§
316(
b)
Regional
Case
Studies,
Part
F:
Northern
CaliforniaChapter
F5:
Recreational
Fishing
Benefits
1
Southern
California
residents
visiting
Southern
California
sites
were
not
included
in
the
model.
The
Agency
is
considering
estimating
a
joint
model
for
Northern
and
Southern
California
for
the
final
rule.

NODA
Version
­
March
12,
2003F5­
1
Chapter
F5:
Recreational
Fishing
Benefits
INTRODUCTION
EPA's
regional
case
study
uses
a
random
utility
model
(
RUM)
approach
to
estimate
the
benefits
of
improved
fishing
opportunities
due
to
reduced
impingement
and
entrainment
(
I&
E)
in
the
Northern
California
region.
The
Northern
California
region
is
defined
based
on
National
Marine
Fisheries
Service
(
NMFS)
regional
boundaries,
and
includes
all
northern
counties
to,
and
including,
San
Luis
Obispo
County.

EPA
included
anglers
intercepted
at
sites
in
the
Northern
California
region
in
the
RUM
model.
When
constructing
each
angler's
choice
set,
EPA
included
all
sites
within
140
miles
of
the
angler's
home
zip
code.
Thus,
sites
from
the
Southern
California
region
were
included
for
some
Northern
California
anglers,
to
allow
anglers
from
Northern
California
to
travel
to
all
substitute
sites
located
within
a
one
day
travel
distance
limit.
Also,
anglers
residing
in
Southern
California
who
were
intercepted
at
Northern
California
sites
were
included
in
the
model.
1
Cooling
Water
Intake
Structures
(
CWISs)
withdrawing
water
from
California
coastal
waters
and
estuaries
impinge
and
entrain
many
of
the
species
sought
by
recreational
anglers.
These
species
include
halibut,
flounders,
striped
bass,
various
bottom
fish
species,
and
other
less
prominent
species.

The
study's
main
assumption
is
that,
all
else
being
equal,
anglers
will
get
greater
satisfaction
and
thus
greater
economic
value
from
sites
with
a
higher
catch
rate.
This
benefit
may
occur
in
two
ways:

<
first,
an
angler
may
get
greater
enjoyment
from
a
given
fishing
trip
with
higher
catch
rates,
yielding
a
greater
value
per
trip;
<
second,
anglers
may
take
more
fishing
trips
when
catch
rates
are
higher,
resulting
in
greater
overall
value
for
fishing
in
the
region.

The
following
sections
focus
on
the
data
set
used
in
the
analysis,
and
the
analytic
results.
Chapter
A­
10
of
the
Proposed
Rule
Case
Study
Report
(
http://
www.
epa.
gov/
ost/
316b)
provides
a
detailed
description
of
the
RUM
methodology
used
in
this
analysis.

F5­
1
DATA
SUMMARY
EPA's
analysis
of
improvements
in
recreational
fishing
opportunities
in
Northern
California
relies
on
data
collected
by
the
National
Marine
Fisheries
Service's
(
NMFS)
Marine
Recreational
Fishery
Statistics
Survey
(
MRFSS)
CHAPTER
CONTENTS
F5­
1
Data
Summary..............................
F5­
1
F5­
1.1
Summary
of
Anglers'
Characteristics.......
F5­
2
F5­
1.2
Recreational
Fishing
Choice
Sets
..........
F5­
5
F5­
1.3
Site
Attributes.........................
F5­
5
F5­
1.4
Travel
Cost............................
F5­
7
F5­
2
Site
Choice
Models...........................
F5­
7
F5­
3
Welfare
Estimates............................
F5­
9
F5­
3.1
Estimating
Changes
in
the
Quality
of
Fishing
Sites.............................
F5­
9
F5­
3.2
Estimating
Losses
from
I&
E
in
Northern
California
..............................
F5­
12
F5­
4
Limitations
and
Uncertainty...................
F5­
17
F5­
4.1
Considering
Only
Recreational
Values.....
F5­
17
F5­
4.2
Charter
Anglers.......................
F5­
17
F5­
4.3
Species
and
Mode
Substitution...........
F5­
18
Glossary.......................................
F5­
19
Acronyms......................................
F5­
20
References.....................................
F5­
21
§
316(
b)
Regional
Case
Studies,
Part
F:
Northern
California
Chapter
F5:
Recreational
Fishing
Benefits
combined
with
the
2000
Add­
on
MRFSS
Cost
Survey
(
NMFS,
2000).
2
The
model
of
recreational
fishing
behavior
relies
on
a
subset
of
the
data
that
includes
only
single­
day
trips
to
sites
located
in
Northern
California,
and
sites
in
the
bordering
county
of
Southern
California.
In
addition,
the
sample
excludes
respondents
missing
data
on
key
variables
(
e.
g.,
home
town),
and
includes
only
private/
rental
boat
and
shore
mode
anglers.
The
Agency
did
not
include
charter
boat
mode
anglers
in
the
model
for
the
NODA,
because
charter
boat
trips
represent
only
thirteen
percent
of
the
total
angling
trips
in
this
region.
As
explained
further
below,
the
welfare
gain
to
charter
boat
anglers
from
improved
catch
rates
is
approximated
based
on
the
regression
coefficients
developed
for
the
boat
anglers.
The
final
sample
used
to
estimate
the
RUM
model
includes
5,385
boat
and
shore
anglers.

F5­
1.1
Summary
of
Anglers 
Characteristics
a.
Fishing
modes
and
targeted
species
A
majority
of
the
anglers
in
the
Northern
California
sample
(
62
percent)
fish
from
either
a
private
or
a
rental
boat
(
see
Table
F5­
1
below).
Approximately
26
percent
fish
from
the
shore;
the
remaining
13
percent
fish
from
a
party
or
charter
boat.
In
addition
to
the
mode
of
fishing,
the
MRFSS
contains
information
on
the
specific
species
targeted
on
the
current
trip.
The
most
popular
species,
targeted
by
25
percent
of
anglers,
is
salmon.
The
second
most
popular
species
group,
targeted
by
21
percent
of
anglers,
is
bottom
fish.
3
Approximately
26
percent
of
anglers
did
not
have
a
designated
target
species.
Of
the
remaining
anglers,
ten,
nine,
six,
two,
and
.5
percent
target
small
gamefish,
flounders,
sturgeon,
big
game
fish,
and
other
species,
respectively.

<
Bottom
fish
species
include
sheephead,
greenlings,
cabezon,
sculpins,
croakers,
sea
basses,
rockfishes,
scorpionfish,
hake,
tomcod,
lingcod,
and
wrasses.
<
Small
game
fish
include
striped
bass,
Pacific
bonito,
Pacific
barracuda,
and
small
tunas
and
mackerels.
<
Flounders
include
California
halibut,
rock
sole,
sanddabs,
starry
flounder,
and
other
flounders.
<
Big
game
fish
include
sharks,
yellowtail,
and
other
jacks.
<
Other
species
include
jacksmelt,
silversides,
northern
anchovy,
herrings,
and
surf
smelt.

The
distribution
of
target
species
is
not
uniform
by
fishing
mode.
For
example,
35%
of
private/
rental
boat
anglers
and
28%
of
charter
anglers
target
salmon,
while
less
than
two
percent
of
shore
anglers
target
salmon.
Forty­
seven
percent
of
shore
anglers
do
not
target
a
particular
species,
while
only
20%
of
private/
rental
boat
anglers,
and
13%
of
charter
boat
anglers
do
not
target
a
species.
Almost
58%
of
charter
boat
anglers
target
bottom
fish
species,
while
only
13%
of
private/
rental
boat
anglers,
and
23%
of
shore
anglers
target
bottom
fish.
Fourteen
percent
of
private/
rental
boat
anglers
target
flounders
(
including
halibut),
while
no
charter
anglers
and
less
than
two
percent
of
shore
anglers
target
flounders.
Twenty­
two
percent
of
shore
anglers
target
small
game
fish,
while
only
7%
of
private/
rental
boat
anglers
and
less
than
two
percent
of
charter
boat
anglers
target
small
game
fish.

2
For
general
discussion
of
the
MRFSS
see
Proposed
Section
316(
b)
Phase
II
Existing
Facilities
Rule
Case
Study
Analysis
or
 
Marine
Recreational
Fisheries
Statistics:
Data
User s
Manual, 
NMFS,
http://
www.
st.
nmfs.
gov.
recreational/
index.
html).

3
Bottom
fish
species
include
sheephead,
greenlings,
cabezon,
sculpins,
croakers,
sea
basses,
rockfishes,
scorpionfish,
hake,
tomcod,
lingcod,
and
wrasses.

NODA
Version
­
March
12,
2003
F5­
2
§
316(
b)
Regional
Case
Studies,
Part
F:
Northern
CaliforniaChapter
F5:
Recreational
Fishing
Benefits
4
Income
was
not
reported
by
most
survey
respondents.
Median
household
income
data
by
zip
code,
from
the
U.
S.
Census
in
2000,
was
used
to
provide
income
information
for
respondents
not
reporting
income.

5
All
costs
are
in
2000$,
which
represent
the
MRFSS
survey
year.
All
costs/
benefits
will
be
updated
to
2002$
later
in
this
analysis
(
e.
g.,
for
welfare
estimation).

6
Calculation
of
trip
cost
and
travel
time
is
explained
in
section
F5­
1.4
below.

NODA
Version
­
March
12,
2003F5­
3
Table
F5­
1:
Species
Group
Choice
by
Mode
of
Fishing
Species
Group
All
ModesPrivate/
Rental
BoatParty/
Charter
BoatShore
FrequencyPercentFrequencyPercent
by
ModeFrequencyPercent
by
ModeFrequencyPercent
by
Mode
Small
game61810.39%
2717.28%
101.32%
34722.20%

Bottom
fish85521.43%
49413.27%
44057.97%
36123.10%

Flounders5459.02%
51613.86%
00.00%
291.86%

Big
game921.52%
792.12%
00.00%
130.83%

Salmon1,32225.33%
1,30034.93%
20927.54%
221.41%

Sturgeon3475.74%
3258.73%
00.00%
221.41%

No
target1,57626.08%
73619.77%
10013.18%
74047.34%

Other300.50%
10.03%
00.00%
291.86%

All
species6,044100.00%
3,72261.58%
759
12.56%
1,563
25.86%

Source:
U.
S.
EPA
analysis.

b.
This
section
presents
a
summary
of
angler
characteristics
for
the
Northern
California
region
as
defined
above.
comparison
the
study
uses
only
the
data
included
in
the
RUM
model,
i.
e.
only
data
for
private/
rental
boat
anglers
and
shore
anglers.
t
anglers
and
1,563
shore
anglers.
marizes
information
on
fishing
trips
and
anglers.

The
average
income
of
the
respondent
anglers
was
$
54,882.4
Ninety­
two
percent
of
the
anglers
are
male.
spent
32
days
fishing
during
the
past
year.
erage
trip
cost
for
surveyed
trips
is
$
23.20
(
2000$),
5
and
the
average
one­
way
travel
time
to
the
site
was
about
half
an
hour.
6
The
average
duration
of
a
fishing
trip
was
4.72
hours.
did
not
include
additional
demographic
statistics
(
e.
g.,
race
and
Anglers 
characteristics
For
this
data
This
data
set
includes
5,285
observations:
3,722
boaTable
F5­
2
sum
The
average
angler
The
av
The
California
data
education).
§
316(
b)
Regional
Case
Studies,
Part
F:
Northern
CaliforniaChapter
F5:
Analysis
of
Recreational
Fishing
Benefits
NODA
Version
­
March
12,
2003F5­
4
Table
F5­
2:
ta
Summary
for
the
Northern
California
Coast
Anglers
Variable
All
ModesPrivate/
Rental
BoatShore
NMeanaStd
DevMinMaxNMeanaStd
DevMinMaxNMeanaStd
DevMinMax
Travel
cost5,28523.2025.220.37157.363,72225.0426.090.61157.361,56318.8322.450.37149.82
Travel
time5,2850.530.590.0083.483,7220.570.610.023.481,5630.430.520.013.36
Male5,2380.920.27013,6920.920.27011,5460.910.2901
Annual
trips5,22432.2640.1503603,68630.0034.4803501,53837.6750.860360
Income5,285$
54,882$
18,047$
0$
145,4253,722$
54,373$
18,384$
0$
142,4591,563$
56,094$
17,163$
0$
145,425
Average
trip
length
in
hours
6,7954.7182.0170.5016.504,1495.0921.7830.5016.5001,8873.4091.1590.5010.50
a
For
dummy
variables
such
as
 
Male 
that
take
the
value
of
0
or
1,
the
reported
value
represents
a
portion
of
the
survey
respondents
possessing
the
relevant
characteristic.

example,
92
percent
of
the
surveyed
anglers
are
males.

Source:
U.
S.
EPA
analysis.
Da
For
§
316(
b)
Regional
Case
Studies,
Part
F:
Northern
CaliforniaChapter
F5:
Recreational
Fishing
Benefits
NODA
Version
­
March
12,
2003F5­
5
F5­
1.2
There
are
126
National
Marine
Fisheries
Service
(
NMFS)
intercept
sites
in
the
Northern
California
region.
These
126
fishing
sites
constitute
the
total
choice
set
for
the
Northern
California
region.
California
region
were
generated
based
on
NMFS
sites
located
within
140
miles
of
the
respondent s
home
zip
code.
Distances
from
unique
zip
codes
to
each
of
the
248
NMFS
sites
located
in
California
were
estimated
using
ArcView
3.2a
software.
A
maximum
of
37
sites
defines
the
choice
set
inclusive
of
the
site
actually
visited
at
the
time
of
the
survey.
In
cases
where
more
than
36
additional
sites
are
within
the
140
mile
distance
limit,
36
sites
are
randomly
drawn
from
the
available
sites.
resulting
choice
sets
for
individual
anglers
range
from
6
to
94
sites,
with
the
average
choice
set
containing
71
sites.
F5­
3
summarizes
the
number
of
sites
available,
and
anglers
intercepted,
for
each
county
in
the
Northern
California
region.

Table
F5­
3:
Number
of
Northern
California
Region
Sites
and
Anglers
by
County
CountyNumber
of
SitesNumber
of
Intercepted
Anglers
a
Alameda12650
Contra
Costa5409
Del
Norte6119
Humboldt11379
Marin11388
Mendocino10233
Monterey12409
San
Francisco12326
San
Luis
Obispo10239
San
Mateo15602
Santa
Clara10
Santa
Cruz10745
Solano2530
Sonoma9256
Total1265,285
a
Includes
intercepted
private/
rental
boat
and
shore
mode
anglers
only.
Charter
boat
anglers
are
not
included
as
no
specific
charter
boat
model
of
site
choice
was
estimated.

Source:
U.
S.
EPA
analysis.

F5­
1.3
This
analysis
assumes
that
the
angler
chooses
between
site
alternatives
by
comparing
his/
her
indirect
utility
function
for
each
alternative
and
choosing
the
one
that
maximizes
utility.
llowing
McConnell
and
Strand
(
1994),
this
assumption
states
that
the
individual
first
chooses
fishing
mode
and
species
and
then,
conditional
on
his
choice,
chooses
the
recreational
site
(
Hicks,
Steinback,
Gautam,
and
Thunberg,
1999).

To
measure
site
quality,
this
analysis
uses
catch
rates
for
the
fish
species
of
concern,
as
well
as
the
presence
of
marinas
and/
or
docks
at
each
site,
and
the
presence
of
piers
or
jetties
at
each
site.
the
most
important
attribute
of
a
fishing
site
from
the
angler s
perspective
(
McConnell
and
Strand,
1994;
Haab
et
al.,
2000).
concern
because
catch
rate
is
a
function
of
fish
abundance,
which
is
affected
by
fish
mortality
due
to
I&
E.
Recreational
Fishing
Choice
Sets
Choice
sets
for
individual
anglers
in
the
Northern
The
Table
Site
Attributes
Fo
Catch
rate
is
This
attribute
is
also
a
policy
variable
of
The
catch
rate
§
316(
b)
Regional
Case
Studies,
Part
F:
Northern
CaliforniaChapter
F5:
Recreational
Fishing
Benefits
7
No­
target
anglers
in
Northern
California
caught
fish
in
all
species
groups
included
in
the
model.

NODA
Version
­
March
12,
2003F5­
6
variable
in
the
RUM
therefore
provides
the
means
to
measure
baseline
losses
in
I&
E
and
changes
in
anglers 
welfare
attributed
to
changes
from
I&
E
resulting
from
the
316(
b)
rule.

To
specify
the
fishing
quality
of
the
case
study
sites,
EPA
calculated
historic
catch
rate
based
on
the
NMFS
catch
rates
from
1996
to
2000.
the
model.
on,
flounders,
small
game
fish,
big
game
fish,
bottom
fish,
and
other
species,
as
described
above.
for
all
species
caught
by
anglers
who
did
not
target
a
specific
species.
7
The
catch
rates
represent
the
number
of
fish
caught
on
a
fishing
trip
divided
by
the
number
of
hours
spent
fishing
(
i.
e.,
the
number
of
fish
caught
per
hour
per
angler).
ated
catch
rates
are
averages
across
all
anglers
by
mode,
target
species,
and
site
over
the
five­
year
period
(
1996­
2000).

The
catch
rate
variables
include
total
catch,
including
fish
caught
and
kept
and
fish
released.
e
NMFS
studies
use
the
catch­
and­
keep
measure
as
the
relevant
catch
rate.
eater
error
may
be
associated
with
measured
number
of
fish
not
kept,
the
total
catch
measure
is
most
appropriate
because
a
large
number
of
anglers
catch
and
release
fish.
h
rate
variables
include
both
targeted
fish
catch
and
incidental
catch.
ple,
small
game
catch
rates
include
fish
caught
by
small
game
anglers,
anglers
targeting
another
species
group
but
who
actually
caught
a
small
game
fish,
and
anglers
who
don t
target
any
particular
species.
eted
species
catch
rates
for
this
analysis,
but
discovered
that
this
approach
did
not
provide
a
sufficient
number
of
observations
to
allow
estimation
of
catch
rates
for
all
fishing
sites
included
in
the
analysis.

Anglers
who
target
particular
species
generally
catch
more
fish
in
the
targeted
category
than
anglers
who
do
not
target
these
species,
because
of
specialized
equipment
and
skills.
summarizes
average
catch
rates
by
species
for
all
sites
in
the
study
area.
ode
anglers
who
target
particular
species,
bottom
fish
anglers
catch
the
largest
number
of
fish
per
hour
(
1.09),
followed
by
anglers
who
target
species
in
the
 
other 
category
(.
83),
those
who
target
small
game
(
0.62),
those
who
target
flounders
(.
55),
those
who
target
big
game
(.
46),
those
who
target
salmon
(.
34),
and
those
who
target
sturgeon
(.
21).
ode
anglers
who
target
particular
species,
anglers
who
target
fish
in
the
 
other 
category
catch
the
largest
number
of
fish
per
hour
(
1.67),
followed
by
anglers
who
target
flounders
(
1.57),
those
who
target
bottom
fish
(
0.95),
those
who
target
big
game
fish
or
sturgeon
(.
52),
those
who
target
small
game
(.
51),
and
those
who
target
salmon
(.
28).

Table
F5­
4:
Average
Catch
Rate
by
Species/
Species
Group
for
the
Northern
California
Coast
Sites
by
Mode
of
Fishing
Species/
Species
Group
Average
Catch
Rate
(
fish
per
angler
per
hour)

All
SitesSites
with
Non
Zero
Catch
Rates
Private/
Rental
BoatShorePrivate/
Rental
BoatShore
Small
game0.150.200.620.51
Other
bottom
fish0.400.661.090.95
Flounders0.160.350.551.57
Big
game0.110.070.460.52
Salmon0.090.020.340.28
Sturgeon0.020.020.210.52
No
target0.250.560.750.87
Other0.120.650.831.67
Source:
U.
S.
EPA
analysis.
Six
species
or
species
groups
were
included
in
They
are
sturgeon,
salm
For
no­
target
anglers,
EPA
calculated
average
catch
The
estim
Som
Although
a
gr
The
total
catc
For
exam
EPA
considered
using
targ
Table
F5­
4
Of
the
boat
m
Of
the
shore
m
§
316(
b)
Regional
Case
Studies,
Part
F:
Northern
California
Chapter
F5:
Recreational
Fishing
Benefits
Some
RUM
studies
have
used
predicted,
rather
than
actual,
catch
rates
(
Haab
et
al.,
2000;
Hicks
et
al.,
1999;
McConnell
and
Strand,
1994).
to
affect
catch
rates;
for
example,
anglers
with
different
levels
of
experience
may
have
different
catch
rates.
al.
(
2000)
compared
historic
catch­
and­
keep
rates
to
predicted
catch­
and­
keep
rates
and
found
that
historic
catch­
and­
keep
rates
were
a
better
measure
of
site
quality.
found
that
using
historic
catch
rates
resulted
in
more
conservative
welfare
estimates
than
predicted
catch
rate
models.
Consequently,
EPA
used
this
more
conservative
approach.

F5­
1.4
EPA
used
ArcView
3.2a
software
to
estimate
distances
from
the
household
zip
code
to
each
NMFS
fishing
site
in
the
individual
opportunity
sets.
locations
from
the
Master
Site
Register
supplied
by
NMFS.
The
Master
Site
Register
includes
both
a
unique
identifier
that
corresponds
to
the
visited
site
used
in
the
angler
survey,
and
latitude
and
longitude
coordinates.
e
sites
the
latitude
and
longitude
coordinates
were
missing
or
demonstrably
incorrect,
in
which
case
the
town
point,
as
identified
in
the
U.
S.
Geological
Survey
(
USGS)
Geographic
Names
Information
System,
was
used
as
the
site
location
if
a
town
was
reported
in
the
site
address.
measured
the
distance
in
miles
of
the
shortest
route,
using
state
and
U.
S.
highways,
from
the
household
zip
code
to
each
fishing
site,
then
added
the
distances
from
the
zip
location
to
the
closest
highway
and
from
the
site
location
to
the
closest
highway.
to
the
visited
site
for
all
modes
is
21.2
miles.
Private/
rental
boat
anglers
traveled
further,
on
average,
to
the
chosen
site
than
shore
anglers,
going
22.91
miles
versus
17.06
miles.

EPA
estimated
trip
 
price 
as
the
sum
of
travel
costs
plus
the
opportunity
cost
of
time
following
the
procedure
described
in
Haab
et
al.
(
2000).
udy
assumed
that
time
spent
 
on­
site 
is
constant
across
sites
and
can
be
ignored
in
the
price
calculation.
ate
anglers 
travel
costs,
EPA
multiplied
round­
trip
distance
by
average
motor
vehicle
cost
per
mile
($
0.325,
2000$).
8
To
estimate
the
opportunity
cost
of
travel
time,
EPA
first
divided
round­
trip
distance
by
40
miles
per
hour
to
estimate
trip
time,
and
used
one
third
of
the
household s
wage
to
yield
the
opportunity
cost
of
time.
ated
household
wage
by
dividing
household
income
by
2,080
(
i.
e.,
the
number
of
full
time
hours
potentially
worked).

EPA
calculated
visit
price
as:

(
F5­
1)

F5­
2
SITE
CHOICE
MODELS
The
nature
of
the
MRFSS
data
leads
to
the
RUM
as
a
means
of
examining
anglers 
preferences
(
Haab,
Whitehead,
and
McConnell,
2000).
among
a
set
of
feasible
sites.
es
that
the
individual
angler
makes
a
choice
among
mutually
exclusive
site
alternatives
based
on
the
attributes
of
those
alternatives
(
McFadden,
1981).
ber
of
feasible
choices
(
J)
in
each
angler s
choice
set
was
set
to
37
sites
within
140
miles
of
the
angler s
home.

An
angler s
choice
of
sites
relies
on
utility
maximization.
j
if
the
utility
(
uj)
from
visiting
site
j
is
greater
than
that
from
vising
other
sites
(
h),
such
that:

(
F5­
2)

Anglers
choose
the
species
to
seek
and
the
mode
of
fishing
in
addition
to
choosing
a
fishing
site.
odes
This
practice
allows
for
individual
characteristics
Haab
et
Hicks
et
al.
(
1999)

Travel
Cost
The
Agency
obtained
fishing
site
For
som
The
program
The
average
one­
way
distance
Based
on
Parsons
and
Kealy
(
1992),
this
st
To
estim
EPA
estim
Anglers
arrive
at
each
NMFS
site
by
choosing
The
RUM
assum
The
num
An
angler
will
choose
site
Available
fishing
m
8
EPA
used
the
2000
government
rate
($
0.325)
for
travel
reimbursement
to
estimate
travel
costs
per
mile
traveled.
This
estimate
includes
vehicle
operating
cost
only.

NODA
Version
­
March
12,
2003
F5­
7
§
316(
b)
Regional
Case
Studies,
Part
F:
Northern
California
Chapter
F5:
Recreational
Fishing
Benefits
include
shore
fishing,
fishing
from
charter
boats,
or
fishing
from
private
or
rental
boats.
The
target
species
or
group
of
species
include
small
game,
bottom
fish,
flounders,
big
game
fish,
salmon,
sturgeon
and
other
fish.
Anglers
may
also
choose
not
to
target
any
particular
species.

Recreational
fishing
models
generally
assume
that
anglers
first
choose
a
mode
and
species,
and
then
a
site.
The
nested
logit
model
is
generally
used
for
recreational
demand
models,
as
it
avoids
the
independence
of
relevant
alternatives
(
IIA)
problem,
in
which
sites
with
similar
characteristics
that
are
not
included
in
the
model
have
correlated
error
terms.
However,
for
the
Northern
California
RUM,
the
estimated
inclusive
value
parameters
for
a
nested
model
of
shore
and
boat
anglers
were
significantly
greater
than
one,
indicating
that
nesting
was
not
appropriate
for
the
data.
Consequently,
EPA
estimated
separate
logit
models
for
boat
and
shore
anglers.
The
Agency
did
not
include
the
angler s
choice
of
fishing
mode
and
target
species
in
the
model,
instead
assuming
that
the
mode/
species
choice
is
exogenous
to
the
model
and
that
the
angler
simply
chooses
the
site.
EPA
used
the
following
general
model
to
specify
the
deterministic
part
of
the
utility
function:
9
where:

v
TCj
SITE­
ATTRIBUTESj
SQRT(
Qjs
)
Flag(
s)
(
F5­
3)

=
the
expected
utility
for
site
j
(
j=
1,...
37);
=
travel
cost
for
site
j;
=
presence
of
marinas
or
docks,
and
piers
or
jetties
at
site
j;
=
square
root
of
the
historic
catch
rate
for
species
s
at
site
j;
10
and
=
1
if
an
angler
is
targeting
this
species;
0
otherwise.

The
analysis
assumes
that
each
angler
in
the
estimated
model
considers
site
quality
based
only
on
the
catch
rate
for
the
targeted
species.
Theoretically,
an
angler
may
catch
any
of
the
available
species
at
a
given
site
(
McFadden,
1981).
If,
however,
an
angler
truly
has
a
species
preference,
then
including
the
catch
variable
for
all
species
available
at
the
site
would
inappropriately
attribute
utility
to
the
angler
for
a
species
not
pursued
(
T.
Haab,
J.
Whitehead,
and
T.
McConnell,
2000).
To
avoid
this
problem,
the
Agency
used
an
interaction
variable
SQRT
(
Qjs
)
×
Flag
(
s),
such
that
the
catch
rate
variable
for
a
given
species
is
turned
on
only
if
the
angler
targets
a
particular
species
(
Flag
(
s)
=
1).
The
Agency
calculated
a
separate
catch
rate
for
no
target
anglers,
using
the
average
of
all
species
caught
by
no
target
anglers.
The
final
model
presented
here
is
a
site
choice
model
that
includes
all
fish
species.
The
analysis
therefore
assumes
that
each
angler
has
chosen
a
mode/
species
combination
followed
by
a
site
based
on
the
catch
rates
for
that
site
and
species.
EPA
estimated
all
RUM
models
with
LIMDEPTM
software
(
Greene,
1995).
Table
F5­
5
gives
the
parameter
estimates
for
this
model.

9
See
Chapter
A­
10
of
the
Proposed
Section
316(
b)
Phase
II
Existing
Facilities
Rule
Case
Study
Analysis
report
for
detail
on
model
specification.

10
The
analysis
used
the
square
root
of
the
catch
rate
to
allow
for
decreasing
marginal
utility
of
catching
fish
(
McConnell
and
Strand,
1994).

NODA
Version
­
March
12,
2003
F5­
8
§
316(
b)
Regional
Case
Studies,
Part
F:
Northern
CaliforniaChapter
F5:
Recreational
Fishing
Benefits
NODA
Version
­
March
12,
2003F5­
9
Table
F5­
5:
Estimated
Coefficients
for
the
Conditional
Site
Choice
Variable
Private/
Rental
Boat
ModelShore
Model
Estimated
Coefficientt­
statisticEstimated
Coefficientt­
statistic
TRIPCST­
0.0896­
53.82­
0.1091­
34.19
SQRT
(
Qsalmon)
5.798023.405.56154.78
SQRT
(
Qsturgeon)
16.697210.10N/
AN/
A
SQRT
(
Qflounder)
2.342610.430.93063.82
SQRT
(
Qsmall
game)
0.82448.042.10388.81
SQRT
(
Qbottom)
1.03206.140.42112.09
SQRT
(
Qbig
game)
3.14765.28N/
AN/
A
SQRT
(
Qno
target)
2.078313.930.46565.58
Marina/
Dock0.41959.38­
1.8644­
17.60
Pier/
Jetty­
0.6808­
17.320.618610.59
Source:
U.
S.
EPA
analysis.

One
disadvantage
of
the
specified
model
is
that
the
model
looks
at
site
choice
without
regard
to
mode
or
species,
whereas
species
selection
is
an
integral
part
of
the
nested
RUM.
across
species
(
i.
e.,
the
value
of
catching,
or
potentially
catching,
a
different
species
is
not
included
in
the
calculation).
Therefore,
improvements
in
fishing
circumstances
related
to
other
species
will
have
no
effect
on
anglers 
choices.

Table
F5­
5
shows
that
all
coefficients
have
the
expected
signs
and
are
statistically
significant
at
the
95th
percentile.
cost
has
a
negative
effect
on
the
probability
of
selecting
a
site,
indicating
that
anglers
prefer
to
visit
sites
closer
to
their
homes
(
other
things
being
equal).
arina/
dock
variable
and
the
negative
coefficient
on
the
pier/
jetty
variable
for
boat
anglers
indicates
that
anglers
fishing
from
boats
are
more
likely
to
choose
sites
with
marinas
or
docks,
and
less
likely
to
choose
sites
with
piers
or
jetties.
these
variables
are
reversed
for
shore
anglers,
indicating
that
shore
anglers
prefer
sites
with
piers
or
jetties,
and
are
less
likely
to
fish
from
marinas
or
docks.

For
all
species,
the
probability
of
a
site
visit
increases
as
the
historic
catch
rate
for
fish
species
increases.
toric
catch
rates
averaged
over
all
species
caught
by
no­
target
anglers
to
characterize
fishing
site
quality
for
no­
target
anglers.
Many
species
can
contribute
to
sites 
perceived
quality
for
no­
target
anglers
because
they
catch
whatever
bites.
no­
target
anglers
select
sites
with
higher
historic
catch
rates.

F5­
3
WELFARE
ESTIMATES
This
section
presents
estimates
of
welfare
losses
to
recreational
anglers
from
fish
mortality
due
to
I&
E,
and
potential
welfare
gains
from
improvements
in
fishing
opportunities
due
to
reduced
fish
mortality
stemming
from
the
316(
b)
rule.

F5­
3.1
To
estimate
changes
in
the
quality
of
fishing
sites
under
different
policy
scenarios,
EPA
relied
on
the
recreational
fishery
landings
data
by
state
and
the
estimates
of
recreational
losses
from
I&
E
on
the
relevant
species
corresponding
to
different
technology
options.
ded
the
recreational
fishery
landings
data
for
the
Northern
California
region.
ated
the
losses
to
recreational
fisheries
using
the
physical
impacts
of
I&
E
on
the
relevant
fish
species
and
the
percentage
of
total
fishery
landings
attributed
to
recreational
fishery,
as
described
in
Section
D.
2
of
the
316(
b)
NODA.
Once
an
angler
chooses
a
target
species
no
substitution
is
allowed
Trip
The
positive
coefficient
on
the
m
The
signs
on
EPA
used
his
In
general,

Estimating
Changes
in
the
Quality
of
Fishing
Sites
The
National
Marine
Fisheries
Service
provi
EPA
estim
§
316(
b)
Regional
Case
Studies,
Part
F:
Northern
California
Chapter
F5:
Recreational
Fishing
Benefits
The
Agency
estimated
changes
in
the
quality
of
recreational
fishing
sites
under
different
policy
scenarios
in
terms
of
the
percentage
change
in
the
historic
catch
rate.
EPA
assumed
that
catch
rates
will
change
uniformly
across
all
marine
fishing
sites
in
Northern
California,
because
species
considered
in
this
analysis
inhabit
the
entire
coast.
EPA
used
five­
year
recreational
landing
data
(
1996
through
2000)
for
state
waters
to
calculate
an
average
landing
per
year
for
each
species
included
in
the
model.
11
EPA
then
divided
losses
to
the
recreational
fishery
from
I&
E
by
the
total
recreational
landings
for
Northern
California
to
calculate
the
percent
change
in
historic
catch
rate
from
eliminating
I&
E
completely.
Table
F5­
6
presents
results
of
this
analysis.
Table
F5­
7
presents
estimated
improvements
in
catch
rates,
over
baseline
losses,
for
the
preferred
technology
option
at
each
facility
in
Northern
California.

11
State
waters
include
sounds,
inlets,
tidal
portions
of
rivers,
bays,
estuaries,
and
other
areas
of
salt
or
brackish
water,
plus
ocean
waters
to
three
nautical
miles
from
shore
(
NMFS
2001a,
http://
www.
st.
nmfs.
gov.
recreational/
index.
html).

NODA
Version
­
March
12,
2003
F5­
10
§
316(
b)
Regional
Case
Studies,
Part
F:
Northern
CaliforniaChapter
F5:
Recreational
Fishing
Benefits
NODA
Version
­
March
12,
2003F5­
11
Table
F5­
6:
Estimated
Changes
in
Catch
Rates
from
Eliminating
All
I&
E
of
Affected
Species
in
Northern
California
Estimated
Recreational
Fishery
Losses
from
I&
E
Total
Recreational
Landings
for
Northern
California
(
fish
per
year)
a
Percent
Increase
in
Recreational
Catch
from
Elimination
of
I&
E
Species
by
Species
Group
Forgone
Recreational
Catch
from
I&
E
Flounders:

California
halibut1,250
Other
flounders127
Total
for
Flounders1,377238,394.578%

Small
game
Fish:

Striped
bass257,431
Total
for
Small
Game257,431459,56356.016%

Bottom
fish:

Cabezon164
Croakers928
Rockfishes18,917
Sculpins38,471
Surfperches182,610
Total
for
Bottom
Fish241,0893,665,5206.595%

Other
fish:

Silversides78,830
Smelts217
Total
for
Other
Fish79,0471,442,3565.480%

Total
for
All
Species
578,9445,795,8339.989%

a.
Source:
The
Marine
Recreational
Fishery
Statistics
Survey,
1996­
2000.
landings
are
calculated
as
a
five
year
average
(
1996­
2000)
for
state
waters.
Total
recreational
§
316(
b)
Regional
Case
Studies,
Part
F:
Northern
CaliforniaChapter
F5:
Recreational
Fishing
Benefits
NODA
Version
­
March
12,
2003F5­
12
Table
F5­
7:
Estimated
Changes
in
Catch
Rates
from
Reducing
I&
E
of
Affected
Species
in
Northern
California,
under
the
Preferred
Policy
Option
Estimated
Change
in
Recreational
Fishery
Losses
under
the
Preferred
Policy
OptionTotal
Recreational
Landings
for
Northern
California
(
fish
per
year)
a
Percent
Increase
in
Recreational
Catch
from
Reduction
in
I&
ESpecies
by
Species
Group
Estimated
Reduction
if
Forgone
Recreational
Catch
Flounders:

California
halibut666
Other
flounders96
Total
for
Flounders762238,394.320%

Small
game
fish:

Striped
bass68,615
Total
for
Small
Game68,615459,56314.931%

Bottom
fish:

Cabezon123
Croakers740
Rockfishes15,099
Sculpins21,611
Surfperches146,077
Total
for
Bottom
Fish183,6513,665,5205.024%

Other
fish:

Silversides62,588
Smelts171
Total
for
Other
Fish62,7601,442,3564.351%

Total
for
All
Species
315,7885,795,8335.449%

a
Source:
The
Marine
Recreational
Fishery
Statistics
Survey,
1996­
2000.
recreational
landings
are
calculated
as
a
five
year
average
(
1996­
2000)
for
state
waters.

F5­
3.2
The
recreational
behavior
model
described
in
the
preceding
sections
provides
a
means
for
estimating
the
economic
effects
of
changes
in
recreational
fishery
losses
from
I&
E
in
Northern
California.
ated
welfare
gain
to
recreational
anglers
from
eliminating
fishery
losses
due
to
I&
E.
timate
represents
economic
damages
to
recreational
anglers
from
I&
E
of
recreational
fish
species
in
Northern
California
under
the
baseline
scenario.
ated
benefits
to
recreational
anglers
due
to
I&
E
reduction
under
the
preferred
policy
option.

EPA
estimated
anglers 
willingness
to
pay
for
improvements
in
the
quality
of
recreational
fishing
due
to
I&
E
elimination
by
Total
Estimating
Losses
from
I&
E
in
Northern
California
First,
EPA
estim
This
es
EPA
then
estim
§
316(
b)
Regional
Case
Studies,
Part
F:
Northern
California
Chapter
F5:
Recreational
Fishing
Benefits
first
calculating
an
average
per
trip
welfare
gain
based
on
the
expected
changes
in
catch
rates
from
eliminating
I&
E.
Table
F5­
8
presents
the
compensating
variation
per
trip
(
averaged
over
all
anglers
in
the
sample)
associated
with
reduced
fish
mortality
from
eliminating
I&
E
for
each
fish
species
group
of
concern.
Table
F5­
8
also
shows
the
per­
trip
welfare
gain
attributable
to
reduced
I&
E
resulting
from
the
preferred
technology
option.
12
Table
F5­
8
shows
that
shore
anglers
targeting
small
game
fish
(
primarily
striped
bass)
have
the
largest
per
trip
gain
($
3.37)
from
eliminating
I&
E
in
Northern
California.
Boat
anglers
targeting
small
game
fish
also
have
a
per­
trip
welfare
gain
of
$
1.19.
Striped
bass
is
a
small
game
species
prized
for
both
its
fighting
skills
and
taste.
While
no­
target
anglers
who
fish
from
boats
have
the
second
highest
welfare
gain
per
trip
($
2.66),
no­
target
anglers
who
fish
from
shore
have
the
lowest
per­
trip
gain.
This
is
likely
to
be
the
case
because
of
the
species
caught
by
no­
target
anglers
from
boats
versus
from
shore.
Over
three­
quarters
of
no­
target
boat
anglers
catch
small
game
fish,
flounders,
or
bottom
fish.
Over
half
of
shore
no­
target
anglers
catch
bottom
fish;
22
percent
catch
other
fish;
and
15
percent
catch
small
game
fish.
In
general,
the
fish
caught
by
no­
target
anglers
from
boats
are
likely
to
be
larger,
and
thus
more
valuable,
than
those
caught
by
no­
target
anglers
from
shore.

Table
F5­
8
also
reports
the
willingness
to
pay
(
WTP)
for
a
one­
unit
increase
in
historic
catch
rate
by
species.
The
value
of
increasing
the
historic
catch
rate
varies
significantly
by
species
and
by
fishing
mode.
For
boat
anglers
who
target
specific
species,
sturgeon
are
the
most
highly
valued
fish,
followed
by
salmon,
big
game
fish,
flounders,
and
bottom
fish.
For
shore
anglers
who
target
specific
species,
salmon
are
most
highly
valued,
followed
by
small
game
fish,
flounders,
and
bottom
fish.
Boat
anglers
have
higher
values
than
shore
anglers
for
flounders
and
bottom
fish.
Most
boat
anglers
who
target
fish
in
the
flounders
category
catch
halibut,
which
are
much
larger
than
other
species
in
the
flounders
category.
Few
shore
anglers
catch
halibut.
Shore
anglers
have
higher
values
for
small
game
fish
than
boat
anglers.
No­
target
anglers
who
fish
from
boats
have
much
higher
values
per
fish
than
no­
target
anglers
who
fish
from
shore.
This
might
be
explained
by
the
fact
that
boat
anglers
generally
catch
more
valuable
species.

12
A
compensating
variation
equates
the
expected
value
of
realized
utility
under
the
baseline
and
post­
compliance
conditions.
For
more
detail
see
Chapter
A­
10
of
the
Proposed
Section
316(
b)
Phase
II
Existing
Facilities
Rule
Case
Study
Analysis..

NODA
Version
­
March
12,
2003
F5­
13
§
316(
b)
Regional
Case
Studies,
Part
F:
Northern
CaliforniaChapter
F5:
Recreational
Fishing
Benefits
NODA
Version
­
March
12,
2003F5­
14
Table
F5­
8:
Per
Trip
Welfare
Gain
from
Eliminating
I&
E
and
from
I&
E
Reductions
under
the
Preferred
Policy
Option
in
Northern
California
Targeted
Species
Group
Per
Trip
Welfare
Gain
(
2002$)
WTP
for
an
Additional
Fish
per
Trip
(
2002$)

Eliminating
I&
EReduced
I&
E
with
Preferred
Technology
Boat
Anglers
Shore
AnglersBoat
AnglersShore
AnglersBoat
AnglersShore
Anglers
Flounders$
0.32$
0.96$
0.02$
0.01$
2.97$
0.99
Small
game
fish
$
1.19$
3.37$
0.32$
0.96$
0.76$
3.55
Bottom
fish$
0.24$
0.11$
0.18$
0.08$
0.75$
0.54
Other
fishaN/
A$
0.58N/
A$
0.46N/
A$
1.10
No
targetb$
2.66$
0.02$
2.48$
0.00d$
8.53$
0.76
SalmoncN/
AN/
AN/
AN/
A$
9.40$
10.66
SturgeoncN/
AN/
AN/
AN/
A$
33.50N/
Ad
Big
game
fishbN/
AN/
AN/
AN/
A$
4.05N/
A
a
Not
targeted
by
boat
anglers
in
the
sample
b
The
value
is
based
on
all
species
caught
by
no­
target
anglers.
c
Not
estimated
due
to
limitations
of
I&
E
data.
d
Mean
value
of
the
welfare
gain
from
reduced
I&
E
under
the
preferred
technology
option
for
no
target
anglers
for
shore
mode
was
negligible.

Source:
U.
S.
EPA
analysis.

EPA
calculated
the
total
economic
value
of
eliminating
I&
E
in
Northern
California
by
combining
the
estimated
per
trip
welfare
gain
with
the
total
number
of
fishing
days
in
the
Northern
California
region.
ation
on
the
total
number
of
fishing
trips
by
state
and
by
fishing
mode;
this
total
number
of
fishing
days
includes
both
single­
and
multiple­
day
trips.
ber
of
fishing
days
by
fishing
mode.

Table
F5­
9:
Recreational
Fishing
Participation
in
2001
by
Fishing
Mode
for
Northern
California
Fishing
Mode
Total
Number
of
Fishing
Days
per
Year
Private
rental
boat920,196
Shore864,178
Charter
boat193,007
Total1,977,381
Source:
NMFS,
http://
www.
st.
nmfs.
gov.
recreational/
index.
html).
NMFS
provided
inform
Table
F5­
9
presents
the
NMFS
num
§
316(
b)
Regional
Case
Studies,
Part
F:
Northern
California
Chapter
F5:
Recreational
Fishing
Benefits
Per
trip
welfare
gain
differs
across
recreational
species
and
fishing
mode.
13
EPA
therefore
estimated
the
number
of
fishing
trips
associated
with
each
species
of
concern
and
the
number
of
trips
taken
by
no­
target
anglers.
EPA
used
the
MRFSS
sample
to
calculate
the
proportion
of
recreational
fishing
trips
taken
by
no­
target
anglers
and
anglers
targeting
each
species
of
concern
and
applied
these
percentages
to
the
total
number
of
trips
to
estimate
species­
specific
participation.
Table
F5­
10
shows
the
calculation
results.
No­
target
anglers
account
for
the
largest
number
of
fishing
days
in
Northern
California
(
616,463
days),
followed
by
anglers
targeting
bottom
fish,
salmon,
small
game
fish,
flounders,
sturgeon,
big
game
fish,
and
other
fish.

The
estimated
number
of
trips
represents
the
baseline
level
of
participation.
Anglers
may
take
more
fishing
trips
as
recreational
fishing
circumstances
change.
However,
EPA
was
unable
to
estimate
a
trip
participation
model
for
Northern
California,
because
the
required
data
were
not
available.
Therefore,
the
welfare
estimates
presented
here
do
not
account
for
likely
increases
in
the
number
of
trips
due
to
elimination
of
I&
E,
and
thus
understate
total
welfare
effects.

13
EPA
used
the
per
trip
values
for
private/
rental
boat
anglers
to
estimate
welfare
gains
for
charter
boat
anglers.

NODA
Version
­
March
12,
2003
F5­
15
§
316(
b)
Regional
Case
Studies,
Part
F:
Northern
CaliforniaChapter
F5:
Recreational
Fishing
Benefits
NODA
Version
­
March
12,
2003F5­
16
Table
F5­
10:
Recreational
Fishing
Participation
by
Species
and
Fishing
Mode
Species
Mode:
Private
Rental
Boats
Number
of
Fishing
Days
Mode:
Shore
Number
of
Fishing
DaysMode:
Charter
Boat
Number
of
Fishing
Days
Total
for
all
Modes
a
Flounders127,53916,0740143,613
Small
Game
fish66,990191,8482,548261,385
Bottom
fish122,110199,625111,886433,621
Other
fish27616,074016,350
No
target181,923409,10225,438616,463
Salmon321,42412,18553,154386,764
Sturgeon80,33312,185092,518
Big
Game
fish19,5087,173026,681
Total
a920,104864,264193,0261,977,395
a
Sum
of
individual
values
may
not
add
up
to
totals
due
to
the
rounding
error.

Source:
U.
S.
EPA
analysis.

Tables
F5­
11
and
F5­
12
provide
total
annual
welfare
estimates
for
baseline
losses
and
reductions
in
baseline
losses,
respectively.
that
fish
must
grow
to
a
certain
size
before
they
will
be
caught
by
recreational
anglers.
discount
factors
separately
for
impingement
and
entrainment
of
each
species
(
see
Section
X.
B.
6
of
the
316(
b)
NODA).
ate
discounted
total
benefits
for
Northern
California,
EPA
calculated
weighted
averages
of
these
discount
factors,
and
applied
them
to
estimated
willingness
to
pay
values.
for
both
a
3
percent
discount
rate
and
a
7
percent
discount
rate.

Table
F5­
11
presents
annual
losses
to
recreational
anglers
from
baseline
I&
E
effects
in
Northern
California.
recreational
losses
(
2002$)
to
Northern
California
anglers,
before
discounting,
from
I&
E
of
flounders,
small
game
fish,
bottom
fish
and
other
fish
are
$
1,432,645
per
year.
scounted
baseline
losses
are
$
1,224,284,
discounted
using
a
three
percent
discount
rate,
and
$
1,050,991,
discounted
using
a
seven
percent
discount
rate.

Table
F5­
12
presents
the
annual
welfare
gain
under
the
preferred
policy
option
for
each
facility
in
the
region.
Total
undiscounted
recreational
losses
are
reduced
by
$
790,560
with
the
preferred
technology.
total
losses
are
reduced
by
$
663,965.
cent
discount
rate,
total
losses
are
reduced
by
$
596,551.
These
values
were
discounted,
to
reflect
the
fact
EPA
calculated
To
estim
Discount
factors
were
calculated
Total
Total
di
Using
a
three
percent
discount
rate,
Using
a
seven
per
§
316(
b)
Regional
Case
Studies,
Part
F:
Northern
CaliforniaChapter
F5:
Recreational
Fishing
Benefits
NODA
Version
­
March
12,
2003F5­
17
Table
F5­
11:
Total
Estimated
Annual
Baseline
Losses
from
I&
E
for
Northern
California
(
2002$)

Species
Total
Losses
before
Discounting
Total
Losses
with
3%
Discounting
Total
Losses
with
7%
Discounting
Flounders$
56,634$
45,307$
35,679
Small
game
fish$
728,909$
634,151$
532,104
Bottom
fish$
77,312$
71,900$
67,261
Other
fish$
9,276$
7,699$
6,772
No
target$
560,514$
465,227$
409,175
Total
recreational
use
losses$
1,432,645$
1,224,284$
1,050,991
Source:
U.
S.
EPA
analysis.

Table
F5­
12:
Total
Estimated
Annual
Reductions
in
Baseline
Losses
under
the
Preferred
Policy
Option
in
Northern
California
(
2002$)

Species
Total
Losses
Before
Discounting
Total
Losses
with
3%
Discounting
Total
Losses
with
7%
Discounting
Flounders$
2,702$
2,189$
1,729
Small
game
fish$
206,584$
183,860$
157,004
Bottom
fish$
59,041$
54,908$
51,366
Other
fish$
7,376$
5,975$
5,458
No
target$
514,857$
471,034$
380,994
Total
recreational
use
losses$
790,560$
663,965$
596,551
Source:
U.
S.
EPA
analysis.

F5­
4
LIMITATIONS
AND
UNCERTAINTY
F5­
4.1
Considering
Only
Recreational
Values
This
study
understates
the
total
benefits
of
improvements
in
fishing
site
quality
because
estimates
are
limited
to
recreational
use
benefits.
Many
other
forms
of
benefits,
such
as
habitat
values
for
a
variety
of
species
(
in
addition
to
recreational
fish),
non­
use
values,
etc.,
are
also
likely
to
be
important.

F5­
4.2
Charter
Anglers
The
Agency
used
values
for
private/
rental
boat
anglers
to
estimate
values
for
charter
anglers.
It
is
not
clear
whether
this
will
result
in
an
overestimate
or
underestimate
of
per
trip
values
for
charter
boat
anglers.
§
316(
b)
Regional
Case
Studies,
Part
F:
Northern
California
Chapter
F5:
Recreational
Fishing
Benefits
F5­
4.3
Species
and
Mode
Substitution
EPA s
estimated
RUM
model
does
not
allow
for
anglers
to
substitute
between
modes
or
species.
The
analysis
therefore
assumes
that
each
angler
has
chosen
a
mode/
species
combination
followed
by
a
site
based
on
the
catch
rates
for
that
site
and
species.
One
disadvantage
of
the
specified
model
is
that
the
model
looks
at
site
choice
without
regard
to
mode
or
species.
Once
an
angler
chooses
a
target
species
and
mode,
no
substitution
is
allowed
across
species
or
mode
(
i.
e.,
the
value
of
catching,
or
potentially
catching,
a
different
species
or
fishing
using
a
different
mode
is
not
included
in
the
calculation).
Therefore,
improvements
in
fishing
circumstances
related
to
other
species
or
modes
will
have
no
effect
on
anglers 
choices,
and
thus
will
not
be
accounted
for
in
the
welfare
estimates.

NODA
Version
­
March
12,
2003
F5­
18
§
316(
b)
Regional
Case
Studies,
Part
F:
Northern
California
Chapter
F5:
Recreational
Fishing
Benefits
GLOSSARY
Cooling
Water
Intake
Structures
(
CWISs)
:
the
total
physical
structure
and
any
associated
constructed
waterways
used
to
withdraw
water
from
waters
of
the
U.
S.
The
cooling
water
intake
structure
extends
from
the
point
at
which
water
is
withdrawn
from
the
surface
water
source
to
the
first
intake
pump
or
series
of
pumps.

Independence
of
Irrelevant
Alternatives
(
IIA):
IIA
means
that
the
relative
odds
of
one
alternative
being
chosen
over
a
second
should
be
independent
of
the
presence
or
absence
of
unchosen
third
alternatives.

Impingement
and
Entrainment
(
I&
E)
:
impingement
is
the
entrapment
of
aquatic
organisms
on
the
outer
part
of
an
intake
structure
or
against
a
screening
device
during
periods
of
intake
water
withdrawal;
entrainment
is
the
incorporation
of
fish,
eggs,
larvae,
and
other
plankton
with
intake
water
flow
entering
and
passing
through
a
cooling
water
intake
structure
and
into
a
cooling
water
system.

Marine
Recreational
Fishery
Statistics
Survey
(
MRFSS)
:
a
long­
term
monitoring
program
that
provides
estimates
of
effort,
participation,
and
finfish
catch
by
recreational
anglers.
The
MRFSS
survey
consists
of
two
independent,
but
complementary,
surveys:
a
random
digit­
dial
telephone
survey
of
households
and
an
intercept
survey
of
anglers
at
fishing
access
sites.
Sampling
is
stratified
by
state,
fishing
mode
(
shore,
private/
rental
boat,
party/
charter
boat),
and
wave,
and
allocated
according
to
fishing
pressure.
Fishing
sites
are
randomly
selected
from
an
updated
list
of
access
sites.

National
Marine
Fisheries
Service
(
NMFS)
:
a
division
of
the
National
Oceanic
and
Atmospheric
Administration
(
NOAA),
NMFS
is
the
primary
fisheries
service
in
the
U.
S.,
responsible
for
fisheries
management
and
marine
ecosystem
health.

Random
Utility
Model
(
RUM):
a
model
of
consumer
behavior.
The
model
contains
observable
determinants
of
consumer
behavior
and
a
random
element.

NODA
Version
­
March
12,
2003
F5­
19
§
316(
b)
Regional
Case
Studies,
Part
F:
Northern
California
Chapter
F5:
Recreational
Fishing
Benefits
ACRONYMS
CWISs:
Cooling
Water
Intake
Structures
I&
E:
impingement
and
entrainment
MRFSS:
Marine
Recreational
Fishery
Statistics
Survey
NMFS:
National
Marine
Fisheries
Service
RUM:
random
utility
model
NODA
Version
­
March
12,
2003
F5­
20
§
316(
b)
Regional
Case
Studies,
Part
F:
Northern
California
Chapter
F5:
Recreational
Fishing
Benefits
REFERENCES
ArcView
3.2a.
Environmental
Systems
Research
Institute,
Inc,
Redlands,
CA,
2000.

Greene,
William
H.
1995.
Limdep
Version
7
User s
Manual.
Plainview,
NY:
Econometric
Software,
Inc.

Haab,
Timothy
C.;
John
C.
Whitehead;
and
Ted
McConnell.
2000.
The
Economic
Value
of
Marine
Recreational
Fishing
in
the
Southeast
United
States:
1997
Southeast
Economic
Data
Analysis.
Final
Report,
July.

Hicks,
Rob,
Scott
Steinback,
Amy
Gautam,
and
Eric
Thunberg.
1999.
Volume
II:
The
Economic
Value
of
New
England
and
Mid­
Atlantic
Sportfishing
in
1994.
U.
S.
Department
of
Commerce,
National
Oceanic
and
Atmospheric
Administration,
National
Marine
Fisheries
Service.
NOAA
Technical
Memorandum
NMFS­
F/
SPO­
38.

Lupi,
Frank,
and
John
P.
Hoehn.
1998.
 
The
Effect
of
Trip
Lengths
on
Travel
Cost
Parameters
in
Recreation
Demand, 
in
Western
Regional
Research
Publication:
W­
133,
Benefits
and
Costs
of
Resource
Policies
Affecting
Public
and
Private
Land.
Eleventh
Interim
Report,
June,
1998.

McConnell,
K.
and
I.
Strand.
1994.
The
Economic
Value
of
Mid
and
South
Atlantic
Sportfishing:
Volume
2.
Cooperative
Agreement
#
CR­
811043­
01­
0
between
the
University
of
Maryland
at
College
Park,
the
Environmental
Protection
Agency,
the
National
Marine
Fisheries
Service,
and
the
National
Oceanic
and
Atmospheric
Administration.

McFadden,
D.
1981.
 
Econometric
Models
of
Probabilistic
Choice. 
In
C.
F.
Manski
and
D.
L.
McFadden,
Eds.,
Structural
Analysis
of
Discrete
Data.
Cambridge,
MA;
MIT
Press.

National
Marine
Fisheries
Service,
Office
of
Science
and
Technology,
Fisheries
Statistics
and
Economics
Division.
1999.
Marine
Recreational
Fisheries
Statistics:
Data
User s
Manual.
http://
www.
st.
nmfs.
gov/
st1/
recreational/
research/
procedures.
html,
July
12.

National
Marine
Fisheries
Service,
Office
of
Science
and
Technology,
Fisheries
and
Economics
Division.
2001a.
Marine
Recreational
Fisheries
Statistics
Survey
(
MRFSS),
Snapshot
Query.
http://
www.
st.
nmfs.
gov/
recreational/
queries/
catch/
snapshot.
html.
Accessed
in
November
2002.

National
Marine
Fisheries
Service,
Office
of
Science
and
Technology,
Fisheries
Statistics
and
Economics
Division.
2001b.
Marine
Recreational
Fisheries
Statistics
(
1996­
2000).
http://
www.
st.
nmfs.
gov/
st1/
recreational/
research
.
Accessed
in
November
2002.

Parsons,
G.
and
M.
J.
Kealy.
1992.
 
Randomly
Drawn
Opportunity
Sets
in
a
Random
Utility
Model
of
Land
Recreation. 
Land
Economics,
68(
1):
pp
93­
106.

NODA
Version
­
March
12,
2003
F5­
21
