Hazard Evaluation Division
Standard Evaluation Procedure
Ecological Risk Assessment
EPA/540/09-86/167
-------
June 1986
r, ;.v_ .
HAZARD EVALUATION DIVISION
STANDARD EVALUATION PROCEDURE
ECOLOGICAL RISK ASSESSMENT
Prepared by
Douglas J. Urban, M.F.S
Norman J. Cook, B.S.
Standard Evaluation Procedures Project Manager
Stephen L. Johnson
Hazard Evaluation Division
Office of Pesticide Programs
United States Environmental Protection Agency
Office of Pesticide Programs
Washington, D.C. 20460
-------
STANDARD EVALUATION PROCEDURE
PREAMBLE
This Standard Evaluation Procedure (SEP) is one of a set of
guidance documents which explain the procedures used to evaluate
environmental and human health effects data submitted to the
Office of Pesticide Programs. The SEPs are designed to ensure
comprehensive and consistent treatment of major scientific topics
in these reviews and to provide interpretive policy guidance
where appropriate. The Standard Evaluation Procedures will be
used in conjunction with the appropriate Pesticide Assessment
Guidelines and other Agency Guidelines. While the documents were
developed to explain specifically the principles of scientific
evaluation within the Office of Pesticide Programs, they may also
be used by other offices in the Agency in the evaluation of
studies and scientific data. The Standard Evaluation Procedures
•will also serve as valuable internal reference documents and will
inform the public and regulated community of important consider-
ations in the evaluation of test data for determining chemical
hazards. I believe the SEPs will improve both the quality of
science within EPA and, in conjunction with the Pesticide Assess-
ment Guidelines, will lead to more effective use of both public
and private resources.
John W. Melone, Director
Hazard Evaluation Division
01
-------
ACKNOWLEDGEMENTS
This document contains important suggestions and contribu-
tions from many past and current professionals in the Ecological
Effects Branch and the "old" Criteria and Evaluation Division
within the Agency's Office of Pesticide Programs. We have, in
many instances, merely compiled, collated, and placed in written
form a large body of knowledge collected over time. We are greatly
indebted to these professionals. We are particularly grateful to
the following individuals for contribution to specific sections:
Richard Lee - Risk Assessment Methods and Aquatic Residues; Dennis
McLane - An Example of Summarizing and Interpreting Exposure Model
Data, and Integrating Exposure and Hazard Data for Risk Assessment;
Ray Matheny - Evaluating Risk to Endangered and Threatened Species
from Pesticide Registration Actions; Charles Bowen III - Tables 8
and 11; Robert Hoist - Spray Drift Model; Richard Stevens - contri-
butions to Mammalian Species, Acute Risks, particularly for irri-
tation and inhalation data; Allen Vaughan - Risk Assessment, Non-
Target Insects; and Daniel Rieder - contributions concerning con-
sultation procedures with the Office of Endangered Species (OES).
We extend a special thanks to Michael W. Slimak, current
Branch Chief of the Ecological Effects Branch, for his helpful
comments and review of this document. We also thank Elizabeth
Collins for her untiring efforts in typing numerous drafts of the
document.
02
-------
PREFACE
This paper is a document that presents state-of-the-art tech-
niques for ecological risk assessment. It presents procedures,
philosophies, and extrapolative techniques as used in the Ecologi-
cal Effects Branch (EEB), in the Hazard Evaluation Division (BED),
in the Office of Pesticide Programs '(OPP), in the U.S. Environ-
mental Protection Agency (U.S. EPA, EPA, or the Agency). It con-
cerns risk assessment procedures pertinent to pesticide uses pro-
posed for registration under the Federal Insecticide, Fungicide,
and Rodenticide Act (the "Act" or "FIFRA"), as amended, 7 U.S.C.
136 et seq. The document also pertains to discussions relative
to the following areas:
0 The Pesticide Assessment Guidelines. Many areas are
mentioned, but three, in particular, are extensively
discussed. These are:
- Subdivision E - Hazard Evaluation: Wildlife and
Aquatic Organisms;
-'Subdivision L - Hazard Evaluation: Nontarget
Insects; and
- Subdivision N - Environmental Fate;
0 40 FR (129), Registration, Reregistration and Classification
Procedures; Thursday, July 3, 1975;
0 40 CFR (158), Data Requirements for Pesticide Registration;
Final Rule; Wednesday, October 24, 1984;
0 40 CFR (154), Special Review of Pesticides; Criteria
and Procedures; Final Rule; Wednesday, November 27, 1985;
0 Proposed draft revisions of restricted use classification
criteria; and
0 Documents, papers, memos, SEPs, or techniques developed
by Agency personnel.
The primary purposes of this document are to provide guidance
and a historical perspective to EEB employees. However, this paper
may prove useful to others (e.g., pesticide applicants or users,
other federal or state agencies, academia, or the general public).
We in EEB point out that being a state-of-the-art document, this
work is obviously not all-inclusive. We did not intend it to be,
but we do plan to update the document as the state-of-the-art
advances.
03
-------
TABLE OF CONTENTS
Page
I. --TRODUCTION 1
II. DEFINITION OF TERMS 1
III. RISK ASSESSMENT METHODS 2
IV. DATA REQUIREMENTS 6
V. APPROACH TO ASSESSING ECOLOGICAL RISK 8
A. Aquatic Risk Assessment 13
1. . Introduction 13
2'. Indicator Species 13
3. Toxicological Hazard Data , 15
4. Aquatic Residues 16
5. Aquatic Non-Target Exposure 23
6. Risk Assessment 25
a. Fish Acute Risk 26
b. Aquatic Invertebrate Acute Risk ... 26
c. Fish and Aquatic Invertebrate
Chronic Risk 27
7. Endangered Species 27
a. Endangered Species Risk Criteria .. 28
b. OES Formal Consultation 28
B. Terrestrial Risk Assessment 29
1. Introduction 29
2. Indicator Species/Test Species 30
3. Terrestrial Residues 30
4. Risk Assessment 32
a. Mammalian Species 33
(1) Acute Risks 33
(2) Subchronic/Chronic Risks .... 39
(3) Secondary Hazards 39
b. Avian Species 42
(1) Acute Risks 42
(2) Subchronic/Chronic Risks .... 49
(3) Secondary Hazards 50
5. Risk Criteria 51
6. Risk Assessment - Non-Target Insects ... 52
;
VI. RISK ASSESSMENT - SHORTCOMINGS AND IMPROVEMENTS .. 53
04
-------
TABLE OF CONTENTS (Continued)
Page
VII. ATTACHMENTS
ATTACHMENT A: Final Criteria for Initiation of
Special Review 56
ATTACHMENT B: Proposed Restricted Use Criteria
for Hazard to Non-Target
Organisms 57
ATTACHMENT C: Model Name: Simulator for Water
Resources in Rural Basins 58
ATTACHMENT D: Model Name: Exposure Analysis
Modeling System 60
ATTACHMENT E: Output from Spray Drift Model 61
ATTACHMENT F: An Example of a More Sophisticated
EEC Using State-of-the-Art
Models 63
ATTACHMENT G: An Example of Summarizing and
Interpreting Exposure Model Data
and Integrating Exposure and Hazard
Data for Risk Assessment 73
ATTACHMENT H: Ecological Effects Branch/HED
Evaluating Risk to Endangered/
Threatened Species From Pesticide
Registration Actions 76
ATTACHMENT I: Pesticides and Crops Considered by
Hoerger and Kenaga (1972) 80
ATTACHMENT J: Techniques for Estimating Pesticide
Residues on Vegetation Immediately
Following Application 84
ATTACHMENT K: Human Risk Approach Using Mammalian
Data 89
CITATIONS/REFERENCES 91
05
-------
ECOLOGICAL RISK ASSESSMENT
I. INTRODUCTION
Since 1970, the regulatory authority and basis for pesticide
risk assessment has rested with the Environmental Protection
Agency (EPA) under the Federal Insecticide, Fungicide, and
Rodenticide Act (FIFRA). Under this Act, EPA must determine
whether a pesticide can be registered for a particular use.
FIFRA, as amended, states that the Administrator shall register a
pesticide if he determines that, "when used in accordance with
widespread and commonly recognized practice it will not generally
cause unreasonable adverse effects on the environment" (P.L.95396,
Sec. 3 (c)(5)(D)). It also states that, "the Administrator may
conditionally amend the registration of such pesticide ... if the
Administrator determines that ... amending the registration ...
would not significantly increase the risk of any unreasonable
adverse effects on the environment" (ibid. Sec. 3(c)(7)(B)).
The term "unreasonable adverse effects on the environment"
means any unreasonable risk to man or the environment, taking
into account the economic, social, and environmental costs and
benefits of the use of any pesticide (ibid. Sec. 2(bb)). Under
FIFRA the process of determining whether or not a risk is unreason-
able (i.e., factoring in benefits along with risks) is a risk
management function. For this discussion, the important term
used in FIFRA is "risk to the environment." In order for the
Administrator to determine if there will be an unreasonable risk
to the environment from the use of a pesticide, an environmental
risk assessment or more specifically, an ecological risk assess-
ment, is required. The term 'ecological' better describes the
broad scope of adverse effects that are of concern from the use
of pesticides. In the past, the adverse effects of greatest
concern have been mortality to single species of non-human,
non-target organisms, including endangered or threatened species.
The Ecological Effects Branch (EEB) in the Hazard Evaluation
Division (BED) in the Office of Pesticide Programs (OPP) has
primary responsibility for developing environmental risk assess-
ments for pesticides. It should be noted that the state-of-the-
art in ecological risk assessment is rapidly evolving and EEB is
striving to broaden our risk assessment concerns to include not
only single species and populations of wildlife, aquatic organisms,
plants, and beneficial insects, but also community and ecosystem
level concerns, as well.
II. DEFINITION OF TERMS
; Risk assessment has been described as estimating the proba-
bility or likelihood of undesirable events such as injury, death,
or decrease in the mass or productivity of game fish, wildlife,
etc. (Suter II, et al., 1983; Rodericks and Tardiff, 1982). We
06
-------
-2-
would like to define ecological risk assessment from the use of
pesticides as estimating the likelihood or probability that ad-
verse effects (e.g., mortality to single species of organisms, or
reductions in populations of non-target organisms due to acute,
chronic, and reproductive effects, or disruption in community and
ecosystem level functions) will occur, are occurring, or have
occurred. This definition encompass-es the final criteria for
initiation of special review (see 40 CFR Part 154: 49005; 49007;
49016 - S 154.7(a)(3), (4), (5), and (6); Attachment A). Community
and ecosystem level functions, while not specifically covered in
the proposed regulations, would be covered under Section 154.7
(a) (6) above.
As stated by Johnson (1982), risk is a function of hazard and
exposure. Similarly, ecological risk is a function of toxicological
hazard and environmental exposure. Toxicological hazard is the
intrinsic quality of a pesticide to cause an adverse effect under
a particular set of circumstances. Toxicological hazard data
would include, for example, laboratory fish, aquatic invertebrate,
or bird LCso values, and effect levels for fish and avian repro-
duction tests. Environmental exposure is a function of two data
components. The first is the estimated amount of the pesticide
residue that will be in the environment and available to non-target
organisms. We call this the estimated environmental concentration
or EEC. The second consists of the numbers, types, distribution,
abundance, dynamics, and natural history of non-target organisms
which will be in contact with these residues. Information on the
proposed label use of the pesticide is essential for such exposure
estimates. We first estimate toxicological hazard and environmental
exposure separately, and then compare them.
III. RISK ASSESSMENT METHODS
Barnthouse, et al., (1982 a, b) describe five methods for
"environmental risk analysis" (i.e., quotient method, analysis
of extrapolation error, fault tree analysis, analytic hierarchy
method, and ecosystem uncertainty analysis). Their "quotient
method" is most similar to EEB's current risk assessment method.
However, there are some subtle differences. In their quotient
method, ah EEC is directly compared to an effect level such as
an LCso value (e.g., 10 ppm/10 ppm). The resultant quotient,
1, can then be compared to some relative quotient ranking to
indicate possible adverse effects to non-target organisms. For
example, where quotient = 0:
0 £ 0.1 = No Adverse Effects
0.1 £ 0 £ 10 = Possible Adverse Effects
0 > 10 = Probable Adverse Effects
07
-------
-3-
We, on the other hand, compare an EEC and an effect level
(e.g., an LCso) based on regulatory risk criteria. The regula- -,
tory risk criteria, as specified in the 1975 Regulations for the"
Enforcement of the FIFRA (40 FR (129): 28260-28265; 28281-28284)
and in Special Review of Pesticides; Criteria and Procedures;
Final Rule (40 CFR Part 154: 49005; 49007; 49016 § 154.7(a)(3),
(4),(5), and (6)) are summarized and presented in Table 1. _£/ I
Many of these risk criteria contain specific safety factorscT
that were derived from a toxicological model presented in the
1975 regulations. The model was designed to provide a safety
factor that would allow for differential variability and sensi-
tivity among fish and wildlife species. It was assumed that
the slope of the dose/response curve for the effects of a pesti-
cide on most fish and wildlife species would be unknown. (It is
impossible to test every non-target species that might be exposed
to a pesticide. ) Therefore, as the 1975 regulations state:
From a cross section of existing dose-response data it
has been estimated that a typical slope is 4.5 probits
per log .cycle, and a minimum slope is about 2 probits
per log' cycle. The latter situation corresponds to a
very variable test population with some individuals
displaying high sensitivity to the toxicant. From
this model it can be estimated that a dose or exposure
10 ti me_s_JLower than the LDso or LCso/woxi'1-d be expected
to lead to a mortality rate of aboutk.0.01^ perCBTrt -
under ^ydt>iSJil_ si ope conditions, but to a mortality
ralre~~ot^4 percent under minimum slope conditions./ A
dose-resp6nse j&-tiroes Icagejc than the LDso or LCso
wpul.d be expected Jto lead to mortality rates of about
~~- 0.1 ''percent andr^lO percent respectively. These figures
were used as the" basis for selecting a safety factor
of 5-10 for setting the classification criteria for
protecting wildlife. These factors would be expected
to provide an ample margin of safety for a typical
species, but only marginal protection to the most
variable species. Even larger safety factors than 10
would be desirable to ensure protection of species in
which even a single death is of special concern, for
instance the death of an endangered species (40 FR (129):
28261). (Also see Attachment H concerning endangered
species.)
!_/ Special Review is a formalized Agency process for determining
whether currently registered pesticides present unreasonable
adverse effects to humans or the environment.
08
-------
-4-
«0
iH
U
fl
1-1
CJ
B
10
iH
I
3
(0
, U
3
2 y 22
A|
o
in
a
A|
O
in
a
V *
w "s
00
in in
sl
C7I
^1
in o
B tJ
m
09
A|
&
Ed
W
Q
JJ
M JO
jj IQ
U-l Qi
H C
-4
o
CD 5 C
> *2 O
<£
>i ffl
~~ . 10 -t} <0 -H
A rH O U4
CD L| Q ~r\
CJ > £X 05 *O
CO CD CD iH 0
ta j K < g
I
45
5
etj
3 CO
0) i-l
«-i O
U4 C _ .
u M « ea
eg.
£
CJ
tu u-i
ca u
o
o
CM
CO 4J
r^'S^
• CD O i—I
, LI LI jj (3
Q) D, C C
4^ _ & Q
•^^ • * wu
O ,'Din^^
S a> •-! a;
"D to Cu
w o to u
** ^-
•• •• •• o CD
80) Q) T* Q 0} +. lj
po M _ b;^ &
4J c a> co
CO
-------
-5-
The following equations describe this model:
(1) log LCk = log LC5o + (probit k - 5)/b
where k = the new percentage mortality
b = the slope
5 = the probit of 50%
(Hill, et al., 1975) '
and the antilog of log LC^ = the estimate of the
dosage of the new percentage mortality,
(2) LCso/LCfc = Safety Factor
For example, the following calculations resulted in the
acute safety factors for wild mammals and birds.
Example (A): if b = 4.5; LCso = 100 ppm; k = 0.1;
probit k - 1.91 ;
then, log LCo.i » 2 + (1.91 -5)/4.5
log LCo.l = 1.31; Antilog (1.31) = 20.4 =
and, LCso/LCo.i = Safety Factor
where 100/20.4 = 4.9 or approximately 5.
Therefore, pesticides which result in residues exceeding l/5th
an LDso or LCso value for non-target organisms with "typical
slopes" will be candidates for restricted use (i.e., use
restricted to certified applicators). It is not stated in the
regulations, but an acceptable interpretation of the situation
is this: based on that cross section of data and accepting the
assumptions of the model, 0.1% (or one in 1000) of the typical
population exposed to the pesticides are likely to die when
the safety factor of 5 is used. These criteria also specify
that the residues are to be determined at the time of maximum
residues (i.e., immediately after application). This adds an
additional (albeit unknown) safety factor to the criteria since
these residues may degrade over time.
It was felt that a slightly higher safety factor was
needed for many aquatic organisms because they cannot easily
limit their exposure to pesticides by moving out of treated
areas or by switching to alternate food items as can birds
and mammals. Therefore, an additional safety factor of 2 was
applied. Consequently, pesticides which result in residues
exceeding l/10th an LCso value for non-target aquatic organisms
with "typical slopes" will be candidates for restricted use.
10
-------
-6-
We can use the previous model to determine the probability of
50 percent mortality in the typical aquatic non-target organism
population.
Example (B): if b = 4.5; LCso = 100 ppm; LC^ = 10;
then, LCso/LCfc = Safety Factor, where 100/10 = 10
log LCfc = 1;
and, log LCj< = log LCso + (probit k - 5)/4.5
1=2+ (probit k - 5)/4.5
probit k = 0.5
therefore, k = 0.00000339767, the new percentage mortality,
As above, we can state that pesticides which result in residues
exceeding l/10th an LCso value for non-target aquatic organisms
with "typical slopes" will be candidates for restricted use.
Again, an acceptable interpretation of this situation is this:
based on that cross section of data and accepting the assumptions
of the model, approximately 0.000034% (or one in 30,000,000) of
the typical population exposed to the pesticide are likely to
die when the safety factor of 10 is used. This obviously
provides a large margin of safety for aquatic organisms with
"typical slopes."
We realize that many theoretical questions can be
raised about the use of risk criteria and safety factors in
general. Currently, we do not use the model to predict the
probability of a pesticide to cause significant acute adverse
effects to non-target organisms. This simple model for
ecological risk assessment does not provide a mechanism for
estimating model uncertainty or the probability of adverse
effects. We have come to view the risk criteria with their
safety factors as "rough" estimates of potential risk to
non-target organisms. We should note that an attempt is
currently being made to re-visit the model based on up-to-
date data bases in both EEB and the Office of Research and
Development (ORD).
IV. DATA REQUIREMENTS
Specific information and testing data are necessary in
order to conduct an ecological risk assessment. FIFRA states
that "The Administrator shall publish guidelines specifying the
kinds of information which will be required to support the
registration of a pesticide ... " (P.L. Sec. 3(c)(2)(A)).
11
-------
-7-
Under FIFRA the Agency is not responsible for producing the
data needed to make an ecological risk assessment. That burden
is placed upon the applicants for registration. OPP has
published regulations which specify the data that are required
for registration (40 CFR Part 158), and guidelines which provide
recommended testing methods that are needed to produce the
required data (Pesticide Assessment Guidelines - Subdivision E).
Further, we are in the process of developing Standard
Evaluation Procedures (SEPs) for each kind of data that is
required for an ecological risk assessment. These SEPs explain
the procedures used to evaluate ecological effects data submitted
to OPP, and ensure comprehensive and consistent treatment of
the science in reviews as well as providing interpretive policy
guidance. SEPs already developed and published through the
National Technical Information Service (NTIS) include:
0 Avian Single-Dose Oral
0 Avian Dietary
0 Acute Toxicity Test for Freshwater Fish;
0 Acute Toxicity Test for Freshwater Invertebrates;
0 Wild Mammal Toxicity Test;
0 Acute Toxicity Test for Estuarine and Marine Organisms
(Estuarine Fish 96-Hour Acute Toxicity Test);
0 Acute Toxicity Test for Estuarine and Marine Organisms
(Shrimp 96-Hour Acute Toxicity Test);
0 Acute Toxicity Test for Estuarine and Marine Organisms
(Mollusc 96-Hour Flow-Through Shell Deposition Study);
0 Acute Toxicity Test for Estuarine and Marine Organisms
(Mollusc 48-Hour Embryo-Larvae Study);
0 Honey Bee - Acute Contact LD$Q; and
0 Honey Bee - Toxicity of Residues on Foliage.
Additional SEPs for reviewing chronic and field testing
data will be completed and include:
0 Avian Reproduction;
0 Fish Early-Life Stage;
Aquatic Invertebrate Life-Cycle;
12
-------
-8-
0 Fish Life-Cycle; and
0 Field Testing for Pollinators.
V. APPROACH TO ASSESSING ECOLOGICAL RISK
Under their definition of risk assessment, Rodericks and
Tardiff (1982) present four procedural steps: (1) the review and
evaluation of hazard data to identify the nature of the hazards;
(2) identifying and evaluating the observed quantitative relation-
ship between dose and response (this frequently requires the impo-
sition of assumptions regarding the quantitative relationship
between the test organisms and the nontarget organisms that will
be protected); (3) the identification of the conditions of expo-
sure (e.g., intensity, frequency, and duration of exposure); and
(4) combining the information on dose-response with that on expo-
sure to derive estimates of the probability that the hazards asso-
ciated with the use of the chemical will be realized under the
conditions of exposure that will be experienced by the non-target
population (s.)- under consideration. The authors state further
that: "Risk assessment involves integration of the information
and analysis associated with these four steps to provide a complete
characterization of the nature and magnitude of risk, and the
degree of confidence associated with this characterization."
In EEB, we generally follow these four steps. As outlined
in Figure 1, toxicological hazard data and exposure data are com-
pared using the regulatory risk criteria. Typically, the toxico-
logical hazard data may consist of acute LD$Q and LCso values, or
chronic no-effect-levels for the most sensitive indicator species.
Exposure data normally consist of model-based estimated environ-
mental concentrations in important media of concern (i.e., water,
soil, non-target organism food items). As the ratio of these in-
put data equals or exceeds the restricted use criteria, a risk is
inferred. If the ratio approaches or exceeds the special review
criteria, then a high risk is inferred. For non-endangered non-
target organisms, this generally means that the ratio approaches
or exceeds one. The criteria for endangered species are more
rigorous. They are presented in Attachment H and consist of
safety factors applied to acute toxicity values such as the
or
We recognize that the ratio method for assessing risk has
numerous weaknesses. For example: (1) it does not adequately
account for effects of incremental dosages; (2) it does not com-
pensate for differences between laboratory tests and field popula
tions; (3) it cannot be used for estimating indirect effects of
toxicants (e.g., food chain interactions); (4) it has an unknown
reliability; (5) it does not quantify uncertainties; and (6) it
does not adequately account for other ecosystem effects (e.g. ,
predator-prey relationships, community metabolism, structural
shifts, etc.). At the present time, therefore, the state-of-the-
13
-------
-9-
FIGURE 1
Flow Chart for Ecological Risk Assessment for Pesticides
INPUTS
TOXICOLOGICAL
HAZARD DATA
(Laboratory)
- Ecotoxicological data
(e.g., LDsos; LCsos;
ECsos; NELs; MATC)
- Human Toxicology data
(e.g.. Rat, Mouse,
Rabbit, LDsos;
, NELs)
(Field)
- Ecotoxicological
data (e.g.,
mortality; sublethal
effects; population
effects)
EXPOSURE DATA
(Laboratory)
Chemical fate and
transport data (e.g.,
1/2-life data; K
-------
-10-
art cannot provide a complete characterization of the magnitude
of risk nor the degree of confidence associated with the charac-
terization.
To further illustrate the approach shown in Figure I consider
the following example: a hypothetical organophosphate insecticide
is applied to cotton at 1 pound active ingredient up to 3 times per
growing season, as a foliar spray when insects appear. Cotton is
a major crop in the U.S. with approximately 8,000,000 acres planted
in 1983 (U.S. Department of Agriculture, 1984). Gusey and Maturgo
(1972) list the following species or groups as utilizing cotton
fields for feeding from June through October: quail, pheasants,
doves, prairie chickens, passerines, rabbits, deer, raccoons, and
antelope. Roach (1973) observed orioles, hummingbirds, bobwhites,
mourning doves, towhees, cardinals, and thrashers using more mature,
shadier cotton fields .in a Mississippi study. In June and July,
before most soil was shaded, purple grackles were observed. Other
birds were observed to a lesser extent. Among mammals, rabbits
and deer browsed incidentally on cotton; cotton rats, house mice,
white-footed mice, cotton mice, beavers, armadillos, raccoons, and
foxes were also observed. A limited number of species of snakes,
lizards, turtles, toads, and frogs were also seen. Therefore,
there is significant potential for exposure of the insecticide to
terrestrial wildlife, especially birds. The most likely result
of exposure to birds is via the diet (versus oral ingestion of the
pesticide per se_). The maximum and typical expected residue con-
centrations of the insecticide immediately after application on
vegetation, insects, soil (after Hoerger and Kenaga, 1972; 1 Ib,
ai/A) and water (see Table 2) are:
Residue (ppm)
Vegetation Type Maximum Typical
Short grass 240 125
Long grass 110 92
Leaves and leafy crops 125 35
Forage, e.g., alfalfa;
also estimate for small insects 58 33
Pods containing seeds; Legumes 12 3
Grain 10 3
Fruit (e.g., cherries, peaches) 7 1.5
Soil Residue (ppm)
0.1 acre-inch in depth 22
Water Residue (ppm)
0.5 acre-foot in depth 0.734
15
-------
-11-
0) CO
*J J2
(0 iH
t-t
•o o
0) •
£o
e -i
h
0)
4J
1C
o o
kl
CO U-l
0)
•D O
O
CO CO
c
c o
1-1 1-1
4J
CO 10
CD O
•O -rt
•r^ iH
o a
•H Qj
4J <
CO
0) 4J
CU O
0)
a c
Ql-iH
10
Cd
00
• p
^»
1 i
O
0)
Q
Water
0
CO
m
m
CM
•H
in
0
CM
4J
»W
on
E
.Q
r- •«• CM o r* •* vo
c*> r* o> IH ^* co r» i
iH iH fH CM
o rH CM CM rn ^r vc
in rM ^r r- co
CM in IH r*- o r<4 ro
ro m in
m t*- ro o f r- •-!
fHrHf-tiHiHCMfOm'S'^'invO
CM in vo r- o CM co
iHCMflM^VOCflCMinCO
foinooocM
co
co
oomoooinomoinomooooooooo
OOOOOOOrHfHiHiHCMCMCMPOI'invOI^COCTt
CJ1
E
o
o
CD ,Q
»>iH
CO
os vo
CM CO
fe •
CO CM
CM ID
II II
0)
4->
10
U-l
O
0
0
O
•H
.0
3
O
4J
jC
en
•H
0)
CM
O
V£)
i—I
CM
II
0>
4J
§
U-l
O
O
O
u-i
I
O
Ul
o
16
-------
-12-
Concerning exposure to aquatic organisms, we can assume
that a typical runoff scenario for cotton growing areas would
include a 10-acre basin draining into a 1-acre pond with an
average depth of 6 feet (USDA, 1982). Further, Wauchope (1978)
suggests thai water-soluble pesticides applied as aqueous
solutions usually show runoff losses of 0.5% (or less) to 3
times this amount (1.5%) if a large,' early runoff event occurs.
Therefore, a generalized maximum runoff figure is 1.5%. Using
the general equation where EEC (ppb) = pesticide loading to the
body of water/weight of the water, a maximum estimated concen-
tration due to runoff would be 9 ppb. An additional source of
the insecticide to the pond is spray drift. There is little
information from which to estimate this source of contamination.
Nigg, et al., (1984) found a mean of 140 ppb in a 0.25 acre pond
three hours after spraying a Florida citrus orchard with 10
Ibs ai/A phenthoate. The sole source of contamination was
determined to be drift. Extrapolating for 1.0 Ib ai/A of the
insecticide, the value would .be 14 ppb. When the drift estimates
are added to runoff estimates, the total initial concentration
of the insecticide in the pond is estimated to be 23 ppb. Thus,
preliminary'EECs for the environment would range from 1.5 ppm
to 240 ppm for terrestrial exposure and 23 ppb to 734 ppb for
aquatic exposure.
The laboratory toxicological hazard data for this hypothe-
tical insecticide is as follows: Bobwhite quail (adult) LD$o =
2 mg/kg and bobwhite quail (14 days old) LCso = 3 ppm; mallard
duck (10 days old) LCso = 30 ppm; avian reproduction NEL = 300 ppm;
bluegill sunfish LCso = 100 ppm; rainbow trout LCso = 57 ppm;
Daphnia magna LCso = 10° PP1"* fathead minnow life-cycle MATC =
50 ppm; Daphnia magna life-cycle NEL = 35 ppm.
When comparing the exposure EECs to the toxicological
hazard data based on the criteria in Table 1, it is clear that
the risk to aquatic organisms is low. The highest EEC, 734
ppb, is less than l/10th the lowest acute LCso value (i.e.,
57 ppm/10 =5.7 ppm) and the lowest chronic value (i.e., 35 ppm/
10 = 3.5 ppm). The risk to avian wildlife, however, is high.
The EECs for common avian diet components like insects, seeds,
grain (i.e., 3 ppm to 58 ppm) are greater than or equal to the
lowest avian acute LCso, 3 ppm. Further, exposure to contam-
inated insects alone poses a risk of unacceptable levels.
The nature of the risk is that significant acute avian
effects are possible. The magnitude of the effects, while
unquantified, could be quite large since cotton is a major crop
in the U.S., especially in the Southeast, Texas, and California.
Avian populations could be adversely affected on a local or
regional level, including members of endangered species.
17
-------
-13-
Finally, to further validate this conclusion, field effects data
are usually required. Specifically, these data are terrestrial
field studies that will investigate and quantify acute effects
to avian populations in actual field situations.
At this time, we cannot provide a reliable estimate of the
probability for significant reductions in populations of non-
target organisms. However, based on an increasing body of
historical field effects data, we believe that the risk assessment
scheme does enable us to distinguish between pesticides that
pose a significant high risk and those that pose a low risk.
A. Aquatic Risk Assessment
1. Introduction
EEB's aquatic hazard assessment process has been discussed
from a regulatory viewpoint elsewhere (Akerman and Coppage, 1979).
This report will concentrate primarily on EEB's risk assessment
procedures and the supporting scientific rationale. In the
aquatic risk- assessment, we examine the potential risks of the
proposed pesticide uses to non-target fish and aquatic inverte-
brates, in both the freshwater and estuarine/marine environments.
These groups of organisms were chosen because there are well
defined testing protocols available for fish, crustaceans,
molluscs, and aquatic insects. In addition, certain species
within these groups are important food and recreation resources,
and consequently are of great economic and aesthetic value.
Risks to non-target algae and aquatic plants are not
addressed here because the Agency has determined that phytotox-
icity data will be requested only on a case-by-case basis.
Examples when such data may be requested are: (1) for pesticides
used in forests and natural grasslands, (2) when hazards are
posed to federally endangered or threatened plants, (3) at the
initiation of a Special Review where a phytotoxicity problem
may exist, (4) where a specific phytotoxicity problem arises
and when general open literature data are not available to
address the problem (see 40 CFR 158.150). When such hazards
are addressed, they are considered in a separate review.
2. Indicator Species
EEB requests certain toxicity data prior to the completion
of an aquatic risk assessment. Considering that there are more
than 2000 species of freshwater and saltwater fish in North
America and tens of thousands of species of aquatic invertebrates,
certain indicator species were selected for the toxicity testing
that were most useful for risk assessment. The selection
criteria included the following:
18
-------
-14-
0 the test species should be one which has demonstrated
sensitivity to the known effects produced by toxic chemi-
cals;
0 the dose-response of the test species to a variety of
pesticides;
* the test species should be ecologically significant, occurring
naturally in large numbers and in widespread habitats;
0 the test species should be aesthetically and/or economically
important;
0 the test species should be readily available for test
purposes; and
0 the test species should have a life-cycle short enough to
permit reasonably short (1 year) life-cycle tests.
While no single species meets all these criteria, test
species should meet a majority of these criteria to qualify.
In addition, we are continually seeking better indicator species.
An example of this is the recent development of early life-stage
toxicity test methodology using three Atherinid fishes (silver-
sides). It was generally recognized that an established
estuarine/marine fish that has been widely used as the indicator
species for toxicity testing, the sheepshead minnow (Cyprinodon
variegatus), is relatively insensitive to known toxic chemicals.
Therefore, EPA's Environmental Research Laboratory at Gulf
Breeze, Florida, proceeded to develop partial life-cycle testing
methods using more sensitive and better representative marine/
estuarine test species (i.e., Menidia beryllina, M. roenidia,
and M. peninsulae) (Goodman, et al., 1985 a, b). When these
methods are finalized and tested, Subdivision E of the Pesticide
Assessment Guidelines will be updated.
The indicator species that EEB has selected for acute
toxicity testing are consistent with the recommendations in:
(1) ASTM Standard E 729-80 for freshwater/marine fish and macro-
invertebrates, (2) Committee on Methods for Toxicity Tests with
Aquatic Organisms, 1975, for freshwater/marine fish and macro-
invertebrates, (3) ASTM Standard E 724-80 for static acute tests
with larvae of bivalve molluscs, and (4) Bioassay Procedures for
the Ocean Disposal Permit Program (anonymous, 1978) for oyster
shell growth tests, static acute tests using mysid and grass
shrimp and Acartia tonsa. More specifically, the species most
often preferred for acute testing are:
0 freshwater/coldwater fish - rainbow trout (Salmo gairdneri)
0 freshwater/warmwater fish - bluegill sunfish (Leporois
macrochirus)
19
-------
-15-
0 freshwater crustacean - Daphnia magna
0 estuarine/marine fish - sheepshead minnow (Cyprinodon
variegatus)
"estuarine/marine shrimp - mysid, penaeid or grass
0 estuarine/marine oyster - eastern oyster (Crassostrea
virginica)
The indicator species that EEB has selected for chronic toxi-
city testing are consistent with the recommendations in (1) National
Water Quality Laboratory Committee on Aguatic Bioassays (1971 a,
b) for fathead minnow and brook trout, (2) Bioassay Procedures for
the Ocean Disposal Permit Programs for mysid and grass shrimp and
sheepshead minnow (anonymous, 1978), (3) ASTM Standard E 1022-84
Practice for Conducting Bioconcentration Tests with Fishes and
Salt Water Bivalve Molluscs, (4) Macek, et al., 1975, for biocon-
centration in bluegill, and (5) Branson, et al., 1975, for biocon-
centration in rainbow trout. More specifically, the species most
often prefer.red for chronic testing are:
0 freshwater fish - rainbow or brook trout Salvelinus
fontinalis), and fathead minnow (Pimephales promelas)
0 freshwater crustacean - Daphnia magna
0 estuarine/marine fish - sheepshead minnow
(Cyprinodon variegatus)
0 estuarine/marine shrimp - mysid (Mysidopsis bahia)
0 estuarine/marine mollusc - eastern oyster
(Crassostrea virginica)
3. Toxicological Hazard Data
The following aquatic toxicological hazard data represent
the full complement of aguatic testing that could be reguested
for an aguatic risk assessment:
Tier 1
(1) 96-hour coldwater fish LCso»
(2) 96-hour warmwater fish LCso?
(3) 48-hour (or 96-hour) freshwater aguatic
invertebrate
Tier 2
(4) 96-hour estuarine/marine fish
(5) 96-hour estuarine/marine shrimp
(6) 48-hour oyster embryo-larvae EC59;
20
-------
-16-
(7) 96-hour oyster shell deposition EC$QJ
(8) Fish early life-stage MATC or Effect/Mo Effect
Level;
(9) Aquatic invertebrate life-cycle MATC or Effect/No
Effect Level;
(10) Fish bioaccumulation factor, e.g., 1000X;
(11) Special aquatic organism test data (e.g., fish
acetylcholinesterase levels).
Tier 3
(12) Fish full life-cycle MATC or Effect/No Effect
Level;
Tier 4
(13) Fish/aquatic invertebrate population effects in
the field;
(14) Simulated and actual field effects data on
aquatic organisms.
These laboratory and field data describe the potential of a
pesticide to cause adverse effects (i.e., mortality, reduction in
growth or impairment of reproduction in aquatic non-target organ-
isms and their populations). They are compared to the estimated
or actual measured pesticide residue in the aquatic environment
in order to estimate the ecological risk to populations of
aquatic non-target organisms from the use of the pesticide.
The actual amount of aquatic toxicological hazard data that
would be required is determined by the "when required" testing
criteria found in the footnotes to Part 158.145 of the Data
Requirements for Pesticide Registration; Final Rule (40 CFR 49
(207): 42894-42895; Wednesday, October 24, 1984) and in the
"when required" paragraphs in Subdivision E of the Pesticide
Assessment Guidelines. The criteria are identical in both
documents. Basically, pesticides that are persistent in any
media, are highly toxic to non-target organisms, require repeat
applications, are applied directly to water, or are likely to
transport to water from the intended use sites, will require
substantial additional data. Also, the fewer criteria that are
met, the fewer data that are generally required.
4. Aquatic Residues
Measured pesticide concentrations in water are usually not
collected by the registrant at this stage of the registration
process in support of a registration action. If such data are
available, they are included in the exposure component of EEB's
ecological risk assessment. Normally, however, EEB must estimate
aquatic exposure to pesticide residues.
21
-------
-17-
Figure 2 AQUATIC ESTIMATED ENVIRONMENTAL CONCENTRATION (EEC) TIER SYSTEM
Level 1
Level 2
Level 3
Level 4
GO TO EEC TABLE FDR DIRECT APPLICATION TO WATER (See Table 2)
DIRECT APPLICATION
MODEL (Lent1C)
EEC < Risk Criteria | No Risk
EEC > Risk Criteria
Modify Variables in Basic Equation (1) From Level 1
1. Chemical Site Dependent Variables;
- drainage basin size (A)
- surface area (A)
- average depth (ft)
- % runoff (decimal)
2. Hod? •
Use Literature References:
- Benchmark Chemicals (e.g., Wauchope, 1978: 462-468;
McElroy, et al., 1976)
-Typical Runoff figures (e.g., Wauchope, 1978: 460-464;
Spencer, et al., 1985)
- Site Characteristics (e.g., U.S. Dept. of Agriculture, 1982)
- Spray Drift (e.g., Nigg, et al., 1984)
SIMPLE
RUNOFF
MODEL
(Lentic)
EEC < Risk Criteria
No Risk
EEC > Risk Criteria
Request EEC Estimation from Exposure Assessment Branch (EAB)
Computer, Simulation Models (SWRRB and EXAMS)
SWRRB predicts pesticide concentration in the runoff (Attachment C_)
EXAMS predicts pesticide concentration in aquatic systems receiving
runoff. It has two versions: (A) steady state input and (B)
pulse loading. The latter is more representative of most agri-
cultural use sites. Both pond and river scenarios available
(Attachment D).
EXAMPLES; (See Attachment F).
COMPUTER
RUNOFF AND
EXPOSURE
MODEL
SIMULATION
(Lentic
and
Lotic)
EEC < Risk Criteria | No Risk
EEC > Risk Criteria
Request Actual Field Residue Monitoring Studies
FIELD DATA
(Lentic)
22
-------
-18-
EEB's current approach to estimating exposure (i.e., how
much pesticide will be in the water [Estimated Environmental
Concentration or EEC]) is similar to the hierarchical or tier
system presented in Subdivision E of the Pesticide Assessment
Guidelines (EPA-540/9-82-024, p. 2). Figure 2 illustrates the
current &EB "Aquatic EEC Tier System." It is generally agreed
that lentic aquatic systems represent more of a worst-case
analysis than lotic systems. Therefore, the lentic system
(e.g., ponds) is often chosen as the aquatic system initially
evaluated. We use state-of-the-art models to develop EECs for
streams and rivers when the analysis in lentic systems indicates
potential for risk. (See Attachment F.)
Pesticide registrants normally supply EPA with data on
the fate and transport of the pesticide in the environment.
These data are used by EEB to, among other things, estimate how
long an EEC calculated in Level 1 or 2 will be present in the
water. The fate data can be divided into five categories of
data requirements and results, as reported in Table 3.
In addition, EEB also receives information on the physical
and chemical' properties of the pesticide. The important
information for consideration in the ecological risk assessment
includes: color, physical state, odor, melting point, bulk,
density or specific gravity, solubility, vapor pressure,
dissociation constant, octanol/water partition coefficient, pH,
molecular weight, and chemical structure.
Initially, EEB considers the worst possible risk situation
(i.e., direct application to water). This worst-case scenario
is described in Figure 2, Level 1. All pesticide uses are
considered under this scenario first, whether or not a direct
application to water is proposed. If there is no risk deter-
mined at Level 1, then no further exposure analyses are necessary.
In Level 1, a "worst-case" EEC is determined by solving the
following mass balance equation:
(1) EEC (ppb) = A (pesticide loading to the body of
water)/B (weight of the water)
Where, A = maximum application rate (Ibs ai/A) x size
of the drainage basin (A) x % runoff
(decimal);
and B = surface area of the body of water (A) x average
depth (ft) x 43,560 ft2/A x 62.36 lbs/ft3
For example, if the application rate is 0.10 Ibs active
ingredient per acre, the drainage basin is estimated to be 1
acre, the percent runoff for a direct application is 100 or 1
when changed to a decimal, the surface area of the body of
23
-------
-19-
Table 3: Environmental Fate and Transport Data
Data Requirements
1. Degradation
a. Hydrolysis
b. Photodegradation
— water
- soil
2. Metabolism
Aerobic Soil
Anaerobic Soil
Anaerobic Aquatic
Aerobic Aquatic
3. Mobility
Leaching
4. Field Dissipation
Soil
Water
Forest
5. Accumulation
Rotational Crop
Irrigated Crop
Fish
Aquatic Non-Target
Reported Results
la. - hydrolytic half-life at
pHs 5, 7 & 9
b. - photolytic half-life in
water and soil
2. - half-life estimates and
residue decline curves
3. - Soil/water relationship
(Kd values); i.e.,
adsorption/desorption
4. - Residue decline curves
5. - Significant residues
accumulated, rates of
accumulation, residue
decline curves
REFERENCE: Environmental Fate Branch, BED, OPP. 1982. Pesticide
Assessment Guidelines Subdivision N, Chemistry:
Environmental Fate. EPA-540/9-82-021.
24
-------
-20-
water is 1 acre, and the average depth of the body of water is
0.5 ft, then equation (1) can be solved as follows:
EEC (ppb) = (0.1 x 1 x !)/(! x 0.5 x 43560 x 62.36) = 73 ppb
This fmmber appears in the third column in Table 2* A
worst-case EEC, like the one calculated above, is then compared
to the lowest acute aquatic ££50 value, chronic effect level or
chronic no effect level (NEL). These values are determined
from the toxicological hazard data outlined above. If the
EEC is less than the acute and chronic aquatic risk criteria,
then EEB presumes that there will be no risk to aquatic organisms
from the proposed pesticide use because there is little likeli-
hood of significant exposure to non-target aquatic organisms.
There would be no need to proceed any further on the EEC Level
System. Alternately, if the EEC is equal to or greater than
these risk criteria, then EEB would proceed to Level 2.
In Level 2, surface runoff (including both water and
sediment) and spray drift are considered the major mechanisms
for pesticide., loss to the aquatic environment. The other
principal mechanisms of pesticide transport away from the
application site (i.e., leaching, degradation and volatilization)
(Rumker, et al., 1975: 104) are not directly included in this
Level 2 estimation, but will be included in Level 3. In some
instances, an estimate of pesticide spray drift can be crucial
in the aquatic risk assessment, especially for pesticides that
are applied by air or by mist blower (see Nigg, et al. , 1984).
Preliminary spray drift scenarios have been developed (see
Attachment E). We estimate that approximately 10% of the
amount of pesticide applied will reach the aquatic environment
via spray drift. It is important to note here that lipophilic
pesticides have a tendency to become tightly bound to soil.
They are, therefore, less likely to be transported via water
runoff, but could be transported via soil erosion. However,
aquatic contamination via spray drift could be an even more
important factor in aquatic risk assessment for these pesticides.
Concerning surface runoff, the same basic equation used in
Level 1 (see equation (1)) is used again in Level 2 to determine
the EEC. However, this time the variables in the equation are
modified to better describe a typical field use site. The
"chemical and site dependent variables," (i.e., drainage basin
size, surface area of the pond, average depth of the pond, and,
runoff percent) are selected from appropriate literature
references (Figure 2). The resultant EEC is again compared
with the aquatic hazard data as modified by the aquatic risk
criteria (Table 1). If the comparison still results in a
conclusion of potential risk to aquatic organisms, then EEB
woul,d proceed to Level 3.
25
-------
-21-
At this point, EEB performs an estimate of an aquatic EEC
using state-of-the-art exposure models (e.g., EXAMS). At the
same time, EEB requests the direct assistance of the Exposure
Assessment Branch (EAB) if any modifications to the model or
use scenario are needed,Scientists from both branches meet
and agree upon the appropriate use site characteristics that
should be entered into the computer runoff and exposure model
in Level 3 (Attachments C and D). All the principal mechanisms
of pesticide transport are considered in this model simulation
including pesticide drift. However, the leaching, or adsorption/
desorption (Kg-) factor is one of the most important parameters.
A number of computer estimated EECs have been completed. The
most appropriate input made for the EXAMS model is the "pulse
loadings" (versus "steady state" or slow build-up from a single
input). These are considered to be more "real" situations for
agrichemical use sites. Pesticides are usually applied at
various times throughout the growing season, and rainstorms,
and irrigation, which result in pesticide runoff, are periodic
in nature. Both lentic (pond) and lotic (river) scenarios have
been evaluated.
It should be noted that these simulations are still being
field validated and are often rebutted by pesticide registrants
either with simulations of their own or arguments that certain
crucial data parameters are biased and need to be changed.
These issues are addressed as they arise, but when disagreements
cannot be resolved, EEB requests actual field residue monitoring
studies (Level 4). Also, at times such studies are offered by
the pesticide applicants in order to verify or refute the model
estimates.
The report of EEC Model results received by EEB often con-
sists of a brief summary of the expected concentrations in dif-
ferent components of soil and water, and a series of tables and
graphs containing the estimated residues over time. In addition,
the report may also contain a page containing a summary of the
chemical properties of the chemical, as shown in Table 4, and a
drift add-on component to the SWRRB/EXAMS model (Attachment E,
Output from Spray Drift Model). The difference between steady
state and pulse loading can be very important. In at least one
case, the pulse loading model produced residue quantities five
times greater than the steady state model. It appears that the
quantity dynamics of the water column and sediments of the pulse
load mode are closer aligned to those conditions found in nature
for agricultural runoffs. Considering these factors, EEB usually
requests at least the pulse loading estimate. Interpretation
errors can result with all these estimates and caution is urged
when they are used in ecological risk assessment.
Perhaps the best approach to date for summarizing and using
the EEC computer model data is found in Attachment G. Of parti-
cular importance is the tabular summary of the SWRRB/EXAMS EEC
26
-------
-22-
Table 4: Summary of Chemical Fate Properties
Common Name:
Chemical Name:
Structure:
Chemical Properties:
Molecular Weight:
Solubility (ppm):
Partitioning:
Hydrolysis (half-life hrs. )
(pH 5? ) _ (pH 7? ) _ (pH 9? )
Kah Knh Kbh
Photolysis (half-life hrs.)
Degradation (half-life hrs. )
Soil (Aerobic )( 25 C )
Soil (Anaerobic) ( 25 C )
Water (Type pH )
Bacteriological
Soil (Type )
Water (Type )
hr K
hr K
hr K
hr K
hr K
Vapor Pressure:
j
Evaporation:
27
-------
-23-
Values. The information on the number of days that the EEC
exceeds a critical risk criteria (e.g., 1/2 LC$Q, which is the
1975 criteria for Presumption of Unacceptable Risk for Aquatic
Organisms in Table 1), and the identification of the day that
the minimum and maximum EECs occurred clearly shows whether
EEB should have a concern for risk to non-target aquatic
organisms, which non-target organisms are at risk, and what
the duration of the risk is. Attachment G is a good written
assessment of risk, integrating exposure data and toxicological
hazard data.
5. Aquatic Non-Target Exposure
Concurrent with the aquatic EEC determination is an
analysis of the use pattern and use site characteristics. This
is done in order to determine which aquatic non-target organisms
are likely to be exposed to the proposed use of the pesticide.
Information on the use pattern and the use site are
collected from a wide variety of sources. The following
contains a number of information sources that are used: proposed
pesticide label; 1978 Census of Agriculture, Vol. 1, Parts 1-50;
EPA Label File; USDA Statistical Reporting Service, Crop Reports
by State; EPA Compendium of Registered Pesticides; EEB Crop
Index and Chemical File; USDA Economic Research Service (Staff
Reports on Pesticide Use for Crop: National, Regional and
State Level); EEB Agricultural Crop Information File; USDA
County Extension Agents; and personal contacts and personal
experience of Agency personnel.
Once the specific treatment areas are identified, then EEB
describes the aquatic non-target organisms that are associated
with the treatment areas and which would likely be exposed to
the pesticide due to their natural history characteristics,
(e.g., food habits, breeding requirements and behavior, resting
behavior). We rely on personal contacts with experts in the
field and on a staff member's personal experience to complete
this analysis, as well as consulting appropriate references:
Lagler, 1956. Freshwater Fishery Biology.
EPA-670/4-73-001, Biological Field and Laboratory Methods
for Measuring the Quality of Surface Waters and Effluents.
Scott and Grossman, 1973. Freshwater Fishes of Canada.
Usinger, 1956. Aquatic Insects of California.
Lee, et al., 1980. Atlas of North American Freshwater
Fishes.
Pflieger, 1975. The Fishes of Missouri.
28
-------
-24-
Trautman, 1981. The Fishes of Ohio.
Hutchinson, Vol. I (1957), II (1967), III (1975). A Treatise
on Limnology.
Hoese and Moore, 1977. Fishes of the Gulf of Mexico.
Holt, Ed. 1969 & 1971. The Distributional History of the
Biota of the Southern Appalachians, Part I: Invertebrates,
and III: Vertebrates.
EPA, 1972, Biota of Freshwater Ecosystems, Water Pollution
Control Research Series 18050 ELD 05/72, Manual No. 1-10.
Galtsoff, Ed. 1957. The Gulf of Mexico, Its Origin, Waters
and Marine Life. Fishery Bull. No. 89, Vol. 55.
Carlander, K.D. 1969. Handbook of Freshwater Fishery Biology.
Vol 1. Iowa State Univ. Press, Ames Iowa.
Carlander, K.D. 1977. Handbook of Freshwater Fishery Biology.
Vol. 2.' Iowa State Univ. Press, Ames Iowa.
Gosner, K.L. 1971. Guide to Identification of Marine and
Estuarine Invertebrates. Wiley-Interscience, N.Y.
Lauff, G.H., ed. 1967. Estuaries. Amer. Assoc. Adv. Sci.
Washington, D.C.
Pennak, R.W. 1953. Freshwater Invertebrates of the United
States. Ronald Press, N.Y.
Smith, R.I., F.A. Piteka, D.P. Abbot and P.M. Wessner. 1970.
Intertidal Invertebrates of the Central California Coast.
Univ. Calif. Press., Los Angeles.
Zingmark, R.G., ed. 1978. An Annotated Checklist of the Biota
of the Coastal Zone of South Carolina. Univ. S. Carolina
Press, Columbia, S.C.
At this point, EEB has reviewed and summarized (1) the
toxicological hazard data for various indicator species of
aquatic non-target organisms and (2) the estimated exposure of
the pesticide from the proposed use. The latter is described
in terms of its residues in the aquatic environment and the
aquatic organisms that are likely to be exposed to the pesticide.
The next step is to actually perform the assessment of risk.
29
-------
-25-
6. Risk Assessment
By relating the known biological and ecological responses
elicited by known concentrations of a pesticide to the actual
or estimated environmental concentration of that pesticide, it
is possible to determine the likelihood of adverse effects and ,
thus, to make an ecological risk assessment. EEB's approach to
using and comparing hazard and exposure data is largely based
upon an approach recommended by an Aquatic Hazards of Pesticides
Task Group of the American Institute of Biological Sciences
(Cairns, Jr., et al. , 1978). Part of the charge to the Task
Group was to develop criteria and rationale for the use of the
basic test data already required (acute toxicity data and
environmental fate data), and to determine the need for addi-
tional testing (life-cycle, early life-stage, accumulation,
simulated and actual field). From the many criteria that were
considered, this Task Group selected six criteria that they
considered to be of primary importance for making decisions
concerning the need for additional data:
0 LCso .1-ess than «) 1 mg/liter (ppm);
0 Estimated environmental concentration greater than (>)
0.01 of the
0 Pesticide used on a major crop or otherwise to be broadly
used;
0 Water solubility value less than «) 0.5 ppm or octanol:
water partition coefficient greater than (» 1000;
0 Half-Life in water greater than (>) 4 days; and
0 Avian safety, mammalian safety, or efficacy test results
produce abnormal, reproductive, and/or other unusual
effects at low dosages or concentrations.
If any one or more of these criteria are met, then additional
testing, beyond the basic test data, is needed. EEB has incorpo-
rated these criteria into Subdivision E of the Pesticide Assessment
Guidelines and into the footnotes of Part 158.145 of the regula-
tions. Many of EEB's pesticide risk assessments end with the con-
clusion that there is a great potential for risk because pertinent
hazard and exposure data are lacking. Therefore, before a more
detailed ecological risk assessment can be completed, additional
data are needed. When all required data have been requested, re-
ceived, reviewed and summarized by EEB, the toxicological hazard
data are compared to the exposure data and the risk is characterized
on the basis of the risk criteria in Table 1. The risk assessment
is generally broken down into acute and chronic risk to fish and
aquatic invertebrates.
30
-------
-26-
a. Fish Acute Risk
In assessing the acute risk of a proposed pesticide to
non-target fish, EEB uses biological response data from acute
bioassays on fish:
0 Coldwater fish LCso;
0 Warmwater fish LCso; and
0 Estuarine/marine fish
The lowest LCso value for each ecosystem which will be
impacted by the pesticide (freshwater and/or estuarine or
marine) is selected from the toxicological data. The typical
approach is to directly compare the aquatic EEC with the
selected LCso value (s). If the EEC is less than l/10th the
aquatic LCso(s), then EEB presumes no acute risk to fish (see
Table 1). If the EEC is between l/10th LCso and 1/2 LC50 (see
Table 1), then EEB presumes that there is risk to fish which
may be mitigated by some precautionary labeling statements
(e.g., "This pesticide is toxic to fish. Drift and runoff from
treated areas may be hazardous to fish in neighboring areas")
or through labeled use restrictions (e.g. , for a mosquito
larvicide, "Do not apply to fish bearing waters").
If, however, the EEC is equal to or greater than a lethal
concentration, which results in significant acute effects on
the population, or the EEC is equal to or greater than 1/2 the
aquatic LCso (see Table 1), then EEB presumes that the acute
risk from the use of the pesticide is significant. The antici-
pated effects are described with as much detail as possible,
sometimes referring to previous significant pesticidal effects
on fish from requested field studies, fish kill data or field
effects from the use of similar pesticides. If the data are
sufficiently compelling, a special review may then be initiated,
b. Aquatic Invertebrate Acute Risk
Biological response data from tests shown here are used
to assess the acute risk of a proposed pesticide to aquatic
invertebrates. These include:
0 Daphnia or aquatic insect
e Shrimp LCso/ECso;
0 Oyster embryo-larvae LCso;
0 Oyster shell growth ECso«
The approach is the same as that described for risk to fish.
31
-------
-27-
c. Fish and Aquatic Invertebrate Chronic Risk
In order to assess the chronic risk of a proposed pesticide
use to fish, the following biological response data are used:
0 Fisft (early life-stage or life-cycle) effect level, NOEL,
or MATC;
0 Aquatic Invertebrate Effect Level and NOEL; and
0 Fish/Aquatic Invertebrate Bioconcentration Factor and
Depuration Rate.
If the EEC is less than or equal to the chronic no observable
effect levels for fish and aquatic invertebrates (see Table 1),
then EEB presumes no chronic risk to fish and aquatic invertebrates.
If the pesticide use causes significant adverse effects on the phy-
siology, growth, population levels or reproductive rates, as indi-
cated in laboratory tests and confirmed in field tests, or the EEC
is equal to or greater than a level which results in significant
chronic effects on aquatic non-target populations, then the presump-
tion is that' the chronic risk from the use of the pesticide is sig-
nificant. At tiroes labeled use restrictions such as "Do not use
in the following counties/states ... " and "This is a Restricted Use
Pesticide" may mitigate the presumed chronic risk. Although there
have been few documented population reductions due to chronic pes-
ticide effects (e.g., reproduction impairment), one of particular
note was reported during the EPA Cancellation hearings for DDT.
Evidence presented in the hearings indicated that
DDT was responsible for the death of lake trout
fry hatched from eggs taken from Lake George, a
tributary of Lake Champlain. It has also been
implicated in excessive mortality of Lake Michigan
coho salmon fry, and salmon eggs from a Maine lake
exhibited lowered hatchability when DDT levels
reached 3 ppm in the eggs. In 1969, residues of
DDT in sea trout in the Laguna Madre (Texas) were
correlated with residues in menhaden, a major food
of the trout. Reproductive impairment had been
observed since 1964 as evidenced by a decline from
30 to 0.2 juvenile trout per acre. After residues
in menhaden declined, the sea trout populations
returned to 1964 levels. (See EPA-540/1-75-022.)
7. Endangered Species
The above discussion on the aquatic risk assessment procedures
did not specifically address risks to endangered species. This is
a normal component of the review, and the approaches to assessing
risk are identical except for two items: (a) the risk criteria,
and (b) consultation with USDI, Office of Endangered Species (see
Attachment H).
32
-------
-28-
a. Endangered Species Risk Criteria
More stringent criteria are used in the risk assessment
for endangered species in order to provide greater protection
to populations of aquatic organisms already severely reduced to
levels where their survival as a speetes is questionablei The
criteria were developed to provide, an estimate of a no-effect-
level or at least a minimal-effect-level based on the data
provided for non-endangered species. If the EEC is less than
l/10th the lowest aquatic acute LCio (when a slope is available)
or less than l/20th the lowest aquatic LCso (when no slope is
available) or less than the lowest aquatic chronic no-effect-
level, then EEB presumes that there will be minimal risk to
endangered aquatic organisms from the use of the pesticide.
If, however, the EEC is greater than the levels set by the
above criteria, then EEB presumes that there will be a risk
to aquatic endangered species and, thus, a formal consultation
with OES is initiated.
b. OES Formal Consultation
This communication is needed to elicit written opinion
from OES concerning our presumed risk, the extent of the risk,
and whether labeling can mitigate the risk. It is typically
requested after an informal consultation with the U.S. Fish and
Wildlife Service (USFWS) or National Marine Fisheries Service
(NMFS) personnel is made and takes the form of a written
communication between EEB and USFWS or NMFS personnel.
Specifically, the request is in the form of a letter from the
Chief of EEB to the NMFS or the region of the USFWS where the
effects may occur. If more than one region is involved, EEB
contacts USFWS Headquarters in Washington, D.C., to arrange
for one of the regions to receive the request for formal
consultation and to contact the other regions.
EEB provides information to the USFWS or NMFS to allow
them to determine if jeopardy may occur to endangered species
from the proposed use. At a minimum the following information
is included, either in the letter or as an addendum:
0 Chemcial name;
0 Type of pesticide (insecticide, herbicide, etc.);
0 Proposed use(s) of the pesticide (e.g., crop/site,
label use rates, methods of application, geographical
location of proposed use site);
0 Toxicity data of the pesticide (only valid data are
included;
33
-------
-29-
0 Exposure information (exposure levels for environmental
compartments applicable to each toxicity category value
provided in (d) are presented);
0 Environmental fate data; and
0 A discussion of the potential hazards (a list of the endan-
gered species EEB has identified as possibly affected is
included).
When the USFWS or NMFS Biological Opinion is received, EEB
determines/ based on the recommendations in the Opinion, the
appropriate course of action. If the Opinion concludes jeopardy,
the viable options are:
0 Label restrictions designed to eliminate risk to endangered
species;
e Recommendation against registration of the pesticide; or
0 Further contact with the USFWS or NMFS for clarification
if there appears to be an inconsistency in the Biological
Opinion.
Also, if EEB disagrees with the Opinion, then EEB reinitiates
consultation with the USFWS or NMFS providing documentation to
support its position.
B. Terrestrial Risk Assessment
1. Introduction
The terrestrial risk assessment involves an examination of
the potential hazards of proposed pesticide uses to non-target
mammalian and avian wildlife. Mammalian and avian wildlife are
given more emphasis in the assessment, primarily, because of two
factors: (1) established protocols for toxicity testing exist
for certain mammalian and avian species; and (2) mammals and birds
are usually the organisms of greatest economic value, if lost or
harmed, because they generally constitute the "game species" of
local, state, and federal governments or agencies. However, two
items must also be considered: (1) it is assumed that when birds
and mammals are "protected" via the risk criteria of the assess-
ment, some "protection" is afforded reptiles and amphibians by
these same criteria; (2) as the state-of-the-art of toxicity test-
ing develops, other organisms, such as reptiles and amphibians,
can be considered more accurately in the risk assessment process.
Non-target insects will be considered in a later section of this
report.
34
-------
-30-
2. Indicator Species/Test Species
The avian indicator organisms used in the terrestrial
toxicity tests are usually the bobwhite quail, ring-necked
pheasant, and mallard duck. The mammals most often used are
domestic ones te^., laboratory rat) *nd are those utilized
in the human risk assessment process.. As needed, certain
non-domestic mammals are used, generally those representative
of areas where pesticide applications are likely to occur.
Also, avian organisms such as red-winged blackbirds, starlings,
mourning doves, sparrows, and Canada geese are also used
depending on use site and pesticide application.
3. Terrestrial Residues
The estimated terrestrial residue profile developed for
pesticides is based primarily upon the works of Hoerger and
Kenaga (1972) and Kenaga (1973). In the earlier article the
authors examined the residue levels from literature sources and
tolerance data of twenty-eight different pesticides in or on
sixty crops.
-------
-31-
Table 5. Maximum Expected Residues and Typical Residues
of Pesticides on Differing Categories of
Vegetation Types (from Hoerger and Kenaga, 1972)
ppm Residue on the Basis of a
Pesticide Dosage of 1 Ib Per Acre
Immediately
After Application
6 Weeks
After Application
Plant
Category
Range Grass
Grass
Leaves and
Leafy Crops
Forage Crops
Pods Containing
Seeds
Grain
Fruit
Upper
Limit
240
110
125
58
12
10
7
Typical
Limit
125
92
35
33
3
3
1.5
Upper
Limit
30
20
20
1.0
1.5
1.5
1.5
Typical
Limit
5
1-5
< 1
< 1
< 1.0
< 1.0
< 0.2
36
-------
-32-
Concerning the estimated residues likely to be found in or
on insects, EEB utilizes the approach proposed by Kenaga (1973).
In his article Kenaga states that residues on insects can be
estimated from residue data for plants with a similar surface
area-to-mass ratio as the insects in question. He indicates
that for small insects the residue data available for dense
foliage situations (alfalfa, clover,.trefoil: 1.0 Ib active
ingredient per acre corresponds to a maximum expected residue
of 58 ppm) are relevant whereas for large insects the data for
seeds and pods (1.0 Ib active ingredient per acre corresponds
to maximum expected residues of 10 to 12 ppm) are pertinent.
EEB finds this approach reasonable, particularly since: (a) in-
sect residue data are lacking for most pesticides; (b) insects
constitute a major portion of the diet of certain non-target
organisms and, therefore, should be considered in a non-target
organism hazard evaluation; and (c) certain residue data for
insects presented by other researchers support Kenaga's proposal.
For example, McEwen, et al., (1972) found that 0.25 Ib active
ingredient Guthion per acre sprayed ultra low volume (ULV) to
control grasshoppers resulted in a residue of 14 ppm Guthion in
grasshoppers on the day of application, and this correlates
with 56 ppm Guthion RUD (Residue from a Unit Dosage).^/
In another spray program involving Toxaphene to control
range caterpillars, these same researchers found a range of 7.2
to 34 ppm Toxaphene in range caterpillars from an application
rate of 1.0 Ib active ingredient Toxaphene per acre (or RUD).
These results, therefore, tend to support Kenaga's hypothesis
regarding foliar/insect residue correlations (10 ppm for large
insects/seeds, pods, fruit and 58 ppm for small insects/dense
.foliage).
Whenever possible, EEB utilizes actual residue data as
supplied by the registrant or found in the literature. Often
such data are lacking, particularly, residues in or on non-target
organism food items such as insects, other invertebrates, seeds,
pods or nuts.
4. Risk Assessment
The next step is to determine the likelihood of exposure
and hazards by correlating the information on residues, food
p_/ Hoerger and Kenaga (1972) define RUD as "Residue from a
Unit Dosage" or:
RUD = actual residue = ppm
treatment rate Ib pesticide/A
37
-------
-33-
items consumed/ and wildlife utilization of crop areas with the
toxicity data available for mammalian and avian species. The
terrestrial toxicity data usually available for assessment are
as follows:
TIER 1
1. The mammalian toxicity data' submitted in support of
(human) toxicology data requirements (e.g., rat acute
oral LDso; acute dermal toxicity; 90-day feeding
studies — rodent and non-rodent);
2. Avian acute oral LDso (upland gamebird or water fowl
species);
3. Avian dietary LCso (upland gamebird); and
4. Avian dietary LDso (upland gamebird);
TIER 2
5. Wild mammal toxicity data (generally, an acute oral
study);
6. Avian reproductive studies (upland gamebird or water
fowl species); and
7. Special studies with avian or mammalian species (e.g.,
non-target mammalian reproduction studies, avian acute
dermal LDso, avian cholinesterase test, avian or mammalian
secondary toxicity);
TIERS 3 and 4
8. Simulated and actual field testing with avian and/or
mammalian species.
With these data EEB correlates the estimated residues in
or on mammalian and avian food items the number of granules
likely to be ingested, or the quantity of pesticide which may
directly contact non-target organisms and develops estimates
concerning potential risks. The various approaches which can
be taken are discussed below.
a. Mammalian Species
(1) Acute Risks
In assessing the acute hazards of pesticide uses to non-
target mammals, the EEB typically utilizes the data from mammalian
(human) toxicity studies, wild mammal LDso studies, simulated and
actual field testing with mammalian species, or special studies
with mammalian species. In most risk assessments, though, the
38
-------
-34-
laboratory rat acute oral LD$Q is exclusively used. EEB correlates
these data with maximum expected numbers of granules, numbers of
seeds, numbers of baits, or vegetative/insect residues to obtain
estimates of hazard. Table 6, for example, provides the typical
approach for determining granular, seed, or bait hazards. Typi-
cally, rat LDso data on the technical grade of the active ingre-
dient are available. These data are -converted first to an LD5Q
based on active ingredients (ai) relative to the animal's body
weight (see columns 1 through 4, Table 6). This LDso *s then used
to develop the lethal amount of product (usually a formulation of
lesser percent active ingredient than technical grade) estimated
to produce such an LDso (see column 5, Table 6). Then the quantity
of pesticide product available to the organism (represented as
milligrams, baits, granules, or seeds) is generated using the
application rate(s) and use site information. The calculations
typically address a one square foot area, but daily feed consump-
tion values and estimates of the multiples of lethal dose which
may be consumed daily also provide a perspective to the potential
risk picture (see columns 6 through 10, Table 6). (It should be
noted however, that granular hazards to mammals should, in most
cases, be mitiimal, especially considering the greater hazards for
avian species. Theoretically, such risks are possible, particu-
larly for small mammals such as insectivores which may accidentally
ingest granules directly or granules that may adhere to earthworms
or soil insects). As can be seen, the final analysis correlates
the quantity of product, seeds, baits or granules available in a
square foot (i.e., 52.11 baits, seeds, or granules) with the
amount of product, seeds, baits or granules needed to produce an
LDso on a Per animal body weight basis (i.e., 20 baits, seeds, or
granules).
These LDso/seed, bait, granular correlations can be carried
further as shown in Table 7. This attachment presents a species
sensitivity profile (a subject discussed in more detail under avian
species) which points out that smaller mammals, simply based on
body weight comparisons, require ingestion of fewer granules than
larger mammals to reach an LD$Q. For example, a 20-day-old eastern
cottontail theoretically has to ingest 91.3 granules of a 10G
(10% ai granular) formulation to reach an LDso- An adult cotton-
tail, however, needs to ingest 1,182 granules of the same 10G
formulation.
Correlation of acute mammalian toxicity data with estimated
or actual residues is usually performed as shown in Table 8. The
rat LDso data must first be converted to an LCso value based upon
the relationship from Lehman (1959), and once this is done, the
correlation of converted LCso an^ residue data can be made. For
example, if the rat LDso, body weight, and food consumption values
are 100 mg/kg ai, 0.40 kg, and 0.02 kg, respectively, the following
relationship exists:
39
-------
-35-
Table 6. Mammalian Toxicity/Use Rate Correlation: 10% ai
Granular Formulation Applied at 1 Ib ai/A
(1)
(2)
(3)
(4)
(5)
Organism
Rat
Body
Weight
0.4 kg
LD50
(rag/
kg ai)
10 mg/kg
LD50
(rog/
animal ai)_/
4 mg/animal
Lethal Amount
of Product
to Produce
LD50/Animal_/
40 mg
(6)
(7)
(8)
(9)
(10)
Quantity
of Product
Sq. Ft.
104.22 mg
Lethal
. Number
of Baits,
Seeds or
Granules/
Animalfy
20
Number
of Baits,
Seeds, or
Granules/
Sq. Ft.Sy
52.11
Daily Feed
Consumption
20,000 mg
Multiples
of Lethal
Dose Which
May Be
Consumed
Daily£/
500X
£/ 10 mg/kg ai x 0.4 kg = 4 mg ai/animal.
3_/ (4 mg/animal)/(10% product) = 40 mg product.
V (1 Ib ai/A)/(10% product) = 10 Ib product/A
(10 Ib prod/A) x (454,000 mg/1 Ib) = 4,540,000 mg prod/A
(4,540,000 mg prod/A)/(43,560 ft2/A) = 104.22 mg/ft2
(40 mg lethal amt. of prod.)/(2 mg granule, seed, or bait
weight) = 20 granules, seeds, or baits
_ (104.22 mg prod./f t2)/( 2 mgs granule, seed, or bait weight) =
52.11 granules, seeds, or baits/ft2
£/ (20,000 mg food cons.)/(40 mg lethal amount prod.) = 500X.
40
-------
-36-
Table 7. Mammalian Species Sensitivity Profile ^
Pesticide 10G (10% ai)/15G (15% ai)
Hazard to Four Species of Non-Target Mammals
Number of Granules Equal to
Body
Weight
Species (g)
Rat 200
Eastern
Cottontail
(Adult) 1100
Weaned Young
20 days old 85
Grey Squirrel
(Adult-
Female) '520
Weaned Young.
10 weeks
old 200
Delmarva Fox£/
Squirrel
(Adult-
Female) 795
Weaned Young
8-10 weeks
old 454
Mg/i/ LD50i/
An imal
(g)V 10G 15G
2.0 215.0 142.8
11.0 1,182.0 785.7
0.85 91.3 60.7
5.2 559.1 371.4
2.0 215.0 142.8
7.95 816.1 567.8
4.54 483.8 324.2
l/5th
10G
43.0
236.5
18.2
111.8
43.0
163.2
96.7
LD50i/
15G
28.5
157.1
12.1
74.2
28.5
113.5
64.8
3/
_/ Utilizing rat LDcQ of 10 mg/kg ai from empirical data
Weight of one 15G granule = 0.093 mg
Weight of one 10G granule = estimated to be same as 15G granule
Weight of pesticide in one granule:
0.093 mg x 15% = 0.0139 mg ai/granule
0.093 mg x 10% = 0.0093 mg ai/granule
_ Rat LDtjQ =10 mg/kg ai
10 mg/kg ai x 0.2 kg body wt. = 2 mg ai/animal
(LDso for Rat on per body wt. basis)
All other values in this column based on the assumption that
each organism has the same sensitivity as the rat (i.e.,
LDso for each organism is 10 mg/kg ai)
_/ Number of 15G granules _ 2 mg ai/animal _ 142.8 granules
required to equal l>V$n ~ 0.014 mg ai/granule
5' l/5th the LDt;n = restricted use classification trigger (see Table 1)
Weight data obtained via personal communication with Gary
Taylor and Dr. Vagan Flyger of the Delmarva Fox Squirrel
Recovery Team
41
-------
-37-
Table 8. Mammalian Food Factor/Residue Calculations
Diet Types
Residue
Contamination
of Diet (ppm)
(From Kenaga,
1973)
Food
Consumed By
Organism
(Hypo-
thetical)
Maximum
Adjusted
Residues
(ppm)
Total
Dietary
Residues
(ppm)
Dense foliage
Small insects
and
Seeds
Seeds
58 ppm
58 ppm
10 ppm
10 ppm
x
x
X
X
100%
50%
50%
100%
58 ppm
29 ppm
5 ppm
10 ppm
58 ppm
34 ppm
10 ppm
Summary;
LD50
Organism (mg/kg/ai)
Estimated
LC50
(ppm ai)
Diet
Situation
Total Dietary
Residue (ppm)
Rat
Same
Same
100 mg/kg
Same
Same
2000 ppm
Same
Same
100% dense
foliage
50% small
insects
50% seeds
100% seeds
58 ppm
34 ppm
10 ppm
42
-------
-38-
food consumption (kg) x residue (mg) - mg/kg/day
body weight (kg) food (kg)
Using this relationship, the reviewer can develop a theoretical
LCsfr for ^the rate of 2&QQ ppro ai aftd then correlate this value
with residues estimated to be on available food items (in some
cases, however, actual LCso data are 'available (McCann, et al. ,
1981), but these are rare since the requirement for such testing
has not been developed. An excellent discussion of this approach,
including an examination of the use of converted LCsos as opposed
to actual LCsos, is presented in the reference just cited. It
shows the problems encountered when utilizing LDso toxicity data
and residue data in the estimation of hazards. In this paper it
shows that out of 17 chemicals tested only one (EPN) had a converted
LCso value (660 ppm) similar to an actual dietary LCso value (603
ppm). Further, it points out that different hazard decisions
were reached 35 percent of the time when the converted LCso values
and the actual LCso values were compared (McCann, et al. , 1981).
It should be noted in Table 8 that food factors are briefly
discussed. 'A more lengthy discussion is presented below under
the avian species risk discussion, but it should be recognized
that for mammalian organisms food factors typically have not been
used in EEB. One possible reason for this is that the main body
of mammalian data used concerns rodents. Rodents, at least the
smaller ones (and these are the ones for which the greater pesti-
cide hazard may exist), are primarily vegetarian (Martin, et al. ,
1951). Thus, the food factor for small rodents usually has been
treated as 100% (i.e., 100% of the organism's total diet is con-
taminated with pesticide; see Table 8). Also, a species sensiti-
vity profile (as developed in Table 7) was not developed in Table
8. This is because a lengthy discussion of such a profile (when
utilizing LCso data) is presented in the avian species section.
Other approaches to a determination of acute mammalian hazards
are possible but in most instances these have not been utilized
in EEB. Mammalian dermal LDso data have been utilized for an
assessment of acute dermal risks by correlating the dermal LDso,
adjusted to a per animal body weight basis (as shown in Table 7
using acute oral LDso data), with the estimated quantity of pesti-
cide per square foot (usually mg ai/ft^ as shown in Table 6).
Some pesticide studies with penned rabbits (to determine dietary
as well as dermal hazards) have been performed, but these are not
common. Development of other mammalian risk assessment techniques
has been minimal, possibly due to a lack of acceptable protocols
or an apparent lack of field mortality (with the exception of
certain pesticides).
43
-------
-39-
Special mammalian studies to assess acute risks are not
commonly used by EEB. However, special tests, which are
usually modifications of existing protocols (e.g. , extending
the observation period in LD$Q studies performed with anti-
coagulant rodenticides), have been utilized for rodenticides
and predacides, but only on a case-by-case basis.
(2) Subchronic/Chronic Risks
The assessment of subchronic or chronic hazards to mammals
correlates the available data from the mammalian toxicity data
submitted in support of human toxicity data requirements (e.g.,
90-day feeding, chronic feeding, oncogenicity, or reproductive
studies) and any other pertinent effects data (such as avian
reproduction effects data), with estimated or actual field
residues data. Such a correlation usually integrates the
available no-effect-levels (NELs) and/or effect levels with
the residues in question. The weakest link in this process,
however, is the accurate determination of residues, particularly
their extent and duration. In most instances, field dissipation
data, especially for residues in or on pertinent mammalian food
items, are riot available. Such dissipation data usually concern
residues in soil, water, and in or on crop parts intended for
human consumption.
It should be noted that, at least historically, few risk
assessments have been performed that address subchronic or
chronic adverse effects to non-target mammals. The emphasis in
this area of terrestrial effects has been on avian species.
The reasons for this are, possibly, that: (1) there is a general
lack of field evidence concerning chronic adverse effects to
mammalian wildlife from long-term exposure to pesticides; and
(2) established protocols which address such effects on non-
target mammalian wildlife are not readily available. Zepp and
Kirkpatrick (1976) and Gilbertson (1975) are two references
where reproductive effects in non-target mammals (i.e., cotton-
tails and minks, respectively) are addressed.
(3) Secondary Hazards
The majority of EEB's secondary risk assessment for
mammals (and birds) concerns pesticides used to control
rodents, carnivores, or other mammalian organisms. These
pesticides, generally categorized as rodenticides and
predacides, can be divided into three groups: (1) acutely
toxic compounds such as 1080, strychnine, and zinc phosphide,
(2) first generation anticoagulants, or those compounds which
require multiple feedings to produce a toxic effect (e.g.,
warfarin, diphacinone, and chlorophacinone), and (3) second
(sometimes called third) generation anticoagulants — those
44
-------
-40-
chemicals, which via a single feeding or limited feedings, produce
an anticoagulant toxic effect ((e.g., brodifacoum, bromadiolone,
and difenacoum) Kaukeinen, 1982; Kaukeinen, 1984). Reviews, there-
fore, focus on these types of compounds because of their strong
potential to produce secondary tpxicity andi risks, However, the
review of secondary toxicity has broadened to include the assess-
ment of the potential secondary hazards associated with the use
of other pesticides (e.g., carbamates and organophosphates). This
is the result of findings by Balcomb (1983) who concluded that the
use of Furadan® 10 granules (10% carbofuran) in a Maryland corn-
field resulted in the poisoning (including one death) of several
raptors which apparently fed on small mammals or birds that had
been killed or immobilized by ingestion of carbofuran granules.
Although these findings on secondary toxicity relate to birds,
EEB considers such toxicity and risks likely for non-target mammals
and birds. Further, the potential for similar effects appears
plausible for organophosphates and, possibly, other classes of
compounds.
Another aspect of secondary toxicity/risk assessment important
to this discussion is that EEB generally recognizes secondary tox-
icity among organisms higher in the food chain. That is, the branch
does not consider mammalian (or avian) effects (including mortality)
from the ingestion of pesticide-contaminated invertebrates such as
insects as secondary toxicity. The distinction may seem minor, but
White, et al., (1979) present the mortality of laughing gull chicks,
chicks, from the ingestion of parathion-contaminated insects carried
to the chicks by adults, as secondary toxicity. EEB considers this
form of toxicity as primary, recognizing, of course, that the actual
primary poisoning occurred at the insect level. With prey and pre-
dators higher in the food chain (e.g., rodents consumed by owls),
however, EEB recognizes toxicity at the predator (owl) level as
secondary. The following shows the distinction:
Primary Versus Secondary Toxicity
LEVEL AT WHICH TOXICITY OCCURS
Primary (A)
Insect
Insect
Rodent
Rodent
Secondary (B)
Quail
Mouse
Owl
Ferret
RECOGNIZED TOXICITY/
HAZARD ASSOCIATED
WITH (B)
Primary
Primary
Secondary
Secondary
45
-------
-41-
In performing a secondary risk assessment for non-target
mammals/ EEB uses the approaches outlined earlier for acute and
subchronic/chronic risks, but carries the assessment further by
developing a larger toxicity data base, for both primary and
secondary (and addressing both acute and chronic) toxicity and
risks. For example, a typical review of a rodenticide proposed
for use in and around agricultural buildings could include data
requirements (for mammals) that address:£/
0 The primary acute toxicity for the target organisms
(rodents);
0 The primary chronic toxicity for the target organisms
(rodents );
0 The primary acute toxicity to non-target mammals (e.g.,
species from several wild mammal families such as felidae,
mustelidae, or canidae);
0 The primary chronic toxicity (if the rodenticide is persis-
tent in the environment) to non-target mammals (e.g.,
species from several wild mammal families such as felidae,
mustelidae/ or canidae);
0 The secondary acute (and chronic if the rodenticide is per-
sistent) toxicity to representative species from several
wild mammal families such as felidae, mustelidae, or cani-
dae; and
0 Residue analyses which address:
- Levels of toxicant in proposed formulations such as
baits;
- The rate of degradation of toxicant in proposed for-
mulation(s) under typical use conditions;
- Levels of toxicant in target organisms feeding on the
proposed formulations;
- The rate of depuration for the toxicant after a rodent
stops feeding on the proposed formulation(s); and
- Body residues and rate of depuration for predators
feeding on contaminated rodents.
These requirements would be in addition to those determined
for avian, aquatic/ and, in some cases/ reptilian and/or
amphibian species.
46
-------
-42-
With>these data EEB would attempt to correlate expected envi-
ronmental and body residues with toxic effect levels found in ttie
toxicity studies. Extrapolation to field situations would be
undertaken, but in most cases field studies would be required to
determine the primary and secondary toxic effects likely under
actual use conditions.
This approach is similar to that presented by Kaukeinen (1982)
where he states:
Parameters associated with such studies in-
clude: 1) prey selection, 2) prey intoxi-
cation, 3) prey residue determination, 4) prey
preparation for predator consumption, 5) pre-
sentation, 6) predator selection, 7) predator
health and handling, 8) predator captive con-
ditions, 9) predator acclimation, 10) predator
intoxication, 11) predator observation and
evaluation, 12) pathology and residue analysis.
As the above discussion indicates, EEB's secondary toxicity/
risks assessment is a highly complex one — one in which the toxicity
data base analyzed and the number of use/exposure variables and varie-
ty of organisms at different trophic levels considered is extensive.
Each facet of information is critical, and testing must be pertinent
to proposed use situations, for as Savarie, P.J., et al., (1978) and
Kaukeinin (1982) point out, extrapolation of laboratory toxicity
results to field situations is extremely difficult, if not question-
able. Thus, with these types of pesticides (i.e., those exhibiting
secondary toxic effects) simulated field and/or actual field testing
is normally a requirement. Such field testing is highly complex,
expensive, time-consuming, and typically requires the employment of
highly trained, experienced personnel. Protocols for such testing
are developed on a case-by-case basis and are pertinent to the pesti-
cide, the proposed use situation, and the non-target organisms
likely to be exposed and potentially affected. Further, development
of such protocols requires close cooperation between registrants
and branch personnel. It should also be noted that to date only a
minimal number of acceptable field studies for mammalian species
have been received. Thus, EEB's efforts are presently directed
towards standardization of testing procedures and secondary risk
assessment methods.
b. Avian Species
(1) Acute Risks
The use of the avian toxicity data in the assessment of acute
risks to avian wildlife is similar to and generally better defined
than for mammals. Use of the avian acute oral LDsg to assess granular,
bait or seed risks is undertaken following the procedures outlined
in Tables 6 and 7. However, the LD$Q data can also be used as shown
in Table 8 (i.e., conversion to estimated LCsgs and correlation of
47
-------
-43-
such LC5QS with actual or estimated residues can be made). Usually,
though, a species sensitivity profile as shown below (see Table
9) is developed. Then conclusions for potential effects to various
non-target avian species can be made.
As an example of the use of the avian acute oral LDso/ Table
9 presents the theoretical number of. granules (of a hypothetical
pesticide) needed to produce an LDso in a variety of bird species.
For a house sparrow, a bird one-fifth the size of a mourning dove,
only 2.8 15G (15% ai) granules are needed to produce an LD$Q. The
mourning dove, however, requires five times the same amount or
approximately 14 granules. The calculations shown are similar to
those in Tables 6 and 7. An excellent discussion on this approach
is presented in three articles (Balcomb, 1979; Balcomb, 1980;
Balcomb, et al. , 1984). The first two discuss the Agency's basis
for classifying certain granular insecticides as "Restricted"
(i.e., for use by certified applicators) based on their toxicity
and palatability to avian species, their exposure on soil surface
following soil incorporation, and reported bird kills. Of parti-
cular interest are the application of Agency risk criteria (i.e.,
one-fifth the avian LCso) to the data base and calculations which
estimate the number of granules required to produce an avian LDso
from available toxicity data (Balcomb, 1979; Balcomb, 1980). The
third paper further examines the use of LDso/granule toxicity data
by correlating field effects data with laboratory toxicity data.
For example, by the use of the acute oral LDso granule data and
the observed field effects information, the researchers concluded
that almost any ingestion of the granular pesticide studied (car-
bofuran) by smaller bird species may be fatal. Larger species,
however, might survive because of the increased quantity of
carbofuran needed to cause mortality and the reversible nature of
carbamate poisoning (Balcomb, et al. , 1984).
When using avian LCso data from avian dietary studies, EEB
typically makes a direct comparison of the LCso value with actual
or estimated residue levels developed from Kenaga (1973). For
example, if the bobwhite quail LCso were 50 ppm ai and the maximum
expected residues in or on seeds were 10-12 ppm ai from a 1 Ib
ai/A application, then the 50 ppm LCso value (or one-fifth the LCso
value (10 ppm) (the hazard criterion which theoretically approaches
the no-effect-level (0.1-10% depending on the dose-response data):
see FR 40(129): 28261; July 3, 1975) is compared with the 10-12
ppm ai maximum expected residue. This approach, of course, is
based on the assumption that all, or 100%, of the quail's diet is
seeds and the total diet is contaminated with the pesticide in
question.
48
-------
-44-
Table 9. Avian Species Sensitivity Profile To A Hypothetical
Pesticide ^/
Pesticide 10G (10% ai)/15G (15% ai)
Hazard to Seven Species of Non-Target Birds
Number of Granules Equal to
Species
Bobwhite (adult)
Bobwhite (14-day)
Robin
Mourning Dove
House Sparrow
Redwing Blackbird
Body
Weight
(g)
200
30
80
100
20
50
Grasshopper Sparrow 13.9
Attwater's £/
Prairie Chicken
(adult)
Prairie Chicken
(14-day)
1000
50
Animal
(g)3/
0.40
0.06
0.16
0.20
0.04
0.10
0.027
2.00
0.10
15G
4.3
0.6
1.7
2.2
0.4
1.1
0.3
21.5
7.1
>0— /
10G
6.4
1.0
2.6
3.2
0.6
1.6
0.4
32.3
1.6
l/5th
15G
0.9
0.1
0.3
0.4
0.1
0.2
0.1
4.3
0.2
LD50_<
10G
1.3
0.2
0.5
0.6
0.1
0 3
0.1
6.5
0.3
Utilizing Bobwhite Quail LDjQ of 2 mg/kg (15G) or 0.3 mg/kg
(converted to ai)
Weight of one 15G granule = 0.093 mg
Weight of one 10G granule = estimated to be same as 15G granule
Weight of pesticide in one granule:
= 0.093 x 15% = 0.0139 mg ai/granule.
= 0.093 x 10% = 0.0093 mg ai/granule.
_' Bobwhite quail LDrQ = 2 mg/kg (15G) = 0.3 mg/kg ai
2 mg/kg (15G) x .2 kg = body wt. = .4 mg/animal (15G)
(LDso for 15G to bobwhite quail on per body wt. basis)
All other values in this column based on the assumption that
each organism has the same sensitivity as the bobwhite
quail (i.e., LDso f°r each organism is 2 mg/kg (15G))
Or
Number of 15G granules
required to equal
0.4 mg/animal
0.093 granule wt.
=4.3
LD5Q = 2 mg/kg (15G) x 15% =0.3 mg/kg ai
.2 kg x 2 mg/kg =0.4 mg/animal (15G)
.2 kg x 0.3 mg/kg ai = 0.06 mg/animal ai
0.06 mg/animal ai
15% = 0.06 mg/animal ai =4.3 granules
• .093 mg granule wt. (15%) (.093)
I/. l/5th LDso = restricted use classification trigger (see Table 1)
Weight data obtained via personal communication with Wayne Shifflet,
Refuge Manager, Attwater's Prairie Chicken Refuge, Aransas, TX
49
-------
-45-
Another approach utilizing the avian LCso data considers
species sensitivity and food factors. This approach develops
the potential acute avian risks for various species via dietary
exposure and incorporates the food consumption habits (based on
summer diets) of each species. In Table 10 the theoretical
values of tliree" spe"cfes""are"'calcuTate"d from the LCso value for
the adult bobwhite quail. These values are developed based upon
the assumption that each species has 'the same sensitivity to a
hypothetical pesticide as the adult bobwhite quail (i.e., the
daily intake which will produce 50% mortality is 2.68 mg/kg/day),
This approach is similar to that used in human risk assess-
ments (see Attachment K) and the one proposed by Kenaga (1973)
for birds. For use with avian species Kenaga (1973) indicated
that:
(a) The classical approach in animal toxicological
research is to determine the average daily intake
of a toxicant which can be consumed by an organism
without adverse effect(s). This quantity of
toxicant consumed is commonly expressed as
"milligrams of toxicant consumed per kilogram of
body weight of the organism per day" or mg/kg/day.
(b) The mg/kg/day data (from the dietary
study) for a particular avian species can be
correlated with the feed consumption/body
weight data of another avian species to
predict a safe dietary (or residue level) of
the pesticide to that species. And after
these correlations are made, then residue
levels in the field (i.e., in or on feed
items) can be compared to the safe levels
determined for each species.
EEB uses this approach to address the potential acute hazards
of a pesticide use to those avian species likely to be exposed
(by typically extrapolating from the bobwhite quail). However,
in utilizing the mg/kg/day data, as proposed in (b) above and
as shown in Attachment K (where a NOEL dietary level for humans
— 0.1 mg/kg/day or 6 mg/person/day or 4 ppm — is determined
from a rat NOEL dietary level 10 mg/kg/day or 4 mg/animal/day
or 200 ppm), EEB chose to predict a toxic (or theoretical LCso)
dietary level of a pesticide for avian species rather than a
"safe" NEL or NOELs. Table 10 presents this approach, and as
indicated above, the initial assumption is that the species
considered will have the same sensitivity to a pesticide via
dietary exposure as bobwhite quail.
50
-------
-46-
!
I
&J
b
•8
•B
ro
CD
•D
1-1
£
*
&
i-H
ro
^
4->
%
<
B
&
(0
£
o
a
CO
to ro
O
O
in
-
S1
&.
s
Species^/
vo
00
VO
CN
00
vo
tN
o
o
o
ro
ro
Oi
ro
00
O
PS
Og
*O ro CD
ro T-I
lJ 00
«0 ro
in
ID -^
to *-»
3 CO
io r-
o a\
ro
•s
§
4J
ro
1
*8
fi-a
CD CO
*
SB03
i-H »J T3
ro u CD
> ro jj
OP -H
CD CO
S2§
~-.P o
4J e co
.c &-^
iH O fl)
flj . 3
5 O\ r-l
•H ro
* CO-
Vj IH oo
ro
to *~
SQ > ro Z
o c10. g
O O vo D
•H la
•O 4-> D)M-I
to ro o>
CD £ 3
o gj j= p &o
•H 3 4J O
-------
-47-
Theoretical (or calculated) LCso values are then developed for the
other species by use of Lehman's (1959) relationship for ppm and
mg/kg/day•£/ As can be seen, this procedure is the same recommended
by Kenaga (1973) and used by toxicologists for human risk assess-
ments. The only difference is that effect levels (LCso values)
are utilized in EEB's approach whereas NOELs, NELs, or safe levels
are used in the other.
Once species sensitivity is established then the next step/
using appropriate food factors, is to correlate actual or expected
residues with the theoretical LCsos. Table 11 presents the food
factor calculations and the correlation of total adjusted residues
with the theoretical LCso values. This approach may appear cum-
bersome, but it does emphasize that, even from one application
rate, different exposures and different potential hazards exist
for organisms with different body weights, food consumption rates,
and/or feeding patterns. A case in point is the LCso and total
adjusted residue values (see Tables 10 and 11) for the Carolina
wren. Because of its relatively large food consumption in relation
to body size, the theoretical LCso value calculated is 7.80 ppm,
a value approximately one-fourth the LCso value of 30.00 ppm for
the reference organism, bobwhite quail. Further, the relative
hazard from residues in or on food items appears much greater,
primarily due to the wren's large consumption of small insects.
The bobwhite quail, however, has a lower dietary residue due to
its greater consumption of plant matter.
Relative to the above discussion on species sensitivity, it
is recognized that the approach is a theoretical one — one used
to estimate potential acute hazards to non-target avian species.
It is considered a logical one, however, for a one-to-one sensiti-
vity ratio is being utilized between organisms of similar size and
food habits and with similar chances for exposure to the pesticide.
Some of the avian research done to date indicates that all possible
variations in species and/or age sensitivity occur from exposure
to the same or similar pesticides (Friend and Trainer, 1974; Hudson,
et al., 1972? Tucker and Haegele, 1971; Schafer, 1972). Other
avian research, however, indicates that although such variations
in species sensitivity to chemicals can occur, smaller species are
generally more sensitive to toxicants than larger species (Hill,
et al., 1975).£/ These latter findings are supported further by
the research of Hill (1971). Hill examined the toxicity of four
See Table 10, footnote 3.
Note that this research represents approximately ten years of
testing with an examination of greater than 130 compounds, of
which 39 were organophosphates.
52
-------
-48-
*
•rH
O
**H
4->
I
rH
S
-H
4->
JB
4J
QJ
E
CO
U
£
I
4J
CO
1
8
Avian Species Aoxicity/Residue
•
rH
rH
3
-9
3
rH
CO CO
1 1
•u c
•iH rf
10\ 4.
,®in -J-.C f
K •-? jJ n
E Q «~
•H ate o
CO QJ
4J N- Jj T
p o c
H fc «
K,*
OW rH
3 O-.
•r~\ CO
*D Q)
«D
E -H
•H OJ E
S "i
1 5
"ft £
t&tf
0) ' Gi O>
r-' o ^*
*—S •^i
^H
& Eg
t^ o vo
• » c
in o eft
rH fSj
n 1- &
Ml Qi Q,
O O O
0 c? 0
-H rH rH
& i I,
a a a
27% 73% 58.0
Beetles, Seeds:
Weevils, Ragweed,
Grasshoppers, Lespedeza,
etc. Corn, etc.
0% 100% 0.0
Seeds: Corn,
Pigweed, etc.
51% 49% 58.0
Beetles, Seeds:
Grass- Crabgrass,
hoppers, Bristle-
8 R g
O n _
m CM
o a>
JJ C >
•r-l -H g
J-H C TJ C
"« ii ft S C
co 3 5 fl) CB
£<§ Ifi £§
CO •
1 •-• CM no
|^
CX
in
*
*r**
in
j|
rH
•
O
E
^r
•
f>**.
I ^
in
I
o
.
o
rH
S
o •
• o
, 00 4J
i " <»
•H . ^ ^
CO >, to
m JJ &*>
IX 0) M CO
•• * .. C
CO CO CO O *•
CO CO TJ 0) 0)
CO CO dP CU -i-l c
& & CO ft AJ
CO •
10 -2 4J
CO rH m
-U )J O\ jj rH "O
CO CO C -H Co
O rH < S (5j
0
00
•
r^
•H
eS
JM
fl3 S
o
•
v4t
^^
••
9
§
M
CD
iM
m
S£ §-E?|
i-i "O O ft)
a) co ffi a: &2
M-l -r-l CO
to jr co • .
OS P CO co J3
.7* 1??' L^"* 1
*-» N H 1
•
j i g
g
Is
U-i
II
= i
co a
rH D.
ao
•g-
«0 X
rH — ~
m c*jp
c O
-«H O
C rH
CO «-'
'gg
•
r— 1 rH
CO
JJ ••
S.
4>
.
CO >H
O T3
& B
• o M
TJ rH C
$ co O
3 4J
o a a Q.4J
*r^
fe^^feT,
*J in r- o
•*-* rH 1— 1 O
II, «|^
•U— — C-H
C dP dP co
co r^ co 73
•H «N r- J>
Q,^ — • «^ .. JJ
\ 4J 0
rH P» m Q) m
CO CN t^ -H Q,
E • • *D X
•3 0 0 8
C ft)
CO X X £ rH
-PEE^S
, Q n c o
•y a a-H *j
0) oo o -n P
rH in rH Q) G,
2 o°-
« 8-S
o •• x
i! rH S
•H ..
•p co co a)
Us SB
CO 'n Q
30) -S
53 8?
"^ CO £ 14 iH
£> * C
r^ co 45 rd i-i
en a) o co 3
rH 3 CQ JJ 0
^- rH Q S
CO 43
(0 > ..
Q) a> co ..
CO ft) rH JJ ft)
C 3 & O rH
ft) T3 E Co QJ
UJ -H «3 rH &
CO X <4-l !B
•
-------
-49-
mosquito larvicides to blue jays, house sparrows, cardinals, wild
bobwhites, and to farm-reared bobwhites in dietary LCso studies.
His results suggest that the larger species were more tolerant to
the pesticides than the smaller ones.2/
Further, the farm-reared bobwhites were not only heavier than
wild-captured ones but also were more tolerant to several of the
pesticides because of their larger size. Therefore, relative to
these two bodies of research, it can be seen that a one-to-one
sensitivity ratio approach is not biased one way or the other,
but gives EEB an opportunity to observe on paper the potential
risks of different dietary residues to different species of birds.
Other avian data, such as simulated and actual field testing
and special studies to assess acute avian risks, are also used.
Simulated and actual field testing or monitoring, plus available
field mortality data, are routinely utilized by EEB. In these
situations, the known rates, available (or estimated) residue data,
and observed effects (toxicological symptoms, as well as mortality)
are all integrated. Usually, for the simulated field testing,
brain acetylcholinesterase data are available, and these provide
another important clue as to whether exposure actually occurred.
As for special studies such as acute dermal LDso or inhalation
tests, few have been done or used by EEB in its risk assessments.
Dermal and inhalation risks have been tentatively linked to cer-
tain carbamates (fenthion and fenitrothion, respectively), and,
of course, avicides are well known for their dermal effects to
birds.
In those cases where acute or chronic avian data are lacking,
then EEB is usually forced to utilize any other available informa-
tion and, particularly, data for mammals. In doing so, the ap-
proaches outlined above for mammals are typically followed.
(2) Subchronic/Chronic Risks
Typically, the assessment of subchronic and/or chronic avian
hazards involves the correlation of NELs and effect levels found
in the avian reproduction studies with actual or expected residue
levels. This process requires a close examination of the avian
reproductive data to see which reproductive parameters were most
affected and whether such effects are likely in the field. It is
a difficult assessment simply because of the number of variables
found not only in the studies themselves but also in the field
I/ Only blue jays were the exception to the size-toxicity
relationship, being more sensitive than the other species
in almost all of the studies performed.
54
-------
-50-
conditions under which the pesticide will be used. The present
laboratory study addresses the effects of chronic low levels of
pesticides on eggs, embryos, and hatchlings. Difficulties have
arisen with the study's fairly common use in examining organophos-
phate and carbamate (or other acutely toxic) compounds which break
down rapid-ly-f--bwt--ate- repeatedly appliedand, therefore, pose a
subchronic or chronic hazard to birds. In these instances the main
reproductive hazard appears to be mortality to the breeding adults
or newly-hatched or young chicks rather than effects on eggs and
developing embryos. Also, the laboratory reproductive study has
been required more readily for persistent herbicides, but for her-
bicides which are minimally toxic, on an acute basis, to mammals
and birds. In these situations, EEB usually puts greater emphasis
on the chronic mammalian data and on requests to the registrants
for pertinent residue, including dissipation, data.
Even with the laboratory reproductive study's shortcomings,
they are requested and utilized in the review process. At this
point, this study is EEB's strongest reference for assessing
chronic avian risks. Field studies (large pen) and monitoring
are also utilized to a much lesser degree, and the variability in
these is even greater, making interpretation of results very dif-
ficult. However, the use of avian reproductive studies, whether
laboratory or field, is well established and a definite, formalized
approach is presented in the Agency's Pesticide Assessment Guide-
lines - Subdivision E.
Other studies or data utilized by EEB include tests where
birds' eggs were sprayed with pesticides. Usually these data are
lacking, but the available ones can show the percent hatching
success for treated eggs versus controls. However, the use of
mammalian data by EEB occurs readily since these data are usually
available and, possibly, are the best chronic data available for
use.
Probably the weakest point in the chronic avian review is
the residue information. As discussed under mammals, pertinent
residue, including field dissipation, data usually are lacking,
thus forcing EEB to extrapolate and estimate the residues likely
to be found over time. Under these conditions, Hoerger and
Kenaga's (1972) and Kenaga's (1973) data are of minimal value
since the residues presented in these articles primarily concern
those found immediately after application of the pesticide.
(3) Secondary Hazards
In assessing the secondary risks of pesticides to non-target
avian species EEB takes the same approach outlined earlier for
determining secondary risks to non-target mammals. Data require-
ments, primary risk, secondary risk, and acute and chronic toxicity
55
-------
-51-
concerns for avian species are similar to those for mammals, and
the full discussion presented earlier is applicable here. The
only major difference concerns those organisms likely to be ex-
posed and those chosen for use in the various toxicity and field
tests. EEB normally requires an upland gamebird and a waterfowl
species (as indicated earlier wider acute risks), but also requires
a variety of other avian species depending upon the proposed use
in question. Most often, raptors (hawks, eagles, owls, and vul-
tures) are used, particularly in secondary toxicity studies where
pesticide contaminated organisms (such as rodents) are fed to the
birds. However, free-ranging individuals are utilized in field
studies designed to address potential secondary risks under typi-
cal pesticide use conditions. Other species utilized in the
primary toxicity tests include: crowned guinea fowl, laughing
doves, pheasants, geese, horned larks, mourning doves, blackbirds,
magpies, pintails, chickens, chukars, sparrows, Gambel's quail,
and domestic turkeys (Tietjen, 1976; Atzert, 1971). Of course,
EEB requires that testing be done on representatives of wild
avian species likely to be exposed under typical use conditions.
With the toxicity data, both laboratory and field, developed
on appropriate avian species, the reviewer correlates these data
with expected environmental and body residue levels (in target and
non-target organisms) to assess the potential impacts. As discussed
in the mammals section, these reviews are highly complex and, in
most cases, the field effects data provide the major information
pertinent to the risk determination. However, these studies are
generated on a case-by-case basis, and to date EEB has received a
minimal number of adequate field studies for review. Consequently,
EEB is working towards standardization of testing methodologies
and secondary risk assessment procedures.
5. Risk Criteria
The above discussion on terrestrial hazard assessment proce-
dures did not include an application of risk, regulatory, or safety
criteria to the LD$Q or LCso data presented because these criteria
were discussed extensively in earlier sections. Tables 7 and 9
show the use of one of these criteria (l/5th the LDso or LC$Q),
but this criterion is only intended for use with non-endangered
wildlife. For endangered wildlife, l/5th the LDjn or LC^o (if a
slope value is available) or l/10th the LD§o or LCso (when no slope
value is present) are used (see Attachment H). Both of these cri-
teria, the nonendangered and endangered, are used by EEB, basically,
to determine NELs or levels at which minimal mortality is likely.
They were developed over the years to provide consistency and are
somewhat supported by LDso/LCso dose-mortality data. Another ap-
proach, which is essentially the same as that discussed, consists
of an examination of the raw dose-mortality data and the dose-mor-
tali.ty curve in order to calculate a minimal-effect level or NEL.
56
-------
-52-
This level can then be compared with the safety criteria (l/5th
LD50/LC5Q, l/5th LDiQ/LCiQ, or l/10th LD^Q/LC^Q ) used. Heathr
et al. , (1972) provide a formula for this technique but point out
that extrapolating from a probit regression line can produce erro-
neous results/ particularly, if there is some curvature to the
Tihe.^/ They indicate that specially designed studies, for exam-
ple, ones to determine LC^o8' are more appropriate. Also, it
should be noted that examination of the dosemortality curve along
with its correlation with actual or measured residues provides EEB
with a better understanding of potential hazards than the mere
use of risk criteria and LDso, LCso* LD^Q or LCio values.
6. Risk Assessment - Non-Target Insects
At present, risk assessment for non-target insects is limited
to honey bees. Sections on non-target aquatic insects and on
insect predators and parasites in Part 158 are reserved, pending
Agency decision as to whether these data requirements should be
established.
The first step in risk assessment is to determine whether
honey bees will be exposed to the pesticide as a result of the
proposed use(s). EEB examines all proposed use situations to
make this determination. Generally, bee exposure may occur in two
major use areas: foliar application to crops attractive to bees,
and adult mosquito control. Because the probability of exposure
can usually be determined, the number of uses which require bee
testing is narrowed considerably. For example, use of granular
formulations and preplant. soil applications does not usually
result in bees being exposed to the pesticide; thus, no testing
would be required in these cases.
If the proposed use will result in bee exposure, the Agency
requires data on acute toxicity to honey bees. If acute tests
show no or low toxicity, no further testing is normally required.
If acute tests show moderate or high toxicity, the Agency requires
data on the extent of residual toxicity of the pesticide to honey
bees.
On the basis of this information, in conjunction with any
other pertinent information, the Agency will do one of two things.
If there is any information which indicates properties of the
The formula presented by Heath, et al., (1972) is:
log LCR = (5 - probit K)/b - log LC5Q
with the antilog of log LC^ the result wanted.
57
-------
-53-
pesticide other than direct toxicity, and which indicates that the
pesticide might cause unique problems in honey bees, the Agency
may require field testing. This testing would be designed to apply
specifically to the potential problem. Normally, however, field
testing is not required. In the usual case, the Agency will use
the acute and residual toxicity data to determine appropriate bee
precaution statements to be placed on the product label.
VI. RISK ASSESSMENT - SHORTCOMINGS AND IMPROVEMENTS
There are a number of weaknesses in EEB's aquatic and terres-
trial risk assessments. As mentioned previously, the ratio or quo-
tient method for assessing risk (1) does not adequately account
for effects of incremental dosages, (2) does not compensate for
differences between laboratory test and field populations, (3) can-
not be used for estimating indirect effects of toxicants (e.g.,
food chain interactions), (4) has an unknown reliability, (5) does
not quantify uncertainties, and (6) does not adequately account
for other ecosystem effects (e.g., predator-prey relationships,
community me.tabolism, structural shifts). Further, we have no
terrestrial exposure model comparable to the aquatic exposure
models (e.g., SWRRB/EXAMS). There is no integrated air component
to the exposure models, and the simple spray drift models need
improvements. The current aquatic exposure models are still being
validated by EPA's Office of Research and Development (ORD). Their
utility is limited somewhat because they do not provide an estimate
of uncertainty of their results. Finally, the risk criteria cur-
rently being used by EEB need a stronger empirical data base for
support. The data base should consist of laboratory and field
effects data.
For 1986 and beyond, EEB is planning to improve in-branch
EEC calculations and continue to develop a better understanding
of available EEC models. In addition, EEB has taken 3 steps to
improve the risk assessment as a whole, and the risk criteria in
particular. First, EEB and ORD have initiated an analysis of
pertinent in-house and published acute and chronic toxicity data
in order to modify the risk criteria. This analysis is based on
the relationship between dose/concentration of the pesticide and
the response of the test organisms.
Second, ORD has two major research projects that have been
designed with EEB's input, to improve the scientific basis for,
and the process of, conducting ecological risk assessments. These
two projects are titled: Field Validation and Ecological Risk
Assessment Research. Under the field validation project, three
field and simulated field studies are designed to determine whether
EEB's predictions of risk of pesticides to non-target organisms
based on our risk criteria are verified in actual field use situ-
58
-------
-54-
ations. The ecological risk assessment project is in the advanced
planning stage, and is scheduled to begin in 1986. It is being
viewed as a long-term research project (5-10 years). The assess-
ment methodology to be developed will contain many components in-
cluding data bases, models, and software to coordinate the compo-
nents. Tlvis integrated system will permit problem solving through
coordinated access to appropriate computational tools, data bases,
and presentation of results. Components that will be contained
in the risk system include:
0 sets of frequently used environmental scenarios, and a
capability to generate new scenarios from large data
bases;
0 data bases of species, populations, and ecosystems at
risk; of environments; of chemical parameters for exposure
and toxicity modeling; and of comparative toxicology for
prediction effects;
0 models of chemical fate to calculate expected residues
based on chemical concentration distribution and organism
behaviors;
0 computational tools for existing specific methods, such
as the ratio method;
0 several models that compute effects based on susceptibility
and exposure, to be used to predict consequences of expo-
sure to toxicants:
- a steady-state model to calculate ultimate effects of
long-term exposure;
- population-specific models that calculate the effects
on populations of particular scenarios of chemical
loadings on their environment;
models for interacting subsets of biotic communities
(predator-prey associations);
- large ecosystem models that include representations
of all major ecological processes for a selected
scenario; and
0 a probability analysis that estimates probabilities asso-
ciated with result in two forms: (1) calculated probabi-
lity values for portions of an analysis for which uncer-
tainties and inherent variation can be quantified and (2)
descriptions of the nature of additional uncertainties.
The risk assessment research program described above will
provide improved capabilities to calculate risk using existing
methods, such as the quotient method or other techniques. The
59
-------
-55-
program proposed ultimately to develop integrated assessment pro-
cedures that will predict ecological consequences for many exposure
scenarios with an associated evaluation of uncertainty.
Finally, EEB is moving in the direction of testing and evalu-
ating ecosystem response in addition to individual responses of
surrogate species. Traditional single-species, clean environment
bioassays will still be required (e.g. , LCso in fish and LCso ^n
birds) because they provide valuable baseline information for com-
parative purposes. Working closely with EPA's ORD, the regulated
community, and academia, EEB will be developing test methods and
analysis schemes that measure ecosystem impact from an integrated,
rather than from a single-species viewpoint. When lower tier
testing indicates potential risk, small-scale or full-scale field
testing will be required of pesticides manufacturers to negate
the potential risk. These field tests will not only measure the
traditional end points (e.g., mortality), they will also assess
impact on populations of organisms and impacts on community
structure and function. These complex test requirements will
hopefully account for the basic resiliency of healthy ecosystems,
identify problems with stressed ecosystems, and help the Agency
better understand the ecological risk of the pesticide.
60
-------
-56-
ATTACHMENT A; Final Criteria for Initiation of Special Review
The Administrator may conduct a Special Review of a
pesticide use if he determines, based on a validated test or
other significant evidence, that the use of the pesticide
(taking into account the ingredients, impurities, metabolites,
and degradation products of the pesticide):
0 May result in residues in the environment of non-
target organisms at levels which equal or exceed
concentrations acutely or chronically toxic to such
organisms, or at levels which produce adverse
reproductive effects in such organisms, as determined
from tests conducted on representative species or
from other appropriate data.
0 May pose a risk to the continued existence of any
endangered or threatened species designated by the
Secretary of the Interior or the Secretary of Commerce
under the Endangered Species Act of 1973, as amended.
0 May result in the destruction or other adverse modifi-
cation of any habitat designated by the Secretary of
the Interior or the Secretary of Commerce under the
Endangered Species Act as a critical habitat for any
endangered or threatened species.
0 May otherwise pose a risk to humans or to the environ-
ment which is of sufficient magnitude to merit a
determination whether the use of the pesticide product
offers offsetting social, economic, and environmental
benefits that justify initial or continued registration.
Reference: 40 CFR 50 (229): § 154.7 (a)(3), (4), (5), and (6).
61
-------
- -.57-
ATTACHMENT B; Proposed Restricted Use Criteria for Hazard to
Non-Target Organisms
(c) (1) All Products. A pesticide product intended for outdoor use
will be considered for restricted use classification if:
(i) When used as according to label directions/ application
results in residues in the diet of exposed mammalian wild-
life/ immediately after application, such that:
(A) The level of residues equals or exceeds l/5th of the acute
dietary LCso; or
(B) The amount of pesticide consumed in one feeding day
(mg/kg/day) equals or exceeds l/5th of the mammalian acute
oral
(ii) When used according to label directions/ application
results/ immediately after application, in residues in the
diet of exposed birds at levels that equal or exceed l/5th
of the avian subacute dietary
(iii) When used according to label directions, application
results in residues in water that equal or exceed l/10th
of the acute LCso for non-target aquatic organisms likely
to be exposed; or
(iv) Under conditions of label use or widespread and commonly
recognized practice/ the pesticide may cause discernible
adverse effects on non-target organisms, such as significant
mortality or effects on the physiology, growth, population
levels or reproduction rates of such organisms, resulting
from direct or indirect exposure to the product ingredients
or residues.
(2) Granular Products. In addition to the criteria of (c)(l)
of this section, a pesticide intended for outdoor use and
formulated as a granular product will be considered for
restricted use classification if:
(i) The formulated product has an acute avian or mammalian oral
LDso of 50 mg/kg or less as determined by extrapolation from
tests conducted with technical material or directly with
the formulated product; and
(ii) It is intended to be applied in such a manner that signifi-
cant exposure to birds or mammals may occur.
t.
(d) Other Evidence. The Agency may also consider evidence such
as field studies, use history, accident data, monitoring
data, or other pertinent evidence in deciding whether the
product or use may pose a serious hazard to man or the
environment that can reasonably be mitigated by restriction
to use by certified applicators.
62
-------
-58-
ATTACHMENT C; Model Name; Simulator for Water Resources in
Rural Basins
Model Acronym; SWRRB
Type of Model; SWRRB is basically a hydrology model combined
with a pesticide runoff model.
Developed by; Agricultural Research Service, Temple TX
Contact; Robert F. Carsel, U.S. EPA Environmental Research
Laboratory, Athens, GA
Application; SWRRB is a hybrid pesticide runoff model to
predict the concentration of pesticide in the runoff and that
available for leaching. The model combines the original
Simulator for Water Resources in Rural Basins (SWRRB) model,
which is a hydrology model, with a pesticide quantification
runoff model. The model utilizes a Soil Conservation Service
(SCS) curve'number technique with actual daily rainfall data.
The model structure is based on the water balance equation,
evapotranspiration, percolation, return flow and pesticide
function which accounts for the loss of applied pesticide to
the atmosphere. The pesticide inputs include soil persistence
and partition constants. The model output data predicts daily
runoff volume and peak rate, sediment yield, evapotranspiration,
percolation, return flow and pesticide concentration in the
runoff. There are seventeen river basins data sets based on
actual field observations available in SWRRB to predict
.pesticides behavior.
Model Accessed Through; EPA National Computer Center (NCC)
Status; The SWRRB model has been evaluated by comparison with
actual pesticide runoff data from the fields in the basins and
with other runoff models. SWRRB predicts quantities of an
event very well while prediction of individual event occurrence,
as with similar models, is somewhat weak. This latter predic-
tion is not used in the determination of estimated pesticide
environmental concentrations.
User's Manual; A manual is available from ERL Athens, GA. It
contains both a user's manual and the hydrology information
for each basin.
63
-------
-59-
(ATTACHMENT C, CONTINUED)
LIST OF AVAILABLE RIVER BASINS
1. Coshocton, Ohio, 115
2. Coshocton, Ohio, 118
3. Coshocton, Ohio, 129
4. Coshocton, Ohio, 130
5. Riesel, Tx. M, SW-2
6. Riesel, Tx., 1, SW-2
7. Riesel., Tx., 2, Y-6
8. Riesel, Tx., 3, Y-8
9. Riesel, Tx., 4, SW-12
10. Tifton, Ga., Z
11. Vega, Tx., M-W-1
12. Watkinsville, Ga., P-l
13. Watkinsville, Ga., M
14. Watkinsville, Ga., P, P-l
15. Watkinsville, Ga., 2, P-2
16. Yazoo, Mississippi
-------
-60-
ATTACHMENT D; Model Name; Exposure Analysis Modeling System
Mode1 Acronym t EXAMS
Type of Model; Hydrologic model to predict "steady-state" or
"pulse-load" behavior of organic toxicants in aquatic ecosystems.
Developed by: U.S. EPA Environmental Research Laboratory,
Athens, Georgia
Contact; Dr. Lawrence Burns, EPA ERL, Athens, GA
Applicationst EXAMS couples the fundamental characteristics of
the environment to the physical and chemical properties of the
toxicant, using process models (mathematical relationships)
appropriate to each aspect of chemical behavior considered by
the model. These include an equilibrium partitioning of the
chemical into (up to 5) ionic species, each of which may occur
as a dissolved, sediment-sorbed, or biosorbed molecule, and a
kinetic treatment of volatilization, transport, direct photo-
lysis, hydrolysis (specific acid, base, and neutral), oxidation,
and bacterial degradation in the water column and bottom
sediments of the ecosystem.
Four general impact environments are available: pond,
river, and oligotrophic and eutrophic lakes. Other impact
environments may be generated where specific situations are to
be analyzed. The environmental parameters are those that have
an effect on the pesticide concentration (e.g., volume of the
environment and flow of the water) and on degradation (e.g.,
biomass and water chemistry).
Two versions of the model are available. The first version,
"steady-state," is used primarily for long-term, constant-input
with calculation of degradation and dissipation during steady-
state and upon cessation of input. The second version allows
for "pulse" loadings as may occur from field runoff during a
rain storm.
Status: This model in all versions is being evaluated.
User's Manual: A manual is available from EPA publications in
Cincinnati, OH, or NTIS, Springfield, VA. (EPA-600/3-82-023)
Model Accessed through: EPA National Computer Center (NCC)
Compiler Language - FORTRAN/TSO (also available through OTS VAX
computer at NCC and POP 1170 at Athens, GA).
65
-------
--61-
ATTACHMENT E; Output from Spray Drift Model
THIS IS A BALLISTIC MODEL WITHOUT DROPLET EVAPORATION.
1. CRITICAL LEVEL 0.10000 PPM
2. APPL. RATE 0.180 LB AI/ACRE
3. HEIGHT 10.00 FT
4. WIND SPEED 10.00 MPH
5. DROPLET TYPE NUMBER 2.0
DRIFT MAY RESULT IN WATER RESIDUES IN THE TOP SIX INCHES EQUAL
TO THE CRIT. LEVEL OF 0.100000 AT A DISTANCE OF 78. FEET OR
24. METERS.
1. CRITICAL LEVEL 0.10000 PPM
2. APPL. RATE 0.180 LB AI/ACRE
3. HEIGHT 10.00 FT
4. WIND SPEED 10.00 MPH
5. DROPLET TYPE NUMBER 3.0
LARGE DROPS CAUSING MINIMAL DRIFT.
1. CRITICAL LEVEL 0.10000 PPM
2. APPL. RATE 0.180 LB AI/ACRE
3. HEIGHT 10.00 FT
4. WIND SPEED 10.00 MPH
5. DROPLET TYPE NUMBER 1.0
DRIFT MAY RESULT IN WATER RESIDUES IN THE TOP SIX INCHES EOUAL
TO THE CRIT. LEVEL OF 0.100000 AT A DISTANCE OF 123. FEET OR
38. METERS.
1. CRITICAL LEVEL 0.01000 PPM
2. APPL. RATE 0.180 LB AI/ACRE
3. HEIGHT 10.00 FT
4. WIND SPEED 10.00 MPH
5. DROPLET TYPE NUMBER 2.0
DROPLETS EVAPORATED BEFORE REACHING THE GROUND. DUE TO DROPLET
EVAPORATION, A LIKELY MAXIMUM DRIFT DISTANCE OF 244. FEET OR
75. METERS IS POSSIBLE YIELDING A CONCENTRATION OF 0.028800 PPM
IN 6 IN. OF WATER.
1. CRITICAL LEVEL 0.00100 PPM
2. APPL. RATE 0.180 LB AI/ACRE
3. HEIGHT 10.00 FT
4. WIND SPEED 10.00 MPH
5. DROPLET TYPE NUMBER 2.0
-------
-62-
(ATTACHMENT E, CONTINUED)
DROPLETS EVAPORATED BEFORE REACHING THE GROUND. DUE TO DROPLET
EVAPORATION, A LIKELY MAXIMUM DRIFT DISTANCE OF 244. FEET OR
75. METERS IS POSSIBLE YIELDING A CONCENTRATION OF 0.028800 PPM
IN 6 IN. OF WATER.
1. CRITICAL LEVEL 0.03000 PPM
2. APPL. RATE 0.180 LB AI/ACRE
3. HEIGHT 10.00 FT
4. WIND SPEED 10.00 MPH
5. DROPLET TYPE NUMBER 2.0
DRIFT MAY RESULT IN WATER RESIDUES IN THE TOP SIX INCHES EOUAL TO
THE CRIT. LEVEL OF 0.030000 AT A DISTANCE OF 187. FEET OR 57.
METERS.
1. CRITICAL LEVEL 0.05000 PPM
2. APPL. RAT'E o.iso LB AI/ACRE
3. HEIGHT .10.00 FT
4. WIND SPEED 10.00 MPH
5. DROPLET TYPE NUMBER 2.0
DRIFT MAY RESULT IN WATER RESIDUES IN THE TOP SIX INCHES EOUAL TO
THE CRIT. LEVEL OF 0.050000 AT A DISTANCE OF 125. FEET OR 38.
METERS.
1. CRITICAL LEVEL 0.05000 PPM
2. APPL. RATE 0.180 LB AI/ACRE
3. HEIGHT 10.00 FT
4. WIND SPEED 10.00 MPH
5. DROPLET TYPE NUMBER 1.0
DROPLETS EVAPORATED BEFORE REACHING THE GROUND. DUE TO DROPLET
EVAPORATION, A LIKELY MAXIMUM DRIFT DISTANCE OF 244. FEET OR
75. METERS IS POSSIBLE YIELDING A CONCENTRATION OF 0.090000 PPM
IN 6 IN. OF WATER.
67
-------
-63-
ATTACHMENT F; An Example of a More Sophisticated EEC Using
State-of-the-Art Models
IA Purposes
To calculate an aquatic EEC for the pesticide for its
proposed new use in field and sweet corn.
II Directions for Use: Corn
III Data Discussion:
There were no studies to be reviewed. This is an analysis
using existing data. Both a runoff scenario and two
aquatic habitats will be examined.
Runoff:
Watersheds in three river basins (Tifton, GA; Yazoo, MS ;
Coshocton, OH) were chosen to determine the possible
quantity of the pesticide in runoff as a function of
meteorology and geography. The model river basins are
part of the Simulator for Water Resources in Rural Basins
(SWRRB).
Four applications at the time of silking were determined
to occur about 1 July in Ohio and at anytime from early
June to late August in Georgia and Mississippi. Applica-
tions were made on a 1- to 8-day schedule as per the label
directions.
A K^ value of 1 was used even though the value has been
reported to vary from 0.58 to 1.34 depending upon the clay
content of the soil.
With each application that was followed by a storm within
1 to 5 days, significant pesticide runoff occurred.
Quantities reached upwards of 0.100 Ib/Acre with values of
0.010 to 0.050 fairly common. Two or three events of this
magnitude occur each year and then no more pesticide
runoff is predicted to occur.
Water Quality Analysis:
From the SWRRB data, a maximum of two runoff quantities 5
days apart (to account for 2 runoff events) were entered
into the EXAMS (Exposure Analysis Modeling System) using
both the Athens ERL Pond and River scenarios. The
chemistry data for the pesticide used in the model is
given in Table 2.
-------
.-64-
(ATTACHMENT Ff CONTINUED)
The first input quantity of 0.00001 kg is to provide a
better graphic plot of later results. Other quantities
of 0.010 and 0.100 and 0.005 and 0.050 kg were those
derived from the SWRRB data. The quantities were not
adjusted to reflect large areas. (At present the effect
of large field runoff is being studied with respect to the
quantity of material that could enter an aquatic system.)
If large field quantities are desired, multiply the outputs
from the EXAMS model by that desired field size - the
results are linear with respect to the pesticide inputs.
It should be noted that the effective pesticide runoff
quantity from a 100 acre field may only be equal to that
released from a 5 to 10 acre field.
The maximum quantity of material that was predicted to
occur .i-n the Athens ERL model pond was 5 ppb dissolved in
the water when .100 kg was introduced into the system.
That quantity found sorbed onto suspended particles was
about 10 ppb (mg/kg dry weight of suspended material).
The pesticide has a calculated half-life of about 15 days
for both dissolved and sorbed suspended material.
In the Athens ERL river model the pesticide does not
exceed 5 ppb at the point of input and dissipates rapidly
to less than 1 ppb by the time it reaches the third water
compartment some 2 km downstream.
IV Conclusions:
The expected environmental concentration in an aquatic
system should be no more than 5 ppb when the runoff input
is 0.100 kg with no other inputs of this quantity for
several weeks.
69
-------
-65-
(ATTACHMENT F. CONTINUED)
Common Name:
Chemical Name:_
Structure:
. No.
Chemical Properties:
Molecular Weight:
365
Solubility -(PETC): 25
Partitioning:
ROW . . 45 _
( 25 °C)
Clay Loam 1.34
Loamy Sand .58
Sandy Loam 1.22
Hydrolysis (halflife hrs.)
long term
(pH J_) 206 hr (pH _7 _ ) study hr (pH _9 _ ) __ 22
M50 days)
hr
.
3.36x10^ _ K 2xl(£i/hr K 3.16xl(£Vhr
Photolysis (halflife hrs.)
_ 194 hrs ________ Kfl
Degradation (halflife hrs.)
Soil (Aerobic) (pH 6-8 ) _ 72-120
Soil (Anaerobic) (pH
)
8.6 x IQll/day (3.5 x 10"4/hr)
hr K 9.6xlQll to 5.7xl(P/nr
hr K
Water (Type Pond Sterile pH
Bac ter iolog ical
Soil (Type ___________ )
t
Water (Type __ )
) 144-400 hr K 4.6xlO"l to 1.7xlO"3/hr
hr K
hr K
Vapor Pressure- 4.3X10Z5. nroHq at 25 -*C
Evaporation- _ _
~
70
-------
-66-
(ATTACHMENT F, CONTINUED)
Table 1: SWRRB Input
K^ (sorption coefficient) * 1.00
Washoff fraction » 10%
Half life on foliage * 2 days
Degradation Rate in soil * 1.7xlCTl /day
Application Efficiency = 75%
Table 2:
EXAMS — Exposure Analysis Modeling System — V2.0: Mode 2
Chemical:
TABLE 1.1. SH2, (NEUIKAL MDLEOJLE, SPECIES II) INPUT EftTA.
MWI* 365 ' SOL « 25.00 YAPR= 4.3000E-05 HENRY= 0.0000
KPS= 1.200 KPB « 0.0000 KOC « 0.0000 KDW - 45.00
KAH1« 3.3600E-06 EAH1= 0.0000 KNH1- 2.0000E-04 ENH1= 0.0000
KBH1« 3.1600E-07 EBH1= 0.0000 KOK1- 0.0000 ECK1« 0.0000
KBACW2= 0.0000 QIW2= 0.0000 KBACS2* 7.0000E-10 QTS2* 0.0000
KDP= 3.5000E-04 RFIAT« 40.00 LAMAX= 0.00
71
-------
-67-
(ATTACHMENT F, CONTINUED)
Table 3. Runoff quantities by Julian date for the three river basins.
Quantities are expressed in Ib ai/acre.
Yazoo MS
1971
180
187-
194
197
201
206
209
210
1972 •
180
182
185
186
187
195
202
1973
180
181
185
187
188
195
202
210
1.00
1.00
1.00
1.00
1.00
.'•'
1.00
1.00
1.00
1.00
1.00
1.00
1.00
Tifton GA
.020
.002
.006
.001
.004
.035
.036
.008
.065
.024
.007
.069
.002
1971
180
183
185
187
194
201
1972
180
187
195
198
202
206
1973
180
187
194
199
202
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
Coshocton OH
.001
-.001
.003
.001
.001
1968
180
187
194
201
206
1969
180
186
187
188
195
201
202
208
1970
180
187
189
194
202
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
.001
.056
.094
.003
.009
.011
72
-------
(ATTACHMENT F, CONTINUED)
-68-
System: POND, AERL...DEVELOPMENT PHASE TEST DEFINITION
Chemical:
Tine
(days)
Initial
0
1
2
3
4
5
Runoff
5
6
7
8
9
10
Runoff
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
40
45
50
Water
Average
Dissolved
Column
Average
Sorbed
(mg/1) (ng/kg)
input 0.000001
5.000E-08
4.812E-08
4.631E-08
4.458E-08
4.291E-08
4.130E-08
input 0.010
5.000E-04
4.B12E-D4
4.632E-04
4.458E-04
4.291E-04
4.131E-04
input '0. 100
5.413E-03
5.210E-03
5.014E-03
4.826E-03
4.646E-03
4.472E-03
4.305E-03
4.144E-03
3.990E-03
3.841E-03
3.698E-03
3.560E-03
3.428E-03
3.301E-03
3.1786-03
3.061E-03
2.947E-03
2.838E-03
2.733E-03
2.632E-03
2.535E-03
2.442E-03
2.352E-03
2.265E-03
2.1B2E-03
2.102E-03
1.743E-03
1.447E-03
1.202E-03
9.
8.
8.
8.
7.
7.
kg
9.
8.
8.
8.
7.
7.
kg
9.
9.
9.
8.
8.
8.
7.
7.
7.
7.
6.
6.
6.
6.
5.
5.
5.
5.
5.
4.
.
.
.
.
.
3.
3.
2.
2.
kg
224E-08
878E-08
545E-08
225E-08
917E-08
621E-08
225E-04
879E-04
546E-04
225E-04
917E-04
621E-04
987E-03
612E-03
251E-03
905E-03
571E-03
251E-03
943E-03
646E-03
361E-03
087E-03
B23E-03
569E-03
325E-03
090E-03
864E-03
647E-03
438E-03
236E-03
043E-03
857E-03
677E-03
505E-03
339E-03
179E-03
025E-03
877E-03
217E-03
670E-03
219E-03
Benthic
Total
Mass
(kg)
. . •
1.000E-06
9.624E-07
9.263E-07
8.916E-07
8.5B2E-07
8.261E-07
1.000E-02
9.625E-03
9.264E-03
8.917E-03
8.583E-03
8.262E-03
1.0B3E-01
1.042E-01
1.003E-01
9.653E-02
9.292E-02
8.944E-02
8.610E-02
8.289E-02
7.980E-02
7.682E-02
7.396E-02
7.121E-02
6.857E-02
6.602E-02
6.357E-02
6.122E-02
5.895E-02
5.677E-02
5.467E-02
5.265E-02
5.071E-02
4.884E-02
4.704E-02
4.530E-02
4.364E-02
4.203E-02
3.487E-02
2.B95E-02
2.405E-02
Average
Dissolved
(ag/1)
0
6
1
1
2
2
2
6
1
1
2
2
2
9
1
2
2
3
3
3
4
4
4
5
5
5
5
5
6
6
6
6
6
6
6
6
6
6
6
5
5
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
*
•
•
•
•
•
*
•
•
•
*
•
•
OOOE-01
391E-10
226E-09
763E-09
254E-09
702E-09
702E-09
395E-06
226E-05
763E-05
254E-05
702E-05
702E-05
50 IE-OS
574E-04
144E-04
665E-04
140E-04
571E--04
962E-04
314E-04
632E-04
915E-04
168E-04
392E-04
589E-04
761E-04
910E-04
037E-04
144E-04
232E-04
303E-04
357E-04
397E-04
423E-04
437E-04
438E-04
429E-04
252E-04
918E-04
492E-04
Average
Sorbed
(ngAg)
0.
1.
2.
3.
4.
4.
4.
1.
2.
3.
4.
4.
4.
1.
2.
3.
4.
5.
6.
7.
7.
8.
9.
9.
9.
.
.
.
.
.
.
.
.
.
.
•
1.
1.
1.
1.
1.
OOOE-01
179E-09
261E-09
253E-09
159E-09
985E-09
985E-09
180E-05
262E-05
253E-05
159E-05
986E-05
986E-05
753E-04
903E-04
956E-04
917E-04
793E-04
589E-04
310E-04
960E-04
545E-04
069E-04
536E-04
949E-04
031E-03
063E-03
090E-03
114E-03
134E-03
150E-03
163E-03
173E-03
180E-03
185E-03
188E-03
188E-03
186E-03
154E-03
092E-03
013E-03
Total
Mass
(kg)
0
9
1
2
3
4
4
9
1
2
3
4
4
1
2
3
3
4
5
5
6
6
7
7
8
8
8
8
9
9
9
9
9
9
9
9
9
9
9
8
8
.OOOE-01
.559E-10
.833E-09
.637E-09
.371E-09
.041E-09
.041E-09
.563E-06
.B34E-05
.637E-05
.372E-05
.041E-05
.041E-05
.421E-04
.353E-04
.207E-04
.986E-04
.696E-04
.341E-04
.925E-04
.452E-04
.927E-04
.351E-04
.729E-04
.064E-04
.359E-04
.616E-04
.839E-04
.029E-04
. 188E-04
.320E-04
.426E-04
.508E-04
.567E-04
.606E-04
.626E-04
.629E-04
.615E-04
.350E-04
.850E-04
.213E-04
73
-------
(ATTACHMENT F, CONTINUED)
-69-
System: POND, AERL DEVELOPMENT PHASE TEST DEFINITION
Chemical:
Time
Water
Average
Dissolved
(days)
(mg/1)
Column
Average
Sorbed
(mg/kg)
Initial input 0.000001
0 . 5.000E-OB
1
2
3
4
5
.B12E-08
.631E-08
.458E-08
.291E-08
.130E-OB
Runoff Input 0.005
5 2.500E-04
6 2.406E-04
7 2.316E-04
8 2.229E-04
9 2.146E-04
10 2.066E-04
Runoff Input D.050
9.
8.
8.
8.
7.
7.
kg
4.
4.
4.
4.
3.
3.
kg
10 2.706E-03
11 2.605E-03
12 2.507E-03
13 2.413E-03
14 2.323E-03
15 2.236E-03
16 2.152E-03
17 2.072E-03
IB
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
.995E-03
.921E-03
.849E-03
.780E-03
.714E-03
.650E-03
.5B9E-03
.530E-03
.474E-03
.419E-03
.367E-03
.316E-03
.268E-03
.221E-03
.176E-03
.133E-03
34 1.091E-03
35 , 1.051E-03
40 8.717E-04
45 7.237E-04
50 6.012E-04
3.
3.
3.
3.
3.
3.
3.
3.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
1.
1.
1.
1.
kg
224E-08
878E-08
545E-08
225E-OB
917E-08
62 IE-OB
613E-04
440E-04
273E-04
113E-04
959E-04
811E-04
993E-03
806E-03
626E-03
452E-03
286E-03
125E-03
971E-03
823E-03
680E-03
543E-03
411E-03
285E-03
162E-03
045E-03
932E-03
823E-03
719E-03
61BE-03
521E-03
428E-03
339E-03
252E-03
169E-03
090E-03
013E-03
939E-03
608E-03
335E-03
109E-03
Benthic
Total
Mass
(kg)
1.
9.
9.
8.
8.
8.
5.
*
.
. .
.
*
5.
5.
5.
4.
4.
4.
4.
4.
3.
3.
3.
3.
3.
3.
3.
3.
2.
2.
2.
2.
2.
2.
2.
2.
2.
2.
1.
1.
1.
OOOE-06
624E-07
263E-07
916E-07
5B2E-07
261E-07
001E-03
813E-03
632E-03
459E-03
292E-03
131E-03
413E-02
210E-02
015E-02
827E-02
646E-02
472E-02
305E-02
144E-02
990E-02
841E-02
698E-02
561E-02
428E-02
301E-02
179E-02
061E-02
947E-02
838E-02
733E-02
632E-02
535E-02
442E-02
352E-02
265E-02
182E-02
102E-02
744E-02
447E-02
203E-02
Average
Dissolved
(mg/1)
0.
6.
1.
1.
2.
2.
2.
3.
6.
8.
1.
1.
1.
4.
7.
.
.
.
.
.
2.
2.
2.
2.
2.
2.
2.
2.
3.
3.
3.
3.
3.
3.
3.
3.
3.
3.
3.
2.
2.
OOOE-01
391E-10
226E-09
763E-09
254E-09
702E-09
702E-09
199E-06
132E-06
819E-06
127E-05
351E-05
351E-05
751E-05
868E-05
072E-04
333E-04
570E-04
7B6E-04
981E-04
157E-04
316E-04
458E-04
584E-04
696E-04
795E-04
881E-04
955E-04
019E-04
072E-04
116E-04
151E-04
179E-04
199E-04
212E-04
218E-04
219E-04
215E-04
126E-04
959E-04
746E-04
Average
Sorbed
(mg/kg)
0.
1.
2.
3.
4.
4.
4.
5.
1.
1.
2.
2.
2.
8.
1.
1.
2.
2.
3.
3.
3.
4.
4.
4.
4.
5.
5.
5.
5.
5.
5.
5.
5.
5.
5.
5.
5.
5.
5.
5.
5.
OOOE-01
179E-09
261E-09
253E-09
159E-09
985E-09
985E-09
902E-06
131E-05
627E-05
080E-05
493E-05
493E-05
765E-05
452E-04
97BE-04
459E-04
897E-04
295E-04
655E-04
980E-04
273E-04
535E-04
768E-04
975E-04
156E-04
315E-04
452E-04
569E-04
668E-04
749E-04
814E-04
865E-04
901E-04
925E-04
938E-04
939E-04
93 1E-04
768E-04
459E-04
066E-04
Total
Mass
(kg)
0
9
1
2
3
4
4
4
9
1
1
2
2
7
1
1
1
2
2
2
3
3
3
3
4
4
4
4
4
4
4
.OOOE-01
.559E-10
.833E-09
.637E-09
.371E-09
.041E-09
.041E-09
.784E-06
.171E-06
.319E-05
.686E-05
.02 IE-OS
.021E-05
.105E-05
. 177E-04
.603E-04
.993E-04
.348E-04
.671E-04
.963E-04
.226E-04
.463E-04
.676E-04
.865E-04
.032E-04
.180E-04
.308E-04
.419E-04
.514E-04
.594E-04
.660E-04
.713E-04
.754E-04
.7B4E-04
.B03E-04
.813E-04
.814E-04
.808E-04
.675E-04
.425E-04
. 106E-04
74
-------
-71-
(ATTACHMENT F, CONTINUED)
-*. * * '
System: POND, AERL DEVELOPMENT PHASE TEST DEFINITION
Chemical: '
Inputs: .000001 kg, day 0; .010 kg, day 5; .100 kg, day 10
0. 108 I B
I BB . B « Water Column
I BB Total Mass (kg)
I BB C » Benthic
I BBB Total Mass (kg)
7.217E-02 I BBB
I BB
I BBBB
I BBBB
I BBBB
3.609E-02 I BBBBB
I BBBBBBB
I BB
I
I BBBBBB
0.000 •ICCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC
0 5 10 15 20 25 30 35 40 45 50
Tine, Days
Compound:
Environment: POND, AERL DEVELOPMENT PHASE TEST DEFINITION
Inputs: .000001 kg, day 0; .005 kg, day 5; .050 kg, day 10
*
2.706E-03 I B
I BB B * Water Column Ave.
I BB Dissolved (mg/1)
I BB C « Benthic Ave. Dissolved
I BBB (mg/1)
1.804E-03 I BBB
I . BB
I BBBB
I BBBB
I * BBBB
9.021E-04 I BBBBB
I BBBBBBB
I BB
I CCCCCCCCCCCCCCCCCCCCCCCCCCCC
I BBBBBB CCCCCCCCCC
0.000 ICCCCCCCCCCCCC
+ ._.. + -.... + .-.. + -......+ -....-+-..... + .,_-. + ...... + .,....+.-....(.
0 5 10 15 20 25 30 35 40 45 50
Time, Days
76
-------
-70-
(ATTACHMENT F. CONTINUED) - .".
System: POND, AERL DEVELOPMENT PHASE TEST DEFINITION
Chemical:
Inputs: .000001 kg, day 0; .010kg, day 5; .100 kg, day 10
5.413E-03 I B
Z BB
I BB B •« Water Column Ave.
I BB Dissolved (mg/1)
X BBB C « Benthic Ave. Dissolved
3.609E-03 I BBB (mg/1)
I BB
I BBBB
I BBBB
I BBBB
1.804E-03 I BBBBB
I BBBBBBB
I BB
I . .CCCCCCCCCCCCCCCCCCCCCCCCCCCC
I BBBBBB CCCCCCCCCC
0.000 ICCCCCCCCCCCCC
+_i_-+_—.4..... +_—_+-_ _. +---...f—-.+--.—+-—.-+-.- _-. +
05 10 15 20 25 30 35 40 45 50
Tine, Days
System: POND, AERL DEVELOPMENT PHASE TEST DEFINITION
Chemical:
Inputs: .000001 kg, day 0; .010 kg, day 5; .100 kg, day 10
9.987E-03 I B B - Water Column Ave.
I BB Sorbed (rag/kg)
I BB C - Benthic Ave.
I BB Sorbed (rag/kg)
I BBB
6.658E-03 I BBB
I BB
I BBBB
I BBBB
I BBBB
3.329E-03 I BBBBB
I BBBBBBB
I BB
I CCCCCCCCCCCCCCCCCCCCCCCCCCCC
I BBBBBB CCCCCCCCCC
0.000 1C CCCCCCCCCCCC
+___„+_._„+__._+____+__—+__—+_—+.._.._+____+_„..-+
0 5 10 15 20 25 30 35 40 45 50
Tine, Days
75
-------
-72-
(ATTACHMENT F. CONTINUED)
,»« •'
': RIVER, AERL DEVELOPMENT PHASE TEST DEFINITION
CHEMICAL:
Dissolved
Hater Benthic
Ave.(?) Ave.(?)
day (ing/1) (mg/1)
Initial'Input
0 1.111E-09 0
1 3.110E-07-1
2 3.053E-07-1
3 1.167E-07-5
4 2.132E-08-9
5 9.696E-07 4
Runoff Input 0
5 4.586E-06 4
6 7.802E-07 4
7 1.054E-06 3
8 B.483E-07 2
9 5.251E-07 2
10 1.854E-07 2
Runoff input 0
10 5.574E-05 2
11 5.559E-07 4
12 2.147E-08 3
13-1.519E-06 3
14-8.992E-07 3
15-5.208E-07 2
16-I.648E-07 2
17-6.443E-09 2
18 8.542E-08 2
19 6.818E-10
20 1.611E-07
21 2.208E-08
22-7.943E-07
23-2.897E-07
24-3.801E-08
25 5.422E-07
26 3.037E-10
27 1.136E-09
28 6.694E-07
29 1.732E-07
30 2.286E-08
31-7.866E-07
32-8.782E-12 1
33 3.347E-13 9
34 3.196E-13 9
35 3.054E-13 9
0.000001
.OOOE-01
.462E-10
.448E-10
.519E-11
.647E-12
.652E-10-
.005 kg
.652E-10-
.042E-09
.200E-09
..789E-09
.564E-09
.438E-09
.050 kg
.438E-09
.667E-08-
.913E-08
.464E-08-
.043E-08-
.725E-08-
.473E-08-
.275E-08-
.114E-08
.985E-08
.862E-08
.767E-08
.714E-08-
.605E-08-
.515E-08-
.414E-08
.372E-08
.308E-08
.215E-08
.182E-08
.134E-08
.121E-08-
.034E-08-
.B72E-09
.430E-09
.015E-09
Dissol.
Mater
Comp. 5
(mg/1)
kg
O.OOOE-
9.481E-
9.230E-
3.520E-
6.419E-
•2.917E-
Sorbed Total Concentration Mass
Benthic Water Benthic Water Benthic
Comp. 6 .Comp. 5 Comp. 6 Comp. 5 Comp. 6
(Big/kg) (mg/1) (mg/kg)
-------
-73-
ATTACHMENT G; An Example of Summarizing and Interpreting
Exposure Model Data and Integrating Exposure
and Hazard Data for Risk Assessment
Aquatic Toxicity
The acute toxicity data indicates that this pesticide is very
highly toxic to aquatic organisms. The most sensitive fish
species is the bluegill which has an LCso of 3.83 ppb. For
aquatic invertebrates, the mysid shrimp had the lowest LCso of
0.5 ppb, whereas the acute value for Daphnia was 1.4 ppb. Two
chronic aquatic studies are available. The 32-day early-life
stage with the fathead minnow which gave a MATC of > 2.5 and
< 6.3 ppb, and the 21-day life-cycle with Daphnia produced a
MATC of > 0.198 and < 0.495 ug/1. All of these values would
place this pesticide in the very highly toxic range.
Aquatic Resj-dues
EAB used the SWRRB-EXAMS models to determine the runoff EECs
for turf. The scenario used was a 1 hectare pond, 2 meters
deep with a runoff basin of 13 hectares. The spray regime for
SWRRB was as follows:
Spray Regime
JULIAN DATES INTERVAL (days)
139 12
151 79
230
1971 133 35
168 71
234
1972 134 34
168 27
195
1973 135 30
166 30
196
1974 140 34
174 20
194
1975 129 48
179 28
205
78
-------
-74-
(ATTACHMENT G, CONTINUED)
The year selected for EXAMS was 1972 for the Tifton Turf I model
SWRRB RUNOFF VALUES FOR THfi TIFTON TURF I BASIN.
Year Day
(Julian)
1972 171
173
177
220
Active Ingredient (Ibs) per Acre
.001 - .004
.002
.001
.007 - .007
.005
.007 - .009
> .009
.013
The concentrations for four portions of the pond were estimated.
They were the chemical dissolved in the water column, chemical
attached to sediment particles in the water column, chemical dis-
solved in the pores of the bottom sediment, and chemical attached
to the bottom sediment. Below is a table which attempts to condense
the results of this computer model.
SWRRB-EXAMS EEC VALUES
Sample
Type
Bottom
Porewater
Water
Dissolved
Bottom
Sediment
Water
Sediment
Particles
No. of
Exceeds
Fish
0
12
149*
149*
Days EEC
1/2 LC50
Daphnia
51
22
149*
149*
Estimated Concentration
(ppb - Day )
Min. Max.
0.0317 - 2
0.2146-149*
33.82-4
55.52-149*
0.787-22
11.67-6
204-22
1210-2
* Hal- a UOT-O ni tron f i~it- 1 4Q Hatrc «•%!•> 1 \r
It should be pointed out that normally significant amounts of
chemical would not be expected to reach ponds through spray drift
when ground application is used. However, this case may be the
exception. Direct application to our model pond would be expected
to result in a concentration of 368 ppb. Hence, only spray drift
l/263th of the direct application would exceed the Daphnia
79
-------
-75-
(ATTACHMENT G, CONTINUED)
Aquatic Hazards
With the exception of the pore water the LCso for both fish (blue-
gill LCsQ-3.83 ppb) and Daphnia (LCso-1-4 ppb) have been exceeded.
The shortest duration of exposure among the remaining three sample
types was the water-dissolved. However, this period is long enough
to be of concern. From day 2 to 18, the concentration was greater
than the 1/2 fish LCso value. Within this period from day 6 to 10
the fish LCso would be exceeded (11.67 ppb day 6 to 4.99 ppb on
day 10).
Hazard would also be expected to benthic invertebrates under the
residue levels proposed by these models. Though the exposure of
daphnids in laboratory studies differs from the type of exposure to
an invertebrate in a natural pond or stream sediment, (estimates
are 33.82 to 204 ppb) the laboratory study has established that in-
vertebrates are extremely sensitive to this chemical (the Daphnia
LCso is 1.4 ppb and MATC values are > 0.198 and < 0.495 ug/1). Un-
like other animals, the primary route of exposure would be expected
to be through their exoskeletons. Many insecticides are lipophilic.
The available product chemistry data indicate this is the nature of
the pesticide. Its solubility in water is only 150 ppb, and the oc-
tanol/water partition coefficient is 6310. These values would indi-
cate a greater affinity for organic material and chitin (exoskeleton
material) than soil or water. Thus, one would expect the organic
runoff material to contain more pesticide than the soil portion.
Also, one would expect invertebrates such as insects and crustaceans
to be attracted to organic material as a potential food source. There-
fore, by moving in and about the residue bearing material the ani-
mal's exoskeleton surface would be expected to contact directly with
the chemical. This exposure would be in addition to that expected
from feeding on contaminated organic material. In addition, benthic
invertebrates would likely be exposed to the chemical dissolved in
the water while feeding at the surface of the bottom sediments or
in the water. Hazard would be expected under these conditions.
Serious adverse effects on aquatic invertebrate populations would
likely affect the higher trophic levels through starvation.
The chronic toxicity data indicates an additional hazard. In this
case the amount dissolved in water will exceed the lower level of
the MATC for Daphnia for the duration of the reported sample (149
days).
Therefore, the concentrations and exposure periods estimated by the
SWRRB-EXAMS model indicate a hazard to both the lentic and benthic
fauna of the pond. This is particularly significant considering only
three applications were assumed for the model and the pesticide label
does not limit the number of applications per year. In addition,
spray drift would be expected to contribute to the hazard, the ex-
tent of which we are unable to estimate at this time.
80
-------
-76-
ATTACHMENT H; Ecological Effects Branch/HED Evaluating Risk To Endangered/
Threatened* Species From Pesticide Registration Actions (see
attached written explanation for each step)
STEP 1
Identification of toxicity of
pesticide to non-target species
1
STEP 2
Screen for toxicityl
Will estimated environmental concentration
(EEC) exceed the "no-effect" cutoff points
for listed species'?**
STEP 3
•STEP 3b
Exposure of Speciesl
Is it possible for any listed
species to be exposed to pro-
jected lethal concentration^?
•
Threat to Habitatl
Is it possible for designated
"critical habitat" to be des-
troyed or adversely modified?
Informal consultation with OES/ttMFS as neededj —
YES/ \JO
HAZARD!
STEP 4
Extent of Hazard)
IND HAZARD!
Formal consultation with OES/NMFS required to determine extent of hazard)
I
STEP 5
Precautionary Labeling|
Can labeling prevent fatality to members of listed species'?!
10
STEP
Label Recommendations|
P 6b
By following labeling there
would not likely be a hazard
to listed species
[Non-labeling alternatives|
* also referred to as listed species
** l/5th the lowest mammalian acute oral LD^o or I£io; l/5th the lowest avian subacute
dietary DCio or LD±QJ 1/1 Oth the lowest aquatic acute LCio* Where LCio or LD^o are not
available, l/10th the mammalian LDso or IC$Q, l/10th the avian !£$Q or LDso or l/20th
the aquatic LCso* Also, herbicides are initially considered to impact listed plants;
consideration is given to whether it is a broadleaf, grasses, or non-selective herbi-
cide. Insecticides are initially considered to impact listed insects.
81
-------
-77-
( ATTACHMENT H, CONTINUED)
Step 1
Identification of toxicity of pesticide to non-target species
From test data submitted or referenced by the registrant, the
toxicological impact, if any, upon non-target species is deter-
mined. Extrapolations are made from the results of basic re-
quired fish and wildlife studies and other validated test data.
Step 2
Screen for toxicity
The question is asked, "will the estimated environmental
concentrations (EEC) exceed the 'no-effect1 cutoff points for
listed species?"
Likelihood of hazard
Fish and wildlife are constantly being exposed to many naturally
occurring compounds that would cause mortality or ecological
disturbance if present at high enough concentrations. Therefore,
even though the chemical is toxic to the organism and there is
the likelihood of exposure to the organism, sufficient concen-
trations of the pesticide must be available to constitute a
hazard. The obvious question: "How does one go about deter-
mining what is a sufficient concentration?"
Since it is impossible to obtain LCso or LD$Q data for listed
species, we must assume that the sensitivity of these species is
similar to that of indicator organisms used in current test pro-
tocols. Although this may or may not be the case, it would seem
appropriate, when using these data for assessing hazard to listed
species, that some "safety factor" be built into the evaluation
process. Since even the loss of one individual of listed species
may be unacceptable, some might argue that all hazard evaluations
should be based on LCj (i.e., lethal concentration required to
kill one percent of the population). However, due to the diffi-
culty in actually determining an LCj, it is proposed that the
more reliable LC^o be used. The following risk criteria for
establishing "no-effect" cutoff points would be:
1. Mammals - Occurs as a residue immediately following appli-
cation in or on the feed of a mammalian listed species
likely to be exposed to such feed in amounts equivalent to
the average daily intake of said species, at levels less
that l/5th the acute oral LDJQ, or LClO» measured in mam-
malian test animals as specified in the Registration Guide-
lines.
82
-------
-78-
(ATTACHMENT H, CONTINUED)
2. Birds - Occurs as a residue immediately following appli-
cation in or on the feed of an avian listed species likely
to be exposed to such feed 'in amounts equivalent to the
average daily intake of said species, at levels less than
l/5th the subacute dietary LCio or LDjQ measured in avian
test animals as specified in the Registration Guidelines.
3. Aquatic Organisms - Results in a maximum calculated concen-
tration in (a) or (b) below of less that l/10th the acute
LCio for aquatic organisms likely to be exposed as measured
in test animals specified in the Registration Guidelines:
(a) following direct application to a 6-inch layer
of water;.or
. (b) in the habitat(s) of concern (habitats of listed
species).
4. Chronic Effects - There are no known reproductive or other
chronic effects to indicator species at levels expected in
the habitat(s) of concern.
Step 3a
Exposure of species
If the answer to 2 above is "yes," proceed to step 3a (Exposure
of Species where the question is asked, "Is it possible for any
listed species to be exposed to projected lethal concentrations?"
A search is made of Branch records and other available sources
of information to identify listed species within proposed treat-
ment areas. Branch reviewers may informally consult with OES
and other persons knowledgeable of current listed species dis-
tribution. Based on the criteria indicated in Step 2, a deter-
mination is made whether or not that use of the pesticide pro-
duct, as proposed, may affect any listed species.
If the answer to Step 2 above is "no," proceed to Step 3b -
Threat to Habitat, and answer the question, "Is it possible
for designated critical habitat to be destroyed or adversely
modified?" Again, as needed, informal consultation may be
made with OES and other pertinent sources to solicit the most
current information on critical habitat. If there is a "No"
answer to the questions asked in Steps 3a and 3b, a "No hazard"
determination is made.
83
-------
-79-
(ATTACHMENT Hf CONTINUED)
Step 4
Extent of Hazard . .
If "Yes" is answered to Steps 3a or 3b, then a formal consul-
tation, as required within the Endangered Species Act, is
initiated by EEB. This consists of a letter of request to
the Chief, Office of Endangered Species, accompanied by a
copy of the Branch review of the pesticide product and other
supporting documentation (i.e., wildlife, fish kills attri-
butable to the use of the pesticide).
OES or NMFS, upon acknowledgement of the consultation request,
prepares a written "biological opinion" within 90 days. They
may request clarification or more data to facilitate the pre-
paration of the written opinion. On occasions they may re-
quest ,an extension of time beyond the 90 days.
Their written opinions summarize the nature of the request
(pesticide toxicological properties, use patterns, listed
species considered and listed species for which there is a
jeopardy and no jeopardy opinion).
Step 5
Precautionary Labeling
When there is a jeopardy opinion the question is asked, "Can
labeling prevent fatality to members of listed species?" If
"Yes" proceed to Step 6a - Label Recommendations with appro-
priate labeling to avert jeopardy to the species identified
within the "biological opinion."
If "No," nonlabeling alternatives must be investigated.
These include, but are not limited to, the following: clari-
fication of the reasonable and prudent alternatives with
OES, involvement of the registrant to seek means to avert
exposure of listed species (which may suggest an alteration
in the use pattern, restriction of the pesticide to specific
sites only and/or use by certified applicators only), suggest
field studies to demonstrate safe usage at prescribed label
rates exposing species most representative of listed species
of concern, refer for Special Review.
84
-------
-80-
ATTACHMENT I:
Pesticides and Crops Considered by Hoerger and
Kenaga (1972)
I. Pesticides Considered; > 20 different pesticides
A. From Literature -
II
B
1. Dicamba, amine salt*
2. 2,4-D
3. Endosulfan
4. Picloram, amine salts
5. Parathion*
6. Methyl parathion*
7. EPN
8. Sulfotepp
9. Malathion
10. Dimethoate*
11. Methoxychlor
12. Captan
B. From Tolerance Data
1. Carbaryl
2. Kelthane
3. Toxaphene*
4. DDT*
5. Malathion*
6. Parathion*
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
7.
8.
9.
10.
11.
12.
* Same pesticide examined twice
Crops Considered
A. From Literature Sources;
1. Grasses; (240 ppm)£/
a. Range grass
b. Range grass (dead undercover)
c. Fodder grass (W. Germany)
From Tolerance Data;
1* Grasses; (110 ppm)
a. • Alfalfa
b. Barley
c. Clover
d. Corn forage
e. Cotton forage
f. Dandelion
g. Sorghum forage
h. Grass
pasture or
range grass
Phosdrin
Diazinon
Demeton*
Trithion
DDT*
Dieldrin
Endrin
Aldrin
Chlordane
Toxaphene*
Ovex
Dioxathion
Methyl parathion*
Demeton*
Dicamba*
Disulfoton
Phorate
Dimethoate*
i. Pea forage
j. Cowpeas
k. Peppermint &
Spearmint hay
1. Sugarbeet tops
Numbers in parentheses are the highest pesticide residue values
found on that crop category, based on an application rate of 1 Ib
per acre.
85
-------
(ATTACHMENT I, CONTINUED)
-81-
2. Leaves & Leafy Crops; (125 ppm) 2. Leaves & Leafy Crops
a. Apples leaves j.
b. Tomato leaves k.
c. Bean leaves 1.
d. Pear leaves m.
e. Spinach n.
f. Chard o.
g. Collard p.
h. Cauliflower g.
leaves
i. Cauliflower
head
3. Forage Crops; (58 ppm)
a. Alfalfa (short)
b. Alfalfa (14-20 inches)
c. Red clover
d. Birdsfoot Trefoil
4. Pod Containing Seeds; (12 ppm)
Watercress
Celery
Celery leaves
Celery stalk
Cabbage
Carrot tops
Turnip greens
Lettuce (mature
& immature)
None examined
a.
b.
c.
d.
Beans:
Snap
Green
French
Red Kidney
5. Fruits; (6.6 ppm)
a. Apricots
b. Cherries
c. Peaches
d. Olives
e. Apples
f. Grapes
g. Strawberries
h. Oranges
(Valencia)
i. Oranges
(Navel)
. Lemons
Forage Crops; (110 ppm)
Same as 1 above
Pods Containing Seeds;
(10 ppm)
a. Shelled cowpeas .
b. Beans
c. Green beans
d. Lima beans
e. Dry beans
f. Black-eyed peas
g. Shelled peas
h. Peas with pods
Fruits; (6 ppm)
a. Blackberries &
Boysenberries
b. Blueberries
c. Cherries
d. Cranberries
e. Currants
f. Dewberries &
Loganberries
g. Gooseberries
h. Grapes
i. Plums
j. Raspberries
k. Strawberries
86
-------
(ATTACHMENT I, CONTINUED)
-82-
6. Grain & Seeds:
None examined
6. Grain & Seeds:
a. Barley
b. Dry beans
c. Dry shelled
lima beans
d. Corn
e. Cottonseed
f. Oats
g. Rice
h. Rye
i. Sorghum
j. Soybeans
k. Vetch seed
1. Wheat
III. Classes of Pesticides Considered
A. Herbicides
(1) Benzoic and Phenylacetic Acids;
(a) Dicamba
(2) Phenoxy Compounds
(a) 2f4-D
(3) Heterocyclic Nitrogen Derivatives (picolinic
acid)
(a) Picloram
B. Fungicides
(1) Chlorinated Compounds
(a) Captan
C. Insecticides
(1) DDT Relatives (diphenyl aliphatics)
(a) DDT
(b) Methoxychlor
(c) Kelthane (Dicofol)
(2) Chlorinated Aryl Hydrocarbons
(containing 6 or more chlorines)
(a) Endrin
(b) Aldrin
(c) Dieldrin
(d) Toxaphene
(e) Endosulfan
(f) Chlordane
87
-------
-83-
(ATTACHMENT I, CONTINUED)
(3) Phosphorous Compounds
(a) Disulfoton (h) Methyl Parathion
(b) Phorate (i) EPN
(c) Phosdrin (j) Trithion
(d) Malathion (k) Dioxathion
(e) Demeton (1) Diazinon
(f) Dimethoate (m) Sulfotepp
(g) Parathion
(4) Sulfonates
(a) Ovex (Ovotran)
(5) Carbamates
.-•• (a) Carbaryl
IV. Pesticides Resulting in Highest Pesticide Residue Values
Based on 1 Ib Per Acre/ According to Crop Category
A. Grasses (240 ppm):
Picloram, amine salts
B. Grasses (110 ppm);
Malathion
C. Leaves and Leafy Crops (125 ppm);
EPN
D. Forage Crops (58 ppm);
Endosulfan
E» Pods Containing Seeds (10 and 12 ppm);
10 ppm; Dimethoate 12 ppm; Methoxychlor
F. Fruits (6.6 and 6.0 ppm);
6.0; Carbaryl 6.6: Endosulfan
G. Grain and Seeds (110 ppm);
Parathion
Reference: Hoerger and Kenaga (1972)
88
-------
-84-
ATTACHMENT J; Techniques for Estimating Pesticide Residues on
Vegetation Immediately Following Application
The following techniques are designed to provide an estimate
of pesticide residues which can occur on types of vegetation which
are utilized by wildlife as forage.
It must be emphasized that actual residue values are always
preferable to "calculated" or "typical" values and should be ob-
tained and utilized whenever possible. Actual residue data become
even more important in marginal cases. However, the techniques
given below should normally be acceptable substitutes in most cases
when residue data are available.
Method A
1. Determine the intended use rate in pounds per acre.
2. Determine on the nomograph the type of vegetation which is
most appropriate for the non-target organism(s) of concern.
3. Place a horizontal straight edge on the appropriate appli-
cation rate in Ib/acre (Column No. 1) and read across to
the most appropriate maximum expected residue column.
Record this expected residue value (ppm).
4. Where subacute LC$Q data are available for a bird or mammal,
determine if the expected residue (ppm) exceeds l/5th of the
LCsO (PPm) as Per the regulation.
Method B
Alternately, where only LD$Q data are available, the following
procedure can be used with reservation and restrictions.
1. Determine the intended use rate in pounds per acre.
2. Determine on the nomograph the type of vegetation which is
most appropriate for the non-target organism(s) of concern.
3. Place a horizontal straight edge on the appropriate appli-
cation rate in Ib/acre (Column No. 1) and read across to
the most appropriate maximum expected residue column.
Record this expected residue value (ppin).
4. Locate on Table 1 the bird species whose body weight most
closely approximates the species of concern. Note the
bird weight and grams of feed eaten per day for this
species.
89
-------
-85-
(ATTACHMENT J, CONTINUED)
5. Using the above values, perform the following calculation:
feed eaten per day (g) x pesticide residues (ppm) = ing pesticide
weight of bird (g) ...... kg bird weight
6. Determine if the daily ingestion amount (mg pesticide/kg
bird weight) exceeds l/5th of the LD$Q (mg/kg) as per the
regulation.
Comments and Cautions
As with any oversimplification or extrapolation process, the
techniques described herein should always be coupled with additional
judgments. Several obvious flaws in the techniques are provided
below. These should not be forgotten or overlooked in the use of
the techniques.
1. The techniques outlined herein cannot be applied to insec-
tivorous bird exposure because insects can be expected to
contain considerably different residues than vegetation
because they will inhale, walk upon, ingest, metabolize,
and otherwise be exposed to greater amounts of pesticide
than would vegetation.
2. Neither can these techniques be effectively used when a
pesticide is applied in a granular form. When a formu-
lation takes this form, first the amount of active ingre-
dient per granule and then the number of granules equiva-
lent to the LDso in mg/kg for a species should be deter-
mined. A judgment can then be made, based on application
rate and LDsg data, as to whether a species is likely to
consume l/5th of its LD5Q in a day.
3. Certain seeds will not be eaten in toto. Instead, only
the inside unexposed portions will be ingested, leaving
the shell, hull or pod.
4. Further exposure beyond the oral route is likely. For
example, compounds known to be dermally toxic can present
additional hazard as a result of locomotion through dew
(and pesticide) covered vegetation. For some toxicants,
this route is as much or more the source of acute or
subacute hazard as the oral route.
5. Young birds may be exposed to relatively greater amounts
of ingested contaminated feed which also may be of smaller
size and therefore contain greater residues. Table 1
refers only to adult birds. In addition, immature birds
of many species go through a growth period when their
diet is almost entirely insectivorous. These birds would be
especially vulnerable during this critical period.
90
-------
-86-
(ATTACHMENT J, CONTINUED)
6. Allowance should be made for at least a 2- to 3-fold
difference in daily rate of food consumption from day
to day/ particularly in the case of large birds or
inclement weather. Thus, if a pesticide had a short-
acting nature (such as strychnine) which could kill
quickly following short, high exposure levels, death
could occur in birds gorging in contaminated feed
which would normally be eaten over a longer time
period. Therefore, the acceptable residues in vegeta-
tion would be 2 to 3 times less than those normally
considered acceptable according to the techniques
contained herein.
7. The "upper limit" residues presented in Kenaga's paper
and employed in the nomograph were stated in terms of
wet weight. The bird consumption values are presented
in-'terms of dry weight. This is a discrepancy, but
drift during application and other factors affecting
residues should essentially offset this weight basis
factor so that further correction need not be applied.
8. The techniques are only appropriate for initial surface
deposited residues and do not account for the systemic
uptake by plants which occurs with some pesticides.
9. The values in the nomograph are not necessarily appro-
priate for large fruit because birds will only peck at
and consume the surface part of the fruit and the
values in the nomograph pertain to the whole fruit.
Since residues will be concentrated on the fruit
surface and not distributed evenly throughout, the
values are not representative of the amounts that
could be expected to be consumed.
10. The techniques are useful only in the evaluation of
acute and short-term subacute hazard. They are not
intended for evaluation of chronic hazard.
11. When multiple applications are made, it is necessary
to know degradation rates to properly assess the
applicability of the techniques.
91
-------
. -87-
.*. » • "
(ATTACHMENT J, CONTINUED)
TABLE 1. Relation of dry feed consumption to body weight of birds,
Bird Species Adult Weight Mean Weight Dry
(g) Feed Eaten Per
Day (g)
Blue tit
Robin (European)
Mourning dove
Bobwhite
Mallard, Pheasant
11
16
100
.170
1200
3.3
2.35
11.2
15.2
50
-------
(ATTACHMENT J.
•CONTINUED) .
MAXIMUM EXPECTED RESIDUES
ON VEGETATION
ppm
-------
-89-
.i» • •
ATTACHMENT K; Human Risk Approach Using Mammalian Data
A. Mammalian Data
1. Older rat 2-year NOEL = 200 ppm
2. Rat body weight = 0.4 kg
3. Rat food consumption = 20 gm or 0.02 kg
4. F. cons./b. wgt. = 5%
B. Calculations
1. Relationship (Lehman, 1959):
% body weight consumed x dietary residue = mg/kg/day consumed
5% x 200 ppm = 10/mg/kg/day = NOEL
2. Human risk calculations:
a. Man body weight = 60 kg
b. Man food consumption = 1500 g or 1.5 kg
c. Food consumption/body weight = 2.5%
d. Apply 100-fold safety factor: i.e., man is
lOOx more sensitive than rat:
i. NOEL x 1/100 = 10 mg/kg/day x 1/100 =0.1 mg/kg/day
•V
ii. Food consumption (%) x residues (ppm) = mg/kg/day
Body weight
2.5% x ppm = 0.1 mg/kg/day
ppm = 4
iii. Also,
1.5 kg x 6 mg = 6 mg/60 kg person/day
60 kg 1.5 kg
[2.5% x 4.0 ppm =0.1 mg/kg/day]
iv. Thus, safe level (or NOEL) established
for man is 0.1 mg/kg/day or 6.0 mg/60 kg
person/day or 4.0 ppm.
-------
-90-
(ATTACHMENT K, CONTINUED)
C. Correlation of NOEL with residue (or tolerance) data
Proposed .
Crop or Established x Food x Daily = Contribution to
Food Item Tolerance Factor Diet NOEL^/ (mg/day/1.5 kg)
(ppm) (%) (kg)
Cottonseed 0.2 ppm x 0.15% x 1.5kg = 0.00045 (0.00045 mg/1.5 kg
= 0.0003 ppro)
These calculations indicate that a tolerance of 0.2 ppm on
cottonseed/ which constitutes only 0.15 percent of man's diet, will
contribute 0.0075 percent (0.0003 ppm/4 ppm or 0.00045 mg/6 mg)
towards the NOEL or ADI for man.2/
£/ Acutally considered an ADI (Acceptable Dietary Intake) by toxi-
cologists.
Tolerances are granted on all crops until, theoretically, the
NOEL or ADI for man is reached.
-------
-91-
CITATIONS/REFERENCES
Akerman, J.W. and D.L. Coppage. 1979. Hazard Assessment Philosophy:
A Regulatory Viewpoint In: Proceedings of a Workshop,
Analyzing the Hazard Evaluation Process. Waterville Valley,
New Hampshire. August 14-18, 1978. Water Quality Section,
American Fisheries Section. Washington, D.C.
Anonymous. 1978. Bioassay Procedures for the Ocean Disposal
Permit Program. EPA-600/9-78-010.
ASTM Standard E 729-80. 1980. Practice for Conducting Acute Toxicity
Tests with Fishes, Macroinvertebrates, and Amphibians.
American Society for Testing and Materials,
1916 Race Street, Philadelphia, PA 19103.
ASTM Standard E 724-80. 1980. Practice for Conducting Static Acute
Toxicity Tests with Larvae of Four Species of Bivalve
Molluscs. American Society for Testing and Materials, 1916
Race Street, Philadelphia, PA 19103.
ASTM Standard E 1022-84. 1984. Practice for Conducting Bioconcentration
Tests with Fishes and Salt Water Bivalve Molluscs. American
Society for Testing and Materials, 1916 Race Street,
Philadelphia, PA 19103.
Atzert, Stephen, P. 1971. A Review of Sodium Monofluoroacetate
(Compound 1080). Its Properties, Toxicology, and Use In
Predator and Rodent Control. Division of Wildlife Services,
Special Scientific Report — Wildlife No. 146. 34 pp.
Balcomb, R., C. A. Bowen II, D. Wright and M. Law. 1984. Effects on
Wildlife of At-Plant Corn Applications of Granular Carbofuran.
J. Wildl. Manage. 48:1353-1359.
Balcomb, R. 1983. Secondary Poisoning of Red-Shouldered Hawks with
Carbofuran. J. Wildl. Manage. 47:1129-1132.
Balcomb, R. 1980. Granular Pesticides: Restricted Classification Based
on the Hazard to Avian Wildlife. Ecological Effects Branch
Internal Report. February 27, 1980.
Balcomb, R. 1979. Eleven Granular Pesticides: Restricted Classification
Based on the Hazard to Avian Wildlife. Ecological Effects
Branch Internal Report. December 21, 1979.
Barnthouse, L.W., S.M. Bartell, D.L. DeAngelis, R.H. Gardner,
R.V. O'Neill, C.D. Powers, G.W. Suter, II, G.P. Thompson,
D.S. Vaughan. 1982a. Preliminary Environmental Risk Analysis for
• Indirect Coal Liguefaction. Draft Report. Oak Ridge National
Laboratory, Oak Ridge Tennessee.
-------
-92-
CITATIONS/REFERENCES (Continued)
Barnthouse, L.W., D.L. De Angelis, R.H. Gardner, R.V. 0'Neillr
C.D. Powers, G.W. Suter, II, D.S. Vaughan. 19825. Methodology for
Environmental Risk Analysis ORNL/TM 8167. Oak Ridge National
Laboratory, Oak Ridge, Tennessee.
Beusch, G.J., W.R. Bontoyan, E.B. Brittin, A.W. Bruns, J.H. Chen,
C.H. Hall, W.L. Jordan, I.N. Matter, and J.A. Shaughnessy.
1982. Pesticide Assessment Guidelines Subdivision D, Product
Chemistry. EPA-540/9-82-018.
Branson, D.R., G.E. Blan, B.C. Alexander, and W.B. Neely. 1975.
Bioconcentration of 2,2',4,4'-tetrachlorobiphenyl in rainbow
trout as measured by an accelerated test. Trans. Am. Fish.
Soc. 104:785-792.
Cairns, Jr., J., K.L. Dickson, and A.W. Maki. eds. 1978. Estimating
the Hazard of Chemical Substances to Aquatic life. ASTM
Special Technical Publication 657. American Society For
Testing' and Materials, 1916 Race Street, Philadelphia, Pa.
19103. pp 239-279.
Committee on Methods for Toxicity Tests with Aquatic Organisms. 1975.
Methods for Acute Toxicity Tests with Fish, Macroinverte-
brates, and Amphibians. U.S. EPA, Ecol. Res. Series. EPA
660/375-009.
Ecological Effects Branch, BED, OPP. 1982. Pesticide Assessment
Guidelines Subdivision E, Hazard Evaluation: Wildlife and
Aquatic Organisms. EPA 540/9-82-024. October.
Environmental Fate Branch, BED, OPP. 1982. Pesticide Assessment
Guidelines Subdivision N, Chemistry: Environmental Fate.
EPA-540/9-82-021.
Friend, M. and D.O. Trainer. 1974. Response of Different-Age Mallards
to DDT. Bull. Environ. Contam. Toxicol. 11:49-56.
Gilbertson, M. 1975. Protocol For Testing Effects of Chemicals
On Mink Reproduction. Environmental Contaminants Control
Branch, Environmental Protection Service, Environment Canada.
Ottawa. Report Number EPS l-EC-75. November.
Goodman, L.R., D.J. Hansen, D.P. Middaugh, G.M. Crise, and J.C.
Moore. 1985a. Method for Early Life-Stage Toxicity tests Using
Three Atherinid Fishes and Results with Chlorpyrifos. Aquatic
Toxicology and Hazard Assessment: Seventh Symposium. ASTM
STP854. R.D. Cardwell, R. Purdy, and R.C. Banner, Eds.
American Society for Testing and Materials, Philadelphia, PA
pp. 145-154.
-------
-93-
CITATIONS/REFERENCES (Continued)
Goodman, L.R., D.P. Middaugh, D.J. Hanson, P.K. Higdon/ and
G.M. Cripe. 19855. Early Life-Stage Toxicity Test with the
Tidewater Silversides (Menidia menidia ) and Chlorine
Produced Oxidants. Environ. Toxicol. Chem. 2:337-342.
Gusey, W.F., and Z.D. Maturgo. 1973. Wildlife Utilization of
Croplands. Environmental Affairs, Shell Oil Chemical.
Houston, Texas. 278 pp.
Heath, Robert G., et al. 1972. Comparative Dietary Toxicities of
Pesticides to Birds. Special Scientific Report — Wildlife
No. 152. February. 57 pp.
Hill, Elwood F. 1971. Toxicity of Selected Mosquito Larvicides to
Some Common Avian Species. J. Wildl. Manage. 35:757-762.
Hill, Elwood F., et al. 1975. Lethal Dietary Toxicities of
Environmental Pollutants to Birds. U.S. Fish and Wildlife
Service. Special Scientific Report — Wildlife No. 191,
February. 57 pp.
Hoist, R.W. and T.C. Ellwanger. 1982. Pesticide Assessment Guide-
lines Subdivision J, Hazard Evaluation: Non-Target Plants.
EPA-540/9-82-020.
Hoerger, F.D., and E.E. Kenaga. 1972. Pesticide Residues on Plants
Correlation of Representative Data as a Basis for Estimation
of Their Magnitude in the Environment. Environmental
Quality. Academic Press, New York, 1:9-28.
Hudson, R.H., R.K. Tucker, and M.A. Haegele. 1972. Effect of Age
on Sensitivity: Acute Oral Toxicity of 14 Pesticides to
Mallard Ducks of Several Ages. Toxicol. Appl. Pharmacol.
22:556-561.
Johnson, E.L. 1982. Risk Assessment in an Administrative Agency.
The American Statistician. August. 36:232-239.
Kaukeinen, D.E. 1984. Potential Non-Target Effects From the Use of
Vertebrate Toxicants In: Proceedings of a Conference On:
The Organization and Practice of Vertebrate Pest Control.
Elvetham Hall, Hampshire, England. August 30-September 3,
1982. ICI Plant Protection Division, Fernhurst, Haslemere,
Surrey, England, GU27 3JE.
Kaukeinen, D.E. 1982. A Review of the Secondary Poisoning Hazard to
Wildlife from the Use of Anticoagulant Pesticides In: Pest
Management. 1:10,12-14;1:16,18-19.
-------
-94-
• *. * *
CITATIONS/REFERENCES (Continued)
Kenaga, E.E. 1973. Factors to be Considered in the Evaluation of the
Toxicity of Pesticides to Birds in Their Environment.
Environmental Quality and Safety. Academic Press, N.Y.
11:166-181.
Kenaga, E.E. 1972. Guidelines for Environmental Study of Pesticides:
Determination of Bioconcentration Potential. Residue Reviews.
44:73-133.
Lehman, A.J. 1959. In: Appraisal of the Safety of Chemicals in Foods,
Drugs, and Cosmetics. Association of Food and Drug Officials
of the United States. Austin, Texas.
Mark, R.J., M.E. Burrows, R.F. Frasny, and B.H. Sleight III. 1975.
Bioconcentration of l^C Pesticides by Bluegill Sunfish
During Continuous Exposure. In: Structure Activity
Correlations in Studies of Toxicity and Bioconcentration
with Aquatic Organisms. Proceedings of a Symposium.
Burlington, Ontario. March 11-13, 1975. G.D. Veith and
D.E. Konaswich, eds. Sponsored by Standing Committee on
Scientific Basis for Water Quality Criteria of the Inter-
national Joint Commission's Research Advisory Board.
Martin, A.C., H.S. Zim and A.L. Nelson. 1951. American Wildlife and
Plants: A Guide to Wildlife Food Habits. Dover Publ. Inc.
N.Y. 500 pp.
McCann, J.A., W. Teeters, D.J. Urban and N. Cook. 1981. A Short-Term
Dietary Toxicity Test on Small Mammals. Avian and Mammalian
Wildlife Toxicity: Second Conference. ASTM Special Technial
Publication 757. December.
McElroy, A.D., et al. 1976. Loading Functions for Assessment of Water
Pollution from Non-Point Sources. EPA-600/2-76-151.
McEwen, Lowell C., et al. 1972. Wildlife Effects from Grasshopper
Insecticides Sprayed on Short-Grass Range. J. Range Manage.
25:188-194.
National Water Quality Laboratory Committee on Aquatic Bioassays.
1971a. Recommended Bioassay Procedure for Fathead Minnow
Pimephales Promelas (Rafinesque) Chronic Tests. In: Biological
Field and Laboratory Methods. U.S. EPA. EPA-670/4-73-001.
pp. 15-24.
1971b. Recommended Bioassay Procedure for Brook Trout
Salvelinus Fontinalis (Mitchell) Partial Chronic Tests
In: Biological Field and Laboratory Methods. U.S. EPA.
EPA-670/4-73-001. pp. 25-33.
-------
-95-
CITATIONS/REFERENCES (Continued)
Nice, Margaret Morse. 1938. The Biological Significance of Bird
Weights. Bird-Banding. IX:1-11.
Nicholson, H.P., H.J. Webb, G.J. Lauer, R.E. O'Brien,
A.R. Grezenda, and D.W. Shanklin. 1962. Insecticide
Contamination in a Farm Pond, Part I - Origin and Duration.
Trans. Am. Fish. Soc. 91:213-217.
Nigg, H.N., J.H. Stamper, R.M. Queen and J.L. Knapp. 1984. Fish
Mortality Following Application of Phenthoate to Florida
Citrus. Bull. Environ. Contain. Toxicol. 32:587-596.
Roach, E.R. 1973. The Effects of the Bollweevil Eradication Experiment
on Wildlife. Masters Thesis. Mississippi State University.
54 pp.
Roderick, J.V. and R.C. Tardiff. 1982. Conceptual Basis for Risk
Assessment. Presented at the Annual American Chemical
Society Meeting. Kansas City, Missouri. September, 14-16,
1982.
Schafer, Edward W. 1972. The Acute Oral Toxicity of 369 Pesticidal
Pharmaceutical and Other Chemicals to Wild Birds. Toxicol.
Appl. Pharmacol. 21:315-330.
Spencer, W.F., M.M. Cliath, J.W., Blair, and R.A. LeMert. 1985.
Transport of Pesticides from Irrigated Fields in Surface
Runoff and Tile Drain Waters. U.S. Department of Agriculture,
Agricultural Research Service, Conservation Research
Report 31. 76 pp.
Stewart, B.A., D.A. Woolhiser, W.H. Wischmeir, J.H. Caro, and
M.H. Frere. 1976. Control of Water Pollution from Cropland:
Volume II — An Overview. EPA-600/2-75-026b. June. 187 pp.
Suter, G.W., D.S. Vaughan and R.H. Gardner. 1983. Risk Assessment
by Analysis of Extrapolation Error: A Demonstration for
Effects of Pollutants on Fish. Environ. Toxicol. and
Chemistry. 2:369-378.
Tietjen, Howard P. 1976. Zinc Phosphide - Its Development As A
Control Agent For Black-Tailed Prairie Dogs. Denver
Wildlife Research Center. Special Scientific Report —
Wildlife No. 195. 14 pp.
Tucker, R.K., and M.A. Haegele. 1971. Comparative Acute Oral Toxicity
of Pesticides to Six Species of Birds. Toxicol. Appl.
'Pharmacol. 20:57-65.
-------
-9.6-
-t "
CITATIONS/REFERENCES (Continued)
Von Rumker, R., G.L. Kelso, F. Horsay, and K.A. Lawrence. 1975.
A Study of the Efficiency of the Use of Pesticide in
Agriculture. EPA-540/9-75-025.
U.S. Department of Agriculture. 1984. Agricultural Statistics 1984.
U.S. Government Printing Office. Washington, D.C. 558 pp.
U.S. Department of Agriculture. 1982. Ponds - Planning Design,
Construction. USDA, Soil Conservation Service. Agricultural
Handbook Number 450. U.S. Government Printing Office.
Washington, D.C. 558 pp.
U.S. Department of Interior. 1964. Pesticide - Wildlife Studies,
1963. A Review of Fish and Wildlife Service Investigations
During the Calendar Year. Fish and Wildlife Service.
Circular 199. August. 130 pp.
U.S. Environmental Protection Agency. 1975. DDT A Review of
Scientific and Economic Aspects of the Decision to Ban Its
Use as a Pesticide. Prepared for: Committee on Appropriations,
U.S. House of Representatives. EPA-540/1-75-022. July. 300 pp.
Wauchope, R.D. 1978. The Pesticide Content of Surface Water
Draining from Agricultural Fields - A Review Journal
of Environmental Quality. 7:459-472.
White, Donald H., Kirke A. King, Christine A. Mitchell,
Elwood F. Hill, and Thair G. Lament. 1979. Parathion Causes
Secondary Poisoning in a Laughing Gull Breeding Colony.
Bull. Environ. Contam. Toxicol. 23:281-284.
Zepp, R.L., Jr., and R.C. Kirkpatrick. 1976. Reproduction in
Cottontails Fed Diets Containing A PCB. J. Wildl. Manage.
40:491-495.
------- |