree species did not become labeled, most likely because they were located farther from the stream channel than the willows. Results from our study provide evidence of strong hydrologic linkage between stream and riparian zone and suggest that N demand by riparian vegetation is a potentially significant sink for streamwater N.

The capacity of a 3^rd-order Ozark Plateau stream (Arkansas, USA) to take up (or remove) nutrient inputs from a rural wastewater treatment plant (WWTP) was examined using nutrient spiraling methods. Short-term nutrient additions often are used to assess nutrient uptake length, where an exponential decline in the concentration of the added nutrient reflects gross nutrient uptake. We applied this quantitative framework using WWTP effluent as a stream nutrient addition, and estimated net nutrient uptake length (S_net), mass transfer coefficient (v_f-net), and uptake rate (U_net) in Columbia Hollow, Arkansas. Water samples were collected at a reference site upstream of the WWTP input and at 6 sites downstream of the WWTP (0.3–2.7 km). Input from the WWTP significantly increased discharge, temperature, conductivity, soluble reactive P (SRP), and NH₄-N, and decreased pH and NO₃-N 0.3 km downstream from the point source. When P additions from the WWTP were low, stored SRP was released from the stream reach to maintain high water-column concentrations. Dissolved inorganic N was not retained in Columbia Hollow. Most or all of the NH₄-N added from the point source was converted to NO₃-N, resulting in net nitrification rates of 7 to 31 g NO₃-N m⁻² d⁻¹. The relationship between dilution-corrected concentrations and distance from the WWTP input indicated no significant nutrient retention, or that several stream kilometers were required before N and P were taken up. U_net typically was >7- to 10-fold higher and v_f-net estimates were 10- to 100-fold lower than values reported for undisturbed streams, indicating low relative nutrient demand. Rather than acting as a nutrient sink, Columbia Hollow appeared to be acting as a short-term storage zone for P and a transformer of N. Thus, the effect of this rural WWTP on the stream was profound, distorting N and P cycling in Columbia Hollow.

Dissolved organic C (DOC) is metabolically important in streams, but its contribution to ecosystem metabolism is not well known because it is a complex mixture of mostly unidentified molecules. The uptake of bioavailable DOC in White Clay Creek (WCC), a 3^rd-order stream in Pennsylvania, was estimated from the results of an experiment using ¹³C-labeled tree-tissue leachate and streambed sediments in recirculating mesocosms. The contribution of DOC in transport to stream metabolism was estimated from measurements of ¹³C-DOC uptake, ¹²C-DOC concentrations, and diel changes in dissolved O₂ in the mesocosms. Eighty percent (±5) of the DOC in the ¹³C-tree-tissue leachate was bioavailable and belonged to 1 of 2 distinct lability classes, readily and intermediately labile. These components made up 88% (±0.6) and 12% (±0.6), respectively, of the biodegradable DOC in the leachate. Uptake mass transfer coefficients for the readily and intermediately labile components were 55 (±24) μm/s and 2.6 (±0.13) μm/s, respectively. Based on our mesocosm measurements, DOC in transport could support 33 to 54% of the bacterial C demand and up to 51% of the community respiration in WCC. Extrapolation of our results to WCC indicates that readily and intermediately labile DOC similar in quality to the ¹³C-DOC would travel 175 and 3692 m downstream in WCC before being taken up by the sediments. These distances represent ∼7% and >150% of the length of the 3^rd-order reach. Our results suggest that readily labile DOC is an important energy source at the reach scale, whereas intermediately labile DOC serves as an energy subsidy from upstream to downstream reaches.

Caridina spp. (Atyidae) are widespread and abundant omnivorous shrimps in tropical Asian streams. Spatial and seasonal variation in C and N stable isotope signatures of 2 species (Caridina cantonensis and C. serrata) and their potential food sources (leaf litter, fine particulate organic matter [FPOM], and periphyton) were investigated in 2 shaded and 2 unshaded Hong Kong streams. The objectives were to identify food sources used by shrimps and to determine whether food use changed according to riparian shading, season, developmental stage, and sex. Isotopic signatures of different species of leaf litter were similar to each other and distinct from FPOM signatures in all streams, whereas periphyton δ¹³C and δ¹⁵N signatures were similar to FPOM signatures. FPOM and periphyton showed seasonal variation in δ¹³C and δ¹⁵N signatures in all streams, but signatures of leaf litter showed relatively minor variation. δ¹³C signatures of FPOM were 7 to 12‰ (shaded streams) and 5 to 7‰ (unshaded streams) higher than those of leaf litter during the dry season, and 5 to 8‰ (shaded streams) and 4 to 7‰ (unshaded streams) higher during the wet than the dry season. δ¹⁵N signatures of FPOM were almost 2× those of leaf litter in all streams and seasons. Periphyton was generally ≤6‰ more ¹³C-enriched and ≤4‰ ¹⁵N-depleted than FPOM. Seasonal variation in δ¹³C signatures of shrimps and their food sources were consistent among streams, and were less ¹⁵N-depleted during the dry season. Dual-isotope multiple-source mixing models indicated that FPOM and periphyton were the main foods of adult (male and female) and juvenile C. cantonensis and C. serrata. Leaf litter contributed <10% to the biomass of C. cantonensis in unshaded streams and 10 to 20% to the biomass of both species in shaded streams. Periphyton contributed >60% to the biomass of C. cantonensis in unshaded streams in the wet season, with the proportion slightly higher during the dry season. In shaded streams, periphyton contributed 35 to 60% to Caridina spp. biomass. Our results indicate that the Hong Kong atyids feed as omnivores, with herbivory as a primary feeding mode supplemented by collection of FPOM and limited direct consumption of allochthonous leaf litter.

We quantified abundance and biomass of the amphipod Gammarus chevreuxi Sexton from benthic cores, sampled monthly over 1 y in the upper reaches of Canal de Mira, the southern arm of the Ria de Aveiro, Portugal. Abundance and biomass showed no clear seasonal patterns but were associated with variation in salinity, dissolved oxygen, and chlorophyll a concentration. Maximum abundance and biomass occurred at relatively low temperature and high food availability, conditions that likely increased survival and/or reproduction. The population showed a semiannual, iteroparous life cycle. Mean life span was ∼6 mo, with overwintering individuals hatched in autumn showing higher longevity than individuals hatched in spring. Breeding was continuous, although juvenile recruitment peaked in early autumn, winter, and early spring. Sudden fluctuations in abundance and size structure of the population may have resulted from immigration during autumn and emigration during winter. Mean fecundity (8 embryos/brood) and intramarsupial loss (0–18%) were low compared with studies of other Gammarus species. Variation in fecundity and intramarsupial loss was mostly explained by size of incubating females. The volume of the embryos was highly variable but not significantly correlated with female size or other reproductive traits. The Hynes and the Morin– Bourassa methods yielded similar estimates of annual production (∼46 g AFDM m⁻² y⁻¹) and P/B̄ ratio (11/y).

We studied microdistributions, survival, and drift of larval hydropsychid caddisflies in the physical and/or chemical presence of 2 types of benthic predators with different foraging modes. Sculpins (Cottus bairdi, C. cognatus) are ambush predators, whereas perlid stoneflies (Acroneuria, Paragnetina) are stalkers, and stoneflies are an intermediate predator consumed by sculpins. Hydropsychid larvae in a northern Michigan stream were significantly more abundant in crevices than on flat surfaces on real cobbles in riffles with both predators. Larval colonization on experimental substrates (bricks) in 7 riffles was greater in crevices (grooves) than on flat surfaces, and the greatest differences in larval densities between these microhabitats occurred where predator densities were highest. Mean size (head capsule width) of larvae in both microhabitats was negatively related to densities of sculpins, but not stoneflies. A multifactorial experiment done in artificial stream channels revealed that caddisflies selected crevices over flat surfaces even in predator-free conditions. Sculpin and stonefly effects on hydropsychid survival were additive, suggesting a lack of multiple predator effects. Crevices provided a refuge from predators; however, stoneflies were more effective than large sculpins (>65 mm total length) at consuming hydropsychids in crevices. Like caddisflies, stoneflies predominantly occupied crevices, but stonefly crevice use and activity were not affected by sculpins, and no stoneflies were consumed by sculpins. Caddisfly drift was ∼3 to 4× greater in the physical presence of each predator than in predator-free channels. However, drift by caddisflies in the physical presence of both predators was lower than expected (i.e., nonadditive), suggesting it is less risky to remain in retreats (i.e., immovable cases) than to drift when both predators are present. Elevated chemical cues of either predator did not trigger drift responses by these sedentary prey. Crevice use and drift appear to be key mechanisms enabling larval hydropsychid caddisflies to coexist with a multispecies complex of predators.

Crayfish can act as keystone species and ecosystem engineers in small streams, but their effects in large rivers are not well known. Two species of crayfish, Orconectes cristavarius and Cambarus chasmodactylus, coexist in the South Fork of the New River in western North Carolina. We used gut-content analyses and an enclosure–exclosure experiment to investigate the influence of both species of crayfish on sediment accumulation and benthic invertebrates. Crayfish guts contained mostly sediment and vegetative detritus. However, C. chasmodactylus guts contained significantly more detritus and animal matter than O. cristavarius guts, and O. cristavarius guts contained significantly more sediment than C. chasmodactylus guts. In the field experiment, sediment volume was lower in open baskets and cage controls, which were exposed to crayfish and benthic-feeding fish, than in enclosures containing only crayfish. Sediment volume was highest in fish/crayfish exclosures. Despite their effect on sediment accumulation, crayfish did not significantly affect the density of any invertebrate taxon. No clear relationship was found between chironomid density and enclosure–exclosure treatment, but chironomid density was positively correlated with sediment volume. Damselflies (Calopteryx maculata) tended to be more abundant in crayfish enclosures than in open treatments and cage controls, and cyclopoid copepods tended to be more abundant in O. cristavarius enclosures than C. chasmodactylus enclosures. The significant differences in the diets of the 2 species of crayfish were not associated with differences in their effects on invertebrates. Our results suggest that these 2 species of crayfish may be functionally redundant in this community, despite differences in diet. Furthermore, the lack of pronounced crayfish effects on invertebrate taxa suggests that the effects of crayfish may not be as strong in large rivers as in small streams.

Invasive species can have large effects on freshwater communities and ecosystems. Potamopyrgus antipodarum, a snail indigenous to New Zealand, has recently colonized North America. We documented the distribution and density of P. antipodarum in the major tributaries of the Madison River in the Greater Yellowstone Ecosystem, examined associations between P. antipodarum and native macroinvertebrates, and experimentally examined how P. antipodarum affected colonization of substrates by other macroinvertebrates. In 1997 and 1998, we sampled benthic macroinvertebrates in the Madison, Firehole, and Gibbon Rivers and Nez Perce Creek in 3 seasons. In 1999, we examined colonization of P. antipodarum and other macroinvertebrates in a field experiment using slate tiles in the Madison River. In September 1997, we found P. antipodarum at 23 of 32 locations. Where they were present, P. antipodarum densities ranged from 22 to 299,000 ind./m². Densities of P. antipodarum declined between September 1997 and March 1998 and increased between March and July 1998. The composition and structure of native macroinvertebrate assemblages covaried with P. antipodarum densities. Whether the differences in macroinvertebrate communities among sites were related to differences in physical characteristics of streams or to interactions with P. antipodarum was unclear. However, in the field experiment, high densities of P. antipodarum were associated with low colonization of other macroinvertebrates; thus, negative interactions between native macroinvertebrates and P. antipodarum may have the potential to influence the large-scale distribution of other macroinvertebrates.

We studied the reproduction and shell size structure of the endangered dwarf wedgemussel (Alasmidonta heterodon) in the Mill River, Massachusetts, at sites of different mussel abundance, and assessed the dispersal capacity of its host fish, the tessellated darter (Etheostoma olmstedi). We quantified 4 phases of reproduction at 5 study sites of contrasting adult mussel abundance (3 low-abundance sites, 1 intermediate, 1 high), including gravidity, glochidial release, host infection rate, and juvenile recruitment. The ratio of gravid to nongravid individuals was higher at the high-abundance site, and comparatively lower at the intermediate- and low-abundance sites. Glochidial release, infection rates on host darters, and juvenile recruitment were directly proportional to mussel abundance. The highest glochidia density (0.12/m³) occurred at the high-abundance site. Mean infection rates during early May to late June ranged from 31% at the high-abundance site to 8% at the intermediate- and 0 to 2% at the 3 low-abundance sites. The high-abundance site showed highest level of juvenile recruitment. Low-abundance sites showed narrow ranges of size classes, probably indicating few year classes. Movement by marked darters was minimal during glochidial release, with 94% of marked darters remaining in locations where they were originally marked. Reproduction by A. heterodon in the Mill River depends on mussel patch density, and mussel dispersal by the host fish may be limited to the immediate vicinity of the infection site. Such a low dispersal capacity may lead to a patchy mussel distribution and may hinder colonization and recovery of this species.

Benthic macroinvertebrate communities were quantified at natural cobble and artificial reef sites in Lake Ontario in 1983 (7 y pre-Dreissena invasion) and in 1991 to 1992 and 1999 to 2000 (1–2 and 9–10 y post-Dreissena invasion, respectively). Overall, the cobble community had higher diversity and abundance than the reef community. In both communities, diversity and abundance of non-Dreissena taxa rose sharply between 1983 and 1991 to 1992 and declined to 1983 levels by 1999 to 2000. However, taxonomic composition (excluding Dreissena and the recent invader Echinogammarus ischnus) was consistent across study years. Between 1983 and 1999 to 2000, Stagnicola catascopium and Physella spp. increased in abundance, several taxa (Musculium partumeium, Bithynia tentaculata, Elimia livescens, Trichoptera, and Chironomidae) decreased in abundance, and changes in the abundance of many taxa were correlated with Dreissena abundance. Dreissena continued to make up >40% of total macroinvertebrate abundance at the cobble site and >60% at the reef site in 1999 to 2000; however, numbers dropped sharply and the size of individual mussels increased as D. bugensis largely replaced D. polymorpha between 1991 to 1992 and 1999 to 2000. Dreissena biomass dropped sharply at the reef site between 1991 to 1992 and 1999 to 2000, but not at the cobble site. The exotic amphipod E. ischnus was abundant in 1999 to 2000, but any effect on the abundance of the related amphipod Gammarus fasciatus was unclear. At the time of this study, the round goby Neogobius melanostomus was not present. We conclude that the transition from D. polymorpha to D. bugensis and processes associated with the ongoing oligotrophication of Lake Ontario are responsible for reduced density of larger-bodied Dreissena in the nearshore region, and that changes in the Dreissena population are largely responsible for changes in the non-Dreissena benthic macroinvertebrate communities.

We assessed the effects of fish predation on zebra mussels (Dreissena polymorpha) in Navigation Pool 8 of the Upper Mississippi River from 13 May to 5 October, 1998. Concrete-block samplers were deployed at 18 randomly chosen sites in the main-channel border, with 6 sites in the upper, middle, and lower segments of the pool. Two blocks, 1 of which was enclosed in a cage to exclude large predatory fishes, were deployed at each site. After 145 d, blocks were retrieved from 12 of the 18 sites, and zebra mussels were found on all blocks. Densities of zebra mussels were higher on caged blocks than uncaged blocks, and the magnitudes of the differences varied spatially. Mean mussel densities on uncaged blocks were reduced by 66%, 86%, and 20% compared to caged blocks in the upper, middle, and lower pool segments, respectively, over the 145-d interval. Mean daily instantaneous zebra mussel mortality rates from large predators ranged from 0.0016 to 0.0138. Similarly, biomass of zebra mussels was higher on caged than uncaged blocks. Mean mussel biomass on uncaged blocks was reduced by 64% pool-wide, relative to biomass on caged blocks. Zebra mussels were consumed by at least 6 fish taxa including redhorse suckers (Moxostoma spp.), common carp (Cyprinus carpio), bluegill (Lepomis macrochirus), quillback carpsucker (Carpiodes cyprinus), flathead catfish (Pylodictis olivaris), and freshwater drum (Aplodinotus grunniens). Fish predation had an important moderating effect on zebra mussel demography in Pool 8.

Predictive models such as River InVertebrate Prediction And Classification System (RIVPACS) and AUStralian RIVer Assessment System (AUSRIVAS) model the natural variation across geographic regions in the occurrences of macroinvertebrate taxa in data from streams that are in reference condition, i.e., minimally altered by human-caused stress. The models predict the expected number of these taxa at any stream site, assuming that site also is in reference condition. A significant difference between the ratio of observed (O) and expected (E) taxa (O/E) and 1.0 indicates that the site is not in reference condition. The standard deviation (SD) of O/E values estimated for a set of reference sites is a measure of predictive-model precision, with a small SD indicating that the model accounts for much of the variability in E that is associated with natural factors such as stream size and elevation. We propose a null model for E that assumes fixed occurrence probabilities for individual taxa across reference sites. The null model explains none of the variability in E caused by natural factors, so the SD of its O/E predictions is the upper limit attainable by any predictive model. We also derive a theoretical lower limit for SD of O/E that is caused only by replicate-sampling variation among predictions from a perfect model. Together, the null-model and replicate-sampling SDs estimate the minimum and maximum precision, respectively, attainable by any predictive model for a given set of reference-site data. A predictive model built from data at 86 reference sites in the Mid-Atlantic Highlands region, USA, had SD = 0.18 for O/E across those sites, while the corresponding null model had SD = 0.20, indicating relatively little gain from the predictive-model effort. In contrast, a model built from 209 sites in North Carolina, USA, had predictive- and null-model SDs of 0.13 and 0.28, respectively, indicating that the North Carolina predictive model had relatively high gain in precision over the null model. Replicate-sampling SDs of O/E for the Mid-Atlantic and North Carolina data were 0.09 and 0.11, respectively, suggesting that the North Carolina predictive model had little room for further improvement, in contrast to the Mid-Atlantic model. The precisions of null-model estimates were lower than those of predictive models, so null models somewhat underestimated the percentages of 447 and 1773 test assemblages from the Mid-Atlantic region and North Carolina, respectively, that differed significantly from reference conditions. The estimates illustrate how a simple and easily built null model provides a lower bound for the prevalence of impaired streams within a region.

Dam removal is often proposed as way to restore ecological integrity to rivers and streams, but ecological responses to dam removals are poorly understood, especially for downstream benthic communities. We examined the responses of benthic macroinvertebrate and algal assemblages in downstream reaches to the removal of a small, run-of-river dam on Manatawny Creek, Pennsylvania. Benthic macroinvertebrates, algae, and habitat characteristics were monitored upstream and downstream of the dam for 4 mo before removal, 3 mo after partial removal (i.e., when the impoundment was largely eliminated but sediment remained trapped behind the remaining structure), and 12 mo after complete dam removal. Macroinvertebrate density, algal biomass, and diatom species richness declined significantly downstream of the dam following complete dam removal, but overall assemblage structure (as indicated by Nonmetric Multidimensional Scaling ordinations) downstream remained similar to upstream control sites throughout the study for both invertebrates and diatoms. Downstream impacts occurred only after the dam structure had been completely removed and sediments had been transported downstream from the former impoundment by high flows. Biotic impacts persisted for the duration of the study (12 mo after complete removal). Our results and other studies of dam removal suggest that downstream sedimentation following dam removal can reduce densities of macroinvertebrates and benthic algae and may reduce benthic diversity, but for small dams such impacts may be relatively minor and will usually be temporary.

Aquatic insect communities were examined in 2 streams at different selection logging intensities in headwater catchments of a northern hardwood forest. Insect communities of these streams were compared to those of a nearby reference stream (no harvesting) over a 2-y pre- and 3-y post-logging period. The experimental catchments were logged by a mechanical harvester and cable skidders, one at a low-intensity (29% basal area removal) and the other at a moderate-intensity (42% basal area removal) harvesting rate. There were no riparian reserves or buffer zones, but logging was conducted in compliance with a riparian code of practice (3-m setback from stream edges) and other best management practices. Changes in community structure, community metrics, or relative abundance of discriminatory taxa attributable to logging impacts were not detected at the low-intensity site. Some deviations from reference and pre-logging trends in community structure, multivariate dispersion, and population levels of discriminatory taxa were detected at the moderate-intensity site after the logging. These deviations were mainly driven by small, but usually significant, increases in abundance of 5 gatherer taxa. The increases in abundance of gatherer taxa appeared to be a response to a significant increase (∼2.5×) in streambed deposition of fine particulate organic material at that site. However, the shifts in community structure and changes in abundance of these taxa at the moderate-intensity site were not larger than some natural changes in abundance among other taxa at the reference site over the 5-y study. The increases in abundance of some taxa at the moderate-intensity site may indicate a logging impact, but the changes were small and there were no indications of reciprocal declines among other taxa. It appears that selection logging at up to 42% basal area removal in compliance with the riparian code of practice and other good management practices largely mitigated harmful alterations to stream habitat and insect communities in these northern hardwood forest catchments.