EPA/600/R-17/277
User's Manual MCnest -
Markov Chain Nest Productivity Model
Version 2.0
Matthew Etterson
Mid-Continent Ecology Division
National Health and Environmental Effects Research Laboratory
Office of Research and Development
U. S. Environmental Protection Agency
6201 Congdon Boulevard
Duluth, MN 55804
September 2017
-------
User's Manual for MCnest v2.0 - September 2017
Blank Page
2
-------
User's Manual for MCnest v2.0 - September 2017
Table of Contents
Table of Contents 3
I. Introduction 5
A. What is MCnest? 5
B. What has changed from previous MCnest releases? 5
C. How to use this Manual 5
D. What does MCnest do? 5
E. Why is MCnest needed? 6
F. What information is needed to use MCnest? 6
II. Quick start guide 7
A. Installation of MCnest 7
B. Sample Data Files 8
C. MCnest startup window 8
D. Running a simulation with no pesticide exposure for a selected species 9
E. Running a simulation with a pesticide exposure 10
1. Test Results 11
2. Exposure Model 13
III. MCnest outputs 17
A. Using the table 17
B. The Output menu 18
1. Phase diagram 19
2. Brood histogram 22
3. Exposure series 22
4. TIM 22
5. Log file 23
6. Save row 24
C. Sort table 24
D. Save table 25
E. Delete rows 25
IV. Systematic description of MCnest controls 25
A. MCnest window 25
1. User-input Controls 26
2. The MCnest menu 27
3. The Toxicology menu 29
4. The Clear menu 29
B. The Life History window 29
C. The Test Results window 31
1. Avian Reproduction Test 32
2. Avian Acute Oral Toxicity Test 33
3. Avian Dietary Toxicity Test 33
4. Toxicity Thresholds 33
D. The TREX window 34
E. The TIM windows 36
F. The Batch window 39
1. Species 40
2. Application Date 41
3
-------
User's Manual for MCnest v2.0 - September 2017
3. Application Rate 41
E. Other MCnest menu options 41
1. Random Numbers 41
2. Load MCnest file 42
3. Import Time-varying parameters 43
V. Literature Cited 44
This model and user's manual have been reviewed in accordance with U.S.
Environmental Protection Agency policy and approved for publication. Mention
of trade names or commercial products does not constitute endorsement or
recommendation for use.
4
-------
User's Manual for MCnest v2.0 - September 2017
I. Introduction
A. What is MCnest?
The Markov chain nest productivity model, or MCnest, is a set of algorithms for integrating the
results of avian toxicity tests with reproductive life-history data to project the relative magnitude
of chemical effects on avian reproduction (Bennett et al. 2005, Bennett and Etterson 2007,
Etterson & Bennett 2013). MCnest was designed and optimized in MATLAB
(www.mathworks.com), but most users will run a compiled version as a standalone program that
does not require the user to have MATLAB software. The mathematical foundation of MCnest
is in the analysis of Markov chains, which provides a flexible template for modeling the variation
in avian breeding cycles among species.
B. What has changed from previous MCnest releases?
There are two versions of MCnest available, MCnest vl.O (Basic MCnest) and MCnest v2.0
(TIM/MCnest). Both versions are available for download at: https://www.epa.gov/chemical-
research/markov-chain-nest-productivitv-model. Version 2.0 extends the basic version to enable
the use of exposure profiles generated using the Terrestrial Investigation Model (TIM, USEPA
2015). MCnest now accounts for background mortality to breeding females as well as the
potential for adult mortality due to pesticide exposure. However, the MCnest breeding season
algorithms themselves have not changed between Basic MCnest (Bennett & Etterson 2013b-d)
and this version, and simulations run in Basic MCnest can be replicated in TIM/MCnest.
C. How to use this Manual
This manual is organized into four major sections. Section I gives some basic background on
MCnest and the justification for its development. Section II is a quick start guide to running
simulations in MCnest that walks users through several examples. Section III describes the
MCnest outputs and their interpretation. Section IV is a complete list of user inputs that can be
modified in MCnest. Parameterization of the TIM model, which can now be invoked by MCnest,
is described in the user's guide to TIM.
D. What does MCnest do ?
The purpose of MCnest is to quantitatively estimate the relative change in the number of
successfully fledged broods per female per year of avian species exposed to a specific pesticide
application scenario. The relative change in the number of successful broods is estimated by
comparing model results based on a defined pesticide application scenario with a no-pesticide
scenario. To express the results in terms of annual reproductive success, the model estimate of
the number of successfully fledged broods per female is multiplied by the expected number of
fledglings per successful nest, which is taken from the literature.
Each model simulation follows the breeding activities of a population of females each day
throughout a breeding season. The breeding activity is described by a series of phases through
which the female transitions (i.e., pair formation, egg laying, incubation, nestling rearing, and
waiting periods prior to starting a new nest attempt). The temporal pattern of breeding activity
of each female varies due to differences in the initiation date of the first nest attempt and due to a
5
-------
User's Manual for MCnest v2.0 - September 2017
specific probability each day that the nest attempt could fail from ecological causes such as
predation or adverse weather. When a nest attempt fails, each female can make a new attempt if
there is time remaining in the breeding season, and for many species, females make a new
attempt after completing a successful brood. If the simulation incorporates one or more pesticide
applications, the pesticide exposure may represent an additional cause of nest failure depending
on the types of pesticide effects observed in tests and the timing of the application relative to the
phase of the nesting attempt for each female. When a nest attempt fails due to pesticide
exposure, each female may make a new attempt if there is time remaining in the breeding season
and pesticide residues decline to levels that would not affect parental well-being. As MCnest
follows each female of the population through the breeding season, it tabulates the number of
nest attempts and successful broods (i.e., broods surviving to fledging) per female in the
population.
As in the previous version of MCnest (Bennett & Etterson, 2013), our goal has been to limit the
number of model inputs so that the model is applicable to as many species as possible. We have
also focused on using model inputs that are readily available from published literature and from
toxicity data submitted as part of the current pesticide registration process under the Federal
Insecticide, Fungicide, and Rodenticide Act (FIFRA). MCnest is designed to accommodate the
structure of quantitative dose-response data from reproduction tests if it becomes available.
E. Why is MCnest needed?
Prior to the adoption of Basic MCnest (Bennett & Etterson 2013) the pesticide registration
process used deterministic risk quotients (RQ) as the primary metric for assessing the potential
risks of pesticide exposure to avian reproduction. While RQs can be used to classify the
potential risk, they do not quantify the magnitude of risk nor provide information on differences
in risk among species or pesticide application scenarios. MCnest offers an approach for more
quantitatively estimating the magnitude of avian reproductive effects when the ultimate goal is to
probabilistically describe the risk of reproductive effects or to estimate the magnitude of effects
to a population.
In population modeling, we need to know how a pesticide application scenario affects the annual
reproductive success of exposed birds compared to a scenario without pesticides. Annual
reproductive success, also known as the fecundity rate, is here defined as the number of
fledglings produced per female in the population over an entire breeding season. It represents
the cumulative production of fledglings during a breeding season, including re-nesting attempts
after nest failures and multiple successful nesting attempts. In MCnest, fecundity can be
estimated as the average number of successfully fledged broods per female per season times the
average number of fledglings per successful brood.
F. What information is needed to use MCnest?
The basic version of MCnest requires data from avian toxicity tests, avian species life-history
profiles, and a pesticide-use scenario that defines the timing and temporal pattern of exposures.
The most common sources of toxicity data are the results from three standard avian toxicity tests:
1) the acute oral LD50 test (USEPA 2012a), 2) the dietary LC50 test (USEPA 2012b), and 3) the
avian reproduction test (USEPA 2012c). When available, additional toxicity data from
6
-------
User's Manual for MCnest v2.0 - September 2017
alternative sources or from alternative interpretations of standard toxicity tests may be useful.
The selection and use of toxicity data in the model is discussed in the MCnest Technical Manual
(Etterson & Bennett 2017).
In the basic version of MCnest, model simulations are run on avian species of interest (i.e., life-
history profiles developed for specific species). A series of life-history parameters is required
for each species of interest. The user can select one of the species for which a suite of default
life-history parameters has been developed, create a new species, or modify the profiles of
existing candidate species by directly inputting the list of life-history parameters. The use of
default species profiles and the creation of new or modified species profiles are discussed later in
the manual.
Finally, MCnest requires that the user define a pesticide-use scenario by parameterizing an
exposure model. With this release, MCnest offers two options for parameterizing exposure, the
Terrestrial Residue Exposure Model (T-Rex, USEPA 2012d) and the Terrestrial Investigation
Model (TIM, USEPA 2015). The first option is identical to that of Basic MCnest and requires, at
a minimum, specifying the date(s) of application, the application rate(s), and the half-life of
residue degradation on foods relevant to the species of interest. The user also has four options
for how the application rate is converted to the ingested daily dose on the day of each
application, which are explained later in Section IV.C. The second option, TIM, requires more
information on physical chemical properties and test results beyond the three standard avian tests
mentioned above. Parameterization of TIM is described in the TIM user's manual and that
information is not repeated here. User's wishing to employ the TIM option will need to cross
reference with the User's Guide for TIM (USEPA 2015).
II. Quick start guide
A. Installation of MCnest
Probably the most efficient way to understand MCnest is to begin using the model with default
life-history data. To begin using the program, extract the MCnest files from "MCnest2.0.zip" to
an appropriate directory on your computer. MCnest2.0.zip should contain six files, all of which
must be in the same directory (Table 2.1).
Table 2.1. Files included in the MCnest v 2.0 release ("MCnest2.0.zip")
File name
Function
MCnest.exe
TIM.MCnest.Species.Library.xlsx
CallTim.bat
StopTim.bat
TIM3.0_Beta_3-18-15b.exe
readme.txt
Main MCnest executable. Run this file to start MCnest
Draft species profiles for use in MCnest
batch file for implementing TIM from Matlab
batch file for interrupting TIM from Matlab
Main TIM executable
Matlab compiler information
If this is your first time using MCnest, you will need to install the MATLAB compiler runtime
library (MCR). It can be downloaded from Mathworks at:
http://www.mathworks.com/products/compiler/mcr/ by selecting the Windows 64-bit R2016b
7
-------
User's Manual for MCnest v2.0 - September 2017
version. Installation of the MCR requires administrator privileges. To install, double-click
"MCR_R2016b_win64_installer.exe" or select "Run..." and browse to the MCR file.
Installation may take several minutes due to the very large file size. Once the MCR is installed,
you may run "MCnest.exe" file without administrator privileges. Be patient; it can take up to 60
seconds to open!
B. Sample Data Files
With this release of MCnest we are including 14 sample data files on seven insecticides (Table
2.2). Each insecticide has two files, one for the toxicity test results and one for TIM parameters.
These files should be extracted from the zip file, but can be placed anywhere on your computer.
For ease of access it is recommended to put them in a subfolder of the directory containing the
MCnest executable.
Table 2.2. Sample data files included with MCnest 2.0.
Insecticide
Class
Mode of Action
carbaryl
carbamate
AChE inhibition
chlorpyrifos
organophosphate
AChE inhibition
indoxacarb
oxadiazine
Voltage-dependent sodium channel blocker
X-cyhalothrin
pyrethroid
Sodium channel modulator
malathion
organophosphate
AChE inhibition
methomyl
carbamate
AChE inhibition
permethrin
pyrethroid
Sodium channel modulator
C. MCnest startup window
Once the MATLAB compiler runtime library is installed, MCnest is called by double-clicking
the executable file (MCnest2.0.exe). Doing so brings up the main MCnest screen (Figure 2.1).
MCnest opens ready to run a control simulation (i.e., no pesticide exposure or effects).
Number of birds: This parameter defines the simulation sample size (number of breeding
females). The associated Flock size parameter defines the level of replication associated with the
output metrics (number of successful broods or number of offspring fledged). For example, the
default values of 10,000 birds with a flock size of 25 results in 400 replicate estimates of the
output metrics.
Species: The Species block contains a dropdown menu of species names for which life-history
profiles have been pre-coded for use in MCnest. The default is the first species in the list,
Canada goose. Specific details about the life-history of species in the library, and functions to
edit life-history parameters to create new species are found on the Life History page, which can
be accessed under the main MCnest menu. More details about editing and creating life-histories
are given in Section III.
8
-------
User's Manual for MCnest v2.0 - September 2017
Figure 2.L Main MCnest window.
^ MCnest Version 2.0
MCnest Toxicology Clear
1 ° II a
Ready
r> ccsure Algorithm
MCnest working directory
Simuation iterations
. .-.i—^ .:-er-:cf J.
Number of birds: 10OOO
Flock size: 25
Output table
Species: Canada goose (giant)
Output
o Broods/Female
Fledglings/Female
Season (edit on Life History page)
Date of first egg of first nest
Day: 20
Month: March
Date of first egg of last nest
Day: 20
Month: April
About MCnest
This is MCnest v2.0
it was compiled on 21-Jul-2017
D. Running a simulation with no pesticide exposure for a selected species
Once the program is loaded, verify that Number of birds edit box and the Flock size are set to
10,000 and 25, respectively and that Exposure Algorithm is set to CONTROL, Under Species,
click American robin. Then click Run in the upper left corner. While the simulation is running,
the Run Status box in the upper right of the main MCnest window will initially display the
message "Simulation running, please wait..." while the Markov chain transition matrix is
constructed. Once the simulation begins to ran, the Run Status box will display the updating
message "Iteration X of Y," where X is the recently completed iteration and Y is the total
number of iterations (breeding females). When the simulation finishes, the window will revert to
a ran time message (Figure 2.2).
9
-------
User's Manual for MCnest v2.0 - September 2017
Figure 22. Sample output from a simulation with 400 replicates of 25 breeding American robins.
^ MCnest Version 2.0
MCnest Toxicology Clear
1 ° II a
Exposure Algorithm:
Simulation iterations
Number of birds: 10000 | Flock size: 25
Season (edit on Life History page)
Date of first egg of first nest
Month: j April - Day: 112
Date of first egg of last nest
Month: July
About MCnest:
This is MCnest v2.0.
It was compiled on 21-Jul-2017.
Run time: 0 m 19 s.
MCnest working directory:
Species: American robin
SPECES
American robin
Output
® Broods/Female
© Fledglings/Female
W:\1 EPA\MCnest\TIM MCnesttMCnestCode\GUI
BROODS B L95 B U95 APPS APP1 RATE 1 crop pesticide Exposure Mortality
1.85 1.52 2.2 On/a n/a unset unset CONTROL 0
The default tabular output to a MCnest simulation gives the species name, the expected number
of successful broods per female (BROODS), the upper and lower 95th percentile confidence
intervals for BROODS (based on the 400 replicates), and the number of pesticide applications
simulated (APPS), which is 0 in this case. As more simulations are run, new output is appended
to this table.
Note that the numerical results on your screen may differ slightly from Figure 2.2 because of
variation in the actual sequences of random numbers utilized by the program. MCnest does give
you the ability to set the random seed so that simulations may be replicated exactly. These
functions are covered in detail below in Section IV.
E. Running a simulation with a pesticide exposure
MCnest is designed to simulate the effects of pesticide applications on the reproductive output of
breeding birds. There are many options in MCnest for how this is done, but all involve
essentially a two-step procedure. Test results must be entered into the model for computing the
surrogate endpoints (Bennett and Etterson 2013) and an exposure model must be parameterized.
10
-------
User's Manual for MCnest v2.0 - September 2017
MCnest offers two choices for exposure models, T-REX (USEPA 2012d) and TIM (USEPA
2015).
1. Test Results
With the MCnest main screen still displaying the American robin simulation, choose Test
Results under the main Toxicology menu. After a short delay, the Test Results window should
appear (Figure 2.3)
Figure 2.3. The Test Results window in MCnest.
* MCnest Version 2,0: Test Results
Half-life on dietary items:
35 days
Avian Reproduction Test
Avian reproduction test source (MRD): I
Test species:
Treatment group data:
Treatment levels:
Average food consumption (g/bird/day):
Average initial female body weight (g): NaN
Average initial male body weight (g): NaN
Average final female body weight (g): NaN
Average final male body weight (g): NaN
Estimated average daily
dose (mg/kg bwt/day):
NaN
NaN
NaN ]
NaN
NaN 1
NaN
NaN
NaN
NaN
NaN J
NaN |
NaN
NaN
NaN
NaN
NaN
NaN
NaN
NaN
NaN |
NaN
NaN
NaN
NaN |
NaN
[" NaN
NaN
NaN
NaN J
NaN
| NaN |
NaN |
NaN |
NaN J
NaN 1
Test NOAELs
Endpoint
NOAEL
Endpoint
NOAEL
Number of eggs laid:
>
% 14-d chicks of eggs set:
» v
% viable eggs of eggs set:
> ~
shell thickness:
> ~
% live 3-wk embryos of viable eggs:
> ~
hatch weight:
> ~
% hatchlings of live 3-wk embyros:
> ~
14-d chick wt:
> ~
% hatchlings of eggs set:
> H
prelaying female weight:
> ~
% 14-d chicks of hatchlings:
> ~
prelaying male weight:
> w
Avian LD50 Test (adults):
Mineau scaling factor:
LD50 (mg/kg bwt}:
Body weight:
Scaled
LD50 test source (MRD):
Test species:
Avian LCSOTest Quveniles):
NaN
NaN
LC50:
Fraction of LC50 (i.e., 0.1 or
Mean body weight (g): |~ NaN
Mean food ingestion rate (g/d): NaN
LC50 source (MRD):
Test species:
Fraction of
LC50 as
dose
(mg/kg bwt):
NaN
MCnest Toxicity Thresholds (mg/kg/d):
Alternative behavioral threshold: 9999
1/10 9999
Adult prelaying body weight: 9999
Eggs laid per hen: 9999
Mean eggshell thickness per pen: 9999
Viable eggs set per pen: 9999
Hatchlings per viable egg per pen: 9999
14-d chicks per hatch ling per pen: 9999
Fraction of juvenile 5-d LC50 9999
A full description of the Test Results controls is given in Section IV. For now, choose the Load
Data option at the top left and navigate to the folder in which you stored the sample data files.
Select the data file for Lambda Cyhalothrin ("LambdaCyhalothrinTestResults.mat") and click
Open. The controls on the TestResults page should now look like Figure 2.4, below. Some
11
-------
User's Manual for MCnest v2.0 - September 2017
additional context for understanding the controls on the TestResults page is presented
immediately below Fig. 2.4. To continue with the examples, skip to 2. Exposure Model.
Figure 2.4. TestResults window with data parameterized for /.-Cyhalothrin
^ MCnest Version 2.0: Test Results
Half-life on dietary items:
Lambda Cyhalothrin
Avian Reproduction Test
Avian reproduction test source (MRID):
Test species: j|
Treatment group data.
Treatment levels:
Measured concentration (mg/kg diet):
Average food consumption (g/bird/day):
Average initial female body weight (g):
Average initial male body weight (g):
Average final female body weight (g):
Average final male body weight (g):
Estimated average daily dose
(mg/kg bwt/day):
1087
1135
1249
Number of eggs laid: 1
% viable eggs of eggs set: 2
% live 3-wk embryos of viable eggs: \2_
% hatchlings of live 3-wk embyros: 2
% hatchlings of eggs set:
% 14-d chicks of hatchlings: 2
2
50.8
1075
1113
1251
3
NaN
NaN
NaN
NaN
NaN
4
NaN
NaN
NaN
NaN
NaN
NaN
5
NaN
NaN
NaN
NaN
NaN
% 14-d chicks of eggs set:
shell thickness: j
hatch weight:
14-d chick wt:
prelaying female weight:
prelaying male weight:
Avian LD50 Test (adults):
Mineau scaling factor: 115
LD50 (mg/kg bwt): 50000
LD50 test source (MRID)
Test species:
Avian LC50 Test (juveniles):
LC50: 2354
Fraction of LC50 (i.e., 0.1 or 0.5): [ 0.5
Mean body weight (g): 17.4
Mean food ingestion rate (g/d): 14.96
LC50 source (MRID):
Test species:
MCnest Toxicity Thresholds (mg/kg/d).
Alternative behavioral threshold:
Body weight:
Scaled LD50:
R2084922.pdf
Mallard
1049
338127587
Fraction of LC50 as
dose (mg/kg bwt):
Northern Bobwhite
9999 1/10 LD50:
Adult prelaying body weight
Eggs laid per hen:
Mean eggshell thickness per pen:
Viable eggs set per pen
Hatchlings per viable egg per pen
14-d chicks per hatchling per pen
3381 2759
10.0792
0.94334
10.0792 j
10.0792 |
10 0792
10.0792 j
Fraction of juveni le 5-d LC50 1011 95
On the TestResults window, results pertaining to the Acute Oral Toxicity Test (USEPA 2012a)
and to the Dietary Toxicity Test (USEPA 2012b) are entered on the right-hand side and results
pertaining to the Avian Reproduction Test (USEPA 2012c) are entered on the left-hand side (Fig.
2.4). For the Avian Reproduction Test and the Dietary Toxicity Test, results must be converted
from dietary concentrations (mg/kg - diet) to doses (mg/kg - body weight), which requires
knowledge of the body weights and food consumption rates of tested animals. Details on these
conversions are provided in the technical manual.
For the Avian Reproduction Test, the No Observed Adverse Effects Levels (NOAELs) are
entered in the bottom left panel, which has drop-down menus for 12 NOAELs, though only 8 are
currently used by the model. These drop-down menus contain only integers, which correspond
to the potential test levels of the Avian Reproduction Test. For each measured test endpoint, the
user should choose the integer corresponding to the NOAEL for that endpoint. For Lambda-
Cyhalothrin, only 2 dietary concentrations (and control) were tested, thus the highest possible
integer would be 2. If the NOAEL is unbounded then the symbol '>' can be chosen to indicate
this and the program will use 9999 for the value of surrogate endpoints to which this test
endpoint is connected.
For the Acute Oral Toxicity Test, scaling factors suggested by Mineau et al. (1996) can be used
to rescale the LD50 from the tested species (typically Northern Bobwhite or Mallard) to the body
12
-------
User's Manual for MCnest v2.0 - September 2017
weight of the modeled species. This is done using an allometric equation with an empirically
determined exponential parameter. Mineau et al. (1996) published values for many insecticides,
primarily organophosphates and carbamates, and suggested an overall mean exponential
parameter among insecticides of 1.15, which is the default value used by MCnest. A value of 1
will result in no rescaling.
For the Dietary Toxicity Test, a choice for the fraction of the LC50 must be made to determine
the surrogate endpoint corresponding to this measured endpoint. The default is 0.5. The choice
of fraction to use is up to the user and discussion of the choice of appropriate value can be found
in Bennett et al. (2005) and Bennett & Etterson (2013b,d).
The toxicity data entered into the TestResults panels are used to calculate the surrogate
endpoints (Fig. 2.4, bottom right panel), which are compared to time-weighted averages of dose
equivalents on a daily basis to determine whether a given nest attempt fails or succeeds. Much
more detail on calculation of surrogate endpoints and the time-weighted averages to which they
are compared are provided by Bennett & Etterson 2013b,c,d.
2. Exposure Model
The examples below parameterize the two exposure models (T-REX and TIM) under the same
scenario for comparison.
First, under the Toxicology menu choose Exposure and then TREX. Once the T-REX page
opens, set Pesticide applications to 4 and the first application to occur on 20 May, with
subsequent applications separated by 7 days. Finally, set all application rates to 0.03 lbs Al/acre,
so that the T-REX page looks like Fig. 2.5, below. MCnest uses algorithms from T-REX
(USEPA 2012d) to translate the application rate (0.03 pounds Al/acre) into doses (expressed as
mg/kg body wt./day) for both adult and juveniles, taking into account the species' typical diet.
This produces the values for "Adult Initial Dose" (1.53 mg Al/kg body weight) and "Juvenile
Initial Dose" (2.28 mg Al/kg bodyweight) that appear in the two associated boxes. These initial
doses will be realized on the application date (May 20) and will decay geometrically with a
residue half-life of 35 days. Click "Done" to close this window and return to the main MCnest
window and set the Exposure Algorithm to T-REX, then click Run. When finished, the main
window should resemble Figure 2.6 (though again, results may differ slightly due to stochastic
variation).
13
-------
User's Manual for MCnest v2.0 - September 2017
Figure 2.5. T-REX window parameterized for A-Cyhalothrin
•Aj MCnest Version 2.0: TREX
Done Reset c
Pesticide applications:
Croc:
Dose calculations
O Set Dose Directly
Half life: 35
Nomogram values:
O Use mean nomogram values
(•) Use maximum nomogram values
O Draw from the nomogram distribution
max
mean
std
%H20 Juv
FIR/BW
short grass
240
85
60.3000
0.8000
1.0800
tall grass
110
36
40.6000
0.8000
1.0800
broad leaf forage
135
45
56.7000
0.8000
1.0800
fruit
15
7
12.4000
08000
1.0800
seeds
15
7
12.4000
0.1000
0.2400
invertebrates
94
65
48
0.8000
1.0800
Reset to default values
1 Adult Nestling |
Apr-02
Apr-30 May-28 Jun-25 Jul-23
Date
Aug-20
Application 1
Adult Initial Dose:
Application 2
Adult Initial Dose:
Application 3
Adult Initial Dose:
Application 4
Adult Initial Dose:
1.5321
0 03 I
1 5321 mg/kg/d
0.03 lb a.i./A
1.5321 mg/kg/d
0-03 lb a.i./A
1.5321 mg/kg/d
003 IbaiJA
Breeding Season (edit on Life History page)
Date of first egg of first nest
Month" April Dav:j-|2
Date of first egg of last nest
Month: July
Dav:l22
Waiting periods:
Waiting period after
pesticide-induced failure:
Waiting period after failure:
Waiting period after fledging:
days
rtav<;
days
Juvenile Initial Dose: 2 2777 mg/kg/d
Month: May v [ Day:: 20. *
Juvenile Initial Dose: 2.2777 mg/kg/d
Month:|May vjl Day|27 j?
Juvenile Initial Dose: 2.2777 mg/kg/d
Month: June v Day: 3 v
Juvenile Initial Dose: 2.2777 mg/kg/d
Month: June v Day: 10 Sv
Toxicity Thresholds (mg/kg/d):
1/10 Adult LD50
Adult body weight
Eggs laid per hen
Eggshell thickness
Viable eggs per egg set
Hatchlings per viable egg
14-d chicks per hatchling
10.0792
10.0792
fraction of juvenile LC50: 101195
3381.27591
10.0792 |
0.94334 I
10.0792
10.0792 I
Figure 2.6. The main MCnest window after running an American robin simulation with exposure
to X-Cyhalothrin.
MCnest Version 2.0
MCnest Toxicology Clear
Exposure Algorithm
ITREX
Run Status
Run time: 0 m 7 j
MCnest working directory:
Simulation iterations
Number of birds:
10000
Flock size.
25
Species
Species: American robin
(•) Broods/Female Q Fledglings/Female
ason (edit on Life History page)
Date of first egg of first nest
Date of first egg of last nest
Month: July
This is MCnest v2.0.
It was compiled on 1-Aug-2017.
SPECIES
American robin
American robin
C:\Users\etter\Documents\TlM_MCnest\TIM_MCnest_2014\Current\MCnest17\GUI
BROODS B L95 B U95 APPS APP1 RATE 1 crop pesticide Exposure Mortality
1.84 1.48 2.2 On/a n/a unset unset CONTROL 0
0.73 0.56 0.88 420-May 0.0300 n/a Lambda C. TREX 0
14
-------
User's Manual for MCnest v2.0 - September 2017
Next, choose Toxicology > Exposure > TIM to bring up the TIM window. On the TIM
window, choose File > Load TIM Parameters and navigate to the folder where the sample data
are located. Choose the file LambdaCyhalothrinTIMParameters.mat and the TIM window
should look like Figure 2.7. Click Close. This will cause an error message to display (Fig. 2.8).
The error message is displayed to warn you that you have loaded a mismatched data set. The
parameter set that you have just loaded was created with Canada Goose as the species. To run
the American Robin simulation, you will need to first set the species to Canada Goose (giant)
and then load the TIM parameters. Once loaded, you can close the TIM window and switch
species back to American Robin. This bug will be fixed in a future version. 15 Sept. 2017.
Figure 2.7. The TIM window parameterized for Lambda-Cyhalothrin.
53 TIM Version 3.0
File
Scenario Definition
Pesticide:
Crop:
Species:
Lambda Cyhalothrin
Number of birds:
Flock size:
Random seed:
American robin
10000
Exposure Routes:
0 dietary
0 drinking water (puddles)
0 drinking water (dew)
0 inhalation (vapor)
0 inhalation (spray)
0 dermal (contact)
0 dermal (spray)
0 spray drift
Droplet spectrum
very fine to fine
Crop Type
[field
Applications
Pesticide Applications:
lbs a
i.iJA
Month:
Day:
Interval:
Application 1:
0.03
May
v 120
Application 2:
0.03
May
,27
j T j
Application 3:
0.03
June
7
Application 4:
0.03
| June
|10
7
Fraction of edge habitat receiving spray drift:
Length of in-field buffer (feet):
Fraction of organic carbon in soit
Soil bulk density (kg/L)
Hour of first application: 8 am
Spray height (m):
Spray duration (min):
Crop height (m):
Crop mass (kg/ha):
Figure 2.8. Warning dialog concerning mismatched species on parameter import for TIM
*'¦ Warning Dialog
~
X
TIM species (Canada goose (giant)) does not match MCnest species (American
robin). TIM will revert to previous parameter set.
Please change MCnest species to Canada goose (giant) and reimport the
desired TIM parameter.
OK
Once you have gone through this process and closed the TIM window without an error, change
Exposure Algorithm on the main MCnest window to TIM and click Run. This will call up a TIM
window, that will immediately minimize. While TIM is running you can watch the TIM spool by
15
-------
User's Manual for MCnest v2.0 - September 2017
clicking on the minimized TIM icon (Fig. 2.9). While TIM is running, MCnest is paused, waiting
for the TIM exposure output. When TIM finishes, the TIM window will close automatically and
MCnest will start running, conditional on the TIM output. Finally, once the MCnest simulation is
complete, the main MCnest window should resemble Figure 2.10.
Figure 2.9. The TIM spool for an American Robin simulation with A-Cyhalothrin
C:\Users\etter\Documents\TIM_MCnest\TIM_MCnest_2014\Current\MCnest17\GUI\TIM3.0_Beta_3-18-15b.exe
On
On
On
On
On
On
On
On
Food Ingestion 1
Puddle 1
Dew 1
Inhalation: Vapor 1
Inhalation: Spray 1
Dermal Contact 1
Dermal Spray 1
Spray Drift 1
Total
Population
Bird
Dead
Hour
Tolerance
Retained Dose
Risk
250
0
2256
71228.869
67.082
0.00000
509
0
2256
44022.556
32.163
0.00000
750
0
2256
47250.294
33.703
0.00000
1000
0
2256
71072.769
52.482
0.00000
1250
0
2256
76210.846
90.675
0.00000
1500
0
2256
43902.472
67.589
0.00000
1750
0
2256
47969.791
91.854
0.00000
2000
0
2256
73004.489
79.494
0.00000
2250
0
2256
81643.381
117.676
0.00000
2500
0
2256
74889.817
60.251
0.00000
2750
0
2256
71648.051
78.531
0.00000
3000
0
2256
69112.458
32.027
0.00000
3250
0
2256
79376.322
86.125
0.00000
3500
0
2256
83881.613
104.344
0.00000
3750
0
2256
40071.580
70.048
0.00000_
16
-------
User's Manual for MCnest v2.0 - September 2017
Figure 2.10. Main MCnest window displaying the results of three simulations
¦#i MCnest Version 2.0
MCnest Toxicology Clear
I Exposure Algorithm:
Simulation iterations
Number of birds: 10000
Flock size: 25
Run Status
TIM run time: 0 m 36 s.
MCnest run time: 0 m 29 s
MCnest working directory:
C:\Users\etter\Documents\TIM_MCnest\TIM_MCnest_2014\Current\MCnest17\GUI
Species: American robin
(•) Broods/Female Q Fledglings/Female
SPECIES
American robin
American robin
American robin
Season (edit on Life History page)
Date of first egg of first nest —
Date of first egg of last nest
Month: | July
This is MCnest v2.0.
It was compiled on 1-Aug-2017.
BROODS B L95 B U95 APPS APP1 RATE 1 crop
1.84 1.48 2.2 On/a n/a unset
0.73 0.56 0.88 4 20-May 0.0300 n/a
0.67 0 48 0.88 420-May 0.0300 Com
pesticide Exposure Mortality
unset CONTROL 0
Lambda C... TREX 0
Lambda C... TIM 0
A quick glance at the Output table suggests that the model predicts an approximately 60%
reduction in the expected number of broods per female due to pesticide exposure compared to
the expected number in the absence of pesticide effects when T-REX is used as the exposure
algorithm. The expected reduction is a bit higher (64%) when TIM is used for exposure, due in
part to the incorporation of other exposure pathways in TIM (inhalation, dermal contact, drinking
water) compared to T-REX, which models only dietary exposure. However, percent reduction
comparisons may be insufficient for understanding potential risks of a pesticide application. For
this, MCnest offers many kinds of output that are stored and accessed in several ways. These are
covered in detail in the next Section III (MCnest outputs).
This is the end of the Quick Start Guide. Congratulations! You now know how to run
elementary simulations in MCnest. However, we recommend you continue through the next
section to see how to extract more information from MCnest simulations. The above three
simulations will be used as examples throughout Section III to demonstrate the diagnostic and
output features of MCnest.
III. MCnest outputs
A. Using the table
The Output table on the main MCnest screen provides a snapshot of the overall results for each
simulation as well as functional access to more detailed information about MCnest results.
Because each MCnest simulation is separate, MCnest does not automatically calculate the
percent reduction in seasonal productivity due to a pesticide exposure relative to a no-pesticide
17
-------
User's Manual for MCnest v2.0 - September 2017
scenario. However, the contents of the Output table can be copied into a spreadsheet, such as
Excel, to calculate comparisons among simulations using the standard windows commands (Ctrl-
C and Ctrl-V).
Additional output functions can be accessed by selecting any cell in the desired output row and
then right-clicking the mouse within the outlined results block. Most of the output is accessed
under the Output submenu; the remaining menu items are for sorting, saving or deleting the
output table. The examples below make use of the three simulations run above in Section II
(Quick Start Guide). Throughout this section on model outputs it is assumed that you have three
similar results available.
B. The Output menu
After selecting the first row, right-click and place the cursor on Output in the results menu to get
access to a sub-menu with 4 choices (Fig. 3.1). When T-REX is run, this menu has 5 choices,
with the new third menu option being Exposure series, whereas when TIM is run, the new third
menu option is TIM (which has, in turn five sub-choices). Below, each submenu choice is
described in turn, with the exception of the TIM output options, which are described in the TIM
User's Guide.
18
-------
User's Manual for MCnest v2.0 - September 2017
Figure 3.1. The Output submenu of the context menu for a. Control simulation, b. a simulation
run using T-REX, and c. a simulation run using TIM.
UCamc wortug drcctofy
A. Control
W '.fCPA'W'Cn«MW,WC n«N/Crm»K:oKfi"T*)^WCne»W4Cf>ertCo
-------
User's Manual for MCnest v2.0 - September 2017
Figure 3.2. Phase diagram for American robin with no pesticide exposure.
PF [MA rfg/ef I M lef I li I IN I llld I BIWp I IWe I IWfl |Q I iDe I I Dp
03
E
<1>
ll
° 0.5
c
o
o
Cl
g
Q_
0
Apr-02 Apr-16 Apr-30 May-14 May-28 Jun-11 Jun-25 Jul-09 Jul-23 Aug-06 Aug-20
Date
The phase diagram is the primary diagnostic tool for understanding how the simulated chemical
exposure was translated by the model into projected reproductive effects. Because there is no
chemical exposure in the example simulation, there are no corresponding diagnostic effects in
the phase diagram. Nevertheless the diagram is useful for tracing the general progress of females
unexposed to pesticides through various breeding phases over the course of the simulated season.
Table 3 provides a more detailed explanation of the phases in the phase diagram. Each of the
phases is defined in greater detail in the technical support document.
Notice that three phases in Table 3.1 do not occur in the phase diagram pertaining to the first
simulation above (Fig. 3.2). These are the doomed incubation period (Id: greenish blue), the
waiting period after pesticide-induced failure (Wp: green), and the phase corresponding to killed
by pesticide (Dp: yellow). These three states can only be entered into after exposure to a
pesticide exceeding the corresponding threshold. However, highlighting the second row in the
Output table and opening the phase diagram will produce a figure (Fig. 3.3) that shows a portion
of females in the waiting period after pesticide induced failure (Wp).
20
-------
User's Manual for MCnest v2.0 - September 2017
Table 3.1. Nest development phases in MCnest
Phase
Legend
abbreviation
Legend color
(approx.)
Pair Formation PF
Rapid Follicle Growth rfg
Overlap of Rapid Follicle Growth and Egg Formation rfg/ef
Egg Formation ef
Incubation I
Nestling N
Doomed Incubation Id
Waiting after pesticide-induced failure Wp
Waiting after ecological failure We
Waiting after successful fledging Wf
Quit Q
Adult female dead due to background causes De
Adult female dead due to pesticide exposure Dp
Hint: you do not need to close the phase diagram for a previous simulation before generating one
for the next. Thus for example, the two phase diagrams could be open concurrently for direct
comparison.
Further guidance on the interpretation of phase diagrams can be found in Appendix A of Bennett
& Etterson 2013c.
Figure 3.3. Phase diagram for an American Robin simulation with X-Cyhalothrin using T-REX
PF rfg/ef I of II lid IWp I We I Wf I lo I Da Dp
Apr-02 Apr-16 Apr-30 May-14 May-28 Jun-11 Jun-25 Jul-09 Jul-23 Aug-06 Aug-20
Date
21
-------
User's Manual for MCnest v2.0 - September 2017
2. Brood histogram
This sub-menu item generates a histogram of the number of successful broods produced per
female in the simulation (Fig. 3.4). It is useful for gaining an intuitive sense for the expected
variance in female reproductive success associated with a simulation scenario.
Figure 3.4. Sample brood histogram for default American robins with no pesticides.
1 ' ' ' '
Number of Broods
3. Exposure series
Figure 3.5 shows the simulated Seasonal Exposure Profile for the second American robin
simulation, in which the threshold (i.e., 0.94 mg/kg/d, per the Test Results input screen; Fig. 2.4)
for number of eggs laid was exceeded immediately on application date and for the remainder of
the breeding season. Note the separate exposure series for adults versus juveniles. This menu
choice is not available with control simulations (no pesticide). When TIM is used, this menu
choice is replaced by the standard TIM outputs (see below and USEPA 2015).
Figure 3.5. Example Seasonal Exposure Profile for American robins with the application dates
and rates set as described in the pesticide examples of Section II (Quick Start Guide).
Adult Nestling
T5
O)
8
Apr-30
May-28
Jun-25
Jul-23
Aug-20
Date
4. TIM
When TIM is used as the exposure algorithm, the Exposure series choice is replaced by TIM,
which has 5 submenu choices (Fig. 3.6). These outputs are described in the TIM technical
document, Appendix A, (https://www.epa.gov/pesticide-science-and-assessing-pesticide-
risks/models-pesticide-risk-assessment#terrestrial).
22
-------
User's Manual for MCnest v2.0 - September 2017
Figure 3.6. The TIM output menu
MCnest Version 2.0
MCnest Toxicology Clear
Exposure Algorithm:
Simulation iterations
Number of birds: 10000
Flock size: 25
Run Status
Ready
MCnest working directoiy:
C:\Users\etter\Documents\TIM_MCnest\TIM_MCnest_2014\Current\MCnest17\GUI
Species: Attwater's Prairie Chicken
SPECIES
American robin
American robin
(•) Broods/Female
Season (edit on Life History page)
Date of first egg of first nest
O Fledglings/Female
Date of first egg of last nest
Month: April
Day: 15
This is MCnest v2.0.
It was compiled on 1-Aug-2017.
Output :
Sort table
Save table
Delete rows
O Type here to search
o o 0
BROODS B L95
1.84 1.48
0.73 0.56
"0.67 0 48
Phase diagram
Brood histogram
TIM >
Log file
Save row
B U95 APPS APP1
2.2 On/a
0.88 420-May
088 420-May
ATE 1 crop
la unset
0.0300 n/a
0.0300 Com
pesticide Exposure Mortality
unset CONTROL 0
Lambda C...TREX 0
Lambda C... TIM 0
Mortality distribution
Dose fractions
TIM input
Dead per hour (file)
Dead per hour (plot)
% w!
I I 5:20 PM _
O ^ H 7/23/2017 ^
5. Log file
This sub-menu choice opens a text file (using the computer's default text editor) and prints the
main results and all of the input parameters used in the simulation (Fig. 3.7). This file is virtual
(it can be regenerated at any time, but it does not exist until generated). Therefore, if it is needed
for documentation, it should be explicitly generated and saved or printed.
23
-------
User's Manual for MCnest v2.0 - September 2017
Figure 3.7. Sample log file for American robin simulation with /.-Cyhalothrin exposure and
using T-REX
2 Log - Notepad
File Edit Format View Help
********Simulation Tags**
Species = American robin
********Summary output***********
Avg. no. fledglings per female = 2.0385; 95 percent CI: [1.57-2.46]
Avg. no. successful broods per female = 0.72802; 95 percent CI: [0.56-0.
Avg. no. nest attempts per female = 0.72802; 95 percent CI: [0.56-0.88]
Avg. nest success = 1; 95 percent CI: [0.36-0.64]
********Simulation Parameters***********
Flock size: 25
Total number of females simulated: 10000
Species name: American robin (detailed life-history below)
Earliest first egg date: April 12.
Latest first egg date: Duly 22.
********CPU Time***********
Simulation initiated at: 2017.Dul.23.11.46.22 AM
Simulation ended at: 2017.Jul.23.11.46.30 AM
Elapsed time: 0 minutes and 7 seconds.
********End of season***********
Method used for modeling end of season: Regular MC.
********Life History***********
Daily probability of leaving pair-formation: 0.25.
Daily probability of nest failure during laying and incubation: 0
Daily probability of nest failure during nestling phase: 0.023.
Length of rapid follicle growth period : 4 days.
Egg-laying interval: 1 days.
Length of incubation phase: 13 days.
Length of nestling phase: 13 days.
Length of doomed incubation phase: 13 days.
Length of waiting period after pesticide-induced failure: 7 days.
O Type here to search
¦ I® a
6. Save row
This sub-menu choice saves the simulation from the selected row to an output file that can later
be opened in MCnest. The file is assigned a name by the program corresponding to the species
name and the date and time that the simulation was run. A confirmation pop-up window is
displayed (Fig. 3.8). Files are saved into the directory specified in the "MCnest working
directory" listed above the ''Output table."
Figure 3.8. Example filename and popup window when a single row is saved to an output file.
Results saved to American robin.2017.Ju .23.11.46.22 AM.
C. Sort table
This menu choice provides functions to sort the displayed Output table using any of the table
headers as sort keys (Fig. 3.9).
24
-------
User's Manual for MCnest v2.0 - September 2017
Figure 3.9. Example of the "Sort" window for sor
A
BROODS
BL95
BU95
APPS
APP 1
RATE 1
crop
pesticide
Exposure
Mortality
V
(•) sort ascending
(•) sort ascending
(•) sort ascending
;ing simulations in the Output table.
I). Save table
This sub-menu choice allows the entire output table to be saved to disk. For this option you are
prompted to provide a name for the table. You are responsible for changing the default name of
"*.mat" to a unique name with a ".mat" extension. It can later be reopened in MCnest, at which
point it would be appended to the bottom of any existing simulations in the Output table.
E. Delete rows
This sub-menu choice deletes the selected row(s) from the table. Multiple rows can be selected
and all selected rows will be deleted. They cannot be recovered.
IV. Systematic description of MCnest controls
The descriptions below contain minimal background and justification for the design of MCnest.
For more details about the algorithms underlying MCnest and the technical justification for
programming choices, please see the MCnest Technical Manual and scientific publications.
MCnest has three basic categories of user controls, drop-down menus, window controls, and the
table context menus. Not all windows contain all types of controls. Most MCnest windows
include various controls that will be familiar to users of windows driven programs, including
push buttons, radio buttons, check boxes, list boxes, text boxes, menus, and context menus. The
presentation below is organized by window, beginning with the main MCnest window.
A. MCnest window
MCnest opens to its main window upon launching and most of the controls are accessed through
this window. The main menu bar contains three choices (MCnest, Toxicology, and Clear). The
MCnest window also contains several user-input controls.
25
-------
User's Manual for MCnest v2.0 - September 2017
1. User-input Controls
Run: The Run button initiates a simulation.
Stop! The Stop button terminates a simulation currently in progress.
Simulation Iterations: This form contains two user-input boxes labeled "Number of birds" and
"Flock size," respectively. These two inputs determine the scale of the simulation and degree of
replication for output statistics. The default values are 10,000 birds with a flock size of 25,
resulting in 400 replicates of 25 birds. This replication is what determines the width of the
confidence limits around model outputs (broods per female, offspring fledged per female).
Differences among outcomes for replicate females arises only from stochasticity induced by the
probabilities of nest survival and female survival.
Species: The species block contains a listbox menu of species names for which life-history
profiles have been pre-coded for use in MCnest. For additional information about the species
and their profiles, see the companion document on Avian Life History Profiles for Use in the
Markov Chain Nest Productivity Model (Bennett & Etterson 2013a). The default is the first
species in the list, i.e., Canada Goose. Specific details about the life history of species in the
library, and functions to edit life-history parameters to create new species are found on the Life
History window, which can be accessed under the main MCnest menu and is described in
Section IV.B (Life History window) below.
Note: all life-history parameters are fixed. Experimental work is under way to explore
systematic and stochastic variation in life history parameters, which may be included in future
versions of the model.
Output: Once simulations have been run, this box allows the user to toggle between the number
of successful broods/female/year and the number of fledglings/female/year in the Output table.
Results for both metrics are captured in the Log file.
Season (edit on Life History page): this is simply a display of the breeding phenology of the
species chosen in the Species listbox. It can be edited on the Life History window, which can be
accessed under the MCnest menu.
About MCnest: status box giving the version and date of compilation of the model. This control
is for display only.
Run Status: status box showing the state of the model and progression through the simulation
when a simulation is running.
MCnest working directory: status box showing the current working directory for MCnest. This is
the directory into which MCnest will save results or initially look for files when you attempt to
load data (you can always navigate to another directory). Clicking in the text box and hitting
return will bring up a browser to change the working directory.
Output table: This frame contains the outputs of individual runs of MCnest. When no previous
runs are available it is empty.
26
-------
User's Manual for MCnest v2.0 - September 2017
In Figure 4.1. the Output table shows the results for the three simulations run in Section II with
the first row highlighted. The first column gives the species name (American robin). The
second column gives the global average (across all 400 flocks) of the number of successful
broods produced per female (1.84). The third and fourth columns give the lower and upper
confidence limit for a 95% confidence interval (1.48 - 2.2) constructed around the global
average number of successful broods per female using the 400 replicates of 25 birds.
Figure 4.1. Main MCnest window showing output of two simulations from Section II (Quick
Start Guide).
MCnest Version 2.0
MCnest Toxicology Clear
Run Stop!
Exposure Algorithm:
Simulation iterations
Number of birds: 10000 Flock size: 25
Run Status
Ready
MCnest working directory:
C:\Users\etter\Documents\TIM_MCnest\TlM_MCnest_2014\Currerit\MCnest17\GUI
Species: Canada goose (giant)
American robin
American robin
(•) Broods/Female
Season (edit on Life History page)
Date of first egg of first nest
Date of first egg of last nest
Month: April
O Fledglings/Female
This is MCnest v2.0.
It was compiled on 1-Aug-2017
BROODS B L95 B U95 APPS APP1 RATE 1 crop pesticide Exposure Mortality
11.84 1.48 2.2 On/a n/a unset unset CONTROL 0
0.73 0.56 0.88 4 20-May 0.0300 n/a Lambda C... TREX 0
0.67 0.48 0.88 4 20-May 0 0300 Com Lambda C... TIM 0
O Type here to search
2. The MCnest menu
Of the three drop-down menus on the main MCnest window, the first two (MCnest and
Toxicology) are the most important, giving access to virtually all of the functionality of MCnest.
The MCnest menu contains 7 choices (Table 4.1).
27
-------
User's Manual for MCnest v2.0 - September 2017
Table 4.1. Menu choices under the main MCnest menu.
Menu Item
Function
Load Species Library
Life History
Batch Mode
Random numbers
Load MCnest File
Import time varying parameters
Exit
Loads a different species library into memory for use in simulations
Loads the life history window for editing species life history values
Allows you to set up multiple simulations to run sequentially along
systematic changes in species identity, application date, and/or
application rate
Allows you to set the random seed and choose among random
number generators
Loads previously saved MCnest results into memory for examination
Allows you to explore the importance of systematic changes in the
value of a parameter over the course of a breeding season
Closes the model
Load Species Library: Although MCnest opens up to the default species profile library, the
menu option allows you to select other species profile libraries, if they exist. Custom species
libraries may be useful for 1) tailoring species profiles to a geographic region of interest; 2)
exploring known variation in model parameters associated with different habitat types, climates,
or background stressors; or 3) sensitivity analysis. Custom species libraries must use the same
structure as the default library provided with MCnest. Further details on the species library and
structure can be found in (Bennett & Etterson 2013a).
Life History: Choosing this menu option loads the Life History window. Controls for the Life
History page are described in Sub-section B (Life History window), below.
Batch Mode: Choosing this menu option loads the Batch mode page. Controls for the Batch
mode page are described in Sub-section D (Batch window), below.
Random Numbers: Choosing this menu option loads the Random Numbers window. Controls
for the Random Numbers page are described in Sub-section E (Random Numbers window),
below.
Load MCnest file: Choosing this menu option loads the results of previous simulations into the
output table on the main MCnest window. Previous results will be appended to the end of any
currently displayed results.
Import time-varying parameters: Choosing this option loads the time-varying parameters page.
NOTE - this feature is still under construction. It currently has limited features available. See
section E.3. below.
Exit: Choosing this option exits MCnest. MCnest can also be closed by clicking the upper right
hand corner of the main MCnest window.
28
-------
User's Manual for MCnest v2.0 - September 2017
3. The Toxicology menu
The toxicology menu has two choices, Test Results and Exposure, which in turn has two
choices, TREX and TIM. Choosing Test Results brings up the Test Results window, which is
covered in detail below in section IV.C. Choosing Exposure > TREX brings up the TREX
page, which is covered in detail below in section IV.D. Finally, choosing Exposure > TIM
brings up the TIM window, which is covered in detail below in the TIM technical
documentation.
4. The Clear menu
The Clear menu is used to remove existing results. It has only three choices.
Summary Table: Choosing this menu option clears the summary table from memory and from
display. It cannot be recovered unless the data were first saved to a file.
Figures: Choosing this menu option closes all open figures. They can be regenerated by
repeating the appropriate commands.
All: Choosing this menu option clears the summary table and closes all open figures. Neither the
table nor figures can be recovered unless the data were first saved to file.
B. The Life History window
The life history window is invoked by selecting the Life History option under the MCnest
menu.
The Life History window has five control frames and a button for returning to the main MCnest
window (Fig. 4.2). The species library contains default parameters for all of the controls on the
Life History page. Thus, if you are using default species you need not use this window. If you
invoke this window with a default species selected, it will display the default parameters for that
species. This can be useful at runtime to see what those values are. However, changing any
parameter on this page will automatically change the default species to "-new-." It will also
make visible a new box on the main MCnest window for naming the new species. The default
name for the new species is "Bird." Mandatory renaming will be removed in a future version of
the model.
29
-------
User's Manual for MCnest v2.0 - September 2017
Figure 4.2. The Life History window with default data for American robin.
MCnest Version 2.0: LifeHistory
Season
Date of first egg of first nest
Day:
Month:
April
12
Date of first egg of last nest
Day:
Month:
July
22
Phase Durations (days)
Rapid Follicle Growth
Egg-laying Interval:
Incubation Period:
Nestling Period:
Waiting period after failure:
Waiting period after fledging:
13
13
Return to Main Screen
Transition Probabilities
Other Life History
Clutch Size:
Incubation Onset
O Penultimate Egg
® Last Egg
Fledglings per successful nest:
2.S
Diet & body weight (exposure)
0.025
0.25
0.023
Body Weight (g):
Short grass:
Tall grass:
Broadleaf forage plants:
Fruit:
Seeds:
Invertebrates:
Sums (must be 1):
~
X
77.3
Adults
Nestlings
0
0
0
0
0
0
0.28
0.3
0
0
0.72
0.7
1
1
Season: This frame allows you to change the length of the breeding season, which is defined in
MCnest by two dates: the date of the first egg in the first and last nests for the season of each
species. Implicit in this definition is the ability of females to complete any nesting attempts
initiated by this later date, assuming the nest is not lost due to ecological or pesticide-induced
failure.
The end date (date of first egg in last nest) must be at least two weeks later than the beginning
date (date of first egg in first nest). If you try to set a breeding season shorter than two weeks,
you will get an error and the end date will be automatically set to two weeks after the beginning
date.
Phase Durations (days): This frame contains eight text boxes for entering the phase durations.
These durations determine the time required to complete various processes in the nesting cycle
and are described in greater detail in the MCnest Technical Manual. With the exception of the
rapid follicle growth period and the egg-laying interval, these parameters must be specified as
integers.
Transition Probabilities: This frame contains three text boxes for model parameters. The first, p,
is the daily probability that a female in the pair formation stage initiates rapid follicle growth for
her first egg of her first nest of the season. This parameter applies only to a female's first nest
attempt of the year. The second two parameters are the daily probabilities of background nest
failure in egg-laying and incubation (pre-hatch) and nestling rearing (post-hatch) periods of the
nest cycle. These failure rates should include all causes of failure except for failures arising from
the pesticide algorithms in MCnest. Pesticide failures are handled separately, as described below
in Sub-section C and in the MCnest Technical Manual (Etterson & Bennett 2017). The final
30
-------
User's Manual for MCnest v2.0 - September 2017
variable in this frame is the daily background rate of adult female survival (again excluding
pesticide mortality). All parameters in this frame are restricted to the interval (0,1).
Other Life History: This frame contains two text boxes and two radio buttons. The first text box
is for average clutch size, the average number of eggs laid in a nest. It must be an integer. The
two radio buttons indicate whether the species generally begins incubation with the penultimate
versus the final egg. Whichever button is selected; all females are assumed to follow the same
pattern. The second text box is for an estimate of the mean number of fledglings per successful
brood. It does not need to be an integer. Clutch size should serve as an upper bound to this
parameter.
Diet & body weight (exposure): You do not need to use this frame unless you are simulating
pesticide effects. Dietary information is only used to estimate dose, based on residue analyses of
different food types. Body weight must be provided in grams. Dietary proportions must be
between 0 and 1 for all categories and they must sum to 1 (down columns). Specifying
proportions outside the range (0,1) will generate an immediate warning and that category will be
set to 1/6 ~ 0.17. However, specifying sums of dietary categories that do not sum to 1 will not
generate errors until you attempt to click Return to Main Screen.
C. The Test Results window
The Test Results window (Fig. 4.3) is invoked by selecting the Test Results option under the
Toxicology menu. The Test Results window has several large frames pertaining to three
different toxicity tests, the Acute Oral Toxicity test (USEPA 2012a), the Dietary Toxicity Test
(USEPA 2012b), and the Avian Reproduction Test (USEPA 2012c) as well as a frame for
surrogate endpoints calculated from these tests results. There are also input boxes at the top of
the window for typing the name of the pesticide and its degradation half-life on dietary items.
The USEPA/OPP default half-life of 35 days will automatically be displayed when opening the
window, though an empirical value may be substituted, if known. Finally, the row of four
buttons at the top of the window provide additional functionality. In particular, the Load data
and Save data functions allow you to load test result data from a previously saved file or to save
the data (once the window is completely filled out) for easy use in the future.
31
-------
User's Manual for MCnest v2.0 - September 2017
Figure 4.3. The Test Results window
MCnest Version 2,0: Test Results
Half-life on dietary items:
35 days
Avian Reproduction Test
Avian reproduction test source (MRID):
Test species:
Treatment group data:
Treatment levels: 1
Measured concentration (mg/kg diet): NaN
Average food consumption (g/bird/day): j NaN
Average initial female body weight (g): NaN
Average initial male body weight (g): NaN
Average final female body weight (g): NaN
Average final male body weight (g): NaN
Estimated average daily
dose (mg/kg bwt/day): I
Endpoint
NOAEL
Endpoint
NOAEL
Number of eggs laid:
-
% viable eggs of eggs set: >
% live 3-wk embryos of viable eggs: >
% hatchlings of live 3-wk embyros: >
% hatchlings of eggs set: >
% 14-d chicks of hatchlings: >
% 14-d chicks of eggs set:
shell thickness:
hatch weight:
14-d chick wt:
prelaying female weight:
prelaying male weight:
3
Avian Acute Oral Toxicity Test:
LD50 test source (MRID):
Test species:
Mineau scaling factor:
LD50 (mg/kg bwt):
Body weight:
Scaled LDSO:
Avian Dietary Toxicity Test:
LC50 source (MRID):
Test species:
LC50:
Fraction of LC50 (i.e., 0.1 or
Mean body weight (g): NaN
Fraction of
LC5Q as
dose
(mg/kg bwt):
Mean food ingestion rate (g/d): NaN
NaN
~1
MCnest Toxicity Thresholds (mg/kg/d):
Alternative behavioral threshold: 9999 1/10
9999
Adult prelaying body weight:
9999
Eggs laid per hen:
9999
Mean eggshell thickness per pen:
9999
Viable eggs set per pen:
9999
Hatchlings per viable egg per pen:
9999 |
14-d chicks per hatch ling per pen:
9999
Fraction of juvenile 5-d LC50
9999
1. Avian Reproduction Test
The Avian Reproduction Test Frame occupies the full left-hand side of the Test Results
window. Within the frame, there are two smaller frames, the first for entering treatment-group
data, and the second for entering the NOAELs from the analyzed test results reported in the
USEPA/OPP Data Evaluation Record (DER). There are also two text boxes at the top of the
frame for entering USEPA/OPP Master Record Identification number (MRID) as well as the
species tested.
In the treatment-group data panel, there are seven rows of input boxes for entering treatment-
level data (except control). They should be self-explanatory and are typically reported in
USEPA/OPP DERs. The first row is for measured concentrations in the feed. The second row is
for entering average food consumption of tested birds, by treatment. The third and fourth rows
are for entering the average initial and weights of males and females, and the fifth and sixth rows
are for entering the average final weights of males and females respectively. The food
consumption and body weight data are used to convert the measured concentrations (mg/kg diet,
row 1) to doses (mg Al/kg body weight/day) displayed in row 7, which cannot be directly edited.
The equation used for calculating daily dose is:
32
-------
User's Manual for MCnest v2.0 - September 2017
Daily dose = Dietary concentration (mg Al/kg diet) x food ingestion rate (g food/day)
Body weight (g).
The conversion of dietary concentrations to daily doses is an approximation because body weight
and food ingestion rates are changing during the course of both the reproduction test and the
LC50 test. Also, studies vary in the degree to which they quantify food spillage during the tests.
For converting dietary concentrations to daily doses, the model user can calculate the average
daily food ingestion rate per bird and the average body weight at the beginning and end of the
test period for each bird. The model user should consult the MCnest Technical Manual for more
background information about the selection of toxicity threshold values prior to use.
In the Test NOAELs panel, there are 12 drop-down boxes for test endpoints from the Avian
Reproduction Test. However, only eight of these are currently used by the model and the
remaining four (%hatchlings of eggs set, %14-d chicks of eggs set, hatch wt, and chick wt) can
be ignored. They may be used by a future version of the model. Each of the remaining eight
NOAELs should be set to the highest level (dietary concentration, row 1 in treatment-group
panel) at which that endpoint did not differ from the control. If no statistical difference from
control was observed in any of the tested dietary concentrations, for a given endpoint, the drop-
down box can be set to '>' to indicate that it is an unbounded NOAEL.
2. Avian Acute Oral Toxicity Test
The acute oral toxicity test frame occupies the upper right of the Test Results window. It has six
input boxes, only five of which are editable. The first two are for the tested species and
referencing the source of the test (MRID). The next two boxes are for rescaling the LD50 to
account for differences in body weight between the tested species and the simulated species
(Mineau et al. 1996). The default value of 1.15 will automatically be displayed when opening the
window. If no scaling is desired, then change this value to 1. Finally, the last editable input box
is for entering the estimated LD50 from the test, which is then converted internally to the scaled
LD50.
3. Avian Dietary Toxicity Test
The dietary toxicity test frame occupies the middle right of the Test Results window. It has
seven input boxes, only six of which are editable. The first two are for the tested species and
referencing the source of the test (MRID). The next box is for entering the estimated LC50 from
the test, followed by the fraction of the LC50 desired for use as a toxicity threshold (default is
0.5, Bennett and Etterson 2013b). The last two are mean body weight and food consumption
rates of tested animals for converting from dietary concentration (mg/kg diet/day) to dose (mg/kg
body weight/day).
4. Toxicity Thresholds
This control frame contains nine text boxes for the toxicity thresholds (values which, if exceeded
by exposure, will trigger nest failure or prevent a bird from initiating a new nest). These values
are calculated from the data entered in the three test results panes and are expressed as daily dose
(mg Al/kg body wt/day). The calculated surrogate endpoints are not directly editable, except for
an auxiliary input box for an alternative behavioral threshold, which can be used if there were
behavioral effects noted at one or more tested dietary concentrations or doses of one of the three
33
-------
User's Manual for MCnest v2.0 - September 2017
tests. If used, this alternative threshold applies only to those phases for which the default
surrogate endpoint for behavior (1/10 LD50) is used (Bennett & Etterson 2006, Bennett &
Etterson 2013b,d). In practice, the model uses the lower of the two values as the surrogate
endpoint for behavior.
D. The TREX window
Figure 4.4. The default Pesticides^window when selecting five pesticide applications^
* MCnest Version 2.0: TREX
Done | [Reset defaults j
P^EHyaJ
Pesticide applications:
Crop:
Application 1
Adult Initial Dose: 0 mg/kg/d Juvenile Initial Dose: 0 mg/kg/d
Application rate: | 0 | to a.i./A Month: January ~] Day:M *1
Residue calculations
© Set Dose Directly
Half life: | 35 days.
Nomogram values:
Use mean nomogram values
o Use maximum nomogram values
Draw from the nomogram distribution
max
mean
std
%H2Q
Juv FIR/BW
short grass
240
85
60.3000
0.8000
1.0800
tall grass
110
36
40.6000
0.8000
1.0800
broad leaf forage
135
45
56.7000
0.8000
1.0800
fruit
15
7
12.4000
0.8000
1.0800
seeds
15
7
12.4000
0.1000
0.2400
invertebrates
94
65
48
0.8000
1.0800
Reset to default values
Application 2
Adult Initial Dose:: 0 mg/kg/d Juvenile Initial Dose:
Application rate: 0 Iba.iVA Month: January ~ Day: 1 ~
Application 3
Adult Initial Dose: q mg/kg/d Juvenile Initial Dose: 0 mg/kg/d
Application rate: | 0 j lb a.i./A Month: [january t| Day:|l
Application 4
Adult Initial Dose: 0 mg/kg/d Juvenile Initial Dose: 0 mg/kg/d
lb a.i./A Month: January
¦Nestling
Mar-01
Application 5
Adult Initial Dose: 0 mg/kg/d Juvenile Initial Dose: 0 mg/kg/d
Application rate: | 0 | lba.i./A Month: jjapimry _jgg] Day:|i ggj
Breeding Season (edit on Life History page
Date of first egg of first nest
Month | Marc ,| Day: [20 -|
Date of first egg of last nest
Month|Apn| • DaV: 120 ~
—Toxicity Thresholds (mg/kg/d): —
1/10 Adult LD50: 9999
Adult body weight: 9999
Eggs laid per hen: 9999
Waiting periods:
Mar-31
Apr-30 May-30
Date
Jul-29
Waiting period after
pesticide-induced failure:
days
Waiting period after failure: 16 davs
Waiting period after fledging: 16 days
Eggshell thickness: 9999
Viable eggs per egg set: 9999
Hatchlings per viable egg: 9999
14-d chicks per hatchling: 9999
fraction of juvenile LC50: 9999
Pesticide applications: This popup menu determines the number of pesticide applications that
will be used. By default, the window opens with 0 applications.
Crop: This input box is for inputting the name of the modeled crop scenario, for future
reference. It's value is reported in the log file.
Residue calculations: The controls in this frame include four radio buttons and a text box. The
first text box is for setting the half-life of pesticide residue degradation of food items (expressed
in days). The default is 35 days, but the value can be set to any positive value. The radio buttons
34
-------
User's Manual for MCnest v2.0 - September 2017
are mutually exclusive (only one can be checked) and indicate how pesticide application rates are
translated into exposure doses. The first choice "Set Dose Directly" allows the user to manually
set the adult and juvenile initial dietary dose of the pesticide on application day. The final three
choices translate the application rate into estimates of adult and juvenile initial dietary doses at
application using the approach in OPP's Terrestrial Residue Exposure model or T-REX (USEPA
2012). The T-REX approach is based on the estimated mean or maximum residue concentration
on each of the six food type categories immediately following an application of 1 pound active
ingredient (AI) per acre. The user can choose Use maximum nomogram values, Use mean
nomogram values, or Draw from the nomogram distribution, using a log-normal distribution
based on the mean and standard deviation. Residue concentrations on food types are translated
into dose using body weight and dietary information on the Life History window. Additional
detail is provided about these calculations in the MCnest Technical Manual. Whichever method
is used to determine the initial dietary dose on the day of application, subsequent exposure will
decay according to the specified half-life. When two or more applications are simulated, it
should be noted that the initial dietary dose presented for each application represents only the
contribution from that application, and does not add the contribution from earlier applications.
The additivity of overlapping decay curves from multiple applications can be observed in the
Seasonal Exposure Profile figure, even though this is not reflected in the adult and juvenile initial
doses shown for each application.
The nomogram values table is also editable. This table lists the default maximum, mean, and
standard deviation nomogram concentrations on each of the six food type categories used in T-
REX Version 1.5. It also includes the default percent moisture value from T-REX and the
juvenile food ingestion rate per body weight (i.e., FIR/BW) used in the calculation of juvenile
doses, as described in the MCnest Technical Manual. All of these values are editable by the
user. If changes are made to any values within the nomogram values table, there is a button to
restore all default values.
NOTE. There is a minor bug in the program that prevents the exposure series from being
calculated and displayed when application rates are set to 0 and the radio button for Draw from
the nomogram distribution is chosen. This will be fixed in a future release. Current
workaround is to set desired application rates to non-zero values before choosing the radio button
for Draw from the nomogram distribution is chosen. 28 July 2017.
Seasonal Exposure Profile (graph): As information on application rate, application date, and
residue half-life are modified in the Pesticides window, a Seasonal Exposure Profile will be
generated automatically. The model user can use this to visualize the exposure profile during the
breeding season of a species and/or to examine the degree of residue carry-over from one
application into the next. After a simulation, the Seasonal Exposure Profile also is available by
right-clicking on a highlighted simulation in the Output table and under "Row," click "Exposure
series."
Application 1: All of the pesticide application frames operate identically to Application 1.
These frames each contain three text boxes and two list boxes. The top two text boxes are for the
adult and juvenile initial dose (i.e., the dose on application day), respectively. These boxes are
not editable when any of the final three radio buttons in the Residue calculations frame are
selected. In this case, initial doses are calculated using the application rate and the dietary
35
-------
User's Manual for MCnest v2.0 - September 2017
information and body weight information for the species selected on the main MCnest page, as
described above and in more detail in the MCnest Technical Manual. The Application rate text
box can be set to any non-negative value. It will not be visible if the "Set Dose Directly" radio
button is selected in the Residue calculations frame, above. Finally, the two listboxes allow the
user to specify the date on which the pesticide is applied to the environment for each application
(i.e., the day on which the initial doses will be experienced).
The Application frames operate independently from each other, though they all use the same
half-life for defining the decay curve. Applications can be specified in any order (e.g.,
Application 2 could be specified as occurring prior to Application 1), and they need not occur
within the window defined by the Breeding Season (i.e., an application may occur prior to or
after the breeding season of some species). Regardless of how many pesticide applications are
chosen on the Pesticides window, the adult and juvenile initial doses shown for each application
are calculated based only on information for that application. However, when more than one
application is specified, the decay curves for each application may overlap. Consequently, the
estimated initial doses at application for birds in the simulation may reflect the combination of
the pesticide residues produced by each application plus any carry-over residues from previous
applications. The additivity of overlapping decay curves from multiple applications can be
observed in the Seasonal Exposure Profile figure, even though this is not reflected in the adult
and juvenile initial doses shown for each application.
Breeding Season: The controls in this frame are redundant with the same controls on the Life
History window. They are not operational on the Pesticides window, but are provided for
reference when setting the dates of pesticide applications.
Waiting Periods: This frame lists the three waiting periods after nest success or failure before
the first egg is laid in a subsequent nest. The two lower periods are redundant with waiting
periods on the Life History profile for each species. The top period is the "Waiting period after
pesticide-induced failure." The default value for this waiting period is set equal to the "Waiting
period after ecological failure" for each species. The waiting period after pesticide failure must
be an integer and can be set to a longer duration than for ecological failure, but it cannot be
shorter.
E. The TIM windows
TIM includes three windows for model input parameters (Figs 4.5-4.7). These are identical to
those presented in the TIM User's Guide (USEPA 2015: Appendix A) and model guidance is the
same. That document is the definitive technical document for TIM and can be found at:
https://www.epa.gov/pesticide-science-and-assessing-pesticide-risks/tim-version-30-beta-
technical-description-and-user-7. The main TIM window (Fig. 4.5) contains a File menu with
five choices (Table 4.3).
36
-------
User's Manual for MCnest v2.0 - September 2017
Table 4.3. Choices available on the File menu on the main TIM window
Menu Choice
Function
Pesticide Parameters Loads the TIM Pesticide Parameters window (Fig. 4.6)
Species
Save TIM
Parameters
Load TIM
Parameters
About
Loads the TIM Species window (Fig. 4.7)
Loads a file browser for saving a *.mat file with the currently entered
TIM parameters (parameters for all TIM windows will be saved)
Loads a file browser for loading a *.mat file with TIM parameters
(parameters for all TIM windows will be loaded)
Displays a message with information about TIM and where to find
further resources
Figure 4.5. The Opening TIM window parameterized with the sample data provided for X-
cyhalothrin
,•* TIM Version 3.0
1 = II a '1^1
rile
Scenario Definition
Pesticide:
Crop:
Species:
Lambda Cyhalothrin
Canada goose (giant)
10000
Number of birds:
Flock size:
Random seed: 12345678
25
Applications
Pesticide Applications: 4
Dates
lbs a.i./A
Application 1: 0.03
Application 2: j 0.03
| May
May
Application 3: 0.03
Application 4: 0.03
Exposure Routes:
m dietary
[71 drinking water (puddles)
[71 drinking water (dew)
(71 inhalation (vapor)
[7] inhalation (spray)
|y| dermal (contact)
[71 dermal (spray)
m spray drift
Method
(Air
Droplet spectrum
[ very fine to fine
Crop Type
field
Day:
Fraction of edge habitat receiving spray drift:
Length of in-field buffer (feet):
Fraction of organic carbon in soil:
Soil bulk density (kg/L)
Hour of first application:
Spray height (m):
Spray duration (min):
Crop height (m):
Crop mass (kg/ha):
37
-------
User's Manual for MCnest v2.0 - September 2017
Figure 4.6. The TIM Pesticide Parameters window
,-A TIM Version 3.0: Pesticide Parameters
Close
Toxicity
Avian acute oral LD50: 50000
1049
229
3.4
B M3.
Fate Parameters
Pesticide half life (puddle):
Koc (L'kg-oc):
Kow:
Henry's law constant (atnVm3-mol}:
Solubility in water (mg a.i./L):
Dislogable foliar residue adjustment factor:
Dermal absorbption fraction:
Diet:
76.2
333200
10000000
1.9e-07
0.005
0.62
Conta minated
Fraction
Half life:
Insects:
Seeds:
Fruit:
Grass:
Broadleaf:
1
35
1
35
1
35
1
35
1
35
38
-------
User's Manual for MCnest v2.0 - September 2017
Figure 4.7. The TIM Species window
15 TIM Version 3.0: Species
a HH
Done
Choose species:
0: Custom Species
Feeding times
Start min:
AM:
PM:
Start max:
4
5
16
19
End min:
End max:
10
Prop min
0.4
Prop max:
0.6
20
21
Gorging
factor:
Body weight
mean
stdev
3514
256.522
3062
3912
Field fidelity parameters
I I Field resident Field fidelity factor:
0.6
Frequency on
Species Type
I Passerine
m Precocial
2. Herbivore
-Dietary proportions (mustsumto 1):
Insects Seeds
Fruit
Grass Broadleaf
Sum
Adults:
0
0
0
1
0
1
Juveniles:
0
0
0
1
0
1
F. The Batch window
The Batch window is invoked by selecting Batch Mode under the main MCnest menu. This
window offers the ability to set up multiple simulations in advance according to three different
types of progressions (Fig. 4.8). If the Exposure Algorithm drop-down menu is set to
CONTROL, then the batch window will show only the top Species panel. If the Exposure
Algorithm is set to either TIM or TREX, then the bottom panels for Application Date and
Application Rate will be visible.
39
-------
User's Manual for MCnest v2.0 - September 2017
Figure 4.8. Batch window with starting conditions set to the exposure scenario described in
Section II._quick start guide.
MCnest Version 2.0: Batch
f^l B 11^1
Exposure:
Species
© All species
Candidate species:
.la Hard
Blue-winged teal
Northern bobwhite
American kestrel
Killdeer
White-winged dove
Mourning dove
Selected species:
r
Override species default waiting period after pesticide failure (Wp)
days.
Application Date
Pesticide Applications: 4
Dates
Ibsa.i./A
Day:
Application 1:
Application 2:
Application 3: [ 0.03
Application 4:
0.03
0.03
0.03
May
May
Application Rate
Geometric series
Arithemetic series
Multiplier 1 Ibs/acr
Repeats: 1
All applications will be
increased geometrically
by multiplying each
successive rate by the
Repeat Interval:
days
Repeats:
Exposure: This display panel notifies you which exposure model is chosen on the main MCnest
window. It cannot be changed here. You must go back to the main MCnest window to change
the exposure algorithm.
Run: This pushbutton starts the simulation through the batches you've set up.
Reset: This pushbutton removes any batches currently designed and returns the window to its
opening state.
Exit: Returns to the main MCnest window without running the batches you've designed.
1. Species
The Species frame contains two radio buttons, two listboxes, two push buttons, and an edit box.
The All species radio button moves all species from the Candidate species listbox to the
Selected species listbox. Deselecting the All species button will restore them to the Candidate
species listbox. The Add and Remove buttons will move selected individual species back and
forth between listboxes. When batching on species, separate simulations will be run on the
species in the Selected species listbox and species in the Candidate species listbox will be
40
-------
User's Manual for MCnest v2.0 - September 2017
ignored. Selecting the radio button Override species default waiting period after pesticide
failure (Wp) will allow you to set a unique value for the waiting period for pesticide failure in
the associated edit box. This value will then apply to all species in the designed batch, unless the
specified value is shorter than the species' value of We (the waiting period after ecological
failure), in which case the species-specific value for We will be used for that species.
Note that if the "Override species default" radio button has been selected and you wish to return
to using the default values for Wp for each of the selected species, the Override species
default... radio button needs to be deselected and then hit Exit.
2. Application Date
The Application Date frame contains three text boxes and additional text boxes arrayed in rows
for up to 5 applications depending on the value of the Pesticide Applications popup menu. The
input boxes within the Dates frame are parameterized exactly the same way they are on the
TREX or TIM windows. At the bottom of the window, the two input boxes Repeat interval and
Repeats determine the number of scenarios that will be run. Using these two pieces of
information, MCnest simulates the application scenario described in the Dates frame. It then
repeats the simulation, shifting all application dates by Repeats days each time until the total
number of desired repeats is reached.
3. Application Rate
The Application Rate frame contains two radio buttons and two text boxes. The radio buttons
determine what kind of progression will occur in changes in application rate. If arithmetic
progression is chosen, then the application rates specified in the Dates frame will be changed by
adding a fixed constant to each application rate with every iteration. If geometric progression is
chosen, then the application rates specified in the Dates frame will be changed by multiplying by
a fixed constant with every iteration. The first input box (labeled either increment or multiplier,
depending on what kind of progression is chosen) indicates the fixed constant that will be either
added to or multiplied by the initial application rates and all subsequent rates. The Repeats
input box determines the number of iterations in the arithmetic or geometric progressions
designed.
Note that the different batch panes are independent and factorial. Every combination of each
iteration will be run. This can add up quickly if batching over all species in the library.
E. Other MCnest menu options
1. Random Numbers
The Random window is invoked by selecting Random Numbers under the main MCnest
window. This window offers the ability to choose among different types of random number
generators and to set the seed for random number generation in MCnest. This may be useful for
exactly repeating a particular simulation in MCnest. The Random window contains one push
button, a popup menu and an input box (Figure 4.9).
41
-------
User's Manual for MCnest v2.0 - September 2017
Figure 4.9. The Random window in MCnest, with no seed specified.
BTj MCnest Version 2.0: Random N...
a MM
Return to Main Screen
Random Numbers
Random Generator:
Seed
mt19937a r
Return to Main Screen: This pushbutton returns you to the main MCnest screen
Random Generator: The Random Generator popup allows you to specify which algorithm is
used to generate random numbers in MCnest simulations. There are six choices (Table 4.4). We
have no reason to believe that any of the algorithms below would offer significantly better
performance than any other, and we recommend using the MATLAB and MCnest default option
called the Mersenne twister (listed as "mtl9937ar") based on its computational efficiency
(speed).
Seed: The "Seed" textbox allows you to specify a numeric seed for the random number
generator. This should be a positive number. To exactly duplicate a previous simulation, find
the seed value at the bottom of the Log file and enter here. If no seed is specified, MCnest will
use the computer's system clock to generate a seed. In this case, the seed is still recorded, so the
simulation may be re-created even when no seed was specified.
Table 4.4. Random number generators available in MATLAB and MCnest1.
Option
MATLAB description
Reference
mtl9937ar Mersenne twister (default)
meg 16807 Multiplicative congruential generator
mlfg6331_64 Multiplicative lagged Fibonacci
generator
mrg32k3a Combined multiple recursive generator
shr3cong Shift-register generator summed with
linear congruential generator
swb2712 Modified subtract with borrow generator Marsaglia and Zaman 1991
Matsumoto and Nishimura 1998
Park and Miller. 1998
Mascagni and Srinivasan 2004
L'Ecuyer et al. 2002
Marsaglia 1999
Adapted from MATLAB online documentation (Mathworks 2012).
2. Load MCnest file
Choosing this menu option loads a standard Windows file browser. You can select and load any
previously saved results file for examination using MCnest output functions, or for comparison
among simulations generated in different sessions. This will work with individual simulations
saved using the Save row option and multiple simulation saved using the Save table option in
the Output table context menu. Any results loaded will be appended to any currently available
results in the Output table.
42
-------
User's Manual for MCnest v2.0 - September 2017
3. Import Time-varying parameters
This is a very new feature with this version of MCnest and should be considered a pilot for future
MCnest development. The motivation for this feature is to provide additional resolution and
realism to MCnest simulations when data are available. Currently, only five parameters can be
specified to vary over time. These are the quitting probabilities (q/ and qs from an absorbing
Markov Chain), the daily nest failure probabilities (mi = pre-hatch and mi= post-hatch) and the
number of fledglings produced from a successful nest (FPSN).
To specify time-varying parameters, the desired temporal variation must first be designed in an
Excel file (Fig. 4.10). In Figure 4.10, the file is displayed with the first column reserved for
entering the date. This is not required, but is allowed for convenience. The next four columns are
for entering date-specific values for the parameters. Note there is only one column for nest
failure probability because in this case the same value will be used for both m\ and mi for that
day (simply by way of example, not asserting anything about American Robin nest survival).
Figure 4.10. An Excel ™ worksheet with time-varying values of daily mortality rates displayed
Conditional Formatting - If3 Insert -
Format as Table ' Delete
Cell Styles » gjp Format •
General
Calibri
Clipboard rs I Font rs I Alignment rs Number Styles Cells Editing
G92 ' i X >/ fx
A
B
C D
1 E
F
G
H i I | J
K
L
M
1
date
m
qf qs
fpsn
97
6-Apr
0.03
0.04
0
98
7-Apr
0.03
0.046667
0
99
8-Apr
0.03
0.053333
0
100
9-Apr
0.03
0.06
0
101
10-Apr
0.03
0.066667
0
102
11-Apr
0.03
0.073333
0
103
12-Apr
0.03
0.08
0
104
13-Apr
0.03
0.086667
0
105
14-Apr
0.03
0.093333
0
106
15-Apr
0.03
0.1 0
107
16-Apr
0.03
0.106667
0
108
17-Apr
0.03
0.113333
0
109
18-Apr
0.03
0.12
0
110
19-Apr
0.03
0.126667
0
111
20-Apr
0.03
0.133333
0
112
21-Apr
0.03
0.14
0
113
22-Apr
0.03
0.146667
0
114
23-Apr
0.03
0.153333
0
115
24-Apr
0.03
0.16
0
116
25-Apr
0.03
0.166667
0
117
26-Apr
0.03
0.173333
0
parameters © 5 |n 1 1 IIH
Ready H M E] 1 + 100%
Figure 4.11. Time-varying parameter window
43
-------
User's Manual for MCnest v2.0 - September 2017
ImportProbabilities2
Choose Excel File
Choose Worksheet:
r^i' a ^
V. Literature Cited
American Ornithologists' Union. 1998. Checklist of North American Birds, 7th ed. American
Ornithologists' Union, Washington, DC, USA.
Bennett RS, Dewhurst IC, Fairbrother A, Hart ADM, Hooper MJ, Leopold A, Mineau P,
Mortensen SR, Shore RF, Springer TA. 2005. A new interpretation of avian and
mammalian reproduction toxicity test data in ecological risk assessment. Ecotoxicology 14:
801-815.
Bennett RS, Etterson MA. 2007. Incorporating results of avian toxicity tests into a model of
annual reproductive success. Integrated Environmental Assessment and Management 3:
498-507.
Bennett RS, Etterson MA. 2013a. Avian Life-History Profiles for Use in the Markov Chain Nest
Productivity Model (MCnest). EPA/600/R-14/009. http://www.epa.gov/chemical-
research/markov-chain-nest-productivitv-model-documentation.
Bennett, R. and M. Etterson. 2006. Estimating pesticide effects on fecundity rates of wild birds
using current laboratory reproduction tests. Human and Ecological Risk Assessment 12:762-
781.
Bennett, R.S and M.A. Etterson. 2013b. On selecting surrogate endpoints for estimating pesticide
effects on avian reproductive success. Integrated Environmental Assessment and
Management 9:600-609.
Bennett, R.S and M.A. Etterson. 2013c. User's Manual for Basic Version of MCnest-Markov
Chain Nest Productivity Model. EPA/600/R/13/034. https://www.epa.gov/chemical-
research/markov-chain-nest-productivity-model-documentation.
Bennett, R.S and M.A. Etterson. 2013d. Technical Manual for Basic Version of the Markov
Chain Nest Productivity Model (MCnest). EPA/600/R/13/033.
https://www.epa.gov/chemical-research/markov-chain-nest-productivitv-model-
documentation.
44
-------
User's Manual for MCnest v2.0 - September 2017
Etterson MA, Bennett RS, Kershner EL, Walk JW. 2009. Markov chain estimation of avian
seasonal fecundity. Ecological Applications 19: 622-630.
Etterson MA, Bennett RS. 2013. Quantifying the effects of pesticide exposure on annual
reproductive success of birds. Integrated Environmental Assessment and Management 9:590-
599.
L'Ecuyer P, Simard R, Chen EJ, Kelton WD. 2002. An Objected-Oriented Random-Number
Package with Many Long Streams and Substreams. Operations Research 50:1073-1075.
2002.
Marsaglia G, Zaman A. 1991. A new class of random number generators. Annals of Applied
Probability 1:462-480.
Marsaglia G. 1999. Random numbers for C: The END? Usenet posting to sci.stat.math.
Available online at:
http://groups.google.com/group/sci.crypt/browse_thread/thread/ca8682a4658al24d/
Mascagni, M, Srinivasan A. 2004. Parameterizing Parallel Multiplicative Lagged-Fibonacci
Generators. Parallel Computing. 30:899-916.
MathWorks 2012. Matlab v2012a. MathWorks, Inc. Natick, MA, USA.
Matsumoto M, Nishimura T. 1998. Mersenne Twister: A 623-dimensionally equidistributed
uniform pseudorandom number generator. ACM Transactions on Modeling and Computer
Simulation 8:3-30.
Mineau P, Collins BT, Baril A. 1996. On the use of scaling factors to improve interspecies
extrapolation of acute toxicity in birds. Regulatory Toxicology and Pharmacology 24:24-29.
Park SK, Miller KM. 1998. Random Number Generators: Good Ones Are Hard to Find.
Communications of the ACM 31:1192-1201.
USEPA. 2012a. OCSPP 850.2100: Avian Acute Oral Toxicity Test; United States Environmental
Protection Agency, Office of Chemical Safety and Pollution Prevention. EPA 712-C-025.
Available: http://www.epa.gov/test-guidelines-pesticides-and-toxic-substances/series-
850-ecological-effects-test-guidelines.
USEPA. 2012b. OCSPP 850.2200: Avian Dietary Toxicity Test; United States Environmental
Protection Agency, Office of Chemical Safety and Pollution Prevention. EPA 712-C-024.
Available: http://www.epa.gov/test-guidelines-pesticides-and-toxic-substances/series-
850-ecological-effects-test-guidelines.
USEPA. 2012c. OCSPP 850.2300: Avian Reproduction Test; United States Environmental
Protection Agency, Office of Chemical Safety and Pollution Prevention. EPA 712-C-023.
Available: http://www.epa.gov/test-guidelines-pesticides-and-toxic-substances/series-
85 0-ecolo gical-effects-test- guidelines.
USEPA. 2015. Technical description and user's guidance document for the terrestrial
investigation model (TIM) Version 3.0 Beta. Available: http://www2.epa.gov/pesticide-
science-and-assessing-pesticide-risks/models-pesticide-risk-assessment#tim.
45
-------
User's Manual for MCnest v2.0 - September 2017
USEPA. U.S. Environmental Protection Agency. 2012d. User's Guide T-REX Version 1.5
(Terrestrial Residue Exposure model). Office of Pesticide Programs. Washington, DC,
USA. (http://www.epa.gOv/oppefedl/models/terrestrial/trex/t rex user guide.htm#app a
Last accessed 11 January 2013).
46
------- |