User's Guide and Technical Documentation


              KABAM version 1.0

(Kow (based) Aquatic BioAccumulation Model)
                     April 7, 2009
          Environmental Fate and Effects Division (EFED)
                 Office of Pesticide Programs
              U.S. Environmental Protection Agency
                    Washington, D.C.
                       1 of 123

-------
Acknowledgements


Author

Kristina Garber


Technical Reviewers

Brian Anderson
Lawrence Burkhard
Paige Doelling
Keith Sappington
Thomas Steeger


Editorial Reviewer

Karen McCormack


QA/QC Officer

Nick Mastrota


QC Testers

Oak Ridge National Laboratories
                                      2 of 123

-------
Table of Contents

1. Introduction	5
  1.1. Model Description	5
  1.2. When to use this model	6
  1.3. Conceptual model	6
  1.4. Model Application	7

2. Input parameters	8
  2.1. Chemical Specific Inputs	8
  2.2. Ecosystem Inputs	13

3. Parameters & Calculations	17

4. Model Results	18

5. Assessing pesticide concentrations in fish tissues for human consumption	24

6. Model assumptions, limitations, and uncertainties	24

Appendix A. Description of bioaccumulation model	26
  A.I. Calculation of fraction of chemical in the water column that is freely dissolved (
-------
Appendix D. Selection of mammal species of concern and corresponding biological parameters
	91
  D.I. Descriptions of mammal species	91
  D.2. Determination of mammalian default parameters for KABAM	92

Appendix E. Selection of bird species of concern and corresponding biological parameters	97
  E.I. Bird family descriptions	97
  E.2. Detailed conceptual model	102
  E.3. Determination of daily food intake	103
  E.4. Definition of default parameters to represent birds in KABAM	104

Appendix F. Description of equations used to calculate the BCF, BAF, BMF, and BSAF values.
	107
  F.I. Bioconcentration	107
  F.2. Bioaccumulation	108
  F.3. Biomagnification	109

Appendix G. Description of equations used to calculate dietary-based and dose-based EECs,
toxicity values, and RQs for mammals and birds consuming contaminated aquatic organisms 110
  G.I. Food ingestion rates	110
  G.2. Drinking water intake rates	110
  G.3. Dose-based EECs	Ill
  G.4. Dietary-based EECs	Ill
  G.5. Adjusted dose-based toxicity values	Ill

Appendix H. Methods for estimating metabolism rate constant (kM)	113
  H.I. Use of Equation Al	113
  H.2. Use of Arnotetal. 2008	114
  H.3. Assumptions and uncertainties	114

Appendix I. References Cited	116
                                       4 of 123

-------
1. Introduction

       1.1. Model Description

KABAM  (Kow  (based)  Aquatic Bio Accumulation  Model)  is  used to estimate potential
bioaccumulation  of hydrophobic  organic  pesticides  in  freshwater  aquatic food  webs  and
subsequent risks to  mammals and birds via consumption of contaminated aquatic  prey. This
model can also be used to estimate pesticide concentrations in fish tissues consumed by humans.
The  model was  designed for use by  the U. S. Environmental Protection Agency  Office of
Pesticide Programs' Environmental Fate and Effects Division  (EFED) scientists. KABAM is
composed of two parts:  1) a bioaccumulation model estimating pesticide concentrations in
aquatic organisms and 2) a risk component translating exposure and toxicological effects of a
pesticide into risk estimates for mammals and birds consuming contaminated aquatic prey.

The  bioaccumulation portion of KABAM is based on an aquatic food web bioaccumulation
model published  by Arnot and Gobas (2004).   This model was originally published in 1993 by
Gobas  (Gobas 1993) and was modified by  Arnot and Gobas  (2004).  The Arnot and Gobas
(2004) model was  selected for  estimating  pesticide bioaccumulation  based on  the  following
reasons: 1) the Gobas 1993 model underlying the Arnot  and Gobas 2004 version is generally
accepted by the scientific community as a reasonable approach for estimating bioaccumulation of
persistent  hydrophobic organic compounds in aquatic systems (Burkhard 1998); 2) the  1993
version of the model has been used by EPA for regulatory  purposes (USEPA 1995, 2000, 2003);
and  3) both Gobas  1993 and Arnot and Gobas 2004 have been published in peer-reviewed
literature.  Although originally developed and applied to the Great Lakes ecosystem for modeling
PCBs and selected  pesticides,  this model has been applied and validated for other ecosystems,
including the Hudson river, Fox river/Green Bay and Bayou D'Indie in Louisiana (USEPA 2003,
Burkhard 2003).  A detailed description of the Arnot and Gobas (2004) model is available in
Appendix A.

The  bioaccumulation  portion  of KABAM  relies  on a pesticide's  octanol-water partition
coefficient (Kow) to estimate uptake and elimination constants through respiration and diet of
aquatic organisms in different trophic levels. Pesticide tissue concentrations in aquatic organisms
are calculated for different trophic levels of a food web through diet and respiration.

In the risk component of KABAM, pesticide concentrations in aquatic organisms are used to
estimate dose- and dietary-based exposures and associated risk quotients for mammals and birds
consuming aquatic  organisms.  The methods used in the risk  component of  KABAM are
consistent with EFED's current modeling approach for assessing risks to terrestrial mammals and
birds described in USEPA 2004a, as implemented in the  T-REX model (version 1.4.1; USEPA
2008a).
                                       5 of 123

-------
       1.2. When to Use this Model

KABAM should be used for pesticides having all of the following characteristics:

       The pesticide is a non-ionic, organic chemical.
       The Log KQW value is between 4 and 8.
       The pesticide has the potential to reach aquatic habitats.

       1.3. Conceptual Model

Conceptually, KABAM represents a freshwater  aquatic ecosystem.   This  ecosystem receives
runoff and spray drift containing pesticides from  sites where pesticides are applied. The aquatic
ecosystem  incorporates seven food  web  components to  describe the trophic transfer of a
pesticide in an aquatic food web.  These include, in  increasing order of trophic level within the
food web: phytoplankton, zooplankton,  benthic invertebrates, filter feeders, small fish, medium
fish and large fish. These  components are referred to within this User's Guide as "trophic
levels." They are not intended to represent discrete trophic levels, but rather generic levels of an
aquatic food  web  (e.g., primary  producers,  primary  consumers, secondary  consumers, and
predators). KABAM also evaluates potential exposures and risks to mammals and birds that feed
upon aquatic animals containing pesticide  residues accumulated through the aquatic food web
(Figure I).
   Figure I. Conceptual model depicting aquatic food web of KABAM. Arrows depict direction of trophic
         transfer of bioaccumulated pesticides from lower levels to higher levels of the food web.
                                         6 of 123

-------
KABAM can represent a specific ecosystem, as defined by the model user. The ecosystem can
be defined by abiotic (e.g., water temperature,  % organic carbon in sediment) and biotic input
parameters  (e.g.,  body weights of aquatic animals, feeding preferences  of fish,  birds, and
mammals).  The model user can  modify these parameters to  match the characteristics of
ecosystems relevant to a specific mesocosm study or a field study.

For general use, the default model ecosystem for KABAM is defined as the EFED standard pond
scenario for the Exposure Analysis Modeling  System (EXAMS). The standard  pond has two
compartments: a water column and a benthic area.  The water column is 20,000,000 liters in
volume and the benthic area has a volume of 500,000  liters.  The  standard pond receives
pesticides in runoff (dissolved in water and sorbed onto eroded soil) and spray drift from a 10-ha
treatment field that is immediately adjacent to  the pond. The treatment field  is represented by
various scenarios using the Pesticide Root Zone Model (PRZM).  The meteorological data
corresponding to the selected PRZM scenario  can influence  the runoff of a pesticide into the
standard pond and also the water temperature of the pond environment.

The default biotic portions of KABAM are designed to be representative of organisms from the
seven trophic levels defined above.  Mammals  and birds of concern are defined by considering
species of mammals and birds relevant to the United States which rely upon aquatic ecosystems
for their food sources.

       1.4. Model Application

The application of KABAM  is referred to in  this  User's  Guide as the "KABAM tool." The
KABAM tool  is implemented in Microsoft Excel 2003. This software program was chosen as
an operating platform  because  it is available  to EFED users and  to the public.  Excel is a
commonly used spreadsheet program that most scientists are familiar with.  Computers suitable
for running the software programs necessary for this tool require no additional hardware.

Once  the KABAM tool is opened, the "Model  Description"  worksheet is displayed. This
worksheet contains the version information, a brief model description,  and a list of references.
Across the  bottom of this Excel   window are several worksheet tabs indicating the various
portions  of the KABAM tool, including "Chemical Specific  Inputs," "Ecosystem Inputs,"
"Parameters & Calculations,"  and   "Results." The  requirements  and  functions of  these
worksheets are explained in more detail below.

The overall format of the KABAM tool was developed for ease of use.  Tables embedded in the
worksheets were designed for clarity of information and for eventual cut and  paste from Excel
into a Microsoft Word document containing a risk assessment. Where necessary, comments are
provided for guidance on selecting input parameters. For more detailed information than is
contained in the spreadsheet concerning the model, input parameters, calculations, and results,
this guidance document should be utilized.
                                        7 of 123

-------
2. Input Parameters

Two types of input parameters  are required to run KABAM:  those  that are specific to the
pesticide and those that  define the aquatic  ecosystem,  including the mammals and birds of
concern. These input parameters  are distinguished by two worksheets that are titled "Chemical
Specific Inputs" and "Ecosystem Inputs."

To run the model, the user is only required to input chemical specific values since default values
are already inserted into the appropriate locations for ecosystem input parameters. These default
values allow the user to run KABAM with reasonable and reliable parameters; however, the user
can select other parameters to explore bioaccumulation of a chemical  and associated potential
risk to mammals and birds that consume aquatic animals.  Guidance for altering input parameters
from the default values is  provided in this User's Guide.

       2.1. Chemical Specific Inputs

The "Chemical  Specific Inputs" worksheet contains three  tables for the user to input data. Tables
1 and 3 require the user  to input chemical-specific values. Table 2 contains default values that
do not require user inputs, but are designed to allow the user flexibility in the case that chemical-
specific data are available for uptake and depuration rate constants in aquatic organisms.

Table 1

Table  1 requires  inputs related to the chemistry and estimated environmental concentrations
(EECs) of the pesticide.  Required  inputs include: 1) pesticide  name, 2) Log K0w, 3) organic
carbon partition coefficient (K0c),  4) sediment pore water concentrations of pesticide residues
(Pore water EEC), and 5)  aqueous concentration of pesticide residues (water column EEC).  This
table contains no default values. The user  should input values for each of these parameters in the
"Value" column of Table  1.

The titles of several tables displayed in the KABAM tool  are designed to automatically insert the
pesticide name as entered in Table 1.

Of all  parameters incorporated  into KABAM, Log  K0w has  the greatest influence on
estimates  of bioaccumulation in aquatic organisms  (see section A.7 of Appendix A).  As a
result,  this parameter is the most important for estimating potential  exposures of mammals and
birds to pesticides through consumption of contaminated aquatic organisms.  Estimates of Log
KOW can be obtained for a pesticide from acceptable or supplemental registrant-submitted studies
(OPPTS Guidelines 830.7550, 830.7560,  830.7570) and  from scientific literature.   One useful
source for locating Log K0w data in the scientific literature is Sangster (2007). Before using data
from this database, the scientist should review the original citation and determine whether the
data are acceptable or supplemental. If no measured values of Log KOW are available, this value
can be estimated using  EPI  Suite software that includes KOWIN  (USEPA  2009), which
considers contributions of the molecule's individual fragments to the overall Log K0w- If a range
of Log KOW values is available, it is suggested that the model user input the high and low Log
KOW values separately in  order to bracket the bioaccumulation potential and its associated risks.
                                        8 of 123

-------
Bioaccumulation potential increases as Log K0w increases. General guidance for evaluating
measured and estimated Log K0w data is available in Appendix B of USEPA 2003.

In Table 1 of the KABAM tool (reproduced below), K0w is automatically calculated as 10 to the
power of the Log K0w value that is entered by the model user. The K0w is used to estimate
uptake and clearance rate constants that define the concentrations of the pesticide in the tissues
of the aquatic organisms.

KOC data  can be obtained from registrant-submitted studies  (OPPTS Guidelines 835.1230,
835.1240) or from  the  scientific literature.  As the K0c value of a chemical increases, the
estimated accumulation of a chemical also increases.  The user should select the KOC value input
into PRZM/EXAMS for deriving  aquatic and  benthic  EECs. Input parameter guidance for
PRZM/EXAMS indicates that the K0c parameter value should be calculated as "the average K0c
from batch experiments" (USEPA 2002). If no scientifically valid estimates of K0care available,
this parameter value can be estimated as 0.35*Kow (USEPA 2004b).
Table 1. Chemical characteristics of Pesticide X.
Characteristic
Pesticide Name
Log KQW
KOW
KOC
(L/kgOC)
Time to steady state
(Ts; days)
Pore water EEC
(ug/L)
Water Column EEC
(ug/L)
Value
Pesticide X
5
100000
25000
30
5
6
Guidance
Required input
Required input
Enter value from acceptable or supplemental study submitted by
registrant or available in scientific literature.
No input necessary. This value is calculated automatically from the Log
KOW value entered above.
Required input
Input value used in PRZM/EXAMS to derive EECs. Follow input
parameter guidance for deriving this parameter value (USEPA 2002).
No input necessary. This value is calculated automatically from the Log
KOW value entered above.
Required input
Enter value generated by PRZM/EXAMS benthic file. PRZM/EXAMS
EEC represents the freely dissolved concentration of the pesticide in the
pore water of the sediment. The appropriate averaging period of the
EEC is dependent on the specific pesticide being modeled and is based
on the time it takes for the chemical to reach steady state. Select the
EEC generated by PRZM/EXAMS which has an averaging period
closest to the time to steady state calculated above. In cases where the
time to steady state exceeds 365 days, the user should select the EEC
representing the average of yearly averages. The peak EEC should not
be used.
Required input
Enter value generated by PRZM/EXAMS water column file.
PRZM/EXAMS EEC represents the freely dissolved concentration of
the pesticide in the water column. The appropriate averaging period of
the EEC is dependent on the specific pesticide being modeled and is
based on the time it takes for the chemical to reach steady state. The
averaging period used for the water column EEC should be the same as
the one selected for the pore water EEC (discussed above).
Note: Table 1 of this User's Guide contains example data for chemical specific characteristics.
                                        9 of 123

-------
The time to steady state (Ts;  in days) is also calculated  automatically by the KABAM tool
according to Equation  1 (Hawker and  Connell  1988).   This equation is consistent with
recommendations  provided  in USEPA  and  OECD guidelines for  fish BCF  studies  for
determining the time to reach steady state (USEPA 1996, OECD 1996).  It should be noted that
there is uncertainty in using this equation for chemicals with Log K0w >6, since this falls outside
of the range of data used to derive this relationship. Alternatively, the time to steady state can be
defined  using empirical data from  available BCF studies that were sufficient to define steady
state. This information can be used to supplement the calculated Ts value.
r   ,.   ,
Equation 1 .
                                         (6.54xW-3)*Kow+553l
                                         - - - -  -
                                                   24
EECs from PRZM (v3.12.2, May 2005) and EXAMS (v2.98.4.6, April 2005) (coupled with the
input shell  pe5.pl, dated Aug 2007) are used in  the  KABAM tool.  EECs generated by
PRZM/EXAMS represent the freely dissolved concentration of the pesticide in the surface and
pore water of the  standard pond. The bioaccumulation portion of KABAM assumes that the
aquatic environment is at steady state. Because the time to reach steady state is pesticide specific,
the appropriate averaging period of the EEC should be determined on a chemical by chemical
basis. Generally, the time to reach steady state can be related to the Log KOW of a chemical, with
increasing time required as the Log K0w increases. Therefore, it is not relevant to use short-term
(peak) estimates of pesticides  in the aquatic environment.  The EEC used to represent  the
concentration of the pesticide in the  pore and surface waters of the aquatic habitat should be
selected so that the averaging period (i.e., 4-d, 21-d, 90-d, 1 year), is consistent with the time to
steady state estimated for that chemical. For example, a chemical with a  Log K0w = 5 would
have an estimated time to  steady state value of 30 days. Since  the standard output file from
PRZM/EXAMS does not include a 30-d average, the next closest averaging period would be
selected (either 21  or 60 days). Therefore, the EEC represented by the 21 -day average would be
selected for this chemical.  In cases where the time to steady state exceeds  365 days, the user
should select the EEC representing the yearly EEC.

In cases where multiple uses of a single pesticide are possible (e.g., cotton, corn, apples), EECs
from the different uses can be modeled to allow for an understanding of the bioaccumulation and
associated risks associated with different uses.

Table 2

KABAM automatically calculates uptake and elimination constants through  respiration (ki and
k2, respectively) and diet (ko and ks, respectively). In using the model with its default parameters
in place, it  is assumed that the elimination of the  pesticide from aquatic organisms through
metabolism does not occur (i.e.,  metabolism rate constant kM = 0).

In Table 2 (reproduced below) of the KABAM tool, the model  user can enter measured rate
constants for uptake and  elimination  constants. These data can be obtained from acceptable or
supplemental studies submitted  by the registrant or from the  literature. For example, k\  and k2
rate constants for fish can be obtained from pesticide BCF studies submitted for the fish (OPPTS
                                        10 of 123

-------
Guideline 850.1730).  However,  caution should be used when altering rate  constants.   For
example, the k2 from a bioconcentration study typically represents a total elimination half life.
However, the k2 in KABAM represents elimination from the gills.  Therefore, incorporation of a
measured k2 into KABAM without consideration of other elimination pathways may result in
erroneous results   In order to run the model, it is  not necessary for the user to alter the
default values inserted into Table 2.  If the model user alters the parameters in Table 2 of the
KABAM tool, they will be highlighted yellow.
Table 2. Input parameters for rate constants, "calculated" indicates that model will calculate rate
constant.
Trophic level
phytoplankton
zooplankton
benthic invertebrates
filter feeders
small fish
medium fish
large fish
ki
(L/kg*d)
calculated
calculated
calculated
calculated
calculated
calculated
calculated
k2
(d-1)
calculated
calculated
calculated
calculated
calculated
calculated
calculated
kD
(kg-food/kg-
org/d)
0*
calculated
calculated
calculated
calculated
calculated
calculated
kE
(d-1)
0*
calculated
calculated
calculated
calculated
calculated
calculated
kM*
(d-1)
0
0
0
0
0
0
0
* Default value isO.
kj and k2 represent the uptake and elimination constants respectively, through respiration.
kD and kE represent the uptake and elimination constants, respectively, through diet.
kM represents the metabolism rate constant.
The  model user should exercise caution when using a value of kM>0, as this approach will
decrease predicted EECs and RQs for mammals and birds.  Initially, kM should be set to 0 as a
screen. The assumption that there is no metabolism of the pesticide within aquatic organisms is
conservative. If no metabolism is observed in available fish BCF studies, then kM should not be
altered.  In cases where metabolism  occurs, this assumption can result in overestimates  of
pesticide accumulation in tissues of aquatic organisms. In cases where the model user has
evidence to indicate that metabolism may occur in fish (i.e., from BCF studies) and RQ values
exceed LOCs, then the user can estimate kM using approaches described in  Appendix H.  This
will  allow the model user to characterize effects  of metabolism on bioaccumulation in  aquatic
ecosystems and associated risks to mammals and birds consuming aquatic organisms.

Table 3

To calculate  risk quotients, user-supplied avian and mammalian toxicity endpoints should be
entered  into Table 3 (reproduced below).  Acceptable  or supplemental registrant-submitted or
open literature studies should be  used  to define  the effects  of the  pesticide  on birds and
mammals.  Required input data include: avian acute  oral LD50 (OPPTS Guideline 850.2100),
avian subacute dietary LCso (OPPTS  Guideline 850.2200), avian reproduction (expressed as a
NOAEC or a NOAEL) (OPPTS  Guideline 850.2300), mammalian  acute  oral  LD50 (OPPTS
Guideline  870.1100),  or subacute dietary LC50  (if available),  and  mammalian  reproduction
NOAEC or NOAEL (OPPTS Guideline 870.3800).
                                       11 of 123

-------
Table 3. Mammalian and avian toxicity data for Pesticide X. These are required inputs.
Animal
Avian
Mammalian
Measure of effect
(units)
LD50 (mg/kg-bw)
LC50 (mg/kg-diet)
NAOEC (mg/kg-diet)
Mineau Scaling
Factor
LD50 (mg/kg-bw)
LC50 (mg/kg-diet)
Chronic Endpoint
units of chronic
endpoint *
Value
50
500
10
1.15
50
N/A
10
ppm
Species
mallard duck
Northern bobwhite quail
mallard duck
Default value for all
species is 1.15 (for
chemical specific values,
see Mineau et al. 1996).
other
other
laboratory rat
If selected species is
"other," enter body
weight (in kg) here.




1.2


*ppm = mg/kg-diet
Note: Table 3 of this User's Guide contains example data for chemical specific characteristics.

In the appropriate cell under the "value"  column of Table 3,  the user should input the lowest
(most sensitive) available toxicity data for each toxicity endpoint. If an endpoint value is not
discrete (i.e., contains a ">" symbol), the whole number should be entered as a discrete value,
keeping in mind that all resulting risk quotient (RQ) values derived using this endpoint are "<".
For the chronic mammalian data, the user must also select the units of the value.  The user should
select units from the drop down menu as either "ppm" or "mg/kg-bw."

Under the "species" column, the user should use the drop down menu to select the appropriate
test species associated with the toxicity value entered in the adjacent cell in the "value" column.
If the test species is not one of the options available in the drop down list, the model user should
select "other" as the test species.  If "other" is selected, the user must enter the  body weight (in
kg) of the test species. In the case that "other" is selected as the test species, a message will
appear in the spreadsheet below Table 3 to alert the user of the need to enter the body weight of
the test species. These data should be obtained from the study report if possible (time weighted
average of control animals). Alternatively, reference body weight values may be obtained from a
variety of sources, including U.S.  EPA 1993 and Dunning 1984. Failure to enter the body weight
of the test species when it is  entered as  "other" will  prevent  calculation of risk quotients that
correspond to that endpoint.

If available, the model user should enter chemical-specific data to represent the avian scaling
factor (see Mineau et al. 1996). If no chemical specific data are available, the  default value of
1.15 should be entered. This value is used to adjust avian dose-based toxicity values based on the
weight of the species of concern (e.g.,  herons) as described in the T-REX User's Guide (USEPA
2008a) and in Appendix G.
                                        12 of 123

-------
       2.2. Ecosystem Inputs
In order to estimate pesticide concentrations in tissues of aquatic organisms, biotic and abiotic
characteristics of the model aquatic ecosystem must be defined. In addition, the mammals and
birds consuming aquatic organisms are also defined as ecosystem inputs.

To run  KABAM, it  is  not necessary  to alter  any  of  the  default parameters in the
"Ecosystem Inputs" worksheet.

If the model user alters  default parameter values,  they will be highlighted yellow in the
KABAM tool.

It may be necessary  for the model user to incorporate alternate ecosystem input values if the
modeling incorporates EECs from a source other than  PRZM/EXAMS (e.g., from a mesocosm
study). In that case, the model user should enter parameter values that correspond to the specific
water body used.

Table 4

Abiotic characteristics of the aquatic ecosystem that are necessary for KABAM are defined  in
Table 4 (reproduced below) of the model tool.  These  characteristics include: concentrations  of
particulate organic carbon  (XPOc), dissolved organic carbon (XDOc), dissolved oxygen (Cox) and
suspended  solids (Css), water temperature (T),  and % organic carbon (OC) content of the
sediment.  The model tool is populated with default values for these parameters, which can be
altered based on  the  needs of the model  user.   Default values relevant to the abiotic
characteristics of the aquatic ecosystem are designed  to be consistent with the OPP  standard
pond scenario used in EXAMS. Brief explanations for these default values as well as guidance
on selecting alternative values are provided in Appendix B.
Table 4. Abiotic characteristics of the model aquatic ecosystem.
Characteristic (symbol; units)
Concentration of Particulate Organic
Carbon (XPOC; kg OC/ L)
Concentration of Dissolved Organic
Carbon (XDOC; kg OC/L)
Concentration of Dissolved Oxygen
(Cox;mg02/L)
Water Temperature
(T; C)
Concentration of Suspended Solids
(Css; kg/L)
Sediment Organic Carbon
(OC; %)
Value
0
0
5.0
15
3.00E-05
4.0%
Guidance*
When using EECs generated by PRZM/EXAMS, use a
value of "0" for both POC and DOC.
Default value is 5.0 mg O2/L when using EECs
generated by PRZM/EXAMS.
Value is defined by the average water temperature of the
EXAMS pond when using EECs generated by
PRZM/EXAMS. Model user should consult output file
of EXAMS to define this value.
Default value is S.OOxlO"5 kg/L when using EECs
generated by PRZM/EXAMS.
Default value is 4.0% when using EECs generated by
PRZM/EXAMS.
*When using pesticide concentrations from monitoring data or mesocosm studies, consult Appendix B of the
User's Guide for specific guidance on selecting values for these parameters.
                                        13 of 123

-------
Table 5
Necessary biotic components of the aquatic ecosystem define characteristics of the sediment and
water column biota. These include body weights and body compositions, specifically % lipids, %
NLOM (non-lipid organic matter), and % water.  These values are defined for the seven trophic
levels of the aquatic ecosystem (phytoplankton, zooplankton, benthic invertebrates, filter feeders,
small fish, medium fish and large fish) modeled by KABAM in Table 5 (reproduced below) of
the tool. Default values for these biotic parameters are displayed in Table 5 below. A description
of how these default parameters were selected is available in Appendix C. In addition, Table 5
allows the model user to define whether organisms within each trophic level respire pore water.
If yes, it is assumed that 5% of the total respired water is from pore water. Default assumptions
related to respiration of pore water for each trophic level are depicted in Table  5 below.
Table 5. Characteristics of aquatic biota of the model ecosystem.
Trophic level
Sediment*
Phytoplankton
Zooplankton
Benthic invertebrates
Filter feeders
Small fish
Medium fish
Large fish
Wet
Weight
(kg)
N/A
N/A
l.OE-07
l.OE-04
l.OE-03
l.OE-02
l.OE-01
l.OE+00
% lipids
0.0%
2.0%
3.0%
3.0%
2.0%
4.0%
4.0%
4.0%
% NLOM
4.0%
8.0%
12.0%
21.0%
13.0%
23.0%
23.0%
23.0%
% Water
96.0%
90.0%
85.0%
76.0%
85.0%
73.0%
73.0%
73.0%
Do organisms in
trophic level respire
some pore water?
N/A
no
no
yes
yes
yes
yes
no
*Note that sediment is not a trophic level. It is included in this table because it is consumed by aquatic organisms
of the KABAM food web.
N/A = not applicable
Table 6

Table 6  (reproduced below) of the  KABAM tool allows the  model user to define the  diet
composition of each of the trophic levels of the aquatic ecosystem. The aquatic trophic levels are
assigned a hierarchy, which is relevant to the assignment of diet composition. The order of the
trophic  levels,  in increasing  hierarchy,  is  as follows:  phytoplankton, zooplankton, benthic
invertebrates, filter feeders, small fish, medium fish,  and large fish. The diet of each aquatic
trophic level is composed  of sediment or water column biota from  lower trophic  levels.  The
KABAM tool does not allow the model user to assign a portion of the  diet of one organism to its
own trophic level or to trophic levels that are higher. The default values defining the diet of each
trophic  level are in  Table  6  below.  An  explanation of how  these default  parameters were
determined is available in Appendix C.
                                        14 of 123

-------
Note that the total diet of each organism within the aquatic food web should equal 100%.  If the
total diet ^ 100%, an error message will appear under Table 6.
 Table 6. Diets of aquatic biota of the model ecosystem.
*Note that sediment is not a trophic level. It is included in this table because it is consumed by aquatic organisms of
the KABAM food web.

Table 7

Table 7 (reproduced below) of the KABAM tool allows the model user to define the mammalian
and  avian  species of concern, as well as their body weights. Species are considered to be of
concern for pesticide exposures through consumption of residues in freshwater aquatic animals
that  serve as prey.

For  mammals, default species include the fog shrew  (Sorex sonomae\ the water shrew  (S.
palustris),  the rice rat (Oryzomys palustris),  the  star-nosed mole  (Condylura  cristata\ the
American mink (Neovison vison), and the Northern river otter (Lontra canadensis). For birds,
default species include sandpipers, rails, herons, kingfisher, ducks, grebes, ibis, rails, cormorants,
osprey,  cranes, bald eagles  (Haliaeetus leucocephalus) and pelicans. Descriptions  of how
mammalian and avian species were  selected,  including their body  weights, are provided in
Appendices D and E, respectively. These appendices also provide descriptions of the  species
themselves as well  as justifications  for  default parameters used to represent the species in
KABAM {i.e., body weight and diet).

The  selected body weight value influences estimates of pesticide exposure through differential
consumption of contaminated food items, as well as dose-based toxicity values. Therefore, the
magnitude of the body weight parameter has an effect on the magnitude of the dose-based RQ.
For  mammals, higher body weight values result in higher dose-based RQs (keeping the diet
constant). As  a result, default body weight values for the fog shrew, water shrew, rice rat, and
star-nosed mole were selected as higher values of relevant ranges  in order to represent size
classes that would be most vulnerable to exposures through bioaccumulation. In order to bound
the risk of accumulated residues to mink and river otter, the lowest and highest body weights of
these species were selected as defaults. For birds, higher body weight results in lower RQs. In
order to bound the risk of accumulated residues to birds, the lowest and highest body weights of
                                         15 of 123

-------
birds with the same diet were selected as defaults. The user can alter the assigned body weights
to represent the low and high end of possible weights in order to bound the potential RQs for a
particular species. Additional data on body  weights of species  of  mammals and  birds  are
provided in Appendices D and E, respectively.
Table 7. Identification of mammals and birds feeding on aquatic
biota of the model ecosystem.
Mammal/Bird #
Mammal 1
Mammal 2
Mammal 3
Mammal 4
Mammal 5
Mammal 6
Birdl
Bird 2
BirdS
Bird 4
Bird5
Bird 6
Name
Fog/Water shrew
Rice Rat/Star-nosed mole
Small mink
Large mink
Small river otter
Large river otter
Sandpipers
Cranes
Rails
Herons
Small osprey
White pelican
Body
weight (kg)
0.018
0.085
0.450
1.800
5.000
15.000
0.02
6.7
0.07
2.90
1.25
7.50
Tables 8 and 9

Tables 8 and 9 (reproduced below) of the KABAM tool allow the model user to define the diet
composition of the mammals and birds of concern that are defined in Table 7. The animal names
entered in Table 7 will appear at the heads of the columns of Tables 8 and 9. The diet of each
mammal and bird species is attributed to a portion of each trophic level of the aquatic ecosystem.
Justifications for the default diets for each mammal and bird species are provided in Appendices
D and E, respectively. Note that the total diet of each mammal and bird should equal 100%.  If
not, an error message will appear under Table 8 or 9.
Table 8. Diets of mammals feeding on aquatic biota of the model ecosystem.
Trophic level in diet
Phytoplankton
Zooplankton
Benthic invertebrates
Filter feeders
Small fish
Medium fish
Large fish
Total
Diet for:
Fog/Water
Shrew
0.0%
0.0%
100.0%
0.0%
0.0%
0.0%
0.0%
100.0%
Rice
Rat/Star-
nosed mole
0.0%
0.0%
34.0%
33.0%
33.0%
0.0%
0.0%
100.0%
Small Mink
0.0%
0.0%
0.0%
0.0%
0.0%
100.0%
0.0%
100.0%
Large Mink
0.0%
0.0%
0.0%
0.0%
0.0%
100.0%
0.0%
100.0%
Small River
Otter
0.0%
0.0%
0.0%
0.0%
0.0%
100.0%
0.0%
100.0%
Large River
Otter
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
100.0%
100.0%
                                        16 of 123

-------
Table 9. Diets of birds feeding on aquatic biota of the model ecosystem.
Trophic level in diet
Phytoplankton
Zooplankton
Benthic invertebrates
Filter feeders
Small fish
Medium fish
Large fish
Total
Diet for:
Sandpipers
0.0%
0.0%
33.0%
33.0%
34.0%
0.0%
0.0%
100.0%
Cranes
0.0%
0.0%
33.0%
33.0%
0.0%
34.0%
0.0%
100.0%
Rails
0.0%
0.0%
50.0%
0.0%
50.0%
0.0%
0.0%
100.0%
Herons
0.0%
0.0%
50.0%
0.0%
0.0%
50.0%
0.0%
100.0%
Small Osprey
0.0%
0.0%
0.0%
0.0%
0.0%
100.0%
0.0%
100.0%
White pelican
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
100.0%
100.0%
If the model user chooses to alter the default diet of a mammal or bird, the model user should
consider the  daily food intake for  determining  appropriate aquatic trophic levels to include
within an animal's diet. The user should verify that the weight of an individual dietary item does
not greatly exceed the daily food intake of the mammal or bird.  This will prevent the user from
simulating a bird or mammal that consumes prey that are much larger than could be reasonably
consumed. This can be determined using allometric equations for estimating daily food intake, as
described in Appendices D and E. In addition, these appendices contain data defining the daily
food intake for several species of birds and mammals.

Pesticide  exposures to  mammals  and birds through consumption of  contaminated aquatic
organisms are determined by weighing the exposure concentration by the contribution of each
food item to the  total diet. While this approach is reasonable for chronic exposures,  it may
underestimate acute exposures resulting  from  consumption of larger trophic level organisms
within short periods of  time. In  order to explore  high-end exposure concentrations and
subsequent risks resulting from  acute exposures, the model user can set the  highest aquatic
trophic level consumed by a bird or mammal to  100%. For example, high-end acute exposures of
cranes (which consume benthic  invertebrates,  filter feeders, and medium fish) to  a  pesticide
could be assessed by setting the crane diet to 100% of medium fish.

3. Parameters & Calculations

Also included in the KABAM tool is a tabularized summary of the relevant parameters for the
bioaccumulation portion of KABAM.  This summary is included in a separate worksheet, titled
"Parameters &  Calculations"  (Table 10  of the KABAM tool) and represents values used to
calculate pesticide tissue concentrations for  the  trophic levels  of the aquatic  ecosystem. This
worksheet is locked (read only) in the KABAM  tool and  cannot be altered by the model user;
however, this worksheet can be printed by the model user or copied into  a risk assessment as a
model output. A full description of the parameters contained in Table 10 of the KABAM tool as
well as the equations used to calculate these parameters can be found in Appendix A.
                                        17 of 123

-------
4. Model Results

The final outputs of  KABAM include Bioconcentration Factors (BCFs),  Bioaccumulation
Factors (BAFs),  Biomagnification Factors  (BMFs),  Biota-Sediment Accumulation  Factors
(BSAFs),  estimates of pesticide concentrations in tissues of aquatic organisms, and RQ values
for mammals and birds consuming contaminated aquatic organisms.

Note that the "results" worksheet of KABAM is locked (read only) and cannot be altered by the
model user, with the exception of format changes (e.g.,  number of decimal places). Also, the
KABAM tool does not automatically account for significant  figures.  The format of numerical
values in the Tool can be altered  by the user to increase or decrease the number of decimal
places.

Table 11 and Figure 1

Table 11 (reproduced below) of the KABAM tool reports pesticide concentrations in tissues of
aquatic organisms on both a total body weight and lipid normalized basis.  The table also reports
contributions of the pesticide concentration in tissue from respiration and from diet.   These
values are useful for understanding the dominant uptake route  of the pesticide that influences
bioaccumulation.  Figure  1 (reproduced below) of the KABAM tool graphically represents the
relative contributions of pesticide  uptake through diet and through respiration to the overall
concentrations of the pesticide in the tissues of the different aquatic animals.
Table 11. Estimated concentrations of Pesticide X in ecosystem components.
Ecosystem Component
Water (total)*
Water (freely dissolved)*
Sediment (pore water)*
Sediment (in solid)**
Phytoplankton
Zooplankton
Benthic Invertebrates
Filter Feeders
Small Fish
Medium Fish
Large Fish
Total
concentration
(jig/kg-ww)
6
6
5
5,000
27,298
21,065
23,678
15,549
34,713
41,050
56,332
Lipid
normalized
concentration
(jig/kg-lipid)
N/A
N/A
N/A
N/A
1364913
702157
789265
777440
867830
1026242
1408297
Contribution
due to diet
(jig/kg-ww)
N/A
N/A
N/A
N/A
N/A
651.72
1,812.95
1,167.92
7,246.79
14,492.66
30,795.48
Contribution
due to
respiration
(jig/kg-ww)
N/A
N/A
N/A
N/A
27,298.25
20,412.98
21,865.01
14,380.88
27,466.40
26,557.01
25,536.39
* Units: ug/L; **Units: ug/kg-dw
Note: Table 11 of this User's Guide contains example results based on example chemical specific data entered in
Tables 1 and 3.
                                        18 of 123

-------
      60000
              Zooplankton
  Benthic
Invertebrates
Filter Feeders     Small Fish

   Trophic Level
Medium Fish
Large Fish
  Figure 1. Total Pesticide Concentration
             per trophic level
                                      I Contribution dueto Respiration (ug/kg-ww)

                                       Contribution due to Diet (ug/kg-ww)
Note: Figure 1 of this User's Guide contains example results based on example chemical-specific data entered in
Tables 1 and 3.

Tables  12 and 13

BCF, BAF, BMP and BSAF are calculated by KABAM (Tables 12 and 13).  These  terms are
intended  to provide a relative measure of the  pesticide concentration in an  organism  to the
pesticide  concentration in sources (i.e., the environment and the diet) of that pesticide. Appendix
F contains the equations used to calculate BCF, BAF, BMF and BSAF.
Table 12. Total BCF and BAF values of Pesticide X in aquatic trophic levels.
Trophic Level
Phytoplankton
Zooplankton
Benthic Invertebrates
Filter Feeders
Small Fish
Medium Fish
Large Fish
Total BCF
Gig/kg-ww)/Qig/L)
4801
3421
3705
2435
4766
4766
4806
Total BAF
Oig/kg-ww)/(fig/L)
4550
3511
3946
2591
5786
6842
9389
Note: Table 12 of this User's Guide contains example results based on example chemical specific data entered in
Tables 1 and 3.
                                           19 of 123

-------
Table 13. Lipid-normalized BCF, BAF, BMF and BSAF values of Pesticide X in aquatic trophic
levels.
Trophic Level
Phytoplankton
Zooplankton
Benthic Invertebrates
Filter Feeders
Small Fish
Medium Fish
Large Fish
BCF
(jig/kg-
lipid)/Oig/L)
240045
114028
123488
121769
119142
119142
120143
BAF
(jig/kg-
lipid)/Oig/L)
227485
117026
131544
129573
144638
171040
234716
BMF
(Hg/kg-
lipid)/(fig/kg-
lipid)
N/A
0.51
.16
.14
.16
.24
.37
BSAF
Oig/kg-
lipid)/(jig/kg-
OC)
11
6
6
6
7
8
11
Note: Table 13 of this User's Guide contains example results based on example chemical specific data entered in
Tables 1 and 3.
Tables 14, 15, and 16

Tables  14,  15,  and 16  (reproduced  below) of the KABAM tool summarize the estimated
exposure values, mammal and bird toxicity values and resulting RQ values, respectively, used to
estimate potential risks to mammals and birds that consume aquatic animals contaminated with
pesticides accumulated through the aquatic food chain.

Table 14 uses the mammalian and avian body weights (entered by the model user) to calculate
the dry food ingestion and drinking water intake rates according to allometric equations specific
to mammals and birds.  The wet food intake is calculated using the dry food intake and the %
water of the diet. Dose-based EECs represent the  sum of pesticide  intake through diet and
through drinking water, accounting for  pesticide concentrations in diet items and in water and
food and water intake rates. Dietary-based EECs represent the sum of pesticide intake through
diet only, without consideration of species specific intake rates or body weights. Descriptions of
the equations used to calculate food intake rates, water intake  rates, dose-based  EECs, and
dietary-based EECs are available in Appendix G.
                                        20 of 123

-------
Table 14. Calculation of EECs for mammals and birds consuming fish contaminated by Pesticide X.
Wildlife Species
Biological Parameters
Body
Weight
(kg)
Dry Food
Ingestion
Rate (kg-dry
food/kg-
bw/day)
Wet Food
Ingestion
Rate (kg-wet
food/kg-
bw/day)
Drinking
Water
Intake
(L/d)
EECs (pesticide intake)
Dose Based
(mg/kg-
bw/d)
Dietary
Based
(ppm)
Mammalian
Fog/water shrew
Rice rat/star-nosed
mole
Small mink
Large mink
Small river otter
Large river otter
0.02
0.1
0.5
1.8
5.0
15.0
0.140
0.107
0.079
0.062
0.052
0.042
0.585
0.484
0.293
0.229
0.191
0.157
0.003
0.011
0.048
0.168
0.421
1.133
13.857
11.921
12.041
9.408
7.844
8.852
23.68
24.64
41.05
41.05
41.05
56.33
Avian
Sandpipers
Cranes
Rails
Herons
Small osprey
White pelican
0.0
6.7
0.1
2.9
1.3
7.5
0.228
0.030
0.147
0.040
0.054
0.029
1.034
0.136
0.577
0.157
0.199
0.107
0.004
0.211
0.010
0.120
0.069
0.228
25.5861
3.6561
16.8571
5.0943
8.1859
6.0108
24.75
26.90
29.20
32.36
41.05
56.33
Note: Table 14 of this User's Guide contains example results based on example chemical specific data entered in
Tables 1 and 3.
Table 15 (reproduced below) of the KABAM tool summarizes the acute and chronic, dose-based
and  dietary-based toxicity values representing effects of a pesticide to  mammals  and birds.
Dietary-based toxicity values are taken directly from user inputs in Table 3, without adjustment.
Available dose-based toxicity  values are adjusted for the  weights of the animal tested (e.g.,
laboratory rat, mallard duck) and of the animal for which the risks are being assessed (e.g., mink,
bald eagle). Methods for adjusting  toxicity values are consistent with those used by  T-REX
(USEPA 2008a). A full description of the methodology for adjusting dose-based toxicity values
is provided in Appendix G.
                                         21 of 123

-------
Table 15. Calculation of toxicity values for mammals and birds consuming fish contaminated by Pesticide
X.
Wildlife Species
Toxicity Values
Acute
Dose Based
(mg/kg-bw)
Dietary Based
(mg/kg-diet)
Chronic
Dose Based
(mg/kg-bw)
Dietary Based
(mg/kg-diet)
Mammalian
Fog/water shrew
Rice rat/star-nosed
mole
Small mink
Large mink
Small river otter
Large river otter
142.87
96.92
63.89
45.18
35.00
26.59
N/A
N/A
N/A
N/A
N/A
N/A
1.05
0.71
0.47
0.33
0.26
0.20
10
10
10
10
10
10
Avian
Sandpipers
Cranes
Rails
Herons
Small osprey
White pelican
25.96
62.10
31.33
54.77
48.27
63.16
500.00
500.00
500.00
500.00
500.00
500.00
N/A
N/A
N/A
N/A
N/A
N/A
100
100
100
100
100
100
Note: Table 15 of this User's Guide contains example results based on example chemical specific data entered in
Tables 1 and 3.

Table 16 (reproduced below) of the KABAM tool presents RQs, which are the ratio of exposure
concentrates to effects values.   RQ values are then compared to Agency levels of concern
(LOCs) for  non-listed and listed mammals and birds. For acute exposures, the LOG  is 0.5 for
(non-listed)  birds and mammals and 0.1 for federally-listed threatened and endangered (listed)
species  of mammals and birds.  For  chronic risk, the  LOG is 1.0 for all species (non-listed and
listed) mammals and birds (USEPA 2004). RQ values that exceed their respective LOG values
appear in red and bold in Table 16.

Dose-based  and dietary-based RQs  are not equivalent. Dietary-based RQs are  calculated by
directly comparing the concentration of a pesticide administered to experimental animals in the
diet in a toxicity study to the concentration estimated in selected food items. These RQs do not
account for the fact that smaller-sized animals need to consume more food relative to their body
weight than  larger animals. The dose-based RQs account for these factors by incorporating the
                                        22 of 123

-------
ingestion rate-adjusted exposure from the various food items to the different weight classes of
assessed animals and the weight class-scaled toxicity endpoints.
Table 16. Calculation of RQ values for mammals and birds consuming fish contaminated by Pesticide X.
Wildlife Species
Acute
Dose Based
Dietary Based
Chronic
Dose Based
Dietary Based
Mammalian
Fog/water shrew
Rice rat/star-nosed
mole
Small mink
Large mink
Small river otter
Large river otter
0.097
0.123
0.188
0.208
0.224
0.333
N/A
N/A
N/A
N/A
N/A
N/A
13.198
16.737
25.643
28.335
30.498
45.296
2.368
2.464
4.105
4.105
4.105
5.633
Avian
Sandpipers
Cranes
Rails
Herons
Small osprey
White pelican
0.986
0.059
0.538
0.093
0.170
0.095
0.049
0.054
0.058
0.065
0.082
0.113
N/A
N/A
N/A
N/A
N/A
N/A
0.247
0.269
0.292
0.324
0.410
0.563
Note: Table 16 of this User's Guide contains example results based on example chemical specific data entered in
Tables 1 and 3.

EECs and RQs for birds are based on the selected body weight of the bird as well as its diet.
Default values for birds were designed to represent birds on the low and high end of weights
with three different diets. Birds consuming benthic invertebrates, filter feeders, and fish include
sandpipers,  ducks and cranes (default birds 1 and 2, which are named sandpipers and cranes,
respectively).  Birds consuming benthic invertebrates and fish include belted kingfisher, rails,
ibis, grebes, double-breasted cormorants, bitterns, egrets, and herons (default birds 3 and  4,
which are named rails and herons, respectively).   Birds  consuming fish include osprey, bald
eagles, and  the white pelican (default birds 5  and 6, which are named small osprey and white
pelican, respectively).  In the case that RQs exceed the LOG  for both birds within a feeding
group, then  it can be assumed that RQs would exceed the LOG for all of the birds within the
feeding category  since birds on the low and high end of the weight ranges have RQs of concern.
In the case that RQs exceed the LOG for the default bird with the high body weight of a feeding
                                        23 of 123

-------
category (i.e., birds 2, 4, and 6), the model user can refine the EECs and RQs to be representative
of specific bird species within a feeding category by entering specific body weights of individual
species  of concern.  Appendix E  contains species  specific  data  on  feeding habits and body
weights of over 40  species of birds, including  some  listed  species,  which consume aquatic
animals from freshwater habitats.
5. Assessing Pesticide Concentrations in Fish Tissues for Human Consumption

It is possible to use KABAM to derive pesticide concentrations in edible tissues of fish that are
relevant to assessments of pesticide  risks to  human health.   Current default values described
above for %  lipid content  of fish  applies to  the whole fish;  however,  not all  fish tissues are
consumed by humans.  Therefore, it  is necessary  to modify the  output of the  pesticide tissue
concentration to  account for a  lower % lipid  composition of edible tissues.  This can  be
accomplished by  entering in  all the relevant default input parameters for KABAM as defined
above. It may be  necessary to explore different body weights of the large fish, based on those
that would be expected to be consumed by humans.

The relevant output is the lipid normalized concentration of the pesticide in the large fish (Table
11).   This value  can be  converted  to the total pesticide  concentration in edible tissues  by
multiplying by the % lipid content of the edible tissues. The default value  for  lipid content in
edible tissue of the  large fish is 3%, based on USEPA 2003. The resulting value represents the
concentration of pesticide in  fish tissue (in |ig/kg-ww) potentially consumed by humans. This
value can then  be used in conjunction with fish  consumption rates to characterize risks of a
pesticide to humans consuming contaminated fish.
6. Model Assumptions, Limitations, and Uncertainties

There are several key assumptions and resulting uncertainties associated with modeling pesticide
concentrations in tissues  of aquatic organisms.  The assumptions involve the equations of the
model itself and the parameterization of those underlying equations. Appendix A describes the
assumptions associated with the equations of the bioaccumulation model. In order to explore
uncertainties associated with  specific parameters and their  influences on  model outputs,  a
sensitivity analysis was conducted (see section A7 of Appendix A). This was used to define the
parameters that have the greatest influence on model outputs (e.g., K0w, water column, and pore
water EECs). Appendices B and C describe the parameterization of the  model, including the
associated assumptions.

In addition,  the use of PRZM/EXAMS for deriving EECs in the surface and pore waters of the
aquatic  ecosystem introduces  the  assumptions  and uncertainties  associated with PRZM and
EXAMS to KABAM.

One major assumption associated  with KABAM concerns the model's assumed steady state.
Given  the  episodic nature of pesticide  applications,  sporadic  peak  exposures  to aquatic
organisms would be expected.  For a chemical with a Log KOW of approximately 5, comparison
                                        24 of 123

-------
of the fish tissue EECs predicted using the steady state and dynamic bioaccumulation modeling
with PRZM/EXAMS/Arnot and Gobas indicates predictions are similar (USEPA 2008b) when a
60-d average was selected  for water and sediment concentrations as input to the  steady state
model. This  result  suggests that steady-state  bioaccumulation modeling can provide useful
predictions of bioaccumulation potential even with highly dynamic exposures, provided proper
consideration of the averaging period  associated with water and sediment  concentrations is
made.

As discussed above, in using KABAM with default settings, it is assumed that the elimination of
the pesticide from aquatic organisms through metabolism does not occur, i.e., the metabolism
rate constant (kM) = 0. In cases where pesticide metabolism does occur, this could overestimate
pesticide bioaccumulation. Appendix H of this guide provides methods for estimating kM for fish
using empirical data provided for specific chemicals (from BCF studies).  This approach can be
used to characterize effects of metabolism, but should be used with caution.

The Arnot and Gobas (2004) model is generally  appropriate for chemicals with Log K0w value
>4 to < 8. Uncertainty increases as the value increases  above 8 because the model has generally
been validated using chemicals with Log K0w values within the range of 4 - 8.  Making
predictions for a chemical with a Log KOW > 8 leads to uncertainty in model outputs because
predictions are based upon extrapolations in its subroutines.

For chemicals with Log Kow < 4, exposure from  food becomes insignificant because uptake and
depuration across the gills controls the residue in the organism.  Thus, there is no need to run a
food web model for these chemicals. In these cases, available BCF data are sufficient to predict
residues in the aquatic species.

It  is assumed that there is  no  predation within a trophic level of the aquatic food web (e.g.,
medium  fish cannot prey upon medium fish). It is also assumed that mammals and birds only
consume organisms from the aquatic system.
                                       25 of 123

-------
Appendix A. Description of Bioaccumulation Model

The bioaccumulation portion of KABAM is based on the model published by Arnot and Gobas
(2004).  The purpose of this  model is to estimate  chemical concentrations  (CB) and  BCF and
BAF values for aquatic ecosystems.   Conceptually, each aquatic organism  is assumed to be a
single compartment. Chemicals enter the organism through respiration and diet and  leave the
organism through respiration  and fecal egestion.  The chemical concentration in the  organism
can also be influenced by the growth of the  organism as well as metabolism of the chemical
within the organism. These processes that define uptake and loss of the chemical from aquatic
organisms are described by rate constants and are incorporated into one equation that is used to
define the concentration of the chemical in organism tissues (Equation Al, see  Table Al). As
uptake constants (i.e., ki and kD) increase, so does the estimated pesticide concentration in an
organism.  As elimination  constants  increase (i.e., k2, kE, kG and kM), estimated  pesticide
concentrations in an organism decrease. However,  for respiration and diet, processes of uptake
and elimination are  linked.  Therefore, factors that would influence uptake constants would also
influence elimination constants, so these cannot be considered independently. In addition, as the
freely dissolved fraction of pesticide in the water (
-------
phytoplankton, which contain pesticide residues.  Tissue residues are calculated for the next five
trophic levels based on their diets of organisms from lower trophic levels.
Table Al. Equation Al, calculation of pesticide tissue residue (CB) for single trophic levels and its associated
parameters (Arnot and Gobas 2004).

FI 11 r *i*K**c+i,*c^)+*D*Stf *ca)
/C fj ~T~ /C p ~T~ /C f-i ~T~ /C j.f

Parameters:
Symbol
CB
CBD
CBR
CDl
cs
CWDP
CWTO
ki
k2
kD
kE
kG
kM
mo
mp
Pi
0
Definition
pesticide concentration in the organism
pesticide concentration in the organism originating from uptake
through diet; CBD = CB when ki = 0
pesticide concentration in the organism originating from uptake
through respiration; CBR= CB when kD = 0
concentration of pesticide in i (prey item)
concentration of the chemical in sediment (dry weight of sediment)
freely dissolved pesticide concentration in pore water of sediment
total pesticide concentration in water column above the sediment
pesticide uptake rate constant through respiratory area (i.e., gills,
skin)
rate constant for elimination of the pesticide through the respiratory
area (i.e., gills, skin)
pesticide uptake rate constant for uptake through ingestion of food
rate constant for elimination of the pesticide through excretion of
contaminated feces
organism growth rate constant
rate constant for pesticide metabolic transformation
fraction of respiratory ventilation involving overlying water
fraction of respiratory ventilation that involves pore-water of
sediment
fraction of diet containing i (prey item)
fraction of the overlying water concentration of the pesticide that is
freely dissolved and can be absorbed via membrane diffusion
Value
calculated
calculated
calculated
calculated
Equation A4
input parameter (from
PRZM/EXAMS)
input parameter (from
PRZM/EXAMS)
Equation A5
Equation A6
Animals: Equation A8;
Phytoplankton: 0
Animals: Equation A9;
Phytoplankton: 0
Animals: Equation A7;
Phytoplankton: 0.1
0
1 -mp
<5%;
0 for organisms with
no contact with pore
water
user defined
Equation A2
Units
g/kg (wet
weight)
g/kg (wet
weight)
g/kg (wet
weight)
g/kg (wet
weight)
g/(kg (dry)
sediment)
g/L
g/L
L/kg*d
d-1
kg food/
(kg org*day)
d-1
d-1
d-1
none
none
none
none
                                        27 of 123

-------
A.I. Calculation of Fraction of Chemical in the Water Column That Is Freely Dissolved
Aquatic ecosystems contain organic  matter suspended  in the water column.  This suspended
organic matter is defined as  dissolved organic carbon  (DOC) and particulate organic carbon
(POC).   It is assumed that once chemicals partition to organic carbon,  they  are no  longer
bioavailable to aquatic organisms.  The fraction of chemical in the water column that is freely
dissolved (O), and thus bioavailable for  uptake by aquatic animals is estimated according to
Equation A2 (Table A2). This equation assumes that equilibrium exists between the pesticide
concentration in the water and in the organic carbon in the water column.  Equation A2 assumes
that partitioning between POC and water and DOC and water can be related to partitioning of the
chemical between octanol  and  water.  These relationships are defined using proportionality
constants (apoc and (XDOC) that are defined from the scientific literature.
Table A2. Equation A2, derivation of available pesticide fraction in water (O) and its associated
parameters (Arnot and Gobas 2004).

Eq 1 O 
\ + (X * a *K } + (X * a *K }
L-r{sipoc u-poc ^-ow ) T V^DOC "DOC ^-ow )

Parameters:
Symbol
XPOC
XDOC
KOW
O
ctpoc
Cfooc
Definition
concentration of particulate organic carbon in water
concentration of dissolved organic carbon in water
octanol water partition coefficient
fraction of the overlying water concentration of the pesticide that is
freely dissolved and can be absorbed via membrane diffusion
Proportionality constant to describe the similarity of phase partitioning
of POC in relation to octanol
Proportionality constant to describe the similarity of phase partitioning
of DOC in relation to octanol
Value
user defined
user defined
user defined
calculated
0.35
0.08
Units
kg/L
kg/L
none
none
none
none
The Arnot and Gobas (2004) approach for calculating the fraction of bioavailable pesticide in the
water column (Equation A2) is different  from the approach used in EPA's Exposure Analysis
Modeling System (EXAMS) (Equation A3, Table A3). The major difference is that the approach
employed by EXAMS accounts for decreases in bioavailable pesticide in the water column due
to sorption to  biota.  The values for aDoc for the two approaches are also slightly different.
Despite these different approaches,  for chemicals with Log K0w values 4-8, the fractions of
bioavailable pesticide in  the water  column estimated by the  two approaches differ by <0.02
(Figure Al). Therefore, utilizing water column  EECs generated by EXAMS is still  consistent
with the results that are generated using the approach described by Arnot and Gobas (2004).
                                       28 of 123

-------
Table A3. Equation A3, derivation of available pesticide fraction in water (F) by EXAMS and its associated
parameters.
E
7 13 F
*"""' * \ + (X *a *K } + (X * a *K } + (x * 0436* pr0-907 }
L^^POC UPOC ^OWJ^^DOC "DOC J^OW)^\ ^ biota ^.HJU ** OW )
Parameters:
Symbol
F
KOW
OCSS
Xss
XPOC
XDOC
Xbiota
ctpoc
Cfooc
Definition
fraction of the overlying water concentration of the pesticide that is
freely dissolved and can be absorbed via membrane diffusion
octanol water partition coefficient
percent organic carbon in suspended sediment
concentration of suspended sediments in water column
concentration of paniculate organic carbon in water
concentration of dissolved organic carbon in water
concentration of biota in water
Proportionality constant to describe the similarity of phase partitioning
of POC in relation to octanol
Proportionality constant to describe the similarity of phase partitioning
of DOC in relation to octanol
Value
calculated
user defined
user defined
user defined
xss*ocss
user defined
user defined
0.35
0.074
Units
none
none
%
kg/m3
kg/m3
kg/m3
kg/m3
none
none
1.0 i


0.9 -


0.8 -
         o
         o
         I -7 H
           0.6 -
           0.5 -
           0.4 -
         re
         ^ 0.2 ^
         ro
         o
           0.1  -
           0.0
              4.0
           4.5
                                                         . . . . Arnot and Gobas (2004)
                                                         	EXAMS
5.0
5.5       6.0
      Log Kow
6.5
7.0
7.5
8.0
 Figure Al. Fraction of bioavailable pesticide in water column estimated using approaches of Arnot and
Gobas (2004) and EXAMS. Values for OCSS, Xss, XDOC and Xbiota are consistent with OPP standard pond
                                    scenario used in EXAMS.
                                           29 of 123

-------
A.2. Calculation of Chemical Concentration in Sediment
Since it is possible for aquatic organisms to be exposed to chemicals through consumption of
contaminated sediment, it is necessary to estimate the concentration of the chemical of concern
in the  sediment. This  is accomplished using Equation A4 (Table A4),  which  uses  the
concentration of the chemical in the  pore  water, the chemical KOC, and the organic carbon
content of the sediment.  This approach is consistent with EXAMS for calculating the fraction of
pesticide sorbed to sediment in the benthic column.
Table A4. Derivation of pesticide concentration in the solid portion of the sediment (Cs) (Arnot and Gobas
2004).
Eq.A4 CS=CSOC*OC
Where: Csoc =CWDP* Koc
Parameters:
Symbol
Cs
Csoc
CWDP
KOC
OC
Definition
concentration of the chemical in sediment (dry weight of
sediment)
normalized (for OC content) pesticide concentration in sediment
freely dissolved pesticide concentration in pore water
organic carbon partition coefficient
percent organic carbon in sediment
Value
calculated
calculated
input parameter
(from
PRZM/EXAMS)
user defined
user defined
Units
g/(kg(dry)
sediment)
g/(kgOC)
g/L
L/kgOC
%
A.3. Calculation of Respiration Uptake (ki) and Elimination (k2) Rate Constants

The respiratory uptake constant (ki) is calculated differently for phytoplankton (Equation A5.1)
and for animals  (Equation A5.2).  For phytoplankton,  ki is dependent upon the K0w of the
chemical as well as 2 constants (A and B) that describe  chemical uptake resistance through the
aqueous and organic phases (respectively) of the plant.  If A and B  are kept constant at 6xlO"5
and 5.5, respectively  as recommended by  Arnot and Gobas (2004), ki for  phytoplankton
increases with increasing K0w, ranging by a factor of 10 from Log K0w 4-8 (Figure A2).

For animals, ki is dependent upon the chemical uptake efficiency of the gills, the ventilation rate,
and the body weight of the organism.  The uptake efficiency of the gills is determined by the
KQW of the chemical, while the ventilation rate of the organism is determined by an allometric
equation that is influenced by the body weight of that organism and the concentration of oxygen
in the water column (Cox) (Table A5). When C0x and organism body weight are kept constant,
Log KOW has little effect on the ki for animals (0.8% change from Log KOW 4-8).  When Log
KQW and body weight are kept constant, Cox strongly affects ki. The value of ki decreases by
50%, as the Cox value increases from 5 to 10 mg/L (Figure A3). The value of ki is also strongly
influenced by the body weight of the organism,  with decreasing ki observed with  increasing
                                       30 of 123

-------
                                                                 -v-7
body weight.  As the body weight of the organism increases from 1x10" to 10 kg (a relevant
range for aquatic organisms, see Appendix C), the ki value spans 3 orders of magnitude (Figure
A3).
Table A5. Equations associated with the derivation of pesticide clearance through the respiratory (gill)
system (kt) and associated parameters (Arnot and Gobas 2004).


A+ /K:
/ Kow
E *G
J?n A S ^ T^CIT nnimnlv 'If 
w
" B
f V1
Where- F  1 9S 1
I Kow j
fw0-65"]
rL - 1400* B
^v ~ 1^uu r
\ ^ox )


Parameters
Symbol
A
B
cox
Ew
Gv
ki
KOW
WB
Definition
constant related to the resistance to pesticide uptake through the
aqueous phase of plant
constant related to the resistance to pesticide uptake through the
organic phase of plant
concentration of dissolved oxygen
pesticide uptake efficiency by gills (fraction)
ventilation rate of fish, invertebrates, zooplankton
pesticide uptake rate constant through respiratory area (i.e., gills, skin)
octanol water partition coefficient
wet weight of the organism
Value
6.0xlO'5 (default)
5.5 (default)
User input
calculated
calculated
calculated
user defined
user defined
Units
days
days
(mg 02)/L
none
L/d
L/kg*d
none
kg
                                       31 of 123

-------
      1.E + 05
      1.E + 04
      1.E + 03
      1.E + 02
      1.E + 01
      1.E + 00
            4.0       4.5       5.0       5.5      6.0       6.5
                                               LogKow

Figure A2. Relationship between Kow and kt for phytoplankton.
   1.0E+07

   1.0E+06

   1.0E+05 -

   1.0E+04
  
   1.0E+03 -

   1.0E+02

   1.0E+01

   1.0E+00
7.0
7.5
8.0
           0           2           4           6           i
                                   Cox (mg 02/L)
Figure A3. Relationship between C0x WB and kt for aquatic animals.
 10
         WB = 1 kg
         WB = 1e-1 kg
         WB = 1e-2kg
         WB = 1e-3kg
          -WB = 1e-4kg
         WB = 1e-5kg
         WB = 1e-6kg
         WB = 1e-7kg
                                          32 of 123

-------
The elimination rate constant for the respiratory system (k2) is related to the respiratory uptake
constants (ki). This is because both  constants are influenced by the same processes related to
respiration.  The value of k2 is also influenced by the partitioning of the chemical between the
organisms and the water.  The organism-water partition coefficient (KBw) is determined by the
body composition of the organism (i.e., lipid, NLOM, and water) and the K0w of the chemical. It
is assumed that the partitioning of the chemical between lipid and water is directly related to the
octanol-water partition coefficient.  It is also assumed that the chemical partitioning between
NLOM and water can be related to the octanol-water partition coefficient using a proportionality
constant (P) (Equation A6, Table A6).
Table A6. Equations involved in the derivation of the respiratory elimination rate constant (k2) and
associated parameters (Arnot and Gobas 2004).

Fa 46 k - :
Where : KBW = VLB * Kow + VNB*j3*K0
w+VwB
Parameters:
Symbol
kl
k2
KBW
KOW
VLB
V-
VWB
P
Definition
pesticide uptake rate constant for chemical uptake through
respiratory area (i.e., gills, skin, membrane permeation)
rate constant for elimination of the pesticide through the
respiratory area (i.e., gills, skin, membrane permeation)
organism-water partition coefficient (based on wet weight)
octanol water partition coefficient
lipid fraction of organism
NLOM (Non Lipid Organic Matter) fraction of animals, NLOC
(Non Lipid Organic Carbon) of plants
water content of the organism
proportionality constant expressing the sorption capacity of
NLOM or NLOC to that of octanol
Value
calculated
(Equation 5)
calculated
calculated
user defined
user defined
user defined
user defined
Phytoplankton:
0.35;
Animals:0.035
Units
L/kg*d
^
none
none
(kg lipid)/
(kg organism
wet weight)
kg NLOM/
(kg organism
wet weight)
kg water/
(kg organism
wet weight)
none
Elimination rate  constants  for phytoplankton  (k2) are calculated using  seven  parameters,
including: K0w, VLB, VNB, VWB, A, B, and P (defined above in Table A6). The parameters A, B
and P are all constants.  If the other parameters are considered in terms of ranges applicable to
KABAM, KOW has the greatest influence on the determination of k2 for phytoplankton.  When
Log KOW values are changed from 4 to 8, the k2 value decreases by three orders of magnitude
(Figure A4).  The lipid fraction of the organism (VLB) and the non-lipid organic carbon fraction
(VNB) influence k2, with decreases in k2 observed as VLB and VNB increase. These two parameters
are related. As VLB decreases, an increase in VNB has a greater effect  on k2.  Likewise, as VNB
                                        33 of 123

-------
decreases, an increase in VLB has a greater effect on k2 (Figure A4).  When the lipid fraction of
organism (VLB) is increased from 0.5 to 3%, the elimination rate constant through respiration,
i.e., k2, decreases by >30% (Figure A5).  A change in VNB from 5% to 20% results in a decrease
in the value of k2 that is >50% (Figure A5).   The water content (Vws) of an organism has a
negligible (<0.5%) effect on k2.
     1.0E + 01 n
     1.0E+00 -
     1.0E-01
     1.0E-02 -
     1.0E-03
                       *
             ''i  ii
                      M
                                X
                                *
      VLB=2.5%
      VLB=2.0%
       VLB=1.5%
       VLB=1.0%
     xVLB=0.5%
x I
  *  *
    MI
          4.0       4.5       5.0       5.5       6.0       6.5       7.0       7.5       8.0
                                            LogKow

Figure A4. Influence of Log KQW on k2 (for phytoplankton) at different % lipid (VLB) (with VNB = 8%).
                                        34 of 123

-------
     1.0E+01
. . . -VNB=20%
	VNB=15%
	VNB= 10%
	VNB=5%
     1.0E+00
          0.0%
                      0.5%
                                                                                    3.0%
Figure A5. Influence of % lipid (VLB) on k2 (for phytoplankton) at different % NLOM (V^) at Log Kow 4.
Note that for all Log Kow values from 4-8, the curves follow a similar trend for different VLB and VNB, but
differ in magnitude of k2.

To determine elimination constant from respiration (k2) for animals, seven input parameters are
required: K0w, WB, VLB, VNB, VWB, Cox, and  P  (all  of which are  defined in Table A6). The
parameter P is a constant representing the proportionality of the sorption of a chemical to NLOM
to the KOW of that chemical. If the other parameters are considered in terms of ranges  applicable
to KABAM, the octanol-water partition coefficient (Kow) and the water content of the organism
(VWB)  have the greatest influence on the determination of k2.   When Log K0w values are
changed from 4 to 8, the k2 value decreases by 4 orders of magnitude (Figures A6 and A7).
Body weight also influences k2, with k2 values decreasing by 2 orders of magnitude as the body
weight is increased from IxlO"7 to 1 kg (this is a range considered relevant to aquatic animals,
see Appendix C for more information) (Figure A6). The lipid fraction of the organism (VLB) and
the non-lipid organic matter fraction (VNB) influence k2, with decreases in k2 observed as these
two values increase. When VLB is increased from 1 to 5%, k2 decreases by 84% (Figure A7).
When VNB is increased from 15 to 40%, a 14% decrease in k2 is  observed.  The water content
(VWB) of an organism has a negligible (<0.5%) effect  on k2. Increasing the concentration  of the
dissolved oxygen (Cox) value from 5 to 10 mg/L results in a decrease of 50% in the value of k2.
                                        35 of 123

-------
      1.0E+02
                                     WB = 10 kg
                                      WB = 1  kg
                                      WB = 1e-1 kg
                                      WB = 1e-2 kg
                                     xWB = 1e-3 kg
                                      WB = 1e-4 kg
                                     + WB = 1e-5kg
                                     -WB = 1e-6
                                     -WB = 1e-7 kg
       1.0E-06
             4.0
Figure A6. Influence of Log KQW on k2 (for animals) at different body weights (WB) (with VLB = 5%,
20%, VWB = 75% and Cox = 10mg/L).
      1.0E+00
         , y  y
           y ^If
ioE-01 ^t^ x*x
                'x;
                ''
1.0E-02
       1.0E-03 -
       1.0E-04
       1.0E-C
4.0
                            v
                         m   v
                          A  x
                            B >   v * w
                               ?   v
                                A   '
                                                     >    y
                                                     1   ' V
                                                      A  x\ .
                           5.0
   6.0
Log Kow
                                                 7.0
8.0
                                         VLB = 5%
                                        VLB = 4%
                                         VLB = 3%
                                         VLB = 2%
                                        xVLB = 1%
Figure A7. Influence of Log KQW on k2 (for animals) at different % lipid composition (VLB) (with WB = 1 kg,
VNB = 20%, VWB = 75% and Cox = 10 mg/L).
                                         36 of 123

-------
A.4. Calculation of Growth Rate Constant
Equations A7.1 and A7.2 provide an approximation of growth of aquatic organisms based on
weight and temperature (Table A7).  Comparing the results of the two equations indicates that
higher temperatures result in higher  growth rate constant (kG) values (Figure A8).  With both
equations, as body weight increases, ko decreases. There is some uncertainty associated with
these equations, since growth rate can be influenced by additional factors, including species and
prey availability.   For KABAM,  it  is  assumed that if the water temperature (T) <  17.5 C
(midpoint between 10 and 25C), equation A.7.1 is used and if T >  17.5 C, equation A.7.2 is
used.
Table A7. Equations involving the derivation of the growth rate constant (ko) and associated parameters
(Arnot and Gobas 2004).
Eq.Al. 1 kG = 0.0005 * WB02 (T7  10 C)
Eq.Al.1 kG = 0.0025 1 * W~  2 (T  25 C}
Parameters:
Symbol
kG
T
WB
Definition
organism growth rate constant
temperature
wet weight of the organism
Value
calculated
user defined
user defined
Units
d'1
C
kg
      1.0E + 00 n
      1.0E-01
      1.0E-02
      1.0E-03 -
      1.0E-04
	T=10oC
	T=20oC
             -7       -6        -5       -4       -3        -2       -1
                                           Log WB(in kg)

Figure A8. Relationship between body weight (WB) and UG at 2 different temperatures.
                                        37 of 123

-------
A.5. Calculation of Dietary Uptake (kD) Rate Constant

In determining uptake and elimination rate constants related to dietary sources of chemicals (kD
and ks, respectively), it is assumed that aquatic organisms are represented by a 2-phase model
that  includes the gastrointestinal tract (GIT)  of the organisms  and the organism itself.  Since
phytoplankton do not consume other organisms, a dietary uptake constant (ko) is not a relevant
rate  constant, and  the elimination rate  constant  due to fecal  elimination (kE) is  considered
insignificant in plants.  Therefore, ko and kE are only calculated for animals.

The  dietary uptake constant  for a  chemical  in animals is influenced by the weight  of the
organism, the feeding rate of the organism (Go), and the dietary pesticide transfer efficiency
(ED).  The feeding rate is  different for  filter feeders compared to other aquatic  organisms.
Empirical dietary pesticide transfer efficiency  (Eo) values vary from 0-100%. Variability in ED
has been attributed  to various factors, including sorption coefficients of chemicals, composition
of diet, and digestibility of diet. Based on several different observations, it is assumed by Arnot
and Gobas (2004) that this value can be related to K0w (Equation A8, Table A8).
                                        38 of 123

-------
Table A8. Equations involving the derivation of the pesticide clearance rate constant through diet (kD) and
associated parameters (Arnot and Gobas 2004).

f~<
!? AQ lr  JT * D

J^y.y-iu n,D - J^D
"B
Where : ED = (3.(M(T7 * Kow + 2.0)^
For animals (except filter feeders) : GD = 0.022 * W '85 * exp(0.06 * T)
For filter feeders : GD = Gv * Css * a
fw0-65}
r< - i /mn*l " B \

^v A" r
V ^ox )

Parameters:
Symbol
cox
Css
ED
GD
Gv
kD
KOW
T
WB
0
Definition
concentration of dissolved oxygen
concentration of suspended solids
dietary pesticide transfer efficiency
feeding rate of organism
ventilation rate of gills
pesticide uptake rate constant for uptake through
ingestion of food
octanol water partition coefficient
temperature
wet weight of the organism
efficiency of scavenging of particles absorbed from water
Value
user defined
user defined
calculated
calculated
calculated
calculated
user defined
user defined
user defined
100
Units
(mg 02)/L
kg/L
%
kg/d
L/d
kg food/(kg
org*day)
none
C
kg
%
For aquatic organisms (non-filter feeders), the dietary uptake constant (ko) is derived using 3
input parameters: octanol-water partition coefficient (Kow), the weight of the organism (WB),
and temperature (T). For chemicals with Log K0w ranging 4-5.5, changes in Log K0w cause
little (<4%) effects to the value of ko; however, for chemicals with Log KOW <5.5, increases in
Log KOW can result in decreases in this value up to an order of magnitude (Figure A9).  Increases
in WB from  IxlO"7  to  1  kg result in an  order of  magnitude decrease in  kD (Figure A9).
Temperature  also affects kD, with an observed increase in kD of 65% when the temperature is
increased from 2.5 to 20C (Figure A10).
                                       39 of 123

-------
      1.0E+00
      1.0E-01
      1.0E-02
      1.0E-03
WB=10kg
WB=1 kg
WB=1e-1 kg
WB=1e-2kg
WB=1e-3kg
WB=1e-4kg
WB=1e-5kg
WB=1e-6kg
WB=1e-7kg
      1.0E-04
Figure A9. Relationship between Log KQW and kD (for non-filter feeders) at different body weights (WB)
(T=10C).
                                             40 of 123

-------
     1.0E+00
      1.0E-01
  9   1.0E-02
      1.0E-03
           4.0      4.5      5.0      5.5     6.0     6.5      7.0      7.5     8.0
-T=20oC

-T=15oC
 T=10oC
-T=5oC
      1.0E-04
Figure A10. Relationship between Log KQW and kD (for non-filter feeders) at different water temperatures
(WB=1 kg).


For filter feeders, the dietary uptake constant (ko) is derived using five input parameters: KOW,
WB, Cox, CSs and o (all of which are defined in Table A.8). For chemicals with Log K0w values
ranging 4-5.5, changes in Log K0w cause little (<5%) effects to the value of kD; however, as the
Log KOW  increases from  6 to 8, the kD for filter feeders decreases by an order of magnitude
(Figure All).   Available National Water Quality  Assessment (NAWQA)  data  for  streams
indicate that suspended sediment  concentrations range  1-281  mg/L  (USGS 2008b).   If the
concentration of suspended sediments (Css) is increased from IxlO"6 to IxlO"4 kg/L, kD for filter
feeders increases by 2 orders of magnitude (Figure A12). Increases in WB from IxlO"4 to IxlO"2
kg (which is a reasonable range  of weights for filter feeders; see Appendix C) result in  an 80%
decrease in kD (Figure All).  Increases in oxygen concentration  (C0x)  from 2  to 8 mg/L results
in a decrease in kD of 80% for filter feeders. Changes  in the scavenging efficiency (o)  of filter
feeders result in proportional changes to the ko value. For example, a 25% decrease in o results
in a 25% decrease in ko.
                                        41 of 123

-------
      1.0E-01
      1.0E-02
      1.0E-03 -
      1.0E-04
                   	WB = 1e-4 kg
                         WB = 1e-3kg
                   .  . . _WB = 1e-2
             4.0       4.5       5.0        5.5        6.0       6.5       7.0       7.5       8.0
                                                  Log Kow

Figure All.  Relationship between Log KQW and filter feeder kD with different WB values.
      1.0E+00
       1.0E-01 -
       1.0E-02
       1.0E-03
       1.0E-04
       1.0E-05
-Css = 1e-6 kg/L
 Css = 1e-5 kg/L
 Css = 1e-4 kg/L
              4.0       4.5       5.0
                 5.5       6.0       6.5
                        Log Kow
7.0      7.5       8.0
Figure A12.  Relationship between Log KOW and filter feeder kD with different Css values.
                                            42 of 123

-------
A.6. Calculation of Dietary Elimination (kE) Rate Constant
The rate constant for elimination of the pesticide through excretion of contaminated feces (kE) is
calculated using the fecal egestion rate (Op), the dietary pesticide transfer efficiency (ED), the
partition coefficient of the pesticide between the gastro-intestinal tract and the organism (KGB),
and the body weight of the organism (We) (Equation A9, Table A9). For filter-feeding and non-
filter feeding aquatic animals, kE  is  calculated in a similar manner, with the exception of the
method of calculating the feeding  rate of an organism (Go). An order of magnitude increase in
either GF, ED, or KGB results in an order of magnitude  increase in fecal egestion rate constant
(k). An order of magnitude increase in WB results in an order of magnitude decrease in kE.
Effects  of changes in individual input parameters used to derive GF, ED, and KGB are explored
below.
Table A9. Equations involving the derivation of the fecal elimination rate constant
parameters (Arnot and Gobas 2004).
* KGB
E F D W
'' B
Where : ED = (3.(M(T7 * Kow + 2.0)^
* P-'K-^'^K-^
LG ( \ ( \ ( \
(\-P }*y
Y _ V1 fcW/ y ND
TS  V SW ) *WD
(kE) and associated

VV \j 1 1 1 ;k T 7" 11 I ;k T 7" 11 I ;k T 7"
I I 	 r* 1 yf- 1 / _i_ I 1 	 r? 1 ^ I/  \ I I 	 ? If/
GF=[(l-e)*FLD+(l-eJV)*FM3+(l-%)*FWD]*GD
For animals (except filter feeders) : GD = 0.022 * W5 * exp(0.06 * T)
For filter feeders : GD = Gv * Css * a
(w-65^
ri - 1400* B
(jv - 14UU
V Uox )
Parameters:
Symbol Definition Va
Cox concentration of dissolved oxygen calcu
Css concentration of suspended solids user d


lue Units
lated (mg O2)/L
efined kg/L
                                        43 of 123

-------
ED
GD
GF
Gv
kE
KGB
KOW
T
VLB
VLD
VLG
VNB
VND
VNG
VWB
VWD
VWG
WB
P
L
EN
w
0
dietary pesticide transfer efficiency
feeding rate of organism
egestion rate of fecal matter
ventilation rate of gills
rate constant for elimination of the pesticide through excretion
of contaminated feces
partition coefficient of the pesticide between the gastro-
intestinal tract and the organism
octanol water partition coefficient
temperature
lipid fraction of organism
overall lipid content of diet
lipid content in the gut
NLOM (Non Lipid Organic Matter) fraction of animals,
NLOC (Non Lipid Organic Carbon) of plants
overall NLOM content of diet
NLOM content in the gut
water content of the organism
overall water content of diet
water content in the gut
wet weight of the organism
proportionality constant expressing the sorption capacity of
NLOM to that of octanol
dietary assimilation rate of lipids
dietary assimilation rate of NLOM
dietary assimilation rate of water
efficiency of scavenging of particles absorbed from water
calculated
calculated
calculated
calculated
for animals:
calculated
for plants: 0
calculated
user defined
user defined
user defined
user defined
calculated
user defined
user defined
calculated
user defined
user defined
calculated
user defined
0.035 for animals
fish: 92%;
aquatic inverts: 75%;
zooplankton: 72%
fish: 60%;
aquatic inverts: 75%;
zooplankton: 72%
freshwater
organisms: 25%
100
%
kg/d
(kg feces)/(kg
organism)*d
L/d
d-1
none
none
C
(kg lipid)/
(kg organism
wet weight)
kg/kg
(kglipid)/(kg
digesta wet
weight)
kg NLOM/
(kg organism
wet weight)
kg/kg
(kg
NLOM)/(kg
digesta wet
weight)
kg water/
(kg organism
wet weight)
kg/kg
(kg water)/(kg
digesta wet
weight)
kg
none
%
%
%
%
44 of 123

-------
       A.6.1. Parameters Affecting GF

The fecal egestion rate GF is calculated using the feeding rate of the organism (GD; kg/day), the
dietary assimilation rates for lipids, NLOM, and water (SL, SN and s\v, respectively) as well as the
contents of the diet (VLD, VND, and VWD)-

For non-filter feeders, the feeding rate (GD)  is calculated using body weight and temperature.
Generally, as body weight and temperature  increase, so does  the feeding rate  of the aquatic
animal (Figure A. 13). An  order  of magnitude increase in  body weight leads to an order of
magnitude increase in the  feeding  rate of the organism. An order of magnitude increase in
temperature leads to a 40% increase  in the feeding rate of non-filter feeders.

1.0E + 00 n
1.0E-01 -
1.0E-02 -
1.0E-03 -
1.0E-04 -
1.0E-05 -
1.0E-06 -
1 np.07
	
	
	 WB = 1e-6 kg
- - - -WB = 1e-3kg
	 WB = 1 e2 kg
 *""*"


	 	 	
                                    10          15         20
                                         Temperature (oC)
25
30
Figure A13.  Relationship between temperature and GD (non-filter feeders) with different WB values.
For filter feeders, the feeding rate (Go) is calculated using four parameters: the concentration of
dissolved oxygen in the water (Cox), body weight (We), the concentration of suspended solids
(Css), and the scavenging efficiency of particles absorbed from water (o).  As with non-filter
feeders, increases in body weight of filter feeders leads to increases in GD (Figures A14-A16).
An order of magnitude increase in body weight leads to an 80% increase in feeding rate (Go). An
increase in dissolved oxygen in the water (Cox) from 2 to 10 mg/L results in a decrease in GD of
80% (Figure A14). Decreases in scavenging efficiency lead to proportional decreases in GD, with
every 10% decrease in scavenging efficiency (o), leading to a 10% decrease in GD (Figure A15).
An order of magnitude increase in the concentration of suspended solids (Css) leads to an order
of magnitude increase in GD (Figure A16).
                                        45 of 123

-------
      1.0E-04
      1.0E-05 -
   Q
   o
      1.0E-06
      1.0E-07
                                                                          	WB = 1e-4kg

                                                                          	WB = 1e-3kg

                                                                          	WB = 1e-2 kg
               2345678

                                               Cox (mg/L)


Figure A14. Relationship between Cox and GD (filter feeders) with different WB values.
             10
      1.0E-04
      1.0E-05
   5
   "5>

   Q
   o
      1.0E-06 -
      1.0E-07
	WB = 1e-4 kg

	WB = 1e-3kg

	WB = 1e-2kg
               0      0.1     0.2     0.3     0.4      0.5      0.6     0.7     0.8     0.9      1

                                           scavenging efficiency


Figure A15. Relationship between scavenging efficiency and GD (filter feeders) with different WB values.
                                             46 of 123

-------
      1.0E-03
      1.0E-04
      1.0E-05 -
   Q
   o
      1.0E-06
      1.0E-07
                          	WB = 1e-4 kg
                          	WB = 1e-3kg
                          	WB = 1e-2kg
          1.00E-06
 1.00E-05
Css (kg/L)
1.00E-04
Figure A16.  Relationship between  concentration of suspended solids (Css) and GD (filter feeders) with
different WB values.
The fecal egestion rate (GF) is calculated using the following parameters: SL, SN, s\v, VLD , VND,
VWD as well as GD, which is discussed above. When the default dietary assimilation rates for
lipids,  NLOM, and water (SL,  SN, and EW, respectively) are used,  changes in lipid, NLOM and
water composition of the diet  (VLD, VND and VWD, respectively) have little effect on GF, when
compared to effects of GD on GF (Figure A. 17).  This indicates that as the feeding rate of an
animal (Go) increases so does its fecal egestion rate (GF), while the composition of the animal's
diet has little effect on the fecal egestion rate.
                                         47 of 123

-------
      1.0E+00
                                                   ----- VLD 1%, VND 10%, VWD89%
                                                         VLD 1%, VND 20%, VWD 79%
                                                         VLD 10%, VND 10%, VWD 80%
                                                         VLD 10%, VND 20%, VWD 70%
           1.0E-05        1.0E-04        1.0E-03       1.0E-02
                                           GD (kg/day)
1.0E-01
1.0E+00
Figure A17. Relationship between GD and GF with different lipid, NLOM, and water compositions in the diet
(VLD VND, and VWD? respectively). Dietary assimilation rates for lipids, NLOM, and water are set to default
values used to represent fish (see Table A9).

When setting the lipid, NLOM, and water composition (VLD, VND and VWD, respectively) of diet
equal for fish, aquatic invertebrates, and zooplankton, the differences in dietary assimilation rates
for lipids, NLOM, and water (SL, SN, and s\v, respectively) of these three groups of animals have
little effect on the fecal egestion rate (Op), when compared to effects of the feeding rate (Go) on
GF (Figure A. 18).  As with the composition of the diet, changes in GD result in greater effects on
GF when compared to changes in the assimilation efficiencies of lipid, NLOM, and water.
                                        48 of 123

-------
1.0E+00

 1.0E-01

 1.0E-02 -

 1.0E-03 -

 1.0E-04

 1.0E-05 -

 1.0E-06

 1.0E-07

 1.0E-08
                                                                       -Fish
                                                                       Aquatic invertebrates
                                                                 	Zooplankton
            1.0E-07    1.0E-06    1.0E-05    1.0E-04    1.0E-03    1.0E-02    1.0E-01    1.0E+00

                                                  GD

Figure A18. Relationship between GD and GF with dietary assimilation rates for lipids, NLOM, and water set
to default values for fish, aquatic invertebrates, and zooplankton (see Table A9). Lipid, NLOM, and water
compositions in the diet (VLD VND, and VWD? respectively) are set to 1,20, and 79%, respectively.
When VLD, VND, VWD are equal (i.e., 33.33%) and when GD is set to a constant number, variations
of SL, SN, and Sw result in equivalent effects to GF, with GF decreasing as assimilation efficiency
increases (i.e., the organism eliminates less  as  its digestion [assimilation efficiency]  becomes
more efficient).  If assimilation efficiency is  set to  1  for lipid,  NLOM,  and water, the fecal
elimination rate of the organism is 0 kg feces/kg org (Figure A. 19).
                                          49 of 123

-------
      0.0004 -,
      0.0003 -
      0.0002 -
      0.0001 -
                                                                     	change in EL
                                                                     	change in EN
                                                                     	change in EW
                   0.1
0.2
0.3
0.4
0.5
Ex
0.6
0.7
0.8
0.9
Figure  A19.   Relationship between  dietary assimilation  efficiencies  (SL,  SN,  and  s\v)  and
(VLD=VND=VWD=33.33%; GB = 1.0x10
others, with the others set to 1.
       kg/day). Each dietary efficiency rate was altered independent of the
In summary, changes in the feeding rate of the organism (Go) have the greatest effect on the
fecal egestion rate of the organism (Op). GD is calculated for non-filter feeders using body weight
and temperature. GD is calculated for filter feeders, using the concentration of dissolved oxygen
in the water (Cox), body  weight (We), the  concentration of suspended  solids (Css),  and the
efficiency of scavenging of particles absorbed from water (o). Changes in these parameter values
would be expected to have the greatest influence on GF, which would result in influences on rate
constant for pesticide elimination through excretion (kE).  Changes in the composition of the diet
and the dietary assimilation rates are expected to have less of an influence on GF when compared
to GD and the parameters used to derive GD.
       A.6.2. Parameters Affecting ED

The efficiency of dietary pesticide transfer (Eo) is based only upon the KOW of the pesticide. As
KOW increases,  ED  decreases.  For chemicals with  Log  K0w 4-8, the  efficiency of dietary
pesticide transfer is 50-3% (Figure A.20).
                                         50 of 123

-------
      0.6 -,
      0.5 -
      0.4 -
      0.3 -
      0.2 -
      0.1 -
      0.0
                  4.5
5.5
6.5
7.5
                                           Log Kow
Figure A.20. Dietary transfer efficiency (ED) vs. Log Kow.
       A.6.3. Parameters Affecting KGB

The equations used to calculate the contents of the gut (i.e., VLG, VNG, and VWG) listed in Table
A.9 can be simplified as follows, using the equation for GF:
                                 VLO =
                                       (\-s }*V  *G
                                       V1   6L/ *LD  ^L
                                       M _ c- I * V  * r?
                                 T7    V    W / ' ND    L
                                 v ^ =	
                                  NO
                                 VWG ='
                                         -F  \*V  *Cr
                                          hW J V WD  ^L
Using these simplified equations, changes  in the feeding rate of the  organism (Go),  the fecal
egestion rate (Op), diet composition, and dietary assimilation of lipid, NLOM, and water can be
explored to understand effects on these parameters on estimations of the lipid composition of the
gut. If SL,  SN, and Sw are all equal and VLD, VND, and VWD are all equal, VLG, VNG, and VWG are
equal.  Changes in feeding rate (Go) and fecal egestion rate (GF) do not affect VLG, VNG or VWG
(Figure  A.21).  As would be expected, changes in VLD result in effects to VLG, with  an order of
magnitude increase in VLD, resulting in an order of magnitude increase in VLG (Figure A.22).
Also, changes in VND result in effects to VNG, with an increase in VND from 10 to 20%, resulting
                                        51 of 123

-------
in a 50% increase in VNG (Figure A. 23). An order of magnitude increase in VWD (keeping VLD
and VND constant) results in slight (approximately  10%) decreases in VLG and VNG and slight
(2%) increases in VWG (Figure A.24).  A decrease in SL from  0.9 to 0.1, results in an order of
magnitude  increase in the lipid content of the gut (VLG) (Figure A.25), but only  slight (<2%)
changes to  the NLOM and water contents of the gut (VNG and VWG, respectively). A decrease in
SN from 0.9 to 0.1 results in an order  of magnitude increase in the NLOM content of the gut
(VNG) (Figure A.26), as well as decreases in the gut composition attributed to lipid  and water. A
decrease in sw from 0.9 to 0.1 results in a 50% increase in the water content of the gut (VWo), as
well as an 80% decrease in the gut composition attributed to lipid and NLOM (Figure A.27). For
invertebrates,  dietary  assimilation  efficiencies  vary significantly, leading to  uncertainty in
assigning one value to this parameter.  Since hydrophobic chemicals are not likely to be stored in
the water of organism tissues, it is assumed that this route is not significant to bioaccumulation.
  1.00E+00 -,
  1.00E-01 -
    o
    X
  1.00E-02
  1.00E-03
                      VLG
                      -VNG
	VWG
        1.E-07      1.E-06     1.E-05      1.E-04     1.E-03      1.E-02     1.E-01      1.E + 00
                                              GD

Figure A21. Relationship between gut contents (VLG VNG, and VWG) and GD.  L N and Ware set to defaults
for fish (Table A9). VLD = 1%, VND = 20%, and VWD = 79%.
                                        52 of 123

-------
      1.0E+00
       1.0E-01
   o
   X
       1.0E-02 -
       1.0E-03
	VLG

	VNG

	VWG
               0      0.01    0.02   0.03    0.04
                                0.05


                                VLD
0.06   0.07    0.08    0.09    0.1
Figure A22.  Relationship between gut contents (VLG VNG, and VWG) and VLD-  L N and W are set to

defaults for fish (Table A9). V, = 20%, VWD = 1-VM,-VLD.
      1.0E+00
       1.0E-01 -
   o
   X
       1.0E-02 -
       1.0E-03
                   	VLG

                   	VNG

                   	VWG
              0.1     0.12    0.14   0.16    0.18    0.2    0.22    0.24    0.26   0.28    0.3


                                                  VND



Figure A23.  Relationship between gut contents (VLG, VNG and VWG) and VM>  L, %, and w are set to

defaults for fish (Table A9). VLD = 1%, VWD = l-VMrVLD.
                                           53 of 123

-------
      1.0E + 00 n
       1.0E-01
   o
   X
       1.0E-02
       1.0E-03
	VLG
	VNG
	VWG
              0.75    0.76    0.77    0.78    0.79    0.8    0.81     0.82    0.83    0.84    0.85
                                                   VWD

Figure A24.  Relationship between gut contents (VLG VNG, and VWG) and VWD-  L N and W are set to
defaults for fish (Table A9). VLD = 1%, VM, = 20%. Note that VLD+VND+VWD does not equal 100%, except
when VWD = 0.79.
     1.0E+00
      1.0E-01
   o
   X
      1.0E-02 -
      1.0E-03
                    	VLG
                    	VNG
                    - - - .VWG
            0      0.1      0.2      0.3      0.4      0.5     0.6     0.7      0.8      0.9       1
                                                   EL

Figure A25.  Relationship between gut contents (VLG VNG, and VWG) and L. EN? and W are set to defaults for
fish (Table A9). VLD = 1%, VM, = 20%, VWD = 79%.
                                            54 of 123

-------
     1.0E+00 -,
      1.0E-01 -
   o
   X
      1.0E-02 -
      1.0E-03
                     	VLG
                     	VNG
                     - - -  .VWG
             0       0.1      0.2      0.3      0.4      0.5       0.6      0.7      0.8      0.9      1
                                                     EN

Figure A26.  Relationship between gut contents (VLG VNG, and VWG) and N. L, and W are set to defaults for
fish (Table A9). VLD = 1%, VM, = 20%, VWD = 79%.
     1.0E+00 -,
      1.0E-01 -
   o
   X
      1.0E-02 -
      1.0E-03
                     	VLG
                     	VNG
                     - - -  .VWG
             0       0.1      0.2      0.3      0.4      0.5       0.6      0.7      0.8      0.9      1
                                                     EW

Figure A27.  Relationship between gut contents (VLG VNG, and VWG) and W. L and N are set to defaults for
fish (Table A9). VLD = 1%, VM, = 20%, VWD = 79%.
                                              55 of 123

-------
The partitioning  of a  chemical between the  gastrointestinal  tract (GIT) and the  organism is
described by KGB-  This partition coefficient is determined using the contents of the gut (VLG,
VNG, and VWG), the contents of the organism's body (VLB, VNB and VWB) as well as the octanol
water partition coefficient (K0w) (Table A.9). An order of magnitude increase in the lipid content
of the body (VLB) results in an order of magnitude decrease in KGB (Figures A.28 and A.29).  An
order of magnitude increase in the NLOM content of the body  (VNB) results in  a decrease in KGB
of approximately 20%. An order of magnitude increase in the lipid content of the  gut (VLG)
results in a 50% increase in KGB (Figure A.28). An order of magnitude increase in the NLOM
content of the gut (VNG), results in an increase  in KGB of 70% (Figure A.29). Changes in the Log
KOW of a chemical from 4 to 8 do not alter KGB-
     1.0E+01
     1.0E+00 -
  CO
  O
     1.0E-01 -
     1.0E-02
                                                        	VLB=0.01,VNB=0.20, VWB=0.79

                                                        	VLB=0.10, VNB=0.20, VWB=0.70

                                                        - - - .VLB=0.01, VNB=0.30, VWB=0.69
                      0.002
                                  0.004
                                               0.006
                                               VLG
                                                           0.008
                                                                        0.01
                                                                                    0.012
Figure A28. Relationship between lipid content of the gut (VLG) and KGB, with different body compositions
(VLB, VMJ, and
                                         56 of 123

-------
     1.0E+01
     1.0E+00 -
   m
   o
     1.0E-01 -
     1.0E-02
                                                         	VLB=0.01,VNB=0.20, VWB=0.79

                                                         	VLB=0.10, VNB=0.20, VWB=0.70

                                                         - - - .VLB=0.01, VNB=0.30, VWB=0.69
                       0.1
                                    0.2
                                                0.3
                                               VNG
                                                             0.4
                                                                         0.5
                                                                                      0.6
Figure A29. Relationship between NLOM content of the gut (VNG) and KGB, with different body compositions
(VLB, VOT, and VWB).
A.7. Overall Sensitivity of Body Concentration of Chemical
Parameters
                                                                to Individual Input
       A.7.1. First Sensitivity Analysis

In order to understand the influence of input parameters on  model predictions of pesticide
concentrations  in tissue  of aquatic organisms  (Ce),  a  sensitivity analysis was conducted.
Parameters were assigned uniform distributions and assumptions of ranges based on data in the
scientific literature.  The range for each parameter is  defined  in Table  A10.   Diets of each
trophic level were varied according to the definitions in Table All. Uniform  distributions were
used to allow unbiased selection of values from set ranges.  Once parameter  assumptions were
assigned, a Monte Carlo simulation was carried out using Crystal Ball 2000.  In this simulation,
10,000 trials were conducted with randomly selected  parameter values  resulting in  predicted
pesticide concentrations in  each of the  seven trophic  levels. The sensitivity of the  model to
specific parameters was defined by the  contribution of each parameter to the variance  of the
estimation of pesticide concentrations in each of the trophic levels.

The  results of this  analysis  indicate that  of all  the  variables in the model, the  Log KOW
contributes the most to variability (<75% of total) in estimates of CB for all animal trophic levels.
For phytoplankton, the water column EEC, concentration of POC  in the water column (XPOc)
                                         57 of 123

-------
and Log K0w contribute the greatest variability in the predicted CB values (38, 28,  and 22%,
respectively).

       A.7.2. Second Sensitivity Analysis

Based on the results of the first sensitivity analysis, a second analysis was conducted  where the
influence of individual  parameters on variability in CB was examined, with fixed  Log KOW
values. In the second sensitivity analysis, the Log K0w was set to values of 4, 5, 6, 7, and 8 and a
Monte Carlo simulation (10,000 trials) was  run for each Log K0w value.   Parameters were
assigned uniform  distributions  and assumptions of  ranges  based  on  data  in the  scientific
literature.  The range for each parameter is defined in Table A10 (with the  exception of Log
KOW).  Diets of each trophic level were varied according to the definitions in Table Al  1.

The contributions of individual parameters at Log K0w values of 4, 5, 6, 7 and 8 to the variability
in the pesticide tissue concentration (CB) of the seven aquatic trophic levels of KABAM are
provided in Tables A12-A18.  The results of this sensitivity analysis indicate that parameters
have different relative importance in estimating CB for the seven trophic levels (e.g.,  the water
column EEC contributes the most variability to the phytoplankton CB, while the pore water EEC
and fraction of respiratory ventilation that involves pore-water of sediment (nip) value  contribute
the most variance to the zooplankton  CB). In  addition, these tables indicate that the  relative
importance of individual parameters to estimates of CB change with Log K0w-   It  should be
noted that several parameters in the Arnot and Gobas (2004) model are linked  (e.g., mP and m0,
diet composition, VLB, VNB, and VWB).  Therefore, sensitivity of CB predictions to one  parameter
implies sensitivity of the predictions to the linked parameters.

This sensitivity analysis  also indicates that some parameters that are fixed in KABAM, including
the constant related to the resistance to  pesticide uptake through the aqueous phase of plant (A),
proportionality constant  expressing the sorption capacity of NLOM to that of octanol (P), and mP
(set to either 0 or 0.05), can contribute >10% of total variability in estimates of  CB.
                                        58 of 123

-------
Table A10. Parameters and associated assumptions used for first and second sensitivity analysis of KABAM.
Para-
meter
A
B
Cox
Css
CWTO
CWTP
Log Kow
Koc
mp
OC
Parameter Description
Constant related to the resistance to
pesticide uptake through the aqueous
phase of plant
Constant related to the resistance to
pesticide uptake through the organic
phase of plant
Concentration of dissolved oxygen
(mg 02/L)
Concentration of suspended solids
(kg/L)
Total pesticide concentration in
water column above the sediment
Freely dissolved pesticide
concentration in pore water of
sediment
Log of octanol-water partition
coefficient
Organic carbon partition coefficient
Fraction of respiratory ventilation
that involves pore-water of sediment
Percent organic carbon in sediment
Trophic Level
Phytoplankton
Phytoplankton
All
All
All
All
All
All
Zooplankton
Benthic Inv.
Filter Feeders
Small Fish
Medium Fish
Large Fish
All
Minimum
of Range
IxlO'5
1
4
2.0xlO"6
0.1
0.1
4
3.5xl03
0
0
0
0
0
0
1%
Maximum
of Range
IxlO'4
10
12
S.OxlO"4
100
100
8
3.5 xlO7
1
1
1
1
1
1
10%
Source/Comments
In Arnot and Gobas 2004, this value is set to a constant of 6.0xlO"5
days. This value was varied by an order of magnitude around the
reported constant value to understand the influence of this parameter
on estimates of bioaccumulation. The reasonable range of values for
this parameter is unknown.
In Arnot and Gobas 2004, this value is set to a constant of 5.5 days.
This value was varied by an order of magnitude around the reported
constant value to understand the influence of this parameter on
estimates of bioaccumulation. The reasonable range of values for
this parameter is unknown.
Minimum is based on 60% of saturation of water with 6 mg/L as
saturation (in 30C water). Maximum is based on solubility limit of
oxygen in cold water (5C; see USGS 2008a).
Based on 5th and 95th percentiles of approximately 38,000
measurements of suspended sediment concentrations in surface
waters of the US provided by NAWQA (USGS 2008b).
Assumed to be reasonable range for EECs expected from
PRZM/EXAMS modeling.
Assumed to be reasonable range for EECs expected from
PRZM/EXAMS modeling.
Assumption that bioaccumulation model can be used for chemicals
with Log Kow 4-8.
Determined based on assumption that Koc can be estimated as
0.35*Kow. In sensitivity analysis, Koc is linked directly to Kow in
order to avoid error in selection of inconsistent values for these
parameters.
Based on full range of parameter values.
Based on full range of parameter values.
Based on full range of parameter values.
Based on full range of parameter values.
Based on full range of parameter values.
Based on full range of parameter values.
In the OPP standard pond used in EXAMS, the default value for this
parameter is 4%. This parameter value is varied by 1 order of
59 of 123

-------
Table A10. Parameters and associated assumptions used for first and second sensitivity analysis of KABAM.
Para-
meter

T
VLB
VNB
VWB
WB
XPOC
XDOC
P
Parameter Description

Temperature (C)
Lipid fraction of organism
NLOM (Non Lipid Organic Matter)
fraction of animals, NLOC (Non
Lipid Organic Carbon) of plants
Water content of the organism
Wet weight (kg) of the organism
Concentration of paniculate organic
carbon in water (kg/L)
Concentration of dissolved organic
carbon in water (kg/L)
Proportionality constant expressing
the sorption capacity of NLOM or
NLOC to that of octanol
Trophic Level

All
Phytoplankton
Zooplankton
Benthic Inv.
Filter Feeders
Fish
All
Phytoplankton
Zooplankton
Benthic Inv.
Filter Feeders
Fish
Phytoplankton
Zooplankton
Benthic Inv.
Filter Feeders
Small Fish
Medium Fish
Large Fish
All
All
All
Minimum
of Range

1
0.5
1.0
0.5
0.4
0.5
-
0.85
0.74
0.69
0.78
0.71
-
IxlO-9
5xlO'6
2xlO-4
IxlO'3
5xlO'3
0.25
2.0xlO'6
5.0xlO"7
0
Maximum
of Range

30
2.0
4.0
12
4
8
-
0.95
0.96
0.83
0.93
0.80
-
IxlO"7
2xlO'3
IxlO"2
5xlO"2
0.6
3.6
5.0xlO'4
5.0xlO"5
1
Source/Comments
magnitude around the OPP standard pond value.
Reasonable range of values for this parameter in the environment.
See Table Cl of Appendix C.
See Table C2 of Appendix C.
See Tables C4-C9 of Appendix C.
See Tables C13-C15 of Appendix C.
See Table C19 of Appendix C.
Set to equal 1- VLB- VWB
Assume 5% deviation from mean (i.e., 90%).
See Section C.2. of Appendix C.
See Table C3 of Appendix C.
See Table C12 of Appendix C.
See Table CIS of Appendix C.
Not a necessary parameter for phytoplankton.
See Section C.2. of Appendix C.
See Table Cll of Appendix C.
See Section C.5 of Appendix C.
See Table C16 of Appendix C.
See Table C17 of Appendix C.
See Section C.5 of Appendix C.
Based on 5th and 95th percentiles of approximately 38,000
measurements of suspended sediment concentrations in surface
waters of the US provided by NAWQA (USGS 2008b).
In the OPP standard pond used in EXAMS, the default value for this
parameter is 5.0xlO"6. This parameter value is varied by 2 orders of
magnitude around the OPP standard pond value.
Designed to represent all values equal to or less than the partitioning
of a chemical between octanol and water.
60 of 123

-------
Table A10. Parameters and associated assumptions used for first and second sensitivity analysis of KABAM.
Para-
meter
L
SN
w
0
Parameter Description
Dietary assimilation rate of lipids
Dietary assimilation rate of NLOM
Dietary assimilation rate of water
Efficiency of scavenging of particles
absorbed from water
Trophic Level
Animals
Animals
Animals
Filter Feeders
Minimum
of Range
0
0
0
0
Maximum
of Range
1
1
1
1
Source/Comments
Based on full range of parameter values.
Based on full range of parameter values.
Based on full range of parameter values.
Based on full range of parameter values.
Table All. Dietary assumptions of aquatic trophic levels used for sensitivity analysis of KABAM.
Trophic Level
Zooplankton
Benthic
Invertebrates
Filter Feeder
Small Fish
Medium Fish
Large Fish
Organism in diet
Phytoplankton
Sediment
Phytoplankton
Zooplankton
Sediment
Phytoplankton
Zooplankton
Benthic
invertebrates
Phytoplankton
Zooplankton
Benthic
invertebrates
Zooplankton
Benthic
invertebrates
Small fish
Small fish
Medium Fish
Minimum
Value
100%
0
0
-
0
0
0
-
0
0
-
0
0
-
0
-
Maximum Value
100%
50%
50%
-
33%
33%
33%
-
50%
50%
-
50%
50%
-
100%
-
Comments
-
-
-
Set to 1- (% diet attributed to sediment + % diet attributed to phytoplankton)
-
-
-
Set to 1- (% diet attributed to sediment + % diet attributed to phytoplankton+%
diet attributed to zooplankton)
-
-
Set to 1- (% diet attributed to phytoplankton+% diet attributed to zooplankton)
-
-
Set to 1- (% diet attributed to zooplankton + % diet attributed to benthic
invertebrates)
It is assumed that large fish consume only smaller fish
Set to 1- (% diet attributed to small fish)
61 of 123

-------
Table A12. Second sensitivity analysis results: Contribution to variance of specific
variables to CB values of phytoplankton at different Log Kow values.
Variable
A
VLB
VWB
Water Column EEC
Xpoc
P
Total
4
<0.1%
<0.1%
5.9%
59.6%
7.0%
26.7%
99.2%
5
<0.1%
<0.1%
3.4%
46.1%
30.9%
18.6%
99.0%
6
0.7%
<0.1%
2.3%
44.9%
39.1%
12.0%
99.0%
7
9.3%
0.2%
0.2%
45.7%
41.7%
2.1%
99.2%
8
18.2%
<0.1%
<0.1%
43.0%
38.0%
<0.1%
99.2%
Table A13. Second sensitivity analysis results: Contribution to variance of specific
variables to CB values of zooplankton at different Log Kow values.
Variable
CQX
mP
Pore Water EEC
T
VLB
VWB
Water Column EEC
WB
XPOC
P
eL
EN
Total
4
<0.1%
4.3%
27.6%
<0.1%
1.2%
20.8%
11.2%
<0.1%
1.7%
32.5%
<0.1%
<0.1%
99.3%
5
<0.1%
23.0%
37.2%
<0.1%
0.4%
14.1%
2.2%
<0.1%
2.5%
20.2%
<0.1%
<0.1%
99.6%
6
0.2%
37.4%
40.8%
1.3%
0.6%
8.8%
0.3%
<0.1%
0.8%
8.4%
<0.1%
0.5%
99.1%
7
2.5%
39.5%
40.6%
7.7%
0.6%
0.6%
<0.1%
0.5%
<0.1%
1.7%
0.2%
1.1%
95.0%
8
4.3%
36.1%
37.4%
19.6%
<0.1%
0.8%
<0.1%
0.6%
<0.1%
<0.1%
<0.1%
0.3%
99.1%
62 of 123

-------
Table A14. Second sensitivity analysis results: Contribution to variance of specific
variables to CB values of benthic invertebrates at different Log Kow values.
Variable
Characteristics of prey*
Cox
Diet composition
mP
OC
Pore Water EEC
T
VLB
VWB
Water Column EEC
XPOC
P
L
EN
Total
4
<0.1%
<0.1%
<0.1%
4.9%
<0.1%
37.2%
<0.1%
4.0%
3.6%
14.4%
2.5%
32.6%
<0.1%
<0.1%
99.2%
5
1.2%
<0.1%
0.2%
16.0%
<0.1%
51.1%
<0.1%
2.5%
1.5%
1.9%
2.7%
21.5%
<0.1%
0.5%
99.1%
6
12.0%
<0.1%
1.2%
5.3%
0.9%
61.4%
1.0%
1.6%
1.1%
<0.1%
0.5%
7.6%
0.7%
5.8%
99.1%
7
12.0%
1.3%
1.0%
<0.1%
<0.1%
61.1%
8.2%
1.0%
0.6%
<0.1%
<0.1%
0.4%
1.2%
5.6%
92.4%
8
6.0%
1.6%
8.6%
<0.1%
<0.1%
59.4%
16.1%
<0.1%
<0.1%
<0.1%
<0.1%
<0.1%
<0.1%
<0.1%
91.7%
*mP, body composition, etc.
Table A15. Second sensitivity analysis results: Contribution to variance of specific
variables to CB values of filter feeders at different Log Kow values.
Variable
Characteristics of prey*
Cox
Css
Diet composition
mP
OC
Pore Water EEC
T
VLB
VWB
Water Column EEC
WB
XPOC
P
eL
EN
0
Total
4
<0.1%
<0.1%
0.2%
<0.1%
3.1%
<0.1%
28.6%
<0.1%
1.5%
11.1%
10.6%
<0.1%
2.0%
42.0%
<0.1%
<0.1%
<0.1%
99.1%
5
1.5%
<0.1%
1.4%
<0.1%
6.9%
<0.1%
42.0%
<0.1%
1.5%
7.0%
1.8%
<0.1%
2.2%
30.7%
0.4%
1.8%
1.9%
99.1%
6
11.8%
<0.1%
2.4%
0.3%
<0.1%
0.9%
52.2%
1.9%
1.3%
4.3%
<0.1%
<0.1%
0.2%
13.7%
2.6%
5.3%
2.1%
99.0%
7
11.7%
2.0%
3.2%
0.6%
0.5%
2.0%
50.5%
13.0%
0.4%
2.8%
<0.1%
<0.1%
<0.1%
1.5%
2.7%
4.6%
3.5%
99.0%
8
2.5%
3.8%
8.1%
1.6%
0.7%
4.1%
42.3%
25.1%
<0.1%
0.4%
<0.1%
0.3%
<0.1%
<0.1%
0.5%
1.2%
7.8%
98.4%
*mP, body composition, etc.
                                          63 of 123

-------
Table A16. Second sensitivity analysis results: Contribution to variance of specific
variables to CB values of small fish at different Log Kow values.
Variable
Characteristics of prey*
Cox
Diet composition
mP
OC
Pore Water EEC
T
VLB
VWB
Water Column EEC
XPOC
P
L
EN
Total
4
<0.1%
0.2%
<0.1%
3.9%
<0.1%
31.8%
<0.1%
2.5%
1.5%
11.6%
2.0%
45.7%
<0.1%
<0.1%
99.2%
5
3.3%
0.2%
0.6%
4.4%
<0.1%
45.3%
0.4%
1.3%
1.0%
1.5%
2.6%
34.3%
1.2%
2.9%
99.0%
6
16.1%
<0.1%
2.2%
0.5%
0.5%
52.1%
0.8%
1.3%
0.3%
0.2%
0.5%
13.8%
4.4%
6.8%
99.5%
7
18.2%
1.3%
2.9%
<0.1%
1.5%
53.0%
9.8%
0.8%
0.2%
<0.1%
<0.1%
1.7%
3.4%
6.1%
98.9%
8
9.2%
2.3%
2.7%
0.6%
2.6%
51.6%
27.7%
0.3%
<0.1%
<0.1%
<0.1%
0.2%
0.9%
0.8%
98.9%
*mP, body composition, etc.
Table A17. Second sensitivity analysis results: Contribution to variance of specific
variables to CB values of medium fish at different Log Kow values.
Variable
Characteristics of prey*
Cox
Diet composition
mp
OC
Pore Water EEC
T
VLB
VWB
Water Column EEC
WB
XpoC
P
L
EN
Total
4
<0.1%
0.3%
<0.1%
1.5%
<0.1%
20.5%
0.2%
0.4%
15.7%
7.7%
<0.1%
1.2%
51.2%
0.2%
0.6%
99.5%
5
2.7%
<0.1%
0.2%
1.5%
<0.1%
33.1%
<0.1%
0.2%
9.1%
1.1%
0.2%
1.7%
43.5%
1.4%
3.4%
98.1%
6
17.3%
<0.1%
0.8%
<0.1%
0.4%
43.3%
1.0%
0.2%
5.0%
0.2%
<0.1%
0.5%
20.9%
2.2%
7.4%
99.2%
7
21.0%
1.1%
0.6%
<0.1%
1.3%
48.5%
11.1%
0.4%
4.2%
<0.1%
<0.1%
<0.1%
4.3%
1.9%
4.7%
99.1%
8
10.0%
2.4%
<0.1%
0.2%
1.9%
49.0%
33.1%
<0.1%
0.5%
<0.1%
<0.1%
<0.1%
0.5%
0.3%
0.7%
98.6%
*mP, body composition, etc.
                                          64 of 123

-------
Table A18. Second sensitivity analysis results: Contribution to variance of specific
variables to CB values of large fish at different Log Kow values.
Variable
Characteristics of prey*
Cox
Diet composition
mP
OC
Pore Water EEC
T
VLB
VWB
Water Column EEC
XPOC
P
L
EN
Total
4
<0.1%
0.9%
0.3%
1.4%
<0.1%
27.0%
1.4%
1.8%
1.1%
9.3%
1.4%
52.0%
<0.1%
2.0%
98.6%
5
5.8%
0.9%
0.8%
0.3%
<0.1%
35.1%
1.5%
0.9%
0.5%
1.2%
1.7%
38.8%
0.9%
10.5%
98.9%
6
22.8%
<0.1%
0.9%
<0.1%
0.4%
41.3%
0.7%
1.3%
0.5%
<0.1%
0.3%
13.8%
1.7%
15.3%
99.0%
7
23.1%
0.9%
0.3%
<0.1%
1.1%
45.1%
10.1%
1.1%
0.6%
<0.1%
<0.1%
1.7%
2.2%
12.8%
99.0%
8
9.7%
2.1%
<0.1%
<0.1%
1.7%
44.9%
35.1%
<0.1%
0.2%
<0.1%
<0.1%
<0.1%
0.4%
4.3%
98.4%
 *mP, body composition, etc.
       A.7.3. Third Sensitivity Analysis

A third sensitivity analysis was conducted to explore the influences of KABAM input parameters
that are controlled by the user (including chemical  specific inputs  and ecosystem inputs), with
the fixed parameters unchanged. In this sensitivity analysis, the Log K0w was set to values of 4,
5, 6, 7, and 8, and  a Monte Carlo simulation (10,000 trials) was run for each Log K0w value.
Parameters were assigned uniform distributions and assumptions of ranges based on data  in the
scientific literature.  The range for each parameter is defined  in  Table  A19.   Diets  of each
trophic level were varied according to the definitions in Table Al 1.

The contributions of individual chemical specific and ecosystem input parameters at Log K0w
values of 4, 5, 6, 7,  and 8 to the variability in the pesticide tissue concentration (CB) of the  seven
aquatic trophic levels of KABAM are provided  in  Tables A20-A26.  As with the second
sensitivity  analysis, the results of this  analysis indicate that parameters have different relative
importance in estimating CB for the seven trophic levels. In addition,  these tables indicate that
the relative importance of individual parameters to estimates of CB change with Log KOW.

This sensitivity analysis indicates that several parameters contribute >10% of variance in  CB of
one or more trophic  levels. These include: water  column EEC,  pore water EEC, particulate
organic carbon (Xpoc), sediment organic carbon (OC), concentration of suspended solids  (Css),
water temperature (T), lipid composition (VLB), diet  composition, and characteristics  of prey
(including body composition, diet composition and  mP). Several of these parameters, including
Xpoc, OC, and Css have default values that were selected to be consistent with the standard pond
used in EXAMS.
                                        65 of 123

-------
One notable  observation resulting from this sensitivity analysis is that at Log K0w 7 and  8,
benthic invertebrate diet composed of sediment contributes >25% of the variance in CB of all
three size classes  of fish.  This indicates  that the proportion of the benthic invertebrate diet
attributed to sediment can influence the estimated pesticide concentrations in fish tissues.

As indicated above, several parameters in the Arnot and Gobas (2004) model are linked (e.g., mP
and mo, diet composition, VLB, VNB and VWB).  Therefore, sensitivity of CB predictions to one
parameter implies sensitivity of the predictions to the linked parameters.
                                        66 of 123

-------
Table A19. Parameters and associated assumptions used for third sensitivity analysis of KABAM.
Para-
meter
Cox
css
CWTO
CWTP
Koc
mp
OC
T
VLB
VNB
Parameter Description
Concentration of dissolved oxygen
(mg02/L)
Concentration of suspended solids
(kg/L)
Total pesticide concentration in
water column above the sediment
Freely dissolved pesticide
concentration in pore water of
sediment
Organic carbon partition coefficient
Fraction of respiratory ventilation
that involves pore-water of sediment
Percent organic carbon in sediment
Temperature (C)
Lipid fraction of organism
NLOM (Non Lipid Organic Matter)
fraction of animals, NLOC (Non
Lipid Organic Carbon) of plants
Trophic Level
All
All
All
All
All
Zooplankton
Benthic Inv.
Filter Feeders
Small Fish
Medium Fish
Large Fish
All
All
Phytoplankton
Zooplankton
Benthic Inv.
Filter Feeders
Fish
All
Minimum
of Range
4
2.0xl(r6
0.1
0.1
3.5xl03
0
0
0
0
0
0
1%
1
0.5
1.0
0.5
0.4
0.5
-
Maximum
of Range
12
5.0xlO'4
100
100
3.5 xlO7
0.05
0.05
0.05
0.05
0.05
0.05
10%
30
2.0
4.0
12
4
8
-
Source/Comments
Minimum is based on 60% of saturation of water with 6 mg/L as
saturation (in 30C water). Maximum is based on solubility limit of
oxygen in cold water (5C; see USGS 2008a).
Based on 5th and 95th percentiles of approximately 38,000
measurements of suspended sediment concentrations in surface
waters of the US provided by NAWQA (USGS 2008b).
Assumed to be reasonable range for EECs expected from
PRZM/EXAMS modeling.
Assumed to be reasonable range for EECs expected from
PRZM/EXAMS modeling.
Determined based on assumption that Koc can be estimated as
0.35*Kow. In sensitivity analysis, Koc is linked directly to K0w in
order to avoid error in selection of inconsistent values for these
parameters.
Based on default parameter values (0 or 0.05).
Based on default parameter values (0 or 0.05).
Based on default parameter values (0 or 0.05).
Based on default parameter values (0 or 0.05).
Based on default parameter values (0 or 0.05).
Based on default parameter values (0 or 0.05).
In the OPP standard pond used in EXAMS, the default value for this
parameter is 4%. This parameter value is varied by one order of
magnitude around the OPP standard pond value.
Reasonable range of values for this parameter in the environment.
See Table Cl of Appendix C.
See Table C2 of Appendix C.
See Tables C4-C9 of Appendix C.
See Tables C13-C15 of Appendix C.
See Table C19 of Appendix C.
Set to equal I-VLB-VWB
67 of 123

-------
Table A19. Parameters and associated assumptions used for third sensitivity analysis of KABAM.
Para-
meter
VWB
WB
Xpoc
XDOC
Parameter Description
Water content of the organism
Wet weight (kg) of the organism at t
Concentration of paniculate organic
carbon in water (kg/L)
Concentration of dissolved organic
carbon in water (kg/L)
Trophic Level
Phytoplankton
Zooplankton
Benthic Inv.
Filter Feeders
Fish
Phytoplankton
Zooplankton
Benthic Inv.
Filter Feeders
Small Fish
Medium Fish
Large Fish
All
All
Minimum
of Range
0.85
0.74
0.69
0.78
0.71
-
IxlO-9
5xlO'6
2xlO-4
IxlO'3
5xlO'3
0.25
2.0xlO"6
5.0xlO"7
Maximum
of Range
0.95
0.96
0.83
0.93
0.80
-
IxlO'7
2xlO'3
IxlO"2
5xlO'2
0.6
3.6
5.0xlO"4
5.0xlO"5
Source/Comments
Assume 5% deviation from mean (i.e., 90%).
See Section C.2. of Appendix C.
See Table C3 of Appendix C.
See Table C12 of Appendix C.
See Table CIS of Appendix C.
Not a necessary parameter for phytoplankton.
See Section C.2. of Appendix C.
See Table Cll of Appendix C.
See Section C.5 of Appendix C.
See Table C16 of Appendix C.
See Table C17 of Appendix C.
See Section C.5 of Appendix C.
Based on 5th and 95th percentiles of approximately 38,000
measurements of suspended sediment concentrations in surface
waters of the US provided by NAWQA (USGS 2008b).
In the OPP standard pond used in EXAMS, the default value for this
parameter is 5.0xlO"6. This parameter value is varied by two orders
of magnitude around the OPP standard pond value.
68 of 123

-------
Table A20. Third sensitivity analysis results: Contributions to variance of specific variables to CB
values of phytoplankton at different Log Kow values.
Variable
VLB
VWB
Water Column EEC
XPOC
Total
4
0.7%
7.6%
80.1%
11.3%
99.7%
5
0.4%
5.1%
55.9%
38.0%
99.4%
6
0.3%
3.3%
51.8%
44.2%
99.6%
7
0.1%
0.5%
52.9%
46.1%
99.5%
8
0.1%
0.1%
52.8%
46.8%
99.6%
Table A21. Third sensitivity analysis results: Contributions to variance of specific variables to CB
values of zooplankton at different Log Kow values.
Variable
Cox
nip
Pore Water EEC
T
VLB
VWB
Water Column EEC
WB
XpoC
Total
4
0.1%
0.1%
0.1%
0.1%
12.3%
1.3%
75.0%
0.1%
10.7%
99.3%
5
O.l%
2.2%
2.2%
O.l%
12.7%
0.8%
48.0%
0.1%
33.5%
99.4%
6
O.l%
22.2%
22.8%
O.l%
13.3%
0.9%
16.0%
0.1%
24.2%
99.4%
7
0.6%
44.0%
42.0%
3.0%
7.2%
0.4%
0.9%
0.1%
1.5%
99.6%
8
3.1%
40.2%
38.8%
15.5%
1.1%
0.1%
0.1%
0.6%
0.1%
99.3%
Table A22. Third sensitivity analysis results: Contributions to variance of specific variables to CB
values of benthic invertebrates at different Log Kow values.
Variable
Characteristics of prey*
Cox
Diet composition
nip
OC
Pore Water EEC
T
VLB
VWB
Water Column EEC
XpoC
Total
4
0.1%
0.1%
0.1%
0.1%
0.1%
0.2%
0.1%
31.1%
0.1%
57.8%
7.4%
96.5%
5
0.1%
0.5%
1.2%
0.8%
0.8%
6.0%
1.3%
38.5%
0.1%
27.9%
22.3%
99.3%
6
0.1%
2.2%
21.1%
0.3%
11.6%
35.1%
4.0%
20.0%
0.5%
1.7%
3.0%
99.5%
7
0.2%
0.2%
34.4%
O.l%
16.6%
41.8%
O.l%
6.3%
0.2%
O.l%
0.1%
99.7%
8
0.1%
0.1%
36.5%
0.1%
18.6%
42.0%
1.7%
0.8%
0.1%
0.1%
0.1%
99.6%
*mP, body composition, etc.
                                          69 of 123

-------
Table A23. Third sensitivity analysis results: Contributions to variance of specific variables to CB
values of filter feeders at different Log Kow values.
Variable
Characteristics of prey*
Cox
Css
Diet composition
mp
OC
Pore Water EEC
T
VLB
VWB
Water Column EEC
WB
XPOC
Total
4
<0.1%
<0.1%
0.3%
0.1%
0.2%
<0.1%
0.2%
<0.1%
24.2%
0.6%
64.8%
<0.1%
9.2%
99.5%
5
2.1%
0.1%
12.6%
0.8%
0.3%
1.9%
8.2%
0.1%
24.0%
0.1%
26.7%
0.1%
22.6%
99.2%
6
7.7%
0.5%
12.5%
2.9%
0.1%
15.6%
39.5%
0.9%
16.5%
0.4%
1.4%
0.1%
1.6%
99.5%
7
14.8%
0.1%
6.3%
2.5%
0.1%
19.7%
45.4%
1.0%
9.5%
0.4%
0.1%
0.2%
0.1%
99.8%
8
5.3%
1.3%
13.0%
7.6%
O.l%
18.4%
39.2%
10.8%
4.0%
O.l%
0.1%
O.l%
0.1%
99.6%
*mP, body composition, etc.
Table A24. Third sensitivity analysis results: Contributions to variance of specific variables to CB
values of small fish at different Log Kow values.
Variable
Characteristics of prey*
Cox
Diet composition
nip
OC
Pore Water EEC
T
VLB
VWB
Water Column EEC
WB
XPOC
Total
4
0.1%
0.1%
O.l%
O.l%
0.1%
0.1%
0.1%
28.7%
0.1%
61.8%
O.l%
8.7%
99.2%
5
3.1%
2.1%
1.1%
0.7%
0.1%
2.4%
4.2%
25.3%
0.2%
34.4%
0.4%
25.4%
99.3%
6
21.0%
5.6%
6.8%
0.1%
6.6%
27.4%
11.2%
13.4%
0.1%
2.7%
0.7%
4.3%
99.7%
7
32.6%
0.6%
9.4%
0.1%
13.3%
38.6%
0.1%
4.8%
0.1%
0.1%
0.2%
0.1%
99.5%
8
28.6%
0.1%
10.0%
O.l%
14.5%
39.4%
6.5%
0.5%
0.1%
O.l%
O.l%
0.1%
99.5%
*mP, body composition, etc.
                                          70 of 123

-------
Table A25. Third sensitivity analysis results: Contributions to variance of specific variables to CB
values of medium fish at different Log Kow values.
Variable
Characteristics of prey*
Cox
Diet composition
mp
OC
Pore Water EEC
T
VLB
VWB
Water Column EEC
WB
XPOC
Total
4
<0.1%
0.1%
0.1%
0.1%
0.1%
0.2%
0.2%
19.3%
2.4%
68.4%
O.l%
8.8%
99.3%
5
4.9%
3.9%
0.3%
0.2%
0.1%
3.0%
9.7%
14.0%
1.8%
35.5%
0.4%
25.8%
99.5%
6
25.2%
8.6%
2.7%
0.1%
6.7%
27.0%
14.8%
6.6%
1.4%
2.4%
0.4%
3.8%
99.6%
7
38.4%
0.9%
2.6%
0.1%
14.0%
39.7%
0.2%
3.0%
0.4%
0.1%
0.1%
0.1%
99.2%
8
30.7%
0.1%
1.4%
O.l%
14.6%
41.2%
11.1%
0.3%
0.1%
0.1%
O.l%
0.1%
99.3%
*mP, body composition, etc.
Table A26. Third sensitivity analysis results: Contribution to variance of specific variables to CB
values of large fish at different Log Kow values.
Variable
Characteristics of prey*
Cox
Diet composition
OC
Pore Water EEC
T
VLB
VWB
Water Column EEC
WB
Xpoc
Total
4
0.2%
0.1%
0.1%
0.1%
0.1%
0.2%
27.5%
0.1%
62.6%
0.1%
7.9%
98.4%
5
5.3%
6.6%
0.4%
0.1%
2.3%
15.9%
19.3%
0.2%
28.8%
0.4%
19.9%
99.1%
6
24.5%
9.7%
2.4%
5.5%
22.7%
17.2%
11.6%
O.l%
2.2%
0.1%
3.6%
99.4%
7
38.0%
1.1%
2.4%
13.3%
36.4%
0.3%
8.0%
0.1%
0.1%
0.1%
0.1%
99.5%
8
30.7%
0.1%
0.1%
13.0%
38.1%
16.0%
1.5%
0.1%
0.1%
0.1%
0.1%
99.3%
*mp, body composition, etc.
                                          71 of 123

-------
Appendix B. Explanation of Defaults and Alternative Values Representing Abiotic
Characteristics of Aquatic Ecosystem

Abiotic characteristics of the aquatic ecosystem that are necessary for KABAM are  defined in
Table 4 of the model tool.   These characteristics include concentrations of paniculate organic
carbon (XPOc), dissolved  organic carbon (XDOc), dissolved oxygen (Cox), suspended solids
(Css), water temperature (T), and % organic carbon (OC) content of the sediment.   The model
tool is populated with default values for these parameters, which can be altered based on the
needs of the model user.   Default values are based on the abiotic characteristics of the aquatic
ecosystem and are designed to be  consistent with the  OPP standard pond scenario used in
EXAMS. Brief explanations for these default values as well as guidance on selecting alternative
values are provided below for each parameter.
B.I. Particulate Organic Carbon (XPOc) and Dissolved Organic Carbon (XDoc)

XPOC and XDOC are entered by the model user in units of kg OC/L. These parameters are relevant
to estimating the available pesticide fraction in water (
-------
aquatic  organism tissues.  Therefore, a decrease in the value  of Cox results in a decrease in
pesticide concentrations in tissues of aquatic organisms.

Cox is entered by the model user in units of mg O2/L. The default value for this parameter is 5.0
mg O2/L, based on the OPP  standard pond. This concentration  does not represent the highest
possible value for Cox (i.e., the limit of solubility of oxygen) and is not expected to result in the
most conservative estimates of pesticide in aquatic animal tissues. However, it is consistent with
the OPP standard pond which is used to derive EECs.

The model  user could explore the  influence of  Cox  on  the predictions of pesticide  tissue
concentrations in aquatic organisms by selecting a higher value, for example the solubility of
oxygen  (potential range: 6-12  mg O2/L). To determine the solubility of oxygen in water at
specific temperatures and pressures, see USGS 2008a.

It may be necessary for the model user to incorporate an alternate Cox value if the modeling
incorporates EECs from a source other than PRZM/EXAMS. In that case, the model user should
enter a Cox value that corresponds to the specific water body used.
B.3. Water Temperature (T)

The water temperature parameter influences calculation of the growth dilution rate constant (kG),
the pesticide uptake rate constant through diet (kD), and the pesticide  elimination rate constant
through excretion of feces (kE). The growth dilution rate constant (kG) is dependent on whether
the temperature is above or below 17.5C. The growth dilution rate constant is higher when the
temperature is above  17.5C compared to when the temperature is below 17.5C (Figure A.8).
Temperature affects the pesticide uptake rate through the dietary uptake rate constant (kD) by
changing the feeding rate of the animal (Go). An increase in temperature results in an increase in
the feeding rate, and with that, an increase in the pesticide uptake constant for the diet (Figure
A. 10). The fecal egestion rate  constant (kE) is affected by  temperature by changing the feeding
rate (Go) as well as the fecal egestion rate (GE) of the animal. An increase in temperature results
in an  increase  in the feeding rate  (Figure A. 13), and with that, an increase in the fecal egestion
rate. The increase in the fecal egestion rate results in an increase in the pesticide rate constant for
pesticide elimination  through  excretion.  In summary, increase  in temperature results  in an
increase  in kD,  kE,   and  kG.  Although kG  and   kE  represent  processes  (i.e.,  pesticide
elimination/dilution) that compete with  ko (i.e., pesticide  uptake), the net  increase  in the two
processes (uptake and elimination/dilution) does not cancel  each other out.

The water temperature of the EXAMS'  pond  varies based  on the selected PRZM scenario.
Therefore, the model user should select the water temperature based on the PRZM scenario used
for deriving EECs. If the modeling incorporates EECs  from a source  other than PRZM/EXAMS,
the water temperature  relevant  to the other EECs should be utilized.
                                        73 of 123

-------
B.4. Concentration of Suspended Solids (C ss)

The concentration of suspended solids (Css) is relevant to filter feeders only.  Css influences the
calculation of the rate constants for pesticide uptake through diet (ko) and pesticide elimination
through excretion of feces (kE). An increase in CSs leads to an increase in the feeding rate of
filter feeders  (Go) which in turn results in an increase in the pesticide uptake through diet (kD).
An increase in Css also leads to an increase in the fecal  egestion rate of filter feeders (Gp) and an
increase in the pesticide elimination through excretion  of fecal matter (kp). Although ko and kp
represent competing processes, the net increase in the two does not cancel each other out.

Css is entered by  the model  user in units of kg/L. The default value for this parameter is
S.OOxlO"5  kg/L, based on the OPP standard pond.  If the  modeling incorporates EECs from a
source other than PRZM/EXAMS, a CSs value relevant to the other EECs should be utilized.
B.5. Sediment Organic Carbon (OC)

Sediment organic carbon (OC) is a parameter that influences organisms that consume sediment.
As  OC  increases, the concentration of the pesticide in the solid component of the sediment
increases to the extent that the pesticide sorbs to organic matter. As the pesticide concentration in
sediment increases,  the  pesticide concentration in  organisms  that consume  sediment  also
increases.

OC is entered by the model user as % of the dry weight of the sediment. The default value for
this parameter is 4.0%, based on  the OPP standard pond.  If the modeling incorporates EECs
from a source other than PRZM/EXAMS, an OC value relevant to  the other EECs should be
utilized.
                                       74 of 123

-------
 Appendix C. Explanation of Default Values  Representing Biotic Characteristics  of the
Aquatic Ecosystem, Including Food Web Structure

The seven trophic levels of the aquatic ecosystem of KABAM are phytoplankton, zooplankton,
benthic invertebrates, filter feeders,  small fish, medium fish, and large fish. In KABAM,  each
trophic level is defined by its  % lipid, % Non Lipid Organic Matter (NLOM), % water, body
weight, and diet. Each of these trophic levels is described within this Appendix, with emphasis
on the information relevant to KABAM and explanations of default parameters used to define
these trophic levels (in Tables 5 and 6 of the KABAM tool). If the model user wishes to explore
the influences  of changes  in parameter values representing the aquatic food web on EECs and
RQs for birds  and mammals, this can be accomplished by  altering parameter values within the
range of reported values for a specific parameter.

Although  the  %  water  composition of  an  aquatic  organism  does  not  influence  the
bioaccumulation of  a  chemical  in that  organism  (see  Appendix A),  it is  an important
consideration for the definition of % lipid and the percent non-lipid organic matter (% NLOM).
Often, tissue analysis results and body weight data in the scientific literature are reported  on a
dry weight basis. For KABAM,  input parameters for body composition are entered on a wet
weight basis.  Therefore,  % water  composition  is discussed in the  sections below since  it is
necessary to understand the water composition of an organism in order to translate the reported
data into input parameters for KABAM.

Lipid composition of  an organism can influence  the  bioaccumulation  of a  chemical  (See
Appendix A), with higher lipid composition leading to higher accumulation.  Since KABAM is
intended for use in ecological risk assessments of pesticides with the  potential to bioaccumulate
in aquatic ecosystems, it is necessary for this tool to serve as a conservative representation of
bioaccumulation. Default parameter values for % lipid were selected from the open literature
and are intended to represent the high-end of available data (75th-90th percentiles).
C.I. Phytoplankton

Phytoplankton are microscopic autotrophic aquatic organisms that derive their nutrition from
photosynthesis.  Groups  of freshwater phytoplankton include algae (green, yellow-green and
golden-brown), cyanobacteria  (blue-green algae), diatoms and  dinoflagellates.  Phytoplankton
can be unicellular, colonial, or filamentous.  These organisms have limited mobility that is based
on water movements; however, some are able to move via  flagella.  An aquatic habitat will
generally contain an assemblage of phytoplanktonic species that vary in proportion over time and
space (Wetzel 1983).

For parameterization of KABAM, it is necessary to define the % water, % lipid and % NLOM
contents of phytoplankton.  The body weight is not a necessary  input for phytoplankton, nor is
the diet composition since these organisms do not consume other organisms.
                                        75 of 123

-------
Since it is assumed that phytoplankton are present in the water column of the aquatic ecosystem
where  photosynthesis  can  occur, it is assumed that phytoplankton do  not reside in  benthic
sediment and do not respire pore water. This should be indicated in Table  5 of the KABAM tool
(i.e., "no" should be entered in the column titled: "Do organisms in trophic level respire some
pore water?").

Aquatic plant tissues are composed of approximately 90% water by weight (Hannan and Dorris
1970; Raven et al. 1999, Sladecek and Sladeckova 1963).  The default parameter for the water
composition of phytoplankton is 90%.

Reported % lipid values for phytoplankton vary from 2-27% of dry weight.  If it is assumed that
phytoplankton are composed of 90% water, then this range of lipid compositions is equivalent to
0.2-2.7% on a wet weight basis (Table Cl).  For KABAM, the default parameter for % lipid
of phytoplankton was selected as  2%  to represent a high-end  estimate (75th  to  90th
percentile of data in Table Cl) of this parameter.

The wet weight of an organism is the sum of the water, lipid, and NLOM content.  If the water
content of phytoplankton is 90% of the wet weight, and the % lipid is known (2%), the NLOM
content of phytoplankton is the % remaining after subtracting the water and lipid content from
100%. Therefore, the default parameter for the NLOM composition of phytoplankton is
8%.
                                      76 of 123

-------
Table Cl. Percent lipid composition of freshwater phytoplankton (under culture conditions) reported in
scientific literature.
the
Species
Not stated
Anabaena sp.
Anabaena sp.
Anabaena sp.
Chamydomonas reinhardtii
Chlamydomonas applanata
Chlamydomonas applanata
Chlorella ellipsoidea
Chlorella pyrenoidosa
Chlorella pyrenoidosa
Chlorella pyrenoidosa
Chlorella pyrenoidosa
Chlorella vulgaris
Chlorella vulgaris
Cryptomonas pyrenoidifera
Microcystis aeruginosa
Nannochloris sp.
Nitzschia palea
Oocystis polymorpha
Ourococcus sp.
Scenedesmus acutus
Scenedesmus obliquus
Selanastrum gracile
Selenasrum capricornutum
Selenasrum capricornutum
Selenasrum capricornutum
Synedra sp.
Synedra sp.
Synedra sp.
Synedra ulna
Mean%
Lipid (dry
weight basis)
Not stated
6.8 (0.4)
5.3 (2.4)
2.2 (0.2)
10.8 (6.2)
18.2
16
13.5
13.4
14.4
16.4
16
12.5
13
8.5 (5.1)
5.8 (2.3)
20.2
22.2
12.6
27
6.4 (2.5)
19
20.8
19.5 (0.2)
16.0 (0.3)
8.0 (0.9)
7.5 (1.6)
13.7 (0.7)
11. 7 (4. 5)
23
Mean%
Lipid (wet
weight basis)
0.5
0.68*
0.53*
0.22*
.08*
.82*
.60*
.35*
.34*
.44*
.64*
.60*
.25*
.30*
0.85*
0.58*
2.02*
2.22*
1.26*
2.70*
0.64*
1.90*
2.08*
1.95*
1.60*
0.80*
0.75*
1.37*
1.17*
2.30*
Source
Oliver and Niimi 1988
Stange and Swackhamer 1994
Stange and Swackhamer 1994
Stange and Swackhamer 1994
Lurling and Van Donk 1997
Shifrin and Chisholm 1981
Shifrin and Chisholm 1981
Shifrin and Chisholm 1981
Shifrin and Chisholm 1981
Shifrin and Chisholm 1981
Shifrin and Chisholm 1981
Shifrin and Chisholm 1981
Shifrin and Chisholm 1981
Shifrin and Chisholm 1981
Lurling and Van Donk 1997
Lurling and Van Donk 1997
Shifrin and Chisholm 1981
Shifrin and Chisholm 1981
Shifrin and Chisholm 1981
Shifrin and Chisholm 1981
Lurling and Van Donk 1997
Shifrin and Chisholm 1981
Shifrin and Chisholm 1981
Stange and Swackhamer 1994
Stange and Swackhamer 1994
Stange and Swackhamer 1994
Stange and Swackhamer 1994
Stange and Swackhamer 1994
Stange and Swackhamer 1994
Shifrin and Chisholm 1981
Average
75th percentile
90th percentile

1.4
1.8
2.1

* Calculated from reported % lipid based on dry weight and assumption that algae wet weight is 90% water.
                                             77 of 123

-------
C.2. Zooplankton

Zooplankton are aquatic animals that are suspended in water. This group is primarily composed
of rotifers, cladocera and  copepods, but also includes protozoa and insects at immature life
stages.  Species of zooplankton primarily consume phytoplankton but also consume detritus,
bacteria,  yeast, and  other (smaller) zooplankton  (Wetzel  1983).  For  parameterization of
KABAM, zooplankton is represented by herbivorous  species that have a diet composed
100% of phytoplankton.

Since it is assumed that zooplankton are present in the water column of the aquatic ecosystem
and do not reside in  the benthic sediment, it is assumed  that zooplankton  do not respire pore
water. This should  be indicated in Table 5 of the KABAM tool (i.e., "no" should be entered in
the column titled: "Do organisms in trophic level respire some pore water?").

Beers (1966) reported water  compositions of several groups of marine zooplankton inhabiting
the Atlantic Ocean.  Average % water composition of these groups ranged 74-96%, with an
average % water composition of 86% corresponding to copepods. Based on this information, the
default % water composition of zooplankton is 85%.  This value is used to translate dry
weight data into equivalent wet weight values.

Reported mean %  lipid values for  zooplankton vary from  6.4-24.3% of dry weight.  If it is
assumed  that zooplankton are composed of 85% water, then this range of lipid compositions is
equivalent to 0.96-3.6% on  a wet  weight basis (Table C2). Based on  this information, the
default % lipid for zooplankton is  set to 3% to represent a high-end (75th to 90th percentile)
estimate of this parameter.

The wet weight of  an organism is the sum of the water, lipid, and NLOM content.  If the water
content of zooplankton is  85% of the wet weight, and the % lipid is known (3%), then NLOM
content of zooplankton is the % remaining after subtracting the water and lipid content from
100%. Therefore, the default parameter for the NLOM composition of zooplankton is 12%.

Wright (1958) provided biomass data for two species of zooplankton (Daphnia longispina and
D. pulex) in a reservoir in Montana,  where the average body weight of zooplankton was 1.3xlO"7
kg-wet weight (assuming 85% water content; range 0.9-1.6xlO"7 kg-wet weight). Acharya et al.
(2005) provided dry body weights for Bosmina freyi that translate to approximately 0.3-3xlO"8
kg-wet weight (assuming 85% water content). Jeppesen et al. (2004) provided body weight data
for Daphnia sp. that translate to approximately 0.67-3.3xlO"7 kg-wet weight (assuming 85%
water content) Based on this information,  the default weight for zooplankton is set to 1x10 7
kg-wet weight, with the intention of being a representative weight of species  of zooplankton.
                                       78 of 123

-------
Table C2. Percent lipid composition of freshwater zooplankton reported in the scientific literature.
Zooplankton identification
Daphnia magna (cladoceran)
Unspecified
Mostly cladocerans, also copepods and
rotifers
Mostly cladocerans, also copepods and
rotifers
Mostly cladocerans, also copepods and
rotifers
Leptodora kindtii
Mostly cladocerans, also copepods and
rotifers
Mostly cladocerans, also copepods and
rotifers
Mostly cladocerans, also copepods and
rotifers
Cyclopedia
Chydorus sphaericus
Bosmina coregoni
Eurytemora affmus
Daphnia hyalina
Mean% Lipid
(dry weight
basis)
6.4-19.7
6.7*
10.8 (3.6)
12.1 (3.0)
12.2 (2.4)
13.1 (1.0)**
13.7 (1.9)
13.9 (1.9)
14.6 (1.0)
15.9 (1.8)**
18.5 (2.8) **
20.5 (1.9) **
23.6 (2.7) **
24.3 (5.3)**
Mean% Lipid
(wet weight
basis)
0.96-3.0*
1.00.33
1.6*
1.8*
1.8*
2.0
2.1*
2.1*
2.2*
2.4
2.8
3.1
3.5
3.6
Source
McKee andKnowles 1987
Morrison et al. 1997
Mitra et al. 2007
Mitra et al. 2007
Mitra et al. 2007
Vijverberg and Frank 1976
Mitra et al. 2007
Mitra et al. 2007
Mitra et al. 2007
Vijverberg and Frank 1976
Vijverberg and Frank 1976
Vijverberg and Frank 1976
Vijverberg and Frank 1976
Vijverberg and Frank 1976
Average
75th percentile
90th percentile
* Calculated
water.
**Expressed
basis.

2.3
2.9
3.3
from reported % lipid based on dry weight and assumption that
as % of total organic matter attributed to lipid. It is assumed that

zooplankton wet weight
this is equivalent to a dry
is 85%
weight
C.3. Benthic invertebrates

The benthic invertebrate trophic  level includes  animals  that inhabit the sediments of aquatic
habitats. Benthic invertebrates include a diverse group of animals, including crustaceans (e.g.,
crayfish,  amphipods), aquatic worms (e.g.,  oligochaetes), aquatic insect larvae (e.g., Diptera,
caddisflies, beetles, mayflies and dragonflies), protozoa, snails, and nematodes. Different species
of benthic invertebrates have a variety of feeding strategies, including herbiovory, detritivory,
and predation upon other benthic invertebrates (Covich et al. 1999).  In order to represent all of
these feeding strategies with the benthic invertebrate trophic level of KABAM, it is assumed that
benthic invertebrates consume organic matter from sediment, phytoplankton, and zooplankton in
equal quantities  Therefore, the default diet composition of benthic invertebrates is 34%
sediment, 33% phytoplankton, and 33% zooplankton.

Since it is assumed  that benthic invertebrates are present in the benthic compartment of the
aquatic ecosystem, it is assumed that benthic invertebrates respire sediment pore water. This
                                        79 of 123

-------
should be indicated in Table 5 of the KABAM tool (i.e., "yes" should be entered in the column
titled: "Do organisms in trophic level respire some pore water?").

Available water composition data for benthic invertebrates include a range of 69-83% water
(Table C3).  The average value of the available data is 76%. Based on this average, the default
value for KABAM representing the % water of benthic invertebrates is 76%.
Tab!
e C3. Water composition (%) of benthic invertebrates reported in the scientific literature.
Organism
Crayfish (Orconectus propinquus)
Hyalella azteca
Mayfly larvae
Diporeia sp.,
Crayfish (Astacus fluviatilis)
Lumbriculus variegatus (bligochaete)
Crayfish (Astacus, Orconectus and Procambarus)
Mean% Water
69
72.5
73
73.1
80.0
81
82.5*
Source
Gewurtz et al. 2000
Lotufoetal. 2001
Gewurtz et al. 2000
Lotufoetal. 2001
Sidwell 1981
Liebig et al. 2005
USDA 2005
     *excluding the shell

Lipid data are available  for various freshwater crustaceans,  oligochaetes, and  insect larvae.
These values indicate a wide range (approximately 1 to 10% of wet weight) of lipid composition
of  benthic  invertebrates  (Tables C4-C9)    The  default  lipid  composition for  benthic
invertebrates is  3%.  This value was selected to be representative of a high-end value (75
percentile) of available lipid compositions for freshwater benthic invertebrates (Table CIO).
th
Table C4. Lipid composition (%) of Hyalella azteca (a freshwater crustacean) reported in the scientific
literature.
Source
Lotufoetal. 2001
Lotufoetal. 2001
Lotufoetal. 2001
Lotufoetal. 2001
Lotufo et al. 2000
Lotufo et al. 2000
Lotufo et al. 2000
Kane Driscoll and Landrum 1997
Lotufo et al. 2000
Lotufo et al. 2000
Kane Driscoll et al. 1997
Kane Driscoll et al. 1997
Mean% Lipid
(dry weight basis)
2.4*
3.2*
6.3*
6.5*
6.9 (0.9)
7.0 (1.1)
7.2 (0.8)
7.5 (1.5)
7.5 (0.9)
7.7 (1.5)
8.2 (0.7)
8.4(0.7)
Mean% Lipid
(wet weight basis)
0.660.03
0.880.04
1.730.25
1.790.41
1.9*
1.9*
2.0*
2.1*
2.1*
2.1*
2.3*
2.3*
Average
75th percentile
90th percentile
6.6
7.6
8.2
1.8
2.1
2.3
               *Calculated from reported % lipid and assumption that dry:wet weight ratio for
               H. azteca is 0.275 (based on Lotufo et al. 2001).
                                         80 of 123

-------
Table C5. Lipid composition (%) of freshwater crayfish (crustaceans) reported in the scientific literature.
Genus
Astacus fluviatilis
Orconectus
Astacus, Orconectus and Procambarus
Undefined
Orconectes propinquus
Orconectes
Procambarus
Procambarus
Mean%
Lipid (wet
weight basis)
0.5
0.86 (0.11)
1.0
1.9 (0.47)
2.4 (0.26)
2.52 (0.16)
2.95 (1.25)
3.02 (1.29)
Source
Sidwell 1981
White et al. 1998
USDA 2005
Morrison et al. 1997
Morrison et al. 2000
Gewurtz et al. 2000
Lin et al. 2004
Lin et al. 2004
Average
75th percentile
90th percentile
1.9
2.6
3.0

Table C6. Lipid
composition (%) ofDiporeia sp. (freshwater crustaceans) reported in the scientific literature.
Source
Landrum et al. 2007
Landrum et al. 2007
Landrum et al. 2007
Kane Driscoll et al. 1997
Kukkonen et al. 2004
Kane Driscoll et al. 1997
Lotufoetal. 2001
Kukkonen et al. 2004
Lotufoetal. 2001
Lotufo et al. 2000
Lotufo et al. 2000
Kane Driscoll and Landrum 1997
Lotufoetal. 2001
Lotufoetal. 2001
Mean% Lipid
(dry weight basis)
10.78 (1.5)
11. 97 (0.38)
17.1 (0.64)
20.1 (4.6)
20.4*
21.3 (6.7)
22.2*
23.0*
23.3*
23.7 (8.5)
23.9 (6.3)
27.2 (1.3)
40.3*
43.1*
Mean% Lipid
(wet weight basis)
2.9*
3.2*
4.6*
5.4*
5.50.7
5.7*
5.970.75
6.21.4
6.271.21
6.4*
6.4*
7.3*
10.850.62
11.591.18

Average
75th percentile
90th percentile
23.5
23.9
36.4
6.3
6.4
9.8
                * Calculated from reported % lipid and assumption that dry: wet weight ratio for
                Diporeia sp. is 0.269 (based on Lotufo et al. 2001).
                                              81 of 123

-------
Table  C7. Lipid composition  (%) of Lumbriculus variegatus (a freshwater oligochaete) reported  in  the
scientific literature.
Source
Croce et al. 2005
Kukkonen et al. 2004
Liebig et al. 2005
Kukkonen et al. 2004
Kukkonen and Landrum 1994
Fisketal. 1998
Kukkonen and Landrum 1994
Fisketal. 1998
Fisketal. 1998
Kukkonen and Landrum 1994
Fisketal. 1998
Fisketal. 1998
Fisketal. 1998
Fisketal. 1998
Mean% Lipid
(dry weight basis)
5.8*
6.3*
8 (0.4)
7.9
9.2 (0.9)
10.5*
11.1 (1.4)
12.1*
13.2*
13.2 (4.3)
15.3*
17.9*
18.9*
19.5*
Mean% Lipid
(wet weight basis)
1.10.1
1.20.13
1.5*
1.50.19
1.7*
2.00.2
2.1*
2.30.2
2.50.3
2.5*
2.90.3
3.40.8
3.60.8
3.70.6
Average
75th percentile
90th percentile
12.1
14.8
18.6
2.3
2.8
3.5
              *Calculated from reported % lipid and assumption that water composition of L.
              variegatus is 81% (Liebig et al. 2005).
Table C8. Lipid composition (%) of other freshwater oligochaetes reported in the scientific literature.
Organism Identification
Tubifex tubifex and
Limnodrilus hoffmeisteri
Ilyodrilus templetoni**
Ilyodrilus templetoni**
Ilyodrilus templetoni**
Ilyodrilus templetoni**
Ilyodrilus templetoni**
Ilyodrilus templetoni**
Oligochaete
Limnodrilus sp.
Oligochaete
Mean% Lipid
(dry weight basis)
5.3*
5.85 (2.28)
6. 11 (0.55)
6.72 (1.59)
7.44 (1.33)
7.35 (1.26)
8.82 (1.60)
9.5 (1.0)
11. 93 (0.16)
12.8 (1.8)
Mean% Lipid
(wet weight basis)
1*
1.1*
1.2*
1.3*
1.4*
1.4*
1.7*
1.8*
2.3*
2.4*
Source
Oliver and Niimi 1988
Lu et al. 2003
Lu et al. 2003
Lu et al. 2003
Lu et al. 2003
Lu et al. 2003
Lu et al. 2003
Landrum et al. 2007
Jonker et al. 2004
Landrum et al. 2007
Average
75th percentile
90th percentile
8.2
9.3
12.0
1.6
1.8
2.3

   *Calculated from reported % lipid and assumption that water composition of L. variegatus is 81% (Liebig et
   al. 2005).
   **Mean of values for/, templetoni is 1.4% (wet weight). When this value is used in calculating the mean and
   percentile values  for the group of oligochaetes, the mean is 1.8% (wet weight). The 75th and 90th percentiles
   are 2.2 and 2.4, respectively.
                                              82 of 123

-------
Table C9. Lipid composition (%) of freshwater insect larvae reported in the scientific literature.
Organism Identification
Chironomus riparius
Hexagenia limbata (mayfly larvae)
H. limbata and H. rigida
Caddisfly larvae
Mayfly larvae
Mean%
Lipid (wet
weight basis)
0.6
1.5 (0.05)
1.50 (0.052)
1.7
2.0 (0.25)
Source
Leonards et al. 1997
Morrison et al. 2000
Gewurtz et al. 2000
Morrison et al. 1997
Morrison et al. 1997
Average
75th percentile
90th percentile
1.5
1.7
1.9

Table CIO. Mean lipid composition (%, wet weight basis) of freshwater benthic invertebrates from data in
Tables C4-C9.
Benthic Invertebrate
Insect Larvae
Hyalella azteca (a freshwater crustacean)
Freshwater oligochaetes (excluding L. variegatus)
Crayfish (freshwater crustaceans)
Lumbriculus variegatus (a freshwater oligochaete)
Diporeia sp. (freshwater crustaceans)
Mean
1.5
1.8
1.8
1.9
2.3
6.3
75th
Percentile
1.7
2.1
2.2
2.6
2.8
6.4
90th
Percentile
1.9
2.3
2.4
3.0
3.5
9.8
The wet weight of an organism is the sum of the water, lipid, and NLOM content. By default, if
the water content of benthic invertebrates is 76% of the wet weight, and the % lipid is known
(default = 3%), the NLOM content of benthic invertebrates is the % remaining after subtracting
the water and lipid content from 100%. Therefore, the  default parameter for  the NLOM
composition of benthic invertebrates is 21%.

The benthic invertebrate trophic level is composed of a wide variety of taxonomic groups.  The
body weights of organisms within this group can vary by orders of magnitude (Table Cl 1).  The
default weight for benthic invertebrates is set to IxlO"4 kg-wet weight, with the intention of
being representative of a midpoint weight of species of benthic invertebrates.
Tabk
Cll. Body weights (wet) of freshwater benthic invertebrates reported in the scientific literature.
Benthic Invertebrate
Amphipods
Mayfly larvae
Chironomids
Caddisfly larvae
Snails
Crayfish
Weight (kg)
O.OSxlO'4
0.16xlO'4*
0.24xlO"4
0.32xlO'4*
0.82xlO"4
IS.OxlO'4
Source
Leonards et al. 1997
Morrison et al. 1997
Leonards et al. 1997
Morrison et al. 1997
Leonards et al. 1997
Morrison et al. 1997
      Converted from reported dry weight to wet weight assuming 75% water content.
                                        83 of 123

-------
C.4. Filter Feeders

Filter feeders  are benthic invertebrates that are distinguished by  their feeding  habits.  These
organisms feed by straining water and extracting organic material such as detritus and plankton.
Examples of freshwater filter feeders include mollusks.  For KABAM, it is assumed that filter
feeders   consume  materials   suspended  in  the  water  column,  including  phytoplankton,
zooplankton, and detritus.  It is also assumed  that filter feeders consume suspended sediment
incidentally. The default composition of the diet of this trophic level is 34% sediment, 33%
phytoplankton, and 33% zooplankton.

Since it is assumed that filter feeders are present in the benthic sediment compartment  of the
aquatic ecosystem, it is also assumed that filter feeders respire sediment pore water. This should
be indicated in Table 5  of the KABAM tool (i.e., "yes" should be entered in the column titled:
"Do organisms in trophic level respire some pore water?").

According to available  data, water composition of freshwater mollusks ranges 78-93% (Table
C12)  The default water content of filter feeders is set to 85%, based on the midpoint of the
range of available data.
Tabl
e C12. Water composition (%) of freshwater mollusks reported in the scientific literature.
Identification
Corbicula strata (freshwater clam)
Corbicula japonica (freshwater clam)
Corbicula sandai (freshwater clam)
Corbicula fluminea (freshwater clam)
lamellibrancha clams (subclass)
Corbicula leana (freshwater clam)
Dreissena polymorpha fzebra mussel)
D. polymorpha
Anodonta anatine (mussels)
Mean % water*
77.6
79.8
80.0
81.4
82
82.1
87
88-93
90.6-92.8
Source
Sidwell 1981
Sidwell 1981
Sidwell 1981
Sidwell 1981
USDA 2005
Sidwell 1981
Bervoets et al. 2005
Hendriks et al. 1998
Hyorylainen et al. 2002
     *It is assumed that this does not include the shell.
Data on lipid content are available for several species of freshwater mollusks. These values range
0.4-4% of wet weight (Tables C13 - CIS).  The default lipid composition for filter feeders is
2%. This value was selected to be representative of a high end (75th percentile of Dreissena sp.
and Corbicula sp.) value of available lipid compositions for freshwater mollusks.
                                        84 of 123

-------
Table  C13. Percent lipid  composition
literature.
of Dreissena sp.  (freshwater  mollusks)  reported in the scientific
Species
D. polymorpha
D. polymorpha
D. polymorpha
D. polymorpha
D. bugensis
D. bugensis
D. polymorpha
D. polymorpha
D. polymorpha
D. polymorpha
D. polymorpha
D. polymorpha
D. polymorpha
Mean % Lipid
(dry weight basis)
4.3*
14
8.5*
9.1
9.0 (1.4)
10 (0.5)
11 (0.6)
10.8*
11.5*
12.3*
12 (4.4)
17
18
Mean % Lipid
(wet weight basis)
0.55
1
1.1
1.2*
1.2*
1.3
1.4*
1.4 (0.1)
1.5 (0.1)
1.6 (0.1)
1.6*
2
2
Source
Bervoets et al. 2005
Hendriks et al. 1998
Kwon et al. 2006
Becker van Slooten and Tarradellas 1994
Marvin et al. 2002
Marvin et al. 2002
Marvin et al. 2002
Kwon et al. 2006
Kwon et al. 2006
Kwon et al. 2006
Marvin et al. 2002
Hendriks et al. 1998
Hendriks et al. 1998
Average
75th percentile
90th percentile
11.3
12.3
16.4
1.4
1.6
1.9

 et al. 2005).
                                                                    of D. polymorpha is 87% (Bervoets
Table C14. Percent lipid composition of Corbicula sp. (freshwater clams) reported in the scientific literature.
Species
C. leana
C. japonica
C. japonica
C. fluminea
C. sandai
C. strata
Mean % Lipid (wet
weight basis)
1.1
1.2*
1.2
1.5
2.4
4.0
Source
Sidwell 1981
Kang et al. 2002
Sidwell 1981
Sidwell 1981
Sidwell 1981
Sidwell 1981
Average
75th percentile
90th percentile
1.9
2.2
3.2

       *Based on reported mean lipid content of 5.8% dry weight and 80% moisture content reported for
       this species by Sidwell 1981.
                                              85 of 123

-------
Table CIS. Percent lipid composition of other freshwater filter feeders reported in the scientific literature.
Identification
Sphaerium striantium (fingernail clam)
Elliptic complanata
Anodonta anatine (mussels)
Anodonta anatine (mussels)
Lamellibrancha (clams)
Anodonta anatine (mussels)
Anodonta anatine (mussels)
Mean % Lipid
(dry weight basis)
8.7
3.2 (1.2)
11. 2 (0.8)
12.2 (0.7)
5.5
10.9 (0.6)
11. 3 (0.9)
Mean % Lipid
(wet weight basis)
0.36
0.48*
0.81
0.98
1.0
1.02
1.05
Source
Rice and White 1987
Marvin et al. 2002
Hyotylainen et al. 2002
Hyotylainen et al. 2002
USDA 2005
Hyotylainen et al. 2002
Hyotylainen et al. 2002
 *Calculated using assumption that filter feeders are 85% water.

The wet weight of an organism is the sum of the water, lipid, and NLOM content. By default, if
the water content of filter feeders is 85% of the wet weight and the % lipid is known (default =
2%), the NLOM content of filter feeders is the % remaining after subtracting the water and lipid
content from 100%. Therefore, the default parameter for the NLOM composition of filter
feeders is 13%.

Reported wet weights of various species of mollusks range 0.2-12 xlO"3 kg. Mean wet weights of
Dreissenapolymorpha have been reported as 0.41  0.26 xlO"3 kg (Van Haelst et al. 1996). Wet
weights of C. fluminea ranged approximately 0.2-2 xlO"3 kg (Andres et al. 1999, Vidal et al.
2002). Hyotylainen et al. (2002) reported wet weights of Anodonta anatine tissue as ranging 4.5-
12.1 xlO"3 kg. Based on this information, the default weight of filter feeders is set to 1 xlO"3
kg, with the intention of being a representative weight of mollusks.
C.5. Fish (Small, Medium and Large Sizes)

There are hundreds of species of fish inhabiting fresh waters of the United States and Canada,
including ponds, lakes, streams, and rivers. Species of bluegill and other sunfish (Lepomis spp.),
bass (Micropterus spp.), and crappie  (Pomoxis spp.) are common inhabitants of fresh warm
water ponds, lakes, and  streams distributed throughout the continental United States (Page and
Burr 1991, Carlander 1977).  As described  below, these species were used to define default
parameters for the small, medium, and large fish  in KABAM.  Although there  are many other
species  of  fish  in ponds of the U.S. (e.g., perch,  minnows), sunfish,  crappie, and bass were
considered representative offish that are found in  freshwaters of the U.S., and thus suitable for
models for defining input parameters for use in KABAM.

Several  bird  and mammal species (e.g., belted kingfisher [Megaceryle  alcyon], northern river
otter) consume amphibians, in addition to fish.  For KABAM, it is assumed that the default fish
also represent, i.e.,  serve as  surrogates for, aquatic-phase amphibians, such as salamanders and
frogs.  This assumption  is consistent with OPP's policy in which exposure and  effects data for
fish are assumed to be representative of aquatic-phase amphibians (USEPA 2004).

Default parameters for  small fish in KABAM are designed  to represent the young-of-year
(YOY), i.e., fish that have hatched within the year, before January 1  of the next year, of sunfish,
bass and  crappie.    YOY of these  species  consume copepods,  cladocerans,  rotifers (i.e.,
                                        86 of 123

-------
zooplankton),  chironomid larvae, and  mayfly larvae (i.e., benthic invertebrates)  (Carlander
1977). Average body weights of YOY of sunfish, bass, and crappie are provided in Table C16.
For  KABAM, it  is assumed  that the small  fish weighs  0.01 kg  and  its diet is  50%
zooplankton and 50%  benthic invertebrates.

Table C16. Average body weights for young of the year fish (Source: Carlander 1977).
Species (scientific name)
Green sunfish (Lepomis cyanellus)
Pumpkinseed (L. gibbosus)
Warmouth (L. gulosus)
Bluegill (L. macrochims)
Redear sunfish (L. microlophus)
Largemouth bass (Microptems salmoides)
White crappie (Pomoxis annularis)
Black crappie (P. nigromaculatus)
Average body weight
(kg)
0.001-0.01
0.002
0.011
0.0001-0.05
0.0006-0.04
0.0002-0.02
0.0002-0.01
0.0005-0.02
The medium fish in KABAM is designed to represent adult sunfish and crappie. These fish reach
sexual maturity between ages 1 and 3, with lifespans >6 years. Their diets include insects, insect
larvae, crustaceans, snails, and other fish (Carlander 1977). Mature fish range in weight,  0.005-
0.579 kg,  depending upon their age (Table  C17; data from Carlander 1977). Although mature
fish display a wide range of weights, most  species weigh approximately 0.1 kg as adults.  For
KABAM, it is assumed that the medium-sized fish weighs 0.1 kg and its diet is 50% benthic
invertebrates and 50% small fish.

Table C17. Average body weights (in kg) of medium fish at different ages.
Species (scientific name)
Green sunfish (Lepomis cyanellus)
Pumpkinseed (L. gibbosus)
Warmouth (L. gulosus)
Bluegill (L. macrochirus)
Redear sunfish (L. microlophus)
White crappie (Pomoxis annularis)
Black crappie (P. nigromaculatus)
lyr
0.01
0.005
0.011
0.014
0.026
0.031
0.037
2yr
0.024
0.034
0.046
0.052
0.081
0.085
0.097
3yr
0.048
0.034
0.046
0.052
0.125
0.123
0.143
4 yr
0.086
0.063
0.085
0.090
0.187
0.181
0.210
5 yr
0.086
0.099
0.163
0.141
0.187
0.346
0.289
6yr
0.132
0.157
0.163
0.141
0.265
0.346
0.363
7yr
-
0.157
-
0.208
-
0.579
0.468
Syr
-
0.157
-
0.208
-
0.579
0.468
- Indicates data were not available

The large fish in KABAM is designed to represent the largemouth bass (Microptems salmoides),
which is a predatory fish commonly found in warm waters throughout the continental United
States.  It is also designed to be representative of large predatory fish that are consumed by
mammals  and birds. The diet of largemouth bass  is composed primarily  of fish,  including
sunfish, crappie, perch,  shad  and smaller-sized  largemouth bass.  Largemouth bass will  also
consume crayfish, especially when no other fish are available.   Largemouth bass become
sexually mature between ages 2-5, with a lifespan reaching beyond 10 years.  Adult largemouth
bass weigh 0.25-3.6 kg, depending upon  their age (Carlander 1977). For  KABAM, it  is
assumed that the large fish weighs 1 kg, and consumes 100% medium-sized fish.
                                       87 of 123

-------
Since small and medium fish consume benthic invertebrates, it is assumed that these fish are
sometimes present in the benthic compartment of the aquatic ecosystem. Therefore, it is assumed
that  small and medium fish respire some pore  water.  It is assumed  that medium fish are
predominantly present in the water column of the aquatic ecosystem, where they are consumed
by large fish. It is assumed that large fish do not respire pore water. This should be indicated in
Table 5 of the KABAM tool (i.e., "yes" should be entered for small and medium fish and "no"
should be entered for large fish in the column titled: "Do organisms in trophic level respire some
pore water?").

Available water composition data for Lepomis sp., Pomoxis sp., and Micropterus sp.  include a
range of 71-80%  water (Table CIS). Although water composition data were not available for
largemouth bass, data do exist for smallmouth bass (M. dolomieu) and are used as a surrogate for
largemouth bass.  The average value of the available data is 73%. Based on this average, the
default value for KABAM representing the % water of all fish is 73%.

Table CIS. Water composition data for fish relevant to small, medium, and large default fish of KABAM.
Species (scientific name)
Black crappie (Pomoxis
nigromaculatus)
Smallmouth bass
(Micropterus dolomieu)
Smallmouth bass
(Micropterus dolomieu)
Smallmouth bass
(Micropterus dolomieu)
Black crappie (Pomoxis
nigromaculatus)
Smallmouth bass
(Micropterus dolomieu)
Black crappie (Pomoxis
nigromaculatus)
Black crappie (Pomoxis
nigromaculatus)
Black crappie (Pomoxis
nigromaculatus)
Smallmouth bass
(Micropterus dolomieu)
Bluegill (L. macrochirus)
Reported Body
Weight (kg)
0.102 (0.007)
0.277 (0.0901)
0.870 (0.0685)
0.326 (0.177)
0.148 (0.021)
0.395 (0.222)
0.114 (0.016)
0.111 (0.015)
0.0798 (0.012)
0.154 (0.0647)
Not reported
Corresponding
Default fish
Medium
Medium-Large
Medium-Large
Medium-Large
Medium
Medium-Large
Medium
Medium
Medium
Medium
unknown
% water
70.7 (0.29)
71.1 (1.26)
71.3 (1.76)
71.9 (1.44)
71.9 (0.84)
72.0 (0.99)
72.1 (0.87)
72.6 (0.38)
73.2 (0.59)
73. 4 (2. 11)
79.5
Source
Sethajintanin et al.
2004
Sethajintanin et al.
2004
Sethajintanin et al.
2004
Sethajintanin et al.
2004
Sethajintanin et al.
2004
Sethajintanin et al.
2004
Sethajintanin et al.
2004
Sethajintanin et al.
2004
Sethajintanin et al.
2004
Sethajintanin et al.
2004
Sidwell 1981
Lipid content of fish reported in the literature varies widely for Lepomis sp., Pomoxis sp., and
Micropterus sp. from 0.5-8% on a wet weight basis, with an average value of 2.9% and a 75th
percentile of 4.0% (Table C19).  Table C19 includes lipid composition data for wild-caught and
laboratory-reared Lepomis sp., Pomoxis sp.,  and Micropterus  sp. Several  lipid  content values
available in the literature cannot be related to the weights of the fish analyzed due to a lack of
information included in the individual studies.  Thus, these lipid contents  cannot be related to
one  of KABAM's default fish. Based on this and the data in  Table  C19, the default lipid
composition for all three fish is set to 4%, to be representative of a high-end value.
                                       88 of 123

-------
Table C19. Lipid composition data for fish relevant to small, medium, and large default fish of KABAM.
Species (scientific name)
Green sunfish
(Lepomis cyanellus)
Bluegill (L. macrochirus)
Largemouth bass
(Micropterus salmoides)
Largemouth bass
(Micropterus salmoides)
Largemouth bass
(Micropterus salmoides)
White crappie
(Pomoxis annularis)
Longear sunfish
(L. megalotis)
Bluegill (L. macrochirus)
Largemouth bass
(Micropterus salmoides)
Largemouth bass
(Micropterus salmoides)
Largemouth bass
(Micropterus salmoides)
Largemouth bass
(Micropterus salmoides)
Bluegill (L. macrochirus)
Bluegill (L. macrochirus)
Smallmouth bass
(Micropterus dolomieu)
White crappie (Pomoxis
annularis)
Bluegill (L. macrochirus)
Black crappie (Pomoxis
nigromaculatus)
Bluegill (L. macrochirus)
Black crappie
(Pomoxis nigromaculatus)
Smallmouth bass
(Micropterus dolomieu)
Black crappie
(Pomoxis nigromaculatus)
White crappie
(Pomoxis annularis)
Smallmouth bass
(Micropterus dolomieu)
Reported Body
Weight (kg)
Not reported
0.012 (0.0012)
Not reported
Not reported
Not reported
Not Reported
Not reported
Not reported
Not reported
Not reported
Not reported
Not reported
Not reported
(juveniles)
Not reported
(adult males)
Not reported
Not Reported
Not reported
(adult females)
0.1 14 (0.0 16)
Not reported
0.111 (0.015)
Not reported
0.148 (0.021)
Not Reported
0.395 (0.222)
Corresponding
Default fish
Unknown
Small
Small
(defined based on
length data)
Small
(defined based on
length data)
Small
(defined based on
length data)
Assume medium
(spawning fish)
Unknown
Unknown
Unknown
Small
(defined based on
length data)
Small
(defined based on
length data)
Small
(defined based on
length data)
Presume small
Presume medium
Unknown
Assume medium
(spawning fish)
Presume medium
Medium
Unknown
Medium
Unknown
Medium
Assume medium
(spawning fish)
Medium-Large
% Lipid
(wet weight)
0.5-2
0.72 (0.46)
0.89 (0.19)*
0.95(0.26)*
0.97 (0.18)*
1
1-2
1-3
1-5
1.3 (0.29)*
1.3 (0.24)*
1.6 (0.47)*
1.7*
1.8*
1.90
2
2.1*
2. 19 (0.51)
2.3
2.54 (1.85)
2.70
2.82 (0.36)
3
3.09 (1.08)
Source
Price and Birge 2006
Liber et al. 1999
Miranda and Hubbard
1994
Miranda and Hubbard
1994
Miranda and Hubbard
1994
Neuman and Murphy
1992
Price and Birge 2006
Price and Birge 2006
Price and Birge 2006
Miranda and Hubbard
1994
Miranda and Hubbard
1994
Miranda and Hubbard
1994
Fischer etal. 1998
Fischer etal. 1998
Kay et al. 2005
Neuman and Murphy
1992
Fischer etal. 1998
Sethajintanin et al.
2004
Sidwell 1981
Sethajintanin et al.
2004
Kay et al. 2005
Sethajintanin et al.
2004
Neuman and Murphy
1992
Sethajintanin et al.
2004
                                           89 of 123

-------
Black crappie
(Pomoxis nigromaculatus)
Black crappie
(Pomoxis nigromaculatus)
Smallmouth bass
(Micropterus dolomieu)
Smallmouth bass
(Micropterus dolomieu)
Smallmouth bass
(Micropterus dolomieu)
Smallmouth bass
(Micropterus dolomieu)
White crappie
(Pomoxis annularis)
White crappie
(Pomoxis annularis)
Smallmouth bass
(Micropterus dolomieu)
Smallmouth bass
(Micropterus dolomieu)
Smallmouth bass
(Micropterus dolomieu)
Smallmouth bass
(Micropterus dolomieu)
White crappie (Pomoxis
annularis)
Bluegill (L. macrochirus)
0.0798 (0.012)
0.102 (0.007)
Not reported
0.154 (0.0647)
0.326 (0.177)
0.870 (0.0685)
Not Reported
Not Reported
0.277 (0.0901)
Not reported
Not reported
Not reported
Not Reported
0.00972 (0.00276)
Medium
Medium
Medium-Large
(defined based on
length data)
Medium
Medium-Large
Medium-Large
Assume medium
(spawning fish)
Assume medium
(spawning fish)
Medium-Large
Medium-Large
(defined based on
length data)
Medium-Large
(defined based on
length data)
Medium-Large
(defined based on
length data)
Assume medium
(spawning fish)
Small
3.11 (1.63)
3.15 (0.40)
3.3 (0.3)
3.33 (1.8)
4.17 (1.34)
4.93 (0.33)
5
5
5.03 (0.358)
5.5 (0.4)
5.6 (0.2)
5.8 (0.4)
6
7.9 (0.14)
Sethajintanin et al.
2004
Sethajintanin et al.
2004
Kwon et al. 2006
Sethajintanin et al.
2004
Sethajintanin et al.
2004
Sethajintanin et al.
2004
Neuman and Murphy
1992
Neuman and Murphy
1992
Sethajintanin et al.
2004
Kwon et al. 2006
Kwon et al. 2006
Kwon et al. 2006
Neuman and Murphy
1992
Carretal. 1997
Average
75th percentile
90th percentile
2.9
4.0
5.5
*Calculated from reported dry weight assuming that fish = 73% water (Table CIS).
The wet weight of an organism is the sum of the water, lipid, and NLOM content.  By default, if
the water content offish is 73% of the wet weight, and the % lipid is known (default = 4%), the
NLOM content of fish is the % remaining after subtracting the water and lipid content from
100%. Therefore, the default parameter for NLOM composition is 23%.
                                       90 of 123

-------
Appendix D. Selection of Mammal Species of Concern and Corresponding Biological
Parameters

Mammal species of concern were defined for use as default species in KABAM. Mammals were
considered to be of concern for pesticide exposures through aquatic bioaccumulation if their
diets incorporated  freshwater aquatic animals. Specific species  were identified using a Field
Guide  to Mammals of North America (Reid 2006).  This guide contains information on the
ranges,  taxonomy,  habits,  feeding preferences, and habitats  of mammals located in the
continental United States, Canada, and Alaska.

A review of this source identified six species of mammals that consume aquatic animals.  These
include the American water shrew  (Sorex palustris), the fog shrew (Sorex sonomae\  the star-
nosed mole (Condylura cristata), the marsh rice rat (Oryzomys palustris), the American  mink
(Mustela vison), and the Northern river otter (Lontra canadensis).  Additional references were
sought to obtain data on the body weights and feeding preferences of these mammals.  These
species are used in KABAM to represent mammals of concern for risks of pesticide exposures
through aquatic bioaccumulation. Descriptions of these species are provided below. Information
from these species descriptions were used to define the default parameters used to represent
mammals in the KABAM tool.
D.I. Descriptions of Mammal Species

       D. 1.1. American Water Shrew (Sorexpalustris)

The distribution of the American water shrew includes  Canada, Alaska, and areas of the
continental United States,  including  the  West  Coast,  Rocky  Mountains, Great  Lakes,
Appalachian,  and New England areas.  These shrews inhabit  areas boarding fast and  slow
moving streams, marshes, creeks, and ponds. This species is primarily insectivorous, consuming
aquatic invertebrates, such as stonefly nymphs, mayflies, caddisflies, and diptera. The American
water shrew is also known to consume other animals, including fish, salamanders, leaches, and
dead mice. Documented body weights range 0.008-0.018  kg, with males weighing more than
females (Beneski and Stinson 1987).

       D. 1.2. Fog Shrew (Sorex sonomae)

Fog shrews inhabit parts of Oregon and California on the Pacific Coast in the "fog belt."  This
species is found in marshes, near streams, and in forests. Their diet includes insects, earthworms,
centipedes, slugs, snails,  and amphibians.  Their weight ranges 0.0055-0.015  kg (Reid 2006,
Smithsonian 2008).

       D.I.3. Marsh Rice Rat (Oryzomyspalustris)

Marsh rice rats are distributed in states along the Gulf of Mexico and East Coast of the United
States. This species inhabits wetlands, marshes, swamps, meadows, and areas along streams. Its
diet includes insects, fiddler crabs, snails, fish, clams, arthropods, wetland plants, seeds, fungus,
                                       91 of 123

-------
baby turtles, bird eggs, and carrion (of mammals and birds). Their weight ranges 0.045-0.080 kg
(Wolfe 1982).

Six subspecies  of O. palustris have been  recognized (Wolfe  1982). One of these subspecies,
Oryzomys palustris natator has been federally listed as endangered since 1991. This subspecies
is known to occur in Florida and has a designated critical habitat (USFWS 1993).

       D. 1.4. Star-nosed Mole (Condylura cristata)

The star-nosed mole is distributed throughout the Eastern and Great Lakes regions of the United
States and Canada.   It inhabits marshy areas and streams. The diet  of this species includes
aquatic  annelids,  aquatic insects,  small fish, mollusks, crustaceans, grubs, and earthworms.
Reported body weights range 0.034-0.085 kg. The weights of these animals do not differ by sex
but by location within their geographic distribution range, with smaller animals being observed
in  the southern parts of the range (e.g.,  Tennessee) (Petersen and Yates  1980, Reid 2006,
Smithsonian 2008).

       D.I.5. American Mink (Mustela vison)

The American mink is distributed throughout the  United States and Canada, except  in the dry
areas  of Arizona, Nevada, California, Utah, and Texas. Mink inhabit wetlands and marshes.
Their diet is composed mostly of fish, amphibians (frogs), crustaceans (crayfish and crabs),
muskrats, and other small mammals. They will also consume squirrels, birds, bird eggs, reptiles,
aquatic insects, earthworms,  and snails if given the opportunity.  Individual  body weights  vary
based on range and sex, with females weighing less than males.  Documented body weights of
this species range 0.45-1.8 kg (Lariviere 1999, USEPA 1993).

       D. 1.6. Northern River Otter (Lontra canadensis)

The historical  distribution of the Northern river  otter includes most of the  United States and
Canada. The current distribution of this species in the United States includes states bordering the
Gulf of Mexico and Great Lakes, the East  Coast, New England, the West Coast and Alaska, as
well  as Canada (Lariviere and Walton 1998, Reid 2006). Northern river otters inhabit lakes,
swamps, marshes, streams, and ponds. The  diet of this species is primarily fish, but also includes
frogs, crayfish, small  mollusks,  reptiles, birds, and fruits. Body  weights  range 5-15 kg,  with
males weighing more than females (USEPA 1993, Lariviere and Walton 1998).
D.2. Determination of Mammalian Default Parameters for KABAM

Tables 7 and 8 of the KABAM tool allow the user to identify six mammal species of concern,
their body weights and their diets. For the purpose of KABAM, mammalian species of concern
include those that consume aquatic animals. Based on the information above, relevant species in
the United States include the American water shrew, the fog shrew, the star-nosed mole, the
marsh rice rat, the American mink, and the Northern river otter. A detailed version (with specific
mammals identified) of the conceptual model of the aquatic ecosystem depicted in Figure I of the
                                       92 of 123

-------
User's Guide is provided in Figure Dl. Default values representing the body weights and diets of
these mammals are described below.
                                                            Rice rat,
                                                            star-nosed
                                                            mole
                                                                       Water column
                                                                     Phytoplankton:
                                                                     Algae,
                                                                     cyanobacteria,
                                                                     diatoms,
                                                                     dinoflaeellates
                                         Large fish:
                                         Largemouth bass
Medium fish:
Sunfish, bluegills,
largemouth bass,
frogs
                                            Zoo plankton:
                                            Cladocera,
                                            copepods,
                                            rotifers
                   Benthic invertebrates:
                   Crayfish, chironomid
                   larvae, mayfly larvae,
                   snails
                                                      Filter feeders:
                                                      Clams, mussels
  Figure Dl. Detailed conceptual model depicting aquatic food web, with mammals included. Arrows depict
 direction of trophic transfer of bioaccumulated pesticides from lower levels to higher levels of the food web.
       D.2.1. Identification of Default Body Weights for Mammalian Species

Body weight and diet are the parameters that distinguish one mammalian species from another
within KABAM. Two pairs of species have similar body weights and diets, such that they can be
grouped together. These pairs are 1) the American water shrew and the fog shrew; and 2) the
star-nosed mole and the marsh rice rat. The American mink and  the northern  river otter are
                                          93 of 123

-------
sufficiently different in body weights to distinguish them as separate  default species in the
model.

The  selected body weight  value influences  the  estimates  of pesticide exposure  through
consumption of contaminated food items, as well as dose-adjusted toxicity values. Therefore, the
magnitude of the body weight parameter has an effect on the magnitude of the RQ.  Since higher
body  weight  values result in  higher dose-based RQs, the higher body weight  values were
selected to represent the four  groups of mammals used in KABAM. In order to bound risk
estimates for  the  two heaviest species of mammals {i.e.., American mink and Northern river
otter), default parameters are set to the minimum and maximum body weights. The following
values are suggested for inclusion in Table 7 of the KABAM tool to represent mammals 1-6:
Mammal #
Mammal 1
Mammal 2
Mammal 3
Mammal 4
Mammal 5
Mammal 6
Name
Fog/Water Shrew
Rice Rat/Star-nosed mole
Small Mink
Large Mink
Small River Otter
Large River Otter
Body
weight (kg)
0.018
0.085
0.450
1.800
5.000
15.000
       D.2.2. Determination of Daily Food Intake

If the weight of a food item (i.e., aquatic trophic level) is less than that of the amount of food
consumed by the mammal in one day, then the food item is a reasonable assignment. In order to
determine whether or not a particular trophic level is relevant to a mammal, the daily food intake
is estimated.

The dry food intake per day (Fdry, kg/day) for a mammal can be calculated according to Equation
Dl  (USEPA 1993).  This value can be converted to represent food intake  per day  on a wet
weight basis (Fwet, kg/day) by assuming that the diet of an organism is 75% water (Equation D2,
see Appendix C for % water of aquatic organisms).

                          Equation Dl.  Fdry = 0.0687 *BW0*22
                        Equation D2.  Fwet =
                                                     dry
                                            1 - (% water of diet)
The resulting wet food intakes per day for the mammalian species of concern for KABAM are
provided in Table Dl. This table presents food intake per day for each species based on the low
and high ranges of the body weights. These wet food intakes can be used to assign appropriate
aquatic animals to the default diets of these mammals.
                                       94 of 123

-------
Table Dl. Low- and high-end body weights and estimated food intake per day of mammals which consume
aquatic animals.
Species
Shrew (Water and Fog)
Shrew (Water and Fog)
Rice Rat, Star-nosed mole
Rice Rat, Star-nosed mole
Mink
Mink
River Otter
River Otter
Body Weight
(kg)
0.006
0.018
0.034
0.085
0.450
1.800
5.000
15.000
Dry Food Intake
per day (kg)
0.001
0.003
0.004
0.009
0.036
0.111
0.258
0.636
Wet Food
Intake per day
(kg)
0.004
0.010
0.017
0.036
0.143
0.446
1.032
2.545
Percent Body
Weight
Consumed Daily
67%
56%
50%
42%
31%
25%
20%
17%
       D.2.3. Definition of Default Diets of Mammals for use in KABAM

       Water/fog shrew

The diets of the American water and fog shrews (see section D.I) include species that would be
classified as benthic invertebrates (e.g., stonefly nymphs, mayflies, and snails) and fish (e.g., fish
and amphibians) according to the trophic levels of KABAM. However, since these species  are
primarily insectivorous, the default diet is assigned as 100% benthic invertebrates.

Based on the daily food intake for these two species (Table E.I), it is reasonable to assume that
these shrews can consume organisms in the small fish category. If interested in the potential
acute  risk to water/fog  shrews from pesticides through consumption of fish/amphibians,  the
model user could define the  diet of these mammals as 100% small fish. Since  this represents  a
higher trophic level in the aquatic ecosystem than the benthic invertebrates, this assumption may
result in a higher RQ.

       Rice rat/star-nosed mole

The diets of the rice rat and the star-nosed mole (see section D.I) include species that would be
classified as benthic invertebrates (e.g., arthropods, snails),  filter feeders (e.g.,  clams) and fish
according to the trophic levels of KABAM.  Based on the daily food intake for these two species
(Table E.I), it is reasonable to assume that individuals of these species could consume organisms
in the small fish category. Since no data are available to define feeding preferences of these
two species, for the purpose of KABAM, the default diet composition of these mammals is
equally distributed among  these three  trophic levels (Le., 34% benthic invertebrates, 33%
filter feeders, and 33% small fish).

       American mink

The diet of the American mink (see section D.I) includes species that would  be classified as
benthic invertebrates (e.g., crayfish) and small/medium-sized fish according to the trophic levels
of KABAM. Based on the  daily food intake  for this species (Table E.I), it  is reasonable to
                                        95 of 123

-------
assume that these mammals could consume organisms in the small and medium fish category.
The default diet for this mammal is 100% medium fish.

      Northern river otter

The diet of the Northern river otter (see section D.I) is primarily fish, but also includes species
that would be classified as benthic invertebrates (e.g., crayfish) according to the trophic levels of
KABAM.  Based on the daily food intake for this species (Table D.I), it is reasonable to assume
that these mammals may consume organisms in the small, medium, and large fish categories.
According to USEPA  1993, river otters have  been documented as including various fish that
would  be  classified in different trophic  levels of KABAM, including sunfish  and  bass.
Therefore, the  default diet for this mammal is 100%  medium fish for the small otter and
100% large fish for the large otter.
                                       96 of 123

-------
Appendix E. Selection of Bird Species of Concern and Corresponding Biological
Parameters

Bird species of concern were identified in order to define default parameters (for body weight
and diet composition) to represent birds in  KABAM.   Bird  species were considered to  be of
concern  for pesticide exposures  through aquatic  bioaccumulation if their  diets incorporated
freshwater aquatic animals. Specific  species were identified using the Smithsonian handbooks'
Birds of North  America  (Eastern and  Western Regions) (Alsop  200la and 200Ib).  These
handbooks  contain information on  the  ranges,  taxonomy,  habits, feeding preferences,  and
habitats of birds located in the continental United States, Canada, and Alaska.

A review of this  source identified over 40 bird species of concern that fall into 11 families. These
families  include: Accipitridae  (eagles,  hawks  and kites),  Alcedinidae (kingfisher), Anatidae
(ducks),  Ardeidae  (herons, egrets  and bitterns), Gruidae  (cranes),  Pelecanidae  (pelicans),
Phalacrocoracidae  (cormorants),  Podicipedidae  (grebes),   Rallidae  (rails),   Scolopacidae
(sandpiper) and Threskiornithidae (ibis). Descriptions of these families are provided below.

It should be noted that this review was not intended to be inclusive  of every relevant species or
family of birds inhabiting North America. Rather, the intention  of this review was to identify
birds  that may consume aquatic  animals containing pesticides  that bioaccumulate in aquatic
ecosystems. Information from identified bird species and families was used to define the default
parameters representing birds in  the KABAM tool.  These  default parameters are  described
below.
E.I. Bird Family Descriptions

       E. 1.1. Accipitridae (Eagles, Hawks and Kites)

Most species of this family prey upon terrestrial rodents; however, several rely  upon aquatic
animals for their diet (Table El).  These species include the osprey (Pandion haliaetus) and the
bald eagle (Haliaeetus leucocephalus) (Alsop 200la and 200Ib).  Ospreys fly over freshwater
and saltwater areas and catch fish from the surface of the water using their feet. Body weights of
osprey range from  1.25 to 2.0 kg (USEPA 1993).   Bald eagles eat fish, rodents, birds, and
carrion. Body weights of adult bald eagles range  3.0 - 5.8 kg (USEPA 1993). An additional
member of this family, the snail kite (Rostrhamus sociabilis), has a subspecies that is  federally
listed as endangered (USFWS 2008).  This species is known  to  occur in wetlands of Florida,
where the bird eats snails (Alsop 200la).  The average body weight of this bird is 0.38 kg
(Dunning 1984).

Table El. Body weights and diets of species of Accipitridae that prey upon aquatic animals.
Species (scientific name)
Snail kite (Rostrhamus sociabilis)
Osprey (Pandion haliaetus)
Bald eagle (Haliaeetus leucocephalus)
Body
weight (kg)
0.381
1.25-2.003
3.00-5.803
Diet
snails2
fish3
fish, rodents, birds, and carrion3
 Running 1984; 2Alsop2001aand2001b;  3USEPA 1993
                                        97 of 123

-------
       E.I.2. Alcedinidae (kingfisher)

One species of this family, the belted kingfisher (Ceryle alcyon) is widely distributed throughout
North America,  inhabiting freshwater areas such as lakes, rivers, and ponds, as well as marine
coastal  areas.   This species  feeds primarily upon fish, but its  diet  also includes amphibians,
insects, and crayfish. Body weights of this species range 0.13-0.22 kg (USEPA 1993; Table E2).

Table E2. Body weights and diets of species of Alcedinidae that prey upon aquatic animals.
Species (scientific name)
Belted kingfisher (Ceryle alcyon)
Body
weight (kg)
0.13-0.221
Diet
primarily fish,
crayfish1
but also amphibians, insects and
 'USEPA 1993
       E.I.3. Anatidae (Ducks)

There are many species of ducks that are widely distributed in North America (Table E3). Ducks
predominantly inhabit freshwater areas such as lakes, rivers, wetlands, and ponds.  Their diets
include a wide  variety  of aquatic organisms,  such as aquatic insects, insect  larvae, snails,
amphibians, fish, crayfish,  mollusks, plankton,  and  aquatic plants  (Alsop 200la and 200Ib).
Body weights of ducks vary based on the species, with a range of 0.3-2.0 kg for ducks inhabiting
freshwater areas (Dunning 1984).

Table E3. Body weights and diets of species of Anatidae that prey upon aquatic animals.
Species (scientific name)
Cinnamon teal (Anas cyanoptera)
Bufflehead (Bucephala alboela)
Wood duck (Aix sponsa)
Hooded merganser (Lophodytes
cuculatus)
Lesser scaup (Aythya affmis)
Common goldeneye
(Bucephala clangula)
Mallard (Anas platyrhynchos)
Red-breasted merganser (Mergus
serrator)
Common merganser (Mergus
merganser)
Body
weight (kg)1
0.36-0.41
0.30-0.55
0.64-0.91
0.54-0.91
0.54-1.05
0.80-1.40
0.72-1.58
0.91-1.31
1.05-2.05
Diet2
seeds, aquatic insects, rice, algae, snails,
crustaceans
aquatic insects and insect larvae, snails, small
fish, seeds
plants, animals, snails, tadpoles, salamanders
fish, crustaceans, aquatic insects, aquatic
animals
plants and animals
mollusks, crustaceans, insects, aquatic plants
plants, insects, mollusks, crustaceans
fish
small fish, mollusks, crustaceans, aquatic
insects, some plants
 Running 1984;  2Alsop 2001a and 2001b

       E.I.4. Ardeidae (Herons, Egrets and Bitterns)

This family includes species of herons, bitterns and egrets, several of which inhabit waters of
North America (Table E4). Their habitats include freshwater areas such  as lakes, rivers, ponds,
wetlands, and streams, as well as marine coastal areas.  These birds wade through water to spear
                                         98 of 123

-------
their food with their beaks. Their diets include fish, crustaceans, amphibians, snakes, crayfish,
and insects (Alsop 2001a and 2001b, USEPA 1993).  Individuals of this family range in weight
from 0.08 to 2.9 kg, depending upon the species (Dunning 1984, USEPA 1993).

Table E4. Body weights and diets of species of Ardeidae that prey upon aquatic animals.
Species (scientific name)
Least bittern (Ixobrychus exilis)
Green heron (Butorides virescens)
Snowy egret (Egretta thula)
Little blue heron (Egretta caerulea)
American bittern (Botaurus
lentiginosus)
Yellow-crowned night heron
(Nyctanassa violacea)
Black crowned night heron (Nycticorax
nycticorax)
Great egret (Ardea alba)
Great blue heron (Ardea herodias)
Body
weight (kg)1
0.08-0.09
0.212
0.35-0.40
0.32-0.45
0.52-1.07
0.72-0.85
0.73-1.01
0.80-1.07
1.87-2.88
Diet2
fish, insects
fish, aquatic invertebrates
crustaceans, insects, fish
small vertebrates, crustaceans, large insects
frogs, small eels, small fish, snakes,
salamanders, crayfish, small rodents, water
bugs
crustaceans, fish, shellfish
fish, mollusks, small rodents, frogs, snakes,
crustaceans, plants, eggs, birds
fish, frogs, snakes, crayfish, large insects
fish, other aquatic animals
 'Dunning 1984; 2Alsop 2001a and 2001b

       E.I.5. Gruidae (Cranes)

Cranes inhabit freshwater wetlands and marshes.  These birds eat fish, frogs, small mammals,
mollusks, crustaceans, and  plants (Alsop 2001a and 2001b).  Two species of cranes, i.e., the
whooping crane (Grus americand) and the Mississippi sandhill crane (Grus canadensis pulld),
are federally listed as endangered (USFWS 2008).  Body  weights  of the whooping crane and
sandhill crane range 2.5-6.7 kg (Dunning 1984) (Table E5).

Table E5. Body weights and diets of species of Gruidae that prey upon aquatic animals.
Species (scientific name)
Whooping crane (Grus americand)
Sandhill crane (Grus canadensis)
Body
weight (kg)1
5.44-6.36
2.45-6.70
Diet2
fish, frogs, small mammals, mollusks,
crustaceans, and plants
plants and animals
 'Dunning 1984;  z Alsop 2001a and 200 Ib

       E.I.6. Pelecanidae (Pelicans)

There is  one species of pelican that inhabits freshwater aquatic habitats of North America: the
American white pelican (Pelecanus erythrorhynchos).  Their habitats  include freshwater areas
such as lakes, rivers, ponds,  wetlands and streams, as well as marine coastal areas.  The diet of
these birds includes fish (Alsop 2001a and 2001b).  The average weight of the American white
pelican is 7.5 kg (Dunning 1984) (Table E6).
                                        99 of 123

-------
Table E6. Body weights and diets of species of Pelecanidae that prey upon aquatic animals.
Species (scientific name)
American white pelican (Pelecanus
erythrorhynchos)
Body
weight (kg)
7.51
Diet
fish2
 'Dunning 1984; 2 Alsop 2001a and 200 Ib

       E.I.7. Phalacrocoracidae (Cormorants)

Of the species  of cormorants  inhabiting  North  America,  the  double-breasted cormorant
(Phalacrocorax auritus) is the most widespread, inhabiting freshwater areas such as lakes, rivers,
ponds, as well as marine coastal areas.   Cormorants dive for their prey, which includes fish,
crustaceans, and  amphibians  (Alsop  200la and  200Ib).   The average weight of the double-
breasted cormorant is 1.8 kg (Dunning 1984).

Table E7. Body weights and diets of species of Phalacrocoracidae that prey upon aquatic animals.
Species (scientific name)
Double-breasted cormorant
(Phalacrocorax auritus)
Body
weight (kg)
1.60-2.041
Diet
fish, crustaceans and amphibians2
 'Dunning 1984; 2Alsop 2001a and 200 Ib

       E.I.8. Podicipedidae (Grebes)

Several species of grebes  reside in the continental United States (Table E8). Their  habitats
include freshwater areas such as lakes,  rivers, ponds, wetlands, and streams, as well as marine
areas. These birds forage for aquatic insects, crustaceans, and fish by diving underwater (Alsop
2001a and 2001b). They range in weight 0.2-1.8 kg (Alsop 2001a and 2001b, Dunning 1984).

Table E8. Body weights and diets of species of Podicipedidae that prey upon aquatic animals.
Species (scientific name)
Eared grebe (Podiceps nigricollis)
Pied-billed grebe (Podilymbus
podiceps)
Horned grebe (Podiceps auritus)
Western grebe (Aechmorphorus
occidentalis)
Clark's grebe (Aechmophoms clarkia)
Body
weight (kg)1
0.22-0.37
0.34-0.55
0.33-0.53
0.80-1.82
1.502
Diet2
aquatic insects
aquatic insects, small fish, crustaceans
fish, crustaceans, aquatic insects
fish
fish
 Running 1984;   2Alsop 2001a and 200Ib

       E.I.9. Rallidae (Rails)

Rail species inhabit freshwater areas such as lakes, rivers, ponds, wetlands and streams as well as
saltwater marshes of North America.  These  species feed upon crustaceans, aquatic insects,
snails, fish, and plants (Alsop 200 la and 200 Ib).  Individuals of this family range in weight from
0.07 to 0.49 kg (Dunning 1984) (Table E9). One species from this family, the clapper rail (Rallus
longirostris) is federally listed as  an endangered species and is known to  occur in Arizona,
California, Nevada, and Utah (USFWS 2008).
                                        100 of 123

-------
Table E9. Body weights and diets of species of Rallidae that prey upon aquatic animals.
Species (scientific name)
Sora (Porzana Carolina)
Virginia rail (Rallus limicola)
King rail (Rallus elegans)
Clapper rail (Rallus longirostris)
Body
weight (kg)1
0.08
0.07-0.12
0.25-0.49
0.25-0.35
Diet2
plants, insects, spiders, small crustaceans,
snails
insects (primarily), worms, crustaceans, small
fish
plants, aquatic invertebrates, aquatic
vertebrates
crabs, crustaceans, worms, amphibians,
reptiles, mollusks, small fish, aquatic insects
 'Dunning 1984;  2Alsop 2001a and 200 Ib

       E.I.10. Scolopacidae (Sandpipers)

Many species of sandpipers inhabit freshwater aquatic habitats of North America (Table E10).
These habitats include lakes,  rivers, ponds, wetlands, and streams.  Their diets include aquatic
invertebrates,  insects,  crustaceans, small  fish,  amphibians, and mollusks  (Alsop  200la and
2001b). Body weights of sandpipers range 0.02- 0.70 kg (Dunning 1984).

Table E10. Body weights and diets of species of Scolopacidae that prey upon aquatic animals.
Species (scientific name)
Least sandpiper (Calidris minutilla)
Spotted sandpiper (Actitis macularia)
Wilson's phalarope (Phalaropus
tricolor)
Greater yellow legs (Tringa
melanoleca)
Willet (Catoptrophorus semipalmatus)
Long-billed curlew (Numenius
americanus)
Body
weight (kg)1
0.022
0.03-0.06
0.072
0.12-0.22
0.22
0.57-0.70
Diet2
insects and larvae, crustaceans
invertebrates, small fish
larvae, crustaceans, seeds
small fish, insects and larvae, crabs, snails
aquatic insects, mollusks, small fish
aquatic insects, larvae, mollusks, crustaceans,
small amphibians
 'Dunning 1984;   2Alsop 2001a and 200Ib

       E. 1.11. Threskiornithidae (Ibis)

Ibis  inhabit freshwater areas  such as lakes, rivers, ponds, wetlands, and streams, as well as
marine coastal areas of North America. These species are wading birds that feed upon crayfish,
aquatic invertebrates, fish, and frogs (Alsop 200la and 200Ib).  Individuals of this family range
in weight from 0.4 to 1.3 kg (Dunning 1984) (Table El 1).

Table Ell. Body weights and diets of species of Threskiornithidae that prey upon aquatic animals.
Species (scientific name)
White-faced ibis (Plegadis chihi)
White ibis (Eudocimus albus)
Body
weight (kg)1
0.43-0.81
0.59-1.28
Diet2
crayfish, aquatic invertebrates, fish, frogs
not stated
 'Dunning 1984; 2Alsop 2001a and 200Ib
                                        101 of 123

-------
E.2. Detailed Conceptual Model

A detailed version of the conceptual model of the aquatic ecosystem depicted in Figure I of the
User's Guide, with specific birds identified, is provided in Figure El.
             Belted
             kingfisher,
             cormorant,
             grebes, ibis,
             herons, rails
                                                                                Water column
                                  Large fish:
                                  Largemouth bass
    Medium fish:
    Sunfish, bluegills,
    largemouth bass,
    frogs
Small fish:
Young of the
year, tadpoles
                                                                            Phytoplankton
                                                                            Algae,
                                                                            cyanobacteria,
                                                                            diatoms,
                                                                            dinoflagellates
                  Zooplankton
                  Cladocera,
                  copepods,
                  rotifers
                         Benthic invertebrates:
                         Crayfish, chironomid
                         larvae, mayfly larvae
                         snails
                            Filter feeders:
                            Clams, mussels
  Figure El. Detailed conceptual model depicting aquatic food web of KABAM. Arrows depict direction of
      trophic transfer of bioaccumulated pesticides from lower levels to higher levels of the food web.
                                            102 of 123

-------
E.3. Determination of Daily Food Intake
If the weight of a food item (i.e., aquatic trophic level) is less than that of the amount of food
consumed by the bird in one day, then the food item is a reasonable assignment. In order to
determine whether or not a particular trophic level is relevant to a bird, the daily food intake is
estimated.

The dry food intake per day (Fdry, kg/day) for a bird can be calculated according to Equation El
(USEPA 1993).  This value can be converted to represent food intake per day on a wet weight
basis  (Fwet, kg/day) by assuming that the diet of an organism is 75% water (Equation E2,  see
Appendix C for % water of aquatic organisms).
                          Equation  El.  Fd  = 0.0582 * BW
                                                            0.651
                        Equation E2.  Fwet =
                                                     F.
                                                      dry
                                             1 - (% water of diet)
Of the bird families described above, body weights range 0.02-7.5 kg. The resulting wet food
intakes per day for birds of concern for KABAM are provided in Table E12. This table presents
food intake values per day for each species based on body weight. These wet food intakes can be
used to assign appropriate aquatic animals to the default diets of these birds.

Table E12. Body weights representative of birds that consume aquatic animals and corresponding daily dry
and wet food intakes.
Family or species
Sandpipers
ducks
cranes
belted kingfisher
rails
ibis
grebes
Double-breasted cormorant
Bitterns, egrets, herons
osprey
Bald eagle
white pelican
Body weight
range (kg)
0.02- 0.70
0.30-2.00
2.45-6.70
0.13-0.22
0.07-0.49
0.43-1.28
0.22-1.82
1.8
0.08-2.90
1.25-2.00
3.00-5.80
7.5
Dry Food Intake
per day (kg)
0.005-0.046
0.027-0.091
0.104-0.201
0.015-0.022
0.010-0.037
0.034-0.068
0.022-0.086
0.085
0.011-0.116
0.067-0.091
0.119-0.183
0.216
Wet Food Intake per
day (kg)
0.018-0.185
0.106-0.366
0.417-0.803
0.062-0.087
0.041-0.146
0.134-0.273
0.087-0.344
0.341
0.045-0.466
0.269-0.366
0.476-0.731
0.864
                                        103 of 123

-------
E.4. Definition of Default Parameters to Represent Birds in KABAM
Based on the species descriptions above, birds can be divided into three groups based on their
diets.  The three diets  include:  1)  filter  feeders, benthic invertebrates  and fish,  2) benthic
invertebrates and fish and 3) fish. (Table E13). These three diets were used to define the default
parameters representing birds in KABAM (Table E14), which are described below.

Table E13. Summary of diets and body weights of families of birds defined as consuming aquatic animals.
Diet
Filter feeders, benthic invertebrates,
fish
Benthic invertebrates and fish
Fish
Family or species
Sandpipers
Ducks
Cranes
Belted kingfisher
Rails
Ibis
Grebes
Double-breasted cormorant
Bitterns, egrets, herons
Osprey
Bald eagle
White pelican
Body weight range (kg)
0.02- 0.70
0.30-2.00
2.45-6.70
0.13-0.22
0.07-0.49
0.43-1.28
0.22-1.82
1.80
0.08-2.90
1.25-2.00
3.00-5.80
7.50
Table E14. Default body weights and diet parameters for use in KABAM to represent birds.
Bird
#
1
2
3
4
5
6
Bird Name
Sandpipers
Cranes
Rails
Herons
Small Osprey
White pelican
Relevant Families/species
Sandpipers, ducks, cranes
Sandpipers, ducks, cranes
Belted kingfisher, rails, ibis,
grebes, double-breasted
cormorant, bitterns, egrets, herons
Belted kingfisher, rails, ibis,
grebes, double-breasted
cormorant, bitterns, egrets, herons
Osprey, bald eagle, white pelican
Osprey, bald eagle, white pelican
Default
weight
(kg)
0.02
6.70
0.07
2.90
1.25
7.50
Default diet
3 3% benthic invertebrates
33% filter feeders
34% small fish
3 3% benthic invertebrates
33% filter feeders
34% medium fish
50% benthic invertebrates
50% small fish
50% benthic invertebrates
50% medium fish
100% medium fish
100% large fish
                                         104 of 123

-------
       E.4.1. Birds Consuming Benthic Invertebrates, Filter Feeders, and Fish

Because  sandpipers, ducks, and  cranes  share similar diets (i.e., benthic invertebrates,  filter
feeders, and fish), they are considered as a group for defining input parameters for KABAM.
Two of the default birds in KABAM (# 1 and 2) represent birds with a similar diet.

Comparison of the daily wet food consumption for sandpipers (Table E12) to the weight of small
and medium fish in KABAM (0.01 and 0.1 kg, respectively)  indicates that not all  of these
species would be expected  to  consume medium-sized fish. Therefore, it  is  assumed  that
sandpipers consume small fish. All  species of cranes are  expected to be able to consume a
medium-sized (0.1  kg) fish in one  day. Therefore, it is assumed  that the diet of cranes  is
composed of medium-sized fish.   Since  the relative proportion of benthic invertebrates,  filter
feeders and fish within the diets of these species is unknown, it is assumed that these prey items
compose an equal share of the diet of these birds.

The 1st default bird  in KABAM has a diet of 33% benthic invertebrates,  33%  filter feeders and
34% small fish.   This bird is intended to represent the low end of birds that  consume benthic
invertebrates, filter feeders, and small fish.  Therefore, the default body weight of 0.02 kg was
selected because it is consistent with the  lowest body weight of birds that have this diet (Table
E13).

The 2nd default bird in KABAM has a diet of 33% benthic invertebrates, 33% filter feeders, and
34% medium fish. This bird is intended to represent the high end of birds that consume benthic
invertebrates, filter feeders, and medium-sized fish.  Therefore, the default body weight of 6.7 kg
was selected (Table E14).

It should be noted that pesticide EECs and subsequent RQs for sandpipers, ducks, and cranes are
bound by KABAM's default birds 1  and 2. RQs for these two default birds are  intended to
represent birds with similar size and feeding habits as sandpipers, ducks, and cranes. These EECs
and RQs can be refined by the model user to represent a specific bird species by entering specific
body weights of individual species of concern and the appropriate species composition of their
diet.

       E.4.2. Birds Consuming Benthic Invertebrates and Fish

Because belted kingfisher,  rails,  ibis,  grebes, double-breasted cormorants, bitterns, egrets, and
herons share similar diets (i.e., benthic invertebrates and fish), they are considered as a group for
defining  input parameters  for KABAM. Two of the default birds in KABAM (# 3 and 4)
represent birds with  a similar diet.

Comparison of the  daily wet food consumption for small  rails, small grebes,  and the belted
kingfisher (Table E12) to the  weight of small and medium-sized fish in KABAM (0.01 and 0.1
kg, respectively) indicates that not all of these species would be  expected to consume medium
fish. Therefore, it is assumed that some of these species consume small fish. Species of rails,
ibis, grebes, bitterns, egrets, herons and the double-breasted cormorant are expected to be able to
consume a 0.1 kg fish per day. Therefore,  it is assumed that the diet of these species is composed
                                       105 of 123

-------
of medium-sized fish. Since the relative proportion of benthic invertebrates and fish within the
diets of these species is unknown, it is assumed that these prey items compose an equal share of
the diet of these birds.

The 3rd default bird in KABAM has a diet of 50% benthic invertebrates and 50%  small fish.
This bird is intended to represent the low end of birds that consume benthic invertebrates and
small-sized fish.  Therefore,  the default body weight of 0.07 kg was selected because it is
consistent with the lowest body weight of birds that have this diet (Table El 3).

The 4th default bird in KABAM has a diet of 50% benthic invertebrates and  50% medium fish.
This bird is intended to  represent the high end of birds that consume benthic invertebrates and
medium-sized fish. Therefore, the default body weight of 2.9 kg was selected (Table E14).

It  should be  noted that  pesticide EECs  and subsequent RQs  for belted kingfisher,  rails, ibis,
grebes, double-breasted cormorants,  bitterns,  egrets,  and herons are  bounded  for KABAM's
default birds 3 and 4. RQs for these two default birds are intended to represent birds with similar
sizes and  feeding habits. These EECs and RQs can be  refined for  specific bird  species  by
entering specific body weights of individual species of concern and entering the appropriate diet.

       E.4.3. Birds Consuming Fish

Because  osprey,  bald  eagles, and white  pelicans  share  similar diets  (i.e., fish), they are
considered as a group for defining input parameters for KABAM. Two of the default birds in
KABAM (# 5 and 6) represent birds with a similar diet.

Comparison of the daily wet food consumption for the lower end body weight (1.25 kg) of these
birds to the weight of medium and large fish in KABAM (0.1 and 1.0 kg, respectively) indicates
that the lower weight individuals of these bird species are able  to consume medium fish, but
unlikely to consume large fish. Therefore, it is assumed that the diet of default bird  #5 (named
osprey), can  be represented by 100% medium-sized fish. Comparison of the daily wet food
consumption  (0.86 kg/day) for the higher end body weight (7.5 kg) of these birds to  the weight
of large fish in KABAM (1.0 kg) indicates that the higher weight individuals of these bird
species are likely to consume large  fish. Therefore, it is assumed that the diet of default bird #6
(named white pelican), can be represented by 100% large-sized fish.

In order to bound EECs and RQs for these three birds, the lowest and highest body weights were
selected to represent KABAM's default birds  5 and 6, respectively, in KABAM.  These EECs
and RQs can be refined for  specific bird  species by entering specific body weights  of individual
species of concern.
                                       106 of 123

-------
Appendix F. Description of Equations Used to Calculate the BCF, BAF, BMF, and BSAF
Values

Bioconcentration, bioaccumulation, and biomagnification factors are calculated in the "results"
worksheet of the KABAM tool using data from the "parameters & calculations" worksheet. The
equations for these calculations are described below.
F.I. Bioconcentration

Bioconcentration is a measure of the amount of pesticide residue in an organism's tissue relative
to the concentration in the organism's environment (USEPA 2008c).  This includes pesticide
uptake through respiration and contact, not through dietary sources.  Bioconcentration factors
(BCFs)  are  calculated  by  considering  pesticide  tissue  concentrations  with respect  to
environmental pesticide concentrations. BCF  values >1  indicate that the concentration in the
organism is greater than that of the medium (e.g., soil or water) from which the pesticide was
taken. BCFs can be calculated on a total organism basis or normalized to the lipid content of the
organism.

KABAM calculates the total (body weight) BCFs of a  chemical for each  aquatic organism
according to Equation  Fl (USEPA 2003). CBCF is calculated using equation Al (see Table A.I
of Appendix A for a full description) where CB = CBCF, when kD = kE = kM = kG = 0. The units of
total BCF values  are expressed as: (jig pesticide/kg wet weight)/(|ig pesticide/L  water).  Total
BCF values account for the total amount of the pesticide in the water (i.e., CWTO).
                            Equation Fl.  Total BCF = 


               Eq. Al  CB = ^*(fflo**C^
                                                        M
KABAM also calculates the lipid-normalized BCFs of a chemical for each aquatic organism
according to Equation F2  (USEPA  2003).  The units  of lipid normalized BCF  values  are
expressed as: (jig pesticide/kg lipid)/(|ig pesticide/L water). VLB represents the fraction of lipid
in the body of the organism for which the BCF is being derived. Lipid normalized BCF values
account for the pesticide concentration that is freely dissolved in the water (i.e., CWTO*).
                   Equation F2.  Lipid normalized BCF =
                                                                 <
                                       107 of 123

-------
F.2. Bioaccumulation

Bioaccumulation is the net uptake of a pesticide from the environment by all possible routes
(e.g.,  respiration,  diet,  dermal) from any source (e.g.,  water, sediment, and other organisms)
(Spacie et al.  1995). Bioaccumulation factors (BAF) are calculated by considering pesticide
tissue concentrations with respect to  environmental  pesticide concentrations. BAF  values >1
indicate that the accumulation in the organism is greater than that of the medium (e.g., soil or
water) from which the pesticide was taken. These factors can be calculated on a total organism
basis or normalized to the lipid content of the organism.

KABAM calculates the  total BAFs of a chemical  for each aquatic  organism according to
Equation F3 (USEPA  2003). The units of total BAF values are expressed as: (jig pesticide/kg
wet weight)/(|ig pesticide/L water). CB is calculated according to Equation Al. Total BAF
values account for the total amount of the pesticide in the water (i.e., CWTO).


                            Equation F3.   Total BAF = 
Lipid-normalized BAFs of a chemical are calculated for each aquatic organism according to
Equation F4 (USEPA 2003). The units of lipid normalized BAF values are expressed as:  (jig
pesticide/kg lipid)/(|ig pesticide/L water). The variable CB is calculated according to Equation
Al. The variable VLB represents the fraction of lipid in the body  of the organism for which the
BCF is being derived. Lipid normalized BAF values account for the pesticide concentration that
is freely dissolved in the water (i.e., CWTO*)-

                                                           (CB/
                                                           (  /v
                    Equation F4.   Lipid normalized BAF =      LB
Accumulation factors are also derived by considering pesticide tissue concentrations with respect
to pesticide  concentrations  in  sediment.  Biota-sediment  accumulation factors  (BSAFs) are
calculated by dividing the lipid normalized concentration of a chemical in an organism by the
chemical concentration in the sediment (dry weight), normalized to the organic carbon content of
the sediment (Equation  F5) (USEPA 2003).  The  variable Csoc represents  the  pesticide
concentration in the sediment, normalized to the organic carbon content of the sediment (units of
g/kgOC).


                                                   (CB/V
                            Equation  F5.  BSAF = ^ '  LB
                                       108 of 123

-------
F.3. Biomagnification

Biomagnification is  the  increase of a  pesticide concentration in the tissue  of an organism
compared to the tissue concentrations of its prey (USEPA 2008b). Biomagnification factors
(BMFs) are calculated by considering lipid normalized pesticide tissue concentrations within an
organism with respect to the lipid normalized concentrations of that pesticide in the prey of the
organism. Factors >1 indicate the occurrence of biomagnification.

KABAM calculates the BMFs of a chemical for each aquatic organism according to Equation
F6  (USEPA 2003).  The units of BMF values are  expressed  as: (jig pesticide/kg lipid)/(|ig
pesticide/kg lipid). The variable CB is calculated according to Equation Al. VLB represents the
fraction of lipid in the body of the organism for which the BMF is being derived. P; represents
the  fraction of diet containing prey item i.  CDi represents the concentration of the pesticide in
prey item i and VLB; represents the fraction of lipid in the body of the prey item i. It should be
noted that although KABAM allows aquatic organisms to consume sediment, uptake of pesticide
through consumption of sediment is not considered in the calculation of BMFs in the model tool.
                         Equation F6.  BMF =
                                        109 of 123

-------
Appendix G. Description of Equations Used to Calculate Dietary-Based and Dose-Based
EECs, Toxicity Values, and RQs for Mammals and Birds Consuming Contaminated
Aquatic Organisms

Exposures of birds and mammals to pesticides  accumulated in tissues of aquatic organisms are
calculated by the KABAM tool. Relevant toxicity data are also calculated by KABAM based on
input data from toxicity studies for birds and mammals. The equations used to estimate exposure
and to adjust toxicity values and to calculate RQs depicted in Tables 14-15 of the KABAM tool
are described below.
G.I. Food Ingestion Rates

Dry food ingestion rates (FIdry) are estimated for mammals and birds using allometric equations
that relate food intake with body weight (Equations Gl and G2, respectively). FI is calculated
in kg dry food/kg-bw day and BW is animal body weight in kg.
                   zr   ,-   r,   JIT     -0687 *BW
                  Equation Gi .  rl,  = -   (mammals)
                    1             dry                    \         1
                                       0 0582* BW0
                  Equation G2.  FI,  =  -  (birds)
                    1              dry                    \     1
Food intake (FI) values are converted from food dry weight/kg-bw day to food wet weight/day
using the wet weight of the assigned diet of each mammal and bird (Equation G3). The variable
Pi represents the fraction of diet of the mammal or bird containing prey item i  (an aquatic
organism). The variable VwBi represents the fraction of water in the body of the prey item i.

                                                   FI
                         Equation  G3. FIwet =        ^
1 - V ( P * V
1  /_j V i  y W
                                                         WBi
G.2. Drinking Water Intake Rates

Drinking water intake rates (DW) for mammals and birds are calculated based on Equations G4
and G5 (USEPA 1993); where BW represents the body weight (in kg) of the animal for which
the drinking water intake is being assessed. Resulting units of DW are L/day.

                    Equation G4.   DW = (o.099*BW090)  (mammals)

                      Equation G5.  DW = (o.059* BW061)  (birds)
                                      110 of 123

-------
G.3. Dose-based EECs

Dose-based EECs are estimated assuming that pesticide intake is a function of the amount of
pesticide contained in the food and drinking water of an animal. The dose-based EEC is derived
according to Equation G6.  In this equation, pesticide intake through food is calculated as the
sum of the products of the fraction of each prey item in the diet (P;) and the  pesticide tissue
residue concentration for each prey item (Ce;; ug/kg-ww).  The sum  of the pesticide residues
ingested  through food  is converted into units of mg pesticide/kg food.   This value is  then
multiplied by the intake rate for wet food (kg food/kg-bw day). The resulting value is in units of
mg pesticide/kg-bw day. Pesticide intake through drinking water is calculated by multiplying the
concentration of the pesticide in water (Cwxo, nig/L) by the water intake (DW in units of L/d)
and dividing by the bodyweight of the mammal or bird of concern. This results in units of mg
pesticide/kg-bw day. The sum of pesticide intake  through diet and through  drinking water is the
dose-based EEC.
                                                              f~^   & 7~~\ T/T/^
  Equation G6.   Dose - based    EEC  = Y  (P. * CB. ) * F/  , +  WTO
    -*                                    ^_J  \Z     57 /   Wei       T-J -TTT
                                                                  nW
G.4. Dietary-based EECs

Dietary-based EECs are estimated assuming that pesticide intake is a function of the amount of
pesticide contained in the food of an animal.  This  differs from  the dose-based EECs in that
pesticide exposure  through  drinking water is not considered. In addition, the dietary-based
exposure value is not adjusted for the relative amount of food consumed per day by animals of
different sizes.  The dietary-based EEC is derived according to Equation G7.  In this equation,
the pesticide intake through food is calculated as the  sum of the products of the fraction of each
prey item in the diet (P;) and the pesticide tissue residue concentration for each prey item (CB;;
ug/kg-ww).

    Equation Gl.  Dietary-based  EEC =y^(P, *  CBi )
G.5. Adjusted Dose-based Toxicity Values

Available dose-based toxicity  values are adjusted  for the weights of the  animal  tested (e.g.,
laboratory rat, mallard duck) and of the animal for which the risks are being assessed (e.g., mink,
bald eagle). These adjustments  are made for mammals and birds according to Equations G8 and
G9,  respectively (USEPA 2006).   In these equations, AT =  adjusted toxicity  value; LD50  or
NOAEL = endpoint reported by toxicity study; TW =  body weight of tested animal (350g rat;
1580g mallard, 178  g Northern bobwhite quail  or weight defined by the model  user  for  an
alternative species);  AW = body weight of assessed  animal;  x = Mineau  scaling  factor.
Chemical specific values for x may be located in Mineau et al.  1996). If no chemical specific
data are available, the default value of 1.15  should be  used  for this parameter. Methods for
adjusting toxicity values are consistent with those used by T-REX (USEPA 2008a).
                                       Ill of 123

-------
                  /             \ f  -L yy  i
Equation G8.  AT = (LDV, orNOAEL)\	      (mammals}
                                \AW,



                          f AW v
   Equation G9.   AT = LDx\ ^- \      (birds )
                      112 of 123

-------
Appendix H. Methods for Estimating Metabolism Rate Constant (kM)

Generally, chemical-specific data are not available to determine the metabolism rate constant
(kM) for aquatic organisms. However, this parameter can be estimated using data from available
bioconcentration factor  (BCF)  studies,  in combination with estimated  rate constants. Two
separate approaches can  be employed to estimate kM.  The first utilizes Equation Al from Arnot
and Gobas 2004. The second utilizes a method described by Arnot et al. 2008. These approaches
are described below.
H.I. Use of Equation Al

In this approach,  Equation Al (see Table Al of Appendix A) is rearranged to solve for kM
(Equation HI). In a BCF study, fish are fed uncontaminated food; therefore, uptake through the
dietary pathway is assumed to be negligible. As a result, it is assumed that kD = 0. BCF studies
with fish involve water-only exposures, so fish do not respire pore water.  As a result, m0 = 1
and nip = 0. To calculate kM, the model user should use the measured  concentration of pesticide
in the test water.  In this  case,  it is assumed that  this value represents the  freely dissolved
pesticide in the water, and therefore, 
-------
          o  Body weight of the fish and water temperature should be set to mean reported
             values from the study. If body weight data are not available for the test fish,
             this approach should not be used and it should be assumed that kM = 0.
          o  This constant is influenced by the % lipid, % NLOM, and % water of the diet
             (VLD, VND and  VWD, respectively). Calculation of this constant requires input of
             diet of the large fish to be 100% medium fish (Table 6 of KABAM tool). If data
             are available from the BCF study report to define  the % lipid, % NLOM, and %
             water of the feed of the test fish, the data  should be entered in the appropriate
             columns of Table 5 of the KABAM tool for the medium fish. Otherwise, if these
             data are not available, the % lipid, % NLOM, and % water of the medium fish can
             be set to the default values of 4, 23, and 73%, respectively.
       kG (d"1) can be estimated from empirical data on body weight over the study period. If kG
       cannot be estimated, the model user can use kG from the large fish.
H.2. Use of Arnot et al. 2008

In this approach, it is assumed that the elimination rate constant measured during the BCF study
(kT) is the sum of  elimination through respiration, fecal elimination  and metabolism of the
pesticide by the fish as well as growth dilution (Equation H3, Arnot et al. 2008). Equation H3
can be rearranged into Equation H4, to solve for kM.

Eg. H3   kT = k2+kE+kG + kM

Eq.H4  kM=kT-k2-kE-kG

Equation H4 can be used to estimate kM from available data from a BCF study.
      kT (d'1) is t
      BCF study.
(d"1) is the total elimination rate constant estimated from the depuration period of the
       As with the first approach, k2 (d"1) can be calculated as ki(empiricai)/KBw (see table A6 of
       Appendix A).
       ks (d"1) can be estimated using the KABAM tool. See discussion above on how to derive
       this constant value.
       kG (d"1) can be estimated from empirical data on body weight over the study period. If kG
       cannot be estimated, the model user can use kG from the large fish.
H.3. Assumptions and Uncertainties

If kM  is calculated as a negative value, it should be assumed that no biotransformation of the
chemical occurs and kM should be set to 0 in Table 2 of the KABAM tool.  Since a negative
biotransformation rate would indicate that the organism is creating the pesticide, it is assumed
that this is not possible for a pesticide.

There is some uncertainty in using the model estimated ko value (using Equation A7), as it may
differ from the growth rate of the test species of the BCF study.
                                       114 of 123

-------
The first approach involves use of total pesticide concentration in fish tissues (Ce; g/kg-ww) and
CWTO (g/L)- It would be appropriate to enter mean values for these parameters into equation H2.
However, variability in these parameters can influence predictions of kM. Therefore, the model
user should explore variability associated with these values by considering standard deviation, as
well as minimum and  maximum values for these  parameters.  This  will result in a range of
relevant kM values.

Both approaches involve use of fish body  composition data (VLB, VNB, and VWB). It would be
appropriate to use mean  values to calculate KBw (and ultimately k2). However, variability in
these  parameters  can influence predictions of kM.   Therefore, the  model user should explore
variability associated with these values by  considering standard deviation, as well as minimum
and maximum values for these parameters.  This approach will result in a range of relevant kM
values.

Both approaches involve  using the  KABAM tool to calculate kE.  This involves the use of diet
composition data  (VLD, VND, and VWD). In the case that data are not available from the study
report  to define the % lipid,  % NLOM,  and % water of the diet  of the test fish,  there is
uncertainty in using default values  for these parameters, as they may differ from the diet of the
test species of the BCF study.
                                       115 of 123

-------
Appendix I. References Cited

Acharya, K., Jack, J.D.  and P.A. Bukaveckas.  2005.  Dietary effects on life history traits of
riverine Bosmina. 2005.  Freshwater Biology, 50: 965-975.

Alsop, III, FJ. 2001a. Birds of North America: East. 1st edition. Smithsonian Handbooks. DK
Publishing, Inc. New York.

Alsop, III, FJ. 2001b. Birds of North America: West. 1st edition. Smithsonian Handbooks. DK
Publishing, Inc. New York.

Andres, S., Baudrimont, M., Lapaquellerie, Y., Ribeyre, F., Maillet, N., Latouche, C. and A.
Boudou.   1999.  Field transplantation of the freshwater bivalve Crobicula fluminea along a
polymetallic contamination gradient (River Lot, France): I. geochemical characteristics of the
sampling sites and cadmium and zinc bioaccumulation kinetics. Environmental Toxicology and
Chemistry, 18(11): 2462-2471.

Arnot, J.A. and F.A.P.C. Gobas. 2004. A food web bioaccumulation model for organic chemicals
in aquatic ecosystems. Environmental Toxicology and Chemistry, 23 (10): 2343-2355.

Arnot, J.A., Macay,  D. and M. Bonnell. 2008. Estimating metabolic biotransformation rates in
fish from laboratory  data. Environmental Toxicology and Chemistry, 27(2): 341-351.

Becker van Slooten, K. and J.  Tarradellas.  1994.  Accumulation,  depuration and growth effects
of tributyltin in  the  freshwater  bivalve  Dreissena  polymorpha  under  field conditions.
Environmental Toxicology and Chemistry, 13 (5): 755-762.

Beers, J.R. 1966.  Studies on the chemical composition of the major zooplankton groups in the
Sargasso Sea off Bermuda. Limnology and Oceanography, 11 (4): 520-528.

Beneski,  J.T. and D.W. Stinson. 1987. Sorex palustris.  Mammalian Species (Publication of the
American Society of Mammalogists), No. 296, p. 1-6.

Bervoets, L.,  Voets,  J., Covaci, A., Chu, S., Qadah, D., Smolders,  R., Schepens, P.  and R. Blust.
2005.  Use of transplanted zebra mussels (Driessenapolymorpha)  to assess the bioavailability of
microcontaminants in Flemish  surface waters. Environmental  Science  and  Technology, 39:
1492-1505.

Burkhard, L.P. 1998. Comparison of two models for predicting bioaccumulation of hydrophobic
organic chemicals in a Great Lakes food web. Environmental Toxicology and Chemistry, 17 (3),
383-393.

Burkhard L.P., Endicott, D.D., Cook, P.M, Sappington, K.G., and E.L. Winchester. 2003.
Evaluation of two methods for prediction of bioaccumulation factors. Environmental Science and
Technology, 37:4626-4634.
                                       116 of 123

-------
Carr,  K.H.,  Coyle,  G.T.  and R.A. Kimerle.   1997.   Bioconcentration of [14C]butyl benzyl
phthalate in bluegill sunfish (Lepomis macrochirus).  Environmental Toxicology and Chemistry,
16 (10): 2200-2203.

Charlander, K.D.  1977. Handbook of freshwater fishery biology,  Volume  2.  The Iowa State
University Press, Ames, IA.

Covich, A.P., Palmer, M.A. and T.A. Crowl.  1999.  The role of benthic invertebrate species in
freshwater  ecosystems;  zoobenthic  species  influence  energy flows  and  nutrient  cycling.
BioScience, 49(2): 119-127.

Croce, V.,  De  Angelis, S.,  Patrolecco, L., Pollesello, S., S.  Valsecchi. 2005.   Uptake and
accumulation of sediment-associated 4-nonylphenol in a benthic invertebrate (Lumbriculus
variegatus, freshwater oligochaete).  Environmental Toxicology and Chemistry, 24 (5): 1165-
1171.

Dunning, J.B. 1984. Body weights of 686 species  of North American  Birds.  Western Bird
Banding Association. Monograph number 1. May 1984.

Fischer, R.U., Congdon, J.D., Difonzo, C.D. and J.M. Aho.  1998. Lipid and reproductive cycles
of bluegills (Lepomis macrochirus) exposed to 35 years of elevated environmental temperatures.
Journal of Thermal Biology, 23 (6): 359-367.

Fisk, A.T., Wiens, S.C., Webster, G.R.B., Bergman, A., D.C.G. Muir.  1998.  Accumulation and
depuration  of sediment-sorbed  Ci2 -  and Cie -  Polychlorinated alkanes  by  oligochaetes
(Lumbriculus variegatus). Environmental Toxicology and Chemistry, 17 (10): 2019-2026.

Gewurtz, S.B., Lazar,  R. and  G.D.  Haffner.   2000.   Comparison  of polycyclic aromatic
hydrocarbon and poly chlorinated biphenyl dynamics in benthic invertebrates of Lake Erie, USA.
Environmental Toxicology and Chemistry, 19 (12): 2943-2950.

Gobas, F.A.P.C. 1993. A model for  predicting the bioaccumulation of hydrophobic  organic
chemicals in aquatic food webs: application to Lake Ontario. Ecological Modeling, 69:1-17.

Hakwer, D.W. and D.W.Connell.  1988.  Influence  of partition  coefficient of lipophilic
compounds on bioconcentration kinetics with fish. Water Research, 22(6),  701-707.

Hannan, H.H. and T.C. Dorris.  1970.  Succession of a macrophyte community in a  constant
temperature river.  Limnology and Oceanography, 15(3), 442-453.

Hendriks,  AJ,  Pieters,  H.  and J.  DeBoer.   1998.   Accumulation  of metals, polycyclic
(halogenated) aromatic hydrocarbons, and biocides in zebra mussel and eel  from the Rhine and
Meuse Rivers. Environmental Toxicology and Chemistry, 17(10): 1885-1898.
                                       117 of 123

-------
Hyotylainen, T., Karels, A. and A. Oikari.  2002. Assessment of bioavailability and effects of
chemicals due to remediation actions with  caging mussels (Anodonta anatind) at a creosote-
contaminated lake sediment site. Water Research, 36: 4497-4504.

Jeppesen, E., Jensen, J.P., S0ndergaard, M., Fenger-Gr0n,  M., Bramm, M.E., Sandby,  K.,
M011er, P.H., and H.U. Rasmussen.  2004.  Impact offish predation on cladoceran body weight
distribution and zooplankton grazing in lakes during winter. Freshwater biology, 49, 432-447.

Jonker,  M.T.O., Hoenderboom,  A.M.,  and A.A.  Koelmans.   2004. Effects  of sedimentary
sootlike materials on bioaccumulation and sorption of poly chlorinated biphenyls. Environmental
Toxicology and Chemistry, 23 (11): 2563-2570.

Kane Driscoll,  S. and  P.P. Landrum.   1997.   A comparison of equilibrium partitioning and
critical body residue  approaches for predicting  toxicity of sediment-associated fluoranthene to
freshwater amphipods.  Environmental Toxicology and Chemistry, 16 (10): 2179-2186.

Kane Driscoll, S., Harkey, G.A. and P.P. Landrum.  1997.  Accumulation and toxicokinetics of
fluoranthene in  sediment bioassays with freshwater amphipods. Environmental  Toxicology and
Chemistry, 16 (4): 742-753.

Kang, Y.-S., Yamamuro, M., Masunaga, S. and J. Nakanishi. 2006. Specific biomagnification of
polychlorinated dibenzo-p-dioxins and dibenzofurans in tufted ducks (Aythya fuligula), common
cormorants (Phalacrocorax carbo) and  their prey from Lake Shinji,  Japan. Chemosphere,  46:
1373-1382.

Kay, D.P., Blankenship, A.L., Coady, K.K., Neigh, A.M., Zwiernik, M.J., Millsap, S.D.,  Strause,
K., Park, C., Bradley, P., Newsted, J.L.,  Jones,  P.O. and J.P.  Giesy.   2005.   Differential
accumulation of polychlorinated biphenyl congeners in the aquatic food web  at the Kalamazoo
River superfund site, Michigan. Environmental Science and Technology, 39: 5964-5974.

Kukkonen, J.,  and P.P. Landrum.  1994.  Toxicokinetics and toxicity of sediment-associated
pyrene to Lumbriculus  variegatus (oligochaeta).  Environmental Toxicology and Chemistry,  13
(9): 1457-1468.

Kukkonen, J.V.K., Landrum,  P.P., Mitra,  S., Gossiaux, D.C., Gunnarsson,  J.  and D.Weston.
2004.  The role of desorption for describing the bioavailability of  select polycyclic aromatic
hydrocarbon and polychlorinated biphenyl  congeners for seven laboratory-spiked sediments.
Environmental Toxicology and Chemistry, 23 (8): 1842-1851.

Kwon, T.-D., Fisher,  S.W., Kim, G.W.,  Hwang, H. and J.-E. Kim. 2006.  Trophic transfer and
biotransformation of polychlorinated biphenyls in zebra mussel, round goby, and smallmouth
bass in Lake Erie, USA. Environmental  Toxicology and Chemistry, 25(4):  1068-1078.

Landrum, P.P.,  Robinson, S.D., Gossiaux, D.C., You,  J., Lydy, M.J., Mitra, S. and T.E.M. Ten
Hulscher. 2007. Predicting bioavailability  of  sediment-associated  organic  contaminants  for
Diporeia spp. and oligochaetes. Environmental Science and Technology, 41: 6442-6447.
                                       118 of 123

-------
Lariviere, S. and L.R.Walton.  1998. Lontra canadensis. Mammalian Species (Publication of the
American Society of Mammalogists), No. 587, 1-8.

Lariviere, S. 1999. Mustela vison. Mammalian Species (Publication of the American Society of
Mammalogists), No. 608, 1-9.

Leonards, P.E.G., Zierikzee, Y., Brinkman, U.A.T., Cofmo, W.P., Van Straalen. N.M., and B.
Van Hattum.  1997. Environmental Toxicology and Chemistry, 16(9): 1807-1815.

Liber,  K.,  Gangl, J.A.,  Cony,  T.D.,  Heinis,  L.J and F.S.  Stay.   1999.   Lethality  and
bioaccumulation of 4-nonylphenol  in bluegill  sunfish in littoral enclosures.  Environmental
Toxicology and Chemistry, 18 (3): 394-400.

Liebig, M.,  Egler, P., Oehlmann, J. and T. Knacker.  2005.  Bioaccumulation  of  14C-17a-
ethinylestradiol by the aquatic oligochaete Lumbriculus variegatus in spiked artificial sediment.
Chemosphere, 59: 271-280.

Lin, H.-I, Berzins, D.W., Myers, L., George, W.J., Abdelghani, A. and K.H. Watanabe.  2004. A
Bayesian approach to parameter estimation for a crayfish (Procambarus spp.) bioaccumulation
model. Environmental Toxicology and Chemistry, 23 (9): 2259-2266.

Lotufo, G.R., Landrum, P.P., Gedeon, M.L., Tigue, E.A., and L.R. Herche.  2000.  Comparative
toxicity and  toxicokinetics of DDT  and its  major  metabolites in freshwater  amphipods.
Environmental Toxicology and Chemistry, 19 (2): 368-379.

Lotufo, G.R., Landrum, P.P., and M.L. Gedeon.  2001.  Toxicity and bioaccumulation of DDT in
freshwater  amphipods in  exposures to spiked sediments.   Environmental Toxicology  and
Chemistry, 20 (4): 810-825.

Lu, X., Reible, D.D., Fleeger,  J.W., and Y. Chai.  2003. Bioavailability of desorption-resistant
phenanthrene  to  the  oligochaete  Ilyodrilus  templetoni.   Environmental Toxicology  and
Chemistry, 22 (1): 153-160.

Lurling, M. and E. Van Donk.  1997. Life history consequences for Daphnia pulex feeding on
nutrient-limited phytoplankton. Freshwater Biology, 38, 693-709.

Marvin, C.H., Howell, E.T., Kloic, T.M.  and E.J. Reiner.  2002.  Polychlorinated dibenzo-/?-
dioxins and dioxinlike polychlorinated biphenyls in sediments and mussels at three sites in the
lower Great Lakes, North America.  Environmental Toxicology and Chemistry, 21 (9): 1908-
1921.

McKee, MJ. and C.O. Knowles.  1987.  Levels of protein,  RNA, DNA, glycogen and lipid
during growth and development of Daphnia magna Straus (Crustacea: Cladocera).  Freshwater
Biology, 18: 341-351.
                                       119 of 123

-------
Mineau,  P., Collins, B.T.  and A. Baril.  1996.  On  the  use of scaling  factors  to  improve
interspecies extrapolation of acute toxicity in birds. Regulatory Toxicology and Pharmacology,
24: 24-29.

Miranda,  L.E.  and  W.D.  Hubbard.  1994.   Length-dependent winter  survival and  lipid
composition of age 0 largemouth bass in Bay Springs Reservoir, Mississippi. Transactions of the
American Fisheries Society, 123: 80-87.

Mitra,  G., Mukhopadhyay, P.K. and S.Ayyappan.   2007.   Biochemical  composition of
zooplankton community grown in freshwater earthen ponds: nutritional implication in nursery
rearing offish larvae and early juveniles. Aquaculture, 272, 346-360.

Morrison,  H.A.,  Gobas, F.A.P.C.,  Lazar,  R.,  Whittle,  D.M., and  GD.  Haffner.   1997.
Development and verification of a benthic/pelagic food web  bioaccumulation model  for PCB
congeners in Western Lake Erie. Environmental Science and Technology, 31: 3267-3273.

Morrison, H.A.,  Whittle, D.M. and G.D.  Haffner.  2000. The  relative importance of species
invasions  and  sediment disturbance in regulating chemical dynamics  in  western Lake  Erie.
Ecological Modeling, 125: 279-294.

Neumann, R.M.  and B.R. Murphy.  1992.  Seasonal  relationships of relative weight to  body
composition in  white  crappie,  Pomoxis  annularis  Rafinesque. Aquaculture  and  fisheries
Management, 23: 243-251.

OECD. 1996. Guidelines for testing of chemicals: Bioconcentration: flow-through Fish test, 305.
Organization for Economic Co-operation and Development. Adopted June 14, 1996.

Oliver, E.G. and AJ.  Niimi.   1988.   Trophodynamic analysis of polychlorinated  biphenyl
congeners and other chlorinated hydrocarbons in the Lake Ontario ecosystem.  Environmental
Science and Technology, 22: 388-397.

Page, L.M. and B.M. Burr.  1991.  A field guide to freshwater fishes; North America,  North of
Mexico. The Peterson Field Guide Series. Houghton Mifflin Company, Boston.

Petersen, K.E. and T.L. Yates. 1980. Condylura crisata. Mammalian Species (Publication of the
American Society of Mammalogists), No. 129, p. 1-4.

Price, DJ.  and WJ. Birge.  2006.  Use of sunfish and stoneroller minnows as real-time in situ
biomonitors of  PCB  contamination  in  freshwater   streams.  Environmental  Science  and
Technology, 40: 4388-4393.

Raven, P.H., Evert, R.F. and S.E. Eichhorn.  1999.   Biology of Plants.  Sixth Edition. W.H.
Freeman and Company /Worth Publishers. New York.

Reid, F.A.  2006.  A field  guide to mammals of North America.  4th edition. Peterson Field
Guides. Houghton Mifflin Company, Boston.
                                       120 of 123

-------
Rice, C.P. and D.S. White.  1987.  PCB availability assessment of river dredging using caged
clams and fish. Environmental Toxicology and Chemistry, 6: 259-274.

Sangster, J.  2007.  LOGKOW A databank of evaluated octanol-water partition coefficients
(Log P).  Available online at: http://logkow.cisti.nrc.ca/logkow/index.jsp. Accessed 29 January
2007.

Sethajintanin, D., Johnson,  E.R.,  Loper,  B.R. and K.A.  Anderson. 2004.  Bioaccumulation
profiles of chemical contaminants in fish from the Lower Willamette River, Portland Harbor,
Oregon.  Archives of Environmental Contamination and Toxicology, 46: 114-123.

Shifrin, N.S. and S. W. Chisholm.  1981.  Phytoplankton lipids:  interspecific  differences and
effects of nitrate, silicate and light-dark cycles.  Journal of Phycology  17(4): 374-384.

Sidwell,  V.  1981.   Chemical  and nutritional composition of finfishes, whales, crustaceans,
mollusks, and their products.  National Oceanic and Atmospheric Administration Technical
Memorandum, NMFS F/SEC-11.

Sladecek, V. and A. Sladeckova. 1963.  Relationship between wet weight and dry weight of the
periphyton. Limnology and  Oceanography, 8 (2): 309-311.

Smithsonian.  2008. North  American Mammals. Smithsonian  Museum  of Natural  History.
Accessed online 22-August-2008 at: http://www.mnh.si.edu/mna/search_name.cfm.

Spacie, A., L. S. McCarty and  G. M. Rand. 1995. Bioaccumulation and  bioavailability in
multiphase  systems. In G.  M. Rand,  editor,  Fundamentals of Aquatic Toxicology: Effects,
Environmental Fate and Risk Assessment 2nd Edition. Taylor and Francis, Washington DC.

Stange, K.  and  D.L.  Swackhamer. 1994.  Factors affecting phytoplankton  species-specific
differences in accumulation of 40 polychlorinated biphenyls (PCBs).  Environmental Toxicology
and Chemistry, 13 (11): 1849-1860.

USD A. 2005. USD A National  Nutrient  Database for Standard Reference, Release 18.  Nutrient
data laboratory Home Page. United States  Department of Agriculture, Agricultural Research
Service. Available online at: http://www.nal.usda.gov/fnic/foodcomp.

USEPA.  1993. Wildlife Exposure factors handbook.  United  States Environmental Protection
Agency. Office of Research and Development. EPA/600/R-93/187.

USEPA.  1995.   Great Lakes Water  Quality  Initiative  Technical Support Document for the
Procedure to Determine Bioaccumulation Factors.   United States  Environmental  Protection
Agency.  Office of Water. Washington, DC.  EPA/820/B-95/005.
                                       121 of 123

-------
USEPA.  1996. Ecological effects test guidelines: OPPTS 850.1730, Fish BCF. United States
Environmental Protection Agency. Prevention, Pesticides and Toxic Substances. EPA 712-C-96-
129.

USEPA.  2000. Methodology for Deriving Ambient Water Quality Criteria for the Protection of
Human Health (2000).  United States Environmental Protection Agency.  Office of Water,
Washington, DC. EPA-822-B-00-004.

USEPA. 2002. Guidance for Selecting Input Parameters in Modeling the Environmental Fate and
Transport of Pesticides, Version II. US Environmental Protection Agency, Washington DC.
Online at: http://www.epa.gov/oppefedl/models water/input_guidance2_28_02.htm.

USEPA.  2003. Methodology for Deriving Ambient Water Quality Criteria for the Protection of
Human Health (2000).  Technical  Support Document Volume  2: Development of National
Bioaccumulation Factors. United States Environmental Protection Agency. Office of Water,
Washington, DC. EPA-822-R-03-030.

USEPA. 2004a. Overview of the Ecological Risk Assessment Process in the Office of Pesticide
Programs. U.S. Environmental Protection Agency, Office of Prevention,  Pesticides and Toxic
Substances, Office of Pesticide Programs, Washington DC. 100 pp. January 23, 2004.

USEPA. 2004b. Exposure Analysis Modeling System (EXAMS): User Manual and System
Documentation. EPA/600/R-00/081, Revision G (May 2004). U.S. Environmental Protection
Agency, National Exposure Research Laboratory, Athens, Georgia.

USEPA.  2008a. User's  Guide: T-REX Version  1.4.1  (Terrestrial Residue Exposure  model).
United States Environmental Protection Agency. Environmental Fate and Effects Division.

USEPA.  2008b. White  paper on methods for assessing ecological  risks of pesticides  with
persistent, bioaccumulative  and  toxic characteristics. Submitted  to the FIFRA Scientific
Advisory Panel for review and comment. United  States  Environmental Protection Agency,
Environmental Fate and Effects Division.

USEPA. 2008c. Terms of Environment: Glossary, Abbreviations and Acronyms. United States
Environmental      Protection      Agency.     Accessed      8-September-2008,      at:
http://www.epa.gov/OCEPATERMS/.

USEPA. 2009. EPI Suite v4.0. United States Environmental Protection Agency. Available for
download at: http://www.epa.gov/oppt/exposure/pubs/episuitedl.htm.

USFWS.  1993. Endangered and Threatened Wildlife and Plants: Designation of critical Habitat
for the Silver Rice Rat. United States Fish and Wildlife Service. 50 CFR Part 17.

USFWS.  2008.   Endangered Species  Program. United  States  Fish and Wildlife Program.
Accessed online 23-July-2008. http://www.fws.gov/endangered/wildlife.html#Species
                                      122 of 123

-------
USGS. 2008a. Dissolved oxygen saturation tables. United States Geological Survey. Available
online at: http://water.usgs.gov/software/DOTABLES/.

USGS. 2008b. National Water-Quality Assessment Program. United States Geological Survey.
http://water.usgs.gov/nawqa/.

Van Haelst, A.G., Zhao, Q., Van Der Wielen, F.W.M., Govers, H.A.J. and P. De Voogt.  1996.
Determination of bioconcentration factors of eight tetrachlorobenzyltoluenes in the zebra mussel
Dreissenapolymorpha. Ecotoxicology and Environmental Safely, 34: 35-42.

Vidal, M.-L., Basseres,  A.  and  J.-F.  Narbonne.   2002.   Seasonal variations of pollution
biomarkers in two  populations of Corbicula fluminea (Miiller).  Comparative Biology and
Physiology Part C, 131: 133-151.

Vijverberg, J. and  T.H.  Frank.   1976.  The chemical  composition and  energy contents  of
copepods and cladocerans in relation to their size.  Freshwater Biology, 6: 333-345.

Wetzel, R.G. 1983.  Limnology. Second Edition. Saunders College Publishing, Harcourt Brace
College Publishers, Fort Worth.

White, P.A., Rasmussen,  J.B.  and C. Blaise.  1998.  Genotoxic substances in the St. Lawrence
system II: extracts of fish and macroinvertebrates from the St. Lawrence and Saguenay Rivers,
Canada. Environmental Toxicology and Chemistry, 17 (2): 304-316.

Wolfe, J.L. 1982. Oryzomys palustris. Mammalian Species (Publication of the American Society
of Mammal ogists), No. 176, p. 1-5.

Wright, J.C.  1958.  The limnology of Canyon Ferry Reservoir, I. phytoplankton-zooplankton
relationships  in  the euphotic  zone during  September  and October 1956.   Limnology and
Oceanography, 3 (2): 150-159.
                                       123 of 123

-------