Macroinvertebrate species composition data are often expressed as proportional abundances when assessing water-quality conditions or responses to disturbance. Proportional abundances represent the probability of belonging to one of many mutually exclusive and exhaustive groups (taxa). Proportional abundances have some unique properties that must be considered when analyzing these data: 1) the probabilities of group membership must sum to 1 and 2) a change in any 1 group affects all other groups. We used multinomial regressions to analyze changes in proportional abundances along gradients of urbanization in 9 metropolitan areas across the USA. Multinomial regression can be used to address multiple nonlinear responses simultaneously, whereas simple linear regressions must be used to analyze linear or polynomial responses of each group independently. We established that: 1) abundance ratios of tolerant and moderately tolerant groups responded consistently (3–5% increase in the ratios for every 1% increase in developed land cover in the watershed) across the urban gradient, 2) functional groups did not change significantly, and 3) ratios based on assemblage metrics were better indicators of environmental disturbance than ratios based on individual taxa. Multinomial regression, with its flexible model form, can capture patterns of species succession along a resource or stressor gradient. Our results also demonstrate that users of multinomial regression may encounter numerical problems with rare taxa, especially when these taxa have a complete separation along the gradient. Consequently, multinomial regressions are more suitable for analyzing aggregations of taxa or taxon traits.

Freshwater mussels (Unionidae) are among the most endangered groups of organisms in the world, and their conservation and recovery are priorities throughout North America, especially the southeastern USA. We used a ribonucleic acid sequencing (RNA-seq)-based approach to develop molecular resources for Villosa lienosa, the little spectaclecase. We sequenced barcoded samples (Illumina HiSeq 2000 platform) and assembled (Trans-ABySS) 778,234 contigs (average length  =  707.5 base pairs [bp]) from 162 million filtered reads. We identified 23,742 unigene hits against the National Center for Biotechnology Information nonredundant database and 36,582 microsatellites with sufficient flanking sequence for primer design. Microsatellite validation indicated a 36% polymorphic rate (16/44 tested markers) in the tested population (26 individuals; mean  =  5 alleles/marker). Analysis of differentially expressed genes between heat-stressed and untreated controls enabled us to identify 604 genes involved in stress-response pathways. Real-time polymerase chain reaction validation of gene-expression results using individual samples confirmed RNA-seq patterns (r  =  0.847, p < 0.001). RNA-seq is a powerful tool for rapid development of molecular resources in nonmodel species, and our study is the first large-scale transcriptome project in freshwater mussels. The validated microsatellite set and stress-associated genes are being used in parentage analysis and health-assessment surveys to support mussel conservation.

One of the primary goals of biological assessment is to assess whether contaminants or other stressors limit the ecological potential of running waters. It is important to interpret responses to contaminants relative to other environmental factors, but necessity or convenience limit quantification of all factors that influence ecological potential. In these situations, the concept of limiting factors is useful for data interpretation. We used quantile regression to measure risks to aquatic life exposed to metals by including all regression quantiles (τ  =  0.05–0.95, by increments of 0.05), not just the upper limit of density (e.g., 90^th quantile). We measured population densities (individuals/0.1 m²) of 2 mayflies (Rhithrogena spp., Drunella spp.) and a caddisfly (Arctopsyche grandis), aqueous metal mixtures (Cd, Cu, Zn), and other limiting factors (basin area, site elevation, discharge, temperature) at 125 streams in Colorado. We used a model selection procedure to test which factor was most limiting to density. Arctopsyche grandis was limited by other factors, whereas metals limited most quantiles of density for the 2 mayflies. Metals reduced mayfly densities most at sites where other factors were not limiting. Where other factors were limiting, low mayfly densities were observed despite metal concentrations. Metals affected mayfly densities most at quantiles above the mean and not just at the upper limit of density. Risk models developed from quantile regression showed that mayfly densities observed at background metal concentrations are improbable when metal mixtures are at US Environmental Protection Agency criterion continuous concentrations. We conclude that metals limit potential density, not realized average density. The most obvious effects on mayfly populations were at upper quantiles and not mean density. Therefore, we suggest that policy developed from mean-based measures of effects may not be as useful as policy based on the concept of limiting factors.

Detritus processing is driven by a complex interplay between macroinvertebrate and microbial activities. Bioturbation/feeding activities of invertebrates in sediments are known to influence decomposition rates. However, direct effects of invertebrates on bacterial communities and detritus processing remain ill-defined, mainly because identifying interactions between invertebrates and sediments is methodologically challenging. We incubated 5 macroinvertebrate species with various bioturbation/feeding traits separately in sediment-filled microcosms inoculated with bacterial communities for 5 d. At the end of the experiment, we assessed: 1) detritus processing (mass loss on ignition [LOI] and dissolved organic C accumulation in the overlying water [absorbance at 280 nm]), 2) bacterial community structure (intergenic spacer analysis [RISA]) and bacterial activity (electron transport system activity [ETSA]), and 3) development of redox potential (Eh) over time (with permanently installed microelectrodes). Invertebrates enhanced bacterial activity and detritus processing, and the magnitude depended on bioturbation/feeding traits. Bacterial community structure differed significantly between microcosms with burrowing invertebrates and microcosms with sediment-dwelling invertebrates. Eh profiles were similar among microcosms with invertebrates with similar bioturbation/feeding traits, but differed among microcosms with invertebrates with dissimilar bioturbation/feeding traits. Our results suggest that bioturbation by aquatic invertebrates mediates detritus processing, Eh dynamics, and structure of the microbial community. These findings highlight the significance of bioturbation and show the utility of spatiotemporal Eh dynamics as footprints reflecting functioning of benthic detrital food webs.

Total benthic invertebrate biomass in shallow offshore waters (depths 10–15 m) increased 17-fold on average following the invasion of dreissenid mussels and implementation of P controls in Lake Simcoe, Ontario, despite a 50% overall decrease in total invertebrate density during the same time period. The increase in total invertebrate biomass at 10- and 15-m depths was primarily dreissenid biomass. Patterns of biomass with depth and dreissenid invasion for individual taxa were typically similar to those for density. Biomass of chironomid and nondreissenid bivalves declined at shallow (10–15 m) sites but increased at the deepest (20–30 m) sites, whereas biomass of oligochaetes declined at all sites. Density and biomass of Amphipoda, Isopoda, and Gastropoda increased at depths ≥10 m, and these taxa were found more frequently in deeper sites after dreissenid invasion than before. Increased habitation of deeper sites by these taxa may be mediated by increased habitat complexity caused by deposition of dreissenid shells, nutrient enrichment of substrate occupied by dreissenids, and improvements in hypolimnetic dissolved O₂ and water clarity observed during dreissenid invasion. Increased length and biomass of profundal chironomids after dreissenid invasion may be the result of taxonomic changes in the chironomid community, which in turn appear to be closely associated with improvements in deep-water O₂ concentrations. Changes in the benthic community described here probably have important consequences for the degree of coupling between nearshore and offshore habitats in Lake Simcoe.

We assessed the relative magnitude of various factors (year, preservation method, continent, investigator, and taxonomic level) affecting prediction of invertebrate dry mass (DM) in light of the variability of assessments of invertebrate density. We developed 34 length (L)–DM relationships for Oligochaeta and 17 freshwater invertebrate families belonging to Mollusca, Crustacea, and Insecta. Comparison of our predicted DM for reference-size individuals with values from 120 other published equations revealed that 31% of predicted values were within our 95% CI and 73% were within a 2× DM range (i.e., between 0.5× and 2×). Interannual differences in exponent (slope) or scale factor (intercept) of L–DM relationships were detected for 6 of the 7 taxa investigated, but represented only 3% of total variance in predicted DM. Similarly, preservation methods and measured body dimension each accounted for a small (0–3%) fraction of total variance. Variation among investigators (12–50%) and continents (1–17%) were more important and might have reflected methodological or regional and latitudinal differences. Increasingly precise taxonomical levels explained progressively lower proportions of the total variance, a result indicating that family or a more precise taxonomic level provided a robust estimate of most invertebrate DM even if the equations were derived from other sites. However, overall variability induced by L–DM relationships was smaller than variability in total invertebrate density among replicate samples (coefficient of variation [CV]  =  19–97%), a result indicating that more effort should be devoted to improving the accuracy of invertebrate density estimates than to developing site-specific L–DM relationships to assess benthic biomass in freshwater.

I collected stoneflies (Plecoptera∶Perlidae) across gradients of elevation and stream size within a 274-km² catchment in Great Smoky Mountains National Park in 1977–1978. I repeated collections of 2 focal species in 2006. Regional climate warmed ∼0.72°C during the interval. Models of stonefly responses to temperature change predicted uphill shifts of 100 to 140 m. One abundant species, Acroneuria abnormis, shifted uphill 60 to 250 m depending on the measure of range shift. Uphill shift was strongly supported by logistic regression and information criteria. The less common Eccoptura xanthenes shifted −61 to 105 m based on several metrics, and no regression model was supported. Data for E. xanthenes, confounded by landscape change and detectability issues, were inadequate to support any conclusion about range shift. Stream acidity has shifted downslope in recent years, counter to the observed uphill shift by A. abnormis, a pattern that supports increasing temperature as the determinant of range limits for this species. Rate of uphill shift by A. abnormis (median elevation, 24 m/decade) was similar to changes by plants and animals in other places.

Freshwater mussels use an array of strategies to transfer their parasitic larvae (glochidia) to fish hosts. We examined the effects of temperature, photoperiod, and female gravidity on mantle lure display and conglutinate release by Ligumia subrostrata (Say, 1831) in 2 laboratory experiments. In the 1^st experiment, we examined the use of these strategies in 4 temperature treatments (5, 15, 25, 35°C) and 3 photoperiods (10∶14, 12∶12, and 14∶10 h light∶ dark). In the 2^nd experiment, we observed infection strategies under ambient conditions with flow-through pond water. Water temperature appeared to be the primary cue governing use of these strategies. Lure display occurred over a protracted period, but the highest display frequency occurred between ∼11 and 20°C. Lure display declined rapidly above this range and ceased altogether >28°C. Release of conglutinates increased coincident with the decrease in lure display but, at ambient temperatures, occurred over a protracted period similar in duration to lure display. Females that were not gravid at the beginning of the experiment did not display lures, and gravid females whose gills were flushed of glochidia displayed for only a short period during which frequency of display was much lower than gravid individuals. Ligumia subrostrata exhibits a temperature-mediated switch between alternate host strategies. We present evidence that lure display is a primary strategy for host infection and conglutinate release is a secondary, bet-hedging strategy to reduce wastage of glochidia that ultimately must be cleared from the gills at the end of the reproductive season.

The importance of understanding the mechanisms underlying insecticide disturbances in natural systems is growing because of increasing global insecticide use. Despite the prevalence of pesticides and the vulnerability of aquatic systems to insecticides, little is understood about the effect of lower concentrations of insecticides (<1 ppm) on aquatic community interactions. Whether insecticide effects are generalizable across different aquatic assemblages and trophic levels also is unclear. Furthermore, few investigators have examined the indirect consequences of insecticides beyond the primary consumer level. We examined how a single application of malathion at 3 concentrations (0, 6, or 40 µg/L) and the presence or absence of zooplankton predators (larval salamanders) affected aquatic communities composed of zooplankton, phytoplankton, periphyton, and 2 geographically distinct amphibian assemblages from Oregon and Pennsylvania. At these concentrations, malathion was directly lethal to certain zooplankton species, causing a shift from cladoceran-dominated assemblages to copepod-dominated assemblages. The decrease in cladoceran abundance released top-down pressure on phytoplankton, allowing an increase in phytoplankton abundance. The increase in phytoplankton was associated with a decrease in periphyton (the major food source of anurans) because of competition between phytoplankton and periphyton. We did not find direct mortality in anurans or salamanders, but the insecticide-mediated reduction in zooplankton indirectly caused a decrease in larval salamander mass. In contrast, anurans exposed to malathion were heavier at metamorphosis. Overall, these results demonstrate that low concentrations of insecticides have indirect consequences on nontarget members of the community across multiple trophic levels, and the indirect insecticide-mediated effects are generalizable across 2 geographically distinct amphibian assemblages.

Intersite differences in benthic communities detected by tests of null hypotheses are used routinely to infer effects of habitat perturbation. The statistical outputs from these tests are often treated as binary results (presence or absence of detectable effects), and the sizes and potential biological importance of detected differences or effects are frequently ignored. This situation can be remedied by measuring standardized effect sizes of detected differences. To demonstrate the benefits of standardized effect sizes, we compared benthic communities in streams draining forested and perturbed catchments based on kick-net samples, and samples from bricks and Hester–Dendy multiplate samplers. We complemented null hypothesis testing by calculating standardized effect sizes (Cohen's d) and their confidence intervals (CIs) to rank 14 benthic metrics and the 3 sampling techniques. Despite having higher variance than metrics from brick or Hester–Dendy samplers, metrics from kick-net samples better separated sites than did metrics from bricks or Hester–Dendy samples. Metrics from brick samples separated sites more often and by a larger number of standard deviations than did metrics from Hester–Dendy samplers. Metrics that included mayfly abundance or richness produced the largest d-values, particularly when calculated from kick-net samples. Metric rankings were inconsistent among techniques. Successional changes over the 30-d study were subtle or absent, but generally consistent among sampling techniques. Differences detected with few replicate kick-net samples were consistent in direction but smaller than differences detected with more replicates. In some cases, differences with large d-values were not detectable with small sample sizes and standard null-hypothesis-testing approaches. These differences were consistently confirmed by addition of replicates. d-values and their CIs add value to data sets, particularly given the small number of replicates common in labor-intensive ecological studies. This approach expresses biological differences in a common currency that can be compared across studies regardless of units of measurement, scale, or technique.

Anthropogenic nutrient enrichment increases the supply ratio of N and P to aquatic ecosystems and can affect the identity of the limiting nutrient. Here we focus on how stream communities change along gradients of N and P supply and stream catchment landuse intensity. We used a survey approach in 41 southern New Zealand tributaries to investigate how much changes in water N and P concentrations are reflected in periphyton C∶nutrient ratios (C∶N or C∶P) and how much food quality (high food quality corresponds to low periphyton C∶nutrient) is reflected in the abundance and taxonomic richness of benthic invertebrate primary and secondary consumers. We measured streamwater nutrient state, periphyton nutrient ratios, biomass (as chlorophyll a in µg/cm²), algal taxon richness, and macroinvertebrate abundance, taxonomic composition, and richness. We also estimated stream habitat and catchment characteristics, such as current velocity, shading, substrate, geology, and landuse intensity. We calculated the Akaike information criterion (AIC) for each possible multiple linear regression model to select the best predictive models for each response variable. C∶nutrient ratios were more strongly negatively related to water-column N than P availability. Neither N nor P availability covaried with periphyton biomass. Lower periphyton C∶N partly explained higher grazer, but not predator, abundance. Increased % runoff from pasture and periphyton N∶P co-occurred with a decrease in invertebrate taxon richness. For example, a 4× increase in periphyton N∶P was related to the loss of ∼½ of invertebrate species, but with high uncertainty (R²  =  0.13). We conclude landuse intensity affects these southern New Zealand streams, and these effects are mediated by agricultural N runoff into streams (among other factors). Further shifts toward high-intensity farming within stream catchments may lead to losses of benthic species at all trophic levels.

The Rio Grande supplies water for endangered biota and a burgeoning human population, but little is known about sources and cycling of nutrients through the food web of this dryland river ecosystem. We described aquatic foodweb structure at 4 sites in the Rio Grande (New Mexico, USA) based on analysis of stable isotopes of C and N as tracers of sources of primary production and trophic positions of consumers, respectively. δ¹³C values obtained from primary producers (e.g., benthic algae and vascular plants) and consumers (macroinvertebrates and fishes) indicated that autochthonous (instream) production predominated as an energy source for fishes, but not for macroinvertebrates. Statistical comparisons of isotopic niche space among functional groups suggested that macroinvertebrates obtained dietary C from both instream and riparian (terrestrial) sources and were more broadly dispersed in isotopic space than were fishes. Macroinvertebrates that fed on riparian sources were less likely to be consumed by fishes. Baseline isotopic ratios changed in nonlinear fashion across sites, as did isotopic values of local consumers, and maximum food-chain lengths. The Rio Grande food web appears to be controlled predominantly at the reach scale by point-source effects related to inflow from tributaries, dams, and wastewater return.

Terrestrial invertebrates (TIs) provide an important trophic subsidy for many stream fishes. However, more information is needed regarding the degree to which different species rely on this subsidy and the potential consequences of altering subsidy levels. Such information is important for understanding foodweb dynamics and predicting patterns of community change as subsidy levels vary among habitats and over time. I manipulated TI inputs to experimental streams and tested for effects on diet and body condition of fishes from 3 different trophic functional groups. Study species included a TI specialist (blackstripe topminnow Fundulus notatus), a benthic invertebrate specialist (orangethroat darter Etheostoma spectabile), and an invertebrate generalist (bigeye shiner Notropis boops). Both F. notatus and N. boops changed diets under TI reduction, but only F. notatus body condition decreased, probably because of its lesser ability to use autochthonous resources. Notropis boops body condition was unchanged by TI reduction. Unlike F. notatus, N. boops was able to incorporate autochthonous C into body tissues, as indicated by muscle δ¹³C values. Etheostoma spectabile diet and body condition were unaffected by TI manipulation. These results suggest that stream fishes in functionally diverse assemblages are not affected equally by TI alteration, and that effects may be predictable based on trophic functional group classification. Such information is useful in predicting patterns of assemblage change with terrestrial subsidy reduction, as often occurs with anthropogenic landscape alteration.

Headwater streams draining row-crop agriculture receive allochthonous inputs of maize detritus and grasses, but organic matter (OM) processing is not well studied in agricultural streams. Agricultural streams in the midwestern USA have incised, trapezoidal channels that retain less particulate OM than forested streams. The 2-stage ditch is a restoration strategy in which small floodplains are constructed and connected to stream channels to increase channel stability and decrease erosion. Microbial decomposition may be higher on restored floodplains because water residence times are longer than on the steep banks of trapezoidal streams. We examined decomposition of maize leaves (Zea mays), native rice cutgrass (Leersia oryzoides), and invasive reed canary grass (Phalaris arundinacea) in 4 restored streams. We measured breakdown rates in the main channel of upstream control reaches (incised, trapezoidal channel), the main channel of downstream treatment reaches (restored with constructed floodplains), steep control banks, and treatment floodplains. OM decomposed faster in channels than on banks and floodplains, and maize decomposed faster (stream k  =  0.0160/d, riparian k  =  0.0040/d) than rice cutgrass (stream k  =  0.0065/d, riparian k  =  0.0018/d) and reed canary grass (stream k  =  0.0036/d, riparian k  =  0.0014/d) probably because lignin and N content differed. Breakdown rates varied among streams because of differences in shredder density (primarily Isopoda: Lirceus and Caecidotea) and water temperature. Floodplain restoration did not affect breakdown rates. Floodplains of 3 streams were inundated longer than steep banks in upstream control reaches, but breakdown rate and inundation duration were not related. However, OM must be retained within the stream to be available for decomposition. Thus, the floodplains may promote the retention of OM, and ultimately, incorporation of maize and grasses into headwater-stream food webs.

Dryland rivers are globally widespread and regionally important, and understanding their ecology is critical for sustaining ecosystem processes and biodiversity. The dry tropics are characterized by episodic summer rainfall. Most of annual river flow occurs in a short period of time, after which rivers typically contract into a series of waterholes of varying permanence and hydrologic connectivity. We investigated how seasonal environmental factors affected macroinvertebrate assemblage composition in dry-tropics rivers at the river, site, and habitat scales. We assessed biophysical characteristics, including water physicochemistry, riparian-zone condition, and macroinvertebrate assemblage composition at 15 sites on 4 unimpounded rivers in the Burdekin catchment, north Queensland, Australia. We used permutational analysis of variance and multiresponse permutation procedures to identify differences between a priori groups, and illustrated the results with principle components analysis (biophysical data) and nonmetric multidimensional scaling (invertebrate assemblage data). Biophysical variables were spatially and temporally variable, and all rivers and most sites differed significantly. Macroinvertebrate assemblages also differed significantly among rivers, sites within rivers, habitats, and seasons. Assemblages from the same habitat but different sites were more similar than assemblages from different habitats within the same site. Thus, the dynamic environment of dryland river systems drives naturally dynamic and variable macroinvertebrate assemblages across a range of scales. The differences among sites from the same river were of particular importance from a monitoring standpoint because they indicated that extrapolating whole-river condition from a few sampled sites would be difficult.

Positive correlations between diversity and stability have been reported for a number of ecosystems and are thought to be caused by a stabilizing effect of differential species' responses to environmental perturbation. Empirical field studies in which investigators tested for diversity–stability relationships are lacking for some taxonomic groups and typically have not included tests of the importance of other potential correlates of diversity or assemblage structure. We sampled stream fish assemblages and associated habitat variables at 36 sites over a 10-y period. Quantitative and qualitative measures of stability were correlated with fish diversity at sites. Fish assemblage composition was correlated with a variety of habitat variables, and diversity was correlated with stream size. We used Akaike's Information Criterion to select the models that best predicted qualitative and quantitative stability. Candidate models contained variables describing diversity, stream size, time between samples, and change in habitat variables over time. Models that included diversity and time between samples were the best predictors of stability. Our results support the existence of diversity–stability relationships, and we showed that other predictors of diversity or habitat change were generally poor predictors of stability.

Near-shore sediments support the most productive and diverse biological communities in lakes and are the sites of intense biogeochemical activity. Understanding what drives temperature and short-term temperature variability in near-shore sediments is important because benthic organisms and processes are temperature-dependent. I compared sediment and near-bottom water temperatures throughout the growing season (May–September) at 5 shallow near-shore sites in Lake Opeongo (Ontario, Canada). These sampling sites were more or less exposed to hydrodynamic forces and had different sediment characteristics but were not influenced by groundwater flow. The sediments were cooler than the overlying water column, especially in spring. Sediment temperature tracked water temperature and, therefore, was variable over short time scales (h–d). Temperature changes in near-shore sediments were smaller and slightly delayed compared to those in the overlying water. Over a whole season, temperature variability measured 5 cm below the sediment surface was, on average, 40 to 60% lower than that observed in the overlying water. The reduction in temperature variability was related to sediment characteristics and differed seasonally. Thermal diffusivity of these near-shore lake sediments ranged over an order of magnitude (0.0006–0.007 cm²/s). Thermal diffusivity increased with decreasing sediment particle size and was lower in spring than in summer and autumn. Overall, these results suggest that wind-driven hydrodynamic forces, which determine the frequency and intensity of upwellings and drive water movements in the shallow parts of Lake Opeongo, also determine the spatial structure and temporal variability of near-shore sediment temperature.

Rapid human population growth engenders landuse changes and alters nutrients, algal biomass (as chlorophyll a [chl a]), and dissolved O₂ concentrations (DO) in streams, but quantitative information on these relationships is limited in regions with a Mediterranean climate. We surveyed macroalgal % cover and chl a and physicochemical factors in spring and summer at 15 stream and estuarine sites in a catchment in southern California with a mosaic of undeveloped, agricultural, and urban areas to examine relationships among land use, nutrients, algae, and DO. We used nutrient diffusing substrata (NDS) at 12 sites to assess the nutrient(s) limiting algal growth. Algal chl a and pH often exceeded suggested or mandated water-quality impairment thresholds at sites affected by human activity, particularly those downstream from a wastewater treatment plant. Total N (TN) affected total and benthic algal chl a in spring and summer, total P (TP) affected benthic algal chl a in summer, and light influenced algae in summer. N was the sole or primary limiting nutrient at ½ of the sites and was colimiting with P at another ⅓ of the sites. Nutrient ratios (molar TN∶TP) were poor predictors of algal responses to nutrient enrichment. Diel changes in dissolved O₂ were related positively to discharge and negatively to algal chl a, particularly when floating macroalgae were present. In June, N, chl a, and macroalgal % cover were positively related to human landuse patterns at the subcatchment scale, whereas in September, P and chl a were related to landuse patterns at more local scales (500 m, 1000 m). Mediterranean streams present a variety of challenges for water-quality managers because of high seasonal and interannual variation in discharge and nutrient flux, high accumulations of algae during the dry season, low nutrient thresholds that generate nuisance algal blooms, and human population growth and associated changes in landuse patterns.

Fish-mediated nutrient recycling influences nutrient dynamics in stream ecosystems, but its consequences for smaller-scale microbial processes in benthic habitats are not well understood. We quantified the effect of nutrient recycling by the grazing fish, Campostoma anomalum, on downstream periphyton in 12 flow-through stream mesocosms. We compared periphyton nutrient ratios and algal biomass (as chlorophyll a [chl a]) between tiles upstream and downstream of enclosures with and without fish to measure nutrient-cycling effects in streams with low (11) and high (177) surface-water dissolved N∶P molar ratios. No upstream–downstream changes in periphyton nutrient ratios were observed in low N∶P streams with or without fish. In high N∶P streams, periphyton C∶N decreased and C∶P and N∶P increased on tiles downstream of enclosures. In high N∶P streams, downstream changes in periphyton nutrient ratios were greater in streams with than without fish, and chl a significantly increased downstream of enclosures with fish. We linked nutrient-recycling effects to downstream microbial processes by comparing bacterial biomass production (BBP), photosynthesis (PS) rates, and the degree of coupling between the 2 processes on tiles downstream of enclosures. We estimated the degree of coupled production between algae and bacteria downstream of enclosures as the covariance between PS and BBP among replicates within each stream (COV_PS–BBP). In high N∶P streams, areal BBP and PS rates and COV_PS–BBP were higher downstream of enclosures with fish. Chl a and COV_PS–BBP declined with increasing periphyton C∶N content, resulting in a positive relationship between COV_PS–BBP and algal biomass across all treatments. Our results indicate that grazing fish alter stream ecosystem N and P dynamics through consumer-mediated recycling pathways, but downstream responses depend on background nutrient regimes. Fish-mediated changes in nutrient dynamics and algal biomass influence reliance of heterotrophs and autotrophs on nutrients recycled within periphyton communities to support benthic production.

Detrital food webs of woodland streams depend on terrestrial litter input and, thus, are susceptible to changes in riparian cover. We assessed effects of litter species richness and quality on decomposition and associated biological communities in temperate deciduous forest and tropical rainforest streams. Three native litter species were incubated in each stream in all combinations (7 litter treatments, 3 richness levels) in coarse- (invertebrate access) and fine-mesh bags (no invertebrate access) and were sampled 5 times over 74 (temperate stream) or 94 d (tropical stream). Decomposition, and fungal biomass, sporulation, and species richness were measured for each treatment. Alnus glutinosa litter was incubated in both streams to assess effects of environmental and biological differences between streams on litter decomposition. Biological colonization (number of fungal species, fungal biomass) and activity (conidial production) were lower in the tropical than the temperate stream, despite its higher water temperature (24 vs 8°C). Mass loss for individual species reached 95% in the temperate and 60% in the rainforest stream. Decomposition rates in mixtures were unaffected by litter richness but could be predicted from their initial N, phenol, and lignin concentrations (leaf quality). In the temperate stream, Alnus decomposition in coarse-mesh bags was positively related to litter richness, and Alnus stimulated decomposition of mixtures. Microbial O₂ consumption, fungal biomass accrual, aquatic hyphomycete sporulation rate and richness, and shredder abundance and richness were insensitive to litter richness. In the temperate stream, presence of tough litter inhibited invertebrate colonization of mixtures, whereas in the tropical stream, presence of soft litter stimulated invertebrate colonization of mixtures. Litter quality (species identity), not richness, was the main controller of decomposition of litter mixtures, and decomposition of litter in mixtures may differ from decomposition of individual species. Thus, disappearance or introduction of key species might affect organic matter processing in streams.

The widespread distributions of aquatic species often contrast with their limited ability to disperse by their own propulsion among wetlands isolated by land. Studies of the potential role of water birds as dispersal vectors have been focused mainly on internal transport (endozoochory). However, many anecdotal observations that small species adhere to flying birds also exist (ectozoochory). We addressed the hypothesis that ectozoochory may contribute to the widespread distributions of aquatic snails (Gastropoda) in several experiments. We tested the likelihood that snails would attach to mallards (Anas platyrhynchos) leaving macrophyte vegetation with high densities of 3 snail species. All species tested (Gyraulus albus, Anisus vortex, and Radix balthica) readily attached to the mallards' bodies. The rate of attachment was proportional to snail density, and the birds' feathers contained most snails. However, ⅔ of the snails detached when mallards subsequently walked for 3 m. Snails of 12 species attached within minutes to any surroundings available when floating in the water, a result indicating that active crawling onto birds may facilitate dispersal. Snails we attached deliberately to duck bills with mud could remain attached for up to 8 h. We measured desiccation tolerance of 13 common aquatic snail species. Almost all snail species survived 48 h of desiccation at 10 to 20°C. The ability to retain water did not differ between species with an operculum and species that form a mucus layer (epiphragm) in their shell openings. Our experiments indicate that aquatic snails possess a range of prerequisites for successful bird-mediated dispersal, but the capacity of snails (and other propagules) to remain attached during flight and successfully colonize new habitats upon arrival must still be assessed.

Trophic interactions can influence the flux of energy and nutrients between donor and recipient ecosystems, thereby altering recipient food webs and communities. We investigated the potential for anadromous alewife to influence filter-feeding insects in lake outflow streams by altering seston and zooplankton export from lakes to streams through size-selective consumption of zooplankton in lakes. We compared spring and summer zooplankton biomass and body size and seston concentration in 6 lakes and their respective outflow streams (3 with alewife access, 3 without alewife access) in Maine, USA. The contribution of seston particles ≥500-µm and mean pelagic zooplankton size decreased from spring to summer in alewife lakes, but not in lakes without alewife. However, we found no seasonal change in zooplankton body size and seston ≥500-µm in the outflow streams of alewife lakes. Furthermore, zooplankton biomass and seston concentration differed between lakes and streams, a result indicating the influence of additional factors on seston movement from lakes to streams. Last, we found no relationship between filter-feeder biomass or community structure and seston quantity, a result indicating that stream communities probably were not strongly regulated by seston availability. Our results illustrate that strong trophic interactions may not be propagated from lakes to streams and suggest the importance of boundaries in modulating the extent to which trophic interactions in one ecosystem can influence the flow of energy to adjacent ecosystems.

Empirical analyses of ecological changes accompanying nutrient enrichment provide one line of evidence for developing protective nutrient criteria. The purpose of my study was to elucidate 2 important aspects of such empirical analyses: 1) how statistical techniques compare regarding types of relationships fit and benchmark values obtained, and 2) how decisions regarding log-transformation of right-skewed nutrient data affect the results. I used data from Great Lakes coastal wetlands describing a suite of water-quality and biotic responses over a large nutrient gradient to conduct side-by-side evaluations of 5 statistical techniques (logistic regression, cumulative probability analysis, linear regression, piecewise regression, classification and regression tree analysis [CART]). With this somewhat noisy data set, differences in goodness-of-fit among techniques that modeled gradual changes vs ones that identified abrupt transitions were remarkably small, providing little evidence for superiority of one over another. However, differences among techniques in nutrient benchmark values and their ecological interpretation were substantial. Log₁₀-transformation of nutrient data had little effect on residuals but shifted the benchmark values for all techniques except CART. Decisions concerning log-transformation ought to be based on implications for deriving criteria rather than perceived statistical assumptions. Multiple statistical techniques and response relationships provide relevant information and no transformation makes all relationships conform to the same pattern, so no cookbook recipe for analyses can be identified. Professional judgment is needed to convey empirical findings toward eventual criteria values regardless of the technique applied.

Variation in N stable isotope (δ¹⁵N) signatures of basal resources can influence interpretation of trophic relationships in ecosystems, and significant variation in δ¹⁵N signatures has been reported in streams and rivers. However, a comprehensive understanding of the main factors driving δ¹⁵N variability is lacking, and this variability confounds the consumer's trophic-level position during δ¹⁵N analysis. We conducted a meta-analysis to examine the variability in δ¹⁵N natural abundance of basal resources and dissolved inorganic N (DIN) in streams and rivers in relation to the environmental factors that may drive this variability. The meta-analysis was based on a literature review over the last 20 y (1989–2009) and contained signatures of δ¹⁵N-DIN (δ¹⁵N-NO₃ and δ¹⁵N-NH₄) and δ¹⁵N-basal resources (δ¹⁵N-detrital compartments, δ¹⁵N-biofilm, δ¹⁵N-algae, and δ¹⁵N-macrophytes) from >100 rivers or streams. Signatures of δ¹⁵N-DIN varied widely (−8.4–19.4‰), and we found fewer values for δ¹⁵N-NH₄ than δ¹⁵N-NO₃, even though NH₄ is assimilated rapidly by basal resources. The range of δ¹⁵N-basal resources was also broad (−4–16‰) within and among compartments. Human land use was the most significant factor explaining variability in δ¹⁵N-DIN and δ¹⁵N-basal resource signatures. We found significant differences between δ¹⁵N signatures of photoautotrophic (i.e., autochthonous) and detrital (i.e., allochthonous) basal resources. Our results point out the difficulty in defining a baseline δ¹⁵N signature of the food web, and provide a basis to explain confounding results in studies using δ¹⁵N analysis to identify trophic linkages in fluvial food webs.