UNITED
STATES
ENVIRONMENTAL
PROTECTION
AGENCY
WASHINGTON
D.
C.,
20460
OFFICE
OF
PREVENTION,
PESTICIDES
AND
TOXIC
SUBSTANCES
PC
Codes:
012501,
012502,
013802,
013803,
013806
DP
Barcode:
D327974
Date:
March
29,
2006
ADDENDUM
TO
RED
CHAPTER
SUBJECT:
Response
to
Registrant
Phase
I
Error
Only
Comments:

Addendum
to
EFED
RED
Chapters
for
Organic
Arsenicals
Accounting
for
Updated
Label
Rates
and
Potential
for
Long
Term
Buildup
in
Soil
[
Cacodylic
Acid
and
its
Sodium
Salt
(
D210451,
D212449,
D255226)
and
Sodium
and
Calcium
Salts
of
Methanearsonic
Acid
(
MSMA
/
DSMA
/
CAMA;
D277233)]

TO:
Lance
Wormell,
Chemical
Review
Manager
Reregistration
Branch
II
Special
Review
and
Reregistration
Division
(
7508C)

FROM:
Keara
Moore,
Chemist
Environmental
Risk
Branch
III
Environmental
Fate
and
Effects
Division
(
7507C)

Thuy
Nguyen,
Senior
Environmental
Scientist
Environmental
Risk
Branch
III
Environmental
Fate
and
Effects
Division
(
7507C)

THROUGH:
Mark
Corbin,
Senior
Environmental
Scientist
Environmental
Risk
Branch
III
Environmental
Fate
and
Effects
Division
(
7505C)

APPROVED
Daniel
Rieder,
Branch
Chief
BY:
Environmental
Risk
Branch
III
Environmental
Fate
and
Effects
Division
(
7507C)
2
Attached,
please
find
the
revised
addendum
to
the
EFED
RED
Chapters
for
Organic
Arsenicals
Accounting
for
Updated
Label
Rates
and
Potential
for
Long
Term
Buildup
in
Soil.
Revisions
have
been
made
to
the
previous
EFED
RED
Addendum
(
DP
Barcode
309100;
February
3,
2006)
to
correct
any
mathematical/
typographical
errors
and
provide
clarification
in
areas
identified
by
the
registrants
in
their
Phase
1
response.
Additional
comments
were
received
regarding
EPA's
methodology
and
environmental
fate
results;
these
comments
will
be
addressed
after
the
60­
day
Phase
3
public
comment
period
scheduled
to
begin
in
April
2006.

The
revisions
made
in
this
document
do
not
alter
any
EFED
conclusions
from
the
previous
assessment.
Where
appropriate,
references
to
"
degradation"
have
been
replaced
with
the
more
suitable
term
"
metabolism."
Language
has
been
added
to
clarify
the
discussion
of
the
potential
for
transformation
of
organic
arsenicals
to
inorganic
arsenic
and
for
transformation
of
dimethylated
arsenical
species
to
monomethylated
species.
A
more
thorough
discussion
of
terrestrial
field
dissipation
data
is
also
included,
presented
in
the
soil
accumulation
section
in
the
environmental
fate
appendix.
3
I.
EXECUTIVE
SUMMARY
This
addendum
updates
the
ecological
risk
assessments
for
the
organic
arsenicals,
which
include
cacodylic
acid
(
DMA;
dimethylarsinic
acid),
sodium
cacodylate
(
DMA­
Na),
monosodium
methanearsonate
(
MSMA),
disodium
methanearsonate
(
DSMA),
and
calcium
acid
methanearsonate
(
CAMA).
Originally,
these
pesticides
were
assessed
as
two
separate
groups:
the
methylarsonate
salts
(
MSMA,
DSMA,
CAMA;
DP
Barcode
D277233)
and
the
cacodylates
(
DMA
and
DMA­
Na;
DP
Barcodes
D210451,
D212449,
D255226).
Although
risk
estimates
are
still
calculated
based
on
the
original
groupings,
these
two
groups
are
being
included
in
a
single
document
because
the
discussion
of
their
fate
characteristics
has
been
combined.
They
all
have
similar
chemical
structures
and
similar
environmental
fate
profiles,
and
they
all
contribute
to
total
arsenic
environmental
loading,
an
important
issue
in
the
current
assessment.
The
updated
drinking
water
assessment
(
DP
Barcode
D309098)
also
considers
these
compounds
as
a
group.

For
MSMA
and
DSMA,
risk
estimates
are
updated
based
on
new
master
labels
that
have
lowered
application
rates
for
most
uses.
The
new
master
label
for
cacodylic
acid
(
dimethylarsinic
acid;
DMA)
had
not
been
completed
at
the
time
of
this
assessment
and
so
has
not
been
included
here.
No
new
toxicity
data
have
been
considered
and
so
calculation
of
risk
is
all
based
on
toxicity
conclusions
from
the
original
assessment.
With
the
lower
application
rates,
most
risk
quotients
(
RQs)
have
dropped
as
well,
but
the
overall
risk
conclusions
have
not
changed.
There
are
no
exceedances
of
levels
of
concern
(
LOC)
for
aquatic
species,
including
fish,
invertebrates,
and
plants.
All
categories
of
terrestrial
exposure
lead
to
exceedances
of
LOCs,
including
small
mammals,
avian
species,
and
upland
and
semi­
aquatic
plants.
Because
of
lower
applications
rates,
in
many
cases
the
terrestrial
exceedances
are
not
as
great
as
in
the
original
assessment
and
there
are
fewer
exceedances
in
the
highest
risk
category.
The
endangered
species
portion
of
this
assessment
has
not
been
updated.
The
original
assessment
found
that
endangered
species
risk
quotients
exceed
the
Agency's
LOC
for
birds,
mammals
and
terrestrial
plants
from
exposure
to
DSMA
and
MSMA.
There
are
still
exceedances
in
each
of
these
categories.
In
the
original
assessment,
however,
one
use
led
to
exceedances
for
freshwater
fish
and
the
new
rates
lead
to
RQs
below
the
LOC.

The
discussion
of
the
environmental
fate
and
transport
of
organic
arsenicals
has
been
expanded.
Greater
attention
is
paid
to
the
possibility
of
transformation
of
these
compounds
to
the
more
toxic
inorganic
arsenic.
Additionally,
the
issue
of
long
term
accumulation
of
total
arsenic
in
soil
after
repeated
applications
of
organic
arsenicals
has
been
considered.
Field
studies,
monitoring
data,
and
modeling
all
indicate
that
accumulation
of
arsenical
in
soil
is
likely.
Based
on
PRZM
modeling
of
soil
concentrations,
maximum
rates
of
MSMA,
DSMA,
and
DMA
on
turf
are
expected
to
exceed
the
ecological
soil
screening
levels
set
by
the
EPA's
Office
of
Solid
Waste
and
Emergency
Response
(
OSWER).
Modeling
suggests
that
applications
of
these
products
to
cotton
will
not
lead
to
soil
level
exceedances.
4
II.
PESTICIDE
USE
CHARACTERIZATION
This
update
is
based
on
master
labels
for
MSMA
and
DSMA.
These
labels,
provided
by
the
registrants
on
11/
14/
05,
have
not
yet
been
finalized
but
are
assumed
to
invalidate
all
previous
labels.
Variance
from
these
rates
could
change
the
conclusions
of
this
assessment.

The
main
agricultural
application
for
both
MSMA
and
DSMA
is
as
a
pre/
post­
emergent
herbicide
on
cotton
applied
prior
to
the
first
bloom.
The
only
other
agricultural
uses
of
MSMA
and
DSMA
supported
by
master
labels
are
in
non­
bearing
orchards,
citrus,
and
vineyards.
Non­
agricultural
uses
of
MSMA
and
DSMA
include
use
on
turf
and
non­
crop
uses.
The
turf
use
is
both
residential
and
commercial,
including
residential
lawns,
sod
farms,
golf
courses,
parks,
and
other
areas.
The
non­
crop
use
includes
drainage
ditch
banks,
rights­
of­
way,
storage
yards
and
similar
areas.

The
maximum
application
rates
for
all
of
these
uses
are
included
in
Table
1.
Although
applied
as
different
parent
compounds,
MSMA
and
DSMA
end
up
as
the
same
chemical
in
the
environment,
monomethyl
arsonic
acid
(
MMA).
Rates
for
both
are
reported
here
in
acid
equivalents
(
ae)
to
represent
the
amount
applied
as
MMA.
Labels
for
all
uses
specify
that
the
maximum
number
of
applications
applies
to
"
either
DSMA
or
MSMA
or
their
combination
per
crop,
per
year."
Exposure
was
therefore
estimated
based
on
the
maximum
MMA
application
rate
for
each
use
for
either
MSMA
or
DSMA.
The
maximum
rate
for
each
use,
marked
in
bold
on
Table
1,
was
considered
in
exposure
estimates.
For
the
non­
crop
use,
the
master
label
single
application
rate
is
lower
than
that
allowed
by
previous
labels,
but
the
number
of
applications
has
increased
so
the
annual
application
rate
has
not
changed.
For
all
other
uses,
the
master
label
annual
application
rate
is
lower
than
the
rates
considered
in
the
previous
assessment.

III.
EXPOSURE
CHARACTERIZATION
A.
Environmental
Fate
and
Transport
Summary
The
environmental
fate
section
has
been
updated
to
discuss
the
organic
arsenicals
(
MSMA,
DSMA,
CAMA,
DMA)
as
a
group
rather
than
in
separate
documents.
It
has
been
expanded
to
include
consideration
of
the
potential
for
long
term
accumulation
of
total
arsenic
in
soil
and
of
metabolism
of
organic
arsenicals
to
inorganic
arsenic.
The
general
conclusions
about
the
fate
and
transport
of
organic
arsenicals
have
not
changed,
but
re­
evaluation
of
the
available
data
has
led
to
revision
of
several
modeling
inputs.
A
detailed
discussion
of
environmental
fate
properties
is
provided
in
Appendix
A
and
summarized
here.
References
to
all
of
the
registrant
studies
and
open
literature
reports
taken
into
account
in
this
fate
characterization
are
included
in
the
Appendix.

MSMA,
DSMA,
and
CAMA
are
salts
of
the
dibasic
weak
acid
MMA.
In
aqueous
solution,
they
dissociate
to
MMA
and
the
associated
companion
cations.
DMA
is
a
weak
acid
with
two
methyl
groups
attached
to
the
central
arsenic
atom,
rather
than
one
as
in
MMA.
These
pesticides
are
all
non­
volatile
solids
that
are
highly
soluble
in
water.
5
Table
1.
Application
Rates
for
MSMA
and
DSMA,
based
on
11/
14
Master
Labels.

Use
Maximum
App.
Rate
(
lb
ae/
A)
Max.
No.
Apps.
Application
Interval
Application
Method
DSMA
1.7
1
n/
a
g
round
or
aerial
Cotton
1.7
2
1
 
3
weeks
ground
(
directed)

Turf
2.5
4
14
days
ground
spray
Orchards,
Citrus,
Vineyards
1
3.7
3
not
specified
ground
(
directed)

Non­
crop
2
3.9
4
10
 
14
days
ground
spray
Grass
for
seed
3
3.3
1
n/
a
n
ot
specified
MSMA
1.7
1
n/
a
g
round
or
aerial
Cotton
1.7
2
not
specified
ground
(
directed)

Turf
­
Sod
Farms4
­
Golf
Course4
3.4
2.2
4
10
 
14
days
ground
spray
Orchards,
Citrus,
Vineyards
1
3.7
3
10
 
14
days
ground
spray
Non­
crop
2
3.9
4
10
 
14
days
ground
spray
Grass
for
seed
3
5.3
1
n/
a
n
ot
specified
1
Non­
bearing
only
­
not
to
be
used
within
one
year
of
harvest.
Restricted
to
spot
treatments
in
Florida.
2
Non­
crop
=
"
drainage
ditchbanks,
rights­
of­
way,
storage
yards
and
similar
areas".
3
Pacific
Northwest
only.
4
Sod
farm
rate
allowed
on
sod
farms
and
established
Bermuda
and
Zoysiagrass.
Golf
course
rate
allowed
on
athletic
fields,
golf
courses,
and
parks.

Environmental
fate
laboratory
studies
show
that
organic
arsenicals
are
stable
under
all
tested
abiotic
conditions;
they
do
not
degrade
by
hydrolysis
or
by
aquatic
or
soil
photolysis.
Arsenicals
can
be
subject
to
microbial
metabolism
in
soil
under
aerobic
and
anaerobic
conditions.
The
occurrence,
rate,
and
products
of
this
metabolism
are
variable,
dependent
on
environmental
conditions.
Persistence
of
applied
parent
compounds
can
range
from
days
to
years,
depending
on
soil
properties
and
ambient
conditions
such
as
soil
moisture,
temperature,
chemical
concentration,
bacterial
population,
and
amount
of
organic
matter.
Regardless
of
the
form
it
takes,
however,
the
total
amount
of
arsenic
present
does
not
change;
these
arsenicals
and
their
transformation
products,
in
combination
with
arsenic
from
the
natural
background
and
from
other
anthropogenic
sources,
maintain
the
total,
immutable
arsenic
load.
Arsenic
from
pesticides
is
not
lost
but
redistributed
and
transformed
throughout
the
environment
(
plants,
animals,
air,
soil,
sediment,
water)
into
other
arsenic
containing
substances.

Metabolism
rates
do
not
appear
to
depend
linearly
on
arsenical
concentration;
the
kinetics
are
therefore
not
necessarily
first­
order
and
so
"
half­
life"
may
not
be
an
appropriate
6
constant
for
all
concentrations.
Despite
the
uncertainty,
first­
order
half­
lives
have
been
calculated
for
modeling
purposes
and
as
a
convenient
measure
to
compare
laboratory
results.
The
estimated
half­
lives,
used
in
EFED's
current
models,
may
underestimate
the
faster
initial
rate
of
metabolism
but
adequately
portray
the
overall
transformation
and
so
are
assumed
to
be
protective
for
chronic
exposure,
a
major
concern
for
arsenicals.
The
modeled
aerobic
soil
half­
life
for
MMA,
based
on
two
studies
with
similar
results,
is
240
days.
No
anaerobic
soil
half­
life
was
determined
for
MMA.
For
DMA,
the
Agency
derived
aerobic
soil
half­
life
is
173
±
115
days
with
a
standard
upper
90%
confidence
limit
on
the
mean
of
240
days.
The
anaerobic
soil
half­
life
for
DMA
was
calculated
to
be
128
±
38
days
with
a
standard
upper
90%
confidence
limit
on
the
mean
of
168
days.

The
effects
of
environmental
factors
on
the
rate
of
arsenical
metabolism
are
complex
and
poorly
defined,
with
different
studies
leading
to
conflicting
results.
An
increase
in
temperature
leads
to
increased
metabolism.
The
observed
influences
of
soil
organic
matter
or
applied
arsenical
concentrations
are
contradictory.
The
effect
of
aerobic
versus
anaerobic
conditions
on
metabolism
rates
is
also
ambiguous.

Potential
metabolites
of
applied
arsenicals
include
volatile
alkylarsines
and
inorganic
arsenic
(
as
arsenate
or
arsenite)
along
with
carbon
dioxide.
Additionally,
DMA
may
be
present
as
a
metabolite
of
MMA
as
well
as
applied
directly.
As
with
the
rate,
the
metabolism
pathway
is
sensitive
to
environmental
conditions
in
indeterminate
ways
with
the
major
metabolites
occurring
in
widely
variable
proportions.
Transformation
to
volatile
alkylarsines,
the
only
metabolism
route
that
would
directly
reduce
soil
arsenic
loading,
has
been
shown
to
be
possible
in
certain
circumstances
but
is
generally
not
expected
to
be
a
major
route
of
dissipation.
A
maximum
of
35%
of
applied
MMA
is
expected
to
be
present
as
DMA
at
any
one
time.
Theoretically,
there
is
some
possibility
for
MMA
to
metabolize
to
DMA,
but
significant
transformation
has
not
been
observed
in
current
acceptable
field
or
laboratory
studies.
Observed
metabolism
of
MMA
and
DMA
to
inorganic
arsenic
has
ranged
from
undetected
after
several
years
to
more
than
80%
transformation
in
several
months.
Generally,
arsenate
[
As(
V)]
is
expected
to
be
the
dominant
species
of
inorganic
arsenic,
but
in
reducing
conditions,
arsenite
[
As(
III)]
may
be
more
stable.

Some
of
the
variability
in
metabolism
processes
is
associated
with
variability
in
sorption,
because
microbial
transformation
is
only
likely
to
occur
while
compounds
remain
dissolved
in
pore
water.
Mobility
of
arsenicals
is
typically
very
low
to
intermediate
and
appears
to
be
independent
of
organic
matter
content.
Instead,
sorption
is
higher
in
soils
with
higher
percentage
of
clay
or
with
more
iron
or
aluminum
content.
One
study
found
by
direct
comparison
that
all
arsenicals
were
more
strongly
sorbed
than
phosphate
in
the
increasing
order:
phosphate
<
DMA
<
arsenate
~
MMA.
The
lowest
non­
sand
Kd
for
MMA
is
11.4
mL/
g.
For
20
tested
soils,
the
range
of
Kds
spans
two
orders
of
magnitude
(
0.5
to
95
mL/
g,
mean
37
mL/
g).
For
DMA,
the
lowest
non­
sand
Kd
from
16
soils
is
8.2
mL/
g
(
range
8.2
to
33
mL/
g,
mean
18
mL/
g).

Surface
Water
Exposure
Conclusions.
Arsenical
pesticides
and
their
metabolites
may
be
transported
to
surface
waters
and
sediments
through
runoff
water,
eroding
soils,
or
drift
7
during
application.
These
routes
of
exposure
are
likely
to
lead
to
local,
temporal
elevations
above
background
arsenic
levels
in
surface
water
bodies.
Tier
I
surface
water
modeling
for
MSMA
and
DSMA
estimated
surface
water
concentrations
as
high
as
360
ppb,
as
MMA.
Limited
targeted
monitoring
has
found
elevated
total
arsenic
levels
in
surface
water
bodies
in
MMA
use
areas.
In
cotton
growing
areas
in
Mississippi,
surface
water
concentrations
of
MMA
up
to
5
ppb
were
detected.
In
golf
courses
in
Florida,
6
of
10
ponds
tested
had
elevated
arsenic
levels
with
a
maximum
of
34
ppb,
as
total
arsenic.
These
monitoring
results
are
discussed
in
more
detail
in
the
organic
arsenical
Drinking
Water
Assessment
(
DP
Barcode
309098).

Soil
Accumulation
Conclusions.
The
relative
immobility
of
arsenicals
along
with
arsenic's
elemental
nature
make
buildup
in
soil
after
repeated
applications
an
important
consideration.
Controlled
field
studies,
monitoring
targeted
to
pesticide
use
areas,
and
soil
modeling
results
all
indicate
that
soil
buildup
is
a
likely
result
of
long
term
organic
arsenical
application.
Arsenic
accumulation
is
likely
to
be
limited
to
the
top
layers
of
soil,
with
studies
suggesting
that
it
is
unlikely
to
occur
at
depths
greater
than
30
cm.
These
conclusions
are
discussed
in
more
detail
in
the
terrestrial
exposure
section.

B.
Aquatic
Exposure
For
surface
water
contamination,
the
estimated
environmental
concentrations
(
EEC)
were
calculated
using
GENEEC2
(
GENeric
Estimated
Environmental
Concentration).
This
Tier­
I
model
uses
the
soil/
water
partition
coefficient
and
degradation
kinetic
data
to
estimate
runoff
from
a
ten
hectare
field
into
a
one
hectare
by
two
meter
deep
"
standard"
pond.
This
model
is
designed
as
a
coarse
screen
and
estimates
conservative
pesticide
concentrations
in
surface
water
from
a
few
basic
chemical
parameters
and
pesticide
label
use
and
application
information.
This
model
is
used
to
screen
chemicals
in
order
to
determine
which
ones
potentially
pose
sufficient
risk
to
warrant
higher
level
modeling.
Additional
information
on
this
model
can
be
found
at:
http://
www.
epa.
gov/
oppefed1/
models/
water/
index.
htm.

The
input
parameters
used
to
estimate
the
surface
water
EECs
for
aquatic
exposure
assessment
were
selected
from
the
environmental
fate
data
submitted
by
the
registrant
and
from
the
open
literature,
and
in
accordance
with
USEPA­
OPP
EFED
water
model
parameter
selection
guidelines,
Guidance
for
Selecting
Input
Parameters
in
Modeling
the
Environmental
Fate
and
Transport
of
Pesticides,
Version
II,
February
28,
2002.
The
maximum
rate
for
each
use,
either
from
MSMA
or
DSMA,
was
modeled.
Both
are
present
in
aqueous
solution
as
monomethyl
arsonic
acid
(
MMA)
and
are
therefore
modeled
in
acid
equivalents
(
ae)
to
represent
the
amount
applied
as
MMA.
With
the
exception
of
the
sorption
coefficient,
which
was
reduced
from
13
mL/
g
to
11.2
mL/
g,
all
modeling
inputs
are
the
same
as
in
the
original
RED
(
DP
Barcode
277233).
The
general
input
values
used
in
the
model
runs
are
included
in
Appendix
B,
along
with
the
model
outputs.
The
results
are
presented
in
Table
2.
8
Table
2.
EECs
for
MMA
in
the
aquatic
environment.

1
The
pesticide
with
the
highest
application
rate
for
each
use
was
modeled
as
the
acid
equivalent
MMA.
2
All
application
rates
are
based
on
ground
application.
The
cotton
use
is
labeled
for
aerial
application
as
well
but
a
higher
rate
is
allowed
for
ground
application.

C.
Terrestrial
Exposure
1.
Animals
Terrestrial
wildlife
are
exposed
to
pesticides
through
the
plant
or
animal
material
that
they
consume
as
food.
The
Kenaga
nomogram
as
modified
by
Fletcher,
a
series
of
tables
that
relate
food
item
residues
to
pesticide
application
rate
based
on
a
database
of
actual
measured
pesticide
residue
values
on
plants,
are
used
to
estimate
exposure
to
terrestrial
organisms.
The
TREX
model
(
version
1.22)
was
used
to
calculate
EECs,
presented
in
Table
3,
for
MMA
from
application
of
MSMA
and
DMSA.
The
equations
and
assumptions
used
by
TREX
are
discussed
in
Appendix
C.

Table
3.
EECs
(
ppm
MMA)
on
terrestrial
food
items
resulting
from
direct
applications
of
DSMA
and
MSMA.

1
The
pesticide
with
the
highest
application
rate
for
each
use
was
modeled
as
the
acid
equivalent
MMA.
Crop
(
Chemical)
1
Application
rate
2
(
lbs
ae/
A)
Number
of
Applications
Peak
EEC
(
ppb
as
MMA)

Cotton
(
DSMA)
1.74
2
84
Non­
crop
(
MSMA)
3.86
4
360
Orchard
(
DSMA)
3.70
3
260
Turf;
max
(
MSMA)
2.23
4
310
Turf;
golf
(
MSMA)
3.35
4
210
Predicted
Maximum
EEC
on
Food
Items
Crop
(
Chemical)
1
Appl.
rate
(
lbs
ae/
A)
No.
of
Appl.
Short
grasses
Broadleaf
plants
&
insects
Seeds
Cotton
(
DSMA)
1.74
2
734
413
46
Non­
crop
(
MSMA)
3.86
4
2557
1438
160
Orchard
(
DSMA)
3.70
3
2071
1165
129
Turf;
max
(
MSMA)
3.35
4
2225
1251
139
Turf;
golf
(
MSMA)
2.23
4
1482
833
93
9
2.
Plants
Exposure
to
upland
and
wetland
plants
is
estimated
using
the
TerrPlant
(
v1.0)
screening
model
and
the
EECs
are
presented
in
Table
4.
TerrPlant
estimates
potential
exposure
using
default
assumptions
for
runoff
and
spray
drift.
It
should
be
emphasized
that
TerrPlant
is
only
used
for
estimating
environmental
concentrations
based
on
a
single
application.
Most
uses
of
MSMA
and
DSMA
allow
for
multiple
applications.
TerrPlant
model
details
are
included
in
Appendix
D.

Table
4.
EECs
(
lb
MMA/
A)
in
off­
site
terrestrial
environments
resulting
from
drift
and
runoff
from
single
applications
of
DSMA
and
MSMA.

1
The
pesticide
with
the
highest
application
rate
for
each
use
was
modeled,
both
as
the
acid
equivalent
MMA.
2
All
application
rates
are
based
on
ground
application.
The
cotton
use
is
labeled
for
aerial
application
as
well
but
a
higher
rate
is
allowed
for
ground
application.
3
Sheet
runoff
+
drift.
4
Channelized
runoff
+
drift.

3.
Soil
Accumulation
Long
term
accumulation
of
pesticides
in
soil
is
not
generally
considered
in
ecological
risk
assessments.
Organic
arsenicals
are
unique,
however,
due
to
the
elemental
nature
of
arsenic.
Arsenic
does
not
degrade;
along
with
arsenicals'
relative
immobility
in
soil,
this
makes
long
term
impacts
an
important
consideration.
Field
studies,
monitoring
of
soil
in
use
areas,
and
modeling
all
suggest
that
it
is
likely
that
applied
arsenicals
will
build
up
in
soil
over
time.
Arsenic
accumulation
is
likely
to
be
limited
to
the
top
layers
of
soil,
with
studies
suggesting
that
it
is
unlikely
to
occur
at
depths
greater
than
30
cm.
A
more
detailed
discussion
of
the
potential
for
arsenic
to
accumulate
in
soil
is
included
in
the
Environmental
Fate
and
Transport
section
in
Appendix
A.

Registrant
terrestrial
field
dissipation
studies
have
measured
the
impact
of
one
season
of
pesticide
applications
on
soil
concentrations.
Two
studies
at
rates
similar
to
the
maximum
labeled
cotton
and
turf
rates
found
that
a
single
year
of
application
is
unlikely
Runoff
(
lb
ae/
A)
Total
Load
Crop
(
Chemical)
1
Appl.
rate2
(
lbs
ae/
A)
Upland3
Semi­
Aquatic4
Drift
only
(
lb
ae/
A)
Upland3
Semi­
Aquatic4
Cotton
(
DSMA)
1.74
0.087
0.87
0.017
0.104
0.887
Non­
crop
(
MSMA)
3.86
0.193
1.93
0.039
0.232
1.969
Orchard
(
DSMA)
3.70
0.185
1.85
0.037
0.222
1.887
Turf;
golf
(
MSMA)
2.23
0.167
1.67
0.033
0.201
1.709
Turf;
max
(
MSMA)
3.35
0.111
1.11
0.022
0.134
1.137
10
to
lead
to
significantly
elevated
soil
arsenic
levels.
Since
most
of
the
applied
arsenic
remained
in
the
top
6
inches
of
soil,
however,
repeated
application
may
lead
to
significant
accumulation.
Higher
application
rates
of
DMA,
possible
in
some
of
the
non­
crop
uses,
led
to
elevated
soil
arsenic
levels
after
a
single
year
of
application.
Few
studies
have
evaluated
the
longer
term
impact
of
repeated
application
of
organic
arsenicals.
Those
that
have
been
conducted
have
conflicting
results,
with
some
reporting
no
arsenic
buildup
despite
very
high
application
rates
and
others
finding
substantial
buildup
at
rates
similar
to
current
labels.
Most
reports
do
not
contain
adequate
explanations
to
account
for
observed
loss
of
arsenic.
Monitoring
of
soil
in
areas
where
arsenicals
are
known
to
have
been
applied,
including
golf
courses
and
roadside
areas,
has
found
significant
increases
of
arsenic
levels
relative
to
background
concentrations.
In
Miami,
an
area
with
low
background
arsenic
levels,
arsenic
was
the
most
common
contaminant
found
in
documented
soil
violations
at
golf
courses.

Soil
accumulation
values
are
generated
by
PRZM
as
part
of
the
process
of
modeling
runoff
concentrations.
PRZM
was
run
with
a
modified
version
of
the
pe4
v01
shell
program
to
estimate
arsenic
accumulation
in
the
top
10
cm
of
soil.
Results
are
presented
in
Table
5.
Maximum
application
rates
for
MMA
and
DMA
on
turf
and
cotton
were
used
and
the
modeling
assumed
median
sorption
(
Kd
=
30
for
MMA
and
Kd
=
20
for
DMA).
The
annual
application
rates
for
DMA
on
non­
crop
areas
and
ornamentals
are
nearly
50%
higher
than
the
turf
rate,
and
so
would
be
expected
to
lead
to
higher
soil
concentrations.
Because
of
limitations
in
the
available
data
and
modeling
capabilities,
soil
concentrations
were
modeled
as
total
arsenic,
rather
than
speciated
forms.
Application
rates
were
therefore
calculated
to
represent
applied
arsenic
and
infinite
half­
lives
were
used
to
capture
all
forms
of
arsenic
that
may
be
present.
Routes
of
dissipation
accounted
for
in
the
modeling
include
runoff,
leaching,
and
soil
erosion;
transformation
to
volatile
species
and
plant
uptake
are
not
accounted
for.
Of
available
scenarios,
PA
turf
and
NC
cotton
scenarios
led
to
the
highest
soil
concentrations.
Because
soil
concentrations
are
based
on
local
application
rather
than
input
from
an
entire
watershed,
use
of
a
percent
cropped
area
(
PCA)
factor
was
unnecessary.

Table
5.
Modeled
soil
concentrations
for
total
arsenic
resulting
from
long
term
application
of
MMA
or
DMA.

Parent
Compound
Use
App.
rate
(
lbs
As/
A)
No.
of
Applications
Chronic
EEC
1
(
ppm,
total
arsenic)

MSMA
Cotton
0.95
2
13
Turf
1.80
4
45
DMA
Cotton
0.65
1
2
Turf
2
(
lawn
edging)
4.18
4
77
1
"
Chronic"
concentrations
are
the
upper
90th
percent
confidence
limit
on
the
annual
average,
or
the
1­
in­
10
year
peak
annual
concentration.
2
Application
rate
based
on
master
label
from
12/
12/
05.
11
IV.
RISK
CHARACTERIZATON
A.
Risk
Estimation
A
risk
quotient
(
RQ)­
based
approach
is
used
in
this
assessment,
comparing
the
ratio
of
exposure
concentrations
to
effects
endpoints
with
predetermined
levels
of
concern
(
LOCs).
Although
risk
is
often
defined
as
the
likelihood
and
magnitude
of
adverse
ecological
effects,
the
risk
quotient­
based
approach
does
not
provide
a
quantitative
estimate
of
likelihood
and/
or
magnitude
of
an
adverse
effect.
The
primary
change
in
these
RQs,
compared
to
those
in
the
2001
RED
(
DP
Barcode
D277233),
is
that
they
are
calculated
using
EECs
based
on
application
rates
from
updated
master
labels.
No
new
toxicity
data
have
been
considered
and
most
toxicity
endpoints
used
in
calculations
have
not
changed.
For
terrestrial
avians
and
small
mammals,
the
LD50s
that
were
used
in
calculating
acute
RQs
was
adjusted
for
the
purity
of
the
test
material.
This
calculation
is
reported
in
the
original
RED,
but
for
those
earlier
calculations,
the
unadjusted
LD50s
were
used.
Further
discussion
of
the
toxicity
can
be
found
in
the
2001
RED.

1.
Non­
target
Aquatic
Animals
Acute
RQs
for
aquatic
organisms,
based
on
the
aquatic
EECs
listed
in
Table
2,
are
shown
below
in
Table
6.
All
calculated
RQs
for
fish
and
aquatic
invertebrates
are
<
0.05
and
below
the
level
of
concern
(
LOC).
(
This
table
updates
Table
10
from
the
2001
RED).

Table
6.
Acute
risk
quotients
for
MMA
exposure
to
aquatic
organisms.

Freshwater
RQ
1
Estuarine/
marine
RQ
1
Crop
(
Chemical)
Peak
EEC
(
ppb)
Fish2
Invert.
3
Fish4
Invert.
5
Cotton
(
DSMA)
84
<
0.05
<
0.05
*
*

Non­
crop
(
MSMA)
360
<
0.05
<
0.05
<
0.05
<
0.05
Orchard
(
DSMA)
260
<
0.05
<
0.05
*
*

Turf;
max
(
MSMA)
310
<
0.05
<
0.05
<
0.05
<
0.05
Turf;
golf
(
MSMA)
210
<
0.05
<
0.05
<
0.05
<
0.05
1
RQ
=
EEC/
LC50
(
fish)
or
EC50
(
invertebrates)
2
LC50s
=
112
ppm
for
DSMA
and
12
ppm
for
MSMA
3
EC50s
=
153
ppm
for
DSMA
and
77.5
ppm
for
MSMA
4
LC50
=
323
ppm
for
MSMA
5
EC50
=
160
ppm
for
MSMA
*
No
toxicity
data
for
DSMA
and
marine/
estuarine
organisms
were
presented,
so
no
RQs
were
calculated.
12
2.
Non­
target
Aquatic
Plants
Risk
quotients
for
aquatic
plants,
based
on
toxicity
values
from
the
original
RED
and
EECs
determined
using
GENEEC2,
are
summarized
below
in
Table
7.
(
This
table
updates
Table
12
from
the
2001
RED).
No
risks
to
non­
endangered
or
endangered
aquatic
plants
exceed
the
level
of
concern
of
1.

Table
7.
Risk
quotients
for
MMA
exposure
to
aquatic
plants.
RQ
1
Crop
(
Chemical)
Peak
EEC
(
ppb)
Non­
endangered
Vascular
spp.
2
Non­
endangered
Non­
vascular3
Endangered
Vascular
spp.
2
Cotton
(
DSMA)
84
<
0.1
<
0.1
<
0.1
Non­
crop
(
MSMA)
360
<
0.1
0.13
<
0.1
Orchard
(
DSMA)
260
<
0.1
0.17
<
0.1
Turf;
max
(
MSMA)
310
<
0.1
0.11
<
0.1
Turf;
golf
(
MSMA)
210
<
0.1
<
0.1
<
0.1
1
RQ
=
EEC
/
EC50
(
non­
endangered
species)
or
NOEC
(
endangered
species)
2
vascular
(
duckweed)
EC50'
s
and
NOECs:
72.7
ppm
&
20.5
ppm
(
DSMA),
53
ppm
&
29
ppm
(
MSMA)
3
non­
vascular
(
algae
or
diatom)
EC50'
s:
1.5
ppm
(
DSMA),
2.8
ppm
(
MSMA)
13
3.
Non­
target
Terrestrial
Animals
Acute
and
chronic
risk
quotients
for
terrestrial
small
mammals
are
summarized
below
in
Table
8.
These
RQs
are
calculated
based
on
toxicity
values
from
the
original
RED,
adjusted
for
the
purity
of
the
test
material,
and
EECs
determined
using
TREX.
(
This
table
updates
Tables
3
and
17
from
the
2001
RED).
These
RQs
are
applicable
to
mammals
with
body
weights
of
35
g
that
consume
green
vegetation
or
insects
equivalent
to
66%
of
their
body
weight
(
herbivores
and
insectivores)
or
seeds
equivalent
to
15%
of
their
body
weight
(
granivores).
The
purpose
of
this
addendum
is
only
to
update
RQs
from
the
2001
RED,
so
RQs
have
not
been
calculated
for
mammals
with
body
weights
of
10
g
or
1000
g,
as
would
be
done
for
a
current
assessment.
Most
of
these
acute
RQs,
excepting
those
for
granivores,
exceed
the
endangered
species
LOC
of
0.1
while
some
also
exceed
the
restricted
use
and
high
risk
LOCs
of
0.2
and
0.5.
All
but
2
of
the
chronic
RQs
exceed
the
chronic
risk
LOC
of
1.

Table
8.
Risk
quotients
for
small
mammals
(
35
g)
from
exposure
to
MMA.

1
Acute
RQ
=
EEC
/
LD50,
corrected
for
body
weight;
LD50s
=
1599
mg/
kg
(
rat)
for
DSMA,
157
mg/
kg
(
rat)
for
MSMA,
adjusted
for
purity
of
test
material.
2
Chronic
RQ
=
EEC/
NOEC,
corrected
for
body
weight;
NOEC
=
100
ppm
(
rat)
for
MMA
***
exceeds
LOCs
for
high
risk
(
0.5),
restricted
use
(
0.2),
and
endangered
species
(
0.1)
**
exceeds
the
LOCs
for
restricted
use
and
endangered
species
*
exceeds
the
LOC
for
endangered
species
*
exceeds
the
chronic
risk
LOC
(
1)
Acute
RQs
1
Chronic
RQs
2
Crop
(
Chemical)
Herbivore
Insectivore
Granivore
Herbivore
Insectivore
Granivore
Cotton
(
DSMA)
0.17*
0.10*
<
0.1
7.3*
4.1*
0.46
Non­
crop
(
MSMA)
6.04***
3.39***
<
0.1
26*
14*
1.6*

Orchard
(
DSMA)
0.48**
0.27**
<
0.1
21*
12*
1.3*

Turf;
max
(
MSMA)
5.25***
2.95***
<
0.1
22*
13*
1.4*

Turf;
golf
(
MSMA)
3.50***
1.97***
<
0.1
15*
8.3*
0.93
14
Acute
RQs
for
avian
species
are
presented
below
in
Table
9.
These
values
are
based
on
toxicity
values
from
the
original
RED,
adjusted
for
the
purity
of
the
test
material,
and
EECs
determined
using
TREX.
(
This
table
updates
Table
4
from
the
2001
RED).
As
with
terrestrial
mammals,
most
RQs,
with
the
exception
of
granivores,
exceed
the
restricted
use
and
endangered
species
LOCs
of
0.2
and
0.1
while
some
also
exceed
the
high
risk
LOC
of
0.5.

Table
9.
Risk
quotients
for
avians
from
exposure
to
MMA.
Acute
RQ
1
Crop
(
Chemical)
Herbivores
Insectivores
Granivores
Cotton
(
DSMA)
0.16*
<
0.1
<
0.1
Non­
crop
(
MSMA)
1.53***
0.86***
0.1*

Orchard
(
DSMA)
0.54***
0.30**
<
0.1
Turf;
max
(
MSMA)
1.33***
0.75***
<
0.1
Turf;
golf
(
MSMA)
0.89***
0.50***
<
0.1
1
RQ
=
EEC
/
LC50;
LC50s
=
4695
mg/
kg
(
DSMA)
and
1667
mg/
kg
(
MSMA),
both
for
northern
bobwhite,
adjusted
for
purity
of
the
test
material.
***
exceeds
LOCs
for
high
risk
(
0.5),
restricted
use
(
0.2),
and
endangered
species
(
0.1)
**
exceeds
the
LOCs
for
restricted
use
and
endangered
species
*
exceeds
the
LOC
for
endangered
species
15
4.
Non­
target
Terrestrial
and
Semi­
Aquatic
Plants
Risk
quotients
for
terrestrial
and
semi­
aquatic
plants
exposed
to
drift
and/
or
runoff
are
summarized
below
in
Table
10.
These
are
based
on
toxicity
values
from
the
2001
RED
and
EECs
determined
using
TerrPlant.
Most
RQs
for
endangered
and
non­
endangered
plants,
both
upland
and
semi­
aquatic,
exceed
the
LOC
of
1
for
exposure
from
runoff
and
drift.
None
of
the
drift
only
RQs
exceed
the
LOC.

Table
10.
Risk
quotients
for
terrestrial
and
semi­
aquatic
plants
from
a
single
application
of
DSMA
or
MSMA.

1
RQ
=
EEC
/
EC25.
For
total
loading
use
seedling
emergence
EC25
(
1.25
and
0.116
lb
ai/
A
for
DSMA
and
MSMA,
respectively).
For
drift
use
vegetative
vigor
EC25
(
0.354
and
0.418
lb
ai/
A
for
DSMA
and
MSMA,
respectively)
2
RQ
=
EEC
/
NOEC.
For
total
loading
use
seedling
emergence
NOEC
(
0.30
and
0.018
lb
ai/
A
for
DSMA
and
MSMA,
respectively).
For
drift
use
vegetative
vigor
NOEC
(<
0.30
and
0.14
lb
ai/
A
for
DSMA
and
MSMA,
respectively)
3
Upland
EEC
based
on
sheet
runoff
+
drift;
Semi­
Aquatic
EEC
based
on
channelized
runoff
+
drift.
*
exceeds
the
LOC
(
RQ
>
1)
for
nontarget
plants
Non­
Endangered
RQs
1
Endangered
RQs
2
Crop
(
Chemical)
Upland3
Semi­
Aquatic
3
Drift
Only
Upland3
Semi­
Aquatic
3
Drift
Only
Cotton
(
DSMA)
<
1
<
1
<
1
<
1
3*
<
1
Non­
crop
(
MSMA)
2*
17*
<
1
13*
109*
<
1
Orchard
(
DSMA)
<
1
1.5*
<
1
<
1
6*
<
1
Turf;
golf
(
MSMA)
1.7*
15*
<
1
11*
95*
<
1
Turf;
max
(
MSMA)
1.1*
9.8*
<
1
7.4*
63*
<
1
16
5.
Soil
Accumulation
Because
it
is
not
typical
for
EFED
to
assess
environmental
risk
resulting
from
accumulation
of
pesticides
in
soil,
there
are
no
established
methods
for
calculating
RQs
based
on
this
type
of
exposure.
The
EPA's
Office
of
Solid
Waste
and
Emergency
Response
(
OSWER)
has
established
procedures
for
dealing
with
the
issue
of
contaminated
soils.
Rather
than
calculating
RQs
resulting
from
accumulation
of
arsenic
in
soil,
EFED
has
compared
the
estimated
soil
concentration
to
ecological
soil
screening
levels
(
SSL)
for
arsenic
set
by
OSWER
in
March,
2005
(
OSWER,
1995).
SSLs
are
defined
as
"
concentrations
of
contaminants
in
soil
that
are
protective
of
ecological
receptors
that
commonly
come
into
contact
with
soil
or
ingest
biota
that
live
in
or
on
soil."
More
information
about
how
these
levels
were
set
can
be
obtained
at
http://
www.
epa.
gov/
ecotox/
ecossl/
pdf/
eco­
ssl_
arsenic.
pdf.

Table
11
presents
the
SSLs
set
to
protect
plants,
birds,
and
mammals.
Insufficient
data
were
available
to
set
an
SSL
for
soil
invertebrates.
Arsenic
SSLs
are
based
on
concentrations
of
total
arsenic.
The
soil
concentrations
estimated
using
PRZM,
also
as
total
arsenic,
are
included
for
comparison.
Application
of
either
MMA
or
DMA
at
the
maximum
rate
for
cotton
does
not
lead
to
exceedance
of
any
SSLs,
but
the
maximum
rates
of
MMA
or
DMA
applied
to
turf
exceed
SSLs
for
all
groups.

Table
11.
Soil
EECs
for
MSMA
and
DMA
compared
to
OSWER
arsenic
SSLs.

Parent
Compound
Use
Chronic
EEC
(
ppm,
total
arsenic)
OSWER
Eco­
SSLs
(
mg/
kg
dry
weight)

MSMA
Cotton
13
Plants
18
Turf
45
Avian
Wildlife
43
DMA
Cotton
2
Mammalian
Wildlife
46
Turf
(
lawn
edging)
77
Soil
Invertebrates
NA
B.
Risk
Description
1.
Risks
to
Aquatic
Organisms
Use
of
MSMA
and
DSMA
present
minimal
risk
to
aquatic
species,
including
fish,
invertebrates,
and
plants.
In
the
original
assessment,
the
one
exception
found
to
this
conclusion
was
risk
to
freshwater
fish
from
the
very
high
application
rate
of
MSMA
to
orchards.
The
lower
application
rates
on
the
current
master
label
have
reduced
this
risk
quotient
below
the
LOC.
17
2.
Risks
to
Terrestrial
Organisms
With
the
lower
application
rates,
most
terrestrial
risk
quotients
(
RQs)
have
dropped
but
the
overall
risk
conclusions
have
not
changed.
Application
of
organic
arsenicals
presents
risk
to
all
categories
of
terrestrial
organisms,
including
small
mammals,
avian
species,
and
upland
and
semi­
aquatic
plants.
Because
of
lower
applications
rates,
in
many
cases
the
exceedances
are
not
as
great
as
in
the
original
assessment.
Additionally,
there
are
fewer
exceedances
in
the
highest
risk
category.

3.
Federally
Threatened
and
Endangered
(
Listed)
Species
Concerns
The
endangered
species
portion
of
this
assessment
has
not
been
updated.
The
original
assessment
found
that
application
of
DSMA
and
MSMA
present
risk
to
endangered
species
for
birds,
mammals
and
terrestrial
plants.
There
are
still
exceedances
in
each
of
these
categories.
The
original
assessment
found
risk
to
endangered
freshwater
fish
from
the
high
application
rate
for
MSMA
on
orchards.
Limiting
this
application
rate
lowered
the
risk
for
freshwater
fish
below
the
LOC.

The
Agency's
preliminary
risk
assessment
for
endangered
species
indicates
that
organic
arsenic
applied
as
a
pesticide
results
in
a
determination
of
"
no
effect"
to
listed
aquatic
animal
and
plant
species
on
an
acute
or
chronic
basis.
RQs
exceed
endangered
species
LOCs
for
terrestrial
plants,
birds,
and
mammals.
These
findings
are
based
solely
on
EPA's
screening
level
assessment
and
do
not
constitute
"
may
affect"
findings
under
the
Endangered
Species
Act.

4.
Uncertainties
The
uncertainties
of
this
risk
assessment
were
discussed
in
the
original
assessments.
Further
discussion
of
the
uncertainty
regarding
the
environmental
fate
and
transport
properties
of
organic
arsenicals
has
been
included
in
Appendix
A,
the
Environmental
Fate
and
Transport
section.
Uncertainties
in
the
modeling
input
parameters
are
discussed
in
the
organic
arsenicals
Drinking
Water
Assessment
(
DP
Barcode
D309098).

One
additional
source
of
uncertainty
is
relevant
to
this
addendum.
All
exposure
estimates
for
MSMA
and
DSMA
in
this
document
are
based
on
application
rates
as
MMA,
the
acid
equivalent
of
the
methanearsonate
salts.
For
determining
terrestrial
risk,
in
some
cases
it
is
unclear
which
form
of
the
parent
compound
was
used
as
the
active
ingredient
used
to
determine
toxicity.
Some
of
these
endpoints
may
have
been
based
on
concentrations
as
MSMA
rather
than
as
MMA.
If
the
toxicity
is
based
on
arsenic
as
MSMA,
the
units
of
the
toxicity
estimate
would
have
to
be
converted
MMA
to
be
appropriately
compared
to
the
exposure
as
MMA.
This
would
lead
to
a
lower
toxicity
value
and
therefore
a
higher
risk
quotient
and
a
higher
estimate
of
risk.
It
is
possible,
then,
that
risk
was
underestimated
in
some
of
the
terrestrial
categories.
Even
if
true,
though,
this
would
not
lead
to
significant
changes
in
the
risk
conclusions.
18
APPENDIX
A:

ENVIRONMENTAL
FATE
OF
ORGANIC
ARSENICALS
19
Physicochemical
Properties
MSMA,
DSMA,
and
CAMA
are
salts
of
the
dibasic
weak
acid
monomethylarsonic
acid
(
MMA;
also
variously
abbreviated
in
other
documents
and
the
published
literature
as
MMAA
or
MAA,
for
methanearsonic
acid).
In
aqueous
solution
these
compounds
dissociate
into
MMA
and
the
associated
ions,
either
sodium
or
calcium.
Cacodylic
acid
(
DMA,
for
dimethylarsonic
acid)
is
a
weak
acid
with
two
methyl
groups
attached
to
the
central
arsenic
atom,
rather
than
one
as
in
MMA.
In
some
formulations,
DMA
is
mixed
with
its
sodium
salt
(
DMA­
Na),
which
in
aqueous
solution
is
also
dissociated
to
DMA
and
a
sodium
ion.
Chemical
structures
and
some
physicochemical
properties
for
MSMA,
DSMA,
DMA,
and
DMA­
Na
are
presented
in
Tables
A1a
and
A1b
below.
This
group
of
compounds
is
referred
to
as
"
organic
arsenicals"
throughout
this
document.
Most
environmental
fate
studies
have
been
conducted
on
MSMA
or
DMA
as
representatives
of
the
monomethylated
and
dimethylated
species,
respectively.

Table
A1a.
Physicochemical
Properties
for
MSMA
and
DSMA
DSMA
MSMA
Molecular
Structure
As
O
O
O
C
H
3
Na
+
Na
+
As
OH
O
O
C
H
3
Na
+

Empirical
Formula
CH3AsNa2O3
CH4AsNaO3
Molecular
Weight
183.92
161.94
CAS
No.
144­
21­
8
2163­
80­
6
PC
Code
013802
013803
Melting
Point
(
º
C)
>
300
116­
121
Density
(
g/
mL)
1.04
1.65
Vapor
Pressure
(
mm
Hg)
1
x
10­
7
7.5
x
10­
7
log
Kow
<
1
<
1
Solubility:
Water
(
mg/
L)
3.4
x
105
104
Methanol
2.6
x
105
16
Hexanol
25
0.005
pKa1,2
(
approx.)
4.0,
9.0
4.0,
9.0
20
Table
A1b.
Physicochemical
Properties
for
Cacodylic
Acid
Cacodylic
Acid
(
DMA)
Sodium
Cacodylate
(
DMA­
Na)

Molecular
Structure
O
||
CH3 
As 
OH
|
CH3
O
||
CH3 
As 
O­
Na+
|
CH3
Empirical
Formula
C2H7AsO2
C2H6AsNaO2
Molecular
Weight
138.0
160.0
CAS
No.
75­
60­
5
124­
65­
2
PC
Code
012501
012502
Melting
Point
(
º
C)
192­
194
77
 
79.5
Density
(
g/
mL)
1.10
1.10
Vapor
Pressure
(
mm
Hg)
Non­
volatile
No
data
Kow
<
0.028
No
data
Solubility:
Water
(
mg/
L)
~
1
to
3
x
106
No
data
Methanol
3.63
x
105
No
data
Hexanol
1.02
x
10­
1
No
data
pKa
6.2
6.2
As
weak
acids,
the
form
in
which
MMA
and
DMA
are
present
is
dependent
on
pH,
which
affects
the
degree
of
association
with
hydrogen
ions
and
therefore
the
charge
of
the
species.
For
MMA,
the
three
possible
species
are
symbolized
H2MMA,
HMMA1­,
and
MMA2­.
Reported
titration
studies
for
MSMA
and
DSMA
yield
base
strengths
for
the
anions
corresponding
to
pKa
values
for
the
acid
MMA
of
4
and
9.
Under
agriculturally
relevant
conditions,
then,
HMMA1­
would
dominate
at
pH
5
to
7
as
approximately
90%
to
99%
of
the
total,
while
at
pH
9,
it
would
be
approximately
equal
to
the
amount
present
as
MMA2­.
DMA
can
be
present
as
neutral
cacodylic
acid
or
as
the
cacodylate
anion
and
has
a
pKa
of
6.2.
At
a
pH
of
5,
then,
neutral
cacodylic
acid
dominates
at
approximately
94%
of
the
total,
while
at
pHs
of
7
to
9,
the
anionic
species
is
dominant,
going
from
87%
of
the
total
to
more
than
99%.
In
the
environment,
chemical
speciation
is
in
dynamic
flux,
depending
on
variations
in
pH
and
affected
by
transport,
mixing,
diffusion,
etc.,
in
accordance
with
principles
of
chemical
equilibria
and
kinetics.
21
Degradation
and
Metabolism
(
Rate
and
Environmental
Conditions)
Environmental
fate
laboratory
studies
show
that
organic
arsenicals
are
stable
under
all
tested
abiotic
conditions.
Registrant
submitted
studies
of
DMA
and
MMA
found
both
compounds
to
be
stable
to
hydrolysis
at
all
pHs
(
MRIDs
42059201
and
42363001).
DMA
and
MMA
were
also
found
to
be
stable
to
photolysis
in
both
aquatic
and
soil
environments
(
DMA:
MRIDs
41662601
&
41662602;
MMA:
MRIDs
41903902
&
41903901).

Organic
arsenicals
can
be
subject
to
microbial
metabolism
in
soil
under
aerobic
or
anaerobic
conditions.
The
occurrence,
rate,
and
products
of
this
metabolism
are
variable,
dependent
on
environmental
conditions.
The
observed
persistence
of
organic
arsenicals
in
aerobic
soil
has
ranged
from
weeks
to
years,
depending
on
soil
properties
and
ambient
conditions
such
as
soil
moisture,
temperature,
chemical
concentration,
and
amount
of
organic
matter.
The
extreme
of
this
range
is
seen
in
several
registrant
submitted
studies
of
microbial
metabolism
which
observed
no
transformation
at
all
(
MRIDs
42616001,
42572601,
43036101).
These
studies
were
determined
to
be
scientifically
sound
but
were
found
to
be
not
fully
acceptable
because
they
provided
no
explanation
for
this
result
in
light
of
well
established
evidence
of
organic
arsenical
metabolism.
Although
these
studies
demonstrate
that
there
are
conditions
in
which
metabolism
does
not
occur,
possibly
due
to
non­
viable
soils,
this
assessment
is
based
on
the
understanding
that
transformation
of
organic
arsenicals
is
an
important,
although
variable,
process.

Adding
to
the
complexity
is
that
metabolism
rates
do
not
appear
to
depend
linearly
on
organic
arsenical
concentration;
in
some
studies,
transformation
decreased
with
increasing
concentration
while
in
others,
concentration
had
no
effect
on
metabolism.
Hence,
the
kinetics
are
not
necessarily
first­
order,
and
"
half­
life"
is
therefore
not
necessarily
an
appropriate
constant
for
all
concentrations.
Keeping
this
uncertainty
in
mind,
first­
order
half­
lives
have
been
calculated
as
required
by
EFED's
current
models
and
also
as
a
convenient
measure
to
compare
results
from
laboratory
studies.
The
Agency
derived
values
for
a
DMA
half­
life
from
four
studies
representing
a
total
of
six
soils.
Together,
these
yielded
an
effective
average
aerobic
soil
half­
life
for
DMA
of
173
+
115
days
with
a
standard
upper
90%
confidence
limit
on
the
mean
of
240
days
(
Woolson
and
Kearney,
1973;
Woolson,
1982;
Gao
and
Burau,
1997;
and
MRID
44767601).
The
average
half­
life
of
DMA
in
three
anaerobic
(
flooded)
soils
was
128
+
38
days
with
a
standard
upper
90%
confidence
limit
on
the
mean
of
168
days
(
Woolson
and
Kearney,
1973).
For
MMA,
two
studies
in
two
soils
both
resulted
in
aerobic
soil
half­
lives
of
approximately
240
days
(
Gao
and
Burau,
1997;
MRID
44767601).
No
anaerobic
soil
half­
life
was
determined
for
MSMA.
The
studies
on
which
these
estimates
are
based
are
discussed
in
more
detail
below
and
the
calculations
are
also
described
in
the
surface
water
modeling
section
of
the
Drinking
Water
Assessment
(
DP
Barcode
D309098).

Some
of
the
variability
in
metabolism
processes
is
associated
with
variability
in
sorption
processes.
Soil
microbial
metabolism
of
organic
arsenicals
only
occurs
while
the
compounds
remain
dissolved
in
pore
water
(
NRC,
2003).
As
the
organic
arsenicals
sorb
22
to
soil,
they
become
less
accessible
to
microbes
and
therefore
less
likely
to
be
metabolized
(
Woolson
and
Kearney,
1973).
Sorption
variability,
discussed
in
more
detail
in
the
mobility
section
(
p.
28),
is
largely
controlled
by
soil
properties
including
the
clay
content,
the
iron
and
aluminum
content,
and
the
soil
pH
(
Wauchope,
1975;
Matera
and
La
Hecho,
2001).
Laboratory
studies
have
shown
that
in
some
situations,
significant
sorption
of
arsenic
compounds
may
occur
within
hours
of
application,
while
in
others,
a
large
portion
of
applied
arsenic
remains
in
water­
soluble
forms
for
days
or
months
after
application
(
Onken
and
Adriano,
1997;
Sarkar
et
al.,
2005).
Remobilization
of
sorbed
arsenic
with
changing
environmental
conditions
is
also
possible
(
Matera
and
La
Hecho,
2001).

The
effects
of
other
environmental
factors
on
the
rate
of
organic
arsenical
metabolism
are
complex
and
poorly
defined
with
different
studies
leading
to
conflicting
results.
An
increase
in
temperature
has
been
shown
to
lead
to
increased
metabolism
(
Akkari
et
al.,
1986;
Gao
and
Burau,
1997)
but
results
on
the
impact
of
soil
organic
matter
or
applied
organic
arsenical
concentrations
are
contradictory
(
Dickens
and
Hiltbold,
1967;
Von
Endt
et
al.,
1968;
Woolson
and
Kearney,
1973;
Woolson,
1982;
Akkari
et
al.,
1986;
Gao
and
Burau,
1997).

The
influence
of
aerobic
versus
anaerobic
conditions
on
metabolism
rates
is
also
ambiguous.
Based
on
a
comparison
of
the
results
of
studies
of
MMA
applied
to
an
aerobic
soil
and
to
an
anaerobic
sediment/
aquatic
system,
the
registrants
conclude
that
aerobic
metabolism
is
more
significant
than
anaerobic
metabolism
(
MAATF,
2005).
After
one
year
of
incubation,
35%
of
the
parent
MMA
remained
in
aerobic
soil
(
MRID
44767601),
while
in
the
anaerobic
system,
only
3%
of
applied
MMA
had
been
metabolized
(
MRID
44767602).
Both
studies
ended
with
a
significant
amount
of
unidentified
arsenic
compounds
bound
to
the
soil
(
10­
25%).
Two
studies
by
Woolson
also
concluded
that
greater
metabolism
occurred
in
aerobic
than
in
anaerobic
conditions.
In
one
study,
an
average
of
24%
of
14C
from
applied
DMA
remained
after
24
weeks
incubation
of
100
ppm
DMA
in
three
aerobic
soils
while
39%
remained
in
flooded
soils
(
Woolson,
1973).
In
the
other
study,
the
amounts
of
applied
DMA
remaining
after
60
days
were
similar
at
soil
moisture
levels
between
77%
and
125%
of
field
capacity
but
in
flooded
soils
(
230%
field
capacity),
more
parent
compound
remained,
indicating
slower
metabolism
in
anaerobic
conditions
(
Woolson,
1982).

Two
more
recent
studies
led
to
the
opposite
result,
finding
that
metabolism
of
organic
arsenicals
is
more
significant
in
anaerobic
conditions.
In
3
soils
incubated
at
30
°
C
for
120
days,
Akkari
found
half­
lives
in
flooded
conditions
(
150%
field
capacity)
of
approximately
1
month
while
in
low­
moisture
conditions,
(
20%
field
capacity),
the
halflives
were
2
to
5
months.
One
of
these
soils
was
also
tested
at
75%
field
capacity
and
found
to
have
a
half­
life
between
the
flooded
and
low­
moisture
soils.
In
this
study,
the
reported
half­
life
also
includes
"
deactivation
by
processes
such
as
salt
formation,
irreversible
absorption,
and
ion­
exchange"
(
Akkari
et
al,
1986).
Gao
and
Burau
(
1997)
incubated
one
soil
at
a
range
of
soil
moistures
for
70
days
and
found
significantly
higher
mineralization
in
flooded
conditions
(>
350
g
H20/
kg
soil).
In
soil
with
moisture
levels
 
350
g
H20/
kg
soil,
88%
to
97%
of
applied
DMA
was
recovered
as
parent
compound,
23
while
for
flooded
soils,
recovered
parent
compound
was
as
low
as
13%,
indicating
a
significant
increase
in
metabolism
in
anaerobic
conditions.

Of
the
studies
reviewed
for
this
document,
the
most
complete,
systematic
investigation
of
the
metabolism
of
organic
arsenicals
under
variable
conditions
is
by
Gao
and
Burau
(
1997).
In
particular,
this
study
is
valuable
because
the
products
are
speciated,
including
trapping
of
volatiles,
with
very
good
recovery.
The
mass
balance
is
greater
than
90%,
even
for
inorganic
arsenic,
so,
unlike
some
studies
reviewed
here,
these
half­
lives
include
only
metabolism
and
not
other
routes
of
dissipation
such
as
sorption.
Gao
and
Burau
were
motivated
by
noticing
differences
in
published
results
on
transformation
processes
(
rates
and
products)
and
by
a
concern
for
the
potential
for
arsenic
to
accumulate
in
soils.
The
extent
to
which
an
applied
organic
arsenical
and
its
arsenic
containing
byproducts
accumulate
locally
depends
on
both
mineralization
and
volatilization.
The
1997
Gao
and
Burau
study
serves
to
systematically
complement,
and,
to
a
great
extent,
unify
existing
data.
The
study
is
designed
to
measure
the
influence
of
four
factors­­
concentration,
soil
moisture,
temperature,
and
soil
amendment
with
organic
carbon
(
cellulose)­­
on
the
rates
and
routes
of
transformation
of
organic
arsenicals
in
soil.
Gao
and
Burau
primarily
tested
DMA
in
the
form
of
sodium
cacodylate,
although
production
of
volatile
arsines
was
measured
for
MMA,
sodium
arsenate
(
AsV),
and
sodium
arsenite
(
AsIII),
and
one
treatment
compared
the
metabolism
rates
of
DMA
and
MMA
under
the
same
conditions.
The
study
was
limited
to
one
California
soil,
a
Sacramento
silty
clay
(
noncalcareous).

Gao
and
Burau's
results
show
the
great
influence
of
soil
environmental
conditions
on
transformation
rates.
After
70
days
of
incubation,
depending
on
conditions,
the
amount
of
parent
cacodylate
remaining
ranged
from
a
high
of
about
97%
to
a
low
of
about
13%
1.
Under
all
conditions,
production
of
volatile
arsines
was
less
than
0.5%
of
the
applied
arsenic.
Arsenite
was
not
a
detected
product
under
these
conditions
and
both
DMA
and
MMA
metabolized
without
detection
of
methylation
or
demethylation.
The
amount
of
parent
compound
remaining
at
the
end
of
the
study
depended
strongly
on
all
tested
factors.
Unless
noted
otherwise,
the
following
tests
were
all
carried
out
with
DMA
at
100
mg
As/
kg
soil
and
soil
moisture
content
of
350
g
water/
kg
soil
(
approximately
1/
3­
bar
suction
[­
0.03
Mpa]).

1)
Soil
Moisture.
As
mentioned
above,
metabolism
of
DMA
was
measured
at
five
soil
moisture
contents
of
50,
250,
350,
450,
and
550
g
water/
kg
soil
[
from
approximately
7%
(­
23.7
MPa
suction)
to
81%
(­
0.0005
MPa)
of
saturation],
at
22
oC.
Mineralization
to
arsenate
increased
with
soil
moisture
from
2.7%
at
the
lowest
soil
moisture
to
86.6%
at
the
highest,
showing
the
tremendous
influence
of
soil
water.

1
Gao
and
Burau
report
some
of
their
results
as
"
percent
mineralized",
calculated
as
100%
minus
the
measured
percent
of
parent
compound
remaining.
The
authors
explain
this
as
an
assumption
for
the
sake
of
consistency.
The
reported
percent
mineralized,
then,
includes
the
<
0.5%
volatilized
as
well
as
any
unrecovered
arsenic.
The
actual
concentrations
of
inorganic
arsenic
resulting
from
degradation
were
measured
directly
with
91%
recovery,
so
9%
is
the
most
unrecovered
arsenic
which
may
be
accounted
for
in
the
reported
demethylation.
In
this
discussion,
in
order
to
avoid
this
uncertainty,
the
percentage
of
parent
compound
remaining
was
determined
based
on
the
reported
percent
mineralized.
24
2)
Temperature.
At
5
and
25
oC,
mineralization
was
approximately
5.2
and
23.8%,
respectively.
This
result
is
roughly
as
would
be
predicted
from
the
Arrhenius
relationship
which
yields
rate
doubling
for
every
10
oC
increase
in
temperature.
[
The
23.8%
mineralization
result
can
be
compared
with
the
result
of
12.1%
mineralization,
from
the
soil
moisture
series,
as
a
measure
of
variability
under
essentially
the
same
conditions,
except
for
an
uncorrected
3
oC
difference
in
temperature.]

3)
Concentration.
At
increasing
cacodylate
concentrations
equivalent
to
arsenic
concentrations
of
10,
30,
and
100
mg
As/
kg
soil
at
25
oC,
decreasing
percentages
of
cacodylate
mineralized
were
approximately
82,
31
and
24%,
respectively,
indicating
that
the
rate
process
is
not
first­
order
in
concentration.
One
possible
reason
for
the
slowing
of
metabolism
at
increasing
concentrations
could
be
toxicity
to
soil
microorganisms,
however,
the
two
upper
concentrations
correspond
to
application
rates
much
higher
than
those
currently
labeled.

4)
Organic
Matter.
Increasing
cellulose
additions
of
0.0
(
unamended),
0.2,
1.0,
and
5.0
g
per
100
g
of
soil
(
original
soil
organic
carbon
content
of
1.8%)
with
sodium
cacodylate
added
at
the
arsenic
equivalent
of
10
mg
As/
kg
soil
at
250C
decreased
mineralization
from
around
77%
down
to
around
49%.
(
As
a
measure
of
variability
in
this
experiment
under
the
same
conditions,
the
first
value
of
77%
can
be
compared
directly
with
the
82%
result
from
the
concentration
experiment
above.)
Some
reports
show
the
same
trend
of
decreasing
mineralization
with
increasing
concentration
of
organic
matter
(
e.
g.,
Woolson,
1982),
but
others
report
the
opposite
trend
(
e.
g.,
Dickens
and
Hiltbold,
1967).
Types
of
added
organic
matter
were
generally
different,
and
could
perhaps
account
for
the
differences.
However,
in
view
of
the
sensitivity
of
metabolism
to
the
other
cited
factors,
lack
of
extremely
careful
control
of
experimental
conditions
could
also
be
a
major
factor
for
the
difference.

5)
Arsenical
Species.
Metabolism
rates
of
MMA
and
DMA
were
compared
by
applying
both
at
100
mg
As/
kg
soil
under
the
same
environmental
conditions.
After
70
days
at
220C,
more
than
twice
as
much
DMA
as
MMA
had
metabolized.
Parent
DMA
remained
at
only
26.7%
of
the
applied,
while
57.1%
of
the
applied
MMA
remained.

Using
the
Gao
and
Burau
data
to
estimate
an
"
effective
half­
life"
for
DMA
(
realizing
that
the
process
is
apparently
not
first­
order
with
concentration),
the
Agency
calculated
halflives
for
each
trial,
based
on
the
initial
and
final
concentrations
of
parent
compound.
These
were
interpolated
as
a
function
of
soil
moisture
at
the
tested
arsenic
equivalent
concentration
of
100
mg/
kg
soil
to
the
standard
75%
of
1/
3­
bar
soil
moisture
content
using
the
soil
moisture
retention
function
(
the
logarithm
of
the
absolute
value
of
soil
tension
vs.
soil
water
concentration)
which
the
authors
gave
for
the
Sacramento
silty
clay
soil.
The
interpolation
gives
a
moisture
concentration
of
approximately
27%,
or
266
g
water
per
kilogram
of
dry
soil,
and
a
corresponding
DMA
half­
life
of
642
days
(
1.8
years).
In
tandem,
the
Agency
then
made
a
simple,
proportionate
adjustment
of
the
rate
constant
corresponding
to
this
half­
life
at
the
100
mg/
kg
concentration
to
the
more
agriculturally
relevant
10
mg/
kg
arsenic
equivalent
(
18
mg/
kg
cacodylic
acid).
This
range
is
relatively
low
when
compared
to
most
study
concentrations
which
were
typically
25
10
to
180
ppm
in
cacodylic
acid.
This
calculation
leads
to
a
normalized
DMA
half­
life
(
at
75%
of
1/
3­
bar
soil
moisture
and
an
arsenic
equivalent
concentration
of
10
mg/
kg)
of
102
days.
The
more
limited
data
for
MMA
was
converted
by
comparison
to
the
DMA
normalization,
resulting
in
a
normalized
half­
life
of
241
days.
Additional
reports
contributing
to
the
discussion
of
organic
arsenical
soil
metabolism
include
two
published
laboratory
studies
authored
by
Woolson
and
others2.
In
the
first,
14C­
cacodylic
acid
was
applied
at
three
concentrations
(
1,
10
and
100
ppm
DMA;
arsenic
equivalent
0.5,
5,
and
54
ppm)
to
each
of
three
soils
of
varying
iron
and
aluminum
content
(
Woolson
and
Kearney,
1973).
Soil
moistures
were
brought
to
75%
of
field
capacity
and
the
temperature
was
25
0C.
For
several
reasons
this
study
does
not
meet
current
guideline
standards
but
was
not
specifically
cited
as
deficient
in
the
1986
Agency
review
and
provides
valuable
information.
The
authors
found
that
"
rate
of
application
had
no
appreciable
effect
on
disappearance"
and
so
reported
the
amount
of
14C
remaining
in
each
soil
after
32
weeks
as
an
average
from
the
three
application
rates.
Based
on
these
final
values,
the
Agency
calculated
assumed
first­
order
aerobic
half­
lives
of
325,
245,
and
106
days,
from
a
loamy
sand,
a
silty
clay
loam,
and
a
clay
loam,
respectively.
These
are
considerably
longer
than
the
half­
lives
reported
in
the
2001
RED
(
117,
89,
and
48
days;
DP
Barcodes
D210451,
D212449,
D255226),
which
were
calculated
based
on
the
14C
remaining
at
24
weeks
rather
than
32
weeks.
Recalculation
based
on
the
complete
length
of
the
study
led
to
these
longer
half­
lives
which
are
more
representative
of
the
long
term
behavior
of
applied
organic
arsenicals
and
closer
to
the
guideline
standard
of
a
study
length
of
one
year.
Less
data
were
reported
for
anaerobic
(
flooded)
soil
metabolism
and
so
half­
lives
were
calculated
based
on
data
from
24
weeks
for
cacodylic
acid
applied
at
100
ppm.
Apparent
anaerobic
half­
lives,
as
reported
in
the
2000
RED,
were
86,
138,
and
159
days.
Due
to
the
different
dataset
used,
these
half­
lives
are
not
directly
comparable
to
the
calculated
aerobic
half­
lives.
They
would
be
expected
to
follow
the
same
trend,
with
the
cacodylic
acid
metabolizing
more
slowly
over
time.
Again,
these
estimated
results
are
dependent
on
professional
judgment
and
interpretation.
The
nature
of
transformation
products
revealed
in
this
and
other
studies
is
integrated
below
in
the
"
metabolites"
subsection.

Woolson
et
al
(
1982)
applied
cacodylate
at
an
initial
arsenic
equivalent
concentration
of
10
ppm
to
a
Mattapeake
silt
loam
soil
treated
under
several
different
conditions
of
soil
moisture
(
77­
230%
of
field
capacity)
with
unamended
soil
and
with
a
variety
soil
amendments.
After
60
days
of
aerobic
incubation
at
25
0C,
the
residual
cacodylic
acid
was
reported
to
average
approximately
15­
30%.
Results
were
similar
under
all
conditions.
Based
on
these
published
data,
the
estimated
average
aerobic
soil
half­
life
(
assumed
first­
order)
for
all
unamended
treatments
was
approximately
20
days.
The
estimated
half­
life
from
amended
soil,
with
a
higher
organic
matter
content,
was
approximately
31
days.

2
These
studies
were
submitted
under
Accession
No.
259582
and
associated
Accession
Nos.
260061
and
260782.
The
study
by
Woolson
and
Kearney
(
1973)
was
specifically
evaluated
at
the
time
but
Woolson
et
al,
1982
was
not.
Both
studies
were
discussed
in
the
1986
Agency
review
and
in
the
2000
RED
for
cacodylic
acid.
26
Again,
none
of
these
published
studies
meet
current
Guideline
and
GLP
criteria.
Halflife
estimates
for
all
were
based
on
single
time
intervals,
rather
than
complete
time
series.
The
Woolson
studies
generally
had
poor
recoveries
with
the
loss
accounted
for
through
inferred
metabolism.
These
studies
appeared
formally
in
peer­
reviewed
scientific
publications,
however,
as
is
normal
for
most
scientific
work
outside
of
the
regulatory
process.
They
have
been
cited
in
the
literature
numerous
times
and
they
were
submitted
by
the
registrant
in
the
past.
These
studies
provide
a
body
of
evidence
on
the
environmental
fate
of
organic
arsenicals
that
cannot
be
dismissed.

One
GLP
aerobic
soil
metabolism
study
is
available
for
MSMA
and
has
been
rated
acceptable
according
to
guideline
standards
(
MRID
44767601).
In
this
study,
14C­
labeled
MSMA
was
applied
to
a
sandy
loam
soil
at
6.1
ppm
(
2.7
ppm
as
arsenic)
and
incubated
at
25
°
C
and
75%
of
field
moisture
capacity
for
1
year.
Based
on
measurements
taken
throughout
the
course
of
the
study,
the
estimated
first­
order
half­
life
for
MMA
was
240
days.
This
half­
life
is
calculated
from
the
percent
of
the
applied
dose
remaining
as
the
parent
compound.
As
much
as
11%
of
the
applied
dose
was
bound
to
soil
and
unextractable,
so
it
is
possible
that
this
half­
life
takes
into
account
some
amount
of
sorption
as
well
as
transformation.
The
metabolism
did
not
occur
at
a
constant
rate;
50%
of
the
parent
compound
had
dissipated
by
60
days,
but
after
this
time,
metabolism
slowed
to
the
extent
that
approximately
35%
of
the
applied
dose
remained
after
one
year.
DMA
was
formed
as
a
metabolite
in
this
study
and
data
on
its
formation
and
decline
led
to
a
half­
life
of
242
days,
as
discussed
in
the
surface
water
modeling
section.

Metabolites
Potential
metabolites
of
applied
DMA
and
MMA
include
volatile
alkylarsines
and
inorganic
arsenic
(
as
arsenate
or
arsenite)
along
with
carbon
dioxide.
Cacodylic
acid
may
be
present
as
a
metabolite
of
MMA
as
well
as
an
applied
parent
compound.
The
formulas
and
names
of
the
principal
species
are
listed
below
in
Table
A2.
The
major
metabolites
identified
in
published
sources
and
registrant
submissions
are
inconstant­­
sometimes
detected
and
sometimes
not.
They
also
occur
in
widely
variable
proportions.
Reasons
for
this
are
unclear,
but
are
likely
associated
with
the
sensitivity
to
the
ambient
conditions,
mentioned
above,
although
some
variability
is
likely
due
to
difficulty
in
analysis
as
well.
Regardless
of
the
form
it
takes,
however,
the
total
amount
of
arsenic
present
does
not
change;
these
organic
arsenicals
and
their
transformation
products,
in
combination
with
arsenic
from
the
natural
background
and
from
other
anthropogenic
sources,
maintain
the
total,
immutable
arsenic
environmental
load.
Arsenic
from
pesticides
is
not
lost
but
redistributed
and
transformed
throughout
the
environment
(
plants,
animals,
air,
soil,
sediment,
water)
into
other
arsenic
containing
substances.

The
only
metabolism
pathway
that
could
directly
reduce
soil
arsenic
loading
is
transformation
to
volatile
alkylarsines.
Microbial
species
capable
of
metabolizing
dissolved
organic
arsenical
compounds
to
gaseous
arsines
have
been
identified
(
Cox
and
Alexander,
1973;
Turpeinen,
2001).
Dimethylarsine
and
trimethylarsine
are
the
most
likely
volatile
products
of
soil
metabolism
(
Cox
and
Alexander,
1973;
Woolson,
1977).
Some
early
studies
of
arsenic
fate
in
soil
concluded
that
volatilization
was
a
major
route
of
dissipation.
Woolson
(
1973)
attributed
losses
of
applied
arsenic
of
up
to
60%
to
27
Table
A2.
Arsenic
species
commonly
found
in
environmental
samples.

"
Organic"
Species
"
Inorganic"
Species
dimethylarsine
(
CH3)
2AsH
arsenate
[
As(
V)]
AsO4
3­

trimethylarsine
(
CH3)
3As
arsenite
[
As(
III)]
AsO3
3­

cacodylic
acid
(
dimethylarsenic
acid,
hydroxydimethylarsine
oxide)
(
CH3)
2AsO(
OH)

volatilization.
This
conclusion
was
inferred
indirectly
from
mass
balance
calculations
and
was
not
confirmed
analytically.
In
a
later
study
designed
to
investigate
the
possibility
of
volatilization,
Woolson
found
that
less
than
0.5%
of
applied
arsenic
was
trapped
as
volatile
arsines
from
unamended
soils,
and
only
1.4%
was
found
in
amended
soils
(
1982).
Several
other
studies
found
volatilization
of
up
to
15%
of
applied
arsenic
from
amended
soils,
but
no
more
than
2.2%
volatilization
was
found
in
unamended
soils
(
Woolson,
1977;
Akins
and
Lewis,
1976;
Cheng
and
Focht,
1979).
More
recent
studies
with
improved
analytical
techniques
have
achieved
mass
balance
without
volatilization
(
Onken
and
Adriano,
1997;
MRID
44767601)
and
other
experiments
designed
specifically
to
consider
volatilization
have
not
found
alkylarsines
produced
at
greater
than
0.5%
of
applied
arsenic
(
Gao
and
Burau,
1997;
Turpeinen
et
al,
2001).
Metabolism
to
volatile
alkylarsines
is
possible
under
certain
conditions
but
is
generally
not
likely
to
be
a
major
route
of
dissipation.
The
possibility
of
volatilization
was
therefore
not
included
in
calculations
on
the
fate
of
applied
organic
arsenical
pesticides.

Other
routes
of
metabolism
include
methylation
and
demethylation.
DMA
has
two
methyl
groups
attached
to
a
central
arsenic,
MMA
has
one,
and
inorganic
arsenic
has
none.
Theoretically,
any
of
the
methyl
groups
on
DMA
or
MMA
are
subject
to
removal,
while
one
methyl
group
could
be
added
to
convert
inorganic
arsenic
to
MMA
which
could
be
further
methylated
to
DMA.
In
fact,
not
all
of
these
transformations
are
likely
to
occur.
There
is
some
uncertainty
associated
with
all
speciation
of
soil
arsenic
due
to
limitations
in
extraction
and
analytical
techniques.
Organic
arsenical
species
may
be
transformed
to
some
extent
during
the
process
of
extraction
and
measurement.
This
is
not
an
issue
when
results
are
reported
as
total
arsenic,
as
done
in
many
studies,
but
those
results
are
less
useful
in
a
discussion
of
metabolites.

Application
of
MMA
may
lead
to
both
DMA
and
iAs
as
end
products.
A
registrant
aerobic
soil
metabolism
study
found
that
after
1
year,
35%
of
MMA
applied
at
6.1
ppm
remained
present
as
the
parent
compound,
32%
was
present
as
DMA,
and
19%
had
evolved
as
carbon
dioxide,
indicating
demethylation
to
inorganic
arsenic
(
MRID
44767601;
Acceptable).
Conversely,
Gao
and
Burau
(
1997)
detected
no
DMA
resulting
from
application
of
MMA
despite
significant
metabolism
to
inorganic
arsenic.
In
the
few
other
laboratory
studies
reviewed
for
this
document
that
speciated
end
products,
variable
DMA
resulting
from
application
of
MMA
was
found
up
to
a
maximum
of
9%
(
MRID
43314801;
MRID
44767602;
Von
Endt
et
al.,
1968).
Field
studies
found
some
DMA
28
after
application
of
MMA,
but
typically
in
very
small
amounts
(
MRID
42526001;
MRID
42616201).
Transformation
to
inorganic
arsenic
is
discussed
further
below.

For
metabolism
of
DMA,
there
are
fewer
studies
available
with
speciated
results.
A
registrant
submitted
aerobic
soil
metabolism
study
detected
MMA
in
soil
samples
following
application
of
DMA,
but
both
the
study
author
and
the
EPA
reviewer
concluded
that
the
MMA
detected
was
an
artifact
of
the
analytical
methodology
(
MRID
42616001;
Unacceptable).
Gao
and
Burau
(
1997)
found
that
after
70
days,
up
to
82%
of
applied
DMA
had
metabolized,
but
only
inorganic
arsenic,
and
not
MMA,
was
detected.
Given
the
relative
stability
of
MMA,
it
is
unlikely
that
MMA
metabolized
too
quickly
to
accumulate,
so
this
result
suggests
that
DMA
metabolizes
directly
to
inorganic
arsenic
by
loss
of
both
methyl
groups,
rather
than
stepwise
going
through
a
monomethylated
intermediate.
As
discussed
above,
other
experiments
have
shown
that
Gao
and
Burau's
results
with
respect
to
MMA
are
not
true
in
all
situations.
Their
results
for
DMA
are
more
conclusive
than
those
for
MMA,
however,
because
DMA's
metabolism
was
investigated
under
at
least
15
sets
of
varying
conditions,
while
MMA
was
only
studied
in
one.
In
another
laboratory
study,
Woolson
et
al.
(
1982)
applied
DMA
at
10
ppm
(
as
arsenic)
in
aerobic
and
anaerobic
soils,
either
unamended
or
amended
with
sewage
sludge,
dairy
manure,
or
hay.
In
unamended
soils,
the
highest
reported
detection
of
MMA
was
at
0.8
ppm
as
arsenic.
In
amended
soils
in
anaerobic
conditions,
detection
of
MMA
went
as
high
as
3.6
ppm
as
arsenic.
A
registrant
submitted
terrestrial
field
dissipation
study
looked
for
MMA
in
soil
after
two
applications
of
DMA
at
24
lb/
A
(
MRID
43485301;
Supplemental).
A
small
amount
of
MMA
(
1.65
ppm
as
MSMA,
equivalent
to
0.76
ppm
as
arsenic)
was
detected
immediately
after
the
second
application
and
was
undetected
by
the
end
of
the
study.
Based
on
the
lack
of
significant
transformation
seen
in
these
studies,
MMA
was
not
treated
as
a
metabolite
of
DMA
in
modeling
potential
exposure
to
both
species,
although
there
is
some
uncertainty
in
this
assumption.

Inorganic
forms
of
arsenic
are
generally
more
toxic
than
the
organic
forms,
so
it
is
very
important
to
assess
the
possibility
of
mineralization
of
organic
arsenical
pesticides
to
arsenate
[
As(
V)]
and
arsenite
[
As(
III)].
Results
of
studies
investigating
conversion
to
inorganic
arsenic
are
extremely
variable,
ranging
from
no
mineralization
in
one
year
to
approximately
80%
mineralization
in
2
months.
The
only
GLP
study
carried
out
for
a
full
year,
as
required
by
FIFRA
guidelines,
found
that
19%
of
MMA
applied
at
6.1
ppm
mineralized
to
inorganic
arsenic.
Mineralization,
as
measured
by
production
of
carbon
dioxide,
was
most
rapid
in
the
first
three
months
and
then
slowed,
with
very
little,
if
any,
occurring
in
the
period
from
9
to
12
months
(
MRID
44767601;
Acceptable).
These
results
are
consistent
with
the
findings
of
several
shorter
term
published
studies
which
determined
that
metabolism
to
inorganic
arsenic
is
most
likely
to
occur
soon
after
application
(
Abdelghani
et
al.,
1977;
Von
Endt
et
al.,
1968;
Dickens
and
Hiltbold,
1967).
All
of
those
studies,
which
lasted
several
weeks
to
several
months,
ended
with
inorganic
arsenic
at
less
than
19%
of
the
applied
amount,
suggesting
that
as
a
reasonable
upper
bound.
29
Two
other
studies,
however,
present
very
different
conclusions.
One
study
applied
DMA
at
the
arsenic
equivalent
of
10
ppm
and
found
that
in
aerobic
conditions,
52%
was
demethylated
after
60
days,
while
in
anaerobic
conditions,
27%
mineralization
occurred.
Mineralization
was
significantly
lower
in
soils
amended
with
sewage
sludge,
manure,
or
hay
(
Woolson,
1982).
In
another
study,
DMA
was
applied
at
10
to
100
ppm
in
a
variety
of
conditions
and
incubated
for
70
days.
The
least
amount
of
mineralization,
all
less
than
7%
of
the
applied,
occurred
in
low
moisture
soils
and
at
low
temperatures.
In
moister
soils
at
22­
25
°
C,
demethylation
ranged
from
approximately
12%
to
approximately
86%
(
Gao
and
Burau,
1997).
In
the
one
treatment
where
MMA
was
applied,
approximately
43%
demethylation
occurred,
relative
to
73%
for
DMA
in
the
same
conditions
(
100
ppm
as
arsenic,
22
°
C,
soil
moisture
approximately
at
field
capacity).
Terrestrial
field
dissipation
studies
also
suggest
that
conversion
to
inorganic
arsenic
can
occur,
with
total
soil
arsenic
levels
remaining
elevated
over
control
values
even
after
applied
parent
compounds
are
no
longer
detectable
(
MRIDs
43485301,
42526001).
These
results
are
discussed
in
more
detail
in
the
soil
buildup
section
(
p.
30).

In
the
one
available
aerobic
soil
laboratory
study
where
there
was
analysis
for
arsenite
[
As(
III)],
a
possible
mineral
transformation
product,
it
was
not
detected
(
Gao
and
Burau,
1997).
Consistent
with
this
result,
and
as
part
of
the
same
study,
applied
arsenite
was
converted
to
arsenate.
It
is
generally
accepted
and
consistent
thermodynamically
that
arsenate
rather
than
arsenite
is
the
prevalent
form
in
aerobic
soils.
There
were
no
reported
tests
for
arsenite
production
from
cacodylic
acid
under
anaerobic
or
flooded
conditions.
However,
comparable
concentrations
of
arsenate
and
arsenite
can
thermodynamically
coexist
under
certain
environmental
conditions,
including
those
found
in
groundwater
and
surface
water
and
such
concentrations
have
been
measured,
as
discussed
in
the
monitoring
section
of
the
Drinking
Water
Assessment
(
DP
Barcode
D309098).

Mobility
In
Air.
Based
on
physical
properties
tabulated
above,
volatilization
of
parent
materials
would
not
expected
to
be
a
significant
route
of
dispersal.
Consistent
with
this
expectation,
volatilization
of
parent
was
not
reported
in
any
lab
study.
However,
as
mentioned
above,
volatile
arsines
produced
by
metabolism
are
part
of
the
global
arsenic
transformation
and
transport
cycle.

In
Soil.
Sorption
to
numerous
diverse
soils
varies
tremendously,
but
indicates
intermediate
to
low
mobility.
Laboratory
studies
have
shown
that
in
some
situations,
significant
sorption
of
arsenic
compounds
may
occur
within
hours
of
application,
while
in
others,
a
large
portion
of
applied
arsenic
remains
in
water­
soluble
forms
for
days
or
months
after
application
(
Onken
and
Adriano,
1997;
Sarkar
et
al.,
2005).
Remobilization
of
sorbed
arsenic
with
changing
environmental
conditions
is
also
possible
(
Matera
and
La
Hecho,
2001).

To
fulfill
data
requirements,
the
registrant
submitted
a
published
non­
FIFRA,
non­
GLP
study
by
Wauchope
(
1975)
for
Agency
review
(
part
of
27
June
1986
review
package,
EPA
Accession
No.
260061)
as
well
as
one
FIFRA
GLP
study
(
MRID
41651906;
30
Supplemental).
Additional
information
about
the
pH
dependency
of
sorption
of
arsenic
was
obtained
from
an
open
literature
report
by
Smith
(
1999).

Wauchope
measured
the
simple
batch
equilibrium
adsorption
of
16
Mississippi
River
alluvial
flood
plain
soils,
none
of
which
were
in
the
"
sand"
textural
class,
two
of
which
had
a
"
clay"
texture,
and
14
of
which
had
a
"
loam"
texture.
The
main
study
objectives
were
to
correlate
sorption
with
soil
properties
and
to
make
direct
experimental
comparison
of
the
relative
sorptions
of
phosphate
(
as
H2PO4
1­
,
a
relatively
immobile
soil
chemical),
cacodylate/
cacodylic
acid,
arsenate/
arsenic
acid,
and
methylarsonate/
methylarsonic
acid
(
at
an
adjusted
pH
of
5.6
for
all
soils
and
chemicals).
Phosphate,
a
large
magnitude
agricultural,
industrial,
and
naturally
occurring
mineral
with
established
relative
immobility,
is
a
well­
suited
benchmark
for
comparing
the
suite
of
organic
arsenicals.
Phosphorous
and
arsenic
are
also
adjacent
periodic
table
group
Vb.

Wauchope
equilibrated
arsenical
slurries
at
a
1:
20
soil:
solution
ratio
in
systems
experimentally
fixed
at
a
pH
of
5.6.
Wauchope
did
not
explicitly
calculate
sorption
coefficients,
so
the
Agency
calculated
simple
soil
sorption
coefficients
(
Kd)
based
on
the
results
at
the
low­
end
initial
concentrations,
3.2
x
10­
4
M
for
cacodylate
and
2.5
x
10­
4
M
for
arsenate
and
MMA.
The
Kds
for
cacodylate
ranged
from
8.2
to
33
mL/
g
with
a
median
of
16
mL/
g
and
for
MMA,
the
range
was
from
17
to
95
mL/
g
with
a
median
of
28
mL/
g.
The
results
were
relatively
independent
of
organic
matter
content.
(
Pseudo
organic
carbon
sorption
coefficients
(
Koc)
range
from
around
700
to
7000
mL/
g
oc,
but
do
not
correlate
with
the
Kd
values
given
above.)
Wauchope
found
that
sorption
was
best
correlated
with
clay
and
iron
and
aluminum
oxide
content.
In
this
respect,
the
two
organic
arsenicals
behaved
like
the
inorganic
arsenate
and
phosphate.
By
direct
comparison,
all
arsenicals
were
more
strongly
sorbed
than
phosphate
in
the
increasing
order:
phosphate
<
DMA
<
arsenate
~
MMA.
Although
arsenicals
sorb
more
strongly
than
phosphate,
phosphate
is
still
expected
to
strongly
compete
for
binding
sites
(
Matera
and
Le
Hecho,
2001).

Additional
Kd
values
for
MMA
were
obtained
from
a
registrant
adsorption/
desorption
study
following
standard
FIFRA
guidelines
(
MRID
41651906).
MMA
was
equilibrated
with
soil
at
a
1:
10
ratio
with
initial
concentrations
ranging
from
1
to
18
ppm.
Four
soils
were
tested
with
average
Kds
of
0.5
mL/
g
(
sand),
11.4
mL/
g
(
silty
loam),
18.7
mL/
g
(
silty
clay),
and
39.4
mL/
g
(
sandy
loam).
Again,
there
was
no
correlation
between
organic
matter
content
and
sorption.
Based
on
these
two
studies,
the
lowest
non­
sand
Kd
values
for
DMA
and
MMA
were
8.2
mL/
g
and
11.4
mL/
g,
respectively.

pH
could
have
a
major
influence
on
sorption
because
of
the
anionic
nature
of
the
tested
chemicals.
Wauchope
experimentally
fixed
the
pH
at
5.6
for
the
16
soil/
water
systems
(
natural
soil
pHs
ranged
from
4.8
to
7.6)
so
correlation
with
pH
cannot
be
determined
from
this
study.
At
a
pH
of
5.6,
DMA
would
be
expected
to
be
present
primarily
in
the
protonated,
uncharged
form.
At
neutral
to
alkaline
pHs,
however,
the
negatively
charged
cacodylate
ion
dominates.
Generally,
anionic
(
negatively
charged)
species
tend
to
be
less
strongly
sorbed
by
soil
surfaces
which
tend
to
maintain
a
negative
(
repelling)
charge;
the
surface
charge
also
tends
to
increase
(
become
more
negative)
at
higher
pHs.
Thus,
at
31
more
nearly
neutral
or
alkaline
pHs,
sorption
coefficients
could
be
considerably
lower
than
those
given
above,
and
mobility
correspondingly
higher.
A
recent
publication
on
arsenate
and
arsenite
sorption
in
Australian
soils
(
Smith
et
al.,
1999)
provides
some
insight
on
the
potential
degree
of
importance
of
pH
on
sorption
of
acid
and
anion
couples.
Although
somewhat
tenuous,
Wauchope's
four
species
might
be
considered
a
homologous
series
with
the
congeners
phosphorus
or
arsenic
at
the
central
core.
Therefore,
Smith's
arsenate
and
arsenite
data
would
serve
as
a
relational
link
to
the
possible
effect
of
pH
on
the
mobility
of
cacodylic
acid/
cacodylate.

In
four
soils
selected
to
vary
widely
in
chemistry
and
mineralogy,
Smith
showed
that
in
the
experimentally
adjusted
pH
range
of
2.0
to
8.5
(
adjusted
with
dilute
nitric
acid
or
sodium
hydroxide)
and
ionic
strength
range
of
0.003
to
0.3
mol/
L
(
adjusted
with
sodium
nitrate)
there
are
complicated
pH
and
ionic
strength
dependencies.
However,
the
Agency
observes
from
the
data
in
the
more
environmentally
relevant
range
of
pHs
from
5
to
8.5
and
ionic
strength
of
0.003
molar,
that
arsenate
decreased
in
sorption
with
increasing
pH
by
a
maximum
factor
of
only
approximately
two.
This
decrease
is
not
dramatic
compared
with
the
much
larger
variability
in
simple,
standard
sorption
Kds
which
Wauchope
measured
above
and
which
Smith
measured
for
a
total
of
10
soils
(
the
four
for
the
detailed
pH
and
ionic
strength
dependencies
plus
six
others
for
a
rudimentary
subset
for
Kd
measurement
and
soil
correlations).
Smith's
simple
Kds
for
arsenate
in
the
10
soils
were
distributed
in
the
range
from
1.7
to
62
mL/
g.

Although
Smith
did
not
report
any
calculated
sorption
coefficients
for
arsenite,
his
comments
and
plotted
data
of
sorbed
amounts
show
arsenite
to
be
moderately
less
sorbed
than
arsenate,
consistent
with
its
lower
negative
charge.
A
more
recent
study
comparing
sorption
of
arsenate
and
arsenite
confirms
this
results,
finding
that
arsenite
is
more
available
to
dissolution
in
water
than
arsenate
(
Onken
and
Adriano,
1997)
However,
in
contrast
with
the
decreasing
sorption
of
arsenate
in
the
four
soils
in
the
pH
range
of
5
to
8.5
and
ionic
strength
of
0.003
M,
Smith
found
sorption
of
arsenite
in
two
soils
was
fairly
constant
from
approximately
pH
2
to
pH
5,
but
increased
significantly
from
approximately
pH
5
to
the
maximum
reported
pH
7
for
arsenite.
Sorption
increased
by
a
maximum
factor
of
approximately
five
in
the
most
sensitive
of
the
two
soils.
Effect
of
ionic
strength
on
sorption
of
arsenite
was
small
and
complex
in
the
tested
ranges
of
pH
and
ionic
strength.
Overall,
the
Agency
concludes
from
Smith's
surrogate
data
in
the
more
environmentally
relevant
range
of
pHs
from
5
to
8.5
and
ionic
strength
of
0.003
molar,
that
sorption
of
cacodylate/
cacodylic
acid
should
not
decrease
(
increase
mobility)
dramatically
with
pH
when
compared
to
the
much
larger
variability
in
soil
sorption
found
with
different
soils.

Soil
Buildup/
Field
Studies
The
relative
immobility
of
arsenicals
along
with
arsenic's
elemental
nature
make
buildup
in
soil
after
repeated
applications
an
important
consideration.
Arsenic
does
not
break
down;
it
can
only
be
redistributed
through
runoff,
leaching,
erosion,
volatilization,
or
plant
uptake.
Even
at
the
lowest
non­
sand
Kds
of
8.2
and
11.4
mL/
g,
sorption
of
arsenicals
would
be
significant,
and
the
range
of
possible
Kds
extends
to
much
higher
values,
especially
if
the
possibility
of
transformation
to
more
strongly
sorbing
inorganic
32
arsenic
is
considered.
As
a
result,
significant
leaching
is
unlikely
in
most
conditions
and
the
potential
for
runoff
is
likely
to
decrease
over
time.
Volatilization
is
also
likely
only
in
specific
circumstances,
leaving
soil
erosion
and
plant
uptake
as
the
sole
routes
of
dissipation
of
organic
arsenicals
applied
to
soil.

Several
terrestrial
field
dissipation
studies
have
been
conducted
by
registrants
to
explore
the
environmental
fate
of
applied
organic
arsenicals.
They
are
discussed
here
with
the
qualification
that,
although
most
have
been
determined
to
be
scientifically
sound,
unless
otherwise
indicated
here,
they
have
not
been
recognized
as
fulfilling
the
data
requirement
because
they
do
not
provide
an
adequate
theoretical
explanation
for
observed
loss
of
arsenic.

In
the
most
recent
study,
rated
supplemental,
DMA
was
applied
twice
at
24
lb
a.
i./
A
to
bare
ground
plots
of
sandy
loam
soil
(
MRIDs
42843101
and
43485301).
The
annual
application
rate
used
in
this
study,
48
lb
a.
i./
A,
is
considerably
higher
than
that
allowed
on
cotton
(
1.2
lb
a.
i./
A/
yr;
12/
12/
05
Master
Label)
but
it
is
similar
to
the
labeled
rate
on
non­
crop
areas
and
ornamentals
(
44
lb
a.
i./
A/
yr;
12/
12/
06
Master
Label).
DMA
in
the
top
6
inches
of
soil
reached
a
peak
concentration
of
32.8
ppm
(
as
DMA)
10
days
after
the
second
application
and
was
non­
detectable
by
243
days.
MSMA
reached
a
peak
concentration
of
1.65
ppm
immediately
following
the
second
application.

Total
arsenic
peaked
at
20.37
ppm
(
as
As),
also
10
days
after
the
second
application.
Following
the
peak,
total
arsenic
decreased
more
slowly
than
DMA
with
little
dissipation
in
the
final
6
months,
reaching
10.68
ppm
on
day
366.
These
values
can
be
compared
to
the
average
arsenic
background
of
2.00
±
0.24
ppm
found
in
the
top
6
inches
of
the
control
plot.
Some
of
the
dissipation
in
the
top
soil
layer
can
be
accounted
for
by
increasing
arsenic
in
the
next
soil
layer.
At
the
6
to
12
inch
depth,
total
arsenic
increased
steadily
from
1.95
ppm
prior
to
DMA
application
to
4.47
ppm
(
as
As)
on
day
366.
Beginning
on
day
182,
slightly
elevated
arsenic
levels
were
observed
in
the
12
to
18
inch
soil
layer
as
well
but
a
statistical
evaluation
found
these
results
to
be
inconclusive.
This
shows
a
significant
persistence
of
applied
arsenic,
with
the
study
author
concluding
that
"
by
[
day
93]
the
organic
arsenicals 
were
largely
mineralized,
leaving
elemental
arsenic
in
the
surface
soil."

Two
other
registrant
studies
monitored
the
terrestrial
field
dissipation
of
MSMA.
In
one,
MSMA
was
applied
twice
to
cotton
in
a
silty
loam
soil
at
2
lb
a.
i./
A
(
MRID
42616201).
In
the
other,
turf
application
was
simulated
by
applying
MSMA
three
times
at
4.95
lb
a.
i./
A
to
bare
ground
sandy
loam
(
MRIDs
42526001
and
4322801).
In
both,
residual
MSMA
decreased
to
undetectable
levels,
by
2
months
in
the
cotton
field
and
by
day
365
in
the
bare
ground
plot.
Both
studies
showed
some
transformation
to
DMA,
but
at
very
low
levels,
peaking
at
less
than
0.5
ppm
in
a
short
time
frame
after
application.
In
the
cotton
study,
soil
arsenic
levels
were
"
variable
with
no
discernable
pattern
of
increase
or
decline."
In
the
bare
ground
study,
total
arsenic
levels
in
the
top
6
inches
of
soil
were
consistently
higher
in
treated
plots
than
in
control
plots
and
higher
than
initial
values.
Total
soil
arsenic
peaked
three
times,
corresponding
to
the
three
applications.
In
the
study
period
following
the
third
peak,
days
41
to
365,
the
average
total
arsenic
in
the
33
treated
plots
was
12.16
±
0.77
ppm
compared
to
an
average
in
the
control
plot
over
the
same
period
of
9.79
±
1.76
ppm.
It
is
difficult
to
quantify
the
impacts
on
total
soil
arsenic,
in
large
part
due
to
the
variability
in
measurements
in
the
control
plot,
but
these
results
suggest
that
arsenic
levels
in
the
treated
plot
were
elevated
over
background
levels
by
at
least
2
ppm.

These
studies
show
that
except
at
high
application
rates,
a
single
year
of
application
is
unlikely
to
lead
to
quantifiable
buildup
of
soil
arsenic,
but
they
show
that
a
large
fraction
of
applied
arsenic
remains
in
the
top
layers
of
soil,
opening
the
possibility
that
soil
buildup
is
a
long
term
concern.
As
a
worst
case
scenario,
all
applied
arsenic
would
remain
in
the
top
soil
layer
with
no
dissipation.
In
the
DMA
field
dissipation
study
(
MRIDs
42843101
and
43485301),
the
worst
case
scenario
would
lead
to
arsenic
levels
in
the
top
6
inches
of
soil
approximately
12.9
ppm3
higher
than
background.
As
discussed
above,
by
the
end
of
the
study,
total
arsenic
in
the
top
layer
of
soil
is
elevated
at
least
8
ppm
over
background
levels,
meaning
that
nearly
two­
thirds
of
the
applied
arsenic
remained
in
the
top
layer
of
soil
at
the
end
of
a
year,
and
additional
arsenic
was
found
from
6
to
12
inches.

At
the
application
rates
used
in
the
studies
of
MSMA
on
cotton
and
turf,
the
worst
case
scenario
would
have
led
to
soil
arsenic
levels
of
approximately
0.9
ppm
and
3.3
ppm
above
background,
respectively3.
The
first
value
is
low
enough
to
be
within
the
natural
range
of
variation
in
measurements
of
background
arsenic
levels,
so
it
is
impossible
to
identify
arsenic
resulting
from
pesticide
application.
For
the
bare
ground
plot,
the
elevated
soil
level,
at
least
2
ppm
greater
than
the
control,
is
a
significant
portion
of
the
worst
case
level
of
3.3
ppm,
showing
that
a
substantial
amount
of
the
applied
arsenic
remained
in
the
top
6
inches
of
soil.
After
repeated
applications
for
multiple
years,
then,
soil
arsenic
levels
could
be
expected
to
increase,
possibly
to
levels
of
concern.

Because
long
term
impacts
are
of
concern,
it
is
important
to
look
at
studies
conducted
over
a
longer
period
of
time.
One
older
registrant
study
applied
MSMA
and
DMA
to
cropped
fields
at
annual
rates
of
up
to
6
lb
ai/
A
and
7.5
lb
ai/
A,
respectively
(
MRID
117165).
After
6
years
soil
arsenic
levels
in
the
top
6
inches
of
soil,
although
somewhat
uncertain
because
of
large
variation
in
residue
levels,
were
as
high
as
14
ppm
and
15.5
ppm,
respectively,
compared
to
an
average
background
level
of
11
ppm.
This
shows
some
buildup,
although
less
than
worst
case
calculations
would
predict.
The
study
authors
hypothesized
that
the
arsenic
loss
was
due
to
volatilization.
The
authors
concluded,
however,
that
"
after
6
annual
applications
all
rates
of
MSMA
and
DMA
resulted
in
poorly
defined
but
significant
buildup
of
arsenic"
in
the
0­
6
inch
layer.

An
incomplete
review
of
the
open
literature
found
several
long
term
field
dissipation
studies
with
some
reporting
no
arsenic
buildup
despite
very
high
application
rates
and
others
finding
substantial
buildup
at
rates
similar
to
current
labels.
Robinson
(
1975)

3
Calculations
of
soil
arsenic
levels
for
worst
case
scenarios
assumed
that
all
applied
arsenic
remained
in
the
top
15­
18
cm
of
soil.
The
applied
rate
was
converted
to
mg
As/
ha
and
then
divided
by
the
mass
of
one
hectare
of
soil
to
a
depth
of
15­
18
cm,
estimated
to
be
approximately
2.25
million
kg.
34
applied
MSMA
for
5
years
at
annual
rates
varying
from
4.4
to
288
kg
MSMA/
ha
and
observed
an
increase
in
total
arsenic
only
for
application
rates
greater
than
36
kg
ai/
ha.
Hiltbold
et
al
(
1974)
applied
MSMA
to
plots
in
three
different
cotton
fields
at
variable
rates.
After
6
years
at
an
application
rate
comparable
to
the
maximum
labeled
MSMA
rate4,
the
amounts
of
soil
arsenic
in
the
top
15
cm,
corrected
for
background
levels,
were
13.0,
13.2,
and
8.0
ppm.
These
levels
represent
57%,
53%,
and
32%
of
the
applied
pesticide,
respectively.
In
the
second
soil,
an
additional
15%
of
the
applied
arsenic
was
detected
in
the
15­
30
cm
soil
layer,
while
in
the
other
soils
no
residual
arsenic
was
detected
below
15
cm.
The
authors
suggest
that
the
unrecovered
arsenic
could
be
accounted
for
through
gaseous
losses.
In
another
field
study,
Woolson
and
Isensee
(
1981)
monitored
soil
arsenic
levels
in
soybean
fields
in
which
DMA,
MSMA,
and
sodium
arsenite
were
applied
annually.
The
data
are
presented
only
as
line
plots,
but
show
that
in
most
cases,
after
6
years,
at
least
half
of
the
applied
arsenic
was
recovered
in
the
top
15
cm
of
soil.
For
MSMA
applied
at
11.2
kg
ai/
ha
(
compare
to
maximum
labeled
MSMA
rate
on
turf
of
17.5
kg
ai/
ha/
yr),
the
line
plot
shows
recovered
soil
arsenic
of
approximately
8
ppm.
DMA
applied
at
11.2
kg
ai/
ha
(
a
higher
application
as
total
arsenic
than
the
same
rate
for
MSMA)
led
to
soil
levels
of
approximately
10
ppm
arsenic
after
6
years.

Additional
information
about
the
potential
for
arsenic
soil
buildup
comes
from
monitoring
studies
in
areas
where
organic
arsenicals
are
known
to
be
used.
One
of
these
studies
measured
soil
arsenic
concentrations
at
5
Florida
golf
courses
that
use
MSMA.
9
samples
from
greens,
tees,
and
fairways
found
arsenic
levels
ranging
from
non­
detectable
to
50
ppm.
The
median
value
was
13
ppm
and
the
90th
percentile
approximately
30
ppm.
The
estimated
mean
background
arsenic
level
in
this
area
is
1.2
ppm
(
DERM,
2002).
Another
study
measured
arsenic
levels
in
roadside
areas
where
the
Louisiana
Department
of
Transportation
applied
arsenicals.
Out
of
559
samples,
42%
had
concentrations
between
21
and
50
ppm
and
22%
had
concentrations
greater
than
50
ppm.
Background
arsenic
in
this
area
was
found
to
be
between
4
and
14
ppm
(
LDOTD,
1984).
Detailed
histories
of
pesticide
applications
are
not
available
in
these
studies
and
other
potential
sources
of
arsenic,
including
historical
land
uses,
are
not
accounted
for.
These
results,
then,
are
inconclusive
as
to
the
source
of
the
arsenic
buildup,
but
they
add
to
the
weight
of
evidence
from
controlled
field
studies
and
modeling.

Soil
accumulation
values
are
generated
by
PRZM
as
part
of
the
process
of
modeling
runoff
concentrations.
PRZM
was
run
with
a
modified
version
of
the
pe4
v01
shell
program
to
estimate
soil
accumulation.
Except
for
exceptions
described
here,
the
modeling
inputs
were
the
same
as
those
included
for
surface
water
modeling
(
DP
Barcode
D309098).
Maximum
application
rates
for
MMA
and
DMA
on
turf
and
cotton
were
used
and
the
modeling
assumed
median
sorption
(
Kd
=
30
for
MMA
and
Kd
=
20
for
DMA).
Because
of
limitations
in
the
available
data
and
modeling
capabilities,
soil
4
Hiltbold
indicates
MSMA
was
applied
at
rates
of
10,
20,
and
40
kg/
ha
for
6
years.
The
total
applied,
as
arsenic,
over
the
course
of
the
study
amounted
to
27,
55,
and
110
kg/
ha,
indicating
average
annual
application
rates,
as
arsenic,
of
4.5,
9,
and
18
kg/
ha/
yr.
The
current
MSMA
label
allows
a
maximum
application
rate
(
on
turf)
of
3.9
lb
ae/
A,
4
times
a
year.
This
is
equivalent
to
an
annual
rate
as
arsenic
of
8.5
kg/
ha/
yr,
comparable
to
the
experimental
average
annual
rate
of
9
kg/
ha/
yr
leading
to
a
total
of
55
kg/
ha.
35
concentrations
were
modeled
as
total
arsenic,
rather
than
speciated
forms.
Application
rates
were
therefore
calculated
to
represent
applied
arsenic
and
infinite
half­
lives
were
used
to
capture
all
forms
of
arsenic
that
may
be
present.
A
wide
range
of
sorption
is
possible;
lower
sorption,
used
to
provide
protective
estimates
for
surface
water
modeling
(
Kd
=
11.4
for
MMA
and
Kd
=
8.2
for
DMA),
would
lead
to
lower
soil
concentrations,
while
greater
sorption
leading
to
higher
concentrations
is
also
possible.
The
possibility
of
transformation
to
volatile
species
was
not
included
in
modeling,
and
modeling
does
not
account
for
the
possibility
of
plant
uptake
of
soil
arsenic.
The
results
presented
are
for
the
scenarios
that
led
to
the
highest
soil
concentrations
(
PA
turf
and
NC
cotton).
Because
soil
concentrations
are
based
on
local
application
rather
than
input
from
an
entire
watershed,
use
of
a
percent
cropped
area
(
PCA)
factor
was
unnecessary.

In
the
top
10
cm
of
soil,
modeling
predicts
that
arsenic
will
accumulate
with
very
little
dissipation
for
several
years
and
then
level
off.
Over
the
long
term,
the
buildup
of
total
arsenic
from
MMA
application
is
predicted
to
reach
chronic
concentrations
of
approximately
13
ppm
and
45
ppm
on
cotton
and
turf,
respectively.
For
DMA
on
turf,
the
highest
application
rate
leads
to
a
modeled
chronic
soil
concentration
of
approximately
77
ppm,
assuming
annual
application
at
that
rate.
For
DMA
on
cotton,
the
chronic
soil
concentration
is
2
ppm.
("
Chronic"
concentrations
are
the
upper
90th
percent
confidence
limit
on
the
annual
average,
or
the
1­
in­
10
year
peak
annual
concentration.)
Considering
another
10
cm,
the
modeled
concentrations
reach
similar
levels,
but
after
longer
periods
time.
If
deeper
soils
are
included,
the
overall
concentrations
would
be
lower.
Most
studies
suggest
that
concentrations
are
highest
in
the
surface
layers
and
buildup
is
typically
limited
to
the
top
30
cm.

The
same
issue
of
arsenic
buildup,
supported
for
soil
by
field
studies,
monitoring,
and
modeling,
is
relevant
to
concentrations
of
arsenic
in
sediment
as
well.
Arsenic
that
reaches
surface
water
is
likely
to
end
up
in
sediments.
A
registrant
study
in
an
aerobic
aquatic/
sediment
system
found
that
after
30
days,
25%
of
applied
DMA
was
found
in
the
sediment
(
MRID
43036101).
For
MMA,
39%
of
the
applied
amount
ended
up
in
sediment
after
30
days,
most
of
it
out
of
the
water
within
the
first
week
(
MRID
43314801;
Acceptable).
In
an
anaerobic
system,
after
1
year
61%
to
95%
of
applied
MMA
was
found
in
the
sediment
(
MRID
44767602;
Acceptable).
For
DMA,
approximately
95%
of
the
applied
amount
was
found
in
the
sediment
after
one
year,
most
of
it
reaching
there
within
the
first
month
(
MRID
42572601).
A
literature
review
of
arsenic
fate
reports
that
in
Lake
Michigan,
arsenic
concentrations
in
water
are
generally
much
less
than
in
the
sediments,
and
refers
to
oceanic
sediments
as
"
the
ultimate
sink
for
arsenic"
(
Woolson,
1977).
36
REFERENCES
Abdelghani,
A.;
Anderson,
A.;
Englande,
A.
J.;
Mason,
J.
W.;
Dekemion,
P.
1977.
Demethylation
of
MSMA
by
soil
microorganisms.
In
Trace
Substances
in
Environmental
Health,
Part
XI
(
Ed:
Hemphill,
D.
D),
University
of
Missouri,
Columbia,
pp.
419­
426.

Akins,
M.
B.;
Lewis,
R.
J.
Chemical
distribution
and
gaseous
evolution
of
arsenic­
74
added
to
soils
as
DSMA­
74As.
Soil
Sci.
Soc.
Am.
J.,
1976,
40,
655­
658.

Akkari,
K.
H.;
Frans,
R.
E.,
Lavy,
T.
L.
Factors
affecting
degradation
of
MSMA
in
soil.
Weed
Sci.
1986,
34,
781­
787.

Bednar,
A.
J;
Garbarino,
J.
R.;
Ranville,
J.
F.;
Wildeman,
T.
R.
Presence
of
organoarsenicals
used
in
cotton
production
in
agricultural
water
and
soil
of
the
Southern
United
States.
J.
Agric.
Food
Chem.,
2002,
50,
7340­
7344.

Cheng,
C.
N.;
Focht,
D.
D.
Production
of
arsine
and
methylarsines
in
soil
and
in
culture.
Applied
Environ.
Microbiol.
1979,
38,
494­
498.

Cox,
D.
P.;
Alexander,
M.
Production
of
trimethylarsine
gas
for
various
arsenic
compounds
by
three
sewage
fungi.
Bull.
Environ.
Contam.
Toxicol.
1973,
9,
84­
88.

DERM,
2002.
Environmental
Quality
Monitoring
at
Five
Municipal
Golf
Courses
in
Miami­
Dade
County.
DERM
Technical
Report;
Florida
Department
of
Environmental
Resources
Management:
Florida.
http://
www.
miamidade.
gov/
derm/
land/
library/
golf_
course.
pdf
Dickens,
R.;
Hiltbold,
A.
E.
Movement
and
persistence
of
methanearsonates
in
soil.
Weeds.
1967,
15(
4),
299­
304.

Gao,
S.;
Burau,
R.
G.
Environmental
factors
affecting
rates
of
arsine
evolution
from
and
mineralization
of
arsenicals
in
soil.
J.
Environ.
Qual.
1997,
26,
753­
763.

Hiltbold,
A.
E.;
Hajek,
B.
F.;
Buchanan,
G.
A.
Distribution
of
arsenic
in
soil
profiles
after
repeated
applications
of
MSMA.
Weed
Sci.
1974,
22,
272­
275.

Kuhlmeier,
P.
D.
Partitioning
of
arsenic
species
in
fine­
grained
soils.
J.
Air
&
Waste
Management
Association.
1997,
47,
481­
490.

Louisiana
Department
of
Transportation
and
Development
(
LDOTD),
1984.
Environmental
and
Health
Effects
of
MSMA
Application
along
Louisiana
Highways.
Prepared
by
Perry,
D.
L.,
Germany,
R.
D.,
and
Flourney,
R.
W.
Enviromed
Laboratories,
Inc.
Baton
Rouge,
LA.
June,
1984.
Rpt.
No.
FHWA/
LA­
84­
1.
37
MAATF,
2005.
The
Environmental
Fate
of
Monosodium
Methane
Arsonate
(
MSMA):
A
Review
of
Important
Processes.
MAA
Research
Task
Force,
Washington,
D.
C.
2005.

Matera,
V;
Le
Hecho,
I.
2001.
Arsenic
behavior
in
contaminated
soils:
Mobility
and
speciation.
In
Heavy
Metals
Release
in
Soils
(
Eds:
Selim,
H.
M.;
Sparks,
D.
M.),
CRC
Press,
Boca
Raton,
pp.
207­
235.

National
Research
Council
(
NRC),
2003.
Bioavailability
of
Contaminants
in
Soils
and
Sediments.
National
Academies
Press,
Washington,
D.
C.
Onken,
B.
M.;
Adriano,
D.
C.
Arsenic
availability
in
soil
with
time
under
saturated
and
subsaturated
conditions.
Soil
Sci.
Soc.
Am.
J.
1997,
61,
746­
752.

Robinson,
E.
L.
Arsenic
in
soil
with
five
annual
applications
of
MSMA.
Weed
Sci.
1975,
23(
5),
341­
345.

Sarkar,
D.;
Datta,
R.;
Sharma,
S.
Fate
and
bioavailability
of
arsenic
in
organo­
arsenical
pesticide
applied
soils.
Part
I.
Incubation
study.
Chemosphere.
2005,
60,
188­
195.

Smith,
E.,
Naidu,
R.,
Alston,
A.
M.
Chemistry
of
arsenic
in
soils:
I.
Sorption
of
arsenate
and
arsenite
by
four
australian
soils.
J.
Environ.
Qual.
1999,
28,
1719­
1726.

Swancar,
A.
Water
Quality,
Pesticide
Occurrence,
and
Effects
of
Irrigation
with
Reclaimed
Water
at
Golf
Courses
in
Florida:
U.
S.
Geological
Survey
Water
Resources
Investigation
Report
95­
4250;
U.
S.
Geological
Survey:
Tallahassee,
1996.
http://
fl.
water.
usgs.
gov/
PDF_
files/
wri95_
4250_
swancar.
pdf
Tu,
C.;
Ma,
L.
Q.;
Zhang,
W.;
Cai,
Y.;
Harris,
W.
G.
Arsenic
species
and
leachability
in
the
fronds
of
the
hyperaccumulator
Chinese
brake
(
Pteris
vittara
L.).
Environmental
Pollution.
2003,
124,
223­
230.

Turpeinen,
R.;
Pantsar­
Kallio,
M.;
Kairesalo,
T.
Influence
of
microbes
on
the
mobilization,
toxicity,
and
biomethylation
of
arsenic
in
soil.
Sci.
Total
Environ.
1999,
236,
173­
180.

Turpeinen,
R.;
Pantsar­
Kallio,
M.;
Kairesalo,
T.
Role
of
microbes
in
controlling
the
speciation
of
arsenic
and
production
of
arsines
in
contaminated
soils.
Sci.
Total
Environ.
2001,
285,
133­
145.

Von
Endt,
D.
W.;
Kearney,
P.
C.;
Kaufman,
D.
D.
Degradation
of
MSMA
by
soil
microorganisms.
Ag.
Food
Chem.
1968,
16(
1),
17­
20.

Wauchope,
R.
D.
Fixation
of
arsenical
herbicides,
phosphate,
and
arsenate
in
alluvial
soils.
J.
Environ.
Qual.
1975,
4(
3),
355­
358.
(
EPA
Accession
No.
260782
and
associated
Nos.
259582
and
260061.)
38
Woolson,
E.
A.;
Kearney,
P.
C.
Persistence
and
reactions
of
14C­
cacodylic
acid
in
soils.
Environ.
Sci.
Technol.,
1973,
7,
47­
50.
(
EPA
Accession
Nos.
259582.)

Woolson,
E.
A.
Fate
of
arsenicals
in
different
environmental
substrates.
Environ.
Health
Perspectives.
1977,
19,
73­
81.

Woolson,
E.
A.;
Isensee,
A.
R.
Soil
residue
accumulation
from
three
applied
arsenic
sources.
Weed
Sci.
Soc.
Am.
J.
1981,
29,
17­
21.

Woolson,
E.
A.;
Aharonson,
N.;
Iadevaia,
R.
Application
of
high­
performance
liquid
chromatography
­
flameless
atomic
absorption
method
to
the
study
of
alkyl
arsenical
herbicide
metabolism
in
soil.
J.
Agric.
Food
Chem.,
1982,
30,
580­
584.
(
EPA
Accession
Nos.
259582,
260061,
260782.)
39
APPENDIX
B:
GENEEC2
INPUTS
Table
B1.
GENEEC2
Input
Parameters
for
DSMA
and
MSMA
Property
Value
Comments
Source
/
MRID
Solubility
(
water)
1
x
106
mg/
L
at
20
E
C
­
2000
RED
(
D210451,
D212449,
255226)

Hydrolysis
t1/
2
0
day
stable
at
all
pHs
42059201
Aquatic
Photolysis
t1/
2
0
day
stable
41662601
Aerobic
Soil
Metabolism
t1/
2
240
days
90th
percentile,
based
on
3
values
Woolson
&
Kearney,
1973;
Woolson
1982;
Gao
&
Burau,
1997;
44767601
Aerobic
Aquatic
Metabolism
t1/
2
480
days
2x240
days.
No
data
available,
2
x
aerobic
soil
t1/
2
since
hydrolysis
assumed
stable
Kd
11.4
mL/
g
lowest
non
sand,
from
20
soils
Wauchope,
1975
41651906
Application
Rate
(
MMA
­
lb
ae/
A)
/
No.
of
Appl
per
year
cotton:
1.74
/
2
(
ground)
orchard:
3.70
/
3
non­
crop:
3.86
/
4
turf
(
max):
3.35
/
4
turf
(
golf):
2.23
/
4
40
APPENDIX
C.
T­
REX
Model
(
T­
REX
Version
1.22,
July
11,
2004)
 
Model
Overview
and
Results
1.
Introduction
This
spreadsheet
based
model
calculates
the
decay
of
a
chemical
applied
to
foliar
surfaces
for
single
or
multiple
applications.
It
uses
the
same
principle
as
the
batch
code
models
FATE
and
TERREEC
for
calculating
terrestrial
estimates
exposure
(
TEEC)
concentrations
on
plant
surfaces
following
application.
A
first
order
decay
assumption
is
used
to
determine
the
concentration
at
each
day
after
initial
application
based
on
the
concentration
resulting
from
the
initial
and
additional
applications.
The
decay
is
calculated
by
from
the
first
order
rate
equation:
CT
=
Cie­
kT
or
in
log
form:

ln
(
C
T
/
Ci)
=
kT
Where:

CT
=
concentration
at
time
T
=
day
zero.

Ci
=
concentration,
in
parts
per
million
(
PPM)
present
initially
(
on
day
zero)
on
the
surfaces.
Ci
is
calculated
based
on
Kenaga
and
Fletcher
by
multiplying
the
Ci
based
on
the
Kenega
nomogram
(
Hoerger
and
Kenaga,
(
1972)
as
modified
Fletcher
(
1994).
For
maximum
concentration
the
application
rate,
in
pounds
active
ingredient
per
acre,
is
multiplied
by
240
for
Short
Grass,
110
for
Tall
Grass,
and
135
for
Broad
leafed
plants/
small
insects
and
15
for
fruits/
pods/
lg
insects.
Additional
applications
are
converted
from
pounds
active
ingredient
per
acre
to
PPM
on
the
plant
surface
and
the
additional
mass
added
to
the
mass
of
the
chemical
still
present
on
the
surfaces
on
the
day
of
application.

k
=
If
the
foliar
dissipation
data
submitted
to
EFED
are
found
scientifically
valid
and
statistically
robust
for
a
specific
pesticide,
the
90%
upper
confidence
limit
of
the
mean
half­
lives
should
be
used.
When
scientifically
valid,
statistically
robust
data
are
not
available
TETT
recommends
the
using
a
default
half­
life
value
of
35
days.
The
use
of
the
35
day
half­
life
is
based
on
the
highest
reported
value
(
36.9
days)
reported
by
Willis
and
McDowell
(
Pesticide
persistence
on
foliage,
Environ.
Contam.
Toxicol,
100:
23­
73,
1987).

T
=
time,
in
days,
since
the
start
of
the
simulation.
The
initial
application
is
on
day
0.
The
simulation
is
designed
to
run
for
365
days.
41
The
program
calculates
concentration
on
each
type
of
surface
on
a
daily
interval
for
one
year.
The
maximum
concentration
during
the
year
are
calculated
for
both
maximum
and
mean
residues.
The
inputs
used
to
calculate
the
amount
of
the
chemical
present
are
in
highlighted
in
light
blue
on
the
spread
sheet.
Outputs
are
in
yellow.
The
inputs
required
are:

Application
Rate:
The
maximum
label
application
rate
(
in
pounds
ai/
acre)
Half­
life:
The
degradation
half­
life
for
the
dominant
process(
in
days)
Frequency
of
Application:
The
labeled
interval,
in
days,
between
repeated
applications
Maximum
#
Application
per
year:
From
the
label
The
calculated
concentrations
are
used
to
calculate
Avian
and
Mammalian
RQ
values.
The
maximum
calculated
concentration
is
divided
by
user
input
values
for
acute
and
chronic
endpoints
to
give
RQs
for
each
type
of
plant
surface.

2.
Avian
Species
For
calculating
dose­
based
RQs
in
birds,
the
maximum
and
mean
Kenaga
residue
values
are
adjusted
for
avian
class
and
food
consumption
based
on
the
following
scaling
factor
(
USEPA,
1993):

FI
(
g/
d)
=
0.648
(
g
bw)^
0.651
For
the
3
avian
weight
classes
considered
(
20,
100
and
1000
g),
this
results
in
%
body
weight
consumption
of:

Weight(
g)
FI
wet
FI
%
bw
consumed
20
4.555599463
22.77799731
114
100
12.98897874
64.94489369
65
1000
58.15338588
290.7669294
29
A.
Dose­
Based
Acute
RQs
Dose­
based
acute
RQs
are
then
calculated
using
the
formula:
RQ
=
adjusted
EEC/
LD50
or
NOAEL
where
the
adjusted
EEC
is
considered
to
be
the
daily
dose
weighted
for
%
body
weight
consumed
of
a
given
food
source.
42
B.
Dietary­
Based
RQs
For
dietary­
based
RQs,
two
values
are
given
for
each
food
group.
First,
the
consumption­
weighted
RQ
for
each
weight
class
(
20,
100,
and
1000g
birds)
is
displayed
and
calculated
using
the
equation:

RQ
=
EEC/((
LC50
or
NOAEC)/(%
bw
consumed))

In
the
second
method,
no
adjustment
is
made
for
consumption
differences
among
the
weight
classes.
This
RQ
is
calculated:

RQ
=
EEC/
LC50
or
NOAEC
3.
Mammalian
Species
A.
Dose­
Based
RQs
For
calculating
dose­
based
RQs
in
mammals,
the
maximum
and
mean
Kenaga
values
are
adjusted
for
mammalian
class
and
food
consumption
(
0.95,
0.66
and
0.15
body
weight
for
herbivores
and
insectivores
and
0.21,
0.15,
and
0.03
body
wt.
for
granivores).
Dosebased
acute
and
chronic
RQs
are
then
calculated
by
dividing
the
adjusted
EECs
(
daily
dose)
by
the
LD50
or
NOAEL.

B.
Dietary­
Based
RQs
Dietary­
based
RQs
are
calculated
using
the
equation:

RQ
=
EEC/((
LC50
or
NOAEC)/(%
bw
consumed))

4.
Graph
A
graph
of
concentration
on
each
plant
surface
vs
time
is
plotted
and
a
concentration
of
concern
line
can
be
added
at
a
user
specified
level.
The
concentration
of
concern
(
e.
g.,
avian
LC50,
mammalian
NOAEL)
label
should
be
entered
in
the
cell
underneath
the
value.
The
graph
automatically
plots
a
line
at
this
concentration
and
the
label
is
extracted
from
that
cell.
The
graph
is
plotted
for
the
first
100
days
post­
application.
Graphs
displaying
acute
and
chronic
LOCs
for
both
birds
and
mammals
are
displayed
in
the
"
Graph"
worksheet.
These
graphs
may
be
useful
as
a
visual
aid
to
communicate
risk
in
your
assessment
and
can
be
copy/
pasted
into
your
document.
To
help
with
scaling
issues
on
the
y
axis,
you
may
want
to
delete
one
of
the
endpoints.

5.
New
Version
Notes
A
new
look
is
used
in
this
update
in
an
effort
to
decrease
confusion
and
increase
transparency
in
the
risk
assessment
process.
This
version
of
T­
REX
(
v1.12)
incorporates
the
ability
to
calculate
EECs
and
RQs
for
maximum
and
mean
residues.
Mean
residues
43
are
calculated
exactly
as
the
maximum
residues
are,
except
the
corresponding
Kenaga
values
are:
85
for
Short
Grass,
36
for
Tall
Grass,
and
45
for
Broad
leafed
plants/
small
insects
and
7
for
fruits/
pods/
lg
insects.
Version
1.22
provides
additional
improvements.

6.
References
Fletcher,
J.
S.,
J.
E.
Nellesson
and
T.
G.
Pfleeger.
1994.
Literature
review
and
evaluation
of
the
EPA
food­
chain
(
Kenaga)
nomogram,
an
instrument
for
estimating
pesticide
residues
on
plants.
Environ.
Tox.
and
Chem.
13(
9):
1383­
1391.

Hoerger,
F.
and
E.
E.
Kenaga.
1972.
Pesticide
residues
on
plants:
correlation
of
representative
dada
as
a
basis
for
estimation
of
their
magnitude
in
the
environment.
IN:
F.
Coulston
and
F.
Corte,
eds.,
Environmental
Quality
and
Safety:
Chemistry,
Toxicology
and
Technology.
Vol
1.
Georg
Theime
Publishers,
Stuttgart,
Germany.
pp.
9­
28.

USEPA.
1993.
Wildlife
Exposure
Factors
Handbook.
Volume
I
of
II.
EPA/
600/
R­
93/
187a.
Office
of
Research
and
Development,
Washington,
D.
C.
20460.
Willis
and
McDowell.
1987.
Pesticide
persistence
on
foliage.
Environ.
Contam.
Toxicol.
100:
23­
73.
44
Appendix
D:
TerrPlant
1.0
Details
Exposure
to
Terrestrial
Plants
including
Wetlands
(
August
8,
2001;
version
1.0)

Terrestrial
plants
inhabiting
dry
and
semi­
aquatic
(
wetland)
areas
may
be
exposed
to
pesticides
from
runoff
and/
or
spray
drift.
Semi­
aquatic
areas
are
low­
lying
wet
areas
that
may
dry
up
at
times
throughout
the
year.

EFED's
runoff
scenario
is
(
1)
based
on
a
pesticide's
water
solubility
and
the
amount
ot
pesticide
present
on
the
soil
surface
and
its
top
one
inch,
(
2)
characterized
as
"
sheet
runoff"
(
one
treated
acre
to
an
adjacent
acre)
for
dry
areas,
(
3)
characterized
as
"
channel
runoff"
(
10
acres
to
a
distant
low­
lying
acre)
for
semi­
aquatic
or
wetland
areas,
and
(
4)
based
on
percent
runoff
values
of
0.01,
0.02,
and
0.05
for
water
solubilities
of
<
10,
10­
100,
and
>
100
ppm,
respectively.

EFED's
Spray
Drift
scenario
is
assumed
as
(
1)
1%
for
ground
application,
and
(
2)
5%
for
aerial,
airblast,
forced
air,
and
spray
chemigation
applications.
The
spray
drift
ratio
used
here
is
in
agreement
with
the
policy
procedures
at
the
time
the
worksheet
was
designed.

Currently,
1)
this
worksheet
is
designed
to
derive
the
plant
exposure
concentrations
from
a
single,
maximum
application
rate
only.
2)
For
pesticide
applications
with
incorporation
of
depth
of
less
than
1
inch,
the
total
loading
EECs
derived
for
the
incorporation
method
will
be
same
as
the
unincorporated
method.

To
calculate
RQ
values
for
Non­
Endangered
Terrestrial
Plants:

Terrestrial
Plants
Inhabiting
Areas
Adjacent
to
Treatment
Site:

Emergence
RQ
=
Total
Loading
to
Adjacent
Area
or
EEC/
Seedling
Emergence
EC25
Drift
RQ
=
Drift
EEC/
Vegetative
Vigor
EC25
Terrestrial
Plants
Inhabiting
Semi­
aquatic
Areas
Adjacent
to
Treatment
Site:

Emergence
RQ
=
Total
Loading
to
Semi­
aquatic
Area
or
EEC/
Seedling
Emergence
EC25
Drift
RQ
=
Drift
EEC/
Vegetative
Vigor
EC25
To
calculate
RQ
values
for
Endangered
Terrestrial
Plants:

Endangered
Terrestrial
Plants
Inhabiting
Areas
Adjacent
to
Treatment
Site:

Emergence
RQ
=
Total
Loading
to
Adjacent
Area
or
EEC/
Seedling
Emergence
EC05
Drift
RQ
=
Drift
EEC/
Vegetative
Vigor
EC05
or
NOAEC
Endangered
Terrestrial
Plants
Inhabiting
Semiaquatic
Areas
Near
Treatment
Site:

Emergence
RQ
=
Total
Loading
to
Semiaquatic
Area
or
EEC/
Seedling
Emergence
EC05
Drift
RQ
=
Drift
EEC/
Vegetative
Vigor
EC05
or
NOAEC
45
Formulas
used
to
calculate
EEC
values
(
8/
08/
01;
version
1.0)

To
calculate
EECs
for
terrestrial
plants
inhabiting
in
areas
adjacent
to
treatment
sites
Un­
incorporated
Ground
Application
(
Non­
granular):

Sheet
Runoff
=
Application
Rate
(
lb
ai/
A)
x
Runoff
Value
Drift
=
Application
Rate
(
lb
ai/
A)
x
0.01
Total
Loading
=
EEC
=
Sheet
Runoff
+
Drift
Incorporated
Ground
Application
with
Drift
(
Non­
granular):

Sheet
Runoff
=
[
Application
Rate
(
lb
ai/
A)/
Incorporation
Depth
(
inch)]
x
Runoff
Value
Drift
=
Application
Rate
(
lb
ai/
A)
x
0.01
Total
Loading
=
EEC
=
Sheet
Runoff
+
Drift
Un­
incorporated
Ground
Application
(
Granular):

Sheet
Runoff
=
EEC
=
Application
Rate
(
lb
ai/
A)
x
Runoff
Value
Incorporated
Ground
Application
without
Drift
(
Granular):

Sheet
Runoff
=
EEC
=
[
Application
Rate
(
lb
ai/
A)/
Incorportion
Depth
(
inch)]

x
Runoff
Value
Aerial/
Airblast/
Spray
Chemigation
Applications:

Sheet
Runoff
=
Application
Rate
(
lb
ai/
A)
x
Runoff
Value
x
Application
Efficiency
of
0.6
Drift
=
Application
Rate
(
lb
ai/
A)
x
0.05
Total
Loading
=
EEC
=
Sheet
Runoff
+
Drift
Runoff
Value
=
0.01,
0.02,
or
0.05
when
the
solubility
of
the
chemical
is
<
10
ppm,
10­
100
ppm,
or>
100
ppm,
respectively
Incorporation
Depth:
Use
the
minimum
incorporation
depth
reported
on
the
label.
