Section
II.
D.
7
­
Page
1
of
35
II.
D.
Appendices:
Drinking
Water
Exposure
7.
Ground
Water
Exposure
Assessment
Based
on
the
February
2005
SAP,
the
Agency
revised
the
groundwater
modeling
effort
to
incorporate
recommendations
regarding
model
scenario
set
up
and
evaluations
against
monitoring
data.
The
major
modeling
change
relates
to
the
way
groundwater
concentrations
are
calculated.
Previously,
OPP
proposed
to
use
only
unsaturated
zone
pore
water
concentrations,
as
this
was
a
readily
available
output
from
the
models.
In
this
revision,
OPP
has
made
modifications
to
the
model
scenario
set
up
in
order
to
capture
spatially
averaged
saturated
zone
concentrations.
In
this
way,
the
models
should
deliver
more
reasonable
estimates
of
concentrations
that
may
be
found
in
drinking
water
in
rural
private
wells.
This
altered
conceptualization,
as
well
as
other
details
of
the
scenarios,
is
described
in
the
sections
that
follow.

Preliminary
results
of
the
model
output
are
given,
and
where
possible
this
output
is
compared
to
monitoring
data.

A.
Groundwater
Scenario
Model
Overview
Figure
II.
D.
7.1
depicts
the
conceptualization
of
the
groundwater
scenario
which
evolved
from
the
advice
of
the
February
2005
SAP.
In
this
conceptualization,
the
pesticide
is
applied
to
the
soil
surface
(
or
plant
canopy)
and
precipitation
or
irrigation
drive
pesticide
through
the
soil
profile
and
into
a
saturated
zone.
Transport
processes
are
simulated
with
three
models
 
PRZM,
RZWQM,
and
LEACHP
 
with
each
model
performing
the
simulation
calculations
differently
(
FIFRA
SAP,
2005).
Particulars
of
the
model
setups
used
to
perform
these
calculations
are
described
below.

All
models
simulated
a
shallow
unconfined
aquifer
with
a
water
table
at
3.5
m
below
the
surface.
Well
screen
length
was
assumed
to
be
1
m,
starting
at
the
water
table
and
extending
to
4.5
m.
The
well
concentration
was
calculated
as
the
average
pore
water
concentration
across
the
length
of
the
screen.
The
models
were
set
up
to
deliver
the
average
pore
water
concentration
in
the
`
saturated'
soil
profile
from
3.5
to
4.5
m.
This
was
accomplished
by
means
depending
upon
the
models'
capabilities,
as
described
later.

The
well
depth
was
chosen
to
ensure
reasonable
conservativeness
in
the
assessment
and
acceptable
runtimes
in
the
models.
Furthermore,
as
some
preliminary
simulations
have
shown,
well
depth
does
not
have
a
critical
influence
in
these
proposed
scenarios.
This
is
because
the
predominant
means
of
degradation
occurs
in
the
top
1
meter
of
the
profile
(
see
below),
and
dispersion
effects
that
would
be
influenced
by
depth
are
considerably
damped
by
using
the
1­
m
spatially
averaged
concentration
for
output.

The
models
consider
that
degradation
rates
change
through
the
soil
profile.
In
general
it
should
be
expected
that
faster
degradation
occurs
in
the
top
of
the
profile
and
Section
II.
D.
7
­
Page
2
of
35
decreases
at
lower
depths.
OPP
used
the
pesticide
aerobic
soil
metabolism
rate
for
the
top
25
cm,
and
a
linearly
declining
rate
with
depth
to
1
m,
below
which
only
abiotic
processes
were
in
effect.
The
pesticide
may
degrade
in
the
upper
reaches
of
the
soil
profile
by
both
abiotic
and
biotic
processes
and
in
the
lower
reaches
and
aquifer
by
abiotic
processes
only
(
Figure
II.
D.
7.2).
This
is
consistent
with
the
default
arrangement
in
the
RZWQM
model
and
is
roughly
equivalent
to
configurations
used
by
Health
Canada
and
the
European
FOCUS
group
(
see
section
on
FOCUS
below).
Also,
temperature
and
soil
water
content
(
as
they
affect
degradation)
can
be
accounted
for
in
LEACHP
and
RZWQM,
but
this
function
does
not
currently
appear
to
be
operating
correctly
in
PRZM.

For
some
pesticides,
well
setbacks
(
Figure
II.
D.
7.1)
are
specified
by
state
or
federal
regulations.
For
such
cases,
the
additional
travel
time
for
a
pesticide
to
reach
a
drinking
water
well
and
the
degradation
that
occurs
during
that
time
is
taken
into
consideration
by
a
plug
flow
model,
as
described
later.

II.
D.
7.
1
­
Depiction
of
general
groundwater
scenario
concept
for
estimating
pesticide
concentrations
in
drinking
water.

Water
Table
Lateral
Groundwater
Movement
100
cm
Set
back
distance
350
cm
Biodegradation
Zone
100
cm
Abiotic
Degradation
Pesticide
Application
Area
Potable
Well
Section
II.
D.
7
­
Page
3
of
35
II.
D.
7.
2
­
Conceptual
model
illustrating
pesticide
degradation
through
the
soil
and
vadose
zone.

B.
Implementation
of
the
Models
With
the
exception
of
LEACHP,
the
application
of
a
fixed
saturated
zone
is
beyond
the
usual
application
of
these
models.
Thus
some
background
investigations
were
required
for
these
models
to
be
parameterized
for
the
conceptual
model.
This
section
describes
the
unique
manners
in
which
each
of
these
models
was
parameterized
in
order
to
fit
this
conceptualization.

1.
PRZM
Although
PRZM
was
not
originally
developed
to
simulate
saturated
conditions,
a
review
of
the
model's
structure
and
coding
showed
that
saturated
conditions
could
be
effectively
simulated
by
redefining
the
field
capacity
parameter.
Soil
compartments
in
PRZM
are
maintained
at
"
field
capacity"
unless
losses
occur
by
evapotranspiration.
Below
the
zone
of
evapotranspiration
(
which
occurs
at
the
bottom
of
the
defined
root
zone),
PRZM
maintains
water
at
field
capacity.
Thus,
a
saturated
zone
can
be
created
in
PRZM
by
setting
the
field
capacity
parameter
(
called
THEFC
in
PRZM)
to
equal
the
porosity.
By
doing
this,
a
constant
water
table
can
be
created
(
Figure
II.
D.
7.3).
Output
concentrations
are
taken
as
the
spatial
average
over
the
depth
from
the
top
of
the
saturated
zone
(
at
3.5
m
below
soil
surface)
to
1
meter
below
the
water
table.
Depending
on
rainfall
and
evapotranspiration
characteristics
of
the
particular
scenario,
this
spatial
averaging
effectively
represents
temporal
averaging
of
from
6
months
to
greater
than
one
year
(
i.
e.,
1
meter
of
water
infiltrating
into
the
water
table
takes
from
6
months
for
Florida
citrus
scenario
to
greater
than
a
year
for
Washington
potato
scenario).

Irrigation,
when
required
by
a
scenario,
is
handled
according
to
recent
OPP
guidance
on
PRZM
irrigation
(
memo
from
D.
Young
and
M.
Corbin
to
the
OPP
Water
Section
II.
D.
7
­
Page
4
of
35
Quality
Tech
Team,
July
2005).
This
guidance
suggests
that
PRZM
has
only
two
appropriate
ways
to
handle
irrigation
 
either
above
the
canopy
or
beneath
the
canopy.
Other
forms
of
irrigation
listed
in
the
PRZM
manual
were
deemed
inappropriate
due
to
hydrologic
improbabilities.
In
addition,
because
the
quantity
of
irrigation
water
is
coupled
to
root
zone
depth
in
PRZM,
rooting
depths
were
set
to
irrigation
depths
rather
than
actual
rooting
depths.
This
is
especially
important
for
deep­
rooting
orchard
crops
for
which
irrigation
water
is
only
applied
to
satisfy
water
demand
in
the
upper
most
soil
layers
(
otherwise,
if
rooting
depth
were
set
to
2
meters
as
in
citrus
irrigation,
water
would
be
unrealistically
applied
to
satisfy
the
entire
2
meter
depth).

II.
D.
7.
3
­
PRZM
scenario
to
simulate
a
fixed
water
table.
Note
the
two
unique
adjustments
to
PRZM
groundwater
scenarios:
1)
"
Field
capacity"
set
to
porosity
to
simulate
a
saturated
zone,
and
2)
root
depth
set
to
irrigation
depth
to
ensure
proper
irrigation.

Simulated
well
screen:
Pore
Water
Concentration
Averaged
over
1
meter
Depth
to
Water
Table
(
3.5
m)

"
Field
Capacity"
set
to
Porosity
Unsaturated
Saturated
Well
Outlet
Represents
Spatial
Average
of
Saturated
Zone
Water
Table
Root
Depth
set
to
Irrigation
Depth
2.
RZWQM
The
current
version
of
RZWQM,
like
PRZM,
was
not
designed
to
handle
a
constant
depth
saturated
zone
(
although
USDA
is
currently
revising
RZWQM
to
handle
this
condition).
Therefore,
OPP
set
up
tile
drains
and
head
gates
to
mimic
a
near­
constant
water
table
depth.
This
is
shown
in
Figure
II.
D.
7.4.
In
developing
the
scenarios,
the
tile
drain
spacing,
lateral
conductivity,
and
drain
diameter
were
adjusted
so
that
a
water
table
formed
just
above
the
head
gate,
thereby
giving
some
flow
out
of
the
drain
at
all
times.
Concentrations
out
of
the
drain
are
calculated
in
RZWQM
by
taking
the
average
concentration
of
the
overlying
saturated
zone
(
from
tile
drain
to
top
of
water
table).
This
method
of
calculation
is
fortuitous
in
that
this
is
the
concentration
that
one
would
expect
from
a
well
screened
from
the
water
table
to
1
meter
below
the
water
table.
This
is
especially
fortuitous
because
RZWQM
does
not
offer
any
other
simple
means
to
calculate
spatially
averaged
concentrations.
We
note
that
USDA­
ARS
has
not
completely
agreed
with
manipulation
of
RZWQM
in
this
way,
but
they
have
not
discounted
it.
The
set
up
is
Section
II.
D.
7
­
Page
5
of
35
currently
being
evaluated
by
USDA­
ARS,
and
it
is
likely
that
ARS
will
be
able
to
modify
RZWQM
to
handle
a
fixed
water
table
condition.

Irrigation
in
RZWQM
is
handled
much
the
same
as
in
PRZM,
in
that
root
zone
depth
is
the
depth
at
which
soil
moisture
depletion
is
considered.
Root
zone
depth
was
set
equal
to
the
desired
depth
of
irrigation,
as
it
was
in
PRZM.

II.
D.
7.
4
­
Formation
of
relatively
constant
water
table
in
RZWQM
using
tile
drains
and
head
gate.
Note
that
tile
drains
and
head
gates
do
not
actually
represent
their
physical
usage,
but
rather
are
implemented
as
a
means
to
control
the
lower
boundary
condition.

Depth
to
Water
Table
set
by
head
gate
(
3.5
m)
Unsaturated
Saturated
Water
Table
Tile
Drain
set
1m
below
head
gate
RZWQM
uses
the
average
concentration
of
water
between
tile
drain
and
water
table
as
the
drain
outlet
concentration
3.
LEACHP
Unlike
the
other
models,
LEACHP
allows
for
a
constant
water
table
depth
as
a
part
of
its
normal
functioning.
Thus
no
special
manipulations
were
required
in
this
regard.

LEACHP
uses
the
Richard's
equation
to
simulate
water
movement.
Five
options
are
available
to
estimate
soil
water
retention
(
water
content
vs.
matric
potential)
and
hydraulic
conductivity
(
water
content
vs.
hydraulic
conductivity).
The
method
used
is
option
5
(
Rawls
and
Brakensiek,
1985)
which
is
based
upon
US
soils,
and
considers
as
independent
variables
particle
size
distribution
(
percent
sand,
silt
and
clay),
soil
organic
matter,
and,
soil
bulk
density.

The
automatic
irrigation
option
was
used.
A
*
name.
sch
file
which
requires
the
depth
where
the
soil
water
sensor
is
located
(
100
mm),
the
threshold
matric
potential
(­
100
kaP)
which
triggers
irrigation,
and
the
replenishment
depth
(
100
cm)
was
created.

4.
Well
Setbacks
Section
II.
D.
7
­
Page
6
of
35
In
some
cases,
well
setbacks
are
required
by
federal
or
state
regulations.
These
setback
requirements
specify
the
nearest
distance
to
a
well
that
a
pesticide
can
be
applied.
For
such
cases,
the
additional
travel
time
to
the
well
allows
for
additional
degradation.
Reductions
in
concentration
are
calculated
in
these
assessments
by
a
plug
flow
approximation
as
follows:







 
=
k
v
L
C
C
exp
0
where
C
=
concentration
at
well
[
mass/
volume]
C0=
concentration
at
point
of
application
[
mass/
volume]
L
=
well
setback
distance
[
length]
v
=
lateral
groundwater
velocity
[
length/
time]
k
=
degradation
rate
in
aquifer
[
time­
1]

The
travel
time
of
the
groundwater
used
in
these
scenarios
is
much
shorter
than
travel
time
would
be
if
only
the
pumping­
induced
gradient
(
cone
of
depression)
of
a
private
drinking
well
was
considered.
OPP
made
conservative
estimates
regarding
natural,
topographically/
stratigraphically­
driven
groundwater
lateral
velocities
(
e.
g.,
velocities
determined
by
elevation
head
and
hydraulic
conductivity
of
the
substrate).
As
an
example,
consider
the
travel
time
through
a
setback
for
unretarded
chemicals
without
regard
to
additional
head
induced
by
well
pumping,
which
can
be
estimated
by
v
r
t
n
=

where
r
=
setback
radius
v
=
natural
lateral
groundwater
velocity
tn
=
travel
time
of
unretarded
solute
due
to
natural
gradient
The
travel
time
of
an
unretarded
chemical
when
only
pumping­
induced
hydraulic
gradient
is
considered
is
determined
from
(
see
for
example
USEPA
1993):

Q
L
r
t
s
w
  
2
=

where
 =
aquifer
porosity
Ls=
screened
length
of
well
Q
=
flow
rate
tw
=
travel
time
of
unretarded
solute
For
setback
estimations,
OPP
assumed
a
high
lateral
velocity
of
0.15
m/
day,
which
is
the
typical
high
end
velocity
that
Russell
et
al.
(
1987)
reported
in
the
Central
Ridge
of
Florida.
For
a
50­
ft
setback,
travel
time
is
about
100
days
due
to
the
natural
topographic
gradient
(
elevation
head).
To
determine
the
well­
induced
travel
time
(
caused
by
gradient
changes
due
to
well
drawdown),
OPP
assumed
that
a
typical
family
uses
101
gallons
per
day
(
0.28
m3/
day)
(
American
Water
Works
Association
estimate),
a
1­
m
screened
well,
and
a
porosity
of
30%.
For
a
50
foot
setback,
the
well­
induced
travel
time
Section
II.
D.
7
­
Page
7
of
35
is
570
days;
the
natural
gradient
travel
is
17%
of
the
well­
induced
travel
time,
which
is
well
within
any
reasonable
expectation
of
the
error
within
such
calculations.
For
1000­
foot
setbacks,
the
natural
gradient
travel
time
is
less
than
1%
of
the
well­
induced
travel
time.
Thus
OPP
neglected
the
velocity
effects
caused
by
private
rural
wells
because
of
their
insignificance.

C.
Standard
Generic
Scenarios
OPP
developed
four
groundwater
scenarios
for
the
Carbamate
Cumulative
assessment
(
Florida
citrus,
Georgia
peanuts,
North
Carolina
cotton,
and
Washington
potato),
and
applied
a
consistent
set
of
rules
to
establish
a
protocol.
This
selection
process
primarily
involved
selection
of
the
soil
characteristics,
selection
of
the
weather,
and
selection
of
appropriate
crop
management
practices.

The
most
appropriate
soils
for
each
crop
were
selected
by
consulting
the
USDA
Soil
Data
Mart
(
http://
soildatamart.
nrcs.
usda.
gov/).
In
cases
where
crops
were
likely
to
be
grown
on
multiple
soils
in
a
region,
the
most
vulnerable
of
the
likely
soils
were
chosen,
as
characterized
by
the
soils
hydrologic
group
(
A,
B,
C,
or
D),
hydraulic
conductivity,
and
organic
matter
content.
Soil
properties
were
taken
from
the
Soils
Data
Mart
and
transformed
into
appropriate
input
parameters
for
each
of
the
models.

Weather
inputs
were
selected
to
represent
the
weather
in
closest
proximity
to
the
scenario
location.
In
the
case
of
RZWQM,
weather
was
generated
from
CLIGEN,
which
is
an
integral
part
of
RZWQM.
For
PRZM
and
LEACHM,
30
years
of
historical
data
was
used
as
obtained
from
the
EPA
Office
of
Research
and
Development
(
http://
www.
epa.
gov/
ceampubl/
tools/
metdata/
index.
htm).

Specific
management
practices
that
might
affect
pesticide
transport
were
included
in
the
models
in
those
cases
where
the
practices
were
likely
to
significantly
affect
pesticide
transport.
Significant
practices
that
were
included
were
the
pesticide
application
method
and
the
application
of
irrigation
water.
Variations
in
tillage
practices
were
ignored
in
the
development
of
these
scenarios,
as
characterization
and
parameterization
of
such
practices
are
difficult
and
would
be
speculative.
Irrigation
and
pesticide
application
practices
were
developed
for
the
models
after
consultation
with
agricultural
extension
agents,
review
of
open
literature,
and
pesticide
label
information.

For
this
carbamate
cumulative
assessment,
the
four
scenarios
developed
by
OPP
(
described
in
section
1D
of
the
main
document
 
see
above)
were
intended
to
represent
high
carbamate
use
in
areas
of
high
groundwater
vulnerability,
and
are
discussed
below.

1.
Florida
Citrus
Most
citrus
production
occurs
on
Florida's
Central
Ridge.
Polk
County
is
typical
of
the
Ridge
and
has
the
highest
acreage
(
101,000
acres
in
year
2000)
in
citrus
production
in
Florida
(
Obreza
and
Collins
2002).
Groundwater
in
this
region
is
particularly
Section
II.
D.
7
­
Page
8
of
35
vulnerable
to
pesticide
contamination
due
to
the
high
water
table
and
sandy
soils
with
low
organic
matter
content.
For
these
reasons,
OPP
chose
the
Polk
County
area
as
representative
of
a
typical
highly
vulnerable
Florida
citrus
area.
Figure
II.
D.
7.5
shows
the
approximate
vicinity
of
the
Florida
citrus
scenario.
Additionally,
the
choice
of
this
location
is
benefited
by
the
availability
of
several
ground
water
studies
(
e.
g.,
Jones
et
al.
1987;
Hornsby
et
al.,
1990,
FL
DEP
2005,
USGS
2005),
which
will
allow
some
evaluation
of
the
scenario
modeling
performance.

II.
D.
7.
5­
Location
of
FL
citrus
scenario.
The
circle
indicates
the
approximate
vicinity.

a.
Soil
Citrus
grows
in
the
Entisols
on
the
Florida
Ridge
which
are
characteristically
low
in
organic
matter
content
and
drain
water
quickly.
Typical
soil
series
used
for
citrus
production
in
Polk
County
are
Candler,
Tavares,
and
Astatula
(
Obreza
and
Collins,
2002).
These
soils
are
predominantly
sand
(>
97%)
and
have
a
low
organic
matter
content
(
0.5
to
2
percent;
US
Soils
data
mart).
In
mature
citrus
groves,
organic
content
is
even
lower
(
0.5
to
1
percent;
Obreza
and
Collins,
2002).
Some
salient
features
of
these
soils
are
given
in
Table
II.
D.
7.1.
These
soils
are
all
in
the
Hydrologic
Group
A,
meaning
there
is
negligible
runoff.

Table
II.
D.
7.1.
Soil
Properties
Candler
Series
Soil
Series
Horizon2
(
cm)
Texture
Sand1
Silt1
Clay1
Saturated
Hydraulic
Conductivity
1
in/
hr
Organic
matter1
(%)
Field
Capacity1
Moist
Bulk
Density2
Avail.
Water
Capacity
(
in/
in)
pH
Section
II.
D.
7
­
Page
9
of
35
Table
II.
D.
7.1.
Soil
Properties
Candler
Series
Soil
Series
Horizon2
(
cm)
Texture
Sand1
Silt1
Clay1
Saturated
Hydraulic
Conductivity
1
in/
hr
Organic
matter1
(%)
Field
Capacity1
Moist
Bulk
Density2
Avail.
Water
Capacity
(
in/
in)
pH
Candler
0
­
80
Sand
97.5
1.25
1.25
6­
50
0.5
­
1.0
0.025­
0.058
1.35­
1.55
0.04­
0.08
4.5­
6
Tavares
0­
8
Sand
97
1.5
1.5
7­
39
0.5­
1.0
0.025­
0.05
1.25­
1.6
0.05­
0.10
3.6­
6
8­
80
Sand
97
1.5
1.5
7­
39
0­
0.5
­­
1.40­
1.70
0.02­
0.05
3.6­
6
Astatula
0­
7
Sand
98.5
0.75
0.75
9­
85
0.5­
1.0
0.025­
0.05
1.25­
1.55
0.04­
0.10
4.5­
6.5
7­
80
Sand
98.5
0.75
0.75
9­
85
0­
0.5
­­
1.45­
1.60
0.02­
0.05
4.5­
6.5
1Obreza
and
Collins,
2002
2USDA
Soils
Data
Mart
b.
Irrigation
Central
Ridge
has
relatively
high
rainfall
(~
50
inches/
year),
but
irrigation
is
necessary
because
of
high
drainage
of
the
characteristic
sandy
soils
on
the
Central
Ridge.
Because
of
the
high
sand
fraction
and
the
associated
low
water
holding
capacity,
irrigation
water
management
is
difficult.
As
a
result,
microirrigation
is
commonly
used,
supplying
only
enough
water
to
satisfy
the
tree
demand
(
Smajstra
and
Harman,
2002;
Parsons
and
Morgan,
2004).
Microirrigation
typically
supplies
10­
20
gallons/
hour
spread
out
over
a
10­
to
18­
ft
area,
with
durations
of
about
4
hours.
Typically,
microirrigation
is
set
to
satisfy
water
in
the
range
of
1
to
2
feet
below
the
surface,
with
irrigation
events
occurring
about
twice
per
week
in
the
spring
and
up
to
3
times
per
week
in
the
summer
(
communication
L.
Parsons,
South
Florida
Agricultural
Extension
Office).
During
the
spring,
soil
moisture
depletion
should
be
no
less
than
1/
3
of
the
available
water
capacity
(
AWC),
while
during
the
remainder
of
the
year,
up
to
2/
3
of
the
available
water
capacity
can
be
depleted
without
severe
effects
(
Boman
et
al.,
2002).
Irrigation
at
50%
of
AWC
was
assumed
for
modeling.

c.
Citrus
Production
and
Crop
profile
About
one­
half
of
citrus
in
Florida
is
grown
on
deep
(
Central
Ridge),
sandy
soil
using
the
unbedded
tree
row
production
technique.
The
remainder
is
grown
on
heavier
and
wetter
soils
in
Florida
Flatwoods.
Citrus
on
the
Central
Ridge
is
planted
along
the
natural
contour.
No
leveling
is
required
for
the
Entisols
because
of
their
natural
drainage
(
Obreza
and
Collins,
2002).
Although
Ridge
citrus
roots
can
penetrate
as
deep
as
15
feet,
most
of
the
roots
are
in
the
top
3
feet.
Because
rooting
depth
is
coupled
to
the
depth
of
irrigation
in
all
of
the
transport
models,
rooting
depth
in
the
models
should
be
set
to
the
appropriate
depth
of
irrigation
(~
2
feet
in
the
case
of
Ridge
citrus).

d.
Water
Table
and
Aquifer
Characteristics
The
water
table
in
the
region
has
an
upper
limit
3.5
to
6
feet
below
the
surface
(
USDA
Soils
Data
Mart).
The
upper
limits
are
generally
reached
in
the
winter
months
(
January
to
December).
Typical
depths
may
be
considerably
deeper.
For
example
in
the
Section
II.
D.
7
­
Page
10
of
35
groundwater
study
conducted
by
Hornsby
et
al.
(
1990)
in
this
area,
water
table
depths
were
reported
to
be
greater
than
20
m.
In
the
study
of
Jones
et
al
(
1987),
depths
of
4
m
were
reported.

Jones
et
al
(
1987)
reported
that
a
representative
typical
high
lateral
groundwater
velocity
in
the
Ridge
area
was
0.15
m/
day.
They
also
observed
that
the
pH
of
the
groundwater
was
typically
4.5,
with
a
range
of
3.5
to
6,
and
temperature
of
20
to
25
C.
The
USGS
Lake
Wales
Ground
Water
Monitoring
Study
(
http://
fisc.
er.
usgs.
gov/
Lake_
Wales_
Ridge/)
reports
surficial
aquifer
pHs
in
the
range
of
4
to
7
(
median
4.9).

2.
GA
Peanuts
The
southern
part
of
Georgia
is
an
area
of
prime
farmland,
suitable
for
field
and
row
crops.
Crops
grown
in
this
region
include
cotton,
peanuts,
beans,
fruit,
tobacco,
potatoes,
sweet
potatoes,
grain
corn,
and
melons
among
others.
Peanuts
were
chosen
as
the
representative
crop
because
Georgia
has
the
highest
peanut
acreage
in
the
U.
S.,
and
acreage
within
Georgia
is
primarily
concentrated
in
the
southwestern
part
of
the
state.
This
area,
lying
in
the
Southern
Coastal
Plain,
has
shallow
groundwater
that
is
susceptible
to
contamination
(
Donohue,
2001)
and
is
used
for
drinking
water
in
some
cases
(
Crandall
and
Berndt,
1996).
For
these
reasons,
OPP
chose
the
area
in
Figure
II.
D.
7.6
for
scenario
development.
In
addition,
a
prospective
groundwater
study
was
conducted
in
this
area
(
MRID
43099601)
which
allows
some
evaluations
regarding
the
suitability
of
the
scenario
parameterization.
Section
II.
D.
7
­
Page
11
of
35
II.
D.
7.
6­
Location
of
the
southern
Georgia
scenario.
The
circle
indicates
the
approximate
vicinity.

a.
Soil
Based
on
soil
data
from
Cook
and
Colquitt
counties
(
Soil
Data
Mart,
USDA,
2005;
Ma
et
al.,
2000),
at
the
center
of
the
relevant
region
(
Figure
II.
D.
7.6),
Tifton
loamy
sand
is
the
dominant
soil
(
23.9%
of
coverage
in
the
region)
and
is
also
a
prime
farmland
soil.
Tifton
is
a
very
deep,
well
drained
soil
on
uplands.
The
subsoil
is
loamy
and
extends
to
a
depth
greater
than
5
feet.
Plinthite
occurs
below
a
depth
of
30
to
50
inches.
Ironstone
nodules
are
present
throughout
the
soil.
Permeability
is
moderate
throughout
the
subsoil.
Available
water
capacity
is
moderate.
Some
properties
of
this
soil
that
are
relevant
for
pesticide
transport
modeling
are
listed
in
Table
II.
D.
72.
This
soil
falls
into
the
Hydrologic
Group
B.

Table
II.
D.
7.2.
Soil
Properties
of
the
Southern
Georgia
Scenario
Horizon
(
cm)
Texture
Saturated
Hydraulic
Conductivity
(
µ
m/
s)
Organic
matter
(%)
Moist
Bulk
Density
Available
Water
Capacity
(
in/
in)
pH
0
­
25
Loamy
sand
42
­
141
0.5
­
1.0
1.3
­
1.55
0.03
 
0.08
4.5
­
6.0
25­
46
Fine
sandy
loam
42
­
141
0.5
­
1.0
1.45
­
1.65
0.08­
0.12
4.5
­
6.
46­
83
Gravely
sandy
clay
loam
4
­
14
0.0
­
0.5
1.5
­
1.7
0.12­
0.16
4.5
­
6.0
83­
162
sandy
clay
1.4
­
4
0.0
­
0.5
1.55
­
1.80
0.1­.
13
4.5
­
5.5
162­
216
sandy
clay
1.4
­
4
0.0
­
0.5
1.65
­
1.85
0.1­
0.12
4.5
­
5.5
b.
Irrigation
Section
II.
D.
7
­
Page
12
of
35
Georgia
peanuts
are
grown
on
dryland
and
on
irrigated
land.
About
50%
of
Georgia
peanut
acreage
is
irrigated.
Typical
irrigation
amounts
may
be
around
1
to
2
inch
per
week
using
center
pivots.
Total
seasonal
use
could
be
10
inches.

c.
Crop
Production
and
Profile
Peanuts
are
typically
rotated
with
cotton
or
a
grass­
type
crop.
Conventional
tillage
is
used
for
almost
all
Georgia
peanut
crops.
Planting
dates
are
from
April
23
and
May
25,
and
harvest
runs
from
early
September
to
early
November.
Plantings
are
typically
single
rows
36
inches
apart.

d.
Water
Table
and
Aquifer
Characteristics
The
water
table
is
typically
at
a
high
of
4
to
6
feet
below
the
ground
surface,
and
some
domestic
wells
draw
from
this
shallow
aquifer.
The
pH
of
the
surficial
aquifer
in
the
Southern
Coastal
Plain
ranges
from
4.1
to
7.4
(
median
5.2)
according
to
a
survey
by
Crandall
and
Berndt
(
1996).

3.
NC
Cotton
North
Carolina
is
ranked
sixth
in
the
nation
in
cotton
acreage
and
seventh
in
production,
generating
5
percent
of
the
United
States
cotton
crop.
Most
of
the
cotton
produced
in
North
Carolina
is
grown
in
the
eastern
half
of
the
state,
or
the
coastal
plain
region.
Three
of
the
four
highest
cotton
producing
counties,
Northampton
(
63045
acres),
Halifax
(
61933
acres),
and
Edgecombe
(
46001
acres),
in
North
Carolina
are
located
in
northeastern
North
Carolina
(
USDA,
Ag
Census,
2002,
http://
www.
nass.
usda.
gov/
census/
census02/
volume1
/
us/
us2appxa.
pdf).
Figure
II.
D.
7.7
shows
the
region
for
which
this
scenario
was
developed.
Crops
grown
in
this
region
include
corn,
peanuts,
tobacco,
soybeans,
small
grains,
cotton,
and
pasture.

The
coastal
plain
is
North
Carolina's
largest
physiographic
province,
covering
45
percent
of
the
state.
The
province
can
be
subdivided
into
two
regions
 
outer
and
inner
coastal
plain.
The
outer
coastal
plain
(
sometimes
called
the
'
tidewater'
region)
consists
of
the
immediate
coast,
barrier
islands,
sounds,
marshes,
lower
river
systems
and
associated
mainland,
and
is
generally
less
than
20
ft
in
elevation.
The
inner
coastal
plain
includes
the
region
from
the
outer
coastal
plain
to
the
Fall
Line.

The
climate
of
North
Carolina's
coastal
plain
province
is
temperate.
Average
high
temperature
during
summer
months
is
in
the
mid­
upper
80s,
while
average
lows
are
near
70
degrees.
During
winter,
average
highs
are
in
the
mid
50s,
while
average
lows
are
in
the
mid
30s.
Temperatures
tend
to
be
more
moderate
n
the
outer
coastal
plain.
Average
rainfall
is
about
51
inches.
Snowfall
is
infrequent
and
generally
averages
less
than
5
inches
per
year
in
the
inner
coastal
plain
and
less
than
2
inches
per
year
in
the
outer
coastal
plain.
Section
II.
D.
7
­
Page
13
of
35
II.
D.
7.
7­
Location
of
the
North
Carolina
scenario.
The
circle
indicates
the
approximate
vicinity.

a.
Soils
Cotton
is
predominately
grown
on
sandy
loam
soils
of
the
coastal
plain.
These
soils
require
subsoiling
(
treatment
to
fracture
and/
or
shatter
soil
with
narrow
tools
below
the
depth
of
normal
tillage
with
a
minimum
mixing
of
the
soil)
to
breakup
naturally
occurring
hardpans.
Dominant
cotton
soils
in
the
three
counties
of
interest
(
Edgecombe,
Halifax,
Northampton)
are
the
Norfolk
loamy
sand
(
Fine­
loamy,
kaolinitic,
thermic
Typic
Kandiudults)
and
Wagram
loamy
sand
(
Loamy,
kaolinitic,
thermic
Arenic
Kandiudults).
The
Norfolk
loamy
sand
was
selected
as
it
is
present
in
both
Edgecombe
and
Northampton
counties,
is
designated
as
prime
farmland,
and
is
an
NRCS
benchmark
soil
(
National
Soil
Handbook,
part
630).
This
soil
is
Hydrologic
Group
B.

Table
II.
D.
7.3.
Soil
Properties
Norfolk
Series
Horizon
(
cm)
USDA
Texture
Saturated
Hydraulic
Conductivity1
in/
hr
Organic
matter1
(%)
Field
Capacity2
(
33
kPa)
Wilting
Point
(
1500
kPa)
2
Moist
Bulk
Density1
Available
Water
Capacity
(
in/
in)
pH
0
­
9
Loamy
sand
42­
141
0.5­
2.0
5.1
2.0
1.55­
1.7
0.06­
0.11
3.5­
6.0
9­
14
Loamy
Sand
42­
141
0.3­
0.8
5.1
2.0
1.55­
1.7
0.06­
0.11
3.5­
6.0
14­
70
Clay
Loam
4­
14
0­
0.5
12
8
1.3­
1.65
0.1­
0.18
3.5­
5.5
70­
100
Clay
loam
4­
14
0­
0.5
13
10
1.2­
1.65
0.12­
0.18
3.5­
5.5
1USDA
Soils
Data
Mart
2Approximated
from
USDA
Norfolk
pedons
NSSL
query
c.
Irrigation
Limited
cotton
acreage
in
North
Carolina
is
irrigated.
The
USDA
Ag
Census
2002
estimates
that
about
3.4%
of
acreage
is
irrigated.
Irrigation
is
not
included
in
the
NC
cotton
model
scenario.
Section
II.
D.
7
­
Page
14
of
35
d.
Cotton
Production
and
Profile
The
majority
of
cotton
is
located
on
the
sandy
loam
soils
of
the
coastal
plain;
these
require
subsoiling
to
break
naturally
occurring
hardpans.
However,
about
20
percent
of
the
cotton
produced
in
the
coastal
plain
is
grown
on
heavier
soils
that
do
not
require
subsoiling.
No­
till
systems
are
gaining
in
popularity
in
these
locales,
as
the
soils
have
higher
levels
of
organic
matter
and
are
often
the
most
productive.
Traditionally,
these
soils
have
been
heavily
tilled,
utilizing
two
disking
operations
followed
by
subsoiling/
bedding.
Strip­
till
is
increasing
dramatically
in
this
area
as
a
method
of
controlling
sand
blasting.
The
heavier
clay
soils
of
the
piedmont
do
not
require
subsoiling,
and
most
of
this
cotton
is
produced
in
no­
till
systems.

Planting
begins
in
mid­
April
and
usually
is
finished
by
the
end
of
May
(
most
active
is
May
1
to
May
29).
Harvesting
begins
at
the
end
of
September
and
ends
mid
December
(
most
active
is
October
15
to
November
15).

4.
WA
Potato
A
major
crop
for
Washington
is
Irish
Potatoes,
grown
primarily
in
the
central
portion
of
the
state
(
Figure
II.
D.
7.8).
Washington
is
the
second
leading
potato­
growing
state
in
the
U.
S.
(
WA
Dept.
of
Ag,
Statistics).
The
highest
potatoproducing
areas
within
the
state
are
located
in
Grant
and
Yakima
counties.
These
are
also
the
counties
with
the
highest
usage
of
carbamates
in
the
northwest.
Grant
County
was
the
top
potato­
producing
county
in
the
nation
in
1998
(
WA
Dept.
of
Ag,
Statistics),
with
almost
50,000
acres
in
potato
production
in
2002
(
USDA,
NASS
2002
Census
of
Agriculture).
Groundwater
is
generally
shallow
in
Grant
County,
with
little
or
no
overlying
confining
layer
in
most
places.
As
is
typical
for
areas
where
potato
production
is
favorable,
the
soils
tend
to
be
coarse­
grained
and
well
drained
(
Sieczka
and
Thornton,
1993).
Sandy,
well­
drained
soils
are
also
likely
to
be
fairly
low
in
organic
carbon;
carbon
(
and
nutrient)
loss
is
exacerbated
by
common
potato
production
practices,
such
as
heavy
tillage
(
Lang
et
al.,
1999).
These
factors
make
Grant
County
especially
susceptible
to
groundwater
pesticide
contamination.
Figure
II.
D.
7.8,
location
of
the
hypothetical
central
Washington
potato
scenario.
Section
II.
D.
7
­
Page
15
of
35
II.
D.
7.
8
­
Washington
State
County
Map.
Grant
County
(
circled)
and
Yakima
County
are
the
major
potato­
producing
counties
in
the
state.

a.
Soil
Potatoes
are
primarily
grown
in
Entisols
(
Torriorthents,
Torripsamments)
and
Aridisols
(
Haplocambids,
Haplodurids)
in
central
Washington,
which
are
generally
characterized
by
low
clay
content,
good
drainage,
and
low
organic
content
(
USDA
Keys
to
Soil
Taxonomy,
2003).
The
most
productive
and
widespread
soil
series
cropped
with
potato
in
Grant
County
(
and
other
areas
within
the
central
Washington
region)
are
Kennewick,
Sagehill,
and
Wiehl
(
USDA­
NRCS,
Soil
Data
Mart).
These
soils
tend
to
be
coarse­
grained
(
sand
fractions
typically
range
from
60­
80%)
with
low
organic
matter
content
(<
1%).
Most
of
these
soils
are
in
hydrologic
group
B.
Several
relevant
soil
properties
are
listed
in
Table
II.
D.
7.4.
Section
II.
D.
7
­
Page
16
of
35
b.
Irrigation
Almost
all
potato
crops
grown
in
the
Pacific
Northwest
(
and
throughout
the
U.
S.)
are
irrigated
(
Thomson
et
al.,
1999),
largely
because
potatoes
are
grown
in
sandy,
welldrained
soils
and
because
of
crop
water
demand.
Information
included
herein
regarding
crop
yields,
production
techniques,
etc.
assume
at
least
some
irrigation
is
conducted.
Therefore
irrigation
water
must
be
included
in
total
water
inputs
to
the
system
(
e.
g.,
precipitation
plus
irrigation
water).
Irrigation
at
50%
of
AWC
is
used
for
this
scenario.

c.
Potato
Production
and
Crop
Profile
Potatoes
are
among
the
most
expensive
major
crops
to
grow.
They
require
greater
amounts
of
fertilizers
(
especially
N)
and
pesticides
than
grain
and
feed
crops,
and
need
more
intensive
management
(
tillage,
equipment,
monitoring).
Potato
confers
very
little
organic
material
to
the
soil.
Although
potatoes
will
grow
on
a
wide
variety
of
soils,
optimal
soils
are
usually
deep,
coarse­
grained,
and
well­
drained.
Sandy
soils
are
particularly
good
for
potato
production.
Generally,
soils
with
little
or
no
slope
are
preferable,
so
that
runoff
is
minimized
(
and
less
water
and
organic
matter
is
lost).
Table
II.
D.
7.4.
Washington
Soil
Properties
for
potato.
Soil
Series
Depth
(
in)
Texture
Sand
(%)
Silt
(%)
Clay
(%)
Saturated
Hydraulic
Conductivity
(
µ
m/
s)
Organic
Matter
(%)
Moist
Bulk
Density
(
g/
cc)
Available
Water
Capacity
pH
Kennewick
0­
9
Loamy
fine
sand
61.5­
79.9
16.6­
35.0
3.5
4.0­
14.0
0.5­
1.0
1.15­
1.45
0.11­
0.17
7.4­
8.4
9­
60
Fine
sandy
loam
­­­
­­­
3.0­
18.0
1.4­
4.0
0­
0.5
1.3­
1.5
0.18­
0.21
7.9­
9.0
Sagehill
0­
8
Very
fine
sandy
loam
59.7
35.3
5.0
14.0­
42.0
0­
0.5
1.2­
1.4
0.18­
0.2
6.6­
8.4
8­
19
Silt
loam
­­­
­­­
2.0­
8.0
14.0­
42.0
0­
0.5
1.3­
1.55
0.18­
0.2
6.6­
8.4
19­
60
Fine
sandy
loam
­­­
­­­
2.0­
8.0
4.0­
14.0
0­
0.5
1.3­
1.6
0.18­
0.2
7.9­
9.0
Wiehl
0­
8
Fine
sandy
loam
66.0
27.5
6.5
14.0­
42.0
0.5­
1.0
1.2­
1.4
0.13­
0.17
7.4­
7.8
8­
18
Silt
loam
­­­
­­­
5.0­
8.0
4.0­
14.0
0­
0.5
1.3­
1.5
0.15­
0.19
6.6­
7.8
18­
25
Gravelly
silt
loam
­­­
­­­
5.0­
8.0
4.0­
14.0
0­
0.5
1.3­
1.5
0.13­
0.17
7.4­
8.4
25­
35
Weathered
bedrock
­­­
­­­
­­­
­­­
­­­
1.6­
1.9
­­­
­­­
Section
II.
D.
7
­
Page
17
of
35
Potato
requires
consistent
amounts
of
water
throughout
its
growing
cycle.
Seasonal
requirements
range
from
~
20­
40
inches.
It
is
advantageous
to
keep
the
soil
near
field
capacity;
soil
should
not
be
allowed
to
get
below
65%
of
field
capacity.
However,
soil
should
not
be
allowed
to
exceed
field
capacity,
or
quality
and
yield
will
become
dramatically
lowered.
The
effective
rooting
depth
of
potato
is
2
feet
(
Sieczka
and
Thornton,
1993).

d.
Water
Table
and
Aquifer
Characteristics
The
water
table
associated
with
this
type
of
soil
in
this
region
has
an
upper
limit
of
1
to
5
feet
and
a
lower
limit
greater
than
6
feet
(
USDA
Soil
Data
Mart).
Groundwater
pH
for
this
region
ranges
from
6.7­
7.8;
typical
pH
for
shallow
ground
water
is
generally
around
7.2
(
personal
communication,
Washington
State
Dept.
of
Ecology).

D.
Chemical
Property
Inputs
Only
two
of
the
N­
methyl
carbamates
 
aldicarb
and
oxamyl
 
were
included
in
the
groundwater
assessment
because
the
use
data
(
appendix
IID.
2)
and
the
relative
toxicity
suggest
that
these
would
be
the
major
contributors
to
ground
water
exposure
in
the
cumulative
assessment.
The
chemical
properties
that
drive
these
assessments
are
given
in
Table
II.
D.
7.5.
These
properties
came
from
an
evaluation
of
registrant­
submitted
studies
and
represent
reasonably
conservative
estimates
of
these
properties.
Other
chemical
properties
are
required
as
inputs
in
order
for
the
models
to
operate,
but
they
have
negligible
effect
on
model
output;
these
properties
can
be
found
in
the
model
input
files.
Properties
for
aldicarb
represent
total
residue
(
parent
aldicarb,
plus
the
degradates
aldicarb
sulfoxide
and
aldicarb
sulfone)
properties
(
i.
e.
half­
lives
represent
the
most
stable
toxic­
relevant
constituent).

Table
II.
D.
7.5.
Relevant
Chemical
Inputs
for
Carbamates.
Koc
Soil
Metabolism
halflife
Hydrolysis
(
acid)
halflife
Hydrolysis
(
neutral)
halflife
Aldicarb
(
total
residue)
10
ml/
g
55
days
500
63
days
oxamyl
6
ml/
g
20
days
stable
8
days
Section
II.
D.
7
­
Page
18
of
35
E.
Results
and
Discussion
1.
Florida
Citrus
a.
Florida
Citrus
/
Aldicarb
Modeling
results
for
the
Florida
citrus
scenario
are
shown
in
Figure
II.
D.
7.9.
The
concentrations
estimated
by
RZWQM
are
spikier
than
those
of
PRZM
and
LEACHP,
but
on
average
the
concentrations
are
quite
similar.
Although
peak
concentrations
are
estimated
to
be
higher
in
RZWQM
than
for
PRZM,
the
overall
average
RZWQM
concentration
is
30%
lower
than
PRZM.
For
the
1
lb/
acre
application
rate,
the
average
concentration
from
PRZM
is
27
ppb
versus
19
ppb
estimated
by
RZWQM.
At
the
labeled
rate
of
5
lb/
acre
and
without
any
setback
considerations,
these
values
would
be
135
and
95
ppb,
respectively.
Setbacks
considerably
reduce
the
concentrations.
The
1000­
ft
set
back
reduces
concentrations
to
about
4%
of
the
application
site
concentrations,
whereas
the
300­
ft
setback
reduces
concentrations
to
about
half
of
the
site
concentration.
With
a
1000­
ft
buffer
concentrations
are
well
below
10
ppb.

II.
D.
7.
9
­
Florida
Citrus
Results
for
RZWQM,
PRZM,
and
LEACHP.
Application
of
aldicarb
at
1
lb/
acre.
For
other
application
rates,
the
output
concentrations
are
proportional.
Distances
in
legend
refer
to
setback
distances
as
specified
by
state
and
federal
regulations.

0
50
100
150
200
4/
1/
91
8/
18/
95
1/
4/
00
5/
22/
04
10/
8/
08
2/
24/
13
7/
13/
17
11/
29/
21
Date
Aldicarb
Residues
(
ppb)
No
Buffer:
RZWQM
No
Buffer:
LEACHP
No
Buffer:
PRZM
300
ft
Buffer:
RZWQM
300
ft
Buffer:
LEACHP
300
ft
Buffer:
PRZM
1000
ft
buffer:
RZWQM
1000
ft
Buffer:
LEACHP
1000
ft
Buffer:
PRZM
Section
II.
D.
7
­
Page
19
of
35
b.
Monitoring
in
Florida
/
Aldicarb
These
results
are
conservative
with
respect
to
a
recent
and
on­
going
groundwater
monitoring
study
on
the
Florida
Central
Ridge
(
http://
fisc.
er.
usgs.
gov/
Lake_
Wales_
Ridge/).
The
USGS
and
the
Florida
Department
of
Agriculture
is
monitoring
31
wells
within
and
around
citrus
groves
on
the
Ridge
(
the
area
of
the
OPP
scenario).
Well
depths
range
from
4
feet
to
110
feet
deep
(
two
thirds
in
the
20
to
60
foot
range),
and
pH
ranged
from
3.9
to
6.9
(
median
about
5).
Concentrations
as
high
as
23
ppb
have
been
recorded
in
one
26­
ft
well,
while
a
4­
ft
well
had
reported
concentrations
as
high
as
21
ppb.
Summarized
results
are
shown
in
FigureII.
D.
7.10.
This
study
is
not
targeted
for
any
specific
pesticide,
but
rather
is
designed
as
a
survey
mechanism
 
that
is,
it
is
not
known
how
much
aldicarb
was
used
nor
is
it
known
how
far
aldicarb
was
used
from
the
wells.

II.
D.
7.
10
­
Monitoring
results
from
the
Lake
Wales
Central
Ridge.
Draft
data
subject
to
revision.
Connected
lines
represent
individual
wells
Aldicarb
+
Metabolites
(
LWRMN)

0
5
10
15
20
25
April
1999
Oct/
Nov
1999
Apr/
May
2000
Oct
2000
Apr
2001
Oct
2001
Apr
2002
Oct
2002
Apr
2003
Oct
2003
Apr
2004
Oct
2004
Sample
Date
Aldicarb+
Ald.
Sulfoxide+
Ald.
Sulfone
Concentration
(
ug/
L)

In
another
monitoring
study,
Jones
et
al.
(
1987)
applied
10
lb/
acre
(
twice
the
current
allowed
label
rate)
to
a
grove
plot
on
the
Central
Ridge
and
observed
aldicarb
residues
in
the
saturated
zone
below
the
plot
as
high
as
100
to
500
ppb.
Concentrations
of
over
100
ppb
were
observed
at
distances
of
up
to
300
feet
from
the
point
of
application
(
Fig
6
of
Jones
et
al,
1987).
With
consideration
that
the
Jones
et
al.
(
1987)
application
rate
was
twice
as
high
as
current
labeled
rates,
our
modeling
scenarios
would
appear
to
be
reasonably
conservative
for
the
types
of
soil
/
ground
water
profiles
modeled.
Section
II.
D.
7
­
Page
20
of
35
The
Florida
Department
of
Environmental
Protection
(
FDEP)
monitors
private
drinking
water
wells
in
rural
areas.
The
monitoring
is
not
comprehensive,
but
instead
is
instituted
when
there
has
been
an
indication
of
a
problem
(
personal
communication,
FDEP).
Total
aldicarb
residues
(
parent,
sulfoxide
and
sulfone
degradates)
as
high
as
47
ppb
were
reported
in
private
drinking
water
wells
in
the
early
1990s
in
the
FDEP
study.
The
concentrations
dropped
off
in
subsequent
years.
The
reduction
in
concentrations
of
aldicarb
may
have
resulted
from
label
changes
which
reduced
application
rates
and
applied
well
setback
requirements.
Specific
reductions
at
home
sites
also
were
also
likely
the
result
of
a
Florida
State
program
to
install
carbon
filters
or
to
pipe
water
in
from
treatment
facilities
when
contamination
was
found.
Other
reasons
for
the
decline
include
the
possibility
of
discontinued
use
in
the
vicinity
of
the
contaminated
areas
(
personal
communication
FDEP).

Jones
and
Estes
(
1995)
reviewed
monitoring
studies
for
aldicarb,
including
studies
conducted
on
Florida
Central
Ridge
citrus.
They
reported
that
119
potable
wells
out
of
4009
sampled
wells
in
Florida
had
aldicarb
residue
detections
and
that
33
of
those
had
concentrations
higher
than
10
ppb.
A
breakdown
of
the
percent
detections
on
the
vulnerable
Central
Ridge
was
not
made
available,
thus
these
values
represent
the
entire
state
(
not
specifically
Ridge
data).

c.
Florida
Citrus
/
Oxamyl
Model
results
from
oxamyl
use
on
the
Florida
citrus
scenario
are
given
in
Figure
II.
D.
7.11.
Again,
RZWQM
predicts
greater
daily
variations
in
concentrations
and
higher
peak
concentrations
than
does
PRZM.
Average
concentrations
are
close,
with
PRZM
giving
32
ppb
and
RZWQM
giving
41
ppb,
which
is
well
within
any
reasonable
expectation
of
certainty
from
these
models.
Oxamyl
concentrations
for
this
scenario
are
higher
than
for
aldicarb,
even
though
oxamyl
has
a
shorter
half­
life.
This
is
because
there
are
no
setback
requirements
for
oxamyl
as
there
are
for
aldicarb.

Monitoring
results
are
much
more
limited
for
oxamyl
than
for
aldicarb.
The
USGS
has
just
started
to
monitor
for
oxamyl
on
the
Florida
Ridge
(
Choquette
2005,
personal
communication),
and
when
these
results
become
available,
they
will
be
evaluated
with
respect
to
model
performance.
Oxamyl
was
monitored
for
but
not
detected
in
the
Florida
Department
of
Health
investigations
(
study
described
above
for
aldicarb).
Section
II.
D.
7
­
Page
21
of
35
II.
D.
7.
11
­
Model
Results
for
Oxamyl
concentrations
with
Florida
citrus
scenario
0
20
40
60
80
100
120
140
160
180
0
2000
4000
6000
8000
10000
12000
days
Oxamyl
Concentration
(
ppb)
RZWQM
PRZM
2.
North
Carolina
Results
for
North
Carolina
are
shown
in
Figure
II.
D.
7.12.
Aldicarb
is
the
only
carbamate
of
interest
in
this
area.
The
RZWQM
simulation
could
not
be
performed
with
this
scenario
due
to
difficulties
in
establishing
a
water
table.
This
problem
is
being
worked
out
with
assistance
from
USDA­
ARS.
PRZM
and
LEACHP
results
were
fairly
close.
OPP
is
searching
for
monitoring
data
for
comparison
with
model
results.
Section
II.
D.
7
­
Page
22
of
35
II.
D.
7.
12
­
PRZM
and
LEACHP
output
for
Aldicarb
Total
Residue
for
NC
cotton.
Application
rate
is
1
lb/
acre.

0
5
10
15
20
25
30
35
40
45
50
5/
1/
1961
11/
26/
1967
6/
22/
1974
1/
16/
1981
8/
13/
1987
3/
9/
1994
date
Aldicarb
Residues
(
ppb)
PRZM
LEACHP
3.
Georgia
Peanut
Results
for
Georgia
Peanuts
are
presented
in
Figure
II.
D.
7.13.
Note
that,
as
with
all
simulations
here,
concentrations
are
normalized
to
1
lb/
acre;
actual
rates
may
be
higher
or
lower,
but
output
will
be
proportional
to
application
rate.
Because
labels
do
not
require
buffers
for
Georgia
Peanuts,
no
adjustments
for
buffers
were
included.
Model
simulations
of
groundwater
concentrations
in
Georgia
are
typically
lower
than
those
in
Florida
(
without
buffers).
This
is
likely
due
to
the
lower
rainfall
and
less
irrigation
used
on
peanuts,
which
causes
pesticide
to
dwell
longer
in
the
upper
1­
meter
degradation
zone.
LEACHP
concentrations
are
somewhat
higher
than
PRZM
or
RZWQM
calculations
in
this
scenario.
We
are
currently
examining
possible
causes
for
this.
OPP
is
searching
for
monitoring
data
for
comparison
with
model
results.
Section
II.
D.
7
­
Page
23
of
35
II.
D.
7.
13
 
PRZM,
RZWQM,
and
LEACHP
results
for
aldicarb
for
the
Georgia
peanut
scenario.
All
concentrations
are
normalized
to
1
lb/
acre.

0
10
20
30
40
50
60
70
80
0
2000
4000
6000
8000
10000
12000
days
Total
Aldicarb(
ppb)
PRZM
RZWQM
LEACHP
4.
Washington
Potato
Both
PRZM
and
RZWQM
simulated
negligible
concentrations
in
the
aquifer,
consistent
with
an
absence
of
detections
from
wells
in
this
area
(
Kirk
Cook,
Washington
State
Dept.
of
Agriculture,
Pesticide
Management
Division,
personal
communication).
This
was
primarily
due
to
the
alkaline
pH
of
the
system
which
enhanced
hydrolysis.
Further
analysis
was
discontinued
for
his
scenario.

F.
European
Approach
to
Ground
Water:
FOCUS
Because
EPA's
proposed
approach
to
addressing
groundwater
issues
is
new,
a
review
of
the
approach
that
the
European
community
uses
is
useful
for
placing
the
EPA's
assessment
into
world
perspective.
The
European
community
addresses
standardization
of
evaluation
of
pesticide
fate
and
transport
through
FOCUS
(
FOrum
for
Co­
ordination
of
pesticide
fate
models
and
their
Use;
information
available
at
http://
viso.
ei.
jrc.
it/
focus/
index.
html).
A
review
of
FOCUS
and
a
comparison
of
EPA's
results
with
those
that
would
be
derived
from
FOCUS
methods
are
presented
here.
Section
II.
D.
7
­
Page
24
of
35
1.
Background
on
European
Regulatory
Modeling
The
FOCUS
groundwater
scenarios
are
a
set
of
nine
standard
combinations
of
weather,
soil
and
cropping
data
which
collectively
represent
agriculture
in
the
European
Union
member
countries
for
the
purposes
of
a
Tier
1
European
Union­
Level
assessment
of
leaching
potential.
The
scenarios
and
their
derivation
are
described
in
detail
in
a
published
report
(
FOCUS,
1995).
The
scenarios
have
been
implemented
as
sets
of
input
files
for
four
simulation
models
­
MACRO,
PEARL,
PELMO
&
PRZM
(
the
reader
should
note
that
PRZM
maintenance
and
improvement
through
FOCUS
is
done
separately
from
PRZM
as
used
by
U.
S.
EPA
regulators;
at
the
present
time
there
is
not
an
exact
match
between
versions
of
FOCUS
PRZM
and
PRZM
used
by
U.
S.
EPA).

The
endpoints
of
concern
evaluated
in
FOCUS
modeling
are
driven
by
different
legislative
directives
than
in
the
United
States.
The
use
of
pesticides
that
potentially
contaminate
the
groundwater
is
banned
by
registration
procedures
at
both
the
European
level
(
Council
Directive
91/
414/
EEC),
and
the
level
of
individual
member
states
(
countries).
The
Directive
places
great
importance
on
the
use
of
models
to
calculate
Predicted
Environmental
Concentrations
(
PECs)
as
a
basis
for
assessing
the
environmental
risks.
Under
the
directive
pesticide
active
ingredients
and
relevant
metabolites
thereof
must
not
exceed
a
concentration
of
0.1
ug/
l
in
groundwater.

In
the
first
tier
of
the
current
procedure,
point
scale
leaching
models
are
combined
with
a
limited
number
of
worst­
case
scenarios
to
assess
PEC
groundwater
in
Europe.

FOCUS
model
shells
were
used
for
running
the
models.
The
FOCUS
shells
guide
the
user
to
select
input
parameter
values
according
to
standardized
procedures.
The
use
of
the
standardized
procedures
means
that
some
key
input
values
may
be
different
than
was
selected
for
the
Carbamate
Cumulative
groundwater
modeling
of
US
scenarios.
For
example,
the
variance
of
pesticide
half­
life
with
depth
(
for
this
carbamate
assessment)
used
the
following
standardized
protocol
for
FOCUS
modeling:

A
"
depth
dependent
correction
factor"
was
applied
to
the
pesticide
degradation
rates
as
follows:
0
 
30
cm
depth
1
30
 
60
cm
depth
0.5
60
 
100
cm
depth
0.3
>
100
cm
depth
0
Therefore,
in
the
simulations
run
for
total
aldicarb
residues,
aldicarb
degradation
rate
was
reduced
by
half
as
aldicarb
residues
were
passing
through
the
30
to
60
cm
depth,
by
70
%
traveling
between
60
and
100
cm
deep,
and
no
degradation
took
place
for
any
residues
that
leached
below
100
cm.

FOCUS
models
may
use
non­
linear
adsorption
of
the
pesticide.
FOCUS­
PRZM
differs
from
the
PRZM
version
currently
in
use
by
EFED
in
this
respect.
Carbamate
modeling
used
the
FOCUS
default
assumptions
and
calculations
as
follows:


normalized
Freundlich
equation
used;
Section
II.
D.
7
­
Page
25
of
35

layer
specific
Kd
calculated
from
the
PRZM
shell
and
inputted
into
PRZM;


Freundlich
exponent
1/
n
is
entered
in
using
the
default
value
of
0.9
for
n
as
specified
in
the
FOCUS
guidance
document.

2.
Interpretation
of
FOCUS
Model
Output
based
on
Legislative
Directives
in
the
European
Union
The
model
shells
rank
the
twenty
mean
annual
concentrations
(
in
soil
pore
water
moving
past
a
depth
of
1
meter)
from
lowest
to
highest.
The
seventeenth
value
(
fourth
highest)
is
used
to
represent
the
80th
percentile
value
associated
with
weather
for
the
specific
simulation
conditions.
Under
the
FOCUS
modeling
system
this
value
is
taken
as
the
overall
90th
percentile
concentration
for
a
combination
of
soil
and
weather
conditions
since
the
standard
scenario
site
soil
characteristics
have
also
been
chosen
to
represent
an
overall
80th
percentile
vulnerability.
In
the
European
Union,
evaluation
of
model
output
is
made
independent
of
the
specific
toxicological
properties
of
the
pesticide.
1
3.
Comparison
of
FOCUS
and
Carbamate
Cumulative
Modeling
Scenarios
Four
of
the
nine
FOCUS
scenarios
were
applicable
to
citrus
production
agriculture:
Piancenza,
Porto,
Sevilla,
and
Thiva.
With
the
possible
exception
of
the
Porto
site,
these
sites
represent
drier
climates
(
Table
II.
D.
7.6)
and
less
permeable
soils
(
Tables
II.
D.
7.7,
8,
9,
and
10)
than
the
scenarios
used
in
the
Carbamate
Cumulative
assessment
other
than
the
low
natural
rainfall
but
heavily
irrigated
Washington
state
potato
scenario.

Table
II.
D.
7.6.
European
Union
climate
zones
and
predominance
of
agriculture
represented
by
the
FOCUS
standard
scenarios
applicable
to
citrus
uses.

Precipitation
(
mm)
Mean
Annual
Temperature
(
°
C)
Arable
land
*
(%)
Total
Area
*
(%)
Representative
Locations
601
to
800
>
12.5
13
11
Sevilla/
Thiva**

1
Regulators
in
Europe
compare
the
resultant
chosen
values
for
pesticide
concentration
in
leachate
against
the
0.1
ug/
l
standard
for
groundwater;
with
the
three
possible
categories
of
outcomes
being
exceedences
at
none
of
the
sites,
at
some
of
the
sites,
or
at
all
of
the
sites.
If
residues
do
not
exceed
01
ug/
l
at
any
of
the
sites
then
the
pesticide
passes
Tier
1
and
may
be
registered
because
of
its
low
probability
of
leaching
to
groundwater
at
concentrations
in
excess
of
0.1
ug/
l.
The
FOCUS
guidance
document
does
offer
the
following
caveat,
however:
"
This
does
not
exclude
the
possibility
of
leaching
in
highly
vulnerable
local
situations
within
specific
Member
States,
but
such
situations
should
not
be
widespread
and
can
be
assessed
at
the
Member
State
level."
When
exceedences
occur
at
some
of
the
sites,
then
registration
may
only
be
approved
in
the
regions
represented
by
the
scenarios
where
the
pesticide
concentrations
were
below
0.1
ug/
l.
Registration
may
not
be
approved
if
the
Tier
1
concentrations
are
above
0.1
ug/
l.
The
FOCUS
guidance
does
provide,
however,
for
the
possibility
that
Tier
1
failures
can
be
overruled
by
substantial
contrary
evidence
(
that
residues
in
groundwater
will
not
in
fact
exceed
0.1
ug/
l
under
normal
use
patterns)
obtained
through
"
lysimeter
or
field
leaching
studies,
monitoring
and
more
refined
modeling."
Section
II.
D.
7
­
Page
26
of
35
801
to
1000
>
12.5
9
8
Piacenza
<
600
>
12.5
4
4
Sevilla/
Thiva
1001
to
1400
>
12.5
3
3
Porto
*
Relative
to
the
area
of
the
European
Union
plus
Norway
and
Switzerland.
**
Although
these
locations
have
less
than
600
mm
of
precipitation,
irrigation
typically
used
at
these
two
locations
brings
the
total
amount
of
water
to
greater
than
600
mm.

Table
II.
D.
7.7.
Soil
parameters
for
FOCUS
Porto
scenario.

depth
classification
pHH2O
Texture
µ
m
om
oc
bulk
density
depth
factor@
cm
<
2
2­
50
>
50
%
%
g
cm­
3
­
0
­
35
loam
4.9
10
48
42
6.6
3.8
0.89
1.0
35
­
60
sandy
loam
4.8
8
31
61
3.7
2.1
1.25
0.5
60
­
100
sandy
loam
4.8
8
31
61
3.7
2.1
1.25
0.3
100
­
120
sandy
loam
4.8
8
31
61
3.7
2.1
1.25
0.0
@
Factor
applied
to
topsoil
degradation
rate.

Table
II.
D.
7.8.
Soil
parameters
for
FOCUS
Piacenza
scenario.

depth
classification
pH
texture
µ
m
om
oc
bulk
density
depth
factor@
cm
<
2
2­
50
>
50
%
%
g
cm­
3
0­
30
loam
7
15
45
40
1.72
1.00
1.3
1.0
30­
40
loam
7
15
45
40
1.72
1.00
1.3
0.5
40­
60
silt
loam
6.3
7
53
40
0.64
0.37
1.35
0.5
60­
80
silt
loam
6.3
7
53
40
0.64
0.37
1.35
0.3
80­
100
sand
6.4
0
0
100
0
0
1.45
0.3
100­
170
sand
6.4
0
0
100
0
0
1.45
0.0
@
Factor
applied
to
topsoil
degradation
rate.

Table
II.
D.
7.9.
Soil
parameters
for
FOCUS
Sevilla
scenario.

depth
classification
pH
texture
µ
m
om
oc
bulk
density
depth
factor@
cm
<
2
2­
50
>
50
%
%
g
cm­
3
­
0­
10
silt
loam
7.3
14
51
35
1.6
0.93
1.21
1.0
10­
30
silt
loam
7.3
13
52
35
1.6
0.93
1.23
1.0
30­
60
silt
loam
7.8
15
51
34
1.2
0.70
1.25
0.5
60­
100
clay
loam
8.1
16
54
30
1.0
0.58
1.27
0.3
100­
120
clay
loam
8.1
16
54
30
1.0
0.58
1.27
0.0
120­
180
clay
loam
8.2
22
57
21
0.85
0.49
1.27
0.0
@
Factor
applied
to
topsoil
degradation
rate.

Table
II.
D.
7.10.
Soil
parameters
for
FOCUS
Thiva
scenario.

depth
classification
pHKCl
texture
%
om
%
oc
bulk
density
depth
factor
<
2
2­
50
>
50
%
%
g
cm­
3
­
Section
II.
D.
7
­
Page
27
of
35
0­
30
loam
7.0
25.3
42.8
31.9
1.28
0.74
1.42
1.0
30­
45
loam
7.0
25.3
42.8
31.9
1.28
0.74
1.42
0.5
45­
60
clay
loam
7.1
29.6
38.7
31.7
0.98
0.57
1.43
0.5
60­
85
clay
loam
7.1
31.9
35.7
32.3
0.53
0.31
1.48
0.3
85­
100
clay
loam
7.1
32.9
35.6
31.5
0.31
0.18
1.56
0.3
100­???
clay
loam
7.1
32.9
35.6
31.5
0.31
0.18
1.56
0.0
@
Factor
applied
to
topsoil
degradation
rate.

4.
Results
of
FOCUS
Modeling
for
Aldicarb
Total
Residues
The
distribution
of
average
annual
concentrations
of
total
aldicarb
residues
in
leachate
is
provided
in
Figure
II.
D.
7.14.
Tier
1
screening
concentrations
were
50
to
80
ug/
l
for
the
four
citrus
production
scenarios
(
Table
II.
D.
7.11).

II.
D.
7.
14
­
Distribution
of
annual
average
concentrations
of
total
aldicarb
concentrations
at
1
meter
depth
after
application
of
1
lb
ai/
A
to
citrus
(
FOCUS
European
regulatory
scenarios
­
PRZM
simulations).

0
20
40
60
80
100
120
0%
20%
40%
60%
80%
100%

Exceedance
Probability
Annual
Average
at
1
m,
ug/
L
Piancenza
Porto
Sevilla
Thiva
There
are
major
factors
which
tend
to
make
the
FOCUS
screening
concentrations
either
more
or
less
"
worst
case"
than
the
Carbamate
Cumulative
output:


FOCUS
concentrations
represent
shallower
water
(
1
m
versus
3.5
m
in
the
Carbamate
Cumulative)
and
concentrations
in
leachate
below
the
specified
depth
rather
than
concentrations
in
groundwater.


FOCUS
concentrations
represent
annual
mean
rather
than
daily
water
concentrations
Section
II.
D.
7
­
Page
28
of
35

The
FOCUS
scenarios
were
very
likely
less
vulnerable
than
the
Carbamate
Cumulative
scenarios
in
terms
of
soils
and
climate
(
soils
much
less
sandy;
precipitation
generally
lighter).

In
spite
of
the
significant
differences
in
the
setup
and
evaluation
of
modeling,
the
FOCUS
modeling
yielded
results
within
the
range
of
Carbamate
Cumulative
modeling
for
aldicarb
residues.
This
is
important
to
know
since
the
FOCUS
models
have
already
been
through
extensive
use
by
regulators
in
Europe
for
several
years
now.
Future
work
may
be
done
with
other
FOCUS
models
such
as
PEARL
and
PELMO
to
confirm
if
similar
results
are
obtained
with
the
other
FOCUS
models.

Table
II.
D.
7.11.
Selected
Tier
1
concentrations
for
total
aldicarb
based
on
a
1
lb
ai/
A
application
rate;
EU
citrus
scenarios.
Tier
1
values
are
calculated
per
the
FOCUS
guidance
document.

European
(
FOCUS)
Site
Name
Tier
1
ground
water
concentration,
ug/
l
Piancenza
79.87
Porto
49.52
Sevilla
40.28
Thiva
50.06
G.
Prospective
Groundwater
Monitoring
(
PGW)
Studies
As
part
of
the
pesticide
registration
process,
registrants
are
sometimes
required
to
perform
prospective
ground
water
studies
(
PGW)
in
order
to
better
evaluate
the
fate
and
transport
of
a
pesticide
as
it
moves
from
a
field
site
into
the
underlying
ground
water.
A
small
number
of
studies
have
been
performed
on
pesticides
that
are
included
in
the
list
of
n­
methyl
carbamate
cumulative
chemicals.
These
include
Oxamyl
in
North
Carolina,
Oxamyl
in
Maryland,
and
Methomyl
in
Georgia.
The
EPA
is
evaluating
model
performance
with
regard
to
these
studies.
The
work
is
ongoing
and
preliminary
results
are
presented
in
this
section.
As
of
this
writing,
only
the
analysis
of
the
North
Carolina
oxamyl
study
has
been
completed
to
enough
of
an
extent
to
present
results.

1.
North
Carolina
A
small­
scale
prospective
groundwater
monitoring
study
was
conducted
for
oxamyl
and
its
oxime
metabolite
in
Tarboro,
North
Carolina.
The
study
is
located
in
the
same
coastal
plain
region
modeled
in
the
North
Carolina
cotton
scenario.
The
study
site
was
chosen
for
its
highly
vulnerable
soil
and
hydrogeologic
characteristics.
The
soil
at
the
site
is
relatively
homogeneous
sand
to
loamy
sand
with
a
layer
of
sandy
loam
to
sandy
clay
loam
at
approximately
two
to
four
feet.
It
correlates
with
the
NRCS
Tarboro
loamy
sand
series,
characterized
by
excessive
drainage
and
negligible
runoff.
The
top
one
foot
of
soil
has
an
average
organic
matter
content
of
0.85%
and
a
pH
of
5.8.
Below
this,
the
organic
matter
content
ranges
from
0.10
to
0.23%
while
the
pH
ranges
from
4.3
to
7.9,
generally
lower
at
the
top
and
increasing
with
depth.
Based
on
undisturbed
soil
samples,
the
average
field
capacity
is
9.6%
in
the
top
two
feet
and
15.1%
from
two
to
four
feet
and
the
bulk
density
at
those
depths
averages
1.42
g/
cm3.
Section
II.
D.
7
­
Page
29
of
35
The
study
site
has
a
history
of
cotton,
soybeans,
peanuts,
tobacco,
and
corn
production.
For
this
investigation,
cotton
was
planted
on
May
22,
1997
and
multiple
applications
of
oxamyl
as
well
as
a
single
application
of
a
conservative
bromide
tracer
were
subsequently
applied.
The
cotton
was
harvested
in
November
and
peanuts
planted
the
following
summer.
Precipitation
was
supplemented
with
overhead
center
pivot
irrigation
to
bring
the
combined
precipitation
and
irrigation
to
56.41
in.,
120%
of
the
historical
mean
precipitation
(
Figure
II.
D.
7.15).

II.
D.
7.
15
­
Precipitation
and
irrigation
throughout
the
study
period.
The
oxamyl
application
period
is
highlighted
in
light
green..

0.0
0.5
1.0
1.5
2.0
2.5
3.0
6/
1/
97
10/
1/
97
1/
31/
98
6/
2/
98
10/
2/
98
Date
Precipitation
+
Irrigation
(
in)

Eight
well
clusters,
including
one
shallow
(
12­
17
feet
screened
interval)
and
one
deep
well
(
17­
21
feet
screened
interval)
each,
were
used
to
monitor
groundwater
and
eight
clusters
of
lysimeters
at
four
depths
were
used
to
monitor
soil
pore
water.
During
the
study,
the
depth
of
the
water
table
ranged
from
10.27
to
17.15
feet,
with
a
mean
depth
of
14.09
feet
below
the
ground
surface.
Potassium
bromide
was
applied
as
a
conservative
tracer
on
July
1,
1997,
and
indicated
rapid
vertical
leaching
and
recharge
to
groundwater.
It
was
detected
in
soil
at
all
locations
at
the
18
to
24
inch
depth
by
the
first
measurement
at
27
days
after
treatment,
and
it
reached
groundwater
in
all
shallow
wells
by
160
days
after
treatment.

In
July,
a
series
of
5
ground
broadcast
applications
of
oxamyl
were
made
on
a
2
acre
plot
at
6
to
8
day
intervals.
The
first
two
applications
were
at
a
rate
of
0.5
lb/
A
and
the
rest
at
1.0
lb/
A.
This
represents
the
maximum
labeled
seasonal
rate
using
the
minimum
application
intervals.
Oxamyl
reached
all
shallow
wells,
initially
detected
between
days
124
and
194
after
treatment.
In
one
well,
oxamyl
persisted
throughout
the
entire
study
period
while
in
the
others
there
were
no
detections
beyond
376
days.
The
maximum
detection
was
3.91
ppb
(
Figure
Section
II.
D.
7
­
Page
30
of
35
II.
D.
7.16).
Oxamyl
was
only
detected
in
5
of
the
deeper
wells,
appearing
by
day
194
after
treatment
and
undetected
by
day
378.
The
range
of
concentrations
detected
at
this
depth
was
0.12
to
1.17
ppb
(
Figure
II.
D.
7.16).

II.
D.
7.
16
 
Oxamyl
concentrations
in
shallow
wells
(
top)
and
deep
wells
(
bottom).
Wells
are
grouped
into
subplots
A,
B,
and
C,
where
A
is
the
most
northern.
Within
each
subplot,
wells
are
listed
upgradient
to
downgradient.
Odd
numbered
shallow
wells
share
a
cluster.

Modeling
work
on
this
data
has
begun
with
an
analysis
of
the
bromide
tracer
in
the
shallow
wells.
The
result
PRZM
and
LEACHP
are
given
in
Figure
II.
D.
7.17
along
with
the
bromide
data
(
RZWQM
simulations
have
not
been
completed
at
the
time
of
this
writing).
Both
PRZM
and
LEACHP
give
similar
responses,
although
LEACHP
simulates
a
later
arrival
time
than
PRZM.
Both
predict
higher
concentrations
than
the
data
show.
However
note
that
lateral
Section
II.
D.
7
­
Page
31
of
35
groundwater
velocities
at
this
location
are
quite
high
(
51
feet
per
year)
and
that
the
groundwater
flow
transverses
across
the
narrow
side
of
the
field.
Thus
advection
and
dispersion
could
have
caused
lower
concentrations
than
those
modeled
In
Figure
II.
D.
7.17,
the
highlighted
data
represent
the
wells
that
were
farthest
downgradient
at
the
site,
and
as
can
be
seen,
these
are
the
wells
with
the
highest
concentrations.

Preliminary
work
with
the
parent
oxamyl
has
shown
that
the
models
predict
substantially
higher
oxamyl
concentrations
than
those
shown
in
Figure
II.
D.
7.16.
OPP
is
examining
possible
causes
and
will
report
results
when
available.
Possibilities
include
an
underestimation
of
the
rate
of
degradation
in
the
models.
For
example,
under
anaerobic
conditions
oxamyl
half
lives
are
on
the
order
of
5
days,
but
this
process
was
not
included
in
the
simulations.
Investigations
will
continue
and
refinements
may
be
necessary,
but
at
the
time
of
this
writing
modeling
results
have
not
compared
well
with
the
monitoring
data.

II.
D.
7.
17
­
Bromide
concentration
and
model
results
of
PRZM
and
RZWQM
for
the
North
Carolina
study.

0
5
10
15
20
25
0
100
200
300
400
500
600
Days
after
application
Conc.
(
ppm)
Bromide
at
15
ft
groundwater
PRZM
Simultaion
LEACHP
2.
Georgia
A
small­
scale
prospective
groundwater
(
PGW)
monitoring
study
was
conducted
in
Cook
County,
Georgia
from
1992­
1993
to
detect
the
presence
of
methomyl
in
ground
water,
soil­
pore
water
and
soil.
The
study
was
located
in
the
coastal
plain
region
 
the
same
region
chosen
for
the
Georgia
peanut
scenario
(
see
above).
The
site
was
selected
for
its
hydrologic
vulnerability:
high
annual
rainfall;
sandy,
low­
organic
matter,
permeable
soil
profile;
low
slope;
shallow
groundwater;
and
extensive
groundwater
recharge.
The
soil
profile
was
relatively
homogenous
with
no
restrictive
layers
and
was
characterized
as
Kershaw
fine
Section
II.
D.
7
­
Page
32
of
35
sandy
loam,
with
texture
characterized
as
sand.
The
mean
concentrations
of
sand,
silt
and
clay
were
95%,
4%,
and
2%,
respectively.
The
mean
organic
matter
content
in
the
upper
0.5
feet
of
the
profile
was
0.9%
and
the
mean
soil
pH
at
the
plow
layer
was
6.4.

The
study
site
had
been
under
agronomic
production
for
five
years
prior
to
the
study
and
was
planted
with
both
sweet
potatoes
and
peanuts.
For
this
investigation,
corn
was
planted
on
August
3,
1992
followed
by
multiple
applications
of
methomyl
and
a
single
application
of
a
conservative
bromide
tracer.
Precipitation
was
supplemented
via
a
traveling­
gun
system
to
bring
the
combined
precipitation
and
irrigation
to
124%
of
the
historical
mean
precipitation,
based
on
the
average
of
30
years
of
meteorological
data
from
two
nearby
weather
stations.

Four
two­
well
clusters,
one
shallow
and
one
deep,
were
installed
to
monitor
groundwater
and
fourteen
lysimeters
were
installed
to
monitor
soil­
pore
water
at
depths
of
3
feet,
6
feet,
9
feet
and
12
feet.
During
the
study,
the
depth
of
the
water
table
ranged
from
7.71
to
22.42
feet,
with
a
mean
depth
of
approximately
15
feet
below
the
ground
surface.
Potassium
bromide
was
used
as
a
tracer.
Transport
of
bromide
to
a
depth
of
42­
48
inches
took
between
22
and
34
days
and
was
detected
at
elevated
levels
62
DAT
in
one
of
the
shallow
wells
(
MW07).
Soil,
soil­
pore
water
and
groundwater
samples
were
taken
from
three
days
prior
to
application
of
methomyl
through
399
days
after
treatment.

Methomyl
was
first
applied
to
postemergent
corn
on
August
13,
1992.
There
were
five
whorl
treatments
spaced
approximately
10
days
apart
and
20
daily
ear
treatments,
conducted
at
0.45
lbs
ai/
acre
each
for
a
total
of
11.25
lbs
ai/
acre.
The
only
detections
of
methomyl
occurred
at
62
and
117
days
after
treatment
(
DAT).
The
maximum
detection
was
0.42
ppb
in
a
shallow
well
at
62
DAT.
Methomyl
was
detected
in
two
of
the
deep
wells
at
62
DAT
and
in
three
of
the
shallow
wells
(
two
at
62
DAT
and
two
at
117
DAT).
Only
one
shallow
monitoring
well,
MW07
had
detections
above
the
detection
limit
at
both
62
and
117
DAT.

The
conductors
of
the
study
concluded
that
no
detectable
concentrations
of
methomyl
were
present
in
the
groundwater
and
attributed
the
higher
concentrations
found
at
62
and
117
DAT
to
be
the
result
of
either
crosscontamination
(
field
or
lab)
or
the
presence
of
methomyl
in
the
soil
and/
or
groundwater
prior
to
the
conduction
of
this
study.
These
detections
could
also
represent
preferential
flow
delivery
to
the
ground
water.
The
presence
of
methomyl
prior
to
the
study
was
sampled
for,
and
at
the
detection
levels
of
0.1
ppb
for
water
and
2.0
ppb
for
soil,
no
methomyl
was
detected.
In
addition,
there
was
no
reported
prior
methomyl
use
at
the
study
site.

Preliminary
LEACHP
modeling
results,
shown
in
Figure
II.
D.
7.18,
indicate
a
plausible
Br
simulation.
We
are
further
investigating
how
well
Section
II.
D.
7
­
Page
33
of
35
methomyl
is
simulated;
no
methomyl
was
detected
in
the
PGW
study,
but
model
simulations
revealed
small
but
detectable
concentrations.
OPP
will
also
evaluate
similar
simulations
with
PRZM
and
RZWQM
using
the
same
data
set.

II.
D.
7.
18
­
Bromide
concentration
and
model
results
of
LEACHP
for
the
Georgia
study.

0
80
160
240
320
400
480
560
640
720
800
0
8
16
24
32
40
Concentration
(
mg/
L)

Days
LeachP­
methomyl
&
leachP_
Br;
PGW
wells
­
BR_
MW1,3,5,7
METHOMYL
LEACHP_
BR
BR_
MW1
BR_
MW5
BR_
MW3
BR_
MW7
H.
Outlook
There
is
no
clear
"
best"
model
to
use
to
assess
pesticide
concentrations
in
groundwater.
All
three
models
give
similar
concentrations
that
are
well
within
the
uncertainty
of
any
environmental
prediction.
Further
evaluations
will
be
required
before
OPP
makes
an
ultimate
decision
on
the
groundwater
model(
s)
to
be
used
for
general
assessments.
In
this
regard,
OPP
will
continue
its
effort
to
evaluate
the
models
on
the
performance
and
usability
will
continue
as
additional
groundwater
data
is
analyzed.
OPP
intends
to
analyze
additional
prospective
groundwater
studies,
and
is
beginning
to
systematically
enter
relevant
data
into
a
database.
Such
an
effort
should
allow
model
evaluation
on
a
broader
scale
than
just
the
available
carbamate
studies
that
were
examined
here.

OPP
is
continuing
to
work
with
USDA­
ARS,
EPA­
ORD,
and
Health
Canada
in
an
effort
to
improve
scenarios
and
models
for
pesticide
groundwater
assessments.
Some
of
the
issues
currently
being
worked
on
include
USDA­
ARS
investigation
into
ways
to
implement
a
fixed
water
table
in
RZWQM,
EPA
effort
to
fix
PRZM's
temperature
routines
in
PRZM,
and
Health
Canada's
efforts
to
develop
consistent
groundwater
Section
II.
D.
7
­
Page
34
of
35
scenarios.
These
collaborative
efforts
should
result
in
a
consistent
and
reasonable
process
for
estimating
pesticide
concentrations
in
groundwater.

I.
References
Boman,
B.,
Morris,
N.
Wade,
M.
2002.
Water
and
Environmental
Considerations
for
the
Design
and
Development
of
Citrus
Groves
in
Florida,
Circular
1419,
Institute
of
Food
and
Agricultural
Sciences,
University
of
Florida,
Gainesville,
FL
Choquette,
A.
F.,
and
Sepulveda,
A.
A.,
2000,
Design
of
a
shallow
ground
water
network
to
monitor
agricultural
chemicals,
Lake
Wales
Ridge,
Central
Florida:
U.
S.
Geological
Survey
Water­
Resources
Investigations
Report
00­
4134,
35
p.

Crandall
C.
A.
and
Berndt,
M.
P.
1996.
Water
Quality
of
Surficial
Aquifers
in
the
Georgia
 
Florida
Coastal
Plain,
Water­
Resources
Investigations
Report
95­
4269,
U.
S.
Geological
Survey,
Tallahassee,
FL.

Donohue,
J.
C.
2001.
Ground­
Water
Quality
in
Georgia
for
2000
Georgia
Department
of
Natural
Resources
,
Environmental
Prot4ection
Division,
Georgia
Geologic
Survey,
Circular
12P,
Atlanta,
GA.

FOCUS.
2000.
Leaching
Models
and
EU
Registration.
European
Commission
Document
4952/
VI/
95.
Also
available
at:
ftp://
ftp.
ei.
jrc.
it/
pub/
FOCUS/
documentation/
FOCUS_
GW_
Report_
Main.
pdf
Hornsby,
A.
G.,
Rao,
P.
S.
C.,
and
Jones,
R.
L.
1990.
Fate
of
Aldicarb
in
the
Unsaturated
Zone
Beneath
a
Citrus
Grove.
Water
Resources
Research,
26(
10),
2287­
2302.

Jones,
R.
L.
and
Estes
T.
L.
1995.
Summary
of
Aldicarb
Monitoring
and
Research
Programs
in
the
U.
S.
A.
Journal
of
Contaminant
Hydrology,
18,
107­
140.

Jones,
R.
L.
Hornsby,
A.
G.,
Rao,
P.
S.
C.
1987.
Movement
and
Degradation
of
Aldicarb
Residues
in
the
Saturated
Zone
Under
Citrus
Groves
on
the
Florida
Ridge.
Journal
of
Contaminant
Hydrology,
1(
3),
265­
285.

Ma,
Q.,
Hook,
J.
E.,
Wauchope,
R.
D.,
Dowler,
C.
C.,
Johnson,
A.
W.,
Davis,
J.
G.,
Gascho,
G.
J.,
Truman,
C.
C.,
Sumner,
H.
R.,
and
Chandler,
L.
D.
2000.
GLEAMS,
OPUS,
PRZM2

,
and
PRZM3
Simulations
Compared
with
Measured
Atrazine
Runoff
.
Soil
Sci.
Soc.
Am.
J.
64:
2070­
2079.

Noling,
2003.
University
of
Florida
IFAS
Extension,
ENY617,
Institute
of
Food
and
Agricultural
Sciences,
University
of
Florida.
(
also
available
at
http://
edis.
ifas.
ufl.
edu/
CH008)
Section
II.
D.
7
­
Page
35
of
35
Obreza,
T.
A.
and
Collins,
M.
E.
2002.
Common
Soils
Used
for
Citrus
Production
in
Florida,
BuSL­
193.
Institute
of
Food
and
Agricultural
Sciences,
University
of
Florida,
Gainesville,
FL
(
also
available
at
http://
edis.
ifas.
ufl.
edu/
SS403)

Parson,
L.
R.
and
Morgan,
K.
T.
2004.
Management
of
Microsprinkler
Systems
for
Florida
Citrus.
HS­
958,
Institute
of
Food
and
Agricultural
Sciences,
University
of
Florida,
Gainesville,
FL
Rawls,
W.
J.
and
D.
L.
Brakensiek.
1985.
Agricultural
management
effects
on
soil
water
retention.
IN:
DeCoursey,
D.
G.
(
ed.).
Proceedings
of
the
1983
Natural
Resources
Modeling
Symposium.
USDA.
Department
of
Agriculture,
Agricultural
Research
Service,
ARS­
30,
p
532
p.

Smajstra,
A.
G.
and
Harman,
D.
Z.
2002.
Irrigated
Acreage
in
Florida:
A
Summary
through
1998.
CIR
1220
,
Institute
of
Food
and
Agricultural
Sciences,
University
of
Florida,
Gainesville,
FL
Smajstrla,
A.
G.
B.
J.
Boman,
D.
Z.
Haman,
F.
T.
Izuno,
D.
J.
Pitts
and
F.
S.
Zazueta
1997.
BUL249.
Basic
Irrigation
Scheduling
in
Florida.
University
of
Florida
IFAS
Extension.

USDA,
2005.
Soil
Data
Mart,
Natural
Resources
Conservation
Service,
United
States
Department
of
Agriculture.
http://
soildatamart.
nrcs.
usda.
gov/.

USEPA,
1993.
Wellhead
Protection:
A
Guide
for
Communities,
EPA/
625/
R­
93/
002,
U.
S.
Environmental
Protection
Agency,
Washington,
DC.

USDA,
2005.
Soil
Data
Mart,
Natural
Resources
Conservation
Service,
United
States
Department
of
Agriculture.
http://
soildatamart.
nrcs.
usda.
gov/.
