-
Title
-
An evaluation of the Washington coastal Rod-and-Reel Survey’s ability to detect Black Rockfish (Sebastes melanops) population abundance changes
-
Date
-
2022 June
-
Creator
-
Hillier, Lisa
-
Identifier
-
Thesis_MES_2022Sp_HillierL
-
extracted text
-
An evaluation of the Washington coastal Rod-and-Reel Survey’s
ability to detect Black Rockfish (Sebastes melanops) population
abundance changes
by
Lisa K. Hillier
A Thesis
Submitted in partial fulfillment
of the requirements for the degree
of Master of Environmental Studies
The Evergreen State College
June 2022
© 2022 by Lisa K. Hillier. All rights reserved.
This Thesis for the Master of Environmental Studies Degree
by
Lisa K. Hillier
has been approved for
The Evergreen State College
by
___________________________
John C. Withey, Ph.D.
Member of the Faculty
Abstract
An evaluation of the Washington coastal Rod-and-Reel Survey’s ability to detect Black Rockfish
(Sebastes melanops) population abundance changes
Lisa K. Hillier
The Washington coastal population of Black Rockfish (Sebastes melanops) supports a popular
recreational fishery and contributes to the diversity of the nearshore ecosystem. Working through
the Pacific Fisheries Management Council process, state fisheries managers at the Washington
Department of Fish and Wildlife (WDFW) set annual harvest guidelines for Black Rockfish.
Fisheries managers depend on survey data to inform assessment models for coastal bottomfish
species, and without reliable, comprehensive, and representative fishery-independent survey data
for stock status analysis, harvest limits may be unsuitable for maintaining sustainable fisheries.
The WDFW implemented a standardized nearshore coastal Rod-and-Reel Survey in 2019 to
address this need, but the survey’s ability to detect changes in Black Rockfish abundance had not
been evaluated. Here, information from published literature, data from WFDW pilot studies, and
data from the Washington Black Rockfish tagging program were used to evaluate the Survey’s
design and determine the effects of location, gear type, bait type, seasonal timing, and sampling
effort on Black Rockfish catch. Results showed that sampling at Rod-and-Reel Survey
designated index stations provided representative estimates of the coastal adult Black Rockfish
population and that the gear type and terminal tackle used were appropriate for sampling semipelagic schooling rockfishes. Spring (March-May) sample timing was biologically suitable for
Black Rockfish off Washington and is desirable due to vessel availability, ocean conditions, and
budget considerations. Analysis of the 2019 and 2021 Rod-and-Reel Survey data concluded that
sampling each of the 125 index stations with 4-drifts per station was adequate for detecting
changes in Black Rockfish population abundance. This comprehensive evaluation affirms that
the WDFW coastal Rod-and-Reel Survey can produce representative abundance indices that can
be utilized by managers and incorporated into Washington Black Rockfish stock assessment
models.
Table of Contents
List of Figures ............................................................................................................................... vi
List of Tables ............................................................................................................................... vii
Acknowledgements .................................................................................................................... viii
1. Introduction ............................................................................................................................... 1
2. Literature Review ..................................................................................................................... 6
2.1 Coastal Management and Exploitation History .................................................................... 6
2.2 Black Rockfish (Sebastes melanops) .................................................................................... 8
2.2.1 Biology ........................................................................................................................... 8
2.2.2 Life History Parameters ................................................................................................. 9
2.2.3 Movement Patterns ...................................................................................................... 10
2.3 WDFW Surveys and Pilot Studies ...................................................................................... 10
2.3.1 Coastal Black Rockfish Tagging Program (1981-2014) ............................................. 10
2.3.2 Seasonal Rod-and-Reel Pilot Study (2010-2013) ........................................................ 14
2.3.3 Rocky Habitat Exploration (2015) ............................................................................... 16
2.3.4 Terminal Tackle Pilot Study (2014 & 2015) ................................................................ 17
2.3.5 Rod-and-Reel vs. Set-line Pilot Study (2015-2017) ..................................................... 20
2.4 Nearshore Survey Considerations and use of CPUE for Abundance Indices ..................... 22
2.5 WDFW Coastal Rod-and-Reel Survey (2019-2021) .......................................................... 24
2.6 Considerations..................................................................................................................... 26
3. Methods .................................................................................................................................... 27
3.1 Survey Location .................................................................................................................. 27
3.2 Survey Terminal Tackle and Gear Type ............................................................................. 31
3.3 Survey Timing .................................................................................................................... 32
3.4 Survey Effort ....................................................................................................................... 32
4. Results ...................................................................................................................................... 34
4.1 Survey Locations ................................................................................................................ 34
4.2 Survey Terminal Tackle and Gear Type ............................................................................. 39
4.3 Survey Timing .................................................................................................................... 42
4.4 Survey Effort ....................................................................................................................... 44
5. Discussion and Conclusions ................................................................................................... 48
6. References ................................................................................................................................ 52
7. Appendices ............................................................................................................................... 61
7.1 Glossary .............................................................................................................................. 61
iv
7.2 Glossary References............................................................................................................ 64
v
List of Figures
Figure 1. Abundance indices derived from the WDFW tagging CPUE analysis and used in the
2015 assessment of the Washington stock of Black Rockfish. ..................................................... 13
Figure 2. Coastwide survey areas for the seasonal rod-and-reel pilot study of Washington coastal
index areas. ................................................................................................................................... 15
Figure 3. Diagram of two terminal tackle types........................................................................... 19
Figure 4. Bar graph from the WDFW gear comparison study by site in 2016 ............................ 21
Figure 5. Bar graph from the WDFW gear comparison study by cell in 2016 ............................ 22
Figure 6. Locations of the historic Black Rockfish tagging program spring surveys off the
Washington coast in CRA2. .......................................................................................................... 29
Figure 7. Abundance indices for index station CPUE for 1998-2014 with lognormal, gamma,
and geometric average plotted ..................................................................................................... 30
Figure 8. The gamma distribution of abundance indices calculated from all surveys and from
surveys only conducted at index stations. ..................................................................................... 35
Figure 9. A scatter plot showing a comparison of CPUE abundance indices .............................. 35
Figure 10. Time series of model spawning output of Black Rockfish in Washington in 2015 ... 37
Figure 11. Time series of stock status of Black Rockfish in Washington in 2015 ...................... 38
Figure 12. CPUE by terminal tackle gear type for survey year 2014. ......................................... 40
Figure 13. CPUE by terminal tackle gear type used in the 2014 WDFW pilot study. ................ 40
Figure 14. CPUE by terminal tackle gear type for survey year 2015. ......................................... 41
Figure 15. CPUE by terminal tackle bait type used in the 2015 WDFW pilot study. ................. 42
Figure 16. Histogram showing CPUE from the fall and spring pilot study surveys from 10
locations over 3 years combined on a natural log scale. ............................................................... 43
Figure 17. CPUE based on 4 drifts and 3 drifts (not independent) at each index location for the
2019, 2020, and 2021 Rod-and-Reel Survey. ............................................................................... 45
Figure 18. Average CPUE calculated for each drift at Index Stations by survey year. ............... 46
Figure 19. Catch Record Areas identified by the Washington Department of Fish and Wildlife
to delineate fishing areas throughout the marine waters of the state. ........................................... 63
vi
List of Tables
Table 1. Table of reference points for a comparison between the 4 Black Rockfish assessment
models to show the effects of the indices ..................................................................................... 39
Table 2. Statistical power (%) with different effect sizes (0.02-0.05) by number of drifts at the
index station. ................................................................................................................................. 47
vii
Acknowledgements
I would like to express my gratitude to the WDFW Fish Program for the support and data for my
thesis. My sincere thanks go out to Dr. Tien-Shui Tsou, Dr. Jason Cope, Lorna Wargo, and
Kathryn Meyer for the guidance, advice, stimulating discussions, and reviews throughout this
process. I would also like to thank my advisor Dr. John Withey for his time, energy, and
expertise. I am extremely grateful to Rob Davis, Kristen Hinton, Dr. Chantel Wetzel, Dr. Kristen
Ryding, and Dr. Yuk Wing Cheng for their advice, statistical expertise, and willingness to share
their knowledge. I would also like to thank my fellow classmates Melissa Sanchez and Erin Stehr
for their kind words of encouragement and our late-night discussions.
viii
1. Introduction
Black Rockfish (Sebastes melanops) is an important component of the nearshore
ecosystem off the Washington coast and a highly desirable species in the coastal recreational
bottomfish (see Glossary) fishery. Their ecological importance and high targetability make
careful management of the stock (see Glossary) essential. Washington’s coastal Black Rockfish
stock is managed through the Pacific Fishery Management Council (PFMC) process which relies
on the best available scientific information to meet the national standards of the MagnusonStevens Fishery Conservation and Management Act (16 U.S.C. § 1801-1891(d) (2014)).
Working through the PFMC process, state fisheries managers at the Washington Department of
Fish and Wildlife (WDFW) set annual harvest guidelines for recreational bottomfish fisheries.
For the PFMC management process to function well, the scientific data available for stock status
analysis must be reliable, comprehensive, and representative of the fish population.
The Washington coastal Black Rockfish population was last fully assessed through the
PFMC process in 2015 (Cope et al. 2016). From this assessment review the PFMC Stock
Assessment Review (STAR) panel report stated that the base model would be greatly improved
by using existing data to derive an index, or by collecting coastwide fishery-independent (see
Glossary) survey data (Advisors 2015). These statements were based on concerns about the
geographic limitations of the WDFW Black Rockfish tagging program data and a
misunderstanding that all data from the tagging program were fishery-dependent (see Glossary).
The STAR panel recommended future research that included definition and measurement of
Black Rockfish habitat, the development of a coastwide fishery-independent survey for
nearshore stocks, and improved catch-per-unit-effort (CPUE) standardization protocols. These
1
recommendations prompted the WDFW to initiate the development of a standardized coastal
nearshore survey that targeted multiple bottomfish species.
The WDFW has contributed to the PFMC process by conducting both fishery-dependent
and independent surveys since the 1980s to assess Black Rockfish populations. Historically,
studies have produced estimates of abundance, growth, survival, and mortality for the proportion
of the Black Rockfish stock found in coastal waters off the central Washington coast between
Grays Harbor and Sea Lion Rock just north of Cape Elizabeth (Wallace et al. 2010). The 2007
coastal Black Rockfish population dynamics model utilized abundance trends from these surveys
to assess the Washington stock and fishery-independent data from the WDFW Black Rockfish
tagging program were used in the 2015 Washington Black Rockfish assessment stock synthesis
(SS) model (Wallace, Cheng, & Tsou 2008; Methot & Wetzel 2013; Cope et al. 2016). In 2019,
the WDFW implemented a fishery-independent coastal Rod-and-Reel Survey to assess multiple
nearshore bottomfish stocks. However, this Survey’s ability to detect changes in abundance of
Black Rockfish for use in the Washington assessment’s base model has not been evaluated.
Prior to the implementation of the 2019 coastal nearshore Rod-and-Reel Survey, the
WDFW completed multiple pilot studies to address issues highlighted in the 2007 and 2015
Black Rockfish assessment reviews. Although the WDFW had been monitoring Black Rockfish
populations through the WDFW Black Rockfish tagging program surveys, and the design
allowed for the collection of both fishery-independent (tag release efforts) and fishery-dependent
(fishery tag recovery) data, the survey design needed adjustment to incorporate multiple
bottomfish stocks and cover a wider geographic range. Targeted species expanded to include
China Rockfish (Sebastes nebulosus), Copper Rockfish (Sebastes caurinus), Quillback Rockfish
(Sebastes maliger), Blue Rockfish (Sebastes mystinus), Deacon Rockfish (Sebastes diaconus),
2
Vermilion Rockfish (Sebastes miniatus), Yelloweye Rockfish (Sebastes ruberrimus), Kelp
Greenling (Hexagrammos decagrammus), and Cabezon (Scorpaenichthys marmoratus) in
addition to Black Rockfish. Several pilot studies were developed to address tagging program
survey design inconsistencies, limitations for targeting multiple bottomfish species, and to
delineate coastal rocky habitat suitable for rockfishes (T. Tsou WDFW, personal
communication). In the fall of 2015, a pilot study to evaluate the effectiveness of rod-and-reel
surveys to capture bottomfish in comparison to a set-line survey was initiated. This pilot study
was conducted over three seasons at several locations off the Washington coast. Additionally, the
WDFW conducted a pilot study to evaluate terminal tackle limitations for capturing schooling
rockfish species. A main goal of this study was to address the issue of consistent and effective
terminal tackle in a rod-and-reel survey for capturing a targeted set of bottomfish. The WDFW
also prioritized efforts to map rocky habitat distribution and quantify abundance off the coast.
Information gained from these pilot studies was used to inform the design of the 2019
standardized coastal nearshore Rod-and-Reel Survey.
When designing the Survey, the WDFW made efforts to address the concerns of the
STAR panel, but a comprehensive evaluation of the design is needed ensure that survey data can
be used to calculate abundance indices for the Washington Black Rockfish stock assessment
model. The Rod-and-Reel Survey must be representative of the adult portion of the Washington
stock, provide a measure of abundance, and be able to detect changes in abundance. If the Rodand-Reel Survey meets these objectives, it may be incorporated into the Black Rockfish base
model in 2023 and in future assessments, improving fishery management. If it does not meet the
objectives, the survey design will be modified by WDFW to ensure that future data collected can
detect abundance changes and an index may be incorporated into later stock assessments.
3
This thesis evaluated the WDFW coastal nearshore Rod-and-Reel Survey to determine
whether it can detect changes in Black Rockfish abundance trends by answering the following
questions:
1.) Are abundance indices derived from surveys at the Rod-and-Reel Survey index station
locations representative of the Washington coastal adult Black Rockfish population?
2.) Does the Rod-and-Reel Survey use terminal tackle and sampling gear that consistently
captures pelagic adult Black Rockfish?
3.) Does the Rod-and-Reel Survey sample at a suitable time of year to consistently capture
Black Rockfish?
4.) Is the Rod-and-Reel Survey effort sufficient to detect changes in population abundance of
Black Rockfish using CPUE estimates?
To address the first question, Rod-and-Reel Survey index station locations were evaluated to
determine if the fish captured at these locations were representative of the Washington coastal
population based on CPUE data from the historic WDFW Black Rockfish tagging program from
1998-2014. The second question was addressed by comparing Black Rockfish CPUE estimates
for terminal tackle trials during the 2014 and 2015 WDFW terminal tackle pilot study. Survey
sampling technique (gear type) was evaluated using information collected during the 2015-2017
WDFW pilot study designed to compare set-line sampling efficiency with rod-and-reel sampling.
Seasonal timing evaluation considered catch rates of Black Rockfish during a WDFW pilot study
conducted from 2010-2013, survey vessel availability, and funding constraints. Finally, survey
effort was evaluated to determine if time spent at each index station was adequate for capturing a
4
representative sample and a power analysis was done to determine if the survey design included
enough locations to be able to detect an informative change in CPUE.
By addressing these four questions this thesis provides fisheries managers with the
information needed to confirm that the WDFW coastal Rod-and-Reel Survey can produce
representative abundance indices for incorporation into the next stock assessment model.
Recommended improvements for abundance indices estimates will be discussed for future Black
Rockfish stock assessments.
5
2. Literature Review
A well-designed survey capable of providing a useful abundance estimate must integrate
information from the target species management and exploitation history, biology, and lessons
learned from research (Walters 2003; Thorson, Stewart & Punt 2012; Kuriyama et al. 2019).
This literature review will begin with a brief synopsis of the management and harvest history of
Black Rockfish in Washington, followed by an overview of the species biology, life history, and
movement behavior. The next section will summarize relevant pilot studies conducted by the
WDFW that were used to inform the Rod-and-Reel Survey design. Finally, there will be a review
of other nearshore survey methods, limitations of these surveys, and information about the use of
CPUE for the development of abundance indices. The design of the WDFW Rod-and-Reel
Survey will be discussed in detail to conclude this section. This review of relevant literature,
pilot studies, and the Rod-and-Reel Survey will provide the framework for conclusions drawn
from the analysis of the Survey design.
2.1 Coastal Management and Exploitation History
Washington’s coastal Black Rockfish stock is managed through the PFMC process. The
PFMC is tasked with sustainable fisheries management of approximately 119 species of fish on
the West Coast of the United States and relies on information from committees, advisory panels,
subpanels, and the public (PFMC 2020). The PFMC’s Science and Statistical Committee reviews
fishery stock assessments and helps the Council evaluate the available scientific information. For
bottomfish management, advice is provided to the PFMC by the Groundfish Management Team
and advisory subpanels which include representatives from fishing industries, the public, and
conservation interests (PFMC 2020). Public meetings to share scientific and management
information and receive comments from stakeholders are also an essential step in the
6
management process. Working through the PFMC process, the WDFW and its partners set
annual harvest guidelines for Black Rockfish for the Washington coastal stock.
Historically, coastal Black Rockfish have been harvested by a variety of methods
including bottom trawl, set-line, trolling, and rod-and-reel (Parker et al. 2000). In 1999, the
WDFW managers closed commercial harvest for all gear types in State waters (0-3 miles
offshore) to increase year-round recreational fishing opportunities for bottomfish (Wallace,
Cheng, & Tsou 2008; Cope et al. 2016). Since then, the recreational fleet, which is comprised of
chartered vessels and independently owned boats, has been responsible for all targeted Black
Rockfish landings. Black Rockfish are managed as part of a rockfish species complex. Coastal
rockfish recreational fishing opportunities were available year-round in Washington until 2017
when the season was truncated to a March through October opening (WDFW 2017). Prior to
2017, managers adjusted the daily bag limits (see Glossary) as a management tool to stay within
annual harvest guidelines. From 1961-1991 recreational harvesters could retain 15 rockfish per
day. This bag limit decreased to 12 fish per day from 1992-1994 and decreased again to 10 fish
per day from 1995-2015, with additional restrictions on the retention of Canary and Yelloweye
rockfish. To help limit encounters of Yelloweye and Canary Rockfish, managers implemented
fishing depth restrictions in 2006. Marine Area 4-Neah Bay (see Glossary), east of the BonillaTatoosh Line, has been limited to 6 rockfish fish per day since 2010, while the western portion of
Marine Area 4-Neah Bay has had a daily bag limit of 10 from 1995 through 2017. In 2017, in
addition to seasonal restrictions, the bag limit was reduced to 7 rockfish (with a limit to the
number of Canary Rockfish retained until 2019) in Marine Areas 1-4 west of the Bonilla-Tatoosh
line and continued the prohibition of Yelloweye Rockfish retention. There are no size restrictions
for the retention of Black Rockfish (WDFW 2021).
7
In 2020, an estimated 137.75 metric tons of Black Rockfish were landed by the
recreational fleet, which was much lower than the past 5-year average of 281.47 metric tons per
year (RecFin 2021). Reduced harvest during 2020 can be attributed to the closure of the coastal
recreational fishery by state managers on March 25, 2020, because of the coronavirus pandemic.
Recreational harvest was re-opened for Black Rockfish on May 26, 2020, but charter vessels
were capacity limited, the halibut fishery opening was delayed, and several ports deferred
opening or have yet to re-open. The Black Rockfish recreational fishery continues to be
extremely popular and must be actively managed to ensure annual harvest guidelines are not
exceeded.
Fisheries with regulations influence species distributions (Schroeder & Love 2002; Currie
et al. 2019). In Washington, the recreational fleet accesses coastal fisheries from a limited
number of primary ports: Neah Bay, La Push, Westport, and Ilwaco. The location of these ports
influences rockfish harvest because of accessibility to rockfish habitat. In addition to fishery
removals, spatial and temporal distributions of Black Rockfish are influenced by species specific
life-history parameters and habitat requirements (Love, Yoklavich, & Thorsteinson 2002;
Johnson et al. 2003; Miller & Shanks 2004). Researchers developing a survey design for Black
Rockfish must consider fisheries impacts when determining spatial sampling coverage, in
addition to species-specific biology, life-history characteristics, and movement patterns.
2.2 Black Rockfish (Sebastes melanops)
2.2.1 Biology
The genus Sebastes includes at least 124 species distributed in temperate zones of the
Pacific and Atlantic oceans (STRI 2015). Black Rockfish are considered a semi-pelagic
schooling species with a range that extends from Amchitka and Kodiak Islands in Alaska to
8
Huntington Beach in Southern California (Love, Yoklavich & Thorsteinson 2002). As adults,
Black Rockfish school in the water column and are generally found above complex rocky
habitats at depths less than 40 fathoms. (Love 2011). In a habitat association study done off the
Oregon coast, researchers found that Black Rockfish showed a relationship with reef structure
and a negative correlation with sand habitats (Easton, Heppell & Hannah 2015). Juveniles of this
species are entirely pelagic, feeding largely on amphipods and copepods and shifting toward a
closer association with the bottom as they age (Miller & Shanks 2004; Studebaker & Mulligan
2009; Studebaker, Cox & Mulligan 2009). As they grow their diet expands to include
zooplankton, isopods, shrimp, octopuses, sand lance, and other fish, including juvenile rockfish
(Love, Yoklavich & Thorsteinson 2002). Age estimates for this species are a maximum of 56
years with a size maximum of 69 cm (27.2 in) and 6 kg (13.3 lb), with females growing larger
than males (Love 2011). In Washington, the state record for the largest Black Rockfish is 4.86 kg
(10.72 lb) for a fish caught in May of 2016 (WDFW 2020).
2.2.2 Life History Parameters
As a large nearshore rockfish species, the biology and life history of Black Rockfish has
been relatively well studied and documented. Like all members of the genus Sebastes, Black
Rockfish bear live young and produce hundreds to thousands of embryos in increasing quantities
as they age (Bobko & Berkeley 2004). Berkeley, Chapman, & Sogard (2004) found that older
Black Rockfish females produce embryos that grew faster and were better at resisting starvation
than embryos from younger females. Female maturity is estimated to be at 50% at about 6-8
years (25-30 cm) and females grow 3-5 cm larger than males (Love 2011; Cope et al. 2016). A
status report from WDFW documents some life history parameters for Black Rockfish off the
Washington coast but highlights unresolved problems with natural and fishery mortality
9
estimates. Natural mortality is challenging to estimate because it can be mixed up with fishing
mortality and is a very important parameter because of its effects on population dynamics
(Wallace, Cheng, & Tsou 2008). Natural mortality estimates were explored during the 2015
Black Rockfish assessment (Cope et al. 2016), but more work may need to be done to consider
how climate change impacts this species (Schwartzkopf et al. 2021; Markel & Shurin 2020).
2.2.3 Movement Patterns
Tagging studies for Black Rockfish indicate that adults tended to have small home ranges
(Parker et al. 2007; Green & Starr 2011). Parker et al. (2007) concluded that during a full year of
monitoring of tagged Black Rockfish, home ranges were 55±9 ha with no seasonal variability.
The WDFW Black Rockfish tagging program summary report for1981-2008 also concluded that
the largest proportion of tagged fish were recovered near the area of release and that there was a
declining tagged fish recovery with increasing distance from the release area (Wallace et al.
2010). For adult fish with small home ranges, and a high association with rugose rocky habitats,
only a few sampling methods can adequately capture adult fish. Additionally, understanding the
home range of a species, what they eat as adults, and general movement patterns, inform
researchers on how to conduct a survey both spatially and temporally to capture a representative
sample of the population.
2.3 WDFW Surveys and Pilot Studies
2.3.1 Coastal Black Rockfish Tagging Program (1981-2014)
In 1981, the WDFW initiated the Black Rockfish tagging program with objectives
focused on producing estimates of abundance, growth, survival, and natural mortality (Wallace
et al. 2010). Wallace et al. (2010) described how Black Rockfish were tagged and released off
10
the Washington coast in four Catch Record Areas (CRA) (see Glossary), with the program
releasing and recovering the highest number of fish (65,714 and 4,783 respectively) in CRA 2
from 1981-2007. Changes to this tagging program happened over time with large modifications
occurring in 1998, including changing the tag type from external to internal. Wallace et al.
(2010) noted that with this tag type modification recovery of tagged fish changed from voluntary
fishery recovery to only dockside fishery recovery by samplers as the tags were no longer
visible. Over the course of the program many vessels (research and recreational charter fleet)
were used as tagging platforms with the WDFW staff onboard following survey sampling
protocols. Data collected for the duration of the program included length, sex (when possible),
and fishing time. Additionally, throughout the program, anglers were always targeting Black
Rockfish although sampling locations were inconsistent and limited (Wallace et al. 2010).
Rocky habitat is not evenly distributed along the Washington coast which hinders
uniform tagging efforts. Because of this patchy habitat distribution, sampling methodology for
the WDFW tagging program changed from ‘local knowledge’ pre-2001, to a more formal
distribution of effort starting in 2001 (Wallace et al. 2010). In 2001, researchers used
georeferenced rocky habitat to create a study area divided into 2° latitudinal blocks, which were
then weighted by proportion of habitat and used to distribute tagging effort (Wallace et al. 2010).
In 2010, the WDFW expanded the traditional tagging program to include additional sites and
added a fall tagging period (WDFW Marine Fish Science Unit 2017). Changes to the program
resulted in increased tag deployments and increased geographic distribution of Black Rockfish
tags from 2010 through 2013. In 2014 a shift in funding stopped tagged Black Rockfish releases,
but dockside tag recovery work for Black Rockfish continued until 2017 (WDFW Marine Fish
Science Unit 2017).
11
The 2015 Black Rockfish assessment (Cope et al. 2016) for the Washington portion of
the stock used the WDFW Black Rockfish tagging program release data from 1986 through 2014
to calculate abundance indices to inform the SS model. In this model, the indices derived from
the tag release portion of the program were based on CPUE of anglers and was calculated by the
angler fishing minutes, by month, and by CRA.
CPUE=
𝐶𝑎𝑡𝑐ℎ (# 𝑜𝑓 𝐹𝑖𝑠ℎ)
𝐸𝑓𝑓𝑜𝑟𝑡 (𝐴𝑛𝑔𝑙𝑒𝑟 𝐹𝑖𝑠ℎ𝑖𝑛𝑔 𝑀𝑖𝑛𝑢𝑡𝑒𝑠)
Assessment authors limited the dataset to the most consistent time for tagging fishing trips
(spring), and to the area where most trips occurred (CRA 2). Cope et al. (2016) used a modified
generalized linear model (delta-GLM; Lo, Jacobson & Squire 1992) to model CPUE where the
positive catch component was modeled using the lognormal and gamma distributions. (Figure 1).
The use of CPUE data is known to have limitations, but if collected properly, standardized, and
used appropriately, CPUE data can be used to create an informative index of abundance (Shono
2008; Maunder & Punt 2004). The 2015 SS model for Washington, specifically the version using
the gamma distribution of the abundance indices, was accepted as the best available science on
the status of the stock in 2015 (Advisors 2015).
12
Figure 1. Abundance indices derived from the WDFW tagging CPUE analysis and used in the
2015 assessment of the Washington stock of Black Rockfish. The three different lines represent
the lognormal, gamma, and geometric averaged distributions used to model the positive catch
component (Cope et al. 2016).
Tagging program data used in the 2015 assessment was provided by the WDFW to the
stock assessment authors for model development. For this thesis work, during examination of the
tagging data used to create the abundance indices that was used in the 2015 Black Rockfish stock
assessment model, an error was found in the calculation of angler fishing minutes. Effort was
calculated using boat minutes, which was defined as the amount of time the boat was onsite
fishing, instead of angler minutes as stated in the 2015 assessment. The number of anglers varied
by boat and by survey. This CPUE calculation error will be discussed further in the results and
discussion sections.
Limitations of the data available from the WDFW Black Rockfish tagging program to
inform abundance indices for multiple bottomfish species, prompted managers to initiate pilot
studies to help inform the design of a new multispecies survey. One of the first components
13
managers evaluated was survey timing. Budget, ocean conditions, and vessel availability prevent
year-round survey effort, so managers initiated the Seasonal Rod-and-Reel Pilot Study (below)
to evaluate seasonal effects on bottomfish catch to help identify the optimum time of year to
sample.
2.3.2 Seasonal Rod-and-Reel Pilot Study (2010-2013)
In 2010, the WDFW initiated a pilot study with a goal of describing seasonal effects on
bottomfish catch during rod-and-reel surveys to address concerns highlighted in the 2007
scientific review of the Northern Black Rockfish assessment (Advisors 2007). Additional details
about the pilot study can be found in the unpublished WDFW report Rod-and-Reel Pilot Study of
Washington Coastal Index Areas, but pertinent information for Black Rockfish is summarized
here. Eleven areas from Cape Flattery WA to Cape Falcon OR were sampled and included the
eastern portion of Marine Area 4, inside the Strait of Juan de Fuca (Figure 2). Surveys began in
the fall of 2010 and occurred in both the spring (March-May) and fall (August-October) seasons
in 2011 and 2012 and concluded in the spring of 2013.
14
Figure 2. Coastwide survey areas for the seasonal rod-and-reel pilot study of Washington
coastal index areas (unpublished WDFW report). Map provided by Rob Davis, WDFW.
15
During this seasonal rod-and-reel pilot study, the terminal tackle (bait types) used to
catch fish were not standardized, but tackle was selected by experienced boat captains to be the
most efficient at catching pelagic rockfishes. Effort at each sampling location was based on a
“set” which was thought of as fishing time with no significant change in effort, gear, or location,
either anchored or unanchored, and could include multiple drifts (WDFW unpublished report).
Data collected for each set included the number of anglers fishing, fishing depth, active fishing
time in minutes, species caught, fork length, and tag information (if tag present). Uninjured fish
not needed for ancillary research projects were released at the site of capture, using a descending
device if warranted. Data analysis and conclusions from this pilot study had not been completed
prior to the writing of this thesis, but the raw data was provided by the WDFW for analysis.
Understanding seasonal movement of a species can inform sample timing when
developing a survey design. Movement patterns of Black Rockfish evaluated by Parker et al.
(2007) showed no seasonal variation, however, females > 39 cm spent more time outside of a
monitored area than males during the reproductive season (Nov., Jan., and Feb.) and both sexes
had the longest absences from monitored areas from April through July. Variable catch rates
based on movement should be considered when designing a survey for abundance estimates.
Optimal timing for a survey varies by target species and this pilot study helped inform WDFW
on which season had higher catch rates for the largest number of the targeted bottomfish species.
In addition to exploring optimal survey timing, WDFW also needed to define the amount of
optimal habitat available for targeted bottomfish off the Washington coast.
2.3.3 Rocky Habitat Exploration (2014-2015)
Rocky habitat has a patchy distribution along the Washington coast and complete high
precision bathymetric maps have historically been unavailable, hampering nearshore surveys for
16
rocky habitat associated species. In 2015, the WDFW developed a systematic survey grid based
on extensive habitat exploration work done in 2014 to facilitate geolocation of rocky habitat in
CRAs 1-4 off the Washington coast. The WDFW unpublished report - Washington Coastal
Survey Grid Design (Appendix A) details the extent, creation, and rational of a survey grid. This
grid system breaks the coastal nearshore into 3,000 meter2 cells for all waters inside the 30fathom contour with additional cells covering known rocky habitats that extended to 40-fathoms
(WDFW unpublished report). By using this grid system WDFW was able to designate grid cells
that included, or did not include rocky habitat, essentially quantifying the amount of rocky
habitat off the nearshore coast. Based on habitat designation by grid cell, surveys could be
stratified by habitat type (WDFW unpublished report).
Rocky habitat in the marine environment includes rocky reefs, boulders, and pinnacles.
The Olympic Coast National Marine Sanctuary explains on their website that “structure” is the
key to rocky reef habitat which create crevices or shelter, and that off the Washington coast
many reefs represent islands of rock surrounded by broad expanses of sand (National Ocean
Service 2022). Black Rockfish are associated with structure, as are many of the other bottomfish
the WDFW targeted with the Rod-and-Reel Survey (Love 2011). Identifying areas of optimal
habitat type helped narrow-down where targeted species were likely to occur off the Washington
coast, but the best way to sample these fish in these habitats still needed to be further explored to
help inform survey methods.
2.3.4 Terminal Tackle Pilot Study (2014 & 2015)
The WDFW initiated a pilot study in 2014 with the objective of identifying terminal
tackle types that would increase the diversity of bottomfish catch by a rod-and-reel survey
without reducing catch rates of Black Rockfish (WDFW unpublished report). Historically,
17
terminal tackle used during the WDFW surveys targeted semi-pelagic schooling species. For
example, the Black Rockfish tagging program used one to three single hook jigs with shrimp
flies and/or artificial worms (Wallace et al. 2010; Figure 3). The terminal tackle pilot study
focused on catch distribution and catch rate comparisons of both pelagic and demersal species by
terminal tackle type. Details about this pilot study can be found in the unpublished WDFW
report Selectivity of Terminal Tackle in Groundfish Rod-and-Reel Surveys Off the Washington
Coast, but pertinent information about this survey as it relates to catch rates of Black Rockfish
by terminal tackle type is summarized here. Comparison surveys for terminal tackle types were
done in March, April, and May of 2014 and 2015 in coastal CRAs 1-4. The methods used for the
terminal tackle surveys conducted in 2014 drastically differed from those done in 2015, making
comparisons between the two years difficult. Differences included types of terminal tackle tested
and how much time each bait type was fished. Analysis of survey data for Black Rockfish
terminal tackle preference was incomplete prior to the writing of this thesis. The WDFW
provided data from this pilot study for analysis.
18
Figure 3. Diagram of the standard “shrimp fly” tackle with two hooks that can be baited with
artificial worms or squid (left) and a diagram of the “mooching rig” (right). The mooching rig
shown here is baited with an artificial worm. (Diagrams courtesy of K. Hinton WDFW).
The effectiveness of terminal tackle used in a hook-and-line survey is entirely dependent
on the behavior of the fish, which can be influenced by many factors including currents, light
level, temperature, fish density, and prey availability (Stoner 2004). Coghlan et al. (2017) used
baited underwater video to find that fish density influenced feeding behavior and noted
behavioral differences based on the duration of baited sampling times. A study on catch rates of
largemouth bass concluded that size was the strongest predictor of capture, regardless of prey
availability (Keiling, Louison & Suski 2020). If behavioral factors are not considered, catch rates
could be based more on behavior variation than abundance trends (Stoner 2004; Kuriyama et al.
2019).
Data collected during the terminal tackle pilot study was used to evaluate changes in
catch rates of targeted bottomfish species. Following this pilot study researchers further
19
questioned whether rod-and-reel was the most efficient and effective gear to catch targeted
bottomfish species (R. Davis WDFW, personal communication). To evaluate this concern a pilot
study based on gear type was initiated by the WDFW in 2016.
2.3.5 Rod-and-Reel vs. Set-line Pilot Study (2016-2017)
From 2016 through 2017 the WDFW explored the effectiveness of set-line gear to survey
nearshore coastal populations of bottomfish (WDFW unpublished report). The primary objective
of this pilot study was to develop a survey capable of providing an index of abundance for
multiple bottomfish species including, but not limited to, Cabezon, Kelp Greenling, Lingcod
(Ophiodon elongatus), and rockfishes. Additionally, a second objective was to compare set-line
survey catch-rates and species diversity with a traditional rod-and-reel survey. Various set-line
gear configurations and hook sizes, adapted from the International Pacific Halibut Commission
(IPHC) Fishery-Independent Set-line Survey (IPHC 2017), were used during the study in an
attempt to optimize catch rates (WDFW unpublished report).
During this pilot study, a rod-and-reel vs. set-line comparison was done at 5 locations
over 5 days in 2016 and another comparison study was done at 31 locations in both spring and
fall of 2017. Detailed methods for both studies can be found in the unpublished WDFW report
Set-line Survey of Washington’s Nearshore Groundfish Species: Method Development and Gear
Selectivity. Results showed that set-line methods developed for both comparison surveys worked
well for highly rugose rocky habitats off the Washington coast. The WDFW also concluded that
species composition for set-line gear was significantly different from the traditional rod-and-reel
surveys. Normalized fishing effort by gear type showed higher set-line catch rates of demersal
species including Cabezon, Big Skate, and Buffalo Sculpin when compared to the rod-and-reel
surveys which were dominated by pelagic schooling rockfishes including a significant difference
20
for Black and Yellowtail rockfish. Black Rockfish catch was significantly higher with rod-andreel gear in both comparison studies (Figures 4 & 5) however, fish caught with set-line gear had
greater mean lengths than those caught with rod-and-reel gear. Results showed that the two
survey gear types are not interchangeable but could be considered complementary if attempting
to survey both pelagic and demersal species (WDFW unpublished report). Huntington and
Watson (2017) also suggest that a combination of a rod-and-reel and set-line surveys would be
robust as an abundance estimation program for diverse bottomfish species and would be a
preferable survey strategy.
Figure 4. Bar graph from an unpublished WDFW report showing the number of individuals
caught per comparison site (general location) of species comprising over 1 % of the total catch
of the gear comparison study in 2016. Black Rockfish catch was plotted on a different y-axis
scale. Error bars represent one standard error of the mean. (WDFW unpublished report)
21
Figure 5. Bar graph from the WDFW unpublished report showing the number of individuals
caught per 3-kilometer squared grid cell of species comprising over 1 percent of the total catch
of the gear comparison study. Black Rockfish catch was plotted on a different y-axis scale. Error
bars represent one standard error of the mean. * Indicates P<0.05 and ** indicates P<0.01
from a Wilcoxon rank-sum test (WDFW unpublished report).
2.4 Nearshore Survey Considerations and use of CPUE for Abundance Indices
A wide variety of survey gear types are employed to collect fishery-independent data for
use in calculating fish population abundance estimates. Surveys utilizing trawl gear,
hydroacoustic, visual (e.g., SCUBA, remotely operated vehicles, submersibles), and hook-andline gear each have advantages and limitations (Koeller 1991; Jagielo et al. 2003; Jones et al.
2012; Keller, Wallace & Methot 2017; Pacunski et al. 2020). For semi-pelagic Black Rockfish
off the coast of Washington, several of these survey methods are unviable, due to the species
high association with rocky habitat and wide depth distribution. Bottom trawl surveys are not
effective in sampling rocky habitats because of gear damage, snags, and habitat impacts.
Hydroacoustic surveys also have limitations for assessing this species because of the limited
ability of the sonar technology to discern targeted fish from the bottom echo return (Parker et al.
22
2007; Patel, Pedersen & Ona 2009). Visual surveys using remotely operated vehicles or
submarines can sample rocky habitats and collect species information, however biological
sampling is lost and behavioral responses to the vehicle may bias results (Stoner et al. 2008;
Laidig, Krigsman & Yoklavich 2013; Pacunski et al. 2016). Surveys using SCUBA have depth
and time limitations that are too great for fully assessing coastal rockfish populations. Hook-andline surveys, if designed appropriately, appear be the best available gear type to sample and
assess coastal Black Rockfish populations.
Over time, research has shed light on the many strengths and weaknesses of rod-and-reel
surveys (Harms, Wallace & Stewart, 2010; Kuriyama et al. 2019). Although this gear type is
well suited for sampling rocky reef habitats and allows for the collection of biological data, other
factors need to be considered. Survey locations, hook saturation, and competition can all bias
abundance estimates, but these may be mitigated through sampling design (Harley, Myers &
Dunn 2001; Kuriyama et al. 2019). Kuriyama et al. (2019) recommended “density-based
sampling” (p. 193) for rod-and-reel surveys which is a survey design that samples areas of higher
probabilities of high densities but concluded that hook-saturation leading to hyperstability, and
competition might be an issue. Hyperstability in rod-and-reel surveys was addressed in
Kuriyama’s dissertation where simulations were used to evaluate CPUE abundance trends for
patchily distributed fish. They concluded that hyperstability was high for surveys of patchy
distribution and effected the relationship between abundance and CPUE. However, hyperstability
decreased with preferential sampling surveys (density-based sampling) and increased sample size
(Kuriyama 2018). They also concluded that changes to CPUE were detectable with preferential
sampling surveys and more sample sites.
A hook-and-line survey conducted in southern California for shelf rockfish since 2003,
23
through a cooperative effort between NOAA Fisheries, Pacific States Marine Fisheries
Commission, and the sportfishing industry, has been used for 14 stock assessments since 2019
(Northwest Fisheries Science Center 2021). An analysis performed on those data in 2010
concluded that when standardized using a Bayesian Generalized Linear Model to account for the
effects of location, fishing time, number of anglers and other statistically significant effects, the
data were adequate for calculating abundance indices (Harms, Wallace & Stewart 2010). Current
sampling protocols for this survey specify that 3 anglers make 5 deployments of a five-hook
sampling rig at each of 200 fixed stations in the Southern California Bight to target multiple
rockfish species (Harms, Benante & Barnhart 2008; Northwest Fisheries Science Center 2021).
Hook saturation and competition was a concern for this survey and researchers stated that they
addressed these problems by increasing the number of drops from 3 to 5 with each angler having
5 hooks per line. They concluded that this would reduce the frequency of returning saturated gear
(Harms, Benante & Barnhart 2008). Information gained for other research and surveys targeting
bottomfish can help to inform survey design. Consequently, the WDFW Rod-and-Reel Survey
was evaluated with these considerations in mind.
2.5 WDFW Coastal Rod-and-Reel Survey (2019-2021)
The coastal nearshore Rod-and-Reel Survey was developed to address the need for a
fishery-independent method to inform multiple bottomfish assessments for Washington stocks.
The following is a synopsis of the survey design, and more specific details can be found in the
WDFW unpublished report Washington Department of Fish and Wildlife’s 2019 Coastal Rodand-Reel Survey Design. The Rod-and-Reel Survey is designed to occur in spring months
(March-May) at 125 index stations in CRAs 1-4 off the Washington coast annually. Index
stations were selected using a survey grid schema. The first step in the selection process was to
24
overlay a one-kilometer squared grid on all waters to 40-fathoms and to identify grid squares
(cells) that contained rocky reef habitat. Each cell was assigned a unique number sequentially
from south to north and west to east. From these grid cells, 114 cells were random-systematically
selected. Selection was done by starting at a random cell and systematically choosing cells at a
preset interval to produce the total number that was feasible to sample in a single survey season.
Of the original selected cells, 7 were eliminated due to hazards or other known location
problems, and 18 were intentionally added to aid in distributing effort more evenly across habitat
by CRA. The final number of one-kilometer squared grid cells selected for the Rod-and-Reel
Survey was 125. Within each cell a single GPS position based on the center of the known rocky
habitat was identified and used as the index station location.
Recreational charter vessels are contracted by WDFW to conduct the survey each year.
Survey crews consist of five hired anglers and 3-4 WDFW scientific staff. To complete the full
annual survey, approximately 21 fishing day trips are required. Each survey day, the lead
biologist and vessel captain work closely to ensure sampling is started at the index station with
boat movement tracked using Coastal ExplorerTM navigation software. Near each index station a
model SBE 19+ V2 conductivity, temperature, and depth instrument (CTD) fitted with a SBE 43
dissolved oxygen sensor and Cyclops-7 fluorescence sensor is deployed to profile the water
column. On board the vessel, data are electronically collected on tablets using input interfaces
constructed in iForms and Microsoft Access then exported to a Microsoft Access database where
it is error checked. Customized queries are used to check for outliers and data isn’t finalized until
all quality control checks are completed at the end of the annual survey, when it is uploaded to
the master survey database.
Standardized fishing rods, reels, and terminal tackle (2 shrimp flies tied on a pre-tied
25
leader above a dropper weight) are supplied by WDFW to each angler. Each angler is also
equipped with a stopwatch, which they start at the beginning of each drift, pause when dealing
with landed catch or servicing gear, and stop at the end of the drift. A “drift” is defined as any
uninterrupted time span that is spent fishing, beginning when the first angler’s hook enters the
water and ending when the last angler’s hook leaves the water for any reason. Drifts last
approximately 8 minutes (R. Davis WDFW, personal communication). At each index station,
five anglers fish from the vessel for a total of 4 drifts. Anglers target rockfish that typically
school above rockpiles. Landed catch is associated to the angler, identified by species, measured,
sexed (when possible), scanned for tags, and recorded by index station, depth, and drift number.
Catch is measured as fork length in centimeters. Fish that are brought out of the water, but lost at
the rail, are documented, and noted as ‘drop off’. Some select bottomfish species receive tags or
are retained for age structure sampling. Additional data collected for each day and index station
includes environmental and ocean conditions, personnel/boat details, location of angler position
on the boat, tag release and recovery, and gear loss. The Rod-and-Reel Survey has been
conducted annually since 2019, however not all index stations were surveyed in 2020 because of
the coronavirus pandemic.
2.6 Considerations
This thesis asks if the WDFW Rod-and-Reel Survey has the ability to detect changes in
coastal Black Rockfish population abundance. The above literature review provides the
background information for an evaluation of the design. Historic and current exploitation,
biological characteristics, and information gained from pilot studies, assessment reviews, and
research were considered during the evaluation and will be further discussed in the conclusion
and discussion section.
26
3. Methods
Informative surveys are designed based on target species specific considerations,
consider sampling gear limitations, and attempt to mitigate for bias. Therefore, the WDFW Rodand-Reel Survey design must be considered from multiple perspectives to determine whether it is
able to detect abundance changes in the Black Rockfish stock. Evaluation of the WDFW Rodand-Reel Survey design focused on 4 components: location, terminal tackle and gear type,
timing, and effort. All analyses described below were performed in R, the statistical computing
language (R Core Team 2021) unless otherwise noted.
3.1 Survey Location
The WDFW Rod-and-Reel Survey limits sampling effort to 125 index stations, of which
66 fall into CRA2 off the Washington coast. To evaluate whether sampling at these selected
index stations produces representative abundance indices for Black Rockfish, a comparison of
two abundance indices was done. The first abundance indices were created by stock assessment
authors from tagging data collected during the WDFW Black Rockfish tagging project (Wallace
et al. 2010) and used in the 2015 Washington Black Rockfish SS model (Cope et al. 2016).
These indices were compared to a second set of abundance indices created by limiting the same
data set to only surveys conducted near index stations. To determine which survey events to
include, all WDFW Black Rockfish tagging program surveys from 1986-2014 and the index
station locations were mapped using ArcGIS Pro (ESRI 2019) (Figure 6). Spring (February-July)
surveys, i.e., tagging or fishing events, that fell within a 500-meter buffer of an index station in
CRA2 were selected for inclusion in the calculation of the index station abundance indices. A
total of 435 surveys met these criteria. Because selected surveys were very sparse before 1998,
and post-1998 no major changes happened in the WDFW Black Rockfish tagging program, only
27
surveys from years 1998-2014 were used in the calculation of the index station abundance
indices. Ultimately, 400 surveys were included in the abundance indices CPUE calculation at
index stations.
28
Figure 6. Locations of the historic Black Rockfish tagging program spring surveys (grey dots)
off the Washington coast in CRA2. Index Stations are displayed with a 500-meter buffer (green
circles), and most are obscured by survey dots. Tagging surveys that fell within the index station
buffer are displayed as orange dots and were included in the index station abundance indices
CPUE analysis.
29
Index station abundance indices were calculated using catch of Black Rockfish per boat1
minute with covariates month and CRA and modeled using a delta-GLM approach (Lo, Jacobson
& Squire 1992). This approach was identical to the that used in the 2015 Washington Black
Rockfish assessment (Cope et al. 2016) and for ease of comparison was plotted in the same way
as in the assessment report to show lognormal, gamma, and geometric average distributions
(Figure 7).
Figure 7. Abundance indices for index station CPUE for 1998-2014 with lognormal, gamma,
and geometric average plotted for ease of comparison to the abundance indices for the WDFW
tagging CPUE analysis from the 2015 assessment of the Washington Black Rockfish stock.
Prior to analysis, the tagging CPUE abundance indices used in the 2015 assessment were
truncated to only surveys conducted from 1998-2014 for comparison to the index station
abundance indices. To determine if the index station abundance indices induced large changes in
the 2015 stock synthesis assessment model, the SS model was run using index station abundance
indices.
1
Effort was calculated using boat minutes in the original stock assessment, not angler minutes. Figure labels were
left as angler minute to avoid confusion during comparison to the original figures (Cope et al. 2016).
30
The base model used in the 2015 assessment used SS version V3.24U-safe (Methot 2014;
Cope et al. 2016). This version was also used for all new model runs which included: 1)
truncation of original tagging abundance indices to years 1998-2014 (with the removal of length
composition data prior to 1998), 2) replacing the original indices with the index station
abundance indices for years 1998-2014 (including only length composition data from index
station surveys), and 3) removal of the entire abundance indices (including all tagging length
composition data). The gamma distributions for each set of indices were used to remain
consistent with the 2015 assessment model. The model run that included indices from index
stations was not re-tuned, and the relative coefficients of variations (CVs) used with the index
station abundance indices were not adjusted from the original model, to see the impacts of the
index station CPUE abundance indices on the model output. When the entire tagging indices
were removed from the model, the model was re-tuned using the methods of Francis (2011). One
issue to note when comparing model run output was the addition of extra variance to the
Washington CPUE indices in the 2015 SS model, reducing the overall influence of the indices
(Cope et al. 2016). In addition to considering model impacts of the index station indices, the
relationship between the two indices was also evaluated.
3.2 Survey Terminal Tackle and Gear Type
In the spring of 2014 and 2015 the WDFW conducted a pilot study to determine which
type of terminal tackle had the highest catch rates of pelagic schooling species and other
recreationally targeted benthic bottomfish (WDFW unpublished report). The survey
methodology differed by year and therefore for this analysis each year was analyzed separately.
The CPUE of Black Rockfish by angler minute by terminal tackle type for each year of the study
was calculated. A Kruskal-Wallis rank sum test with a post hoc Dunn’s test (Dunn 1964) was
31
used to determine which terminal tackle types were significantly different for Black Rockfish.
Concerns over species specific tackle selectivity coupled with rod-and-reel survey
standardization issues prompted the initiation of a pilot study to compare set-line vs rod-and-reel
survey gear methods (WDFW unpublished report). From 2016 to 2017 the WDFW examined the
effectiveness of set-line gear to capture bottomfish in rocky habitats. Analyses using the
Wilcoxon rank-sum test, Kruskal Wallis H test, and a two-way analysis of similarity (ANOSIM)
were done by the WDFW to determine catch rates by species based on gear type and season
(WDFW unpublished report). These analyses performed by WDFW were abridged for only
Black Rockfish by gear type and presented in this thesis.
3.3 Survey Timing
Determining the catchability of multiple bottomfish species of ecological and fishery
importance was a primary research objective during the seasonal rod-and-reel pilot study
conducted from 2010 through 2013 (WDFW unpublished report). During this study 10 areas
were surveyed in the spring and fall for three years. To determine if Black Rockfish catchability
was higher by season, catch of fish by angler minute, by area, by season, and by year was
evaluated using a linear mixed effects model which allowed for nested random effects. For this
analysis CPUE at each location was compared by season over three years. The package lme4 was
used in R for this modeling (Bates et al. 2015).
3.4 Survey Effort
The Rod-and-Reel Survey design relies on 4 replicate passes or “drifts” at each index
station for the station survey to be complete (WDFW unpublished report). Data from the Rodand-Reel Surveys completed in 2019, 2020, and 2021 were used to assess effort at each index
32
station. A t-test was used to evaluate the difference in CPUE estimates by the number of drifts
conducted by location for three survey years.
To evaluate if the number of index stations surveyed each year in the Rod-and-Reel
Survey was high enough to be able to detect changes in Black Rockfish abundance a power
analysis was done. A power analysis for a two-sample Z test was done to evaluate the risk of
making a type II error (failure to reject the null hypothesis, when it is in fact false) using an
online power calculator2. Rod-and-Reel Survey data collected in 2019 and 2021 were used for
this analysis. Data from 2020 was excluded because not all index stations were surveyed due to
the coronavirus pandemic.
2
The online power calculator used for this analysis is available at https://ytliu0.github.io/Stat_Med/ power2.html.
33
4. Results
The WDFW nearshore coastal Rod-and-Reel Survey was evaluated to determine whether
it can detect changes in Black Rockfish population abundance by analysis of 1) sampling
locations, 2) terminal tackle and gear efficacy, 3) survey timing, and 4) survey effort. Each of
these components was analyzed using an approach based on the type of available data from
historic surveys, pilot studies, and the WDFW Rod-and Reel Survey. All data were provided by
the WDFW.
4.1 Survey Locations
Abundance indices derived from data collected during the WDFW Black Rockfish
tagging program throughout CRA 2 was compared to abundance indices calculated from surveys
conducted at Rod-and-Reel Survey index stations within CRA 2, for the period 1998-2014.
Overall, both abundance indices show similar trends over time (Figure 8). The single largest
divergence, in 2014, may be attributed to the lower number of surveys completed near index
stations CRA2 as the survey effort that year expanded and was distributed differently along the
coast. Comparing individual CPUE estimates from each year to each other showed that the point
estimates from all survey locations tended to be higher than those for surveys only conducted at
index stations for Black Rockfish (Figure 9).
34
Figure 8. The gamma distribution of abundance indices calculated from all surveys and from
surveys only conducted at index stations.
Figure 9. A scatter plot (each dot represents one year from 1998-2014) showing a comparison of
CPUE abundance indices (using the gamma distributions) of all surveys with surveys at index
stations. The blue line is a one-to-one line.
35
While the two abundance indices followed the same general abundance trends, there
could have been a significant difference when informing the stock assessment model. To
determine this, the two indices were compared using output from the SS model used in the 2015
Washington Black Rockfish assessment (Methot & Wetzel 2013; Cope et al. 2016). Differences
in model output were compared for 4 model runs. Model runs included 1) the base model
originally used in the 2015 WA assessment, 2) truncating the tagging CPUE abundance indices
to the years 1998-2014 with removal of all length composition data pre-1998, 3) replacing the
tagging CPUE abundance indices with the CPUE indices from index stations and including only
length composition data from index station surveys, 4) removal of the abundance indices from
the base model. No tuning or modifications of the coefficient of variation (CV) error were done
with the truncated model run or the index station abundance indices model run. The model run
with no abundance indices was tuned using Francis weighting as in the 2015 assessment (Cope et
al. 2016). The stock synthesis results for spawning output and stock status are shown in Figures
10 & 11.
36
Figure 10. Time series of spawning output of Black Rockfish in Washington for 4 model runs
with ~95% asymptotic intervals. The dark blue circles show the original model run from the
2015 assessment, the light blue triangles show the same data truncated to the years 1998-2014
with pre-1998 length compositions removed, yellow plus signs show the model run with the
indices from the Index Stations and length composition data only from Index Stations and the red
x line shows the model run with no indices included.
37
Figure 11. Time series of stock status, shown as fraction of unfished, of Black Rockfish in
Washington for 4 model runs with ~95% asymptotic intervals. The dark blue circles show the
original model run from the 2015 assessment, the light blue triangles show the same data
truncated to the years 1998-2014 with pre-1998 length compositions removed, yellow plus signs
show the model run with the indices from the Index Stations and length composition data only
from Index Stations and the red cross line shows the model run with no indices included.
Stock synthesis model diagnostics were used to evaluate impacts the indices had on the
model output. Reference points used for comparison of model runs show a very small change in
unfished spawning biomass and the status of stock depletion remained unchanged. A summary of
the model diagnostics is shown in Table 1.
38
Table 1. Table of reference points for a comparison between the 4 Black Rockfish assessment
models to show the effects of the indices. The Base Model reference points have no modification
and are from the 2015 Washington Black Rockfish accepted stock synthesis model (Cope et al.
2016).
Unfished Spawning Biomass (mt)
Unfished age 3+ biomass (mt)
Spawning Biomass 2015
Unfished recruitment (R0)
Depletion 2015
Base
Model
1356
9119
582
2102
0.43
Base
1998-2014
1365
9121
582
2096
0.43
Index
1998-2014
1401
9226
601
2102
0.43
Base
No Indices
1422
9076
584
1950
0.41
Methodology used to select index stations for the Rod-and-Reel Survey in CRA 2 were also used
for selecting index stations in the other three coastal marine areas (CRA 1, 3, and 4). Based on
the above analysis, index stations in CRA2 show representative abundance trends of the Black
Rockfish adult population and it can be inferred that index stations selected in each of the other
three coastal CRA would be representative.
4.2 Survey Terminal Tackle and Gear Type
The CPUE of Black Rockfish by angler minute by terminal tackle type for each year of
the WDFW terminal tackle pilot study was explored. The 2014 pilot study data histogram of
CPUE by terminal tackle type (Figure 12) and box plots of CPUE by gear type (Figure 13) were
examined. Unfortunately, because large differences in consistent terminal tackle testing in 2014
(different areas fished with different bait type, different fishing times per bait type) were
discovered the dataset was not used to statistically evaluate terminal tackle preference by Black
Rockfish, but shrimp flies were the preferred terminal tackle by anglers for catching Black
Rockfish. Lessons learned from the 2014 pilot study informed the 2015 survey methodology.
39
Figure 12. CPUE by terminal tackle gear type for survey year 2014.
Figure 13. CPUE by terminal tackle gear type used in the 2014 WDFW pilot study.
40
Data from the 2015 survey were explored through a histogram (Figure 14) and box plots
of CPUE by gear type (Figure 15). Survey methodology in 2015 included only three terminal
tackle gear types: shrimp flies, mooching rigs with artificial bait, and 2 shrimp flies baited with
squid. For 2015, there was a significant difference in CPUE based on terminal tackle type
(Kruskal-Wallis, H2=113.1, P < 0.001). Post-hoc comparison using a Dunn’s test showed that
CPUE using shrimp flies were significantly different than both mooching rigs with artificial bait
(P < 0.001) and shrimp flies with squid (P < 0.001). There was no significant difference between
CPUE with mooching rigs with artificial bait compared to shrimp flies with squid (P = 0.807).
Figure 14. CPUE by terminal tackle gear type for survey year 2015.
41
Figure 15. CPUE by terminal tackle bait type used in the 2015 WDFW pilot study.
Based on the analysis of Black Rockfish CPUE data from the WDFW terminal tackle pilot study,
the most effective terminal tackle for catching Black Rockfish was shrimp flies.
From 2015 to 2017 the WDFW conducted a series of studies to examine the effectiveness
of set-line gear to capture bottomfish in rocky habitats. From analysis done on these survey data,
WDFW managers concluded highly valuable and abundant Black Rockfish are better monitored
with rod-and-reel gear while set-line gear encounters substantially more demersal species
(WDFW unpublished data). A cost analysis was also provided by the WDFW for the comparison
pilot study. The 5-day comparison study cost approximately $13K to complete for the rod-andreel component and $36K for the set-line effort (R. Davis WDFW, personal communication).
4.3 Survey Timing
Data from the seasonal rod-and-reel pilot study were explored through a histogram of
42
CPUE by season (Figure 16). Seasonal timing based on CPUE of Black Rockfish was analyzed
using a linear mixed-effects model (Laird & Ware 1982). This model was chosen because it
works well for data that are collected and summarized in groups and have a symmetrical
distribution. Analysis evaluated the relationship between CPUE of surveys done in the spring vs.
fall at 10 locations over three years. The season in which the surveys were conducted did not
have a significant effect on Black Rockfish CPUE (fixed effect of season P=0.72, accounting for
the random effect of year).
Figure 16. Histogram showing CPUE from the fall and spring pilot study surveys from 10
locations over 3 years combined on a natural log scale.
Seasonal survey timing decisions for the Rod-and-Reel Survey were highly dependent on
charter vessel availability, ocean conditions, and financial feasibility (T. Tsou WDFW, personal
communication). In general, charter vessel operators cater to the recreational fishing fleet and
43
make their income March through October by taking recreational harvesters out to fish. The
WDFW contracts with multiple charter vessel operators at the beginning of the bottomfish
fishing season to conduct the survey, as the weather is variable, and the demand is lower in the
early spring (R. Davis WDFW, personal communication).
4.4 Survey Effort
The effect of survey effort per index station, with effort defined as the number of drifts
conducted (4 vs. 3), was evaluated for data collected during the 2019, 2020, and 2021 Rod-andReel Surveys. Survey data was explored by year and by the number of sampling drifts preformed
at each index station (Figures 17). Current survey design requires 4 drifts across the habitat at
each index station. An analysis of the number of drifts at index stations for each survey year was
done using a t-test of the difference between the CPUE based on 3 drifts and the CPUE based on
4 drifts. Results showed that the difference between CPUE based on 3 or 4 drifts was
significantly different for all survey years 2019, 2020, and 2021 (t124 = -3.03, P = 0.001, t45= 3.70, P < 0.001, and t123=-3.57, P < 0.001, respectively). Average CPUE by drift was plotted to
see if there were trends in the data by year (Figure 18). This visual representation of average
CPUE for each drift shows that additional drifts yielded less and less CPUE.
44
Figure 17. CPUE based on 4 drifts and 3 drifts (not independent) at each index location for the
2019, 2020, and 2021 Rod-and-Reel Survey.
45
Figure 18. Average CPUE calculated for each drift at Index Stations by survey year.
To evaluate if the number of index stations surveyed each year in the Rod-and-Reel
Survey was high enough to be able to detect changes in Black Rockfish abundance a power
analysis was done. Rod-and-Reel Survey data collected in 2019 and 2021 were used for this
analysis. Data from 2020 was excluded because the survey was not completed due to the
coronavirus pandemic. Using a power calculation for a two-sample Z test, sample means from
the 2019 and 2021 surveys, a difference in the mean samples for the null hypothesis of zero, and
a difference of means in the alternative hypothesis of 0.05, beta was calculated, with sample size
of 125 (# of index stations) and standard error of 0.0172 for a two-sided test. From these
calculations, the power of the hypothesis test to detect a difference of 0.05 in CPUE was 89%.
Alternative differences were explored for the current survey methodology which uses 4 drifts per
index station and for the first 3 drifts at each index station. Results from this exploration is
shown in Table 2.
46
Table 2. Statistical power (%) with different CPUE effect sizes (0.02-0.05) by number of drifts at
the index station for a two-sided two-sample Z test.
3 Drifts
0.02
21%
Effect Size
0.03
0.04
42%
64%
4 Drifts
25%
48%
72%
0.05
83%
Sample
Size
125
Standard
Error
0.0172
SD
2019
0.11306
SD
2021
0.15509
89%
125
0.0157
0.103099
0.141542
47
5. Discussion and Conclusions
Designing a survey capable of detecting a change in relative population abundance
requires consideration of multiple species-specific factors. In 2019, the WDFW implemented the
standardized coastal nearshore Rod-and-Reel Survey which samples 125 index stations annually.
Analysis of abundance indices calculated from data from the historic WDFW Black Rockfish
tagging program data found no difference between the complete survey data series and surveys
conducted only at index stations. Replacing the tagging abundance indices with the indices from
index stations in the 2015 SS assessment model had minimal impacts on the stock assessment,
with no change to the overall status of stock depletion. However, extra variance was added to the
indices by the 2015 stock assessment authors, reducing the impact of the indices to the model.
These results indicate that the index station locations selected in CRA 2 for the Rod-and-Reel
Survey were representative of the population. Index station selection for the Rod-and-Reel
Survey was conducted in the same way for all 4 coastal CRAs, using a grid system over rocky
habitat. Selecting locations using the grid system reduced the chance of sampling only easily
accessible habitat based on departure port. If index stations selected in CRA 2 represent the
population, the same should hold true for CRAs 1, 3, and 4 because they were selected in the
same methods.
During exploration of the WDFW Black Rockfish tagging program data that was used in
the 2015 Washington Black Rockfish SS model, a notable error was discovered. Assessment
authors stated that catches of Black Rockfish were recorded by angler minute, month, and CRA
(then referred to as ‘punch card area’) and abundance indices were calculated from this data for
1986-2014 (Cope et al. 2016). However, data referred to as “angler minute” was actually the
amount of time a boat was at the location for each sampling event with anglers fishing. The
48
number of anglers fishing from the boat was not included in the data calculation and varied from
2 to 21 anglers. This oversight impacted CPUE estimates, including the CPUE scale and should
be corrected if the data series were to be used in the future assessments.
Species-specific characteristics of Black Rockfish, considered a semi-pelagic schooling
species with high association with rocky habitat, must be considered when selecting survey gear
and terminal tackle. Pilot study data analyzed by the WDFW confirmed that Black Rockfish
were better, and more cost effectively, sampled using rod-and-reel gear for fishing shoreward of
40 fathoms along the Washington coast. Results from analysis of data from the 2015 terminal
tackle pilot study showed a significant increase in Black Rockfish CPUE using shrimp flies when
compared to the 2 other terminal tackle types tested. The 2014 pilot study tested 5 different
terminal tackle types for Black Rockfish CPUE, with shrimp flies showing the highest catch
rates but results were difficult to interpret statistically.
Gear type and terminal tackle used for the WDFW coastal Rod-and-Reel Survey are
similar to those used in the Southern California Bight for the Shelf Rockfish Rod-and-Reel
Survey, and that survey data has been used to create abundance indices for use in multiple stock
assessment models (Harms, Benante & Barnhart 2008; Harms, Wallace & Stewart 2010). Based
on the information presented here, the gear type and terminal tackle specified in the Rod-andReel Survey methodology were well suited for sampling Black Rockfish despite modifications to
the survey design to increase catch of other targeted bottomfish species. Surveys using rod-andreel methods have an added benefit of the ability to collect biological samples. Length frequency
and age information collected during surveys provide valuable information on growth that can
also be used to inform assessment models.
The timing of the spring (March through May) Rod-and-Reel Survey is not solely based
49
on Black Rockfish catchability. Other factors contributing to designating this time frame
included survey vessel availability, seasonal weather conditions, and cost. Through the analysis
of data from a pilot study sampling Black Rockfish in the spring and fall, which showed no
significant difference in CPUE based on season, it can be concluded that spring is a suitable
season for sampling Black Rockfish from both a biological and a cost/feasibility perspective.
The number of index stations included in the survey were initially based on funding and
survey capabilities (WDFW unpublished report). To determine whether the number of stations
was appropriate for detecting a change in CPUE abundance, a power analysis was completed.
The number of index stations included in the design of the survey (125) is adequate for detecting
a change in abundance of at least 0.05 CPUE with statistical power of at least 80%, for both 3
and 4 drifts across the station. This effect size (0.05 CPUE) is sufficient to detect a difference of
about half a standard deviation (of the CPUE data from index stations, see Table 2).
To determine if the number of drifts conducted at index stations during the Rod-and-Reel
Survey are appropriate for capturing a representative sample of Black Rockfish, CPUE by drift
was evaluated. Analysis showed that incorporating CPUE estimates from the 4th drift at a station
lowered the overall CPUE estimate. This reduction was seen in both years with a complete Rodand-Reel Survey. The post hoc power analysis showed that both 3 and 4 drifts have a power
greater than 80% to detect a change of 0.05 in CPUE, however 4 drifts provided a slightly higher
power to detect smaller changes in CPUE, due to smaller standard error. These results suggest
that the 4th drift may not be needed for capturing a representative sample of Black Rockfish and
that 3 drifts may accomplish the same task. Hook saturation and competition have been a
concern for hook and line surveys (Harms, Benante & Barnhart 2008; Kuriyama et al. 2019;
Rodgveller, Lunsford & Fujioka 2008). However, hook saturation does not appear to be a major
50
concern based on the number of anglers (5) and drifts (4) for the Rod-and-Reel Survey, as CPUE
estimates decrease with each drift across the habitat. Additionally, this schooling species, that
swims in the water column above rocky habitat (Love 2011), may be able to out compete fish
that are more tightly associated with the bottom for descending bait. Because data is collected for
this survey by drift, analysis should be repeated after an additional full survey year has been
completed to determine if the 4th drift can be removed from the survey design allowing for the
addition of index stations.
A fishery-independent survey was recommended by the PFMC STAR panel review in
2015 to inform stock assessment models, and the WDFW responded by initiating the coastal
nearshore Rod-and-Reel Survey in 2019. By thoroughly evaluating the survey design and
answering the four research questions posed, fisheries managers now have the information
needed to affirm that the WDFW Rod-and-Reel Survey can produce representative abundance
indices for Washington’s Black Rockfish stock for incorporation into future stock assessment
models.
51
6. References
Advisors, STAR Panel. (2007). Black Rockfish (Northern) STAR Panel Report. Pacific States
Marine Fisheries Commission. Portland OR. May 21-25, 2007. 14 p.
Advisors, STAR Panel. (2015). Stock Assessment Review (STAR) Panel Report for Black
Rockfish, 20 July 2015- 24 July 2015. NMFS, Northwest Fisheries Science Center,
Newport, OR 97365. 18 p.
Bates, D., Maechler, M., Bolker, B., & Walker, S. (2015). Fitting Linear Mixed-Effects Models
Using lme4. Journal of Statistical Software, 67(1), 1-48.doi:10.18637/jss.v067.i01.
Berkeley, S. A., Chapman, C., & Sogard, S. M. (2004). Maternal age as a determinant of larval
growth and survival in a marine fish, Sebastes melanops. Ecology, 85(5), 1258-1264.
Bobko, S.J. & Berkeley, S.A. 2004. Maturity, ovarian, cycle, fecundity, and age-specific
parturition of black rockfish (Sebastes melanops). NOAA Fish. Bull. 102:418-429.
Cope, J. M., Sampson, D., Stephens, A., Key, M., Mirick, P. P., Stachura, M., Tsou, T.,
Weyland, P., Berger, A., Buell, T., Councill, E., Dick, E., Fenske, K., Monk, M., &
Rodomsky, B. (2016). Assessments of California, Oregon and Washington stocks of
black rockfish (Sebastes melanops) in 2015. Northwest Fisheries Science Center, Seattle,
Washington, USA. 1-402
Currie, J. C., Thorson, J. T., Sink, K. J., Atkinson, L. J., Fairweather, T. P., & Winker, H. (2019).
A novel approach to assess distribution trends from fisheries survey data. Fisheries
Research, 214, 98-109.
52
Environmental Systems Research Institute (ESRI). (2019). ArcGIS Pro Desktop. Redlands, CA:
Environmental Systems Research Institute.
Dunn, O.J. (1964). Multiple comparisons using rank sums. Technometrics 6:241-252.
Easton, R. R., Heppell, S. S., & Hannah, R. W. (2015). Quantification of habitat and community
relationships among nearshore temperate fishes through analysis of drop camera video.
Marine and Coastal Fisheries, 7(1), 87-102.
Francis, R. C. (2011). Data weighting in statistical fisheries stock assessment models. Canadian
Journal of Fisheries and Aquatic Sciences, 68(6), 1124-1138.
Green, K. M., & Starr, R. M. (2011). Movements of small adult black rockfish: implications for
the design of MPAs. Marine Ecology Progress Series, 436, 219-230.
Harley S.J., Myers R.A., & Dunn A. (2001). Is catch-per-unit-effort proportional to abundance?
Can. J. Fish. Aquat. Sci. 58(9): 1760–1772.
Harms, J. J. H., Benante, J. A., & Barnhart, R. M. (2008). The 2004-2007 hook and line survey
of shelf rockfish in the Southern California Bight: estimates of distribution, abundance,
and length composition. U.S. Dept. Commer., NOAA Tech. Memo. NMFS-NWFSC-95,
110 p.
Harms, J. H., Wallace, J. R., & Stewart, I. J. (2010). Analysis of fishery-independent hook and
line-based data for use in the stock assessment of bocaccio rockfish (Sebastes
paucispinis). Fisheries Research, 106(3), 298-309.
53
Huntington, B. E., & Watson, J. L. (2017). Tailoring Ecological Monitoring to Individual Marine
Reserves: Comparing Longline to Hook-and-Line Gear to Monitor Fish Species. Marine
and Coastal Fisheries, 9(1), 432-440.
International Pacific Halibut Commission (IPHC). (2017). Retrieved 10 8, 2020, from
https://iphc.int/management/science-and-research/fishery-independent-setline-surveyfiss/61-fiss-design-and-implementation
Jagielo, T., Hoffmann, A., Tagart, J., & Zimmermann, M. (2003). Demersal groundfish densities
in trawlable and untrawlable habitats off Washington: implications for the estimation of
habitat bias in trawl surveys. Fishery Bulletin, 101(3), 545-565.
Johnson, S. W., Murphy, M. L., & Csepp, D. J. (2003). Distribution, habitat, and behavior of
rockfishes, Sebastes spp., in nearshore waters of southeastern Alaska: observations from
a remotely operated vehicle. Environmental Biology of Fishes, 66(3), 259-270.
Jones, D., Wilson, C. D., De Robertis, A., Rooper, C., Weber, T., & Butler, J. L. (2012).
Evaluation of rockfish abundance in untrawlable habitat: combining acoustic and
complementary sampling tools. Fishery Bulletin. 1095 p.
Keiling, T. D., Louison, M. J., & Suski, C. D. (2020). Big, hungry fish get the lure: Size and food
availability determine capture over boldness and exploratory behaviors. Fisheries
Research, 227, 105554.
Keller, A. A., Wallace, J. R., & Methot, R. D. (2017). The Northwest Fisheries Science Center's
west coast groundfish bottom trawl survey: history, design, and description. 47 p.
54
Koeller, P. A. (1991). Approaches to improving groundfish survey abundance estimates by
controlling the variability of survey gear geometry and performance. Journal of
Northwest Atlantic Fishery Science, 11. 51-58.
Kuriyama, P. (2018). Multispecies Management and Assessment in the US West Coast
Groundfish Fishery (Doctoral Dissertation). 132 p.
Kuriyama, P. T., Branch, T. A., Hicks, A. C., Harms, J. H., & Hamel, O. S. (2019). Investigating
three sources of bias in hook-and-line surveys: survey design, gear saturation, and
multispecies interactions. Canadian Journal of Fisheries and Aquatic Sciences, 76(2),
192-207.
Laidig T.E., Krigsman L.M., & Yoklavich M.M. (2013). Reactions of fishes to two underwater
survey tools, a manned submersible and a remotely operated vehicle. Fish. Bull. 111: 54–
67.
Laird, N. M., & Ware, J. H. (1982). Random-effects models for longitudinal data. Biometrics,
963-974.
Lo, N., L. Jacobson & J. Squire. (1992). Indices of relative abundance from fish spotter
data based on delta-lognormal models. Can. J. Fish. Aquat. Sci. 49: 2515-2526.
Love, M. S., Yoklavich, M., & Thorsteinson, L. K. (2002). The rockfishes of the northeast
Pacific. Univ of California Press. 441 p.
Love, M. S. (2011). Certainly more than you want to know about the fishes of the Pacific Coast:
a postmodern experience. Really Big Press. 650 pp.
55
Magnuson-Stevens Fishery Conservation and Management Act (16 U.S.C. § 1801-1891(d)
(2014)).
Markel, R. W., & Shurin, J. B. (2020). Contrasting effects of coastal upwelling on growth and
recruitment of nearshore Pacific rockfishes (genus Sebastes). Canadian Journal of
Fisheries and Aquatic Sciences, 77(6), 950-962.
Maunder, M. N., & Punt, A. E. (2004). Standardizing catch and effort data: a review of recent
approaches. Fisheries research, 70(2-3), 141-159.
Methot, R. D. (2014). User manual for stock synthesis. Model Version 3.24s: updated November
21, 2013. NOAA Fisheries, Seattle, USA. 153 p.
Methot Jr, R.D., & Wetzel, C. R. (2013). Stock synthesis: a biological and statistical framework
for fish stock assessment and fishery management. Fisheries Research, 142, 86-99.
Miller, J. A., & Shanks, A. L. (2004). Evidence for limited larval dispersal in black rockfish
(Sebastes melanops): implications for population structure and marine-reserve
design. Canadian Journal of Fisheries and Aquatic Sciences, 61(9), 1723-1735.
National Ocean Service. (25 March 2022). Olympic Coast National Marine Sanctuary (Revised
July 26, 2017). https://olympiccoast.noaa.gov/living/habitats/rockyreef/rockyreef.html.
Northwest Fisheries Science Center. (2021, May 7). Southern California Shelf Rockfish Hook
and Line Survey. https://www.fisheries.noaa.gov/west-coast/science-data/southernCalifornia-shelf-rockfish-hook-and-line-survey.
56
Pacunski, R., Lowry, D., Hillier, L., & Blaine, J. (2016). A comparison of groundfish species
composition, abundance, and density estimates derived from a scientific bottom-trawl and
small remotely-operated vehicle for trawlable habitats. Wash. Dept. of Fish and Wildlife
Rep. #FPT 16-03. 35 p.
Pacunski, R., Lowry, D., Selleck, J., Beam, J., Hennings, A., Wright, E., Hillier, L., Palsson, W.,
& Tsou, T. (2020). Quantification of bottomfish populations, and species-specific habitat
associations, in the San Juan Islands, WA employing a remotely operated vehicle and a
systematic survey design. Report to Washington Department of Fish and Wildlife FPT,
20-07. 53 p.
Parker, S.J., Berkeley, S. A., Golden, J.T., Gunderson, D.R., Heifetz, J., Hixon, M.A., Larson,
R., Leaman, B.M., Love, M.S., Musick, J.A. & O’Connell, V.M. (2000). Management of
Pacific rockfish. Fisheries, 25(3), 22-30.
Parker, S.J., Rankin, P.S., Olson, J.M., & Hannah, R.W. (2007). Movement patterns of black
rockfish Sebastes melanops in Oregon coastal waters. In Heifetz, J., DiCosimo, J.,
Gharrett, A.J., Love, M.S., O'Connell, V.M., and Stanley, R.D. (editors), Biology,
Assessment, and Management of North Pacific Rockfishes. Alaska Sea Grant College
Program. 39-57.
Patel R., Pedersen G., & Ona E. (2009). Inferring the acoustic dead-zone volume by split-beam
echo sounder with narrow-beam transducer on a noninertial platform. J. Acoust. Soc. Am.
125(2): 698–705.
Pacific Fishery Management Council. (2015). November 2015 Briefing Book (modified 21 May
2020). Agenda Item l.3.a, Supplemental ODFW Report: Oregon Black Rockfish
57
Assessment. https://www.pcouncil.org/documents/2015/11/agenda-item-i-3-asupplemental-odfw-report-oregon-black-rockfish-stock-assessment.pdf/
Pacific Fishery Management Council. (2020, May 13). Pacific Fishery Management Council:
Managing Pacific coast Fisheries. Retrieved from: https://www.pcouncil.org/navigatingthe-council/.
R Core Team. (2021). R: A language and environment for statistical computing. R Foundation
for Statistical Computing, Vienna, Austria. https://www.R-project.org/.
RecFIN. (2021, December 1). Report generated using state agency recreational fishery data. Data
good through 2004-03 to 2021-09. https://www.recfin.org/
Rodgveller, C. J., Lunsford, C. R., & Fujioka, J. T. (2008). Evidence of hook competition in
longline surveys. Fish.Bull. 106:364-374.
Schwartzkopf, B. D., Ciannelli, L., Garza, J. C., & Heppell, S. A. (2021). Growth of juvenile
black rockfish (Sebastes melanops) during estuarine residence. Environmental Biology of
Fishes, 104(7), 851-865.
Schroeder, D. M., & Love, M. S. (2002). Recreational fishing and marine fish populations in
California. California Cooperative Oceanic Fisheries Investigations Report, 182-190.
Shono, H. (2008). Application of the Tweedie distribution to zero-catch data in CPUE
analysis. Fisheries Research, 93(1-2), 154-162.
58
Smithsonian Tropical Research Institute (STRI). (1 May 2022). Shorefishes of the Eastern
Pacific online information system. Genus: Sebastes, Rockfishes (copyright 2015-STRI).
https://biogeodb.stri.si.edu/sftep/en/thefishes/taxon/5294
Stoner, A. W. (2004). Effects of environmental variables on fish feeding ecology: implications
for the performance of baited fishing gear and stock assessment. Journal of Fish
Biology, 65(6), 1445-1471.
Stoner A.W., Ryer C.H., Parker S.J., Auster P.J., & Wakefield W.W. (2008). Evaluating the role
of fish behavior in surveys conducted with underwater vehicles. Can. J. Fish. Aquat. Sci.
65(6): 1230–1243.
Studebaker, R. S., Cox, K. N., & Mulligan, T. J. (2009). Recent and historical spatial
distributions of juvenile rockfish species in rocky intertidal tide pools, with emphasis on
black rockfish. Transactions of the American Fisheries Society, 138(3), 645-651.
Studebaker, R. S., & Mulligan, T. J. (2009). Feeding habits of young-of-the-year black and
copper rockfish in eelgrass habitats of Humboldt Bay, California. Northwestern
Naturalist, 90(1), 17-23.
Thorson, J. T., Stewart, I. J., & Punt, A. E. (2012). Development and application of an agentbased model to evaluate methods for estimating relative abundance indices for shoaling
fish such as Pacific rockfish (Sebastes spp.). ICES Journal of Marine Science, 69(4), 635647.
Wallace, F.R., Cheng, Y.W. & Tsou, T.S. (2008). Status of the Black Rockfish Resource north of
Cape Falcon, Oregon to the U.S.-Canadian border in 2006. In Pacific Fishery
59
Management Council. 2009. Appendix: Status of the Pacific Coast Groundfish Fishery
Through 2007 and Recommended Biological Catches for 2009: Stock Assessment and
Fishery Evaluation. Pacific Fishery Management Council, Portland, Oregon.
Wallace, F., Tsou, T.S., Cheng, Y.W. & Wargo, L. (2010). Summary of the coastal black
rockfish tagging program 1981-2008. State of Washington Department of Fish and
Wildlife. Technical report No. FPT 11-02. 101p.
Walters, C. (2003). Folly and fantasy in the analysis of spatial catch rate data. Canadian Journal
of Fisheries and Aquatic Sciences, 60(12), 1433-1436.
Washington Department of Fish and Wildlife: Marine Fish Science Unit. (2017). BN 2013-2015
Report to the Legislature: Status of Rockfish Research and Stock Assessment Programs.
Washington Department of Fish and Wildlife, Olympia WA. 20 p.
Washington Department of Fish and Wildlife. (2017) Washington Sport Fishing Rules: Effective
July 1, 2017-June 30, 2018. Olympia, Washington http://wdfw.wa.gov/fishing/
regulations.
Washington Department of Fish and Wildlife. (2020) Black Rockfish. Retrieved November 8,
2020 from https://wdfw.wa.gov/species-habitats/species/sebastes-melanops
Washington Department of Fish and Wildlife. (2021) Washington Sport Fishing Rules: Effective
July 1, 2020-June 30, 2021. Olympia, Washington.
https://wdfw.wa.gov/sites/default/files/publications/02264/wdfw02264.pdf
60
7. Appendices
7.1 Glossary
Bag limit- “the maximum number of fish or game animals permitted by law to be taken by one
person in a given period” (Merriam-Webster 2022).
Bottomfish- As defined by Washington State law these include Pacific cod, Pacific
tomcod, Pacific hake (or whiting), walleye pollock, all species of dabs, sole and flounders
(except Pacific halibut), lingcod, ratfish, sablefish, cabezon, greenling, buffalo sculpin,
great sculpin, red Irish lord, brown Irish lord, Pacific staghorn sculpin, wolfeel, giant
wrymouth, plainfin midshipman, all species of shark, skate, rockfish, rattail, and surf
perches (excluding shiner perch) (WAC 220-16-340).
Catch Record Areas (CRA)- also called Marine Areas (MA) and Punch Card Areas (PCA) –
Areas identified by the Washington Department of Fish and Wildlife to delineate fishing
areas throughout the marine waters of the state. Shown on the map below as in white
circles. (WDFW 2022) https://www.eregulations.com/washington/fishing/marine-arearules-definitions (Figure 19).
Fishery-dependent- “data that are collected from commercial sources (fishermen or dealer
reports) and recreational sources (individual anglers, party or charter boats)” retrieved
from Atlantic States Marine Fisheries Commission (2022A).
Fishery-independent- “data that are collected by scientists conducting resource monitoring
projects” retrieved from Atlantic States Marine Fisheries Commission (2022B).
Groundfish- “a marine fish (such as cod, haddock, pollack, or founder) of commercial
importance” (Merriam-Webster 2022).
61
Population- “a fish population is defined as a group of individuals of the same species or
subspecies that are spatially, genetically, or demographically separated from other
groups (Wells & Richmond 1995)”. (Pope, Lochmann, & Young 2010, p. 325).
Stock- “a fish stock is a management unit grouped by genetic relationship, geographic
distribution, and movement patterns” retrieved from Atlantic States Marine Fisheries
Commission (2022C).
62
Figure 19. Catch Record Areas identified by the Washington Department of Fish and Wildlife to
delineate fishing areas throughout the marine waters of the state.
63
7.2 Glossary References
Atlantic States Marine Fisheries Commission. (2022A). Fisheries-Science 101. Retrieved April
18, 2022 from http://www.asmfc.org/fisheries-science/fisheries-science101#:~:text=Fishery- dependent%
Atlantic States Marine Fisheries Commission. (2022B). Fisheries-Science 101. Retrieved April
18, 2022 from http://www.asmfc.org/fisheries-science/fisheries-science101#:~:text=Fishery- independent%
Atlantic States Marine Fisheries Commission. (2022C). Fisheries-Science 101. Retrieved April
18, 2022 from http://www.asmfc.org/fisheries-science/fisheries-science101#:~:text=stock%
Merriam-Webster. (n.d.). Bag limit. In Merriam-Webster.com dictionary. Retrieved April 18,
2022, from https://www.merriam-webster.com/dictionary/bag%20limit.
Merriam-Webster. (n.d.). Groundfish. In Merriam-Webster.com dictionary. Retrieved April
18, 2022, from https://www.merriam-webster.com/dictionary/groundfish
Pope, K. L., Lochmann, S. E., & Young, M. K. (2010). Methods for assessing fish
populations. In: Hubert, W.A, Quist, M. C., eds. Inland Fisheries Management in North
America, 3rd edition. Bethesda, MD: American Fisheries Society: 325-351., 325-351.
Wash. Admin. Code § 220-16-340. Definitions-Bottomfish (Recodified). Vol.22-07, 18 March
2017.
64
Washington Department of Fish and Wildlife. (2022) Washington Sport Fishing Rules: Effective
July 1, 2021-June 30, 2022. Olympia, Washington.
https://wdfw.wa.gov/sites/default/files/publications/02264/wdfw02264.pdf
Wells, J. V., and M. E. Richmond. 1995. Populations, metapopulations, and species
populations: what are they and who should care? Wildlife Society Bulletin 23:458–462.
65
