Using Multiple Metal Mixture Models to Predict Toxicity of Riverine Sediment Porewater to the Benthic Life Stage of Juvenile White Sturgeon (Acipenser transmontanus).

Five metal mixture dose-response models were used to predict the toxicity of porewater to young sturgeon at areas of interest in the Upper Columbia River (WA, USA/BC, Canada) and to evaluate these models as tools for risk assessments. Dose components of metal mixture models included exposure to free metal ion activities or metal accumulation by biotic ligands or humic acid, and links of dose to response used logistic equations, independent joint action equations, or additive toxicity functions. Laboratory bioassay studies of single metal exposures to juvenile sturgeon, porewater collected in situ in the fast-flowing Upper Columbia River, and metal mixture models were used to evaluate toxicity. The five metal mixture models were very similar in their predictions of adverse response of juvenile sturgeon and in identifying copper (Cu) as the metal responsible for the most toxic conditions. Although the modes of toxic action and the 20% effective concentration values were different among the dose models, predictions of adverse response were consistent among models because all doses were tied to the same biological responses. All models indicated that 56% ± 5% of 122 porewater samples were predicted to have <20% adverse response, 25% ± 5% of samples were predicted to have 20% to 80% adverse response, and 20% ± 4% were predicted to have >80% adverse response in juvenile sturgeon. The approach of combining bioassay toxicity data, compositions of field porewater, and metal mixture models to predict lack of growth and survival of aquatic organisms due to metal toxicity is an important tool that can be integrated with other information (e.g., survey studies of organism populations, life cycle and behavior characteristics, sediment geochemistry, and food sources) to assess risks to aquatic organisms in metal-enriched ecosystems. Environ Toxicol Chem 2024;43:62-73. Published 2023. This article is a U.S. Government work and is in the public domain in the USA.

Time-dependent accumulation of Cd, Co, Cu, Ni, and Zn in natural communities of mayfly and caddisfly larvae: Metal sensitivity, uptake pathways, and mixture toxicity.

Conceptual and quantitative models were developed to assess time-dependent processes in four sequential experimental stream studies that determined abundances of natural communities of mayfly and caddisfly larvae dosed with single metals (Cd, Co, Cu, Ni, Zn) or multiple metals (Cd + Zn, Co + Cu, Cu + Ni, Cu + Zn, Ni + Zn, Cd + Cu + Zn, Co + Cu + Ni, Cu + Ni + Zn). Metal mixtures contained environmentally relevant metal ratios found in mine drainage. Free metal ion concentrations, accumulation of metals by periphyton, and metal uptake by four families of aquatic insect larvae were either measured (Brachycentridae) or predicted (Ephemerellidae, Heptageniidae, Hydropsychidae) using equilibrium and biodynamic models. Toxicity functions, which included metal accumulations by larvae and metal potencies, were linked to abundances of the insect families. Model results indicated that mayflies accumulated more metal than caddisflies and the relative importance of metal uptake by larvae via dissolved or dietary pathways highly depended on metal uptake rate constants for each insect family and concentrations of metals in food and water. For solution compositions in the experimental streams, accumulations of Cd, Cu, and Zn in larvae occurred primarily through dietary uptake, whereas uptake of dissolved metal was more important for Co and Ni accumulations. Cd, Cu, and Ni were major contributors to toxicity in metal mixtures and for metal ratios examined. Our conceptual approach and quantitative results should aid in designing laboratory experiments and field studies that evaluate metal uptake pathways and metal mixture toxicity to aquatic biota.

Bioaccumulation and Toxicity of Cadmium, Copper, Nickel, and Zinc and Their Mixtures to Aquatic Insect Communities.

We describe 2 artificial stream experiments that exposed aquatic insect communities to zinc (Zn), copper (Cu), and cadmium (year 2014) and to Zn, Cu, and nickel (year 2015). The testing strategy was to concurrently expose insect communities to single metals and mixtures. Single-metal tests were repeated to evaluate the reproducibility of the methods and year-to-year variability. Metals were strongly accumulated in sediments, periphyton, and insect (caddisfly) tissues, with the highest concentrations occurring in periphyton. Sensitive mayflies declined in metal treatments, and effect concentrations could be predicted effectively from metal concentrations in either periphyton or water. Most responses were similar in the replicated tests, but median effect concentration values for the mayfly Rhithrogena sp. varied 20-fold between the tests, emphasizing the difficulty comparing sensitivities across studies and the value of repeated testing. Relative to the single-metal responses, the toxicity of the mixtures was either approximately additive or less than additive when calculated as the product of individual responses (response addition). However, even less-than-additive relative responses were sometimes greater than responses to similar concentrations tested singly. The ternary mixtures resulted in mayfly declines at concentrations that caused no declines in the concurrent single-metal tests. When updating species-sensitivity distributions (SSDs) with these results, the mayfly responses were among the most sensitive 10th percentile of available data for all 4 metals, refuting older literature placing mayflies in the insensitive portion of metal SSDs. Testing translocated aquatic insect communities in 30-d artificial streams is an efficient approach to generate multiple species effect values under quasi-natural conditions that are relevant to natural streams. Environ Toxicol Chem 2020;39:812-833. Published 2020 Wiley Periodicals, Inc. on behalf of SETAC. This article is a US government work, and as such, is in the public domain in the United States of America.

Potential Toxicity of Dissolved Metal Mixtures (Cd, Cu, Pb, Zn) to Early Life Stage White Sturgeon ( Acipenser transmontanus) in the Upper Columbia River, Washington, United States.

The Upper Columbia River (UCR) received historical releases of smelter waste resulting in elevated metal concentrations in downstream sediments. Newly hatched white sturgeon hide within the rocky substrate at the sediment-water interface in the UCR for a few weeks before swim-up. Hiding behavior could expose them to metal contaminants, and metal toxicity could contribute to population declines in white sturgeon over the past 50 years. This study evaluates whether there is a link between the toxicity of dissolved metals across the sediment-water interface in the UCR and the survival of early life stage (ELS) white sturgeon. Toxicity of dissolved metal mixtures is evaluated using a combination of previously collected laboratory and field data and recently developed metal mixture toxicity models. The laboratory data consist of individual metal (Cd, Cu, Pb, and Zn) toxicity studies with ELS white sturgeon. The field data include the chemical composition of surface and pore water samples that were collected across the sediment-water interface in the UCR. These data are used in three metal accumulation and two response models. All models predict low toxicity in surface water, whereas effects concentrations greater than 20% are predicted for 60-72% of shallow pore water samples. The flux of dissolved metals, particularly Cu, from shallow pore water to surface water likely exposes prime ELS sturgeon habitat to toxic conditions.

Understanding the captivity effect on invertebrate communities transplanted into an experimental stream laboratory.

Little is known about how design and testing methodologies affect the macroinvertebrate communities that are held captive in mesocosms. To address this knowledge gap, we conducted a 32-d test to determine how seeded invertebrate communities changed once removed from the natural stream and introduced to the laboratory. We evaluated larvae survival and adult emergence in controls from 4 subsequent studies, as well as corresponding within-river community changes. The experimental streams maintained about 80% of the invertebrates that originally colonized the introduced substrates. Many macroinvertebrate populations experienced changes in numbers through time, suggesting that these taxa are unlikely to maintain static populations throughout studies. For example, some taxa (Tanytarsini, Simuliidae, Cinygmula sp.) increased in number, grew (Simuliidae), and possibly recruited new individuals (Baetidae) as larvae, while several also completed other life history events (pupation and emergence) during the 30- to 32-d studies. Midges and mayflies dominated emergence, further supporting the idea that conditions are conducive for many taxa to complete their life cycles while held captive in the experimental streams. However, plecopterans were sensitive to temperature changes >2 °C between river and laboratory. Thus, this experimental stream testing approach can support diverse larval macroinvertebrate communities for durations consistent with some chronic criterion development and life cycle assessments (i.e., 30 d). The changes in communities held captive in the experimental streams were mostly consistent with the parallel changes observed from in situ river samples, indicating that mesocosm results are reasonably representative of real river insect communities. Environ Toxicol Chem 2018;37:2820-2834. Published 2018 Wiley Periodicals Inc. on behalf of SETAC. This article is a US government work and, as such, is in the public domain in the United States of America.

Disentangling the effects of low pH and metal mixture toxicity on macroinvertebrate diversity.

One of the primary goals of biological assessment of streams is to identify which of a suite of chemical stressors is limiting their ecological potential. Elevated metal concentrations in streams are often associated with low pH, yet the effects of these two potentially limiting factors of freshwater biodiversity are rarely considered to interact beyond the effects of pH on metal speciation. Using a dataset from two continents, a biogeochemical model of the toxicity of metal mixtures (Al, Cd, Cu, Pb, Zn) and quantile regression, we addressed the relative importance of both pH and metals as limiting factors for macroinvertebrate communities. Current environmental quality standards for metals proved to be protective of stream macroinvertebrate communities and were used as a starting point to assess metal mixture toxicity. A model of metal mixture toxicity accounting for metal interactions was a better predictor of macroinvertebrate responses than a model considering individual metal toxicity. We showed that the direct limiting effect of pH on richness was of the same magnitude as that of chronic metal toxicity, independent of its influence on the availability and toxicity of metals. By accounting for the direct effect of pH on macroinvertebrate communities, we were able to determine that acidic streams supported less diverse communities than neutral streams even when metals were below no-effect thresholds. Through a multivariate quantile model, we untangled the limiting effect of both pH and metals and predicted the maximum diversity that could be expected at other sites as a function of these variables. This model can be used to identify which of the two stressors is more limiting to the ecological potential of running waters.

In vivo isotopic fractionation of zinc and biodynamic modeling yield insights into detoxification mechanisms in the mayfly Neocloeon triangulifer.

Diversity and biomass of aquatic insects decline in metal-rich aquatic environments, but the mechanisms by which insects from such environments cope with potentially toxic metal concentrations to survive through adulthood are less well understood. In this study, we measured Zn concentrations and isotopes in laboratory-reared diatoms and mayflies (Neocloeon triangulifer) from larval through adult stages. The larvae were fed Zn-enriched diatoms, and bio-concentrated Zn by a factor of 2.5-5 relative to the diatoms but maintained the same Zn-isotopic ratio. These results reflect the importance of dietary uptake and the greater rate of uptake relative to excretion or growth. Upon metamorphosis to subimago, Zn concentrations declined by >70%, but isotopically heavy Zn remained in the subimago bodies. We surmised that the loss of isotopically light Zn during metamorphosis was due to the loss of detoxified Zn and retention of metabolically useful Zn. Through the transition from subimago to imago, Zn concentrations and isotope ratios were virtually unchanged. Because the decrease in Zn body concentration and increase in heavier Zn are seen in the subimagos relative to the larvae, the compartmentalization of Zn must be occurring within the larvae. A biodynamic model was constructed, allowing for isotopic fractionation and partitioning of Zn between metabolically essential and detoxified Zn reservoirs within larvae. The model provides a consistent set of rate and fractionation constants that successfully describe the experimental observations. Specifically, metabolically essential Zn is isotopically heavier and is tightly held once assimilated, and excess, isotopically light Zn is sequestered, detoxified, and ultimately lost during the metamorphosis of larvae to subimagos. To our knowledge, this is the first documentation of in vivo isotopic fractionation in insects, offering an improved understanding of the mechanisms and rates by which the N. triangulifer larvae regulate excess Zn in their bodies.

Larval aquatic insect responses to cadmium and zinc in experimental streams.

To evaluate the risks of metal mixture effects to natural stream communities under ecologically relevant conditions, the authors conducted 30-d tests with benthic macroinvertebrates exposed to cadmium (Cd) and zinc (Zn) in experimental streams. The simultaneous exposures were with Cd and Zn singly and with Cd+Zn mixtures at environmentally relevant ratios. The tests produced concentration-response patterns that for individual taxa were interpreted in the same manner as classic single-species toxicity tests and for community metrics such as taxa richness and mayfly (Ephemeroptera) abundance were interpreted in the same manner as with stream survey data. Effect concentrations from the experimental stream exposures were usually 2 to 3 orders of magnitude lower than those from classic single-species tests. Relative to a response addition model, which assumes that the joint toxicity of the mixtures can be predicted from the product of their responses to individual toxicants, the Cd+Zn mixtures generally showed slightly less than additive toxicity. The authors applied a modeling approach called Tox to explore the mixture toxicity results and to relate the experimental stream results to field data. The approach predicts the accumulation of toxicants (hydrogen, Cd, and Zn) on organisms using a 2-pKa bidentate model that defines interactions between dissolved cations and biological receptors (biotic ligands) and relates that accumulation through a logistic equation to biological response. The Tox modeling was able to predict Cd+Zn mixture responses from the single-metal exposures as well as responses from field data. The similarity of response patterns between the 30-d experimental stream tests and field data supports the environmental relevance of testing aquatic insects in experimental streams. Environ Toxicol Chem 2017;36:749-762. Published 2016 Wiley Periodicals Inc. on behalf of SETAC. This article is a US government work and, as such, is in the public domain in the United States of America.

Metal mixture modeling evaluation project: 2. Comparison of four modeling approaches.

As part of the Metal Mixture Modeling Evaluation (MMME) project, models were developed by the National Institute of Advanced Industrial Science and Technology (Japan), the US Geological Survey (USA), HDR|HydroQual (USA), and the Centre for Ecology and Hydrology (United Kingdom) to address the effects of metal mixtures on biological responses of aquatic organisms. A comparison of the 4 models, as they were presented at the MMME workshop in Brussels, Belgium (May 2012), is provided in the present study. Overall, the models were found to be similar in structure (free ion activities computed by the Windermere humic aqueous model [WHAM]; specific or nonspecific binding of metals/cations in or on the organism; specification of metal potency factors or toxicity response functions to relate metal accumulation to biological response). Major differences in modeling approaches are attributed to various modeling assumptions (e.g., single vs multiple types of binding sites on the organism) and specific calibration strategies that affected the selection of model parameters. The models provided a reasonable description of additive (or nearly additive) toxicity for a number of individual toxicity test results. Less-than-additive toxicity was more difficult to describe with the available models. Because of limitations in the available datasets and the strong interrelationships among the model parameters (binding constants, potency factors, toxicity response parameters), further evaluation of specific model assumptions and calibration strategies is needed.

Expanding metal mixture toxicity models to natural stream and lake invertebrate communities.

A modeling approach that was used to predict the toxicity of dissolved single and multiple metals to trout is extended to stream benthic macroinvertebrates, freshwater zooplankton, and Daphnia magna. The approach predicts the accumulation of toxicants (H, Al, Cd, Cu, Ni, Pb, and Zn) in organisms using 3 equilibrium accumulation models that define interactions between dissolved cations and biological receptors (biotic ligands). These models differ in the structure of the receptors and include a 2-site biotic ligand model, a bidentate biotic ligand or 2-pKa model, and a humic acid model. The predicted accumulation of toxicants is weighted using toxicant-specific coefficients and incorporated into a toxicity function called Tox, which is then related to observed mortality or invertebrate community richness using a logistic equation. All accumulation models provide reasonable fits to metal concentrations in tissue samples of stream invertebrates. Despite the good fits, distinct differences in the magnitude of toxicant accumulation and biotic ligand speciation exist among the models for a given solution composition. However, predicted biological responses are similar among the models because there are interdependencies among model parameters in the accumulation-Tox models. To illustrate potential applications of the approaches, the 3 accumulation-Tox models for natural stream invertebrates are used in Monte Carlo simulations to predict the probability of adverse impacts in catchments of differing geology in central Colorado (USA); to link geology, water chemistry, and biological response; and to demonstrate how this approach can be used to screen for potential risks associated with resource development.

Predicting the toxicity of metal mixtures.

The toxicity of single and multiple metal (Cd, Cu, Pb, and Zn) solutions to trout is predicted using an approach that combines calculations of: (1) solution speciation; (2) competition and accumulation of cations (H, Ca, Mg, Na, Cd, Cu, Pb, and Zn) on low abundance, high affinity and high abundance, low affinity biotic ligand sites; (3) a toxicity function that accounts for accumulation and potency of individual toxicants; and (4) biological response. The approach is evaluated by examining water composition from single metal toxicity tests of trout at 50% mortality, results of theoretical calculations of metal accumulation on fish gills and associated mortality for single, binary, ternary, and quaternary metal solutions, and predictions for a field site impacted by acid rock drainage. These evaluations indicate that toxicity of metal mixtures depends on the relative affinity and potency of toxicants for a given aquatic organism, suites of metals in the mixture, dissolved metal concentrations and ratios, and background solution composition (temperature, pH, and concentrations of major ions and dissolved organic carbon). A composite function that incorporates solution composition, affinity and competition of cations for two types of biotic ligand sites, and potencies of hydrogen and individual metals is proposed as a tool to evaluate potential toxicity of environmental solutions to trout.

Assessing time-integrated dissolved concentrations and predicting toxicity of metals during diel cycling in streams.

Evaluating water quality and the health of aquatic organisms is challenging in systems with systematic diel (24 h) or less predictable runoff-induced changes in water composition. To advance our understanding of how to evaluate environmental health in these dynamic systems, field studies of diel cycling were conducted in two streams (Silver Bow Creek and High Ore Creek) affected by historical mining activities in southwestern Montana. A combination of sampling and modeling tools was used to assess the toxicity of metals in these systems. Diffusive Gradients in Thin Films (DGT) samplers were deployed at multiple time intervals during diel sampling to confirm that DGT integrates time-varying concentrations of dissolved metals. Site specific water compositions, including time-integrated dissolved metal concentrations determined from DGT, a competitive, multiple-toxicant biotic ligand model, and the Windemere Humic Aqueous Model Version 6.0 (WHAM VI) were used to determine the equilibrium speciation of dissolved metals and biotic ligands. The model results were combined with previously collected toxicity data on cutthroat trout to derive a relationship that predicts the relative survivability of these fish at a given site. This integrative approach may prove useful for assessing water quality and toxicity of metals to aquatic organisms in dynamic systems and evaluating whether potential changes in environmental health of aquatic systems are due to anthropogenic activities or natural variability.

Zinc isotope investigation of surface and pore waters in a mountain watershed impacted by acid rock drainage.

The pollution of natural waters with metals derived from the oxidation of sulfide minerals like pyrite is a global environmental problem. However, the metal loading pathways and transport mechanisms associated with acid rock drainage reactions are often difficult to characterize using bulk chemical data alone. In this study, we evaluated the use of zinc (Zn) isotopes to complement traditional geochemical tools in the investigation of contaminated waters at the former Waldorf mining site in the Rocky Mountains, Colorado, U.S.A. Geochemical signatures and statistical analysis helped in identifying two primary metal loading pathways at the Waldorf site. The first was characterized by a circumneutral pH, high alkalinity, and high Zn/Cd ratios. The second was characterized by acidic pHs and low Zn/Cd ratios. Zinc isotope signatures in surface water samples collected across the site were remarkably similar (the δ(66)Zn, relative to JMC 3-0749-L, for most samples ranged from 0.20 to 0.30‰±0.09‰ 2σ). This probably suggests that the ultimate source of Zn is consistent across the Waldorf site, regardless of the metal loading pathway. The δ(66)Zn of pore water samples collected within a nearby metal-impacted wetland area, however, were more variable, ranging from 0.20 to 0.80‰±0.09‰ 2σ. Here the Zn isotopes seemed to reflect differences in groundwater flow pathways. However, a host of secondary processes might also have impacted Zn isotopes, including adsorption of Zn onto soil components, complexation of Zn with dissolved organic matter, uptake of Zn into plants, and the precipitation of Zn during the formation of reduced sulfur species. Zinc isotope analysis proved useful in this study; however, the utility of this isotopic tool would improve considerably with the addition of a comprehensive experimental foundation for interpreting the complex isotopic relationships found in soil pore waters.

Modeling precipitation and sorption of elements during mixing of river water and porewater in the Coeur d'Alene River basin.

Reddish brown flocs form along the edge of the Coeur d'Alene River when porewater drains into river water during the annual lowering of water level in the basin. The precipitates are efficient scavengers of dissolved elements and have characteristics that may make metals associated with them bioavailable. This work characterizes the geochemistry of the porewater and models the formation and composition of the flocs. Porewater is slightly acidic, has suboxic to anoxic characteristics, tends to have higher alkalinity, and contains elevated concentrations of many constituents relative to river water. Laboratory mixing experiments involving porewater and river water were done to produce the precipitates. Thermodynamic predictions using PHREEQC indicate that predicted amounts of ferrihydrite and gibbsite agree with removal of Fe and Al. Predictions of element removal by adsorption onto ferrihydrite are consistent with observed removal using a combination of surface complexation constants for the generalized two-layer model (As and Se), alternative surface constants derived from experiments at high sorbate-to-sorbent ratios (Cd, Co, Cu, Ni, Pb, and Zn), and adjusted surface constants to fit experimental data (Cr, Mo, and Sb). This new set of surface complexation constants needs further testing in other contaminated systems.

Importance of sediment-water interactions in Coeur d'Alene Lake, Idaho, USA: management implications.

A field study at Coeur d'Alene Lake, Idaho, USA, was conducted between October 1998 and August 2001 to examine the potential importance of sediment-water interactions on contaminant transport and to provide the first direct measurements of the benthic flux of dissolved solutes of environmental concern in this lake. Because of potential ecological effects, dissolved zinc and orthophosphate were the solutes of primary interest. Results from deployments of an in situ flux chamber indicated that benthic fluxes of dissolved Zn and orthophosphate were comparable in magnitude to riverine inputs. Tracer analyses and benthic-community metrics provided evidence that solute benthic flux were diffusion-controlled at the flux-chamber deployment sites. That is, effects of biomixing (or bioturbation) and ground-water interactions did not strongly influence benthic flux. Remediation efforts in the river might not produce desired water-quality effects in the lake because imposed shifts in concentration gradients near the sediment-water interface would generate a benthic feedback response. Therefore, development of water-quality models to justify remediation strategies requires consideration of contaminant flux between the water column and underlying sediment in basins that have been affected by long-term (decadal) anthropogenic activities.

Modeling metal removal onto natural particles formed during mixing of acid rock drainage with ambient surface water.

Studies have examined partitioning of trace metals onto natural particles to better understand the fate and transport of trace metals in the environment, but few studies have compared model predictions with field results. We evaluate the application of an empirical modeling approach, using surface complexation parameters available in the literature, to complex natural systems. In this work, the equilibrium speciation computer program PHREEQC was used along with the diffuse double-layer surface complexation model to simulate metal removal onto natural oxide particles formed during the mixing of acid rock drainage with ambient surface water. End-member solutions sampled in the Coeur d'Alene (CdA) Mining District in September 1999 from the Bunker Hill Mine and the South Fork Coeur d'Alene (SFCdA) River were filtered and mixed in several ratios. Solution chemistry was determined for end-members and mixed solutions, and X-ray diffraction (XRD) was used to determine the mineralogy of precipitate phases. Predicted amounts of Fe precipitates were in good agreement with measured values for particulate Fe. Surface area and reactive site characteristics were used along with surface complexation constants for ferrihydrite (Dzombak, D. A.; Morel, F. M. M. Surface Complexation Modeling: Hydrous Ferric Oxide; John Wiley & Sons: New York, 1990) to predict ion sorption as a function of mixing fraction. Comparisons of model predictions with field results indicate that Pb and Cu sorption are predicted well by the model, while As, Mo, and Sb sorption are less well-predicted. Additional comparisons with particulate metal and Fe data collected from the CdA Mining District in 1996 and 1997 suggest that sorption on particulate Fe, including amorphous iron oxides and schwertmannite, may be described using universal model parameters.