-
Title
-
Eng
Fishery Management Past and Present: Updating the Management of Impacts on ESA-listed Fish Species Using Genetic Stock Identification Tools In-Season to Validate Pre-Season Fishery Model Predictions
-
Date
-
2008
-
Creator
-
Eng
Iverson, Christina A
-
Subject
-
Eng
Environmental Studies
-
extracted text
-
Fishery Management Past and Present: Updating the
Management of Impacts on ESA-Listed Fish Species Using
Genetic Stock Identification Tools In-Season to Validate PreSeason Fishery Model Predictions
by
Christina A. Iverson
A Thesis
Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
December 2008
© 2008 by Christina Iverson. All rights reserved.
This Thesis for the Master of Environmental Studies Degree
by
Christina A. Iverson
has been approved for
The Evergreen State College
by
______________________
Amy Cook, PhD
Member of the Faculty
______________________________
Tom Good, PhD
NOAA-NMFS
________________________
Maria Bastaki, PhD
Member of the Faculty
_____________________________
Date
ABSTRACT
Fishery Management Past and Present: Updating the Management of
Impacts on ESA-Listed Fish Species Using Genetic Stock Identification Tools
In-Season to Validate Pre-Season Fishery Model Predictions
By Christina Iverson
The Washington Department of Fish and Wildlife, pursuant to North of Falcon
agreements made under the Pacific Salmon Treaty, monitors annual impacts on
ESA-listed Puget Sound Chinook salmon populations during fisheries held in
Washington State waters. A Fishery Regulation Assessment Model, or FRAM, is
used by fishery managers to predict and assess harvest-related impacts on ESAlisted Puget Sound Chinook salmon stocks. Yet, beginning in 1998 genetic
analysis was used to estimate stock-specific fishery impacts independent of the
standard management regime, FRAM. In-season genetic samples from Puget
Sound Chinook salmon captured as bycatch were obtained from the sockeye
directed purse seine fishery in Marine Areas 7 and 7A, the Chinook directed
recreational fishery in Marine Area 7 and the Chinook directed gill net fishery in
Marine Areas 7B and 7C. The processing of these samples using genetic stock
identification (GSI) techniques allowed fishery managers to report on the actual
stocks present and impacted in those fisheries. Since 1998 genetic samples were
obtained in 2006 from Marine Areas 7 and 7A purse seine and recreational
fisheries and from these fisheries again in 2007, with the addition of gill net
samples from Marine Areas 7B and 7C. These in-season samples were processed
using GSI and compared to pre-season FRAM stock impact predictions. Results
from the 2007 Marine Area 7 and 7A net fishery demonstrate how FRAM preseason predictions of stock impacts can be very different from how this fishery
functions in-season. FRAM predicted over 61% of that fishery‟s Chinook salmon
mortality would be of Puget Sound origin. GSI data indicated a 4% total
contribution from Puget Sound Chinook salmon. Additionally, fisheries in
Marine Areas 7 and 7A have historically been managed together. These two
fishing areas are geographically isolated from each other by the San Juan Islands.
The frequency of Chinook salmon observed during commercial fisheries in these
two distinct geographical locations from 1997 through 2007 were found to differ
significantly through statistical analysis, which may suggest that a different
management strategy may be needed to more thoroughly monitor the needs of
ESA-listed stocks in these two areas.
TABLE OF CONTENTS
List of Figures …... ………………………………………………… VI
List of Tables. ……………………………………………………… VII
Acknowledgments …………………………………………………....VIII
1
Introduction .............................................................................................. 9
1.1
The Endangered Species Act ………………………………..... 12
1.2
The History of Puget Sound Chinook Salmon ESA Listings.. 13
1.3
The National Oceanic and Atmospheric Administration and
National Marine Fisheries Service.............................................18
1.4
The Magnuson-Stevens Act of 1976 .........................................19
1.5
The History and Drafting of the Pacific Salmon Treaty........ 21
1.6
The Washington Department of Fish and Wildlife..................28
1.7
The Department of Fisheries and Oceans, Canada ……….…31
2
Fishery Management Tools
...................................................... 32
2.1
Population Assessment Methods................................................33
2.2
Population Forecasting ……………………………………..….35
2.3
Predicting The Northern Diversion Rate …………................. 35
2.4
Setting Yearly Harvest Exploitation Rate, Rules and
Regulations ……………………………………………….…… 37
2.5
Fishery Regulation Assessment Model -FRAM .......................41
2.6
Genetic Stock Identification ….................................................. 45
2.7
Marine Protected Areas ……………....................................... 47
2.8
Critical Habitat Designation ……………………………......... 48
3
Methods ……………………………………........................................... 51
3.1
Study Site .................................................................................... 51
3.2
Data Analysis .............................................................................. 52
4
Results ..…………………………………………………....................... 54
4.1
Comparison of Chinook Bycatch Observed in Marine Areas 7
and 7A …………………………………………………………. 54
4.2
Genetic Stock Identification Data for 1998, 2006 and 2007 ....56
4.3
Comparison of FRAM Output and GSI Data for 2007……... 57
4.4
Stock Composition Trends Over Time in the Marine Area
7B and 7C Bellingham Bay Gill Net Fishery ……………....... 59
4.5
Comparison of Effects by Gear Type and Species Targeted.. 61
5
Discussion ……………………………………………………………....64
6
Recommendations and Suggested Future Research ............................68
References........................................................................................................... 72
Appendix A: Acronyms and Definitions …………………………….………. 91
v
FIGURES
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
Figure 11
Figure 12
Figures 13
Figure 14
Map of NOAA’s Puget Sound Chinook Salmon ESUs ……....17
Convention Waters Fishing Area for 1937 Convention …..…22
Percentages of Commercial Salmon Catches from Canadian
waters versus the Diversion Rate into United States
Waters...........................................................................................23
Map of Salmon Migration Routes Originating in Washington,
Oregon and British Columbia by Species …………………….24
DFO’s Forecasted Johnstone Strait Diversion of 2007 Fraser
Sockeye First Estimate, Based on May & June Average
Temperature …………………………………………………....36
Maps of NOAA Fisheries Northwest Region Critical Habitat
Designations for West Coast Salmon and Steelhead in
Washington - August 2005......................................................... 49
Map of Study Site WDFW Marine Areas 7, 7A, 7B
and 7C …………………………………………………..………51
Assignment of stock of origin of Chinook salmon bycatch
samples obtained during Sockeye and Pink purse seine
fisheries in Marine Areas 7 and 7A for 1998, 2006
and 2007 ……..………………………………………..………..57
2007 FRAM predicted Puget Sound Chinook mortality versus
DNA analysis confirmed Puget Sound mortality ……....…….58
1998 Puget Sound Chinook salmon genetically sampled from
Purse Seine, Gill Net and Recreational Fisheries in Marine
Areas 7, 7A, 7B and 7C ……………………..……………….....62
2007 FRAM predictions for recreational Puget Sound Fall
Chinook mortality in the Marine Area 7 fisheries versus inseason GSI data ………………………………………………...63
2007 FRAM predictions for recreational Puget Sound Spring
and Summer Chinook mortality in Marine Area 7 fisheries
versus in-season GSI data …………………………...……..….63
Data Distribution Histograms for Raw Data of Marine Area 7
Observations from 1997-2007………….……………………....90
Data Distribution Histograms for Raw Data of Marine Area
7A Observations 1997-2007....……………….………………....90
vi
TABLES
Table 1
Table 2
Table 3
Table 4
Table 5
Table 6
Table 7
Table 8
Table 9
Table 10
Table 11
Table 12
Table 13
Table 14
Number of populations of each Pacific salmon species that are
extinct or still surviving in California, Nevada, Oregon,
Washington, Idaho and Southern British Columbia................13
NOAA’s Endangered Act Species Status of West Coast Salmon
and Steelhead …………………………………….......................16
Chinook Bycatch Observed 1997-2007 as Reported by WDFW
Marine Area ................................................................................55
ONCOR Results from 2007 DNA analysis of Marine Areas
7B/7C Bellingham Bay Chinook Directed Gill Net Fishery
Samples ………………………………………………..…..……60
1998 Purse Seine Bycatch Proportions as Reported by
WDFW…………………………………………………………..76
1998 Gill Net Bycatch Proportions as Reported by
WDFW ………………………………………………………….77
1998 Recreational Catch as Reported by WDFW …………...78
2006 Bycatch Data as Reported by WDFW ……………..….. 79
2007 Bycatch Data as Reported by WDFW……………..……80
2007 Marine Area 7B/7C Gill Net Genetic Data as Reported
by ONCOR …………………………………………………......81
ONCOR Assignments of Individuals from Reporting Groups
for 2007 Gill Net Data in Proportions and Percentages…...…83
Shapiro-Wilk normality test results using R Version 2007......88
t-Test: Paired Two Sample for Means performed with R
Version 2007……………………………………………….....…88
t-Test: Paired Two Sample for Means performed with MS
Excel ………….…………………………………...…………….89
vii
ACKNOWLEDGEMENTS
Thanks to Jeromy Jording for providing the data from the WDFW Observer
database and his overall support. Thanks to Amy Cook for her advice, easy going
attitude, all her guidance, and support. Thanks to Maria Bastaki and Greg Stewart
for help with my statistical analyses. Thanks to Tom Good for all his invaluable
comments and advice over the years. Editing and suggestions from all three of my
advisors greatly improved the quality of my final written work. Thank you to the
entire staff of the WDFW Genetics Unit, including Ken Warheit, Denise
Hawkins, and Cherrill Bowman. A special thanks to Cheryl Dean for showing me
how process genetic samples in the lab and to Scott Blankenship for showing me
how to assign genotypes and for passing on everything he knew about the project.
Thank you to the Puget Sound Salmon Management Observer Program which
helped collect genetic samples and observational data during commercial fishing
operations in the San Juan Islands from 1991 to 2008 throughout Marine Area 7
in its entirety. Thanks to Russel Barsh whom I interviewed about the history of
the San Juan Islands, and to Robert Kehoe of the PSVOA for his time and insight
into this complicated issue. Thanks to all the commercial fishermen that
participated in my research by sharing their local ecological knowledge about the
San Juan Islands, the salmonid stocks of the Pacific Northwest, and their feedback
on fishery management strategies they have experienced over the years.
viii
1
Introduction
Puget Sound Chinook salmon were listed under the Endangered Species Act
(ESA) as a threatened species on March 24, 1999. Their threatened status was
reaffirmed on June 28, 2005. The evolutionarily significant units (ESU) included
in this listing encompass all naturally spawned populations of Chinook salmon
from rivers and streams flowing into Puget Sound including the Straits of Juan De
Fuca from the Elwha River, eastward, including rivers and streams flowing into
Hood Canal, South Sound, North Sound and the Strait of Georgia in Washington,
as well as twenty-six artificial propagation programs.
Washington State Department of Fish and Wildlife (WDFW) fishery
managers currently use a tool called a fishery regulatory assessment model
(FRAM) to predict a yearly cap on incidental mortalities of ESA-listed salmon
stocks during fisheries targeting other species. Specifically, in this paper, FRAM
is used to assess the mortality of ESA-listed Puget Sound Chinook salmon stocks
encountered as bycatch during fall commercial sockeye and pink fisheries in
WDFW Marine Areas 7 and 7A. In 2007 FRAM estimated over 61% mortality
on Puget Sound Chinook salmon stocks (Blankenship 2007). However, DNA
analysis of fin clip samples of Chinook salmon obtained from vessels during fall
sockeye and pink fisheries revealed a 4% confirmed Puget Sound Chinook
presence in the fishery. Therefore, I plan to explore in this paper how the sole use
of FRAM, which WDFW fishery managers currently use, to estimate Puget
9
Sound Chinook salmon bycatch mortality in Areas 7 & 7A, is in need of some
examination.
Additionally, WDFW Marine Area 7 is located in very near proximity to the San
Juan Islands Salmon Preserve (Figure 7) and several Marine Protected Areas.
Marine Area 7 is also located at the southern most tip of the San Juan Islands. As
adult salmonids of Puget Sound origin return to their natal streams to spawn this
area is the last possible open water location for their interception by the United
States commercial fishing fleet before reaching fresh water. WDFW Observer
data collected during fisheries from 1991 through 2007 suggests that the
frequency of Chinook salmon observed in Marine Area 7 during commercial
fisheries is lower than in 7A. I would like to argue here in this paper that
managing Area 7 & 7A together may not be the best strategy, and demonstrate
why not.
Incorporating genetic information into the management of ESA-listed
Chinook salmon is highly necessary to offer a more accurate picture of the status
of these threatened stocks (OSU 2008). In January of 2006 the Salmon Spawning
and Recovery Alliance, Washington Trout, the Native Fish Society and the ClarkSkamania Fly fishers sent a letter of notification to the National Oceanic and
Atmospheric Administration (NOAA) asking them to “reinitiate ESAconsultation on the Puget Sound Comprehensive Chinook Management Plan:
Harvest Management Component, a Resource Management Plan, or RMP, that
was co-developed by the state and tribes for fisheries affecting Puget Sound
Chinook salmon (Beardslee 2006). They believed that incidental mortality rates
10
on Chinook salmon from listed and highly threatened ESU‟s were still too high.
The data collected and modeled for mortality on these listed, and considered weak
stocks thus far have had great variation. Estimated mortality of listed Chinook
salmon stocks either directly, or indirectly has been between 22%-76% annually
(Beardslee 2006). If this estimated range of mortality, which is provided to
NOAA, does fall within the actual range of mortality experienced by these weak
stocks, then certainly these estimates should be cause for alarm. If FRAM has
been over predicting mortality on listed Puget Sound Chinook salmon ESUs, then
unnecessary lawsuits such as the one mentioned above will continue to happen.
Additionally, lawsuits such as these might not be the best way to approach the
problem. Conversely, if the mortality estimates provided by the use of FRAM are
underestimating the actual impacts on weak and federally listed stocks then it will
be very difficult to monitor, as mandated by the ESA, exactly what is happening
to stocks listed for protection. Making adjustments to the current FRAM could
result in producing more accurate estimates for mortality of ESA-listed Puget
Sound Chinook salmon ESUs, and fishery managers would be able to provide
those working to recover weak stocks actual information on how well these stocks
are rebounding. This paper will use the preferred combination of direct and
indirect methods of study, as suggested in Iverson (1996) to examine fishery
modeling limitations through a comparison to actual GSI sampling data.
The goal of this paper is to 1) make a compelling argument in favor of a
revision of the current standard management regime of the WDFW, the sole use
of FRAM modeling pre-season to estimate mortality of ESA-listed Puget Sound
11
Chinook salmon stocks encountered during commercial fisheries in Marine Areas
7 & 7A, and 2) demonstrate how different the Chinook salmon bycatch
frequencies have been consistently over the last ten years between Area 7 & 7A,
making the case that these two areas should be managed separately.
1.1
The Endangered Species Act
"Nothing is more priceless and more worthy of preservation than the rich
array of animal life with which our country has been blessed."
— President Nixon, upon signing the Endangered
Species Act
The Endangered Species Act of 1973 (ESA) was signed on December 28,
1973, and provides for the conservation of species which are endangered, or
threatened throughout all or a significant portion of their range, and the
conservation of the ecosystems on which they depend. The ESA replaced the
Endangered Species Conservation Act of 1969; it has been amended several
times. A species is defined as “endangered” if it is in danger of extinction
throughout all or a significant portion of its range. A species is defined as
“threatened” if it is likely to become an endangered species within the foreseeable
future.
There are approximately 1,880 species listed under the ESA. Of these
species, approximately 1,310 are found in part or entirely in the United States and
its waters. NOAA's National Marine Fisheries Service (NMFS) and the U.S. Fish
12
and Wildlife Service (USFWS) share responsibility for implementing the ESA.
The USFWS manages land and freshwater species, and NMFS manages marine
and “anadromous” species. Currently NMFS has jurisdiction over approximately
60 listed species (NOAA 2008).
1.2
The History of Puget Sound Chinook Salmon ESA Listings
Scientists estimate nearly 1,383 genetically-isolated Pacific salmon
populations once spawned from California to southern British Columbia.
However, due to dam building and other alterations of lakes and rivers, 406 or
29% of the salmon populations have become extinct in the last 240 years (Osborn
2008).
Table 1
Common Name
Scientific Name
Extinct Surviving
Steelhead
Oncorhynchus mykiss
131
436
Chinook
Oncorhynchus tshawytscha
159
237
Sockeye
Oncorhynchus nerka
34
38
Coho
Oncorhynchus kisutch
50
135
Chum
Oncorhynchus keta
23
89
Pink
Oncorhynchus gorbuscha
9
42
Table 1. Number of populations of each Pacific salmon species that are extinct or still
surviving in California, Nevada, Oregon, Washington, Idaho and southern British Columbia.
(Osborn 2008).
13
As previously mentioned, Puget Sound Chinook salmon were listed as a
threatened species on March 24, 1999; their threatened status was reaffirmed on
June 28, 2005. An evolutionarily significant unit, or ESU, of Pacific salmon is
considered to be a "distinct population segment" and thus a "species" under the
Endangered Species Act. The threatened Puget Sound ESU includes all naturally
spawned populations of Chinook salmon from rivers and streams flowing into
Puget Sound including the Straits of Juan De Fuca from the Elwha River,
eastward, including rivers and streams flowing into Hood Canal, South Sound,
North Sound and the Strait of Georgia in Washington, as well as twenty-six
artificial propagation programs which encompass: the Kendal Creek Hatchery,
Marblemount Hatchery (fall, spring yearlings, spring sub-yearlings, and summer
run), Harvey Creek Hatchery, Whitehorse Springs Pond, Wallace River Hatchery
(yearlings and sub-yearlings), Tulalip Bay, Issaquah Hatchery, Soos Creek
Hatchery, Icy Creek Hatchery, Keta Creek Hatchery, White River Hatchery,
White Acclimation Pond, Hupp Springs Hatchery, Voights Creek Hatchery, Diru
Creek, Clear Creek, Kalama Creek, George Adams Hatchery, Rick‟s Pond
Hatchery, Hamma Hamma Hatchery, Dungeness/Hurd Creek Hatchery, and the
Elwha Channel Hatchery Chinook hatchery programs (NOAA 2008).
Over the past several decades, wild populations of salmon throughout the West
Coast have declined to dangerously low levels. In 1991 NMFS began a series of
comprehensive status reviews of salmon populations throughout Washington,
Oregon, Idaho, and California. Ten of the seventeen West Coast Chinook salmon
ESUs, including the Puget sound Chinook ESU, have been listed as endangered or
14
threatened under the ESA (Table 2). The locations of watersheds from which the
ESA-listed Chinook salmon ESUs originate are such that 53%, or roughly 30
miles, of riparian habitat immediately inland from the Puget Sound and
surrounding the natal freshwater rivers and streams is owned privately (Figure 1).
Privately owned land is the most difficult to regulate and monitor for habitat and
fish population health and species recovery (NOAA 2008).
15
Table 2
Table 2. http://www.nwr.noaa.gov/ESA-Salmon-Listings/upload/snapshot0208.pdf
16
Figure 1. NOAA WEBSITE- http://www.nwr.noaa.gov/ESA-Salmon-Listings/Salmon
Populations/Maps/upload/chinpug.pdf.
17
1.3
The National Oceanic and Atmosphere Administration and National
Marine Fisheries Service
The National Oceanic and Atmospheric Administration‟s (NOAA)
MISSION STATEMENT: Stewardship of living marine resources through
science-based conservation and management and the promotion of healthy
ecosystems (NOAA 2008).
NOAA's National Marine Fisheries Service (NMFS) is the federal agency that is
responsible for the stewardship of our nations living marine resources and their
habitat. The NMFS is responsible for the management, conservation and
protection of living marine resources within the United States' Exclusive
Economic Zone (EEZ), which is defined as water three to 200 mile offshore.
Using the tools provided through the Magnuson-Stevens Act (See Section 1.4),
NMFS assesses and predicts the status of fish stocks, ensures compliance with
fisheries regulations and works to reduce wasteful fishing practices. Under the
Marine Mammal Protection Act (MMPA), first established in 1972, and the ESA,
the NMFS aims to recover protected marine species without impeding economic
and recreational opportunities. Through the use of regional offices and staff the
NMFS is able to work directly with communities on fishery management issues
(NOAA 2008). The NMFS also plays an advisory role in managing living marine
resources located in coastal areas that are under state jurisdiction. It provides
18
scientific and policy leadership in the international arena, and implements
international conservation and management measures as necessary.
1.4
The Magnuson-Stevens Act of 1976
The Magnuson-Stevens Fishery Conservation and Management Act
(MSA) is the primary law governing marine fisheries management in United
States federal waters. The Act was first enacted in 1976 and amended in 1996
(NOAA 2008). The Magnuson-Stevens Act (MSA) aided in the development of
the domestic fishing industry by phasing out foreign fishing within the EEZ. In
order to manage the fisheries, and promote conservation, the MSA created eight
regional fishery management councils. Under Section 302 of the MagnusonStevens Act (SEC. 302. REGIONAL FISHERY MANAGEMENT COUNCILS
16 U.S.C. 1852 97-453, 101-627,104-297) the following councils apply
specifically to Pacific salmonid populations:
The Pacific Council --The Pacific Fishery Management Council
shall consist of the States of California, Oregon, Washington, and
Idaho and shall have authority over the fisheries in the Pacific
Ocean seaward of such States.
The North Pacific Council --The North Pacific Fishery
Management Council shall consist of the States of Alaska,
Washington, and Oregon and shall have authority over the fisheries
in the Arctic Ocean, Bering Sea, and Pacific Ocean seaward of
Alaska.
19
The Western Pacific Council --The Western Pacific Fishery
Management Council shall consist of the States of Hawaii,
American Samoa, Guam, and the Northern Mariana Islands and
shall have authority over the fisheries in the Pacific Ocean seaward
of such States and of the Commonwealths, territories, and
possessions of the United States in the Pacific Ocean area.
The 1996 amendments to the MSA focused on rebuilding over-fished fisheries,
protecting essential fish habitat and reducing bycatch. Congress added new
habitat conservation provisions to that act in recognition of the importance of fish
habitat to productivity and sustainability of U.S. marine fisheries. The re-named
Magnuson-Stevens Act mandated identification of essential fish habitat (EFH) for
managed species. The act also requires measures to conserve and enhance the
habitat needed by fish to carry out their life cycles. Congress defined EFH as
"those waters and substrate necessary to fish for spawning, breeding, feeding, or
growth to maturity." An additional EFH guideline used to interpret the provided
EFH definition, and which applies specifically to the ESA-listing of Puget Sound
Chinook salmon is:
-necessary means the habitat required to support a sustainable fishery and
the managed -species' contribution to a healthy ecosystem (NOAA-NMFS
2008).
20
1.5
The History of the Drafting of the Pacific Salmon Treaty
The original document that addressed the need to fairly allocate
transboundary salmonid resources between the United States and Canada was the
Fraser River Convention, which was ratified in 1937 (Shepard et al., 2005).
Around the 1960‟s, and toward the end of the Convention period, when
negotiations were well underway for the subsequent 1985 Pacific Salmon Treaty,
a sudden shift in ocean conditions contributed to a marked increase in the average
Johnstone Strait diversion rate (Miller 2002). By the late1970‟s the Canadian
fishing fleet was taking full advantage of this newly discovered phenomenon now
known as the “Northern diversion rate”. This phenomenon is observed through a
change in expected migratory patterns of the returning adult salmonids on their
journey back to natal spawning grounds through the Straits in the San Juan
Islands and around West Vancouver Island (Section 2.3 and Figure 2). Once
discovered, the Canadian purse seine fleet began to target returning salmon
outside the original 1937 Convention Waters in the Georgia Strait, especially
during higher “northern diversion rate” years, in order to increase their catches
and thus their bargaining power at the international treaty table (Brown 2005;
Miller 2002; Shepard et al., 2005).
21
Indicates Northern
Diversion Route
Expected Route
Convention Waters
Figure 2. Convention Waters Fishing Area for 1937 Convention. Source: North American Pacific
Salmon: A Case of Fragile Cooperation. Kathleen A. Miller. 2002.
Between the years of 1953-1976 the diversion rate averaged 16.4 percent. From
1977 through 1985, the diversion rate average increased to 46 percent (Miller
2002). This increase in interception opportunities shift surely strengthened
Canada‟s hand in the negotiations which eventually lead to the 1985 Treaty. The
Canadian fishermen took advantage of unusually high diversion rates in 1978,
1980, 1981, and 1983 which resulted in a substantial increase in their overall
share of the salmon harvest (Figure 3).
22
---- Canadian Commercial Catch as % of Total
Commercial Catch
____ Diversion Rate Through Johnstone Strait
Figure 3. Percentages of Commercial Salmon Catches from Canadian waters versus the Diversion Rate into
United States Waters. Source: North American Pacific Salmon: A Case of Fragile Cooperation. Kathleen A.
Miller. 2002.
Due to the life cycle migration patterns of west coast salmonids (Figure 4)
fishermen from the state of Alaska and the country of Canada, both situated north
of Washington and in colder more nutrient rich waters, were perfectly placed to
intercept great numbers of homeward bound West Coast salmonids during
spawning season (Pearcy 1992; Miller 2002; Quinn 2005; Dominquez 2007).
Thus, Alaska was apprehensive about signing any rights away during the
negotiations of the Pacific Salmon Treaty with Canada.
23
Figure 4. Salmon Migration Routes of Pacific Salmon. Source: Salmon Ecology Key
Slides - Prof. Larry Dominguez. The Evergreen State College 2007.
By 1984 the state of Washington alone had been spending $87 million a
year on salmon management and hatchery production. $800 million had been
invested in fish passage and hatchery production to mitigate the damage to
Columbia River stocks due to the construction of several hydroelectric facilities,
and $750 million in restoration due to the 1980 Northwest Power Act, a
Congressional mandate (Blumm 1994). At this point the state began to believe
24
their hard work was not “paying off” as they had foreseen. They were not seeing
a “return on these investments”, which they predicted would quickly be observed
through increased Columbia River salmonid populations. The 1984 Secretary of
Energy, Donald P. Hodel wrote to the Secretary of State George Shultz:
“Much of this substantial investment…is severely jeopardized by
continued uncontrolled harvest of Columbia River Chinook runs by
Canadian and Alaskan fishermen. It is imperative that if this investment is
to achieve corresponding regional benefits, the United States and Canada
must soon reach accord on an interception treaty. Continued decline of
the Columbia‟s salmon runs may only lead to further regional hardship…”
Eventually a treaty was signed in 1985 that represented a new era of cooperation
between the two countries. This new Pacific Salmon Treaty (PST) would forever
require an annual re-evaluation of stocks and regulations for each country. In July
of 1999 the two countries signed a revised Pacific Salmon Agreement which was
developed through cooperation by the U.S. and Canadian federal governments,
tribes, state governments, and sport and commercial fishing groups.
It is simply not possible to successfully monitor and recover declining
transboundary populations of marine organisms if both sides of the international
border have unique management strategies; therefore a treaty, such as the PST,
was the best possible solution for both sides to such a complicated issue. Russel
Barsh, an Ecologist with the Center for the Study of Coast Salish Environments,
has been studying the life history of the salmonid populations which originate
near the Canadian-United States border for many years. More recently his
25
research has included an examination of the habitat use and behaviors of juvenile
salmonid outmigrants throughout the San Juan Islands. Results from his studies
thus far show that juvenile outmigrants from both Canadian origin stocks and
United States origin stocks are completely intermixed throughout the islands. “As
far as juveniles are concerned, they all congregate and feed in the islands in mixed
groups from both sides of the international border before heading out to the open
ocean. It is therefore presumably safe to assume that adults returning to natal
spawning grounds would also stop in the islands, commingling and feeding before
continuing inland and entering the senescence life history phase. ” (Barsh 2008).
This research highlights the largely still unknown life history characteristics of
returning Pacific salmonids, as fishermen and fishery managers have speculated
for years about what adult salmon did upon reaching this ecosystem. The widely
accepted belief was that they quickly and independently migrated through the
islands from open water in segregated runs, as data collected through the use of
WDFW test fisheries had previously suggested.
Since the signing of the 1985 PST, in order to help fulfill conservation
goals and ensure that each country has the right to reap the benefits created from
its own fisheries enhancement efforts, the PST has been implemented by an eightmember bilateral Pacific Salmon Commission (PSC). The PSC includes
representatives of federal, state and tribal governments from both countries. The
PSC does not regulate salmon fisheries. However, it does provide regulatory
advice and recommendations in a forum which fosters the ability for the two
26
countries to reach a mutual agreement about transboundary resource issues such
as salmonid harvest opportunities.
According to the 1999 Annex to the Pacific Salmon Treaty, the U.S. share
of the Total Allowable Catch (TAC) of sockeye salmon in Marine Areas 7/7A is
16.5%, with the non-treaty share of the U.S. TAC being 32.3%. These U.S. and
non-treaty share proportions will remain in effect through 2010 (WDFW 2008).
However, since a new threat to the health of sockeye populations from the Fraser
River was discovered in 2001, the abilities of both countries to fully harvest their
PST derived sockeye harvest shares has been impacted by conservation concerns
for “late run” sockeye. This stock group has been entering the Fraser River in
August, four to six weeks earlier than they normally do, in September, and
suffering a nearly 95% combined migration and pre-spawning mortality rate
(Cook 2004). These fish have begun to head directly into freshwater instead of
remaining in the Strait of Georgia for several weeks as was customary, and
milling around before heading to spawning grounds. What has been identified by
UBC biologists at this time as the cause of the significantly increased mortality
rate is a rapid increase in kidney parasite infections, which impairs these fishes
ability to regulate the vital physiological adjustments necessary when
transitioning from salt to fresh water (Cook 2004). The higher water temperature
allows the parasite, Parvicapsula, to proliferate at a much faster rate than
previously observed, and result in pre-spawning mortality. Because the “late run”
sockeye timing overlaps with abundant summer sockeye runs present in WDFW
Marine Areas 7 and 7A fisheries, limiting harvest seasons to reduce impacts on
27
these late run sockeye, which now have dramatically reduced populations,
requires additional new regulations on the number of harvestable summer runs of
sockeye.
1.6
The Washington Department of Fish and Wildlife
The Washington Department of Fish and Wildlife (WDFW), previously
known as the Washington Department of Fisheries, is the state agency responsible
under legislative mandate to manage all marine and freshwater species and to
“preserve, protect and perpetuate” fish populations and at the same time to
“enhance and improve recreational and commercial fishing in this state”
(WDFW 2008). The WDFW, known in 1977 as the Washington Department of
Fisheries, initiated a comprehensive and long-range research effort to address the
specific needs of managing Washington‟s naturally-produced salmon runs. This
became known as the Wild Salmon Production Evaluation unit (WSPE). It was
created to measure production, survival, and fisheries contribution of wild origin
salmon stocks. Since its creation, WSPE has continued to measure survival at
three long-term monitoring stations located around the state. The WSPE monitors
wild salmon populations in Puget Sound, the Washington coast and lower
Columbia River. Regional biologists and the Hatchery/Wild Interactions Unit
monitor the freshwater production of wild origin salmon populations at additional
sites statewide (WDFW 2008). WDFW defines a fish stock as:
-A stock is a group of fish of the same species that spawn in the same
location at the same time with little interbreeding with other groups.
28
-Basic unit of assessment for productivity, extinction probability, and
recovery plan.
The goal of the WDFW‟s Wild Salmonid Policy is: to protect, restore, and
enhance the productivity, production, and diversity of wild salmonids and their
ecosystems to sustain ceremonial, subsistence, commercial, and recreational
fisheries, non-consumptive fish benefits, and other related cultural and ecological
values (WDFW 2008).
Under the framework of the Wild Salmon Policy, there are components that must
be monitored for the program to be considered a success. Fish Populations,
Escapement, Genetics, Harvest Management, and Hatcheries must be monitored
and measured against the standards set by the Wild Salmon Policy. Specifically,
for the purposes of this paper the Harvest Management Policy Statement reads:
The fisheries will be managed to meet the spawning escapement policy as well as
genetic conservation and ecological interaction policies. The Harvest
Management performance standards that must be met are as follows:
-Harvest management will be responsive to annual fluctuations in abundance
of salmonids, and will be designed to meet any requirements for sharing of
harvest opportunity.
-The allowable incidental harvest impact on populations shall be addressed in
existing preseason and in-season planning processes...
-Where a population is not meeting its desired spawner abundance level, the
State, in managing the non-treaty harvest, may give priority to non-treaty
fisheries that can minimize their impacts on weak stocks and increase their
29
harvest on healthy stocks by: (1) using gears that can selectively capture and
release stocks with minimal mortality, or (2) avoid impacts by eliminating
encounters with weak populations (proven time/area closures, gear types).
This must be done consistent with meeting treaty and non-treaty allocations
and in accordance with agreed mass marking policies (NOAA-NMFS 2008).
Currently the WDFW also is responsible for drafting Harvest Management Plans
which outline objectives to guide the Washington co-managers in planning annual
harvest regimes, as they affect ESA-listed Puget Sound Chinook salmon. The
current Plan under review by NOAA Fisheries for approval applies to
management years 2004 - 2009. These objectives include total U.S. exploitation
rate ceilings, and spawning escapement goals. This Plan describes the technical
derivation of these objectives, and how these guidelines are applied to annual
harvest planning. The Plan guides the implementation of fisheries in Washington,
and it considers the total harvest impacts of all fisheries, including those in Alaska
and British Columbia, to assure that conservation objectives for ESA-listed Puget
Sound Chinook salmon ESUs are achieved. The accounting of total fisheryrelated mortality includes incidental harvest rates such as mortality rates for fish
encountered as “bycatch” in fisheries which are directed at other salmon species,
and for non-landed Chinook salmon mortality, or mortalities that result from
hook-and-release fisheries. The fundamental intent of the Plan is to enable
harvest of strong, productive stocks of Chinook salmon and other salmon species
and to minimize harvest of “weak” or critically depressed Chinook salmon stocks.
As mentioned, the Puget Sound ESU currently includes many “weak”
30
populations. A “weak” population is defined as an ESU which is not meeting
escapement goals as currently set. Providing adequate conservation of weaker
stocks necessitates foregoing some harvestable surplus of healthy stocks. Some of
the WDFW policies that specifically apply to commercial and recreational harvest
management are as follows:
-The Department will support harvest strategies that promote optimum
long-term sustainable harvest levels.
-The Department will support monitoring programs which gather
biological, discard, and bycatch data from each of the fisheries.
-The Department will take a precautionary approach in the management
of species where the supporting biological information is incomplete
and/or the total fishery-related mortalities are unknown.
-The Department will support consideration of the use of risk-averse
management tools to protect the resources in the face of management
uncertainty.
-The Department will support management measures which conserve,
restore, and enhance the quality of essential fish habitats upon which
Council-managed fisheries resources depend (WDFW 2008).
1.7
The Department of Fisheries and Oceans, Canada
The Department of Fisheries and Oceans, Canada (DFO) is responsible for
managing sockeye, pink, chum, coho and Chinook salmon within their territorial
waters and EEZ. They uphold PST obligations with the United States and the
31
British Columbian tribes through participation in the PSC. They do so through
the use of some of the same tools U.S. fishery managers use, such as test fisheries
(Section 2.1). Commercial fisheries in Canada are also managed under the ITQ
system (Section 2.1). The DFO uses salmon management advisory boards to
oversee operational issues associated with salmon fisheries, such as pre-season
planning and appropriate enforcement. The advisory board is also compelled to
follow conservation guidelines and other policy directives, and the scientific
advice provided by the Pacific Scientific Advice Review Committee, stock
assessment reports, and other policy documents to guide their planning.
DFO uses ongoing fisheries reform initiatives and new commitments such
as their complimentary Wild Salmon Policy and the implementation of markselective fisheries to also attempt to recover weak salmon populations which
originate in Canadian waters. Additionally, new fishery management tools such
as Genetic Stock Identification (GSI) is actively being used in Canada to manage
coho salmon fisheries off the west coast of Vancouver Island (OSU 2008).
2
Fishery Management Tools
The extent of intermingling of stocks of marine fish is often complicated. For
example, in the North Pacific Ocean, where there is no obvious physical barrier to
widespread migration and intermingling of salmon, Oncorhynchus sp., stocks
intermingle over broad oceanic areas (Pearcy 1992; Iverson 1996; Quinn 2005;
Barsh 2008). Rational fisheries management requires knowledge of the extent to
32
which exploited populations comprise a discrete, and self-sustaining stock
(Iverson 1996).
2.1
Population Assessment Methods
Studies which use either indirect or direct sampling techniques, or a
combination of both, are generally performed by fisheries biologists to determine
how large a population unit, or subpopulation is that they are attempting to
manage. One direct method is tagging individual fish to determine the exact
extent of movement of individuals, or groups of fish. Fish are marked at a
specific location and at a particular time with specialized tags, such as coded wire
tags, which bear identifying information and are surgically implanted into the
snouts of young hatchery fish (WDFW 2008). Tagged fish are recaptured at a
later date and the coded wire tag (or other tag) is recovered, identified, and
recorded. This tool is used to suggest the extent of movement of these fishes
(Marshall 1998; Hall 2001; Quinn 2005). Coded wire tag data has been used
since 1986 to gather distribution and migration data about both hatchery and wild
populations under the assumption that hatchery fish behavior is identical to wild
fish behavior. Recent studies have begun to attempt to test the validity of this
assumption (Barnett-Johnson et al., 2007; OSU 2008).
Indirect methods include counts of body parts, body proportions,
physiological attributes, parasite fauna used as natural markings, and genetics
(Iverson 1996; Dominquez 2007). These indirect measurements may also include
spawning surveys to count the number of redds, fish nests constructed, or the
33
counting of carcasses on a spawning ground to calculate the number of
reproductively successful adults of a population.
WDFW also uses a tool called a “test fishery” to predict the size of a
returning run of salmon. (The term “run” is synonymous with the term stock for
this discussion). This requires fishery biologists to operate fishing vessels to
simulate fishing methods employed by fishermen at various locations to predict
the time a salmon stock will “peak”, and to gather the age and sex ratio present in
the stock. These data are gathered by using net gear and counting and sexing all
fish captured each fishing day, per set, for a period of several weeks. During this
process the number of fish caught will increase until it reaches a “peak”, or
maximum number, and this will indicate the majority of the fish for that run have
then passed through that fishing area for the season. This information is used to
forecast what is called “run timing”, helping to predict where the fish will be
located as they continue to migrate from test fishing grounds to natal spawning
grounds. This knowledge can help fishery managers both examine the size and
strength of a run, the potential reproductive health indicated by the sex ratio, and
the average age of sexual maturity for a run. It also allows managers to change
fishing regulations if information from test fisheries being conducted indicates
that the number of fish actually returning differs substantially from pre-season
estimates (WDFW 2008).
34
2.2
Population Forecasting
The annual process of setting Washington State fishing seasons begins
each year with a pre-season forecast of the abundance of various individual fish
stocks. These forecasts are based on estimates of the number of juvenile wild
salmon produced in a river system and counted as outmigrating juveniles, surveys
of adult fish which have returned to spawn, counts of fish returning to hatcheries
to spawn, and samples from fisheries in "terminal" areas, or the waters near the
home streams where fish are returning to spawn. When compiled, these numbers
give WDFW fishery managers an estimate of the size and strength of the
upcoming season‟s fish populations. This pre-season forecast estimate is then
added to a base of information on the historic run-size strength and predicted
fishery impacts for the various fish populations. The primary tool used to develop
this base of information for Chinook salmon has been coded wire tags (WDFW
2008).
2.3
Predicting the Northern Diversion Rate
Occasionally salmon will return in higher abundance to spawning grounds
located in Washington and British Columbia through the Strait of Georgia, rather
than the historically customary route through the Strait of Juan de Fuca. The
cause of this change in migratory behavior is still largely unknown. Many
scientists speculate that it is temperature related (Groot and Quinn 1987; Miller
2002; Folkes 2007). To measure and test this hypothesis, the Department of
Fisheries and Oceans (DFO) in Canada has set up temperature monitoring stations
35
along the coast of the Strait of Georgia to examine the relationship between an
increase in sea surface temperature and an increase in the “northern diversion
rate” of Fraser River sockeye salmon (Folkes 2007). Data collected thus far
seems to indicate that in years where the sea surface temperature is warmer than
average, the sockeye returning to the Fraser River will choose to return through
Johnstone Strait and the Strait of Georgia, rather than through the Strait of Juan de
Fuca (Figure 5).
Figure 5. 2007 Northern diversion rate forecast. The relationship between Kains
Island SST and estimated proportion of the Fraser sockeye run that diverts through Johnstone
Strait (1977-2006). The relationship was fit using a General Additive Model (GAM)
with binomial error and a logic link function. Data labels represent year (i.e. 83- 1983).
(Folkes 2007).
36
This predicted information is generated by DFO and provided to WDFW fishery
managers in the months prior to the pre-season forecasting, in order to help
WDFW predict the size of the salmon runs returning to U.S. waters. This
information is also helpful when fishery managers make pre-season calculations
about potential impacts to ESA-listed bycatch species as a function of the total
days a fishery is open for net fishing in these areas.
2.4
Setting Yearly Harvest Exploitation Rate, Rules and Regulations
The WDFW participates each year in setting the non-treaty commercial
and recreational fishing regulations. Harvest rules are built on a foundation of
historical scientific surveys, computer model predictions and joint deliberations
involving representatives of treaty tribes, several states, the federal government
and the public. Using data collected annually from thousands of stream and
harvest surveys and inputting these data into the computer modeling program
FRAM, the WDFW works each year with tribal co-managers, other governments
and the public to set fishing seasons. The goal is to protect weak wild fish
populations while providing harvest opportunities on healthy wild and hatchery
origin stocks. Setting successful harvest regulations requires fishery managers to
pay acute attention to overarching conservation goals, which were designed to
ensure that enough fish survive annual harvest in order to spawn and perpetuate
the long-term viability of each run. These goals are set based on what is believed
to be the best available scientific information available on the number of fish a
37
given stream is capable of supporting, and the number of "recruits," or new fish
that can be produced by each pair of spawning adults.
Admittedly, managing salmon fisheries in the state of Washington is one
of the most complex natural resource challenges in the country. This is due to the
life history characteristics of Pacific salmonids, behavior patterns and
geographical factors. As previously mentioned Pacific salmon are highly
migratory, passing from freshwater streams and major rivers, out to the Puget
Sound, up along the coast of British Columbia and as far north as Alaska before
returning to natal streams to spawn. This means that salmonid survival rates
depend biologically on habitat conditions over thousands of miles of fresh and
saltwater. It also means that politics which dictate harvest rates in Alaska and
Canada can affect the number of salmon that return to Washington waters
(Shepard et al., 2007). Taking into account the fact that Washington State fishing
activities involve several species that migrate over thousands of miles and across
international boundaries, the WDFW participates in three separate harvest
management panels:
-The Pacific Salmon Commission (PSC), which consists of
representatives of Alaska, Washington, Oregon and Canada, the treaty
Indian tribes of Washington and the Columbia River and the federal
government.
-The Pacific Fisheries Management Council (PFMC) which includes the
principal fisheries officials from the states of California, Oregon,
Washington and Alaska, the regional director of the National Marine
38
Fisheries Service and eight private citizens appointed by the U.S.
Secretary of Commerce from lists submitted by each state governor,
jointly manages coastal fisheries, including salmon and ground fish from
three to 200 miles off shore. The season setting process occurs in a series
of public meetings.
-The North-of-Falcon (NOF) public planning forum in which federal,
state and tribal fish managers meet in tandem with PFMC deliberations on
ocean seasons, to set recreational and commercial salmon fisheries for
waters within three miles of the coast of Washington and northern Oregon,
as well as Puget Sound. The North of Falcon season setting process occurs
in a series of public meetings each spring, attended by federal, state, tribal
and commercial fishing industry representatives and concerned citizens.
Fishing season options are developed each year in the late winter and early spring.
After fishing seasons are set each April, the WDFW and tribes continue to
monitor in-season activity and stock impacts as they are occurring “on-the-water”.
This is performed using the sampling techniques (discussed in Section 2.1 and
2.6) such as Test Fisheries and genetic stock identification tools. Fishery
managers must make frequent in-season re-assessments about which regulations
should be adjusted, and how, according to the “real-time” data collected and
analyzed during fishery operations.
The objective for annual, pre-season fishery planning is to develop a
fishing regime that will assure that exploitation rates that do not exceed the
objectives established for each WDFW management unit. As the Puget Sound
39
ESU has many stocks listed for ESA protection, annual target rates that emerge
from WDFW pre-season planning aim to fall well below their respective ceiling
rates. While these ESA-listed stocks are rebuilding, annual harvest objectives
will intentionally be conservative, even for relatively strong and productive
populations (WDFW 2008). These harvest thresholds are intentionally set above
the level at which a population may become demographically unstable, or subject
to further loss of genetic integrity. If abundance (i.e., escapement) is forecast to
fall to or below this threshold, harvest impacts will be further constrained, by
what fishery managers call Critical Exploitation Rate Ceilings, so that escapement
will exceed the low abundance threshold.
Quantification of recent stock productivity (i.e., recruitment and survival)
is subject to uncertainty and bias through the sampling methods employed and
discussed in Section 2.1. The implementation of harvest regimes is also subject to
management error. WDFW fishery managers specifically consider these sources
of uncertainty and error, and must make in-season adjustments to manage the
consequent risk that harvest rates will exceed appropriate levels. The productivity
of each stock is re-assessed annually, and harvest objectives are modified as
necessary, to reflect current population status (WDFW 2008).
Washington State and Canada currently participate in the Total Allowable
Catch (TAC) quota system for limiting take during fisheries. The TAC system
applies to all fisheries held within each countries‟ EEZ. Those opposed to this
quota system state that if the fishery is simply closed once the TAC is reached,
this causes fishermen to “race against each other” to harvest a larger share of the
40
TAC than their competitors in the fishery (Runolfsson 1997). Such behavior,
which has fishermen fishing to a maximum sustainable yield (MSY) level each
year in a fishery over time has been documented to drive healthy populations to
extinction over relatively short periods of time (Cook 2006; Iverson 1996). Thus,
continuous adjustment of a fisheries‟ TAC is necessary because of the inherent
biological variability in fisheries, and their ecological interrelationships. The
ability of fishery managers to set TAC at a sustainable level should continue to
improve over time as they employ research methods to understand how ecosystem
populations and interactions vary annually, and through the use of diligent and
long-term monitoring efforts.
2.5
Fishery Regulation Assessment Model - FRAM
A Fishery Regulation Assessment Model (FRAM) developed by WDFW
fishery managers is currently used by the Pacific Fishery Management Council
(PFMC) to annually estimate impacts of proposed ocean and terminal fisheries on
salmon stocks. This tool has been used in different variations since the 1970‟s
(MEW 2006). FRAM is a single-season modeling tool. The Chinook version
evaluates impacts on most stock groups originating from the north-central Oregon
coast, Columbia River, Puget Sound, and Southern British Columbia. The FRAM
produces a variety of output reports that are used to examine the impacts of
proposed fisheries for compliance with management objectives, allocation
arrangements, ESA compliance, and domestic and international legal obligations.
Only recently has FRAM begun to be used for assessing compliance with
41
Chinook agreements in international fisheries management forums. The FRAM is
a discrete, time-step, age-structured, deterministic computer model used preseason to predict the impacts from a variety of proposed fishery regulation
mechanisms for a single management year. It produces point estimates of fishery
impacts by stock for specific time periods and age classes. The FRAM performs
bookkeeping functions to track the progress of individual stock groups as the
fisheries in each time step exploit them (MEW 2006).
Currently, 33 stock groups are represented in the Chinook FRAM. Each
of these groups have both marked and unmarked components to permit
assessment of mark-selective fishery regulations. For most wild stocks and
hatchery stocks without marking or tagging programs, the cohort size of the
marked component is zero; therefore, the current version of FRAM has a virtual
total of 66 stock groups for Chinook. Stocks or stock-aggregates represented in
the FRAM were chosen based on the level of management interest, their
contribution rate to PFMC fisheries, and the availability of representative coded
wire tag (CWT) recoveries in the historical CWT database (MEW 2006). The
FRAM includes pre-terminal and terminal fisheries in southeast Alaska, Canada,
Puget Sound, and off the coasts of Washington, Oregon, and California. There
are 73 fisheries in Chinook FRAM. The intent is to encompass all fishery impacts
to modeled Chinook salmon stocks in order to account for all fishing-related
impacts and thereby improve model accuracy (MEW 2006). Terminal fisheries in
Chinook FRAM are aggregations of gears and management areas.
42
Major assumptions and limitations of FRAM:
CWT fish accurately represent the modeled stock. Many “model” stocks
are aggregates of stocks that are represented by CWT‟s from only one
production type, usually hatchery origin. For example, in nearly all cases
wild stocks are aggregated with hatchery stocks and both are represented
by the hatchery stock‟s CWT data. Therefore, for each modeled stock
aggregate, it is assumed that the CWT data accurately represent the
exploitation rate and distribution pattern of all the untagged fish in the
modeled stock.
Length at age of Chinook is stock specific and is constant from year to
year. Von Bertalanffy (1934) growth functions are used for Chinook in
determining the proportion of the age class that is of legal size in size-limit
fisheries. Parameters for the growth curves were estimated from data
collected over a number of years. It is assumed that growth in the year to
be modeled is similar to that in the years used to estimate the parameters.
Stock distribution and migration is constant from year to year and is
represented by the average distribution of CWT recoveries during the base
period. Fishery managers currently lack data on the annual variability in
distribution and migration patterns of Chinook salmon stocks. In the
absence of such estimates, fishery-specific exploitation rates are computed
relative to the entire cohort. Differences between the distribution and
migration pattern of stocks during the base period and the year being
modeled will decrease the accuracy of the estimates of stock composition
and stock-specific exploitation rates for a modeled fishery.
There are not multiple encounters with the gear by the fish in a specific
time/area/fishery stratum. Within each time/area/fishery stratum, fish are
assumed to be vulnerable to the gear only once. The catch equations used
in the model are discrete and not instantaneous. Potential bias in the
estimates may increase with large selective fisheries or longer time
intervals, both of which increase the likelihood that fish will encounter a
gear more than once.
While it is difficult to directly test the validity of these assumptions, results of
validation exercises provide one assessment of how well these assumptions are
met and the sensitivity of the model to the assumptions (MEW 2006).
Additionally, one study conducted in Canada looking at behavior and subsequent
mortality rates of adult Chinook salmon caught and released from purse seine
43
fisheries in Johnstone Strait revealed that mortality is based on several variables
such as size, landing procedure, landing time, catch size and degree of external
injury (Candy et. all 1996). This study also confirmed that Chinook caught and
released from purse seine fisheries can be recaptured in the same fishery, or
concurrent fisheries utilizing different gear types, within the original capture site
vicinity.
The WDFW FRAM relies heavily upon spawner survey data and
escapement numbers. Unfortunately, these data and escapement numbers can
often be inaccurate (Knudsen 2000). This is due to the fact that escapement
numbers are provided by the use of combination of both indirect methods such as
counting parts, or spawned adult salmon at spawning grounds, counting redds,
etc. and direct methods such as recovering coded wire tags during each fishery
(Section 2.1). It is impossible to sample 100% of the stock composition present in
a fishery using these sampling methods. Further, the ability of FRAM, or any
fishery model, to predict reality is limited by our lack of full understanding of
ecological processes controlling populations, our inability to measure those
processes accurately, and to incorporate all the relevant processes in a single
model. Currently, there are no existing modeling approaches that could produce
an unambiguous risk classification for weak salmon stocks (Wainwright and
Waples 1998).
An effort to assess the accuracy of pre-season FRAM reporting with the
Genetic Stock Identification data collected in-season after the close of the season
could significantly help the pre-season forecasting for the subsequent year. Using
44
the pre-season estimates and in-season data to readjust the model inputs could
help fine tune the error provided by both methods of estimation (Blankenship
2007).
2.6
Genetic Stock Identification
Genetic Mixed Stock Analysis (MSA), or Genetic Stock Identification, (GSI) are
used interchangeably in WDFW fishery management reports. This genetic tool
has been used infrequently over the last ten years as an in-season tool in an
attempt to validate FRAM pre-season predictions of mortality on stocks of
Chinook salmon caught as bycatch in Puget Sound commercial fisheries which
target other salmon species. Data provided from GSI analysis can offer an actual
impact per stock as encountered in a fishery, rather than theorized impacts
predicted by fishery models which have been solely relied upon in the past. There
are known issues with the GSI approach however. ONCOR, a computer
simulation program WDFW uses to run GSI simulations and theorize stock
impacts, has been known to incorrectly assign a small proportion of the overall
fish sampled to any of the represented populations sampled during its simulations
(Blankenship 2007). This error, although accounted for in the probable error of
running any computer simulation, could result in fishery managers reporting
inaccurate impacts on ESA-listed fish populations. The benefit to using GSI in
light of this error is that errors reported are usually very small percentages,
usually within a confidence interval (Blankenship 2007). This type of error also
does not result in a false assignment of mortality to a stock which was not present
45
in the fishery, as the simulation can only assign mortality to stocks present in the
samples collected during the fishery and provided for laboratory analysis. Lastly,
the Genetic Analysis of Pacific Salmonids (GAPS) database must contain an
exhaustive collection of DNA profiles in order to match the sampled fish from the
fishery back to the known stock of origin. This database is not entirely complete,
but is revised every year to include each new West Coast salmonid stock
genetically identified. A coast wide effort to complete the GAPS profiles of all
Chinook salmon stocks has been initiated by genetics labs from Alaska to
California in order to strengthen the abilities of fishery managers to report stock
impacts in their entirety (Blankenship 2007; OSU 2008).
The draft 2006-2008 Research and Data Needs for the PFMC identifies as
its highest priority the development of GSI for fisheries management applications.
The report states: “Advances in genetic stock identification, and other techniques
may make it feasible to use a variety of stock identification technologies to assess
fishery impacts and migration patterns: The increasing necessity for weak-stock
management puts a premium on the ability to identify naturally reproducing
stocks and stocks that contribute to fisheries at low rates. The CWT marking
system is not suitable for these needs. The Council should encourage efforts to
apply these techniques to management” (OSU 2008).
A DNA analysis was initially conducted on bycatch from WDFW Marine
Areas 7, 7A, 7B and 7C commercial and recreational fisheries in 1998 by Anne
Marshall of the WDFW genetics unit. This original analysis was performed by
identifying allozyme genotypes which are detectable in collected fin tissues
46
(Marshall 1998). The GAPS database is currently being used to perform a more
comprehensive genetic analysis on fin clipped tissues using microsatellite analysis
and SNP markers (Warheit 2006, Blankenship 2007). This new microsatellite
analysis provides the probability that each fish originated from the genetic stock
identity stored in the GAPS database. Fish that show a weak overall probability,
roughly less than 70%, of matching a known stock of origin present in the GAPS
database are excluded from mortality assessments (Blankenship 2007).
A general lack of funding prevented genetic samples from being obtained
and analyzed in the years following 1998. Since the 1998 genetic analysis, GSI
data was also collected and analyzed for 2006 and 2007. However, forecasts for
poor sockeye returns limited the days the Marine Area 7 and 7A commercial
fisheries were open in these years, thus sample sizes for 2006 and 2007 fell
drastically below the desired numbers (Hawkins and Adicks 2007). As such,
these data are limited in what they can be used to infer about the presence and
distribution of Chinook salmon stocks in Marine Area 7 and 7A fisheries.
2.7
Marine Protected Areas
Since the 1998 WDFW Marine Protected Areas (MPA) policy was created, the
Director of the Washington Department of Fish and Wildlife has been using
marine protected areas as one of the agency's working tools for resource
protection and management. The Director has been responsible for plan
development and implementation to manage consumptive and non-consumptive
uses. The creation of a MPA is not delayed until all the habitat and population
47
assessment questions are answered with scientific studies because the recovery of
depressed populations often depends on a timely establishment of these sites.
WDFW relies on existing data to determine populations of concern and the
selection of MPAs (WDFW 2008). Many fish resources require major reductions
in harvest pressure and protection from removal as bycatch to establish productive
populations of adults (Cook 2006; WDFW 2008). MPAs provide an important
tool fisheries managers can use to recover species from past practices of overharvesting, and prevent future over harvest (Bohnsack 1993; 1996). They also
provide areas for the collection of baseline data on populations found within the
site, provide reference areas, and protect endemic or sensitive populations and
habitats. Lastly, they facilitate integrated management of all resources within the
established MPA. The WDFW Commission‟s approach to implementing and
designating MPAs specifically in the Puget Sound includes:
-Designed MPAs are needed in Puget Sound to protect a variety of
species, to promote the recovery of some over-harvested species and to
protect important habitats.
Current MPAs that encompass WDFW Marine Area 7:
-Yellow and Low Islands Marine Preserve (closed to salmon fishing)
-Shaw /Friday Harbor (open to salmon fishing)
-Argyle Lagoon (open to salmon fishing)
-False Bay (open to salmon fishing)
48
2.8
Critical Habitat Designation
The ESA requires the federal government to designate “critical habitat”
for any species it lists under the ESA; in this case, salmon and steelhead
populations (Figure 6). “Critical habitat” is defined as: (1) specific areas within
the geographical area occupied by the species at the time of listing, if they
contain physical or biological features essential to conservation, and those
features may require special management considerations or protection; and (2)
specific areas outside the geographical area occupied by the species if the agency
determines that the area itself is essential for conservation (WDFW 2008)
Figure 6. Critical Habitat Map of Washington. http://www.nwr.noaa.gov/Salmon-Habitat/CriticalHabitat/upload/WA-CH-map.pdf.
49
Critical habitat designations must be based on the best scientific information
available, in an open public process, within specific timeframes. Before
designating critical habitat, careful consideration is given to the economic
impacts, impacts on national security, and other relevant impacts of specifying
any particular area as critical habitat. The Secretary of Commerce may exclude an
area from critical habitat if the benefits of exclusion outweigh the benefits of
designation, unless excluding the area will result in the extinction of the species
concerned (WDFW 2008). Under Section 7 of the ESA, all federal agencies must
ensure that any actions they authorize, fund, or carry out are not likely
to jeopardize the continued existence of a listed species, or destroy or adversely
modify its designated critical habitat. A critical habitat designation does not set
up a preserve or refuge, and applies only when federal funding, permits, or
projects are involved; critical habitat requirements also do not apply to citizens
engaged in activities on private land that do not involve a federal agency.
50
3
Methods
3.1
Study Site
The study was conducted in WDFW Marine Area 7, which encompasses Areas 7,
7A, 7B and 7C (Figure 7).
U.S. - Canada Border
Area 7A
Area 7B
*
*
Area 7C
Area 7
Legend
* Used to
indicate the
close
proximity of
the San Juan
Islands
Salmon
Preserve to
the fishing
area.
Figure 7. Study Area: WDFW Marine Areas 7, 7A, 7B, and 7C. Courtesy of S. Blankenship, WDFW.
2007.
51
Pursuant to the North of Falcon (NOF) fishery monitoring agreements,
DNA samples were collected during the months of July – September of 2006 and
2007 by WDFW observers placed aboard random fishing vessels throughout each
12 hour commercial fishery open in Marine Areas 7, 7A, 7B and 7C (Hawkins
and Adicks 2007). Using sterilized scissors, a small fin-clip (< 10 grams), of
dorsal fin tissue from: the first five Chinook salmon caught per set were collected
for purse seine vessels targeting sockeye, all Chinook salmon caught for gillnet
vessels targeting Chinook salmon, and each landed Chinook salmon for
recreational fishermen targeting Chinook salmon. Fin clips were placed in a
sterile vial filled with ethanol to avoid cross contamination. In addition, scale
samples were also obtained for the first five Chinook sampled aboard purse seine
and gillnet vessels and processed separately for age and stock of origin
determination in the WDFW scale lab. Additional data collected included latitude
and longitude of the fishing vessel, time of capture, fish length and sex (if
possible), and the presence or absence of adipose fin (used to indicate a hatcheryorigin fish). The DNA sample vials were delivered to the WDFW genetics
laboratory within 12 hours of collection and were processed within 48 hours as
described below in Section 3.2.
3.2
Data Analysis
The following methods were performed to obtain and analyze genetic
material in the WDFW genetics laboratory; DNA extraction, Polymerase Chain
Reaction (PCR) Amplification and Genotyping. This process began with the
52
extraction of DNA from fin tissue samples and the purification of the obtained
DNA material using Macherey-Nagel silica membrane kits. PCR reactions were
run using MJ Research PTC-200 and AB 9700 thermal cyclers. The 13
microsatellite DNA loci which currently comprise the coast wide Chinook salmon
DNA screening protocol (Ogo-2, Ogo-4, Oki-100, Omm-1080, Ots-3M, Ots-9,
Ots-201b, Ots-208b, Ots-211, Ots-212, Ots-213, Ots-G474, and Ssa-408) were
screened using an ABI-3730 DNA Analyzer with in-lane size standards (ABIGeneScan-500 liz) and GeneMapper 3.7 software. Allele binning and naming
was accomplished using MicrosatelliteBinner-v.1h, where MicrosatelliteBinner
creates groups (bins) of alleles with similar mobilities (alleles with the same
number of repeat units), and the upper and lower bounds of the bins are
determined by identifying clusters of alleles separated by gaps (nominally 4.0
base pairs in size) in the distribution of allele sizes. Each bin is then named as the
mean allele size for the cluster rounded to an integer (Hawkins and Adicks 2007).
The mixed stock analysis program used the 13 standardized microsatellite
loci from the GAPS consortium (GAPS v2.1 dataset; release date Aug. 25, 2006),
which contains genetic data for 167 stocks categorized into 44 regional reporting
units by the Pacific Salmon Commission (PSC). Estimates of stock of origin for
each individual were generated using a Bayesian procedure based on the
probability that a genotype from the fishery samples was derived from a
stock/population, given the baseline allele frequencies for that population.
Genotype probabilities were generated using the algorithm of Rannala and
Mountain (1997). A Markov chain procedure was used to refine the fishery
53
proportion estimates derived from the genotype probabilities. The stock
contribution estimates are the mean posterior probabilities from the Markov
chain. Estimates of error for the stock assignments were generated through
simulation. One thousand datasets were constructed, each containing 100
individuals, where the stock composition was identical to that of the stock
composition estimated from the actual bycatch samples. Each simulated sample
was analyzed as described above, with the mean and variance of stock
assignments recorded. This procedure allowed us to calculate an error (as a
standard deviation) for the stock composition estimated from the actual bycatch
samples. GMA (Kalinowski 2003) software was used to estimate stock
composition and its associated error, and Genclass2 (Piry et al., 2004) to conduct
individual-based assignments for the 2006 data (Warheit 2006). ONCOR
software was used for the individual-based assignments of the 2007 data
(Blankenship 2007).
4
Results
4.1
Comparison of Chinook Bycatch Observed in Marine Areas 7 and 7A
With the exception of year 2004, Marine Area 7A consistently provided a larger
sample size of Chinook bycatch than Marine Area 7 over the ten years observed
(Table 3). (The validity of the 2004 outlier was verbally verified with the Puget
Sound Salmon Management Unit supervisor. He confirmed that data from 2004
was entered into the database correctly from the original observer datasheet). A
Shapiro-Wilk normality test was performed on the raw data collected over the ten
54
years from Marine Areas 7 and 7A. The normality test provided an insignificant
result, with a p-value of 0.1523 (Tables 12-14). Thus, the null hypothesis, that
these data came from a normally distributed data set, could not be rejected. These
data were then considered to fit a normal distribution curve and were analyzed as
such. A t-Test: Paired Two Sample for Means using a one tailed t-Test, to test if
the number of Chinook salmon observed in Marine Area 7 differed significantly
from the number of Chinook salmon observed as bycatch per year in Marine Area
7A over the ten year period, was performed using both MS Excel and R Version
2007. The results from both programs yielded significant results, with a p-value
of 0.00899 (Tables 12-14). Thus the number of Chinook salmon observed per
year in Marine Area 7A does differ significantly from the number of Chinook
salmon observed in Marine Area 7 the ten year period.
Table 3
Chinook Salmon Bycatch Observed 1997-2007
As Reported by WDFW Marine Area
Year
1997
1998
2000
2001
2002
2003
2004
2005
2006
2007
Area 7
69
67
56
78
67
159
56
8
116
2
Area 7A
498
219
70
109
81
811
6
599
467
149
Table 3. WDFW Observer MS Access Database 2007.
55
4.2
Genetic Stock Identification Data for 1998, 2006 and 2007
The Chinook salmon genetic data collected and analyzed for stock of origin in
1998, 2006 and 2007 demonstrate a consistency in stock composition over time
(Figure 8, Tables 5-9). The analysis of these data revealed that Canadian Chinook
salmon stocks present were consistently dominant in these three years, and the
presence of Puget Sound Chinook salmon stocks was very low. Further statistical
analysis of these data was not possible due to the small and varied sample sizes,
the differing methods used to provide these values, and the natural variance
present in the MSA analysis process (Blankenship 2007). A Chi-squared analysis
was not possible as these data did not meet requirements to run such a test.
Additionally, GSI data was historically combined for Marine Areas 7 and 7A. As
Marine Area 7 consistently provided significantly smaller sample sizes these data
may have been combined to increase the overall sample sizes for power in
statistical analyses (Table 3, 5-7).
56
Chinook Sa lm on Byca t ch Ge ne t ica lly Sa m ple d Fr om M a r ine Ar e a s
7 a nd 7 A in T a r ge t e d Socke ye / Pink Pur se Se ine Fishe r ie s
100
98
90
Percent ag e In Fi shery
80
93
84
70
60
1998
50
2006
40
2007
30
20
10
13
0
Canada
Puget Sound- United
States
Other- United States
Count ry of Origin of By ca t ch St ock
Figure 8. Assignment of stock of origin of genetic samples obtained from Chinook salmon bycatch during
commercial sockeye and pink purse seine fisheries in Marine Areas 7 and 7A for 1998, 2006 and 2007.
WDFW Memos 2006, 2007.
4.3 Comparison of Fishery Regulation Assessment Model (FRAM) Predictions
and Genetic Stock Identification (GSI) Data for 2007
The 2007 pre-season non-treaty FRAM mortality estimate for Chinook salmon as
bycatch in the sockeye directed purse seine fishery for Marine Areas 7 and 7A
was 1,931. Of the 1,931 Chinook salmon 1,184, or 61.32%, were expected to be
specifically of Puget Sound origin (Figure 9). The total Puget Sound Chinook
salmon bycatch ceiling computed by FRAM was 1,421* (* computed as 1,184 x
120% = 1,421 cap (Blankenship 2007). Due to the unusually low number of
returning sockeye salmon in 2007, the purse seine fishery in Marine Areas 7 and
7A was not opened for commercial fishing of sockeye. However, it was an odd
numbered calendar year, and pink salmon spawn in large numbers during odd
57
numbered years. A commercial fishery targeting pink salmon was open in Marine
Areas 7 and 7A during 2007. As the sockeye and pink salmon fisheries overlap
by a few statistical weeks on the WDFW fishery regulation calendar, fishery
managers operated under the assumption that the composition of bycatch stocks
present in the pink salmon purse seine fishery should closely mimic that which
would be sampled during those same statistical weeks of the sockeye salmon
purse seine fishery in Marine Areas 7 and 7A (Blankenship 2007). Genetic
samples collected during the 2007 pink directed purse seine commercial fishery
revealed that the Puget Sound Chinook salmon presence during the Marine Area 7
and 7A fishery was 4%, and significantly lower than the pre-season FRAM
predicted mortality (Figure 8, Tables 9-10). This discrepancy between the preseason FRAM prediction and the in-season GSI data may be accounted for by the
shortened fishery season, and lack of a commercial sockeye directed fishery.
2 0 0 7 FRAM Pre dict e d Puge t Sound Chinook Sa lm on
Mort a lit y v s. GSI Puge t Sound Chinook Mort a lit y
100
Percent age In Fishery
90
80
70
61
60
50
40
30
20
4
10
0
FRAM
2007
DNA
Figure 9. 2007 FRAM predicted percentage of Puget Sound Chinook salmon mortality in purse seine
fisheries from Marine Area 7 and 7A and GSI percentage reported from samples collected during these
fisheries. (Blankenship 2007).
58
4.4 Stock Composition Trends Over Time in the Marine Area 7B and 7C
Bellingham Bay Gill Net Fishery
Genetic samples obtained during the 2007 Marine Areas 7B and 7C
Bellingham Bay Chinook Directed Gill Net Fishery were simultaneously
processed by myself, and the staff of the WDFW genetics laboratory for this
analysis (Table 10). Results suggest that the catch composition was consistent
with the data processed in the WDFW genetics laboratory by Anne Marshall in
1998. The catch composition reported in 1998 was 98% Puget Sound Chinook
salmon, and 2% Canadian origin Chinook salmon (Table 6). The 2007 results
maintained a strong Puget Sound Chinook salmon stock composition of 98%, and
2% of Oregon origin Chinook stocks were present (Table 9-10). Of the 98%
Puget Sound Chinook salmon present in the 1998 sample, it was not possible to
detect the exact Chinook salmon stocks these represented, as the GAPS database
was not complete for all Puget Sound Chinook stocks, hatchery or wild, at that
time. Since 1998 the GAPS database has been expanded to include a more
comprehensive collection of wild and hatchery West Coast salmonids, including
Puget Sound Chinook salmon stocks previously excluded from the GAPS
database. A distinction between North and South Puget Sound Chinook
populations can also now be made, in an attempt to try to isolate geographical
impacts as well. Due to the GAPS database expansion the 2007 Bellingham Bay
Gill Net fishery results were reported to include, with finite precision, which
specific Puget Sound Chinook salmon stocks were impacted. The results are
listed in Table 4.
59
Table 4
WDFW REPORTING GROUP
L_Columbia_R._fall
N_Oregon_Coast
Mid_and_Upper_Columbia_R._spring
SSE_Alaska
Rogue_River_Oregon
Central_BC_Coast
U_Skeena_River
Central_Valley_fall
S_Puget_Sound
NSE_Alaska
E_Vancouver_Island
L_Fraser_River
Central_Valley_spring
Washington_Coast
N_California/S_Oregon_Coast
Mid_Fraser_River
N_Thompson_River
W_Vancouver_Island
Mid_Oregon_Coast
L_Columbia_R._spring
Nass_River
Straits_Juan_de_Fuca
L_Skeena_River
California_Coast
Hood_Canal
U_Columbia_R._summer/fall
Snake_River_spring/summer
Klamath_River
S_BC_Mainland
N_Gulf_Coast
Taku_R.
S_Thompson_River
Deschutes_River_fall
N_Puget_Sound
U_Stikine_R.
Snake_River_fall
Willamette_River
U_Fraser_River
L_Thompson_River
Central_Valley_winter
Mid_Columbia_R._tule_fall
Proportion
0.00
0.00
0.00
0.00
0.02
0.00
0.00
0.00
0.37
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.07
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.54
0.00
0.00
0.00
0.00
0.00
0.00
0.00
Percentage
1.53%
36.84%
7.39%
54.24%
Table 4. ONCOR Results from 2007 DNA analysis of Marine Areas 7B/7C Bellingham
Bay Chinook Directed Gill Net Fishery Samples. Iverson and Blankenship 2008.
60
The resolution of the WDFW reporting groups can also now be broken down to
provide even more precise population impacts. Data from the individual fish each
reporting group is comprised of can be separated out and subsequently assigned
back to their exact stock of origin. For example, in the South Puget Sound
reporting group above in Table 4, Chinook salmon originating from the Voights
Creek Hatchery contributed 7%, Issaquah Hatchery 2%, Nisqually River 6%,
Soos Creek Hatchery 9%, and Cedar River 13% respectively (Table 11).
4.5
Comparison of Effects by Gear Type and Species Targeted
As expected, when directly targeted by fishers Puget Sound Chinook salmon
mortality greatly increases (Figure 10). (These data do not reflect a markselective fishery regulation for recreational fisheries as these fisheries have only
recently begun to be implemented, and were not used in 1998). Of the gear types
discussed here, those gear types allowed to directly target Chinook salmon in
Marine Areas 7/7A/7B/7C are those fishing recreationally and those using gill
nets.
61
1 9 9 8 Puge t Sound Chinook Sa lm on Ge ne t ica lly Sa m ple d From Purse Se ine ,
Gill Ne t a nd Re cre a t iona l Fishe rie s in Ma rine Are a s 7 , 7 A, 7 B a nd 7 C
100
98
90
Per centage In Fisher y
80
Sock eye
Directed
Purse Seine
Chinook
Bycatch
84
79
70
San Juan
Islands
Recreational
Catches
60
50
40
30
Chinook
Directed Gill
Net Fishery
20
21
10
2
13
3
0
Canadian Origin
Puget Sound- United
States Origin
Other- United States
Origin
Figure 10. Total Fishery Impacts Accounted for by genetic sampling on Puget Sound Chinook in 1998 for
Marine Areas 7, 7A, 7B and 7C. (Marshall 1998).
Upon comparing the pre-season FRAM predicted percentages of mortality of
specific Puget Sound Chinook salmon stocks to the data actual in-season genetic
samples obtained from the 2007 recreational fisheries, it appears that FRAM was
unable to predict precise stock impacts with an error of less than 6-15% (Figures
11 and 12).
62
2007 Recreational Catch Proportions Reported In-Season with GSI
versus FRAM Predicted Proportions of Puget Sound Fall Chinook Landed
0 .1 4 0
0 .1 2 0
0 .1 1 6
Proport ions
0 .1 0 0
0 .0 8 0
0 .0 5 5
0 .0 6 0
0 .0 4 0
0 .0 2 0
0 .0 0 0
GSI
FRAM
Figure 11. 2007 FRAM pre-season predictions for recreational Puget Sound Fall Chinook mortality in the
Marine Area 7 fisheries versus in-season GSI data reported. (Blankenship 2007).
2 0 0 7 Re cr e a tiona l Ca tch Pr opor tions Re por te d I n- Se a son w ith GSI
v e r sus FRAM Pr e dicte d Pr opor tions of Puge t Sound Spr ing a nd Sum m e r
Chinook La nde d
0.250
0.208
Proport ions
0.200
0.150
0.100
0.056
0.050
0.000
GSI
FRAM
Figure 12. 2007 FRAM pre-season predictions for recreational Puget Sound Spring and Summer Chinook
mortality in Marine Area 7 fisheries versus in-season GSI data reported. (Blankenship 2007).
63
5
Discussion
These analyses suggest that the sole use of FRAM pre-season to predict
mortality of ESA-listed Chinook salmon stocks in-season during both targeted
and non-target fisheries in Marine Areas 7, 7A, 7B and 7C may not provide the
finest resolution possible on impacts to ESA-listed Puget Sound Chinook salmon
stocks as federally mandated. These analyses also suggest that Marine Area 7 and
7A could be managed separately, as the Chinook salmon stocks present in these
two geographically distinct areas have significantly differed in observed
abundance consistently over the last ten years. As often hypothesized by fishers
and fishery managers alike, the Chinook salmon stocks present in Marine Areas 7
and 7A visually appear to be returning to different geographically located
spawning locations. As such, the historical practice of lumping data from Marine
Areas 7 and 7A together in order to increase sample sizes to have the ability to
perform more powerful statistical analyses has prevented this from being
examined in more detail. It should be noted that the 1998 data was examined
separately for Marine Area 7A stock composition due to the larger sample size
(n=219), (Marine Area 7 n=67, Total Marine Area 7/7A used in reported
statistical analyses was n=286) and the variation was determined to be statistically
insignificant from the total sample of n=286 (Table 5).
Since GSI technologies for identifying Chinook salmon stocks have now
developed to a point where GSI is useful for in-season fishery management, we
may now begin to use this tool on a regular basis to modify how stock impacts are
reported in-season when combined with the pre-season modeling methods
64
previously used. When GSI is used we get „real-time‟ data on the stocks
impacted in a fishery which was not possible in the past using modeling
techniques alone.
Still, GSI analysis is not a panacea. As stated by Wainwright and Waples
(1998), „there is no single, easy method for conducting salmonid risk evaluations
over broad geographic areas: differences in species biology, natural resource
management, and the degree and methods of population monitoring require that
different considerations be emphasized for different species and geographic
areas‟. There are still no perfect methods to assess stock specific mortality and
population risks. Reporting errors (such as those mentioned in Section 2.6) were
detected when a GSI analysis was performed to predict stock specific mortality
estimates with the 2007 Bellingham Bay data I acquired. I was able to detect
some biases in the assigned reporting groups which were assigned by the GSI
software ONCOR during my simulations (Blankenship 2007). This was
discovered when the GSI output was compared to the known origin of the
samples, and before the 1000 simulations were run (Table 10).
Additionally limitations of collecting GSI samples due to the short time
the fishery is open, and in recent years, the complete closure of some commercial
fisheries due to low population estimates of returning stocks of sockeye salmon
have severely limited the data available for GSI analysis. Also, small sample
sizes of fish are insufficient to estimate stock proportion from stocks which are
occurring as a small percentage of the overall mixture collected (i.e. stocks less
65
than 5% of mixture). This may make it difficult to detect these fish as even
present in the fishery with the use of these tools.
Finally, at this time the GAPS genetic baseline is still incomplete for
Puget Sound salmonid populations. Between-region estimates such as those
reported as Canadian Chinook salmon versus United States Chinook salmon for
pre-season FRAM predictions of mortality would likely be correct; however
estimates for specific stocks within the US population could still be inaccurate.
Within-region assignments would be affected by stocks not being present in the
GAPS baseline at the time of GSI analysis, or being represented by a significantly
small sample size in the analysis (Warheit 2006).
However, the long-term goal of using GSI in addition to FRAM is not
only to provide finer scale stock impacts, but to increase the information available
to managers on the temporal and spatial distribution of specific West Coast
salmonid stocks. If GSI data confirms that substantial variation in temporal and
spatial distribution exists, this may allow commercial fishermen access to
relatively abundant stocks of salmon while protecting weak stocks. The next step
in applying GSI technologies to fisheries management is to explore and map the
distributions and migration patterns of stocks in Council-managed fisheries Coast
Wide. However, the most significant advancement will ultimately come from an
improved understanding of stock-specific marine distributions and migration
pathways in relation to submarine topography and oceanic conditions. This will
facilitate another much needed step toward a future of ecosystem-based
management for salmonids (Pitcher 2001).
66
The primary objective of implementing GSI in the yearly NOF monitoring
process would be to improve information on spatio-temporal distribution of West
Coast Chinook salmon for use as precise in-season ESA-listed stock impact
management. In addition, the information gathered will also start to answer
questions about the relative distributions and abundance of Chinook salmon. This
information is vital to reducing weak stock impacts. Such finite information will
greatly reduce the level of uncertainty associated with historically derived stock
assessments. When genetic samples obtained during a fishery confirm that there
are so few Puget Sound Chinook salmon present in a Chinook salmon directed
fishery (Figures 11 and 12), continuously reporting an error even as minimal as 615% each year could potentially have seriously detrimental effects on severely
depressed stocks. For some severely depressed stocks a reporting error of 6-15 %
could feasibly represent the entire remaining population.
It is essential to collect time- and location-specific genetic samples, scale
samples and oceanographic data during each open commercial and recreational
fishery (Barnett-Johnson et al., 2007). These data collections will develop a
complete database of stock distributions through GSI analysis for comparison
with the historically used CWT database, but with fewer assumptions, such as
having fewer hatchery indicator stocks representing natural production stocks, and
much higher resolution in space and time. This will enable fishery managers to
precisely examine migration routes, evaluate the presence and duration of large
congregations of fish, relate fish distributions to ocean conditions, and generally
67
expand the range of information available on Pacific salmon. Compilation of such
a database will require that GSI sampling continue for several years (OSU 2008).
The use of GSI for stock specific distribution patterns and abundance of
Puget Sound Chinook salmon populations could also benefit the recovery of other
ESA-listed species in Puget Sound, such as the Southern Resident Killer Whale
(Orca orcinus). Southern Resident Orca populations primarily located in the
Puget Sound and San Juan Islands were recently listed for protection under the
ESA. The risk assessment was provided by results from a prey selection study on
the resident San Juan Islands Orca population during the summers of 2004-2007.
During this study a team of researchers followed whales and collected fish scales
and remains after observed feeding events. The scales were then examined using
GSI to identify the main prey of these Orcas. The research found that Chinook
salmon were the significantly preferred prey species (Hempelmann et al., 2008).
The recovery plan identified reduced prey availability as a possible risk to the
population (Hempelmann et al., 2008).
6
Recommendations and Suggested Future Research
As the overall Puget Sound Chinook salmon abundance has been
significantly lower over the past ten years of observing in Marine Area 7, I
suggest the following; provide these weak stocks with adequate recovery habitat,
and the time necessary to propagate a surplus of new recruits. This may require
that this fishing area be closed to all fisheries for a period of up to five years.
This amount of time will allow Puget Sound Chinook salmon populations
68
outmigrating as juveniles in year one and two of the closure to completely mature
and return to spawn, as Chinook salmon are sexually mature as early as age three,
but can spawn at age four or five as well (Quinn 2005). Data collected from
spawning surveys completed at spawning grounds of ESA-listed Puget Sound
Chinook salmon populations would provide the baseline of the current
reproductive success of these weak stocks. Over the next five years, even with all
things being constant, such as continued „poor‟ ocean conditions providing an
inadequate food supply and high harvest rates in Alaska and Canada, etc., fishery
managers should expect to see some improvement in the number of spawned
adults if in fact the use of fishery closures will benefit a weak population in the
long run. Fishery managers have been closing fisheries for many years, as
needed, to limit impacts to weak and ESA-listed stocks. Therefore, the costbenefit analysis for continuing to keep Marine Area 7 open at this time does not
make sense. To use the fishery closure tool as it was intended, to limit impacts on
ESA-listed populations, I believe that closing Marine Area 7 for a period of five
years would be using this tool for such purpose.
Additionally, closing Marine Area 7 in its entirety to certain gear types which are
directly targeting Chinook salmon also would benefit the recovery of ESA-listed
Puget Sound Chinook salmon populations. As discussed in Sections 4.1.4 and
4.1.5 (See Figure 10), the gear types which appear to be causing the most impact
to ESA-listed Chinook populations would be the gillnet and recreational fisheries.
These two gear types are currently allowed to directly target Chinook salmon
populations, therefore it would make sense to close all Chinook salmon directed
69
fishing of these two gear types in Marine Area 7 for a period of at least five years.
Fish stocks can be highly resilient, thus a closure period of even five years could
allow marginally weak stocks to build a surplus of recruits and new age classes to
naturally supplement the population (Cook 2006; Quinn 2005). For some
severely depressed stocks, it may be too late to naturally recover these
populations in this manner. However, limiting further impacts on all fish stocks
can only benefit the overall health of the Puget Sound marine ecosystem. As it
has been documented that wild origin fish have higher survival and reproduction
rates in the natural environment than hatchery fish (Jonsson et al., 2003; McIsaac
1990) the cost-effective way to rebuild ESA-listed populations in Puget Sound
would be to limit impacts to wild origin populations where the benefits clearly
outweigh the costs.
With tools already in place such as MPAs and the designation of Critical
Habitat for ESA-listed species, we have already begun to use these new tools to
aid in our efforts to recover weak fish populations (Bohnsack 1993; 1996).
Natural fish populations are resilient, and, given a chance, they can rebound
unbelievably fast (Quinn 2005). Now that we have the tools to examine salmon
stock impacts at the fine scale of genetic stock of origin, we need to use these
tools responsibly for the recovery of those species.
Oregon State University has already begun a pilot program called Project
CROOS, which uses the knowledge and skills of local commercial fishermen to
gather much needed fish distribution data. Fishermen are chartered to collect
samples while fishing their normal and accustomed troll fisheries; during fishery
70
closures, they work with OSU researchers to gather genetic data on fish present in
much of the coastal waters of Oregon. Washington has recently begun to suggest
collaboration with OSU to expand this project coast wide in an effort to collect
fish migration data for all west coast salmonid stocks (Blankenship 2007).
Additionally, as we become more educated as a society about the food we
consume, the ability to verify the catch location, or home basin, of a fish could be
used to market more abundant stocks. This may lead to an increase in the market
value of sustainably harvested fish to local Pacific Northwest salmon fishermen.
GSI is a tool that offers such verification in a relatively quick and efficient
manner through projects such as Project CROOS.
The development of future fishery management models depends on results
of the continued study as well as sustained genetic sampling efforts over the next
several years. Understanding aspects of the life history of fish stocks will be of
increasing importance in the management of existing marine resources.
Describing migratory and distribution patterns, habitat use, age, growth, mortality,
age structure, sex ratios, and reproductive biology will be essential information
for natural resource managers to optimize sustainability and harvest opportunities
of these resources. The improved understanding of ocean distributions that will
result from conducting GSI studies over a period of years will help us characterize
discrete stocks and design much needed stock-specific management measures.
Many factors, both natural and human-related, affect the status of fish stocks,
protected species and ecosystems (NRC 1996). Although these factors cannot all
be controlled, newly available technology and fishery management tools such as
71
GSI enable natural resource management agencies charged with the task of
monitoring impacts to ESA-listed species to have the ability to closely monitor
and adjust protection as necessary.
References
Barnett-Johnson, R., Churchill B. Grimes, Chantell F. Royer, Christopher J.
Donohoe (2007). Identifying the contribution of wild and hatchery Chinook
salmon (Oncorhynchus tshawytscha) to the ocean fishery using otolith
microstructure as natural tags. Canadian Journal of Fisheries and Aquatic
Sciences, Vol. 64, pp. 1683-1692.
Barsh, R. (2008). Interview: The History Of The Salish Sea. Friday Harbor,
Washington: 5.
Beardslee, K., Svend Brandt-Erichsen, Gary Loomis (2006). Conservation Groups
to Sue Feds over Puget Sound Salmon Harvest. W. T. Salmon Spawning and
Recovery Alliance, Native Fish Society, and Clark-Skamania Flyfishers. Press
Release.
Bertalanffy, L. von, (1934).Untersuchungen über die Gesetzlichkeit des
Wachstums. I. Allgemeine Grundlagen der Theorie; mathematische und
physiologische Gesetzlichkeiten des Wachstums bei Wassertieren. Arch.
Entwicklungsmech., Vol. 131: pp 613-652.
Blankenship, S. (2007). WDFW Model Predictions data, stock composition of
Marine Areas 7 and 7A data, and DNA Verification Abilities Interview. WDFW
Olympia, Wa.
Blumm, M. C., Bodi, F. Lorraine (1994). SALMON LAW & HISTORY: Sources
and Analysis, Northwestern School of Law of Lewis and Clark College &
American Rivers Northwest.
Bohnsack, J. A. (1993). Marine reserves. They enhance fisheries, reduce
conflicts, and protect resources. Oceanus Vol. 36: pp 63-71.
Bohnsack, J. A. (1996). Marine reserves, zoning, and the future of fishery
management. Fisheries Vol. 21: pp 14-16.
Brown, D. (2005). Salmon Wars: The Battle for the West Coast Salmon Fishery
Harbour Publishing.
72
Candy, J. R., E. W. Carter, T. P. Quinn, B. E. Riddell. (1996). Adult Chinook
Salmon Behavior and Survival after Catch and Release from Purse-Seine Vessels
in Johnstone Strait, British Columbia. North American Journal of Fisheries
Management Vol. 16: pp 521-529.
Cook, A. (2006). Fisheries Lectures - MES PEEP Fall 2006. TESC. Olympia,
Wa.
Cook, M. (2004). The Mystery of the Disappearing Sockeye Salmon. UBC
Reports. Vol. 50, No. 4. University of British Columbia. BC, Canada.
Dominquez, L. (2007). Salmon Migration Routes Originating in Washington,
Oregon and British Columbia.: Power Point Slides from Salmon Ecology. Salmon
Ecology Fall 2007. TESC Olympia, Wa.
Folkes, M. (2005, 2007). Pre-Season northern diversion forecast of Fraser
sockeye via Johnstone Strait Memorandum. S. B.-D. o. F. a. O. Canada.
Groot, C., T.P. Quinn. (1987). Homing migration of sockeye salmon,
Oncorhynchus nerka, to the Fraser River. Fishery Bulletin. Vol. 85, No. 3, pp.
455-469.
Hall, C. A. (2001). The effects of upwelling on intertidal fish in the Monterey
Bay. Marine Biology Department-UCSC. Santa Cruz, CA, University of
California, Santa Cruz, Ca. BA: 36.
Hawkins, D., Kyle Adicks. (2007). Chinook Funding Proposal: Genetic Stock
Identification of Chinook Bycatch in the Washington Marine Areas 7 and 7A (San
Juan Islands) Sockeye and Pink Fisheries. Submitted to US Chinook Technical
Committee for Funding Under the Budget Increment Associated with the US
Letter of Agreement. WDFW Genetics Laboratory. Olympia, Wa.
Hempelmann, J., M. Bradley Hanson, Robin W. Baird, Candice Emmons,
Gregory S. Schorr, John Sneva, Don Van Doornik, Katherine Ayres, Samuel K.
Wasser, Kenneth C. Balcomb, Kelley Balcomb-Bartok, and Michael J. Ford.
(2008). Genetic Species and Stock Identification of Southern Resident Killer
Whale Prey. Poster. Coastwide Salmonid Genetics Meeting. Olympia, Wa.
Iverson, E. S. (1996). Living Marine Resources: Their Utilization and
Management New York, Chapman & Hall.
Jonsson, N., Bror Jonsson, Lars Peter Hansen. (2003). The marine survival and
growth of wild and hatchery-reared Atlantic salmon. Journal of Applied Ecology
Vol. 40: pp 900-911.
73
Kalinowski, S. T. (2003). Genetic Mixture Analysis 1.0., Department of Ecology,
Montana State University. Bozeman, Mt.
Knudsen, E.E. (2000). Managing Pacific salmon escapements: the gaps between
theory and reality. Sustainable Fisheries Management: Pacific salmon. Lewis
Publishers, Boca Raton, Fl. pp 237-272.
Marshall, A. R. (1998). Genetic Stock Composition Analysis of Three 1998 North
Puget Sound Chinook Fishery Samples Memo. Olympia, Genetics Unit of
Washington Department of Fish and Wildlife. Olympia, Wa.
McIsaac, D. (1990). Factors affecting the abundance of 1977-1979 brood wild fall
Chinook salmon (Oncorhynchus tshawytscha) in the Lewis River, Washington.
Ph. D. dissertation, University of Washington. Seattle, Wa.
MEW (2006). FISHERY REGULATION ASSESSMENT MODEL (FRAM) - An
OVERVIEW for CHINOOK and COHO. Report. MEW Members: Henry Yuen
(USFWS); Andy Rankis (NWIFC); Larrie LaVoy (WDFW); Jim Packer
(WDFW); Curt Melcher (ODFW); Ethan Clemons (ODFW); Robert Conrad
(NWIFC); C. Dell Simmons (NMFS); Rishi Sharma (CRITFC); Allen Grover
(CDFG).
Miller, K. (2002). North American Pacific Salmon: A Case of Fragile
Cooperation. E. a. S. I. G. N. C. f. A. Research.
NOAA-NMFS. (2008). "Essential Fish Habitat." from
http://www.nwr.noaa.gov/Salmon-Habitat/Salmon-EFH/Index.cfm.
NOAA-NMFS. (2008). "Wild Salmon Policy." 2008, from
http://wdfw.wa.gov/fish/wsp/joint/final/fwsp01.htm.
NOAA. (2008). "NOAA Fisheries Information ", from
http://www.nmfs.noaa.gov/msa2005/.
National Research Council (NRC). 1996. Upstream: salmon and society in the
Pacific Northwest. National Academy Press, Washington, D.C.
Osborn, L. (2008). Retrieved 2008 from www.currentresults.com.
OSU. (2008). "ProjectCROOS: Pilot Program to Apply Genetic Stock
Identification in Pacific Salmon Fisheries in 2007." from
http://projectcroos.com/files/docts/gsi2007proposal.pdf.
Pearcy, W. G. (1992). Ocean Ecology of North Pacific Salmonids. Seattle
University of Washington Press - Washington Sea Grant Program.
74
Piry S., A. Alapetite, J.-M. Cornuet, D. Paetkau, L. Baudouin, A. Estoup. (2004)
GeneClass2: A Software for Genetic Assignment and First-Generation Migrant
Detection. Journal of Heredity. Vol. 95: pp 536-539.
Pitcher, T. J. (2001). Fisheries managed to rebuild ecosystems? Reconstructing
the past to salvage the future. Ecological Applications. Vol. 11(2): pp 601-617.
Quinn, T. P. (2005). The Behavior and Ecology of Pacific Salmon and Trout,
University of Washington Press.
Rannala B., Joanna L. Mountain. (1997). Detecting immigration by using
multilocus genotypes. Proceedings of the National Academy of Sciences. Vol.
94, No. 17 pp. 9197-9201.
Runolfsson, B. (1997). Fencing the Oceans: A Rights-Based Approach to
Privatizing Fisheries. Regulation. 20.
Shepard, M. P., Argue A. W. (2005). The 1985 Pacific Salmon Treaty: Sharing
Conservation Burdens and Benefits. UBC Press.
Wainwright, T. C., Robin S. Waples (1998). Prioritizing Pacific Salmon Stocks
for Conservation: Response to Allendorf et al. Conservation Biology, Vol. 12, No.
5, pp. 1144-1147.
Warheit, K., S. Blankenship (2006). Stock composition of Chinook bycatch
within Area 7/7A non-treaty sockeye fishery. Washington Department of Fish and
Wildlife Molecular Genetics Laboratory Memo. Olympia, Wa.
WDFW. (2008). "Critical Habitat Designation." 2008, from
http://www.nwr.noaa.gov/Salmon-Habitat/Critical-Habitat/Index.cfm.
WDFW. (2008). "Fishery Management and Fish Science ", 2008, from
http://wdfw.wa.gov/fish/management/science_management.html;
http://wdfw.wa.gov/fish/wild_salmon_monitor;
http://wdfw.wa.gov/recovery/sciencebasedmanagement_files/frame.htm.
WDFW. (2008). "Harvest Management Information." 2008, from
http://wdfw.wa.gov/factshts/harvest.htm.
WDFW. (2008). "Marine Protected Areas." 2008, from
http://wdfw.wa.gov/fish/mpa/puget_sound/intro.htm;
http://wdfw.wa.gov/fish/mpa/puget_sound/gao.htm;
http://wdfw.wa.gov/fish/mpa/puget_sound/index.htm.
75
Table 5
1998 Purse Seine Bycatch Proportions as Reported by WDFW
MAJOR REGIONAL STOCK GROUP/
REGIONAL STOCK SUB-GROUP
COLUMBIA/SNAKE
Lower Columbia
Lower Columbia
Upper Columbia
Upper Columbia
% (SD)
% (SD)
3 (2)
SP
FA
SP
SU & FA Snake FA
Snake SP & SU
0 (0)
1 (1)
1 (1)
1 (1 )
0 (1)
WASHINGTON COASTAL & STRAIT
0 (0 )
PUGET SOUND
Skagit SP Skagit/Still. SU & FA (wild)
0
Other Puget Sound SU "-~A~
Other Puget Sound SP
B.C.-FRASER RIVER
Lower Fraser SP & SU
1;
Lower Fraser FA
Thompson 3U
Mid-Fraser SP & 3U
Upper Fraser SP
B.C.-VANCOUVER ISLAND/MAINLAND COAST
West Vancouver Island FA
Upper Georgia Strait SU & FA
Lower Georgia Strait SU & FA
TOTAL
0
(0 )
13 (2)
11
(1)
(3)
1 (1)
82
(3)
0 (0)
6
66
(2 )
(4)
9 (3 )
1
(1)
2 (2)
1 (1)
1 (2 )
-1.
100
(1)
100
MA 7 n = 67, MA 7A n = 219, n = 286 total. Anne Marshall WDFW Memo 1998.
76
Table 6
1998 Gill Net Bycatch Proportions as Reported by WDFW
MAJOR REGIONAL STOCK
GROUP/
REGIONAL STOCK SUB-GROUP
COLUMBIA/SNAKE
Lower
Lower
Upper
Upper
Columbia SP
Columbia FA
Columbia SP
Columbia SU & FA
Snake FA
Snake SP & SU
WASHINGTON COASTAL &
STRAIT
PUGET SOUND
Skagit SP
Skagit/Still.
SU & FA (wild)
Other Puget Sound SU & FA
Other Puget Sound SP
% (SD)
0
(0 )
0
(0)
0
(0 )
0
(0)
0
(0)
0
(0)
0
0
(0)
98 (1)
0
(0)
1 (1)
0
(0 )
0
(1)
Thompson SU
0
(0 )
Mid-Fraser SP ~ SU
0
(0)
TOTAL
(0 )
(0)
Lower Fraser SP & SU
Lower Fraser FA
West Vancouver Island FA
Upper Georgia Strait SU & FA
Lower Georgia Strait SU & FA
0
98 (1)
B.C.-FRASER RIVER
Upper Fraser SP
B.C.-VANCOUVER ISLAND/MAINLAND COAST
% (SD)
H
1 (1)
1.
0
(0)
0
(1)
(1.)
__ 1 (1)
lob
100
MA 7B n = 140, MA 7C n = 160, n = 300 total. Anne Marshall WDFW Memo 1998.
77
Table 7
1998 Recreational Catch as Reported by WDFW
MAJOR REGIONAL STOCK GROUP/
REGIONAL STOCK SUB-GROUP
%
(SD)
COLUMBIA/SNAKE
Lower Columbia SP
Lower Columbia FA
0
(0)
0
(0)
Upper Columbia SP
0
(0)
0
(0)
0
0
(0)
(0)
Upper Columbia SU & FA
Snake FA
Snake SP & SU
WASHINGTON COASTAL & STRAIT
PUGET SOUND
Skagit SP Skagit/Still. SU & FA (wild)
0
(0)
79
(5)
10
(6)
o. (0)
Other Puget Sound SU & FA
79
(5)
Other Puget Sound SP
0
(0)
B.C.-FRASER RIVER
Lower Fraser SP & SD
0
(SD
)
(0)
%
0
(0)
5
(4)
5
(4)
0
0
(0)
(0)
1;
Lower Fraser FA
Thompson 3U
Mid-Fraser SP & SU
Upper Fraser SP
B.C.-VANCOUVER ISLAND/MAINLAND COAST
11
West Vancouver Island FA
0
(0)
Upper Georgia Strait SU & FA
0
(1)
Lower Georgia Strait SU & FA
-11.
(7)
TOTAL
100
(7)
100
MA 7 n = 138 total. Anne Marshall WDFW Memo 1998.
78
Table 8
2006 Bycatch Data as Reported by WDFW
Pacific Salmon Commission
Reporting Groups
Area 7A
Non-Treaty Fishery
N = 44
Area 7A
Treaty Fishery
N = 275
S. Thompson River
76%-84%
82%-86%
E. Vancouver Island
0%-4%
5%-7%
Lower Fraser River
10%-18%
2%-4%
Middle Fraser River
-
1%-3%
N. Thompson River
0%-4%
1%-3%
North Puget Sound
-
1%-3%
W. Vancouver Island
-
0%-2%
South Puget Sound
0%-2%
0%-1%
Upper Columbia River summer/fall
-
0%-1%
Washington Coast
-
0%-1%
CA Central Valley - spring
-
0%-1%
Individual Fish Assigned to
Puget Sound
Treaty Fishery (Area 7A)
N = 2; Samish Hatchery
N = 1; S. Prairie Creek or
Voights Creek Hatchery
Ken Warheit WDFW Memo 2006.
Table 8. Stock composition estimates are based on a sample size n = 44 individual Chinook
salmon; therefore, one fish is approximately equal to 2%. Of the 167 stocks present in the
baseline, only a small number were estimated to occur in the bycatch sample. The following
mixed-stock proportions are percent ranges that incorporate the mean estimate and its associated
error: (1) South Puget Sound, 0-2%; (2) Lower Fraser River, 10-18%; (3) South Thompson River,
76-84%; (4) North Thompson River, 0-4%; and (5) East Vancouver Island, 0-4%. Therefore,
these 44 samples were composed of fish from Puget Sound (0-2%) and stocks from Canada (98100%).
79
Table 9
2007 Bycatch Data as Reported by WDFW
Pacific Salmon Commission
Reporting Groups
Area 7/7A
Non-Treaty
Fishery
Proportions
N = 115
Rogue River
0.009
Upper Col. River Summer/Fall
0.010
Washington Coast
0.005
Hood Canal
0.009
Straits of Juan de Fuca
0.013
South Puget Sound
0.009
N = 1; Clear Creek Hatchery
North Puget Sound
0.008
N = 3; S.F. Skokomish, Elwha
Hatchery, Marblemount Hatchery
Lower Fraser River
0.204
South Thompson River
0.615
Upper Fraser River
0.009
East Vancouver Island
0.091
Central BC Coast
0.011
SSE Alaska
0.009
Individual Fish Assigned to
Puget Sound
Non-Treaty Fishery (Areas 7/ 7A)
MA 7/7A total n = 115. Scott Blankenship. WDFW Report 2007.
80
Table 10
2007 Marine Area 7B/C Gill Net Genetic Data as Reported by ONCOR
Populations
Actual Value
GSI
SIM
L_Columbia_R._fall
0
0.00
0.00
0.00
0.01
N_Oregon_Coast
0
0.00
0.00
0.00
0.00
Mid_and_Upper_Columbia_R._sprin
g
0
0.00
0.00
0.00
0.00
SSE_Alaska
0
0.00
0.00
0.00
0.01
Rogue_River
0.0153
0.01
0.01
0.00
0.04
Central_BC_Coast
0
0.00
0.00
0.00
0.01
U_Skeena_River
0
0.00
0.00
0.00
0.00
Central_Valley_fall
0
0.00
0.00
0.00
0.00
0.3683
0.59
0.07
0.45
0.72
NSE_Alaska
0
0.00
0.00
0.00
0.00
E_Vancouver_Island
0
0.00
0.00
0.00
0.01
L_Fraser_River
0
0.00
0.00
0.00
0.00
Central_Valley_spring
0
0.00
0.00
0.00
0.00
Washington_Coast
0
0.00
0.00
0.00
0.01
N_California/S_Oregon_Coast
0
0.00
0.00
0.00
0.00
Mid_Fraser_River
0
0.00
0.00
0.00
0.00
N_Thompson_River
0
0.00
0.00
0.00
0.00
W_Vancouver_Island
0
0.00
0.00
0.00
0.00
Mid_Oregon_Coast
0
0.00
0.01
0.00
0.02
L_Columbia_R._spring
0
0.00
0.00
0.00
0.01
Nass_River
0
0.00
0.00
0.00
0.00
Straits_Juan_de_Fuca
0
0.00
0.00
0.00
0.01
L_Skeena_River
0
0.00
0.00
0.00
0.00
California_Coast
0
0.00
0.00
0.00
0.00
0.074
0.11
0.05
0.03
0.21
U_Columbia_R._summer/fall
0
0.00
0.00
0.00
0.01
Snake_River_spring/summer
0
0.00
0.00
0.00
0.00
Reporting Groups
S_Puget_Sound
Hood_Canal
ST
DEV
(95% INT)
81
Klamath_River
0
0.00
0.00
0.00
0.00
S_BC_Mainland
0
0.00
0.00
0.00
0.00
N_Gulf_Coast
0
0.00
0.00
0.00
0.00
Taku_R.
0
0.00
0.00
0.00
0.01
S_Thompson_River
0
0.00
0.00
0.00
0.01
Deschutes_River_fall
0
0.00
0.00
0.00
0.00
0.5424
0.27
0.06
0.16
0.39
U_Stikine_R.
0
0.00
0.00
0.00
0.00
Snake_River_fall
0
0.00
0.00
0.00
0.00
Willamette_River
0
0.00
0.00
0.00
0.00
U_Fraser_River
0
0.00
0.00
0.00
0.00
L_Thompson_River
0
0.00
0.00
0.00
0.00
Central_Valley_winter
0
0.00
0.00
0.00
0.00
Mid_Columbia_R._tule_fall
0
0.00
0.00
0.00
0.00
N_Puget_Sound
MA 7B/7C n = 64 total. WDFW Genetic Data and generated ONCOR Output. Christina Iverson and Scott
Blankenship 2008.
82
Table 11
ONCOR Assignments of Individuals from Reporting Groups for
2007 Gill Net Data in Proportion and Percentage
POPULATION ESTIMATES
Alsea_R
Andrew_Cr
Andrew_CryH
Andrew_MacH
Andrew_MedH
Applegate_Cr
Atnarko_H
Babine_H
Battle_Cr
Big_Boulder_Cr
Big_Qual_H
Birkenhead_H
Bulkley_R
Butte_Cr_Sp
Carson_H
Chetco_R
Chickam_WhitH
Chickamin_R
Chilko_R
Clear_Cr
Clear_Cr_H
Clearwater_R
Cole_Rivers_H
Conuma_H
Coos_H
Coquille_R
Cowichan_H
Cowlitz_H_fa
Cowlitz_H_sp
Cripple_Cr
Damdochax_R
Deadman_H
Deer_Cr_sp
Dungeness_R
Ecstall_R
Eel_R
Elk_H
Elwha_H
Elwha_R
Feather_H_fa
Feather_H_sp
Forks_Cr_H
GeorgeAdams_H
Hamma_Hamma_R
Hanford_Reach
Hoh_R
Proportion
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.02
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.01
0.00
0.00
0.00
%
1.53%
0.69%
83
Humptulips_H
Hupp_Sp_H
Imnaha_R
John_Day_R
Kalama_H_sp
Keta_R
Kilchis_R
Kincolith_R
King_Cr
King_Salmon_R
Kitimat_H
Klamath_R_fa
Klinaklini_R
Klukshu_R
Kowatua_Cr
Kwinageese_R
L_Adams_H
L_Deschutes_R
L_Kalum_R
L_Sauk_R
L_Tahltan_R
L_Thom_R
Lewis_H_sp
Lewis_R_f
Louis_Cr
Lyons_Ferry_H
M_Shuswap_H
Makah_H
Hoko_H_Fa
Marble_H
Marblemount_H_sp
Marblemount_H_su
Maria_Slough
McKenzie_H
Methow_R
Mill_Cr_sp
Millicoma_R
Minam_R
Morkill_R
N_Santiam_H
Nakina_R
Nanaimo_H_f
Necanicum_H
Nechako_R
Nehalem_R
Nestucca_H
Newsome_Cr
NF_Nooksack_H
NF_Stilliguamish_H
Nicola_H
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
84
Nitinat_H
Owegee_R
Porteau_Cove_H
Puntledge_H_f
Queets_R
Quesnel_R
Quilayute_R
Quinsam_H
Raft_R
Rapid_R_H
Robertson_H
Russian_R
S_Coos_H
S_Prairie_Cr
S_Umpqua_H
Sacramento_H
Salmon_R_f_Fraser
Salmon_R_f_OR
Samish_H
Sandy_R
Sarita_H
Secesh_R
Siletz_R
Situk_R
Siuslaw_R
Sixes_R
Skykomish_R
Snoqualmie_R
Sol_Duc_H
Soos_H
Spius_H
Spring_Cr_H
Stanislaus_R
Stillaguamish_H
Stuart_R
Suiattle_R
Sustut_R
Swift_R
Tahini_R
Tahsis_R
Tatsatua_Cr
Torpy_R
Tranquil_R
Trask_R
Trinity_H_f
Trinity_H_sp
Tucannon_H
Tucannon_R
Tuolumne_R
U_Cascade_R_Sp
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.48
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.02
0.00
0.09
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
47.94%
1.88%
8.93%
85
U_Chilcotin_R
U_Deschutes_R
U_Nahlin_R
U_Sauk_R
U_Skagit_Su
U_Yakima_Sp
Umpqua_H
Voights_H
W_Chilliwack_H
Wallace_H
Wannock_H
Warm_Springs_H
Wells_H
Wenatchee_H_sp
Wenatchee_R_sp
Wenatchee_R_s/f
WF_Yankee_Frk
White_H
Wilson_R
Yaquina_R
L_Skagit_R_Fa
U_Sauk_R_SpSu
Skykomish_H_Su
Skykomish_R_Su
Nisqually_R_SuFa
Bear_Cr_SuFa
Cedar_R_SuFa
Grovers_Cr_H
Issaquah_Cr_SuFa
Issaquah_H_SuFa
UW_H_SuFa
NF_Skokomish_R_Fa
SF_Skokomish_R_SuFa
Hoh_R_SpSu
Quinault_NFH_Fa
Quinalt_R_Fa
Chehalis_R_Fa
Elochoman_R_Fa
Abernathy_NFH_Fa
Abernathy_Cr_Fa
Coweeman_R_Fa
Green_R_Fa
Lewis_R_Fa
Lewis_R_LFa
Washougal_R_Fa
Klickitat_R_Su
L_Yakima_Fa
Marion_Drain_Fa
Yakima_bright_Fa
Priest_Rapids_H_Fa
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.07
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.03
0.02
0.00
0.00
0.06
0.00
0.13
0.00
0.00
0.02
0.00
0.01
0.06
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
6.64%
2.82%
1.60%
6.09%
12.74%
2.16%
0.82%
5.88%
86
Umatilla_H_Fa
American_R_Sp
Little_Naches_Sp
Naches_Sp
Twisp_R_Sp
TOTALS
0.00
0.00
0.00
0.00
0.00
1.00
99.99%
Christina Iverson and Scott Blankenship. WDFW Genetic Data and ONCOR Output 2008.
87
Table 12
Shapiro-Wilk normality test results using R Version 2007
Chinook in 7
69
67
56
78
67
159
56
8
116
2
Chinook in 7A
498
219
70
109
81
811
6
599
467
149
Diff
-429
-152
-14
-31
-14
-652
50
-591
-351
-147
> shapiro.test(Fish_Diff$Diff)
Shapiro-Wilk normality test results:
data: Fish_Diff$Diff
W = 0.8859, p-value = 0.1523
Table 13
t-Test: Paired Two Sample for Means performed with R Version 2007
> t.test(Fish_Diff$Diff, alternative='less', mu=0.0,
conf.level=.95)
One Sample t-test results:
data: Fish_Diff$Diff
t = -2.8867, df = 9, p-value = 0.00899
alternative hypothesis: true mean is less than
0
95 percent confidence
interval:
-Inf -85.07786
sample
estimates:
mean of x =
-233.1
88
Table 14
t-Test: Paired Two Sample for Means performed with MS Excel according to Section 12.4
of Applied Statistics with Microsoft Excel. (Keller 2001).
t-Test: Paired Two Sample for
Means
Mean
Variance
Observations
Pearson Correlation
Hypothesized Mean Difference
df
t Stat
P(T<=t) one-tail
t Critical one-tail
P(T<=t) two-tail
t Critical two-tail
Chinook in 7
67.8
2103.511
10
0.456
0
9
-2.886
0.00899
1.833
0.0179
2.262
Chinook in 7A
300.9
74527.43
10
89
Figure 13. Data Distribution Histograms for Raw Data of Marine Area 7
Observations from 1997-2007.
Figure 14. Data Distribution Histograms for Raw Data of Marine Area 7A
Observations 1997-2007.
90
APPENDIX A
ACRONYMS:
NOAA-National Oceanic and Atmospheric Administration
USFWS- United States Fish and Wildlife Service
WDFW-Washington State Department of Fish and Wildlife
NMFS-National Marine Fisheries Service (fisheries branch of NOAA)
PMFC-Pacific Marine Fisheries Commission
DFO-Department of Fisheries and Oceans, Canada
PSVOA- Purse Seine Vessel Owners Association
MSA- Magnuson-Stevens Fishery Conservation and Management Act
ESA- Endangered Species Act
ESU- Evolutionarily Significant Units
TAC- Total Allowable Catch
DEFINITIONS: (As Listed in SEC. 3. 16 U.S.C. 1802 Of the MagnusonStevens Act)
Anadromous species - species of fish which spawn in fresh or estuarine waters of
the United States and which migrate to ocean waters.
Bycatch - fish which are harvested in a fishery, but which are not sold or kept for
personal use, and includes economic discards and regulatory discards. Such term
does not include fish released alive under a recreational catch such as a mark
selective fishery.
Commercial fishing - fishing in which the fish harvested, either in whole or in
part, are intended to enter commerce or enter commerce through sale, barter or
trade.
Conservation and management - refers to all of the rules, regulations,
conditions, methods, and other measures (A) which are required to rebuild,
restore, or maintain, and which are useful in rebuilding, restoring, or maintaining,
any fishery resource and the marine environment; and (B) which are designed to
assure that-(i) a supply of food and other products may be taken, and that recreational
benefits may be obtained, on a continuing basis;
(ii) irreversible or long-term adverse effects on fishery resources and the marine
environment are avoided; and
(iii) there will be a multiplicity of options available with respect to future uses of
these resources.
91
