Stream flow is a controlling element in the ecology of rivers and streams. Knowledge of the natural flow regime facilitates the assessment of whether specific hydrologic attributes have been altered by humans in a particular stream and the establishment of specific goals for stream‐flow restoration. Because most streams are ungaged or have been altered by human influences, characterizing the natural flow regime is often only possible by estimating flow characteristics based on nearby stream gages of reference quality, i.e., gaged locations that are least disturbed by human influences. The ability to evaluate natural stream flow, that which is not altered by human activities, would be enhanced by the existence of a nationally consistent and up‐to‐date database of gages in relatively undisturbed watersheds. As part of a national effort to characterize stream‐flow effects on ecological condition, data for 6785 U.S. Geological Survey (USGS) stream gages and their upstream watersheds were compiled. The sites comprise all USGS stream gages in the conterminous United States with at least 20 years of complete‐year flow record from 1950–2007, and for which watershed boundaries could reliably be delineated (median size = 578 km2). Several hundred watershed and site characteristics were calculated or compiled from national data sources, including environmental features (e.g., climate, geology, soils, topography) and anthropogenic influences (e.g., land use, roads, presence of dams, or canals). In addition, watersheds were assessed for their reference quality within nine broad regions for use in studies intended to characterize stream flows under conditions minimally influenced by human activities. Three primary criteria were used to assess reference quality: (1) a quantitative index of anthropogenic modification within the watershed based on GIS‐derived variables, (2) visual inspection of every stream gage and drainage basin from recent high‐resolution imagery and topographic maps, and (3) information about man‐made influences from USGS Annual Water Data Reports. From the set of 6785 sites, we identified 1512 as reference‐quality stream gages. All data derived for these watersheds as well as the reference condition evaluation are provided as an online data set termed GAGES (geospatial attributes of gages for evaluating stream flow).

Human impacts on watershed hydrology are widespread in the US, but the prevalence and severity of stream‐flow alteration and its potential ecological consequences have not been quantified on a national scale. We assessed streamflow alteration at 2888 streamflow monitoring sites throughout the conterminous US. The magnitudes of mean annual (1980–2007) minimum and maximum streamflows were found to have been altered in 86% of assessed streams. The occurrence, type, and severity of streamflow alteration differed markedly between arid and wet climates. Biological assessments conducted on a subset of these streams showed that, relative to eight chemical and physical covariates, diminished flow magnitudes were the primary predictors of biological integrity for fish and macroinvertebrate communities. In addition, the likelihood of biological impairment doubled with increasing severity of diminished streamflows. Among streams with diminished flow magnitudes, increasingly common fish and macroinvertebrate taxa possessed traits characteristic of lake or pond habitats, including a preference for fine‐grained substrates and slow‐moving currents, as well as the ability to temporarily leave the aquatic environment.

Understanding the extent to which natural streamflow characteristics have been altered is an important consideration for ecological assessments of streams. Assessing hydrologic condition requires that we quantify the attributes of the flow regime that would be expected in the absence of anthropogenic modifications. The objective of this study was to evaluate whether selected streamflow characteristics could be predicted at regional and national scales using geospatial data. Long-term, gaged river basins distributed throughout the contiguous US that had streamflow characteristics representing least disturbed or near pristine conditions were identified. Thirteen metrics of the magnitude, frequency, duration, timing and rate of change of streamflow were calculated using a 20-50 year period of record for each site. We used random forests (RF), a robust statistical modelling approach, to develop models that predicted the value for each streamflow metric using natural watershed characteristics. We compared the performance (i.e. bias and precision) of national-and regional-scale predictive models to that of models based on landscape classifications, including major river basins, ecoregions and hydrologic landscape regions (HLR). For all hydrologic metrics, landscape stratification models produced estimates that were less biased and more precise than a null model that accounted for no natural variability. Predictive models at the national and regional scale performed equally well, and substantially improved predictions of all hydrologic metrics relative to landscape stratification models. Prediction error rates ranged from 15 to 40%, but were 25% for most metrics. We selected three gaged, non-reference sites to illustrate how predictive models could be used to assess hydrologic condition. These examples show how the models accurately estimate predisturbance conditions and are sensitive to changes in streamflow variability associated with long-term land-use change. We also demonstrate how the models can be applied to predict expected natural flow characteristics at ungaged sites.

Data collected from 172 sites in 20 major river basins between 1993 and 1995 as part of the US Geological Survey's National Water-Quality Assessment Program were analyzed to assess relations among basinwide land use (agriculture, forest, urban, range), water physicochemistry, riparian condition, and fish community structure. A multimetric approach was used to develop regionally referenced indices of fish community and riparian condition. Across large geographic areas, decreased riparian condition was associated with water-quality constituents indicative of nonpoint source inputs-total nitrogen and suspended sediment and basinwide urban land use. Decreased fish community condition was associated with increases in total dissolved solids and rangeland use and decreases in riparian condition and agricultural land use. Fish community condition was relatively high even in areas where agricultural land use was relatively high (>50% of the basin). Although agricultural land use can have deleterious effects on fish communities, the results of this study suggest that other factors also may be important, including practices that regulate the delivery of nutrients, suspended sediments, and total dissolved solids into streams. Across large geographic scales, measures of water physicochemistry may be better indicators of fish community condition than basinwide land use. Whereas numerous studies have indicated that riparian restorations are successful in specific cases, this analysis suggests the universal importance of riparian zones to the maintenance and restoration of diverse fish communities in streams.

To examine the relations between fish community function and stream size, we classified 429 lotic freshwater fish species based on multiple categories within six species traits: (1) substrate preference, (2) geomorphic preference, (3) trophic ecology, (4) locomotion morphology, (5) reproductive strategy, and (6) stream size preference. Stream size categories included small streams, small, medium, and large rivers, and no size preference. The frequencies of each species trait category were determined for each stream size category based on life history information from the literature. Cluster analysis revealed the presence of covarying groups of species trait categories. One cluster (RUN) included the traits of planktivore and herbivore feeding ecology, migratory reproductive behavior and broadcast spawning, preferences for main‐channel habitats, and a lack of preferences for substrate type. The frequencies of classifications for the RUN cluster varied significantly across stream size categories (P = 0.009), being greater for large rivers than for small streams and rivers. Another cluster (RIFFLE) included the traits of invertivore feeding ecology, simple nester reproductive behavior, a preference for riffles, and a preference for bedrock, boulder, and cobble–rubble substrate. No significant differences in the frequency of classifications among stream size categories were detected for the RIFFLE cluster (P = 0.328). Our results suggest that fish community function is structured by large‐scale differences in habitat and is different for large rivers than for small streams and rivers. Our findings support theoretical predictions of variation in species traits among stream reaches based on ecological frameworks such as landscape filters, habitat templates, and the river continuum concept. We believe that the species trait classifications presented here provide an opportunity for further examination of fish species' relations to physical, chemical, and biological factors in lotic habitats ranging from small streams to large rivers.

ABSTRACT1. Detailed research into the ecological impacts of inter-basin water transfers (IBTs) is virtually nonexistent on a global scale. However, a growing awareness of the serious nature of such impacts-for example, the loss of biogeographical integrity, the loss of endemic biotas, the frequent introduction of alien and often invasive aquatic and terrestrial plants and animals, the genetic intermixing of once separated populations, the implications for water quality, the frequently drastic alteration of hydrological regimes, the implications for marine and estuarine processes, climatic effects, and the spread of disease vectors, amongst many others-demands a most urgent and world-wide appraisal of all current planning and research strategies.2. This paper first defines the types of extant IBTs, and details some case studies for three widely separated regions of the world, namely: south-eastern Australia, southern Africa, and the central and south-western parts of the United States of America. In doing so, it highlights the chronic paucity of ecological data on their impacts, while simultaneously emphasising their extreme complexities.3. Finally, we call for an international meeting on such schemes, as a matter of priority and extreme urgency, in order to assess the extent of IBTs, their geographical distribution, and their ecological and sociological impacts and implications.

Two‐pass backpack electrofishing data collected as part of the U.S. Geological Survey's National Water‐Quality Assessment Program were analyzed to assess the efficacy of single‐pass backpack electrofishing. A two‐capture removal model was used to estimate, within 10 river basins across the United States, proportional fish species richness from one‐pass electrofishing and probabilities of detection for individual fish species. Mean estimated species richness from first‐pass sampling ([pcirc],s1) ranged from 80.7% to 100% of estimated total species richness for each river basin, based on at least seven samples per basin. However, [pcirc],s1 values for individual sites ranged from 40% to 100% of estimated total species richness. Additional species unique to the second pass were collected in 50.3% of the samples. Of these, cyprinids and centrarchids were collected most frequently. Proportional fish species richness estimated for the first pass increased significantly with decreasing stream width for 1 of the 10 river basins. When used to calculate probabilities of detection of individual fish species, the removal model failed 48% of the time because the number of individuals of a species was greater in the second pass than in the first pass. Single‐pass backpack electrofishing data alone may make it difficult to determine whether characterized fish community structure data are real or spurious. The two‐pass removal model can be used to assess the effectiveness of sampling species richness with a single electrofishing pass. However, the two‐pass removal model may have limited utility to determine probabilities of detection of individual species and, thus, limit the ability to assess the effectiveness of single‐pass sampling to characterize species relative abundances. Multiple‐pass (at least three passes) backpack electrofishing at a large number of sites may not be cost‐effective as part of a standardized sampling protocol for large‐geographic‐scale studies. However, multiple‐pass electrofishing at some sites may be necessary to better evaluate the adequacy of single‐pass electrofishing and to help make meaningful interpretations of fish community structure.

We developed and evaluated empirical models to predict biological condition of wadeable streams in a large portion of the eastern USA, with the ultimate goal of prediction for unsampled basins. Previous work had classified (i.e., altered vs. unaltered) the biological condition of 920 streams based on a biological assessment of macroinvertebrate assemblages. Predictor variables were limited to widely available geospatial data, which included land cover, topography, climate, soils, societal infrastructure, and potential hydrologic modification. We compared the accuracy of predictions of biological condition class based on models with continuous and binary responses. We also evaluated the relative importance of specific groups and individual predictor variables, as well as the relationships between the most important predictors and biological condition. Prediction accuracy and the relative importance of predictor variables were different for two subregions for which models were created. Predictive accuracy in the highlands region improved by including predictors that represented both natural and human activities. Riparian land cover and road-stream intersections were the most important predictors. In contrast, predictive accuracy in the lowlands region was best for models limited to predictors representing natural factors, including basin topography and soil properties. Partial dependence plots revealed complex and nonlinear relationships between specific predictors and the probability of biological alteration. We demonstrate a potential application of the model by predicting biological condition in 552 unsampled basins across an ecoregion in southeastern Wisconsin (USA). Estimates of the likelihood of biological condition of unsampled streams could be a valuable tool for screening large numbers of basins to focus targeted monitoring of potentially unaltered or altered stream segments.