oEPA
United States Office of Water EPA-820-D-24-002
Environmental Protection 4304T December 2024
Agency
DRAFT
Comparison of Aquatic Life Protective Values Developed for Pesticides under
the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) and
the Clean Water Act (CWA)
Technical Support Document
Prepared by:
U.S. Environmental Protection Agency
Office of Water & Office of Pesticide Programs
Washington, DC
December 2024
-------
Notices
This draft technical support document, "Comparison of Aquatic Life Protective Values Developed for
Pesticides under the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) and the Clean Water Act
(CWA)" presents analyses supporting the EPA's efforts to develop a common approach for assessing
potential pesticide toxicity effects under the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA)
and the Clean Water Act (CWA). This document was developed to support an evaluation to assess
whether aquatic life benchmarks developed by the Office of Pesticide Programs in support of
registration decisions for pesticides under FIFRA are appropriate to serve as CWA aquatic life
304(a)aquatic life values, either as 304(a)(1) recommended criteria or 304(a)(2) informational
benchmarks.
This technical support document does not substitute for the CWA or the EPA's regulations; nor is it a
regulation itself. The document does not impose legally binding requirements on the EPA, states, tribes,
or the regulated community, and might not apply to a particular situation based upon the
circumstances.
The EPA may update this document in the future. This document has been reviewed in accordance with
EPA policy and approved for publication. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
ii
-------
Authors, Contributors, and Reviewers
Christine M. Bergeron, Office of Water, Office of Science and Technology, Health and Ecological Criteria
Division, Washington, DC
Thomas Steeger, Office of Chemical Safety and Pollution Prevention, Office of Pesticide Programs,
Environmental Fate and Effects Division, Washington, DC
Douglas (Ethan) Harwood, Office of Chemical Safety and Pollution Prevention, Office of Pesticide
Programs, Pesticide Re-evaluation Division, Washington, DC
Kathryn Gallagher, Office of Water, Office of Science and Technology, Health and Ecological Criteria
Division, Washington, DC
-------
Executive Summary
The U.S. Environmental Protection Agency (EPA) has undertaken an effort to harmonize aquatic
effects assessment methods for pesticides to provide a common basis for evaluating the effects of these
chemicals on water quality under the Clean Water Act (CWA) and the Federal Insecticide, Fungicide, and
Rodenticide Act (FIFRA). The EPA previously solicited public input through a public comment period and
held national and regional stakeholder meetings early in the process of considering harmonized aquatic
effects assessments for pesticides, under a project entitled "OW/OPP Common Effects Methodology"
(docket number: EPA-HQ-OPP-2009-0773). In 2012, the FIFRA Scientific Advisory Panel reviewed the EPA
analyses regarding potential approaches and made recommendations for the EPA to move toward a
harmonized approach under FIFRA and CWA. EPA scientists from both the Office of Water (OW) and the
Office of Pesticides Programs (OPP) have worked together to develop the analyses currently being
released for public comment. This continued collaborative effort within the EPA ensures development of
protective aquatic life values using current science while minimizing duplicative work within the agency
and promoting consistency in aquatic effects assessments for pesticides. The EPA evaluated insecticides
and herbicides from different chemical classes and with different modes of action to determine whether
the OPP aquatic life benchmarks (ALBs) developed in support of registration decisions for pesticides
under FIFRA are similarly protective as potential CWA 304(a)(1) recommended criteria, and other
criteria-related values, and may thus be appropriate to serve as CWA aquatic life 304(a) aquatic life
protective values, either as 304(a)(1) recommended criteria or 304(a)(2) informational benchmarks.
Currently, OPP ALBs and CWA 304(a)(1) recommended Aquatic Life Criteria (ALC) values are
developed using parallel, but different, rigorously peer-reviewed methods to generate values protective
of aquatic communities for both acute and chronic effects. There are several similarities in the
approaches, for example both approaches: 1) use the Office of Research and Development's
ECOTOXicology (ECOTOX) Knowledgebase to identify open literature toxicity studies for chemicals, 2)
have similar toxicity data quality review approaches, and 3) use the same assessment endpoints
including acute survival and chronic survival, growth, and reproductive effects.
The two main differences in the approaches are regarding primary data sources and the methods to
calculate protective values. For data sources, OPP ALBs are extracted from the most recent publicly
available pesticide ecological risk assessments and largely use registrant-submitted studies
supplemented with open literature studies from ECOTOX. CWA 304(a)(1) ALC have historically been
developed based on data-quality reviewed, publicly available information primarily collected from
ECOTOX to fulfill the eight aquatic taxa minimum data requirements (MDRs) per the "Guidelines for
Deriving Numerical Water Quality Criteria for the Protection of Aquatic Organisms and Their Uses" (U.S.
EPA 1985) ("Guidelines") in addition to high quality, reviewed data collected under FIFRA in the form of
registrant-submitted data, when available.
Both the OW ALC and OPP benchmark approaches result in robust aquatic life protective values. In
support of pesticide registration decisions under FIFRA, the regulatory thresholds and ALBs are typically
based on high quality data for the most sensitive acceptable aquatic plant (vascular and nonvascular),
freshwater invertebrate and vertebrate species (acute and chronic data) for each taxon tested. For
purposes of implementing section 304(a)(1) of the CWA, ALC recommended values are typically
determined by regression analysis based on the four most sensitive genera averages in the data set to
calculate the 5th percentile value of the distribution represented by the tested genera averages to
determine values estimated to be protective of approximately 95% of aquatic genera.
iv
-------
There are currently 23 pesticides with ALC, more than half of which (15) are for pesticides no longer
in commerce. OPP ALBs for pesticides and selected degradates, however, address many of the currently
registered pesticides in commerce. Currently, there are over 750 OPP ALBs publicly available.
The EPA investigated several methods to compare OPP ALB to values that the OW has calculated for
pesticides based on ALC methods or other criteria-related approaches. The analyses show that OPP ALBs
and CWA aquatic life 304(a) values are similar (within a factor of 10). Specifically, case studies for eight
pesticides from several classes with existing 304(a)(1) ALC recommendations demonstrate that the most
sensitive OPP ALB for a given pesticide is generally somewhat lower (more sensitive) and mostly within a
factor of two of its existing 304(a)(1) ALC recommendations. The EPA also conducted analyses
comparing the OPP ALBs to criteria-related values for 26 pesticides and 5 herbicides from several
different classes derived using conservative methods or safety/assessment factors (e.g., using the Great
Lakes Initiative (GLI) approach developed for application in Great Lake states) for when data are limited
and MDRs are not met. These analyses also show that OPP ALBs are similar to values (within 5-10X)
developed using these criteria-related approaches applied when toxicity data are limited. This range in
values is approximately the same as the inherent variability observed in repeated toxicity tests on a
single species conducted within the same laboratory or across laboratories (Chapman 1998; Duke and
Taggart 2000; Fairbrother 2008; Raimondo et al. 2007; Raimondo et al. 2010). These draft analyses
support the conclusion that the OPP ALBs are similarly protective and appropriate for use in establishing
CWA aquatic life 304(a) protective values for pesticides.
v
-------
Table of Contents
1 Overview of the EPA Effort to Harmonize Pesticide Aquatic Effects Assessments 1
2 The EPA Methods for Aquatic Effects Assessments for Pesticides 1
2.1 Background 1
2.2 Similarities Between OPP and OW Approaches 1
2.3 Differences Between OPP and OW Approaches 2
2.4 Overview of the EPA OPP Aquatic Life Benchmarks 3
2.5 Overview of the EPA CWA National Recommended 304(a)(1) Aquatic Life Criteria 6
2.5.1 Acute Criteria 6
2.5.2 Chronic Criteria 7
2.5.3 ALC Duration and Frequency Aspects 7
3 Examination of Potential Approaches for Harmonizing OPP Benchmarks and CWA Section 304(a)
Effects Assessment Methods for Pesticides 8
3.1 Comparison of OPP Aquatic Life Benchmarks and CWA Aquatic Life Criteria for Pesticides 8
3.1.1 Acute Value Comparison 8
3.1.2 Chronic Value Comparison 11
3.2 Comparison of OPP Aquatic Life Benchmarks and Alternative Criteria-Related Approaches
When Data are Insufficient to Develop Aquatic Life Criteria 12
3.2.1 Overview of Great Lakes Initiative Approach to Develop Criteria-Related Values to Compare
to OPP ALBs 13
3.2.2 Overview of a Modified Guidelines Methods to Develop Criteria-Related Values to
Compare to OPP ALBs 14
3.2.3 Insecticidal Pesticides 14
3.2.3.1 Acute Values 14
3.2.3.1.1 Result of Analyses using Acute Modified Guidelines Methods and GLI Methodology
for Insecticides 15
3.2.3.2 Chronic Values 20
3.2.4 Herbicidal Pesticides 25
3.2.3.3 Acute Values 26
3.2.3.4 Chronic Values 27
3.3 Analysis of Ratios of OPP Aquatic Life Benchmarks and Alternative Criteria-Related Approaches
to Develop Aquatic Life Criteria 28
4 Summary and Conclusions 32
5 References 34
Appendix A: Acute Sensitivity Distributions for Pesticides and Various Protective Values 35
Appendix B: Acute and Chronic Sensitivity Distributions for Herbicides and Various Protective Values .. 47
vi
-------
1 Overview of the EPA Effort to Harmonize Pesticide Aquatic Effects Assessments
The Environmental Protection Agency's (EPA) objective for this effort is to harmonize aquatic effects
assessment methods to provide a common basis for evaluating the effects of pesticides on water quality
under the Clean Water Act (CWA) and the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA).
This collaborative effort within the EPA ensures development of protective aquatic life values using
current science while minimizing duplicative work within the agency and promoting consistency in
aquatic effects assessments for pesticides. The analyses presented in this document support leveraging
the most sensitive Aquatic Life Benchmarks (ALB) developed by the Office of Pesticide Programs (OPP) in
support of registration decisions for pesticides as CWA aquatic life 304(a) protective values, either as
304(a)(1) recommended criteria or 304(a)(2) informational benchmarks. Case studies for select
insecticides and herbicides from different chemical classes and with different modes of action described
in this document demonstrate that the EPA's process used to develop OPP ALBs generates protective
values for registered pesticides and would be equivalent to values developed as CWA 304(a)(1) ALC if
sufficient data were available to fill the eight minimum data requirements (MDRs) and generate criteria
using the Agency's traditional criteria development approach (U.S. EPA, 1985).
2 The EPA Methods for Aquatic Effects Assessments for Pesticides
2.1 Background
To develop aquatic effects assessments under the CWA and FIFRA, the EPA uses parallel but different
rigorously peer-reviewed methods to generate protective values for both acute and chronic effects.
Currently, EPA has derived national recommended 304(a)(1) Ambient Water Quality Criteria for the
protection of aquatic life ("Aquatic Life Criteria" or ALC) for 23 pesticides1. For FIFRA purposes, the EPA
assesses potential ecological risks of pesticides considering both terrestrial and aquatic effects data,
including data underlying the ALBs, combined with exposure modeling and monitoring data. OPP ALBs
are used by states and other stakeholders to evaluate water monitoring data and prioritize resources
with respect to priority pollutants to ensure protection of aquatic life. The EPA currently has over 750
registered pesticides and selected degradates with at least one OPP ALB (i.e., freshwater or
estuarine/marine vertebrate acute and chronic, freshwater or estuarine/marine invertebrate acute and
chronic, vascular aquatic plants acute and chronic, or nonvascular aquatic plants acute and chronic)
along with a link to the relevant Agency action (e.g., ecological risk assessment)2.
2.2 Similarities Between OPP and OW Approaches
There are many similarities in the OPP and OW approaches to generate values protective of aquatic
communities. First, both approaches rely on the EPA Office of Research and Development (ORD)
Ecotoxicology (ECOTOX3) Knowledgebase to identify and collect open literature toxicity studies for
1 See EPA's Aquatic Life Criteria table at https://www.epa.gov/wqc/national-recommended-water-quality-criteria-
aquatic-life- criteria-table.
2 See EPA's Aquatic Life Benchmarks at https://www.epa.gov/pesticide-science-and-assessing-pesticide"
risks/aquatic-life-benchmarks-and-ecological-risk.
3 ECOTOX (https://cfpub.epa.gov/eCOtox/) is a publicly available database summarizing the ecological effects of single
chemicals to aquatic and terrestrial plants and animals. ECOTOX was developed by EPA's Office of Research and Development's
Mid-Continental Ecology Division (ORD/MED), which routinely conducts literature searches for pesticides undergoing
Registration Review as well as for litigation-related endangered species assessments.
1
-------
chemicals. Relevant literature for ECOTOX is retrieved using a comprehensive search strategy designed
to locate literature worldwide on the toxicity of chemicals to a wide range of aquatic animal and aquatic
plant species. ECOTOX also includes unpublished registrant-submitted data in response to FIFRA testing
requirements for many chemicals. Second, the data quality review approaches for toxicity data have
been harmonized for FIFRA and CWA aquatic effects assessments. For FIFRA assessments, similar to
registrant-submitted studies, data identified in the open literature undergo review as specified in 2011
OPP guidance document entitled Evaluation Guidelines for Ecological Toxicity Data in Open Literature4 to
ensure they are consistent with standards specified in the Information Quality Act and subsequent
guidelines developed by the Office of Management and Budget (OMB) on ensuring and maximizing the
quality, objectivity, utility and integrity of information disseminated by federal agencies5. For ALC, the
EPA reviews studies identified in the open literature according to its Standard Operating Procedures for
Systematic Review of Ecological Toxicity Data6, which is consistent with the data quality review
procedures in the Guidelines for Deriving Numerical National Water Quality Criteria for the Protection of
Aquatic Organisms and their Uses7 (referred to as the "Guidelines"; U.S. EPA 1985) and subsequent
harmonization with FIFRA procedures. Lastly, the same assessment endpoints are used under both
statutes, and typically include acute survival and chronic survival, growth and/or reproductive effects.
2.3 Differences Between OPP and OW Approaches
There are two main differences in the OPP ALB and OW ALC approaches to generate protective values
regarding data sources and the methods to calculate protective values.
Data Sources: Under FIFRA, primary data sources are registrant-submitted studies in response to data
requirements identified in Title 40, Part 158 of the Code of Regulations (40CFR158)8. These studies are
conducted under rigorous Good Laboratory Practice (GLP) standards as specified in 40CFR1609. The
studies undergo extensive review and analysis by the EPA; these reviews are captured in data evaluation
records (DERs), and the results of these unpublished10 studies are subsequently captured in ECOTOX. In
addition, open literature studies identified in ECOTOX are used that meet the standards specified by
OMB regarding data quality using protocols identified in the EPA guidance11. Typically, the EPA uses
ECOTOX to identify whether more sensitive toxicity endpoints are available than those derived from
registrant-submitted studies. Similar to registrant-submitted studies, DERs are compiled for open
literature studies as well.
Unlike FIFRA, the CWA does not give the EPA authority to require collection of test data (i.e., issue data
call-ins) to ensure that toxicity data are available to fulfill the MDRs for establishing ALC (U.S. EPA, 1985).
The EPA has historically developed ALC for pesticides using registrant-submitted test data under FIFRA
and publicly available open literature provided through ECOTOX. The EPA reviews the studies obtained
4 https://www.epa.gov/pesticide-science-and-assessing-pesticide-risks/evaluation-guidelines-ecological-toxicity-
data-open
5 https://obairTiawhitehouse.archives.gov/sites/default/files/oirTib/fedreg/ireproduciblle2.pdf
6 https://www.epa.gOv/wqc/aquatic-life-criteria-and-methods-toxics#sop; (U.S. EPA 822-R-24-008)
7 https://www.epa.gov/sites/default/files/2016-02/docuirnents/guidelines-wateir-quallitY-ciriteiria.pdf
8 https://ecfr,io/Title-40/sp40.26.15s.e
9 https://ecfr.io/Titlle-40/pt40.26.160
10 Because of confidential business information (CBI) and other protections set forth in FIFRA Section 10, the actual
studies containing raw data are not public.
11 https://www.epa.gov/pestijcijde-scijence-and-assessijng-pestijcijde-irijslks/evalluatijon-guijdellijnes-ecollogijcall-toxijcijtY-
data-open
2
-------
in the open literature based on the Office of Water's Standard Operating Procedures for Systematic
Review of Ecological Toxicity Data and these reviews are captured in OW DERs, for acceptability review
before use in ALC.
Effects Assessments and Protective Value Development: Under FIFRA, the EPA typically bases its
regulatory thresholds and benchmark values on the most sensitive tested species as described in the
2004 Overview of the Ecological Risk Assessment Process in the Office of Pesticide Programs
("Overview Document"). These values are frequently based on standardized test guidelines that are
generally intended to meet toxicity testing requirements under FIFRA. For evaluating acute aquatic
effects on freshwater taxa to make regulatory judgments under FIFRA, the EPA generally requires testing
on one warm water fish, one cold water fish, and one aquatic invertebrate. Data are also required on
aquatic vascular and multiple species of non-vascular plants. For evaluating chronic effects, the EPA
requires chronic toxicity data for a freshwater fish and an invertebrate; ideally the species tested in the
chronic toxicity tests should have corresponding acute toxicity data. Estuarine/marine (saltwater) test
requirements include acute toxicity studies of a single fish and several invertebrates (crustacean;
mollusc); estuarine/marine chronic toxicity tests are conditionally required depending on how the
chemical is used. Depending on chemical/physical characteristics of a compound, the EPA conditionally
requires subchronic and chronic toxicity testing of benthic freshwater and estuarine/marine
invertebrates. As noted in the regulations, data routinely required under Part 158 may not always be
sufficient to assess whether there are unreasonable adverse effects on the environment. Therefore, the
Agency retains the right to call-in additional data to inform its regulatory decisions. FIFRA section
3(c)(2)(B) provides authority to issue data call-ins for additional information needed to support a
registration. Also, under 40 CFR Part 158.30(b) and 40 CFR Part 158.75, the EPA may require additional
information to better characterize the potential risks.
To develop ALC recommendations under section 304(a) of the CWA, the EPA typically generates a
sensitivity distribution of genus average data of publicly available, high-quality data to estimate values
protective of approximately 95% of aquatic genera, following the methods described in the Guidelines.
For evaluating acute effects on freshwater taxa, the Guidelines approach recommends eight MDRs:
three vertebrates (a salmonid, another bony fish, and an amphibian or another family of fish), five
invertebrates (a planktonic crustacean, a benthic crustacean, an insect, a species from a phylum other
than Chordata or Arthropoda, and a species from another order of insect or another phylum not already
represented). Chronic and estuarine/marine test requirements are of similar scope with different test
species.
2.4 Overview of the EPA OPP Aquatic Life Benchmarks
The EPA regulatory decisions related to pesticides under FIFRA consider both terrestrial and aquatic
effects data, including those data underlying OPP ALBs, combined with exposure modeling and
monitoring data in a comprehensive risk assessment. OPP ALBs are developed using toxicity values
based on registrant-submitted scientific data as well as scientifically defensible acute and chronic
aquatic life toxicity tests available in the open scientific literature that are reviewed by the EPA and used
in the Agency's most recent publicly available ecological risk assessments for new pesticide registrations,
risk assessments for currently registered pesticides, or in preliminary Problem Formulations written in
support of the OPP Registration Review process13.
12 https://www.epa.gov/sites/default/files/2014-ll/documents/ecorisk-overview.pdf
13 https://www.epa.gov/pesticide-reevaluation/registration-review-process
3
-------
For conventional pesticides, the EPA reviews studies according to criteria outlined in Standard
Evaluation Procedure manuals and testing methods (e.g., OCSPP Test Guidelines and Organization for
Economic Cooperation and Development (OECD) Test Guidelines and Guidance Documents14) accepted
by the scientific community and determines if they are acceptable for use in the regulatory process. This
determination is based on the design and conduct of the experiment from which the data were derived,
and an evaluation of whether the data fulfill the purpose(s) of the data requirement. In evaluating
experimental design, the EPA considers whether generally accepted methods were used, sufficient
numbers of measurements were made to achieve statistical reliability, and suitable controls were built
into all phases of the experiment. In an effort to reduce duplicative testing and to create a framework
for sharing of data, the OECD member countries have agreed that tests conducted in accordance with
OECD Test Guidelines15 and principles of GLP in one country must be accepted by other OECD countries
for assessment purposes through the Mutual Acceptability of Data16 agreement.
The EPA evaluates the conduct of each study in terms of whether it was conducted in conformance with
the design, good laboratory practices (GLP as described in the 40CIFR160) were observed, and results are
reproducible. Scientifically sound studies that meet guideline specifications are classified as
"acceptable" and can be used quantitatively to derive regulatory thresholds and fulfill testing
requirements as specified in 40CFR158 §158.70. Studies that are scientifically sound but do not meet
guideline specifications are classified as supplemental. Depending on the extent to which a study
deviates from guideline specifications, it may be used quantitatively to derive risk estimates or used
qualitatively to provide supporting evidence for the quantitative regulatory thresholds derived from
studies classified as acceptable. Studies that are not considered scientifically sound are classified as
"invalid" and have no utility in assessing toxicity or risk.
Under FIFRA, as amended, the EPA is required to review currently registered pesticides (i.e., Registration
Review) on a 15-yr cycle to ensure that data meet current testing requirements and to ensure that
pesticides can continue to be used without causing unreasonable risks to human health and the
environment. Also, if new uses of registered pesticides are proposed, then the pesticide is evaluated to
determine whether the data are sufficient to support the new use as specified in the 40CFR158. As a
result, regulatory thresholds (i.e., benchmarks) may change as additional data are identified. Also, as
new pesticides are approved by EPA, new regulatory thresholds are added to the OPP ALBs. The EPA's
goal is to update these benchmarks and assessments on an annual basis.
OPP ALBs developed for ecological risk assessments are regulatory threshold concentrations below
which pesticides are not expected to harm aquatic life and are considered by the EPA to be protective of
community-level effects. The EPA may further refine risk assessments for federally listed threatened and
endangered species assessments under the Endangered Species Act (ESA) based on the full distribution
of toxicity data for a given genus, using point estimates, species sensitivity distribution approaches, or
probabilistic methods. At a minimum, benchmarks are based on the single most sensitive quantitative
14 OECD Guidelines for the testing of chemicals are a collection of the most relevant internationally agreed testing methods
used by governments, industry and independent laboratories to assess the safety of chemicals. They are primarily used in
regulatory safety testing and subsequent chemical notification and registration. The set of Test Guidelines is updated on a
regular basis to keep pace with progress in science and countries' regulatory needs. OECD-wide networks of national
coordinators and national experts provide input from scientists in government, academia, and industry.
L - ~
15 ht!Bg://wwwj)egll|b[a!;org/^^
systems 20745761
16 https://www.oecd.org/chemicalsafety/testing/mutualacceptanceofdatamad.htm
4
-------
endpoint from the freshwater invertebrate and vertebrate data sets to calculate respective freshwater
invertebrate and vertebrate acute and chronic benchmarks. Benchmarks are also identified for vascular
and non-vascular aquatic plants. In the majority of cases, ecological risk assessments rely on a suite of
registrant-submitted standardized toxicity tests with aquatic vascular and non-vascular plants and acute
and chronic toxicity studies conducted with freshwater fish and invertebrates. These studies are
performed on a limited number of species (one warm water fish and one cold water fish, and one
freshwater invertebrate, and aquatic vascular and non-vascular plants, as noted above) intended to be
representative of a broader number of taxonomic groups. However, in addition to data required to
support the registration of a pesticide, open literature studies are identified through ECOTOX. In
situations where additional data are available, decisions are made regarding the quality and utility of
such information for use in assessing ecological risks (e.g., a review of the validity and reliability of study
protocols), which is consistent with the Agency's risk assessment and open literature guidance
documents. The extent to which such additional data are either employed or rejected is described
through a transparent, concise review (i.e., DER).
Acute toxicity benchmarks rely on regression-based median lethal or median effect concentrations (i.e.,
LC5o or EC50 values) where the measurement endpoint is typically lethality. Chronic benchmarks are
based on hypothesis-based testing to identify no-observed adverse effect concentrations (NOAECs)
where the measurement endpoint may be lethality (i.e., impaired survival), growth and/or reproduction.
While 95% confidence intervals are typically available for the regression-based endpoints, similar
measures of dispersion are not typically available for NOAECs, and the nature of the effect and its
magnitude (i.e., percent of impairment relative to controls) at the statistically significant (p<0.05) lowest
observed adverse effect concentration (i.e., LOAEC) can vary widely. The variability of chronic NOAEC
values within and across species is in part due to the spacing between test concentrations, differences in
the duration of exposure, and differences in the measurement endpoint (i.e., survival versus growth
versus reproduction).
OPP ALBs are derived by multiplying the most sensitive toxicity values (i.e., the lowest acceptable
toxicity value for the most sensitive species within a taxonomic group) by their respective non-listed
species Level of Concern (LOC) ratio. These LOCs are used by EPA to indicate potential risk to non-target
organisms from the use of pesticides and the need to consider regulatory action. The LOC differs
according to taxon and exposure duration (i.e., acute versus chronic):
¦ Acute risk LOC of 0.5 is based on LC50 or EC50 data for acute effects for aquatic animals (this is
the same as the OW factor of 2 applied to the Final Acute Value; the effect of which is to adjust
the acute median lethal concentration to a minimum effect level in the range of acceptable
control mortality);
¦ Chronic risk LOC of 1.0 based on NOAEC data for chronic effects for all animals (same
adjustment factor under CWA acute ALC); and,
¦ Aquatic plant risk LOC of 1.0.
As noted, the OPP ALBs represent a threshold below which exposure to that pesticide would not be
considered to represent a risk of concern for non-listed species. Although the acute toxicity values for
aquatic animals are typically based on studies ranging in duration between 48 and 96 hours, for risk
assessment purposes, the toxicity values for aquatic plants and the acute toxicity values for aquatic
animals are compared to peak exposure values whereas chronic toxicity values for aquatic animals are
typically compared to 21-day and 60-day exposure estimates for invertebrates and fish, respectively. As
discussed in the 2014 Overview Document, the EPA is also responsible for assessing potential risk to
federally listed threatened or endangered species and their designated critical habitat. Under Section
5
-------
7(a)2 of the ESA, federal agencies are required to consult with the U.S. Fish and Wildlife Service and/or
the National Marine Fisheries Service (collectively referred to as the "Services") to ensure that actions
they fund, authorize, permit, or otherwise carry out will not jeopardize the continued existence of any
listed species or adversely modify designated critical habit. For the EPA, the actions under FIFRA include
the registration and registration review of pesticides. When assessing effects to listed species, the EPA
uses a different process to evaluate potential impacts to individuals or populations of listed species. It
may use lower, more conservative, LOCs to identify potential effects to individual listed species and/or
alternative toxicity thresholds to identify potential impacts to populations of species. The EPA also
evaluates potential effects to a species' habitat or food sources, which may also involve developing
toxicity thresholds for these types of effects. The intent is to avoid jeopardy to the population and/or
adverse modification of designated critical habitat and to formally consult with the Services in situations
where there is a likely to adversely affect (LAA) determination.
2.5 Overview of the EPA CWA National Recommended 304(a)(1) Aquatic Life Criteria
The EPA typically uses the approach laid out in the Guidelines to develop national recommended 304(a)
ALC. The Guidelines specifies a general assessment goal of "protection of aquatic organisms and their
uses" that includes in part "prevention of unacceptable long-term and short-term effects" on aquatic
species assemblages and important aquatic species. Similar to aquatic assessments under FIFRA, ALC
consist of two sets of values intended to be protective of aquatic life, acute and chronic criteria. To
develop ALC, the EPA reviews studies according to its Standard Operating Procedures for Systematic
Review of Ecological Toxicity Data [https://www.epa.aov/svstem/files/documents/2024-09/eco-data-
aualitv-svstematic-review-sop-and-ders.pdf). which reflects the Guidelines data quality review
procedures and harmonization with procedures under FIFRA, as noted above. Acute criteria, or Criterion
Maximum Concentrations (CMCs), are intended to protect against acute effects from short-term
exposures. Chronic criteria, or Criterion Continuous Concentrations (CCCs), are intended to protect
against chronic effects on survival, growth, and reproduction that occur from longer-term exposure. The
criteria include three parts: a) chemical concentration, or magnitude; b) limitations on acceptable
duration of exposure; and c) limitations on frequency of allowable exceedance of the specified
concentration.
The EPA's Guidelines provide that other scientifically-defensible approaches and data may be used
contingent on the nature of the chemical and other factors (e.g., the dominant route of exposure).
Consideration of MDRs can be adjusted in the problem formulation phase of the evaluation, based on
best professional judgement and data availability. In more recent ALC (2012 onward), a specific
discussion of available acute and chronic data on endangered and threatened species or closely related
surrogates is discussed in the criteria document. Surrogate species are the most phylogenetically-related
taxonomic level possible to account for the anatomical and physiological traits conserved across taxa
that influence species and taxa sensitivity to a pollutant. The EPA consults with the Services on EPA's
approval of state water quality standards that may affect listed species on a state-by-state basis under
CWA section 303(c).
2.5.1 Acute Criteria
The Guidelines recommends ALC be developed using acute toxicity data for a minimum of eight family
MDRs with the intention of encompassing varied chemical sensitivities across organisms present in
aquatic ecosystems. The eight MDRs for freshwater ALC include: 1) a salmonid; 2) a second fish family in
the class Osteichthyes; 3) a third chordate family; 4) a planktonic crustacean; 5) a benthic crustacean; 6)
an insect; 7) a family in a phylum other than Chordata or Arthropoda; and 8) a family from any insect
6
-------
order or phylum not represented. The eight MDRs for the development of estuarine/marine ALC are: 1
and 2) two families in the phylum Chordata; 3) a family in a phylum other than Arthropoda or Chordata;
4) either the Mysidae or Penaeidae family; 5, 6, and 7) three other families not in the phylum Chordata
(may include Mysidae or Penaeidae, whichever was not used above); and 8) any other family.
For an acute criterion, or CMC, the four genus mean values nearest to the 5th centile of the sensitivity
distribution are used to calculate the Final Acute Value (FAV). The criteria calculation is conducted using
a censored log-triangular distribution that accounts for the number of genera for which there are data.
The use of the factor of two to reduce the FAV to the acute criterion is based on analysis of 219 acute
toxicity tests on a range of chemicals, as described in the Federal Register on May 18, 1978 (43 FR
21506-18). For each of these tests, mortality data were used to determine the highest test
concentration that did not cause mortality greater than that observed in the control for that particular
test, which would be between 0 and 10% for an acceptable acute test. This analysis was re-analyzed by
the EPA recently with the addition of new data, yielding the same general conclusion (U.S. EPA 2014).
Thus, dividing the LC5o-based FAV by two decreases potential acute effects to a level comparable to
control mortality levels. Therefore, the acute criterion is expected to protect 95% of genera in aquatic
ecosystems from acute effects.
2.5.2 Chronic Criteria
When chronic values are available for the minimum eight families, the Final Chronic Value (FCV) is
calculated by developing a genus sensitivity distribution in the same manner as for the FAV using the
four GMCVs nearest to the 5th centile of the sensitivity distribution. The chronic values are currently
typically based on 20% to 10% effect levels (e.g., EC2oto ECio) on survival, growth, or reproduction.
Chronic toxicity tests used in criteria calculations generally range from one to two months in duration. If
chronic values are not available for genera within eight families, but are available for at least one fish,
one invertebrate, and one acutely sensitive species, then the chronic criteria may be calculated by
dividing the FAV by a final acute-to-chronic ratio (FACR), based on the available paired acute and chronic
values. The FCV is equal to the chronic criterion (CCC). Similar to OPP ALBs there is no additional factor
applied because the genus mean averages are based on low effect levels, ECio or EC2o-
Lastly, if the FAV and/or the FCV calculated following the Guidelines approach is not determined to be
protective of a commercially or recreationally important species, the FAV and/or the FCV can be
lowered to address the acute or chronic species mean value for that species, applying the duration and
frequency criteria components as described above.
2.5.3 ALC Duration and Frequency Aspects
Duration recommendations for ALC are typically based on assumptions described in the Guidelines, with
one hour being the typical acute criteria duration and four days the typical chronic criteria duration for
water column-based criteria. The duration and frequency components of ALC are intended to ensure
adequate protection of aquatic life, by establishing limits for exposures at criteria concentrations
(magnitudes).
One-hour acute criteria are typically based on 48- and 96-hour toxicity test results, for invertebrates and
vertebrates, respectively, the same general toxicity test designs used in FIFRA tests. The one-hour
average duration for acute criteria was specified in the Guidelines because some substances can cause
toxicity rapidly, noting that "one hour is probably an appropriate averaging period because high
concentrations of some materials can cause death in one to three hours. Even when organisms do not
die within the first hour or so, it is not known how many might have died due to delayed effects of this
short of an exposure."
7
-------
Four-day chronic criteria values are typically based on 20- to 60-day toxicity tests, for invertebrates and
vertebrates respectively, the same toxicity test designs used in FIFRA tests. The Guidelines state that an
averaging period of four days was appropriate for use with the chronic criterion because for reasons
including that "for some species it appears that the results of chronic tests are due to the existence of a
sensitive life stage at some time during the test, rather than being caused by either long-term stress or
long-term accumulation of the test material in the organism." This four-day duration component of the
chronic water column criterion is also consistent with U.S. EPA (1991) which describes how National
Pollutant Discharge Elimination System (NPDES) permit limits are derived and considers the default four-
day chronic averaging period as "the shortest duration in which chronic effects are sometimes observed
for certain species and toxicants." In conclusion, four-day averaging "should be fully protective even for
the fastest acting toxicants." Thus, the EPA typically recommends four-day averaging periods for chronic
ALC. (There are two exceptions: recently updated ALCfor ammonia and selenium have longer
recommended chronic water column criterion duration periods to reflect the behavior of these specific
chemicals in organisms and the environment).
The frequency component of criteria has remained the same across all water column ALC, with criteria
recommended not to be exceeded more than once in three years. The rationale behind the frequency
component of criteria is based on the recovery of the aquatic ecosystem after an exceedance event
based on an analysis of published studies.
3 Examination of Potential Approaches for Harmonizing OPP Benchmarks and CWA Section
304(a) Effects Assessment Methods for Pesticides
The current effort to harmonize effects assessments under FIFRA and CWA section 304(a) for pesticides
compares the relative magnitude of values derived by OPP in support of registration decisions for
pesticides under FIFRA and CWA ALC (or similar criteria-related values). The EPA examined the values
resulting from the different approaches that have enough data to develop both OPP ALBs and ALC for
the same pesticide. In addition, the EPA investigated two approaches for creating criteria-related values
when data are insufficient to develop ALC and compared these values to OPP ALBs. The Guidelines
describe the methodology the EPA has traditionally used for deriving aquatic life water quality criteria
(also referred to as the Tier I approach) (U.S. EPA. 1985, 40CFR132. Appendix A).
3.1 Comparison of OPP Aquatic Life Benchmarks and CWA Aquatic Life Criteria for Pesticides
3.1.1 Acute Value Comparison
There are eight currently registered pesticides with separate acute ALC and OPP ALBs (i.e., acrolein,
azinphos methyl [guthion], carbaryl, chloropyrifos, diazinon, lindane, malathion, and
pentachlorophenol). For these eight pesticides, OPP ALBs have been updated more recently than ALC. In
seven out of the eight examples, acute ALC values are similar to (acrolein) or higher than the acute OPP
ALBs (carbaryl, chloropyrifos, diazinon, lindane, malathion, and pentachlorophenol) (Table 3.1). The
exception was azinphos methyl (guthion), which has an ALC that is lower than the OPP ALBs; however,
this ALC was developed using an approach that pre-dates the Guidelines methodology. The ALC and OPP
ALB for these eight pesticides are on average within a factor of two in these examples with the
exception azinphos methyl (guthion) (8x) which was not derived following the Guidelines methods, as
noted.
Figure 3.1 illustrates a comparison of the outcomes of OW and OPP methodologies and resulting acute
values for diazinon, a data-rich organophosphate pesticide. To derive acute ALC, genus averages of
toxicity tests (GMAVs) are ordered from most sensitive to least sensitive and the Guidelines algorithm
(regression) is used to develop the final acute value based on the acute LC5o data. The final acute value is
8
-------
then divided by two to yield the acute criterion value. The acute OPP ALB is derived by identifying the
lowest species toxicity text (by taxon) and dividing that by two. In general, the resulting OPP ALB is
somewhat lower than the ALC, because the lowest toxicity test used in deriving the OPP ALB is typically
averaged with other toxicity tests for the same genus in calculating the ALC. However as noted above,
the fold-difference between OPP ALB and ALC is within the typical variability observed in laboratory
toxicity tests. For diazinon, the OPP ALB is lower than the ALC because the most sensitive toxicity test is
one of 24 toxicity tests that are averaged together to derive the lowest genus average (GMAV).
9
-------
100,000.00
10,000.00
1,000.00
M 100.00
c
o
c
N 10.00
ro
1.00
0.10
o o
o
~ AD
~ ~ O
~ ~
~
I Lowest Genus Average
' Lowest Species Toxicity Test
Salmonid - GMAV
A
Other Fish-GMAV
¦
Lowest Test
¦ Arthropod - GMAV
~
Mollusk - GMAV
~
Other Invert. - GMAV
A Amphibian - GMAV
O
Plant-GMAV
OPP: Lowest Value/2
GMAV = Genus Mean Acute Value
- OW: Acute Criteria (n=26)
0.01
0.0
0.1
0.2
0.3
0.4 0.5 0.6
Acute Sensitivity Centile
0.7
0.8
0.9
1.0
Figure 3.1: Comparison of OPP and OW acute effects assessment method for diazinon, a data rich organophosphate insecticide.
OW-generated values shown in green (genus averages in green boxes, criteria value is green dotted line.) OPP-generated values are black square (lowest toxicity
test) and associated OPP ALB (red solid line). Note: The lowest species toxicity test (OPP approach) is one of 24 tests making up the lowest genus average (OW
approach).
10
-------
3.1.2 Chronic Value Comparison
There are five currently registered pesticides with separate chronic ALC and OPP ALBs (i.e., acrolein,
carbaryl, chloropyrifos, diazinon, and pentachlorophenol). Azinphos methyl (guthion), lindane, and
malathion have acute OPP ALBs and ALC but no chronic ALC. As with the acute values, OPP ALBs have
been updated more recently than ALC. In four out of the five examples ALC values are the same as
(diazinon and chlorpyrifos) or higher than OPP ALBs (i.e., chronic ALC are somewhat higher for carbaryl,
chloropyrifos, and pentachlorophenol) (Table 3.1). The chronic ALC is lower than the OPP ALB only for
acrolein where an Acute-to-Chronic Ratio (ACR) was used to derive the chronic ALC because of data
limitations whereas the OPP ALB uses the most sensitive NOAEC. The factor difference between the ALC
and OPP ALBs in these examples is less than 4X. This larger fold-difference for some chronic criteria
compared to the acute criteria is due in part to the use of NOAECs in OPP ALB value calculations, while
for CWA ALCs the EC2oor maximal acceptable toxicant concentration (MATC) has generally been used in
the past in chronic criteria development. (For acute values the OPP ALB and ALC are based on the same
effect metric, LC5o.)
In summary, there are limited examples to compare both chronic ALC and OPP ALBs for the same
pesticide. However, in most cases where there was sufficient data and current OW methodology (i.e.,
Guidelines) was used to derive the chronic criteria, chronic OPP ALBs were similar to or slightly lower
than chronic ALC. Generally, the relative difference between the OPP ALB and ALC values were within a
factor of approximately 2X across all acute and chronic criteria with Guidelines-based ALC.
11
-------
Table 3.1. Comparison of the EPA's Acute and Chronic CWA Aquatic Life Criteria (ALC) and OPP
Aquatic Life Benchmarks (ALB).
Pesticide
ALC
date
ALB
date
Acute
ALC
(M-g/ L)
Acute ALB
(Hg/L)
Ratio
ALB/ALC
(acute)
Chronic
ALC
(Hg/L)
Chronic
ALB
(Hg/L)
Ratio
ALB/ALC
(chronic)
ALC derived using Guidelines methodology
acrolein
2009
2016
3
3.5 v
1.2X
0\N~OPP
3
7.1'
2.4X
ALB > ALC
carbaryl
2012
2022
2.1
0.85 1
0.40X
ALC > ALB
2.1
0.5 1
0.24X
ALC > ALB
chlorpyrifos
1986
2022
0.083
0.05 1
0.60X
ALC > ALB
0.041
0.04'
0.96X
0\N~OPP
diazinon
2005
2016
0.17
0.105 1
0.62X
ALC > ALB
0.17
0.17'
OW=OPP
lindane
1995
2016
0.95
0.5 1
0.52X
ALC > ALB
NA
2.9V
No ALC
pentachlorophenol
1995
2020
191
7.4V
0.39X
ALC > ALB
15 1
6.9 u
0.46X
ALC > ALB
ALC derived using pre- Guidelines methodology
azinphos methyl
(guthion)
1986
2016
0.01
0.08'
8X
ALB > ALC
NA
0.25 1
No ALC
malathion
1986
2016
0.1
0.049 1
0.49
ALC > ALB
NA
0.06 1
No ALC
vVertebrate value is the most sensitive ALB
' Invertebrate value is the most sensitive ALB
1 ALC at pH 7.8
3.2 Comparison of OPP Aquatic Life Benchmarks and Alternative Criteria-Related Approaches When
Data are Insufficient to Develop Aquatic Life Criteria
To examine harmonizing approaches when there are not enough data to develop ALC according to a
strict adherence to the Guidelines approach, the EPA explored two alternative approaches for deriving
criteria-related values to compare with OPP ALBs: 1) the Great Lakes Initiative (GLI) approach to develop
Tier II values (U.S. EPA 1995), and 2) modified Guidelines methods for criteria development for
pesticides when MDRs are not met. For information related to the derivation of the alternative
approaches, see the document "Supporting Information for Comparison of OPP Aquatic Life
Benchmarks, OW Aquatic Life Criteria and Alternative Criteria-Related Approaches When Data are
Insufficient to Develop Aquatic Life Criteria."
12
-------
3.2.1 Overview of Great Lakes Initiative Approach to Develop Criteria-Related Values to
Compare to OPP ALBs
In 1995, as part of the Great Lakes Initiative (GLI), the EPA published an alternate method, called Tier II
benchmarks (U.S. EPA 1995, 40 CFR Part 132 Appendix A), for deriving aquatic life protection values for
Great Lakes states using assessment (or safety) factors for chemical pollutants when the Guidelines
MDRs are not met. These Tier II values are protective aquatic life values that are expected to be equal to
or below ALC values calculated using the Guidelines MDRs.
When fewer than eight MDRs are fulfilled, a GLI Tier II acute Secondary Maximum Concentration (SMC)
can be calculated using the lowest genus mean acute value (GMAV) adjusted by an assessment factor
(or Final Acute Value Factor; FAVF) to account for uncertainty due to missing data. Lower extrapolation
factors correspond to datasets with higher numbers of MDRs fulfilled (Table 3.2). This value is then
divided by two to derive the SMC. In addition, the GLI Tier II methodology specifies that datasets must at
least contain acute toxicity data for a genus in the family Daphnidae. FAVFs are intended to be
conservative.
FAVFs for a given chemical were calculated by first constructing 199 datasets of multiple minimum data
sets (n=8) from all acceptable acute data for that chemical. Each value in a dataset represented a
randomly sampled EC/LCso for each of the eight MDR families described in the Guidelines. Final Acute
Values (FAVs) were calculated for each of the 199 datasets (for each chemical) following Guidelines
methodology. For each dataset, the lowest LC5o was divided by the calculated FAV and this process was
repeated with sequential removal of one randomly determined LCsofor subsets of n=7 through n=l.
When n=l, the LC5o must be from a genus in the family Daphniidae. Data subsets of increasing (2
through 7) size included the LC5o for Daphniidae plus one or more LC5oS from additional families
randomly selected from the full dataset of n=8. Chemical specific FAVFs were then calculated as the 95th
centile of the 199 LCso/FAV ratios at each n. Finally, a set of final FAVFs for all chemicals was calculated
as the median of all chemical specific FAVFs for subsample sizes n=l through n=7 for all tested chemicals
(Host et al. 1995), and these FAVFs, or secondary acute factors (SAF), were recommended to be applied
to all datasets that were missing one or more MDR groups (Table 3.2).
Table 3.2 Summary of the Minimum Data Requirements (MDRs) and Corresponding Final Acute Value
Factors (FAVF) based on Great Lakes Initiative (GLI) Methodology.
# Minimum Data Requirements (MDRs) Satisfied
Final Acute Value Factor
1
21.9
2
13
3
8
4
7
5
6.1
6
5.2
7
4.3
The GLI Tier II methodology allows for the calculation of a chronic value for chemicals where fewer than
three ACRs have been experimentally determined. A default ACR of 18 is used when empirically derived
ACRs are not available. The default ACR of 18 was derived in a manner as to provide a level of protection
similar to that intended in the Tier I methodology. Once ACRs have been determined, the Secondary
13
-------
ACR (SACR) is then calculated as the geometric mean of the three ACRs. If no ACRs are available, the
SACR is 18 by default. The Secondary Chronic Value (SCV) is the SAV divided by the SACR.
The EPA conducted analyses using the GLI Tier II methodology for 21 pesticides with uses that are
widespread based on monitoring reports from the U.S. Geological Survey (USGS) and labeled use
patterns and compared the values to EPA's ALC using Tier I methodology (where possible) and EPA's
OPP ALB and concluded:
For the organophosphate and carbamate chemicals examined, the GLI Tier ll-derived values are
usually lower than ALBs (by approximately 4X), as expected since assessment factors were
applied (see Tables 3.4 and 3.5).
There is high variability among the chemicals that make up the GLI Final Acute Value Factor
¦ For example, FAVFs for individual chemicals at n=l range from 1.87 to 1,068 to comprise
the GLI assessment factor at n=l of 21.9 (the median of all individual chemical FAVFs).
The GLI methodology is resource intensive because a literature search and quality assurance review of
all studies is required to determine the number of MDRs fulfilled to know which assessment factor is
appropriate.
3.2.2 Overview of a Modified Guidelines Methods to Develop Criteria-Related Values to
Compare to OPP ALBs
As part of this harmonization effort, the EPA investigated modifying the Guidelines method for criteria
development using insecticide and herbicide case studies where MDRs were not met to derive criteria-
related values. For insecticidal pesticides, the EPA developed criteria-related values by using an
Invertebrate-only Genus Sensitivity Distribution (GSD) or Acute-to-Chronic Ratio (ACR) approach. For
herbicidal pesticides, the EPA developed criteria-related values by using a Modified GSD or ACR
approach including plants. This differs from EPA's standard approach to deriving ALC under the
Guidelines because plants are not typically included in the GSD, due to the paucity of aquatic plant data
for most chemicals in the open literature. For ALC development, plants are typically examined only to
see if they are the most sensitive taxa, as noted below. The GSD values were then compared to OPP
ALBs.
3.2.3 Insecticidal Pesticides
3.2.3.1 Acute Values
Case studies of 21 insecticidal pesticides that either had enough data to fulfill the eight MDRs to develop
ALC or Tier I (illustrative ALC example) value or a Tier II value (based on the GLI methodology) were
developed through prior analyses. Using these 21 pesticides, the EPA explored calculating sensitivity
distributions (SD) with only invertebrate taxa to generate the HC0s values to compare with the
freshwater invertebrate acute OPP ALB. The objective was to develop a method for insecticidal
pesticides that focused on protecting 95% of sensitive taxa (i.e., the 5th centile hazard concentration;
HC05) of concern. This method conserved resources by decreasing focus on non-target, insensitive taxa
that may inappropriately weight the data resulting in under-protection of sensitive group. However, this
method produces a value that is inherently more conservative than ALC because it reduces the number
of genera (sample size) used in the calculation, as part of the "N" in the denominator of the criteria
calculation.
The EPA explored the data using the GSD model following the Guidelines algorithm with all genus-level
invertebrate data as this methodology is most analogous to ALC development, including dividing the
acute invertebrate-only GSD calculation output by factor of two, to yield a low-effect acute values, as is
done in standard ALC calculations. Generally, the case study analyses used all data underlying the OPP
14
-------
ALBs even if it would be rejected or classified as qualitative according to the Guidelines due to standard
data quality guideline shortcomings (e.g., in a very few cases, issues such as chemical purity was <90% or
did not meet study duration requirements but the data were still used for completeness of
comparisons). Case study analyses were performed on mostly carbamate (C) and organophosphate (OP)
pesticides to compare the invertebrate-only GSD analyses to ALC and Tier II values and OPP ALBs. The
analysis focused on invertebrates since of the 16 OPs and four carbamates evaluated, the majority (94%)
of acute invertebrate OPP ALBs were lower (more sensitive) than their corresponding acute vertebrate
OPP ALBs by factors ranging from 4.5 to 2,614. The OP insecticide methamidophos was the one
exception where the ratio was 0.73. Across all the OPs and carbamates evaluated, the chronic
invertebrate OPP ALB was more sensitive than their corresponding chronic fish OPP ALB by factors
ranging between 2.6 and 5,812. EPA's evaluation resulted in three groupings of pesticides based on their
available data:
1. Data-rich pesticides either have an ALC or have sufficient data to be able to develop ALC based
on the methodology in the Guidelines, because all eight MDRs are met.
o These chemicals are carbamate insecticides (carbaryl, methomyl* propoxur*),
organophosphate insecticides (OPs) (malathion, diazinon, chlorpyrifos, dichlorvos*) and
the herbicide acrolein. *Methomyl, propoxur and dichlorvos do not have 304(a) criteria
but have sufficient data to develop an illustrative ALC example for the purposes of these
analyses only.
2. Data-limited pesticides values were developed using modified invertebrate-Genus Sensitivity
Distribution (GSD) values and GLI Tier II values.
o These include carbamates (oxamyl) and OPs (dimethoate, phosmet, acephate, terbufos)
3. Data-insufficient pesticides do not have sufficient data to generate modified invertebrate-only
GSD values but values were developed using the GLI Tier II methodology.
o These include the following pesticides: OPs (methamidophos, profenfos), a pyrethroid
(fenpropathrin), and other pesticides (fenbutatin-oxide, methoxyfenozide, norflurazon,
propargite, pyridaben)
3.2.3.1.1 Result of Analyses using Acute Modified Guidelines Methods and GLI Methodology for
Insecticides
Similar to the comparison above for pesticides that have 304(a) national recommended ALC, in this
analysis, for data-rich pesticides (where ALC or illustrative ALC values can be derived because all eight
MDRs are filled) there is less than a factor of two difference between the OPP ALB and ALC for seven out
of the eight pesticides. The difference between the OPP ALB and ALC for the eighth pesticide, carbaryl, is
2.5X, with the OPP ALB being lower than the ALC. The invertebrate-only (GSD) analyses showed that for
data-rich pesticides, the OPP ALB and invertebrate-only GSD values are all within a factor of two except
for malathion (8.5X) which does not use the toxicity study that the OPP ALB is based on due to
acceptability criteria under the Guidelines. In six of the eight examples, the invertebrate-only GSD value
is lower than the ALC including all of the taxa. This is largely due to the smaller sample size, or "N."
Reducing the "N" decreases the calculated FAV even if the most sensitive data is the same for the ALC
and invertebrate-only GSD calculations. The two exceptions are malathion which uses a methodology
that pre-dates the Guidelines to derive the ALC and acrolein which has a large invertebrate-only GSD
value because vertebrates, which are excluded from the invertebrate-only analysis, are more sensitive
than invertebrates. The analyses with data-rich pesticides demonstrates that there are only small the
difference between the OPP ALB and ALC values when similar data is used to derive the values. (See
Table 3.3 and Figures A.1-A.7 for more information on these chemicals and relative values.)
15
-------
For data-limited pesticides, the analyses show that there is more variability between the OPP ALB and
the invertebrate-only GSD values. The values for two pesticides are within a factor of two (oxamyl and
acephate). The values for the remaining three pesticides are greater than a factor of six where the OPP
ALBs are higher than the conservative invertebrate-only GSD values. For terbufos and dimethoate, the
OPP ALB is a factor of 6-10X higher than the invertebrate-only GSD values. For phosmet, the factor
difference is 58X due to the use of different data in the invertebrate-only GSD value and the OPP ALB.
Because the eight MDRs could not be filled to derive ALC values, the GLI Tier II methodology was applied
to these pesticides. As with the invertebrate-only GSD values, the GLI Tier II calculated values were
variable (1.5-7.7X, with the exception of phosmet at 22X) and all lower than the OPP ALB. With both
methodologies the variability is a result of the conservative approaches, including the use of "safety"
factors, and, in some cases, the use of different data compared to the OPP ALB. (See Table 3.4 and
Figures A.8-A.12 for more information).
For data-insufficient pesticides, an invertebrate-only GSD value could not be calculated because a
minimum of four invertebrate genera are required to use the Guidelines methodology. As expected, the
GLI Tier II calculated values were variable (3.1-8X) and all lower than the OPP ALB except for
methoxyfenozide, which used different data than the OPP ALB. This variability can be attributed to the
large factors applied when few MDRs are met. (See Table 3.5 for more information.)
16
-------
Table 3.3. Comparison of acute values for data-rich pesticides in which all eight of the minimum data
requirements (MDRs) are met per the Guidelines (OPP ALBs [lowest LC5o/2], OW ALC or illustrative
ALC example, and analysis of invertebrate-only data).
Magnitude relative to ALB is the OPP ALB/OW value; a ratio < 1 means the OPP ALB value is lower than
the OW value, a ratio > 1 means the OPP ALB is higher than the OW value.
Chemical sensitivity distributions presented in Appendix A.
OW ALC or illustrative ALC
OW Genus-level
Invertebrate-only
HCos/21
(# of genera, magnitude
relative to ALB)
Pesticide
Most Sensitive OPP ALB
example (Year published, #
(Year published, species)
of genera, magnitude
relative to ALB)
Carbamates
Carbaryl
0.85 ng/L
2.11 M-g/L
1.54 ng/L
(2022; Pteronarcella
(2012, 47 genera, 0.40X)
(20 genera, 0.55X)
badia)
Methomyl2
2.5 ng/L
4.326 ng/L
2.55 ng/L
(2010; Daphnia magna)
(illustrative example
calculated for this analysis,
8 genera, 0.58X)
(6 genera, 0.98X)
Propoxur2
5.5 ng/L
4.6 ng/L
2.66 ng/L
(2009; Daphnia magna)
(illustrative example
calculated for this analysis,
11 genera, 1.2X)
(5 genera, 2.IX)
OPs
Malathion
0.049 ng/L
0.1 M-g/L
0.418 ng/L
(2016; Ceriodaphnia
(1986, "Gold Book", 0.49X)
(29 genera, 0.12X)
dubia)
Diazinon
0.105 ng/L
0.170 ng/L
0.097 ng/L
(2016; Ceriodaphnia
(2005, 20 genera, 0.61X)
(11 genera, 1.1X)
dubia)
Chlorpyrifos
0.05 ng/L
0.083 ng/L
0.029 ng/L
(2022; Daphnia magna)
(1986, 15 genera, 0.60X)
(15 genera, 1.7X)
Dichlorvos2
0.0334 ng/L
0.032 ng/L
0.023 ng/L
(2021; Daphnia pulex)
(illustrative example
calculated for this analysis,
12 genera, 1.1X)
(6 genera, 1.5X)
Other
Acrolein
3.5 ng/L
3.0 ng/L
22.87 ng/L
(contact
herbicide)
(2023; Xenopus laevis)
(2009, 14 genera, 1.2X)
(7 genera, 0.68X)
Note the magnitude
comparison is with the
invertebrate ALB of <15.5 ng/L.
MDR=minimum data requirement; NA=not applicable
1 Uses Guidelines methodology for calculating the FAV.
17
-------
2No 304(a) ALC recommendation available but has sufficient data to develop an illustrative ALC example for the purposes of
these analyses only.
Table 3.4. Comparison of acute values for data-limited pesticides (OPP ALBs [lowest LC5o/2], GLI Tier II
calculated values, and analysis of invertebrate-only data).
Magnitude relative to ALB is the OPP ALB/OW value; a ratio < 1 means the OPP ALB value is lower than
the OW value. Chemical sensitivity distributions presented in Appendix A.
Pesticide
Most Sensitive OPP ALB
(Year published, species)
OW GLI Tier II value
(# of MDRs filled,
magnitude relative to
ALB)
OW Genus-level
Invertebrate-only HCos/21
(# of genera, magnitude
relative to ALB)
Carbamates
Oxamyl
90 ng/L
(2016; Chironomus plumosus)
17.3 ng/L
(6 MDRs filled, 5.2X)
57.35 ng/L
(4 genera, 1.6X)
OPs
Acephate
550 ng/L
(2007; Daphnia magna)
364.7 ng/L
(7 MDRs filled, 1.5X)
1,069 ng/L
(6 genera, 0.51X)
Dimethoate
21.5 ng/L
(2016; Pteronarcys californica)
3.5 ng/L
(5 MDRs filled, 6.IX)
2.15 ng/L
(4 genera, 10X)
Phosmet
4.32 ng/L
(2023; Daphnia magna)
0.20 ng/L
(5 MDRs filled, 22X)
0.074 ng/L
(4 genera, 58X)
Terbufos
0.085 ng/L
(2023; Daphnia magna)
0.011 ng/L
(5 MDRs filled, 7.7X)
0.014 ng/L
(4 genera, 6.IX)
MDR=minimum data requirement; NA=not applicable
1 Uses Guidelines methodology for calculating the FAV.
18
-------
Table 3.5. Comparison of acute values for data-insufficient pesticides where a GLI Tier II value could
be calculated but there were insufficient data to calculate either a Tier I acute value or genus-level
invertebrate value.
Magnitude relative to ALB is the OPP ALB/OW value; a ratio < 1 means the OPP ALB value is lower than
the OW value.
Pesticide
Most Sensitive
OPP ALB
(Year published,
species)
OW GLI Tier II value
(# of MDRs filled,
magnitude relative to
ALB)
OW Genus-level
Invertebrate-only
HCos/2
OPs
Methamidophos
13 M-g/L
(2016; Daphnia
magna)
2.58 ng/L
(4 MDRs filled, 5X)
NA
(1 genus)
Profenofos
0.465 ng/L
(2008; Daphnia
magna)
0.077 ng/L
(6 MDRs filled, 6X)
NA
(2 genera)
Other
Fenpropathrin
(Synthetic Pyrethroid)
0.0015 ng/L
(2021; Hyalella azteca)
0.00025 ng/L
(5 MDRs filled, 6X)
NA
(2 genera)
Fenbutatin Oxide
(Organotin Acaricide)
0.85 ng/L
(2009; Oncorhynchus
mykiss)
0.173 ng/L
(3 MDRs filled, 4.9X)
NA
(1 genus)
Methoxyfenozide
(Insect Growth Regulator;
Diacylhydrazine)
28.5 ng/L
(2013; Chironomus
riparius)
231.3 ng/L
(3 MDRs filled, 0.12X)
NA
(1 genus)
Norflurazon
(Pyridazine Herbicide)
4,050 ng/L
(2023; Oncorhynchus
mykiss) Note the lowest
ALB is for nonvascular
plants (6.03 ng/L), but the
GLI Tier II value is based on
O. mykiss so the vertebrate
ALB is used in this
comparison
506.3 ng/L3 MDRs
filled, 8X)
NA
(1 genus)
Propargite
(OS Miticide)
1 Hg/L
(2021; Daphnia
magna)
2.231 Hg/L(3 MDRs
filled, 3.IX)
NA
(1 genus)
Pyridaben
(Nicotinamide Inhibitor)
0.265 ng/L
(2023; Daphnia
magna)
0.033 Hg/L(3 MDRs
filled, 8X)
NA
(1 genus)
MDR=minimum data requirement; NA=not applicable
19
-------
3.2.3.2 Chronic Values
Similar to the acute insecticide analyses, the EPA compared chronic OPP ALBs to current ALC or
illustrative ALC examples if one could be derived for this analysis, as well invertebrate-only GSD values,
and GLI Tier II calculated values.
The EPA examined three data-rich pesticides, one carbamate and two organophosphate insecticides, to
compare chronic values for both ALC and invertebrate-only GSD values to OPP ALB. These data-rich
pesticides still did not meet the MDRs to be able to calculate values using the invertebrate-only GSD, so
the acute-to-chronic ratio (ACR) methodology was applied.
Comparison of OPP and OW Acute to Chronic Ratio (ACR) Approaches for Individual Taxa
There are some differences in how OPP and OW have historically calculated ACRs. The requirements of
the historical OW approach are summarized from the Guidelines as follows:
At least three ACRs are required, including one invertebrate, one fish, and at least one acutely
sensitive freshwater species (for the derivation of a freshwater chronic value).
Acute and chronic toxicity data should be from tests performed in the same laboratory using the
same dilution water. When multiple acute values from the same laboratory are available, the
geometric mean of those values is used.
The maximum acceptable toxicant concentration (MATC), which is the geometric mean of the
no-observe effect (NOEC) and lowest-observed effect (LOEC) concentrations, is used as the
chronic value when it is available. When an MATC is not available, the best acceptable chronic
value (e.g., a less than value) is used.
ACRs can be calculated using data for a marine/estuarine species and applied towards the
calculation of a freshwater chronic values, provided the other requirements described above are
met.
The OPP ACR approach is similar in that the acute and chronic values should be paired from studies
performed under similar conditions (e.g., same dilution water, test conditions). However, one major
difference is that the OPP approach uses empirical no-observed adverse effect concentrations (NOAEC)
values, not MATCs. In addition, the OPP ACR approach offers greater flexibility than the OW approach in
that OPP does not require acute and chronic tests to be from the same laboratory. In cases where
multiple acute and/or chronic values are available for the same species but paired acute and chronic
data from the same laboratory/dilution water are not available, then the geometric mean of all reliable
acute and/or chronic values should be used under OPP (U.S. EPA 2005).
Final ACR Determination
The OW ALC approach uses individual ACRs to determine a final ACR (FACR) by calculating the geometric
mean of all acceptable ACRs. The Guidelines specify that if the ACRs appear to increase or decrease as
the species mean acute values (SMAVs) increase, the FACR should be calculated as the geometric mean
for those species whose SMAVs are close to the final acute value (FAV). This can occur for chemicals
where acute sensitivities vary greatly across taxonomic groups based on the chemical mode of action.
OPP ACRs are calculated for species within a particular taxonomic group are applied to acute values
from other species within the same taxonomic group.
For purposes of this comparison, the Guidelines methodology for calculating an FACR was applied to
species-specific ACRs calculated following both OW and OPP methodologies.
20
-------
The chronic ACR analyses found that:
Differences in individual ACR methodology between OW and OPP (e.g., the ALC Guidelines approach
of paired acute and chronic laboratory tests) reduces number of taxa available to use. See Tables 3.6
-3.8.
Invertebrate chronic OPP ALBs are similar to criteria-related values calculated using FACRs with OW
and OPP methodology (most within a factor of 2, all within a factor of 5). See Table 3.9.
Table 3.6. Diazinon Acute-to-Chronic Ratios (ACRs) by species and calculation method.
Genus
Species
ACR
Notes
OW-ACR
OPP-ACR
Invertebrates
Ceriodaphnia
dubia
1.112
1.709
Daphnia
magna
NA
5.190
Americamysis
bahia
1.586
2.295
Fish
Salvelinus
fontinalis
>903.8
>1,315
Pimephales
promelas
196.2
279.6
Jordanella
floridae
23.84
30.43
Cyprinodon
variegatus
>2,979
3,590
All Taxa
53.01
50.33
All Invertebrates (FACR)
1.328
2.731
Table 3.7. Carbaryl Acute-to-Chronic Ratios (ACRs) by species and calculation method.
Genus
Species
ACR
Notes
OW-ACR
OPP-ACR
Invertebrates
Ceriodaphnia
dubia
1.328
1.609
Daphnia
magna
1.581
2.5
Americamysis
bahia
0.8530
1.178
Qualitative ACRC
Fish
Gila
elegans
NA
3.108
ELS chronic test
Pimephales
promelas
23.82
42.86
Life cycle chronic test
Pimephales
promelas
6.256
9.326
ELS chronic test
Ptychocheilus
lucius
NA
2.944
ELS chronic test
All Taxa3
3.684
4.361
All Taxa Final (FACR)b
2
2.463
All Invertebrates
2
2.006
OW-FACR rounded up to 2 per Guidelines.
ELS=fish early life stage
a Of the two ACRs for P. promelas, only the life cycle test was included in this calculation.
bOW and OPP all taxa final ACRs do not include ACR for P. promelas (>10x spread and acutely insensitive), or qualitative ACR
for A bahia.
cAs described in the carbaryl ALC document, these ACRs are treated as qualitative because control survival and number of
young produced per female did not meet ASTM test requirements (U.S. EPA 2012).
21
-------
Table 3.8. Malathion Acute-to-Chronic Ratios (ACRs) by species and calculation method.
Genus
Species
ACR
Notes
OW-ACR
OPP-ACR
Invertebrates
Daphnia
magna
5.942a
13.22
Qualitative OW-ACR developed, because an
ACR following the Guidelines could not be
calculated, as there were no acute and
chronic studies from same
study/laboratory/test water. The resulting
qualitative ACR was included because no
other ACRs for invertebrate taxa were
available.
Fish
Oncorhynchus
mykiss
NA
4.074
Gila
elegans
NA
15.46
Pimephales
promelas
NA
63.18
Ptychocheilus
lucius
NA
5.440
Jordanella
floridae
15.98
40.58
Lepomis
macrochirus
15.27
21.60
Oryzias
latipes
NA
48.60
OPP ACR should be considered qualitative
due to chronic test duration (14-d)
Oreochromis
mossambica
NA
NA
ACR not calculated because no NOAEC was
available
Channa
punctata
NA
4.234
OPP ACR should be considered qualitative
due to chronic test duration (15-d)
Cyprinodon
variegatus
8.5
12.75
All Taxa FACRa
10.54
12.61
All Invertebrates
5.942
13.22
a OPP all taxa ACR does not include qualitative ACRs for O. latipes or C. punctata or ACR for P. promelas (>10x spread and
acutely insensitive). OW all taxa ACR does include the "qualitative" D. magna ACR.
Table 3.9. Final Freshwater Chronic Values.
OPP ALB (ng/L)
Criteria-re
ated values (|ig/L
Fish
Invertebrate
Using OW-
ACR (All
Taxa)
Using
OPP-
ACR (All
Taxa)
Using OW-ACR
(Invertebrates)
Using OPP-
ACR
(Invertebrates)
Carbaryl
6.8
0.5
2.110a
2.103
1.537
1.532
Diazinon
<0.55
0.17
0.1699s
0.1244
0.09675
0.07085
Malathion
8.6
0.06
0.0847b
0.0708
0.1407
0.0632
aALC chronic value
b Illustrative value calculated for this report, no 304(a)(1) ALC chronic value available
22
-------
Despite having relatively robust data sets for the chronic analyses, ACRs were required to derive chronic
ALC and chronic invertebrate-only GSD values for carbaryl, diazinon, and malathion. The differences in
the resulting criteria-related values using the OW or OPP ACR methodologies are minimal (less than 2X
difference). As with the acute values, the invertebrate-only chronic GSD values are generally lower than
the ALC incorporating all the available taxa because of the small sample size (N) used in the calculation,
which results in a lower value. Lastly, the lowest chronic OPP ALB are similar (diazinon and malathion) or
within a factor of four (carbaryl) of the ALC (see Tables 3.9 and 3.10).
When three ACRs could not be fulfilled, the chronic GLI Tier II methodology was applied. As expected,
the GLI Tier II calculated values were variable (1.3-42X) and all lower than the OPP ALB with one
exception (methoxyfenozide which uses different data to calculate the values). The GLI Tier II calculated
values include a high default ACR of 18 when data are missing in order to be able to calculate a value
using that method (See Table 3.10 for more information.)
Table 3.10. Comparison of chronic values for insecticidal pesticides (chronic OPP Aquatic Life
Benchmarks (NOAEC), OW ALC / Illustrative ALC example or GLI Tier II calculated values, and
invertebrate-only HCos values).
Magnitude relative to ALB is the OPP ALB/OW value; a ratio < 1 means the OPP ALB value is lower than
the OW value, a ratio >1 means the OPP ALB is higher than the OW value.
Note: For GLI Tier II calculated values, a default ACR of 18 is used when empirically derived ACRs are not
available.
OW ALC / Illustrative ALC
OW Invertebrate-only
Most Sensitive OPP ALB
example or
HCos
Pesticide
(Year published,
Tier II value
(# of ACRs filled,
species)
(# of ACRs filled,
magnitude relative to ALB)
magnitude relative to
ALB)
Carbaryl
0.5 ng/L
2.1 M-g/L
1.54 ng/L
(2022; estimated NOAEC
(ALC, 0.24X)
(See Table 3.7 for ACRs,
value for Pteronarcella
0.32X)
badia calculated using
the ACR for Daphnia
magna)
Oxamyl
27 ng/L
(2016; Daphnia magna)
2.4 ng/L
(GLI Tier II; 1 ACR, 11X)
NA
Diazinon
0.17 ng/L
0.17 ng/L
0.097 ng/L
(2016, Daphnia magna)
(ALC, IX)
(See Table 3.6 for ACRs,
1.8X)
Malathion1
0.06 ng/L
0.08 ng/L
0.14 ng/L
(2016, Daphnia magna)
(illustrative ALC example
calculated for this analysis;
0.75X)
(See Table 3.8 for ACRs,
0.43X)
Acephate
150 ng/L
(2007, Daphnia magna)
40.5 ng/L
(GLI Tier II; 0 ACRs, 3.7X)
NA
23
-------
OW ALC / Illustrative ALC
OW Invertebrate-only
Most Sensitive OPP ALB
example or
HCos
Pesticide
(Year published,
Tier II value
(# of ACRs filled,
species)
(# of ACRs filled,
magnitude relative to ALB)
magnitude relative to
ALB)
Dimethoate
0.5 ng/L
(2016, estimated NOAEC
value for Pteronarcys
californica calculated
using the ACR for
Daphnia magna)
0.3 ng/L
(GLI Tier II; 2 ACRs, 1.7X)
NA
Methamidophos
4.5 ng/L
(2016, Daphnia magna)
0.42 ng/L
(GLI Tier II; 1 ACR, 11X)
NA
Phosmet
0.75 ng/L
(2023, Daphnia magna)
0.02 ng/L
(GLI Tier II; 2 ACRs, 38X)
NA
Profenofos
0.2 ng/L
(2016, Daphnia magna)
0.013 ng/L
(GLI Tier II; 1 ACR, 15X)
NA
Terbufos
0.03 ng/L
(2023, Daphnia magna)
0.0014 ng/L
(GLI Tier II; 2 ACRs, 21X)
NA
Fenbutatin Oxide
0.31 ng/L
(2009, Oncorhynchus
mykiss). Note the
vertebrate ALB is lower than
the invertebrate ALB (16
Pg/L)
0.06 ng/L
(GLI Tier II; 2 ACRs, 5.IX)
NA
Fenpropathrin
<0.0015 ng/L
(2021, Hyalella azteca)
0.000036 ng/L
(GLI Tier II; 1 ACR, 42X)
NA
Methomyl
0.6 ng/L
(2020, Daphnia magna)
0.47 ng/L
(GLI Tier II; 2 ACRs, 1.3X)
NA
Methoxyfenozide
3-1 Hg/L
(2019, Chironomus
riparius)
25.5 ng/L
(GLI Tier II; 1 ACR, 0.12X)
NA
Norflurazon
770 ng/L
(2023, Oncorhynchus
mykiss). Note the lowest
ALB is for nonvascular plants
(5.33 |ag/L), but the GLI Tier II
value is based on O. mykiss so
the vertebrate ALB is used in
this comparison
56.3 ng/L
(GLI Tier II; 0 ACRs, 14X)
NA
9 M-g/L
0.56 ng/L
NA
(2021, Daphnia magna)
(GLI Tier II; 1 ACR, 16X)
Note the lowest ALB is for
Propargite
nonvascular plants (1.27
pg/L), but the GLI Tier II value
is based on D. magna so the
invertebrate ALB is used in
this comparison
24
-------
Pesticide
Most Sensitive OPP ALB
(Year published,
species)
OW ALC / Illustrative ALC
example or
Tier II value
(# of ACRs filled,
magnitude relative to ALB)
OW Invertebrate-only
HCos
(# of ACRs filled,
magnitude relative to
ALB)
Pyridaben
0.044 ng/L
(2023, Daphnia magna)
0.004 ng/L
(GLI Tier II; 1 ACR, 11X)
NA
1No 304(a) ALC recommendation available but has sufficient data to develop an illustrative ALC example for the purposes of
these analyses only.
3.2.4 Herbicidal Pesticides
Two herbicide classes (i.e., triazines and organophosphorous herbicides) encompassing five case studies
with varying amounts of available toxicity data were used to compare OPP ALBs to criteria-related
values. The herbicide class of triazines included atrazine, propazine, and simazine; the
organophosphorus herbicides included bensulide and glyphosate. The EPA's objective with the
herbicidal pesticide case studies was to perform similar analyses to that for the insecticidal pesticides to
develop three CWA section 304(a) ALC values to compare to OPP ALBs: 1) an ALC-equivalent value when
all MDRs are met; 2) a modified HC0s value when MDRs may not be met; and 3) a Tier II value using the
GLI methodology. This process was complicated by the fact that the Guidelines do not specify MDRs or
standard acceptable toxicity tests with vascular and nonvascular aquatic plants. In the Guidelines
approach, if toxicity data are available for aquatic plants the relative sensitivities of aquatic plants to
animals are compared. The Guidelines state that, in most cases, "the results of tests with plants usually
indicate that criteria which adequately protect animals and their uses will probably also protect aquatic
plants and their uses." However, for herbicidal pesticides, the most sensitive taxa are often aquatic
plants. Aquatic vascular and nonvascular toxicity studies are somewhat different than tests with aquatic
animals, due to the issue of plants' potential ability to regrow after acute exposure to a toxic substance
is ended. Further the lines between acute effects and chronic effects are less clear with plants, as some
of the same endpoints can be used between acute and chronic effects for endpoints. However, as
primary producers at the base of the aquatic food chain, plants should be considered systematically in
assessment paradigms. There are endpoints derived from plant studies which correspond to the acute
LC5o and chronic NOAEC values from animal studies (i.e., the plant equivalent 50% inhibitory
concentration or IC5o and the NOAEC). The major difference though is that both the plant IC5o and
NOAEC endpoints are derived from the same study whereas in for aquatic animals, the two endpoints
come from different studies. For these analyses, to be consistent with what was done for aquatic
animals but different from how ALC have traditionally used plant toxicity data, the LC5o or IC5o (lethal or
inhibitory concentration collectively referred to as "acute values") and the animal and plant NOAEC
values (collectively referred to as "chronic values") were directly incorporated into the analyses (e.g.,
sensitivity distributions). The data and general methods used to develop the three different ALC or
criteria-related values for herbicide pesticides are described below.
Data for these analyses came from two sources. First, the EPA determined the acceptable toxicity tests
from the FIFRA re-registration documents for the five herbicides. Next, the EPA performed an ECOTOX
search and supplemented the re-registration data with toxicity tests deemed acceptable by ECOTOX to
fulfill the MDRs prescribed in the Guidelines.
Acute ALC-equivalent values were determined if all the eight animal MDRs were met by following the
Guidelines algorithm and including the plant toxicity data (LC5o or IC5o) combined with the animal data to
determine the FAV. If the MDRs were met for chronic toxicity data, the same methods were applied as
25
-------
the acute data. If all the MDRs were not met, but ACRs could be developed, then a chronic value was
developed using the ACR method. The ACRs were determined on animal data using the less constrained
OPP methodology (see above) and applied to the plant GMAV if that was the lowest value.
If the eight animal MDRs were not met, the EPA developed a modified value by using the Guidelines
methodology incorporating all available acceptable animal and plant toxicity data regardless of MDR
status. For this method, a minimum number of four GMAVs are required for both acute and chronic
values. As for animals, low sample size (number of genera; N) in criteria calculations has a large impact
on criteria estimates, which was intended in the Guidelines to incorporate the uncertainty in developing
criteria with fewer data.
In addition, if the eight animal MDRs were not met, EPA developed values using the GLI methodology.
For the acute value, the GLI assessment factor is dependent on the number of animal MDRs met and, in
this analysis, was applied to the lowest LC5o (or IC5o) regardless of taxa type (animal or plant). The
variable assessment factors in GLI methodology were also intended to address uncertainty associated
when less data are available, when MDRs are not met. Similarly, the chronic value using the GLI
methodology is dependent on the number of animal ACRs that can be derived, but the extrapolation
factor was applied to the lowest NOAEC regardless of taxa type (animal or plant).
3.2.3.3 Acute Values
Our analyses found that the difference between the lowest OPP ALB (plant) and the ALC-equivalent
value or modified HC0s/2 value ranged from a factor of 1.2 to 5.9 with an average of 3.3. The OPP ALB
was lower than (atrazine) or close to (simazine) for the two herbicides with enough data to derive an
illustrative ALC value. The OPP ALB was higher than the conservative GLI Tier II calculated or modified
HC05/2 values. The factor differences for the illustrative ALC or modified HC0s/2 values were also smaller
than the GLI Tier II approach (8-16X). (See Table 3.11 for more information.)
Table 3.11. Comparison of acute values for herbicidal pesticides (plant OPP ALB, OW illustrative ALC -
equivalent or Tier ll-equivalent values, and modified HCos/2 values).
Magnitude relative to ALB is the OPP ALB/ OW value; a ratio < 1 means the OPP ALB value is lower than
the OW value.
Chemical sensitivity distributions presented in Appendix B.
Pesticide
OPP Most Sensitive ALB
(Year published, species)
OW Illustrative ALC
example or Tier II values
(# of MDRs filled,
magnitude relative to
ALB)
OW Modified HC05/2
(# of MDRs filled,
# of genera available,
magnitude relative to
ALB)
Chlorotriazines
Atrazine1
< 1 Hg/L
(2016; Oscillatoria lutea;
nonvascular plant)
5.7 ng/L
(illustrative ALC example
calculated for this
analysis; 8 MDRs filled,
0.18X)
NA
Propazine
24.8 ng/L
(2022; Navicula pelliculosa;
nonvascular plant)
1.55 ng/L
(GLI Tier II; 4 MDRs filled,
16X)
4.2 ng/L
(3 MDRs, 7 genera, 5.9X)
26
-------
Pesticide
OPP Most Sensitive ALB
(Year published, species)
OW Illustrative ALC
example or Tier II values
(# of MDRs filled,
magnitude relative to
ALB)
OW Modified HC05/2
(# of MDRs filled,
# of genera available,
magnitude relative to
ALB)
Chlorotriazines
Simazine1
6 Hg/L
(2023; Arthrospira
platensis; nonvascular
plant)
5.2 ng/L
(illustrative example
calculated for this
analysis; 8 MDRs filled,
1.2X)
NA
Organophosphorus Herbicides
Bensulide
140 ng/L
(2016; Lemna gibba;
vascular plant)
10.7 ng/L
(GLI Tier II; 4 MDRs filled,
13X)
53.21 ng/L
(4 MDRs, 9 genera, 2.6X)
Glyphosate
11,900 ng/L
(2016; Lemna gibba;
vascular plant)
1,607 ng/L
(GLI Tier II; 4 MDRs filled,
7X)
4,908 ng/L
(4 MDRs, 9 genera, 2.4X)
MDR=minimum data requirement; ACR=Acute to Chronic Ratio; NA=not applicable
1No 304(a) ALC recommendation available but has sufficient data to develop an illustrative ALC example for the purposes of
these analyses only.
3.2.3.4 Chronic Values
As of August 2024, OPP publishes the available vascular and nonvascular plant NOAECs on their Aquatic
Life Benchmarks table17 and those ALB values were used to compare to the OW-derived chronic values
for herbicides, except for atrazine which does not have a chronic plant NOAEC listed. Also, in the case of
bensulide, the chronic invertebrate ALB of 11 ng/L was used in the comparison with the OW-derived
chronic values as it was lower than the plant ALB. There were not enough MDRs to derive an illustrative
ALC-equivalent value for these pesticides, so GLI Tier II values were calculated instead. The analyses
indicate that the difference between the lowest OPP ALB and the modified HC0s value ranged from a
factor of 1.3 to 7.9 with an average of 3.9. The range in the factor difference for the GLI Tier II approach
values was larger than for the modified HC0s approach (1.3-36X). Similar to the acute values, the OPP
ALB was higher than the conservative OW-derived Tier II and modified HC05 values in most case studies.
(See Table 3.12 for more information.)
17 https://www.epa.gov/pesticide-science-and-assessing-pesticide-risks/aquatic-life-benchmarks-and-ecological-
risk
27
-------
Table 3.12. Comparison of chronic values for herbicidal pesticides (chronic OPP Aquatic Life
Benchmarks [NOAEC], GLI Tier II calculated values, and modified HC05 values). Magnitude relative to
ALB is the OPP ALB/ OW value; a ratio < 1 means the OPP ALB value is lower than the OW value.
Chemical sensitivity distributions presented in Appendix B.
Note: For GLI Tier II values, a default ACR of 18 is used when empirically derived ACRs are not available.
OWTier II value
(# of ACRs filled,
magnitude relative to
ALB)
OW Modified HC05
Most Sensitive OPP ALB
(# of MDRs filled, # of
Pesticide
(Year published and
species)
genera available,
magnitude relative to
ALB)
Chlorotriazines
Propazine
6.5 ng/L
0.18 ng/L
2.3 ng/L
(2022; Navicula
(GLI Tier II; 1 ACR filled,
(3 MDRs, 7 genera,
pelliculosa; nonvascular
36X)
2.8X)
plant)
Simazine
1 Hg/L
0.77 ng/L
0.8 ng/L
(Arthrospira platensis;
(GLI Tier II; 2 ACRs filled,
(3 MDRs, 13 genera,
nonvascular plant)
1.3X)
1.3X)
Organophosphorus Herbicides
Bensulide
11 Hg/L
0.88 ng/L
1.4 ng/L
(2016; Daphnia magna;
(GLI Tier II; 1 ACR filled,
(2 MDRs, 7 genera,
invertebrate)
12.5X)
7.9X)
Glyphosate
1,300 ng/L
316 ng/L
5,087 ng/L
(Lemna gibba; vascular
(GLI Tier II; 2 ACRs filled,
(2, MDRs, 6 genera,
plant)
4. IX)
0.26X)
MDR=minimum data requirement; ACR=Acute to Chronic Ratio; NA=not applicable
3.3 Analysis of Ratios of OPP Aquatic Life Benchmarks and Alternative Criteria-Related Approaches
to Develop Aquatic Life Criteria
OPP ALBs were compared to three criteria-related values by developing a ratio of one to the other. The
comparative analysis focused on two classes of insecticides (organophosphates and carbamates) and
two classes of herbicides (triazines and organophosphorus herbicides), with robust databases. The three
different OW-estimated, criteria-related methods are:
1. Methodology for Aquatic Life Criteria (ALC) as outlined in the Guidelines. (Either existing ALC or
Tier I (ALC-equivalent) values when there is sufficient data to be able to develop a value based
on the methodology in the Guidelines.)
2. Great Lakes Initiative (GLI) Extrapolation Factors to calculate Tier II benchmarks (see Section
3.2.1)
3. Modified HC0s methods to conduct case studies with insecticides and herbicides where MDRs
may not be met using an Invertebrate-only and Herbicide-modified Genus Sensitivity
Distribution (see Section 3.2.2).
Table 4.1 summarizes the mean ratios of the OPP ALBs to the corresponding criteria-related value using
the three different methods (i.e., ALC Tier 1, Invertebrate-only GSD HCos, and the GLI factor approach)
for two different classes of insecticides (i.e., carbamates and organophosphates) with a common mode
28
-------
of action (i.e., acetylcholine esterase inhibition) as well as a broader group of chemistries [e.g.,
organochlorines, organometallics, synthetic pyrethroids, N-phenyl heterocycles, diacrylhydrazines,
sulfite esters, and phenols) referred to as "Other" which include insecticides, herbicides, and fungicides
representing multiple modes of action.
Given that there were no statistically significant differences between the three groups of pesticides for
acute ratios derived using the three approaches, ratios were averaged across the three chemical groups
and the means along with their standard deviations and standard errors are reported in Table 3.13.
Similar analyses were conducted to the GLI factor approach.
When comparing across the three methods (i.e., ALC Tier 1, Invertebrate GSD HC0s, and the GLI factor
approach) of estimating acute values, there was no statistical difference (p>0.05) for the three chemical
groups combined. Although the combined mean acute ratios reported in Table 3.13 across the three
methods range from 0.69 to 5.29, high variability likely reduced the ability to detect statistically
significant differences between the three methods. When all the estimated acute ratios are averaged
(across the three methods and three classes of insecticides) the mean ratio is 2.75. The average
estimated chronic values for OPs and carbamates are close to one when comparing the ALC and GSD
HCos methods relative to OPP ALBs. The mean chronic ratio is higher for the GLI because it includes a
high default ACR of 18 when data are missing in order to be able to calculate an ACR using that method.
When all the chronic values are averaged (across the three methods and the three groups of insecticide,
including the ALC chronic methodology), the mean chronic ratio is 4.81.
Table 4.1. Summary of Ratios of Acute and Chronic OPP ALBs to Corresponding Criteria Based on Tier 1
ALC, Invertebrate-only Genus Sensitivity Distribution (GSD) or Acute-to-Chronic Ratio (ACR) and Great
Lakes Initiative (GLI) Tier II Methods for Two Classes of Insecticides.
Acute (OPP ALB/Criteria-related value)
Chemical Class
Tier 1 ALC
(Mean ± SD)
Invertebrate GSD
(Mean ± SD)
GLI Tier II
(Mean ± SD)
Carbamates (4)
0.71 ±0.43 (3)
1.20 ±0.74 (4)
5.2(1)
Organophosphates (11)
0.70 ±0.27 (4)f
3.07 ±3.73 (7)f
5.29 ±2.32 (5)f
Other(9)
0.69 ±0.41 (3)
NAft
5.17 ±3.12 (6)
Combined C& OP & OT (24)
0.70 ±0.32 (10)
2.39 ±3.07 (11)
5.22 ±2.52 (12)
Mean Ratio across methods
and insecticide classes
2.75 ± 2.19 (8)
Chronic (OPP ALB/Criteria-related value)
Chemical Class
Tier 1 ALC
(Mean ± SD)
Invertebrate ACR
(Mean ± SD)
GLI Tier II
(Mean ± SD)
Carbamates (4)
0.24(1)
0.325 (1)
11.25 (1)
Organophosphates (11)
0.61 ±0.45 (3)
1.09 ±0.94 (2)
15.07 ± 13.20 (6)
Other(9)
2.37 (1)
NA
7.54 ±6.56 (8)
Combined C& OP & OT (24)
0.89 ±0.90 (5)
0.84 ±0.80 (3)
10.80 ±9.88 (15)
Mean Ratio across methods
and insecticide classes
4.81 ±5.76 (8)
Sample sizes are shown in parentheses (n).
Excludes guthion (since utilized old methodology) and phosmet (initially utilized different data which resulted in
erroneous factor); had guthion and phosmet been retained, the mean Tier 1 ALC would be 2.16 ± 3.27, the mean
invertebrate GSD would be 9.98 ± 19.9, and the mean GLI Tier II would be 8.01 ± 6.97.
"Excludes acrolein because vertebrates are more sensitive than invertebrates.
29
-------
Given the limited number of case studies with the herbicides, statistical comparisons were constrained,
and unlike the carbamates and the OPs, the chlorotriazines and OP herbicides do not have a common
mode of action; therefore, the ratios by method were not combined in Table 4.2. However, when all the
estimated ratios are averaged (across the three methods and two classes of herbicides) the mean acute
ratio is 7.47 and the mean chronic ratio is 8.07. Both the average acute and chronic ratios for
chlorotriazines and OP herbicides for ALC and the modified HC0s are within or close to a factor of 5, but
the ratios for the GLI method are higher due to the conservative nature of the methodology (use of
assessment factors growing in magnitude as the number of MDRs met decreases) and the modification
to include the herbicide data. The illustrative Tier 1 ALC are the most comparable to the OPP ALB
because assessment factors are not applied as in the GLI Tier II calculations, and the impact of small
sample sizes is not magnified as in the modified HC0s.
30
-------
Table 4.2. Summary of Ratios of Plant OPP ALB to Corresponding Criteria Based on Tier 1ALC,
Modified GSD and GLI Tier II Methods for Two Classes of Herbicides.
Acute (OPP ALB/Criteria-related value)
Chemical Class
Illustrative Tier 1 ALC
Modified GSD HC05/2
GLI Tier II
Mean ± SD (n)
Mean ± SD (n)
Mean ± SD (n)
Chlorotriazine (3)
0.67 ±0.71 (2)
5.90 (1)
16.0 (1)
Organophosphates (2)
NA
4.53 ±4.70 (2)
10.23 ±3.21 (2)
Mean Ratio across
methods and herbicide
7.47 ± 5.87 (4)
classes
Chronic (OPP ALB/Criteria-related value)
Chemical Class
Illustrative Tier 1 ALC
Modified GSD HC05
GLI Tier II
Mean ± SD (n)
Mean ± SD (n)
Mean ± SD (n)
Chlorotriazine (2)
NA
2.04 ± 1.11 (2)
18.70 ± 24.61 (2)
Organophosphates (2)
NA
4.08 ±5.40 (2)
7.45 ±7.14 (2)
Mean Ratio across
methods and herbicide
8.07 ± 7.43(4)
classes
Sample sizes are shown in parentheses (n).
Even for pesticides with some of the most robust data sets (i.e., OPs and carbamates), it is challenging to
identify studies that meet the MDRs specified in the Guidelines. Although the 2012 FIFRA SAP
recommended the use of sensitivity distributions, identifying suitable studies with which to populate
such distributions can also be challenging and could result in the need to use studies identified as
qualitative studies, due to their data quality or other shortcomings, versus use of only quantitative
studies in sensitivity distributions, as is strongly preferred. Some variability in the ratios (i.e., large
differences in the values between OPP ALBs and criteria-related values) can be attributed to the
inclusion of qualitative toxicity data to calculate criteria-related values in the methods where MDRs
were not met (the invertebrate-only or herbicide modified GDS or GLI Tier II approach). Another source
of variability in the ratios can be attributed to the conservative nature of the alternative criteria-related
methodologies used when the MDRs are not met. While the GLI Tier II approach is designed to be
conservative (i.e., result in a low value) depending on the number of MDRs available, the invertebrate-
only and herbicide modified GDS approach is also inherently more conservative than the ALC approach
due to the low genera sample size (N) used in the Guidelines algorithm. The analysis also suggests that
as more data become available, the resulting values from ALC are not substantially different than the
OPP ALB. In fact, in most of the limited cases, OPP ALBs are within a factor of two lower for a given
pesticide than the corresponding Tier I ALC. Importantly, the ratios between ALC and criteria-related
values and OPP ALBs for both acute and chronic insecticides and herbicides are no larger than
differences observed due to natural variability in toxicity responses and the intra- and interlaboratory
variability reported in the open literature (5-10X; Chapman 1998; Duke and Taggart 2000; Fairbrother
2008; Raimondo etal. 2007; Raimondo etal. 2010).
31
-------
4 Summary and Conclusions
The EPA's objective under the Common Effects Project is to harmonize aquatic effects assessment
methods for pesticides to provide a common basis evaluating the effects of these chemicals on water
quality for both under the CWA and the FIFRA using best available information. After spending several
years investigating the underlying methods, current evaluations have focused on maximizing efficient
use of resources by leveraging existing work by OW and OPP and comparing the relative magnitudes of
the effects values.
To attain this objective, EPA scientists collaborated to compare existing OPP ALBs and ALC, and other
criteria-related values. However, in most cases, pesticide data are lacking to fulfill MDRs and for the
development of ALC using the Guidelines methodology. For this effort, three investigative approaches of
CWA section 304(a) ALC and alternative values were developed to compare with OPP ALBs.
1. Methodology to develop Aquatic Life Criteria or illustrative equivalent (Tier I) values;
2. Great Lakes Initiative methodology and assessment factors to develop Tier II values; and
3. Modified HC0s methodology for Invertebrate-only Genus Sensitivity Distribution (GSD) for
insecticides and including plants in the GSD for herbicides and/or Acute-to-Chronic Ratio
(ACR) approaches for both.
The comparative analysis focused primarily on two classes of insecticides (organophosphates and
carbamates) and two classes of herbicides (triazines and organophosphorus herbicides), with relatively
robust databases and compared the ratios of acute and chronic OPP ALBs to corresponding alternative
criteria-related values.
In summary, the results of the EPA's analyses indicate that OPP ALB and ALC values are similarly
protective of aquatic life. Most importantly, comparisons between OPP ALBs and ALC, derived using
longstanding methods established to develop FIFRA and CWA protective values, respectively, indicate
there is little difference between these values (most within a factor of 2), and the ALB is often somewhat
lower. The alternate criteria-related approaches (invertebrate-only and herbicide modified GSD and the
GLI) result in lower values as compared to the OPP ALBs due to data limitations and application of
assessment (safety) factors, or application of a low sample size factor ("N") in the calculations lowers
these values to account for uncertainty. However, when all the ALB/ALC ratios are averaged across the
three ALC and criteria-related methods and classes of insecticides, the factor differences for the acute
and chronic values are similar (within a factor of four). For herbicides, the ALB/ALC ratios averaged
across the three ALC and criteria-related methods also indicate the values are similar and within an
order of magnitude (mean acute ratio of approximately 7 and chronic ratio of approximately 8). These
differences in values all fall within the natural variability in toxicity responses and the intra- and
interlaboratory variability reported in the open literature (5-10X; Chapman 1998; Duke and Taggart
2000; Fairbrother 2008; Raimondo et al. 2007; Raimondo et al. 2010). Importantly, although the
pesticides investigated in this comparative analysis have relatively robust datasets, most pesticides have
more constrained data sets with little to no additional toxicity information from the open literature (i.e.,
beyond studies submitted to the EPA in support of FIFRA registration or re-registration actions),
meaning there generally would be insufficient data to meet the MDRs to develop ALC recommendations
based on the Guidelines. The EPA develops informational aquatic life benchmarks under CWA Section
304(a)(2) for pollutants, typically when there are insufficient toxicity data available to develop
recommended water quality criteria under CWA Section 304(a)(1).
32
-------
The draft analyses presented here show that CWA section 304(a) ALC and criteria-related values and
OPP ALBs are similarly protective of aquatic life. If the EPA were to use the most sensitive animal and
plant OPP ALBs as CWA aquatic life 304(a) protective values, either as 304(a)(1) recommended criteria
or 304(a)(2) informational benchmarks, it would provide states and Tribes with information they can
consider in their water quality standards to manage potential effects of most registered pesticides on
aquatic life. This would satisfy the goals of the Common Effects Project by simplifying risk
communications through harmonizing aquatic life toxicity assessment approaches across the EPA, save
federal government resources, and provide information for environmental protection. With the addition
of new pesticide CWA section 304(a) aquatic life recommended protective values for most pesticides in
commerce, which are updated regularly to include the latest science, states and Tribes would be able to
consider these values in their state water quality protection programs, such as for monitoring and for
developing water quality criteria. The EPA proposes that the CWA aquatic life 304(a) protective values
would use the Guidelines recommended standard frequency and duration (one hour acute, 4-day
chronic duration; frequency of not to be exceeded more than once in three years) if applied in state
criteria or for monitoring purposes.
33
-------
5 References
Chapman, P. M., A. Fairbrother, and D. Brown. (1998) A critical evaluation of safety (uncertainty) factors
for ecological risk assessment. Environmental Toxicology and Chemistry 17(1): 99 -108
Duke, L.D. and M.Taggart. (2000) Uncertainty Factors in Screening Ecological Risk Assessment.
Environmental Toxicology and Chemistry. 19(6): 1668 - 1680.
Fairbrother, A. (2008) Risk Management Safety Factor. In Encyclopedia of Ecology, Jorgensen S. R. and
B. D. Fath Eds, Vol 4 Ecotoxicology: Elsevier, Oxford, UK.
Host, G. E, R. R. Regal and C. E. Stephan. 1995. Analysis of acute and chronic data for aquatic life. Draft
report. United States Environmental Protection Agency, Washington D C, March 16, 1995
Mayer, F.L. and M.R. Ellersieck. 1986. Manual of acute toxicity: Interpretation and data base for 410
chemicals of freshwater animals. Resource Publication 160. U. S. Fish and Wildlife Service. Department
of the Interior, Washington, D.C., 579 p.
Raimondo, S., P. Mineau, and M. G. Barron. (2007) Estimation of Chemical Toxicity to Wildlife Species
Using Interspecies Correlation Models. Environ. Sci. Technol. 47: 5888-5894.
Raimondo, S., C. R. Jackson and M. G. Barron. (2010) Influence of Taxonomic Relatedness and Chemical
Mode of Action in Acute Interspecies Estimation Models for Aquatic Species. Environ. Sci. Technol 44:
7711-7716.
U.S. EPA. 1985. Guidelines for derving numerical national water critera for the protection of aquatic
organisms and their uses. United States Environmental Protection Agency. Stephan, C.E., D.I. Mount,
D.J. Hansen, J.H. Gentile, G.A. Chapman and W.A. Brungs. PB85-227049. National Technical Information
Service, Springfield, VA. U.S.
U.S. EPA. 1991. Technical support document for water quality-based toxics control. Office of Water, U.S.
Environmental Protection Agency, EPA 505/2-90-001, Washington, D.C. 143 pp.
http://www.epa.gov/npdes/pubs/owm0264.pdf.
U.S. EPA. 1995. Final Water Quality Guidance for the Great Lakes System. Federal Register Vol 60, No
56: 15366 - 15425 https://www.eoviinfo.eov/content/plke/IFR-1995-03-23/pdf/95-6671.pdf
U.S. EPA. 2004. Overview of the Ecological Risk Assessment Process in the Office of Pesticide Programs,
U.S. Environmental Protection Agency, Endangered and Threatened Species Effects Determinations.
https://www.epa.eov/sites/defaullt/filles/2014-ll/docuinrients/ecoirislk-oveirview.pdf
U.S. EPA. 2005. Use of acute-to-chronic ratios in support of ecological risk assessment of pesticides.
Memo to Steve Bradbury, Director, Environmental Fate and Effects Division. Office of Prevention,
Pesticides, and Toxic Substances. June 7, 2005.
U.S. EPA. 2014._Examination of the FAV-CMC adjustment factor using the U.S. EPA ACE database.
Unpublished report.
U.S. EPA 2015. Tier II Aquatic Life Community Benchmarks. Unpublished Report prepared by Cadmus for
U.S. EPA's Office of Water and Office of Pesticide Programs.
34
-------
Appendix A: Acute Sensitivity Distributions for Pesticides and Various Protective Values
100,000.00
10,000.00
1,000.00
buo
3.
Arthropod
o
Annelid
A
Mollusk
¦
Salmonid Fish
~
Other Fish
~
Amphibian
CMC
Genus-level Invertebrate
OPP Invert. Benchmark
OPP Fish Benchmark
Anodonta
A ~
~ ~
~ ~
~ ,
Cambarus *
Pontoporeia
~ ~
~ ~ ~ on
Aplexa
Orconectes
Procambarus
Notonecta
~ Asellus
_ . _Mysis_
Lumbriculus
OPP Fish Benchmark= 110 pg/L
-Q
u
Simocephalus
Claassenia ^yale"a
Daphnia
Gammarus
Pteronarcella
kwala
Ceriodaphnia
Ptexonaj;cys_
FAV/2 = Criterion Maximum Concentration = 2.1 pg/L
Isogenus
-Genus-level Invertebrate HC05/2 = 1.537 pg/L
^ OPP Invertebrate Benchmark = 0.85 ng/L
0.4 0.5 0.6
Sensitivity Centile
Figure A.l. Carbaryl genus-level sensitivity distribution. Symbols represent Genus Mean Acute Values
(GMAVs) calculated using all quantitative data from the aquatic life criteria document for carbaryl (U.S.
EPA 2012), and additional data from the OPP benchmark document for carbaryl (U.S. EPA 2007).
35
-------
1,000,000
Arthropod
¦
Salmonid Fish
~
Other Fish
~
100,000 -
~
Other Invertebrate
Genus-level Invertebrate
OPP Invert. Benchmark
_ . _
OPP Fish Benchmark
~
10,000 -
~
¦
FAV/2 = "Criterion Maximum Concentration"
~
CUD
3
. M.
OPP Fish Benchmark = 1,850 jig/L
i-
3 1,000 J
X
o
Q_
o
1-
CL
Aedes
100
Daphnia
Gammarus
Pteronarcys
10 -
OPP Invertebrate Benchmark = 5.5 ug/L
FAV/2 = "Criterion Maximum Concentration" = 4.6 ng/L
1
Genus-level Invertebrate FIC05/2 = 2.7 ng/L
0.0
0.1 0.2 0.3
0.4 0.5 0.6
o
1
° J
00
o J
Lo
1-4 J
o
Sensitivity Centile
Figure A.2. Propoxur genus-level sensitivity distribution. Symbols represent Genus Mean Acute Values
(GMAVs) calculated using all available data from the Office of Pesticide Program's registration review
document for propoxur (U.S. EPA 2009) and an ECOTOX search conducted by Office of Water in 2013.
Propoxur does not have a recommended 304(a) aquatic life criteria. The "Criterion Maximum
Concentration" is an illustrative example calculated for these analyses.
36
-------
1,000,000.00
CUD
Arthropod
o
Other Invertebrate
A
Mollusk
¦
Salmonid Fish
~
Other Fish
~
Amphibian
CMC
Genus-level Invertebrate
OPP Invert. Benchmark
OPP Fish Benchmark
Villosa v
V
Utterbackia A A
Lumbriculus.
Limnodrilus
Tubifex
Lampsilis
~
~ Asellus
to°K
' / Elliptio
Orconectes
Chironomus
Arctopsyche
Palaemonetes
Simocephalus
Daphnia
Pteronarcys^
Drunella
\ >»~
\ .D * V
^ a ^Notonecta
Cypridopsis
~ ~D
==<
Peltodytes
Atherix
» Hydropsyche
K
^ Lestes
OPP Fish Benchmark = 2.05 ng/L
1.00 A
_ .. ... Claassenia
^Limnephilus
V- Ceriodaohnia
^ Genus-level Invertebrate HC05/2 = 0.4180 [ig/L
' \ Gammarus
Isoperla
FAV/2 = Criterion Maximum Concentration (CMC) = 0.1 fig/L
OPP Invertebrate Benchmark = 0.049 ng/L
0.4 0.5 0.6
Sensitivity Centile
Figure A.3. Malathion genus-level sensitivity distribution. Symbols represent Genus Mean Acute Values
(GMAVs) calculated using all available data from the aquatic life criteria document for malathion (U.S.
EPA 1986), data from the OPP re-registration eligibility assessment document (U.S. EPA 2010) and
supplemented with an ECOTOX search conducted by Office of Water in 2010. Note that the 1986 aquatic
life criteria for malathion, was calculated by applying a lOx safety factor to a sensitive LC5o of 1.0 ng/L.
37
-------
10,000.00 A
1,000.00 -
CUD
3
c
o
c
N
ro
Arthropod
o
Other Invertebrate
A
Mollusk
¦
Salmonid Fish
~
Other Fish
~
Amphibian
CMC
Genus-level Invertebrate
OPP Invert. Benchmark
OPP Fish Benchmark
~ Chironomus
Pteronarcys
Gammarus Hyalella
Simocephalus
100 Daphnia
Ceriodaphnia
Lumbriculus ^
~ D^O
Dugesia
A O
Gilia
OPP Fish Benchmark = 45 ng/L
--FAV/2 = Criterion Maxmium Concentration = 0.17 [ig/L
OPPJ.nvertebrate Benehmark = 0.105 jig/L
Genus-level Invertebrate HC05/2 = 0..0968ng/L
0.4 0.5 0.6
Sensitivity Centile
Figure A.4. Diazinon genus-level sensitivity distribution. Symbols represent Genus Mean Acute Values
(GMAVs) calculated using all quantitative data from the aquatic life criteria document for diazinon (U.S.
EPA 2005), data from the OPP re-registration eligibility assessment document (U.S. EPA 2007) and
supplemented with an ECOTOX search conducted by Office of Water in 2010.
38
-------
10,000.00
, , 100.00
GUO
3
in
o
Arthropod
o
Mollusk
¦
Salmonid Fish
~
Other Fish
~
Amphibian
Genus-level Invertebrate
OPP Benchmark
CMC
OPP Fish Benchmark
10.00 A
>
Q.
_o
-C
u
Chironomus
1.00 -L
Amblema
V
X
Aplexa
~
¦
Eriocheir
Procambarus
~ o
Pteronarcys ^ ¦
Orconectes
~
~
X
Hyalella
0.10 H
Classenia « * *Peltodytes
Pteronarcella
Gammarus
Daphnia
OPP Fish Benchmark - 0.9 fig/L
\ Gammarus
_ _ Daphnia FAV/2 = Criteric
Simocephalus
FAV/2 = Criterion Maximum Concentration = 0.083 ng/L
OPP Invertebrate Benchmark = 0.05 \ig/l
Genus-level Invertebrate HC05/2 = 0.0290 pig/L
Ceriodaphnia
0.4 0.5 0.6
Sensitivity Centile
Figure A.5. Chlorpyrifos genus-level sensitivity distribution. Symbols represent Genus Mean Acute
Values (GMAVs) calculated using all available data from the aquatic life criteria document for
chlorpyrifos (U.S. EPA 1986), data from the OPP re-registration eligibility assessment document (U.S.
EPA 2000) and supplemented with an ECOTOX search conducted by Office of Water in 2010.
39
-------
10,000.00
o
¦
~
Arthropod
Mollusk
Salmonid Fish
Other Fish
~ ~
n A Lumbriculus
1,000.00
~
Other Invertebrate
FAV/2 = "Criterion Maximum Concentration"
~
Genus-level Invertebrate HC05/2
OPP Invertebrate Benchmark
O
¦
¦
100.00
OPP Vertebrate Benchmark
OPP Vertebrate Benchmark = 50 |ig/L
_J
OX)
3
O 10.00
>
o
u
5
1.00
Gammarus
Simocephalus
0.10
Daphnia
Pteronarcys
OPP Invertebrate Benchmark = 0.035 ng/L
*
Genus-level Invertebrate HC05/2 = 0.023 ng/L *
FAV/2 = "Criterion Maximum Concentration" = 0.032 |ig/L//
0.01
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Sensitivity Centile
Figure A.6. Dichlorvos genus-level sensitivity distribution. Symbols represent Genus Mean Acute Values
(GMAVs) calculated using all available data from the Office of Pesticide Program's registration review
document for dichlorvos (U.S. EPA 2009) and an ECOTOX search conducted by Office of Water in 2013.
Dichlorvos does not have a recommended 304(a) aquatic life criteria. The "Criterion Maximum
Concentration" is an illustrative example calculated for these analyses.
40
-------
1,000 A
3
c
'q)
o
1-
u
<
Arthropod
o
Mollusk
¦
Salmonid Fish
~
Other Fish
~
Amphibian
Genus-level Invertebrate
OPP Invert. Benchmark
CMC
OPP Vertebrate Benchmark
Daphnia
Peltoperla
~ Chironomus
Tanytarsus o 1=1
Aplexa
O Physa
Gammarus
Genus-level invertebrate HC05/2 = 22.87 ng/L
10 A
OPP Invertebrate Benchmark = <15.5 ng/L
OPP Vertebrate Benchmark = 3.5 [ig/L
FAV/2 = Criterion Maximum Concentration = 2.96 pig/L
0.4 0.5 0.6
Sensitivity Centile
Figure A.7. Acrolein genus-level sensitivity distribution. Symbols represent Genus Mean Acute Values
(GMAVs) calculated using all quantitative data from the aquatic life criteria document for acrolein
(U.S. EPA 2009) and data from the OPP re-registration eligibility assessment document (U.S. EPA
2009).
41
-------
10,000
Arthropod
¦ Salmonid Fish
~ Other Fish
FAV/2 = "Criterion Maximum Concentration"
¦
~
1,000
Genus-level Invertebrate
OPP Invert. Benchmark
- ¦ - OPP Fish Benchmark
Gammarus
_l
CUD
Isogenus
OPP Fish Benchmark = 160 pg/L
">
E
o
-C
4-»
-------
100,000
10,000
Arthropod
O Arthropod (Qualitative)
¦ Salmonid Fish
~ Other Fish
¦ GLI Tier II Acute Value
Genus-level Invertebrate
OPP Invertebrate Benchmark
OPP Fish Benchmark
OPP Fish Benchmark = 2,100 ng/L
ClO
E
05
X
O
Chironomus
O Gammarus
Daphnia
° Echinogammarus
OPP Invertebrate Benchmark = 90 ng/L
Genus-level Invertebrate FIC05/2 = 57.35 [ig/L
GLI Tier II Acute Value = 17.3 jig/L
0.4 0.5 0.6
Sensitivity Centile
Figure A.9. Oxamyl genus-level SD. Symbols represent Genus Mean Acute Values (GMAVs) calculated
using all available data from an Office of Water data analysis in 2015, supplemented the Office of
Pesticide Programs (OPP) registration review document for oxamyl (U.S. EPA 2009).
43
-------
1,000,000
CUD
3.
OJ l,t
O
OJ
£
h
Arthropod
O Arthropod (Qualitative)
¦ Salmonid Fish
~ Other Fish
O Other Fish (Qualitative)
Genus-level invertebrate FIC05/2
OPP Invertebrate Benchmark
OPP Fish Benchmark
Tier I Acute Value
Daphnia
OChironomus
Gammarus
' Pteronarcys
OPP Fish Benchmark = 3,100 ng/L
OPP Invertebrate Benchmark = 21.5 |ig/L
GLI Tier II Acute Value = 3.5 ng/L
0.4 0.5 0.6
Sensitivity Centile
Genus-level invertebrate HC05/2-2.15 (ig/L
0.7 0.8 0.9 1.0
Figure A.10. Dimethoate genus-level sensitivity distribution. Symbols represent Genus Mean Acute
Values (GMAVs) calculated using all available data from an Office of Water data analysis in 2015,
supplemented the Office of Pesticide Programs (OPP) registration review document for dimethoate
(U.S. EPA 2008).
44
-------
10,000.00 -j
1,000.00 -j
W> 100.00
a;
E
O 10.00
1.00
Arthropod
o
Arthropod (Qualitative)
¦
Salmonid Fish
~
Other Fish
- GLI Tier II Value
Genus-level Invertebrate HC05/2
OPP Invertebrate Benchmark
OPP Fish Benchmark
Caecidotea
Streptocepahlus
Daph.nia..
OPP Fish Benchmark = 35 fig/L
OPP Invertebrate Benchmark = 4.32 |ig/L
Gammarus
0.10 -j
GLI Tier II Acute Value = 0.20 pg/L
Genus-level Invertebrate HC05/2 = 0.0740 pg/L
0.01 H
0.0
0.4 0.5 0.6
Sensitivity Centile
Figure A.11. Phosmet genus-level sensitivity distribution. Symbols represent Genus Mean Acute Values
(GMAVs) calculated using all available data from an Office of Water data analysis in 2015, supplemented
the Office of Pesticide Programs (OPP) registration review document for phosmet (U.S. EPA 2009).
45
-------
10,000,000
1,000,000
100,000
tuo
2.10,
Q)
4-»
CD
-C
Q. 1,
CD
U
<
000
000
100
10
0.0
Arthropod
Arthropod (Qualitative)
Salmonid Fish
Salmonid Fish (Qualitative)
Other Fish
Other Fish (Qualitative)
Amphibian
- Genus-level arthropod
OPP Invertebrate Benchmark
OPP Fish Benchmark
Tier II Acute Value
n
Gammarus
Daphnia8,
#_ _ Skwala
^Isogenus
O Ephermeridae Pteronarcella
(Family)
Chironomusb
Genus-level invertebrate HC05/2 = 1,069 [ig/L
OPP Invertebrate Benchmark = 550 |ig/L
GLI Tier II = 364.7 \xg/L
Notes:
, for a 70% formulation
a - Geometric mean of quantitative LC50 and an LC5C
used as the invertebrate benchmark value,
b - Listed in OPP benchmark document but not in the Tier II benchmark document,
c - Geometric mean of /?. catesbelanaqualitative SMAV and
R. clamitons SMAV representing as the OPP amphibian benchmark,
d - Geometric mean of quantitative SMAV for O. mykiss and
o.l
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Sensitivity Centile
Figure A. 12. Acephate genus-level sensitivity distribution. Symbols represent Genus Mean Acute Values
(GMAVs) calculated using all available data from an Office of Water data analysis in 2015, supplemented
the Office of Pesticide Programs (OPP) registration review document for phosmet (U.S. EPA 2007).
46
-------
Appendix B: Acute and Chronic Sensitivity Distributions for Herbicides and Various
Protective Values
100,000
10,000
1,000
bo
3: ioo
a;
c
*N
fU
S 10
AA^'
aa#a++++aAA
~ A
A A
A A O
~ A A '
AA'
Chlorella
Scenedesmus
¦ Anabaena
Ankistrodesmus
Raphidocelis
¦ Navicula
A""
" Pseudanabaena
~ Amphibian
O insect
A Invertebrate
A Fish
+¦ Mollusk
Nonvascular plant
¦ Vascular plant
(FAV/2) - ALC (illustrative example calculated for this analysis
Nonvascular Plant ALB
Oscillatoria (nondefinitive value, less than value)
7
(FAV/2) - ALC (illustrative example calculated for this analysis) = 5.7 p.g/L
Nonvascular Plant ALB=<1 ng/L
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Acute Sensitivity Centile
0.7 0.8
0.9
1.0
Figure B.l. Atrazine acute genus-level sensitivity distribution. Symbols represent Genus Mean Acute
Values (GMAVs) calculated using all available data registration review document (U.S. EPA 2016)
supplemented with data obtained by an ECOTOX search (November 2021). Atrazine does not have a
recommended 304(a) aquatic life criteria, however an illustrative example ALC was calculated for this
analysis.
47
-------
10,000 q
1,000
CUD
3: ioo
a;
c
"m
CD
Q.
O
~
Amphibian
~
Fish
Nonvascular plant
¦
Vascular plant
Modified HC05/2
GLI Tier II
Nonvascular Plant ALB
Vascular Plant ALB
Xenopus (non-definitive, greater than value) -
Lepomis (non-definitive, greater than value)
Oncorhynchus A
X
Navicula r Raphidocelis
^ Navicula
Anabaena
Vascular Plant ALB = 100 pg/L
Nonvascular Plant ALB = 24.8 pg/L
10
Modified HC05/2 - 4.2 pg/L
GLI Tier II = 1.6 pg/L
0.0
0.1
0.2
0.3
0.4
0.5
0.6
Acute Sensitivity Centile
0.7
0.9
1.0
Figure B.2. Propazine acute genus-level sensitivity distribution. Symbols represent Genus Mean Acute
Values (GMAVs) calculated using all available data registration review document (U.S. EPA 2016)
supplemented with data obtained by an ECOTOX search (November 2021).
48
-------
10,000,000
1000,000
100,000
^ 10,000
ClJD
Q)
.E 1,000
N
05
E
to
100
10
~ Amphibian
O Fish
A Insect
A Invertebrate
+- Mollusk
Nonvascular plant
¦ Vascular plant
(FAV/2) - ALC (illustrative example calculated for this analysis
1 1 Nonvascular Plant ALB
Vascular Plant ALB
A + ~
O A
A O O
A O
Raphidocelis
Selenastrum m ~ Lemna
_ Navicula
T "
, *- Vallisneria
Anabaena
" Arthrospira
Vascular Plant ALB = 67 |ig/L
Nonvascular Plant ALB = 6 |ig/L
: ^"Arthrospira
7
(FAV/2) - ALC (illustrative example calculated for this analysis) = 5.1 |ig/L
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Acute Sensitivity Centile
0.7
0.8
0.9
1.0
Figure B.3. Simazine acute genus-level sensitivity distribution. Symbols represent Genus Mean Acute
Values (GMAVs) calculated using all available data registration review document (U.S. EPA 2009)
supplemented with data obtained by an ECOTOX search (November 2021). Simazine does not have a
recommended 304(a) aquatic life criteria, however an illustrative example ALC was calculated for this
analysis.
49
-------
10,000
1,000
00
3 100
01
¦D
"5
<~>
c
CD
CO
~
Fish
~
Invertebrate
Nonvascular plant
¦
Vascular plant
Modified HC05/2
- -
GLI Tier II
.
Vascular Plant ALB
Invertebrate ALB
.1
Lemna
Anabaena (non-definitive, greater than value)
v
1
Skeletonema
~
Navicula (non-definitive, less than value)
^- Raphidocelis
j- Invertebrate ALB = 290 ng/L
j- Vascular Plant ALB = 140 pg/L
j- Modified HC05/2 = 53.2 |ig/L
10
- GLI Tier II = 10.7 |ig/L
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Acute Sensitivity Centile
0.7
0.9
1.0
Figure B.4. Bensulide acute genus-level sensitivity distribution. Symbols represent Genus Mean Acute
Values (GMAVs) calculated using all available data registration review document (U.S. EPA 2009)
supplemented with data obtained by an ECOTOX search (November 2021).
50
-------
1,000,000
100,000
00
1
nj
I/)
o
-C
Q.
_>-
(5
10,000
~
Fish
~
Insect
X
Invertebrate
Nonvascular plant
¦
Vascular plant
- Modified HC05/2
GLI Tier II
Vascular Plant ALB
Invertebrate ALB
r Raphidocelis
» A
1,000
0.0
- Navicula
- Lemna
^ j- inverteorate alb = iz,iuu |ig/L
Anabaena
Invertebrate ALB = 12,100 (ig/L
Vascular Plant ALB = 11,900 pg/L
Modified HC05/2 = 4,908 |jg/L
- GLI Tier II = 1,607 pg/L
r.
0.1 0.2 0.3 0.4 0.5 0.6
Acute Sensitivity Centile
0.7
0.9
1.0
Figure B.5. Glyphosate acute genus-level sensitivity distribution. Symbols represent Genus Mean Acute
Values (GMAVs) calculated using all available data registration review document (U.S. EPA 2009).
51
-------
10,000
1,000
^7 ioo
M
3
01
c
N 10
TO
CL
o
~
Amphibian
A
Fish
+
Invertebrate
Nonvascular plant
¦
Vascular plant
Modified HC05
GLI Tier II
Nonvascular Plant ALB
Vascular Plant ALB
Navicula w
flapHfcfoceTis
i
Lemna
Daphnia
Pimephales A
~ Xenopus
i
Anabaena
Vascular Plant ALB = 22 (ig/L
Nonvascular Plant ALB = 6.5 ng/L
Modified HC05 =2.3[lg/L
GLI Tier II = 0.18 pg/L
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Chronic Sensitivity Centile
0.9
1.0
Figure B.6. Propazine chronic genus-level sensitivity distribution. Symbols represent Genus Mean
Chronic Values (GMCVs) calculated using all available data registration review document (U.S. EPA
2016) supplemented with data obtained by an ECOTOX search (November 2021).
52
-------
10,000
1,000
^ 100
W>
Q)
C
*N 10 J
fO
£
175
Amphibian
Fish
Invertebrate
Nonvascular plant
Vascular plant
- Modified HC05
GLI Tier II
Vascular Plant ALB
V
~
Navicula
^-Typha J
~
Pontederia
r Lemna
x ^
^ Myriophyllum
- Vallisneria (non-definitive, less than value)
Raphidocelis
- Vascular Plant ALB = 58 ng/L
- Arthrospira
Modified HC05 = 0.80 fig/L
- GLI Tier II = 0.77 ^g/L
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Chronic Sensitivity Centile
0.9
1.0
Figure B.7. Simazine chronic genus-level sensitivity distribution. Symbols represent Genus Mean
Chronic Values (GMCVs) calculated using all available data registration review document (U.S. EPA
2016) supplemented with data obtained by an ECOTOX search (November 2021).
53
-------
10,000
1,000
^ 100
bJO
~
Fish
~
Invertebrate
Nonvascular plant
¦
Vascular plant
Modified HC05
GLI Tier II
Vascular Plant ALB
Invertebrate ALB
C
CD
CO
10
Anabaena
v,
/
Skeletonema r RaPhidocelis
/'
Navicula (non-definitive, less than value)
Lemna (non-definitive, less than value) j~ Vascular Plant ALB = 42 ^g/L
A Daphnia (non-definitive, less than value)
L
Invertebrate ALB = 11 ng/L
Modified HC05 = 1.4 fig/L
1 _ .
-GLI Tier II = 0.88 (ag/L
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Chronic Sensitivity Centile
0.8
0.9
1.0
Figure B.8. Bensulide chronic genus-level sensitivity distribution. Symbols represent Genus Mean
Chronic Values (GMCVs) calculated using all available data registration review document (U.S. EPA
2016) supplemented with data obtained by an ECOTOX search (November 2021).
54
-------
100,000
10,000
Fish
Invertebrate
Nonvascular plant
Vascular plant
Modified HC05
GLI Tier II
Lowest NOAEC, Vascular Plant
- Navicula
- Anabaena
- Lemna
Raphidocelis
Lowest NOAEC, Vascular Plant = 7,560 pig/L
CUD
3.
- Modified HC05 = 5,087 ng/L
fU
i/>
O
"q. 1,000
_>¦
r GLI Tier II = 316 ng/L
100
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Chronic Sensitivity Centile
0.8
0.9
1.0
Figure B.9. Glyphosate chronic genus-level sensitivity distribution. Symbols represent Genus Mean
Chronic Values (GMCVs) calculated using all available data registration review document (U.S. EPA
2009).
55
------- |