II.
D.
3
­
Page
1
of
26
II.
D
Appendices:
Drinking
Water
Exposure
3.
Drinking
Water
Treatment
Effects
on
N­
methyl
Carbamate
Pesticides
Based
on
available
data,
this
section
provides
an
analysis
of
the
effects
of
water
treatment
methods
on
N­
methyl
carbamates,
including
aldicarb,
aldicarb
sulfone,
aldicarb
sulfoxide,
carbaryl,
carbofuran,
formetanate
HCL,
methiocarb,
methomyl,
oxamyl,
pirimicarb,
propoxur,
and
thiodicarb.
This
current
review
of
data
is
an
update
to
the
previous
literature
review,
which
was
used
to
support
OPP's
Water
Treatment
Science
Policy
(
USEPA,
2000;
USEPA
2001)

An
evaluation
of
laboratory
and
field
monitoring
data
indicate
that
N­
methyl
carbamates
may
be
effectively
removed
from
drinking
water
by
lime
softening
and
activated
carbon.
With
the
exception
of
aldicarb,
lime
softening
processes
degrade
N­
methyl
carbamates
through
alkalinecatalyzed
hydrolysis.
Abiotic
hydrolysis
studies
conducted
in
pH
9
buffer
solutions
indicate
that
carbamates
are
susceptible
to
degradation
during
lime
softening.
In
addition
to
lime
softening,
sorption
on
activated
carbon
using
granular
activated
carbon
(
GAC)
or
powdered
activated
carbon
(
PAC)
appears
to
be
effective
in
removing
N­
methyl
carbamates
from
drinking
water.
Other
treatment
methods,
such
as
chlorination,
chloramination,
chlorine
dioxide,
and
potassium
permanganate,
appear
tp
be
only
effective
in
oxidation
of
N­
methyl
carbamate
compounds
containing
a
methylthio
group
(
CH3­
S­),
e.
g.,
methiocarb
and
aldicarb.
These
compounds
are
expected
to
oxidize
to
sulfoxide
and
sulfone.

Available
USGS
water
treatment
plant
monitoring
data
indicate
that
N­
methyl
carbamates
and
their
degradation
products
have
a
low
detection
frequency
in
raw
and
finished
water
samples.
The
low
detection
frequency,
coupled
with
sample
handling
issues,
prevent
an
estimate
of
pesticide
removal
through
typical
water
treatment
plants.
However,
an
analysis
of
occurrence
for
carbamate
degradation
products
in
the
monitoring
study
and
the
laboratory
studies
suggest
they
were
formed
through
both
environmental
and
water
treatment
processes.

A.
Environmental
Fate
Data
Pertinent
to
Treatment
Effects
With
the
exception
of
aldicarb,
the
N­
methyl
carbamates
are
rapidly
degraded
through
alkaline
catalyzed
hydrolysis.
As
a
group,
they
are
generally
non­
persistent
and
mobile
in
aerobic
soil
and
form
oxidative
degradation
products.
Table
II.
D.
3.1
provides
a
summary
of
the
environmental
properties
of
the
n­
methyl
carbamate
pesticides.
Based
on
environmental
fate
data,
N­
methyl
carbamates
are
expected
to
degrade
via
base­
catalyzed
hydrolysis
during
lime
softening
as
well
as
by
oxidation
when
water
is
treated
with
oxidative
disinfectants.

Table
II.
D.
3.1:
Environmental
Fate
Data
for
Selected
N­
methylcarbamates1
Pesticide
Abiotic
Hydrolysis
(
pH=
9)
at
25oC
Aerobic
Soil
Metabolism
Soil
Partitioning
II.
D.
3
­
Page
2
of
26
Halflife
(
days)
Degradation
Products
Halflife
(
days)
Degradation
Products
Coefficient
Kf
Aldicarb
197
2.3
aldicarb
sulfoxide
aldicarb
sulfone
<
1
Aldicarb
sulfoxide
2.3
Stable
<
1
Aldicarb
sulfone
Stable
<
1
Carbaryl
0.13
1­
naphthol
4
1­
naphthol
2­
3
Carbofuran
0.63
carbofuran
phenol
150­
321
3­
ketocarbofuran
carbofuran
phenol
0.72
(
median)

Formetanate
Hydrochloride
1
N'­(
3­
hydroxy
phenol)
­
N,
Ndimethyl
formamide
hydrochloride
6.4
3­
formaminophenyl
methylcarbamate,
3­
dimethylaminomethylene
iminophenol
hydrochloride,
3­
aminophenylcarbamate,
3­
aminophenol
<
3.43
3­
formamido­
phenyl
methyl­
carbamate
Methiocarb
<
1
Mesurel
phenol
Mesurel
sulfoxide
phenol
No
Data
No
Data
Methomyl
30
S­
methyl­
Nhydroxythioacetimidat
e
30­
45
No
Data
<
1.5
Oxamyl
0.125
Oxime
11­
27
Oxime
Dimethyl
oxamic
acid
<
1
Pirimicarb
Stable
7­
294
5,6­
dimethyl­
2­
methylamino­
4'­
hydroxypyrimidine;
5,6­
dimethyl­
2­
methylamino­
4'­

hydroxypyrimidincarbamate
Propoxur
1.6
112­
180
<
1
Thiodicarb
0.5
Methomyl
1.5
methomyl
2­
14
1­
Data
were
obtained
from
most
recent
Re­
registration
Eligability
Document
assessments
or
EFED
One­
Liner
Database
B.
Review
of
Available
EPA
Studies
B.
1
Preliminary
Water
Treatment
Effects
for
Selected
N­
methyl
Carbamates
(
Miltner,
R.
J.
June,
2005.
Status
Report:
Summary
of
ORD/
WSWRD
Studies
to
Control
N­
methyl
Carbamates
in
Drinking
II.
D.
3
­
Page
3
of
26
Water.
USEPA/
ORD/
WSWRD.
Cincinnati,
OH)
etermination
of
Timing
of
Carbamate
Applications
Background
/
Purpose
EPA's
Office
of
Research
and
Development
/
Water
Supply
and
Water
Research
Division
(
ORD/
WSWRD)
conducted
bench­
scale
screening­
level
water
treatment
studies
to
assess
the
effect
of
common
water
treatment
processes
on
removal
of
selected
N­
methyl
carbamates.
Screening­
level
treatment
processes
included
coagulation/
clarification,
lime
softening,
adsorption
to
PAC,
and
oxidation
with
chlorine,
chloramines,
chlorine
dioxide,
and
potassium
permanganate.
The
data
are
considered
preliminary
because
they
have
not
been
subject
to
a
formal
peer
review
process.
A
qualitative
summary
of
the
preliminary
data
is
presented
in
a
summary
report,
which
will
be
followed
by
a
more
comprehensive
final
report.

Materials
and
Methods
Bench­
scale
water
treatment
studies
were
conducted
at
room
temperature.

Non­
radiolabeled
carbamates
were
spiked
into
these
waters
at
concentrations
ranging
from
44
to
88
µ
g/
L.
Concentrations
of
carbamates
were
determined
using
GC/
MS.
Degradation
(
transformation)
products
were
not
identified.
In
all
studies,
control
samples
were
held
for
the
duration
of
the
process
reaction
times
to
ensure
that
there
was
no
background
loss
of
the
carbamates
that
could
be
attributed
to
the
treatment.

Coagulation,
Clarification
and
PAC
Adsorption
Coagulation/
clarification
studies
were
conducted
in
jar
tests
using
raw
surface
water
from
the
Winton
Woods
Lake
with
alum
coagulant
doses
adjusted
to
control
turbidity
under
typical
full­
scale
drinking
water
treatment
conditions.
These
studies
also
were
conducted
with
and
without
PAC
to
assess
the
impact
of
activated
carbon
on
pesticide
adsorption.
Alum
(
aluminum
sulfate)
is
the
most
commonly
used
coagulant.
Winton
Woods
Lake
is
not
a
source
of
drinking
water,
but
is
typical
of
surface
waters
with
regard
to
levels
and
fluctuations
in
turbidity
and
total
organic
carbon
(
TOC).
Prior
to
spiking
carbamates
into
these
waters,
preliminary
studies
were
conducted
to
determine
proper
alum
doses
for
control
of
turbidity.
Hydrodarco
B,
a
commonly
employed
PAC,
was
used.
PAC
doses
spanned
from
10
mg/
L,
which
is
typical
of
doses
used
for
taste
and
odor
(
T&
O)
control,
to
60
mg/
L,
which
is
relatively
atypical
and
high.
30
mg/
L
is
representative
of
high­
end
doses
of
PAC.

Softening
Lime
softening
studies
were
conducted
in
jar
tests
using
Great
Miami
Aquifer
water.
Lime
doses
raised
pH
according
to
typical
full­
scale
conditions.
In
II.
D.
3
­
Page
4
of
26
addition
to
pH,
alkalinity,
turbidity,
calcium
hardness,
magnesium
hardness
and
total
hardness
were
monitored.
The
Great
Miami
Aquifer
is
the
source
of
drinking
water
for
the
Cincinnati
Water
Works'
Bolton
Water
Treatment
Plant.
Prior
to
spiking
carbamates
into
these
waters,
preliminary
studies
were
conducted
to
determine
the
conditions
that
would
control
calcium
hardness,
magnesium
hardness
and
total
hardness.
Although
lime
softening
was
used
in
these
studies,
it
was
found
that
lime/
soda
softening
would
be
required
to
control
calcium
hardness
and
turbidity
at
the
higher
pHs
required
for
control
of
magnesium
hardness.
Settling
took
place
for
one
hour.
To
prevent
loss
of
the
carbamates
at
the
high
pHs
of
softening,
samples
were
acidified
upon
collection.

Oxidation
Oxidation
studies
using
chlorine,
chloramine,
chlorine
dioxide,
and
potassium
permanganate
were
conducted
in
laboratory
waters.
For
chlorine,
chloramine
and
chlorine
dioxide,
studies
were
conducted
over
24
hours
and
at
doses
that
would
be
high
for
drinking
water
treatment.
Doses
were
Recommended
Maximum
Disinfection
Residuals
(
RMDLs)
as
defined
in
the
Disinfectant/
Disinfection
Byproduct
(
D/
DBP)
Rule.
For
permanganate,
studies
were
conducted
over
6
hours
and
the
dose
was
1
mg/
L.
Six
hours
is
representative
of
time
through
a
water
treatment
plant
in
the
presence
of
permanganate,
and
doses
beyond
1
mg/
L
may
impart
undesired
color
in
finished
drinking
water.
These
oxidation
studies
were
designed
to
assess
the
extent
of
these
reactions
at
relatively
high
doses
and
in
lab
waters
(
very
low
TOC,
very
low
ionic
strength)
offering
no
significant
demand
on
the
oxidant.
These
studies
were
controlled
by
sampling
non­
oxidant
treated
samples
held
to
assess
background
stability
during
the
reaction
time.
Where
these
reactions
are
significant,
follow
up
studies
will
examine
reaction
rates.

QA
/
QC
Studies
These
studies
were
conducted
at
ORD's
laboratories
in
Cincinnati
under
contract
68­
C­
00­
159
with
the
University
of
Cincinnati
(
UC).
Battelle
Laboratories
in
Columbus
OH,
a
subcontractor
to
UC,
performed
the
GC/
MS
analyses
after
developing
methods
for
the
combination
of
carbamates
(
aldicarb,
carbaryl,
methiocarb,
oxamyl
and
propoxur).
Significant
QA/
QC
was
a
part
of
this
effort.
A
stability
study
was
conducted
to
define
sample
holding
times
and
required
preservatives.
All
samples
were
acidified
upon
collection.
A
study
was
conducted
to
find
the
reducing
agent
(
sulfite)
that
would
not
cause
analytic
interference.
Dilution
series
in
the
various
waters
were
submitted
as
blinds.
The
QA/
QC
program
required
recovery
checks,
replication
and
analyses
of
blanks
with
each
sample
set.
Replication
included
both
bench­
scale
treatment
samples
and
laboratory
splits
of
submitted
treatment
samples.

B.
2
Softening
and
Chlorination
Screening
Studies
for
Select
Pesticides
(
Speth
and
Pisigan.
2001.
Softening
and
Chlorination
Screening
II.
D.
3
­
Page
5
of
26
Studies
for
Selected
Pesticides.
USEPA/
ORD/
WSWRD.
Cincinnati,
OH)

Background
/
Purpose
The
Office
of
Pesticide
Programs
(
OPP)
requested
that
the
ORD/
WSWRD
generate
drinking
water
treatment
data
for
a
select
group
of
pesticides.
Specifically,
OPP
requested
data
on
lime
softening
and
chlorination
processes
and
chose
ten
pesticides
to
study
for
each
process.
Methomyl
was
the
only
N­
methyl
carbamate
studied
with
softening.
Aldicarb
was
the
only
N­
methyl
carbamate
studied
with
chlorination.

These
studies
were
conducted
at
ORD's
laboratories
in
Cincinnati
under
contract
68­
C­
99­
211
with
the
IT
Corporation.
IT
Corp
subcontracted
the
analytical
measurements
to
Environmental
Health
Laboratories
(
EHL)
and
Environmental
Micro
Analysis,
Inc.
(
EMA).
EHL
analyzed
aldicarb
and
EMA
analyzed
methomyl.

Materials
and
Methods
Based
on
analytical
methods,
pesticides
were
broken
into
four
groups
for
the
softening
study,
and
three
groups
for
the
chlorination
study.
Aldicarb
and
methomyl
were
measured
by
EPA
Method
531.1.
Pesticides
were
spiked
into
the
ground
water
at
concentrations
near
100
ug/
L.
In
all
studies,
control
samples
were
held
for
the
duration
of
the
process
reaction
times
to
ensure
that
there
was
no
background
loss
of
the
carbamates
that
could
be
attributed
to
the
treatment.
Degradation
(
transformation)
products
were
not
searched
for.

These
studies
were
conducted
with
Great
Miami
Aquifer
water
collected
from
the
Cincinnati
Water
Works'
Bolton
Water
Treatment
plant.

QA/
QC
Studies
Prior
to
conducting
the
softening
and
chlorination
studies,
QA/
QC
studies
were
completed.
The
pesticides
were
spiked
into
the
water
at
concentrations
near
100
µ
g/
L
in
several
analytic
groupings
and
analyzed
to
confirm
that
the
groundwater
matrix
did
not
interfere
with
the
analyses,
the
spiking
procedure
was
adequate,
the
grouping
of
pesticides
did
not
cause
analytical
interferences,
and
the
analytical
precision
was
acceptable.
Preservation,
shipping
and
stability
were
also
defined.
Thiosulfate
was
determined
to
be
the
dechlorinating
agent
that
would
not
cause
analytic
interference.
All
samples
for
carbamates
were
acidified
upon
collection.

Lime
softening
The
softening
screening­
level
studies
consisted
of
jar
tests
with
Great
Miami
Aquifer
water.
The
lime
pellets
used
in
the
study
were
also
obtained
from
the
II.
D.
3
­
Page
6
of
26
Bolton
plant.
In
preliminary
studies,
varying
amounts
of
lime
were
added
to
a
series
of
jars
to
determine
the
lime
dose
that
matched
the
effluent
pH
of
the
Bolton
plant's
lime
contactor
basin
(
calcium
softening)
and
to
achieve
a
pH
of
11
(
magnesium
softening).
Soda
ash
was
not
considered
for
this
limited
study
because
elevated
calcium
levels
were
not
important
in
the
final
magnesiumsoftened
water.
Floc
would
be
present
for
any
possible
pesticide/
floc
complexation,
and
the
pH
would
be
elevated
for
base
catalyzed
reactions.

From
preliminary
studies,
the
calcium­
softening
dose
was
determined
to
be
150
mg/
L.
This
was
the
same
dose
as
that
used
in
the
Bolton
Waterworks
and
resulted
in
the
same
final
pH
as
that
seen
in
the
Bolton
plant.
The
magnesium­
softening
dose
was
determined
to
be
300
mg/
L.

All
jars
were
allowed
to
settle
for
two
hours
before
the
softened
samples
were
collected.
Softening
samples
were
acidfied
upon
collection
to
prevent
base
catalyzed
loss
of
the
carbamates.
Two
hours
of
settling
was
chosen
because
it
resulted
in
adequate
settling,
and
it
gave
a
reasonable
amount
of
time
at
a
high
pH
to
observe
base­
catalyzed
degradation.

Chlorination
In
the
chlorination
studies,
chlorine
was
dosed
under
added
at
uniform
formation
conditions
(
UFC)
(
Summers
et
al.,
1996)
to
pesticide­
spiked
Bolton
water.
UFC
represents
mean
national
distribution
system
conditions
for
chlorination
and
were
developed
for
studies
examining
the
formation
and
control
of
disinfection
byproducts
in
drinking
water.
They
are:
a
chlorine
residual
of
1
mg/
L
at
20
degrees
C,
24
hours
and
pH
8.
A
chlorine
demand
study
was
conducted
to
determine
the
appropriate
chlorine
dose
to
give
this
residual
under
these
conditions.

Data
Analyses
For
both
the
softening
and
chlorination
studies,
a
one­
way
analysis
of
variance
(
anova)
model
was
used
to
determine
whether
the
pesticide
was
lost
through
treatment.
Treatments
were
compared
at
the
5%
significance
level
using
a
t­
test.
Control
samples
were
also
evaluated
to
determine
whether
the
pesticide
naturally
degraded,
or
was
lost,
over
the
process
reaction
time.

B.
3
EPA/
ORD
Monitoring
Data
(
Miltner
et
al.,
1989,
Treatment
of
Seasonal
Pesticides
in
Surface
Waters,
Journal
AWWA)

In
full­
scale
water
plant
treatment
studies,
Miltner
et
al.
(
1989)
monitored
a
number
of
pesticides
through
three
full­
scale
water
treatment
plants
following
pesticide
application
and
Spring­
time
runoff.
Conventional
treatment
(
coagulation,
clarification
and
filtration,
lime
or
lime/
soda
softening,
adsorption
II.
D.
3
­
Page
7
of
26
onto
PAC,
adsorption
onto
GAC,
and
oxidation
by
chlorine
were
monitored
at
the
full­
scale.
Samples
were
collected
before
and
after
treatment
processes
with
the
downstream
sample
always
collected
after
the
upstream
sample
based
on
the
utility's
expertise
with
time
of
travel
through
the
plants,
and
their
experience
with
responding
to
Spring­
time
runoff
events.
PAC
doses
ranged
up
to
the
maximum
fed
for
T&
O
control
in
order
to
assess
adsorption
capacities.
The
N­
methyl
carbamate
carbofuran
was
one
of
the
monitored
pesticides.

Miltner
et
al.
supplemented
the
full­
scale
monitoring
with
bench­
scale
studies
wherein
pesticides,
including
carbofuran,
were
spiked
into
field
waters
in
EPA's
laboratories
in
Cincinnati.
Field
waters
included
those
from
the
monitored
fullscale
plants
and
Ohio
River
water.
Coagulation/
clarification,
adsorption
onto
PAC,
and
oxidation
by
chlorine
and
chlorine
dioxide
were
studied
at
the
benchscale
Coagulant
doses,
chlorine
doses
and
chlorine
dioxide
doses
were
representative
of
full­
scale
treatment.
Coagulation
was
by
alum.
PAC
doses
ranged
from
those
typical
of
T&
O
control
to
those
relatively
atypical
and
high
in
order
to
assess
adsorption
capacities.
Hydrodarco
B
and
WPH
PACs
were
studied;
both
are
commonly
used
in
drinking
water
treatment.
Bench­
scale
studies
were
conducted
at
room
temperatures.

Additionally,
adsorption
isotherms
were
conducted
at
the
bench
scale;
the
isotherms
for
carbofuran
are
also
reported
in
Speth
and
Miltner
(
1990).

B.
4
EPA/
ORD
Adsorption
Isotherm
Studies
(
Speth
and
Miltner,
1998,
Technical
Note:
Adsorption
Capacity
of
GAC
for
Synthetic
Organics,
Journal
AWWA;
Speth
and
Miltner,
1990,
Technical
Note:
GAC
Adsorption
Capacity
for
SOCs,
Journal
AWWA)

Adsorption
isotherms
measure
the
maximum
capacity
of
activated
carbon
to
remove
contaminants
from
water.
In
these
studies,
Filtrasorb
400,
a
GAC
commonly
used
in
drinking
water
treatment,
was
employed.
The
studies
were
conducted
at
room
temperature.
Two
waters
were
studied:
laboratory
water
to
assess
maximum
capacities
without
competition
from
organics
in
field
waters,
and
Ohio
River
water
to
assess
lowered
capacities
in
the
presence
of
competing
organics.
These
studies
included
the
N­
methyl
carbamates
aldicarb,
carbofuran,
methomyl
and
oxamyl.
Control
samples
were
reacted
without
carbon
for
the
duration
of
the
process
reaction
times
to
ensure
that
there
was
no
background
loss
of
the
carbamates
that
could
be
attributed
to
the
adsorption.
All
carbamate
samples
were
acidified
upon
collection.

B.
5
Results
and
Discussion
A
qualitative
summary
of
water
treatment
data
is
shown
in
Table
II.
D.
3.2.
A
"
No"
designation
indicates
no
control
was
observed
for
the
particular
treatment
process.
II.
D.
3
­
Page
8
of
26
A
"
YES"
designation
indicates
that
pesticide
control
was
observed.
For
the
process­
specific
discussions
that
follow,
the
appropriate
references
are
as
given
in
Table
II.
D.
3.2.

Table
II.
D.
3.2:
Qualitative
Assessment
of
Water
Treatment
Data
for
N­
methyl
Carbamates
Activated
Carbon
Oxidation
Pesticide
Coagulation
Softening
GAC
PAC
Chlorine
Chloramine
KMnO4
ClO2
Aldicarb
No
(
a)
No
(
a)
Yes
(
b)
Yes
(
a)
Yes
(
a,
c)
Yes
(
a)
Yes
(
a)
Yes
(
a)

Carbaryl
No
(
a)
Yes
(
a)
X
Yes
(
a)
No
(
a)
X
No
(
a)
No
(
a)

Carbofuran
No
(
d)
Yes
(
d)
Yes
(
b,
d)
Yes
(
d)
No
(
d)
X
X
No
(
d)

Methiocarb
No
(
a)
Yes
(
a)
X
Yes
(
a)
Yes
(
a)
Yes
(
a)
Yes
(
a)
Yes
(
a)

Methomyl
X
Yes
(
c)
Yes
(
e)
X
X
X
X
X
Oxamyl
No
(
a)
Yes
(
a)
Yes
(
b)
Yes
(
a)
No
(
a)
X
No
(
a)
No
(
a)

Propoxur
No
(
a)
Yes
(
a)
X
Yes
(
a)
No
(
a)
No
(
a)
No
(
a)
No
(
a)

X
­
Indicates
no
data
were
available
to
assess
treatment
process.
(
a)
Miltner,
2005;
(
b)
Speth
and
Miltner,
1990;
(
c)
Speth
and
Pisigan,
2001;
(
d)
Miltner
et
al.,
1989;
(
e)
Speth
and
Miltner,
1998
Coagulation
and
Clarification
Coagulation
and
clarification
was
not
found
to
be
effective
for
the
control
of
Nmethyl
carbamates.
They
are
water
soluble
and
not­
well
sorbed
to
the
particulates
that
are
targeted
for
removal
by
coagulation
and
clarification.
Given
analytic
precision,
coagulated
and
clarified
N­
methyl
carbamate
concentrations
could
not
be
differentiated
from
raw
water
N­
methyl
carbamate
concentrations.
Coagulation
and
clarification
provided
expected
results
with
a
drop
in
the
pH
in
clarified
waters
and
control
of
turbidity
to
levels
below
2
ntu
in
clarified
waters.

Adsorption
onto
PAC
Table
II.
D.
3.3
summarizes
the
control
provided
by
PAC
giving
the
approximate
percent
removal
at
10
mg/
L,
a
common
PAC
dose
for
T&
O
control,
and
at
30
mg/
L,
an
upper­
end
dose
for
T&
O
control.
These
results
indicate
good
removal
at
typical
T&
O
control
doses
and
better
removal
when
doses
approach
the
upper
end
of
what
treatment
plants
may
be
able
to
feed.
Carbofuran,
carbaryl,
methiocarb
and
propoxur
have
higher
molecular
weights
and
rings
structures
that
may
account
for
their
better
control
than
oxamyl
and
aldicarb
with
lower
molecular
weights
and
branched
structures.

Table
II.
D.
3.3.
Adsorption
of
N­
methyl
Carbamates
onto
PAC
Percent
Removal
Pesticide
10
mg/
L
PAC
30
mg/
L
PAC
II.
D.
3
­
Page
9
of
26
Carbaryl
62
75
Methiocarb
61
74
Carbofuran
57
83
Propoxur
44
57
Oxamyl
25
39
Aldicarb
20
38
Adsorption
onto
GAC
Aldicarb,
oxamyl,
carbofuran
and
methomyl
were
found
to
be
strongly
adsorbed
onto
GAC.
The
data
were
regressed
using
the
Freundlich
adsorption
model.
The
Feundlich
K
value
gives
the
adsorption
capacity
when
the
equilibrium
concentration
(
concentration
in
water)
is
1
ug/
L.
The
resulting
order
of
adsorbability
was
carbofuran
(
16830
ug
adsorbed
per
gram
of
carbon)
>
aldicarb
(
8270)
>
methomyl
(
4780)
>
oxamyl
(
1740).
Although
adsorbability
depends
on
several
factors,
carbofuran's
higher
molecular
weight
and
included
ring
structure
may
account
for
its
higher
adsorption
capacity.
Aldicarb
and
carbofuran
were
also
studied
in
coagulated
and
clarified
Ohio
River
water
where
adsorption
capacities
would
be
expected
to
be
lower
as
other
organics
would
compete
with
the
carbamates
for
adsorption
sites.
The
resulting
order
of
adsorbability
was
carbofuran
(
13065)
>
aldicarb
(
4160).

A
K
value
of
300
ug/
gram
is
generally
considered
a
cost
effective
measure
of
adsorbability
indicating
several
years
of
GAC
bed
life
at
typical
full­
scale
operating
conditions.
These
K
values
suggest
control
of
carbamates
to
non­
detect
levels
for
several
years
before
GAC
replacement
when
GAC
contactors
would
be
utilized
downstream
of
coagulated
and
clarified
Ohio
River
water.
GAC
is
a
designated
Best
Available
Technology
for
many
pesticides
regulated
in
drinking
water,
including
the
N­
methyl
carbamates
carbofuran
and
oxamyl.

Softening
With
the
exception
of
aldicarb,
all
the
N­
methyl
carbamates
degraded
during
lime
softening
(
Table
II.
D.
3.4).
One
exception
was
aldicarb,
which
only
degraded
22
percent
at
pH
11.25.
Because
aldicarb
does
not
undergo
rapid
alkaline
catalyzed
hydrolysis
(
t1/
2=
197
days
at
pH
9,
Table1),
this
observation
was
expected.

For
studies
with
Great
Miami
Aquifer
water,
the
calcium
concentration
decreased
to
the
greatest
extent
under
calcium­
softening
conditions.
The
percent
removal
of
magnesium
was
less
than
10
percent
under
calcium­
softening
conditions.
Under
magnesium­
softening
conditions,
the
magnesium
was
reduced
to
greater
than
90
percent.
The
increased
calcium
in
the
magnesium­
softening
samples
as
compared
to
the
calcium­
softening
samples
was
not
surprising
because
of
the
water's
initial
alkalinity,
indicating
that
this
water
requires
lime
soda
ash
softening
to
control
calcium
at
magnesium­
softening
conditions.
This
was
confirmed
in
other
ORD/
WSWRD
studies.
II.
D.
3
­
Page
10
of
26
During
calcium
softening,
removal
of
N­
methyl
carbamates
ranged
from
87
to
99
percent,
with
the
exception
of
aldicarb.
During
magnesium
softening,
their
removal
ranged
99
to
100
percent,
with
the
exception
of
aldicarb.
In
studies
by
Speth
and
Pisigan
(
2001)
thiodicarb
demonstrated
90
percent
removal
during
magnesium
softening
1,
although
no
removal
during
calcium
softening.
The
similarity
in
structure
between
methomyl
and
thiodicarb
may
have
resulted
in
their
similar
behavior
under
higher­
pH
magnesium­
softening
conditions.
Thiodicarb
is
not
an
N­
methyl
carbamate,
but
is
a
dimer
of
methomyl
and
degrades
to
methomyl
(
Table
II.
D.
3.1).
Its
larger
structure
may
have
prevented
it
from
being
degraded
under
calcium­
softening
conditions.

Although
the
studies
for
most
of
these
N­
methyl
carbamates
were
done
at
the
bench­
scale
in
jar
tests,
carbofuran
was
monitored
in
full­
scale
treatment
plants
practicing
magnesium
softening
(
Miltner
et
al.,
1989).
Results
were
similar
to
those
of
the
other
N­
methyl
carbamates.

Although
the
mechanism
of
pesticide
removal
was
not
incorporated
into
these
studies,
base­
catalyzed
degradation
likely
explains
the
results.
Possible
transformation
products
are
given
in
Table
II.
D.
3.1.

Table
II.
D.
3.4.
Control
of
N­
methyl
Carbamates
by
Lime
Softening
Calcium
Softening
Magnesium
Softening
Carbamate
Ref.
pH
Lime
Dose,
mg/
L
%
R
pH
Lime
Dose,
mg/
L
%
R
aldicarb
carbaryl
carbofuran
methiocarb
methomyl
oxamyl
propoxur
(
a)
(
a)
(
d)
(
a)
(
c)
(
a)
(
a)
10.26
10.26
10.26
10.15
10.26
10.26
130
130
130
150
130
130
6
99
99
87
99
99
11.25
11.25
10.9,
11.1
11.25
11.2
11.25
11.25
220
220
220
300
220
220
22
99
100
99
99
99
99
%
R
=
percent
removal
(
a)
Miltner,
2005,
Great
Miami
Aquifer
(
c)
Speth
and
Pisigan,
2001,
Great
Miami
Aquifer
(
d)
Miltner
et
al.,
1989,
Maumee
River,
Sandusky
River
Oxidation
Oxidation
studies
with
chlorine,
chloramine,
chlorine
dioxide,
and
potassium
permanganate
showed
that
aldicarb
and
methiocarb
degraded
through
chemical
oxidation
with
each
oxidant.
Generally,
control
of
aldicarb
and
methiocarb
by
chlorine,
chloramine
and
chlorine
dioxide
exceeded
95
percent
removal,
where
as
control
by
permanganate
was
only
31
and
23
percent
respectively.
This
would
be
expected
as
permanganate
is
a
weaker
oxidant
than
the
others
and
generally
used
for
T&
O
control
and
manganese
control
rather
than
for
disinfection.
In
contrast,
carbofuran,
carbaryl,
oxamyl,
and
propoxur
were
not
oxidized
when
treated
with
II.
D.
3
­
Page
11
of
26
the
various
oxidants.
Given
analytic
precision,
oxidized
N­
methyl
carbamate
concentrations
could
not
be
differentiated
from
control
N­
methyl
carbamate
concentrations.
One
possible
explanation
is
that
methiocarb
and
aldicarb
contain
a
methylthio
group
(
CH3­
S­)
which
is
prone
to
oxidize
during
oxidative
disinfection
processes.
Although
not
identified
in
these
studies,
sulfoxide
and
sulfone
are
the
expected
transformation
products
of
aldicarb
and
methiocarb.

Aldicarb's
reaction
with
chlorine
was
examined
under
different
conditions
in
two
studies.
Both
studies
utilized
pH
8,
room
temperature
and
24­
hour
reaction
times.
Results
were
similar
in
laboratory
water
with
a
higher
4
mg/
L
chlorine
dose
(
Miltner,
2005)
and
in
Great
Miami
Aquifer
water
under
UFC
conditions
with
a
lower
chlorine
dose
of
1.6
mg/
L
(
Speth
and
Pisigan,
2001).

In
studies
by
Speth
and
Pisigan
(
2001)
under
UFC
conditions,
molinate,
a
related
carbamate,
also
showed
99
percent
removal
when
oxidized
by
chlorine.
Because
molinate
has
an
ethylthio
group,
the
oxidation
process
is
expected
to
yield
sulfoxide
and
sulfone
transformation
products.

B.
6
Other
Treatment
Processes
Although
these
studies
did
not
include
membranes,
these
are
expected
to
be
effective
in
removing
N­
methyl
carbamates.
Reverse
osmosis
membranes,
and
likely
ultrafiltration
membranes,
would
physically
remove
these
carbamates
from
drinking
water.

C,
Open
Literature:
Oxidation
Studies
C.
1.
Chlorine
Miles
et
al.
(
1988)
studied
the
fate
of
carbaryl
and
propoxur,
at
1.0
mg/
L,
in
phosphate­
buffered
water
treated
with
10
mg/
L
of
hypochlorite
at
pH
7
in
the
dark.
The
chlorine
degradation
of
carbaryl
was
slightly
faster
than
that
for
propoxur.
The
first­
order
kinetic
half­
lives
of
carbaryl
were
3.5
days
at
pH
7.0
and
0.05
days
at
pH
8.0.
For
propoxur,
the
half­
lives
were
9.2
days
at
pH
7.0
and
0.29
days
at
pH
8.0.
Structural
elucidation
of
the
chlorination
products
was
not
performed.

In
another
separate
laboratory
experiment,
Miles
and
Oshiro
(
1990)
monitored
the
degradation
of
0.1
uM
of
methomyl
by
1
uM
chlorine.
Methomyl,
which
contains
a
methylthio
group,
was
transformed
to
methomyl
sulfoxide
and
Nchloromethomyl
which
were
eventually
converted
to
acetic
acid,
methanesulfonic
acid,
and
dichloromethyl
amine.
II.
D.
3
­
Page
12
of
26
Miles
(
1991)
found
that
0.01mM
of
aldicarb
in
0.1
mM
phosphate
buffered
solution
reacted
with
0.1
­
10
mM
hypochlorite
at
pH
6
­
9.
Aldicarb
initially
underwent
oxidation
to
form
aldicarb
sulfoxide
and
upon
further
chlorine
oxidation,
aldicarb
sulfoxide
was
converted
to
aldicarb
sulfone.
An
additional
chlorine
addition
reaction
led
to
the
formation
of
N­
chloroaldicarb
sulfone,
which
can
decompose
to
an
acid
and
dichlorodimethylamine.
Mason
et
al
(
1990)
also
found
that
chlorination
of
aldicarb
led
to
formation
of
ten
degradation
products.
Four
of
the
degradation
products
were
identified
as
aldicarb
sulfoxide,
aldicarb
sulfoxide
oxime,
aldicarb
sulfone,
and
aldicarb
oxime.
Both
aldicarb
sulfoxide
and
aldicarb
sulfone
were
also
found
by
Miles
(
1991).
A
summary
of
the
chlorination
degradation
products
of
methomyl
and
aldicarb
are
shown
in
Table
II.
D.
3.5.

Table
II.
D.
3.5.
Chlorination
Products
of
Methomyl
and
Aldicarb
Carbamate
Chlorine
Transformation
Products
Reference
Methomyl
methomyl
sulfoxide,
N­
chloromethomyl,
acetic
acid,
methanesulfonic
acid,
dichloromethylamine
Miles
and
Oshiro,
1990
Aldicarb
aldicarb
sulfoxide,
aldicarb
sulfone,
N­
chloroaldicarb
sulfone,
aicd
form,
dichloromethylamine
aldicarb
sulfoxide,
aldicarb
sulfoxide
oxime,
aldicarb
sulfone,
aldicarb
oxime
Miles,
1991
Mason
et
al,
1990
C.
2.
Ozone
and
Advanced
Oxidation
Processes
Advanced
oxidation
processes
(
AOPs)
are
those
which
utilize
hydroxyl
radicals
for
chemical
oxidation.
Ozone
has
a
higher
oxidation
potential
(
2.07
V)
than
does
an
oxidant
like
chlorine.
The
hydroxyl
radical
produced
by
AOPs
has
an
oxidation
potential
of
2.8
V.
Therefore,
contaminants
in
drinking
water
like
carbamates
can
be
oxidized
by
the
ozone
molecule
or
by
the
hydroxyl
radical.
The
hydroxyl
radical
can
be
created
by
AOPs
combining
processes
(
ozone/
UV,
ozone/
H2O2,
ozone/
H2O2/
UV).
Ozonation
at
higher
pH
(
greater
than
pH
8)
also
creates
hydroxyl
radicals.
With
the
inclusion
of
UV,
photolysis
can
also
contribute
to
the
reaction
mechanism.
Because
these
are
very
reactive
processes,
the
concentrations
of
pesticides
like
carbamates
can
be
significantly
lowered
in
drinking
water,
but
a
significant
number
of
transformation
products
can
be
created.
Ozonation
and
AOPs,
however,
are
used
by
less
than
2
percent
of
drinking
water
treatment
plants.

Ikehata
and
Gamal
El­
Din
(
2005)
reviewed
the
literature
for
ozonation
and
advanced
oxidation
processes
for
their
control
of
pesticides,
including
the
Nmethyl
carbamates
aldicarb,
carbaryl,
carbofuran,
fenobucarb,
isoprocarb,
methomyl,
metolcarb,
oxamyl
and
propoxur.
Where
conditions
were
similar
to
those
of
drinking
water
treatment,
the
following
may
be
summarized
as
follows:
II.
D.
3
­
Page
13
of
26

Ozonation
at
1
mg/
L
of
applied
ozone
at
pH
8.3
resulted
in
complete
conversion
of
aldicarb
at
9.5
ug/
L
,
with
aldicarb
sulfoxide
identified
as
a
byproduct.
In
the
ozonation
of
river
water
at
pH
7.2,
the
order
of
reactivity
of
N­
methyl
carbamates
was:
carbaryl
>
propoxur
>
fenobucarb
>
isoprocarb
>
metolcarb.
It
was
proposed
that
the
isopropoxy
group
on
propoxur
was
more
susceptible
to
ozone
attack
than
the
sec­
butyl
or
isopropyl
groups
on
fenobucarb
and
isoprocarb,
respectively.


In
ozone/
UV
reactions
with
propoxur,
the
hydroxyl
radical
accounted
for
over
half
the
reaction,
whereas
the
ozone
molecule
accounted
for
less
than
one­
fifth,
and
photolysis
for
less
than
one­
third.


Ozonation
byproducts
of
carbaryl
were
found
to
be:
1­
naphthol,
naphthoquinone,
phthalic
anhydride
and
N­
formylcarbamate
of
1­
naphthol.


Ozonation
of
carbofuran
was
improved
when
combined
with
UV
(
ozone/
UV).


Fast
reaction
kinetics
were
reported
for
the
ozonation
of
oxamyl
and
methomyl.


Mineralization
of
n­
methyl
carbamates
by
ozone
or
AOPs
under
drinking
water
treatment
conditions
has
not
been
demonstrated.

D.
Field
Monitoring
Data
D.
1
Pesticide
Occurrence
Data
from
the
USGS/
EPA
Pilot
Reservoir
Monitoring
Program
(
Bloomquist
et
al.,
2001)

EPA's
Office
of
Pesticide
Programs
(
OPP)
analyzed
USGS/
EPA
pilot
reservoir
monitoring
data
to
assess
concentrations
of
carbamates
and
their
degradation
products
in
raw
and
finished
waters
and
to
examine
trends,
when
possible,
in
whole
plant
water
treatment
effects
on
pesticide
removal
and
transformation
(
Blomquist
et
al,
2001).
Reservoir
("
outfall")
samples,
although
collected,
were
not
considered.
Analysis
and
summary
statistics
for
carbamates
are
discussed
below:

Factors
Affecting
Interpretation
of
Reservoir
Monitoring
Data

The
USGS/
EPA
pilot
reservoir
monitoring
study
was
designed
to
provide
pesticide
occurrence
data
in
raw
and
finished
water
(
Blomquist
et
al.,
2001).
Data
analysis
objectives
included
the
following:
provide
pesticide
data
for
human
and
ecological
exposure
assessment;
assessment
of
II.
D.
3
­
Page
14
of
26
sampling
frequency
for
capturing
peak
pesticide
concentrations
in
community
water
system
(
CWS);
assessment
of
general
relationships
of
pesticide
concentrations
in
finished
drinking
water
in
relation
to
source
water
and
overall
understanding
of
whole
system
water
treatment
effects;
establishment
of
relationships
of
pesticide
concentrations
in
reservoirs
to
watershed
and
reservoirs
characteristics;
and
validation
and
testing
of
watershed
and
reservoir
models.


Raw
and
finished
water
were
not
temporally
paired
to
account
for
travel
time
of
the
pesticide
and
its
transformation
products
through
the
water
treatment
plant.
(
The
lack
of
temporal
pairing
limits
a
direct
linkage
of
pesticide
removal
by
treatment
and
degradation
and
formation
of
degradation
products
during
water
treatment.)
Although
the
water
samples
were
not
temporally­
paired,
the
results
were
expected
to
show
a
general
relationship
in
concentration
differences
between
raw
and
finished
water
samples.
Temporal
variability
in
pesticide
concentrations
was
expected
to
be
lower
when
compared
to
flowing
water
bodies
because
the
source
water
was
derived
from
reservoirs.
Additionally,
water
samples
were
taken
on
the
same
time
scale
(
hours)
as
the
water
treatment
cycles
for
the
water
utilities.

An
analysis
of
sample
pairing
time,
expressed
as
a
percentage
of
the
reservoir
flow­
through
time,
indicated
that
timing
and
sequencing
of
pairing
were
not
consistent
among
the
reservoirs
(
Figure
II.
D.
2.1).
The
quality
assurance
plan
(
QAP)
for
the
pilot
reservoir
monitoring
program
did
not
specify
a
pairing
sampling
strategy
of
raw
and
finished
water
samples.
The
water
sampling
time
and
sequence
were
at
the
discretion
of
the
person(
s)
sampling
the
water
treatment
plants.

In
some
cases,
finished
water
samples
were
taken
before
the
"
paired"
raw
water
samples.
This
situation
indicates
that
the
slug
of
sampled
finished
water
was
ahead
of
the
"
paired"
raw
water
sample,
which
leads
to
negative
percentage
in
sample
pair
times
relative
to
average
water
plant
flow
time.
Negative
sample
pair
times
were
found
in
all
the
treatment
plants
except
TX.

Raw
water
samples
were
also
taken
prior
to
the
finished
water
samples.
Under
these
conditions,
the
percentage
in
sample
pairing
time
relative
to
average
water
plant
flow
time
would
be
positive.
The
mean
percentage
of
sampling
pairing
time
typically
accounted
for
less
than
20%
of
the
water
flow
through
time
in
the
plant.
An
exception
to
this
observation
is
was
the
NY
reservoir
where
the
mean
percentage
of
sample
pairing
time
accounted
for
57%
of
water
treatment
flow
through
time.
Exact
raw
and
finished
water
sample
pairing
(
100%)
was
found
for
three
water
samples
at
the
LA
water
treatment
plant.
II.
D.
3
­
Page
15
of
26
Figure
II.
D.
2.
1
­
Range
of
Sample
Pairing
Time
(
expressed
as
a
percent
of
average
water
treatment
plant
flow­
through
time)
Among
the
Water
Treatment
Plants.
Box
represents
90th
and
10th
percentiles;
whiskers
represent
95th
and
5th
percentiles;
bar
in
box
is
median;
points
are
outliers.

Water
Treatment
Plant
CA
IN
LA
MO
NC
NY
OH
PA
SC
SD
TX
Sampling
Pairing
Time
(%
of
Average
Flow­
through
time)

­
150
­
100
­
50
0
50
100
150
There
were
high
deviations
in
the
sample
pairing
time
among
the
water
treatment
plants.
The
most
variable
sampling
pairing
times
were
associated
with
the
LA
water
treatment
plant.
In
most
cases,
the
standard
deviation
was
equal
to
or
higher
than
the
mean
percentage
of
sample
pairing
time.


Pesticide
concentrations
in
finished
water
samples
were
determined
in
"
unquenched"
water
samples.
The
lack
of
quenching
of
free
chlorine
in
finished
water
samples
does
not
eliminate
the
possibility
of
continued
chemical
oxidation
during
storage
and
analysis.
Water
treatment
plant
processes
in
the
reservoir
monitoring
study
employ
pre­
and
post­
disinfection
treatment
processes,
using
chlorine
as
the
disinfectant.
Hence,
the
absence
of
quenching
may
limit
the
detection
and
definitive
quantification
of
pesticides
and
transformation
products
prone
to
oxidation
during
storage.
Recoveries
in
matrix­
spiked
finished
water
samples
and
sample
storage
times
are
presented
to
assess
the
impact
of
non­
quenching
pesticide
stability
(
Table
II.
D.
3.8).
II.
D.
3
­
Page
16
of
26

Ancillary
data
on
weather,
pesticide
use,
and
watershed
vulnerability
need
to
be
considered
when
interpreting
occurrence
data.
Sampling
was
extended
through
2000
because
of
extreme
drought
conditions
in
the
northeastern
United
States
and
California
during
the
1999
sampling
season.
Lower
than
average
rainfall
may
have
impacted
pesticide
runoff
and
resulted
in
fewer
detections
of
pesticides.


Three
analytical
methods
(
2001,
9002,
and
9060)
were
used
in
the
reservoir
monitoring
study.
Method
2001
was
conducted
using
C­
18
solid
phase
extraction
and
gas
chromatography/
mass
spectrometry
(
GC/
MS)
(
Zaugg
et
al.,
1995).
This
method
has
been
validated
and
is
approved
for
use
in
the
National
Water
Quality
Assessment
(
NAWQA)
program.
Methods
9002
and
9060
were
provisional
(
under
development
and
validation)
during
the
course
of
the
study,
but
are
now
currently
approved
by
USGS.
Method
9002
(
now
referred
to
as
method
2002)
is
was
conducted
using
C­
18
solid
phase
extraction
and
GC/
MS
(
Sandstrom
et
al.,
2001).
Method
9060
(
now
referred
2060)
is
was
conducted
using
solid
phase
extraction
and
high
performance
liquid
chromatography/
mass
spectrometry
(
HPLC/
MS)
(
Furlong
et
al.,
2001).
These
methods
were
used
to
expand
information
on
occurrence
of
pesticides
and
degradation
products.


For
Method
9060,
the
data
from
March
1,
1999
to
December
31,1999
have
been
flagged
with
supplemental
USGS
data
quality
(
Written
Communications
from
Joel
Blomquist,
USGS
9/
23/
02).
The
flagging
was
performed
because,
in
1999,
the
analytical
demand
for
Method
9060
exceeded
the
instrumental
analysis
capacity.
This
delay
led
to
longstorage
times
(
180
days)
in
refrigerators
of
unextracted
samples.
Interpretation
of
the
1999
data
requires
careful
consideration
of
median
half­
life
of
the
pesticides
during
storage
of
unquenched
samples
with
respect
to
sample
holding
times.


The
qualified
"
estimate"
designation,
according
to
USGS
protocol,
has
been
extended
for
data
from
USGS
Method
9060
to
account
for
background
concentrations
of
pesticides
in
blank
water
samples
(
Verbal
Communication
J.
Blomquist,
10/
7/
02).
Because
background
concentrations
in
blank
water
samples
approached
minimum
reporting
limits
(<
0.003
ug/
L)
for
several
compounds,
the
qualified
estimate
designation
(
E)
is
limited
to
concentrations
greater
than
or
equal
to
0.003
ug/
L.
Concentrations
below
0.003
ug/
L
are
not
considered
as
estimated
concentrations.
This
designation
was
adopted
to
ensure
reliable
estimates
of
detections
above
background
concentrations.

Methods
of
Data
Analysis
II.
D.
3
­
Page
17
of
26
Data
from
USGS/
EPA
Reservoir
Monitoring
Data
(
Blomquist
et
al.,
2001)
were
reformatted
in
an
a
spreadsheet
to
accommodate
formatting
requirements
for
Statistical
Analysis
Systems
(
SAS
is
a
Trademark
of
SAS
Institude.
Inc.,
Cary
NC).
Sampling
dates
in
the
original
data
set
were
modified
to
facilitate
translation
of
date
variables.
After
the
modification
step,
EXCEL
data
sets
for
USGS
schedules
2001,
9060,
and
9002
were
merged
into
a
common
data
set
using
a
SAS
program.

Summary
Statistics
Summary
statistics,
including
the
mean,
median,
standard
deviation,
minimum,
and
maximum
pesticide
concentrations
in
raw
and
finished
water
samples,
were
determined
using
Statistical
Analysis
Systems
(
SAS)
procedures.

Pesticide
Concentration
Difference
for
Raw
and
Finished
Water
Samples
Raw
water
removal
percentages
were
not
calculated
because
of
their
known
instability
in
finished
water,
low
detection
frequencies,
and
low
concentrations
in
raw
and
finished
water
samples.

Assessment
Criteria
for
Assessing
Pesticide
Transformation
A
set
of
criteria
were
used
to
identify
possible
pesticide
transformation
during
water
treatment.


Pesticide
degradation
products
detected
in
only
finished
water
samples.


Parent
pesticides
in
only
raw
water
samples.


Co­
occurrence
of
parent
pesticides
in
only
raw
water
samples
and
detection
of
degradation
product
in
only
finished
water
samples
for
the
same
paired
water
sample.


Low
recoveries
of
matrix
spikes
from
finished
water
in
combination
with
high
recoveries
in
matrix
spikes
from
raw
water.

The
premise
is
that
at
least
one
of
these
conditions
may
indicate
possible
pesticide
transformation
during
water
treatment.

Water
Treatment
Trains
and
Basic
Water
Quality
Data
Although
the
water
quality
parameters,
including
pH,
hardness,
and
total
organic
carbon,
varied
among
the
12
reservoirs
(
Table
II.
D.
3.6),
the
physical
construct
of
the
treatment
train
processes
was
similar.
II.
D.
3
­
Page
18
of
26
Source
Water

Prechlorination
(
Preoxidation)

Coagulation
and
Clarification

Filtration

Post
Oxidation

Clear
well
Table
II.
D.
3.6:
Average
Water
Quality
Parameters
for
Raw
Water
at
Candidate
Reservoirs
Water
Quality
Properties
Water
Systems
Avg
Flow
Through
Time
(
hr)
pH
Alkalinity
mg/
L
as
CaCO3)
Hardness
(
mg/
L
as
CaCO3)
TOC*
(
mg/
L)
MO
26
7.9
­
9.2
63­
120
90
­
145
4.7
TX
10
7.7
100
108
4­
8
OH
23
7.7
95
126
5.2
OK
NA
7.9­
8.8
137
150
5.8
CA
3.25
7.5
91
250
6­
8
IN
8.75
8.2
128
200
4
SD
12­
13
9.2
32
NA
NA
SC
4
6.9
17
15
3.8
NC
NA
7
12
NA
NA
LA
NA
NA
NA
NA
NA
NY
0.29
7.8­
9.0
40­
100
140
4.4
PA
7­
9
7.2
7.2
172
2­
3
NA­
Not
available
*
TOC=
Total
Organic
Carbon
The
average
water
flow­
through
time
at
each
treatment
plant
was
less
than
24
hours.
The
most
common
treatment
practices
included
prechlorination
and
post
disinfection,
coagulation,
and
pH
adjustment
processes.
Chlorine
and
chlorine
dioxide
were
the
most
common
disinfectants
used
in
the
prechlorination
process,
while
chlorine
and
chloramines
were
the
most
common
disinfectants
used
in
the
post
disinfection
process.
The
most
common
coagulants
used
in
the
treatment
trains
were
aluminum
salts
and
polymers.
The
data
also
shows
that
pH
was
adjusted
by
adding
lime
and
sodium
hydroxide.
Several
of
the
treatment
plants
used
activated
carbon
in
the
treatment
train.
Powdered
activated
carbon
was
used
in
the
PA,
NY,
SC
and
,
IN
water
utilities,
while
GAC
was
used
at
the
MO,
OK
and
OH
water
utilities.

Monitoring
Results
for
Carbamate
Pesticides
Twenty­
eight
carbamates
pesticides
and
their
degradation
products
were
analyzed
in
the
reservoir
monitoring
study
(
Table
II.
D.
3.7).

Table
II.
D.
3.7.
Selected
carbamate
pesticides
and
their
degradation
products
included
in
the
reservoir
study,
USGS
Analytical
Schedules.
Pesticide
IUPAC
Name
Degradates
Alidcarb
2­
methyl­
2­(
methylthio)
propionaldehyde
O­
methylcarbamoyloxim
Aldicarb
sulfone,
Aldicarb
sulfoxide
Carbaryl
1­
naphthyl
methylcarbamate
1­
naphthol,
1,4­
naphthoquinone
Molinate
S­
ethyl
azepane­
1­
carbothioate
II.
D.
3
­
Page
19
of
26
Table
II.
D.
3.7.
Selected
carbamate
pesticides
and
their
degradation
products
included
in
the
reservoir
study,
USGS
Analytical
Schedules.
Pesticide
IUPAC
Name
Degradates
Methomyl
S­
methyl­(
EZ)­
N
 
(
methylcarbamoyloxy
thioacetimidate
Methomyl
oxime
Triallate
S­
2,3,3­
trichloroallyl
di­
isopropylthiocarbamate
Butylate
S­
ethyl
di­
isobutylthiocarbamate
Oxamyl
N,
N­
dimethyl­
2­
methylcarbamoyloxyimino­
2­(
methylthio)
acetamide
Oxamyl
oxime
Pebulate
S­
propyl
butyl(
ethyl)
thiocarbamate
Methiocarb
4­
methylthio­
3,5­
xylyl
methylcarbamate
Propoxur
2­
isopropoxyphenyl
methylcarbamate
Bendiocarb
2,3­
isopropylidenedioxyphenyl
methylcarbamate
Thiobencarb
S­
4­
chlorobenzyl
diethyl(
thiocarbamate)
Cycloate
S­
ethyl
cyclohexyl(
ethyl)
thiocarbamate
Benomyl
methyl­
1­[(
butylamino)
carbonyl]­
Hbenzimidazol
2­
ylcarbamate
Propham
isopropyl
phenylcarbamate
Carbufuran
2,3­
dihydro­
2,2­
dimethylbenzofuran­
7­
yl
methylcarbamate
3­
hydroxycarbofuran
Quality
Assurance
and
Quality
Control
Assessment
The
carbamate
pesticides
and
their
degradation
products
were
analytes
on
the
2001,
9002,
and
9060
USGS
analytical
schedules.
Methods
9002
and
9060
were
undergoing
QA/
QC
testing
during
the
monitoring
program.

In
1999,
water
samples
for
Method
9060
were
stored
for
extremely
long
periods
of
time
(
median
holding
time
of
90
days)
prior
to
extraction
(
Written
Communications
from
Joel
Blomquist,
USGS
9/
23/
02).
The
USGS
conducted
a
comparison
of
storage
times
with
median
half­
lives
in
organic
blank
water,
ground
water,
and
surface
water.
The
data
indicated
that
the
carbamate
pesticides
and
their
degradation
products
measured
by
Method
9060
have
median
half­
lives
ranging
from
4.5
days
for
methomyl
oxime
to
28.9
days
for
propham,
which
are
much
shorter
than
the
median
sample
holding
time
of
90
days
in
1999.
This
long
median
storage
time
relative
to
the
median
half­
lives
is
expected
to
create
a
systematic
bias
toward
lower
concentrations.
Any
detections
in
raw
and
finished
water
samples,
however,
should
be
considered
as
strong
evidence
of
pesticide
occurrence.
Reliable
quantification
of
carbamate
concentrations
in
finished
water
may
be
compromised
if
carbamate
pesticides
undergo
oxidation
and
hydrolysis
in
finished
drinking
water.

For
the
carbamate
pesticides,
the
mean
recovery
in
raw
water,
ranged
from
2%
­
140%,
the
maximum
recovery
ranged
from
17%
­
194%,
and
the
minimum
II.
D.
3
­
Page
20
of
26
recovery
ranged
from
0%
­
96%
(
Table
II.
D.
3.8).
For
finished
water,
the
mean
recovery
ranged
from
1%
­
134%,
the
maximum
recovery
ranged
from
9%
­
195%,
and
the
minimum
recovery
ranged
from
0%
­
72%.
The
mean
recovery
values
for
the
carbamate
pesticides
in
finished
water
samples
were
significantly
lower
(
P

0.05;
t­
test)
than
the
mean
recovery
values
for
the
raw
water
samples.

Table
II.
D.
3.8:
Summary
statistics
of
fortified
laboratory
set
and
matrix
samples
for
carbamate
pesticides
from
USGS
methods
2001
and
9060
(
decimal
percentage)
Compound
No.
of
Samples
Mean
%
recovery
Standard
Deviation
Maximum
Recovery
Median
Recovery
Minimum
Recovery
Schedule
Number
Raw
Water
Samples
Propoxur
31
78%
19%
103%
82%
29%
9060
3­
Hydroxycarbofuran
32
78%
23%
116%
83%
28%
9060
Aldicarb
32
11%
15%
49%
2%
0%
9060
Aldicarb
sulfone
32
23%
31%
179%
17%
0%
9060
Aldicarb
sulfoxide
32
36%
23%
83%
40%
0%
9060
Bendiocarb
32
69%
30%
123%
80%
0%
9060
Benomyl
32
73%
37%
148%
77%
0%
9060
Butylate
34
124%
30%
192%
115%
86%
2001
Carbaryl
23
139%
30%
185%
142%
88%
2001
Carbofuran
23
140%
28%
194%
142%
96%
2001
Carbofuran
32
84%
21%
124%
91%
29%
9060
Imidacloprid
31
124%
37%
193%
131%
29%
9060
Methomyl
32
77%
30%
144%
78%
0%
9060
Molinate
34
106%
22%
190%
100%
86%
2001
Oxamyl
32
62%
31%
116%
68%
0%
9060
Oxamyl
oxime
31
2%
4%
17%
0%
0%
9060
Triallate
33
105%
17%
174%
104%
82%
2001
Finished
Water
Samples
Propoxur
28
74%
22%
101%
80%
0%
9060
3­
Hydroxycarbofuran
30
71%
28%
117%
79%
0%
9060
Aldicarb
30
4%
11%
56%
0%
0%
9060
Aldicarb
sulfone
30
12%
12%
39%
6%
0%
9060
Aldicarb
sulfoxide
30
29%
40%
142%
10%
0%
9060
Bendiocarb
30
63%
31%
109%
77%
0%
9060
Benomyl
30
35%
45%
122%
3%
0%
9060
Butylate
31
31%
50%
153%
0%
0%
2001
Carbaryl
21
122%
41%
163%
135%
26%
2001
Carbofuran
26
134%
32%
189%
132%
63%
2001
Carbofuran
30
75%
24%
104%
81%
0%
9060
Imidacloprid
30
113%
36%
148%
123%
0%
9060
Methomyl
30
27%
34%
100%
7%
0%
9060
Molinate
31
26%
42%
111%
0%
0%
2001
Oxamyl
30
48%
32%
99%
54%
0%
9060
Oxamyl
oxime
28
1%
3%
9%
0%
0%
9060
Triallate
31
41%
43%
117%
18%
0%
2001
Summary
Statistics
II.
D.
3
­
Page
21
of
26
The
carbamate
pesticides
and
their
degradation
products
exhibited
extremely
low
detection
frequencies
in
raw
and
finished
water
samples
(
Table
II.
D.
3.9).
Out
of
the
28
compounds
analyzed
by
USGS
Methods
2001,
9002,
9060,
18
carbamate
pesticides
or
degradation
products
were
detected.
The
summary
statistics
were
identified
by
analytical
schedules
because
several
carbamate
pesticides
are
analytes
on
the
Method
9060
analytical
schedule.
As
previously
mentioned,
the
data
for
Method
9060
in
1999
should
be
considered
as
minimum
concentrations.
Additionally,
the
lack
of
a
detection
is
not
conclusive
evidence
that
the
compound
was
not
present
because
carbamates
are
unstable
in
finished
water
samples.

Other
considerations
are
associated
with
analytical
recoveries
of
the
carbamates.
There
was
a
difference
in
analytical
recovery
of
carbamate
pesticides
between
raw
and
finished
water,
with
lower
recoveries
lower
in
the
finished
water.
For
USGS
Method
2001,
carbaryl
and
carbofuran
were
considered
permanently
qualified
estimates
because
of
variable
recoveries
(
Zaugg,
et
al.
1995).
Aldicarb,
aldicarb
sulfone,
aldicarb
sulfoxide,
and
oxamyl
oxime
were
classified
as
qualified
estimates
for
Method
9060
because
recoveries
were
biased
outside
the
acceptable
range
of
median
recoveries
(<
60%
and
>
120%)
(
Furlong
et
al.,
2001)
The
USGS
defines
estimated
concentrations
when
low
analytical
recoveries
are
observed
or
when
qualitative
detections
of
concentrations
are
less
than
the
statistically
determined
limit
of
detection
(
LOD).

Table
II.
D.
3.9:
Summary
Table
for
Detections
of
Carbamates
and
their
Degradation
Products
Raw
Water
Finished
Water
Pesticide
sched
ule
N
No.
detects
Pct
detects
Max
Conc.
(
ug/
L)
Mean
Conc.
(
ug/
L)
N
No.
detects
Pct
detects
Max
Conc.
(
ug/
L)
Mean
Conc.
(
ug/
L)
1,4­
Naptho­
quinone
9002
316
3
0.95%
0.0054
0.0036
220
1
0.45%
0.0025
0.0025
1­
Napthol
9002
316
3
0.95%
0.2280
0.0806
220
0
3­
Hydroxycarbofuran
9060
311
0
224
1
0.45%
0.0094
0.0094
3­
keto­
carbofuran
9060
311
0
224
0
4­
Chlorobenzylmethyl
sulfo
9002
316
0
220
0
Aldicarb
9060
311
0
224
0
Aldicarb
sulfone
9060
311
1
0.32%
0.0074
0.0074
224
0
Aldicarb
sulfoxide
9060
311
0
224
0
Bendiocarb
9060
311
0
224
1
0.45%
0.0042
0.0042
Benomyl
9060
309
20
6.47%
0.0382
0.0247
223
2
0.90%
0.2150
0.1102
Butylate
2001
323
1
0.31%
0.0022
0.0022
227
0
Carbaryl
2001
323
7
2.17%
0.0465
0.0137
227
2
0.88%
0.0041
0.0035
9060
311
2
0.64%
0.0063
0.0059
224
0
Carbofuran
2001
323
2
0.62%
0.0188
0.0155
227
3
1.32%
0.0095
0.0082
9060
311
2
0.64%
0.0100
0.0074
224
1
0.45%
0.0041
0.0041
Cycloate
9002
316
0
220
0
9060
311
0
224
0
EPTC
2001
323
25
7.74%
0.0373
0.0154
227
11
4.85%
0.0286
0.0135
Methiocarb
9060
311
0
224
0
II.
D.
3
­
Page
22
of
26
Table
II.
D.
3.9:
Summary
Table
for
Detections
of
Carbamates
and
their
Degradation
Products
Raw
Water
Finished
Water
Pesticide
sched
ule
N
No.
detects
Pct
detects
Max
Conc.
(
ug/
L)
Mean
Conc.
(
ug/
L)
N
No.
detects
Pct
detects
Max
Conc.
(
ug/
L)
Mean
Conc.
(
ug/
L)
Methomyl
9060
311
0
224
0
Methomyl
oxime
9060
311
0
224
0
Molinate
2001
323
1
0.31%
0.0023
0.0023
227
0
Oxamyl
9060
311
0
224
0
Oxamyl
oxime
9060
307
0
220
2
0.91%
0.0139
0.0135
Pebulate
2001
323
0
227
0
Propham
9060
311
0
224
0
Propoxur
9060
307
1
0.33%
0.0048
0.0048
220
1
0.45%
0.8230
0.8230
Thiobencarb
2001
323
0
227
0
Triallate
2001
323
1
0.31%
0.0022
0.0022
227
0
Parent
carbamate
compounds
found
in
raw
water
samples
included
benomyl,
butylate,
carbaryl,
carbofuran,
EPTC,
molinate,
propoxur,
and
triallate.
EPTC
had
the
highest
detection
frequency,
7.74%
(
25
detections
in
323
samples)
in
raw
water
and
4.85%
(
11
detections
in
227
samples)
in
finished
water.
With
the
exception
of
benomyl
and
EPTC,
the
reported
concentrations
for
parent
carbamates
and
their
degradation
products
were
considered
qualified
estimates.

Parent
carbamate
detections
were
found
in
the
raw
water
of
the
OK,
PA,
SC,
MO,
NC,
LA,
NY,
and
SD
water
treatment
plants
(
Table
II.
D.
3.10).
In
finished
water,
parent
carbamate
detections
were
found
in
OK
for
bendiocarb,
MO
and
NY
for
benomyl,
MO
and
PA
for
carbaryl,
PA
for
carbofuran
(
Method
2001
and
9060),
and
PA
and
SD
for
EPTC.
No
carbamate
detections
were
found
in
the
TX
water
treatment
plant.

Table
II.
D.
3.10:
Summary
statistics
for
water
type,
year,
and
water
utility
for
carbamate
pesticides
and
their
degradation
products
(
ug/
L)
Nondetects
Conc.
Estimated
Conc.
Measured
Chemical
schedu
le
State
Year
Water
Type
No.
LOD
Range
No.
Range
No.
Range
9002
OK
2000
Raw
19
0.008
1
0.0054
.
.
9002
PA
2000
Finish
10
0.008
1
0.0025
.
.
9002
PA
2000
Raw
10
0.008
1
0.0037
.
.
1,4­
Napthoquinon
e
9002
SC
2000
Raw
23
0.008
1
0.0017
.
.
9002
MO
1999
Raw
19
0.005
1
0.228
.
.
9002
PA
2000
Raw
10
0.005
1
0.006
.
.
1­
Napthol
9002
SC
1999
Raw
19
0.005
1
0.0077
.
.
3­
Hydroxycarbofuran
9060
NC
1999
Finish
9
0.0623
1
0.0094
.
.

Aldicarb
sulfone
9060
NY
1999
Raw
9
0.16
1
0.0074
.
.
Bendiocarb
9060
OK
1999
Finish
9
0.0612
1
0.0042
.
.
9060
MO
2000
Finish
15
0.0219
1
0.215
.
.
9060
NY
2000
Finish
9
0.0219
1
0.0053
.
.
Benomyl
9060
OK
1999
Raw
13
0.0219
7
0.0187­
.
.
II.
D.
3
­
Page
23
of
26
Table
II.
D.
3.10:
Summary
statistics
for
water
type,
year,
and
water
utility
for
carbamate
pesticides
and
their
degradation
products
(
ug/
L)
Nondetects
Conc.
Estimated
Conc.
Measured
Chemical
schedu
le
State
Year
Water
Type
No.
LOD
Range
No.
Range
No.
Range
0.0382
9060
OK
2000
Raw
12
0.0219
6
0.0243­
0.0363
1
0.0242
9060
PA
1999
Raw
11
0.0219
2
0.0259­
0.03
01
.
.

9060
PA
2000
Raw
8
0.0219
3
0.0089­
0.0163
.
.

9060
SC
2000
Raw
23
0.0219
1
0.006
.
.
Butylate
2001
NY
2000
Raw
10
0.002
1
0.0022
.
.
2001
MO
1999
Finish
10
0.003
1
0.0041
.
.
2001
MO
2000
Raw
18
0.003­
0.041
1
0.008
.
.

2001
NC
1999
Raw
9
0.003
1
0.0039
.
.
2001
OH
2000
Raw
9
0.003
1
0.012
.
.
2001
PA
2000
Finish
10
0.003­
0.041
1
0.0028
.
.

2001
PA
2000
Raw
7
0.003­
0.041
4
0.0054­
0.0465
.
.

9060
PA
2000
Raw
10
0.0628
1
0.0054
.
.
Carbaryl
9060
SC
1999
Raw
19
0.0628
1
0.0063
.
.
2001
PA
2000
Finish
8
0.003­
0.03
3
0.0075­
0.0095
.
.

2001
PA
2000
Raw
9
0.003­
0.05
2
0.0121­
0.0188
.
.

9060
PA
2000
Finish
9
0.0566
1
0.0041
.
.
Carbofuran
9060
PA
2000
Raw
9
0.0566
2
0.0048­
0.01
.
.
2001
LA
2000
Raw
10
0.002­
0.03
1
0.011
.
.

2001
NY
1999
Raw
6
0.002­
0.01
.
.
5
0.0101­
0.0373
2001
NY
2000
Raw
3
0.002­
0.007
3
0.0021­
0.0033
5
0.0057­
0.0132
2001
PA
2000
Finish
10
0.002
1
0.0015
.
.
2001
PA
2000
Raw
10
0.002
1
0.0017
.
.
2001
SD
1999
Finish
6
0.002
.
.
4
0.0128­
0.0286
2001
SD
1999
Raw
6
0.002
.
.
4
0.0134­
0.0362
2001
SD
2000
Finish
5
0.002
.
.
6
0.0048­
0.0186
EPTC
2001
SD
2000
Raw
5
0.002­
0.01
.
.
6
0.0093­
0.0302
Molinate
2001
SC
1999
Raw
20
0.004
1
0.0023
.
.
Oxamyl
oxime
9060
OK
1999
Finish
8
0.0644
2
0.0131­
0.0139
.
.

9060
IN
1999
Finish
10
0.0594
.
.
1
0.823
Propoxur
9060
MO
1999
Raw
18
0.0594
1
0.0048
.
.
Triallate
2001
CA
1999
Raw
7
0.001
1
0.0022
.
.
II.
D.
3
­
Page
24
of
26
With
the
exception
of
1,4­
napthoquinone,
the
degradation
products
of
carbamates
were
detected
in
finished
or
raw
water
samples.
There
was
no
evidence
of
cooccurrence
of
parent
carbamate
pesticides
and
degradation
products
in
raw
and
finished
water
samples.
Although
3­
hydroxycarbofuran
and
oxamyl
oxime
were
detected
in
finished
water
samples,
there
are
no
concurrent
detections
of
carbofuran
or
oxamyl
in
"
paired"
raw
water
samples
to
conclude
transformation
during
treatment.
Although
1,4­
napthoquinone
was
detected
in
a
finished
and
raw
water
at
the
PA
site.
There
was
no
co­
occurrence
of
carbaryl
in
the
paired
finished
or
raw
water
sample
to
associate
1,4­
napthoquione
formation
to
carbaryl
degradation.
Because
aldicarb
sulfone
(
an
oxidative
degradation
product
of
aldicarb)
and
1­
napthol
(
hydrolysis
degradate
of
carbaryl)
were
only
detected
in
raw
water,
they
are
most
likely
environmental
degradation
products.

Results
and
Discussion
The
USGS
monitoring
data
for
selected
drinking
water
treatment
plants
on
reservoirs
indicates
a
low
detection
frequency
(<
1%
of
samples)
of
carbamate
pesticides
in
raw
and
finished
drinking
water.
The
carbamates
with
the
highest
detection
frequencies
(
6.5
to
7%
of
samples)
in
raw
water
samples
were
benomyl
and
EPTC.
In
finished
water
samples,
carbofuran
and
EPTC
had
the
highest
detection
frequencies
(
1
to
4%
of
samples).
Because
the
monitoring
program
was
not
targeted
to
carbamate
use
areas
and
sample
handling
issues,
the
reported
detection
frequencies
and
concentrations
may
underestimate
the
occurrence
for
carbamates
in
raw
and
finished
drinking
water.

Because
carbamate
degradation
products
were
detected
in
both
raw
and
finished
water
samples,
carbamate
degradation
products
were
formed
through
environmental
processes.
However,
laboratory
studies
indicate
that
degradation
products
can
be
formed
through
water
treatment
processes.

The
uncertainties
associated
with
monitoring
data
limit
a
clear
definitive
analysis
of
water
treatment
effects.

E.
LITERATURE
CITED
Bloomquist,
J.
D.,
J.
M.
Denis,
J.
K.
Cowles,
J.
A.
Hetrick,
R.
D.
Jones,
and
N.
B.
Birchfield.
2001.
Pesticides
in
Selected
Water­
Supply
Reservoirs
and
Finished
Drinking
Water,
1999­
2000:
Summary
of
Results
from
a
Pilot
Monitoring
Program.
USGS
Open­
File
Report
01­
456.

Blomquist,
J.
D.,
2001.
Transmittal
of
Preliminary
Digital
Data
Sets
From
the
USGSUSEPA
Program
"
Pesticides
in
Water­
Supply
Reservoirs
and
Finished
Drinking
Water­
A
Pilot
Monitoring
Program."
USGS,
Baltimore,
MD.
II.
D.
3
­
Page
25
of
26
Blomquist,
J.
D.,
Personal
Communications
on
April
28,
2000.

Furlong,
E.
T.,
D.
B.
Anderson,
S.
L.
Werner,
P.
P.
Soliven,
L.
J.
Coffey,
and
M.
R.
Burkhardt.
2001.
Methods
of
analysis
by
the
U.
S.
Geological
Survey
National
Water
Quality
Laboratory­
Determination
of
pesticides
in
water
by
graphitized
carbon­
based
solid­
phase
extraction
and
high­
performance
liquid
chromatography/
mass
spectrometry.
USGS
Water
Resource
Investigations
Report
01­
4134.

Ikehata,
K.
and
M.
Gamal
El­
Din,
2005.
Aqueous
Pesticide
Degradation
by
Ozonation
and
Ozone­
Based
Advanced
Oxidation
Processes:
A
Review
(
Part
1).
Ozone
Science
and
Engineering.
27:
83.

Mason,
Yael,
E.
Choshen,
and
C.
Chaim
Rav­
Acha.
1990.
Carbamate
Insecticides:
Removal
from
Water
by
Chlorination
and
Ozonation.
Wat
Res.
24:
11­
21.

Miles,
C.,
Oshiro.
1990.
Environ.
Toxicol.
Chem.
9:
535.

Miles,
Carl.
J,
M.
L.
Trehy,
and
R.
A.
Yost.
1988.
Degradation
of
N­
methylcarbamate
and
Carbamoyl
Oxime
Pesticides
in
Chlorinated
Water.
Bull.
Environ.
Contam.
Toxicol.
41:
838­
843.

Miles,
Carl
J.
1991.
Degradation
of
Aldicarb,
Aldicarb
Sulfoxide,
and
Aldicarb
sulfone
in
Chlorinated
Water.
Environ.
Sci.
Technol.
25:
1774­
1779.

Miltner,
R.
J.,
D.
B.
Baker,
T.
F.
Speth,
and
C.
A.
Fronk.
1989.
Treatment
of
Seasonal
Pesticides
in
Surface
Waters.
Jour.
AWWA.
81:
43.

Miltner,
Richard.
J.
June,
2005.
Status
Report:
Summary
of
ORD/
WSWRD
Studies
to
Control
n­
methyl
Carbamates
in
Drinking
Water.
USEPA/
ORD/
WSWRD.
Cincinnati,
OH
Sandstrom,
M.
W.,
M.
E.
Stroppel,
W.
T.
Foreman,
M.
P.
Schroeder,
2001.
Methods
of
analysis
by
the
U.
S.
Geological
Survey
National
Water
Quality
Laboratory­
Determination
of
moderate
use
pesticides
and
selected
degradates
in
water
by
C­
18
solid
phase
extraction
and
gas
chromatography/
mass
spectrometry.
USGS
Water
Resource
Investigations
Report
01­
4098.

Speth,
Thomas
F.
and
R.
Pisigan.
2001.
Softening
and
Chlorination
Screening
Studies
for
Selected
Pesticides.
USEPA/
ORD/
WSWRD.
Cincinnati,
OH.

Speth,
T.
F.
and
R.
J.
Miltner.
1998.
Technical
Note:
Adsorption
Capacity
of
GAC
for
Synthetic
Organics,
Jour.
AWWA.
90:
171.

Speth,
T.
F.
and
R.
J.
Miltner.
1990.
Technical
Note:
GAC
Adsorption
Capacity
for
SOCs.
Jour.
AWWA.
82:
72.
II.
D.
3
­
Page
26
of
26
Summers,
R.
S.,
S.
M.
Hooper,
H.
M.
Shukairy,
G.
Solarik,
and
D.
M.
Owen.
1994.
Assessing
DBP
Yield
under
Uniform
Conditions.
Jour.
AWWA.
88:
6:
80.

U.
S.
EPA,
2000.
Progress
Report
on
Estimating
Pesticide
Concentrations
in
Drinking
Water
and
Assessing
Water
Treatment
Effects
on
Pesticide
Removal
and
Transformation:
A
Consultation.
FIFRA
Scientific
Advisory
Panel
(
SAP),
September
29,2000.
http://
www.
epa.
gov/
scipoly/
2000/
september/
sept­
00
 
sap­
dw­
0907.
pdf).

U.
S.
EPA.,
2001.
Office
of
Pesticide
Programs
Science
Policy:
The
Incorporation
of
Water
Treatment
Effects
on
Pesticide
Removal
and
Transformations
in
Food
Quality
Protection
Act
(
FQPA)
Drinking
Water
Assessments.
USEPA/
OPP,
Washington,
DC.
http://
www.
epa.
gov/
oppfead1/
trac/
science/
water_
treatment.
pdf
Zaugg,
S.
D.
,
M.
W.
Sandstrom,
S.
G.
Smith,
and
K.
M.
Fehlberg.
1995.
Methods
of
analysis
by
the
U.
S.
Geological
Survey
National
Water
Quality
Laboratory­
Determination
of
pesticides
in
water
by
C­
18
solid
phase
extraction
and
capillary­
column
gas
chromatography/
mass
spectrometry
with
selected
ion
monitoring.
USGS
Open
File
Report
95­
181.
