SBREFA
PANEL
SUPPORT
FOR
COOLING
WATER
PHASE
III
REGULULATIONS
UNDER
SECTION
316(
B)
OF
THE
CLEAN
WATER
ACT
For
the
United
States
Small
Business
Administration
Office
of
Advocacy
FINAL
REPORT
Prepared
by:

Jonathan
Skolnik
Paul
Nguyen
Mary
Chou
Jack
Faucett
Associates
4550
Montgomery
Avenue
Suite
300
North
Bethesda,
MD
20814
301­
961­
8800
All
Comments
and
Questions
Should
be
Directed
to:
Jonathan
Skolnik,
Project
Director
Skolnik@
jfaucett.
com
January
9,
2004
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
1
­
TABLE
OF
CONTENTS
1.
EXECUTIVE
SUMMARY.................................................................................................
3
2.
PHASE
III
BACKGROUND
..............................................................................................
6
3.
INTRODUCTION
TO
THE
BENEFITS
ESTIMATE
METHODOLOGY..........................
8
3.1.
Analysis
of
EPA's
Methodology
for
Estimating
Benefits
in
Phase
II
...............................
9
3.2.
EPA's
Methodology
for
Calculating
Benefits
................................................................
10
3.3.
Analysis
of
Benefits
using
RUM
approach
....................................................................
15
3.4.
Analysis
of
Benefits
using
HRC
approach
.....................................................................
17
3.5.
Review
of
Case
Studies
.................................................................................................
18
3.5.1
Ohio
River
Case
Study.........................................................................................
18
3.5.2.
Brayton
Point
Case
Study
...................................................................................
20
3.5.3.
J.
R.
Whiting
Case
Study
.....................................................................................
22
3.6.
Summary
of
Benefit
Uncertainties.................................................................................
26
4.
REVIEW
AND
CRITIQUE
OF
PHASE
III
BENEFITS
ESTIMATES..............................
29
5.
REVIEW
AND
CRITIQUE
OF
COST
ESTIMATES
FOR
PHASE
III
............................
38
6.
PROPOSED
PHASE
III
REGULATORY
ALTERNATIVES
...........................................
45
6.1.
Current
Regulatory
Strategy
..........................................................................................
45
6.2.
Overview
of
Phase
III
Regulatory
Options
....................................................................
46
6.3.
Costs
and
Benefits
of
the
Three
Regulatory
Options......................................................
47
7.
CONCLUSION:
REGULATORY
OPTIONS
&
RECOMMENDATIONS
......................
66
LIST
OF
FIGURES
AND
TABLES
Table
1:
Phase
III
Timeline
........................................................................................................
7
Table
2:
Intake
Flow
(
MGD)
of
Phase
II
Case
Study
Facilities.................................................
10
Table
3:
Overview
of
EPA's
Methodology
of
Benefits
Valuation
............................................
11
Table
4:
Number
of
EPA
Case
Study
Facilities
Extrapolated
to
all
In­
Scope
Facilities
.............
13
Figure
1:
Production
Foregone
based
at
Ohio
River
Facilities
..................................................
14
Table
5:
Reduction
of
Impinged
Fish
Prior
to
and
After
Deterrent
Net
....................................
23
Table
5:
Estimate
of
Economic
Losses
of
I&
E
at
the
J.
R.
Whiting
...........................................
24
Table
6:
Estimate
of
Economic
Losses
of
I&
E
at
the
J.
R.
Whiting
Facility
using
the................
25
Mid­
Point
Average
with
the
Angling
Index...............................................................................
25
Table
7:
Summary
of
Analysis
and
Recommendations
.............................................................
26
Table
8:
Eight
Regions
New
Regions
for
the
Benefit
Analysis
.................................................
29
Table
9:
EPA
North
Atlantic
Region
Benefit
Estimate
.............................................................
34
Figure
2:
Salem
Case
Study
Benefits........................................................................................
35
Figure
3:
North
Atlantic
Region
Benefits
.................................................................................
35
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
2
­
Table
10:
EPA
North
Atlantic
Region
Benefits
Estimate..........................................................
37
Table
11:
Average
Annualized
Costs
per
Facility.....................................................................
41
Table
12:
Revised
Compliance
Costs
for
Phase
III
based
on
Higher
Cost
Estimates
in
Phase
II
NODA
..............................................................................................................................
43
Table
13:
Revised
Permitting
and
Monitoring
Costs.................................................................
44
Table
14:
Revised
Annual
Average
Annual
Costs
per
Facility..................................................
44
Table
15:
Regional
Benefits
As
Calculated
by
EPA
for
Phase
II
..............................................
48
Table
16:
Phase
III
Estimated
Commercial
and
Recreational
Benefits
by
Region.....................
49
Table
17:
Portion
of
Gallons
Withdraw
by
Facility
Size
Category............................................
50
Table
18:
Total
Estimated
Benefits
Scaled
By
Facility
Size
Categories
....................................
50
Figure
3:
MGD
to
GPM
Conversion
Formula
and
Example
.....................................................
52
Table
12:
Gallons
per
Minute
Conversion
by
Facility
Size
Category........................................
52
Table
18:
Compliance
Technology
Scenarios...........................................................................
53
Figure
4:
EPA
Initial
Capital
Cost
Equations
...........................................................................
54
Table
19:
Distribution
of
Compliance
Technologies.................................................................
55
Table
20:
Initial
Capital
Cost
Estimates
for
Regulatory
Option
1
.............................................
55
Figure
5:
EPA
Operating
and
Maintenance
(
O&
M)
Cost
Formulas
..........................................
56
Table
21:
O&
M
Cost
Estimates
for
Regulatory
Option
1..........................................................
57
Table
22:
Total
Average
Monitoring,
Record
Keeping,
&
Reporting
Costs
...............................
57
Per
Facility
(
Dollars).................................................................................................................
57
Table
23:
Regulatory
Option
One
Summary
of
Costs
&
Benefits
.............................................
59
Table
24:
Comparison
of
Costs
&
Benefits
of
Different
Thresholds
Under
Option
One............
59
Table
25:
Proportion
of
Benefits
Attributed
to
Impingement
....................................................
60
Table
26:
Estimated
Benefits
of
Regulatory
Option
Two..........................................................
61
Table
27:
Regulatory
Option
Two
Costs
..................................................................................
61
Table
28:
Regulatory
Option
Two
­
Summary
of
Cost
and
Benefits
.........................................
62
Table
29:
Regulatory
Option
Two
­
Costs
&
Benefits
of
Different
Thresholds
.........................
62
Table
30:
Regulatory
Option
Three
­
Estimated
Costs..............................................................
63
Table
31:
Estimated
Benefits
Changes
Resulting
from
Not
Monitoring
....................................
64
Table
32:
Cost
and
Benefits
of
Regulatory
Option
Three..........................................................
64
Table
33:
Regulatory
Option
Three
­
Benefits
Achieved
and
Facilities
Exempted
....................
65
Over
Different
Thresholds
........................................................................................................
65
Table
34:
Summary
of
Benefit­
Cost
Ratios
...............................................................................
66
Table
35:
Benefits
and
Facilities
Exemption
Comparisons
of
Each
Regulatory
Option.............
67
Table
36:
Cost­
Benefit
of
Preferred
Regulatory
Alternative
.....................................................
67
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
3
­
1.
Executive
Summary
The
purpose
of
this
study
is
to
provide
general
support
to
a
SBREFA
panel
to
be
convened
for
the
Phase
III
rulemaking
of
Section
316(
b)
of
the
Clean
Water
Act.
This
includes
an
analysis
of
the
methodologies
used
to
estimate
the
benefits
and
costs
of
the
regulation,
the
development
of
alternative
regulatory
strategies,
and
the
development
of
rough
estimates
of
the
benefits
and
costs
of
these
alternatives.

According
to
Environmental
Protection
Agency
(
EPA),
there
are
an
estimated
925
facilities
within
the
scope
of
Phase
III
regulations
and
785
of
those
facilities
are
expected
to
incur
costs
of
upgrading
to
technologies
that
reduce
impingement
and
entrainment
(
I&
E)
of
fish.
An
extensive
review
of
relevant
documents
developed
by
the
EPA
and
comments
submitted
by
industry
members
for
the
Phase
I
and
II
rulemaking
was
conducted
to
provide
the
foundation
for
estimating
the
economic
impact
of
Phase
III.
Study
staff
solicited
and
integrated
input
from
EPA
and
industry
experts
into
the
analysis.
These
activities
culminated
in
the
development
of
this
final
report,
which
serves
as
a
tool
for
SBREFA
panel
members
and
EPA
staff
in
measuring
potential
impacts
and
identifying
possible
regulatory
alternatives
to
the
proposed
Phase
III
regulations.

This
report
includes
an
in­
depth
review
of
the
methods
used
by
EPA
to
quantify
the
impacts
of
the
proposed
rule
followed
by
a
cost­
benefit
sensitivity
analysis
of
three
regulatory
options
for
Phase
III.
These
regulatory
alternatives
explore
the
potential
to
maximize
the
benefits
of
reducing
I&
E
while
limiting
financial
burden
on
smaller
entities.
The
three
options
are
briefly
summarized
below:

Regulatory
Option
1:
Raise
the
threshold
from
2
million
gallons
per
day
(
MGD)
design
intake
flow
(
DIF).

Regulatory
Option
2:
Facilities
with
DIF
less
than
50
MGD,
need
only
comply
with
the
impingement
reduction
requirements.

Regulatory
Option
3:
Facilities
with
DIF
<
50
MGD
need
only
meet
impingement
reduction
requirements
and
do
not
have
to
conduct
baseline
and
compliance
monitoring.

The
first
step
was
to
analyze
the
methodology
EPA
used
to
estimate
the
benefits
and
costs
of
the
regulation.
This
analysis
uncovered
a
number
of
areas
where
benefits
were
either
overestimated
or
uncertain
and
costs
were
overestimated
or
uncertain.
Some
of
the
major
areas
identified
in
the
cost
and
benefit
methodology
include
the
following:

 
EPA
extrapolated
national
baseline
losses
using
outdated
data
from
only
a
small
number
of
sites.
 
EPA
case
study
sites
were
not
randomly
selected.
 
EPA's
valuation
of
nonuse
benefits
make
up
of
over
97
percent
of
the
total
benefits
estimate
and
are
speculative.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
4
­
 
EPA
assumes
a
linear
relationship
between
the
amount
of
fish
saved
from
I&
E
and
increases
in
commercial
and
recreational
fish
harvests.
 
EPA
assumes
all
I&
E
fish
are
Age
One
Equivalents
(
the
age
when
fish
are
assessed
an
economic
value)
while
experts
contend
most
impinged
fish
are
"
young­
of­
the­
year"
fish
that
may
not
survive
until
year
one.
 
Two­
year
monitoring
program
will
cost
facilities
more
than
EPA
has
estimated.

EPA
has
reviewed
portions
of
this
study,
particularly
the
sections
critiquing
the
quality
of
the
case
study
data.

The
second
step
was
to
analyze
the
benefits
and
costs
of
the
alternatives
designed
with
the
goal
of
minimizing
adverse
impacts
on
small
businesses.
The
results
of
benefit­
cost
analysis
of
each
option
indicate
that
removing
entrainment
reduction
requirements
is
not
necessarily
the
most
cost
effective
approach
to
minimizing
economic
impact
on
small
businesses
and
maximizing
benefits.
A
summary
of
the
benefit­
cost
ratios
is
provided
in
the
table
below.
Each
benefit­
cost
ratio
represents
the
amount
of
estimated
return
for
each
dollar
spent
in
compliance.
For
example,
the
ratio
of
0.37
for
the
2­
10
MGD
size
category
under
Option
One
signifies
an
estimated
37
cent
return
in
the
form
of
reduced
I&
E
for
each
dollar
spent
to
comply
with
the
rule.

Regulatory
Options:
Benefit­
Cost
Ratios
Facility
Size
Category
(
DIF)
Option
1:
Raising
the
Compliance
Threshold
Option
2
Option
3
Baseline
Ratios
10+
MGD
20+
MGD
50+
MGD
250+
MGD
2
­
10
0.37
0.14
0.19
10
­
20
0.72
0.72
­
­
­
0.32
0.40
20
­
50
1.00
1.00
1.00
­
­
0.53
0.59
50
­
250
1.95
1.95
1.95
1.95
­
1.95
1.95
250+
4.90
4.90
4.90
4.90
4.90
4.90
4.90
Overall
Benefit­
Cost
Ratio
For
Each
Option
1.85
2.05
2.20
2.90
4.90
2.21
2.36
After
performing
an
analysis
on
each
option,
it
was
determined
that
the
preferred
regulatory
alternative
was
Regulatory
Option
One.
This
option
examined
the
cost
effectiveness
of
raising
the
threshold
to
that
of
10,
20,
50
or
250
MGD.
Results
indicate
that
raising
the
compliance
threshold
from
the
proposed
2
MGD
to
that
of
20
MGD
would
eliminate
spending
on
facilities
where
the
costs
of
the
regulation
outweigh
the
benefits.
Facilities
with
DIF
greater
than
20
MGD
withdraw
the
most
water
and
cause
the
highest
amount
of
impingement
and
entrainment.
Smaller
facilities
take
in
less
water,
cause
less
impingement
and
entrainment,
and
do
not
face
proportionally
lower
compliance
costs.
It
is
concluded
that
it
is
not
cost­
effective
to
require
smaller
facilities
to
retrofit
their
intake
systems.
As
the
table
below
illustrates,
raising
the
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
5
­
threshold
to
20
MGD
exempts
approximately
44
percent
of
facilities
while
achieving
95
percent
of
the
benefits.
Benefit­
Costs
of
Raising
the
Compliance
Threshold
Facility
Size
Category
Total
Compliance
Cost
(
Millions)
Total
Benefit
(
Millions)
Benefit/
Cost
Ratio
Percent
of
Total
Benefits
Achieved
Percent
of
Facilities
Exempted
10+
MGD
$
296.18
$
608.15
2.05
97.8%
27.0%

20+
MGD
$
266.83
$
587.10
2.20
94.4%
43.7%
50+
MGD
$
169.02
$
489.61
2.90
78.7%
76.8%
250+
$
54.36
$
266.25
4.90
42.8%
94.5%

It
should
be
noted
the
benefit­
cost
analyses
results
presented
in
this
study
are,
at
best,
rough
estimates.
Because
of
EPA
obligations
to
protect
confidential
business
data,
exact
costs
and
benefits
of
these
options
may
be
only
be
fully
calculated
by
EPA.
The
results
presented
in
this
study
provide
a
strong
case
for
further
examination
of
the
identified
alternatives.
It
is
suggested
that
raising
the
compliance
threshold
will
still
retain
a
significant
amount
of
benefits.
Further
research
should
be
conducted
to
explore
this
option.

More
detailed
explanations
of
the
methodology
and
results
of
this
analysis
can
be
found
in
the
later
sections
of
this
report.
The
report
is
structured
into
six
other
sections.
The
following
list
offers
a
preview
of
each
section:


Section
Two
provides
a
brief
overview
of
the
Phase
III
rulemaking
process
and
proposed
regulation.


Section
Three
includes
an
in­
depth
review
and
analysis
of
the
EPA's
cost­
benefit
analysis
and
methodology
used
in
Phase
I
and
II.


Sections
Four
and
Five
examines
and
critiques
the
approach
used
in
developing
cost
and
benefit
estimates
for
Phase
III.


Section
Six
offers
an
overview
of
three
regulatory
options
for
Phase
III
accompanied
by
a
cost­
benefit
valuation
of
each.


Section
Seven
summaries
the
findings
of
the
regulatory
option
analysis
and
presents
a
recommendation
for
a
preferred
regulatory
alternative.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
6
­
2.
Phase
III
Background
EPA
is
developing
regulations
under
Section
316(
b)
of
the
Clean
Water
Act.
Section
316(
b)
requires
that
the
location,
design,
construction
and
capacity
of
cooling
water
intake
structures
reflect
the
best
technology
available
for
minimizing
adverse
environmental
impact.
More
than
1,500
industrial
facilities
use
large
volumes
of
cooling
water
from
lakes,
rivers,
estuaries
or
oceans
to
cool
their
plants,
including
steam
electric
power
plants,
pulp
and
paper
makers,
chemical
manufacturers,
petroleum
refiners,
and
manufacturers
of
primary
metals
like
iron
and
steel
and
aluminum.

Cooling
water
intake
structures
cause
adverse
environmental
impact
by
pulling
large
numbers
of
fish
and
shellfish
or
their
eggs
into
a
power
plant
or
factory
cooling
system.
There,
the
organisms
may
be
killed
or
injured
by
heat,
physical
stress,
or
chemicals
used
to
clean
the
cooling
system.
Larger
organisms
may
be
killed
or
injured
when
they
are
trapped
against
screens
at
the
front
of
an
intake
structure.

The
final
new
facility
rule
(
Phase
I)
established
requirements
applicable
to
the
location,
design,
construction,
and
capacity
of
cooling
water
intake
structures
at
new
facilities
that
withdraw
at
two
(
2)
million
or
more
gallons
per
day
(
MGD)
and
use
at
least
twenty
five
(
25)
percent
of
the
water
they
withdraw
solely
for
cooling
purposes.
EPA
proposed
Phase
I
regulations
governing
new
facilities
that
employ
a
cooling
water
intake
structure
on
July
20,
2000.
EPA
took
final
action
on
the
Phase
I
regulations
on
November
9,
2001.

The
proposed
existing
facilities
rule
for
large
electric
power
generating
facilities
(
Phase
II),
will
establish
requirements
applicable
to
existing
facilities
that
both
generate
and
transmit
electric
power
or
that
generate
electric
power
for
sale
to
another
entity
for
transmission;
use
one
or
more
cooling
water
intake
structures
to
withdraw
water
from
waters
of
the
U.
S.;
have
or
require
a
National
Pollutant
Discharge
Elimination
System
(
NPDES)
permit
issued
under
section
402
of
the
CWA;
and
meet
proposed
flow
thresholds.
Existing
electric
power
generating
facilities
subject
to
this
proposal
would
include
those
that
use
cooling
water
intake
structures
to
withdraw
fifty
(
50)
million
gallons
per
day
(
MGD)
or
more
and
that
use
at
least
twenty­
five
(
25)
percent
of
water
withdrawn
solely
for
cooling
purposes.

The
proposed
Phase
III
rules,
the
subject
of
this
study,
will
cover
the
remaining
facilities
including
small
electric
generating
and
existing
industrial
facilities.
On
November
22,
2002,
EPA
and
the
plaintiffs
in
Riverkeeper,
Inc.
v.
Whitman
jointly
submitted
a
Second
Amended
Consent
Decree
to
the
U.
S.
District
Court,
Southern
District
of
New
York,
which
the
court
signed
on
November
25,
2002.1
The
Second
Amended
Consent
Decree
extended
the
deadlines
for
EPA
to
propose
and
take
final
action
on
regulations
under
Section
316(
b)
of
the
Clean
Water
Act
to
minimize
the
adverse
environmental
impact
of
cooling
water
intake
structures
at
industrial
facilities.

Under
the
Second
Amended
Consent
Decree,
the
new
deadlines
for
rulemaking
are
as
follows:

1
U.
S.
District
Court,
Southern
District
of
New
York
No.
93
Civ.
0314
(
AGS)
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
7
­
 
Phase
II
regulations
governing,
at
a
minimum,
existing
utilities
and
non­
utility
power
producers
that
employ
a
cooling
water
intake
structure
(
CWIS),
and
whose
flow
levels
exceed
a
minimum
threshold
to
be
determined
by
EPA:
Final
action
by
February
16,
2004.
(
EPA
proposed
Phase
II
regulations
on
February
28,
2002.)

 
Phase
III
regulations
governing,
at
a
minimum,
existing
facilities
that
employ
a
cooling
water
intake
structure,
that
are
not
covered
by
the
Phase
II
rule,
and
whose
intake
flow
levels
exceed
a
minimum
threshold
to
be
determined
by
EPA:
Proposal
by
November
1,
2004.
Final
action
by
June
1
2006.

The
table
below
outlines
the
key
activities
in
the
Phase
III
rulemaking
including
dates
and
brief
summaries
of
the
Regulatory
Flexibility
processes
and
outreach
meetings.

Table
1:
Phase
III
Timeline
Date
Event
August
30,
2002
EPA
completed
an
"
initial
small
entity
impact
summary"
to
determine
the
potential
economic
impacts
of
regulatory
options
on
small
entities,
as
required
by
the
Regulatory
Flexibility
Act,
5
U.
S.
C.
§
601,
et
seq.

The
findings
of
the
initial
summary
were
that
some
regulatory
options
for
the
Phase
III
Regulations
may
have
a
significant
economic
impact
on
a
substantial
number
of
small
entities.
September
14,
2002
EPA
staff
distributed
a
detailed
background
document
to
potential
Small
Entity
Representatives
(
SERs),
to
trade
associations
helping
EPA
identify
SERs,
and
to
SBA
and
OMB.

The
background
document
includes
several
preliminary
regulatory
alternatives
for
the
Phase
III
Regulations
and
an
assessment
of
the
potential
economic
impacts
of
one
of
these
alternatives
on
small
entities.
October
1,
2002
EPA
managers
and
staff
held
an
outreach
meeting
with
potential
SERs,
trade
associations,
SBA,
and
OMB
to
review
and
discuss
the
outreach
materials
and
seek
comments
that
EPA
would
consider
in
developing
a
"
Convening
Document"
for
the
SBAR
panel.

In
light
of
the
extension
in
the
deadlines
for
the
Phase
III
rule,
EPA
did
not
convene
a
SBR
panel,
but
intends
to
do
so,
if
necessary,
by
February
2004.
April
4,
2002
A
notice
was
published
in
the
Federal
Register
announcing
that
the
EPA
submitted
a
revised
Information
Collection
Request
(
ICR)
supporting
development
of
the
§
316(
b)
regulations
to
the
Office
of
Management
and
Budget
(
OMB).

The
renewal
would
allow
EPA
to
conduct
the
following
activities:
 
continue
collecting
essential
technical
and
economic
data
on
potential
Phase
III
facilities,
specifically
in
the
offshore
and
coastal
oil
and
gas
extraction
facilities
(
OCOGEFs)
and
seafood
processing
industries
not
previously
considered
in
Phase
I
or
II;
 
clarify
responses
EPA
received
on
questionnaires
distributed
in
1998
and
2000;
and
 
request
that
facilities
provide
information
on
existing
316(
b)
environmental
studies.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
8
­
3.
Introduction
to
the
Benefits
Estimate
Methodology
The
review
and
critique
of
the
United
States
Environmental
Protection
Agency's
(
EPA)
estimate
of
benefits
based
on
proposed
regulations
in
Phase
III
of
the
Section
316(
b)
rulemaking
is
presented
in
two
separate
parts.
This
section
is
an
analysis
of
the
approach
outlined
in
the
initial
background
documents
provided
by
the
EPA.
Section
4
examines
recent
revisions
to
this
methodology
as
described
in
the
Phase
II
NODA
dated
March
19,
2003.

In
the
document,
"
Background
Information
for
Potential
Small­
Entity
Representatives
on
EPA's
Rulemaking
Project,"
the
EPA
used
supporting
information
from
Phase
I
and
Phase
II
rules
to
estimate
potential
environmental
impacts
associated
with
regulating
Cooling
Water
Intake
Structures
(
CWIS)
in
Phase
III.
The
EPA
cited
findings
from
the
following
three
case
studies
conducted
in
Phase
II
to
illustrate
potential
benefits
in
Phase
III:

1.
Facilities
along
the
Ohio
River
(
Miami
Fort
Power
Plant)
2.
Brayton
Point
Generating
Station
in
Somerset,
Massachusetts
3.
J.
R.
Whiting
Plan,
located
on
Lake
Erie
in
Michigan
The
EPA
also
referred
to
studies
of
entrainment
conducted
at
five
Hudson
River
power
plants
in
1980
and
2000
in
illustrating
the
potential
benefits
of
Phase
III
regulations.
The
results
of
these
studies
show
a
20
to
40
percent
reduction
of
certain
fish
species
during
the
20­
year
period.
However,
EPA
did
not
conduct
a
case
study
of
impinged
and
entrained
(
I&
E)
fish
at
the
Hudson
River
facilities.
As
the
benefits
of
reducing
I&
E
at
Hudson
River
are
uncertain,
an
analysis
of
I&
E
fish
in
Hudson
River
are
not
included.

This
section
of
the
report
focuses
on
areas
where
EPA's
methodology
for
calculating
national
baseline
economic
benefits
may
result
in
either
uncertainty
or
an
overestimate
of
benefits.
For
the
purposes
of
this
analysis,
the
term
"
uncertainty"
is
used
to
describe
steps
in
EPA's
methodology
that
can
lead
to
either
an
underestimate
or
overestimate
of
benefits.
This
study
does
not
include
a
discussion
of
underestimates
since
the
EPA
has
already
identified
potential
underestimates
in
an
explanation
of
their
methodology;
the
following
paragraph
provides
a
description
of
some
of
these
underestimates.

In
an
introduction
to
the
Phase
II
case
studies,
EPA
stated
several
reasons
for
why
their
analysis
is
"
likely
to
lead
to
potentially
significant
underestimates
of
baseline
losses
in
most
cases,
and
therefore
underestimates
of
regulatory
benefits."
In
regards
to
data
limitations,
EPA
stated
that
much
of
the
facility­
provided
biological
monitoring
data
used
in
the
case
studies
was
dated
to
20
or
more
years
ago,
before
activities
under
the
Clean
Water
Act
improved
aquatic
conditions.
As
a
result,
EPA
suggested
that
the
current
number
of
I&
E
losses
are
likely
to
be
much
greater
due
to
an
increased
population
of
fish
species
living
in
cleaner
habitats.
In
terms
of
recreational
and
commercial
benefits,
EPA
stated
only
selected
species
were
evaluated
because
I&
E
or
valuation
data
was
limited;
as
a
result,
only
the
values
of
a
portion
of
I&
E
fish
were
considered.
Also,
EPA
noted
the
value
assigned
to
forage
species
might
be
understated
because
the
full
ecological
value
of
the
species
as
part
of
the
food
web
was
not
considered.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
9
­
The
analysis
of
the
three
case
studies
listed
above
will
provide
specific
examples
for
ways
in
which
EPA's
methodology
may
result
in
either
uncertainty
or
an
overestimate
of
benefits.
This
Section
of
the
report
is
divided
into
six
sub­
sections
including:

1.
A
critique
of
EPA's
methodology
for
estimating
benefits
in
phase
III
2.
An
analysis
of
EPA's
methodology
for
calculating
Benefits
3.
An
analysis
of
Benefits
estimated
using
RUM
approach
4.
An
analysis
of
Benefits
estimated
using
HRC
approach
5.
A
review
of
case
study
facilities
6.
The
section
concludes
with
a
summary
of
the
benefits
methodology
3.1.
Analysis
of
EPA's
Methodology
for
Estimating
Benefits
in
Phase
II
This
section
includes
an
analysis
and
critique
of
EPA's
methodology
for
estimating
benefits
in
Phase
II,
which
were,
as
noted
in
the
introduction,
used
to
approximate
benefits
in
Phase
III.
EPA's
methodology
is
complex
and
based
on
several
assumptions
and
generalizations.
As
a
result,
there
is
uncertainty
regarding
the
accuracy
of
EPA's
benefits
at
individual
facilities,
as
well
as
facilities
nationwide.
While
the
EPA
highlights
steps
in
their
valuation
process
where
an
assumption
might
lead
to
an
underestimate
of
benefits,
a
discussion
of
potential
overestimates
is
negligible.
This
section
aims
to
highlight
places
in
which
those
uncertainties
and
overestimates
may
exist
and
require
further
analysis
or
investigation.
As
EPA's
estimate
of
benefits
is
based
on
data
gathered
from
the
Phase
II
case
studies,
this
section
begins
with
a
discussion
of
possible
biases
in
EPA's
selection
of
these
case
studies,
and
is
followed
by
an
analysis
of
EPA's
method
of
evaluating
and
valuing
reduction
of
I&
E.

Uncertainty:
In
the
selection
of
case
studies
EPA
did
not
consider
facilities
that
withdraw
less
than
50
MGD.

EPA
suggested
the
potential
benefits
of
regulating
existing
facilities
in
Phase
III
are
similar
to
the
benefits
derived
from
case
studies
conducted
in
Phase
II.
Based
on
EPA's
preliminary
smallentity
impact
analysis
for
Phase
III,
there
are
approximately
707
in­
scope
electric
generating
and
manufacturing
facilities,
with
105
considered
small
based
on
SBA
size
standards.
2
The
average
design
intake
flows
of
these
small­
entity
facilities
are
13.8
MGD
for
electric
generators
and
27
MGD
for
manufacturers.
As
a
result,
the
majority
of
potentially
regulated
small
entities
in
Phase
III
have
facilities
with
CWIS
that
withdraw
less
than
50
MGD.
The
case
study
facilities
in
Table
1
show
that
the
design
intake
flows
of
three
Phase
II
case
studies
are
97
to
1,955%
percent
2
From
review
of
the
document,
"
Background
Information
for
Potential
Small­
Entity
Representatives
on
EPA's
Rulemaking
Project:
Proposed
Regulations
to
Establish
Requirements
for
Cooling
Water
Intake
Structures
at
Section
316(
b)
Phase
III
Facilities,"
dated
September
17,
2002.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
10
­
greater
than
50
MGD.
It
is
likely
that
the
benefits
of
regulating
facilities
with
larger
intakes
are
different
from
the
benefits
of
regulating
facilities
with
smaller
intakes.
To
determine
the
extent
of
this
difference,
a
better
methodology
would
be
to
conduct
case
studies
that
determine
the
benefits
of
regulating
facilities
with
similar
intake
capacity
volumes
as
those
facilities
potentially
regulated
in
Phase
III.

Table
2:
Intake
Flow
(
MGD)
of
Phase
II
Case
Study
Facilities
Facility
Intake
(
MGD)
Percent
Greater
than
50
MGD
Miami
Fort
Plant
98.7
97%
J.
R.
Whiting
Plant
308
516%
Brayton
Point
Station
925
­
1,130
1,955%

Uncertainty:
EPA's
selection
of
case
studies
may
be
biased.

EPA's
selection
of
case
study
facilities
may
be
biased
because
these
facilities
were
chosen
based
on
the
availability
of
biological
monitoring
data,
instead
of
random
sampling.
In
comments
to
EPA
regarding
Phase
II,
the
Utility
Water
Act
Group
(
UWAG)
stated,
"
those
facilities
with
the
most
complete
data
sets
tend
to
be
the
`
high
profile'
sites
that
developed
the
data
to
allay
perceptions
about
potential
environmental
impacts.
The
many
sites
that
do
not
have
extensive
data
sets
may
be
more
representative
of
typical
CWIS
impacts
that
facilities
that
have
been
heavily
scrutinized
and
therefore
have
amassed
extensive
data
and
information."
When
benefits
from
a
small
number
of
case
studies
which
are
not
representative
of
the
population
are
extrapolated
to
estimate
national
benefits,
biases
at
the
individual
facility
level
will
be
greatly
magnified.
In
order
to
provide
impartial
results,
a
better
methodology
would
be
to
conduct
case
studies
at
randomly
selected
facilities.

Uncertainty:
EPA
does
not
account
for
density
dependence
in
estimating
benefits
of
reducing
I&
E
losses
based
on
replacement
and
restoration
costs.

In
their
comments
to
EPA
regarding
Phase
II,
UWAG
highlighted
the
existence
of
density
dependence
in
which
reductions
in
population
size
caused
by
natural
environmental
fluctuations
must
result
in
increased
survival,
growth,
or
fecundity
of
the
remaining
individuals.
UWAG
cited
the
Connecticut
Yankee
and
mid­
Hudson
power
plants
as
examples
in
which
long­
term
monitoring
demonstrated
the
negligible
effects
of
once­
through
facilities
because
density
dependence
and
other
mechanisms
were
operating
to
mitigate
individual
losses.
UWAG
also
cited
a
study
by
the
Ohio
River
Ecological
Research
Program
which
demonstrated
that,
in
general,
conditions
upstream
and
downstream
of
each
plant
studied
are
similar,
and
conditions
over
the
past
20
years
have
generally
improved;
this
is
the
opposite
of
what
one
would
expect
if
I&
E
losses
were
adversely
affecting
fish
populations.

3.2.
EPA's
Methodology
for
Calculating
Benefits
The
EPA
used
a
complex
methodology,
based
on
several
assumptions,
to
calculate
the
benefits
of
reducing
economic
losses
of
I&
E
at
the
case
study
facilities.
In
addition,
the
method
of
calculating
benefits
varied
among
case
study
facilities.
The
uncertainties
and
potential
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
11
­
overestimates
of
benefits
in
this
methodology
are
highlighted.
An
outline
of
the
steps
in
EPA's
methodology
is
provided
below
in
Table
2.

Table
3:
Overview
of
EPA's
Methodology
of
Benefits
Valuation
Step
Description
Step
1
Compile
and
estimate
I&
E
losses
by
life
stage.

Step
2
Extrapolate
stage­
specific
losses
to
losses
of
age
1
equivalent
fish.

Step
3
Extrapolate
age
equivalents
of
exploited
fish
species
to
pounds
of
fish
lost
to
commercial
and
recreational
fisheries.

Step
4
Extrapolate
age
1
equivalents
of
forage
fish
to
losses
in
production
of
commercial
and
recreational
fish.

Step
5
Calculate
baseline
I&
E
/
I&
E
with
Regulation.

Step
6
Calculate
differences
in
losses.

Step
7
Calculate
value
of
economic
losses
from
I&
E
using
two
types
of
methods:
 
Use
demand
driven
approach
with
benefits
transfer
techniques
 
Use
supply
driven
techniques
such
as
the
Random
Utility
Model
(
RUM),
Habitat­
Replacement
Cost
(
HRC)
analysis
and
societal
revealed
preference.

Step
8
Estimate
costs
as
benefits.

Step
9
Extrapolate
data
from
case
study
facilities
to
national
baseline
losses.

Uncertainty:
EPA's
use
of
non­
standardized,
facility­
monitored
I&
E
data
in
their
case
studies.

There
are
no
nationwide
standards
for
monitoring
I&
E
fish
at
CWIS
facilities.
Since
much
of
the
I&
E
data
EPA
used
to
estimate
economic
losses
in
Phase
II
were
collected
by
the
facilities
themselves,
there
may
be
discrepancies
based
on
each
facility's
method
and
location
for
sampling
the
water.
In
addition,
many
facilities
monitored
I&
E
for
particular
fish
species;
as
a
result,
the
data
may
show
a
greater
abundance
of
that
species
than
a
randomly
drawn
sample.
The
EPA
stated,
"
there
are
often
substantial
uncertainties
and
potential
biases
in
the
I&
E
estimates;
comparison
of
results
between
studies
is
therefore
very
difficult
and
sometimes
impossible,
even
among
facilities
that
impinge
and
entrain
the
same
species."
In
order
to
obtain
an
unbiased
set
of
I&
E
data,
the
EPA
should
develop
guidelines
that
specify
the
way
in
which
I&
E
fish
should
be
gathered
and
monitored
over
a
period
of
time.
With
these
standards,
EPA's
method
of
extrapolating
data
from
individual
facilities
to
all
in­
scope
facilities
will
have
more
validity.
It
is
unclear
if
EPA's
use
of
facility­
monitored
I&
E
will
lead
to
an
overestimate
or
underestimate
of
benefits.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
12
­
Uncertainty:
EPA's
estimate
of
nonuse
benefits
as
50
percent
of
estimated
recreational
fishing
benefits
may
be
overstated.

In
the
benefits
transfer
technique,
the
EPA
calculated
the
mean
annual
economic
value
of
I&
E
based
on
an
estimate
of
commercial,
recreational,
nonuse
and
forage
values.
The
EPA
described
nonuse
benefits
as
the
value
individuals
ascribe
to
improved
environmental
quality
apart
from
any
past,
present
or
anticipated
future
use
of
the
resource
in
question.
The
EPA
also
defined
nonuse
benefits
as
passive
use
values
that
may
include
the
concept
that
some
ecological
services
are
valuable
apart
from
any
human
uses
or
motives.
Examples
of
these
ecological
services
may
include
improved
reproductive
success
for
aquatic
and
terrestrial
wildlife,
increased
diversity
of
aquatic
and
terrestrial
species,
and
improved
conditions
for
recovery
of
I&
E
species.
To
estimate
the
value
of
nonuse
benefits,
EPA
used
a
50
percent
rule
of
thumb,
in
which
nonuse
was
calculated
as
50
percent
of
estimated
recreational
fishing
benefits.
3
However,
EPA
noted
that
"
using
the
50
percent
rule
poses
several
concerns
and
includes
several
limitations."
Such
limitations
include
the
dated
nature
and
key
differences
in
the
studies
from
which
the
50
percent
estimate
was
derived.
To
address
this
issue,
EPA
stated
that
they
"
intend
to
revisit
the
body
of
research
on
this
topic
and
re­
evaluate
how
to
apply
benefits
transfer
in
developing
estimates
of
nonuse
value
benefits
in
the
future."
In
response,
UWAG
stated
that
EPA
made
"
no
effort
to
investigate
the
similarity
of
these
situations
to
CWIS
improvements."
UWAG
suggested
that
marginal
improvements
to
fisheries
being
studied
may
not
be
unique
or
create
substantial
awareness,
thus
minimizing
the
nonuse
value.
UWAG
cited
an
analysis
from
the
Triangle
Economic
Research
(
TER),
which
concluded
that
nonuse
values
are
not
zero
but
likely
to
be
negligible.

Uncertainty:
EPA's
estimate
of
costs
as
benefits.

The
EPA
equated
the
costs
of
reducing
I&
E
with
the
benefits.
A
representative
from
UWAG
stated,
the
"
assumption
that
benefits
equal
costs
is
fundamentally
flawed.
Consumer
surplus,
not
cost,
is
the
basic
measuring
concept
for
estimating
benefits."
A
representative
from
the
Edison
Electric
Institute
(
EEI)
also
questioned
EPA's
use
of
costs
to
approximate
benefits.
The
EEI
representative
stated,
"
benefits
are
associated
with
people's
willingness
to
pay
to
have
certain
products
or
services
while
costs
are
the
opportunity
costs
of
the
forgone
resources
used
to
produce
that
product
or
service.
There
is
no
reason
to
expect
that
costs
would
be
a
good
proxy
for
benefits.
There
is
no
basis
in
economic
theory
or
practice
for
using
replacement
costs
to
approximate
benefits."

Uncertainty:
EPA's
use
of
economic
losses
at
a
small
number
of
facilities
to
extrapolate
to
national
baseline
losses.

EPA
extrapolated
baseline
loss
estimates
at
individual
case
study
facilities
to
estimate
national
baseline
losses.
While
the
method
of
extrapolation
is
reasonable
in
estimating
national
benefits,
the
use
of
data
from
only
one
facility
to
extrapolate
to
20
to
80
facilities
may
greatly
magnify
uncertainties
in
the
calculation
of
benefits
at
that
one
facility.

3
See
EPA
316(
b)
Case
Studies
Chapter
E4­
3,
http://
www.
epa.
gov/
waterscience/
316b/
casestudy/
che4.
pdf
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
13
­
From
I&
E
monitored
at
41
case
study
facilities,
the
EPA
used
data
from
13
facilities
to
extrapolate
national
baseline
economic
losses
for
539
in­
scope
facilities.
The
method
EPA
used
to
select
these
13
reference
facilities
is
unclear
and
should
be
more
thoroughly
addressed.
Inscope
facilities
are
any
of
the
539
utility
and
non­
utility
steam
electric
power
generating
facilities
identified
in
EPA's
2000
Section
316(
b)
Industry
Survey
as
being
potentially
covered
by
this
proposed
rule.
As
shown
in
Table
3
below,
losses
in
four
waterbody
types
were
extrapolated
based
on
data
from
one
case
study
facility.
The
EPA
should
consider
using
an
average
of
data
from
more
than
one
facility
to
use
in
extrapolating
to
national
losses.

Table
4:
Number
of
EPA
Case
Study
Facilities
Extrapolated
to
all
In­
Scope
Facilities
Type
of
Waterbody
#
of
Case
Study
Facilities
Facility
Used
as
a
Basis
for
EPA's
Extrapolation
No.
of
In­
Scope
Facilities
within
Waterbody
Type
Estuary
 
Non
Gulf
4
1
(
Salem)
78
Estuary
 
Gulf
Coast
4
1
(
Big
Bend)
30
Freshwater
30
9
(
Ohio
River
Studies)
393
Great
Lakes
1
1
(
J.
R.
Whiting)
16
Ocean
2
1
(
Pilgrim)
22
TOTAL
41
13
539
Overestimate:
EPA's
use
of
the
Flow
Index
to
estimate
national
baseline
losses.

EPA
used
two
methods
of
extrapolation
to
estimate
national
baseline
economic
losses:
the
Flow
Index
and
the
Angling
Index.
The
Flow
Index
extrapolation
is
based
on
the
assumption
that
impingement
and
entrainment
are
strictly
proportional
to
flow.
This
may
not
be
the
case,
as
noted
in
a
research
report
prepared
by
D.
G.
Heimbuch,
Ph.
D.
for
UWAG.
4
In
this
report,
the
author
illustrates
a
case
in
which
I&
E
at
three
Ohio
River
facilities
is
not
proportional
to
flow.
In
Figure
1,
available
data
from
five
of
the
nine
Ohio
River
facilities
are
graphed
to
show
the
relationship
between
production
foregone
due
to
I&
E
and
flow.
The
data
points
show
that
I&
E
losses
are
not
strictly
proportional
to
flow.
As
a
result,
the
use
of
the
Flow
Index
may
overestimate
benefits
as
an
increase
in
intake
flow
may
not
always
be
commensurate
with
an
increase
in
losses
of
I&
E.

4
Based
on
report
prepared
for
UWAG
by
D.
G.
Heimbuch,
Ph.
D.
entitled,
"
Review
of
EPA's
Methods
for
Extrapolating
Baseline
Loss
Estimates
at
Case
Study
Facilities
to
National
Baseline
Loss
Estimates."
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
14
­
Figure
1:
Production
Foregone
based
at
Ohio
River
Facilities
Kyger
Creek
Miami
Fort
WC
Beckford
Phillip
Sporn
Tanners
Creek
0
1,000,000
2,000,000
3,000,000
4,000,000
5,000,000
0
200
400
600
800
1000
1200
Intake
Flow
(
MGD)
I&
E
Production
Foregone
(
lbs.)

In
their
comments
to
EPA
regarding
Phase
II,
UWAG
cited
an
ongoing
study
conducted
by
Oak
Ridge
National
Laboratory
(
ORNL)
on
behalf
of
the
Electric
Power
Research
Institute
(
EPRI)
to
survey
multiple
Texas
and
Tennessee
reservoirs
in
an
attempt
to
find
a
relationship
between
reservoir
withdrawals
and
species
composition
or
abundance.
ORNL's
analysis
showed
no
relationship
between
intake
rate
and
fish
population
responses
at
reservoir
withdrawal
rates
of
1
to
5,200
MGD.
This
finding
also
suggests
that
EPA's
use
of
the
Flow
Index
to
extrapolate
to
national
losses
of
all
in­
scope
facilities
may
result
in
an
overestimate
of
losses
as
the
relationship
between
intake
and
production
foregone
is
not
strictly
proportional.

Overestimate:
EPA's
use
of
the
Angling
Index
to
estimate
national
baselines
losses.

In
addition
to
the
Flow
Index,
the
EPA
used
an
Angling
Index
to
extrapolate
to
national
economic
losses.
The
Angling
Index
estimates
the
level
of
recreational
angling
pursued
by
populations
within
120
miles
of
each
facility.
UWAG
cited
the
following
three
factors
that
may
inflate
the
Angling
Index:

1.
Method
duplicates
counting
for
facilities
within
120
miles
of
each
other.
2.
The
EPA
may
not
have
accurately
evaluated
fishermen's
choices
of
where
they
would
go
fishing.
3.
Loss
of
fish
does
not
automatically
translate
into
benefit
if
suddenly
protected.

EPA's
use
of
the
Angling
Index
varied
depending
on
the
waterbody
type.
For
example,
since
the
facilities
in
the
Ohio
River
are
located
in
proximity
to
one
another,
the
EPA
treated
the
29
Ohio
River
facilities
as
one
large
facility
for
the
purpose
of
calculating
the
angling
index;
this
eliminated
the
problem
of
multiple
counting
of
angling
days
in
the
case
of
the
Ohio
River
facilities.

The
EPA
used
two
supply
driven
techniques
to
estimate
baseline
economic
losses:
the
Random
Utility
Model
(
RUM)
and
the
Habitat
Replacement
Cost
(
HRC).
The
RUM
approach,
discussed
in
this
section,
evaluates
changes
in
consumer
valuation
of
water
resources
expected
from
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
15
­
reductions
in
I&
E­
related
fish
losses.
The
HRC
technique,
discussed
in
the
following
subsection,
assigns
economic
value
to
I&
E
losses
based
on
the
combined
costs
of
implementing
restoration
actions
to
produce
the
organisms
that
were
lost,
administering
the
program
and
monitoring
the
production
resulting
from
restoration
actions.
Economic
losses
from
the
Ohio
River
case
study
were
calculated
using
RUM,
while
the
Brayton
Point
and
J.
R.
Whiting
losses
were
calculated
using
HRC
techniques.
The
following
analysis
by
valuation
approach
indicates
how
EPA's
methodology
may
lead
to
uncertainty
or
an
overestimate
of
benefits.

3.3.
Analysis
of
Benefits
using
RUM
approach
The
Random
Utility
Model
(
RUM),
combined
with
a
trip
frequency
model,
estimates
the
number
of
days
fished
when
site
or
individual
characteristics
change.
RUM
assigns
a
value
to
each
fish
by
specie
based
on
data
provided
in
phone
and
site
surveys
of
recreational
fishermen.
The
model
takes
into
account
the
number
of
trips,
cost
per
trip,
the
numbers
and
types
of
fish
caught,
and
the
changes
in
trip
frequency
due
to
increasing
or
decreasing
catch
rates.
Changes
in
economic
value
occur
because
of
these
trip
frequency
fluctuations.
The
main
assumption
in
the
model
is
that
anglers
will
have
greater
satisfaction
and
economic
value
from
sites
where
catch
rates
are
higher.
For
example,
an
angler
may
chose
to
travel
three
hours
to
a
site
where
he
catches
five
fish
per
hour
rather
than
traveling
a
shorter
distance
for
site
that
yields
only
one
fish
an
hour.
The
EPA
uses
these
dollar
values
in
conjunction
with
commercial
pricing
data
to
estimate
benefits.

The
following
is
a
simplified
step­
by­
step
overview
of
the
EPA's
RUM
valuation
of
I&
E
losses:

1.
Use
RUM
to
determine
the
dollar
value
for
different
species
as
assessed
by
surveys
of
recreational
fishermen.
2.
Quantify
I&
E
losses
by
species
for
a
given
area.
3.
Calculate
how
many
of
these
I&
E
fish
can
be
attributed
as
a
loss
to
either
recreational
or
commercial
uses.
4.
Multiply
recreational
values
and
market
prices
by
the
number
of
fish
lost
to
either
recreational
or
commercial
uses
to
estimate
economic
losses
from
I&
E.

While
logical,
the
RUM
model
includes
some
assumptions
that
may
lead
to
either
uncertainty
or
an
overestimate
of
benefits.
The
following
sections
outline
these
issues.

Overestimate:
EPA's
use
of
data
from
surveys
of
recreational
fishermen
to
value
recreational
fish.

The
RUM
valuation
of
recreational
fish
values
may
lead
to
an
overestimate
of
benefits.
These
value
estimates
are
based
on
surveys,
which
may
have
recall,
non­
response
and
sampling
biases.
Phone
and
onsite
surveys
collected
specific
information
regarding
the
anglers'
mode
of
fishing,
length
of
the
completed
trip,
species
targeted
on
that
trip,
number
of
fish
caught,
the
number
of
times
they
have
fished
in
the
past
two
to
twelve
months
and
the
average
cost
of
the
trip.
This
information
served
as
input
into
the
model
and
was
expanded
to
represent
the
dollar
values
of
fish
caught
recreationally.
Any
biases
in
the
survey
data
that
may
have
overrepresented
the
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
16
­
opinions
of
avid
recreational
fisherman
would
ultimately
lead
to
an
overestimate
of
the
value
of
the
fish.
The
following
three
subsections
provide
a
more
detailed
discussion
of
these
biases.

Overestimate:
Recall
Bias
in
Survey
Recall
bias
is
the
inability
of
a
respondent
to
remember
exactly
the
number
of
trips
and
fish
he
or
she
caught
in
the
past.
EPA
acknowledged
that
studies
have
shown
that
avid
fishermen
surveyed
have
a
tendency
to
overstate
his
or
her
historical
catch
rates
and
trip
frequency.
Combined
with
the
assumption
that
higher
catch
rates
equals
higher
"
satisfaction"
or
higher
value,
this
bias
alone
can
inflate
the
RUM
valuations.

Overestimate:
Non­
response
Bias
in
Survey
Non­
response
bias
is
the
phenomenon
where
a
person
declines
to
participate
in
a
survey
due
to
lack
of
interest
and
the
belief
that
the
survey
does
not
pertain
to
them.
The
first
of
two
surveys
used
by
EPA,
the
National
Marine
Fisheries
Service
(
NMFS)
Marine
Recreational
Fishing
Statistics
Survey
(
MRFSS),
is
a
long­
term
monitoring
program
that
combines
a
random
digit­
dial
telephone
survey
of
households
with
a
survey
of
anglers
at
randomly
picked
fishing
sites.
The
phone
surveys,
though
random,
may
not
have
accurately
represented
the
population.
The
surveys
were
geared
toward
avid
fisherman
because
they
contained
questions
on
species
targeted
and
the
gear
and
methods
used.
Less
frequent
or
less
avid
anglers
may
not
have
chosen
to
take
the
survey
because
they
assumed
it
did
not
pertain
to
them.
In
addition,
only
participants
agreeing
to
the
follow­
up
survey
were
included.
Because
the
respondents
most
likely
to
participate
in
these
surveys
are
avid
anglers,
the
sampling
may
not
be
random
and
the
values
assessed
may
be
overstated.
The
structure
and
methods
used
in
surveying
may
inadvertently
over
represent
the
activities
of
avid
users
and
downplay
other
participant
values.
Because
of
this
non­
response
bias,
EPA's
extrapolation
of
economic
values
to
fit
those
of
the
general
population
may
only
actually
be
the
values
of
avid
users,
who
value
fish
much
more
than
the
general
population
and
less
frequent
participants.

Overestimate:
Sampling
Effects
in
Survey
Similar
to
the
non­
response
and
recall
biases,
sampling
effects
is
the
phenomenon
where
surveying
may
allow
avid
participants
to
have
an
unequal
influence
on
the
overall
outcome
of
responses.
There
exists
a
bias
in
using
recreational
surveys
because
avid
participants
are
the
majority
of
people
surveyed
and
may
claim
to
go
fishing
much
more
than
the
average
person.
Because
of
the
dates,
times,
and
season
in
which
surveying
was
conducted,
there
may
have
been
more
avid
anglers
onsite
than
average
fisherman.
Data
collected
therefore
may
have
not
been
representative
of
the
aggregate
population
because
the
activity
levels
of
avid
fisherman
are
mostly
likely
overrepresented.
This
overrepresentation
may
assign
a
much
higher
value
to
the
fish
than
actually
valued
by
the
general
public.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
17
­
Overestimate:
EPA's
assumption
that
all
I&
E
fish
saved
will
be
a
benefit
to
either
recreational
or
commercial
users.

EPA
attributed
the
saved
I&
E
fish
as
a
gain
to
either
recreational
or
commercial
users;
this
may
lead
to
an
overestimate
of
benefits
as
not
all
I&
E
fish
may
be
caught
by
recreational
or
commercial
fishermen.
No
information
or
studies
were
provided
to
substantiate
the
underlying
assumption
that
an
increased
number
of
save
rates
led
to
a
proportional
increase
in
commercial
profit
or
recreational
catch
rates.
Increased
fish
yields
may
require
additional
infrastructure
investments
such
as
larger
nets,
ships,
and
more
processing
plants
before
any
of
the
fish
are
brought
to
market.
Direct
increases
in
fish
yield
may
not
translate
into
an
equal
growth
in
profit
margin
as
a
firm
may
choose
not
to
invest
in
more
production
capacity.

Uncertainty:
EPA
assigns
a
proportion
of
total
I&
E
losses
to
either
recreational
or
commercial
uses.

To
distinguish
losses,
EPA
used
the
Fishing
Statistics
Survey
(
MRFSS)
and
fish
yield
data
gathered
from
surveys
and
industry
reporting
to
estimate
the
total
amount
of
fish
caught
by
type
for
an
area.
By
knowing
how
many
fish
were
caught
by
either
recreational
or
commercial
users,
EPA
then
used
a
proportion
to
determine
how
much
of
the
I&
E
fish
lost
were
attributable
to
each
use.
For
example,
if
30
percent
of
one
type
of
fish
were
harvested
commercially
at
one
site,
EPA
contended
that
30
percent
of
I&
E
fish
of
that
same
type
could
be
assigned
as
a
loss
to
commercial
fishing.
The
remaining
70
percent
of
I&
E
fish
were
assigned
to
recreational
fishing
losses.
Similar
to
EPA's
assumption
that
I&
E
reductions
lead
directly
to
increased
catch
rates
and
fishery
yields,
the
proportion
estimates
may
overstate
the
benefits
by
assigning
the
greater
proportion
a
higher
value.
For
example,
in
the
Ohio
River
case
study,
a
Black
Crappie
is
assigned
a
recreational
value
between
$
1
and
$
5.02.
The
price
may
be
inflated
because
of
biases
built
into
the
RUM
analysis
explained
above.
Benefits
may
be
overestimated
if
more
fish
are
assigned
to
recreational
catches
than
actually
caught.

3.4.
Analysis
of
Benefits
using
HRC
approach
The
HRC
method
is
another
approach
to
valuing
losses
of
aquatic
resources
that
result
from
I&
E
of
organisms
by
a
cooling
water
intake
structure
(
CWIS).
The
HRC
method
focuses
on
replacing
natural
habitats
to
provide
an
environment
where
fish
can
reproduce
and
serve
as
prey
to
other
species.
This
method
considers
a
broad
range
of
ecological
and
human
services
associated
with
I&
E
losses
that
are
usually
undervalued
or
ignored
by
conventional
approaches,
such
as
commercial
and
recreational
impact
valuation
methods
that
only
account
for
those
species
and
life
stages
that
can
be
valued
directly
(
ex.
those
species
targeted
by
recreational
or
commercial
anglers).
On
the
other
hand,
the
HRC
method
defines
the
value
of
all
I&
E
losses
in
terms
of
the
expenditures
that
would
be
required
to
replace
all
organisms
lost
to
I&
E
at
a
CWIS
through
enhanced
natural
production
in
the
environment.

A
few
examples
of
ecological
and
human
services
addressed
by
HRC
include:

 
Decreased
number
of
popular
species
that
are
not
fished;
 
Decreased
number
of
special
status
(
e.
g.,
threatened
or
endangered)
species;
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
18
­
 
Increased
number
of
exotic
or
disruptive
species
that
compete
well
in
the
absence
of
species
lost
to
I&
E;
 
Decreased
local
biodiversity;
 
Disruption
of
predator­
prey
relationships;
 
Disruption
of
age
class
structures
of
species;
and
 
Disruption
of
natural
succession
processes.

The
following
are
the
steps
for
conducting
a
HRC
valuation
of
I&
E
losses:

1.
Quantify
I&
E
losses
by
species.
2.
Identify
habitat
requirements
of
I&
E
species.
3.
Identify
potential
habitat
restoration
actions
that
could
benefit
I&
E
species.
4.
Consolidate,
categorize,
and
prioritize
identified
habitat
restoration
alternatives.
5.
Quantify
the
expected
increases
in
species
production
for
the
prioritized
habitat
restoration
alternatives.
6.
Scale
the
habitat
restoration
alternatives
to
offset
I&
E
losses.
7.
Estimate
"
unit
costs"
for
the
habitat
restoration
alternatives.
8.
Develop
total
cost
estimates
for
I&
E
losses.

Uncertainty:
EPA's
choice
of
a
restoration
alternative
based
on
the
species
with
the
lowest
per
production
benefit
value
In
the
HRC
method,
species
with
the
lowest
per
unit
production
benefit
value
determines
the
amount
of
that
restoration
required.
For
example,
if
1
million
year­
one
gizzard
shad
and
1
million
year­
one
emerald
shiners
are
lost
to
I&
E
every
year,
if
wetland
restoration
is
the
most
effective
and
cost­
effective
restoration
alternatives
for
both
species,
and
if
local
wetland
restorations
have
been
documented
to
produce
500
gizzard
shad
per
acre
per
year
but
only
100
emerald
shiners
per
acre
per
year,
then
offsetting
the
I&
E
losses
of
both
species
requires
10,000
acres
(
not
2,000
acres)
of
successful,
sustained
wetland
restoration.
It
is
uncertain
if
restoration
of
10,000
acres
might
result
in
an
overpopulation
of
gizzard
shads,
which
might
have
a
negative
impact
on
emerald
shiners,
as
well
as
other
species.
This
can
lead
to
an
overestimate
of
benefits
if
a
smaller
number
of
acres
or
a
cheaper
restoration
alternative
is
more
effective
in
sustaining
the
original
population
of
fish
species.

3.5.
Review
of
Case
Studies
In
addition
to
the
uncertainties
and
potential
overestimates
of
EPA's
methodology
as
described
in
the
sections
above,
the
following
review
of
three
case
studies
highlights
specific
steps
in
EPA's
method
of
valuing
I&
E
losses
at
individual
facilities
that
may
result
in
a
miscalculation
of
benefits.

3.5.1.
Ohio
River
Case
Study
EPA
examined
the
impact
of
I&
E
at
nine
CWIS
along
a
500­
mile
stretch
of
the
Ohio
River.
The
RUM
approach
was
used
to
value
the
loss
of
I&
E
at
facilities
along
the
Ohio
River.
The
valuation
of
benefits
from
I&
E
reduction
may
be
an
overestimate
due
to
biases
in
data
collected
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
19
­
from
a
survey
of
recreational
anglers;
as
noted
in
Section
4.1
above,
survey
data
may
be
subject
to
recall,
non­
response
and
sampling
biases.
In
addition,
EPA's
assumption
that
all
I&
E
fish
will
be
caught
by
either
recreational
or
commercial
values
may
lead
to
an
overestimate
of
benefits.

Data
collected
from
these
nine
facilities
were
extrapolated
to
20
other
facilities
along
the
same
river
and
then
extrapolated
again
to
value
I&
E
losses
in
all
393
in­
scope
facilities
with
CWIS
in
freshwater.
EPA's
method
of
extrapolating
the
number
of
I&
E
species
found
in
9
case
study
facilities
to
the
20
other
facilities
located
along
the
Ohio
River
is
problematic
and
may
result
in
either
uncertainty
or
an
overestimate
of
I&
E
species
for
the
following
reasons.
5
Uncertainty:
Data
used
in
the
Ohio
River
case
study
dates
from
1977
to
1979
is
therefore
outdated.

The
use
of
outdated
data
from
1977
to
1979
to
estimate
current
I&
E
may
result
in
an
overestimate
or
underestimate
of
benefits.
As
the
proportion
and
number
of
fish
species
in
the
Ohio
River
have
changed
in
the
past
20
years
for
reasons
other
than
CWIS,
the
use
of
such
outdated
data
leads
to
uncertainty
in
EPA's
estimate
of
benefits
from
reducing
I&
E
in
freshwater.

Overestimate:
Some
I&
E
fish
may
have
been
impinged
because
of
cold
shock
during
the
winter
season,
rather
than
CWIS.

EPA's
method
of
determining
the
number
of
gizzard
shad
can
lead
to
an
overestimate
of
potentially
I&
E
species.
Studies
have
shown
that
during
extremely
cold
winters,
gizzard
shads
may
suffer
from
cold
shock
in
which
their
bodies
become
fatigued
and
they
eventually
die.
During
these
low
temperatures,
more
fish
are
impinged
as
they
do
not
have
the
strength
to
swim
away
from
the
CWIS
intakes.
As
a
result,
the
number
of
impinged
fish
during
these
winter
seasons
increases.
If
these
fish
are
expected
to
die
from
the
cold
shock,
then
the
number
of
I&
E
species
measured
at
that
site
may
result
in
an
overestimate
of
the
number
of
species
that
die
from
impingement.

Uncertainty:
Facilities
used
inconsistent
sampling
methods.

Sampling
methods
and
timing
varied
across
each
CWIS.
Variation
in
the
time
span
of
monitoring
and
the
time
intervals
in
which
fish
were
collected
from
the
traveling
screens
may
have
affected
I&
E
estimates.
It
is
important
to
recognize
such
uncertainties
because
these
data
sets
were
the
basis
for
determining
I&
E
benefits
which
were
extrapolated
to
20
other
CWIS
in
the
region.
For
example,
at
the
Miami
Fort
Plant,
I&
E
data
was
collected
during
April
1977
through
March
1978,
while
data
at
Philip
Sporn
Plant
was
collected
between
March
1978
and
February
1979.
Not
only
did
the
seasons
vary,
so
did
the
methods.
For
example,
the
Philip
Sporn
Plant,
whose
bi­
weekly
collections
lasted
28
hours
with
fish
removed
from
the
traveling
5K.
J.
Harman
of
West
Virginia
University
prepared
a
report
entitled,
"
Review
of
the
Biological
Validity
of
EPA's
Methods
for
Baseline
Loss
Estimates
at
Case
Study
Facilities
and
Extrapolation
to
National
Estimates,"
dated
August
4,
2002.
In
this
report,
Harman
specifically
reviews
EPA's
methodology
of
extrapolation
for
the
J.
R.
Whiting
facility
and
the
Ohio
River
Watershed
Study.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
20
­
screens
in
4
hour
intervals.
In
contrast,
Miami
Fort
samples
were
collected
only
once
a
week
for
two
different
12
hour
blocks
of
time.

3.5.2.
Brayton
Point
Case
Study
Brayton
Point
Station
is
located
on
Mount
Hope
Bay
in
the
Town
of
Somerset,
Massachusetts.
It
is
located
in
the
upper
embayment
of
Narragansett
Bay.
There
are
eight
units
at
Brayton
Point.
Units
1­
3
are
coal­
fired
steam­
electric
generators
(
from
Taunton
River).
Currently
trash
racks
and
continuously­
rotating
traveling
screen
are
used
to
divert
fish
from
the
CWIS;
according
to
the
EPA,
neither
technology
is
particularly
effective.
In
1994,
Unit
4
was
changed
from
a
closed­
cycle
cooling
system
to
a
once­
through
cooling
system
(
from
Lee
River).
At
Unit
4,
an
angled
screen
assembly
with
fish
buckets
and
fish
diversion/
return
system
was
installed
to
reduce
impingement
mortality.

EPA
noted
that
the
conversion
of
Unit
4
to
a
once­
through
cooling
system
resulted
in
a
41
percent
increase
in
coolant
flow
and
increased
thermal
discharge
to
the
bay.
A
subsequent
analysis
of
fisheries
data
conducted
by
the
Rhode
Island
Division
of
Fish
and
Wildlife
showed
an
87
percent
reduction
in
finfish
abundance
in
Mt.
Hope
Bay
coincident
with
the
Unit
4
modification.

Uncertainty:
EPA
assumption
that
a
closed­
cycle
cooling
system,
rather
than
a
once­
through
can
reduce
the
loss
of
fish
species
by
87
percent.

EPA
suggests
that
the
use
of
a
closed­
cycle
cooling
system,
rather
than
a
once­
through,
can
reduce
the
loss
of
fish
species
by
87
percent.
However,
it
is
possible
that
the
installation
of
a
closed­
cycle
cooling
system
at
a
randomly
chosen
facility
located
in
an
estuary
waterbody
type
may
not
experience
similar
reductions
in
losses.
The
controversy
in
recent
decades
about
the
decline
in
fish
populations
at
Mt.
Hope
Bay
suggests
that
this
decline
is
significant
and
rather
unusual
given
the
rapid
amount
of
reduction.
In
order
to
determine
if
an
87
percent
reduction
in
losses
is
an
appropriate
estimate
for
Phase
III
facilities,
the
EPA
should
conduct
a
case
study
of
a
randomly
chosen
facility,
with
an
intake
of
less
than
50
MGD,
which
has
converted
from
a
oncethrough
to
a
closed­
cycle
cooling
system.
The
cost
of
installing
a
closed­
cycle
cooling
system
at
a
smaller
facility
should
also
be
considered
as
it
may
not
be
economically
feasible
for
that
facility
to
absorb
associated
costs.

In
addition
to
determining
the
efficiency
of
a
closed­
cycle
cooling
system
and
the
costs
of
installation
at
smaller
facilities,
EPA
should
provide
more
concrete
support
for
its
approach
to
estimating
the
economic
benefits
of
reducing
I&
E
at
Brayton
Point.
There
are
several
instances
where
EPA
estimates
can
lead
to
an
overestimate
of
benefits.

Uncertainty:
EPA
chose
restoration
alternatives
based
on
available
data.

Based
on
the
HRC
method
of
valuation,
EPA
identified
potential
habitat
restoration
actions
that
could
benefit
I&
E
species,
six
potential
habitat
restoration
alternatives
were
suggested.
However,
only
four
were
discussed
and
suggested
partially
because
only
those
four
could
be
quantified
with
the
HRC
valuation.
This
suggests
that
there
is
some
deficiency
in
using
the
HRC
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
21
­
method
for
valuation.
The
inability
to
assign
a
value
to
potential
habitat
restoration
techniques
leads
to
a
habitat
restoration
chosen
based
on
valuation
capabilities
instead
of
actual
efficiency
in
reducing
the
number
of
I&
E
species.
In
cases
such
as
this,
EPA
may
consider
selecting
a
different
method
for
quantifying
I&
E
losses.

In
consolidating,
categorizing,
and
prioritizing
identified
habitat
restoration
alternatives,
EPA
mentioned
that
water
quality,
one
of
the
restoration
methods
unable
to
be
quantified,
was
found
to
be
a
preferred
alternative
but
no
project
was
identified.
The
EPA
should
provide
further
explanation
regarding
the
constraints
in
using
this
method
in
order
to
support
its
preferred
restoration
alternatives.

Uncertainty:
EPA
assumed
all
I&
E
fish
will
be
adequately
replaced
given
the
implementation
of
four
habitat
restoration
alternatives.

After
the
expected
increases
in
species
production
for
the
prioritized
habitat
restoration
alternatives
were
quantified,
EPA
estimated
losses
based
on
four
preferred
habitat
restoration
alternatives.
The
preferred
restoration
alternatives
are:

1.
Restore
submerged
aquatic
vegetation
(
SAV)
2.
Restore
tidal
wetlands
3.
Create
artificial
reefs
4.
Install
fish
passageways
to
increase
access
to
suitable
spawning
habitats
SAV
restoration
accounted
for
only
0.13
percent
of
total
I&
E
losses
for
all
fish
species
impinged
and
entrained
at
Brayton
Point,
tidal
wetland
restoration
accounted
for
13.75
percent,
artificial
reef
development
accounted
for
0.80
percent,
and
installed
fish
passageways
accounted
for
1.59
percent.
As
a
result,
unidentified
habitat
restoration
alternatives
accounted
for
83.72
percent
of
total
I&
E
losses
for
all
fish
species.
The
EPA
assumed
that
all
these
fish,
which
constitute
the
majority
of
fish
impinged
and
entrained,
will
be
adequately
replaced
given
implementation
of
the
four
quantifiable
restoration
alternatives.
This
can
lead
to
further
uncertainty
about
the
actual
value
of
reducing
I&
E
at
Brayton
Point
based
on
the
HRC
valuation
method.

Overestimate:
EPA's
sampling
of
I&
E
fish
to
value
the
SAV
restoration
alternative.

To
determine
the
value
of
SAV
restoration
alternatives,
the
EPA
calculated
species
abundance
based
on
six
2­
minute
samplings
at
Buzzards
Bay
(
one
of
three
sampling
locations)
during
late
summer
during
which
time
eelgrass
is
the
most
abundant.
These
samplings
would
show
the
high­
end
of
species
abundance
for
the
fish,
which
might
overestimate
the
value
of
SAV,
and
lead
to
an
overestimate
of
benefits.

Uncertainty:
EPA
assumed
fish
abundancy
is
the
same
in
restored
and
undisturbed
habitats.

In
adjusting
sampling
results
to
estimate
annual
average
increase
in
production,
EPA
assumed
that
there
is
no
difference
between
restored
and
undisturbed
SAV
habitats.
It
seems
unlikely
that
restoration
will
produce
the
same
abundance
of
fish
as
when
the
habitat
was
undisturbed.
A
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
22
­
better
methodology
would
be
to
conduct
a
study
of
fish
abundancy
in
restored
SAV
habitats
to
determine
if
this
approximation
is
true.
It
is
uncertain
if
this
assumption
leads
to
either
an
underestimate
or
overestimate
of
benefits.

3.5.3.
J.
R.
Whiting
Case
Study
The
J.
R.
Whiting
Plant,
located
in
Michigan
on
Lake
Erie,
has
an
intake
flow
of
308
MGD.
Before
installation
of
a
deterrent
net
in
1980
to
reduce
impingement,
some
21.5
million
age
1
equivalents
were
lost
to
impingement
at
the
facility
each
year.
These
losses
were
reduced
by
nearly
90
percent
with
application
of
the
deterrent
net.
While
this
case
study
illustrates
the
potential
benefits
of
impingement
reduction
technologies,
the
case
study
fails
to
provide
definitive
proof
that
such
results
can
be
obtained
at
other
facilities,
especially
facilities
of
significantly
smaller
size.

EPA
estimated
the
economic
value
of
I&
E
fish
by
adding
the
costs
of
a
benefits
transfer
technique
(
demand­
driven)
and
a
streamlined
HRC
approach
(
supply­
driven).
Based
on
this
methodology,
the
total
baseline
economic
loss
determined
to
be
$
1,235,000
for
impingement
and
$
1,703,000
for
entrainment.
However,
this
may
be
an
overestimate
of
the
actual
costs
of
reducing
I&
E.
The
EPA
stated
that
to
meet
the
schedule
of
the
316(
b)
existing
sources
rule,
a
streamlined
HRC
valuation
of
I&
E
losses
at
the
J.
R.
Whiting
facility
was
required.
This
streamlined
approach
resulted
in
the
use
of
several
assumptions
based
on
available
and
comparable
data
that
might
result
in
an
overestimate
of
benefits.
The
following
provides
examples
of
assumptions
EPA
made
in
an
effort
to
complete
the
case
study
by
the
deadline
set
for
the
existing
sources
rule.
The
examples
are
accompanied
by
recommendations
for
gathering
additional
data
to
make
alternative
calculations
that
might
produce
a
better
estimate
of
the
benefits.

Overestimate:
EPA's
use
of
data
to
determine
the
efficiency
of
a
deterrent
net
in
reducing
I&
E.

EPA's
method
for
calculating
the
difference
between
impingement
prior
to
and
after
the
installation
of
the
deterrent
net
may
result
in
an
overestimate
of
the
number
of
losses
reduced
from
the
net.
EPA
used
impingement
monitoring
data
from
2
years
(
1978
 
1979)
to
estimate
the
annual
number
of
fish
impinged
without
a
net,
and
data
from
10
years
(
1981­
1991)
to
estimate
the
annual
number
of
fish
impinged
with
a
net.

An
estimate
of
the
potential
benefits
of
installing
a
deterrent
net
based
on
a
comparison
of
average
annual
impingement
between
the
2
year
period
prior
to
1980
and
10
year
period
after
1980
may
produce
skewed
results.
The
reason
is
based
on
EPA's
description
of
Lake
Erie
as
having
undergone
drastic
biological
changes
during
the
past
20
years,
from
1980
to
2000.
EPA
noted
that
water
clarity
has
improved
dramatically
as
a
result
of
stricter
water
pollution
controls
as
well
as
filtering
by
expanding
populations
of
introduced
zebra
mussel.
As
a
result,
impingement
monitoring
from
1981
to
1991
might
reflect
not
just
the
effects
of
the
deterrent
net,
but
also
changes
in
the
density
of
certain
fish
species
due
to
the
biological
changes.
A
better
estimate
of
the
reduction
of
losses
from
the
deterrent
net
may
be
a
comparison
of
annual
impingement
from
the
2
year
period
prior
to
the
installation
of
the
net
(
1978­
1979)
to
the
2
year
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
23
­
period
after
installation
of
the
net
(
1981­
1982).
The
results
of
the
comparison
of
age­
1
equivalent
fish
are
shown
in
the
Table
5
below.
Based
on
a
comparison
of
equal
monitoring
periods
within
a
5
year
time
period
(
1978­
1982),
the
results
show
that
the
net
barrier
reduces
impingement
of
age
1
equivalents
by
86
percent,
rather
than
92.5
percent.

Table
5:
Reduction
of
Impinged
Fish
Prior
to
and
After
Deterrent
Net
J.
R.
Whiting
Facility
Annual
Impingement
Reduction
of
I&
E
from
Net
2
years
prior
to
net
21,493,215
­­­
10
years
after
net
1,612,966
92.5
percent
2
years
after
net
3,018,332
86.0
percent
Uncertainty:
EPA's
estimate
of
I&
E
fish
that
would
otherwise
be
caught
by
recreational
or
commercial
fishermen.

For
fish
caught
both
recreationally
and
commercially,
EPA
assumed
that
50
percent
of
the
estimated
catch
of
the
I&
E­
impacted
fish
are
assigned
to
a
loss
in
commercial
landings
and
the
remaining
50
percent
of
the
estimated
total
number
of
losses
due
to
I&
E
are
assigned
to
the
recreational
landings.
For
the
two
species,
channel
catfish
and
white
bass,
the
recreational
values
are
7
to
8
times
their
respective
commercial
values.
As
a
result,
if
these
species
are
actually
caught
primarily
for
commercial
purposes,
than
the
economic
benefits
of
reducing
I&
E
are
overstated.

Uncertainty:
EPA
chose
restoration
alternatives
based
on
available
data.

EPA
chose
coastal
wetland
restoration
as
the
preferred
restoration
alternative
because
the
J.
R.
Whiting
facility
is
located
near
some
wetlands,
and
there
is
readily
available
information
regarding
restoration
by
wetlands.
This
method
of
choosing
the
restoration
alternative
by
association
and
availability
of
data
may
result
in
the
selection
of
an
alternative
that
is
not
as
effective
as
others.

Uncertainty:
Impacts
of
coastal
wetland
restoration
in
Green
Bay
(
Lake
Michigan)
were
used
as
estimates
for
impacts
of
restoration
for
the
J.
R.
Whiting
site
(
Lake
Erie).

Data
from
Green
Bay
(
Lake
Michigan)
coastal
wetlands
was
used
to
approximate
impacts
of
coastal
wetland
restoration
at
the
J.
R.
Whiting
facility.
The
EPA
stated
"
because
of
the
close
match
between
the
physical
habitats
of
southern
Green
Bay
and
western
Lake
Erie,
and
the
confirmation
of
similar
species
between
the
sites,
EPA
estimated
densities
for
each
southern
Green
Bay
species
and
used
them
as
a
proxy
for
direct
measurements
of
potential
increased
production
following
wetlands
restoration."
EPA
stated
that
the
restoration
efforts
in
Green
Bay
would
produce
the
same
densities
of
fish
species
as
restoration
efforts
in
Lake
Erie.
This
assumption
might
lead
to
an
overestimate
if
the
density
of
fish
species
is
greater
in
Lake
Erie
than
in
Green
Bay.
EPA
should
follow
its
own
recommendation
of
taking
"
direct
measurements
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
24
­
of
densities
of
each
species
before
and
after
actual
wetland
habitat
restoration
in
western
Lake
Erie
to
test
this
assumption
and
improve
the
reliability
of
the
HRC
valuation
for
J.
R.
Whiting."

Overestimate:
EPA
chose
the
upper
bound
instead
of
a
mid­
point
average
to
calculate
the
required
scale
of
restoration.

The
EPA
chose
90th
and
50th
percentile
to
bound
the
estimate
of
the
required
scale
of
restoration.
Based
on
the
HRC
methodology,
the
number
of
acres
to
be
restored
by
a
particular
restoration
alternative
is
determined
by
the
specie
that
requires
the
largest
number
of
acres
for
potential
restoration
of
all
I&
E
species.
However,
the
100th
percentile
of
offsetting
all
losses
of
rainbow
smelt
was
considered
by
EPA
to
be
unjustifiable
given
the
assumptions
used
in
the
streamlined
approach.
Therefore,
the
90th
percentile
upper
bound
was
used
as
an
approximation
of
economic
losses
based
on
the
HRC
method
in
calculating
the
range
of
losses.
The
use
of
the
upper
bound,
instead
of
the
mid­
point
average
between
the
upper
(
90th
percentile)
and
lower
(
50th
percentile)
limits,
is
contrary
to
EPA's
use
of
the
mid­
point
in
approximating
economic
losses
based
on
the
benefits
transfer
approach.
Table
5
shows
the
difference
between
EPA's
estimate
of
the
range
using
the
90th
upper
bound
percentile
and
an
alternative
estimate
using
the
mid­
point
between
the
upper
and
lower
bound
estimates.

Table
5:
Estimate
of
Economic
Losses
of
I&
E
at
the
J.
R.
Whiting
Facility
using
the
Mid­
Point,
rather
than
the
Upper
Bound
Method
Valuation
Approach
Range
Impingement
Entrainment
Benefits
Transfer
Approach
Average
$
351,118
$
41,493
High
$
1,210,000
$
1,669,000
HRC
Method
Low
NA
NA
EPA
Range
$
351,000
­
$
1.2
million
$
41,000
­
$
1.7
million
Benefits
Transfer
Approach
Average
$
351,118
$
41,493
High
$
1,210,000
$
1,669,000
Low
$
200,000
$
200,000
HRC
Method
Average
$
705,000
$
934,500
Alternative
Range
$
351,000
­
$
705,000
$
41,000
­
$
935,000
Overestimate:
EPA
chose
to
exclude
the
Angling
Index
in
extrapolating
losses
at
J.
R.
Whiting
to
losses
in
all
Great
Lake
waterbody
types.

The
EPA
used
estimates
from
the
J.
R.
Whiting
case
study
to
determine
the
benefits
of
reducing
I&
E
at
all
facilities
in
the
Great
Lakes
waterbody
type.
Extrapolation
using
the
J.
R.
Whiting
case
study
can
significantly
magnify
overestimates
in
the
methodology
as
the
case
study
was
conducted
using
a
streamlined
approach
because
of
limited
time.
Unlike
EPA's
method
of
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
25
­
extrapolating
individual
facility
baseline
economic
losses
to
the
national
level
for
all
facilities
in
that
waterbody
type,
the
EPA
made
two
choices
in
calculation
that
resulted
in
a
larger
estimate
of
benefits.
These
two
choices
include:

1.
Use
of
the
upper
bound
range
of
I&
E
costs.
2.
Use
of
only
the
flow­
based
index
instead
of
both
the
angling­
based
and
flow­
based
losses.

For
oceans
and
the
Great
Lakes,
EPA
extrapolated
the
HRC
findings
using
only
the
MGD­
based
index
because
the
angler­
based
index
was
not
relevant
since
only
a
small
share
of
the
impacted
fish
are
recreational.
In
addition,
EPA
used
the
upper
bound
range
of
I&
E
costs
as
the
best
estimate
for
baseline
losses.
An
alternative
best
estimate
of
I&
E
for
all
Great
Lakes
was
calcuated
using
the
mid­
point
range
and
both
the
angling­
based
and
flow­
based
indexes;
results
show
that
national
economic
losses
of
I&
E
for
all
Great
Lake
facilities
are
2
to
3
times
smaller
using
this
approach.
As
shown
in
Table
6,
this
alternatives
method
leads
to
I&
E
of
$
13,000
and
$
14,300,
as
compared
to
EPA's
estimate
of
$
31,500
and
$
43,400,
respectively.

Table
6:
Estimate
of
Economic
Losses
of
I&
E
at
the
J.
R.
Whiting
Facility
using
the
Mid­
Point
Average
with
the
Angling
Index
(
Dollars)

Alternative
EPA
Chosen
Impingement
Entrainment
Impingement
Entrainment
Facilities
Mid­
point
High­
Point
Mid­
point
Highpoint
Midpoint
High­
Point
Midpoint
Highpoint
J.
R.
Whiting
$
797
$
1,235
$
873
$
1,703
$
797
$
1,235
$
873
$
1,703
All
Other
In­
Scope
$
12,231
$
18,966
$
13,398
$
26,154
$
19,523
$
30,271
$
21,385
$
41,745
Total
$
13,028
$
14,271
$
31,506
$
43,448
Overestimate:
EPA's
use
of
I&
E
data
at
J.
R.
Whiting
to
extrapolate
I&
E
losses
in
all
Great
Lake
waterbody
types.

Impingement
and
entrainment
data
from
the
J.
R.
Whiting
facility,
located
in
western
Lake
Erie,
was
used
to
represent
all
facilities
in
the
Great
Lakes.
6
This
method
may
result
in
an
overestimate
of
the
number
of
I&
E
species
because
western
Lake
Erie
is
noted
to
be
the
most
biologically
productive
area
among
the
entire
area
covered
by
the
Great
Lakes.
Several
studies
dated
from
1979
to
1993
support
this
finding.
The
type
and
value
of
species
found
in
western
Lake
Erie
varies
from
other
parts
of
the
Great
Lake
System.
For
example,
in
Lake
Erie,
most
of
the
clupeid
(
soft­
finned
bony
fish
that
have
oily
flesh,
a
narrow
body,
and
a
forked
tail,
e.
g.

6
K.
J.
Harman
of
West
Virginia
University
prepared
a
report
entitled,
"
Review
of
the
Biological
Validity
of
EPA's
Methods
for
Baseline
Loss
Estimates
at
Case
Study
Facilities
and
Extrapolation
to
National
Estimates,"
dated
August
4,
2002.
In
this
report,
Harman
specifically
reviews
EPA's
methodology
of
extrapolation
for
the
J.
R.
Whiting
facility
and
the
Ohio
River
Watershed
Study.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
26
­
herrings,
sardines,
menhadens,
and
shad)
grouping
is
represented
by
gizzard
shad,
while
in
other
Great
Lakes,
these
groups
are
dominated
by
the
alewife
species.
While
the
gizzard
shad
is
valued
at
$
0.15/
lb,
the
alewife
has
no
commercial
value.
As
a
result,
an
estimate
of
the
economic
benefits
of
saving
gizzard
shad
from
I&
E
in
Western
Lake
Erie
will
result
in
an
overestimate
of
benefits
when
applied
to
the
entire
Great
Lake
system.

Uncertainty:
EPA's
use
of
outdated
I&
E
data
at
J.
R.
Whiting
to
extrapolate
to
all
Great
Lake
facilities.

Another
concern
regarding
EPA's
method
of
extrapolation
in
the
J.
R.
Whiting
case
study
is
EPA's
use
of
outdated
I&
E
data,
with
the
most
recent
data
for
entrainment
dating
to
1979
and
the
most
recent
data
for
impingement
dating
to
1991.
The
composition
and
abundance
of
fish
within
the
Great
Lakes
System
has
changes
over
the
past
10
to
20
years;
based
on
data
collected
by
the
United
State
Fish
and
Wildlife
Services
(
USFWS)
 
Sandusky
Biological
Station
located
in
Lake
Erie,
gizzard
shad
have
declined
in
abundance
from
16
percent
to
4
percent
based
on
the
difference
between
data
gathered
during
the
five
years
periods
of
1978­
1982
and
1996­
2001.
The
Station
used
a
catch
per
unit
effort
(
CPUE,
fish/
hour)
of
age­
0
fish
during
bottom
trawls
at
their
East
Harbor
station.
As
a
result,
the
use
of
outdated
data
in
estimating
potential
benefits
can
lead
to
either
uncertainty
or
an
overestimate
of
the
number
of
I&
E
species
as
abundance
has
changed
in
recent
years.

3.6.
Summary
of
Benefit
Uncertainties
Table
7
on
the
following
page
provides
a
summary
of
the
uncertainties
and
possible
overestimates
in
EPA's
methodology
of
calculating
potential
benefits
of
regulating
facilities
in
Phase
III
of
the
316(
b)
rulemaking.
Recommendations
for
addressing
some
of
these
uncertainties
are
also
provided.

Table
7:
Summary
of
Analysis
and
Recommendations
Impact
on
Benefits
Critique
Recommendation
EPA's
Selection
of
Case
Studies
Uncertainty
EPA
did
not
consider
facilities
that
withdraw
less
than
50
MGD.
Conduct
case
studies
of
facilities
with
intakes
of
less
than
50
MGD.
Uncertainty
EPA's
selection
of
case
studies
is
biased.
Conduct
case
studies
at
randomly
selected
facilities.
Uncertainty
EPA
does
not
account
for
density
dependence
in
estimating
benefits
of
reducing
I&
E
losses
based
on
replacement
and
restoration
costs.
Consider
a
broader
range
of
facilities
to
determine
longer­
term
impacts
of
CWIS
on
fish
species.

EPA's
Methodology
for
Calculating
Benefits
Uncertainty
EPA's
use
of
non­
standardized,
facilitymonitored
I&
E
data
in
their
case
studies.
Develop
standards
for
monitoring
I&
E
so
that
data
is
not
biased
and
inconsistent.
Uncertainty
EPA's
estimate
of
nonuse
benefits
as
50
percent
of
estimated
recreational
fishing
benefits
may
be
overstated.
Estimate
nonuse
benefits
in
the
context
of
CWIS
improvements.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
27
­
Impact
on
Benefits
Critique
Recommendation
Uncertainty
EPA's
estimate
of
costs
as
benefits.
Alternative
approach
has
not
been
determined.
Uncertainty
EPA's
use
of
economic
losses
at
a
small
number
of
facilities
to
extrapolate
to
national
baseline
losses.
Consider
using
an
average
of
I&
E
from
more
than
one
facility
to
extrapolate
to
national
losses.
Overestimate
EPA's
use
of
the
Flow
Index
to
estimate
national
baseline
losses.
Consider
data
gathered
from
facilities
in
which
I&
E
losses
are
not
strictly
proportional
to
flow.
Overestimate
EPA's
use
of
the
Angling
Index
to
estimate
national
baseline
losses.
Alternative
approach
has
not
been
determined.

Analysis
of
Benefits
using
the
RUM
Approach
Overestimate
EPA's
use
of
data
from
surveys
of
recreational
fishermen
to
value
recreational
fish
may
be
biased
due
to
recall,
non­
response
and
sampling
effects.
Alternative
approach
has
not
been
determined.

Overestimate
EPA's
assumption
that
all
I&
E
fish
saved
will
be
a
benefit
to
either
recreational
or
commercial
users.
Collect
data
on
the
percentage
of
I&
E
fish
caught
recreationally
and
commercially
to
determine
the
value
of
I&
E
losses.
Uncertainty
EPA
assigns
a
proportion
of
total
I&
E
losses
to
either
recreational
or
commercial
users.
Collect
data
on
the
percentage
of
I&
E
fish
caught
recreationally
and
commercially
to
determine
the
value
of
I&
E
losses.
Analysis
of
Benefits
using
the
HRC
Approach
Overestimate
EPA's
choice
of
a
restoration
alternative
based
on
the
species
with
the
lowest
per
production
benefit
value.
Conduct
studies
to
measure
the
costs
and
benefits
of
specific
restoration
alternatives
to
determine
the
most
efficient
method
of
restoring
the
population
of
I&
E
fish
species.
Ohio
River
Case
Study
(
Miami
Fort
Plant)
Uncertainty
Data
used
in
study
dates
from
1977­
1979
is
outdated.
Sample
I&
E
data
along
the
Ohio
River
to
gather
current
data.
Overestimate
Some
I&
E
fish
may
have
been
impinged
because
of
cold
shock
during
the
winter
season,
rather
than
CWIS.
Measure
I&
E
at
CWIS
during
different
seasons
throughout
the
year.

Uncertainty
Facilities
used
inconsistent
sampling
methods.
Standardize
methods
for
gathering
I&
E
data.
Brayton
Point
Case
Study
Uncertainty
EPA's
assumption
that
a
closed­
cycle
cooling
system,
rather
than
a
once­
through
can
reduce
the
loss
of
fish
species
by
87
percent.
To
test
this
assumption,
conduct
a
case
study
of
a
randomly
chosen
facility,
with
an
intake
of
less
than
50
MGD,
which
has
converted
from
a
once­
through
to
a
close­
cycle
cooling
system.
Uncertainty
EPA
chose
restoration
alternatives
based
on
available
data.
Conduct
studies
to
determine
the
effectiveness
of
all
restoration
alternatives
considered
(
for
each
waterbody
type).
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
28
­
Impact
on
Benefits
Critique
Recommendation
Uncertainty
EPA
assumed
all
I&
E
fish
will
be
adequately
replaced
given
the
implementation
of
four
habitat
restoration
alternatives.
Conduct
studies
to
measure
the
adequacy
of
the
selected
restoration
alternatives
in
restoring
the
population
of
I&
E
fish
species.
Overestimate
EPA's
sampling
of
I&
E
fish
to
value
the
SAV
restoration
alternative.
Conduct
sampling
throughout
the
year.
Uncertainty
EPA
assumed
fish
abundancy
is
the
same
in
restored
and
undisturbed
habitats.
To
test
this
assumption,
conduct
a
study
that
measures
the
abundancy
of
fish
species
in
restored
SAV
habitats
and
undisturbed
habitats.
J.
R.
Whiting
Case
Study
Overestimate
EPA's
use
of
data
to
determine
the
efficiency
of
a
deterrent
net
in
reducing
I&
E.
Determine
the
efficiency
of
a
deterrent
net
using
a
comparable
set
of
data.
Uncertainty
EPA's
estimate
of
I&
E
fish
that
would
otherwise
be
caught
by
recreational
or
commercial
fishermen.
Collect
data
on
the
percentage
of
fish
caught
recreationally
and
commercially
to
determine
their
appropriate
values.
Uncertainty
EPA
chose
restoration
alternatives
based
on
available
data.
Conduct
studies
to
determine
the
effectiveness
of
all
restoration
alternatives
considered.
Uncertainty
Impacts
of
coastal
wetland
restoration
in
Green
Bay
(
Lake
Michigan)
were
used
as
estimates
for
impacts
of
restoration
for
the
J.
R.
Whiting
site
(
Lake
Erie).
To
test
this
assumption,
conduct
a
study
that
measures
the
abundancy
of
fish
species
in
restored
and
undisturbed
coastal
wetland
habitats.
Overestimate
EPA
chose
the
upper
bound,
instead
of
a
mid­
point
average
to
calculate
the
required
scale
of
restoration
Chose
a
consistent
mid­
point
average
to
calculate
the
required
scale
of
restoration.
Overestimate
EPA
chose
to
exclude
the
Angling
Index
in
extrapolating
losses
at
J.
R.
Whiting
to
losses
in
all
Great
Lake
waterbody
types.
Use
a
consistent
method
of
extrapolating
facility­
based
data
to
national
baseline
losses.

Overestimate
EPA's
use
of
I&
E
data
at
J.
R.
Whiting
to
extrapolate
I&
E
losses
in
all
Great
Lake
waterbody
types.
Use
data
from
more
than
one
facility
to
estimate
I&
E
losses
for
all
facilities
located
in
the
Great
Lakes.
Uncertainty
EPA's
use
of
outdated
I&
E
data
at
J.
R.
Whiting
to
extrapolate
to
all
Great
Lake
facilities.
Use
data
collected
by
the
USFWS
to
estimate
I&
E
at
the
J.
R.
Whiting
facility
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
29
­
4.
Review
and
Critique
of
Phase
III
Benefits
Estimates
A
Notice
of
Data
Availability
(
NODA),
published
in
the
Federal
Register
on
March
19,
2003
provides
a
revised
methodology
developed
by
the
Environmental
Protection
Agency
(
EPA)
to
estimate
the
dollar­
values
of
national
benefits
resulting
from
proposed
Phase
II
regulations.
The
NODA
outlines
areas
where
the
benefits
estimation
methodology
was
revised
in
response
to
individual
and
industry
comment.
EPA
will
use
the
same
approach
to
estimate
Phase
III
benefits.
This
section
presents
a
brief
summary
and
critique
of
EPA's
proposed
revised
approach.

The
New
Regional
Approach
In
the
initial
cost­
benefit
analysis
of
the
Phase
II
rule,
the
EPA
derived
national
benefits
by
extrapolating
benefits
estimates
from
eight
case
study
water
bodies.
In
the
revised
methodology,
EPA
proposes
to
analyze
benefits
on
the
regional­
scale.
The
regional
approach
estimates
monetary
values
for
nonuse
fish
and
improved
recreational
and
commercial
catch.
The
estimated
benefits
from
all
the
regions
will
then
be
aggregated
to
form
national
benefits
estimation.
The
eight
study
regions
and
the
number
of
facilities
in
each
region
are
presented
in
the
table
below.
Note
that
the
number
of
facilities
in
the
table
below
references
the
distribution
of
Phase
II
facilities.
Regional
distribution
of
Phase
III
facilities
was
unavailable.
More
detailed
information
regarding
Phase
III
facilities
is
presented
in
Section
5.

Table
8:
Eight
Regions
New
Regions
for
the
Benefit
Analysis
Region
Number
of
Phase
II
Facilities
Great
Lakes
55
Interior
U.
S.
(
Aggregate
of
facilities
not
located
on
the
coast)
372
North
Atlantic
21
Mid
Atlantic
44
South
Atlantic
14
Gulf
of
Mexico
23
Northern
California
8
Southern
California
11
TOTAL
FACILITIES
548
In
the
revised
methodology
EPA
proposes
to
use
the
following:

1.
Updated
Market
data
to
derive
commercial
fishing
benefits.
2.
Regional
Random
Utility
Models
(
RUM)
based
on
non­
market
use
studies
and
surveys
to
derive
recreational
fishing
benefits,
and
3.
A
new
"
Benefits
Transfer
Process"
to
determine
non­
use
benefit
values
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
30
­
Revised
Benefits
Methodology
As
previously
stated,
the
EPA
is
developing
benefits
estimates
of
I&
E
reduction
for
each
of
the
eight
regions.
Each
regional
estimate
will
involve
calculation
of
recreational,
commercial,
and
nonuse
values.
The
aggregated
sum
of
all
the
regions
will
represent
the
national
benefits
of
the
proposed
rule.
The
following
is
the
four­
step
process
proposed
by
the
EPA
for
estimating
direct
use
benefits
for
the
regions.
After
a
brief
explanation
of
each
step,
the
process
will
be
critiqued
and
be
assigned
one
or
more
of
the
following
three
categories
that
describe
key
flaws
in
the
approach:

1.
Overestimate
2.
Underestimate
3.
Uncertainty
Step
One:
Estimate
regional
impingement
and
entrainment
losses
To
calculate
impingement
and
entrainment
(
I&
E)
the
agency
will
use
data
from
facilities
with
I&
E
data
and
extrapolate
those
figures
to
derive
data
for
all
other
facilities.
The
figures
for
each
facility
in
the
region
are
then
to
be
summed
to
estimate
a
regional
total.

Uncertainty:
Lack
of
timely
and
reliable
data.

While
the
regional
approach
to
estimating
national
benefits
can
potentially
be
an
improvement
in
methodology,
the
NODA
does
not
address
concerns
regarding
the
lack
of
timely
and
reliable
data
sources
for
I&
E
data.
The
agency
is
merely
soliciting
I&
E
studies
from
facilities,
which
is
data
that
is
not
representative
of
the
total
population
of
facilities.
At
the
time
of
publication
of
the
NODA,
EPA
cites
it
only
has
"
received
approximately
20
studies
from
inland
facilities."
7
Without
a
complete
set
of
reliable
data
from
each
facility
the
final
analysis
will
contain
a
large
degree
of
uncertainty,
because
the
sample
will
not
be
random
and
therefore
be
subject
to
selfselection
bias.

The
issue
concerning
the
lack
of
standards
for
I&
E
monitoring
and
reporting
is
also
not
addressed
in
the
NODA.
EPA
relies
solely
on
reporting
from
individual
facilities.
If
each
facility
uses
its
own
procedure
of
I&
E
recording,
the
resulting
benefits
analysis
may
be
unreliable.

Overestimate:
Impingement
and
age
one
equivalents
In
the
original
analysis
EPA
assumed
that
all
fish
impinged
were
Age
One.
This
is
the
age
where
fish
are
given
a
commercial
economic
value.
In
response
to
comment
that
this
assumption
causes
a
gross
overestimate
of
forgone
yield
for
commercial
fisheries,
EPA
states
it
will
now
"
assume
instead
that
ages
of
impinged
fish
are
Age
1
and
older,
and
follow
an
age
distribution
that
is
implied
by
the
associated
survival
rates 
The
effect
of
this
adjustment
is
that
a
higher
proportion
of
impinged
fish
are
assumed
to
survive
until
harvest.
As
a
result
of
the
adjustment,

7
Federal
Register
Vol.
68,
No
3,
March
19,2003
pg.
13543
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
31
­
the
estimate
of
forgone
yield
associated
with
impingement
increases
by
a
factor
ranging
from
three
to
ten "
8
According
to
comments
submitted
by
EA
Engineering,
Science
and
Technology
on
behalf
of
UWAG,
EPA's
attempt
to
revise
the
approach
introduces
much
more
error
into
the
analysis.
According
to
experts,
impingement
is
dominated
by
the
Age
Zero
fish
class,
or
what
is
called
"
young­
of­
the­
year"
(
YOY),
not
by
Age
One
fish.
Experts
contend
that
EPA
should
lower
production­
forgone
estimates
because
the
survival
rate
of
YOY
fish
is
much
lower
than
the
survival
rate
of
all
other
fish
classes.
Despite
proposing
to
use
a
vague
and
unexplained
"
age
distribution"
and
"
survival
rate"
filter
to
refine
the
estimate,
by
assuming
all
impinged
fish
are
Age
One
the
benefits
estimate
will
incorrectly
give
YOY
fish
the
same
value
as
Age
One
fish.
The
result
is
a
potential
overstatement
of
benefit
estimates.

Step
Two:
Estimate
benefits
to
recreational
fishing
For
the
final
rule
analysis,
EPA
states
that
the
agency
will
use
regional
random
utility
models
(
RUM)
instead
of
the
models
based
on
water
body
type
used
in
the
original
analysis.
These
models
will
use
existing
studies
to
determine
economic
value
for
the
quality
of
recreational
fishing.
The
models
will
continue
to
use
the
assumption
that
the
greater
the
catch
rates
the
higher
the
economic
value
of
a
fishing
trip.
Higher
catch
rates
are
believed
to
yield
higher
satisfaction,
which
in
turn
causes
anglers
to
assign
higher
values
to
subsequent
trips.
The
models
measure
an
angler's
decision
to
visit
a
site
as
a
function
of
the
cost
of
visiting
the
site
(
travel
cost
and
lost
wages),
the
fishing
quality
(
catch
rate),
and
the
attribute
of
the
site
(
i.
e.,
boat
launches,
docks,
etc.).
The
following
data
sources
are
cited
for
intended
use:

 
National
Marine
Fisheries
Service
(
NMFS)
Marine
Recreational
Fishing
Survey
(
1994,
1997,
2000);
 
1995
Michigan
Recreational
Anglers
Survey,
 
2000
National
Survey
of
Recreation
and
Environment;
and
 
2000
National
Survey
of
Fishing,
Hunting,
and
Wildlife
Associated
Recreation.

Overestimate:
RUM
data
and
survey
bias
The
same
concerns
from
the
critique
of
EPA's
Phase
II
use
of
survey
data
in
the
RUM
models
applies
to
the
implementation
of
the
new
regional
RUM
Models.
The
regional
RUM
models
are
still
based
on
surveys
conducted
by
different
agencies
for
different
purposes.
There
is
a
potential
for
overestimates
in
recreational
values
due
to
the
following
survey
biases:

1.
Sampling
Bias:
This
is
caused
by
fact
that
the
all
anglers
within
a
region
do
not
have
the
opportunity
to
participate
in
the
survey.
2.
Avidity
Bias:
The
resulting
benefits
analysis
may
be
overstated
because
the
survey
may
be
skewed
because
an
over­
sampling
of
avid
anglers,
who
report
higher
values
to
each
trip
and
inflated
catch
rates.

8
Federal
Register
Vol.
68,
No
3,
March
19,2003
pg.
13546
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
32
­
3.
Recall
Bias:
Survey
participants
may
overstate
their
level
of
fishing
trips
because
they
may
not
remember
and/
or
anglers
may
state
that
they
go
more
than
they
actually
do.
Surveys
that
seek
to
measure
respondent
behaviors
may
include
using
travel
diaries
or
other
methods
of
recording
the
destination,
costs,
and
dates
of
each
trip.
Simply
asking
a
respondent
for
an
estimation
of
trip
cost
and
frequency
will
bias
the
value
estimate.

Overestimate:
RUM
Estimation
of
Opportunity
Costs
RUM
models
benefit
estimations
include
modules
that
represent
the
opportunity
cost
experienced
by
anglers.
Opportunity
costs,
in
this
case,
are
the
lost
wages
of
anglers
who
go
fishing.
According
industry
analysis,
existing
studies
cite
a
one­
third
of
the
wage
rate
as
the
appropriate
model
input,
however
in
the
North
Atlantic
Region
Case
study
in
the
NODA,
the
full
wage
rate
of
respondents
appears
to
be
used.

Also,
trip
costs
and
lost
wages
are
entirely
dependant
on
the
survey
responses.
For
example,
an
accurate
response
to
question
29
from
the
MRFSS
Economic
Survey,
"
About
how
much
money
could
you
have
earned
if
you
hadn't
taken
this
trip?"
is
a
crucial
component
to
the
RUM
Model.
It
is
unclear
if
wages
lost
as
a
result
of
a
fishing
trip
was
pre­
tax.
If
not,
this
would
also
contribute
to
an
overestimation
of
recreational
benefits.
As
mentioned
above,
these
surveys
include
a
number
of
biases
that
introduce
error
into
the
estimate.

Uncertainty:
Non­
Standardized
RUM
Inputs
It
should
be
noted
that
that
EPA
does
use
the
one­
third
of
wage
rate
rule
in
inputting
the
opportunity
cost
into
the
RUM
model
for
the
Northern
California
Region
case
study,
but
as
noted
above,
not
in
the
Atlantic
Region
case
study.
Non­
standardized
inputs
for
RUM
models
among
the
different
regions
will
yield
an
inaccurate
national
benefits
estimate.

Step
Three:
Estimate
Benefits
to
Commercial
Fishing
EPA
will
continue
to
base
commercial
fishing
benefit
estimations
on
the
gross
loss
of
revenue
associated
to
I&
E.
EPA
states
it
will
update
the
market
data
(
cost
per
pound
of
fish)
in
order
to
ensure
the
final
analysis
is
reflective
of
current
market
prices.

Uncertainty:
Linear
fish
stock
and
harvest
relationship
EPA
states
it
will
relax
the
original
assumption
that
recreational
or
commercial
anglers
harvest
every
sparred
fish
due
to
CWIS
modifications.
The
agency
states
it
will
use
a
linear
assumption
between
fish
stock
and
harvest.
For
example,
if
"
10
percent
of
the
current
commercially
targeted
stock
is
harvested,
then
10
percent
of
any
increase
in
stock
due
to
this
rule
would
be
harvested."
9
The
ramifications
of
such
an
assumption
are
unclear.
There
are
no
sources
cited
to
substantiate
this
approach
and/
or
the
10
percent
figure
used
in
the
example.
According
to
economic
theory,
there
is
a
decline
in
marginal
utility
by
adding
to
the
supply.
This
means
that
at
some
point
adding
more
fish
to
the
population
does
not
translate
into
high
levels
of
harvest
and
commercial
9
Federal
Register
Vol.
68,
No
3,
March
19,2003
pg.
13547
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
33
­
revenue
because
consumer
demand
for
the
fish
will
not
necessarily
increase
and
prices
will
fall
due
to
higher
supply.

Step
Four:
Nonuse
Values
Calculating
nonuse
values
is
one
of
the
most
complex
processes
in
the
benefits
estimation
methodology.
These
values
cannot
be
observed
because
there
is
no
specific
action
or
event
involved.
To
calculate
the
dollar­
values
for
nonuse
organisms
such
as
forage
fish
and
other
species
that
are
not
caught
by
commercial
and
recreational
fishing,
EPA
will
use
a
process
called
"
benefits
transfer."
The
components
of
EPA
benefits
transfer
methodology
are
described
briefly
below.

First
EPA
uses
survey
information
to
assess
respondents'
values
for
habitats
such
as
eelgrass
and
wetlands.
These
lands
play
a
significant
role
in
the
production
of
fish
and
shellfish.
Second,
agency
then
estimates
the
amount
of
each
habitat
that
would
be
needed
to
replace
fish
and
shellfish
that
are
lost
to
I&
E.
EPA
then
combines
the
habitat
cost
information
with
I&
E
data
to
estimate
nonuse
values.
EPA
estimates
nonuse
values
for
the
North
Atlantic
Region
to
amount
to
at
least
$
76
million
and
at
most
$
140
million.

Uncertainty:
Survey
validity
(
Implications
for
all
three
uses)

EPA
uses
the
responses
for
each
of
the
previously
mentioned
surveys
to
provide
the
foundation
for
the
entire
analysis.
This
said,
it
must
be
noted
that
theses
surveys
were
not
explicitly
conducted
to
calculate
any
types
of
dollar
values
for
fish.
EPA
uses
the
survey
to
determine
the
values
assessed
to
habitats.
UWAG
analysis
of
one
study
used
by
EPA
to
derive
nonuse
estimates
revealed
that
the
authors
themselves
contend
that
the
study
was
more
useful
in
calculating
relative
values
rather
than
dollar
estimates.

"
However,
we
believe
that
that
relatives
values
of
resources
[
derived
from
the
contingent
choice
survey]
are
more
reliable
than
are
the
dollar
estimates
of
values
and
recommend
relative
values,
rather
than
dollar
values,
be
used
in
the
process
of
selecting
management
actions"
(
Opaluch
et
all.
1995
as
cited
in
"
UWAG
Comments
on
Benefits
Estimations
in
EPA's
Proposed
Phase
II
316
(
b)
Rule,"
prepared
by
Triangle
Economic
Research,
June
2,
2003)

The
surveys
used
employ
a
hypothetical
series
of
questions
regarding
habitat
preferences,
not
fish
values.
The
Office
of
Budget
and
Management
(
OMB),
who
provides
guidelines
on
costbenefit
analysis
of
federal
regulations
states
that
the
hypothetical
situation
should,
"
correspond
closely
with
the
policy
context
to
which
estimates
will
be
applied."
10
Because
the
surveys
were
not
originally
intended
to
assess
the
economic
or
dollar
value
of
fish
they
may
not
be
a
reliable
source
of
information
to
base
the
entire
recreational
benefits
analysis.

10
OMB
Draft
2003
Report
to
Congress
on
the
Costs
and
Benefits
of
Federal
Regulations)
68
Fed.
Reg.
5492
February
3,
2003
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
34
­
Uncertainty:
Translating
Habitat
Values
into
Fish
Values
EPA
uses
an
"
indirect"
approach
to
estimating
the
nonuse
values
of
fish
by
basing
the
estimates
on
survey
respondents'
valuation
of
new
habitats
and
then
using
biological
assumptions
about
how
much
of
that
habitat
is
needed
to
produce
fish.
The
NODA
cites
neither
previous
literature
nor
new
studies
to
substantiate
such
an
approach.
Comments
from
Triangle
Economic
Research
on
behalf
of
UWAG
raise
this
concern,
"
The
presumption
that
values
for
eelgrass
and
wetlands
can
be
transformed
into
nonuse
values
for
fish
is
unfounded.
Economic
values
are
largely
tied
to
the
intrinsic
value
of
resources
rather
than
service
flows
from
those
resources."

Previously
EPA
had
used
a
"
50
percent
rule
of
thumb"
to
estimate
nonuse
benefits.
This
involved
assigning
50
percent
of
the
combined
recreational
and
commercial
use
figures
to
the
nonuse
category.
The
table
below,
taken
from
the
NODA,
shows
all
three
benefit
estimates
for
the
North
Atlantic
region.
The
Nonuse
value
makes
up
over
95
percent
of
the
total
benefits
estimate.
This
implies
a
20­
fold
increase
in
non­
use
benefits.

Table
9:
EPA
North
Atlantic
Region
Benefit
Estimate
Benefit
Category
Benefit
($
1,000)
Percentage
Commercial
$
280
0.35%
Recreational
$
3,070
3.89%
Nonuse
$
75,640
95.76%
TOTAL
$
78,990
100.00%

As
stated
above,
the
translation
of
the
economic
values
assessed
by
the
survey
for
the
habitat
(
eelgrass
and
wetlands)
and
the
value
of
fish
is
unclear
and
supporting
literature
is
not
cited
the
NODA.

A
comparison
from
the
case
studies
illustrates
the
inconsistency
and
uncertainty
of
this
process.
When
compared,
the
benefits
of
the
original
case
study
of
the
Salem
facility
in
located
in
the
Delaware
estuary
are
dramatically
different
from
those
of
the
North
Atlantic
Case
Study
provided
in
the
NODA.
The
two
charts
below
illustrate
the
differential
between
the
original
benefits
methodology
and
the
benefits
transfer
process
used
in
the
NODA.
Note
that
the
nonuse
value
in
the
Salem
case
is
20
percent
of
the
total
benefits
estimate
while
the
value
for
the
North
Atlantic
Region
is
95.76
percent.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
35
­
Figure
2:
Salem
Case
Study
Benefits
Commerical
39%

Recreation
41%
Nonuse
20%

Source:
Chapter
B­
3
Delaware
Estuary
Benefits
Analysis
http://
www.
epa.
gov/
waterscience/
316b/
casestudy/
chb4.
pdf
Figure
3:
North
Atlantic
Region
Benefits
Nonuse
95.76%
Commerical
0.35%
Recreation
3.89%

Source:
68
Federal
Register
p.
13579
of
the
NODA
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
36
­
Overestimate:
Possible
double
counting
of
forage
fish
In
the
original
benefits
analysis
EPA
included
the
value
of
forage
fish
in
the
commercial
benefits
analysis.
Forage
fish
were
valued
as
a
food
source
for
commercially
valued
fish
species.
In
the
revised
analysis
forage
fish
are
assigned
a
separate
value
under
the
nonuse
category.
The
NODA
does
not
state
whether
forage
fish
values
have
been
excluded
from
the
commercial
benefit
estimations.
This
double
counting
would
overestimate
the
national
benefits
of
reduced
I&
E.

Overestimate:
Linear
stock
and
harvest
relationship
As
previously
stated
EPA
plans
to
use
a
linear
approach
to
measuring
benefits
of
reduced
I&
E.
The
example
used
is
that
that
is
a
10
percent
reduction
in
I&
E
will
lead
to
a
10
percent
increase
in
fish
yields
for
commercial
fishing.
Literature
cited
by
experts
show
that
natural
variations
in
fish
population
are
usually
larger
than
3
percent.
In
its
June
2003
comments,
UWAG
states
that
EPA
has
no
data
on
the
extent
to
which
flow
reduction
technologies
will
reduce
I&
E
losses,
but
instead
appears
to
assume
that
such
reductions
will
be
roughly
equivalent
to
the
percent
reduction
in
flow.
Although
UWAG
agrees
that
flow
reduction
will
produce
appreciable
reductions
in
entrainment,
and
may
also
reduce
impingement,
UWAG
does
not
agree
that
EPA
may
assume
that
such
options
will
produce
concomitant
reduction
in
I&
E.
A
report
prepared
for
EPRI,
Impacts
of
Volumetric
Flow
Rate
of
Water
Intakes
on
Fish
Populations
and
Communities
(
March
2003),
shows
the
following:

 
the
extent
to
which
volume
of
water
withdrawn
or
intake
rate
is
related
to
I&
E
is
highly
site­
specific
and
not
necessarily
linear;
and
the
relationship
between
intake
flow
volume
and
effects
at
the
population
level
and
higher
are
even
less
well
correlated.

Overestimate:
Effect
on
stock
caused
by
reduced
I&
E
The
NODA
relies
on
several
studies
to
help
substantiate
the
general
levels
of
nonuse
benefit
estimates.
However,
the
majority
of
these
studies
are
targeted
toward
measuring
improvements
in
water
quality
and
do
not
directly
relate
to
increases
in
population.
Some
experts
believe
that
water
quality
improvements
increase
the
viability
or
survival
rates
of
the
entire
fish
stock,
not
the
individual
fish.
Nonuse
values
calculated
by
EPA
are
based
on
per
fish
value
and
do
not
address
the
concept
of
overall
stock
impact.
Experts
believe
that
unlike
water
quality
improvements,
"
for
most
species
I&
E
reductions
will
have
very
little
impact
on
stock."
11
Conclusions
After
an
analysis
of
EPA's
revised
benefits
methodology
the
following
recommendation
is
offered:

To
avoid
overestimating
benefits
on
that
national
level,
the
nonuse
category
should
be
estimated
more
conservatively.
The
nonuse
benefit
value
of
the
North
Atlantic
region,
as
calculated
by
the
Phase
II
method,
makes
up
over
95
percent
or
75.6
million
of
the
78.9
11
See
Triangle
Economic
Research
June
2,
2003
Comments
on
316(
b)
NODA
June
3,
2003,
p
14
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
37
­
million
dollar
benefits
estimate.
The
revised
method
appears
to
greatly
exceed
the
50
percent
rule
used
in
the
original
analysis,
where
50
percent
of
the
total
recreational
and
commercial
benefit
is
said
to
represent
the
nonuse
benefit.
If
the
50
percent
rule
was
reinstated
in
the
revised
analysis
than
the
nonuse
benefits
would
be
much
less
or
$
1.67
million
for
the
Atlantic
Region.
The
total
benefits
for
the
region
would
be
approximately
$
5
million
versus
the
$
79
million
cited
in
the
NODA.

The
following
table
compares
the
Northern
Atlantic
Region
benefits
provided
in
the
NODA
to
what
the
benefits
estimates
would
be
if
the
nonuse
values
were
removed
or
if
the
50
percent
rule
was
applied.
A
per
facility
benefit,
based
on
information
provided
by
21
facilities
in
the
North
Atlantic
Region
responding
to
EPA's
survey,
is
also
provided.

Table
10:
EPA
North
Atlantic
Region
Benefits
Estimate
(
Millions
of
Dollars)

Benefit
Category
Revised
Phase
II
NODA
Benefit
Benefit
Without
Nonuse
50%
Nonuse
Rule
Commercial
$
0.280
$.
280
$
0.280
Recreational
$
3.070
$
3.070
$
3.070
Nonuse
$
75.64
­
$
1.675
TOTAL
BENEFIT
$
78.99
$
3.350
$
5.025
Per
Facility
Benefit
$
3.761
$
0.159
$
239.3
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
38
­
5.
Review
and
Critique
of
Cost
Estimates
for
Phase
III
of
Section
316(
b)
Rulemaking
In
the
Phase
III
Background
document,
dated
September
17,
2002,
EPA
included
a
preliminary
estimate
of
facility­
level
compliance
costs
for
316(
b)
Phase
III.
EPA
used
the
range
of
technologies
and
costs
developed
in
the
Phase
II
rule
proposal
(
published
April
9,
2002)
to
provide
high
and
low
estimates
of
possible
costs
to
Phase
III
facilities.
In
the
Background
document,
it
is
stated
that
"
EPA
is
revising
the
cost
methodology
to
better
reflect
retrofitting
costs
and
other
factors
identified
in
comments
on
the
Phase
II
rule
proposal.
It
is
expected
that
the
revised
methodology
will
produce
higher
cost
estimates."
Based
on
correspondence
with
EPA
staff,
the
revised
methodology
will
not
be
available
for
review
until
2004.
As
a
result,
the
recommendations
below
are
intended
to
assist
EPA
in
developing
this
revised
cost
methodology.
It
should
be
noted
that
EPA
did
provide
some
revised
costing
formulas
and
cost
curves,
but
they
were
incomplete
and
therefore
could
not
be
used.

The
following
section
provides
an
overview
of
EPA's
cost
methodology
proposed
in
the
Phase
III
Background
document.
The
cost
methodology
includes
calculation
of
capital
costs,
operation
and
maintenance
(
O&
M)
costs,
and
permitting
and
monitoring
(
P&
M)
costs.
As
this
preliminary
Phase
III
estimate
was
primarily
based
on
the
methodology
utilized
in
the
Phase
II
rule
proposal,
the
recommendations
provided
to
EPA
in
the
development
of
the
revised
Phase
III
estimate
will
be
based
on
the
Phase
II
rule
proposal,
the
Phase
II
NODA
and
industry
comments
in
response
to
the
Phase
II
NODA.

Methodology
for
Calculating
Capital
and
Operation
and
Maintenance
(
O&
M)
Costs
The
capital
and
O&
M
costs
of
installing
and
retrofitting
to
comply
with
the
Phase
III
requirements
were
determined
using
methodology
similar
to
that
described
in
the
Technical
Development
Document
for
the
Proposed
Section
316(
b)
Phase
II
Existing
Facility
Rule.
The
following
is
a
step­
by­
step
description
of
the
methodology
used
to
calculate
annualized
total
technology
costs
for
each
facility:

Step
1:
Determine
the
compliance
technology
requirements
based
on
calculation
baseline.

For
assigning
compliance
costs
on
a
facility­
level
basis,
EPA
first
assessed
the
intake
technologies
in­
place
and
compared
these
to
compliance
requirements,
thereby
determining
if
the
facility
would
require
technology
upgrade
costs.

Step
2:
Determine
which
of
the
following
three
scenarios
applies
if
the
facility
requires
upgraded
technology:

The
three
technologies
proposed
for
complying
with
Phase
III
requirements
include:

(
1)
installation
of
fish
handling
and/
or
return
systems
without
replacement
of
existing
screens,
(
2)
replacement
of
existing
coarse­
mesh
screens
with
fine­
mesh
screens,
or
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
39
­
(
3)
replacement
of
existing
coarse­
mesh
screens
with
fine­
mesh
in
addition
to
installation
of
fish
handling
and/
or
return
systems.

Step
3:
Select
the
proper
range
of
the
appropriate
"
initial
capital
cost"
equation.

The
EPA
utilized
a
set
of
four
simple
equations
to
represent
the
capital
costs
of
one
type
of
intake
technology;
each
of
these
four
equations
is
valid
over
a
narrow
range
of
flow.
The
equation
that
represents
the
capital
costs
of
a
particular
technology
at
a
facility
is
determined
by
that
facility's
DIF.

These
equations
apply
a
46
percent
escalation
factor
to
technology
costs
to
account
for
mobilization/
demobilization,
engineering,
site
work,
electrical,
controls,
contractor
overhead/
profit,
and
contingency.

Step
4:
Estimate
additional
retrofit
costs
necessary
to
modify
and
augment
the
structure
that
is
in
place
at
the
facility.

 
For
facilities
upgrading
their
intake
technologies,
an
additional
retrofit
factor
of
30
percent
was
applied,
regardless
of
the
compliance
scenario.
 
For
facilities
retrofitting
with
a
fine
mesh
screen,
capital
costs
were
further
escalated
using
a
set
of
construction
factors
based
on
compliance
requirements
and
screen
flow
sizing.
EPA
doubled
the
construction
factor
for
nuclear
facilities
as
these
facilities
would
not
be
able
to
use
blasting
and
high­
impact
equipment.
 
To
account
for
the
inherent
uncertainty
in
retrofitting
an
existing
intake
structure
an
additional
allowance
factor
of
5
percent
was
applied
to
the
overall
capital
costs.

Step
5:
Apply
the
appropriate
State­
specific
capital
cost
factor
to
account
for
the
nature
of
regional
variation
in
capital
costs.

To
adjust
totals
costs
to
reflect
varying
costs
of
capital
and
labor
across
different
states,
EPA
included
a
overall
state­
specific
capital
cost
multiplier.
The
multiplier
ranges
between
0.75
and
1.06.

Step
6:
Escalate
the
"
initial
capital
cost"
equation
according
to
the
equation
provided
below.

To
calculate
total
capital
costs,
the
"
initial
capital
cost"
equation
should
be
revised
to
include
the
additional
factors
considered
in
Steps
4
and
5.
The
resulting
equation
is:

(
initial
capital
cost
equation)
*
(
1
+
retrofit
factor
+
construction
factor
+
allowance)
*
(
state
factor)
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
40
­
Step
7:
Estimate
O&
M
costs
using
the
appropriate
O&
M
cost
curve
equation
based
on
the
design
intake
flow.

As
with
capital
costs,
the
EPA
developed
O&
M
cost
curves
for
each
of
the
three
technology
upgrade
scenarios
described
above.
O&
M
costs
for
each
of
the
technology
scenarios
are
dependent
on
DIF.
No
additional
scaling
factors
were
applied
to
the
O&
M
cost
equations.

Facilities
not
requiring
any
additional
CWIS
compliance
modification
received
costs
of
zero
dollars.
For
each
facility
receiving
costs
greater
than
zero,
capital
costs
were
annualized
assuming
a
10
year
useful
life
of
the
compliance
equipment
and
a
seven
percent
discount
rate.
Both
costs
were
summed
to
derive
the
annualized
total
technology
cost
for
each
facility.

Methodology
for
Calculating
Permitting
and
Monitoring
(
P&
M)
Costs
As
noted
in
the
Phase
III
Background
document,
EPA
has
not
yet
developed
administrative
costs
specific
to
the
Phase
III
facilities,
and
realistic
estimates
cannot
be
made
until
regulatory
options
are
better
developed.
Instead,
for
this
initial
analysis,
EPA
used
the
range
of
costs
developed
for
Phase
II
facilities
to
provide
high
and
low
estimates
of
possible
costs
to
Phase
III
facilities
(
see
the
Economic
and
Benefit
Analysis
Proposed
Section
316(
b)
Phase
II
Existing
Facility
Rule).

For
each
Phase
II
facility,
the
primary
determinant
of
the
magnitude
of
administrative
costs
is
the
source
waterbody
type.
Administrative
costs
include
one­
time
and
annual
costs
of
completing
a
wide
range
of
activities
associated
with
the
submission
of
the
initial
post­
promulgation
NPDES
permit
application,
subsequent
NPDES
permit
renewals,
and
monitoring.
The
annualized
Phase
II
permitting,
re­
permitting,
and
monitoring
costs
were
summed
for
each
Phase
II
facility
and
the
highest
and
lowest
annualized
total
permitting
and
monitoring
cost
for
each
source
waterbody
category
was
determined.
These
costs
were
then
expressed
in
2002
dollar
values
using
the
most
recent
employment
cost
index
for
Professional
Specialty
and
Technical
Occupations
published
by
the
U.
S.
Bureau
of
Labor
Statistics.
Each
Phase
III
facility
assigned
CWIS
compliance
technology
costs
greater
than
zero
was
assigned
a
high
and
low
annualized
total
permitting
and
monitoring
cost
based
on
its
waterbody
type.

Preliminary
Estimate
of
Total
Annualized
Compliance
Costs
for
Small
Entities
For
each
facility,
the
annualized
total
technology
(
capital
and
O&
M)
cost,
and
the
high
and
low
annualized
total
permitting
and
monitoring
(
P&
M)
costs
were
summed
to
derive
the
high
and
low
total
costs
for
each
facility.
To
conduct
an
initial
small­
entity
impact
analysis,
EPA
calculated
the
range
of
costs
for
firms
potentially
subject
to
Phase
III.
Table
11
below
shows
the
range
of
average
annualized
technology
and
P&
M
costs
for
facilities
affected
by
the
Phase
III
rulemaking.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
41
­
Table
11:
Average
Annualized
Costs
per
Facility
(
Dollars)

Firm/
Facility
Type
#
of
Inscope
Facilities
Average
Annualized
Capital
Cost
Average
Annual
O&
M
Cost
Average
Total
Annualized
Technology
Cost
(
Capital
+
O&
M)
Average
Total
Annualized
Technology
+
Low
P&
M
Cost
Average
Total
Annualized
Technology
+
High
P&
M
Cost
All
Electric
Gen.
Facility
Types
149
$
7,077
$
2,313
$
9,391
$
153,879
$
167,025
All
Manufacturing
Facility
Types
558
$
3,900
$
1,199
$
5,099
$
69,604
$
74,659
Total
Facility
Types
707
$
4,570
$
1,434
$
6,004
$
87,365
$
94,125
Summary
of
Changes
to
Cost
Analysis
in
Phase
II
NODA
As
previously
stated,
the
Phase
III
cost
methodology
is
based
on
the
Phase
II
approach.
As
result,
changes
made
in
the
Phase
II
approach
will
be
included
in
the
Phase
III
methodology.
Several
major
changes
were
made
to
the
cost
analysis
in
the
Phase
II
rule
proposal.
EPA
used
new
information
to
revise
the
capital
and
operation
and
maintenance
(
O&
M)
costs
for
several
compliance
technologies,
including
those
used
as
the
primary
basis
for
the
proposed
regulatory
option.
Based
on
a
revised
estimate,
total
capital
costs
increased
by
66
percent
and
total
O&
M
costs
increased
by
48
percent.
The
cost
differences
were
primarily
due
to
the
expanded
range
of
technology
options
considered
for
the
NODA
and
the
"
best
performing
technology"
criteria
used
to
assign
cost
modules
to
model
facilities.

Summary
of
Cost
Methodology
Critiques
Similar
to
Section
3.2,
which
critiqued
the
approach
used
to
estimate
the
benefits
of
the
proposed
regulations,
this
section
will
offer
a
critique
of
the
cost
methodology.
The
following
three
categories
will
used
to
describe
key
flaws
in
a
number
of
areas:

1.
Overestimate
2.
Underestimate
3.
Uncertainty.

Uncertainty:
EPA's
definition
and
methods
for
determining
the
"
Calculation
Baseline"

The
degree
to
which
I&
E
should
be
reduced
and
the
best
performing
technology
for
reducing
I&
E
is
based
on
a
reasonable
estimate
of
the
baseline.
A
facility
with
baseline
design
is
one
that
has
not
used
locational
strategies,
technologies
or
operational
measures
to
reduce
I&
E.
The
proposed
performance
standards
are
percentage
reductions
from
this
baseline.
As
a
result
total
compliance
costs
are
dependent
on
the
strategy
used
to
develop
the
baseline.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
42
­
In
the
Phase
II
proposal,
the
EPA
defined
the
calculation
baseline
as
"
an
estimate
of
impingement
mortality
and
entrainment
that
would
occur
at
your
site
assuming
you
had
a
shoreline
cooling
water
intake
structure
with
an
intake
capacity
commensurate
with
a
oncethrough
cooling
water
system
and
with
no
impingement
and/
or
entrainment
reduction
controls."
Because
commenters
described
this
definition
as
too
vague,
the
EPA
invited
comments
on
a
more
descriptive
definition
and
an
alternative
"
As
Built"
approach
included
in
the
Phase
II
NODA.

The
Utility
Water
Act
Group
(
UWAG)
provided
several
suggestions
for
enhancing
the
two
approaches
in
the
Phase
II
NODA.
However,
UWAG
determined
that
the
right
structure
for
performance
evaluation
depends
on
a
number
of
factors,
including
the
following:

 
What
is
the
nature
and
extent
of
existing
impingement
or
entrainment
or
both;
 
Whether
the
assessment
is
for
impingeable
or
entrainable
organisms;
 
Whether
the
technology
consists
of
collection
systems,
diversions,
behavioral
barriers,
location
changes,
or
flow
changes;
 
Whether
the
facility
has
data
to
establish
the
baseline
configuration;
and
 
Whether
the
facility
proposes
to
use
restoration
or
request
alternative
limits
based
on
the
benefit­
cost
test.

UWAG
concluded
that
the
EPA
should
give
permitees
and
permit
writers
the
flexibility
to
develop
appropriate
site­
specific
performance
evaluation
requirements
that
consider
relevant
factors
including
the
ones
listed
above.

Underestimate:
EPA
has
proposed
a
broader
range
of
compliance
technologies
that
may
be
applicable
to
facilities
potentially
affected
by
Phase
III.

During
the
development
of
the
Phase
II
NODA,
EPA
evaluated
new
and/
or
additional
costs
for
a
wider
range
of
intake
technologies
identified
as
having
the
potential
to
meet
the
proposed
regulation
requirements
without
the
expense
and
energy
penalty
associated
with
capacityreduction
technologies
such
as
cooling
towers.
A
total
of
11
cost
modules
were
developed
for
assessing
model­
facility
compliance
costs.

Due
to
the
expanded
range
of
cost
modules
and
the
decision
to
assign
costs
based
on
the
"
best
performing
technology"
instead
of
the
least
cost
approach,
total
capital
costs
for
the
preferred
option
increased
by
66
percent
and
total
O&
M
costs
increased
by
48
percent.
The
increased
costs
estimated
in
the
Phase
II
NODA
also
reflect
revised
downtime
estimates
for
technology
installation
as
follows:

 
Additional
unscheduled
downtimes
of
between
two
and
eight
weeks
for
the
installation
of
various
non­
recirculating
compliance
technologies;
and
 
At
nuclear
power
plants,
cooling
system
conversions
are
more
likely
to
require
a
net
downtime
of
seven
months
instead
of
four
months,

This
expanded
set
of
compliance
options
modules
should
be
considered
as
potential
options
in
determining
compliance
costs
for
facilities
potentially
affected
by
the
Phase
III
rulemaking.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
43
­
Similar
to
the
impact
of
the
expanded
set
of
methodologies
to
preliminary
Phase
II
cost
estimates
in
the
rule
proposal,
it
is
likely
that
an
increased
variety
of
options
will
also
result
in
an
increase
in
Phase
III
compliance
costs.
The
table
below
shows
revised
cost
estimates
based
on
a
66
percent
increase
in
average
annualized
capital
costs
and
a
48
percent
increase
in
average
annual
O&
M
costs.
A
66
percent
increase
in
total
capital
cost
is
about
equal
to
a
66
percent
increase
in
average
annualized
capital
costs.
The
range
of
P&
M
costs
remain
the
same.

Table
12:
Revised
Compliance
Costs
for
Phase
III
based
on
Higher
Cost
Estimates
in
Phase
II
NODA
(
Dollars)

Firm/
Facility
Type
Avg
Annnualized
Capital
Cost
REVISED
Avg
Annualized
Capital
Cost
(
66%
greater)
Average
Annual
O&
M
Cost
REVISED
Avg
Annual
O&
M
Costs
(
48%
greater)
REVISED
Average
Total
Annualized
Technology
+
Low
P&
M
Cost
REVISED
Average
Total
Annualized
Technology
+
High
P&
M
Cost
Total
Facility
Types
$
4,570
$
7,585
$
1,434
$
2,122
$
91,069
$
97,829
Uncertainty:
Definition
of
"
significant
greater"
when
conducting
cost­
cost
and
costbenefit
tests
to
determine
if
a
facility
may
choose
a
site­
specific
alternative.

In
the
Phase
II
rule
proposal,
a
facility
may
choose
a
site­
specific
alternative
to
demonstrate
use
of
best
technology
available
(
BTA)
for
minimizing
adverse
environmental
impact
at
its
site.
If
a
facility
chooses
this
alternative,
the
facility
must
demonstrate
that
costs
of
compliance
with
the
applicable
performance
standards
would
be
"
significantly
greater"
than:

 
the
costs
considered
by
the
Administrator
when
establishing
the
performance
standards;
or
 
the
benefits
of
complying
with
the
applicable
performance
standards
at
its
site.

In
the
Phase
II
NODA,
the
EPA
requested
comments
on
whether
the
Agency
should
adopt
a
quantitative
definition
of
"
significantly
greater,"
and
if
so,
what
specific
ratio
would
be
appropriate.

Based
on
comments
from
industry
representatives,
any
cost
greater
than
the
costs
estimated
by
EPA
or
the
benefits
of
compliance
should
provide
facilities
with
the
option
of
choosing
a
sitespecific
alternative
to
demonstrate
use
of
BTA
for
minimizing
adverse
environmental
impact
at
its
site.

Underestimate:
A
two­
year
verification
monitoring
program
will
cost
permittees
more
than
EPA
has
estimated.

UWAG
developed
its
own
estimate
for
an
impingement
verification
program
that
EPA
had
estimated
to
cost
only
$
16,985.
UWAG
used
the
same
hourly
rates
used
by
EPA,
but
added
in
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
44
­
labor
and
travel
costs
associated
with
getting
to
and
from
the
site
for
a
contractor
located
150
miles
away
and
also
included
the
costs
of
identifying
some
specimens
in
the
contractor's
lab.
UWAG
estimate
for
a
year's
worth
of
monthly
impingement
monitoring
is
$
38,724,
over
twice
the
amount
estimated
by
EPA.
As
a
result,
monitoring
costs
for
Phase
III
facilities
may
be
over
twice
the
amount
estimated
in
the
Phase
III
background
document
as
the
methodology
used
to
calculated
costs
is
based
on
Phase
II
estimates.
Based
on
Tables
B1­
2,
B1­
3
and
B1­
4
in
the
document,
Phase
II
Economics
and
Benefits
Analysis,
Part
B:
Costs
and
Economic
Impacts,
impingement
monitoring
constitutes
1.42
percent
of
total
P&
M
costs.
As
a
result,
Table
13
below
shows
the
revised
high
and
low
P&
M
costs
given
a
100
percent
increase
in
the
impingement
monitoring
program.

Table
13:
Revised
Permitting
and
Monitoring
Costs
(
Dollars)

Firm/
Facility
Type
Low
P&
M
Costs
High
P&
M
Costs
REVISED
Low
P&
M
Costs
REVISED
High
P&
M
Costs
Total
Facility
Types
$
81,361
$
88,122
$
82,513
$
89,369
Recommendations
for
Revised
Cost
Estimate
 
Average
Annual
Cost
Per
Facility
Based
on
the
analysis
conducted
above,
the
following
provides
a
revised
estimate
of
compliance
costs
per
facility.
The
revision
is
based
on
higher
capital
costs,
higher
operation
and
maintenance
costs,
and
higher
permitting
and
monitor
costs.
The
revisions
are
presented
in
Table
14
below.

Table
14:
Revised
Annual
Average
Annual
Costs
per
Facility
(
Dollars)

Firm/
Facility
Type
REVISED
Avg
Annualized
Capital
Cost
(
66%
greater)
REVISED
Avg
Annual
O&
M
Costs
(
48%
greater)
REVISED
Low
P&
M
Costs
REVISED
High
P&
M
Costs
REVISED
Average
Total
Annualized
Technology
+
Low
P&
M
Cost
REVISED
Average
Total
Annualized
Technology
+
High
P&
M
Cost
Total
Facility
Types
$
7,585
$
2,122
$
82,513
$
89,369
$
92,221
$
99,077
In
Section
XII
of
its
June
2003
comments
to
the
Phase
II
NODA
and
Section
XXII
of
its
August
2002
comments
to
the
Phase
II
rule
proposal,
UWAG
describes
EPA's
analysis
of
costs
and
benefits
as
"
virtually
impossible
to
assess"
and
"
impossible
to
assess
in
detail."
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
45
­
6.
Proposed
Phase
III
Regulatory
Alternatives
This
section
discusses
the
current
approach
of
regulating
cooling
water
intake
structures
based
on
information
from
the
Phase
I
rule
and
the
Phase
II
proposed
rule.
The
chapter
will
also
outline
several
possible
regulatory
options
for
Phase
III
and
provide
a
cost­
benefit
analysis
that
will
evaluate
which
options
may
provide
cost
and
environmentally
effective
solutions
while
minimizing
impacts
on
small
businesses.
The
chapter
will
be
structured
as
follows:


Section
6.1
describes
summarize
the
details
of
the
most
current
Phase
III
proposed
requirements;


Section
6.2
provides
an
overview
of
three
regulatory
options

Section
6.3
explains
the
methodology
and
results
of
a
cost­
benefit
analysis
for
each
option;
and

Section
6.4
summarizes
the
analysis
and
presents
a
recommendation.

6.1.
Current
Regulatory
Strategy
This
section
outlines
the
current
regulatory
requirements
of
Phase
III.
Facilities
having
design
intake
flows
(
DIF)
no
more
than
2
MGD,
and
those
using
less
than
25
percent
of
their
design
intake
flow
for
cooling
purposes
(
regardless
of
DIF),
would
not
be
subject
to
the
specific
Phase
III
requirements.
Instead,
requirements
for
such
facilities
would
continue
to
be
addressed
by
the
permitting
authority
on
case­
by­
case
basis
using
best
professional
judgment.

The
flow
threshold
of
2
MGD
is
taken
as
a
starting
point;
other
MGD
threshold
values
will
be
considered.
Other
thresholds
will
also
be
considered
for
the
percentage
of
intake
water
used
for
cooling
purposes
and
for
capacity
utilization
at
electric
generators.
EPA
is
seeking
information
about
intermittent
or
seasonal
withdrawal
of
water
by
manufacturing
facilities
(
analogous
to
capacity
utilization
at
electric
generating
plants).
EPA
will
also
consider
ways
to
reduce
the
burden
of
permit
application
requirements
for
facilities
with
DIF
less
than
50
MGD.

These
are
the
performance
standards
assumed
as
a
starting
point
for
discussion,
and
assumed
for
purposes
of
estimating
costs
in
the
next
section.

 
DIF
less
than
50
MGD:

o
Water
source
an
estuary,
tidal
river,
ocean,
or
one
of
the
Great
Lakes:
reduce
impingement
(
80%
to
95%)
and
entrainment
(
60%
to
90%)
o
Other
water
sources:
reduce
impingement
(
80%
to
95%)

 
DIF
50
MGD
or
more,
electric
generators
with
capacity
utilization
<
15%

o
All
water
sources:
reduce
impingement
(
80%
to
95%)
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
46
­
 
DIF
50
MGD
or
more:
electric
generators
with
capacity
utilization
at
least
15
percent
and
all
other
facilities.

o
Water
source
an
estuary,
tidal
river,
ocean,
or
one
of
the
Great
Lakes:
reduce
impingement
(
80%
to
95%)
and
entrainment
(
60%
to
90%)
o
Water
source
a
freshwater
lake
or
reservoir:
reduce
impingement
(
80%
to
95%),
and
if
increasing
DIF
must
not
disrupt
natural
thermal
stratification
or
turnover
pattern
o
Water
source
a
freshwater
stream
or
river
and
DIF
is
not
more
than
5
percent
of
mean
annual
stream
or
river
flow:
reduce
impingement
(
80%
to
95%)
o
Water
source
a
freshwater
stream
or
river
and
DIF
is
more
than
5
percent
of
mean
annual
stream
or
river
flow:
reduce
impingement
(
80%
to
95%)
and
entrainment
(
60%
to
90%)

A
facility
that
already
meets
the
performance
standards,
or
one
that
has
reduced
its
intake
capacity
commensurate
with
the
use
of
a
closed­
cycle,
recirculating
system
would
have
already
applied
best
technology
available
for
minimizing
adverse
environmental
impact.

A
facility
would,
with
the
permitting
authority's
approval,
be
able
to
employ
restoration
measures
in
lieu
of
or
in
combination
with
intake
structure
technologies
or
operational
measures,
if
it
demonstrated
to
the
permitting
authority
that
it
is
maintaining
the
fish
and
shellfish
within
the
waterbody,
including
community
structure
and
function,
to
a
level
comparable
to
those
which
would
result
from
employing
design
and
construction
technologies
or
operational
measures
to
meet
the
performance
standards
for
reducing
impingement
mortality
and
(
if
applicable)
entrainment.

In
addition,
if
the
costs
to
implement
these
performance
standards
at
a
facility
are
significantly
greater
than
the
costs
EPA
considered
in
establishing
them
or
the
benefits
of
complying
with
such
standards,
a
facility
may
request
and
obtain
alternative,
site­
specific
standards.

6.2.
Overview
of
Phase
III
Regulatory
Options
The
remainder
of
this
study
examines
the
effects
of
three
modifications
to
the
regulatory
approach
described
above:

Regulatory
Option
1:
The
threshold
(
2
MGD
design
intake
flow)
for
applicability
of
specific
performance
standards,
may
be
set
higher
based
on
economic
tests
or
small
business
impacts,
provided
this
does
not
produce
financial
inequity
between
Phase
II
and
Phase
III
facilities.
The
thresholds
to
examine
will
be
10,
20
and
50
MGD.

Alternatively,
the
exemption
could
be
set
higher
based
on
the
biology
involved,
and
possibly
the
proportionality
between
the
size
of
the
DIF
and
the
affected
water
body
(
particularly
for
rivers
and
streams).
With
regard
to
financial
inequity,
this
is
unlikely
to
be
an
issue
for
manufacturing,
since
it
is
highly
unlikely
that
two
manufacturers
in
a
given
industry
would
be
located
in
the
same
competitive
marketplace.
This
may
be
an
issue
for
the
utilities.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
47
­
Regulatory
Option
2:
Facilities
with
DIF
<
50
MGD
and
located
on
sensitive
waterbodies
would
have
the
same
requirements
as
other
small
facilities
to
reduce
impingement
mortality
only.
In
other
words,
all
facilities
with
DIF
<
50
MGD,
need
only
comply
with
the
impingement
requirements.

Regulatory
Option
3:
Facilities
with
DIF
<
50
MGD
need
only
meet
a
set
of
specific
technological
requirements
that
are
set
by
rule
to
address
only
impingement
(
no
entrainment
requirements).
The
range
of
the
specified
technologies
can
vary
by
the
DIF,
and
the
type
of
affected
water
bodies.
These
will
be
chosen
to
reflect
the
biology
of
each
separate
facility/
water
body.
Baseline
monitoring
and
compliance
monitoring
would
not
be
required.
Alternatively,
compliance
monitoring
could
be
required
to
confirm
that
the
technology
chosen
is
meeting
the
presumed
performance
requirements.

6.3.
Costs
and
Benefits
of
the
Three
Regulatory
Options
The
following
sections
provide
the
details
of
the
methodology
and
results
of
a
cost­
benefit
analysis
of
the
three
regulatory
options
outlined
in
the
Section
5.
For
each
regulatory
option,
there
will
be
a
brief
summary
of
each
option,
followed
by
three
separate
sections.
The
sections
are
as
follows:

1.
Benefits
Estimate,
2.
Cost
Estimate,
and
3.
Summary
and
Comparison
of
Cost­
Benefit
analysis
Regulatory
Option
One
­
Exempt
Smaller
Facilities
The
first
proposed
regulatory
option
for
the
Phase
III
Rulemaking
explores
the
possibility
of
raising
the
2
million
gallon
per
day
(
MGD)
threshold
for
compliance
to
that
of
10,
20,
or
50
MGD.
The
purpose
of
this
exercise
is
to
illustrate
the
cost­
effectiveness
of
exempting
facilities
of
smaller
sizes.
These
facilities
withdraw
less
water,
cause
less
impingement
and
entrainment
of
fish,
and
do
not
face
proportionally
lower
compliance
costs.

Estimating
the
Benefits
of
Regulatory
Option
One
This
section
will
outline
the
methodology
and
preliminary
findings
of
this
benefit
analysis.
These
estimates
are
based
on
the
limited
information
gathered
from
the
March
19,
2003
Notice
of
Data
Availability
for
the
Phase
II
Rule,
supplemental
reports,
industry
comment
and
analysis,
and
correspondence
with
EPA
staff.
Assumptions
were
made
in
cases
where
there
was
no
clear
existing
or
available
data.
These
assumptions
will
be
clearly
identified.
The
process
of
developing
benefits
estimates
for
the
three
MGD
thresholds
involved
the
following
steps:

1)
Determine
the
number
of
affected
facilities
in
each
facility
size
category;
2)
Derive
an
estimated
total
benefits
dollar
amount;
and
3)
Scale
the
benefits
to
facility
size.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
48
­
Step
One:
Defining
the
Phase
III
Affected
Entity
Universe
The
Phase
II
NODA
states
the
future
analysis
of
both
this
phase
and
Phase
III
rulemaking
will
be
based
not
on
water
body
type,
but
on
a
regional
basis.
Previously,
EPA
extrapolated
national
cost­
benefit
estimates
from
case
studies
on
different
water
body
types.
The
new
approach
involves
developing
cost
and
benefit
figures
for
eight
different
regions.
Those
eight
regions
include:

1)
Great
Lakes;
2)
Interior
U.
S.;
3)
North
Atlantic;
4)
Mid
Atlantic;
5)
South
Atlantic;
6)
Gulf
of
Mexico;
7)
Northern
California;
and
8)
Southern
California.

Data
describing
the
number
of
potentially
affected
facilities
was
provided
by
EPA
and
are
based
on
survey
response.
EPA
extrapolated
survey
results
and
projected
that
there
are
approximately
925
facilities
within
the
scope
of
the
Phase
III
rule.
Of
this
figure,
EPA
staff
estimated
that
785
would
incur
the
costs
of
installing
new
technology
to
come
into
compliance
with
the
rule.
Approximately
17
percent
or
134
of
the
785
are
electric
generators.
The
remaining
651
are
manufacturers.
Availability
of
detailed
data
concerning
the
exact
number
of
facilities,
current
intake
flow
configuration,
and
technology
requirements
for
compliance,
was
limited
due
to
EPA
obligations
to
protect
confidential
business
information
(
CBI).
Without
further
information,
it
is
assumed
that
each
of
the
785
facilities
will
need
to
upgrade
with
one
or
both
of
the
technologies
to
comply
with
the
proposed
Phase
III
rule.
By
using
information
provided
by
EPA,
it
was
determined
that
approximately
41
percent
of
all
facilities
would
be
required
to
upgrade
to
impingement
technologies
and
the
remaining
59
percent
would
require
both
impingement
and
entrainment
technologies.
These
technologies
will
be
explained
in
Section
6.

Step
Two:
Estimating
Benefits
The
EPA
has
not
released
a
national
benefits
estimate
for
the
Phase
III
rulemaking
process.
The
estimate
used
in
this
exercise
is
based
on
the
two
regional
case
studies
for
the
North
Atlantic
and
Northern
California
detailed
in
the
March
19,
2003
Phase
II
NODA.
Table
15
below
outlines
the
benefit
estimates
from
the
NODA.

Table
15:
Regional
Benefits
As
Calculated
by
EPA
for
Phase
II
(
Dollars)
Benefit
Type
Northern
California
North
Atlantic
Recreational
$
790,560
$
280,000
Commercial
$
22,755
$
3,070,000
Nonuse
Not
Calculated
By
EPA
$
75,640,000
TOTAL
$
813,315
$
78,990,000
Source:
Phase
II
NODA,
40
CFR
Part
125,
March
19,2003
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
49
­
Without
any
additional
data
regarding
the
benefit
values
for
other
regions
or
a
breakdown
as
to
the
number
of
the
785
facilities
that
fall
within
which
region,
a
national
benefits
estimate
was
derived
by
using
the
average
benefit
from
the
two
regions
and
extrapolating
the
data
to
all
8
regions.
Again,
because
EPA
has
not
published
benefit
estimates
for
Phase
III,
the
estimate
developed
in
this
analysis
is
based
on
Phase
II
estimates
as
outlined
in
the
March
19,
2002
NODA.
These
Phase
II
benefit
figures
were
extrapolated
to
all
eight
defined
regions.
The
table
below
provides
benefit
estimates
on
the
regional
and
national
scale
for
Phase
III.
The
total
national
benefit
of
Phase
III
is
estimated
at
$
621.77
million.

Table
16:
Phase
III
Estimated
Commercial
and
Recreational
Benefits
by
Region
(
Millions
of
Dollars)

Benefit
Type
Northern
CA
Region
North
Atlantic
Region
Total
Benefit
Average
Benefit
Per
Region
Total
National
Benefit
(
8
Regions)

Recreational
$
0.079
$
0.28
$
1.07
$
0.535
$
4.282
Commercial
$
0.023
$
3.07
$
3.09
$
1.55
$
12.371
Nonuse
*
$
75.65
$
75.65
$
75.65
$
605.120
TOTAL
$
621.773
*
Nonuse
for
the
Northern
California
Region
was
not
calculated
in
the
March
19,
2002
NODA
It
should
be
noted
that
industry
comments
to
the
Phase
II
NODA
raised
concerns
that
the
benefit
estimate
for
nonuse
values
may
be
overstated.
For
example
in
Table
2
above,
the
nonuse
value
for
the
North
Atlantic
Region
makes
up
over
95
percent
of
the
total
benefit.

Step
Three:
Scaling
Benefits
to
Facility
Size
Category
It
is
assumed
that
there
is
a
proportional
relationship
between
the
amount
of
cooling
water
intake
and
I&
E.
The
benefits
of
reducing
I&
E
were
scaled
to
represent
the
different
design
intake
flow
of
each
facility.
This
scaling
involves
two
steps.
First,
the
total
amount
of
water
withdrawn
per
day
must
be
converted
to
an
average
flow
for
each
of
the
facility
size
categories:
2­
10;
10­
20;
20­
50;
50­
250,
and
250+
MGDs.
According
to
correspondence
with
the
EPA,
the
ratio
of
average
flow
to
design
flow
is
0.48.
Table
17
below
shows
the
percentage
of
total
gallons
withdrawn
per
day
by
facility
size
category
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
50
­
Table
17:
Portion
of
Gallons
Withdraw
by
Facility
Size
Category
Facility
Size
Category
(
MGD)
Estimated
No.
Of
Facilities
Avg.
Design
Intake
Flow
(
MGD)
Avg.
Total
Flow
(
MGD)
Total
Gallons
(
MGD)
Percent
of
Total
Gallons
2
­
10
212
6
3
611
2.19%
10
­
20
131
15
7
943
3.39%
20
­
50
260
35
17
4,368
15.68%
50
­
250
139
150
72
10,008
35.92%
250+
43
578
132
5,676
42.82%

Total
785
21,606
100.00%

The
average
design
flow
is
calculated
by
taking
the
midpoint
of
the
upper
and
lower
bound
of
each
facility
size
category.
For
example,
in
the
10
to
20
MGD
category,
the
midpoint
of
10
and
20
is
15.
The
average
total
flow
is
then
multiplied
by
the
number
of
facilities
to
yield
estimated
total
gallons
per
day
withdrawal.
The
far
right
column
shows
the
percentage
of
total
gallons
withdrawn
that
can
be
attributed
to
each
DIF
category.
For
example,
approximately
16
percent
of
all
water
drawn
can
be
attributed
to
those
facilities
withdrawing
between
20
and
50
million
gallons
per
day.
It
should
be
noted
that
there
was
no
upper
bound
information
for
the
250+
MGD
size
category.
Therefore,
the
figure
selected
is
based
on
the
simple
average
of
the
ratios
of
the
midpoint
to
the
lower
bound
of
each
of
the
for
other
facility
size
categories.
However,
different
averages
affect
the
percentage.
For
example,
when
using
a
smaller
average
the
percent
of
total
is
slightly
less.
If
500
MGD
is
used
the
percentage
drops
slightly
to
32.7.
If
an
average
of
650
MGD
is
used
the
proportion
rises
to
45.7
percent.
These
flow
percentages
are
applied
to
national
benefits
estimate
to
arrive
at
each
DIF
category's
proportion
of
the
benefits.
The
results
are
show
in
Table
11
below.

Table
18:
Total
Estimated
Benefits
Scaled
By
Facility
Size
Categories
Facility
Size
Category
(
MGD)
Percent
of
Total
Gallons
Benefit
Value
(
Millions
of
Dollars)

2
­
10
2.19%
13.63
10
­
20
3.39%
21.05
20
­
50
15.68%
97.49
50
­
250
35.92%
223.36
250+
42.82%
266.25
Total
100.00%
621.77
The
table
above
shows
that
the
larger
facility
size
categories
withdraw
the
most
water
and
therefore
could
yield
the
most
benefits
if
regulated.
Combined,
the
three
largest
groups
(
20­
50,
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
51
­
50­
250,
and
250+)
withdraw
almost
95
percent
of
all
water
withdrawn
and
therefore
raising
the
threshold
to
20
MGD
can
attain
almost
95
percent
of
the
total
benefits
while
exempting
44
percent
of
the
facilities.

Cost
of
Regulatory
Option
1
While
explaining
the
step­
by­
step
cost
estimation
process
outlined
in
Section
5,
the
estimates
for
Regulatory
Option
1
will
be
developed.
The
same
methodology
will
be
used
for
each
subsequent
option.
It
should
be
noted
that
EPA
is
in
the
early
stages
of
developing
a
revised
methodology
to
estimate
costs,
however
no
estimates
have
been
produced
for
Phase
III
at
this
time.
Moreover,
the
problems
associated
with
confidential
business
data
will
likely
lead
to
the
result
that
only
EPA
will
be
able
to
fully
calculate
these
estimates.
The
approach
used
in
this
analysis
is
based
upon
information
gathered
from
communication
with
EPA
staff
and
following
documents:

1.
Phase
II
NODA
40
CFR
125
(
March
19,
2003);

2.
Part
B:
§
316(
b)
Phase
II
Economic
Benefits
Analysis
(
EBA)
Costs
and
Economic
Impacts
from
"
Economic
and
Benefits
Analysis
for
Proposed
Section
316(
b)
Existing
Facilities
Rule,"
U.
S
Environmental
Projection
Agency,
Office
of
Science
and
Technology
Engineering
and
Analysis
Division
(
February
28,
2002);
and
3.
Supplemental
Information
on
Methods
for
Estimating
Costs
of
Cooling
Water
Intake
Structure
Technologies
 
Proposed
Regulations
to
Establish
Requirements
for
Cooling
Water
Intake
Structures
at
Section
316(
b)
Phase
III
Facilities
(
September
27,
2002).

As
previously
discussed,
total
compliance
cost
is
the
aggregate
of
the
following
three
factors:

1.
Capital
Costs;
2.
Operating
and
Maintenance
(
O&
M)
Costs;
and
3.
Permitting
and
Monitoring
(
P&
M)
Costs.

Each
of
these
factors
involves
a
set
of
assumptions
and
cost
curves
that
take
into
account
technology,
retrofit,
and
construction
costs.
The
following
analysis
both
describes
the
costing
methodology
and
presents
the
total
costs
of
compliance
for
the
facility­
size
categories.

Step
One:
Determine
total
design
intake
flow
for
each
facility
EPA
defines
design
intake
flow
as
the
value
assigned
during
the
facilities
design
that
measures
the
total
volume
of
water
withdrawn
per
day.
The
first
step
in
developing
compliance
cost
estimates
is
converting
design
intake
flow
(
DIF)
from
million
gallons
per
day
units
(
MGD)
to
gallons
per
minute
(
GPM).
The
conversion
formula
provided
by
EPA
and
an
example
MGD
to
GPM
conversion
using
25
MGD
is
presented
in
Figure
4
below.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
52
­
Figure
3:
MGD
to
GPM
Conversion
Formula
and
Example
Source:
EPA,
"
Supplemental
Information
on
Methods
of
Estimating
Costs
of
Cooling
Water
Intake
Structure
Technologies,"
September
27,
2002.

Using
the
above
conversion
formula,
Table
12
shows
the
results
of
converting
all
average
facility
DIF
to
GPM.
For
example,
facilities
in
the
10­
20
MGD
size
category
on
average
are
designed
to
withdraw
on
average
15
million
gallons
of
water
per
day.
This
roughly
equates
to
10,417
gallons
per
minute.

Table
12:
Gallons
per
Minute
Conversion
by
Facility
Size
Category
Facility
Size
Category
(
MGD)
Estimated
No.
of
Facilities
Average
Design
Intake
Flow
(
MGD)
Gallons
Per
Minute
(
GPM)

2
­
10
212
6
4,167
10
­
20
131
15
10,417
20
­
50
260
35
24,306
50
­
250
139
150
104,167
250+
43
578
401,388
Step
Two:
Cost
Formulas
For
each
of
the
three
total
cost
components
(
Initial
Capital,
O&
M,
and
P&
M),
EPA
developed
a
set
of
formulas
to
calculate
costs.
The
following
defines
each
cost
component
and
presents
the
formulas
used.

Initial
Capital
Costs
Initial
Capital
costs
are
the
costs
incurred
by
a
facility
for
purchasing
and
installing
technologies
that
reduce
I&
E.
These
technologies
include
fish
handling
and
return
systems,
mesh
screens,
and
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
53
­
nets.
The
table
below
describes
some
of
these
technologies
and
their
targeted
area
of
impingement
or
entrainment
reduction.
For
example,
fish
handling
and
return
systems
primarily
reduce
impingement.

Table
18:
Compliance
Technology
Scenarios
Technology
Targeted
Reduction
Area
Fish
Handling
and
Return
System
Impingement
Fine
Mesh
Traveling
Screens
with
Fish
Handling
and
Return
Both
I&
E
Passive
fine
mesh
screens
at
shoreline
Both
I&
E
Fish
Barrier
Net
Impingement
Add
double
entry,
single
exit
with
fin
mesh
and
fish
handling
and
return
system
Both
I&
E
For
each
of
the
different
technologies
scenarios
above,
EPA
has
developed
a
set
mathematical
formula
to
calculate
the
initial
capital
costs.
These
formulas
are
scaled
to
reflect
changing
costs
due
to
different
total
design
intake
flow
(
in
GPM).
The
formulas
used
in
this
analysis
are
detailed
in
the
table
below.
The
"
X"
variable
is
the
gallons
per
minute
estimate
for
a
given
facility.
Installing
Fishing
handling
and/
or
Return
Systems
reduces
impingement
(
Formulas
A­
D).
Installing
Fine
Mesh
Screens
reduces
entrainment
(
Formulas
E
through
H).
Installing
fine
mesh
screens
and
fish
handling
systems
to
existing
screens
address
both
impingement
and
entrainment.
For
Regulatory
Option
1,
one
formula
from
each
of
the
technology
upgrades
is
applied.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
54
­
Figure
4:
EPA
Initial
Capital
Cost
Equations
Source:
EPA,
"
Supplemental
Information
on
Methods
of
Estimating
Costs
of
Cooling
Water
Intake
Structure
Technologies,"
September
27,
2002.

The
resulting
dollar
values
for
each
Impingement,
Entrainment,
or
both
I&
E
technologies
cost
equation
are
then
multiplied
by
a
number
of
escalation
factors.
These
factors
include
construction
costs
(
30
percent
for
entrainment
technologies
and
15
percent
for
both
I&
E
technologies),
retrofitting
costs
(
30
percent
for
all
technologies),
and
a
total
cost
allowance
of
5
percent.
Based
on
information
provided
by
the
EPA
regarding
technology
costing
modules
applied
to
model
facilities
for
the
Phase
II
NODA,
it
is
projected
that
approximately
49
percent
of
Phase
III
facilities
will
only
adopt
impingement
technologies
and
the
remaining
51
percent
of
facilities
adopt
both.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
55
­
Table
19:
Distribution
of
Compliance
Technologies
Module
#
of
Facilities
Target
1
118
I
2
76
I&
E
3
23
I&
E
4
68
I&
E
5
63
I
6
0
I&
E
7
17
I&
E
8
9
I
9
31
I&
E
10
0
I&
E
11
59
I&
E
Total
Percent
Impingement
Only
190
41%
Entrainment
Only
0
0%
Both
Impingement
&
Entrainment
274
59%
Source:
EPA,
"
Technology
Costing
Modules
Applied
to
Model
Facilities,"
316(
b)
Phase
II
NODA
Cost
Modules.

The
table
below
presents
the
resulting
estimates
for
Initial
Capital
Costs
for
Regulatory
Option
1.
Using
these
above
initial
capital
cost
formulas,
estimates
for
capital
costs
were
developed
for
each
of
the
three
facility
size
categories
outlined
in
Regulatory
Option
One.
Those
categories
are
2­
10,
10­
20,
and
20­
50
MGD.
For
example,
in
the
20­
50
MGD
size
category
there
are
approximately
260
facilities.
Of
this
amount,
107
facilities
are
assumed
to
adopt
only
impingement
technologies
and
153
facilities
adopt
both
I&
E
technologies.
The
table
below
outlines
the
cost
estimation
process
and
results.
For
example,
the
initial
capital
costs
of
compliance
are
estimated
at
$
77.07
million
for
facilities
in
the
20­
50
MGD
facility
size
category.

Table
20:
Initial
Capital
Cost
Estimates
for
Regulatory
Option
1
(
Millions
of
Dollars)

Facility
Size
Category
(
MGD)
No.
of
Facilities
Impingement
Technology
Costs
I&
E
Technology
Costs
Total
Initial
Capital
Costs
2
­
10
212
$
4.69
$
16.65
$
21.34
10
­
20
131
$
3.40
$
15.97
$
19.37
20
­
50
260
$
25.98
$
51.10
$
77.07
50
­
250
139
$
16.29
$
85.76
$
102.05
250+
43
$
11.85
$
41.75
$
53.60
Total
785
$
62.21
$
211.29
$
273.43
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
56
­
Operating
and
Maintenance
(
O&
M)
Costs
Similar
to
the
capital
cost
estimates,
the
O&
M
costs
are
determined
by
using
the
same
GPM
figures
for
each
facility
size
category
and
applying
them
to
the
corresponding
formula.
EPA
developed
O&
M
cost
based
on
information
gathered
from
technology
manufactures
and
vendors.
The
following
table
presents
the
O&
M
formulas
used
in
this
analysis.

Figure
5:
EPA
Operating
and
Maintenance
(
O&
M)
Cost
Formulas
Source:
EPA,
"
Supplemental
Information
on
Methods
of
Estimating
Costs
of
Cooling
Water
Intake
Structure
Technologies,"
September
27,
2002.

By
using
these
cost
curves,
the
following
O&
M
cost
estimates
were
developed
for
Regulatory
Option
1.
Again,
each
facility
size
category's
GPM
figure
is
the
`
X"
variable
and
determines
which
formulas
should
be
applied.
For
example,
in
the
20­
50
MGD
category,
which
draws
24,306
GPM,
the
corresponding
formulas
"
B,"
F,"
or
"
J"
were
applied.
The
table
below
shows
estimated
O&
M
costs
for
each
of
the
facility
size
categories.
The
estimated
total
O&
M
costs
for
Phase
III
facilities
in
the
20­
50
MGD
Facility
Category
is
approximately
$
2.66
million.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
57
­
Table
21:
O&
M
Cost
Estimates
for
Regulatory
Option
1
(
Millions
Dollars)

Facility
Size
Category
(
MGD)
No.
of
Facilities
Impingement
Technology
Costs
Both
I&
E
Technology
Costs
Total
O&
M
Costs
2
­
10
212
$
0.22
$
0.86
$
1.08
10
­
20
131
$
0.15
$
0.74
$
0.88
20
­
50
260
$
0.44
$
2.22
$
2.66
50
­
250
139
$
0.30
$
2.66
$
2.95
250+
43
$
0.18
$
1.21
$
1.39
Total
785
$
1.28
$
7.68
$
8.96
Permitting
and
Monitoring
(
P&
M)
Costs
According
to
EPA
staff,
the
agency
has
not
yet
developed
administrative
costs
specific
to
Phase
III
facilities.
P&
M
costs
include
costs
of
obtaining
NPDES
permits
and
monitoring,
recording
keeping,
and
reporting
of
I&
E
data.
EPA's
initial
Phase
II
analysis
used
a
high
and
low
estimate
for
P&
M.
For
this
Phase
III
analysis,
P&
M
estimates
have
been
developed
using
the
documents
entitled,
"
316(
b)
Phase
II
Economic
Benefit
Analysis,
Part
B:
Costs
and
Economic
Impact."
The
following
table
presents
the
average
monitoring
costs.

Table
22:
Total
Average
Monitoring,
Record
Keeping,
&
Reporting
Costs
Per
Facility
(
Dollars)

Activity
Low
High
Average
Impingement
Sampling
$
16,985
$
21,623
$
19,304
Entrainment
Sampling
$
37,369
$
44,044
$
40,707
Entrainment
Sampling
Capital
and
O&
M
Costs
$
8,300
$
8,300
$
8,300
Visual
or
remote
inspections
$
9,094
$
9,094
$
9,094
Remote
monitoring
and
O&
M
costs
$
250
$
250
$
250
Verification
study
$
6,427
$
6,427
$
6,427
Yearly
Status
Report
$
15,656
$
15,656
$
15,656
Total
$
99,738
Source:
EPA,
"
316(
b)
Phase
II
Economic
Benefit
Analysis,
Part
B:
Costs
and
Economic
Impact,"
Table
B­
4.

Assuming
the
costs
for
the
last
four
items
can
be
divided
between
I&
E
based
on
the
proportions
inherent
in
the
first
three
items,
it
is
estimated
that
impingement­
monitoring
activities
account
for
approximately
$
25
thousand
of
the
total
average
monitoring
cost.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
58
­
Permitting
Costs
The
proposed
rule
would
require
existing
facilities
to
conduct
a
number
of
activities
as
part
of
submitting
an
initial
post­
promulgation
NPDES
permit
application.
Activities
include:

 
collecting
source
water
baseline
biological
characterization
data
and
source
water
body
flow
information;
 
conducting
an
impingement
and
entrainment
characterization
study;
and
 
evaluating
potential
cooling
water
intake
structure
effects.

According
to
the
EPA,
"
Some
of
these
activities
would
be
required
regardless
under
current
case­
by­
case
cooling
water
intake
structure
permitting
procedures 
12"
Because
permitting
costs
appear
to
be
required
regardless
of
Phase
III
regulation,
this
analysis
does
not
include
any
permitting
costs.
The
cost­
benefit
results
would
change
significantly
if
permitting
costs
were
determined
to
be
higher
as
a
result
of
new
Phase
III
requirements.
Increases
in
overall
costs
would
result
in
a
lower
benefit
to
cost
ratio.

Summary:
Cost­
Benefit
Comparison
of
Regulatory
Option
One
Regulatory
Option
One
seeks
to
explore
the
potential
advantages
of
raising
the
threshold
of
compliance
from
2
MGD
to
that
of
10,
20,
or
50
MGD.
This
section
presents
a
summary
and
overall
analysis
of
the
cost­
benefit
exercise
for
option.
Section
2.0
and
2.2
developed
cost
and
benefit
calculations
for
Regulatory
Option
One.
Table
19
compiles
total
cost
estimates
developed
and
compares
these
estimates
with
the
benefits.
For
example,
the
total
estimated
cost
of
compliance
for
facilities
withdrawing
between
20
and
50
MGD
is
$
173.62
million
and
the
benefit
is
125.70
million.
The
far
right
column
entitled
"
Benefit/
Cost
Ratio"
in
the
table
below
represents
the
cost­
effectiveness
or
return
from
each
dollar
spent.
For
instance,
in
the
2­
10
MGD
category
the
benefit
to
cost
ratio
is
0.37.
For
each
dollar
spent
in
complying
with
Phase
III
regulations,
the
return
is
only
0.37
cents
in
reduced
I&
E
benefits.
A
ratio
less
than
to
one
means
that
the
costs
outweigh
the
benefits.

12
"
§
316(
b)
Phase
II
Economic
Benefits
Analysis
(
EBA)
Costs
and
Economic
Impacts
from
"
Economic
and
Benefits
Analysis
for
Proposed
Section
316(
b)
Existing
Facilities
Rule,"
U.
S
Environmental
Projection
Agency,
(
February
28,
2002);
Part
B1­
1.3,
Pg.
B1­
9
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
59
­
Table
23:
Regulatory
Option
One
Summary
of
Costs
&
Benefits
Facility
Size
Category
By
DIF
(
MGD)
Estimated
No.
of
Facilities
Estimated
Total
Benefits
(
Millions
of
Dollars)
Estimated
Total
Costs
(
Millions
of
Dollars)
Benefit/
Cost
Ratio
2
­
10
212
$
13.63
$
37.15
0.37
10
­
20
131
$
21.05
$
29.36
0.72
20
­
50
260
$
97.49
$
97.80
0.99
50
­
250
139
$
223.36
$
114.66
1.95
250+
43
$
266.25
$
54.36
4.90
TOTAL
785
$
621.77
$
333.33
1.85
The
results,
presented
above,
illustrate
that
this
option
becomes
cost­
effectiveness
only
when
the
threshold
is
raised
to
exempt
facilities
withdrawing
less
than
20
MGD.
This
indicates
that
it
is
not
cost­
effective
to
regulate
facilities
of
smaller
sizes.
For
example,
in
the
10­
20
MGD
group
the
return
for
each
dollar
spent
is
.72
cents,
while
the
20­
50
MGD
group
experiences
a
return
of
approximately
one
dollar.

Aggregate
benefit­
cost
ratios
for
exempting
larger
facilities
are
provided
in
the
table
below.
The
table
summarizes
the
findings
of
this
analysis
by
comparing
the
proportion
of
benefits
achieved
to
the
proportion
of
facilities
that
would
be
exempt
if
the
thresholds
were
raised
to
10,
20,
50
or
250+
MGD.
For
example,
almost
95
percent
of
the
total
estimated
benefits
of
the
proposed
Phase
III
regulation
can
be
achieved
by
raising
the
threshold
to
20
MGD.
At
the
same
time,
approximately
79
percent
of
facilities
would
be
exempt
and
would
not
incur
any
costs.

Table
24:
Comparison
of
Costs
&
Benefits
of
Different
Thresholds
Under
Option
One
(
Millions)

Facility
Size
Category
Total
Compliance
Cost
(
Millions)
Total
Benefit
(
Millions)
Benefit/
Cost
Ratio
Percent
of
Total
Benefits
Achieved
Percent
of
Facilities
Exempted
10+
MGD
$
296.18
$
608.15
2.05
97.8%
27.0%
20+
MGD
$
266.83
$
587.10
2.20
94.4%
43.7%
50+
MGD
$
169.02
$
489.61
2.90
78.7%
76.8%
250+
$
54.36
$
266.25
4.90
42.8%
94.5%
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
60
­
Regulatory
Option
Two
 
Impingement
Only
This
section
outlines
the
results
the
cost
and
benefit
analysis
for
Regulatory
Option
Two.
In
this
option,
facilities
withdrawing
less
than
50
MGD
per
day
are
required
to
reduce
only
impingement
mortality.
In
other
words,
all
facilities
with
DIF
less
than
50
MGD
need
only
to
comply
by
adopting
impingement
technologies
and
conducting
impingement­
related
monitoring
activities.

Benefits
of
Regulatory
Option
Two
Isolating
benefits
attributed
strictly
to
impingement
was
a
difficult
process.
Total
dollar
value
benefits
are
not
broken
down
into
impingement
or
entrainment
categories
for
nonuse
or
recreational
benefits.
The
approached
used
here
to
isolate
the
dollar
value
of
"
impingement
benefits"
or
value
of
fish
saved
from
impingement
is
based
on
the
portion
of
commercial
revenue
loss
caused
by
impingement.
Case
study
data
taken
from
the
Phase
II
NODA
for
the
North
Atlantic
and
Northern
California
regions
is
presented
in
Table
24
below.
The
table
shows
impingement
is
attributed
to
11.27
percent
of
the
average
total
commercial
revenue
lost.

Table
25:
Proportion
of
Benefits
Attributed
to
Impingement
Annual
Gross
Commercial
Fishing
Revenue
Loss
in
NORTH
ATLANTIC
Region
(
Dollars)

IMPINGEMENT
ENTRAINMENT
TOTAL
Impingement
Percentage
ESTUARY
$
60,874
$
567,746
$
628,620
OCEAN
$
64,800
$
639,888
$
704,688
TOTAL
$
125,674
$
1,207,634
$
1,333,308
9.43%
Annual
Gross
Commercial
Fishing
Revenue
Loss
in
NORTHERN
CALIFORNIA
Region
(
Dollars)
IMPINGEMENT
ENTRAINMENT
TOTAL
ESTUARY
$
39,082
$
81,039
$
120,121
OCEAN
$
2,920
$
32,096
$
35,016
TOTAL
$
42,002
$
113,135
$
155,137
27.07%

AVERAGE
IMPINGEMENT
PORTION
18.25%
Source:
Phase
II
NODA
(
40
CFR
Part
125,
pp.
13556­
13562)

By
using
the
18.25
percent
proportion
as
a
general
indicator
of
benefits
attributable
to
adopting
impingement
technologies,
the
benefits
of
Regulatory
Option
Two
can
be
calculated.
The
table
below
shows
the
adjusted
benefits
for
facilities
with
DIF
of
less
than
50
MGD.
18.25
percent
of
the
total
benefits
for
each
size
category
are
assigned
to
the
facility
categories
less
then
50
MGD.
The
benefit
estimates
for
the
50­
250
and
250+
facility
size
categories
remain
the
same
since
Regulatory
Option
Two
still
requires
that
facilities
withdrawing
greater
than
50
MGD
be
required
to
adopt
impingement,
entrainment
or
both
I&
E
reduction
technologies.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
61
­
Table
26:
Estimated
Benefits
of
Regulatory
Option
Two
Facility
Size
Category
(
MGD)
Estimated
Benefit
(
Millions
of
Dollars)
2­
10
$
2.49
10­
20
$
3.84
20
­
50
$
17.79
50­
250
$
223.36
250
+
$
266.25
TOTAL
$
513.73
Costs
of
Regulatory
Option
Two
Calculating
the
costs
of
Regulatory
Option
Two
involves
isolating
the
costs
incurred
by
facilities
when
purchasing,
installing,
operating,
and
monitoring
only
impingement
technologies.
The
same
methodology
presented
in
earlier
in
this
is
used.
According
to
EPA,
of
the
estimated
925
Phase
III
facilities,
785
are
expected
to
incur
costs
of
implementing
impingement
or
both
I&
E
technologies.
For
the
purposes
of
this
analysis,
it
is
assumed
that
all
facilities
will
be
required
to
adopt
impingement
technologies.
The
table
below
summaries
the
Capital,
O&
M,
and
P&
M
costs
for
each
facility­
size
category.
For
example,
in
the
10­
20
MGD
category
the
total
costs
of
compliance
is
estimated
to
be
approximately
$
12
million.
It
should
also
be
noted
that
the
total
costs
for
the
largest
two
size
categories,
50­
250
MGD
and
250+
MGD,
remain
the
same
as
in
Option
1
because
these
facilities
will
still
be
required
to
adopt
impingement
and
entrainment
technologies
and
conduct
the
required
monitoring
activities
under
Regulatory
Option
Two.

Table
27:
Regulatory
Option
Two
Costs
(
Millions
of
Dollars)

DIF
(
MGD)
Number
of
Facilities
Initial
Capital
Costs
O&
M
Costs
P&
M
Costs
TOTAL
COSTS
2
­
10
212
$
11.44
$
0.53
$
5.51
$
17.48
10
­
20
131
$
8.29
$
0.35
$
3.41
$
12.05
20
­
50
260
$
25.98
$
1.08
$
6.76
$
33.82
50
­
250
139
$
102.05
$
2.95
$
9.66
$
114.66
250+
43
$
49.98
$
1.39
$
2.99
$
54.36
Total
785
$
197.74
$
6.31
$
28.33
$
232.37
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
62
­
Summary:
Cost­
Benefit
Comparison
of
Regulatory
Option
Two
Regulatory
Option
Two
explores
the
effect
of
requiring
facilities
withdrawing
less
than
50
MGD
(
DIF
<
50
MGD)
to
only
upgrade
with
impingement
technologies.
Facilities
larger
than
50
MGD
will
still
be
required
to
upgrade
to
impingement,
entrainment
or
both
I&
E
technologies.
The
table
below
summarizes
the
cost
and
benefit
estimates
developed
in
this
section.
The
resulting
benefit­
cost
ratios,
provided
in
the
far­
right
column,
illustrate
the
cost­
effectiveness
of
this
regulatory
option.
The
results
indicate
that
the
costs
outweigh
the
benefits
in
all
facility
size
categories
except
the
50­
250
MGD
and
the
250+
MGD.
The
overall
benefit­
cost
ratio
is
2.21.
For
each
dollar
spent
to
comply
with
this
option,
approximately
$
2.21
is
returned
in
the
form
of
reduced
I&
E
benefits.

Table
28:
Regulatory
Option
Two
­
Summary
of
Cost
and
Benefits
(
Millions
of
Dollars)

Facility
Size
Category
(
MGD)
Total
Costs
(
Millions)
Total
Benefits
(
Millions)
Benefit­
Cost
Ratio
2
­
10
$
17.48
$
2.49
0.14
10
­
20
$
12.05
$
3.84
0.32
20
­
50
$
33.82
$
17.79
0.53
50
­
250
$
114.66
$
223.36
1.95
250+
$
54.36
$
266.25
4.90
Total
$
232.37
$
513.73
2.21
The
resulting
analysis
for
Regulatory
Option
Two
illustrates
that
overall
the
option
has
a
positive
return.
However,
option
is
not
become
cost­
effective
until
the
largest
two
categories
are
regulated
(
50­
250
and
250+
MGD).
The
table
below
shows
the
relationship
between
the
proportions
of
benefits
achieved
and
facilities
exempted
if
the
compliance
threshold
were
raised.
For
example,
under
the
requirements
of
this
option
if
the
threshold
was
raise
to
50
MGD,
approximately
96
percent
of
the
benefits
would
be
achieved
while
77
percent
of
all
facilities
would
be
exempted.

Table
29:
Regulatory
Option
Two
­
Costs
&
Benefits
of
Different
Thresholds
(
Millions
of
Dollars)

Facility
Size
Category
Total
Compliance
Cost
Total
Benefit
Benefit/
Cost
Ratio
Percent
of
Total
Benefits
Achieved
Percent
of
Facilities
Exempted
10+
MGD
$
214.89
$
511.24
2.38
99.5%
27.0%

20+
MGD
$
202.84
$
507.40
2.50
98.8%
43.7%

50+
MGD
$
169.02
$
489.61
2.90
95.3%
76.8%

250+
$
54.36
$
266.25
4.90
51.8%
94.5%
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
63
­
Cost
and
Benefits
of
Regulatory
Option
Three
This
section
will
outline
the
estimated
costs
and
benefits
for
Regulatory
Option
Three.
Similar
to
regulatory
option
two,
this
option
requires
all
facilities
withdrawing
less
than
50
MGD
to
only
upgrade
to
meet
impingement
requirements.
However,
Regulatory
Option
Three
does
not
require
facilities
withdrawing
less
than
50
MGD
(
DIF
<
50)
to
conduct
baseline
and
compliance
monitoring.
The
following
sections
will
adjust
the
cost
and
benefit
estimations
to
reflect
these
requirements.

Regulatory
Option
Three
Costs
This
section
will
summarize
the
cost
estimates
for
Regulatory
Option
Three.
The
only
difference
in
costs
between
option
three
and
two
is
the
absence
of
the
monitoring
costs
in
facilities
withdrawing
less
than
50
MGD.
The
largest
two
facility
size
categories,
50­
250
MGD
and
250+
MGD,
will
still
be
required
to
conduct
monitoring
activities.
The
table
below
outlines
the
adjusted
costs.
For
example,
for
the
20­
50
MGD
size
category,
the
total
estimated
compliance
cost
is
$
27.06
million.
This
is
down
from
a
total
of
$
33.82
million
seen
under
the
requirements
for
Regulatory
Option
2.

Table
30:
Regulatory
Option
Three
­
Estimated
Costs
(
Millions
of
Dollars)

Facility
Size
Category
(
MGD)
Number
Of
Facilities
Initial
Capital
Costs
O&
M
Costs
P&
M
Costs
Total
Costs
2
­
10
212
$
11.44
$
0.53
­
$
11.97
10
­
20
131
$
8.29
$
0.35
­
$
8.65
20
­
50
260
$
25.98
$
1.08
­
$
27.06
50
­
250
139
$
102.05
$
2.95
$
9.66
$
114.66
250+
43
$
49.98
$
1.39
$
2.99
$
54.36
Total
785
$
197.74
$
6.31
$
12.65
$
216.70
Benefits
of
Regulatory
Option
Three
There
is
no
data
available
to
develop
an
estimation
of
the
change
in
benefits
as
a
result
of
not
conducting
baseline
monitoring
and
compliance
monitoring.
It
is
assumed
that
a
facility
that
regularly
monitors
and
ensures
that
equipment
is
operating
properly
will
experience
less
impingement
losses.
For
this
sensitivity
analysis,
10
percent
loss
in
effectiveness
was
chosen
to
represent
the
change
in
overall
benefits
resulting
from
not
monitoring.
The
table
below
provides
the
adjusted
benefit
estimates.
The
benefits
for
the
largest
size
categories
remain
the
same
because
only
facilities
withdrawing
less
than
50
MGD
will
be
exempt
from
the
monitoring
requirement.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
64
­
Table
31:
Estimated
Benefits
Changes
Resulting
from
Not
Monitoring
(
Millions
of
Dollars)

Facility
Size
Category
(
MGD)
Estimated
Benefit
of
Impingement
Technologies
Total
Estimated
Benefit
Without
Monitoring
(
10
Percent
Less
Effectiveness)
2­
10
$
2.49
$
2.24
10­
20
$
3.84
$
3.46
20
­
50
$
17.79
$
16.01
50­
250
$
223.36
$
223.36
250+
$
266.25
$
266.25
TOTAL
$
513.73
$
511.32
Note:
Benefit
estimates
remain
the
same
for
the
50­
250
and
250+
MGD
categories
Summary:
Cost­
Benefit
Comparison
of
Regulatory
Option
Three
The
previous
two
sections
outlined
the
cost
and
benefits
of
Regulatory
Option
Three.
This
regulatory
option
requires
facilities
withdrawing
less
than
50
MGD
to
only
comply
by
installing
impingement
technology.
Monitoring
is
not
required.
This
section
will
briefly
summarize
the
cost
and
benefit
data
developed
in
this
analysis.
The
table
below
presents
both
sets
of
data
and
provides
a
benefit­
cost
ratio
to
indication
cost­
effectiveness.

Table
32:
Cost
and
Benefits
of
Regulatory
Option
Three
Facility
Size
Category
(
MGD)
Total
Costs
(
Millions)
Benefits
(
Millions)
Benefit/
Cost
2
­
10
$
11.97
$
2.24
0.19
10
­
20
$
8.65
$
3.46
0.40
20
­
50
$
27.06
$
16.01
0.59
50
­
250
$
114.66
$
223.36
1.95
250+
$
54.36
$
266.25
4.90
Total
$
216.70
$
511.32
2.36
The
resulting
benefit­
costs
ratios
for
Regulatory
Option
Three
show
that
the
costs
outweigh
the
benefits
of
I&
E
reduction
in
all
categories
except
in
the
50­
250
MGD
and
250+
MGD
categories.
It
should
be
noted
that
the
dramatic
change
between
the
20­
50
and
50­
250
MGD
benefit­
cost
ratios
is
due
to
the
fact
that
under
the
option
the
largest
facilities
will
still
be
required
to
adopt
all
technologies
and
conducting
monitoring
activities.
Therefore,
the
benefits
remain
the
same
for
the
50­
250
and
250+
categories.
The
overall
effectiveness
of
this
option
is
measured
by
the
benefit­
cost
ratio
of
2.61.
That
is
$
2.61
are
returned
in
the
form
of
reduced
I&
E
benefits
for
each
dollar
spent
in
compliance.
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
65
­
The
purpose
of
this
study
is
to
explore
the
effect
of
raising
compliance
thresholds
on
the
overall
cost­
effectiveness.
The
table
below
compares
the
results
of
raising
the
flow
threshold
to
that
of
10+,
20+,
50+
and
250+.
For
example,
in
the
10+
category
only
facilities
withdrawing
more
than
10
MGD
would
be
required
to
comply
with
the
requirements
of
Option
Three.
The
remaining
facilities
(
DIF
<
10
MGD)
would
make
no
changes.
Analysis
shows
that
if
the
compliance
threshold
is
raised
to
50
MGD,
approximately
97
percent
of
total
benefits
would
be
achieved
while
exempting
77
percent
of
all
facilities.

Table
33:
Regulatory
Option
Three
­
Benefits
Achieved
and
Facilities
Exempted
Over
Different
Thresholds
Facility
Size
Threshold
(
MGD)
Benefits
(
Millions)
Costs
(
Millions)
Benefit­
Cost
Percent
of
Total
Benefits
Achieved
Percent
of
Total
Facilities
Exempted
10+
501.63
$
186.61
2.69
99.7%
27.0%
20+
499.50
$
182.33
2.74
99.3%
43.7%
50+
489.61
$
169.02
2.90
97.3%
76.8%
250+
266.25
$
54.36
4.90
52.9%
94.5%
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
66
­
7.
Conclusion:
Regulatory
Options
&
Recommendations
In
the
previous
sections
of
this
report,
rough
estimates
of
the
costs
and
benefits
of
the
different
regulatory
scenarios
for
Phase
III
have
been
developed.
The
analysis
provided
should
be
considered
a
preliminary
sensitivity
analysis.
More
refined
estimates
based
on
confidential
business
data
are
currently
being
developed
by
EPA.
The
findings
of
this
analysis
should
be
used
as
a
tool
to
indicate
where
it
may
be
more
cost­
effective
to
exempt
facilities
of
smaller
sizes
from
all
or
part
of
the
regulatory
requirements.
Because
of
EPA
responsibilities
to
protect
confidential
business
data,
exact
results
of
the
costs
and
benefits
of
these
options
may
be
only
be
fully
calculated
by
EPA.
That
said,
it
should
be
noted
that
EPA
does
not
make
decisions
solely
based
on
cost­
benefit
analyses.
Costs
are
weighed
against
a
variety
factors
such
overall
economic
impacts
and
long­
range
environmental
and
human­
health
benefits.
This
cost­
benefit
exercise
provides
an
initial
exploration
of
the
potential
of
alternative
strategies
for
minimizing
negative
economic
impact
to
small
entities
while
still
achieving
a
significant
amount
of
environmental
benefits.
This
section
will
summarize
the
findings
of
the
cost­
benefit
analysis
for
the
three
options
and
conclude
with
some
recommendations.

The
analysis
for
each
of
the
three
regulatory
options
shows
that
the
largest
increases
in
costeffectiveness
can
be
achieved
by
raising
the
thresholds
to
deregulate
smaller
facilities.
The
table
below
summarizes
the
benefit­
cost
ratios
for
each
regulatory
option.
Regulatory
Option
3
appears
to
be
the
most
cost
effective
option
with
a
return
of
$
2.36
for
each
dollar
spent.
However,
deregulating
smaller
facilities
(
Option
1)
provides
almost
the
same
level
of
costeffectiveness
while
eliminating
cumbersome
regulations
on
small
entities
that
provide
few
benefits
per
dollar
invested.

Table
34:
Summary
of
Benefit­
Cost
Ratios
DIF
(
MGD)
Baseline
Ratios
Option
1:
Raising
Compliance
Thresholds
Option
2
Option
3
2
­
10
0.37
10+
MGD
20+
MGD
50+
MGD
250+
MGD
0.14
0.19
10
­
20
0.72
0.72
­
­
­
0.32
0.40
20
­
50
1.00
1.00
1.00
­
­
0.53
0.59
50
­
250
1.95
1.95
1.95
1.95
­
1.95
1.95
250+
4.90
4.90
4.90
4.90
4.90
4.90
4.90
Overall
Benefit­
Cost
Ratio
For
Each
Option
1.85
2.05
2.20
2.90
4.90
2.21
2.36
In
the
summary
sections
for
each
regulatory
option,
a
table
was
provided
that
compared
the
amount
of
benefits
achieved
with
the
number
of
facilities
exempted.
These
figures
are
compiled
into
Table
35
below.
Analysis
for
each
option
explore
the
potential
of
achieving
the
largest
amount
of
benefit
while
exempting
the
largest
amount
of
facilities
in
the
10,
20,
50
or
250+
MGD
thresholds.
For
example,
analysis
for
Regulatory
Option
One
shows
that
almost
95
percent
of
the
benefits
can
be
achieved
while
exempting
44
percent
of
all
facilities
when
the
SBEFA
PANEL
SUPPORT
FOR
PHASE
III
OF
SECTION
316(
B)
FINAL
REPORT
­
67
­
threshold
is
raised
to
20
MGD.
Under
Option
Two,
98.77
percent
of
the
benefits
can
be
achieved
while
exempting
almost
77
percent
of
total
facilities
when
the
threshold
is
raised
to
20
MGD.

Table
35:
Benefits
and
Facilities
Exemption
Comparisons
of
Each
Regulatory
Option
Percent
of
Benefits
Achieved
Percent
of
Facilities
Exempted
DIF
Threshold
(
MGD)
Option
1
Option
2
Option
3
10+
97.81%
99.52%
99.56%
27.0%
20+
94.42%
98.77%
98.89%
43.7%
50+
78.74%
95.31%
95.75%
76.8%
250+
42.82%
51.83%
52.07%
94.5%

Preferred
Regulatory
Alternative:
Raise
the
Threshold
to
20
MGD
In
this
examination
of
the
costs
and
benefits
of
the
three
regulatory
options
for
Phase
III,
eliminating
the
entrainment
technology
retrofit
requirements
was
shown
to
be
not
the
most
cost­
effective
course
of
action.
In
order
to
limit
financial
burden
to
the
largest
number
of
businesses
and
maximize
the
benefits
of
reducing
I&
E,
it
is
recommended
that
the
compliance
threshold
for
Phase
III
be
set
at
20
MGD.
As
Option
1
in
Table
33
above
shows,
almost
95
percent
of
the
benefits
can
be
achieved
and
44
percent
of
facilities
can
be
exempted.
The
table
below
summarizes
these
costs
and
benefits
of
this
approach.
For
each
dollar
spent
in
compliance
yields
a
return
of
$
2.20.

Table
36:
Cost­
Benefit
of
Preferred
Regulatory
Alternative
(
Millions
of
Dollars)

Exemption
Number
of
Affected
Facilities
Number
of
Exempted
Facilities
Benefits
Costs
Benefit­
Cost
Ratio
DIF
Less
than
20
MGD
442
343
$
587.10
$
266.83
2.20
