EPA-600/3-84-099
October 1984
Guidelines for Deriving Numerical Aquatic Site-Specific
Water Quality Criteria by Modifying National Criteria
Anthony R. Carlson3, William
David
by
A. 3rungsb, Gary A. Chapnanc, and
J. Hansen^
a 'J.S. EPA, Environmental Research Laboratory, Duluth, Minnesota
^ U.S. EPA, Environmental Research Laboratory, Narragansett, Rhode Island
c U.S. E?A, Environmental Research Laboratory, Corvallis, Oregon
d U.S. EPA, Enviroa-nental Research Laboratory, Gulf Breeze, Florida
LXVI^ONMENTAL RESEARCH LABORATORY
Oi-TlC:-; OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
-------
NOTICE
This document has been reviewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication. Mention of trade names
or commercial products does net constitute endorse-
ment or recommendation for use.
ii
-------
INTRODUCTION
Relationship to the National Guidelines
These Guidelines for Deriving Numerical Aquatic Site-Specific Water
Quality Criteria by Modifying National Criteria (hereinafter referred to as
the Site-Specific Guidelines) are the next steps evolving from the Guidelines
for Deriving Numerical National Water Quality Criteria for the Protection of
Aquatic Life and Its Uses (U.S. Environmental Protection Agency, 1983) (here-
inafter referred to as the National Guidelines).
In that the Site-Specific Guidelines follow from the National
Guidelines, an understanding of the National Guidelines and the national
criteria document for the material of interest is a prerequisite for
understanding and use of the Site-Specific Guidelines. The derivation of a
site-specific criterion for freshwater or saltwater aquatic life will
generally evolve from national criteria that are available for a limited
number o? chemicals (Appendix 1). Site-specific criteria derived by these
guidelines nay be the same as, or higher or lower than national criteria.
In the absence of a national criterion, additional data may be generated
so that the minimum data set requirements of the National Guidelines are met
and a national or site-specific criterion may be calculated.
The national water quality criteria have been developed using guideline
procedures that have undergone extensive scientific review regarding their
general applicability. States may choose to apply these criteria directly or
to modify them according to site-specific criteria guidelines. Whenever
decisions are sought regarding modification of these criteria, the assistance
of those biologists, chemists, hydrologists, and toxicologists most
knowledgeable of the local species and conditions is essential to the proper
evaluation of exposure assessment and population at risk.
1
-------
Rationale for the Site-Specific Guidelines
National criteria may be underprotective or overprotective because: (1)
The species at the site are more or less sensitive than those included in the
national criteria data set. (2) The physical and/or chemical characteristics
of the water at the site alters the biological availability and/or toxicity
of the material. Therefore, it is appropriate that the individual
Site-Specific Guidelines procedures address each of these conditions
separately, as well as the combination of the two.
Site-specific criterion derivation may be justified because species at
the sits may be more or less sensitive than those in the national criterion
document. For example, the national criteria data set contains data for
trout, salmon, or penaeid shrimp, aquatic species that have been shown to be
especially sensitive to some materials. Because these or other sensitive
species may not occur at a particular site, they may not be representative of
those species that do occur there. Conversely, there may exist at a site
untested sensitive species that are ecologically or economically important
and would need to be protected. Secondly, differences in physical and
r.henica"! characteristics of water have been demonstrated to ameliorate or
enhance the biological availability and/or toxicity of chemicals in
freshwater and saltwater environments. Alkalinity, hardness, pH, suspended
solids and/or salinity influence the concentrations) of the toxic form(s) of
some heavy metals, ammonia and other chemicals. For some materials, hardness
or pH-dependent national criteria are available for fresh water. No
salinity-dependent criteria have been derived because most of the saltwater
data for heavy metals has been developed in high salinity waters. However,
in so-ne estuarine sites where salinity may vary significantly, the
2
-------
development of salinity-dependent site-specific criteria for metals of local
interest nay be appropriate.
The effect of seasonality on the physical and chemical characteristics
of water and subsequent effects on biological availability and/or toxicity of
a material, may also justify seasonally dependent site-specific criteria.
The major implication of seasonally dependent criteria is whether or not the
"most sensitive" time of the year coincides with that time for which the flow
is the basis for waste treatment facilities design or NPDES permits. That
is, if the physical and chemical characteristics of the water during low flow
seasons increases the biological availability and/or toxicity of the chemical
of concern, the permit limitations may be more restrictive than if the
converse relationship were to apply.
Definition of Site
Since the rationales for the Site-Specific Guidelines are usually based
on potential differences in species sensitivity, physical and chemical
characteristics of the water, or a combination of the two, the concept of
site cust be consistent with this rationale.
A site nay be a single point source discharge or quite large. If water
quality effects on toxicity are not a consideration, the site will be as
large as a generally consistent biogeographic zone permits. In this case,
for example, large portions of the Chesapeake Bay, Lake Michigan, or the Ohio
River may each be considered as one site because their respective aquatic
communities do not vary substantially. Unique populations or less sensitive
use within sites may justify a designation as a distinct site (site within a
site). When sites are large, the necessary data generation can be more
economically supportable.
3
-------
If the selected species of a site are toxicologically comparable to
those In the national criteria data set for a material of interest, and
physical and/or chemical water characteristics are the only factors
supporting modification of the national criteria, then the site would be
defined on the basis of expected changes in the material's biological
availability and/or toxicity due to physical and chemical variability of the
site water.
Two additional considerations in defining a site are: 1) viable
connunities must occur, or be historically documented, in order to select
resident species for use in deriving site-specific criteria, and 2) the site
oust contain acceptable quality dilution water if site water will be required
for testing (to be discussed later in these Guidelines).
For the purpose of the Site-Specific Guidelines, the term "selected
resident species" is defined as those species that commonly occur in a site
including those that occur only seasonally (migration) or intermittently
(periodically returns or extends its range into the site). It is not
intended to include species that were once present in that site and cannot
return due to physical habitat alterations.
Selection of a resident species should be designed to account for
differences between the sensitivities of the selected resident species and
those in the national data set. There are several possible reasons for this
potential difference. The principal reason is that the resident communities
in a site may represent a more or less narrow mix of species due to a limited
range of natural environmental conditions (e.g., temperature, salinity,
habitat, or other factors affecting the spatial distribution of aquatic
species). The number of resident species will generally decrease as the size
of the site decreases.
4
-------
A second potential reason for a real difference in sensitivity could be
the absence of most of the species or groups of species (e.g., families) that
are traditionally considered to be sensitive to certain, but not all,
materials (e.g., trout, salmon, saltwater penaeid shrimp, and Daphnia magna).
Predictive relative species sensitivity does not apply to all materials, and
the assumption that sensitive species are unique rather than representative
of equally sensitive untested species is tenuous. A final reason could be
that the resident species may have evolved a genetically based greater
resistance to high concentrations of a material, but no data have been
presented to demonstrate such a genetic difference. A few instances of
increased resistance have been suggested but they may be due to an
acclimation of individual organisms to a stress. However, such an
acclimation, should it occur, would be transitory.
Assumptions
There are numerous assumptions associated with the Site-Specific
Guidelines, most of which also apply to and have been discussed in the
National Guidelines. A few need to be emphasized. The principal assumption
is that the species sensitivity ranking and toxicological effect endpoints
(e.g., death, growth, or reproduction), derived from appropriate laboratory
tests will be similar to those in site situations. Another assumption is
that protection of all of the site species all of the time is not necessary
because aquatic life can tolerate some stress and occasional adverse
effects.
It is assumed that the Site-Specific Guidelines are an attempt to
protect more correctly the various uses of aquatic life by accounting for
toxicological differences in species sensitivity or the biological
5
-------
availability, and/or toxicity of a material at specific sites. Modification
of the data set must always be scientifically justifiable and consistent with
the assumptions, rationale, and spirit of the National Guidelines.
Site-specific criteria are designed to be used by the States to develop
water quality standards, mixing zone standards, or toxicity based effluent
standards. The development of such standards should take into account
additional factors such as the use of the site, and social, legal, and
economic considerations as they impact the site, the environmental and
analytical chemistry of the material, the extrapolation from laboratory data
to site situations, and the relationship between the species for which data
are available and the species in the body of water which is to be protected.
Heavy Metal Speciation
The national criteria for metals are established primarily using
laboratory d*ta in which reported effect concentrations have been analyzed
primarily as total, total recoverable, or acid extractable metal
concentrations. Consequently, the national criteria are expressed as total
recoverable metal. Metals exist in a variety of chemical forms in water.
Available toxicological data have demonstrated that some forms are much more
toxic than others. Most of the toxicity appears to reside in the soluble
fraction and, potentially, in the easily labile, nonsoluble fraction. The
national criteria values may be unnecessarily stringent if applied to total
metal measurements in waters where total metal concentrations include a
preponderance of metal forms which are highly insoluble or strongly
complexed. Derivation of criteria based on metal forms is not possible at
this time because adequate laboratory or field data bases do not exist in
which net^l toxicity is partitioned among the various metal forms. Analysis
6
-------
of total and soluble metal concentrations when soluble metal is added to site
water may indicate that the metal is rapidly converted to insoluble forms or
to other forms with presumed low biological availability- Under these
circumstances, derivation of a site-specific criterion based on site-water
effect in either the indicator or resident species procedures will probably
result in less stringent criteria values.
Use of the indicator species or resident species procedures is
encouraged for derivation of site-specific criteria for those metals whose
biological availability and/or toxicity is significantly affected by
variation in physical and/or chemical characteristics of water. Measurement
of both total recoverable and soluble metal concentrations during toxicity
testing is recommended.
Plant anc Other Data
In the published criteria documents, no national criterion is based on
plant data or "Other Data" (e.g. flavor impairment, behavioral, etc*). For
some materials, observed effects on plants occurred at concentrations near
the criterion. The following procedures do not contain techniques for
handling such data, but if a less stringent site-specific criterion is
derived, those data may need to be considered.
PROCEDURES
There are three procedures in these Site-Specific Guidelines for
modifying the national criterion which is composed of both a maximum
concentration and a 30-day average concentration. These procedures are:
A. The recalculation procedure for the derivation of a site-specific
criterion to account for differences in resident species sensitivity to a
material.
7
-------
B. The indicator species procedure for the derivation of a site-specific
criterion for a material to account for differences in biological
availability and/or toxicity of a material caused by physical and/or
chemical characteristics of a site water.
C. The resident species procedure for the derivation of a site-specific
criterion to account for differences in resident species sensitivity and
differences in the biological availability and/or toxicity of a material
due to physical and/or chemical characteristics of a site water.
The following is the sequence of decisions to be made before any of the
above procedures are initiated:
• Define the site boundaries.
• Determine from the national criterion document and other sources if
physical and/or chemical characteristics are known to affect the
biological availability and/or toxicity of a material of interest.
• If data in the national criterion document and/or from other sources
indicate that the range of sensitivity of the selected resident
species to the material of interest is different from that range for
the species in the national criterion document and variation in
physical and/or chemical characteristics of the site water is not
expected to be a factor, use the recalculation procedure (A).
• If data in the national criterion document and/or from other sources
indicate that physical and/or chemical characteristics of the site
water may affect the biological availability and/or toxicity of the
material of interest, and the selected resident species range of
/
sensitivity is similar to that for the species in the national
criterion document, use the indicator species procedure (B).
8
-------
• If data in the national criterion document and/or from other sources
indicated that physical and/or chemical characteristics of the site
water -nay affect the biological availability and/or toxicity of the
material of interest, and the selected resident species range of
sensitivity is different from that for the species in the national
criterion document, use the resident species procedure*
The following Figures 1 and 2 are generalized flow charts for these
Guidelines .
X. Recalculation procedure for the derivation of a site-specific criterion
to account for differences between selected resident species and other
spec ies«
1. SuTnary: This recalculation procedure allows modifications in the
national data set on the basis of eliminating data for species that
are aot resident at that site. When the recalculation procedure for
the site-specific Final Acute Value results in a reduction in the
national data base below the minimum data set requirements,
additional resident species testing in laboratory water is
necessary.
2. Rationale: This procedure is designed to account for any real
difference between the sensitivity range of species represented in
the national data set and species found at a site.
3. Conditions:
• If acute toxicity data for resident species are insufficient to
meet the minimum data set requirements of the National Guide-
lines, additional acute toxicity data in laboratory water for
untested resident species would be needed before a calculation of
the site-specific criterion could be made.
9
-------
r i c u r e i
Generalized Flow Chart for Oerlvlng Nuerrlcel 81te-Speolflo Criteria
for th» Proteotlon of Aquatlo Llf» md Its U»«».
This It an Illustration of procedures that oan be used to
derive the Conovntrst ion of a two-pert or Iter I on
Derivation of the 90-day Averaoe ConoentratI on la
Illustrated In Figure C.
To Hodlfg
Haxlaus
ConoentratIon
HAXIHUH
CONCENTRATION
REVIEW
REASONS FOR
MODIFICATION
SIte-Speolf lo
Final Route
Value
Water
CharaoterIstIo
To Hodlfg
90 - Dag
Averaoe
Conoentratlor
Both Speoles
Sensatlvltg
and Water
Charaoterlstlo
Dlfferenoes
Complete Mlnlaua
Oata Set with
Resident Speoles In
Site Water
Test Indloator
Speoles In Site and
Laboratory
Waters to Cet a
"Water Effeot Ratio
Used to Modify
National Value
Reoaloulate Final
Aoute Value
(If Neoesaary
Conplete Hlnlaua
Data Set with
Resident Speoles In
Laboratory Water I
-------
figure: e
Generalized Flow Chart for Deriving Numerical 81te-SpeoIflo Criteria
for the Proteotlon of Aquatlo Life and Its Uitt.
This Is an Illustration of procedures that oan be used to
derive the 96-d»y Awrso* Concentration of a two-part
orlterlon. Derivation of the HikImuh ConoentratI on Is
Illustrated In Figure 1.
REVIEW
REASONS FOR
MODIFICATION
To Mod If y
NamI rum
ConoentratI on
See Flguro I
Flna
Residue
To Modify
30 - Oay
Average
ConoentratIon
Final Plant
Value
Othi
Data
Final Chronlo
Value
If National Final
Plant Value Drives
90 - Day Average
Concentration -
Conduot Plant Testis)
In Site Water
If Not* Retain
National Final
Plant Value
For Lipid Soluble
Materials Adjust
National Final
Residue Value for
X Haxlaua Lipids
In Site Speoles -
Caloulate a
Residue Value
For Non-1Ipld Soluble
Materials (e.g.Meroury
Conduot Tests with
Resident Speoles In
Site Water and
Caloulate a Residue
Value
Aooept National
Aoute-Chronlo Ratio
and Olvlde Into
SIte-SpeolfIo
Final ftcute Value
Test Indicator Species
In Site and Laboratory
Waters to get a "Water
Effect Ratio" - use to
Modify National Value
Lowest Important or
B(ologIoaI Iu
Significant Value for
Resident Speoles
-)i
Deters!ne
Slte-Speolflo Final
Aoute-Chronlo Ratio
and Olvlde Into
Slte-Speolflo
Final Route Value
81te-Speolflo
Final Plant
Value
Slte-Speolf lo
Final Residue
Value
81te-SpeoIf lo
Final Chronlo
Value
61te-SpeoIflo
Other Data
CHOOSE L0WC6T
OF FOUR AS
SI - DAY
AVERAGE
CONCENTRATION
-------
Certain families or organisms have been specified to be
represented In the National Guidelines acute toxicity minimum
data set (e.g., Salmonidae in fresh water and Penaeidae or
Mysidae in salt water). If this or any other requirement cannot
be met because the family or other group (e.g., insect or benthic
crustacean in fresh water) is not represented by resident
species, select a substitute(s) from a sensitive family
represented by one or more resident species and meet the 8 family
minimum data set requirement. If all the families at the site
have been tested and the minimum data set requirements have not
been met use the most sensitive resident family mean acute value
as the site-specific Final Acute Value.
Due to the emphasis this procedure places on resident species
testing when the minimum data set has not been met, there may be
difficulty in selecting resident species compatible to laboratory
testing. Some culture and/or handling techniques may need to be
developed.
chronic testing is required by this procedure since the
national acute-chronic ratio will be used with the site-specific
Final Acute Value to obtain the site-specific Final Chronic
Value.
For the lipid soluble chemicals whose national Final Residue
Values are based on Food and Drug Administration (FDA) action
levels, adjustments in those values based on the percent lipid
content of resident aquatic species is appropriate for the
derivation of site-specific Final Residue Values.
12
-------
For lipid-soluble materials, the national Final Residue Value is
based on an average 11 percent lipid content for edible portions
for the freshwater chinook salmon and lake trout and an average
of 10 percent lipids for the edible portion for saltwater
Atlantic herring. Resident species of concern may have higher
(e.g.. Lake Superior siscowet, a race of lake trout) or lower
(e.g., many sport fish) percent lipid content than used for the
national Final Residue Value.
For some lipid-soluble materials such as polychlorinated
biphenyls (PCB) and DDT, the national Final Residue Value is
based on wildlife consumers of fish and aquatic invertebrate
species rather than an FDA action level because the former
provides a more stringent residue level (see National Guidelines
for details). Since the data base on the effects of ingested
aquatic organisms on wildlife species is extremely limited, it
would be inappropriate to base a site-specific Final Residue
Value on resident wildlife species. Consequently, site-specific
modification for chose materials is based on percent lipid
content of resident species consumed by humans.
For the lipid-soluble materials whose national Final Residue
Values are based on wildlife effects, the limiting wildlife
species (mink for PCB and brown pelican for DDT) are considered
acceptable surrogates for resident avian and mammalian species
(e.g., herons, gulls, terns, otter, etc.)* Conservatism is
appropriate for those two chemicals, and no less restrictive
modification of the national Final Residue Value is appropriate.
13
-------
The site-specific Final Residue Value would be the same as the
national value.
4. Details of Procedure:
• If the minimum data set requirements are met as defined in the
National Guidelines or through substitution of one or more sensi-
tive resident family(ies) for non-resident family(ies) or
group(s) required in the National Guidelines, calculate a site-
specific Final Acute Value using all available resident species
data in the national document and/or from other sources. If all
the fa-iilies at the site have been tested and the minimum data
set requirements have not been met use the most sensitive resident
family mean acute value as the site-specific Final Acute value.
• If the minimum data set requirements are not met, satisfy those
requirements with additional testing of resident species in
Laboratory water.
• If all species in a family at the site have been tested, then
their Species Mean Acute Values should be used to calculate the
site-specific Family Mean Acute Value and data for non-resident
species in that family should be deleted from that calculation.
If all resident species in that family have not been tested, the
site-specific Family Mean Acute Value would be the same as the
national Family Mean Acute Value.
• To derive the site-specific maximum concentration divide the
site-specific Final Acute Value by 2.
• Divide the site-specific Final Acute Value by the national Final
Acute-Chronic Ratio to obtain the site-specific Final Chronic
Value.
14
-------
• When a site-specific Final Residue Value can be derived for lipid
soluble materials controlled by FDA action levels> the following
recalculation equation would be used:
site-specific Final Residue Value =
FDA action level
(mean normalized BCF from criterion document) (appropriate % lipids)
where the appropriate percent lipid content is based on consumed
resident species. A recommended method to determine the lipid
content of tissues is given in Appendix 2.
• For PCB and DDT whose national Final Residue Values are based on
wildlife consumers of aquatic organisms, no site-specific
modification procedure is appropriate.
• In the case of mercury (a non-lipid-soluble material), a
site-specific Final Residue Value can be derived by conducting
acceptable bioconcentration tests with edible aquatic resident
species using accepted test methods (Appendix 2) or the national
value can be accepted as the site-specific value. For a
saltwater residue value, use a bivalve species (the oyster is
preferred), and for a freshwater value, use a fish species.
These taxa yield the highest known bioconcentration factors for
metals. The following recalculation equation would be used:
site-specific Final Residue Value 3 FDA action level
site-specific BCF
• The lower of the site-specific Final Chronic Value and the
site-specific Final Residue Value becomes the site-specific
30-day average concentration unless plant or other data indicate
a lower value is appropriate. If a problem is identified,
15
-------
judgment should be used in establishing the site-specific
cri terion.
imitations:
Whatever the results of this recalculation procedure may be, a
decision should be made as to whether the numerical differences,
if any, are sufficient to warrant changes in the criterion.
The number of families used to calculate any Final Acute Value
significantly affects that value. Even though the four lowest
Family Mean Acute Values (most sensitive families) are most
important in that calculation, the smaller N is, the lower the
Final Acute Value. Consequently, if none of the four most
sensitive families are changed or deleted, any reduction in N
will result in a lower Final Acute Value. Changes in or
deletions of any of the four lowest values, regardless of whether
X is changed, may result in a higher or lower Final Acute Value.
Site-specific or national Final Residue Values based on FDA
action levels may not precisely protect that use since the FDA
action levels are adverse (i.e., loss of marketability).
Bioaccumulation, except in field studies, does not add to the
laboratory-derived bioconcentration factors because the
laboratory procedures preclude food chain uptake. Consequently,
some residue levels obtained by laboratory studies of
bioconcentration (direct uptake of the material from water) may
underestimate potential effects encountered at a site. The
magnitude of site-specific bioconcentration factors obtained in
the laboratory, therefore, may be insufficient to protect the
public from the effects of the ingested material of concern.
16
-------
Indicator species procedure for the derivation of a site-specific
criterion for a material to account for differences in the biological
availability and/or toxicity of a material due to physical and/or
chemical characteristics of a site water.
1. Summary: This procedure is based on the assumption that physical
and/or chemical characteristics of water at an individual site may
influence biological availability and/or toxicity of a material.
Acute toxicity in site water and laboratory water is determined using
species resident in the site, or acceptable nonresident species, as
indicators or surrogates for species found at the site. The
difference in toxicity values, expressed as a water effect ratio, is
used to convert the national maxijnu-i concentration for a material to
a site-specific maximum concentration fron which a site-specific
Tinal Acute Value is derived.
This procedure also provides three ways to obtain a site-specific
Final Chronic Value. It may be (1) calculated (no testing required)
if an applicable Final Acute-Chronic Ratio for a given material is
available in the national criteria document. This ratio is simply
divided into the site-specific Final Acute Value to obtain the
site-specific Final Chronic Value; (2) obtained by performing two
acute and chronic toxicity tests which include both a fish and
invertebrate species (resident or non-resident) in site water.
Acute-chronic ratios are calculated for each species, and the
geometric mean of these ratios is then divided into the site-specific
Final Acute Value to obtain the site-specific Final Chronic Value;
and (3) obtained by performing chronic toxicity tests with at least
one fish and one invertebrate (resident or non-resident) in both
-------
laboratory water and site water and calculating a geometric mean
chronic water effect ratio which is used to modify the national Final
Chronic Value.
2. Rationale: This procedure is designed to compensate for site water
which may markedly affect the biological availability and/or toxicity
of a material. Major factors affecting aquatic toxicity values of
many materials, especially the heavy metals, have been identified.
For example, the carbonate system of natural waters (pH, hardness,
alkalinity, and carbon dioxide relationships) has been the most
studied and quantified with respect to effects on heavy metal
biological availability and/or toxicity in freshwater; however, the
literature indicates that in natural systems organic solutes,
inorganic and organic colloids, salinity and suspended particles also
play an important but less quantifiable role in the biological
availability and/or toxicity of heavy metals to aquatic life.
This procedure provides a means of obtaining a site-specific Final
Chronic Value for a material when the acute-chronic ratios in the
national criteria document are thought to be inapplicable to
site-specific situations.
3. Conditions:
• There is no reason to suspect that the resident species
sensitivity is different from those species in the national data
set.
• The toxic response seen in the tests used in the development of
the national water quality criterion would be essentially the
same if laboratory test water required in this procedure had been
used instead.
13
-------
• Differences in the toxicity values of a specific material
determined in laboratory water and site water may be attributed
to chemical (e.g., complexing ligands) and/or physical (e.g.,
adsorption) factors that alter the biological availability and/or
toxicity of the material.
• Selected indicator species directly integrate differences in the
biological availability and/or toxicity of a material. They
provide a direct measure of the capacity of a site water to
increase or decrease toxicity values relative to values obtained
in laboratory water.
• National Final Acute-Chronic Ratios for certain materials can be
used to establish site-specific Final Chronic Values.
• A site-specific acute-chronic ratio, obtained in site water
testing, reflects the integrated effects of the physical and/or
chemical characteristics of water on toxicity values.
• The water effect ratio concept used in this procedure for
modifying national Final Acute Values to site-specific situations
is also applicable to modifying national Final Chronic Values to
site-specific situations.
4. Details of Procedure:
• Test at least two indicator species, a fish and an invertebrate,
using laboratory dilution water and site dilution water according
to acute toxicity test procedures recommended in Appendix 2. For
each species, use organisms from the same population and conduct
the tests at the same time and, most importantly (except for the
water source) under similar conditions (e.g., temperature,
lighting, etc.)- Measure the concentration of the material in
19
-------
Che acute toxicity tests; the concentration must be within the
solubility limits of the material. To avoid solubility problems,
species selected for testing should be among the most sensitive
to the material of interest (screening tests may be necessary)*
Conpare the laboratory and site water LC50 values for each
indicator species to determine if they are different (P<0.05)
(see statistical procedure in Appendix 3). If the LC50 values
are not different, then the national maximum concentration is the
site-specific maximum concentration. If the LC5Q values are
different, calculate the water effect ratio for each species
according to the following equation:
Water Effect Ratio - Site Water LC50 Value
Laboratory Water LC50 Value
Determine if the two ratios are statistically different (P<0.05)
(see Appendix 3).
If the two ratios are not different calculate the geometric mean
of the water effect ratios. The site-specific maximum concentra-
tion can be calculated by using this geometric mean water effect
ratio in the following equation: site-specific maximum concen-
tration = water effect ratio x the national maximum concentration
(or x the national maximum concentration adjusted to a water
characteristic of the laboratory water when appropriate).
If the two ratios are different, additional tests may have to be
conducted to confirm or refute the data. In such cases
professional judgment is appropriate in determining if some or
none of the ratio data can be used to modify the national maximum
concentration.
20
-------
The site-specific maximum concentration is multiplied by 2 to
obtain the site-specific Final Acute Value which is used to
calculate the site-specific Final Chronic Value.
If the national Final Acute-Chronic Ratio for the material of
interest was used to establish a national Final Chronic Value,
the site-specific Final Chronic Value may be calculated using the
acute-chronic ratio in the following equation:
Site-Specific Chronic Value = Site-Specific Acute Value
Final Acute-Chronic Ratio
If the national Final Acute-Chronic Ratio was not used to
establish a national Final Chronic Value, the national Final
Chronic Value may be used as the site-specific Final Chronic
Value, or it may be measured by performing 2 acute and 2 chronic
tests, (Appendix 2) using site water. Test at least one fish and
one invertebrate species, and conduct an acute test using site
water of similar quality. These data are used to calculate an
acute-chronic ratio for each species. If these ratios are within
a factor of 10, the geometric mean of the 2 acute-chronic ratios
(the site-specific Final Acute-Chronic Ratio) is used to
calculate the site-specific Final Chronic Value using the
following equation:
Site-Specific Final Chronic Value =
Site-Specific Final Acute Value
Site-Specific Final Acute-Chronic Ratio
After an acute/chronic ratio is determined for one species and if
that ratio is within the range of the values used to establish
the national acute-chronic ratio, it is recommended that the
21
-------
site-specific ratio be used in recalculating the national ratio.
This recalculated ratio would then be used as the site-specific
Final Acute-Chronic Ratio in the above equation.
• A site-specific Final Chronic Value can be obtained by testing
indicator species for chronic toxicity. Test at least two
indicator species, a fish and an invertebrate, using laboratory
dilution water and site dilution water according to chronic
toxicity test procedures recommended in Appendix 2. For each
species, use organisms from the same population, and conduct
tests at the same time and most importantly (except for the water
source) under similar conditions (e.g., temperature, lighting).
The concentration of the material in the toxicity tests must be
within the solubility limits of the material. To avoid
solubility problems, species selected for testing should be among
the most sensitive to the material of interest (screening tests
may be necessary).
Compare the laboratory and site water chronic values for each of
the indicator species to determine if they are reasonably
different (limits of chronic values do not overlap).
If for a species the chronic values are not different, the water
effect ratio = 1.0.
If the chronic values are different, calculate the water effect
ratio for each species according to the following equation:
Chronic Water Effect Ratio =
Chronic Value in Site Water
Chronic Value in Laboratory Water
22
-------
Calculate the geometric mean of the water effect ratios for the
species tested.
If the nean water effect ratio is not different from 1.0, the
national Final Chronic Value is the site-specific Final Chronic
Value.
If the mean water effect ratio is different from 1.0, the
site-specific Final Chronic Value can be calculated by using the
following equation: site-specific Final Chronic Value = Chronic
Water Effect Ratio x the national Final Chronic Value (or the
national Final Chronic Value adjusted to a water quality
characteristic of the laboratory water when appropriate).
The site-specific Final Chronic Value is used in the
determination of the site-specific 30-day average concentration.
• The lower of the site-specific Final Chronic Value and the
recalculated site-specific Final Residue Value (as described in
the recalculation procedure) becomes the site-specific 30-day
average concentration unless plant or other data (including data
obtained from the site-specific tests) indicates a lower value is
appropriate. If a problem is identified, judgment should be used
in establishing the site-specific criterion.
Limitations:
• If filter feeding organisms are determined to be among the most
sensitive to the material of interest from the national criteria
document and/or other sources, and members of the same group are
important components of the site food web, a member of that
group, preferably a resident species, should be tested in order
to discern differences in the biological availability and/or
-------
toxicity of the material of interest due to ingested
particulates.
Site water for testing purposes should be obtained under typical
conditions and can be obtained at any time of the day or season.
Storm or flood impacted water is unacceptable as test water in
the acute tests used to calculate water effect ratios and
acute/chronic ratios but is acceptable test water for short
periods of time in long-terra chronic tests used to calculate
these ratios. There are some special cases when storm impacted
wacer is acceptable in acute toxicity testing for use in criteria
development. Tor example, an effluent discharge may be allowed
only during high water periods, or a non-point source of a
chemical pesticide may be of most concern during storm-related
runoff events.
Site water must not be influenced by effluents containing the
material of interest or effuents that may impact the material's
bioavailability and/or toxicity. The site water should be used
as soon as possible after collection in order to avoid signifi-
cant changes in its physical and chemical characteristics. If
diurnal cycles in water characteristics (e.g., carbonate systems,
salinity, dissolved oxygen) are known to affect a material's
biological availability and/or toxicity markedly, use of on-site
flow-through testing is suggested; otherwise transport of water
to off-site locations is acceptable. During transport and
storage, care should be taken to maintain the quality of the
water; however, certain conditions of the water such as pH and
dissolved oxygen concentration may change and the degree of these
changes should be measured and reported.
-------
• Seasonal site-specific criteria can be derived if monitoring data
are available to delineate seasonal periods corresponding to
significant differences in water characteristics (e.g., carbonate
systems, salinity, turbidity).
• The frequency of testing (e.g., the need for seasonal testing)
will be related to the variability of the physical and chemical
characteristics of site water as it is expected to affect the
biological availability and/or toxicity of the material of
interest. As the variability increases, the frequency of testing
will increase.
• With the exception that storm or flood impacted water may be used
in chronic toxicity tests, the limitations on the use of
indicator species to derive a site-specific Final Chronic Value
are the same as those for site-specific modification of a
national Final Acute Value.
C. Resident species procedure for the derivation of a site-specific
criterion to account for differences in resident species sensitivity and
differences in biological availability and/or toxicity of a material due to
variability i?. physical and chemical characteristics of a site water.
1. Summary: Derivation of the site-specific maximum concentration and
site-specific 30-day average concentration would be accomplished
after the complete acute toxicity minimum data set requirements have
been met by conducting tests with resident species in site water.
Chronic tests may also be necessary.
2. Rationale: This procedure is designed to compensate concurrently for
ar.y real differences between the sensitivity range of species
represented in the national data set and for site water which nay
25
-------
markedly affect the biological availability and/or toxicity of the
material of interest.
3. Conditions:
• Develop the complete acute toxicity minimun data set using site
water and resident species.
4. Details of Procedure:
• Complete the acute toxicity minimum data set test requirements
using site water and derive a site-specific Final Acute Value.
• The guidance for site water testing has been discussed in the
indicator species procedure (B).
• Certain families of organisms have been specified in the National
Guidelines acute toxicity minimum data set (e.g., Salmonidae in
fresh water and Penaeidae or Mysidae in salt water); if this or
any other requirement cannot be met because the family or other
group (e.g. insect or benthic crustacean) in fresh water is not
represented by resident species, select a substitute(s) from a
sensitive family represented by one or more resident species and
meet the 8 family minimum data set requirement. If all the
families at the site have been tested and the minimum data set
requirements have not been net use the most sensitive resident
family mean acute value as the site-specific Final Acute Value.
• To derive the site-specific maximum concentration divide the
site-specific Final Acute Value by two.
• The site-specific Final Chronic Value can be obtained as
described in the indicator species procedure (B). An exception
is that a chronic water effect ratio should not be used to
calculate a Final Chronic Value.
26
-------
• The lower of the site-specific Final Chronic Value and the
recalculated site-specific Final Residue Value (as described in
the recalculation procedure) becomes the site-specific 30-day
average concentration unless plant or other data (including data
obtained from the site-specific tests) indicates a lower value is
appropriate. If a problem is identified, judgment should be used
in establishing the site-specific criterion.
5. Limitations:
• The frequency of testing (e.g., the need for seasonal testing) will
be related to the variability of the physical and chemical charac-
teristics of site water as it is expected to affect the biological
availability and/or toxicity of the material of interest. As the
variability increases, the frequency of testing will increase.
• Many of the limitations discussed for the previous two procedures
would also apply to this procedure.
This draft of the Site-Specific Guidelines was written by Anthony R.
Carlson, William A. Brungs, Gary A. Chapnan, and David J. Hansen under the
direction of the Site-Specific Criteria Committee of George S. Baughman,
Gillian A. Brungs, Anthony R. Carlson, Ronald G. Garton, David J. Hansen,
Douglas A. Lipka, Alan B. Rubin, and Rosemarie C. Russo. John H. Gentile,
Robert L. Spehar, and Charles E. Stephan provided review and comments. These
efforts were supported by the U.S. Environmental Protectin Agency's
Environmental Research Laboratories in Athens, Georgia; Corvallis, Oregon;
Duluth, Minnesota; Gulf Breeze, Florida; and Narragansett, Rhode Island. The
Office of Water Regulations and Standards' Criteria and Standards Division
and the Office of Research and Developments Office of Environmental
Processes and Effects Research also supported these efforts.
27
-------
REFERENCES
U.S. Environmental Protection Agency. 1983. Guidelines for deriving numer-
ical national water quality criteria for the protection of aquatic life
and its uses- Draft July 5, 1983. U.S. EPA, Environmental Research
Laboratories at Duluth, MN; Gulf Breeze, FL; Narragansett, RX; and
Corvallis, OR.
28
-------
APPENDIX 1
FRESHWATER AND SALTWATER NATIONAL CRITERIA LIST
(x - criteria are available)
Chemical
Aldrin
Ammonia
Dieldrin
Chlordane
DDT -S Metabolites
Endosulfan
Endrin
Heptachlor
Lindane
Toxaphene
Arsenic(III)
Cadmium
Chlorine
Chror.ium(VI)
Chromium(III)
Copper
Cyanide
Lead
Mercury
Nickel
Seleniura(IV)
Silver
Zinc
Freshwater
x
x
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Saltwater
x
x
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
29
-------
APPENDIX 2
TEST METHODS
The following procedures are recommended for conducting tests with aquatic
organises, including fishes, invertebrates, and plants. These procedures are
the state-of-the-art based on currently available information.
Because all details are not covered in the following procedures,
experience in aquatic toxicology, as well as familiarity with the pertinent
references listed, is needed for conducting these tests satisfactorily.
Requirements concerning tests to determine the toxicity and bioconcentra-
tion of a material in aquatic organisms are given in the National Guidelines.
A. ACUTE TESTS:
American Public Health Association, American Water Works Association, and
Water Pollution Control Federation. 1980. Standard methods for the
examination of water and wastewater. 15th ed. American Public Heath
Association, Washington, D.C. 1134 p.
American Society for Testing and Materials. 1980. Standard practice for
conducting acute toxicity tests with fishes, macroinvertebrates, and
amphibians. Standard E 729-80, American Society for Testing and
Materials, Philadelphia, Penn. 25 p.
American Society for Testing and Materials. 1980. Standard practice for
conducting static acute toxicity tests with larvae of four species of
bivalve molluscs. Standard E 724-80, American Society for Testing and
Materials, Philadelphia, Penn. 17 p.
30
-------
3. CHRONIC TESTS:
American Public Health Association, American Water Works Association, and
Water Pollution Control Federation. 1980. Standard methods for the
examination of water and wastewater. 15th ed. American Public Health
Association, Washington, D.C. 1134 p.
Anerican Society for Testing and Materials- Proposed standard practice for
conducting toxicity tests with early life stages of fishes. S. C.
Schi-nmel (Task. Group Chairman). American Society for Testing and
terials, Philadelphia, Penn. (latest draft).
Anerican Society for Testing and Materials. Proposed standard practice for
conducting Daphnia magna renewal chronic toxicity tests. R. M.
Colotto (Task Group Chairman). Anerican Society for Testing and
Materials, Philadelphia, Penn. (latest draft).
Anerican Society for Testing and Materials. Proposed standard practice for
conducting Daphnia magna chronic toxicity tests in a flow-through
svsten. *1. J. Adans (Task Group Co-chairman). American Society for
Testing and Materials, Philadelphia, Penn. (latest draft.)
American Society for Testing and Materials. Proposed standard practice for
conducting life cycle toxicity tests with saltwater oysid shrimp.
Susan Gentile and Charles McKenny (Task Group Co-chairman). American
Society for Testing and Materials, Philadelphia, Penn. (latest
draft.)
Benoit, D. A. 1982. User's guide for conducting life-cycle chronic
toxicity tests with fathead minnows (Plaiephales promelas).
EPA-600,/3-81 -011, U.S. EPA, Environmental Research Laboratory, Duluth,
Minn -
31
-------
c. FISH LIPID ANALYSIS PROCEDURE:
Approximately 10 g tissue is homogenized with 40 g anhydrous sodium
sulfate in a Waring blender. The mixture is transferred to a Soxhlet
extraction thimble and extracted with a 1:1 mixture of hexane and methylene
chloride for 3-4 hours. The extract volume is reduced to approximately 50
ml and washed into a tared beaker, being careful not to transfer any
particles of sodium sulfate which may be present in the extract. The
solvent is removed in an air stream and the sample is heated to 100° C for
15 minutes before weighing the sample.
The lipid content is calculated as follows:
% lipid = total residue - tare weight x 100
tissue weight
U.S. Environmental Protection Agency, Environmental Research
Laboratory-Duluth, Duluth, MN 55804.
D. HIOCONCEN'TRATIOK FACTOR (BCF) TEST:
American Society for Testing and Materials. Proposed standard practice for
conducting bioconcentration tests with fishes and saltwater bivalve
molluscs. J. L. Hamelink and J. G. Eaton (Task Group Co-chairmen).
American Society for Testing and Materials, Philadelphia, Penn.
(latest draft.)
Veith, G. D., D. L. DeFoe, and B. V. Bergstedt. 1979. Measuring and
estimating the bioconcentration factor of chemicals in fish. J. Fish.
Res. Board Can. 36: 1040-1048.
32
-------
PLANT TESTS:
American Public Health Association, American Water Works Association, and
Water Pollution Control Federation. 1980. Standard methods for the
examination of water and wastewater. 15th ed. American Public Health
Association, Washington, D.C. 1134 p.
Lockhart, W. L. and A. P. Blouw. 1979. Phytotoxicity tests using the
duckweed Lemna minor, pp. 112-118, IN: Toxicity tests for freshwater
organisms. E. Scherer (ed.), Can. Spec. Publ. Fish. Aquat- Sci. 44.
(Canadian Government Publishing Centre, Supply and Services Canada,
Hull, Quebec, Canada K1A 059.)
Joubert, G. 1980. A bioassay application for quantitative toxicity
¦neasureients, using the green algae Selenastrum capricornutum. Water
Res. 14: 1759-1763.
Miller, W. E., J. C. Greene, and T. Shiroyana. 1978. The Selenastrum
capricornutum Printz algal assay bottle test - Experimental design,
application, and data interpretation protocol. EPA-600/9-78-018,
Environmental Research Laboratory-Corvallis, Corvallis, Oreg. 125 p.
Steele, R. L., and G. B. Thursby. A toxicity test using life stages of
Champia parvula [Rhodophyta]. Presented at the Sixth Symposium on
Aquatic Toxicology. Sponsored by the American Society for Testing and
Materials Committee E-47 on Biological Effects and Environmetal Fate.
13-14 October 1981. American Society for Testing and Materials,
Philadelphia, Penn.
U.S. Environmental Protection Agency. 1974. Marine algal assay procedure;
bottle test. Eutrophication and Lake Restoration Branch, National
Environmental Research Center, Corvallis, OR. 43 p.
-------
APPENDIX 3
The following problems are addressed and examples are given:
(1) how to determine if two LC50 values are statistically significantly
different, and
(2) how to determine if the difference between two pairs of LC50 values is
statistically significant.
To determine if two LC50 values are statistically different (at p .05):
(a) Obtain the 95% confidence Units for both LC50 values.
(b) If the confidence intervals do not overlap the two values are different.
(c) If one confidence interval encompasses the other the values are not
dif ferent.
(d) If the confidence intervals partly overlap the values may be different.
To ascertain if they are different further statistical analysis must be
do ne.
If the above procedure does not indicate whether or not the LC50 values
are statistically significantly different, examine the confidence interval of
either the ratio or the difference of the two values. If the confidence
interval of the ratio brackets one, the two LC50 values are not statistically
significantly different; if the confidence interval does not bracket one, then
there is a statistical difference. The difference between two LC50 values is
not statistically significant if the confidence interval of the difference
includes zero; if the confidence interval does not cover zero, then the
difference is statistically non-zero.
The following exanple demonstrates how the ratio of the LC50 values can be
compared when the estimated LC50 values are obtained by the Trimmed
34
-------
Spearman-Xarber Method. (See Hamilton et al. 1977 for a discussion of the
Trimmed Spearaan-Karber Method, including calculation of the variance.) The
example presents a difference between laboratory and site LC50 values that is
statistically significant.
Table la gives the estimated LC50 values with 95% confidence intervals for
both the lab and site measurements. The LC50 values are obtained by using the
Trimmed Spearman-Karber Method on the natural logarithm of the concentrations-
To determine if there is a statistically significant difference, it is
essential to work, with the metric in which the analysis was performed. In the
example the metric is the natural logarithm of the concentration. The LC50
values in Table la were obtained from the results in Table lb, which gives
loge LC50 values and variances.
The calculations for the ratio and its 95% confidence interval are given
in Table lc. Since the confidence interval does not cover one, the laboratory
and site LC50 values are statistically significantly different.
To compare two pairs of LC50 values several different procedures are
possible. The procedure that follows shows one way to compare the ratios of
the "LC50 values. Specifically, the variable that is examined is the difference
of the ratio of LC50 values:
l°Se ^C^Qslte i ~ loge ^^site 2
loge LC50lab ± loge LC50lab 2.
(As stated before, it is necessary to work in the metric in which the analysis
was performed. Since the Trimmed Spearman-Karber estimate is usually obtained
from an analysis of the logarithm of the dose, the ratio above should be of the
logarithms of the LC50 values.)
The following four steps may indicate whether or not the difference is
significant (at p _< .05) without calculating the confidence interval of the
d i fference:
35
-------
(1) Obtain the 95% confidence limits for both LC50 values.
(2) If the confidence intervals do not overlap the two values are different.
(3) If one confidence interval encompasses the other the values are not
different.
(4) If the confidence intervals partly overlap the values may be different.
To ascertain if they are different further statistical analysis must be
done.
If the above four steps do not indicate whether or not the difference of
the ratios is statistically significant, the confidence interval of the
difference should be examined. If the confidence interval of the difference
brackets zero, the difference is not statistically significant; if the
confidence Interval does not cover zero, the difference is statistically
signi fleant.
An example is given in Tables 2a-2c. Table 2a gives the estimated LC50
values with 95% confidence intervals for two sets of site and lab measurements.
These results were obtained from Table 2b which gives the results in natural
log units based on the Trimmed Spearman-Karber Method of estimation.
Table 2c demonstrates how to determine if the difference is statistically
significant. In this example, the difference is not significant. Note that
this result means that there is no evidence that there is a difference; it does
not mean that two ratios are necessarily identical.
References:
Hamilton, M. A., R. C. Russo, and R. V. Thurston. 1977. "Trimmed
Spearnan-Karber Method for Estimating Median Lethal Concentrations in
Toxicity Bioassays". Environ. Sci. Technol. 11(7): 714-719. Correction
12(4): 417 (1978).
36
-------
Ku, H. H. 1966. "Notes on the Use of Propagation of Error Formulas". J. of
Research of the National Bureau of Standards - C. Engineering and
Instrument 70C: 331-263—341-2 73.
Tables la-c Analysis of Lab and Site LC50 Values
Table la LC50 Values
Source Estimated LC50 95% Confidence Interval
Lab
Site
75
130
(55,104)
(100,169)
Table lb Loge LC50 Value
Source Logp, LC50
Lab 4.32
Site 4.87
Variance
.0256
.0169
Table lc Calculation of Ratio of Site to Laboratory LC50 Values* and 95%
Confidence Intervals
(i) Ratio = loge LC50 site/loge LC50 lab - 4.87/4.32 = 1.13
(ii) Variance of ratio »
/lQSe LC50slteV
\l°Se '-c5°lab /
- AiSz)2 (*
[i.32f ^(4
^ /variance loge LCSOq-jt-p + variance logP LC50
(loge LC50lab)2
[ (loge LC50site)2
lahj
0169 + .0256
87)2 (4.32)2
= .0026
37
-------
(iii) Confidence limit » 2 x (variance of difference)
= 2 x (.0026)1/2 = .10
(iv) Confidence interval ¦ ratio + confidence limit
= 1.13 + .10 = (1.03, 1.23)
(v) Since the confidence interval does not bracket one, the ratio of
site to laboratory LC50 values is statistically significant at
a<.05.
* \ote that in this example the ratios are of loge LC50 values since the
Trimmed Spearraan-Karber Method of estimating LC50 values was used. This
method estimates the LC50 based on the logarithm of the concentration, so
the logarithm of the LC50 should be used here.
Tables 2a-c Analysis of the Lab and Site LC50 Values for Two Species
Table 2a LC50 Values
Source Estimated LC50
Species 1
Lab
Si te
Species 2 Lab
Si te
75
130
60
90
95% Confidence Interval
(55,104)
(100,169)
(48, 75)
(67,122)
Table 2b Loge LC50 Values
Source
Species 1 Lab
Site
Species 2 Lab
Site
Log0 LC50
4.32
4.87
4.10
4.50
Variance
.0256
.0169
.0121
.0225
38
-------
Table 2c Calculation of Difference of Ratios Between Field and Site LC50
Values* and 95% Confidence Intervals
(i) Difference =
loge l*C50siŁe \ - loge LC50gjte 2
loge LC50iab i loge LC50iab 2
4.87 4.50
4.32 4.10
1.13 - 1.10 - .03
(ii) Variance of difference -
varianceAoge LC5°sirp A + variance A°Se *^50 ^ 2
Vl08e LC5Qlab 1 j \l0®e L"°lab 2
( u • A°Se ^C50 \
(where variance / e eiroi is found as
in Table lc (ii))-
- .0026 + .0022 = .0049
(iii) Confidence limit » 2 x (variance of difference)^^
= 2 x (.0049)1/2 = .14
(iv) Confidence interval = difference + confidence limit
= .03 + .14 (-.11, .17)
(v) Since the confidence interval does bracket zero, there is not enough
evidence to reject the hypothesis that the ratios are different.
* Note that in this example the ratios are of loge LC50 values since the
Trimmed Spearman-Karber Method of estimating LC50 values was used. This
method estimates the LC50 based on the logarithm of the concentration, so
the logarithm of the LC50 should be used here.
39
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1 REPORT NO 2
EPA-600/3-84-099
3. RECIPIENT'S ACCESSION NO.
PB* 5 12110 1
4. TITLE AND SUBTITLE
Guidelines for Deriving Numerical Aquatic Site-Specific
Water Quality Criteria by Modifying National Criteria
5R6Pc°tRcT3|?Ti984
6. PERFORMING ORGANIZATION CODE
?. AUTHOR(S)
Anthony R. Carlson, William A. Brungs, Gary A. Chapman,
and David J. Hansen
8. PERFORMING ORGANIZATION REPORT NO.
9 PERFORMING ORGANIZATION NAME AND ADDRESS
U.S. Environmental Projection Agency
Environmental Research Laboratory-Duluth
6201 Congaon 3oulevard
Duluth, Minnesota 55804
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
1?. SPONSORING AGENCY NAME AND ADDRESS
Office of Research and Development
Environmental Research Laboratory
U.S. Environmental Protection Agency
Duluth, Minnesota 55804
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA-600/03
15. supplementary notes
16 ABSTRACT
A major goal of the U.S. Environmental Protection Agency is to directly link
regulatory decisior.-making regarding priority water bodies to the capacity of those
water bodies to receive wastewater discharges and still maintain acceptable water
quality. To assist states in achieving this goal in a consistent* cost-effective
manner, the Office of Research and Development (ORD) has developed a new approach to
water quality criteria derivation with the report "Guidelines for Deriving Numerical
Aquatic Site-Specific Water Quality Criteria by Modifying National Criteria."
These guidelines provide a series of protocols for modifying national water quality
criteria to reflect local environmental conditions. The national criteria, because
they are to protect the biological integrity of all water bodies, serve as benchmarks
and may require adjustments for site-specific applications- The new protocols take
into account site-specific variations in species composition, physical factors, and
chemical water quality variables. Consideration of local conditions assures that
criteria for a given water body are tailored specifically to its aquatic life and
uses.
17. KEY WORDS AND DOCUMENT ANALYSIS
a DESCRIPTORS
b. IDENTIFIERS/OPEN ENDEDTERMS
c. COSATi Field/Group
IB DISTRIBUTION STATEMENT
Release to public
19 SECURITY CLASS (This Report/
unclassified
21. NO. OF PAGES
42
20 SECURITY CLASS (This page)
unclassified
22. PRICE
Forti 2220 — 1 (R«v. 4_77) previous edit
------- |