SETAC 2009 Abstracts
Session: Fate and Effects of Metals: General
A Review of the Mechanisms of Al Toxicity to Fish
Robert C. Santore, Adam C. Ryan, Paul Paquin, Robert W.
Gensemer, William Adams
Abstract:
Aluminum toxicity data (i.e. LC50s and LT50s) for fish vary over wide ranges,
even within a given species or for closely related species. Some of the variability
in observed toxicity may be due to water chemistry conditions that influence the
bioavailability of dissolved Al. These bioavailability effects appear to be
comparable to effects seen for other metals (e.g. pH, NOM, and hardness all
modify Al toxicity). However, additional variability in the toxicity data may not be
strictly related to these bioavailability effects. Some of this additional variability
can be attributed to transient effects associated with disequilibrium from freshly
prepared Al solutions that are over-saturated with respect to a solid phase (e.g.,
amorphous gibbsite). Transient effects are associated with observed toxicity
seconds after a solution is prepared, and those effects typically disappear after
aging for even a few minutes. Consideration of multiple mechanisms of Al
toxicity can help explain some of the variability evident in toxicity data that cannot
be explained with typical bioavailability relationships. In this poster we review
how the experimental design and exposure conditions used in Al toxicity tests
very likely determined the mechanisms by which aluminum was toxic and, hence,
influenced the magnitude of the resulting LC50, LT50, or other toxicity endpoint.
Experiments that demonstrated toxicity at low pH frequently showed aluminum
toxicity at levels that were consistent with toxicity mechanisms associated with
dissolved metal concentrations. Under these conditions, bioavailability
relationships that are typical for metals (e.g., pH, NOM, hardness) are also
important for aluminum. Experiments conducted at higher pH where Al solubility
is fairly low were characterized by toxicity at fairly low levels of Al, marginally
above solubility limits, but toxicity was only observed in freshly prepared
solutions. As these solutions aged, solubility reduced the aqueous aluminum
concentrations and toxicity was no longer evident. Additional data at
circumneutral pH showing effects at concentrations well beyond solubility limits
(by orders of magnitude) are likely due to a third mechanism based on toxicity
resulting from particulate aluminum. Consideration of these different
mechanisms provides an alternative view of why observed aluminum toxicity to
fish is highly variable.
Session: Fate and Effects of Metals: General
Influence of Hardness and DOC on Aluminum Toxicity to Freshwater Organisms at Circumneutral pH
Patricio Rodriguez, Jose Arbildua, German Villavicencio, Robert
W. Gensemer, William Stubblefield, Robert C. Santore, William Adams
Abstract:
It is well known that increasing water hardness and dissolved organic carbon
(DOC) concentrations mitigate the toxicity of aluminum (Al) to freshwater
organisms at acidic pH (i.e., pH ? 6). However, recent studies suggest that
hardness and DOC can mitigate Al toxicity at circumneutral to basic pH (i.e., pH
6 or greater), even though the Al in test solutions exists primarily in a
supersaturated insoluble form (e.g., Al hydroxide) under these conditions. This
study further quantifies the influence of these water quality characteristics on Al
toxicity by conducting acute tests with Ceriodaphnia dubia (48-hr) and chronic
tests with Pseudokirchneriella subcapitata (72-hr) at water hardness values
ranging from 10.6 - 120 mg/L as CaCO3, at DOC concentrations ranging from 0 -
4 mg/L, and at pH values from 6 - 8. All Al toxicity results were expressed as total
nominal Al because both dissolved and monomeric Al concentrations remained
consistently low (near solubility limits as dictated by pH) in all treatments, and so
dose-responses were best expressed as a function of total nominal Al. At all
DOC concentrations, increasing hardness reduced Al toxicity to P. subcapitata at
pH 6 and 7, but toxicity increased with increasing hardness at pH 8. Al toxicity to
C. dubia was also significantly reduced with increasing hardness at pH 6, but
mortalities were less than 50% in all hardness treatments at pH 7 and 8 even
with total nominal Al concentrations as high as 5,000 ?g Al/L (DOC = 0) or
15,000 ?g Al/L (DOC = 2 and 4 mg/L). For both species, Al toxicity consistently
decreased with increasing DOC concentrations at all pH levels tested. These
results support the mitigating effects of hardness and DOC on Al toxicity to
freshwater organisms at circumneutral pH, but at least for algae, hardness
effects differed substantially between pH 6 and 8. The mechanisms explaining
hardness or DOC effects on Al toxicity in highly supersaturated solutions are as
yet unknown, but may be useful in the development of regulatory criteria or
toxicity prediction models.
Sessions: Metals Bioavailability Models: From Theory to Practice
Using the Cu BLM as a Water Quality Criteria Tool: Description of a Fixed Monitoring Benchmark Approach
Adam C. Ryan, Robert C. Santore, Paul R. Paquin, Charles Delos,
Luis Cruz, Dave Moon, Lareina Guenzel, Sarah Johnson, Blake
Beyea
Abstract:
The Biotic Ligand Model (BLM) was recently used as the basis for an update to
the U.S. EPA water quality criteria for copper (Cu). The BLM has been widely
tested and shown to be predictive of Cu toxicity to a number of freshwater
organisms under a variety of environmental conditions. Application of the BLM for
calculating water quality criteria (WQC) typically results in time-variable WQC.
Time-variable WQC are not unique to the BLM, and would result from any criteria
approach that depends on water quality (such as ammonia criteria or hardness
equations for metals). However, the widespread use of the BLM has renewed
interest in developing an approach that can consider variability when setting
permit limits or other regulatory guidelines. We have developed a fixed
monitoring benchmark (FMB) approach, which is a probability-based method that
incorporates time variability in BLM-predicted instantaneous water quality criteria
(IWQC) and in-stream Cu concentrations. The phrase “fixed monitoring
benchmark” is used, rather than “fixed site criteria” because it reflects the
purpose of this value, being that it is a benchmark that can be used to evaluate
compliance with WQC. This approach relies upon the distribution of toxic units
(TU), where each TU is calculated for a single sample using the Cu concentration
and IWQC for this sample. The calculation of TU requires that BLM input
parameters and the measured dissolved Cu concentration are available for each
sample. The distribution of TU values for all of the samples collected at a site is
then used to estimate the probability that an in-stream Cu concentration equals
or exceeds its associated IWQC (i.e. the probability that TU 1). Although
understanding whether the current Cu concentrations are above or below IWQC
values is useful, the intent of the FMB approach is to determine a revised Cu
distribution, defined such that the resulting low exceedence frequency is
consistent with the level of protection that is intended by EPA WQC guidelines.
This approach is straightforward and makes use of time variable BLM-based
WQC in regulatory contexts where a single criterion value is consistent with past
practices and established methods. Some states may be interested in adopting
FMBs as site-specific criteria to establish a more technically sound water quality
target as a basis for permitting and assessment decisions (e.g., at sites where
modifying the previous hardness-dependent criteria is a high priority).
Sessions: Metals Bioavailability Models: From Theory to Practice
Evaluation and Refinement of a Freshwater Biotic Ligand Model for Lead
Adam C. Ryan, Robert C. Santore, Paul R. Paquin, Martin Grosell,
Kevin V. Brix, Robert Gerdes, Edward M. Mager
Beyea
Abstract:
The biotic ligand model (BLM) for lead (Pb) was refined with recently available
Pimephales promelas and Ceriodaphnia dubia toxicity data. The data
considered here, evaluated the effects of pH, dissolved organic carbon (DOC)
concentration, Ca2+, and Na+ on Pb toxicity. A clear effect of DOC concentration
on Pb toxicity was demonstrated, and this effect was predicted reasonably well
by the Pb BLM. The data also demonstrated that Ca2+ was clearly protective of
Pb toxicity for P. promelas, but this was not as apparent for C. dubia. An
increase in the apparent binding strength of Ca2+ at the biotic ligand improved
BLM predictions. The effect of pH on Pb toxicity was not clear, partly due to the
presence of confounding variables and the limited pH range tested, but
introduction of PbOH+ as a bioavailable species improved BLM predictions.
Saturation index calculations suggested that many of the Pb toxicity studies that
were previously published have reported effect concentrations that were in
excess of solubility. Toxicity data utilized in this work were for sensitive species
and life stages, and corresponded to low levels of Pb that were generally below
solubility limits.
Session: BSAG: Relating laboratory and field bioaccumulation; importance of abiotic and biotic parameters, and use of trophic magnification factors
In-situ and Laboratory-derived BSAFs for PCBs, Dioxins and Furans in New York-New Jersey Harbor
Kevin J. Farley, Robin L. Miller, James R. Wands
Abstract:
BSAFs for PCB homologs and dioxin/furan congeners were computed using: (1)
US EPA/Army Corps of Engineers dredged material testing data for Nereis, (2)
NOAA sediment laboratory exposures for Armandia, and (3) in situ
measurements of sediment and un-classified worms that were collected as part
of the NY-NJ Harbor Contaminant Assessment Reduction Project (CARP).
Computed BSAFs were found to vary widely (0.05 to 5 kg (organic carbon)/kg
(lipid)) as a function of both chemical hydrophobicity and sample location.
Variations in BSAF with hydrophobocity generally followed expectations with diand
tri-CBs displaying low BSAFs, tetra- through hepta-CBs showing increased
BSAF values (suggesting dietary exposure as a dominant exposure route), and
higher chlorinated PCBs and dioxin/furan congeners exhibiting decreased BSAFs
(suggesting reductions in transfer through biological membranes for the larger,
more hydrophobic compounds). Spatial variations in BSAFs also showed
consistent trends for all three datasets, with the cleaner, outer-harbor sites
exhibiting higher BSAF values and the more contaminated, inner-harbor sites
showing reduced BSAF values. Reasons for spatial variations in BSAFs are not
known. However, differences in organic matter/food quality and contaminant
concentrations are offered as possible explanations.
Session: Fate and Effect of Metals: Toxicity and Regulatory Implications
Sediment Toxicity in the Upper Columbia River (USA) in Relation to Acid Volatile Sulfide and Simultaneously Extracted Metals
D. Scott Becker, Rick Cardwell, Robert C. Santore, Paul R. Paquin
Abstract:
The Upper Columbia River (UCR) extends for approximately 150 miles in
Washington State, from the Grand Coulee Dam to the U.S.-Canadian border.
Construction of the dam in 1942 created Lake Roosevelt, which accounts for
much of the UCR. Sediments in parts of the UCR have been influenced by a
range of anthropogenic sources, including releases from mining, milling and
smelting processes (e.g., smelter slag). Several divalent metals have been found
to be associated with the slag, including cadmium, copper, lead, and zinc. A
remedial investigation initiated in 2005 includes sediment toxicity evaluations as
an important component. A key consideration in these evaluations is the
bioavailability of the divalent metals. As part of the remedial investigation, 50
stations were sampled throughout the length of the UCR, and toxicity was
evaluated using the standard 28-d amphipod test based on Hyalella azteca, a
test species known to be sensitive to metals toxicity. A variety of physical and
chemical variables were also measured in the sediments, including variables
related to bulk sediments [i.e., metals concentrations, acid volatile sulfide (AVS),
simultaneously extracted metals (SEM), total organic carbon (TOC)] and
porewater (i.e., dissolved metals concentrations, various other constituents). This
presentation describes preliminary evaluations of the 2005 sediment toxicity data
with respect to bioavailability of the divalent metals in UCR sediments. Included
are comparisons of the sediment toxicity results with analyses based on AVS,
SEM, and TOC concentrations in bulk sediments, as well results of a preliminary
application of the biotic ligand model (BLM) to dissolved metals concentrations in
porewater. The results of these evaluations indicate that although bulk metals
concentrations in some sediments are relatively high (i.e., they exceed sediment
quality guidelines), the bioavailability of the metals is limited, based on the
general lack of lethal effects found in the amphipod test, as well as the relatively
close agreement between the observed sublethal effects (i.e., reduced amphipod
biomass) and toxicity predictions based on the AVS/SEM and BLM analyses.
Session: Fate and Effects of Metals: BLM
The Development of an Estuarine and Marine Biotic Ligand Model for Copper
Dr. Adalto Bianchini, Robert C. Santore
Abstract:
Water parameters such as pH, hardness, ionic composition, and sulfide and
dissolved organic matter concentrations have been shown to influence copper
bioavailability and toxicity. The protective effects of these various water chemistry
parameters have been modeled in the freshwater Biotic Ligand Model (BLM) for
copper. This model simultaneously accounts for the speciation and complexation
of dissolved metal and competitive binding of metal and other cations at the site
of action. It has now been adopted (U.S., Australia/New Zealand Environmental
Protection Agencies) or is in the process of being adopted (Environment Canada,
the European Union) as a legal tool for environmental regulation of metals in the
aquatic environment. Despite the important differences in the amount of possible
copper ligands (Cl-, SO4
2-, NOM, S2O3
2-, sulfide, Br-, and B(OH)4
-) or competitors
(Na+, Mg2+, Ca2+, K+, and Sr2+) in freshwater, brackish, and seawater, copper
toxicity studies and attempts to validate the BLM model in estuarine and marine
animals are scarce. At present, we are modeling toxicity and water chemistry
data from several experimental studies to further develop and evaluate a BLM
version for estuarine and marine environments. Data are being compiled from
toxicity studies with different invertebrate species (copepods, isopods and crabs)
in a wide range of salinity (freshwater to seawater) in either the absence or the
presence of different concentrations of dissolved organic matter from different
sources. These data are being modeled using a recently developed marine BLM,
which uses a model of copper speciation including binding to natural organic
matter that is suitable for high-salinity environments. The biotic ligand parameters
were calibrated to accumulation data for two marine invertebrates. The marine
BLM for copper has been tested using toxicity data from several estuaries in the
U.S. However, most of the available toxicity data were collected using organisms
that require test conditions near full strength marine salinities (such as Mytilus).
Since this work will focus on toxicity data for organisms acclimated over a range
of salinities, it will extend the applicability of the BLM to variable salinity
environments.
Session: Equilibrium Partitioning (EqP) Approach – Considerations for Use in Risk Assessments
Bioavailability of Metals in Sediments of the Upper Columbia River, Washington
Paul R. Paquin, Robert C. Santore, Mark Velleux, Kevin J. Rader, D. Scott Becker, Rick Cardwell
Abstract:
The Upper Columbia River (UCR) extends for approximately 150 miles in
Washington State, from Grand Coulee Dam to the U.S.-Canadian border.
Construction of the dam in 1942 led to the formation of an impoundment, Lake
Roosevelt, which covers most of the UCR site. Sediments in parts of the UCR
have elevated concentrations of some metals, including cadmium, copper, lead,
and zinc. A remedial investigation was therefore initiated in 2005 to investigate
various aspects of the UCR site, including sediment chemistry and toxicity.
Sediment toxicity was evaluated with the standard 28-d test with Hyalella azteca
(a metal-sensitive amphipod), using sediment samples from 50 locations
throughout the UCR study area. A variety of physical-chemical measurements
were also made on the sediments used for toxicity testing, including bulk
sediment measurements [i.e., metals concentrations, acid volatile sulfide (AVS),
simultaneously extracted metals (SEM), total organic carbon (TOC)] and
porewater (i.e., dissolved metals concentrations, various other constituents). This
presentation summarizes the results of an initial evaluation of the 2005 sediment
chemistry and toxicity data as well as a second similar, though somewhat
smaller, UCR dataset collected in 2004. Although SEM exceeded AVS in many
of the UCR sediment samples, sublethal effects were only reported for a limited
number of samples. In general, when SEM exceeds AVS, there is a tendency for
porewater metal concentrations to increase relative to levels expected when
excess AVS is present. For the UCR data, the porewater metal concentrations
were at times elevated to levels that exceeded their respective hardness-based
chronic water quality criteria (WQC). The related sums of interstitial water toxic
units (IWTU) were often > 1, even when effects were not observed. When the
analysis was repeated with IWTUs evaluated on the basis of benchmarks based
on the biotic ligand model (BLM) (i.e., WQC or effect levels for sensitive species),
to account for the bioavailability of porewater metals, the results were more
consistent with the absence of effects that was observed for most samples.
Session: Aquatic Toxicology - General - Part 2
Influence of Dissolved Organic Carbon on Chronic Toxicity of Copper to Juvenile Freshwater Mussels (Villosa iris) and Cladocerans (Ceriodaphnia dubia)
Ning Wang, William Brumbaugh, Chris Ingersoll, James Kunz, Ray
Arnold, Joseph Gorsuch, Chris Mebane, Robert C. Santore
Abstract:
Our previous study indicated that dissolved organic carbon (DOC) influenced the
acute toxicity of copper to juvenile freshwater mussels, and that the biotic ligand
model (BLM), used in the derivation of the U.S. EPA acute water quality criteria
for copper, reliably predicted the acute toxicity of copper to mussels. However,
little is known about the influence of DOC on the chronic toxicity of copper to
mussels. The objectives of this study were to (1) assess chronic 28-d toxicity of
copper to 2-month-old rainbow mussel (Villosa iris) at four nominal
concentrations of DOC (0.5, 2.5, 5, and 10 mg C/L) in flow-through diluter
systems, and (2) evaluate the BLM prediction of chronic copper toxicity to
mussels. A 7-d static-renewal, copper toxicity test with daphnids (Ceriodaphnia
dubia) was also conducted concurrently at the four DOC concentrations to
determine the sensitivity of mussels relative to this commonly tested species. An
extract of Suwannee River natural organic matter was mixed with diluted well
water (hardness 100 mg/L, pH 8.3) to prepare a series of DOC concentrations for
copper toxicity testing. At the end of the tests, 20% effective concentrations
(EC20s) for dissolved copper increased from 12 to 58 ?g/L for mussel survival
and growth (length), from 11 to 272 ?g/L for daphnid survival, and from 10 to 35
?g/L for daphnid reproduction across DOC concentrations of 0.5 to 10 mg/L. The
EC20s for mussel survival were generally below the EC20s for daphnid survival
at the various DOC concentrations. However, the EC20s for mussel survival or
growth were equal to or greater than the EC20s for daphnid reproduction. The
results of this study indicate that (1) increasing DOC concentrations reduced the
chronic toxicity of copper to mussel survival and growth, and (2) a threshold
based on the reproduction endpoint of daphnids would be protective of long-term
survival and growth of rainbow mussel, and perhaps of other species of
freshwater mussels. The BLM predictions of chronic copper effective
concentrations for the two test species are ongoing.
Session: Fate and Effects of Metals: Environmental Fate and Modeling
Performance Assessment of Chemical Speciation Models
Kevin J. Rader, Richard F. Carbonaro, Kevin J. Farley, Dominic DiToro,
Abstract:
Accurate determination of metal aqueous speciation and partitioning to solids is
essential to a mechanistic description of their transport, toxicity, and fate in
aquatic environments. Some of the key ligands and binding phases for metals in
surface waters and sediments include dissolved organic carbon (DOC),
particulate organic carbon (POC), and oxides of iron and manganese. Speciation
models such as MINEQL+, Visual MINTEQ, and WHAM6 have been developed
to quantify metal speciation between these phases. The algorithms of these
models have been incorporated into risk assessment model such as the Biotic
Ligand Model (BLM) and Unit World Model for Metals as well as reactive
transport models such as TICKET and PHREEQC. The accuracy of these
speciation algorithms cannot be taken for granted. While they have, in many
instances, undergone rigorous calibration and testing in highly controlled
laboratory systems, the amount of testing against field speciation data is usually
not as extensive. Because of the additional complications introduced, it is
important to undertake these performance assessments using field data to
determine the reliability of speciation models and ways in which they can be
improved. The work presented here examines the application of speciation
models to field datasets for the purpose of performance assessment. Data from
the Trinity River in Texas and other systems are included. Inconsistencies and
limitations are highlighted including poor performance when models are applied
outside of the calibration range, difficulty in handling competition effects, and
failure to consider pertinent processes. Potential methods of model refinement
are suggested.
Session: Fate and Effects of Metals: Environmental Fate and Modeling
Are We Ready to Incorporate Manganese Oxides into Unit World Models for Metals?
Richard F. Carbonaro, Kevin J. Farley Kevin J. Rader
Abstract:
Manganese(III,IV) hydroxides and oxides (hereon after referred to as Mn oxides),
play an important role in metal cycling in natural systems, acting as adsorbents
and oxidants. In the natural environment, Mn oxides are typically present as
surface coatings, and fine-grained aggregates with large surface areas. As a
result, small amounts of Mn oxides on a mass basis may be capable of
significantly influencing the distribution of metals between the dissolved and
particulate phases. Partitioning of metal to Mn oxides has been examined
experimentally for the past thirty years by a number of investigators. Based on
this work, metal partitioning has been shown to be dependent on the specific
mineral phase of Mn oxide, the solution chemistry, and the reactivity of the metal.
Several surface complexation models have been developed to describe the
partitioning of metal onto Mn oxides. However, the formulation of a global
partitioning model for Mn oxides has been limited by several factors. First,
manganese is known to form more than 30 known oxides and hydroxide solid
phases, each exhibiting unique adsorptive properties. Second, although a
relatively large number of proton-binding and metal partitioning data sets are
available for Mn oxides, only a few data sets have concurrent proton titration data
and metal adsorption data. Finally, the various surface complexation models
proposed for metal sorption onto Mn oxides have been calibrated for a limited
number of metals and are mostly incompatible with one another. Recent work to
incorporate a manganese cycle and Mn oxide partitioning model into the
TICKET-UWM (unit world model) for metals will also be discussed.
