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.

-------