United States Science Advisory Board EPA-SAB-EPEC-00-006
Environmental (1400A) February 2000
&EPA
AN SAB REPORT: REVIEW
OF THE BIOTIC LIGAND
MODEL OF THE ACUTE
TOXICITY OF METALS
www.epa.gov/s
PREPARED BY THE ECOLOGICAL
PROCESSES AND EFFECTS
COMMITTEE OF THE SCIENCE
ADVISORY BOARD
-------
February 28, 2000
EPA-SAB-EPEC-00-006
Honorable Carol M. Browner
Administrator
U.S. Environmental Protection Agency
1200 Pennsylvania Ave., NW
Washington, DC 20460
Subj ect: Review of the Biotic Ligand Model of the Acute Toxicity of Metals
Dear Ms. Browner:
At the request of the EPA Office of Water, the Ecological Processes and Effects Committee
(EPEC) of the Science Advisory Board (SAB) met on April 6-7, 1999 to review the Biotic Ligand
Model (BLM) for predicting the acute toxicity of metals to aquatic organisms. The BLM has been
developed to improve the estimation of the bioavailable fraction of dissolved metals, such as copper
and silver, that may pose a risk to aquatic organisms in surface waters. The Agency proposes to
incorporate the BLM in its approach to establishing water quality criteria that will be protective of
aquatic organisms.
The relevant scientific questions surrounding use of the BLM include: the extent to which it
realistically models metals chemistry, and thus bioavailability, of metals in the water column; the extent
to which it accurately captures the major exposure routes and mechanisms of toxicity of metals to water
column organisms; and the extent to which the BLM represents an improvement over existing methods
for developing and adjusting water quality criteria. The Committee's evaluation of these points,
summarized in the accompanying SAB report, form the basis for responding to the specific charge
questions posed by the Office of Water.
In general, the Committee found that the BLM can significantly improve predictions of the acute
toxicity of certain metals across a range of water chemistry parameters. The theory and empirical
validations performed to date are an important step toward a geochemically and biologically robust
approach for incorporating bioavailability concepts into water quality criteria. The Committee
concluded that the model could be a practical aid in site-specific water quality regulation and
assessment, complementing and in some cases providing a ready alternative to current empirical (e.g.,
Water-Effect Ratio) approaches. Over the longer term, the Committee sees potential broader
applications, including: the development and application of sediment quality guidelines; site-specific risk
-------
assessments and remediations; natural resource damage assessments; chemical registrations; and
product risk assessments.
-------
At the same time, the Committee observed that there has not yet been sufficient time to validate
the model in a number of areas, including: a broad range of aquatic organisms; longer term exposures; a
wide variety of metals; or a comprehensive range of water chemistry parameters and naturally occurring
field conditions. Particular areas in need of further validation were identified. We also provide
recommendations regarding appropriate use of the BLM pending this additional validation.
We encourage the Agency to continue its efforts to incorporate the best science and models
into its approach to deriving and adjusting water quality criteria. We hope our review of the BLM will
be helpful in that regard, and we look forward to a response from the Assistant Administrator for
Water.
Sincerely,
/signed/
Dr. Joan M. Daisey, Chair
Science Advisory Board
/signed/
Dr. Terry F. Young, Chair
Ecological Processes and
Effects Committee
Science Advisory Board
/signed/
Dr. Charles A. Pittinger, Acting Chair
for Review of the BLM
Ecological Processes and
Effects Committee
Science Advisory Board
-------
NOTICE
This report has been written as part of the activities of the Science Advisory Board, a public
advisory group providing extramural scientific information and advice to the Administrator and other
officials of the Environmental Protection Agency. The Board is structured to provide balanced, expert
assessment of scientific matters related to problems facing the Agency. This report has not been
reviewed for approval by the Agency and, hence, the contents of this report do not necessarily
represent the views and policies of the Environmental Protection Agency, nor of other agencies in the
Executive Branch of the Federal government, nor does mention of trade names or commercial products
constitute a recommendation for use.
Distribution and Availability: This Science Advisory Board report is provided to the EPA
Administrator, senior Agency management, appropriate program staff, interested members of the
public, and is posted on the SAB website (www.epa.gov/sab). Information on its availability is also
-------
provided in the SAB's monthly newsletter {Happenings at the Science Advisory Board). Additional
copies and further information are available from the SAB Staff.
-------
U.S. ENVIRONMENTAL PROTECTION AGENCY
SCIENCE ADVISORY BOARD
ECOLOGICAL PROCESSES AND EFFECTS COMMITTEE
Review of the Biotic Ligand Model
April 6-7, 1999
ACTING CHAIR
Dr. Charles A. Pittinger, Procter and Gamble Co., Ivorydale Technical Center, Cincinnati, OH
MEMBERS
Dr. Miguel F. Acevedo, University of North Texas, Denton, TX
Dr. William J. Adams, Kennecott Utah Copper Corp., Magna, UT
Dr. Lisa Alvarez-Cohen, University of California-Berkeley, Berkeley, CA
Dr. Steven M. Bartell, Cadmus Group, Inc., Oak Ridge, TN
Dr. Kenneth W. Cummins, Tarpon Bay Environmental Lab, South Florida Water Management
District, Sanibel, FL
Dr. Carol A. Johnston, Natural Resources Research Institute, Univ. MN, Duluth, MN
Dr. Paul A. Montagna, Marine Science Institute, University of Texas at Austin, Port Aransas, TX
Dr. Frieda B. Taub, School of Fisheries, University of Washington, Seattle, WA
Dr. Terry F. Young (EPEC Chair), Environmental Defense Fund, Oakland, CA
FEDERAL EXPERTS
Dr. Samuel N. Luoma, U.S. Geological Survey, Menlo Park, CA
SCIENCE ADVISORY BOARD STAFF
Ms. Stephanie Sanzone, Designated Federal Officer, EPA Science Advisory Board (1400A), 1200
Pennsylvania, NW, Washington, DC
111
-------
Ms. Mary L. Winston, Management Assistant, EPA Science Advisory Board (1400A), 1200
Pennsylvania, NW, Washington, DC
TABLE OF CONTENTS
1. EXECUTIVE SUMMARY 1
2. INTRODUCTION 4
2.1 Background 4
2.2 Charge to the Committee 5
3. GENERAL COMMENTS 7
4. RESPONSE TO SPECIFIC CHARGE QUESTIONS 11
REFERENCES CITED R - 1
APPENDIX A: ACRONYMS AND ABBREVIATIONS A - 1
IV
-------
1. EXECUTIVE SUMMARY
The BLM is a mechanistic bioavailability model which considers the influence of both biotic and
abiotic (organic and inorganic) ligands in the calculation of the bioavailability of metals to aquatic
organisms. The U.S. EPA's Office of Water requested that the Ecological Processes and Effects
Committee (EPEC) of the Science Advisory Board review the current state of validation of the BLM
relative to the current dissolved metal concentration criteria. Specifically, the Committee was asked to
comment on the model's proposed application for deriving and adjusting aquatic life criteria for metals
as part of the Agency's proposed integrated approach for the water column, sediments and interstitial
water. This report summarizes conclusions reached by the Committee following review of the draft
document, Biotic Ligand Model of the Acute Toxicity of Metals, and Agency presentations and
discussions at the April 6-7, 1999 review meeting.
In general, the Committee found that the BLM can significantly improve predictions of the acute
toxicity of certain metals across an expanded range of water chemistry parameters compared to the
WER. The theory and empirical validations performed to date were viewed as an important step
toward a geochemically and biologically robust approach for incorporating bioavailability concepts into
water quality criteria. The Committee concluded that the model could be a practical aid in site-specific
water quality regulation and assessment, complementing and in some cases providing a ready alternative
to current empirical (e.g., Water-Effect Ratio) approaches. Over the longer term, the Committee sees
potential broader applications, including: the development and application of sediment quality guidelines;
site-specific risk assessments and remediations; natural resource damage assessments; chemical
registrations; and product risk assessments.
At the same time, the Committee observed that there has not been sufficient time to validate the
model with what would be considered a broad range of aquatic organisms, a wide variety of metals
partitioned across key environmental compartments, or a comprehensive range of water chemistry
parameters and naturally occurring field conditions. Particular areas in need of further validation were
identified. It was recommended that the model's use in regulatory contexts be guided by the Agency to
ensure that any forthcoming applications are supported by and conform with the then-current state of
validation.
The Committee responded to four specific charge questions posed by the Agency:
1. Does the BLM improve the Agency's ability to predict toxicity to water column
organisms due to metals (copper and silver) in comparison to the currently
applied dissolved metal concentration criterion?
In comparison to the currently applied dissolved metal concentration criterion, the BLM has to
date been shown to predict with reasonable accuracy (generally within a factor of two of measured
1
-------
values) the acute toxicities of copper and silver to fish. Currently, the model has been shown to predict
acute toxicity, to a limited number of water column organisms, for selected metals, under equilibrium
conditions.
2. Is the scientific and theoretical foundation of the model sound?
The scientific underpinnings of the BLM appear to be sound. The strength of the model lies in
the fact that it is built upon a mechanistic paradigm with a strong physiological basis, i.e., predicting
binding at the site of action (gill in fish) and the mechanism of acute toxicity (blockage of sodium and
calcium channels) for metals (copper and silver).
3. In comparison to the current WER adjustment for aquatic life criteria, will the
application of the BLM as a site-specific adjustment reduce uncertainty
associated with metals bioavailability and toxicity?
Application of the BLM would not necessarily reduce uncertainty relative to empirical data.
However, its predictiveness over a wide range of environmental conditions makes the BLM a more
versatile and effective tool for deriving site-specific WQC compared to the WER.
4. Are the data presented for validation of the BLM sufficient to support the
incorporation of the BLM directly into copper and silver criteria documents?
It appears premature to use the BLM to revise the protocol for deriving national ambient water
quality criteria at this time, primarily because the model has not yet been validated for a sufficiently
diverse set of aquatic organisms and endpoints, coupled with the full range of water quality conditions.
However, the Committee concluded that the BLM could have current, practical applications for
calculating site-specific modifications to ambient copper criteria, and to a lesser extent silver, as an
alternative or complementary method to the current water-effects ratio (WER) approach.
Finally, the Committee provided recommendations for further research to provide additional
validation in the following areas:
a) Prediction of chronic and sub-acute toxicities, not currently supported by the BLM.
b) Broadening the supporting database to include greater taxonomic and functional
diversity and additional comparisons with the water effect ratio method.
c) Gaining better mechanistic and kinetic understanding, i.e., distinguishing relative
differences in binding affinity and toxicity mitigation among hardness cations (e.g., Ca,
Mg, Mn) and with other "biotic ligands" besides the fish gill, and evaluating
predictability of the model under non-equilibrium water quality conditions.
-------
d) Distinguishing the events and kinetics of DOC complexing with divalent cations and
biological uptake to improve interpretations of model predictions.
e) Applicability to multiple metals, and sensitivity analyses for varying water chemistry
conditions.
-------
2. INTRODUCTION
2.1 Background
The Biotic Ligand Model (BLM) is a model that incorporates metal speciation and the
protective effects of competing cations to predict metal binding at the fish gill or other site of action of
acute metal toxicity in aquatic organisms (i.e., the "biotic ligand") (Figure 1). The Agency has proposed
that the BLM be included in an integrated approach to metals management, including establishment of
metals water quality criteria. National ambient water quality criteria (WQC) consist of 3 components:
the concentration of the pollutant that will protect 95% of aquatic species; a time period over which
exposure is to be averaged; and the allowable frequency for exceeding the criteria. The allowable
concentrations of the pollutant generally are based on laboratory toxicity tests using a specified array of
test species, and are expressed in terms of a criterion maximum concentration (CMC) to protect
against acute (short-term) effects and a criterion continuous concentration (CCC) to protect against
chronic (long-term) effects.
Dissolved Metals—In 1993, the Agency issued technical guidance specifying that national
aquatic life criteria for metals be derived using the relationship between toxicity and dissolved metal,
rather than total recoverable metal (EPA, 1993), in order to more accurately reflect the bioavailable
fraction of metal in test solutions. (The change to dissolved criteria requires the application of a
conversion factor to convert criteria previously expressed in terms of total recoverable metal.) In an
effort to further refine the assessment of metals bioavailability in the water column, the Agency has
subsequently supported development of the Biotic Ligand Model (BLM), which considers biotic
ligands, as well as abiotic organic and inorganic ligands, in the calculation of bioavailability of metals. In
the current review, the SAB is asked whether the BLM will improve toxicity predictions in comparison
to the dissolved metal concentration
criterion.
Figure 1. Biotic Ligand Model
4
-------
Site-Specific Considerations—In the case of cationic metals, toxicity is hardness-dependent
because hardness ions compete with metal ions for binding sites on respiratory membranes. Thus, a
site-specific correction for hardness is incorporated into the national WQC for these metals by
expressing the CMC and CCC in the form of equations that yield the criteria values (in micrograms per
liter) as a function of total hardness. As an example, the equations for copper are as follows:
= g0.8545(ln hardness)- 1.702
= e0.9422(lnhardness) -1.700
In addition, the Agency has created three procedures by which national WQC may be modified to
reflect site-specific conditions relating to: differences in sensitivities of species at the site relative to
species used to derive the national criterion; differences in site water chemistry relative to laboratory
test water; and simultaneous consideration of both types of differences.
The Agency's current guidance on making the second type of correction, effects of site water
chemistry on toxicity, is called the Water-Effects Ratio (WER). Guidance on applying the WER
approach for metals was most recently issued in 1994 (EPA, 1994). In the current review, the SAB is
asked whether a model-based approach (the BLM) would be an improvement over the empirically
based WER for making site-specific adjustments of WQC for metals.
2.2 Charge to the Committee
The Committee met in Washington, DC on April 6-7, 1999 to review the draft document,
Biotic Ligand Model of the Acute Toxicity of Metals (EPA, 1999); the BLM is proposed for
incorporation into the Agency's approach for deriving aquatic life criteria for metals in the water
column. The Charge to the Committee included the following questions:
a) Does the BLM improve the Agency's ability to predict toxicity to water column
organisms due to metals (copper and silver) in comparison to the currently applied
dissolved metal concentration criterion?
b) Is the scientific and theoretical foundation of the model sound?
c) In comparison to the current WER adjustment for aquatic life criteria, will the
application of the BLM as a site-specific adjustment reduce uncertainty associated with
metals bioavailability and toxicity?
d) Are the data presented for validation of the BLM sufficient to support the incorporation
of the BLM directly into copper and silver documents?
-------
The Committee's response to each of these questions is contained in the sections that follow.
The Committee's comments on the suitability of the BLM for applications involving sediments and
interstitial (pore) water are contained in a companion SAB report on the Metals Mixtures Equilibrium
Sediment Guideline (ESG) (EPA-SAB-EPEC-00-007) and summarized briefly in Section 3.
-------
3. GENERAL COMMENTS
The BLM is a mechanistic bioavailability model that has to date been shown to significantly
improve our ability to predict the acute toxicity of certain metals across an expanded range of water
chemistry parameters. The theory and empirical validations performed to date are an excellent
beginning in the development of a geochemically and biologically robust approach for incorporating
bioavailability concepts into water quality criteria. The model, which will be an important component of
the Agency's integrated metals methodology (discussed further in EPA-SAB-EPEC-00-005, SAB,
2000), opens the way for what may be the next generation of water quality criteria for metals in the
form of tissue-residue based guidelines. In the near-term, the model can be a practical aid in site-
specific water quality regulation and assessment, complementing and in some cases (e.g., in predicting
acute toxicity of copper to water-column organisms) providing a ready alternative to current empirical
(e.g., WER) approaches.
The distinguishing feature of the model, in contrast to approaches based only upon estimation of
free metal ions as the toxic species, is its capability to predict the competition of the free metal ion with
other cations (e.g., Ca, H) and other ligands (DOC) for binding with the "biotic ligand" (the site of
membrane transport and route of direct uptake of freely dissolved metals). The presence of these
cations and ligands in solution can mitigate toxicity in a predictable fashion based on their relative
concentrations and strengths of binding. The model allows changes in toxicity under equilibrium
conditions to be estimated across ranges of key water quality parameters (pH, alkalinity, hardness, and
DOC). Furthermore, through the model's ability to integrate the binding site density of the biotic ligand,
conditional stability constants for the metal-ligand complex and competing cations, and measured or
postulated water quality conditions, the acute lexicological effects of a metal in a broad range of waters
can be normalized to a common metric (e.g., gill-metal LC50). This unifying feature offers a powerful
and consistent approach to comparing potential effects of metals among differing surface waters and
changing conditions within a single water body.
The Committee emphasizes, however, that a fundamental assumption of the BLM is that toxicity
is driven by exposure to dissolved metal alone, rather than combined toxic effects from dissolved and
dietary exposures. While this may be largely true for acute effects, understanding of dietary exposure
(via uptake of metal-DOC complexes or bioaccumulated metals in aquatic prey species) may be
necessary to predict chronic effects for particular metals. Suspended particulates include algae and
detritus which are consumed by aquatic organisms (such as grazers and filter feeders) and may be a
major dietary route of exposure for some aquatic organisms.
Applications of the BLM that appear to be adequately supported for regulatory use (e.g.,
certain WQC, NPDES and TMDL calculations) are specifically described in responses to the charge
questions below. These are distinguished from applications which the Committee believes will require
further validation (e.g., for interstitial water; for silver) or, in other cases, critical evaluation as to
whether the BLM is appropriate and sufficiently robust (e.g., for predicting chronic toxicity; for a broad
7
-------
range of metals). Because the issues of adequate validation and the directions of further research are
pervasive across the charge questions, the following discussion summarizes the Committee's
recommendations in this area.
Additional Validation Needs
As the authors have acknowledged in the review materials, additional validation and verification
of the BLM is necessary before the full potential of the model can be realized. Given the relative
newness of the BLM, there has not been sufficient time to validate the model with what would be
considered a broad range of aquatic organisms, a wide variety of metals partitioned across key
environmental compartments, or a comprehensive range of water chemistry parameters and naturally-
occurring field conditions. Certain underlying physical, chemical, and ecological questions (e.g., the
composition and comparability of different sources of DOC, the significance of non-respiratory uptake
mechanisms) are distinct from the BLM model per se, yet will be fundamental to establishing
appropriate guidelines for the model's application and interpretation. There is a suite of factors that
should be investigated over time that may improve the model's predictability and improve our
understanding of its strengths or limitations.
This does not mean that the model should not be used at all in its present state, but rather that it
is not ready for indiscriminate use by the federal, regional, tribal or state authorities. The model's use in
regulatory contexts should be guided by the Agency to ensure that any forthcoming applications are
supported by and conform with the current state of validation. Development of specific guidance by the
Agency on the model's use, with flexibility to allow expanded applications in the future, would be
helpful in this regard.
Outstanding needs principally relate to:
a) Chronic and Sub-Acute Toxicity
The model does not account for chronic or sub-acute toxicity, although this may be
incorporated through future research. For metals that have a small acute-to-chronic ratio (ACR), such
as copper with an ACR of 2x, the model may predict chronic toxicity with simple adjustments in model
parameters. For other metals with moderate to large ACRs, and where the mode of toxicity appears to
be different for acute and chronic effects, there is some doubt as to the applicability of the model.
Furthermore, there is a need to evaluate acute toxicity contributed by non-dissolved metal species, and
acute mechanisms of action not directly related to impairment of physiological function at the gill
surface. Other methods may be required to adequately address other mechanisms of metal toxicity.
These could include mechanisms of toxicity unrelated to the gill; or toxicity linked to chronic uptake at
the gill at concentrations lower than those that affect the gill itself (e.g., damage to dermal surfaces such
as fin erosion, internal effects to the liver, gut epithelia, reproductive processes, or energetics that have
implications for populations).
-------
b) Broad Taxonomic Applicability
We recommend that additional testing be performed with a wider range of organisms
(freshwater and marine, vertebrate and invertebrate, pelagic and benthic, representing multiple
functional groups) and that additional studies be undertaken to compare water-effect ratios with
predicted toxicity values generated by the BLM (i.e., model verification with independent data sets).
Further, we recommend that a sensitivity analysis be performed with the model to identify those
variables to which the model is most sensitive.
c) Mechanistic and Kinetic Understanding
The model relies upon equilibrium assumptions, yet the importance of kinetic exchanges of
metal between ligands in the micro-environment near the gill membrane is unknown. Better
understanding of the relationship between biotic ligand metal concentration and expressed toxicity will
help to improve the model's predictiveness. Distinguishing relative differences in binding affinity and
toxicity mitigation among hardness cations (e.g., Na, Ca, Mg) will provide further improvements to the
model. Further elucidation of these differences could provide insight into the validity and applicability of
current, simplifying assumptions and models for metal toxicity that use hardness as a variable in site-
specific water quality criteria calculations. There are also outstanding questions regarding the
applicability of the model to other "biotic ligands" besides the fish gill (i.e., for non-piscine taxa of water
column-dwelling organisms), particularly invertebrates and marine species, which may require collection
of binding site density and conditional stability constant information specific to other taxa.
In addition to assumptions of molecular equilibrium at the biotic ligand, some data (e.g.,
pertaining to diluter cycling and toxicant delivery time; Ma et al, 1999) also raise questions of the
predictability of the model under non-equilibrium water quality conditions, such as would be expected
under dynamic field conditions (e.g., mixing zones, stream confluences).
d) DOC Complexing
Distinguishing the events and kinetics of dissolved organic carbon (DOC) complexing with
divalent cations and biological uptake will contribute greatly to interpretations of model predictions.
DOC is not a uniform constituent; DOC generation in surface waters is the result of a complex set of
degradation reactions which produce a myriad of dissolved organic ligands with different functionalities.
The BLM, however, only allows DOC to be specified as percent humic acid, with the remainder of
DOC modeled as fulvic acid. In testing the model's ability to predict copper toxicity, the authors
assumed DOC to be 10 percent humic acid. However, the percent of DOC present as humic acid will
vary with system type.
In addition, Dahm (1981) demonstrated that there is a distinct, two-phase depletion of organic
material introduced into surface water systems: phase 1 involves physical complexing with cations on
-------
the order of hours, and phase 2 involves biological (microbial) uptake and metabolism on the order of
several days. Yet, the model does not take microbial uptake and decomposition into account.
e) Applicability to Multiple Metals and Water Chemistry Conditions
There is a need to validate the applicability of the model to other metals besides copper, and
further work is needed to understand and resolve the larger uncertainties associated with silver toxicity
relative to copper. It may be important to incorporate additional environmental ligands, such as
suspended particulates, that are known to sorb divalent cations. The model should be tested under a
broader range of water chemistry (e.g., pH, Ca, Mg, DOC) conditions, including a range of sources of
DOC (e.g., from different waters, from interstitial versus overlying water, from allochthonous versus
autochthonous sources, etc.), and conditions where several water chemistry parameters are varied
simultaneously. A sensitivity analysis of the effects of varying these different water chemistry
parameters would be useful when adequate data are generated.
Seasonal and diurnal shifts in water chemistry parameters such as pH and temperature will
influence metals bioavailability, thus complicating efforts to obtain representative samples of site water
for toxicity tests and to determine which water chemistry data should be used to parameterize the
BLM. For example, Brick and Moore (1996) found that total acid-soluble metals concentrations,
including copper and zinc, in river water samples increased 2-3 fold at night relative to the day. The
Committee recommends, therefore, that further consideration be given to how water chemistry
conditions are characterized for use in the BLM.
f) Use of BLM-based WQC in Sediment Guidelines
In addition to the charge questions about the BLM addressed in this report, the Agency also
requested advice on the potential use of the BLM-derived or adjusted WQC in the interstitial water
component of the Metals Mixtures Equilibrium Sediment Guidelines (ESG). The Committee's
comments on use of BLM-based WQC in sediment guidelines are contained in a companion document
(EPA-SAB-EPEC-00-005, SAB, 2000). In short, the Committee concluded that because the
chemistry of interstitial water is not the same as the chemistry in the water column, it would be
inappropriate to substitute the BLM-adjusted water column criterion for the water quality criterion in
the ESG equation without additional validation experiments. The BLM, if validated for application to
interstitial water, would allow consideration of additional site-specific chemistry conditions that affect
metals bioavailability; the currently proposed IW component of the ESG is based on the water quality
criteria Final Chronic Value for each metal, corrected only for site-specific hardness.
10
-------
4. RESPONSE TO SPECIFIC CHARGE QUESTIONS
Charge Question 1. Does the BLM improve our ability to predict toxicity to water
column organisms due to metals (copper and silver) in comparison to the currently
applied dissolved metal concentration criterion?
In theory, the BLM should provide a more accurate measure of bioavailable metal than does
the use of operationally defined "dissolved" metal. The evidence collected to date for copper and silver
supports this theory for acute toxicity. In comparison to the currently applied dissolved metal
concentration criterion, the BLM has been shown to predict with reasonable accuracy (generally within
a factor of two of measured values) the acute toxicities of copper and silver to fish. However, this
question bears some qualification to accurately reflect the model's current state of validation and
verification. Currently, as the authors have attested, the model has been shown to predict acute
toxicity, to a limited number of water column organisms, for selected metals, under equilibrium
conditions. Preliminary results with Daphniapulex and with metals other than copper and silver are
promising, indicating that the model may well have broad applicability to a range of divalent metals and
taxa. The mechanistic framework that the model provides for predicting bioavailability could lead to
significant improvements over current criteria based upon dissolved metal concentrations.
Charge Question 2. Is the scientific and theoretical foundation of the model sound?
The scientific underpinnings of the BLM appear to be sound. The strength of the model lies in
the fact that it is built upon a mechanistic paradigm with a strong physiological basis, i.e., predicting
binding at the site of action (gill in fish) and the mechanism of acute toxicity (blockage of sodium and
calcium channels) for metals (copper and silver). Additionally, the model incorporates water quality
parameters such as DOC and pH for the first time, and it allows for prediction of toxicity under a
broader range of environmental variables. It is the strength of the science behind the model that makes
it attractive. While the model is built upon a strong fundamental and theoretical basis, it requires
additional validation and verification as discussed above.
Charge Question 3. In comparison to the current WER adjustment for aquatic life
criteria, will the application of the BLM as a site-specific adjustment reduce
uncertainty associated with metals bioavailability and toxicity?
Because the BLM is deterministic, its results may be infinitely precise, although possibly
inaccurate. To evaluate the extent to which the BLM will reduce uncertainty relative to the WER
approach, it is necessary to examine the variability (uncertainty) of model calculations, given realistic
variations in its input parameters. The resulting distribution of model results could then be compared
with the variability of the empirical data to begin answering this question.
11
-------
Although application of the BLM may not reduce uncertainty relative to empirical data, its
predictiveness over a wide range of environmental conditions makes the BLM a more versatile and
effective tool for deriving site-specific WQC than the WER. The reasons are twofold. First, empirical
data often are more convincing than modeled results, but may not be feasible to collect for all relevant
environmental conditions. The BLM can consider the full range of environmental variables (pH, DOC,
hardness, and alkalinity) that may exist at a given site over the course of a year (or longer). Second, the
model can be executed repeatedly in practical applications. WERs are limited (and imprecise) in that
they represent the water chemistry present at the time the water samples were collected and as such
cannot define the range of bioavailability conditions (e.g., that associated with seasonal or episodic
changes in water chemistry).
While there are remaining validation needs, as described in this report, the Committee supports
use of the BLM as an alternative to the WER method for developing site-specific WQC in some cases
(see Charge Question 4). We recommend that the model be utilized in parallel with the WER at sites in
order to generate paired data sets using both approaches. These additional comparisons between the
two approaches will be valuable for empirical verification, and the potential exists that at a given site,
organisms and/or field conditions will need to be tested for which the BLM has not been validated.
Current applications for the BLM and WER, in addition to adjusting WQC for site-specific
water chemistry, include effluent permitting under the National Pollutant Discharge Elimination System
and the Total Maximum Daily Load procedure. Over the longer term, when more fully validated, the
BLM may be useful in the development and application of sediment quality guidelines; in site-specific
risk assessments and remediation; in natural resource damages assessment; in chemical registrations;
and product risk assessments.
Charge Question 4. Are the data presented for the validation of the BLM sufficient to
support the incorporation of the BLM directly into copper and silver criteria
documents?
To address this question, it is important to clarify what is meant by "incorporation into the water
quality criteria documents." There are several possible interpretations, including: (1) using the model to
directly establish over-arching national water quality criteria (e.g., CMC and CCC calculations); and
(2) modifying those criteria for site- or regional-specific risk assessments or permitting. We
recommend somewhat different approaches for each.
a) National Ambient Water Quality Criteria
The current water quality criteria documents provide equations for tailoring metal criterion
values to different waters based only upon varying water hardness. Clearly, there are convincing data
and an emerging consensus that other water chemistry factors are or can be important in modifying
metal toxicity. The Office of Water's Water Quality Criteria and Standards Plan (EPA, 1998)
12
-------
states that "EPA will update the aquatic life criteria derivation methodology to reflect new science and
modeling capabilities." Given the importance of the decisions based upon these criteria, it is generally
understood that major revisions should be performed only when the Agency is convinced that the
revisions, including underlying models, are scientifically robust.
Current national ambient water quality criteria are based upon a statistical approach which, at
least theoretically if not empirically, incorporates the lexicological responses of a broad range of taxa to
contaminants. The criteria, however, do not excel at predicting responses across a wide range of field
conditions. Research on the BLM has underscored these limitations. On the other hand, it would be
impractical to generate databases representing the same taxonomic diversity, coupled with the full
range of water quality conditions for a BLM-based criterion for every metal. Thus, it appears both
premature and somewhat impractical to use the BLM model to revise the protocol for developing
national ambient water quality criteria at this time. It is possible, however, that significant enhancements
in the model's accuracy and applicability will be made in the near future so that the BLM can be
incorporated into the CMC and CCC protocols.
b) Site-Specific Modifications to Ambient Water Quality Criteria
While it may be premature to incorporate the BLM into the documents to derive over-arching
national water quality criteria, practical applications of the model for calculating site-specific
modifications would be feasible in some cases and have precedent. The Water-Effects Ratio method of
adjusting water quality criteria to specific waters, for example, was described in the 1994 Interim
Guidance on the Determination and Use of Water-Effect Ratios (WER) for Metals (EPA, 1994).
Similar guidance could be provided for prudent applications of the BLM to copper, and to a lesser
extent to silver, as described below.
Copper
With several important caveats, the data presented for the validation of the BLM are sufficient
to support its use for site-specific modification of the copper criterion. These caveats, listed below,
generally reflect the need to avoid applying the model to water quality conditions or regulatory
scenarios that are beyond the scope of the current validation effort. The Committee recognizes that
additional studies are underway, and the level of validation is expanding to allow additional applications.
Current limitations include the following:
(1) The model should presently be reserved for calculation of site-specific modifications to
acute toxicity criteria, i.e., in estimating the criterion maximum concentration. The data are not
sufficient to support the current use of the BLM in estimating chronic water quality criteria (i.e.,
the criterion continuous concentration). However, we recognize that for copper the CMC and
CCC are not widely different and the ACR is typically less than 2.0.
13
-------
(2) The model should presently be reserved for calculation of acute copper criteria for the
protection of freshwater organisms excluding algae and macrophytes. It should not
14
-------
be used in estuarine or marine applications until representative datasets have been used to
validate the model's applicability for these waters.
(3) Caution should be used in estimating acute copper criteria when the calculations would be
based solely upon predictions or measurements related to non-piscine biotic ligands (e.g.,
invertebrate respiratory structures). This is because there are less data currently available for
the metal-ligand complex (e.g., binding site densities and conditional stability constants) in
tissues of invertebrate species than for fish species. Additional data with sensitive species other
than daphnids would help to eliminate this concern.
Silver
The Committee is less confident in the accuracy of silver acute toxicity predictions than for
copper and we do not recommend that this approach be incorporated into the silver criteria document
and used to derive the CMC at this time. The principal reason for reservation at this time is that the
mechanisms by which silver binds to the biotic ligand under different conditions is not as well
understood as for copper (see comments below). The relationship between silver bioaccumulated at
the gill and silver toxicity is also unclear. The approach does appear to have merit as a complementary
method to the existing WER approach for site specific modifications of WQC, but should be used with
caution. The Committee recognizes that the WER approach also has limitations, namely that it utilizes a
limited number of species and does not properly account for temporal variations in water quality. Thus,
EPEC supports the continued development of the BLM. Parallel use of the model and WER
approaches initially would be useful to provide additional validation of the BLM relative to empirical
WERs.
There is no question that water chemistry parameters other than hardness, the sole parameter
currently recognized in the criteria document, do significantly impact silver toxicity. However,
quantitative predictions of acute silver toxicity using the model are more challenging than those with
copper for a number of reasons. First, it appears that the level of silver accumulation on the gill varies
over time, such that static equilibrium conditions are not necessarily achieved (Wood et al, 1999).
Second, only a relatively small fraction of the measured total accumulation of silver at the gill may be
associated with binding to physiologically active sites. The potential for accumulation of silver at other
physiologically important binding sites not represented by the model is possible. Third, as the Paquin et
al. paper in the review materials (EPA, 1999) indicates, variations exist in the strength of DOC binding
sites with silver concentration. For these reasons, it has been difficult to demonstrate a consistent
correlation of effects with total gill silver under a range of water quality conditions. Fourth, complexes
of the ligand with silver chloride appear to contribute slight toxicity in addition to silver-ligand
complexing. While this contribution has been incorporated into the present model, the toxicity of silver
chloride appears to vary among fish species, particularly at high chloride concentrations (Figure 7, p. 3-
97).
15
-------
The Committee agrees with Paquin et al. (in EPA, 1999) that independent verification of the
model in predicting acute silver toxicity with additional datasets is recommended to identify other
potential inorganic ligands (e.g., sulfide) that form complexes with silver, and to assess the accuracy of
the model under a broader spectrum of water quality conditions and aquatic organisms. We encourage
the regulatory and scientific communities as they further apply the BLM model to collect paired datasets
of water-effect ratio measurements and BLM measurements for copper, silver, and other metals to
allow for model improvement and validation.
16
-------
REFERENCES CITED
Dahm, C. N. 1981. Pathways and mechanisms for removal of dissolved organic carbon from leaf
leachate in streams. Can. J. Fish. Aquat. Sci. 38: 68-76.
Brick, C.M. and J.N. Moore. 1996. Diel Variations of Trace Metals in the Upper Clark Fork River,
Montana. Envir. Sci. Technol. 30:1953-1960.
Ma, H. S.D. Kim, O.K. Cha, and H.E. Allen. 1999. Effect of Kinetics of Complexation by Humic
Acid on Toxicity of Copper to Ceriodaphnia dubia. Envir. Toxicol. & Chem. 18:828-837.
Science Advisory Board (SAB). 2000. An SAB Report: Review of an Integrated Approach to Metals
Assessment in Surface Waters and Sediments. EPA-SAB-EPEC-00-005. USEPA Science
Advisory Board, Washington, DC
U.S. Environmental Protection Agency. 1993. Office of Water Policy and Technical Guidance on
Interpretation and Implementation of Aquatic Life Metals Criteria. Martha Prothro, Acting
Assistant Administrator for Water.
U.S. Environmental Protection Agency. 1994. Interim Guidance on Determination and Use of Water-
Effect Ratios for Metals. Office of Water and Office of Research and Development. February
1994.
U.S. Environmental Protection Agency. 1998. Water Quality Criteria and Standards Plan: Priorities
for the Future (Interim Final). EPA Office of Water, EPA 822-R-98-003.
U.S. Environmental Protection Agency. 1999. Integrated Approach to Assessing the Bioavailability
and Toxicity of Metals in Surface Waters and Sediments: Addendum (Biotic Ligand Model of
the Acute Toxicity of Metals). Office of Water and Office of Research and Development,
Washington, DC. Presented to the EPA Science Advisory Board, April 6-7, 1999.
Wood, C.M., R.C. Playle, and C. Hogstrand. 1999. Physiology and modeling of mechanisms of silver
uptake and toxicity in fish. Envir. Toxicol. & Chem. 18:71-83.
R- 1
-------
APPENDIX A: ACRONYMS AND ABBREVIATIONS
ACR Acute-Chronic Ratio
BLM Biotic Ligand Model
CCC Criterion Continuous Concentration
CMC Criterion Maximum Concentration
DOC Dissolved Organic Carbon
ESG Equilibrium Sediment Guidelines
NPDES National Pollutant Discharge Elimination System
TMDL Total Maximum Daily Load
WQC Water Quality Criteria
WER Water-Effect Ratio
A-l
------- |