Quantifying Variability in Four US Streams Using a Long-Term Data Set: Patterns in Water Quality Endpoints.

Temporal and spatial patterns of variability in aquatic ecosystems can be complex and difficult to quantify or predict. However, understanding this variability is critical to making a wide range of water quality assessment and management decisions effectively. Here we report on the nature and magnitude of spatial and temporal variation observed in conductivity, total phosphorus, and total nitrogen during a 15-year study of four U.S. stream systems receiving pulp and paper mill effluent discharges. Sampling locations included mainstem sites upstream and downstream of effluent discharge, as well as tributary sites. In all four stream systems, variability in conductivity as measured by the coefficient of variation was typically in the range of 10-50%, and was as low or lower than the variability in nutrient endpoints. The effect of effluent discharge was relatively minor overall, except in some site-specific instances. Some relatively large differences between tributary and mainstem variability were also observed. Flow variation tended to have a more consistent and larger effect on conductivity variation compared to the nutrient endpoints. After removing flow effects, significant relatively complex trends over time were observed at several sites. Changes in variability during the study also were observed. This paper highlights the importance of long-term studies to accurately characterize water quality variability used in water quality management decision-making.

Quantifying Variability in Four U.S. Streams Using a Long-Term Dataset: Patterns in Biotic Endpoints.

Effective water resources assessment and management requires quantitative information on the variability of ambient and biological conditions in aquatic communities. Although it is understood that natural systems are variable, robust estimates of long-term variation in community-based structure and function metrics are rare in U.S. waters. We used a multi-year, seasonally sampled dataset from multiple sites (n = 5-6) in four streams (Codorus Creek, PA; Leaf River, MS; McKenzie and Willamette Rivers, OR) to examine spatial and temporal variation in periphyton chlorophyll a, and fish and macroinvertebrate metrics commonly used in bioassessment programs. Within-site variation of macroinvertebrate metrics and benthic chlorophyll a concentration showed coefficient of variation ranging from 16 to 136%. Scale-specific variability patterns (stream-wide, season, site, and site-season patterns) in standardized biotic endpoints showed that within-site variability patterns extended across sites with variability greatest in chlorophyll a and lowest in Hilsenhoff's Biotic Index. Across streams, variance components models showed that variance attributed to the interaction of space and time and sample variance accounted for the majority of variation in macroinvertebrate metrics and chlorophyll a, while most variation in fish metrics was attributed to sample variance. Clear temporal patterns in measured endpoints were rare and not specific to any one stream or assemblage, while apparent shifts in metric variability related to point source discharges were seen only in McKenzie River macroinvertebrate metrics in the fall. Results from this study demonstrate the need to consider and understand spatial, seasonal, and longer term variability in the development of bioassessment programs and subsequent decisions.

A review of regulatory modeling frameworks supporting numeric water quality criteria development in the United States.

The US Environmental Protection Agency (USEPA) has a long history of leveraging environmental models and integrated modeling frameworks to support the regulatory development of numeric ambient water quality criteria for the protection of aquatic life and human health. Primary modeling types include conceptual, mechanistic, and data-driven empirical models; Bayesian and probabilistic models; and risk-based modeling frameworks. These models and modeling frameworks differ in their applicability to and suitability for various water quality criteria objectives. They require varying knowledge of system processes and stressor-response relationships, data availability, and expertise of stakeholders. In addition, models can be distinguished by their ability to characterize variability and uncertainty. In this work, we review USEPA recommendations for model use in existing regulatory frameworks, technical support documents, and peer-reviewed literature. We characterize key attributes, identify knowledge gaps and opportunities for future research, and highlight where renewed USEPA guidance is needed to promote the development and use of models in numeric criteria derivation. These outcomes then inform a decision-based framework for determining model suitability under particular scenarios of available knowledge, data, and access to technical resources. Integr Environ Assess Manag 2023;19:191-201. © 2022 SETAC.

Ambient water quality criteria derived using probabilistic risk assessment.

National recommendations for numeric human health ambient water quality criteria (AWQC) for toxic substances are derived by the US Environmental Protection Agency (USEPA) using a deterministic approach that combines point estimates for exposure, toxicity, and acceptable risk. In accordance with the Clean Water Act, states, territories, and authorized tribes must either adopt these recommendations or modify and replace them with criteria using an alternative, scientifically defensible method. Recent reports have criticized the deterministic approach, stating that it suffers from compounded conservatism by selecting upper percentiles or maximum values for multiple inputs and that it cannot directly determine what portion of the population a given criterion protects. As an alternative, probabilistic risk assessment (PRA) has been promoted as a more transparent and robust method for deriving AWQC. Probabilistic risk assessment offers several advantages over the deterministic approach. For example, PRA uses entire data distributions rather than upper-percentile point estimates to specify exposures, thereby reducing compounded conservatism. Additionally, because it links acceptable risk targets with specific segments of the exposed population, PRA-based AWQC demonstrably protects multiple subsets of the population. To date, no study has quantitatively compared deterministic and PRA approaches and resulting AWQC using national inputs consistent with USEPA guidance. This study introduces a PRA method for deriving AWQC and presents case studies to compare probabilistically derived AWQC with USEPA's 2015 recommendations. The methods and results of this work will help federal and state regulators, water quality managers, and stakeholders better understand available approaches to deriving AWQC and provide context to assumption- and method-specific differences between criteria. Integr Environ Assess Manag 2023;19:501-512. © 2022 The Authors. Integrated Environmental Assessment and Management published by Wiley Periodicals LLC on behalf of Society of Environmental Toxicology & Chemistry (SETAC).

Detection and accumulation of environmentally-relevant glyphosate concentrations delivered via pulse- or continuous-delivery on passive samplers.

Glyphosate is the most commonly used herbicide globally, which has contributed to its ubiquitous presence in the environment. Glyphosate application rates and delivery to surface water vary with land use. Streams in urban and agricultural catchments can experience continuous delivery of low concentrations of glyphosate and aminomethylphosphonic acid (AMPA), while their presence in forest streams occurs as an episodic pulse following silvicultural application. We assessed whether trace concentrations of glyphosate delivered as a 1-day pulse (mimic silvicultural applications) followed by flushing with deionized water would affect the detection of glyphosate or AMPA on novel passive samplers, Polar Organic Chemical Integrative Sampler with Molecular Imprinted Polymer (POCIS-MIP), compared with continuous delivery (mimic agricultural or urban applications). Within each delivery type, POCIS-MIP were exposed to seven treatment concentrations of Rodeo (equivalent to 0.0 to 1.84 µg glyphosate L-1). Experimental results demonstrate POCIS-MIP can detect differences in relative glyphosate concentrations above 0.115 µg L-1 (pulse-delivery) or 0.23 µg L-1 (continuous-delivery), but were unable to distinguish trace concentrations (i.e., < 0.115 or 0.23 µg L-1). Our results suggest POCIS-MIP may better retain glyphosate when delivered as a pulse than when delivered continuously, but both underestimated actual treatment concentrations by 46 to 56%. There is a need to demonstrate the field applicability of passive sampling methods to improve environmental monitoring of silvicultural herbicides, and our results demonstrate passive samplers were unable to distinguish lower concentrations, suggesting a limited utility for determining trace concentration levels such as those experienced during or immediately after silvicultural application.

Deriving Human Health and Aquatic Life Water Quality Criteria in the United States for Bioaccumulative Substances: A Historical Review and Future Perspective.

Methods used to derive water quality regulations for persistent, bioaccumulative, and toxic substances (PBTs) in the United States have evolved substantially over the past 50 yr, leveraging current understandings and assumptions about the nature and magnitude of partitioning and accumulation of substances in water, sediments, and organisms. In the United States and across the world, environmental regulations continue to evolve into more refined water quality criteria protective of aquatic life and human health. The present review provides historical context on the establishment of aquatic life and human health water quality criteria in the United States by compiling information from regulatory agencies and peer-reviewed literature on methods used to characterize and quantify bioaccumulation of substances in aquatic organisms and humans. Primary data needs and assumptions for various methods, as well as their application in setting criteria by the US Environmental Protection Agency over the past half century, are highlighted. Our review offers an important retrospective on the data and methods used to derive water quality criteria for PBTs and provides commentary on the future of US criteria development. Environ Toxicol Chem 2021;40:2394-2405. © 2021 SETAC.

eDNA as a tool for identifying freshwater species in sustainable forestry: A critical review and potential future applications.

Environmental DNA (eDNA) is an emerging biological monitoring tool that can aid in assessing the effects of forestry and forest manufacturing activities on biota. Monitoring taxa across broad spatial and temporal scales is necessary to ensure forest management and forest manufacturing activities meet their environmental goals of maintaining biodiversity. Our objectives are to describe potential applications of eDNA across the wood products supply chain extending from regenerating forests, harvesting, and wood transport, to manufacturing facilities, and to review the current state of the science in this context. To meet our second objective, we summarize the taxa examined with targeted (PCR, qPCR or ddPCR) or metagenomic eDNA methods (eDNA metabarcoding), evaluate how estimated species richness compares between traditional field sampling and eDNA metabarcoding approaches, and compare the geographical representation of prior eDNA studies in freshwater ecosystems to global wood baskets. Potential applications of eDNA include evaluating the effects of forestry and forest manufacturing activities on aquatic biota, delineating fish-bearing versus non fish-bearing reaches, evaluating effectiveness of constructed road crossings for freshwater organism passage, and determining the presence of at-risk species. Studies using targeted eDNA approaches focused on fish, amphibians, and invertebrates, while metagenomic studies focused on fish, invertebrates, and microorganisms. Rare, threatened, or endangered species received the least attention in targeted eDNA research, but are arguably of greatest interest to sustainable forestry and forest manufacturing that seek to preserve freshwater biodiversity. Ultimately, using eDNA methods will enable forestry and forest manufacturing managers to have data-driven prioritization for conservation actions for all freshwater species.

Fathead minnow response to broad-range exposure of ß-sitosterol concentrations during life-cycle testing.

The ß-sitosterol concentration in pulp and paper mill effluents is typically greater than that of other phytosterols and has been shown to cause a variety of effects in fish. The authors exposed fathead minnow (Pimephales promelas) to low (22 ± 0.93 µg/L), medium-low (70 ± 2.1 µg/L), medium-high (237 ± 5.5 µg/L), and high (745 ± 16.2 µg/L) concentrations of ß-sitosterol as well as negative (water), positive (ethynyl estradiol, 16 ± 0.58 ng/L), and carrier (0.6 mL/L acetone) controls. Fish were monitored over a full life cycle for population-level endpoints including growth and survival, reproductive endpoints (e.g. fecundity, sex steroids and vitellogenin, gonado-/hepatosomatic indices, and gonad histology). No significant differences were seen in fish growth, mortality, or reproduction with ß-sitosterol exposure, although a trend for lower egg production in ß-sitosterol exposures relative to the water control may be related to the acetone carrier. All ethynyl estradiol-exposed fish were smaller, showed female characteristics, and did not spawn. Sex steroid and vitellogenin were highly variable with no detectable treatment-related differences. Gonadal tissue showed no ß-sitosterol-related differences in reproductive development and spawning capability, although most ethynyl estradiol-exposed males had ovarian tissue and were not spawning-capable. The results indicate that ß-sitosterol exposure had little apparent impact on fathead minnow survival, growth, and reproduction even at concentrations >10 times that of typical effluents, although small sample size and variability precluded fully evaluating treatment responses on sex steroids and vitellogenin.

Spatial and temporal patterns of periphyton chlorophyll a related to pulp and paper mill discharges in four US receiving streams.

Nutrients in pulp and paper mill effluent (PPME) have been implicated in increased periphyton chlorophyll a (chl a) downstream of discharges. These findings are largely based on short-term studies conducted in artificial stream channels or mesocosms and often in oligotrophic systems, and it is unclear if long-term chl a patterns in higher-nutrient systems would show similar response. We conducted a long-term study of 4 receiving waters (Codorus Creek, Pennsylvania; the Leaf River, Mississippi; and the McKenzie and Willamette rivers, Oregon) in which periphyton samples and associated data on water quality (nitrogen and phosphorus concentrations, pH, color, and conductivity) and 2 physical habitat variables (depth and current velocity) were collected over an 8-y period from multiple sites upstream and downstream of PPME discharges. Study streams represented different ecoregions, warm- and coldwater systems, gradients of in-stream effluent concentration (<1-33%), and mill process types. General Linear Models examining the main and interaction effects of site, season, and year on periphyton chl a for each of the 4 streams showed periphyton chl a downstream of the PPME discharge in Codorus Creek and the McKenzie River was greater at some, but not all upstream sites, suggesting these differences may be due to factors other than PPME. Mean periphyton chl a ranged from <1 to 285 mg/m2 across streams, with relatively consistent site patterns across seasons and years. Overall, chl a in the spring and summer was greater than in the fall in Codorus Creek and on sand substrates in the Leaf River, with overall differences across years seen on rare occasions in the Leaf and Willamette rivers. Regression models examining environmental-chl a relationships explained 45.4% and 30.2% of variation in periphyton chl a in the McKenzie River and Codorus Creek, respectively, and <10% in the Leaf and Willamette rivers. Physical variables (stream depth and current velocity) were the most important model variables in the McKenzie River, while total nitrogen and color were of greatest importance in Codorus Creek. The findings of this study demonstrate the inherent variability of chl a standing crops, highlight the complexity of lotic periphyton communities, and reiterate the importance of long-term, multi-season studies in elucidating spatial and temporal patterns.

Patterns of macroinvertebrate assemblages in a long-term watershed-scale study to address the effects of pulp and paper mill discharges in four US receiving streams.

Changes in macroinvertebrate communities exposed to pulp and paper mill effluent (PPME) have been seen in mesocosm and short-term field studies. However, long-term patterns of macroinvertebrates in PPME receiving streams have not been examined. We conducted a study of 4 PPME receiving streams (Codorus Creek, PA; the Leaf River, MS; and the McKenzie and Willamette rivers, OR) over 9 y to assess temporal patterns in macroinvertebrate community structure and metrics related to PPME discharge. Study streams represented different ecoregions, warm-/cold-water systems, gradients of PPME concentration (<1%-33%), and mill process types. Bray-Curtis similarity and nonmetric multidimensional scaling showed significant community differences across sites in Codorus Creek, but differences were related to stream temperature patterns and not PPME. In the other study streams, seasonal community differences across years were greater than differences across sites. General linear models were used to examine spatial and temporal variation in macroinvertebrate metric response (% dominant taxa, density, richness, Hilsenhoff Biotic Index [HBI], Simpson's Index, and ash-free dry mass). Mean HBI scores indicated that the macroinvertebrate community reflected fair to very good water quality conditions, with water quality typically classified as "good" at most sites. Significant site differences in macroinvertebrate metric response were uncommon in the Leaf, McKenzie, and Willamette rivers but were seen in all metrics in Codorus Creek, where metric response was spatially variable. In the McKenzie River, there was an increase in mean HBI scores at sites downstream of the mill relative to 1 of the 2 upstream sites. However, significant differences were seen only between 1 upstream and downstream site, and HBI scores at all downstream sites consistently reflected "good" water quality. Significant annual differences in metric response were typical in all rivers. Water quality (pH, conductivity, total nitrogen) and habitat (velocity, depth, substrate composition) variables explained community structure patterns in the Leaf and McKenzie rivers, but macroinvertebrate-environment relationships were weak in the other 2 streams. The results of this study indicate that macroinvertebrate community structure is temporally variable and reiterate the importance of long-term studies for accurate determination of the effects of point sources such as PPME on receiving systems.

Patterns of fish community structure in a long-term watershed-scale study to address the aquatic ecosystem effects of pulp and paper mill discharges in four US receiving streams.

Physiological changes have been seen in individual fish exposed to pulp and paper mill effluent (PPME), but it is unclear whether community-level changes are seen in fish in PPME receiving waters. We conducted a study of 4 PPME receiving streams (Codorus Creek, PA, USA), the Leaf River (Forrest and Perry Counties, MS, USA), and the McKenzie and Willamette rivers (Lane County, OR, USA) over 9 y to assess temporal patterns in the type and relative abundance of fish species and measures of community structure and function related to PPME discharge. We used boat and backpack electrofishing to sample large- and small-bodied fish from the McKenzie and Willamette rivers, boat electrofishing to sample large-bodied fish from the Leaf River, and backpack electrofishing to sample the entire fish community from Codorus Creek. Study streams represented different ecoregions, warm- and coldwater systems, gradients of PPME concentration (<1%-33%), and mill process types. Bray-Curtis similarity and nonmetric multidimensional scaling showed high variation in fish communities across sites, seasons, and years. Significant site differences in fish communities were seen in most streams and community types, but distinct separation of sites was seen only in Codorus Creek and unrelated to PPME discharge. No seasonal differences were seen in fish community structure in any stream, with only weak annual patterns in large-bodied fish in the Leaf River and small-bodied fish in the McKenzie River. General linear models were used to examine spatial and temporal variation in fish metric response (abundance, species richness, Simpson's diversity, % dominant species, standing crop, % DELT, % intolerant, % omnivore, % piscivore). Significant site differences in metric response were largely limited to Codorus Creek and unrelated to PPME. Significant reductions of % dominant taxa of small-bodied fish and % large-bodied piscivores were also observed downstream of the PPME discharge on the McKenzie River relative to upstream sites. Seasonal changes in fish metric response were rare, and changes with year were variable. The relationship between fish community structure and water quality variables (pH, color, conductivity, total phosphorus, total nitrogen) was weak in all streams for all community types. The results of this study show that PPME exposure has little effect on fish communities in these streams and aid in addressing management strategies. The high spatial and temporal variability reiterate the importance of long-term studies to elucidate patterns in receiving waters.

A long-term, multitrophic level study to assess pulp and paper mill effluent effects on aquatic communities in four US receiving waters: lessons learned.

Lessons learned from the development, implementation, and initial 8 y of study findings from a long-term study to assess the effects of pulp and paper mill effluents on receiving waters are summarized as a conclusion to a series of articles (this issue) on study findings. The study, based on industry-defined information needs, was developed via a science-based experimental design into a long-term (>10 y) watershed-scale monitoring program that integrated in-stream population/ community assessment, laboratory chronic bioassays, and fathead minnow full life-cycle assays as well as water quality and effluent quality monitoring and habitat assessment in addressing the presence of effluent effects. The 4 study streams (Codorus Creek, PA; Leaf River, MS; and the McKenzie and Willamette rivers, OR) represented both bleached and unbleached kraft mill processes and effluent concentrations that ranged from near typical for the United States (0.4%) to very high (Codorus Creek= 32%). Following 8 y of monitoring, the weight of evidence suggests an absence of biological differences at stations downstream of the mill discharges for periphyton or macroinvertebrates and, with the exception of 1 of 9 large-bodied fish and 1 of 7 small-bodied fish community structure metrics for 1 river (McKenzie), an absence of differences for fish communities. Laboratory bioassay and fathead minnow full-life cycle tests supported a substantial "margin of safety" in that, depending on the effluent, adverse responses did not occur until effluent concentrations were from 2 times to more than 150 times in-stream concentrations. The incorporation of a watershed spatial scale illustrated that each sample site tended to be unique over the 28 to 50 km monitored segments with respect to habitat and that knowledge of these variables permitted accurate evaluations of effluent effects. Similarly, the multiyear study framework provided information regarding the natural seasonal and year-to-year variability in fish communities and consequently a better understanding of how potential effluent effects signals could be expressed within this variability. The study incorporated an adaptive management strategy that provided for study design and monitoring modifications over time as a way of benefiting from practical experience and knowledge gained through time and to optimize the use of study resources. Results from this initial 8 y of monitoring, to our knowledge, represent the longest-known population/community-level assessment of the in-stream effects of pulp and paper mill effluents. Beyond the lessons learned with respect to effluent effects are those related to the esign and conduct of long-term watershed-scale studies that may be of use to others in developing watershed assessment or management programs.