Page
1
of
8
Memorandum
From:
Lynn
Zipf
USEPA/
OW/
OST
ph:
(
202)
566
1429
zipf.
lynn@
epa.
gov
To:
304(
m)
Record
(
EPA
Docket
Number
OW­
2003­
0074)

Date:
December
03,
2003
Re:
Revisions
to
EAD's
Toxic
Weighting
Factor
Methodology
parameters
The
purpose
of
this
memorandum
is
to
detail
the
revisions
to
the
Toxic
Weighting
Factor
(
TWFs)
methodology
as
well
as
set
forth
which
TWFs
will
be
revised
for
the
final
Effluent
Guidelines
Program
Plan
for
2004/
2005.

The
TWFs
are
intended
for
use
in
the
cost­
effectiveness
(
CE)
analysis
of
effluent
guidelines.
TWFs
account
for
differences
in
toxicity
among
the
pollutants
of
concern
and
provide
the
means
to
compare
mass
loadings
of
different
pollutants
on
the
basis
of
their
toxic
potential.
For
example,
a
mass
loading
of
a
pollutant
in
pounds
per
year
(
lb/
yr)
may
be
multiplied
by
a
pollutant­
specific
weighting
factor
to
derive
a
"
toxic­
equivalent"
loading
(
lb­
equivalent/
yr).
The
cost­
effectiveness
(
in
dollars
per
pound­
equivalent
removed)
of
implementing
various
treatment
options
may
be
compared
by
summing
mass
load
reductions
on
a
toxic­
equivalent
basis
and
dividing
the
sum
into
the
total
estimated
cost.

TWFs
are
derived
from
chronic
aquatic
life
criteria
(
or
toxic
effect
levels)
and
human
health
criteria
(
or
toxic
effect
levels)
established
for
the
consumption
of
fish.
For
carcinogenic
substances,
EPA
sets
the
human
health
risk
level
at
10­
5
(
i.
e.,
protective
to
a
level
allowing
1
in
100,000
excess
lifetime
cancer
cases
over
background).
In
the
TWF
method
for
assessing
waterbased
effects,
these
toxicity
levels
of
pollutants
of
concern
are
compared
to
a
benchmark
value
that
represents
the
toxicity
level
of
a
specified
pollutant.
EPA
selected
copper,
a
toxic
metal
commonly
detected
and
removed
from
industrial
effluent,
as
the
benchmark
pollutant.
EPA
used
copper
in
previous
TWF
calculations
for
the
cost­
effectiveness
analysis
of
effluent
guidelines.
Although
EPA
revised
the
water
quality
criterion
for
copper
in
1998
(
to
9.0
micrograms
per
liter
[

g/
L]),
the
TWF
method
uses
the
former
criterion
(
5.6

g/
L)
to
facilitate
comparisons
with
costeffectiveness
values
calculated
for
other
regulations.
The
former
criterion
for
copper
(
5.6

g/
L)
was
reported
in
the
1980
Ambient
Water
Quality
Criteria
for
Copper
document
(
U.
S.
EPA,
1980).

To
calculate
TWF
values,
EPA
adds
TWFs
for
aquatic
life
effects
and
for
human
health
effects
for
each
pollutant
of
concern.
EPA
uses
chronic
effects
on
aquatic
life
and
human
health
Page
2
of
8
TWF

5.6
AQ

5.6
HHOO
effects
from
ingesting
contaminated
organisms
(
HHOO)
as
the
basis
for
TWFs.
The
calculation
is
performed
by
dividing
aquatic
life
and
human
health
criteria
(
or
toxic
effect
levels)
for
each
pollutant,
expressed
as
a
concentration
in
micrograms
per
liter
(

g/
L),
into
the
former
copper
criterion
of
5.6

g/
L:

where:
TWF
=
toxic
weighting
factor
AQ
=
chronic
aquatic
life
value
(

g/
L)
HHOO
=
human
health
(
ingesting
organisms
only)
value
(

g/
L)

Human
Health
Values
EPA
addresses
potential
human
health
toxicity
for
TWFs
using
the
HHOO,
human
health
(
ingesting
organisms
only)
criterion
or
toxic
effect
level.
For
PWFs,
the
HHWO,
human
health
(
ingesting
water
and
organisms)
criterion
or
toxic
effect
level
is
used.
EPA
uses
the
following
hierarchy
to
determine
human
health
values,
in
order
of
priority:

1.
Calculated
human
health
criteria
using
EPA's
Integrated
Risk
Information
System
(
IRIS)
(
U.
S.
EPA,
1998/
1999e)
oral
reference
doses
(
RfDs)
or
oral
cancer
potency
slope
factors
(
SFs)
in
conjunction
with
adjusted
3
percent
lipid
bioconcentration
factor
(
BCF)
values
derived
from
Ambient
Water
Quality
Criteria
Documents
(
U.
S.
EPA,
1980).
Three
percent
is
the
mean
lipid
content
of
fish
tissue
reported
in
the
study
from
which
the
average
daily
fish
consumption
rate
of
6.5
grams
per
day
(
g/
day)
was
derived
(
U.
S.
EPA,
1991).

2.
Calculated
human
health
values
using
current
IRIS
RfDs
or
SFs
and
representative
unadjusted
BCF
values
for
common
North
American
species
of
fish
or
invertebrates
or
estimated
BCF
values.

3.
Calculated
human
health
values
using
RfDs
or
SFs
from
EPA's
Health
Effects
Assessment
Summary
Tables
(
HEAST)
(
U.
S.
EPA,
1997)
or
EPA's
Region
III
Risk­
Based
Concentration
(
RBC)
Table
(
U.
S.
EPA,
1998)
used
in
conjunction
with
adjusted
3
percent
lipid
BCF
values
derived
from
Ambient
Water
Quality
Criteria
Documents.

4.
Calculated
human
health
criteria
using
current
RfDs
or
SFs
from
HEAST
or
EPA's
Region
III
RBC
table
and
representative
BCF
values
for
common
North
American
species
of
fish
or
invertebrates
or
estimated
BCF
values.
Page
3
of
8
HHOO
(
µ
g/
L)

RfD
(
mg/
kg/
d)
x
70
kg
x
1,000
µ
g/
mg
0.0065
kg/
d
x
BCF
(
L/
kg)

HHOO
(
µ
g/
L)

70
kg
x
Risk
Level
(
dimensionless)
x
1,000
µ
g/
mg
SF
(
mg/
kg/
d)

1
x
0.0065
kg/
d
x
BCF
(
L/
kg)

HHWO
(
µ
g/
L)

RfD
(
mg/
kg/
d)
x
70
kg
x
1,000
µ
g/
mg
2
L/
d

[
0.0065
kg/
d
x
BCF
(
L/
kg)]

HHWO
(
µ
g/
L)

70
kg
x
Risk
Level
(
dimensionless)
x
1,000
µ
g/
mg
SF
(
mg/
kg/
d)

1
x
[
2
L/
d

[
0.0065
kg/
d
x
BCF
(
L/
kg)]]
5.
Criteria
from
the
Ambient
Water
Quality
Criteria
Documents.

6.
Calculated
human
health
values
using
RfDs
or
SFs
from
data
sources
other
than
IRIS,
HEAST,
or
the
Region
III
RBC
Table.

This
hierarchy
is
based
on
Section
2.4.6
of
the
Technical
Support
Document
for
Water
Quality­
based
Toxics
Control
(
U.
S.
EPA,
1991),
which
recommends
that
the
most
current
risk
information
from
IRIS
be
used
when
estimating
human
health
risks.
This
document
also
recommends
using
an
average
daily
fish
consumption
rate
of
6.5
grams,
an
average
daily
water
intake
of
2
liters,
and
an
average
adult
body
weight
of
70
kilograms.
In
cases
where
a
chemical
has
both
an
RfD
and
SF
from
sources
at
the
same
level
of
the
hierarchy,
the
human
health
values
are
calculated
using
the
SF,
which
always
results
in
the
more
stringent
value
of
the
two.
When
a
chemical
has
both
an
RfD
and
SF
but
these
values
are
from
different
levels
of
the
hierarchy,
the
value
that
is
from
the
source
higher
on
the
hierarchy
is
used.
The
carcinogenic
risk
level
is
10­
5
for
TWFs,
whereas
a
risk
level
of
10­
6
is
used
for
PWFs.
The
following
equations
are
used
to
calculate
human
health
values:

For
Toxicity
Protection
(
ingestion
of
organisms
only)

For
Carcinogenicity
Protection
(
ingestion
of
organisms
only)

For
Toxicity
Protection
(
ingestion
of
water
and
organisms)

For
Carcinogenicity
Protection
(
ingestion
of
water
and
organisms)

Revisions
for
the
Human
Health
TWF
Methodology
Page
4
of
8
The
following
revisions
need
to
be
made
to
the
TWF
methodology.
The
TWF
methodology
needs
to
reflect
the
changes
that
were
made
in
the
Methodology
for
Deriving
Ambient
Water
Quality
Criteria
for
the
Protection
of
Human
Health
(
EPA­
822­
B­
00­
004)
as
well
as
those
proposed
in
the
Draft
Final
Guidelines
for
Carcinogen
Risk
Assessment
(
External
Review
Draft,
February
2003).
Specifically,
but
not
limited
to:

°
Average
daily
fish
consumption
rate
of
6.5
grams
needs
to
reflect
updated
information
in
Methodology
for
Deriving
Ambient
Water
Quality
Criteria
for
the
Protection
of
Human
Health
(
EPA­
822­
B­
00­
004).
EPA
recommends
a
default
fish
intake
rater
of
17.5
grams/
day
to
adequately
protect
the
general
population
of
fish
consumers.

°
Update
BCFs
to
BAFs.
In
order
to
prevent
harmful
exposures
to
waterborne
chemicals
through
the
consumption
of
contaminated
fish
and
shellfish,
national
304(
a)
criteria
for
the
protection
of
human
health
must
address
the
process
of
bioaccumulation
in
aquatic
organisms.
For
deriving
national
304(
a)
criteria
to
protect
human
health,
EPA
accounts
for
potential
bioaccumulation
of
chemicals
in
fish
and
shellfish
through
the
use
of
national
bioaccumulation
factors
(
BAFs).
A
national
BAF
is
a
ratio
(
in
L/
kg)
that
relates
the
concentration
of
a
chemical
in
water
to
its
expected
concentration
in
commonly
consumed
aquatic
organisms
in
a
specified
trophic
level.
For
more
information
on
how
to
derive
BAFs,
see
Methodology
for
Deriving
Ambient
Water
Quality
Criteria
for
the
Protection
of
Human
Health
(
EPA­
822­
B­
00­
004).

°
Update
terminology
and
methodology
for
carcinogens.
See
Methodology
for
Deriving
Ambient
Water
Quality
Criteria
for
the
Protection
of
Human
Health
(
EPA­
822­
B­
00­
004)
and
Draft
Final
Guidelines
for
Carcinogen
Risk
Assessment
(
EPA
2003).

°
Update
Rfds/
cancer
potency
information
as
new
data
becomes
available
in
IRIS
or
HEAST.

Chronic
Aquatic
Life
Values
When
selecting
chronic
aquatic
toxicity
values,
EPA
uses
national
water
quality
criteria,
when
available.
When
these
criteria
are
not
available,
other
values
representative
of
the
chemical's
chronic
toxicity
are
used.
EPA
uses
the
following
hierarchy
to
select
the
appropriate
chronic
values:

1.
National
chronic
freshwater
quality
criteria
2.
Lowest
reported
measured
maximum
allowable
toxicant
concentration
(
MATC),
lowest­
observed­
effect
concentration
(
LOEC),
or
no­
observed­
effect
concentration
(
NOEC)

3.
Lowest
reported
measured
chronic
growth
or
reproductive
toxicity
test
concentration
Page
5
of
8
4.
Estimated
chronic
toxicity
concentration
from
a
measured
acute:
chronic
ratio
for
a
less
sensitive
species,
quantitative
structure­
activity
relationship
(
QSAR)
model,
or
default
acute:
chronic
ratio
of
10:
1
National
Chronic
Freshwater
Quality
Criteria
National
chronic
water
quality
criteria
are
the
first
choice
for
values
because
they
represent
a
consideration
of
a
chemical's
toxicity
to
a
diverse
genera
of
aquatic
life
and
have
been
published
by
EPA.
The
derivation
of
EPA
criteria
values
is
described
in
EPA
Office
of
Water's
criteria
documents
for
specific
pollutants
(
U.
S.
EPA,
1980).
"
Criteria"
is
defined
as
the
4­
day
average
concentration
of
toxicants
at
which
a
diverse
genera
of
aquatic
organisms
and
their
uses
should
not
be
unacceptably
affected,
provided
that
these
levels
are
not
exceeded
more
than
once
every
3
years.

Lowest
Reported
MATC,
LOEC,
or
NOEC
Concentration
The
term
"
chronic"
involves
a
stimulus
that
continues
for
a
long
time,
often
for
periods
of
several
weeks
to
years,
depending
on
the
organism's
reproductive
life
cycle.
Chronic
aquatic
tests
measure
the
effects
of
long­
term
exposure
to
a
chemical.
The
biological
response
to
the
exposure
is
typically
of
relatively
slow
progress
and
long
continuance.
Citing
rapid
developments
in
test
methodology,
EPA
recommends
several
7­
day,
short­
term
period
exposure
duration
test
methods
(
U.
S.
EPA,
1989a).
Test
endpoints
include
such
variables
as
survival
percentage,
hatchability,
and
normal
larvae
weight
and
length.
Chronic
tests
of
longer
exposure
duration
measure
endpoints
such
as
growth
and
reproduction.

EPA
uses
chronic
aquatic
test
data
to
identify
three
concentration
levels
of
potential
significance:
the
no­
observed­
effect
concentration
(
NOEC),
the
lowest­
observed­
effect
concentration
(
LOEC),
and
the
maximum
allowable
toxicant
concentration
(
MATC).
The
NOEC
is
the
highest
toxicant
concentration
to
which
test
organisms
have
been
exposed
with
results
of
no­
observed
adverse
effect.
The
NOEC
may
be
statistically
determined
using
hypothesis
testing,
or
it
may
be
derived
from
the
inhibition
concentration,
which
is
an
estimate
of
the
toxicant
concentration
that
will
result
in
a
given
percentage
reduction
in
biological
measurement
of
the
test
organisms.
The
LOEC
is
the
lowest
toxicant
concentration
at
which
a
chronic
effect
on
a
test
organism
has
been
observed.
The
MATC
is
the
geometric
mean
of
the
NOEC
and
LOEC
and
is
meant
to
represent
the
threshold
level
where
chronic
effects
will
begin
to
occur.
MATC
values
are
selected
first,
followed
by
LOEC
values,
and
lastly
by
NOEC
values.

Lowest
Chronic
or
Reproductive
Test
Concentration
For
chemicals
that
do
not
have
chronic
aquatic
life
criteria,
MATCs,
LOECs,
or
NOECs,
EPA
obtains
chronic
effect
concentrations
from
readily
available
sources
of
chronic
toxicity
test
data.
The
preferred
information
source
is
the
EPA's
Assessment
Tools
for
the
Evaluation
of
Risk
Page
6
of
8
(
ASTER)
(
U.
S.
EPA,
1998/
1999a),
which
combines
the
Aquatic
Toxicity
Information
Retrieval
database
(
AQUIRE)
(
U.
S.
EPA,
1998/
1999b),
the
EPA
Environmental
Research
Laboratory­
Duluth
(
ERL­
Duluth)
fathead
minnow
database
(
U.
S.
EPA,
1998/
1999c),
and
tables
of
toxicity
test
results
from
water
quality
criteria
documents.
The
ASTER
system
differentiates
between
AQUIRE
test
data
that
are
likely
to
be
of
good
quality
and
AQUIRE
test
data
that
are
of
unknown
quality,
according
to
the
following
criteria:

°
Test
pH
within
range
of
6.5
­
8.5
°
Review
code
1
(
methodology
section
cites
published
or
well­
documented
procedures;
satisfactory
control;
measured
concentration;
temperature,
pH,
dissolved
oxygen,
and
hardness
are
reported)
or
2
(
one
or
more
of
the
following
may
occur:
control
mortality
not
reported;
no
solvent
control
when
a
solvent
is
used
in
the
test;
unmeasured
concentration;
water
chemistry
variables
not
reported
or
incomplete)

°
No
use
of
formulations
or
carriers
°
Measured
values
and
flow­
through
exposure
only
for
tests
on
fish
(
no
static
exposure)

°
Measured
values
only
for
invertebrates
or
plants
(
exposure
may
be
static
or
flow­
through)

Test
results
from
the
ERL­
Duluth
fathead
minnow
database
are
assumed
to
be
of
good
quality.
However,
test
results
reported
in
water
quality
criteria
documents
are
assumed
to
be
of
unknown
quality.

EPA
selects
the
lowest
reported
concentration
 
from
a
chronic
growth
or
reproductive
test
on
a
North
American
native
fish
or
invertebrate
or
from
a
biologically
significant
(
i.
e.,
chlorophyll
production)
EC
50
test
for
an
algal
species
 
from
the
pool
of
test
data
likely
to
be
of
good
quality
or,
alternatively,
from
the
pool
of
data
of
unknown
quality.
If
appropriate
test
data
are
not
available
from
ASTER,
other
primary
or
secondary
information
sources
are
consulted.

Estimated
Chronic
Toxicity
Concentration
EPA
uses
estimated
chronic
toxicity
concentrations
when
measured
values
are
unavailable.
The
first
option
for
estimating
a
chronic
toxicity
concentration
is
to
use
acute
toxicity
concentrations
and
a
measured
acute:
chronic
ratio
(
ACR).
ACRs
are
based
on
measured
acute
and
chronic
pollutant
concentration
values
for
the
same
species.
The
calculated
ACR
is
applied
to
the
acute
aquatic
toxicity
criterion
or
toxic
effect
level
selected
for
the
pollutant
of
concern.
EPA
uses
this
method
in
instances
where
an
ACR
is
available
for
a
species
that
has
a
Page
7
of
8
measured
chronic
toxicity
concentration
that
is
greater
than
the
acute
criterion
or
the
representative
acute
toxic
effect
level
for
the
selected
pollutant.
These
instances
arise
when
chronic
toxicity
test
data
are
available
for
less
sensitive
species
only.
The
acute
aquatic
toxic
effect
level
(
used
if
national
acute
water
quality
criteria
are
not
available)
is
typically
the
lowest
reported
acute
aquatic
bioassay
test
concentration
(
24­
to
96­
hour
median
lethal
concentration
(
LC
50))
for
a
North
American
resident
species
of
fish
or
invertebrate.
As
with
chronic
toxic
effect
levels,
a
test
result
of
good
quality
is
selected
ahead
of
a
test
result
of
unknown
quality.

The
second
option
for
estimating
a
chronic
toxicity
test
concentration
is
to
use
ERLDuluth's
QSAR
model
(
U.
S.
EPA,
1998/
1999d).
QSAR
derives
statistically
based
relationships
between
physical­
chemical
properties
and
biological
activity.
The
QSAR
model
uses
measured
toxicity
test
results
for
compounds
with
similar
chemical
structures
and
properties
to
estimate
MATC
values
for
compounds
whose
chemical
structure
and
properties
are
known
or
may
be
estimated.

The
final
option
for
estimating
a
chronic
toxicity
concentration
is
to
apply
an
assumed
ACR
of
10:
1
to
the
acute
aquatic
toxic
effect
concentration.
The
ACR
of
10:
1
is
based
on
a
recommendation
in
EPA
Office
of
Water's
Technical
Support
Document
for
Water
Quality­
based
Toxics
Control
(
U.
S.
EPA,
1991)
for
estimating
chronic
toxicity
when
no
data
are
available.
The
recommendation
assumes
that
the
chronic
toxicity
value
is
10
times
lower
than
the
acute
value.

Revisions
for
the
Aquatic
Life
TWF
Methodology
EPA's
Office
of
Science
and
Technology
is
currently
forming
an
Aquatic
Life
Guidelines
Revisions
Workgroup
of
Agency
scientist
to
identify,
review,
evaluate,
and
revise
the
existing
Guidelines
for
Deriving
Water
Quality
Criteria
for
the
Protection
of
Aquatic
Life
(
1985).
The
revisions
identified
by
this
workgroup
should
be
incorporated
in
the
TWF
methodology
for
aquatic
life.

List
of
TWFs
that
may
be
reviewed,
developed
and/
or
updated
prior
to
final
plan:

(
1)
Nitric
acid
(
7697372)
ranks
5th
among
all
TRI
compounds
for
TRI
hazard
and
9th
among
all
TRI
compounds
for
TRI
pounds.
(
These
rankings
are
based
on
RSEI
model
output
for
water
discharges.)

(
2)
Hydrogen
fluoride
(
7664393)
ranks
18th
for
TRI
hazard
and
35th
for
TRI
pounds.

(
3)
Nitrate
compounds
(
NA)
ranks
19th
for
TRI
hazard
and
1st
for
TRI
pounds.

(
4)
Mercury
compounds
(
NA)
ranks
59th
for
TRI
hazard
and
174th
for
TRI
pounds.
(
Per
the
2/
24/
03
memorandum
from
Susan
Keane,
ABT
to
Lynn
Zipf,
EPA,
methylmercury
rather
than
mercury
may
be
the
most
appropriate
representative
compound.
Although
an
EAD
TWF
is
available
for
mercury,
a
TWF
is
not
available
for
methylmercury.)
Page
8
of
8
(
5)
Chlorine
dioxide
(
10049044)
ranks
74th
for
TRI
hazard
and
47th
for
TRI
pounds.

(
6)
1,3­
Phenylenediamine
(
108452)
ranks
81st
for
TRI
hazard
and
93rd
for
TRI
pounds.

(
7)
Mercury
(
7439976)
ranks
96th
for
TRI
hazard
and
223rd
for
TRI
pounds.
(
See
also
(
4),
mercury
compounds
above.)

(
8)
Review
sodium
nitrite
TWF.

(
9)
Review
benzo(
a)
pyrene
TWF.

(
10)
PACs.

(
11)
Review
TWF
for
aniline.

(
12)
Develop
TWF
for
vanadium.

For
those
chemicals
that
have
RfDs
or
cancer
potency
information
available
in
IRIS
or
HEAST
but
do
not
have
TWFs,
EPA
may
develop
TWFs
as
feasible.

References:

Draft
Final
Guidelines
for
Carcinogen
Risk
Assessment
(
External
Review
Draft,
February
2003).
U.
S.
Environmental
Protection
Agency,
Risk
Assessment
Forum,
Washington,
DC,
2003.

Methodology
for
Deriving
Ambient
Water
Quality
Criteria
for
the
Protection
of
Human
Health
U.
S.
Environmental
Protection
Agency,
Office
of
Water,
Washington,
DC,
October
2003.
