1
OFFICE
OF
PREVENTION,
PESTICIDES
AND
TOXIC
SUBSTANCES
UNITED
STATES
ENVIRONMENTAL
PROTECTION
AGENCY
WASHINGTON
D.
C.,
20460
PC
Code:
056502
DP
Barcode:
308564
Date:
July
19,
2005
MEMORANDUM
SUBJECT:
Review
of
Surface
Water
Monitoring
Data
(
Summary
Report1
and
Original
Monitoring
Reports)
Submitted
to
Support
PCNB
Reregistration.

TO:
Jill
Bloom,
Chemical
Review
Manager
Special
Review
and
Re­
Registration
Division
FROM:
Elizabeth
Behl,
Branch
Chief
Cheryl
A.
Sutton,
Ph.
D.,
Environmental
Scientist
Environmental
Risk
Branch
IV
Environmental
Fate
and
Effects
Division
(
7507C)

OPP
has
completed
its
review
of
monitoring
data
cited
in
the
document
"
A
Summary
of
Water
Quality
Monitoring
data
for
Pentachloronitrobenzene"
(
MRID
46327001;
referred
to
subsequently
as
the
"
Summary
Report")
submitted
in
support
of
the
re­
registration
of
the
fungicide
pentachloronitrobenzene
(
PCNB).
EFED
previously
reviewed
the
Summary
Report
and
reported
its
overall
conclusion
in
the
executive
summary
of
the
document
Revised
Environmental
Fate
and
Ecological
Risk
Assessment
of
Pentachloronitrobenzene
(
PCNB)
(
DP
Barcode
D291276;
dated
2/
15/
2005).
In
response
to
a
request
from
the
registrants
EFED
agreed
to
review
the
original
monitoring
reports
from
each
study
(
cited
individually
below);
these
documents
were
provided
to
EFED
on
June
24,
2005.

The
Summary
Report
contains
information
on
five
surface
water
monitoring
studies
conducted
by
different
scientists
in
which
PCNB
was
an
analyte.
This
document
reviews
the
original
reports
on
these
five
studies,
provides
OPP's
conclusions
on
these
data,
and
comments
on
conclusions
reached
in
the
Summary
Report
based
on
these
same
data.
Although
ground
water
and
sediment
were
also
sampled
in
some
studies,
this
review
focuses
on
the
surface
water
monitoring,
at
the
request
of
SRRD.

1
A
Summary
of
Water
Quality
Monitoring
data
for
Pentachloronitrobenzene,
7/
14/
2004,
Cohen,
S.
Z.
et
al.,
48p.
MRID
463270­
01
2
I.
Reviews
of
Surface
Water
Monitoring
Data
by
Site
1.
Hudson
National
Golf
Course
a.
Reports
provided:

i.
Environmental
Monitoring
Report
­
1996
(
1/
14/
1997;
4/
5/
1997):
Describes
monitoring
conducted
in
1996.
ii.
1997
Annual
Report
Monitoring
Program
Hudson
National
Golf
Club
Croton­
on­
Hudson
Volume
1
­
Report
(
4/
3/
1998):
Describes
monitoring
conducted
in
1997
and
on
1/
27/
1998.
iii.
1998
Annual
Report
Monitoring
Program
Hudson
National
Golf
Club
Croton­
on­
Hudson
(
7/
31/
2000):
Describes
monitoring
conducted
in
1998.
iv.
1999
Annual
Report
Monitoring
Program
Hudson
National
Golf
Club
Croton­
on­
Hudson
(
1/
16/
2001):
Describes
monitoring
conducted
in
1999.

b.
Pesticide
use
information
Application
records
were
provided
for
1996.
The
only
usage
data
provided
for
other
years
was
in
the
Summary
Report.
Specific
details
are
presented
in
Table
1.

Table
1:
Pesticide
Usage
Data
for
Hudson
National
Golf
Club
Application
date
Rate
Areas
treated
11/
11/
1996
50
lbs
0.75
WSP
PCNB
(
8
oz/
m;
units
not
described)
All
greens
11/
13/
1996
25
lbs.
0.75
WSP
PCNB
(
4
oz/
m;
units
not
described)
All
tees
11/
3/
1997
"
at
half
rate"
(
9
lb.
ai/
ac?)
Tees
11/
20/
1997
"
at
half
rate"
(
9
lb.
ai/
ac?)
Greens
12/
3/
1997
"
at
half
rate"
(
9
lb.
ai/
ac?)
Tees
12/
5/
1997
"
at
half
rate"
(
9
lb.
ai/
ac?)
Greens
1998
Not
applied
­­
1999
Not
applied
­­

The
9
lb.
ai/
ac
rate
was
reported
in
summary.
How
that
correlates
with
the
total
quantity
used
(
24­
36
lbs)
reported
in
the
monitoring
report
for
this
time
period
could
not
be
determined.
A
total
of
24
pounds
of
product
was
applied
to
tees;
36
pounds
was
applied
to
greens.
Applications
(
total
pounds)
were
split
on
two
separate
dates.
PCNB
was
also
used
in
February
of
1996;
however,
since
monitoring
did
not
occur
until
10
months
later,
that
is
not
included
in
Table
1
above.

c.
Sampling
design
This
study
was
not
specifically
designed
to
evaluate
impacts
from
the
use
of
PCNB
on
water
quality.
Instead,
sampling
took
place
throughout
the
year
to
evaluate
water
quality
impacts
from
all
pesticide
and
fertilizer
applications
to
the
golf
course.
The
3
surface
water
sampling
that
occurred
within
two
months
of
application
of
PCNB
was
quite
limited.
Table
2
below
contains
only
surface
water
data
collected
in
sampling
periods
during
each
year
that
might
be
relevant
to
the
impact
of
PCNB
use
on
water
quality.
For
example,
in
1997,
there
were
a
total
of
six
surface
water
sampling
events:
three
representing
"
dry"
weather"
(
i.
e.,
baseflow
conditions)
and
three
representing
"
wet"
weather
(
i.
e.,
runoff
events).
In
1997,
only
the
"
dry"
sampling
on
12/
12/
1997
and
the
"
wet"
sampling
on
12/
30/
1997
occurred
after
treatment;
only
these
data
are
included
in
Table
2.
An
additional
round
of
sampling
occurred
in
January
1998,
which
is
also
included
here
as
it
appears
to
represent
impacts
from
the
1997
late
winter
treatment.

Table
2:
Summary
of
surface
water
monitoring
data
from
Hudson
National
Golf
Club
relevant
to
PCNB
application.
Sampling
date
posttreatment
Number
of
locations
sampled
(
typically
one
sample
collected
per
location)***
Detections
Comments
(
e.
g.
stormflow
versus
composite
sample;
if
runoff
is
from
treated
areas)

12/
12/
1996
5
sites
sampled
detection
at
one
site
(
SW3a:
0.12
ppb)
Others
sites
<
0.05
ppb
Stormflow
sampled.
Last
treatment
was
one
month
earlier.

12/
11/
1997
6
sites
sampled
<
DL
"
Dry"
sampling;
represents
base
flow
rather
than
runoff.
Relevance
unknown.
12/
30/
1997
6
sites
sampled
Detections*
at
5
sites
(
3.9;
3.7;
5.3;
1.2;
1.1
ppb)
SW5
site
<
0.5
ppb
Some
duplicates
analyzed
as
well
with
concentrations
in
same
range
(
e.
g.,
one
at
6.4
ppb)
"
Wet"
sampling;
represents
storm
runoff
event.

1/
27/
1998
5
sites
sampled
Detections
at
4
sites
(
trace**;
0.6;
trace;
trace)
SW4
site
<
0.5
ppb
No
rainfall
occurred
on
the
sampling
date.
Detections
result
from
Nov/
Dec
1997
application.

5/
21/
1998
to
end
of
1998
Samples
continued
to
be
collected
and
analyzed
for
PCNB
throughout
1998.
Not
detected
when
not
used.
Not
relevant.

Sampling
continued
throughout
1999
and
through
2003,
despite
lack
of
use.
Samples
continued
to
be
collected
and
analyzed
for
PCNB
throughout
1999.
Not
detected
when
not
used.
Not
relevant.

*
Multiple
additional
precipitation
events
occurred
between
the
original
PCNB
applications
and
this
sampling
event;
PCNB
concentrations
in
runoff
from
these
events
may
have
been
higher.
**"
trace"
concentrations
were
less
than
reporting
limit
(
0.5
ppb)
but
greater
than
the
limit
of
detection.
***
Site
SW6
is
a
background
sample
and
isn't
included
in
sampling
and
detections
4
Sample
analysis
Only
PCNB
was
analyzed
and
reported
in
the
site­
specific
monitoring
data
reports
cited;
the
PCNB
degradate
pentachloroaniline
(
PCA),
a
residue
of
concern
with
respect
to
water
quality,
was
not
included.
No
information
was
provided
on
PCA
analysis,
thus
we
could
not
determine
how
the
concentrations
were
estimated
and
what
the
associated
error
or
uncertainty
is.
The
Summary
Report
provides
PCA
concentrations
based
on
an
analysis
of
archived
spectra,
however
which
samples
contained
PCA
(
from
which
locations
or
from
which
sampling
period).
No
QA/
QC
information
was
provided
on
PCA
analysis.

Table
3:
Summary
information
on
analytical
methods
used
for
PCNB
and
QA/
QC
used
for
surface
water
samples
collected
from
Hudson
National
Golf
Club.
Analytical
method
Quantitation
limit,
Detection
limit
Percent
recovery
for
compound
of
interest
Description
of
QA/
QC
(
spikes,
blanks,
duplicates,
etc.)
EPA
method
8080
GC/
ECD
(
1996
samples)
PQL:
not
available
DL:
0.05
ppb
Not
provided
Not
described
SDWA
method
525.2
(
1997,
1998,
and
1999
samples)
MDL:
0.5
ppb
(
although
identified
as
MDL,
this
appears
to
be
the
reporting
limit)
104.5%
for
blank
46­
53%
for
matrix
spike
surrogate
(
reported
in
1997
for
SS1­
3):
SS1:
59­
97
%
recovery
SS2:
80­
100
%
recovery
SS3:
105­
135
%
recovery
Reported
in
1997:
"
poor
recovery
of
surrogate
sample";
matrix
spikes
and
duplicate
samples
were
analyzed.
Field
and
lab
blanks
also
analyzed.

d.
Summary
PCNB
was
reported
in
surface
water
samples
collected
at
Hudson
National
Golf
Course
at
concentrations
ranging
from
0.12
ppb
to
6.4
ppb.
PCA
values
are
reported
in
the
Summary
Report,
but
these
results
are
highly
uncertain
based
on
a
lack
of
QA/
QC
information.

Following
application
in
the
fall
of
1996,
five
sites
were
sampled
in
one
sampling
event.
Following
application
in
1997,
approximately
the
same
locations
(
six
sites)
were
sampled
in
three
sampling
events
(
including
one
in
January
1998),
only
one
of
which
represented
a
storm
event.
Roughly
17
samples
potentially
relevant
to
the
use
of
PCNB
were
collected;
this
total
includes
six
"
dry"
samples,"
the
relevance
of
which
is
unknown.
In
both
years
applications
were
made
to
tees
and
greens.

The
highest
reported
detections
were
associated
with
storm
events;
samples
were
collected
roughly
one
month
post­
treatment.
Multiple
precipitation
events
that
were
not
sampled
occurred
between
pesticide
application
and
the
storm
event
that
was
sampled.
PCNB
use
was
suspended
at
the
course
following
detections
in
five
of
six
monitoring
locations
on
12/
30/
1997,
after
applications
in
early
November
and
December
of
that
year.
5
Sampling
a
month
later
(
in
late
January)
of
base
flow
(
not
a
storm
event)
resulted
in
additional
detections
at
low
levels
in
four
of
five
locations
sampled.

Only
two
years
of
this
multi­
year
study
are
relevant
to
assessment
of
water
quality
impacts
resulting
from
PCNB
use
on
golf
courses,
as
PCNB
was
only
applied
in
those
years.
Very
few
of
the
total
surface
water
samples
collected
and
analyzed
in
any
given
year
when
PCNB
was
applied
(
1996
and
1997)
are
relevant
to
PCNB
applications.
For
example,
in
1996,
Table
4
of
the
Summary
Report
indicates
that
39
samples
were
collected.
Only
five
of
these
were
collected
following
an
application
of
PCNB;
the
remainder
were
collected
before
PCNB
was
used
at
the
course.
Similarly,
in
1997,
samples
were
collected
in
only
one
runoff
sampling
event
post­
treatment;
the
other
five
sampling
events
took
place
before
PCNB
was
applied.

The
Summary
Report
included
results
of
31
samples
collected
in
1997
and
analyzed
for
PCNB
prior
to
the
use
of
that
compound
on
the
golf
course;
thus,
the
nondetections
on
PCNB
in
these
samples
were
to
be
expected
and
are
of
no
relevance
with
respect
to
water
quality
impacts
from
PCNB
use.
Surface
water
sample
collection
and
analysis
continued
throughout
1998
and
1999,
although
PCNB
use
was
discontinued
after
the
December
5,
1997
application.
The
sampling
totals
for
1998­
2003
(
showing
many
samples
collected
and
analyzed
with
no
PCNB
detected)
provided
in
Table
4
of
the
Summary
Report
are
misleading
with
regard
to
long­
term
impacts
of
PCNB
resulting
from
continuous
use,
given
that
PCNB
use
stopped
after
December
5,
1997.
The
study
does
show
impacts
from
the
Fall
treatment
(
1997)
on
water
quality
for
1.5
to
2.5
months
following
application
(
treatment
in
November/
December
with
resulting
detections
in
December/
January).
Unfortunately,
since
no
additional
samples
were
analyzed
for
PCNB
after
January
(
when
the
decision
was
made
to
discontinue
use)
this
dataset
cannot
be
used
to
draw
conclusions
about
the
duration
of
environmental
impacts
of
PCNB
Use.

Uncertainties:

With
a
change
in
contractors
in
1997,
some
sampling
locations
were
moved.
Although
the
sampling
locations
are
described
in
the
Summary
Report
and
in
the
sitespecific
monitoring
reports,
it
is
still
difficult
to
understand
the
hydraulic
connection
between
the
treated
areas
and
the
sampling
points.
The
degree
to
which
untreated
areas
contribute
to
the
water
body
being
sampled
is
also
unclear.
The
data
provided
do
not
support
the
conclusion
in
the
Summary
Report
that
these
monitoring
data
represent
"
significant
areas
of
treated
turf."
The
impact
of
moving
the
sampling
locations
is
an
additional
uncertainty.

With
a
change
in
contractors
in
1997,
methods
used
for
PCNB
analysis
changed.
The
detection
limit
associated
with
PCNB
analytical
methods
increased
an
order
of
magnitude
(
0.05
to
0.5
ppb)
from
1996­
1997.
Recoveries
for
PCNB
were
not
provided
for
the
earlier
method,
but
ranged
from
59%
to
135
%
for
the
later
method,
based
on
surrogate
standards.
The
laboratory
reports
from
1997
indicate
some
uncertainty
in
the
values
reported.
6
Multiple
precipitation
events
that
were
not
sampled
occurred
between
the
time
when
PCNB
was
applied
and
the
time
when
the
storm
event
that
was
sampled
occurred.
PCNB
concentrations
may
have
been
significantly
higher
in
runoff
from
these
interim,
unsampled
precipitation
events.
The
difference
in
what
is
represented
by
samples
collected
in
"
dry"
versus
"
wet"
sampling
rounds
is
also
unclear.
The
report
indicates
that
"
dry"
samples
represent
baseflow,
but
the
source
of
baseflow
isn't
described
(
potentially
a
combination
of
ground
water
and
in­
stream
flow
from
up­
gradient
runoff
unrelated
to
the
treated
area).
Because
of
this
interpretation
of
these
results
is
unclear.

Concentrations
of
PCA
were
reported
in
the
Summary
Report
based
on
a
re­
analysis
of
archived
mass
spectra.
These
values
not
supported
by
data
provided
in
the
monitoring
report
or
laboratory
reports
and
are
of
uncertain
quality.
No
data
are
provided
on
PCA
method
detection
limit,
recovery,
or
QA.

2.
Monterey
Peninsula
a.
Reports
provided:

i.
Stormwater
Monitoring
of
Golf
Course
Watersheds
of
the
Monterey
Peninsula
­.
(
April
1996)
ii.
Stormwater
Monitoring
of
Golf
Course
Watersheds
of
the
Monterey
Peninsula
­.
(
June
1996)
iii.
Stormwater
Monitoring
of
Golf
Course
Watersheds
of
the
Monterey
Peninsula
­
fall
and
winter
1996­
1997.
(
July
1997)
iv.
Stormwater
Monitoring
of
Golf
Course
Watersheds
in
the
Del
Monte
Forest
 
fall
and
winter
1997­
1998.
(
January
1999)
v.
Stormwater
Monitoring
of
Golf
Course
Watersheds
in
the
Del
Monte
Forest
 
fall
and
winter
1998­
1999.
(
May
2000)
vi.
Stormwater
Monitoring
of
Golf
Course
Watersheds
in
the
Del
Monte
Forest
 
fall
and
winter
1999­
2000.
(
September
2000)

b.
Pesticide
use
information
No
PCNB
usage
data
were
provided
for
the
monitoring
conducted
at
Monterey
Peninsula/
Del
Monte
Forest.
The
Summary
Report
states
that
the
consultant
was
able
to
"
confirm
that
putting
green
surfaces
were
treated
with
PCNB
at
the
Cyprus
Point
Club
and
the
Spyglass
Hill
Golf
Course
in
Pebble
Beach,
CA
at
the
time
of
the
study."
No
information
on
application
rates
or
timing
was
provided.
Usage
data
appear
to
have
been
collected
retrospectively,
following
detections,
and
were
not
reported
for
any
golf
courses
other
than
the
two
mentioned
above.
7
c.
Sampling
design
This
study
was
not
specifically
designed
to
evaluate
impacts
from
the
use
of
PCNB
on
water
quality.
Instead,
sampling
took
place
to
evaluate
water
quality
impacts
from
all
pesticides,
fertilizer,
surfactants,
oil
and
grease
and
other
compounds.
Water
samples
were
analyzed
for
approximately
59
compounds
(
including
organochlorines,
organophophates,
dithiocarbamates)
in
five
different
drainages.
Some
locations
were
selected
to
be
unaffected
by
golf
course
runoff,
some
sites
had
mixed
influences
and
are
located
downstream
of
runoff
from
all
sources
(
including
urban
residential
use).

Because
the
PCNB
application
dates
were
not
available,
it
is
impossible
to
confirm
that
the
surface
water
sampling
that
occurred
followed
application
of
PCNB.
Table
4
below
contains
results
of
all
surface
water
data
collected
in
the
study.
As
no
information
on
application
rates,
timing,
or
location
was
provided,
these
results
are
not
representative
of
the
impact
of
PCNB
use
on
water
quality,
and
provide
only
anecdotal
information.

Table
4:
Summary
of
surface
water
monitoring
data
from
Monterey
Peninsula/
Del
Monte
Forest.
Relevance
to
PCNB
is
unclear.
Sampling
date
(
not
known
if
post­
treatment;
no
usage
data
available)
Number
of
locations*
sampled
(
typically
one
sample
collected
per
location)
Detections
Description
(
stormflow
versus
composite
sample;
is
runoff
from
treated
areas?)
1996
(
April
and
June)
Not
relevant
PCNB
not
reported
as
analyte
­­

9/
26/
1996
5
<
DL
"
dry"
9/
13/
1996
0
10/
29/
1996
7
(
of
9
stations)
<
DL
Storm
event
1/
15/
1997
9
1
detection
(
0.13
ppb;
Fanshell
beach)
Storm
event.
Fanshell
beach
represents
runoff
from
Cyprus
Beach
golf
course
12/
5/
1997
9
<
DL
Storm
event
1/
2/
1998
9
<
DL
Storm
event
11/
7/
1998+
11/
30/
1998
9
<
DL
Storm
event
3/
19/
1999
9
<
DL
Storm
event
11/
7/
1999
7
(
of
9)
1
detection
(
1.6
ppb;
Spy
Glass
Beach)
Storm
event
Spy
Glass
beach
represents
runoff
from
the
Spyglass
golf
course
and
a
residential
community.
1/
24/
2000
9
<
DL
Storm
event
*
PCNB
application
could
not
be
determined
at
all
locations
from
the
data
provided;
thus,
some
of
the
locations
sampled
may
not
be
relevant.

d.
Sample
analysis
Only
PCNB
was
analyzed;
the
PCNB
degradate
PCA
was
not
included
in
this
study
as
an
analyte.
PCNB
was
not
included
as
analyte
until
fall/
winter
1996­
7.
8
Table
5:
Summary
information
on
analytical
methods
and
QA/
QC
used
for
surface
water
samples
collected
from
Monterey
Peninsula/
Del
Monte
Forest.
Analytical
method
Quantitation
limit,
Detection
limit
Percent
recovery
for
compound
of
interest
Description
of
QA/
QC
(
spikes,
blanks,
duplicates,
etc.)
EPA
method
8080
(
GC/
ECD)
Reporting
limit:
0.05
ppb
Not
specified
Not
specified
e.
Summary
PCNB
was
detected
in
two
samples
from
two
different
locations
in
these
reports.
Concentrations
were
0.13
ppb
and
1.6
ppb
and
were
associated
with
storm
events.
PCA
was
not
analyzed
during
this
study.

Given
the
lack
of
information
provided
on
PCNB
usage
(
location
and
timing),
results
from
this
study
cannot
be
used
to
draw
broad
conclusions
about
the
potential
for
PCNB
impacts
on
water
quality
from
use
according
to
the
label.
The
lack
of
detections
and
concentrations
reported
could
result
from
a
multitude
of
factors
ranging
from
lack
of
application
of
PCNB
to
advanced
water
quality
management
at
these
courses,
which
reduces
overall
pollutant
discharge.
As
both
these
hypotheses
are
plausible,
but
highly
uncertain,
this
study
can
be
used
only
anecdotally
and
does
not
add
to
the
weight
of
evidence
on
PCNB­
specific
impacts
(
or
lack
thereof)
on
water
quality.
Overall
water
quality
assessed
in
this
study
was
clean,
or,
in
the
study
author's
words,
"
uncontaminated
by
typical
indicators
of
urban
or
agricultural
runoff."

Uncertainties:

Inadequate
usage
data
were
provided.
The
relevance
of
samples
for
assessing
PCNB
impacts
on
water
quality
is
highly
uncertain.
No
QA/
QC
practices
are
described
for
these
monitoring
data
with
respect
to
collection
or
analysis.
Recovery
data
were
not
provided.

3.
Echo
Falls
a.
Reports
provided:

i.
Laboratory
Results
for
years
that
fenarimol
was
detected
in
water
samples
collected
at
Echo
Falls
Golf
Course
(
1990,
1992­
96),
Evans
Hamilton,
Inc.,
internal
report
to
E&
TS.
ii.
Echo
Falls
Golf
and
Country
Club
Surface
Water
Quality
Monitoring
(
Internal
review
sheets)
9
b.
Pesticide
use
information
The
Summary
Report
states
that
"
PCNB
is
applied
two
times
a
year.
No
other
application
information
is
available."
No
information
on
the
timing
or
location
of
PCNB
use
was
available.

c.
Sampling
design
The
only
information
on
sampling
design
is
provided
in
the
Summary
Report.
The
study
was
required
by
Snohomish
County,
WA,
as
a
condition
for
development
approval;
it
was
not
specifically
designed
to
evaluate
impacts
from
the
use
of
PCNB
on
water
quality.
Baseline
monitoring
began
in
1990;
"
operational"
monitoring
began
in
1992.
Two
samples
were
analyzed
for
PCNB
on
April
25,1996.
Although
the
Summary
Report
indicates
that
monitoring
continues,
no
data
were
provided
for
after
1996.
The
temporal
connection
between
the
time
of
application
and
sample
collection
could
not
be
determined
from
the
information
provided.
The
spatial
connection
between
the
location
treated
and
the
sampling
points
could
not
be
determined
from
the
information
provided.

Table
6:
Summary
of
surface
water
monitoring
data
from
Echo
Falls
Golf
and
Country
Club
potentially
relevant
to
PCNB
application.
Sampling
date
(
not
known
if
post­
treatment;
no
usage
data
available)
Number
of
locations
sampled
(
typically
one
sample
collected
per
location)
Detections
Description
(
storm
flow
versus
composite
sample;
is
runoff
from
treated
areas?)
April
25,
1996
1
(
Anderson
Creek)
0
Not
specified
One
sample
was
also
collected
at
Evans
Creek.
However,
the
Summary
Report
indicates
that
it
was
a
background
sampling
location,
and
therefore
it
is
not
included
in
Table
6
above.

d.
Sample
analysis
Only
PCNB
was
analyzed;
the
PCNB
degradate
PCA
was
not
included
as
an
analyte.

Table
7:
Summary
information
on
analytical
methods
and
QA/
QC
used
for
surface
water
samples
collected
from
Echo
Falls
Golf
and
Country
Club.
Analytical
method
Quantitation
limit,
Detection
limit
Percent
recovery
for
compound
of
interest
Description
of
QA/
QC
(
spikes,
blanks,
duplicates,
etc.)
Not
specified
0.05
ppb
Not
specified
Detailed
information
not
provided.
Some
indication
that
method
blanks
and
spikes
were
conducted.
10
e.
Summary
No
information
on
the
timing,
rate
or
location
of
PCNB
use
was
provided.
A
single
sample
was
collected
(
and
one
background
sample);
PCNB
was
not
detected,
but
it
is
not
clear
if
an
application
had
occurred
as
of
the
sampling
date.
It
is
not
known
if
the
collected
sample
represents
base
flow
or
storm­
related
runoff.
PCA
was
not
analyzed
during
this
study.

One
of
the
two
documents
submitted
for
this
study
contained
no
information
about
PCNB.
Without
any
information
on
PCNB
usage,
these
results
do
not
provide
information
upon
which
to
base
conclusions
about
the
potential
for
PCNB
impacts
on
water
quality
to
occur
from
use
according
to
the
label.

Uncertainties:

No
usage
data
were
provided.
The
single
sampling
location
was
not
described.
No
QA/
QC
practices
are
described
for
these
monitoring
data
with
respect
to
collection
or
analysis.
Recovery
data
were
not
provided.

4.
Twin
Cities,
MN
a.
Reports
provided:
i.
Quantity
and
Quality
of
Runoff
from
Four
Golf
Courses
in
the
Twin
Cities
Metropolitan
area,
J.
M.
Barten,
September
1995
ii.
Laboratory
Report
3358,
Interpoll,
August
16,
1994
b.
Pesticide
use
information
PCNB
was
only
applied
at
the
various
golf
courses
during
the
month
of
October
1994;
specific
details
are
presented
in
Table
8.

Table
8:
PCNB
Usage
Data
for
golf
courses
in
the
Twin
Cities
Metropolitan
area.
Application
date
Rate*
Areas
treated
Fraction
of
total
area
treated
10/
24/
1994
12
oz/
1000
sq
ft
(
total
29.4
lbs)
All
greens
(
0.9
acres)
at
Baker
National
golf
course
<
2%
treated
(
total
area
46.7A)
10/
20/
1994
8
oz/
1000
sq
ft
(
total
15.7
lbs)
Greens
(
1.8
acres
of
greens)
at
Meadowbrook
golf
course
<
2%
treated
(
total
area
93.7A)

10/
24/
1994
4
oz/
1000
sq
ft
(
total
16.3
lbs)
Greens
(
0.4
acres)
at
Woodhill
Country
Club
<
1.3%
treated
(
total
area
30.9A)
10/
10/
1994
6
oz/
1000
sq
ft
(
total
19.5
lbs)
Greens
(
3
acres)
at
Minikada
Club
2%
treated
(
total
area
137A)
10/
20/
1994
6
oz/
1000
sq
ft
(
total
13.1
lbs)
Tees
(
2
acres)
at
Minikada
Club
<
2%
treated
(
total
area
137A)
10/
22/
1994
6
oz/
1000
sq
ft
(
total
169.9
lbs)
Fairways
(
26
acres)
at
Minikada
Club
19%
treated
(
total
area
137A)
*
Since
the
formulation
is
not
specified,
it
is
not
possible
to
calculate
the
rate
in
pounds
a.
i./
Acre.
11
c.
Sampling
design
Automatic
samplers
were
installed
at
four
golf
courses
and
analyzed
for
the
presence
of
nutrients,
fungicides,
herbicides,
and
heavy
metals.
Flow­
weighted
composite
samples
were
collected
at
each
site;
however,
it
is
unclear
what
time
period
these
composites
represent.
Due
to
costs,
fungicides
were
analyzed
only
in
the
1994
season.
Samples
were
collected
from
intermittent
streams
except
at
Woodhill
Country
Club,
where
samples
were
collected
from
"
a
large
pit."
At
the
other
locations,
small
events
were
not
represented,
as
the
sample
volumes
collected
were
too
small
for
analysis.
Sampling
was
in
October;
however,
sample
collection
dates
were
not
provided
in
the
monitoring
report,
nor
was
there
an
indication
of
the
time
period
over
which
samples
were
composited.
The
Summary
Report
does
indicate
that
samples
were
collected
10/
17
and
10/
18
following
a
rainfall
event.
PCNB
was
applied
to
only
one
golf
course
prior
to
that
event;
thus,
only
one
potentially
relevant
sample
was
collected,
potentially
representing
application
to
greens
at
the
Minikada
club.
Rainfall
information
is
aggregated
by
month,
so
it
was
not
possible
to
determine
from
the
report
if
storm
events
occurred
following
application
and
prior
to
the
event
sampled
a
week
later.

Table
9:
Summary
of
surface
water
monitoring
data
from
golf
courses
in
the
Twin
Cities
Metropolitan
area
that
are
potentially
relevant
to
PCNB
application.
Sampling
date
Number
of
locations
sampled
(
one
sample
collected
per
location)
Detections
Description
(
stormflow
versus
composite
sample;
is
runoff
from
treated
areas?)
10/
17/
1994
and
10/
18/
1994
1
Minikada
club
0
Monitoring
report
indicates
that
samples
were
composites;
however
the
summary
report
indicates
that
samples
represented
discrete
storm
events.

d.
Sample
analysis
Only
PCNB
was
analyzed;
the
PCNB
degradate
PCA
was
not
included
as
an
analyte.

Table
10:
Summary
information
on
analytical
methods
and
QA/
QC
used
for
surface
water
samples
collected
from
golf
courses
in
the
Twin
Cities.
Analytical
method
Quantitation
limit,
Detection
limit
Percent
recovery
for
compound
of
interest
Description
of
QA/
QC
(
spikes,
blanks,
duplicates,
etc.)
EPA
method
8080*
0.12
ppb
Not
specified
Detailed
information
not
provided.
Some
indication
that
spikes
and
duplicate
samples
were
collected.
*
Method
indicated
in
summary
report;
not
included
in
monitoring
report.
12
e.
Summary
A
single
sample
was
collected
after
treatment
of
greens
at
the
Minikada
club;
PCNB
was
not
detected.
PCA
was
not
analyzed
during
this
study.

While
the
Summary
Report
indicates
that
this
sample
represents
a
discrete
period
during
a
storm
event,
the
original
data
report
indicates
that
samples
were
composited,
but
does
not
indicate
over
what
time
period.
Since
base
flow
was
negligible
at
all
sites,
except
at
the
Minikada
golf
course,
the
issue
of
sample
compositing
is
especially
important.
The
description
of
the
Minikada
Club
in
the
monitoring
data
report
indicates:
"
The
1168
acre
watershed
above
the
course
delivered
over
564
acre­
feet
of
water
to
the
course,
over
half
during
non­
rainfall
periods.
The
watershed
contains
a
wetland
complex
upstream
of
the
course
which
stores
water
and
releases
it
after
stormwater
flows."
The
catchment
area
of
the
single
sampling
location
at
this
site
is
not
described
in
the
monitoring
data
report,
but
information
is
provided
in
the
Summary
Report.
The
spatial
relationship
between
the
monitoring
location
and
the
treated
greens
(
tees
and
fairways
were
not
treated
with
PCNB
by
the
time
of
sampling)
is
unclear.

Detectable
concentrations
of
some
fungicides
were
observed
in
this
study,
but
concentrations
were
very
low,
with
an
estimated
99.5
%
remaining
on
the
turf,
according
to
the
study
authors.
The
study
authors
concluded
that
"
the
low
pollutant
export
rate
appears
to
be
a
function
of
the
management
practices
used
on
the
golf
courses.
These
practices,
which
promote
rainfall
infiltration
and
limit
applications
to
chemicals,
particularly
phosphorous,
include
soil
aeration,
soil
fertility
testing,
application
of
organic
matter,
and
maintenance
of
dense
vegetation."
Three
of
the
four
courses
(
including
the
one
for
which
one
potentially
relevant
sample
was
collected)
had
applied
for
certification
as
members
of
the
Audubon
cooperative
sanctuary
system.
Because
of
this
the
study
authors
concluded
that
it
is
possible
that
"
the
courses
selected
for
study
are
superior
relative
to
the
management
practices
applied
to
them."
The
relevance
of
these
data
for
national
assessment
is
questionable.

Uncertainties:

The
single
sampling
location
was
not
described.
The
compositing
was
not
fully
described.
Recovery
data
were
not
provided.
The
relationship
between
the
areas
treated
and
the
sampling
location
is
unclear.
The
application
rate
is
also
unclear.

5.
Padilla
Bay
a.
Reports
provided:

i.
Potential
for
agricultural
pesticide
runoff
to
a
Puget
sound
estuary:
Padilla
bay,
Washington,
J.
R.
Mayer
and
N.
R.
Elkins,
Bull.
Env.
Contam.
Toxicol
(
1990)
45:
215­
222
13
b.
Pesticide
use
information
PCNB
use
information
indicates
that
it
is
applied
"
to
farmlands."
Cropland
and
pasture
comprise
72%
of
land
use
in
the
Padilla
bay
watershed
(
forest
20%,
urban
areas
8%).
Fourteen
different
pesticides
are
listed
as
applied
in
the
watershed
to
farmlands
"
early
each
summer
(
mid­
June)."

Table
11:
Pesticide
Usage
Data
for
PCNB
use
on
farmlands
in
Padilla
Bay
watershed.
Application
date
Rate
Areas
treated
Not
specified.
Monitoring
report
indicates
application
was
in
"
summer"
1­
18
lbs
ai/
A
Farmland
locations
are
not
specified.
Specific
crops
to
which
PCNB
was
applied
are
not
specified.

c.
Sampling
design
The
purposes
of
this
study
were
to
identify
and
quantify
pesticide
runoff
in
the
agricultural
environment
and
to
assess
the
ecological
significance
of
pesticide
runoff
into
the
bay,
specifically
to
an
eelgrass
community
of
Puget
Sound.
Sampling
stations
were
located
on
Padilla
Bay
(
estuarine)
and
on
three
sloughs
which
discharge
into
the
bay
and
which
transport
the
bulk
of
runoff
into
the
bay.
This
was
a
two­
year
investigation.
Cropland
and
pasture
comprise
72%
of
land
use
in
the
Padilla
bay
watershed
(
forest
20%,
urban
areas
8%).
Sampling
occurred
at
17
stations
(
11
surface
water,
4
estuarine,
2
sediment).
Stations
were
sampled
four
times;
two
of
the
four
sampling
events
were
characterized
as
pre­
pesticide
application
(
May
and
June
1997,
and
April
1998);
these
are
not
included
in
the
table
below.
Samples
represent
storm
events.

Table
12:
Summary
of
surface
water
monitoring
data
(
freshwater
only)
from
sites
in
Padilla
Bay
watershed.
Relevance
to
PCNB
(
agricultural
use)
is
unclear.
Sampling
date
Number
of
locations
sampled
(
one
sample
collected
per
location)
Detections
Description
(
stormflow
versus
composite
sample;
is
runoff
from
treated
areas?)
6/
22/
1997;
6/
29/
1996;
7/
7/
1997
(
post­
application)
11
0
Large
storm
on
6/
21/
1997
8/
7/
1998;
8/
9/
1998
(
postapplication
11
0
Two
"
minor
rainfall
events"
8/
5/
1998
and
8/
6/
1998
PCNB
was
not
detected.
It
was
reported
that
in
1997,
dicamba
was
detected
in
water
samples
at
all
locations
and
2,4­
D
was
found
in
9
of
11
slough
samples.
No
pesticides
were
detected
in
1998,
which
was
described
as
a
"
dry"
year.

d.
Sample
analysis
Only
PCNB
was
analyzed;
the
PCNB
degradate
PCA
was
not
included
as
an
analyte.
PCNB
was
not
detected.
14
Table
13:
Summary
information
on
analytical
methods
and
QA/
QC
used
for
surface
water
samples
collected
from
sites
in
Padilla
Bay
watershed.
Analytical
method
Quantitation
limit,
Detection
limit
Percent
recovery
for
compound
of
interest
Description
of
QA/
QC
(
spikes,
blanks,
duplicates,
etc.)
EPA
method
508
0.01
ppb
Not
specified
Duplicate
samples
were
collected.

e.
Summary
The
purposes
of
this
study
were
to
identify
and
quantify
pesticide
runoff
in
the
agricultural
environment
and
to
assess
the
ecological
significance
of
pesticide
runoff
into
the
bay,
specifically
to
an
eelgrass
community
of
Puget
Sound.

PCNB
use
was
indicated
as
agricultural.
The
timing
and
location
of
PCNB
applications
were
not
provided.
PCNB
was
not
detected.
PCA
was
not
analyzed
during
this
study.

Uncertainties:

The
spatial
relationship
between
the
monitoring
location
and
the
application
sites
is
not
provided.
Timing
of
PCNB
application
is
not
specified.

II.
Overall
Conclusions
by
EFED
OPP
has
thoroughly
reviewed
the
surface
water
monitoring
data
provided
and
concludes
that
only
one
of
the
studies
(
from
Hudson
National
Golf
Course)
provides
a
measure
of
useful
information
on
the
surface
water
quality
impact
of
PCNB
applications
to
turf
(
specifically
in
a
golf
course
setting).
One
other
study
provides
some
anecdotal
information,
but
the
three
remaining
studies
do
not
add
any
information
relevant
to
impacts
of
PCNB
use
on
surface
water
quality.

Four
of
these
five
studies
were
designed
to
assess
impacts
from
golf
course
application
of
a
wide
variety
of
pesticides.
The
fifth
was
designed
to
evaluate
the
impact
of
a
large
number
of
agricultural
pesticides
and
nutrients
on
surface
water
and
estuarine
water.
As
a
result
of
the
broad
focus
on
multiple
stressors
and
the
somewhat
unusual
timing
of
PCNB
application
(
applied
in
late
fall
rather
than
the
spring/
summer)
most
samples
collected
in
these
studies
occurred
during
periods
of
time
when
PCNB
was
not
used.
That,
coupled
with
usage
information
which
is
severely
lacking
in
detail,
makes
the
majority
of
the
sampling
not
relevant
for
assessing
impacts
on
surface
water
quality
from
PCNB
use.

Although
some
of
these
monitoring
studies
collected
samples
for
multiple
years,
PCNB
use
was
limited
to
only
one
or
two
years.
When
it
was
used
on
golf
courses,
it
appears
to
have
been
applied
to
greens
and
tees
only
(
with
the
exception
of
one
location
which
used
PCNB
on
fairways,
but
sampling
occurred
prior
to
that
application).
A
15
multi­
year
study
design
would
be
needed
to
provide
information
on
long­
term
exposure
to
PCNB,
thus
the
duration
of
the
studies
which
form
the
basis
of
the
Summary
Report
could
only
provide
data
relevant
for
assessing
acute
exposure
as
a
result
of
the
study
design.

The
level
of
QA/
QC
reported
in
the
Summary
Report
and
in
individual
monitoring
reports
is
weak.
Method
recovery
information
is
provided
for
only
one
study,
and
the
investigators
in
that
study
indicated
that
PCNB
data
were
somewhat
uncertain.
PCA
results
are
essentially
not
provided.

III.
Conclusions
from
the
Summary
Report
(
by
the
consultant)
and
OPP's
Response
Summary
Report
Conclusion
 
"
an
overwhelming
majority
of
the
hundreds
of
results
from
the
monitoring
studies
discussed
herein
were
"
ND".

OPP
Response
From
analysis
of
the
monitoring
reports
described
above,
only
a
small
number
of
samples
were
collected
which
represent
monitoring
following
application
of
PCNB
(
approximately
30
samples).
The
majority
of
the
analyses
reported
as
"
ND"
were
for
samples
collected
when
PCNB
was
not
used
(
the
Hudson
National
study);
in
other
cases
data
were
not
provided
to
confirm
that
application
occurred
prior
to
sampling
at
a
location
which
would
intercept
runoff
from
a
treated
area.
The
data
provided
do
not
support
the
conclusion
in
the
Summary
Report.

Summary
Report
Conclusion
 
When
PCNB
or
PCA
were
detected,
concentrations
were
almost
always
below
5
ppb,
and
usually
below
1
ppb.
The
peak
concentrations
in
the
draft
EFED
chapter
are
typically
one
to
two
orders
of
magnitude
higher
or
more
for
similar
exposure
periods.

OPP
Response
At
the
Hudson
National
course,
concentrations
ranged
from
trace
to
6.4
ppb
in
samples
collected
following
application.
These
values
are
in
the
range
of
model
estimates
(
refined
to
account
treatment
of
tees
and
greens
only)
when
used
at
the
maximum
label
application
rate
(
32.7
lb
ai/
A
applied
twice
at
a
7­
day
interval).
OPP
modeling
estimated
a
1­
in­
10
year
peak
concentration
of
13.4
ppb
for
the
index
reservoir
(
drinking
water
assessment)
and
5.0
ppb
for
the
standard
farm
pond
(
ecological
exposure
assessment).
Thus,
OPP
model
predictions
were
in
the
same
range
as
the
monitoring
at
this
site,
event
though
the
OPP
modeling
assumed
a
higher
PCNB
application
rate
than
reported
for
this
course.
16
Summary
Report
Conclusion
 
Many
of
the
measurement
points
in
the
summary
are
"
close
to
being
edge­
of­
field
waters,
thus
representing
greater
concentrations
than
would
be
obtained
from
the
index
reservoirs."

OPP
Response
None
of
the
monitoring
sites
can
be
conclusively
stated
to
represent
edge­
of­
field
concentrations,
based
on
the
information
provided.
Hudson
National
data
come
the
closest
to
this
characterization,
but
the
edge­
of­
field
descriptor
also
implies
that
the
field
is
treated.
Spatial
and
temporal
usage
information
is
not
provided
at
this
scale
to
support
this
statement.

Summary
Report
Conclusion
 
Eight
studies
were
done
in
6
states
in
several
regions.
PCNB
use
is
documented
in
7
of
the
8
studies;
five
provide
data
on
runoff.

OPP
Response
The
five
studies
provided
on
runoff
were
conducted
in
five
different
states.
PCNB
use
is
only
documented
in
two
of
the
five
studies
at
a
level
that
is
greater
than
anecdotal.

Summary
Report
Conclusion
 
Six
of
the
8
studies
are
multi­
year
studies.

OPP
Response
While
the
studies
may
be
conducted
over
multiple
years,
PCNB
use
was
only
for
a
single
year
in
two
of
the
five
studies,
and
for
two
years
in
another.
In
the
two
remaining
studies,
the
years
of
use
are
not
provided.
These
data
are
not
adequate
as
a
basis
for
estimating
long
term
impacts
on
water
quality
of
PCNB
use
resulting
from
annual
use.

Summary
Report
Conclusion
 
"
It
cannot
be
said
that
these
studies
are
mere
snapshots
in
time
of
a
highly
localized
situation
where
PCNB
use
was
not
documented."

OPP
Response
OPP
concludes
that,
based
on
a
detailed
analysis
of
the
monitoring
studies,
four
of
five
of
these
studies
are
snapshots
in
time
of
a
highly
localized
situation
where
17
PCNB
use
was
not
documented.
The
Hudson
National
study
did
document
use
better,
but
represents
a
snapshot
of
four
sampling
events
associated
with
two
applications
of
PCNB
based
on
a
spatially
limited
monitoring
dataset.

Summary
Report
Conclusion
 
The
monitoring
studies
combine
in
a
weight­
of­
evidence
approach
to
raise
significant
questions
about
the
EPA
concentration
predictions.

OPP
Response
OPP
modeling
assumed
that
100%
turf
was
treated
at
the
maximum
rate
and
frequency
(
two
applications
of
32.67
lb
ai/
A
at
a
7­
day
interval)
allowed
on
the
label
(
excluding
the
one
higher,
geographically
restricted
rate
allowed
in
CA,
OR
and
WA).
The
concentrations
estimated
by
modeling
represent
runoff
from
these
treated
fields
into
a
reservoir
or
small
pond
from
all
precipitation
events
that
occur
over
a
30­
year
period,
including
the
first
events
that
occur
after
treatment.
The
model
output
values
were
then
refined
to
account
for
partial
land
area
treatment
(
i.
e.,
for
application
only
to
specific
parts
of
a
golf
course,
such
as
tees,
greens,
and
possibly
fairways).
The
data
collected
in
this
monitoring
study
indicate
that
PCNB
can
be
transported
off­
site
into
adjacent
surface
water
bodies.
The
peak
1­
in­
10­
year
model
estimate
for
the
reservoir
(
13.4
ppb),
when
adjusted
for
application
to
tees
and
greens,
is
within
a
factor
of
two
of
the
peak
value
(
6.4
ppb)
measured
in
one
of
two
years
at
the
Hudson
National
site.
The
peak
1­
in­
10­
year
model
estimate
for
the
standard
pond
(
5.0
ppb),
when
adjusted
for
application
to
tees
and
greens,
is
approximately
22%
lower
than
the
peak
value
measured
in
one
of
two
years
at
the
Hudson
National
site.
However,
in
general,
given
the
lack
of
spatially
explicit
data
on
the
areas
treated
and
the
temporal
disconnect
between
the
timing
of
application
and
precipitation
events,
one
cannot
directly
compare
OPP's
surface
water
predictions
with
these
monitoring
data.

The
monitoring
data,
taken
as
a
whole,
do
not
add
to
the
weight­
of­
evidence
to
discount
the
model
estimates.
