Jump to main content.

National Information
Region 8 Risk Assessment
Ecological Risk Assessment pages

Toxicity Assessment

On This Page

Introduction

Ecological Toxicity Reference Values (TRVs) are species-specific and chemical-specific estimates of an exposure level that is not likely to cause unacceptable adverse effects on growth, reproduction, or survival.  There are three different types of TRVs:

Most ecological TRVs are based on studies of laboratory animals exposed to varying levels of chemical for varying durations of exposure.  These studies are designed to establish the quantitative relationship between exposure level/duration and the occurrence of adverse effects. The results are usually referred to as "dose-response curves." 

Laboratory-based dose-response curves have the advantage that most variables can be well-controlled, and the resulting dose-response data are usually easy to analyze and interpret.  However, one limitation of this approach is that dose-response data from laboratory animals exposed under laboratory conditions may not be a good surrogate for populations of wild receptors exposed under field conditions.  Therefore, whenever it is feasible, it is usually desirable to seek field-based methods to confirm predictions based on laboratory toxicity data (see weight of evidence discussion in the Risk Characterization section).

Species-Specific vs. Group-Specific TRVs

As discussed in the section on Exposure Assessment, most ecological risk assessments do not focus on risks to individual species, but seek to evaluate risks to groups of related receptors (e.g., raptors, passerine birds, small omnivorous mammals, large herbivorous mammals, piscivorous birds, fish, benthic invertebrates).  However, there may be important differences in sensitivity between the individual species present within a group and the selection of a TRV to represent the group must take this potential variability in sensitivity between species into account.  Conceptually, if reliable TRVs were available for all of the species in the group, these could be arrayed in rank order to form a species sensitivity distribution (SSD), and the TRV selected to represent the group would be from the low end (e.g., 5th percentile) of the SSD.  A good example of this approach is the method used by EPA to derive national Ambient Water Quality Criteria (AWQC) values for aquatic receptors (see Sources for TRVs section below).

When this approach is used, if the HQ is less than 1, it may be concluded that most or all of the species in the group are protected.  If the HQ exceeds 1, it is not correct to assume that all species in the group are at risk.  Rather, it should be concluded that the most sensitive species of the group may be at risk, and it is unknown whether other (less sensitive) species of the group are or are not at risk.  In cases where the HQ for the group is only moderately above 1, some (perhaps many) of the individual species may not be at risk, with the threat being restricted only to the most sensitive species.  As the HQ for the group becomes larger, it is expected that more and more species in the group would be at risk.

Unfortunately, in real life, toxicity data are generally not available to define a reliable SSD for all the species of concern at a site, or even for all of the species in any particular group.  In these cases, the toxicity data are often evaluated on the basis of much broader taxonomic groups (e.g., all birds, all mammals).  The basic strategy used to select the TRV for the broad group is the same as described above:  the TRV is intended to be at the low end of the SSD for the broad group. 

It is important to recognize that this approach adds a second assumption, namely that inter-species variability in TRVs within a small group (e.g., raptors) is similar to the inter-specific variability within the broad group (e.g., all birds).  Conceptually, it is possible that this assumption is not valid.  For example, it might be true that all waterfowl tend to have TRVs near the high end of the all-bird SSD, while all passerines have TRVs near the low end of the all-bird SSD.  In most cases, it will not be known whether this hypothetical situation exists (if the data were good enough to answer the question, then it would be possible to develop small group-specific SSDs and TRVs).  The usual approach is to assume that the SSD for the broad group is a good surrogate for the SSD for each small group, and to employ a single TRV (based on the data for the broad group) to estimate the TRV for the most sensitive members of each small group.

[Top of page]

You will need the free Adobe Acrobat Reader to view some of the files on this page. See EPA's PDF page to learn more.

Sources of Ecological Toxicity Reference Values (TRVs)

As noted above, most TRVs are based on dose-response data published in the toxicity literature.  Numerous groups have derived TRVs for various groups of receptors, as summarized below.  In addition, the University of Tennessee has assembled a database of ecological benchmarks called SADA (Spatial Analysis and Decision Assistance). exit epa The SADA website contains links to many useful source documents, organized by medium:  surface water, sediment, and soil.

Surface Water TRVs for Aquatic Receptors

Toxicity values for the protection of aquatic life from contaminants in surface water are available from several sources.  EPA's Current National Recommended Water Quality Criteria, along with any state-specific water criteria, should be utilized as the primary source for surface water TRVs.  The Current National Recommended Water Quality Criteria table lists aquatic life, human health, and nutrient criteria for 158 pollutants as guidance for states and tribes in adopting water quality standards for surface waters.  Details on the basis of these values (chemical-specific criteria documents from the 1980s) may be found in the corresponding EPA Ambient Water Quality Criteria (AWQC) Reference Documents.  The Aquatic Life criteria list chemical concentration goals to protect surface water for aquatic life use.

Other documents that present data and aquatic TRVs include the following:

Bulk Sediment TRVs for Benthic Invertebrates

Toxicity values for the protection of benthic invertebrates from contaminants in sediment are available from several sources.  These TRVs include a variety of different types, depending on the nature of the data (e.g., no effect levels, threshold effects levels, probable effects levels) used to derive the TRVs.  Currently, EPA does not provide national recommendations for which set of TRVs are most appropriate for use in ecological risk assessment, but the consensus-based values derived by MacDonald et al. (2000) are usually preferred:

Alternative sediment-based TRVs that may be used when values are absent from MacDonald et al. (2000) include the following:

[Top of page]

Soil TRVs for Terrestrial Plants and Soil Organisms

Toxicity values for the protection terrestrial plants and soil organisms from contaminants in soil are available from several sources.  The Ecological Soil Screening Level (Eco-SSL) for plants should be utilized as the primary source for terrestrial plant and soil organism TRVs:

Other reports which contain TRVs for plants and/or soil organisms include the following:

Wildlife TRVs for Ingestion of Contaminants

There are three primary sources of dose-based wildlife TRVs (listed in order of preference below):

What to Do if No TRV Can Be Located

If a TRV for a particular chemical for a particular receptor species or group is not available in one of these sources, it may be possible to develop a TRV based on a review of published toxicity studies.  This task should not be undertaken without the help of an expert toxicologist.  The basic methods for performing this type of assessment are detailed in Ecological Soil Screening Level (Eco-SSL) Guidance and Documents.

[Top of page]

Chemical-Specific Resources and Guidance

Bioavailability

The consequences of exposure to a chemical depend on many things, including the degree to which the chemical is absorbed into the body from the environmental medium.  This is referred to as bioavailability.  A detailed discussion of bioavailability and how it may be measured at a site is provided in the human health bioavailability section of this Web site.  Nearly all of the basic concepts and approaches discussed in this section are also applicable to evaluation of exposures of wildlife receptors by the oral route.  Further information is provided in:

Some ecological receptors may also be exposed by direct contact with contaminated media (e.g., fish exposed to water, benthic organisms exposed to sediment, plants exposed to soil), and bioavailability may also be important in these situations.  However, rather than measuring bioavailability and adjusting risk estimates for these types of receptors, it is usually most efficient to perform toxicity tests using site-specific media (e.g., expose fish to site water, grow plants in site soil) (see section on Site-Specific Toxicity Testing).  One of the main advantages of this approach is that if bioavailability of a chemical in a site medium is low, this will be reflected in lower-than-expected toxicity in the test, and these results can be used to improve the accuracy of the risk assessment.

[Top of page]

Local Navigation

Jump to main content.