Multiple lines of evidence point to pesticides as stressors affecting invertebrate communities in small streams in five United States regions.

Multistressor studies were performed in five regions of the United States to assess the role of pesticides as stressors affecting invertebrate communities in wadable streams. Pesticides and other chemical and physical stressors were measured in 75 to 99 streams per region for 4 weeks, after which invertebrate communities were surveyed (435 total sites). Pesticides were sampled weekly in filtered water, and once in bed sediment. The role of pesticides as a stressor to invertebrate communities was assessed by evaluating multiple lines of evidence: toxicity predictions based on measured pesticide concentrations, multivariate models and other statistical analyses, and previously published mesocosm experiments. Toxicity predictions using benchmarks and species sensitivity distributions and statistical correlations suggested that pesticides were present at high enough concentrations to adversely affect invertebrate communities at the regional scale. Two undirected techniques-boosted regression tree models and distance-based linear models-identified which pesticides were predictors of (respectively) invertebrate metrics and community composition. To put insecticides in context with known, influential covariates of invertebrate response, generalized additive models were used to identify which individual pesticide(s) were important predictors of invertebrate community condition in each region, after accounting for natural covariates. Four insecticides were identified as stressors to invertebrate communities at the regional scale: bifenthrin, chlordane, fipronil and its degradates, and imidacloprid. Fipronil was particularly important in the Southeast region, and imidacloprid, bifenthrin, and chlordane were important in multiple regions. For imidacloprid, bifenthrin, and fipronil, toxicity predictions were supported by mesocosm experiments that demonstrated adverse effects on naïve aquatic communities when dosed under controlled conditions. These multiple lines of evidence do not prove causality-which is challenging in the field under multistressor conditions-but they make a strong case for the role of insecticides as stressors adversely affecting invertebrate communities in streams within the five sampled regions.

Ecological consequences of neonicotinoid mixtures in streams.

Neonicotinoid mixtures are common in streams worldwide, but corresponding ecological responses are poorly understood. We combined experimental and observational studies to narrow this knowledge gap. The mesocosm experiment determined that concentrations of the neonicotinoids imidacloprid and clothianidin (range of exposures, 0 to 11.9 µg/liter) above the hazard concentration for 5% of species (0.017 and 0.010 µg/liter, respectively) caused a loss in taxa abundance and richness, disrupted adult emergence, and altered trophodynamics, while mixtures of the two neonicotinoids caused dose-dependent synergistic effects. In 85 Coastal California streams, neonicotinoids were commonly detected [59% of samples (n = 340), 72% of streams], frequently occurred as mixtures (56% of streams), and potential toxicity was dominated by imidacloprid (maximum = 1.92 µg/liter) and clothianidin (maximum = 2.51 µg/liter). Ecological responses in the field were consistent with the synergistic effects observed in the mesocosm experiment, indicating that neonicotinoid mixtures pose greater than expected risks to stream health.

Sediment Sources and Sealed-Pavement Area Drive Polycyclic Aromatic Hydrocarbon and Metal Occurrence in Urban Streams.

Metals and polycyclic aromatic hydrocarbons (PAHs) are common pollutants in urban streambed sediment, yet their occurrence is highly variable and difficult to predict. To investigate sources of PAHs and metals to streambed sediment, we sampled pavement dust, soil, and streambed sediment in 10 urban watersheds in three regions of the United States and applied a fallout-radionuclide-based sediment-source analysis to quantify the pavement dust contribution to stream sediment (%dust). We also mapped the area of sealcoated pavement in each watershed (%sealed) to investigate the role of coal-tar pavement sealant (CTS) as a PAH source. Median total and carbon-normalized total PAH concentrations were significantly higher in streambed sediment in the Northeast (54.3 mg/kg and 2.71 mg/gOC) and Southeast (5.37 mg/kg and 1.36 mg/gOC), where CTS is commonly used, than in the Northwest (2.11 mg/kg and 0.071 mg/gOC), where CTS is rarely used. Generalized additive models indicated that %sealed and in some cases %dust significantly affected total PAH concentrations in streambed sediments. The %dust was a significant variable for common urban metals: Cu, Pb, and Zn. These findings advance our quantitative understanding of the role of pavement dust as a source and a vector of contaminants to urban streams.

New-generation pesticides are prevalent in California's Central Coast streams.

Pesticides are widely recognized as important biological stressors in streams, especially in heavily developed urban and agricultural areas like the Central California Coast region. We assessed occurrence and potential toxicity of pesticides in small streams in the region using two analytical methods: a broad-spectrum (223 compounds) method in use since 2012 and a newly developed method for 30 additional new-generation fungicides and insecticides. At least one pesticide compound was identified in 83 of the 85 streams sampled. About one-half (48%) of the 253 pesticides measured were detected at least once and 27 were detected in 10% or more of samples. Three of the top 4, and 6 of the top 10 most frequently detected compounds (chlorantraniliprole, dinotefuran, boscalid, thiamethoxam, clothianidin and the fluopicolide degradate 2,6-dichlorobenzamide) were analyzed by the new method. Pesticide mixtures were common, with two or more pesticide compounds detected in 81% of samples and 10 or more in 32% of samples. The pesticide count at a site was relatively consistent over the 6-week study. Four sites with mixed land-use in the lower basin (<5 km from the sampling site) tended to have the highest pesticide counts and the highest concentrations. Potential toxicity (assessed by comparison to benchmarks) to invertebrates was much more common than potential toxicity to fish or plants and was associated with a wide array of insecticides. The common occurrence of new-generation pesticides highlights the need to continuously update analytical methods to keep pace with changing pesticide use for a fuller assessment of pesticide occurrence and effects on the environment.

Multiple in-stream stressors degrade biological assemblages in five U.S. regions.

Biological assemblages in streams are affected by a wide variety of physical and chemical stressors associated with land-use development, yet the importance of combinations of different types of stressors is not well known. From 2013 to 2017, the U.S. Geological Survey completed multi-stressor/multi-assemblage stream ecological assessments in five regions of the United States (434 streams total). Diatom, invertebrate, and fish communities were enumerated, and five types of potential stressors were quantified: habitat disturbance, excess nutrients, high flows, basic water quality, and contaminants in water and sediment. Boosted regression tree (BRT) models for each biological assemblage and region generally included variables from all five stressor types and multiple stressors types in each model was the norm. Classification and regression tree (CART) models then were used to determine thresholds for each BRT model variable above which there appeared to be adverse effects (multi-metric index (MMI) models only). In every region and assemblage there was a significant inverse relation between the MMI and the number of stressors exerting potentially adverse effects. The number of elevated instream stressors often varied substantially for a given level of land-use development and the number of elevated stressors was a better predictor of biological condition than was development. Using the adverse effects-levels that were developed based on the BRT model results, 68% of the streams had two or more stressors with potentially adverse effects and 35% had four or more. Our results indicate that relatively small increases in the number of stressors of different types can have a large effect on a stream ecosystem.

Is there an urban pesticide signature? Urban streams in five U.S. regions share common dissolved-phase pesticides but differ in predicted aquatic toxicity.

Pesticides occur in urban streams globally, but the relation of occurrence to urbanization can be obscured by regional differences. In studies of five regions of the United States, we investigated the effect of region and urbanization on the occurrence and potential toxicity of dissolved pesticide mixtures. We analyzed 225 pesticide compounds in weekly discrete water samples collected during 6-12 weeks from 271 wadable streams; development in these basins ranged from undeveloped to highly urbanized. Sixteen pesticides were consistently detected in 16 urban centers across the five regions-we propose that these pesticides comprise a suite of urban signature pesticides (USP) that are all common in small U.S. urban streams. These USPs accounted for the majority of summed maximum pesticide concentrations at urban sites within each urban center. USP concentrations, mixture complexity, and potential toxicity increased with the degree of urbanization in the basin. Basin urbanization explained the most variability in multivariate distance-based models of pesticide profiles, with region always secondary in importance. The USPs accounted for 83% of pesticides in the 20 most frequently occurring 2-compound unique mixtures at urban sites, with carbendazim+prometon the most common. Although USPs were consistently detected in all regions, detection frequencies and concentrations varied by region, conferring differences in potential aquatic toxicity. Potential toxicity was highest for invertebrates (benchmarks exceeded in 51% of urban streams), due most often to the neonicotinoid insecticide imidacloprid and secondarily to organophosphate insecticides and fipronil. Benchmarks were rarely exceeded in urban streams for plants (at 3% of sites) or fish (<1%). We propose that the USPs identified here would make logical core (nonexclusive) constituents for monitoring dissolved pesticides in U.S. urban streams, and that unique mixtures containing imidacloprid, fipronil, and carbendazim are priority candidates for mixtures toxicity testing.

Multi-region assessment of chemical mixture exposures and predicted cumulative effects in USA wadeable urban/agriculture-gradient streams.

Chemical-contaminant mixtures are widely reported in large stream reaches in urban/agriculture-developed watersheds, but mixture compositions and aggregate biological effects are less well understood in corresponding smaller headwaters, which comprise most of stream length, riparian connectivity, and spatial biodiversity. During 2014-2017, the U.S. Geological Survey (USGS) measured 389 unique organic analytes (pharmaceutical, pesticide, organic wastewater indicators) in 305 headwater streams within four contiguous United States (US) regions. Potential aquatic biological effects were evaluated for estimated maximum and median exposure conditions using multiple lines of evidence, including occurrence/concentrations of designed-bioactive pesticides and pharmaceuticals and cumulative risk screening based on vertebrate-centric ToxCast™ exposure-response data and on invertebrate and nonvascular plant aquatic life benchmarks. Mixed-contaminant exposures were ubiquitous and varied, with 78% (304) of analytes detected at least once and cumulative maximum concentrations up to more than 156,000 ng/L. Designed bioactives represented 83% of detected analytes. Contaminant summary metrics correlated strong-positive (rho (ρ): 0.569-0.719) to multiple watershed-development metrics, only weak-positive to point-source discharges (ρ: 0.225-353), and moderate- to strong-negative with multiple instream invertebrate metrics (ρ: -0.373 to -0.652). Risk screening indicated common exposures with high probability of vertebrate-centric molecular effects and of acute toxicity to invertebrates, respectively. The results confirm exposures to broad and diverse contaminant mixtures and provide convincing multiple lines of evidence that chemical contaminants contribute substantially to adverse multi-stressor effects in headwater-stream communities.

Inclusion of Pesticide Transformation Products Is Key to Estimating Pesticide Exposures and Effects in Small U.S. Streams.

Improved analytical methods can quantify hundreds of pesticide transformation products (TPs), but understanding of TP occurrence and potential toxicity in aquatic ecosystems remains limited. We quantified 108 parent pesticides and 116 TPs in more than 3 700 samples from 442 small streams in mostly urban basins across five major regions of the United States. TPs were detected nearly as frequently as parents (90 and 95% of streams, respectively); 102 TPs were detected at least once and 28 were detected in >20% samples in at least one region-TPs of 9 herbicides, 2 fungicides (chlorothalonil and thiophanate-methyl), and 1 insecticide (fipronil) were the most frequently detected. TPs occurred commonly during baseflow conditions, indicating chronic environmental TP exposures to aquatic organisms and the likely importance of groundwater as a TP source. Hazard quotients based on acute aquatic-life benchmarks for invertebrates and nonvascular plants and vertebrate-centric molecular endpoints (sublethal effects) quantify the range of the potential contribution of TPs to environmental risk and highlight several TP exposure-response data gaps. A precautionary approach using equimolar substitution of parent benchmarks or endpoints for missing TP benchmarks indicates that potential aquatic effects of pesticide TPs could be underestimated by an order of magnitude or more.

Common insecticide disrupts aquatic communities: A mesocosm-to-field ecological risk assessment of fipronil and its degradates in U.S. streams.

Insecticides in streams are increasingly a global concern, yet information on safe concentrations for aquatic ecosystems is sparse. In a 30-day mesocosm experiment exposing native benthic aquatic invertebrates to the common insecticide fipronil and four degradates, fipronil compounds caused altered emergence and trophic cascades. Effect concentrations eliciting a 50% response (EC50) were developed for fipronil and its sulfide, sulfone, and desulfinyl degradates; taxa were insensitive to fipronil amide. Hazard concentrations for 5% of affected species derived from up to 15 mesocosm EC50 values were used to convert fipronil compound concentrations in field samples to the sum of toxic units (∑TUFipronils). Mean ∑TUFipronils exceeded 1 (indicating toxicity) in 16% of streams sampled from five regional studies. The Species at Risk invertebrate metric was negatively associated with ∑TUFipronils in four of five regions sampled. This ecological risk assessment indicates that low concentrations of fipronil compounds degrade stream communities in multiple regions of the United States.

Prioritizing river basins for intensive monitoring and assessment by the US Geological Survey.

The US Geological Survey (USGS) is currently (2020) integrating its water science programs to better address the nation's greatest water resource challenges now and into the future. This integration will rely, in part, on data from 10 or more intensively monitored river basins from across the USA. A team of USGS scientists was convened to develop a systematic, quantitative approach to prioritize candidate basins for this monitoring investment to ensure that, as a group, the 10 basins will support the assessment and forecasting objectives of the major USGS water science programs. Candidate basins were the level-4 hydrologic units (HUC04) with some of the smaller HUC04s being combined; median candidate-basin area is 46,600 km2. Candidate basins for the contiguous United States (CONUS) were grouped into 18 hydrologic regions. Ten geospatial variables representing land use, climate change, water use, water-balance components, streamflow alteration, fire risk, and ecosystem sensitivity were selected to rank candidate basins within each of the 18 hydrologic regions. The two highest ranking candidate basins in each of the 18 regions were identified as finalists for selection as "Integrated Water Science Basins"; final selection will consider input from a variety of stakeholders. The regional framework, with only one basin selected per region, ensures that as a group, the basins represent the range in major drivers of the hydrologic cycle. Ranking within each region, primarily based on anthropogenic stressors of water resources, ensures that settings representing important water-resource challenges for the nation will be studied.

Biofilms Provide New Insight into Pesticide Occurrence in Streams and Links to Aquatic Ecological Communities.

Streambed sediment is commonly analyzed to assess occurrence of hydrophobic pesticides and risks to aquatic communities. However, stream biofilms also have the potential to accumulate pesticides and may be consumed by aquatic organisms. To better characterize risks to aquatic life, the U.S. Geological Survey Regional Stream Quality Assessment measured 93 current-use and 3 legacy pesticides in bed sediment and biofilm from 54 small streams in California across a range of land-use settings. On average, 4 times as many current-use pesticides were detected in biofilm at a site (median of 2) as in sediment (median of 0.5). Of 31 current-use pesticides detected, 20 were detected more frequently in biofilm than in sediment and 10 with equal frequency. Pyrethroids as a class were the most potentially toxic to benthic invertebrates, and of the 9 pyrethroids detected, 7 occurred more frequently in biofilm than sediment. We constructed general additive models to investigate relations between pesticides and 6 metrics of benthic community structure. Pesticides in biofilm improved fit in 4 of the 6 models, and pesticides in sediment improved fit in 2. The results indicate that the sampling of stream biofilms can complement bed-sediment sampling by identification of more current-use pesticides present and better estimation of ecological risks.

Legacy and Current-Use Contaminants in Sediments Alter Macroinvertebrate Communities in Southeastern US Streams.

Sediment contamination of freshwater streams in urban areas is a recognized and growing concern. As a part of a comprehensive regional stream-quality assessment, stream-bed sediment was sampled from streams spanning a gradient of urban intensity in the Piedmont ecoregion of the southeastern United States. We evaluated relations between a broad suite of sediment contaminants (metals, current-use pesticides, organochlorine pesticides, polychlorinated biphenyls, brominated diphenyl ethers, and polycyclic aromatic hydrocarbons), ambient sediment toxicity, and macroinvertebrate communities from 76 sites. Sediment toxicity was evaluated by conducting whole-sediment laboratory toxicity testing with the amphipod Hyalella azteca (for 28 d) and the midge Chironomus dilutus (for 10 d). Approximately one-third of the sediment samples were identified as toxic for at least one test species endpoint, although concentrations of contaminants infrequently exceeded toxicity benchmarks. Ratios of contaminant concentrations relative to their benchmarks, both individually and as summed benchmark quotients, were explored on a carbon-normalized and a dry-weight basis. Invertebrate taxa measures from ecological surveys tended to decline with increasing urbanization and with sediment contamination. Toxicity test endpoints were more strongly related to sediment contamination than invertebrate community measures were. Sediment chemistry and sediment toxicity provided moderate and weak, respectively, explanatory power for the similarity/dissimilarity of invertebrate communities. The results indicate that current single-chemical sediment benchmarks may underestimate the effects from mixtures of sediment contaminants experienced by lotic invertebrates. Environ Toxicol Chem 2020;39:1219-1232. Published 2020. This article is a U.S. Government work and is in the public domain in the USA.

Daily stream samples reveal highly complex pesticide occurrence and potential toxicity to aquatic life.

Transient, acutely toxic concentrations of pesticides in streams can go undetected by fixed-interval sampling programs. Here we compare temporal patterns in occurrence of current-use pesticides in daily composite samples to those in weekly composite and weekly discrete samples of surface water from 14 small stream sites. Samples were collected over 10-14 weeks at 7 stream sites in each of the Midwestern and Southeastern United States. Samples were analyzed for over 200 pesticides and degradates by direct aqueous injection liquid chromatography with tandem mass spectrometry. Nearly 2 and 3 times as many unique pesticides were detected in daily samples as in weekly composite and weekly discrete samples, respectively. Based on exceedances of acute-invertebrate benchmarks (AIB) and(or) a Pesticide Toxicity Index (PTI) >1, potential acute-invertebrate toxicity was predicted at 11 of 14 sites from the results for daily composite samples, but was predicted for only 3 sites from weekly composites and for no sites from weekly discrete samples. Insecticides were responsible for most of the potential invertebrate toxicity, occurred transiently, and frequently were missed by the weekly discrete and composite samples. The number of days with benthic-invertebrate PTI ≥0.1 in daily composite samples was inversely related to Ephemeroptera, Plecoptera, and Trichoptera (EPT) richness at the sites. The results of the study indicate that short-term, potentially toxic peaks in pesticides frequently are missed by weekly discrete sampling, and that such peaks may contribute to degradation of invertebrate community condition in small streams. Weekly composite samples underestimated maximum concentrations and potential acute-invertebrate toxicity, but to a lesser degree than weekly discrete samples, and provided a reasonable approximation of the 90th percentile total concentrations of herbicides, insecticides, and fungicides, suggesting that weekly composite sampling may be a compromise between assessment needs and cost.

Multi-region assessment of pharmaceutical exposures and predicted effects in USA wadeable urban-gradient streams.

Human-use pharmaceuticals in urban streams link aquatic-ecosystem health to human health. Pharmaceutical mixtures have been widely reported in larger streams due to historical emphasis on wastewater-treatment plant (WWTP) sources, with limited investigation of pharmaceutical exposures and potential effects in smaller headwater streams. In 2014-2017, the United States Geological Survey measured 111 pharmaceutical compounds in 308 headwater streams (261 urban-gradient sites sampled 3-5 times, 47 putative low-impact sites sampled once) in 4 regions across the US. Simultaneous exposures to multiple pharmaceutical compounds (pharmaceutical mixtures) were observed in 91% of streams (248 urban-gradient, 32 low-impact), with 88 analytes detected across all sites and cumulative maximum concentrations up to 36,142 ng/L per site. Cumulative detections and concentrations correlated to urban land use and presence/absence of permitted WWTP discharges, but pharmaceutical mixtures also were common in the 75% of sampled streams without WWTP. Cumulative exposure-activity ratios (EAR) indicated widespread transient exposures with high probability of molecular effects to vertebrates. Considering the potential individual and interactive effects of the detected pharmaceuticals and the recognized analytical underestimation of the pharmaceutical-contaminant (unassessed parent compounds, metabolites, degradates) space, these results demonstrate a nation-wide environmental concern and the need for watershed-scale mitigation of in-stream pharmaceutical contamination.

Projected urban growth in the southeastern USA puts small streams at risk.

Future land-use development has the potential to profoundly affect the health of aquatic ecosystems in the coming decades. We developed regression models predicting the loss of sensitive fish (R2 = 0.39) and macroinvertebrate (R2 = 0.64) taxa as a function of urban and agricultural land uses and applied them to projected urbanization of the rapidly urbanizing Piedmont ecoregion of the southeastern USA for 2030 and 2060. The regression models are based on a 2014 investigation of water quality and ecology of 75 wadeable streams across the region. Based on these projections, stream kilometers experiencing >50% loss of sensitive fish and invertebrate taxa will nearly quadruple to 19,500 and 38,950 km by 2060 (16 and 32% of small stream kilometers in the region), respectively. Uncertainty was assessed using the 20 and 80% probability of urbanization for the land-use projection model and using the 95% confidence intervals for the regression models. Adverse effects on stream health were linked to elevated concentrations of contaminants and nutrients, low dissolved oxygen, and streamflow alteration, all associated with urbanization. The results of this analysis provide a warning of potential risks from future urbanization and perhaps some guidance on how those risks might be mitigated.

Chemical and Physical Controls on Mercury Source Signatures in Stream Fish from the Northeastern United States.

Streams in the northeastern U.S. receive mercury (Hg) in varying proportions from atmospheric deposition and legacy point sources, making it difficult to attribute shifts in fish concentrations directly back to changes in Hg source management. Mercury stable isotope tracers were utilized to relate sources of Hg to co-located fish and bed sediments from 23 streams across a forested to urban-industrial land-use gradient within this region. Mass-dependent isotopes (δ202Hg) in prey and game fish at forested sites were depleted (medians -0.95 and -0.83 ‰, respectively) in comparison to fish from urban-industrial settings (medians -0.26 and -0.38 ‰, respectively); the forested site group also had higher prey fish Hg concentrations. The separation of Hg isotope signatures in fish was strongly related to in-stream and watershed land-use indicator variables. Fish isotopes were strongly correlated with bed sediment isotopes, but the isotopic offset between the two matrices was variable due to differing ecosystem-specific drivers controlling the extent of MeHg formation. The multivariable approach of analyzing watershed characteristics and stream chemistry reveals that the Hg isotope composition in fish is linked to current and historic Hg sources in the northeastern U.S. and can be used to trace bioaccumulated Hg.

Survey of bioaccessible pyrethroid insecticides and sediment toxicity in urban streams of the northeast United States.

Pyrethroids are a class of widely-used insecticides that can be transported from terrestrial applications to aquatic systems via runoff and tend to sorb to organic carbon in sediments. Pyrethroid occurrence is detrimental to stream ecosystems due to toxicity to sediment-dwelling invertebrates which are particularly at risk of pyrethroid exposure in urban streams. In this work, 49 streams located in watersheds in the northeastern United States were surveyed for nine current-use pyrethroids using two extraction methods. Total sediment concentrations were determined by exhaustive chemical extraction, while bioaccessible concentrations were determined by single-point Tenax extraction. Total and bioaccessible pyrethroid concentrations were detected in 76% and 67% of the sites, and the average sum of pyrethroids was 232 ng/g organic carbon (OC) for total and 43.8 ng/g OC for bioaccessible pyrethroids. Bifenthrin was the most commonly detected pyrethroid in streambed sediments. Sediment toxicity was assessed using 10-d Hyalella azteca bioassays, and 28% and 15% of sediments caused a decrease in H. azteca biomass and survival, respectively. A temperature-based focused toxicity identification evaluation was used to assess pyrethroids as the causal factor for toxicity. The concentrations of pyrethroids was only weakly correlated with the degree of urban land use. Sediment toxicity was predicted by total and bioaccessible pyrethroid concentrations expressed as toxic units. This work suggests that bioaccessibility-based methods, such as Tenax extraction, can be a valuable tool in assessing sediment toxicity.

Effects of urban multi-stressors on three stream biotic assemblages.

During 2014, the U.S. Geological Survey (USGS) National Water-Quality Assessment (NAWQA) project assessed stream quality in 75 streams across an urban disturbance gradient within the Piedmont ecoregion of southeastern United States. Our objectives were to identify primary instream stressors affecting algal, macroinvertebrate and fish assemblages in wadeable streams. Biotic communities were surveyed once at each site, and various instream stressors were measured during a 4-week index period preceding the ecological sampling. The measured stressors included nutrients; contaminants in water, passive samplers, and sediment; instream habitat; and flow variability. All nine boosted regression tree models - three for each of algae, invertebrates, and fish - had cross-validation R2 (CV R2) values of 0.41 or above, and an invertebrate model had the highest CV R2 of 0.65. At least one contaminant metric was important in every model, and minimum daytime dissolved oxygen (DO), nutrients, and flow alteration were important explanatory variables in many of the models. Physical habitat metrics such as sediment substrate were only moderately important. Flow alteration metrics were useful factors in eight of the nine models. Total phosphorus, acetanilide herbicides and flow (time since last peak) were important in all three algal models, whereas insecticide metrics (especially those representing fipronil and imidacloprid) were dominant in the invertebrate models. DO values below approximately 7 mg/L corresponded to a strong decrease in sensitive taxa or an increase in tolerant taxa. DO also showed strong interactions with other variables, particularly contaminants and sediment, where the combined effect of low DO and elevated contaminants increased the impact on the biota more than each variable individually. Contaminants and flow alteration were strongly correlated to urbanization, indicating the importance of urbanization to ecological stream condition in the region.

Linking the Agricultural Landscape of the Midwest to Stream Health with Structural Equation Modeling.

Multiple physical and chemical stressors can simultaneously affect the biological condition of streams. To better understand the complex interactions of land-use practices, water quality, and ecological integrity of streams, the U.S. Geological Survey National Water Quality Assessment Project is conducting regional-scale assessments of stream condition across the United States. In the summer of 2013, weekly water samples were collected from 100 streams in the Midwestern United States. Employing watershed theory, we used structural equation modeling (SEM) to represent a general hypothesis for how 16 variables (previously identified to be important to stream condition) might be inter-related. Again, using SEM, we evaluated the ability of this "stressor network" to explain variations in multimetrics of algal, invertebrate, and fish community health, trimming away any environmental variables not contributing to an explanation of the ecological responses. Seven environmental variables-agricultural and urban land use, sand content of soils, basin area, percent riparian area as forest, channel erosion, and relative bed stability-were found to be important for all three-community metrics. The algal and invertebrate models included water-chemistry variables not included in the fish model. Results suggest that ecological integrity of Midwest streams are affected by both agricultural and urban land uses and by the natural geologic setting, as indicated by the sand content of soils. Chemicals related to crops (pesticides and nutrients) and residential uses (pyrethroids) were found to be more strongly related to ecological integrity than were natural factors (riparian forest, watershed soil character).