![[logo] US EPA](../gif/logo_epaseal.gif){kind=logo}
OPP-OW Common Effects Assessment Methodology
Toward a Common Effects Characterization Methodology
Scoping Document
OPP-OW Common Effects Assessment Methodology
April 2009
On this page
- The Need for a Common Effects Characterization Methodology
- Statutory Requirements
- Current Effect Characterization Processes
- Previous Comparisons of Effects Characterization Methodologies
- Developing a Proposed Common Effects Characterization Methodology
Purpose
This document provides the background for a proposal to derive a common effects characterization methodology for use in ecological assessments of chemicals by EPA to meet the mandates of the Clean Water Act (CWA) and the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA). The resulting framework will integrate the Agency's aquatic effects characterization methods and provide a common basis for achieving the water quality protection goals established under the CWA and FIFRA statutes. The framework will also enable the Agency to more transparently communicate to States, the public and other stakeholders how approaches used by EPA offices to characterize effects in aquatic ecosystems are integrated and implemented.
The resulting common effects characterization methodology will focus on "data limited" situations in which the available data are considered insufficient by the Office of Water (OW) for deriving ambient water quality criteria (AWQC) for the protection of aquatic life according to OW 1985 Guidelines, but are considered sufficient for deriving risk quotients used in pesticide registration by the Office of Pesticide Programs (OPP). The methods developed, when sufficiently validated and vetted, could then be used by EPA, States and Tribes to derive scientifically defensible water quality criteria that can in turn be used to set water quality standards in a manner consistent with aquatic effects assessments conducted by OPP and OW. These methods could also be used to interpret aquatic ecological risks associated with pesticide exposure information (e.g., monitoring data). Formal revisions to existing OW criteria derivation guidelines or the current taxa-specific OPP assessment methodologies are not being proposed as part of this process.
The resulting common effects characterization will also enable the OPP to advance its ability to characterize risk by assessing uncertainties in interspecies sensitivity more routinely for pesticides in aquatic ecosystems, as is currently done by OW in deriving AWQC. This will be used to complement OPP's current approach of characterizing effects for different taxa.
The Need for a Common Effects Characterization Methodology
Recent surface water monitoring reports for certain chemicals (e.g. pesticides and some classes of pharmaceuticals) have raised concerns about the possible risk these chemicals pose to aquatic life. Legitimate issues are being raised about the need for more timely Federal guidance to allow States to gauge whether these chemicals represent a concern for aquatic life. One impediment to the timely development of Federal guidance is the limited toxicity data that may be available to meet the minimum data requirements for establishing national recommended ambient water quality criteria (AWQC) for the protection of aquatic life. EPA also has different minimum data requirements used to support effects assessment activities by the Office of Water (OW) under the CWA and by the Office of Pesticide Programs (OPP) under FIFRA. The quantity of data required under the pesticide registration process, a major source of these toxicity data, may be insufficient to meet the more extensive OW data requirements for deriving AWQC. Furthermore, there are differences in the effects characterization approaches used by EPA under the CWA and FIFRA . At issue is the relationship between the effects characterization methods used for pesticide registration actions under FIFRA versus those used to derive water quality criteria used by States and Tribes when adopting water quality standards under the CWA.
OPP toxicity benchmarks and OW AWQC are both developed with high quality data pursuant to parallel but somewhat different rigorously peer-reviewed assessment methodologies. The opportunity being addressed by EPA is how best to build on the substantial high quality science developed under both programs to develop a consistent and common set of effects characterization methods that integrates these approaches for regulators to use in different programs at both the Federal and State level. If successful, this common tiered effects characterization methodology and resultant advisory values would allow Federal and State risk managers to make environmentally protective and scientifically defensible, timely decisions about chemicals that may be found in ambient water in a consistent manner while meeting the mandates of both CWA and FIFRA.
Statutory Requirements
The Clean Water Act (CWA) Regulatory Framework
The objective of the CWA is to restore and maintain the chemical, physical, and biological integrity of the nation's waters. CWA section 101(a)(2) sets an interim national goal of achieving, wherever attainable, water quality which provides for the protection and propagation of fish, shellfish, and wildlife and provides for recreation in and on the water (which is often referred to simply as "fishable and swimmable").
To assist in achieving the objective and goals of the CWA, States and Tribes (hereafter, States) have the primary responsibility under CWA section 303(c) for establishing water quality standards for their surface waters. State water quality standards consist of designated uses, water quality criteria, and an anti-degradation policy. Examples of designated uses that States adopt for their surface water include recreation, protection and propagation of aquatic life, public water supply, navigation, or agriculture. The water quality criteria that States adopt must protect the designated uses of their waters and be based on sound scientific rationale. Water quality criteria can be expressed in narrative, such as "no toxics in toxic amounts", or numerically.
CWA section 304(a) requires EPA to develop and publish AWQC to be used as guidance that can be adopted by States to protect designated uses. These criteria must reflect the latest scientific knowledge and do not consider the social and economic impacts, or the technological feasibility, of achieving that concentration of a chemical in the surface water. Aquatic life criteria define water column concentrations to protect against adverse ecological effects to aquatic life as a result of exposure from direct contact or ingestion of the chemical of concern in the water column. EPA interprets aquatic life criteria that protect the designated use for "fishable" waters as criteria that protect the aquatic assemblage in a waterbody – not the protection and preservation of every individual of every species in a waterbody. EPA has Guidelines1 for the derivation of aquatic life criteria in order to provide uniformity and transparency. The aquatic life criteria are usually derived from laboratory toxicity test results using specific standard procedures described in the Guidelines, but the Guidelines also have general provisions for deviating from these procedures as warranted by the available science.
When adopting water quality criteria into their water quality standards, States can either use the CWA section 304(a) criteria recommended by EPA, modify these recommendations to reflect site-specific conditions, or derive other criteria based on scientifically defensible methods (40 CFR part 131.11(b)). EPA has made specific procedures available to States to derive site-specific criteria when the State determines that the national recommended values may be under- or over-protective. This determination can be made when the species at the site are more or less sensitive than those included in the national criteria data set, or when the physical and/or chemical characteristics of the site alter the biological availability and/or toxicity of the chemical. More information regarding these procedures (i.e., the recalculation procedure, the water-effect ratio procedure and the resident species procedure) can be found in the Water Quality Standards Handbook2. CWA section 303(c) then requires EPA to review new or revised State standards and approve them if they meet CWA requirements.
The Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) Regulatory Framework
The objective of FIFRA is to provide federal control of pesticide distribution, sale and use. All pesticides used in the United States must be registered by EPA. Registration assures that pesticides will be properly labeled and that, if used in accordance with specifications, will not cause unreasonable adverse effects on human health and the environment3. Each registration specifies the crops and sites on which a pesticide may be applied, and each use must be supported by data. Categories of required data include product chemistry, environmental fate, residue chemistry, dietary and non-dietary hazards to humans, and hazards to domestic animals and nontarget organisms. All studies must be conducted under conditions that meet Good Laboratory Practice (GLP) regulations.
FIFRA requires EPA to develop human health and ecological risk assessments to evaluate the potential for adverse effects from the registered use of pesticides. To that end, EPA evaluates toxicity data provided by the registrant and obtained from other sources. The statute provides EPA with authority to issue "Data Call-ins" that require production of effects and exposure data from the registrant to support the registration process of a particular pesticide4 . FIFRA (CFR 40 Part 158.630. subpart G) identifies terrestrial and aquatic nontarget plant and animal data requirements for particular pesticide products; these studies include both acute and chronic effects testing. FIFRA (CFR 40 Part 158.1300 Subpart N) specifies environmental fate data requirements; such data are used to estimate the movement of the pesticide in the environment and to derive potential terrestrial and aquatic exposure estimates that may result from the registered use of the pesticide. These exposure numbers are in turn integrated with effects data to provide estimates of the potential for adverse effects, i.e., risk to nontarget organisms.
The two main regulatory processes in OPP are registration and reevaluation. Registration focuses on decisions that allow new pesticide products to enter the marketplace or that allow registrants to make changes to the way existing pesticide products are sold, distributed, or used. Reevaluation is the review process for currently-registered pesticides. The goal of this process is to ensure that they meet current health, safety, and environmental standards. FIFRA does require that EPA consider "economic, social and environmental costs and benefits of the use of any pesticide" 5 in making the risk management decision of whether the adverse effect is "unreasonable". OPP's ecological risk characterizations, which include, environmental fate assessments and ecological effect assessments, are independent of economic considerations.
Challenges
Currently, OW has 47 established national recommended water quality criteria for the protection of aquatic life 6. Of these, 16 are ingredients in currently registered pesticides. Meanwhile, OPP has registered roughly 1000 active ingredients representing approximately 27,000 formulated products 7. Although both OW and OPP operate under different statutory requirements, they have the mutual goal of protecting the environment. The processes followed by OW and OPP have undergone substantial peer review and endorsements by the scientific community. However, State and local regulators have commented on seeming discrepancies in the effects assessment approaches used by OW and OPP as well as resultant differences in AWQC values developed for the ecosystem and the OPP toxicity benchmark values developed for specific taxa.
Current Effect Characterization Processes
Current Process in Aquatic Life Criteria Derivation
Aquatic life criteria consist of two concentrations 8 – the Criterion Maximum Concentration (CMC), that is intended to protect against severe acute effects, and the Criterion Continuous Concentration (CCC) that is intended to protect against longer term effects on survival, growth, and reproduction. The acute criterion limits peak exposures by requiring that 1-hour averages of exposure concentrations not exceed the CMC more often than once in three years on average. The chronic criterion limits more prolonged exposures by requiring that 4-day averages of exposure concentrations not exceed the CCC more often than once in three years on average.
When deriving new or revised national recommended criteria, EPA searches the ECOTOX 9 database and its extensive literature holdings compiled from monthly searches of online abstracting services (American Chemistry Society's STN-CAS and Cambridge Scientific Abstracts), manual searches of table of contents from high-impact journals, and the bibliographies of review articles. The target of these searches are documents containing information regarding lethal and sublethal adverse effects on, and bioaccumulation by, freshwater and/or saltwater aquatic plants and animals, as well as chronic feeding studies and long-term field studies using wildlife species that regularly consume aquatic organisms. To ensure comprehensive coverage of the literature for these chemicals, a chemical specific search is also conducted using STN-CAS, Cambridge Scientific Abstract, Dissertation Abstracts, Science Direct, and Toxline. EPA also considers any relevant information submitted by the public, reviewers, and other government agencies.
The acute criterion is based on available acute values: median lethal concentrations (LC50) or median effect concentrations (EC50 for severe acute effects such as immobilization) from acute toxicity tests (48- to 96-hours long) meeting certain data quality requirements. To compute an acute criterion, the Guidelines require that acceptable acute values be available for at least eight families to address a range of taxonomic diversity. These minimum data requirements include three vertebrates (a salmonid, another bony fish, and another vertebrate) and 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.
For each genus, a Genus Mean Acute Value (GMAV) is calculated by first taking the geometric mean of the available acute values within each species (Species Mean Acute Value, SMAV) and then the geometric mean of the SMAVs within the genus. The GMAVs are then ranked and a regression analysis is performed on the four most sensitive GMAVs resulting in an estimate of the concentration of the pollutant corresponding to a cumulative probability of 0.05 (the 5th percentile of the species sensitivity distribution). This is the Final Acute Value (FAV), which may be lowered to equal the SMAV of an important, sensitive species if appropriate. The FAV is then divided by 2 to derive the acute criterion value (CMC) that is expected to fall below where any acute adverse effects to organisms are observed.
The chronic criterion is based on available chronic values, either:
the geometric mean of the highest no observed effect concentration (NOEC) and lowest observed effect concentration (LOEC) for effects on survival, growth, or reproduction in chronic tests, or
in some recent criteria (e.g., ammonia) the EC20, the concentration reducing organism performance by 20% based on concentration-effect regression analyses.
Chronic tests for invertebrate species are required to include the entire life-cycle, but for fish species partial life-cycle and/or early life-stage test may be accepted.
If chronic values are available for the minimum eight families as described in the acute criterion above, the chronic criterion (CCC) is calculated in the same manner as the FAV. If chronic values are available for at least one fish, one invertebrate, and one acutely sensitive species, then the chronic criterion may be estimated by dividing the FAV by a Final Acute Chronic Ratio based on the available paired acute and chronic values. A chronic criterion may not be calculated if fewer chronic values are available. Alternatively, the chronic criterion may be based on plant toxicity data if aquatic plants are more sensitive than aquatic animals. The important point though is that the full complement of chronic tests for the eight families is not mandatory for deriving the CCC.
In 1995 as part of the Great Lakes Initiative (GLI), EPA published an alternate method (also known as the Tier II methodology) for calculating Tier II values when the minimum data requirements for deriving national recommended CWA section 304(a) aquatic life criteria are not met (40 CFR part 132, App. A. XII). These Tier II values represent estimates of ambient chemical concentrations that would likely be at or below AWQC values had the AWQC data requirements been fully met. These Tier II values may be used as water quality criteria for State or Tribal water quality standards in the Great Lakes States. In this method, the lowest GMAV in the database is divided by a Secondary Acute Factor (SAF) dependent on the number of satisfied minimum data requirements to derive a secondary acute value. This secondary acute value is then divided by two in order to derive the acute criterion.
This method also outlines an approach to calculate a chronic criterion when fewer than 3 appropriate Acute to Chronic Ratios (ACR) are available, where a default value of 18 is used to replace each missing ACR. In this approach, the geometric mean of the experimentally determined ACRs and the default ACRs is used to calculate a Secondary Acute to Chronic Ratio (SACR). If no experimentally determined ACRs are available, the SACR is equal to 18. The secondary acute value is divided by the SACR to derive the secondary chronic value. Although the GLI Tier II method was intended for use in those States that were a part of the Great Lakes Initiative, it has been used in other States to derive water quality standards that have been approved by EPA.
Current Process in Pesticide Registration Actions
EPA incorporates measures of exposure and effects for the pesticide active ingredient in the risk assessment process for all regulatory decisions10. Available technical grade active ingredient and formulated product information, environmental fate data, and toxicological data may come from a number of sources, including FIFRA section 6 (a)(2) adverse effect incident data, open literature data retrieved through the ECOTOX database, and direct guideline study submissions required in support of registration. EPA searches the public literature for documents containing information on lethal and sub-lethal effects on and bioaccumulation by freshwater and estuarine/marine animals and plants using a methodology modeled on the approach used for derivation of AWQC and similar in scope.
The typical assessment endpoints for pesticide ecological risk assessments are reduced survival for both aquatic and terrestrial animals from direct acute exposures and survival, growth, and reproductive impairment for both aquatic and terrestrial animal species from direct chronic exposures. As noted in the OPP Overview 11 document, which describes the process used to conduct ecological risk assessment, OPP evaluates data on sublethal effects in addition to direct effects on survival, growth and reproduction. For terrestrial plants, the assessment focuses on the maintenance and growth of standing crop or biomass. The assessment endpoints for aquatic plants focus on algal growth rates and biomass (reproduction) measurements, as well as similar measurements for vascular plants.
In the majority of cases, current testing requirements for aquatic organisms include but are not limited to a suite of required aquatic toxicity studies for effects characterization. These test requirements are defined for each chemical class by use category in Data requirements for Pesticides as described in title 40 of the Code of Federal Regulations, Part 158 12 and are performed on a limited number of laboratory test organisms in the following broad taxonomic groupings:
Freshwater fish: testing required on one coldwater fish and one warmwater fish; these data are also used as surrogates for aquatic-phase amphibians, (typical test species: bluegill sunfish (Lepomis macrochirus), rainbow trout (Oncorhynchus mykiss ), and fathead minnow (Pimephales promelas))
Freshwater invertebrates: testing required on one freshwater species (typical test species: waterflea (Daphnia magna)),
Estuarine/marine fish: testing required on one estuarine/ marine species (Typical test species: sheepshead minnow (Cyprinodon variegatus )),
Estuarine/marine invertebrates: testing required on one estuarine/marine mollusk and one estuarine marine invertebrate (typical test species: Eastern oyster (Crassostrea virginica) and mysid shrimp (Americamysis bahia)), and
Aquatic plants: testing required on algal and vascular aquatic plants (typical test species: duckweed (Lemna gibba), green algae, cyanobacteria, freshwater diatom, marine diatom)
For baseline assessments, OPP selects endpoints from these broad taxonomic groups that are the most sensitive acute and chronic toxicity values from all available test data, including open literature and registrant submissions. If additional toxicity data for more species of organisms in a particular group are available, the selection need not be limited to the species listed above, but may include the most sensitive toxicity values for other species from all sources that meet data quality standards13 . Standard evaluation protocols14, the guidance document entitled the Rejection Rate Analysis 15, OPP Guidance on use of open literature16, and best professional judgment are used to determine whether studies satisfy testing requirements.
Data gaps are considered as a source of uncertainty in the risk assessment conclusions, and the risk assessment discusses the potential for additional data to affect the risk conclusions. In the absence of data, adverse effects are assumed to occur until a study is submitted to indicate otherwise. Regardless of the extent of data beyond the regulation-required set of toxicity studies, baseline risk assessments rely on selection of the lowest acute and chronic toxicity values from the most sensitive species tested in acceptable studies. Implicit in the use of the lowest toxicity values for the most sensitive species is the presumption that these toxicity values afford protection not only for the individual surrogate species but for other untested taxa as well. However, the extent to which the most sensitive laboratory test species is representative of the sensitivities of naturally occurring aquatic species is uncertain. This issue applies to both the OPP and OW effects characterization methodologies.
In baseline ecological risk assessments, the effects characterization describes the types of effects a pesticide can produce in an organism and how those effects change with varying pesticide exposure levels. The following measurement endpoints are often used to characterize effects in the baseline assessment. More refined assessments may use the full species sensitivity distribution for a given taxa or other toxicity endpoints, as appropriate.
Acute aquatic animal assessment: most sensitive (lowest) tested effect concentration for 50% of the animals tested (EC50) or the lethal concentration for 50% of the animals tested (LC50) for freshwater fish and invertebrates or estuarine/marine fish and invertebrates from acute toxicity tests. These assessments also evaluate sublethal effects as discussed in the Overview document17.
Chronic aquatic animal assessment: most sensitive (lowest) no observed adverse effect concentration (NOAEC) for freshwater fish and invertebrates or estuarine/marine fish and invertebrates from early life- stage or full life-cycle tests.
Aquatic plant assessment: most sensitive (lowest) EC50 for both aquatic vascular and nonvascular plants.
The ecological risk assessment process used by OPP is flexible in that it allows for refinements that take advantage of the available data. While baseline assessments rely on the most sensitive toxicity values for each taxa evaluated, refined assessments can utilize all data in species sensitivity distributions by taxa18. Refinements have been performed using point estimate approaches (deterministic effect assessments) or estimates that account for the variability and uncertainty of the data (probabilistic approaches)19. The methodologies used for both approaches have been vetted through the FIFRA Scientific Advisory Panel process and the results of such analyses have been utilized in the pesticide regulatory process by OPP.
In response to recommendations and input from stakeholders, OPP developed a webpage of non-regulatory aquatic toxicity "OPP Aquatic Benchmarks" 20. OW and OPP agreed that these values can be used by States and others to help in monitoring activities and evaluating potential risks of pesticides in the aquatic environment, if an aquatic life criterion is not available. These taxa-specific benchmarks, based on toxicity data used by OPP in assessments for pesticide registration decision-making, are considered estimates of the concentrations below which pesticides are not expected to have the potential for adverse effects for the particular taxa for which those data serve as surrogates.
The aquatic benchmarks 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 Level of Concern (LOC) ratio. The LOC differs according to taxon, as follows:
- Acute risk LOC of 0.5 for acute effects for aquatic animals
- Chronic risk LOC of 1.0 for chronic effects for all animals
- Aquatic plant LOC of 1.0
A number of States have used these values as interim measures of whether pesticide residues detected in surface water are sufficient to warrant additional action in terms of resources to refine monitoring efforts.
Current Processes in Other Programs
When proposing a new process, it may also be useful to look internationally 21 to determine whether other countries have an approach that warrants further investigation.
In Canada, Canadian Water Quality Guidelines (CWQGs) are recommended as levels that should protect aquatic resource uses (e.g., freshwater life, agricultural water uses for irrigation and livestock, raw water for drinking water supply, recreational water quality and aesthetics, and industrial water supplies) and result in no observable adverse effects on aquatic ecosystems over the long term. The goal of these values is to protect all life stages during an infinite exposure to water. CWQG are calculated from the lowest observed effects level (LOEL) from a chronic study using a non-lethal assessment endpoint for the most sensitive life stage of the most sensitive species tested. The LOEL is divided by 10 as a safety factor to derive the guideline value. If chronic data are lacking, guidelines may be derived from acute data by using acute-to-chronic ratios or by dividing acute values (LC50 or EC50) by 100 for persistent chemicals or by 20 for non-persistent chemicals.
In Australia and New Zealand, water quality guidelines are calculated using NOEC data with a minimum data set of 5 species representing different taxonomic groups. These guidelines recommend different levels of ecosystem protection for pristine environments (99% of species sensitivity distribution) and general use (95%).
Global coordination of pesticide registration efforts, including global work sharing and global joint reviews, has yielded a common assessment approach among established regulatory authorities (e.g. U.S., Canada, EU and its member states, Australia, New Zealand). The U.S. has worked with NAFTA and OECD for many years to reach agreement on data requirements, guidelines for conducting studies, and review of studies. Thus, the methodology described above as OPP's effect assessment/characterization methodology is also used by other regulatory authorities around the world for pesticide regulatory decision-making.
Previous Comparisons of Effects Characterization Methodologies
In recent years there have been a number of efforts to compare OW and OPP effects characterization methodologies, recognizing the need to coordinate the approaches to the extent possible while acknowledging that the two offices regulate under different statutes.
An internal workgroup compared EPA's assessment approaches for pesticides under the CWA and FIFRA. Two comparative analyses were performed. The first analysis compared the CWA and FIFRA methods using the data available in three national recommended AWQC documents for pesticides. The second analysis directly compared the existing national recommended AWQC for two pesticides to the FIFRA Aquatic Benchmarks. These analyses showed that CWA and FIFRA approaches yielded advisory values that were within a factor of 3 for these chemicals, and that results were unbiased relative to each other (that is, neither approach consistently estimated the lower value). For these pesticides, toxicity data sets approached or exceeded the minimum data requirements necessary for national recommended AWQC development. These findings are generally consistent with previous USEPA analyses (such as those done by in 1995 as part of the Great Lakes Initiative) where typical distributions of species sensitivity in data sets used to derive acute criteria were compared to the lowest LC50 or EC50 value. This analysis, indicated as data sets become larger and the types of species in the data set vary, the FIFRA method will be, on average, within a factor of 2 to 3 of the corresponding aquatic life criterion. Uncertainty, however, in the relative difference between these values can range up to 10-50 fold. However this has not been evaluated to see if it is applicable for pesticides.
Over the past several years, OPP and OW have worked collaboratively on analysis of effects data for several pesticides, i.e., copper, atrazine, and acrolein. The aquatic toxicity data reviewed by both offices were largely the same and the process used in evaluating the adequacy of the data under each program office underscored a common and consistent approach. The collaborative effort resulted in efficiencies in data review and characterization and enabled both programs to more clearly articulate how their effects endpoints relate.
With the participation of the Association of American Pesticide Control Officials/State FIFRA Issues Research and Evaluation Group (AAPCO/SFIREG), a pilot project was established in EPA Region 5 to evaluate the development of community-level aquatic benchmarks for the protection of aquatic life for a variety of targeted pesticides using the OW AWQC Guidelines and/or the GLI Tier II methodology. EPA Region 5 awarded regionally-supported special project funds to the Minnesota Department of Agriculture to implement the project together with the Minnesota Pollution Control Agency. The project also aims to explore ways to develop and complement the current aquatic plant evaluation component of the GLI Tier II methodology. The project is exploring methods for deriving non-regulatory benchmarks using both available and novel approaches. The product of this effort will be a matrix with values derived by existing methods, including: OW Guidelines-derived values, GLI Tier II derived values, and OPP aquatic benchmarks.
Developing a Proposed Common Effects Characterization Methodology
One goal of a common effects characterization methodology is to improve the tools and approaches available to States and stakeholders to derive scientifically defensible water quality criteria that can in turn be used to set water quality standards in a manner consistent with aquatic effects assessments conducted by OPP and OW in compliance with both the CWA and FIFRA. This may lead to a methodology that does not utilize a recommended set of minimum data requirements historically used by OW to establish AWQC. It may also lead to different methods than those used historically by OPP for aquatic effects assessment. Formal revisions to existing OW Guidelines or the current taxa-specific OPP assessment methodologies are not being proposed as part of this process.
This effort to derive a common effects characterization methodology for use in ecological assessments of chemicals by EPA will draw on the lessons learned from both the current process in the two programs, as well as the previous comparisons of the two effects characterization methodologies. The goal is to develop a common approach that will depend on the availability and quality of the data to derive advisory values that will ensure the protection of the environment. As in the past, OPP and OW will continue to pilot efforts to work together on effects assessments for chemicals with extensive datasets.
EPA is proposing, as a first step in this project, the development of a series of white papers that evaluate the suitability of currently available effects characterization methods for use with datasets of varying size and different content. These would include:
An exploration of the application of safety or application factors, including a review of the current GLI Tier II approach to reflect the current state of the science.
Generation of a dataset using synthetic data. For example, Species Sensitivity Distributions (SSDs) in Web-ICE 2.0 22 (Interspecies Correlation Estimation).
Analysis of available methods for bridging data, for example, using models such as Quantitative Structure-Activity Relationships (QSARs), information on the relative sensitivity of species, or the mode of action of chemicals.
A hybrid of the above.
For each approach, the white paper would present the scientific basis and a critical review of its application for effects characterization. Each white paper would describe strengths and limitations of the approach qualitatively, and, where possible, quantitatively. In addition, each white paper would include one or more case studies to demonstrate "proof of concept".
Each white paper would address key issues that are common to all approaches, as well as specific to any given approach. Examples of several key issues common to all include:
The general applicability of each approach across classes of substances, modes of action and taxonomic groups,
The overall uncertainty associated with application of each method under different data availability scenarios, and
The relationship of the predicted results from each method when applied to existing chemicals for which AWQC and OPP toxicity benchmarks are available.
Following the development of the white papers, EPA would present its findings to a diverse set of stakeholders in order to solicit informed scientific input as to the soundness and applicability of the tools and/or approaches. The white papers would then be modified based on stakeholder comment and presented to management for further consideration. The expected outcome of this process would be a subset of effects characterization methods that appear most promising for further refinement and validation. It is then expected that they would be subject to a rigorous external peer review mechanism, for example, a review conducted by a joint meeting of the EPA Science Advisory Board and the FIFRA Scientific Advisory Panel. The resulting effects characterization methodology would then serve as a common basis for deriving AWQC advisory values for use by States, assessing ecological risks by OPP, and for interpreting pesticide monitoring data for which AWQC are not yet available. While EPA is proposing this approach, EPA welcomes suggestions from stakeholders whether this is the best approach to move toward a common effects characterization methodology.
References
Stephan, C.E, D.I. Mount, D.J. Hanson, J.H. Gentile, G.A. Chapman, and W.A. Brungs. 1985. Guidelines for Deriving Numerical National Water Quality Criteria for the Protection of Aquatic Organisms and Their Uses. EPA PB85-227049.
Water Quality Standards Handbook: Second Edition. 1994. Section 3.7, Site-specific Aquatic Life Criteria. Pp. 3-38 – 3-45.
Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) https://www.epa.gov/agriculture/lfra.html
OPP's data requirements are set forth in Title 40 of the Code of Federal Regulations (CFR) Part 158, but EPA has authority to ask for additional data or waive requirements, as appropriate. FIFRA section 3(c)(2)(B) provides the authority to require additional data on currently registered products. These data must be "required to maintain in effect an existing registration of a pesticide."
FIFRA section 2(bb)
USEPA 2008. Current National Recommended Water Quality Criteria. https://www.epa.gov/waterscience/criteria/wqctable/index.html
Pestbank 2008. (CD-ROM) Contains pesticide product data on the 27,000 pesticide products currently registered in the US and pesticide tolerance data from EPA's National Pesticide Information Retrieval System (NPIRS). Includes product names, registration numbers, and sites of application. Available from SilverPlatter.
There may also be separate acute criteria for freshwater and saltwater exposure if enough data exist for a given pollutant.
There are over 5,000 journals, 2,000 published books, 3,000 government reports, and 600 theses represented in the literature holdings of ECOTOX. U.S. EPA. 2007. Ecotoxicity Database (ECOTOX) Mid-Continent Ecology Division, National Health and Environmental Effects Research Laboratory. U.S. Environmental Protection Agency, Office of Research and Development. http://cfpub.epa.gov/ecotox/.
As part of the risk assessment, EPA also examines available information to determine the need to expand beyond the focus on the active ingredient to consider pesticide formulation, inert ingredients, or degradates.
U.S. EPA. 2004. Overview of the Ecological Risk Assessment Process in the Office of Pesticide Programs. Office of Prevention, Pesticides, and Toxic Substances. Office of Pesticide Programs. Washington, D.C. January 23, 2004. Support Document 1: Study Classification used by EFED in Data Evaluation Records (DERs) https://www.epa.gov/oppfead1/endanger/consultation/ecorisk-overview.pdf
40CFR158 Subpart G (158.630 Terrestrial And Aquatic Non-Target Organism Data Requirements; 158.660 Non-target Plant Protection Data Requirements), revised July 1, 2008.
Ibid U.S. EPA 2004
USEPA. Harmonized Test Guidelines. Series 850 Ecological Effects Test Guidelines—Public Drafts https://www.epa.gov/oppts/pubs/frs/home/draftguidelines.htm
USEPA. 1994. Pesticide Reregistration Rejection Rate Analysis Ecological Effects. Office of Prevention, Pesticides and Toxic Substances. EPA 738-R-94-035.
USEPA, 2004, Procedures for the Inclusion of Open Literature Searches in Pesticide Screening Level Risk Assessments for Ecological Effects, in Overview Document (Ibid U.S. EPA 2004), appendix 71 b.
Ibid U.S. EPA 2004.
For example: USEPA. 2002. Environmental Fate and Ecological Risk Assessment for the Reregistration Eligibility Decision on Endosulfan. DP Barcode D238673. U.S. Environmental Protection Agency, Office of Pesticide Programs, Environmental Fate and Effects Division, Washington, DC.
USEPA 2004. FIFRA SAP on Refined (Level II) Terrestrial and Aquatic Models for Probabilistic and Ecological Asses of Pesticides. https://www.epa.gov/scipoly/sap/meetings/2004/033004_mtg.htm
OPP Aquatic Benchmark Table https://www.epa.gov/oppefed1/ecorisk_ders/aquatic_life_benchmark.htm
Kent, R.A., T. Crommentuijn, C.W. Hickey, and F.J. Keating Jr. 2003. Review of Current Approaches and Future Directions in National Ambient Water-Quality Criteria for Aquatic Life Protection: Canada, Netherlands, New Zealand, USA. In Reevaluation of the State of the Science for Water-Quality Criteria Development. M.C. Reiley et al., editors. Society of Environmental Toxicology and Chemistry. ISBN 1-880611-30-9. Pp. 143-170.
Raimondo, S, D. N. Vivian and M. G. Brown. 2007. Web-based Interspecies Correlation Estimation (Web-ICE) for Acute Toxicity: User Manual. Version 1.1. EPA/600/R-07/071. Gulf Breeze. FL. https://www.epa.gov/ceampubl/fchain/webice/