-
extracted text (extracttext:extracted_text)
-
PREY BIOMASS ABUNDANCE, DISTRIBUTION, AND AVAILABILITY TO
THE ENDANGERED STELLER SEA LION (EUMETOPIAS JUBATUS)
POPULATION AT
UGAMAK ISLAND, ALASKA, 1995-99.
by
Kathryn Chumbley
A Thesis
Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
May 2007
This Thesis for the Master of Environmental Studies Degree
by
Kathryn Chumbley
has been approved for
The Evergreen State College
by
______________________________________
Dr. Gerardo Chin-Leo
Member of the Faculty
________________________________________
Dr. Amy Cook
Member of the Faculty
_________________________________________
Lowell Fritz
Fishery Biologist, National Marine Fisheries Service
Alaska Fishery Science Center, National Marine Mammal Laboratory
________________________
Date
© Copyright May 2007
ABSTRACT
Prey biomass abundance, distribution, and availability to the endangered
Steller sea lion (Eumetopias jubatus) population at
Ugamak Island, Alaska, 1995-99.
Kathryn Chumbley
A 70% overall decline in the number of Steller sea lions (Eumetopias
jubatus) has occurred throughout most of Alaska since the 1970s. Data collected
from population abundance surveys conducted since 1990 indicate that the decline
continued in the Gulf of Alaska and Aleutian Islands at about 5% per year through
2000. Ugamak Island (54o 12.45 N, 164o 46.6 W), formerly the largest Steller sea
lion rookery site in Alaska, is located in the eastern Aleutian Islands in Unimak
Pass. The western stock of Steller sea lions (west of 144° W) was listed as
endangered under the Endangered Species Act in 1997 and includes Ugamak
Island. Seasonal and geographical changes in prey resource distribution,
abundance, and availability to young foraging Steller sea lions is one of the
potential causes of the population decline in the endangered western stock of
Steller sea lions.
The National Marine Mammal Laboratory in conjunction with the Alaska
Maritime National Wildlife Refuge and the University of Alaska, Fairbanks
conducted hydroacoustic and trawl surveys within the 20 nm critical habitat area
surrounding Ugamak Island, AK from 1995 through 1999 during summer and
winter seasons to assess a relative index of prey resource biomass available to
Steller sea lions. Reseach associated with this thesis included cruise participation,
hydroacoustic data analysis, and survey variance estimation using geostatistical
analysis software, Estimation of Variance (EVA), to increase accuracy of survey
variance estimates. Estimates of relative biomass collected during prey
assessment surveys ranged from 23.8 kg/m2 to 331.08 kg/m2 during summer
surveys and 300.81 kg/m2 to 1930.88 kg/m2 during winter surveys. Survey
variance ranged from 17.4 to 52.1% for summer surveys and 19.0 to 39.4% for
winter surveys. Bottom trawls, mid-water trawls, and longline surveys were also
conducted within the Ugamak Island study area.
Prey survey results show that Steller sea lion prey species, as well as species
important to those prey, were present with the 20 nm fishery management area
surrounding Ugamak Island. Results also show that prey species diversity is
higher in the Ugamak Island area than in other surveyed areas. Bottom trawls
conducted during summer months were predominately composed of walleye
pollock (Theragra chalcogramma), Pacific halibut (Hippoglossus stenoepis), rock
sole (Lepidopsetta spp.), and sculpins (Cottidae). No bottom trawls were
conducted during winter months. Mid-water trawls were conducted on an
opportunistic basis during summer and winter months and species caught included
walleye pollock, gadids, sandlance (Ammodytes hexapterus), capelin (Mallotus
villosus), hexagramids, euphausiids and other invertebrates. Longline surveys
were conducted during summer and winter months in areas unsuitable for bottom
trawling. Stomach contents collected from Pacific halibut and Pacific cod (Gadus
macrocephalus) on longline surveys were composed of gadids, small schooling
fishes, demersal fish, cephalopods, crustaceans, mollusks and benthic
invertebrates.
Ongoing satellite telemetry and foraging research shows that Ugamak Island
is an important diving and potential foraging habitat for young Steller sea lions
and that most of the satellite tagged sea lions stayed close to the island (within 20
nm) and dove to depths ranging between 0 and 50 meters of water. The results of
the satellite telemetry research demonstrate that the nearshore marine
environment around Ugamak Island is an important and potentially critical part of
the habitat of foraging young Steller sea lions. Results from this study provide
baseline data needed to explore the relationships between biomass density
changes and the effects on endangered Steller sea lions foraging at Ugamak
Island, Alaska.
Table of Contents
List of Figures ……………………………………………….……….…
List of Tables …………………………………………………………...
Acknowledgements ……………………………………………………..
INTRODUCTION………………………………………………………
Identification of the Problem……………………………………….
Description of the Study Area………………………………………
Prey Assessment Surveys at Ugamak Island…………………….…
METHODS………………………………………………………………
Acoustically derived relative biomass estimation……………….....
Estimation Variance Analysis (EVA) ………………………….…..
Transitive theory in One Dimension (1D) ………………….…
Covariograms …………………………………………………
EVA variance estimates and random sample variance estimates ….
Trawl surveys ………………………………………………………
Midwater Trawls and Neuston Tows ……………………….…
Midwater Trawls ………………………………………………
Bottom Trawls …………………………………………………
Longline Surveys ……………………………………………..……
Diet composition of predatory fish …………………………….
RESULTS ……………………………………………………………….
Acoustically derived relative biomass estimation at Ugamak Island
Trawl surveys ……………………………………………………….
Midwater trawl surveys ………………………………………..
Neuston Tows ………………………………………………….
Longline Surveys ……………………………………………………
Diet composition of predatory fish ………………………….…
DISCUSSION ……………………………………………………..….…
Acoustically derived relative biomass estimation at Ugamak Island
Estimation Variance Analysis (EVA) ………………………….
Trawl Surveys ………………………………………………………
Midwater trawl surveys ………………………………………..
Bottom Trawl surveys ………………………………………….
Standardized abundances and CPUE …………………………..
Length Frequency Distribution …………………………………
Species diversity ……………………………………………….
Longline Surveys ……………………………………………………
Steller sea lion food habits research ……………………………..…
Predatory fish stomach collections …………………………….
Scat collections …………………………………………………
Foraging distribution of Steller sea lions ………………………….
Juvenile Steller sea lion foraging research ……………………
General Recommendations …………………………………………
Literature Cited ………………………………………………………….
List of Appendices ………………………………………………………
vi
vii
viii
ix
1
1
7
9
11
13
15
18
19
21
21
22
22
27
29
30
32
32
47
47
50
50
50
60
60
61
62
62
63
64
65
66
68
69
69
71
73
73
78
79
89
List of Figures
Fig. 1. Map of Alaska and Ugamak Island Steller sea lion rookery sites… 9
Fig. 2. Map of hydroacoustic transects, midwater trawl stations, and nekton tow
stations at Ugamak Island, Alaska, 1995-99 ……………………………… 12
Fig. 3. Modified Issacs-Kidd Midwater Trawl…………………………… 24
Fig. 4. Modified Herring Trawl ………………………………………….. 25
Fig. 5. Neuston Net …………………………………………………….… 26
Fig. 6. Bottom trawls, longline sets and hydroacoustic transects at Ugamak
Island, AK 1995-99 ……………………………………………………..… 28
Fig. 7. Summary of Relative Biomass Density by Hydroacoustic Transect and
Year Ugamak Island, Alaska, Summer 1995-98 ……………………..…. 34
Fig. 8. Summary of Relative Biomass Density by Hydroacoustic Transect and
Year Ugamak Island, Alaska, Winter 1997-99 ………………………..... 35
Fig. 9. Vertical Distribution of Biomass Density at Ugamak Island, AK 1995-99
………………………………………………………………………….…. 36
Fig. 10. Survey Variance Estimation Covariograms, Summer 1995-98, Ugamak
Island, Alaska ……………………………………………………………... 39
Fig. 11. Survey Variance Estimation Covariograms, Winter 1997-99, Ugamak
Island, Alaska…………………………………………………………….. 41
Fig. 12. Hydroacoustic Transect Cumulates, Summer 1995-98, Ugamak Island,
Alaska……………………………………………………………………... 42
Fig. 13. Hydroacoustic Transect Cumulates, Winter 1997-99, Ugamak Island,
Alaska…………………………………………………………………….. 44
vii
List of Tables
Table 1. Estimates of relative biomass density (Q) and relative
estimation error by season and year at Ugamak Island,AK,
summer 1995-1999……………………………………………… 45
Table 2. Survey variance using EVA and random variance estimates
from SMMOCIsurveys at Ugamak Island, AK, summer and winter
1995-99. ………………………………………………………… 46
Table 3. Midwater and neuston trawls at Ugamak Island, AK,
summer and winter 1995-99. …………………………………… 48
Table 4. Longline set log at Ugamak Island, AK, 1995-98. ………… 52
Table 5. Frequency of ocurrence of prey in the diets of Pacific halibut
and Pacific cod at Ugamak Island, Alaska from 1995-98. ……… 53
viii
Acknowledgements
Many thanks to Dr. Gerardo Chin-Leo (The Evergreen State College,
MES program) for being the primary reader of this thesis and for his many years
of support, excellent review, encouragement, and patience during a very long
thesis project. I also thank Dr. Amy Cook, (The Evergreen State College, MES
program), for participating as a reader and for providing timely and excellent
comments on this thesis. I would also like to thank Lowell Fritz (National Marine
Fisheries Service, Alaska Fishery Science Center, National Marine Mammal
Laboratory) for participating as a thesis reader and providing excellent feedback
and support while I was balancing both work and school.
I would also like to thank Vernon Byrd (Alaska Maritime National
Wildlife Refuge, USFWS, Homer, AK), Dr. Brenda Norcross (University of
Alaska, Fairbanks, AK), Dr. John Piatt (US Geological Survey, Anchorage, AK),
and Dr. Richard Merrick (National Marine Fisheries Service, Northeast Fishery
Science Center, Woods Hole, MA) for the design and implementation of the
SMMOCI cooperative research project of which this thesis is a part, and to their
support of my thesis.
Many people contributed to this project by participating in research cruises
during summer and winter, collecting data, providing logistical support, and
without whose participation this project would have been impossible. NMML
researchers and participants included Dr. Richard Merrick, Jim Thomason, Dr.
Rolf Ream, John Sease, Dr. Patience Brown, Carolyn Kurle, Mike Strick, Charles
Hutchinson, Lisa Baraff, Chris Gburski, and Dr. Marcus Horning. AMNWR
researchers and participants included Don Dragoo, Jeff Williams, Steve Ebberts,
Susan Woodword, and T. Bittner. UAF (Fairbanks) researchers and participants
included Dr. Brenda Norcross, Dr. Franz Mueter, Brenda Holladay, Elizabeth
Chilton, Charlene Zabriskie, Ed Roberts, and Pat Lovely who provided bottom
ix
trawl surveys and data. Researchers and participants from Alaska Department of
Fish and Game included Dr. Lorrie Rea and Dennis McAllister. A sincere thank
you to all for enduring rough seas, getting up in the middle of the night to fish,
endless hours of line transect data collection, and being away from home for
weeks at a time.
I’d like to thank Neal Williamson (National Marine Fisheries Service,
Alaska Fishery Science Center, Midwater Assessment and Conservation
Engineering Division, Seattle, WA). for his invaluable advice on hydroacoustic
data and survey variance estimation analysis and his support. Jan Benson, Lowell
Fritz, and Bruce Robson (National Marine Fisheries Service, Alaska Fishery
Science Center, Seattle, WA) provided much needed advice with ArcMap and I
thank them for their help and patience with a novice mapper. Jeff Breiwick
(National Marine Mammal Laboratory, Alaska Ecosystem Program,Seattle, WA)
provided invaluable assistance with Word formatting for this thesis.
I would also like to thank Pat Livingston, Geoff Lang, MeiSun Yang, and
the members of National Marine Fisheries Service, Alaska Fishery Science
Center, Resource Ecology and Fisheries Management Division Food Habits
Laboratory, Seattle, WA who provided predatory fish stomach content analysis
as did Dr. Brenda Norcross and Brenda Holladay (UAF, Fairbanks). Annette
Brown and Morgan Busby (National Marine Fisheries Service, Alaska Fishery
Science Center, Fisheries and Oceanography Coordinated Investigations (FOCI)
Division, Seattle, WA) provided identification of larval fish and zooplankton.
Funding for this project was provided by the National Marine Fisheries
Service, Alaska Fishery Science Center, National Marine Mammal Laboratory,
Alaska Ecosystem Program in Seattle, WA. I thank Dr. Tom Loughlin and Dr.
Doug DeMaster for funding and use of SMMOCI data for this thesis project.
x
Vessel support was provided by the Alaska Maritime National Wildlife Refuge
(USFWS) vessel M/V Tiglax, based in Homer, AK. Many thanks to the Captain
and crew of the M/V Tiglax for their enthusiasm and dedication to this project.
Finally, I’d like to thank my family and friends for without their unwavering
support and patience, I wouldn’t have made it through a very long project.
xi
INTRODUCTION
Identification of the Problem
Marine life in the North Pacific Ocean has undergone major changes in
population dynamics in recent decades. While walleye pollock (Theragra
chalcogramma) and salmon (Oncorhynchus spp.) populations have increased
dramatically, stocks of crab and Pacific herring (Clupea pallasii) have declined.
From 1950 to 1995, a period witnessed by an explosive growth and change in
fisheries resources, a concurrent marked decline in Alaskan Steller sea lion
(Eumetopias jubatus) populations occurred in Alaska (Merrick et al. 1987,
Alverson 1992). During the same time period, some marine mammal populations
increased, while others declined. These changes are most evident in the western
Gulf of Alaska and the eastern Bering Sea where the Steller sea lion population
has undergone a dramatic decline. The National Marine Fisheries Service
(NMFS), following analysis of population abundance data, estimated over a 75%
decrease in numbers of Steller sea lions in these areas since the late 1970s
(Loughlin et al. 1992; Loughlin and York 2000). Genetic evidence suggests that
the Alaskan Steller sea lion population consists of a western stock (which has
declined) and an eastern stock (which has increased over the past two decades to
about the same size as the current western stock) (Bickham et al. 1996; Loughlin
1997). The NMFS listed Steller sea lions as threatened under the Endangered
Species Act (ESA) in 1990. Results from genetics studies prompted the NMFS
to relist the western stock as endangered and the eastern stock as threatened under
the Endangered Species Act (ESA) in 1997.
One of the prominent hypothesized causes for the population decline in
Steller sea lions is that available food resources (i.e., species abundance and
composition of forage fishes) have changed, and specifically that this change is
affecting nursing females with pups on the rookeries (Merrick, et al. 1987;
1
Merrick et al. 1995). Steller sea lion food habits research has been of interest for
several years (Sea Grant 1993). Since that time there is even more interest in
comparing the western versus eastern sea lion stocks and determining effects of
food availability on both. Some diet studies show a strong positive correlation
between differences in Steller sea lion diet diversity by area and the degree of
population decline in those areas while other studies do not show a correlation
between diet diversity and population decline (Merrick et al 1997, Sinclair and
Zeppelin 2002, Wynne et al. 2005). Merrick et al. (1997) report in the 1990’s that
as diet diversity decreased Steller sea lion populations decreased, suggesting that
Steller sea lions require a variety of prey species for survival (Merrick et al 1997;
Sinclair and Zeppelin 2002). However, recent studies in the Kodiak area show
that the sea lion population decline is continuing in that area even though diet
diversity is high (Wynne et al. 2005).
Steller sea lions eat a variety of fish and invertebrates, including species of
primary and secondary importance to Alaskan commercial fisheries. Sea lion
prey species that are targets of prime commercial fisheries in Alaskan waters,
include walleye pollock, Atka mackerel (Pleurogrammus monopterygius), Pacific
cod (Gadus macrocephalus), flatfishes (Pleuronectidae), rockfishes (Sebastes
spp.), shrimps (Pandalidae), Pacific herring, and salmon. Other prey of pinnipeds
include those that are also the prey of the commercially harvested groundfish
species. These prey include capelin (Mallotus villosus), Pacific sandlance
(Ammodytes hexapterus), eulachon (Thaleichthys pacificus) and cephalopods.
Pinnipeds are opportunistic feeders and thus tend to eat whatever is most
abundant and accessible. Their diets are based more on availability of prey than
preference for a specific food item (Pitcher 1980; Pitcher 1981; Kajimura 1985;
Merrick et al. 1997; Sinclair and Zeppelin 2002). Utilization of a given prey item
2
may differ among individuals due to age or reproductive status (Frost and Lowry
1986). This and the variability in prey abundance results in seasonal, annual, and
regional dietary differences among individuals of the same species. The potential
for competition between commercial fisheries and marine mammals exists if
increased seasonal or age-specific energetic demands of marine mammals
coincide with temporal and spatial scarcity of prey resulting from removal by
commercial fisheries (Mueter and Norcross 1998).
Commercial fisheries have been affected by the listing of Steller sea lions
under the ESA despite the fact that the extent to which commercial fishery
harvests influence pinniped populations is unknown (Lavigne 1982; Swartzman
and Haar 1983; Harwood and Croxall 1988; Loughlin and Merrick 1989;
Alverson 1992; NMFS 1992). Until adequate fisheries data are available,
uncertainty about interactions between fisheries and marine mammals forces
resource managers to be conservative in their management plans and policies,
erring on the side of the declining and endangered marine mammal populations.
NMFS has implemented conservation measures to encourage Steller sea lion
recovery, including time and area restrictions for potentially competitive fisheries.
NMFS has also established 3, 10, and 20 nautical mile (nm) management zones
around sea lion rookeries and haul-outs because commercially important species
such as walleye pollock and Atka mackerel were found in sea lion stomachs
during diet studies conducted in the Gulf of Alaska and Bering Sea. While more
primary prey availability is important, the need for seasonal availability of
primary prey in the right locations is a vital component for foraging juvenile and
adult female Steller sea lions.
The relationships between prey fish abundance, harvest, and impacts on
pinniped populations are not well defined. However, nutritional stress in Steller
sea lions has been correlated with large commercial walleye pollock harvests
(Calkins and Goodwin 1988; Lowry et al. 1989). Sea Grant (1993) reported that
the importance of walleye pollock in the diet of Steller sea lions may be biased in
3
that it is an assessment based on stomach contents of sea lions from the late
1970's and early 1980's when there was an explosive increase of walleye pollock
in the Gulf of Alaska (Pitcher 1981). Abundance of walleye pollock not only
increased in the diets of Steller sea lions, but also precipitated the increase in
commercial fishing effort. When there was no commercial fishery for walleye
pollock in the 1950's, there were also no walleye pollock in sea lion diets
(Mathiesen et al. 1962; Thorsteinson and Lensink 1962; Fiscus and Baines 1966).
In the central and eastern Aleutian Islands (west of Ugamak Island) the diet of sea
lions is dominated by Atka mackerel, the most abundant prey in the area.
Changes in Stelller sea lion diet between the early 1970's and the 1990's reflect
the nature of sea lions as opportunistic feeders and may be more indicative of the
availability of prey rather than an indicator of sea lion feeding preferences.
Measures to establish buffer zones around Steller sea lion rookeries, in
conjunction with the research emphasis on walleye pollock and Atka mackerel,
tend to obscure the significance of other known pinniped prey such as flatfishes,
Pacific herring, cod, salmon, capelin, Pacific sandlance, and cephalopods.
Interdecadal changes in pinniped consumption of these species, as well as walleye
pollock and Atka mackerel, are reflected in the stomach contents of Steller sea
lions. Steller sea lions around Kodiak Island consumed mainly small forage
fishes such as capelin, with cephalopods as a secondary food source, between
1973 and 1978 (Pitcher 1981). From 1985 to 1986, sea lion diets included no
capelin but were dominated by walleye pollock, octopus and flatfishes (Calkins
and Goodwin 1988, Merrick and Calkins, 1996).
Although demersal fish availability was poorly understood, it was a
significant component of juvenile Steller sea lion diets in the Gulf of Alaska in
1985-86 (Calkins and Goodwin 1988, Merrick and Calkins 1996). Changes in the
4
species composition of the juvenile groundfish community may have been
reflected in sea lion diets when walleye pollock replaced capelin as the major
species in the diet. A groundfish community dominated by gadids and flatfishes
coincided with a marked decline in a shrimp dominated crustacean community
which occurred in the nearshore zones around Kodiak Island in the late 1970's
(Anderson and Piatt 1999). The decline in shrimp occurred concurrently with a
decline in capelin and other forage fishes. This decline has been demonstrated to
affect commercial fish species in the eastern North Pacific Ocean (Hollowed and
Wooster 1995) and some researchers indicate that the change may have been part
of the regime shift that occurred in the Gulf of Alaska during the late 1970's
(Royer 1989, Ebbesmeyer et al. 1991, Trenberth and Hurrell 1994). Other
researchers show that pollock population abundance is naturally highly cyclic,
that forage fish species likely were not the dominate prey species in the fish
community and that shifts in abundance were likely not affected by a regime shift
(Fritz and Hinckley 2005).
The extent and causes of the changes in prey availability and their effects on
Steller sea lion populations and the nearshore marine ecosystem are unknown.
Mortality of juvenile pinnipeds due to decreased availability of suitable food is
hypothesized as a cause of the Steller sea lion population decline (Loughlin and
Merrick 1989; Sea Grant 1993; Merrick 1995) as is a decrease in natality (Holmes
and York, 2003). To test this hypothesis NMFS and USFWS initiated a study to
assess the availability of small fishes, which are the principal prey of juvenile
Steller sea lions. Prey assessment surveys were conducted in cooperation with the
U.S. Fish and Wildlife Service (USFWS) Aleutian Maritime National Wildlife
Refuge (AMNWR), the Biological Research Division of the U.S. Geological
Survey (USGS), and the University of Alaska Fairbanks, Alaska (UAF) under the
auspices of a cooperative research group named Seabird, Marine Mammal,
Oceanographic Coordinated Investigations (SMMOCI). Surveys of fish species
availability were conducted in conjunction with NMFS/NMML (National Marine
5
Mammal Lab) population abundance surveys of Steller sea lions at major
rookeries in the Gulf of Alaska. The emphasis of the study was on small prey
species abundance near sea lion rookeries in concurrence with the theory that poor
survival of sea lion pups is causing the decline in the population, and that juvenile
survival may be less successful because of reduced availability of forage fish.
SMMOCI surveys were initiated as a 5 year pilot study to investigate
functional response on sea lion rookeries and seabird colonies related to changes
in the nearshore marine ecosystem (V. Byrd, pers. comm., USFWS/AMNWR).
Summer surveys were initiated due to a concern about nutritional health and stress
on adult female sea lions during the breeding season, nutritional health and stress
on breeding seabirds, and how these factors may contribute to the declines in each
population. Vessel cost and availability was also a factor in the decision to
conduct summer surveys. Winter surveys were conducted to assess prey
resources available to young sea lions during their first year of foraging, resource
availability to the general population of Steller sea lions during winter, as well as
resource abundance and availability differences between seasons.
The main objectives of this study were:
1.
To assess distribution and abundance of juvenile and subadult life stages of
commercially important fishes which serve as potential prey for Steller sea lions
within the 20 nm critical habitat area surrounding Ugamak Island.
2.
To determine distribution and abundance of non-commercially important
species of fish which serve as potential prey for Steller sea lions within the 20 nm
critical habitat area surrounding Ugamak Island.
6
3.
To determine interannual fluctuations in availability of commercially and
non-commercially important demersal fishes around sea lion rookeries.
4.
To utilize geostatistical methods to increase accuracy of hydroacoustic survey
variance estimation.
Description of the Study Area
Ugamak Island (54o 12.45 N, 164o 46.6 W) was formerly one of the largest
Steller sea lion rookeries in the world. Ugamak Island lies near the western edge
of Unimak Pass in the Fox Islands area of the eastern Aleutian Islands (EAI), at
the downstream end of the Alaska Coastal Current (Fig. 1). The pass is broad,
about 18 km at its most narrow spot, and is relatively shallow (mostly <100 m).
The island is subject to oceanographic influences from water masses of both the
Gulf of Alaska (GOA) and the Bering Sea. Water flow in Unimak Pass is largely
governed by tidal processes which may push GOA water north or Bering Sea
shelf break water south through Unimak Pass and adjacent passes (Kinder and
Schumacher 1981, Schumacher et al. 1982, Hood 1986). Warm, low-salinity
Alaska Coastal Current water hugs the coast of Unimak Island and winds around
into the Bering Sea without crossing the passes. Waters around Ugamak Island
and nearby Aiktak Island are well-mixed by tidal upwelling in the passes (Haney
et al. 1991). They are characterized by temperatures and salinities that are
intermediate between shelf break water and GOA waters, and have weak vertical
property gradients (Haney et al. 1991).
Euphausiids, particularly, Thysanoessa inermis, completely dominate the
biomass of zooplankton and form large, dense aggregations in passes and straits
in the study area (Troy et al. 1991). Shelf species of forage fish, such as capelin
and sandlance, are relatively scarce perhaps in part because shelf habitat around
the islands is rather limited. Troy et al. ( 1991) found that in tidally mixed water
around the islands, juvenile pollock (age 0+) overwhelmingly dominate (99.7%)
7
in trawl catches during fall. Farther offshore in GOA water, lanternfishes
(Myctophidae) are the most abundant species of forage fish (Troy et al. 1991).
The Steller sea lion rookeries at Ugamak Island are located in small bays on
the southeast and northeast ends of the island. The water is relatively shallow
near the rookery, and then drops off to about 50 m with some deeper areas of up
to 150 m, within 5 miles of the rookery. Moderate to strong tidal currents, with
an average maximum flow of 1.3 to 4 knots, and small tidal ranges of about 0.5 m
are typical in the Ugamak Island area. The waters surrounding Ugamak Island are
characterized by very rough bottom topography (Mueter and Norcross 1998).
8
Fig. 1. Map of Alaska and Ugamak Island Steller sea lion rookery sites.
Prey Assessment Surveys at Ugamak Island
9
The main objective of the SMMOCI surveys was to describe the nearshore
marine ecosystem including assessing potential prey biomass available to Steller
sea lions and seabirds in the vicinity of sea lion rookeries and haulouts and
seabird colonies. Research conducted at Ugamak Island as a part of the SMMOCI
surveys includes hydroacoustic line transects, mid-water and bottom trawl
surveys, marine mammal and seabird line transect surveys, and collection of
oceanographic data (temperature and salinity).
From 1995-99 the National Marine Mammal Laboratory’s (NMML)
Alaska Ecosystem Program conducted SMMOCI prey assessment survey research
onboard the USFWS vessel M/V Tiglax during both the Steller sea lion breeding
(June-July) and non-breeding seasons (March) in Alaska in the regions from the
Kenai peninsula to the western Aleutian Islands. SMMOCI surveys were
conducted during June-July 1995-98 and March 1997-99. Many sites were
surveyed including Marmot, Sugarloaf, Chowiet, Atkins, Kasatochi, Kiska,
Buldir, Agattu, and Ugamak islands as well as Cape Sarichef and the Unimak
Pass area. Additional surveys were conducted by USFWS at Buldir, Kasatochi,
and Aiktak islands as well as by USGS in the Barren islands, however, only
surveys from Ugamak Island are reported in this thesis.
The nearshore marine components of the study include: (1) biomass
estimates of potential seabird and marine mammal prey within 20 nm of the
breeding areas, (2) identification of common prey in the area, (3) assessing
oceanographic characteristics of water masses nearby, (4) characterizing bottom
fauna, (5) recording the feeding distribution of birds and marine mammals, and
(6) assessing food web relationships by analyzing stomach contents of fish and
birds (Byrd et al. 1997).
This thesis will focus on relative biomass indices from hydroacoustic
10
transects, variance estimation from hydroacoustic transect surveys, and trawl and
longline surveys from Ugamak Island during SMMOCI studies. Ugamak Island
study objectives were to:
•
Describe the nearshore marine ecosystem including assessing relative
biomass indices within the 10-20 nm fishery management area
surrounding the Ugamak Island.
•
Determine interannual and seasonal fluctuations in relative biomass
surrounding the Ugamak Island Steller sea lion rookery.
•
Utilize geostatistical methods to better estimate variance in hydroacoustic
survey biomass at Ugamak Island.
METHODS
Hydroacoustic prey assessment surveys
The USFWS vessel, M/V Tiglax, was used on all surveys conducted
during this study. Hydroacoustic data were collected along a series of parallel
transects within 20 nm radius of Ugamak Island to estimate the distribution and
relative biomass index of potential prey resources (Fig. 2).
11
Fig. 2. Map of hydroacoustic transects, midwater trawl stations, and nekton tow
stations at Ugamak Island, Alaska, 1995-99.
12
The vessel operated at 10 knots (kts) during hydroacoustic transects. Data were
collected using the vessel’s BioSonics 102 hydroacoustic system, with hull
mounted (4 m deep) 38 and 120 kHz transducers operated in a multiplexing
(alternating between transducers) mode (BioSonics 1994). The system was run in
multiplexing mode to obtain separate estimates of total biomass (large and small
targets) using the 120 kHz transducer and estimates of large target biomass using
the 38 kHz transducer. All data were echo integrated in real time using BioSonics
Echo Signal Processing (ESP) software running on the ships’ computer.
Transects were standardized, such that subsequent surveys covered the same areas
and bathymetry. All transect legs (7) were surveyed once during daylight hours.
Additionally, the central transects (3) were also surveyed at night, on an
opportunistic basis, for a total of 10 transect lines equaling approximately 160 km
of transects per survey. However, night transects were less consistent than
daylight transects and are not used in the analysis.
Acoustically derived relative biomass estimation
Targets Per Unit Surface Area (TPUSA) were integrated by the ESP
software program using the reports from the Run Table (RE output table) and
values were reported in kg/m2. The RE table contains all the integration report
information collected by ESP on a per depth stratum and per report basis. Each
report listed in this file includes individual depth stratum results and contains
calculated values for relative biomass density collected at 1 minute intervals along
each transect. Transect lengths ranged from 8 nm (Transect 1) to 18.7 nm
(Transects 3 and 5). Data collected from the 120 kHz transducer was used in the
analysis as it provided a better index of relative biomass density available to all
predators. Relative biomass density is an estimate of the amount of acoustically
detected biomass encountered during transect survey for all species ranging from
13
zooplankton to fish. Relative biomass estimated by the 120 kHz transducer
ranged from zooplankton species to fish species but were not discernable to
species. In order to differentiate between species extensive trawling would be
required to verify and identify prey to species.
Data were analyzed post-survey using additional ESP software and
EXCEL. Analysis of data provided average bottom depth, TPUSA (kg/m2) and
mean relative biomass density. TPUSA provided a relative index of biomass by
averaging the biomass density (kg/m2) of each sample obtained from each one
minute time segment from each depth strata sampled on each transect for all
transects on a survey.
Relative biomass density values were then graphed for each depth strata in
order to detect areas where the bottom may have been integrated, as well as other
data anomalies. Integrated bottom signal and other anomalous data were edited
by hand since the ESP program did not have the capability to edit them
automatically (J. Piatt, USGS, pers. comm.). Sources of anomalous data can
include surface bubbles extending below the depth of the transducer or poor
weather conditions caused by winds or sea state resulting in pitching and rolling
of the ship.
After anomalous data points were removed the edited survey data were
analyzed to detect the horizontal location of concentrations of relative biomass
density geographically in the survey area. For each transect, edited data were
averaged by depth strata to obtain a single data point to represent seasonal and
annual variability within the study area by depth strata, transect, year and season.
To represent annual and seasonal vertical relative biomass density
distribution in the survey area data were averaged by depth strata across all
transects. These average data were then summed to obtain an average relative
14
biomass density for each survey year and season. These summed data were then
divided by the depth strata midpoints (e.g. 5m, 10m, 15m, etc.) to obtain a
weighted average relative biomass density data for each survey year and season.
Weighted average relative biomass densities were graphed to compare vertical
variability between years and seasons.
Estimation Variance Analysis (EVA)
Ecological analysis generally includes investigations of the dispersion and
patterns associated between species at different places and times- patterns that
reflect spatial dependence rather than independence (Pielou 1977). Both Ricklefs
(1973) definition of ecology as “the study of the natural environment, particularly
the interrelationships between organisms and their surroundings” and that of
McNaughton and Wolf (1973) - “the scientific study of the relationships”, imply
spatial and temporal dependence (Rossi et al. 1992). The concepts of spatial and
temporal dependence or continuity should be readily apparent to the ecologist.
Examples of these concepts include vegetation species and densities that are
generally different on north-facing vs. south-facing slopes, increased density of
grasshoppers during hot, dry periods, and plants in greenhouse experiments that
are routinely rotated to eliminate micro climatic and micro environmental effects.
Additionally, distance from a major seed, a predator, or an herbivore source or
temporal features of a system such as diel trends in temperature, radiation,
salinity, or thermocline can affect a species distribution and behavior (Robertson
1987).
Interpolation of data is key to ecological field studies. Ecologists who
infer mean values for particular variables within a given experiment or time
increment implicitly interpolate values for all points not measured. If
assumptions regarding sampling independence and normality are met then
parametric statistics provide optimal estimates of variance around unbiased
15
means. These variance estimates, based on normally distributed data, are widely
used to describe attributes of experiments and to test hypotheses about ecological
processes (Robertson 1987). However, assumptions about sample independence
are difficult to meet in ecological field studies due to autocorrelation of sampled
data points: samples collected close to one another are often more similar to one
another than are samples collected farther away, whether in space or time
(Robertson 1987).
Because of the prevalence of autocorrelated data in field studies, estimates
of variance around interpolated points may differ substantially from overall
population variance. As a result imprecise estimates of sample values within the
unit sampled and a biased estimate of treatment effects in experimental systems
can occur (Trangmar et al. 1985, Sokal and Rohlf 1981). The recent
development of regionalized variable theory, for applications in geology
(Matheron 1971, Journel and Huijbregts 1978, Krige 1981) and soil science
(Burgess and Webster 1980a) provides an elegant means for describing
autocorrelation in data, and a means to use this autocorrelation information to
derive precise, unbiased estimates of sample values within the sampling unit.
These estimates incorporate the detailed spatial patterns with known variance for
each interpolated point. Spatial variability in particular has long been difficult to
quantify in ecologically meaningful ways and the development of this theory is of
considerable interest to ecologists (Robertson 1987).
In recent years there has been increased attention on the design of acoustic
surveys and estimates of survey variance. Two approaches which have been
commonly adopted are: (1) a stratified random sample design relying on classical
statistics for variance estimation (Jolly and Hampton 1990) and (2) a systematic
sampling design using a grid of parallel transects and employing techniques
16
commonly used in geology from a field of statistics known as “geostatistics”
(Petitgas 1993). The first approach is design-based requiring data from a random
sample of transects. The second approach is a statistical model-based approach
that assumes a non-random model of spatial structure (Petitgas 1993). Many
practitioners of acoustic survey assessment acknowledge the statistical validity of
the random sampling approach, but prefer to employ a grid of parallel transects at
a fixed intertransect distance knowing that the abundance estimate will be more
precise. The problem with this approach is that classical statistics do not provide
an estimator for the variance in a systematic survey (Williamson and Traynor
1996).
The theory of geostatistics offers a solution to this problem. Geostatistics
is a branch of applied statistics that focuses on the detection, modeling, and
estimation of spatial patterns. The theory makes use of the observed spatial
structure evident in the correlations in the sampled data and incorporates this
structure into the calculation of variance, which is a two step process: (1) defining
the degree of autocorrelation (or similarity between neighboring data points)
among the measured data points, and (2) interpolating values between measured
points based on the degree of autocorrelation encountered. Autocorrelation is
evaluated by means of the semi-variance statistic γ(h) = 1/2N(h) ∑N(h)I=1 [z(xi) z(xi+h)]2 where z(xi) is the measured sample value at point xi, z(xi+h) is the sample
point value at point xi+h, and N(h) is the total number of sample point contrasts or
couples for the interval in question. The resulting plot of γ(h) vs. all h’s evaluated
is termed the semi-variogram; the shape of this plot describes the degree of
autocorrelation present (Robertson 1987).
The one dimensional (1D) procedure proves to be very appropriate for
acoustic surveys performed along regularly spaced parallel transects (Petitgas
1993). In echo integration surveys of pelagic marine biomass, the measured
backscattered acoustic energy is summed over all individual samples made
through the water column and averaged along unit distances of the ship’s course.
17
Thus, the structural information present on an echogram is not used when
performing biomass estimation (Petitgas and Levenez 1996). Since it is common
that dense targets will constitute a large percentage of the biomass, survey
reliability largely depends on encountering a sufficient number of these targets
(Petitgas and Levenez 1996). If the researcher is primarily interested in global
estimation (i.e. survey abundance and its variance), Petitgas (1993) recommends
the use of the transitive method in 1D. Since acoustically sampled data are
serially correlated along each transect, the information from the transect can be
represented by a single point or cumulate. A cumulate is defined as the product of
the average acoustic return multiplied by the length of the transect. A matrix of
the transect data in two dimensions now becomes a set of n transect cumulates in
1D (Williamson and Traynor 1996). The 1D transitive method models
patchiness of biomass and spatial structure to calculate sample variance. This
method is not generally well known, but is excepted by fisheries acousticians as
valid. The 1D transitive method was used to calculate survey variance for all of
the Ugamak Island survey transects during this study.
Transitive theory in One Dimension (1D)
The transitive theory in one dimension (1D) was applied to the Ugamak
Island survey data using the Estimation Variance (EVA) software provided by
Petitgas and Prampart (1993). EVA software provides a mechanism to
characterize data structure and to estimate variance (Williamson and Traynor
1996). The 1D theory was developed to assess total quantity present over an area
by sampling on a regular grid pattern. In general, the origin of the grid is not
determined by the variable values to be surveyed. As described in systematic
18
sampling design of Cochran (1977), the origin of the grid may be considered
randomly and uniformly located within the survey area (Petitgas 1993). In the
transitive theory, a fixed spatial distribution showing a fixed total quantity is
sampled by a grid of random origin (Petitgas 1993). The transects are oriented
parallel to the y-axis of the grid and traverse the entire width of the grid area.
Hydroacoustic data systematically collected along parallel transects are
continuously sampled, therefore, the data can be cumulated without making an
error on the value of the cumulates (Petitgas 1993) .
Backscattered acoustic data from the Ugamak Island surveys were
analyzed from edited relative biomass density output tables of the BioSonics Echo
Signal Processing (ESP) program. Each transect was summed by depth strata
then totaled to give a cumulate relative biomass density over all depth strata for
each transect. Each cumulate value represents a unit of relative biomass per
transect line and the data set is considered to be one dimensional. Using EVA, an
estimate of total relative biomass, Q, is obtained by multiplying the sum of the
cumulates by the intertransect distance (Petitgas 1993). Seven daytime transects
were conducted at Ugamak Island during most survey years and the inter-transect
distance was equal to 3 nm. The quantity to estimate is Q = ∫-∞+∞ q(x)dx. It is
estimated by the discrete summation:
Q* = a '7 k=1 q (x0 + ka)
where x0 is the (random) origin of the grid and where a is the inter-transect
distance.
Covariograms
The transitive 1D covariogram is a type of non-centered covariance that is
used as a tool to describe the spatial structure of the population of interest
(Petitgas and Prampart 1993, Williamson and Traynor 1996). To calculate an
19
estimation variance of the relative biomass density (Q), a model is fitted to the
covariogram of the raw data. The models available in the EVA software to fit a
covariogram are the exponential, spherical, Gaussian, and triangular with a nugget
effect optional in each selection. The shape of the covariogram describes the
degree of autocorrelation in the data in 1D. A model was then fit to the
covariogram of the transect cumulate data for each Ugamak Island survey.
Given f(x) to represent relative biomass density for transect location x, the
total abundance Q = ∫f(x)dx. Its estimator is:
Q* (x0) = a 'f(x0 + ia)
where a is the distance between transects and x0 is the random starting point. The
covariogram function is defined as g(h) = ∫f(x) f(x + h)dx. The experimental
covariogram value at lag k is calculated by summing all non-zero products f(x) f
(x +ka), i.e. all transect cumulates k intertransect distances apart. Note that the
behavior of g(h) is both a function of the values f(x) and the number of non-zero
products f(x)f(x +ka).
As described in Williamson and Traynor (1996), the Ugamak Island data
were best fit using a spherical or exponential model or some combination of the
two. Guidelines followed in fitting a model to the raw covariogram values
included:
1. If a single model provided the best fit, the sill was set to the raw covariogram
value at lag 0 and the range was set equal to the width of the survey area.
2. If a combination of models best fit the data, the sum of the sills was set equal
to the 0 lag covariogram value and the range for one of the models
matched the field length.
3. For some data sets, the range in the second model was set to roughly coincide
20
with the size of any aggregates discernible within the field; for others, the
range was selected to provide the best fit.
4. In choosing model parameters, emphasis was placed on fitting the first few
lags as closely as possible (Williamson & Traynor 1996).
EVA variance estimates compared to random sample variance estimates
The variance of such estimation writes simply in 1D:
σQ2 = a '+
n= -
g(na) - ∫+∞-∞ g(h) dh,
where g(h) is the transitive covariogram model of the cumulates (Petitgas 1993).
Relative estimation error is defined as:
σQ ÷ Q
Confidence intervals using EVA are derived as ± 2 x the relative estimation error.
For comparison purposes, (relative) confidence intervals were also
estimated using classical random sample methods for each Ugamak Island survey.
((σ/m) ÷ √ nb ) x 2
where m equals mean values of the cumulates and b equals the number of
transects.
Trawl surveys
Midwater and bottom trawls, as well as long line surveys, were conducted
during SMMOCI survey years at Ugamak Island from 1995-98. These surveys
assessed the types of potential prey within the foraging range of Steller sea lions
and seabirds, in the midwater and bottom portions of the Ugamak Island
nearshore marine habitat.
21
Midwater Trawls and Neuston Tows
Midwater Trawls
Both midwater trawls and neuston tow sampling were conducted by
NMML during SMMOCI hydroacoustic surveys near Ugamak Island between
1995 and 1999. Midwater trawls were conducted on an opportunistic basis in
conjunction with hydroacoustic survey transects. When echo sign of potential
prey was detected along transects the midwater trawl was deployed to verify
potential prey species. Additionally, the trawl was deployed during periods of no
detected echo signal return to verify that no potential prey species were present.
A total of 5 opportunistic midwater trawls and 4 Neuston surface tows
were conducted to verify echo sign encountered near the Ugamak Island Steller
sea lion rookery. When acoustic echo sign was evident during transects, and
reachable by the trawl sampling gear, the transect was paused and the trawl was
deployed. During 1995-96 a 2 m Isaac Kidd midwater trawl (IKMT) was towed
for 15-30 minutes at 2-3 knots (Byrd et al. 1997) (Fig. 3). From 1997 to 1998 a
modified midwater herring trawl was used to sample echo sign and was towed for
approximately 15 minutes at the same speed (Fig. 4). The herring trawl net
design consisted of a 30 ft. wide mouth opening, 1/8 inch (3.2 mm) codend mesh,
and a double warp headrope. A netsounder system was placed on the trawl
headrope for the duration of the tow to indicate the depth and configuration of the
net, and in order to fish the net through the layer of potential prey species. Tows
were limited to 15 minutes in duration at equilibrium (depth of the target layer) in
order to identify the target signal return and assess species composition. Tows
were also deployed using both the IKMT and the herring trawl, and when no
targets were encountered in order to ground truth acoustic echo signal return
22
accuracy. Midwater trawls were only utilized to verify acoustic echo sign during
transects and not designed to serve as a quantitative midwater prey assessment
survey tool.
During occasions when echo signal was detected near the surface, when
large aggregations of seabirds were seen feeding at the surface, or when the seas
were too rough for trawling, a Neuston plankton net, measuring 30 cm by 49 cm,
was deployed (Fig. 5). Neuston tows were deployed at Ugamak Island during
summer 1998 and winter 1997 and 1999. The net was deployed vertically off the
starboard side of the vessel using a small winch and towed at a speed of 2-3 kts.
for a period of 15 minutes. Samples collected from midwater trawls and neuston
tows were identified to the lowest taxonomic level possible, counted, measured,
and preserved for later identification if unidentified in the field.
23
Fig. 3. Modified Issacs-Kidd Midwater Trawl
24
Fig. 4. Modified Herring Trawl
25
Fig. 5. Neuston Net
26
Bottom Trawls
A series of bottom trawls were conducted by University of Alaska
Fairbanks (UAF) researchers during June-July 1995-98 SMMOCI surveys at
Ugamak Island (Fig. 6). The surveys targeted small demersal fishes within the
foraging range of the Ugamak Island Steller sea lion rookery. Bottom trawls were
conducted from the M/V Tiglax, but separately from the hydroacoustic surveys, at
depth-stratified stations radiating out from Ugamak Island. Juvenile and subadult
stages of demersal fishes were the primary target species of the bottom trawl net.
Sampling followed a random sampling design incorporating 3 depth strata. The
depth strata chosen were based on previous study results and were designated as
10-40 m, 40-70 m and over 70 m (Norcross et al. 1995a). Bottom trawl survey
methods are fully described in Holladay et al. 2000, Mueter and Norcross 1998,
and Norcross et al. 1995a.
27
Fig. 6. Bottom trawls, longline sets and hydroacoustic transects at Ugamak
Island, AK 1995-99.
28
Longline Surveys
Many areas surrounding the SMMOCI study locations were not surveyed
with bottom trawls due to rough bottom substrate, including areas surrounding
Ugamak Island. Due to the amount of untrawlable area, longline surveys were
initiated in 1996 to sample large predatory bottom fishes. Bottom fish predator
species were used as a sampling mechanism to assess prey fish species abundance
occurring near the bottom in the areas that exhibited undesirable trawl substrate.
Stomach contents of opportunistic bottom predators were assumed to reflect the
abundance of common prey species available to foraging Steller sea lions in the
same location (Fahrig et al. 1993, Mueter and Norcross 1998).
Longline surveys were conducted near Ugamak Island sea lion rookery
during June-July 1996-98 and during March 1997-98 (Fig. 6). A single skate 100
hook herring baited longline was used to sample hard bottom rocky substrates
where it was difficult to tow a small bottom trawl. The skate was deployed about
4 km from the rookery, sets ranged from 32 m to 62 m in depth, and 80% of sets
were allowed to sample for less than 3 hours to prevent digestion of predator fish
stomach contents.
Predatory fish caught by longline were identified to species, measured,
weighed, and sexed. The mouth and gills of each predator were checked for signs
of regurgitation of prey contents. If regurgitation was evident the stomach from
the predator was not collected. Stomachs were excised, placed in a cloth bag with
a sample identification label which included vessel name, cruise number, haul
location, haul number, date, predator species name, predator length and weight,
and collectors initials. Stomach samples were preserved in 10% buffered
formaldehyde solution for later laboratory analysis.
29
Diet composition of predatory fish
Traditional dietary analysis methods include estimation of volume or weight,
counts, and frequency of occurrence of individual prey items (Cortes 1977;
Hyslop 1980). Analysis of fish stomachs from longline surveys were conducted
by the Institute of Marine Sciences, University of Alaska Fairbanks (UAF) in
summer 1996-97 and by the Resource Ecology and Fisheries Management
(REFM) division of NMFS at the Alaska Fisheries Science Center (AFSC) in
Seattle, Washington in winter 1997 and during both winter and summer 1998. No
longline surveys were conducted in winter 1999 due to very poor weather
conditions and time constraints. Pacific halibut (Hippoglossus stenoepis) and
Pacific cod (Gadus macrocephalus) stomach samples were analyzed for diet
composition at Ugamak Island. Stomach contents at the time of capture were
considered to be representative of the diet composition of the bottom predator
species in the sampling area.
Prior to laboratory processing, the formalin was neutralized from each
collection of stomach samples, then rinsed in freshwater. An estimated stomach
fullness to the nearest 10% (volume) was determined for each stomach sample
prior to removing all stomach contents for identification. All prey items were
removed from the stomachs, blotted on paper towels to remove excess moisture,
and weighed (0.01 g resolution) to obtain total stomach content weight. Wet
weights (0.01 g resolution) were recorded for individual fish and crab species
while percent volume was estimated for other species. Prey items were counted
and identified to the lowest possible taxonomic level. Prey items occurred in
different stages of digestion and were identified based on remaining hard parts
including shells, bones, scales, and fish otoliths. Prey taxa were placed into the
30
following grouped categories for analysis (Mueter and Norcross 1998, Holladay
et al. 2000):
(1) Cephalopods: octopus and squid
(2) Crustacea: barnacles (Balanus spp.), Amphipoda, Isopoda, Mysidacea, shrimp,
decorator crab (Oregonia gracilis), Pygmy cancer crab (Cancer oregonensis), lyre
crab (Hyas lyratus), hermit crab (Paguridae), and unidentified decapods
(3) Other benthic invertebrates: unidentified tunicates; unidentified gastropods
and bivalves; bryozoans; ribbon worms; polychaetes; brittle stars; sea urchins; and
invertebrate fragments
(4) Gadidae: walleye pollock (Theragra chalcogramma), Pacific cod (Gadus
macrocephalus) and unidentified gadids
(5) Small schooling fishes: Pacific herring (Clupea pallasii); Pacific sandlance
(Ammodytes hexapterus) and capelin (Mallotus villosus)
(6) Pleuronectidae: flathead sole (Hippoglossoides elassodon), rock sole
(Lepidopsetta spp.), butter sole (Pleuronectes isolepis), and unidentified flatfishes
(7) other demersal fishes: rockfishes (Scorpaenidae), pricklebacks (Lumpenus
spp.) unidentified poachers (Agonidae), skates (Rajidae), sculpins (Cottidae)
including Triglops spp., northern sculpin (Icelinus borealis), yellow Irish Lord
(Hemilepidotus jordani), and unidentified sculpins.
(8) Unidentified fishes: This category could possibly include members of other
fish categories.
(9) Seabird: crested auklet (Aethia cristatella).
In a comparative review of fish stomach contents analysis methods, Hyslop
(1980) found that recording the number of stomachs containing one or more
individuals in each food category as a percentage of all stomachs, commonly
referred to as frequency of occurrence, was the simplest method. Even though
frequency of occurrence gives little indication of the relative amount of each food
31
category, it does present a qualitative assessment of population-wide food habits
(Hyslop 1980; Cortés 1977). Frequency of occurrence of each prey category was
used to measure relative importance of prey in the diet of a predator species at the
rookery, and is defined as:
# of stomachs in which prey was found ÷ Total # of stomachs
where the number of stomachs in which prey was found for a stomach collection
sample event is divided by the total number of stomachs collected during the
same sampling event. The frequency of occurrence approach considers only
presence-absence of a prey taxon and is not an estimation of the number or weight
of prey species consumed.
The indices used above were meant as simple descriptors of diet composition
and were assumed to reflect prey availability of commonly consumed prey groups
near the rookery. Rigorous statistical comparisons of diet composition or prey
availability among rookeries was not possible because the variance of the
estimators was not known (Mueter and Norcross 1998; Holladay et al. 2000).
RESULTS
Acoustically derived relative biomass estimation at Ugamak Island
The relative biomass densities reported during Ugamak Island hydroacoustic
prey assessment surveys should be considered to be a relative index of midwater
biomass, rather than an absolute estimate.
32
Mean relative biomass densities were calculated for Ugamak Island for
summers 1995-98 (Fig. 7) and winters 1997-99 (Fig. 8) (Table 1). Average
relative biomass densities were calculated annually and seasonally for each
transect by depth strata. Geographic distribution of relative biomass density
during summer surveys was the highest in 1995 on Transect 7 (.0354 kg/m2) and
lowest on Transect 5 (.0003 kg/m2) in 1996 (Fig. 7). During winter surveys
geographic distribution of relative biomass density was much higher than summer
relative density with the highest density on Transect 4 (.1750 kg/m2) in 1999. The
lowest relative density during winter occurred on Transect 5 (.00009 kg/m2) in
1998 (Fig 12). Transect 1 was not surveyed during winter 1999.
Vertical distribution of relative biomass density at Ugamak Island during
summer surveys ranged from 16 m in 1996 to 32 m in 1995. During winter
surveys vertical distribution
of relative biomass density ranged from 37 m in 1997 to 44 m in 1999 (Fig. 9).
Although vertical distribution of relative biomass was located deeper in the water
column during winter surveys than during summer surveys it was still within the
average diving range of young Steller sea lions at Ugamak Island (0-50 m)
(Fadely et al. 2005).
33
0 .0 4
T ra n s 1
T ra n s 2
0 .0 3 5
T ra n s 3
T ra n s 4
T ra n s 5
0 .0 3
T ra n s 7
2
Average Density (kg/m )
T ra n s 6
0 .0 2 5
0 .0 2
0 .0 1 5
0 .0 1
0 .0 0 5
0
1995
1996
1997
1998
Y ear
Fig. 7. Summary of Relative Biomass Density by Hydroacoustic Transect and Year Ugamak Island, Alaska, Summer 1995-98.
34
0.18
Trans 1
Trans 2
Trans 3
Trans 4
Trans 5
Trans 6
Trans 7
0.16
2
Average Density (kg/m )
0.14
0.12
0.1
0.08
0.06
0.04
0.02
0
1997
1998
1999
Year
Fig. 8. Summary of Relative Biomass Density by Hydroacoustic Transect and Year Ugamak Island, Alaska, Winter 1997-99.
35
0
Summer
5
Winter
10
15
20
Depth (m)
25
30
35
40
45
Winter
1995
1996
Summer
1997
Year
1998
1999
Fig. 9. Vertical Distribution of Biomass Density at Ugamak Island, AK 1995-99
36
Estimation Variance Analysis (EVA)
For each survey, estimates of relative biomass density (Q) and relative
estimation error (σQ/Q*) were calculated (Table 1). Relative biomass density was
highest during winters 1997 and 1999 and lowest during summer 1996. Relative
estimation errors were highest during summer 1998 and winters 1997 and 1998,
and about the same during summers 1995, 1996, and 1997 and winter 1999 (Table
1). The Ugamak Island survey variance estimates using EVA ranged from 17.4 to
52.1% for summer surveys and 19.0 to 39.4% for winter surveys (Table 2).
Classical random statistical sampling estimates showed dramatic differences from
the EVA results. Summer surveys ranged from 49 to 112% while winter surveys
ranged from 46 to 108% (Table 2).
Cumulate values were calculated for each survey transect by summing the
voltage returns in each depth strata then calculating a total voltage return (Q) for
each transect (Table 1). A histogram of the transect cumulates was graphed for
each survey and a model was fitted to the raw covariogram values. Raw
covariogram values are influenced by the amount of relative biomass present,
which varies interannually and interseasonally. The shape of the covariogram
plot describes the degree of autocorrelation in the data in 1D. A curved line on
the covariogram indicates patchiness in the distribution of relative biomass
whereas a straight line indicates a more uniform distribution in relative biomass.
Covariogram models of relative biomass estimates during summers 1995 and
1998 show more sharply curved lines, indicating a more patchy relative biomass
distribution than those of summers 1996 and 1997 (Fig. 7). During winters 1997
and 1998 the covariogram models show sharper curved lines and thus more
patchy relative biomass distribution than during winter 1999 (Fig. 8). The
covariogram model for summer 1997 shows the most uniform distribution of
relative biomass of all of the Ugamak Island surveys during summer or winter
(Figs. 6 and 7).
37
Histograms from transect cumulate data show that during summer 1995
relative biomass was higher on transects 4, 5, and 6 than on other transects; in
summer 1996 relative biomass was higher in transects 1, 2 and 4; in summer 1997
relative biomass was higher in transects 2 and 3; in summer 1998 relative biomass
was higher in transect 4 (Fig. 9).
During winter surveys transect cumulate data show that during winter
1997 relative biomass was higher on transects 3 and 4; in winter 1998 relative
biomass was higher on transects 3, 4 and 7; in winter 1999 relative biomass was
higher on transect 3 but more uniformly distributed throughout the survey area
than in previous winter surveys (Fig. 10).
38
Ugamak – Summer 1995
Ugamak – Summer 1996
Fig. 10. Survey Variance Estimation Covariograms, Summer 1995-98, Ugamak Island, Alaska
39
Ugamak – Summer 1997
Ugamak – Summer 1998
Fig. 10. Survey Variance Estimation Covariograms, Summer 1995-98, Ugamak Island, Alaska
40
Ugamak – Winter 1997
Ugamak – Winter 1998
Ugamak – Winter 1999
Fig. 11. Survey Variance Estimation Covariograms, Winter 1997-99, Ugamak Island, Alaska
41
Ugamak – Summer 1995
Ugamak – Summer 1996
Fig.12. Hydroacoustic Transect Cumulates, Summer 1995-98, Ugamak Island, Alaska
42
Ugamak – Summer 1997
Ugamak – Summer 1998
Fig 12. Hydroacoustic Transect Cumulates, Summer 1995-98, Ugamak Island, Alaska
43
Ugamak – Winter 1997
Ugamak – Winter 1998
Fig. 13. Hydroacoustic Transect Cumulates, Winter 1997-99, Ugamak Island, Alaska
44
Ugamak – Winter 1999
Table 1. Estimates of relative biomass density (Q) and relative estimation error
by season and year at Ugamak Island,AK, summer 1995-1999.
Survey
Year
Spacing
Season (nm)
Number of
Transects
1995
1996
1997
1998
summer
summer
summer
summer
3
3
3
3
7
7
7
7
265.66
23.87
197.94
331.08
.0873
.0097
.0953
.2607
1997
1998
1999
winter
winter
winter
3
3
3
7
7
6
783.85
300.81
1930.88
.1969
.1680
.0924
Q
Relative
est. error
45
Table 2. Survey variance using EVA and random variance estimates during
surveys at Ugamak Island, AK, Summer and Winter 1995-99.
(“----“ indicates no survey).
__________________________________________________________________
Summer
Winter
__________________________________________________________________
Year
EVA
Random
Variance Est.
Random
EVA
Variance Est.
__________________________________________________________________
1995
1996
1997
1998
1999
46
17.4%
19.46%
19.1%
52.1%
------
50%
49%
56%
112%
-------
----------39.4%
34.0%
19.0%
----------108%
86%
46%
Trawl surveys
Midwater trawl surveys
In summer 1995 infrequent echo sign was seen on hydroacoustic survey transects at
Ugamak Island, particularly during the day, with the exception of what was believed to be
occasional patches of Pacific herring (Clupea pallasii) and capelin (Mallotus villosus). Pacific
herring was identified at the surface twice during survey transects. During night transects, the
scattered day time echo sign coalesced to form long bands or layers which were rarely seen on
the vessel’s 50 kHz sounder, suggesting that the return echo signal was either from zooplankton
or fishes without a swim bladder. Unfortunately, trawling during these transects was not
conducted due to insufficient wire on the net reel to reach the target species layer which was
located near the bottom. Past experience from other midwater tows at other locations indicated
that the echo sign was probably from 0-aged fish (usually gadids) or euphausiids. The only
other significant echo sign believed to be something other than zooplankton was observed
outside of the Ugamak Island study area at the southeast corner of the Chowiet Island study area.
Very strong sign was seen at the bottom (>150 m) on both the BioSonics 120 kHz system and
ship's 50 kHz system. An attempt was made to sample this layer, however, was unsuccessful due
to the depth of the echo sign layer and the amount of wire on the net reel. Tows made in the
same area by the NOAA ship RV Miller Freeman in April 1995 found a similar aggregation of
fish and identified it as age-1 and age-2 walleye pollock (Theragra chalcogramma).
In summer 1995-96 the Isaacs Kidd Midwater trawl (IKMT) was used during midwater
trawl sampling. In the summer 1996 well defined midwater layers were seen at Ugamak Island.
As in 1995, significant midwater fish sign was seen at Ugamak during the day, and was believed
to be due to occasional patches of Pacific herring (Clupea pallasii) and capelin (Mallotus
villosus). However, no midwater tows were conducted at Ugamak Island in 1996 due to the
depth of the target species layer and the limitations of the wire on the net reel (Table 3)
47
Table 3. Midwater and neuston trawls at Ugamak Island, AK, summer and winter 1995-99.
Cruise
95-1
Haul
#
MT6
Haul
Date
7/15/1995
Begin Lat/Long
54 08.70N 164 52.6W
End Lat/Long
54 10.50N 164 54.60W
Location
Avatanak/Ugamak
Net
IKMT
Species
Theragra chalcogramma
Ammodytes hexapterus
Common Name
Walleye pollock
juvenile sandlance
Euphausiids
Calanus copepods
Amphipod
Crab zoea
Gadids, age 0 & 1
# Fish
Hexagrammos sp.
Euphausiids
1
1
95-1
MT7
7/17/1995
54 16.16N 164 55.20W
54 17.42N 164 57.08W
Ugamak
IKMT
97-1
M03
3/8/1997
54 13.85N 164 45.44W
54 45.65N 164 45.11W
Ugamak
HerringTrawl
Hexagrammos sp.
97-1
M04
3/10/1997
54 20.58N 164 48.71W
54 20.60N 164 50.59W
Ugamak
HerringTrawl
Triplops forficata
97-1
N02
3/8/1997
54 14.13N 164 46.86W
54 14.13N 164 47.42W
Ugamak
Neuston
Hexagrammos sp.
Hexagrammos sp.
12
97-1
N03
3/10/1997
54 21.68N 164 53.18W
54 21.68N 164 51.93W
Ugamak
Neuston
Hexagrammos sp.
Hexagrammos sp.
1
97-2
M01
7/11/1997
54 08.43N 164.32.00W
54 07.67N 164 32.68W
Ugamak
HerringTrawl
Ammodytes hexapterus
juvenile sandlance
larval flatfish
Larval gadids
Euphausiids
1
98-1
M01
3/20/1998
54 13.00N 164 28.80W
54 12.67N 164 30.06W
Ugamak
HerringTrawl
98-2
N02
7/8/1998
54 4.07N 164 52.07W
54 04.37N 164 53.78W
Ugamak
Neuston
99-1
N01
3/21/1999
54 12.84N 164 54.04W
54 12.73N 164 55.15W
Ugamak
Neuston
48
Mallotus villosus
Ammodytes hexapterus
2
1
larval capelin
15
Larval fish
Pteropods,Amphipods
1
Larval sandlance
15
During 1997-99 surveys the modified midwater herring trawl was
introduced and enabled sampling of older aged fishes as well as age-0 gadids,
euphausiids, and jellyfish that were captured using the IKMT net. In summer
1997, one midwater trawl was made with the herring trawl at Ugamak Island to
identify echo sign observed in the upper 30 m. The catch from this tow included
larval gadids, flatfish, Pacific sandlance, capelin, euphausiids, and jellyfish.
Larval fishes obtained were frozen for later identification in the laboratory.
During the summer 1998 cruise one midwater tow, two neuston surface tows and
one vertical plankton tow were conducted to verify echo sign. Indication of echo
sign at Cape Sarichef, northwest of Ugamak at the entrance to the Bering Sea,
prompted a midwater tow at 50 m depth. Samples collected from this tow
included very large jellyfish and 0-age walleye pollock. One neuston tow was
conducted at Ugamak Island to verify echo sign near the surface and produced a
catch of pteropods, amphipods, and larval fishes. Otherwise relatively little echo
sign was seen during Ugamak Island survey transects (2) (Table 3).
Ugamak Island surveys were expanded to include winter survey transects
during February/March 1997-99. During the winter 1997 survey at Ugamak
Island two midwater trawls were made with the modified herring trawl and one
with a neuston net. The midwater trawls caught a variety of fish, including adult
walleye pollock, hexagramid fish species, as well as euphausiids and a few jelly
fish.
Winter 1998 surveys rarely indicated strong echo sign during the day and on
few occasions at night. Night time transects at Ugamak revealed faint, scattered
echo sign of zooplankton and fish after 1-2 am. A tow with the herring trawl on a
layer of widely scattered stronger echo sign revealed a catch composition of adult
walleye pollock, larval capelin, and euphausiids.
The winter 1999 survey once again was characterized by sparse echo sign
during the day and on few occasions at night. Night time transects at Ugamak
showed faint scattered echo sign mostly likely that of zooplankton and small fish.
49
A Neuston tow on a vertical layer of strong echo sign at Ugamak showed it was
composed of larval Pacific sandlance and worms (Fig. 2) (Table 3). Very rough
weather prevented midwater trawling during the 1999 survey.
Neuston Tows
Neuston plankton tows were conducted at Ugamak Island during summer
1998 due to large concentrations of feeding seabirds on the surface of the water
and on acoustic signal return of prey concentration near the surface. Species
caught during this tow included pteropods, larval fish and amphipods. During
winter 1997 the neuston tows were made to identify the echo sign being fed upon
by murres and auklets. Catches were generally composed of juvenile hexagramid
fishes, euphausiids, and copepods. During winter 1999 very rough weather
conditions prevented midwater trawling and instead a Neuston tow was made to
verify echo sign with the catch composed of larval Pacific sandlance (Fig. 2)
(Table 3).
All larval fish samples collected from tows were either identified in the
field or by fisheries scientists at the University of Alaska Fairbanks (UAF), or
NMFS Recruitment Processes Program of the Alaska Fishery Science Center in
Seattle, Washington.
Longline Surveys
Diet composition of predatory fish
Large predatory groundfish, such as Pacific cod and Pacific halibut were
utilized as sampling tools in untrawlable areas around Ugamak Island during
50
surveys. Stomach contents of these groundfish predators were assumed to reflect
the abundance of common prey species in the vicinity of Ugamak Island.
Generally, each longline set caught between 0 and 34 fish (Table 4).
Fishes, particularly gadids and osmerids, were a major component of Pacific
halibut diets near the rookery at Ugamak Island in 1996, with 79% frequency of
occurrence. In summer 1996, 5 Pacific halibut and 16 Pacific cod were caught at
Ugamak Island by longline surveys. The Pacific halibut caught averaged 131 cm
in length (range 116-138 cm). Pacific cod sampled by longline averaged 65 cm in
length (range 56-70 cm). Proportionally more Pacific cod contained prey than did
Pacific halibut and both consumed demersal and pelagic prey species (Holladay et
al. 2000). In summer 1996 Pacific cod at Ugamak Island consumed a relatively
large frequency of occurrence of crustacea (81%), other demersal fish (50%), and
small schooling fish (50%), followed by mollusks (44%), other invertebrates
(31%) and flatfish (6%) (Fig. 14). Pacific halibut consumed other demersal fish
(80%), gadids (40%), cephalopods (20%) and mollusks (20%) during summer
1996 (Fig. 15) (Table 5).
In summer 1997 stomach samples were collected from 17 Pacific halibut
and 6 Pacific cod at Ugamak Island during longline surveys. Two additional
halibut were caught during the same time frame by baited hook and line and one
additional cod was caught by beam trawl at Ugamak. The Pacific cod averaged
67 cm in size (range 58.5-75 cm) with no empty stomachs reported. Pacific cod
from this collection consumed crustacea (100%), mollusks (57%), small
schooling fish (43%), benthic invertebrates (42%), other demersal fish (28%) and
51
Table 4. Longline set log at Ugamak Island, AK, 1995-98.
Tow
Date
Time
Area
Beginning
Lat. (N)
LL-1
6/27/95
3:30
Ugamak
54 11.64
164 49.08
6/30/96
4:30
Ugamak
54 07.20
164 28.80
L01
3/10/97
6:10
Ugamak
54 07.20
L1
7/11/97
5:40
Ugamak
LL2
3/20/98
17:15
Ugamak
52
Long. (W)
Lat. (N)
--
Ending
Long. (W)
Dur.
(min)
Depth
(m)
min.
Depth
(m)
max
Pred.Sp.
#Stomachs
--
480
32
Empty
0
54 07.20
164 28.20
120
40
Empty
0
164 28.80
54 07.20
164 28.20
140
40
54 06.60
164 28.50
54 06.60
164 28.19
120
49
54 43.80
164 28.20
54 36.00
164 28.80
120
42
Halibut
Pcod
Halibut
Pcod
Halibut
Pcod
1
8
1
8
4
10
60
62
Table 5. Frequency of occurrence of prey in the diets of Pacific halibut and Pacific cod at Ugamak Island, Alaska from 1995-98
Cephalopod
Crustacea
Mollusca
Other
benthic
invert
56 - 70
(65)
0%
81%
44%
31%
19%
50%
6%
50%
0%
1
116 138
(131)
0%
20%
20%
0%
40%
0%
0%
80%
0%
1
1
(98)
0%
0%
0%
0%
0%
0%
0%
0%
0%
P. Cod
8
1
52 -75
(68)
0%
85%
14%
71%
0%
28%
14%
42%
14%
6/1997
P. Cod
7
0
58.5 75
(67)
14%
100%
57%
42%
0%
43%
0%
28%
28%
6/1997
P.
halibut
23
4
69 160.5
(112)
9%
13%
0%
13%
35%
13%
0%
0%
17%
3/1998
P.
halibut
4
2
94 149
(118)
0%
0%
0%
0%
50%
0%
50%
0%
0%
Date
Predator
N
N
Empty
6/1996
P. Cod
16
0
6/1996
P.
halibut
5
3/1997
P.
halibut
3/1997
Length
range
(cm)
(mean)
Gadidae
Sm.
school
fish
Pleuronectidae
Other
demersal
fish
Unid.
fish
53
Cephalopod
Crustacea
Mollusca
Other
benthic
invert
53 - 76
(61)
10%
40%
10%
20%
20%
10%
0%
50%
20%
12
69 125
(96)
10%
10%
0%
10%
80%
10%
0%
10%
10%
0
46 - 57
(52)
67%
99%
0%
67%
0%
0%
0%
33%
0%
Date
Predator
N
N
Empty
3/1998
P. Cod
10
0
6/1998
P.
halibut
22
6/1998
P. Cod
3
54
Length
range
(cm)
(mean)
Gadidae
Sm.
school
fish
Pleuronectidae
Other
demersal
fish
Unid.
fish
cephalopods (14%) (Fig. 14). Pacific halibut averaged 112 cm in size (range 69160 cm) with 4 empty stomachs reported. Pacific halibut contained
proportionally more prey than Pacific cod and consumed gadids (35%), small
schooling fish (13%), crustacea (13%), other benthic invertebrates (13%), and
cephalopods (9%) (Fig. 15) (Table 5).
During summer 1998 stomach samples were collected from 22 Pacific
halibut and 3 Pacific cod at Ugamak Island during longline surveys. The Pacific
cod averaged 52 cm in size (range 46-57 cm) with no empty stomachs reported.
Pacific cod at Ugamak Island consumed a relatively large frequency of
occurrence of crustacea (81%), other invertebrates (67%), cephalopods (67%),
and other demersal fish (33%) (Fig. 14). Pacific halibut averaged 96 cm in size
(range 69-125 cm) with 12 empty stomachs reported during summer 1998.
Pacific halibut consumed gadids (80%), small schooling fish (10%), other
demersal fish (10%), cephalopods (10%), crustaceans (10%) and other
invertebrates (10%) (Fig. 15) (Table 5).
Longline surveys during winter 1997 collected stomach samples from 1
Pacific halibut and 8 Pacific cod at Ugamak Island. The Pacific cod averaged 68
cm in size (range 52 -75 cm) with 1 empty stomach reported. Pacific cod
consumed a relatively large frequency of occurrence of crustacea (85%), followed
by other invertebrates (71%), other demersal fish (42%), small schooling fish
(28%), flatfish (14%), and mollusks (14%) (Fig. 16). The one Pacific halibut
caught during winter 1997 measured 98 cm in length had an empty stomach (Fig.
17) (Table 5).
During winter 1998 stomach samples were collected from 4 Pacific halibut
and 10 Pacific cod at Ugamak Island during longline surveys. The Pacific cod
averaged 61 cm in size (range 53 - 76 cm) with no empty stomachs reported.
55
56
M
ol
lu
sc
a
H
al
ib
ut
C
ep
ha
lo
po
d
C
ru
O
s
th
ta
er
ce
In
a
ve
rte
br
at
es
S
an
df
is
h
Fl
at
fis
he
s
H
er
rin
At
g
ka
M
ac
k
er
O
th
el
er
de
m
er
sa
S
m
l
.S
ch
oo
lin
g
Sa
lm
on
Pc
od
Po
llo
ck
G
ad
id
ae
% Frequency Occurrence
H
al
ib
ut
C
ep
ha
lo
po
d
C
ru
O
st
th
ac
er
ea
In
ve
rte
br
at
es
M
ol
lu
sc
a
H
er
rin
At
g
ka
M
ac
ke
O
re
th
er
l
de
m
er
Sm
sa
l
.S
ch
oo
lin
g
Sa
nd
fis
h
Fl
at
fis
he
s
Sa
lm
on
Pc
od
Po
llo
ck
G
ad
id
ae
%Frequency of Occurrence
(A)
120
% FO Pcod 1996
% FO Pcod 1997
100
% FO Pcod 1998
80
60
40
20
0
Species Group
(B)
90
80
%FOScats 1996
% FO Scats 1997
70
% FO Scats 1998
60
50
40
30
20
10
0
Species Group
Fig. 14. Pacific Cod stomach contents from longline surveys (A) and Steller
sea lion scat samples (B) Summer 1996-98 at Ugamak Island, Alaska.
M
ol
lu
sc
a
H
al
ib
ut
C
r
u
O
st
th
ac
er
ea
In
ve
rte
br
at
es
Sa
nd
fis
h
Fl
at
fis
he
s
C
ep
ha
lo
po
d
H
er
rin
A
g
tk
aM
ac
ke
O
re
th
l
er
de
m
er
s
S
al
m
.S
ch
oo
lin
g
Sa
lm
on
P
co
d
Po
llo
ck
G
ad
id
ae
% Frequency Occurrence
H
al
ib
ut
C
r
us
O
th
ta
er
ce
In
a
ve
rte
br
at
es
M
ol
lu
sc
a
H
er
rin
At
g
ka
M
ac
k
O
er
th
el
er
de
m
er
Sm
sa
l
.S
ch
oo
lin
g
Sa
nd
fis
h
Fl
at
fis
he
s
C
ep
ha
lo
po
d
Sa
lm
on
Pc
od
Po
llo
ck
G
ad
id
ae
% Frequency Occurrence
(A)
90
80
% FO Halibut 1996
% FO Halibut 1997
% FO Halibut 1998
70
60
50
40
30
20
10
0
Species Group
(B)
90
80
%FO Scats 1996
% FO Scats 1997
% FO Scats 1998
70
60
50
40
30
20
10
0
Species Group
Fig. 15. Halibut stomach contents from longline surveys (A) and Steller
sea lion scat samples (B) Summer 1996-98 at Ugamak Island, Alaska.
57
58
M
ol
lu
sc
a
H
al
ib
ut
C
ep
ha
lo
po
d
C
ru
O
st
th
ac
er
ea
In
ve
rte
br
at
es
S
an
df
is
h
Fl
at
fis
he
s
H
er
rin
At
g
ka
M
ac
k
er
O
th
el
er
de
m
er
sa
S
m
l
.S
ch
oo
lin
g
S
al
m
on
Pc
od
Po
llo
ck
G
ad
id
ae
%Frequency Occurrence
H
al
ib
ut
C
ep
ha
lo
po
d
C
r
u
O
st
th
ac
er
ea
In
ve
rte
br
at
es
M
ol
lu
sc
a
H
er
rin
At
g
ka
M
ac
ke
O
re
th
l
er
de
m
er
sa
Sm
l
.S
ch
oo
lin
g
Sa
nd
fis
h
Fl
at
fis
he
s
Sa
lm
on
Pc
od
Po
llo
ck
G
ad
id
ae
%Frequency Occurrence
(A)
90
% FO Pcod 1997
80
% FO Pcod 1998
70
60
50
40
30
20
10
0
Species Group
(B)
90
% FO Scats 1997
80
% FO Scats 1998
70
60
50
40
30
20
10
0
Species Group
Fig. 16. Pacific cod stomach contents from longline surveys (A) and Steller
sea lion scat samples (B) Winter 1997-98 at Ugamak Island, Alaska.
M
ol
lu
sc
a
H
al
ib
ut
C
ep
ha
lo
po
d
C
ru
O
st
th
a
er
ce
In
a
ve
rte
br
at
es
S
an
df
is
h
Fl
at
fis
he
s
H
er
rin
A
g
tk
aM
ac
ke
O
re
th
l
er
de
m
er
sa
S
l
m
.S
ch
oo
lin
g
S
al
m
on
P
co
d
P
ol
lo
ck
G
ad
id
ae
% Frequency Occurrence
M
ol
lu
sc
a
H
al
ib
ut
C
ep
ha
lo
po
d
C
ru
O
s
th
ta
er
ce
In
a
ve
rte
br
at
es
S
an
df
is
h
Fl
at
fis
he
s
H
er
rin
At
g
ka
M
ac
k
O
er
th
el
er
de
m
er
sa
S
m
l
.S
ch
oo
lin
g
Sa
lm
on
P
co
d
Po
llo
ck
G
ad
id
ae
% Frequency Occurrence
(A)
60
1997 % FO Halibut
50
1998 % FO Halibut
40
30
20
10
0
Species Group
(B)
90
80
1997 % FO Scats
1998 % FO Scats
70
60
50
40
30
20
10
0
Species Group
Fig. 17. Halibut stomach contents from longline surveys (A) and Steller sea
lion scat samples (B) Winter 1997-98 at Ugamak Island, Alaska.
59
Pacific cod at Ugamak Island consumed a relatively large frequency of
occurrence of other demersal fish (50%) crustacea (40%), gadids (20%), other
invertebrates (20%), small schooling fish (10%), cephalopods (10%), and
mollusks (10%) (Fig. 16) (Table 5). Pacific halibut averaged 118 cm in size
(range 94-149 cm) with 2 empty stomachs reported during winter 1998. Pacific
halibut consumed gadids (50%) and flatfish (50%) (Fig. 17) (Table 5).
DISCUSSION
Acoustically derived relative biomass estimation at Ugamak Island
For many years several species of seabirds and pinnipeds in the GOA and
Bering Sea have exhibited signs of food stress such as low productivity, low
recruitment, die-offs, and population declines (Piatt and Anderson 1996). Along
the Alaska Peninsula, and in the EAI, the continental shelf narrows and there is
less shelf habitat for commonly occurring forage fish species that are important to
piscivorous seabirds and marine mammals, such as sandlance and capelin. The
EAI area is notable for its dominance by only a few superabundant species (e.g.
euphausiids, pollock, tufted puffins, shearwaters), a conspicuous scarcity of bank
seabirds (e.g. murres and kittiwakes), a once large population of Steller sea lions,
and a dynamic marine environment in the island passes between the GOA and
Bering Sea. The food web found there is also typical of those found in the middle
and outer shelf domains of the Bering Sea. The advection of fish and plankton to
relatively stationary consumers appears to be an important phenomenon in this
and other ecosystems (e.g. the northern Bering and Chukchi seas; Piatt et al. 1991,
60
1992).
Pelagic juvenile (age 0+) pollock are usually the dominant prey of
piscivorus seabirds in this area during summer (Byrd et al. 1997). Juvenile
pollock advected by prevailing currents in this area represent a fundamentally
distinct food resource for puffins and other seabirds. The closest known
spawning stocks are in the Shumagin Islands, the eastern Bering Sea shelf north
of Unimak Island, and Bogoslof Island (Piatt and Hatch, unpubl. data, Piatt
pers.comm.). While pollock recruitment depends largely on predation and
environmental variability, the functional relationship between young pollock and
their predators is not clear. Recognizing that there are interannual fluctuations in
food webs within habitats, the Ugamak Island system may not reflect historical
patterns of forage fish abundance (Byrd et al. 1997).
Estimation Variance Analysis (EVA)
Geostatistics offer the ecologist a variety of tools to organize and
summarize spatial patterns. Applied statistical methods, such as geostatistics, are
needed in ecological studies for modeling the strength and areal extent of spatial
correlations. The geostatistical toolbox contains many instruments for
characterizing not only the spatial continuity inherent in an organism’s
distribution or the spatial dependence of suspected environmental components,
but also the spatial interdependence between the organism and its environment.
Geostatistical estimation of spatial patterns are more appropriate than
classical random estimation of variance methods due to spatial dependency of the
data. Ecological studies that produce data not amenable to standard parametric
statistical treatment because of spatial or temporal autocorrelation may
significantly benefit from geostatistical analysis. Autocorrelation is a potential
problem in many if not most field sampling strategies, and its presence should be
routinely evaluated. The application of geostatistics to studies with data that
61
exhibit autocorrelation and to studies dealing explicitly with spatial or temporal
patterning may substantially aid their interpretation (Robertson 1987). The EVA
method of variance estimation is proving to be useful for monitoring abundance
estimations over the years, on systematic surveys.
Relative biomass densities from Ugamak Island SMMOCI surveys were
separated by large areas of empty water during some surveys and increasing or
decreasing the intertransect spacing was considered. If relative biomass occurring
between 0 and 3 nm was more uniformly distributed then increasing the
intertransect spacing to greater than 3 nm could have been considered. Higher
relative biomass produced by more uniformly distributed echo sign was seen
during the winter 1999 survey. During most Ugamak Island surveys, however,
relative biomass was more dispersed and patchy between 0-3 nm and if any
intertransect spacing changes were made it would have been better to narrow the
spacing to less than 3 nm. With a greater time and budget allotment for the
SMMOCI project a decrease in transect spacing may have been an option worth
consideration. However, ship-based surveys are expensive and a decrease in
transect spacing would have added more transects to the survey area, decreased
the already limited amount of time for trawling, limited the time to complete an
entire survey of the area, and consequently cost more in ship time and money.
Trawl Surveys
Midwater trawl surveys
Opportunistic midwater trawling during hydroacoustic transect surveys is
a common method of verifying echo sign encountered (Gunderson 1993,
MacLennan and Simmonds 1992). Therefore, midwater trawling was not
62
intended as a quantitative assessment but rather a qualitative assessment of
midwater prey resources during SMMOCI surveys. Midwater trawl survey
results demonstrated that there were prey species present that are important to
Steller sea lions and piscivorous seabirds within the 20 nm NMFS fishery
management area. Prey captured during midwater trawling included species
important to predatory fish species such as pollock, age 1 gadids, hexagramids,
and sandlance, as well as species which are important to those prey (copepods,
euphausiids, age 0 gadids). Additionally these important prey species are present
within the 0 to 50 m depth range of foraging juvenile Steller sea lions. With
increased ship time and budget midwater trawl surveys could have provided a
more quantitative assessment of midwater prey resources available to foraging
Steller sea lions and seabirds.
Bottom Trawl surveys
In 1995 a total of 15 bottom trawls were conducted at Ugamak Island by
the University of Alaska Fairbanks (UAF) as part of the SMMOCI study. During
1996 and 1997 there were 17 and 22 bottom trawls conducted, respectively (Fig.
6). The number of quantitative tows collected at Ugamak from 1995-97 are
indicated in Table A1.
Results from the UAF bottom trawls showed the most abundant species in
the total catch from all years combined were rock sole (Lepidopsetta spp., 4339
specimens, 46.2% of total catch), walleye pollock (Theragra chalcogramma,
1456 specimens, 15.5% of total catch), Pacific halibut (Hippoglossus stenolepis,
927 specimens, 9.9% of total catch), northern sculpin (Icelinus borealis, 618
specimens, 6.6% of total catch), Triglops spp. (332 specimens, 3.5% of total
catch), Gymnocanthus spp. (327 specimens, 3.5% of total catch), Pacific cod
(Gadus macrocephalus, 223 specimens, 2.4% of total catch), slim sculpin
(Radulinus asprellus, 98 specimens, 1.0% of total catch), and arrowtooth flounder
63
(Atherestes stomias, 91 specimens, 1.0% of total catch). Relative catch
composition differed considerably among years. Rock sole was the most abundant
species in 1995. However in 1996, catch of walleye pollock (782 specimens,
55.2% of total catch) exceeded that of rock sole. Although in much lower
abundance than walleye pollock, Pacific cod were found in the same locations
(Fig. 6). Many of these species were present in the diet of Steller sea lions
(Merrick et al. 1997, Sinclair and Zeppelin 2002).
Standardized abundances and CPUE
Catch composition was summarized to illustrate species composition of
bottom trawl catches by year. Interannual differences in catch composition could
only be illustrated in a broad sense due to sampling design and variability in
sampling effort in different depth strata. Catch abundances were standardized to
the number of fish caught per 1000 m2 (Mueter and Norcross 1998) (Table A2).
Differences in relative species composition were seen between years at Ugamak
Island. Gadids dominated catches in 1996, however, few were caught in 1995 or
1997. (Mueter and Norcross 1998; Holladay et al. 2000).
In summer 1995 Ugamak Island bottom trawl surveys caught in order of
abundance, Pacific halibut, rock sole, northern sculpin, and walleye pollock.
Ugamak Island had among the highest abundances of age-0 walleye pollock, with
the average density of over 70 fish per 1000 m2 at stations sampled near the island
in 1996. Pacific cod were also abundant, however, they were found at much
lower abundance than walleye pollock (Mueter and Norcross 1998) . In 1997, 30
fish taxa were caught in bottom trawls (Holladay et al. 2000).
64
Sculpin species occurred in low abundances at Ugamak Island and had no
clear spatial patterns of abundance. Arrowtooth flounder were sparse or not
observed near the Ugamak Island rookery area (Mueter and Norcross 1998) .
In 1997 the most commonly caught fishes at Ugamak Island were rock
sole (76% of tows), Triglops spp. (71% of tows), and northern sculpin (41% of
tows). The five most abundant taxa at Ugamak accounted for 79% of regional
abundance, and included rock sole (30% CPUE), Triglops spp. (13% CPUE),
walleye pollock (12% CPUE), poachers (12% CPUE), and northern sculpin (11%
CPUE). Poachers were also among the most abundant group at Ugamak
(Fig. A25) (Holladay et al. 2000).
Length Frequency Distribution
Length and frequency distributions were compared among years for the
most abundant species caught during bottom trawl sampling. Differences in
abundance between Ugamak Island and other rookeries were accompanied by
clear differences in size and age composition of the most abundant species
(Mueter and Norcross 1998). Most age 1 and older halibut were caught at
Ugamak Island, along with large numbers of age 0 halibut.
Only gadids of age 0, ranging from 15 to 50 mm in size, were caught
during the survey in 1995 and 1996. A large number of age 0 walleye pollock
with a mode of 23 mm fork length were caught in 1996. Pacific cod were also
smaller and more numerous in 1996 at Ugamak Island (Mueter and Norcross
1998).
The bottom trawl sampling gear primarily selected age-0 and age-1 fishes
which were identifiable as clearly separated modes (Mueter and Norcross 1998).
Age-0 gadids were caught by the sampling gear and were smaller and more
numerous in 1996 at Ugamak Island. Size composition differences in sculpins
among years and rookeries existed but were less obvious. Length frequency
65
measurements by species category were not reported for bottom trawling surveys
in 1997.
Species diversity
Species diversity from bottom trawl sampling at Ugamak Island, as
measured by Simpson's complement, was quite variable but did not differ
significantly among years (F=0.133, p=0.875 in 1996; F=2.071, p=0.097 in 1997)
when compared by depth stratum (Mueter and Norcross 1998, Holladay et al.
2000). However, species diversity was significantly different among regions in
1997 within the <40 m depth stratum (F=5.305, p<0.05). Species diversity was
significantly higher at Ugamak Island (0.53 ± 0.14) within the < 40 m depth
stratum than in other study areas. Within the 40-70 m (F=1.143, p=0.357) or
over the 70 m (F=2.607, p=0.067) depth stratum, species diversity was not
significantly different in 1997.
Rough bottom topography and untrawlable or marginally trawlable bottom
substrate was one of the main problems encountered during bottom trawl
sampling, and common in the immediate vicinity of all rookeries sampled during
the SMMOCI study including Ugamak Island. Trawling was limited to even
bottom and small grain substrates, such as gravel, which severely restricted
sampling. Therefore, bias in the data may have occurred as not all fish
communities available to foraging sea lions were sampled. The composition of
juvenile fish communities on rough untrawlable bottom near rookeries is likely to
be different from fish composition at trawlable sites (Mueter and Norcross 1998) .
This situation was confirmed in 1996 by use of a video camera lowered to the
bottom in both trawlable and untrawlable nearshore areas. In reef type substrate
66
or rocky bottom areas video observations did not show many juvenile fishes
comparable in size to fishes caught in bottom trawl samples. Most of the fishes in
this substrate were adults outside of the size range appropriate for juvenile sea
lion prey. Video transects in the Kodiak Island area also showed very few small
fishes on rocky bottom substrate compared to other bottom types (Norcross and
Mueter 1999). Therefore, bias in trawl sampling for juvenile fishes may be
relatively small (Mueter and Norcross 1998).
Bottom trawl sampling gear caught fishes in a very small size range and
which were much smaller in size than those found in the stomach collections from
juvenile Steller sea lions in the mid-1980s. Therefore, the species composition in
the bottom trawl catches may not adequately represent the composition of the fish
community which serves as potential prey for sea lions. The larger size range of
fish consumed by sea lions was difficult to sample. However, the species
composition of bottom trawl catches can serve as an index for the availability of
potential prey in future years. High abundances of rock sole and halibut at
Ugamak Island may have been due to depth and sediment effects. Age 0 and age
1 rock sole and Pacific halibut prefer sandy substrates and shallow depth
(Norcross et al. 1995a, Norcross et al. 1997).
Holladay et al. (2000) suspected that due to species behavior, the bottom
trawl may not have accurately assessed species abundance and distribution of
some fish taxa. For example, Pacific sandlance have a varied distribution; school
in surface water, and bury in beach and possibly deep water sediments (Hart
1980). Generally, the bottom trawl drags over the surface of the sediments and
rarely digs into them. Pacific sandlance may have been collected only when the
bottom trawl dug into the sediments and collected buried individuals. Schools of
age-0 Pacific cod and walleye pollock caught by the bottom trawl had extremely
patchy local distributions (Brenda Norcross, Institute of Marine Science,
University of Alaska, pers. comm. Sept. 17, 2003), and infrequent catches of these
schools may introduce data bias. Juvenile rockfishes, which are generally
67
associated with rocky areas (Kreiger 1993; O'Connell and Carlile 1994) were
avoided in the sampling design (Holladay et al. 2000).
Longline Surveys
Due to the difficulty of sampling the appropriate size range of fishes,
predatory demersal fish sampling began by using longline sets in 1996. In July
1996, predatory fish stomach samples were collected from rough bottom areas
and reflected prey similar in size range to those found in stomach collections of
juvenile sea lions from the 1980's (Mueter and Norcross 1998).
Longline surveys of predatory demersal fish provided a means to sample
rough untrawlable areas in the vicinity of the Ugamak Island sea lion rookery. As
with other sampling methods, longline sampling has its limitations. Unlike
trawling, during which species in the path of the trawl are collected, longline
sampling is biased toward predatory species that are already hungry and once
caught may have time to digest prey items in their stomachs until the set is
retrieved. Although the SMMOCI longline sets were only allowed to soak on
average for less than 3 hours, the digestion factor could lead to an underestimate
of predatory fish diets.
The result that longline sampling did provide however, is that there were
prey species in untrawlable areas around Ugamak Island that are important to
Steller sea lions. Prey species that were found in the stomachs of Pacific cod and
Pacific halibut from those areas included demersal fish, small schooling fish such
as sandlance and gadids, and cephalopods, all of which are potentially important
prey for young Steller sea lions.
68
Steller sea lion Food Habits Research
Predatory fish stomach collections
The high correlation between diet diversity and population change
supports the hypothesis that diet is linked with the Steller sea lion population
decline in Alaska (Merrick et al. 1997). Seabird abundance has also declined in
the areas surveyed. Diet diversity has also been suggested as a potential cause of
the declining population of northern fur seals (Callorhinus ursinus) in the Bering
Sea (Sinclair et al. 1996, Sinclair et al. 1994).
Low abundances of potential sea lion prey in both trawl samples and
predatory fish diets in the eastern part of the study area coincided with the highest
observed declines in sea lion populations between 1994 and 1996 (Richard
Merrick, NMFS, Northeast Fisheries Science Center, pers.comm.) suggesting a
potential link between the availability of bottom fish as prey and sea lion declines
(Mueter and Norcross 1998).
Steller sea lion stomach collections from the mid-1980s showed walleye
pollock as the most common fish prey consumed by juvenile sea lions in the Gulf
of Alaska (Calkins and Goodwin 1988, Merrick and Calkins 1996). Walleye
pollock in the diet of juvenile sea lions ranged in size from 70 mm to 550 mm,
with age 1 pollock at 208 mm as the average size consumed. The number of
pollock over 250 mm was very small (Merrick and Calkins 1996). Small forage
fish, such as capelin, Pacific herring, and Pacific sandlance, were found in 25% of
juvenile sea lion stomachs and were the second most common prey. Flatfish
occurred in the diet of 17.6% of adult sea lions but not in the diets of juvenile sea
lions.
In contrast to the diet composition of juvenile sea lions in the mid-1980s,
collections by Mueter and Norcross (1998) found demersal fishes, particularly
flatfish composed of age-0 rock sole and Pacific halibut. In 1996, age-0 walleye
pollock, ranging in size from 14 to 57 mm in length (average = 28.8 mm)
69
dominated the species composition at Ugamak Island. They concluded that
walleye pollock and other age-0 fish collected potentially serve as food for sea
lions in the following year rather than at the time of sampling.
Small forage fish (capelin, Pacific herring, Pacific sandlance) were the
second most common prey group found in juvenile sea lion stomachs during the
1980s, but were rarely found in SMMOCI bottom trawl collections. Since they
are midwater fish, no Pacific herring were caught and only 68 Pacific sandlance
were caught in 3 years of sampling. However, Pacific sandlance may not have
been adequately sampled by the trawl gear used, due to their elongate body shape
and tendency to bury in the sediment (Mueter and Norcross 1998). The
occurrence of small forage fish at Ugamak Island was confirmed by their presence
in the diet of predatory demersal species such as Pacific cod and Pacific halibut
which were sampled by longline in those areas (Mueter and Norcross 1998).
Based on other collections, three groups of commercially important
species which dominated bottom trawl collections or were important components
of halibut diets served as potential prey for sea lions are gadids, osmerids, and, to
a lesser extent, flatfishes (Merrick and Calkins 1996, Merrick et al. 1997). All of
these groups were abundant in bottom trawls and/or halibut stomachs from the
Ugamak Island region (Table 5, Table A1).
The limited number of sampling years preclude analysis of trends in
interannual fluctuations of commercially important species around Steller sea lion
rookeries (Mueter and Norcross 1998). Comparisons of prey species abundance
collected from bottom trawl samples and prey species consumed by predatory
fishes were hampered by the fact that species caught in bottom trawls were in
fairly good condition and were able to be identified to species level while
predatory fish stomach samples contained specimens exhibiting varying degrees
70
of digestion. For example the prey species category in stomach content analysis
samples referred to as gadids is composed of a combination of both pollock and
cod, while pollock and cod are much easier to differentiate in bottom trawl
samples where samples are freshly caught and digestion is not a factor.
Scat collections
Data on diet composition from stomach samples of juvenile sea lions are
not available for the years of this study for direct comparison with those from the
1980s. Due to the declining population and endangered status of Steller sea lions
in the western stock, stomach collections for food habits research are no longer
possible. Scat (fecal) sample collection is a common method currently utilized to
study pinniped diet, however, the inability to differentiate scat material collected
on rookeries by sea lion age group makes differentiating juvenile sea lion diets
from other age group diets impossible (Bigg and Olesiuk 1990, Merrick et al.
1997, Sinclair and Zeppelin 2002). Regardless, scat material is collected from
rookeries and haulouts and gives a general view of seasonal sea lion food habits.
Steller sea lion diet data analyzed from scat samples collected from 1995 to 1998
at Ugamak Island show similarities to SMMOCI predatory demersal fish stomach
food habits data. Scat samples were collected during summer and winter with
walleye pollock dominating as the most prevalent prey species. During summer
scat collections walleye pollock had the highest frequency of occurrence (51%
FO) followed by salmon (48% FO) and Pacific herring (33% FO). During winter
scat collections walleye pollock was the primary prey species (81% FO), followed
by sandfish (64% FO) and Pacific cod (36% FO) (Merrick et al. 1997, Sinclair
and Zeppelin 2002). Scat samples collected during winter from nearby Aiktak
Island, between 1995 and 1997, contained walleye pollock (83% FO), Pacific cod
(16% FO) and Irish lord (12% FO).
71
Pacific cod showed a summer diet preference for crustacea and other
benthic invertebrates but also contained other demersal fish (28-50% FO) and
small schooling fish (43-50% FO) (Figs. A15-A17). Winter Pacific cod diets
were comprised of mostly crustacea and other benthic invertebrates but also
included other demersal fish (42-50% FO), gadids (20% FO), and small schooling
fish (10-28% FO) (Figs A18-A19) (Table 5).
Although sample sizes were much smaller, longline surveys showed that
Pacific halibut summer diets contained other demersal fish (33-80% FO) and
gadids (walleye pollock and cod, 35-80% FO) (Figs. A20-A22). During winter
halibut diets contained gadids (50% FO) and flatfishes (50% FO) (Figs. A23A24).
Again, species identification comparisons between scat sample specimens,
stomach sample specimens, and bottom trawl specimens were problematic.
Species identification during scat analysis is done using bone fragments. There
are particular bone elements that are diagnostic to species in some prey fishes,
such as pollock and cod. These samples are actually easier to identify to species
by bone elements than pollock and cod specimens found in stomach samples of
predatory fishes. Most of the specimens in predatory fish stomachs have
undergone varying amounts of digestion, but not complete digestion, and thus
have tissue attached whereas bone fragments in scat samples are generally devoid
of tissue. Therefore, the prey species identified from predatory fish stomach
collections are reported as gadids with fewer reported to species level of pollock
or cod and prey species identified from scat collections are more often reported as
pollock, cod or gadids. While species diversity is higher at Ugamak Island than in
other study areas, it is unknown whether or not diversity or abundance are high
enough to sustain foraging juvenile Steller sea lions (Merrick et al. 1997, Sinclair
72
and Zeppelin 2002).
Foraging distribution of Steller sea lions
Reduced prey availability for foraging sea lions may be linked to
environmental changes or commercial fishing impacts, or both, and are a possible
cause of the Steller sea lion population decline (Loughlin and Merrick 1989;
Merrick 1995). Ugamak Island has historically supported one of the largest
Steller sea lion rookeries in the world (Loughlin et al. 1984, Merrick et al. 1988).
Satellite telemetry research conducted by Merrick and Loughlin (1997) showed
that adult female sea lions foraged for approximately the same amount of time in
both summer (breeding) and winter (non-breeding) seasons. However, the time
used during foraging was different between the two seasons. During winter adult
females spent more time at sea, dove deeper, and had greater home ranges than
did adult females during summer. During the breeding season adult females
stayed closer to shore (more time within 20 nm radius) due to dependent young on
the rookery (Merrick and Loughlin 1997).
Juvenile Steller sea lion foraging research
Use of dive and time-depth recording instrumentation has become a
common research method for studying foraging ecology of pinnipeds (Kooyman
et al. 1983; Gentry and Kooyman 1986). The instruments and data may be
retrieved after the animal returns from feeding trips, as is common with use of the
time-depth recorders (TDR) or the data may be transmitted to a satellite and
accessed by the researchers, as in the satellite-linked time depth recorder
(SLTDR) or the newer satellite dive recorder (SDR) instruments (Goebel et al.
1991; Boyd et al. 1994; Merrick et al. 1994; Werner and Campagna 1995).
73
While relatively few data characterizing the general foraging ecology of
young Steller sea lions are available, several studies have occurred in the past few
years focusing on juvenile and young of the year animals using satellite-linked
dive recording instruments.
In terms of prey availability, the most critical time for foraging juvenile
Steller sea lions is most likely to be in the winter when availability of juvenile
fishes is restricted due to the seasonal movements of many species into deeper
waters and off the continental shelf (Merrick and Loughlin 1997; Mueter and
Norcross 1998). Merrick and Loughlin (1997) described diving and trip behavior
of endangered western stock adult and juvenile Steller sea lions. Based on their
telemetry data, and due to concerns over prey availability, NMFS enacted
fisheries management measures to reduce spatial overlap of potential sea lion
foraging areas and impact of fisheries in those areas. Although the precise date of
weaning in Steller sea lions is unknown, it has been hypothesized that weaning
most likely occurs for most animals when pups reach 10-12 months of age, just
prior to the next summer breeding season (Calkins and Pitcher 1982; Trites and
Porter 2002). Diving ability increases throughout the first year of age and
animals 10-12 months of age are capable of diving up to 288 m, although the
average dive depth for this age group is 16.6 m (Loughlin et al. 2002).
Young of the year and juvenile sea lions from Ugamak Island, nearby
Aiktak Island, and vicinity have been instrumented during both winter (Nov.Mar.) and summer (Apr.-July) months (Loughlin et al. 2002; Fadely et al. 2005).
Fadely et al. (2005) found that a majority of foraging young of the year and
juvenile sea lions captured at or in the vicinity of Ugamak Island during winter
months stayed close to shore, and traveled less than 15 nm from shore. They
made dives within the 0-50 m depth range with the majority of dives less than 20
74
m in depth. Dive patterns during summer months showed that young of the year
and juvenile sea lions traveled farther from shore and dove deeper during May
trips, but returned to similar nearshore diving patterns as during winter months for
the months of June and July (Fadely et al. 2005).
Juvenile and young of the year sea lion diving patterns associated with
nearshore trips could be attributed to several different scenarios. Young, not yet
weaned, sea lions could be fully or partially dependent on their mothers and
potentially following them on foraging trips (Trites and Porter 2002). Learning
to forage, independent foraging, or play behavior in association with other young
animals are other possibilities for the nearshore diving patterns displayed by
young sea lions (Fadely et al. 2005).
While it may be tempting to conclude that the nearshore diving patterns of
young of the year Steller sea lions within the Ugamak Island fishery management
area indicate that adequate prey is available, there is no historic (pre-decline era)
baseline prey resource data available for comparison. Biomass abundance and
prey species composition in the Ugamak Island area during the pre-decline era is
unknown, precise impacts of fisheries in the area during pre-decline years are
unknown as well as the effects of localized prey depletion, and the pre-decline
diving and foraging patterns of sea lions at Ugamak Island are unknown.
Knowing the foraging habits and dive patterns exhibited by young Steller sea
lions in the Ugamak Island area and abundance of prey species available locally
by depth range may facilitate detecting or predicting changes in sea lion foraging
patterns, and ultimately the impact those changes could have on juvenile sea lion
survival. Whether the reasons for young Steller sea lion diving activity are
precisely defined at present or not, it is important to know prey resource
abundance and distribution within the nearshore marine environment around
Ugamak Island as it is an important, and perhaps critical, part of the foraging
habitat of young endangered Steller sea lions during winter and summer months.
75
What is known is that Ugamak Island was formerly the largest Steller sea
lion rookery in Alaska and supported approximately 11,000 sea lions during the
breeding season before the population decline began. As of 2004 there were
approximately 1,400 sea lions at Ugamak Island during the summer breeding
season. Between 1994 and 2000 the Steller sea lion population at Ugamak Island
decreased by 21%. Since 2000 the same population has experienced a 40%
increase. Additionally, juvenile sea lions are the age group experiencing the
sharpest decline in the population and could be driving the overall current
population decline in the western stock of endangered Steller sea lions.
Prey species important to Steller sea lions, as well as the prey of those fish
species, were present in the 20 nm fishery managment area surrounding Ugamak
Island during the study. Prey diversity was higher at Ugamak Island during this
study than at other locations surveyed.
Decline in optimal habitat conditions has contributed to the decline of
many species. Detecting changes in habitat is of high ecological significance for
endangered species in general. Indices of relative biomass collected during this
study contribute to the habitat information for the nearshore waters surrounding
Ugamak Island and are the first indices of this type collected not only in the
habitat around Ugamak Island but in the western stock of Steller sea lions.
Results from this study provided baseline data needed to explore the relationships
between biomass density changes and the effects on endangered Steller sea lions
foraging at Ugamak Island, Alaska. Dedicated systematic prey surveys, such as
those conducted during this study, can be used as a monitoring tool on an annual
and seasonal basis to assess a relative index of prey biomass and species
composition available to foraging endangered Steller sea lions, seabirds, and other
species dependent on the nearshore waters surrounding Ugamak Island for
76
survival.
Because walleye pollock are important prey for predators, and support an
enormous commercial fishery, over fishing of pollock has been implicated as a
source of predator population declines (Merrick et al. 1987, Piatt and Anderson
1996). With proposed increased fishing quotas in the central and western
Aleutian islands for important prey species such as Atka Mackerel, proposed
fishery removal experiments, and possible changes in no trawl zones around
rookeries and haulouts, SMMOCI surveys provided an important baseline of data
prior to experimentation. The SMMOCI prey assessment surveys were the only
joint oceanographic and fisheries research surveys being consistently conducted
in the vicinity of Steller sea lion rookeries and haulouts and in NMFS regulated
fishery management areas during 1995-99. In addition to surveys conducted by
NMML, replicate and comparable surveys have been conducted by AMNWR at
nearby Aiktak Island (pers. comm. V. Byrd, USFWS/AMNWR) and by the Gulf
of Alaska Apex Predator Prey study (GAP) in the Kodiak archipelago (pers.
comm. R. Foy University of Alaska Fairbanks). This research is providing
important descriptive information about the nearshore marine ecosystem
including relative prey biomass estimates in these areas from year to year.
Results from the SMMOCI surveys show that these studies can describe
nearshore marine ecosystem components and may ultimately help reveal patterns
that demonstrate the response of top-level predators to fluctuations in prey
resources available to marine mammals and seabirds (Byrd et al. 1997).
Additionally, this research may characterize area conditions adequately enough to
be utilized as a yearly monitoring survey tool for management of declining and
endangered sea lions and seabirds (pers. comm. V. Byrd, USFWS/AMNWR).
However, a longer time series and more ship time for trawling is needed for more
accurate biomass estimations on future surveys at Ugamak Island.
77
General Recommendations
•
Although it is difficult to link direct cause and effect between
closed fishing areas and Steller sea lion population increase or
decrease correlation analyses should be explored.
•
Fishery regulatory management measures should continue in order
to detect changes in prey species diversity and abundance in the 20
nm vicinity of the Ugamak Island rookery.
•
Dedicated summer and winter prey abundance surveys should be
resumed in the Ugamak Island fishery management area as a long
term monitoring tool in trend site areas where Steller sea lion
populations are both stable and declining.
•
Satellite telemetry research being conducted on foraging Steller sea
lions should be accompanied with simultaneous prey surveys.
78
Literature Cited
Alverson, D. L. 1992. A review of commercial fisheries and the Steller sea lion
(Eumetopias jubatus): the conflict arena. Rev. Aquat. Sci. 6:203-256.
Anderson, Paul. J. and John F. Piatt. 1999. Community reorganization in the Gulf
of Alaska following ocean climate regime shift. Mar. Ecol. Progress
Series. Vol. 189:117-123.
Bickham, J. W.; Patton, J. C., and T. R. Loughlin. 1996. High variability for
control-region sequences in a marine mammal: Implications for
conservation and biogeography of Steller sea lions (Eumetopias jubatus) .
J. Mammal. 77(1): 95-108.
Bigg, M.A. and P.F. Olesiuk. 1990. An enclosed elutriator for processing marine
mammal scats. Mar. Mammal Sci. 6:350-355.
BioSonics. 1994. Model Echo Integrator and Model 281 Dual-Beam Processor
Operator’s Manual. BioSonics, Inc. 3670 Stone Way North, Seattle, WA
98103.
Boyd, I. L., J. P. Y. Arnould, T. Barton, and J. P. Croxall. 1994. Foraging
behaviour of Antarctic fur seals during periods of contrasting prey
abundance. J. Animal Ecol. 63:703-713.
Burgess, T. M. and R. Webster. 1980a. Optimal interpolation and isarithmic
mapping of soil properties. I. The semi-variogram and punctual kriging.
Journal of Soil Science 31:315-331.
Byrd, G. V., R. L. Merrick, J. F. Piatt, and B. L. Norcross. 1997. Seabird, Marine
Mammal, and Oceanography Coordinated Investigations (SMMOCI) near
Unimak Pass, Alaska. In Proceedings of the International Symposium on
the Role of Forage Fishes in Marine Ecosystems, University of Alaska Sea
Grant College Program, Rep. No. 97-01. Pg. 351-361.
Calkins, D. G. and E. Goodwin. 1988. Investigation of the declining sea lion
population in the Gulf of Alaska. Unpubl. Rept., AK Dept. Fish & Game,
Anchorage, AK, 76 pp.
Calkins, D. G. and K. W. Pitcher. 1982. Population assessment, ecology and
trophic relationships of Steller sea lions in the Gulf of Alaska. P. 455-546.
79
In Environmental Assessment of the Alaskan Continental Shelf. Final
Reports Vol. 19.
Cochran, W. 1977. Sampling techniques. John Wiley and Sons, third. Ed. New
York, NY.
Cortés, E. 1997. A critical review of methods of studying fish feeding based on
analysis of stomach contents: application to elasmobranch fishes. Can. J.
Fish. Aquat. Sci. 54: 726-738.
Ebbesmeyer, C., D. R. Cayan, D. R. McLain, F. H. Nichols, D. H. Peterson, and
K. T. Redmond. 1991. 1976 step in the Pacific climate: forty
environmental changes between 1968-1975 and 1977-1984. In:
Betancourt, J.L., and V.L. Sharp (eds) Proceedings of the Seventh Annual
Pacific Climate (PACLIM) Workshop, April 1990. California Department
of Water Resources Interageny Ecological Studies Program Tech. Rep. 26:
129-141.
Fadely, B. S., B. W. Robson, J. T. Sterling, A. Greige, and K. A. Call. 2005.
Immature Steller sea lion (Eumetopias jubatus) dive activity in relation to
habitat features of the eastern Aleutian Islands. Fisheries Oceanography
14: 243-258.
Fahrig, L., G. R. Lilly, and D. S. Miller. 1993. Predator stomachs as sampling
tools for prey distribution: Atlantic cod (Gadus morhua) and capelin
(Mallotus villosus). Can. J. Fish. Aquat. Sci. 50(7): 1541-1547.
Fiscus, C. H. and G. A. Baines. 1966. Food and feeding behavior of Steller and
California sea lion. J. Mammal. 47:218-223.
Fritz, L.W. and S. Hinckley. 2005. A critical review of the regime shift-“Junk
Food”-nutritional stress hypothesis for the decline of the western stock of
Steller sea lion. Mar. Mammal Sci. 21(3):476-518.
Frost, K. J. and L. F. Lowry. 1986. Sizes of walleye pollock, Theragra
chalcogramma, consumed by marine mammals in the Bering Sea. Fish.
Bull. (US) 84:192-197.
80
Gentry, R. L. and G. L. Kooyman. 1986. Methods and dive analysis. In Fur
seals, maternal strategies on land and at sea (R. L. Gentry and G. L.
Kooymanm eds.), p. 28-40. Princeton Univ. Press, Princeton, NJ.
Goebel, M. E., J. L. Bengston, R. L. DeLong, R. L. Gentry and T. R. Loughlin.
1991. Diving patterns and foraging locations of female northern fur seals.
Fish. Bull. 89:171-179.
Gunderson, D. R. 1993. Surveys of Fisheries Resources. John Wiley & Sons,
Inc., New York, NY. 248 p.
Gunderson, D. R. and I. E. Ellis. 1986. Development of a plumb staff beam trawl
for sampling demersal fauna. Fish. Res. 4:35-41.
Haney, J.C., D.W. Hood, S.S. Saupe, and D.M. Troy. 1991. Physical processes
and hydrography. In LGL Alaska Research Assoc. Inc., Marine birds and
mammals of the Unimak Pass area: Abundance, habitat use and
vulnerability. Final rep. to Minerals Management Service (Contract 1435-0001-30564) OCS Study MMS 91-0038, Anchorage, AK, pp. 2-1 to 244.
Hart, J. L. 1980. Pacific fishes of Canada. Bull. 180. Fisheries Res. Board of
Canada. Ottawa, Canada, 740 p.
Harwood, J. and J. P. Croxall. 1988. The assessment of competition between seals
and commercial fisheries in the North Sea and the Antarctic. Mar.
Mammal Sci. 4:13-33.
Holladay, B. A., B. L. Norcross, and F. Mueter. 2000. Forage fish abundance
and distribution at Forrester Island, Alaska. Final Contract Report.
Institute of Marine Sciences. School of Fisheries and Oceans, Univ.
Alaska Fairbanks, Fairbanks, AK. 75 pp.
Hollowed, A. B., and W. S. Wooster. 1995. Decadal-scale variations in the
eastern subarctic Pacific: II. Response of Northeast Pacific fish stocks. In
R.J. Beamish (ed.) Climate change and northern fish populations. Can.
Spec. Publ. Fish. Aquat. Sci. 121:373-385.
Holmes, E.E., and A.E. York. 2003. Using age structure to detect impacts on
threatened populations: A case study with Steller sea lions. Conservation
Biology 17:1794-1806.
81
Hood, D.W. 1986. Physical and chemical processes. In J.C. Truett and P.C.
Craig (eds.), Evaluation of environmental information for the Unimak
Pass area, Alaska. Univ. Alaska Inst. Marine Sci. Occas. Publ. 2,
Fairbanks, AK, pp. 183-222.
Hyslop, E. J. 1980. Stomach content analysis - a review of methods and their
application. J. Fish Biol. 17: 411-429.
Jolly, G. M and I. Hampton. 1990. A stratified random transect design for
acoustic surveys of fish stocks, Can. J. Fish. Aquat. Sci., 47(7): 12821291.
Journel, A. G. and C. J. Huijbregts. 1978. Mining geostatistics, Academic Press,
London England.
Kajimura, H. 1985. Opportunistic feeding by the northern fur seal, Callorhinus
ursinus, p. 300-318. IN: J.R. Beddington, R.J.H. Beverton, and D.M.
Lavigne (eds.) Marine mammals and fisheries. Allen and Unwin Publ.,
London, England.
Kinder, T.H. and J.D. Schumacher, 1981. Circulation over the continental shelf
of the southeastern Bering Sea. In D.W. Hood and J.A. Calder (eds.),
Eastern Bering Sea shelf: oceanography and resources. University of
Washington Press, Seattle, WA, vol. 1. U.S. Dept. Commer., NOAA,
OMPA p. 53-75.
Kooyman, G. L., J. O. Billups, and D. W. Farell. 1983. Two recently developed
recorders for monitoring diving activity of marine birds and mammals. In
Experimental biology at sea (A. G. Macdonald and I. G. Priede, eds.) p.
187-214. Academic Press, New York, NY.
Krebs, C. J. 1989. Ecological methodology. Harper & Row, New York, NY.
Krieger, K. J. 1993. Distribution and abundance of rockfish determined from a
submersible and by bottom trawling. Fish Bull. (US) 91: 87-96
82
Krige, D. G. 1981. Lognormal-de Wijsian geostatistics for ore evaluation. South
African Institute of Mining and Metallurgy Monograph Series.
Geostatistics 1. South Africa Institute of Mining and Metallurgy,
Johannesburg, South Africa.
Lavigne, D. M. 1982. Marine mammal-fishery interactions: a report of IUCN
workshop. Trans. N. Am. Wildl. Conf. 47:312-321.
Loughlin, T. R. 1997. Using the phylogeographic method to identify Steller sea
lion stocks. In Molecular genetics of marine mammals (A. Dizon, S. J.
Chivers, and W. F. Perrin, eds.), p. 159-171. Spec. Publ. 3 of the Soc.
Mar. Mammal.
Loughlin, T. R and A. E. York. 2000. An accounting of the sources of Steller sea
lion mortality. Mar. Fish Rev. 62(4):40-45.
Loughlin, T. R. and R. L. Merrick. 1989. Comparison of commercial harvest of
walleye pollock and northern sea lion abundance in the Bering Sea and
Gulf of Alaska, p. 679-700. IN: Proceedings of the international
symposium on the biology and management of walleye pollock,
November 14-16, 1988, Anchorage, AK. University of Alaska Sea Grant
Report AK-SG-89-01.
Loughlin, T. R., A. S. Perlov, V. A. Vladimirov. 1992. Range-wide survey and
estimation of total number of Steller sea lions in 1989. Mar. Mamm. Sci.
83:220-239.
Loughlin, T.R., D.J. Rugh, and C.H. Fiscus. 1984. Northern sea lion distribution
and abundance: 1956-80. J. Wildl. Manage. 48:729-740.
Loughlin, T. R., J. T. Sterling, R. L. Merrick, J. L. Sease, and A. E. York. 2002.
Diving behavior of immature Steller sea lions (Eumetopias jubatus). Fish
Bull. (101)566-582.
Lowry, L. F., K. J. Frost and T. R. Loughlin. 1989. Importance of walleye pollock
in the diets of marine mammals in the Gulf of Alaska and Bering Sea, and
implications for fishery management, p. 701-726. IN: Proceedings of the
international symposium on the biology and management of walleye
pollock, November 14-16, 1988, Anchorage, AK. University of Alaska
Sea Grant Report AK-SG-89-01.
MacLennan, D. N., and E. J. Simmonds. 1992. Fisheries Acoustics. Chapman &
Hall, London, U.K., 325 p.
83
Matheron, G., 1971. The theory of regionalised variables and its applications, Les
Cahiers du Centre de Morphologie Mathématique, no.5, ENSMP, Paris.
211 pp.
Mathiesen, O. A., R. T. Baade and R. J. Lopp. 1962. Breeding habits, growth and
stomach contents of the Steller sea lion in Alaska. J. Mammal. 43:469477.
McNaughton, S. J. and L. L. Wolf. 1973. General ecology. Holt, Rinehart &
Winston, New York, NY.
Merrick, R. L. 1995. The relationship of the foraging ecology of Steller sea lions
(Eumetopias jubatus) to their population decline in Alaska. Ph.D. diss.,
171 p. Univ. Washington, Seattle, WA.
Merrick, R. L., and T. R. Loughlin. 1997. Foraging behavior of adult female and
young-of-the-year Steller sea lions in Alaskan waters. Can. J. Zool.
75:776-786.
Merrick, R. L. and D. G. Calkins. 1996. Importance of juvenile pollock in the diet
of Gulf of Alaska Steller sea lions. In: R. D. Brodeur, P. A. Livingston, T.
R. Loughlin, and A. B. Hollowed (eds.) Ecology of juvenile walleye
pollock (Theragra chalcogramma). U.S. Dep. Commer., NOAA Tech.
Rep. NMFS 126.
Merrick, R. L., M. K. Chumbley, and G. V. Byrd. 1997. Diet diversity of Steller
sea lions (Eumetopias jubatus) and their population decline in Alaska: a
potential relationship. Can. J. Fish. Aquat. Sci. 54: 1342-1348.
Merrick, R. L., T. R. Loughlin and D. G. Calkins. 1987. Decline in abundance of
the northern sea lion, Eumetopias jubatus, in Alaska, 1956-1986. Fish.
Bull. (US) 85:351-365.
Merrick, R. L., R. Brown, D. G. Calkins and T. R. Loughlin. 1995. Comparison of
Steller sea lion (Eumetopias jubatus) pup masses from rookeries with
decreasing and increasing populations. Fish. Bull. U.S. 93: 752-757.
Merrick, R. L., T. R. Loughlin, G. A. Antonelis and R. Hill. 1994. Use of
satellite-linked telemetry to study Steller sea lion and northern sea lion
foraging. Pol. Res. 13:105-114.
84
Merrick, R. L., P. Gearin, S. Osmek, and D. Withrow. 1988. Field studies of
northern sea lions at Ugamak Island, Alaska during 1985 and 1986
breeding seasons. U.S. Dept. Commer., NOAA Tech. Memo. NMFS
F/NWC-143, 60 pp.
Mueter, F. J. and B. L. Norcross. 1998. Availability of commercial fish species
as food for marine mammals Year II. Final Contract Report, NOAA
Saltonstall-Kennedy, NOAA Cont. # NA66FD0043, 92 pp.
NMFS. 1992. Recovery plan for the Steller sea lion (Eumetopias jubatus).
Prepared by the Steller Sea Lion Recovery Team for the National Marine
Fisheries Service, Silver Spring, MD. 92 pp.
Norcross, B. L. and F. J. Mueter. 1999. The use of an ROV in the study of
juvenile flatfish. Fisheries Research 39:241-251.
Norcross, B. L., F. J. Mueter and B. A. Holladay, 1997. Habitat models for
juvenile pleuronectids around Kodiak Island, Alaska. Fish. Bull. (U.S.) 95:
504:520.
Norcross, B. L., F. J. Mueter, M. Frandsen and B. A. Holladay. 1995a.
Availabiity of commercial fish species as food for marine mammals. Final
Contract Report, NOAA Saltonstall-Kennedy, 121 pp.
O’Connell, V. M. and D. W. Carlile. 1994. Comparison of a remotely operated
vehicle and a submersible for estimating abundance of demersal shelf
rockfish in the Eastern Gulf of Alaska. No. Am. J. Fish. Mgmt. 14:196201.
Petitgas, P. 1993. Geostatistics for fish stock assessments: a review and an
acoustic application. ICES Journal of Marine Science, 50:285-298.
Petitgas P. and J. J. Levenez. 1996. Spatial organization of pelagic fish:
echogram structure, spatio-temporal condition, and biomass in Senegalese
waters. ICES J. Mar. Sci. 53:147-153.
Petitgas, P. and A. Prampart. 1993. EVA: a geostatistical software on IBM-PC
for structure characterization and variance computation. ICES CM
1993/D: 65, 55 pp.
Pielou, E. C. 1977. Mathematical ecology. John Wiley & Sons, New York, NY.
85
Piatt, J.F. and P. Anderson. 1996. Response of common murres to Exxon Valdez
oil spill and long-term changes in the Gulf of Alaska marine ecosystem.
In S.D. Rice, R.B. Spies, D.A. Wolf, and B.A. Wright (eds.) Proceedings
of the Exxon Valdez Oil Spill Symposium. Am. Fish. Soc. Symp. 18:720737.
Piatt, J.F., J.L. Wells, A. MacCharles, and B. Fadely. 1991. The distribution of
seabirds and their prey in relation to ocean currents in the southeastern
Chukchi Sea. Can. Wildl. Serv. Occas. Pap. 68:21-31.
Piatt, J.F., A. Pinchuk, A. Kitaysky, A.M. Springer, and S.A. Hatch. 1992.
Foraging distribution and feeding ecology of seabirds at the Diomede
Islands, Bering Strait. U.S. Fish Wildl. Serv. Final Rep. For Minerals
Managment Service, OCS Study MM 92-0041. Anchorage, AK.
Pitcher, K. W. 1981. Prey of the Steller sea lion, Eumetopias jubatus, in the Gulf
of Alaska. Fish. Bull. (US) 79:467-472.
Pitcher, K. W. 1980. Food of the harbor seals, Phoca vitulina richardsi, in the
Gulf of Alaska. Fish. Bull. (US) 78:544-549.
Pitcher, K. W. and D. G. Calkins. 1981. Reproductive biology of Steller sea
lions in the Gulf of Alaska. J. Mammal. 62:599-605.
Ricklefs, R. E. 1973. Ecology. Chiron Press, Newton, MA.
Robertson, Philip G. 1987. Geostatistics in Ecology: Interpolating with known
variance. Ecology, 68(3). pp.744-748.
Rossi, Richard E., David J. Mulla, Andre G. Journel, and Eldon H. Franz. 1992.
Geostatistical tools for modeling and interpreting ecological spatial
dependence. Ecological Monographs, 62(2), pp. 277-314.
Royer, T. C. 1989. Upper ocean temperature variability in the northeast Pacific
Ocean: is it an indicator of global warming? J. Geophys. Res. 94: 1817518183.
Schumacher, J.D., C.A. Pearson, and J.E. Overland. 1982. On the exchange of
water between the Gulf of Alaska and the Bering Sea through Unimak
Pass. J. Geophys. Res. 87:5785-5795.
86
Sea Grant. 1993. Is it food? Addressing marine mammal and sea bird declines:
workshop summary. March 11-14, 1991, Fairbanks, AK. University of
Alaska Sea Grant Report AK-SG-93-01.
Sinclair, E. H. and T. K. Zeppelin. 2002. Seasonal and spatial differences in diet
in the western stock of Steller sea lions (Eumetopias jubatus). J.
Mammol., 83(4): 973-990.
Sinclair, E., T. Loughlin, and W. Pearcy. 1994. Prey selection by northern fur
seals (Callorhinus ursinus) in the eastern Bering Sea. Fish. Bull. U.S.,
92:144-156.
Sinclair, E. H., G. A. Antonelis, B. W. Robson, R. R. Ream, and T. R. Loughlin.
1996. Northern fur seal predation on juvenile walleye pollock. In:
Ecology of juvenile walleye pollock, Theragra chalcogramma. R. D.
Brodeur, P. A. Livingston, T. R. Loughlin, and A. B. Hollowed, (eds.).
NOAA Tech. Rep., NMFS, 126:167-178.
Sokal, R. R. and F. J. Rohlf. 1981. Biometry. Second edition. W.H. Freeman,
San Francisco, CA, USA.
Swartzman, G. L. and R. T. Haar. 1983. Interaction between fur seal populations
and fisheries in the Bering Sea. Fish. Bull. (US) 81:121-132.
Thorsteinson, F. V. and C. J. Lensink. 1962. Biological observations of Steller sea
lions taken during an experimental harvest. J. Wildl. Manage. 26:353-359.
Trangmar, B. B., R. S. Yost, and G. Uehara. 1985. Application of geostatistics
tospatial studies of soil properties. Pages 45-94 in N.C. Brady, editor.
Advances in agronomy. Volume 38. Academic Press, New York, NY,
USA.
Trenberth, K. E. and J. W. Hurrell. 1994. Decadal atmosphere-ocean variations in
the Pacific. Climate Dynamics 9: 303-319.
Trites, A. W. and B. T. Porter. 2002. Attendance patterns of Steller sea lions
(Eumetopias jubatus) and their young during winter. J. Zool. London.
256:547-556.
87
Troy, D.M., J.C. Haney, and T.A. Carpenter. 1991. Zooplankton abundance and
distribution. In LGL Alaska Research Assoc. Inc., Marine birds and
mammals of the Unimak Pass area: Abundance, habitat use and
vulnerability. Final rep. to Minerals Management Service (Contract 1435-0001-30564) OCS Study MMS 91-0038, Anchorage, AK, pp. 3-1 to 340.
Werner, R. and C. Campagna. 1995. Diving behaviour of lactating southern sea
lions (Otaria flavescens) in Patagonia. Can. J. Zool. 73:1975-1982.
Williamson, Neal J. and Jimmie J. Traynor. 1996. Application of a onedimensional geostatistical procedure to fisheries acoustic surveys of
Alaskan pollock. ICES Journal of Marine Science, 53: 423-428.
Wynne, K., R.J. Foy and C.L. Buck. 2005. Gulf Apex Predator-prey Study
(GAP) Final Report. NOAA Grant NA 16FX1270, University of Alaska
Fairbanks, Kodiak, AK. 241pp.
88
List of Appendices
Fig. A1. Transect 1, Strata 1-9, Summer 1995-98, Ugamak Island, Alaska
91
Fig. A2 Transect 2, Strata 1-9 Summer 1995-98 Ugamak Island, Alaska
92
Fig. A3. Transect 3, Strata 1-11, Summer 1995-98 Ugamak Island, Alaska
93
Fig. A4. Transect 4, Strata 1-10, Summer 1995-98 Ugamak Island, Alaska
94
Fig. A5. Transect 5, Strata 1-7, Summer 1995-98, Ugamak Island, Alaska
95
Fig. A6. Transect 6, Strata 1-8 (m), Summer 1995-98 Ugamak Island, Alaska
96
Fig. A7. Transect 7, Strata 1 - 9 (m), Summer 1995-98 Ugamak Island, Alaska
97
Fig. A8. Transect 1, Strata 1-5, Winter 1997-99 Ugamak Island, Alaska
98
Fig. A9. Transect 2, Strata 1-8, Winter 1997-99 Ugamak Island, Alaska
99
Fig. A10. Transect 3, Strata 1-8, Winter 1997-99 Ugamak Island, Alaska
100
Fig. A11. Transect 4, Strata 1-11, Winter 1997-99 Ugamak Island, Alaska
101
Fig. A12. Transect 5, Strata 1-8, Winter 1997-99 Ugamak Island, Alaska
102
Fig. A13. Transect 6 Strata 1-8, Winter 1997-99 Ugamak Island, Alaska
103
Fig. A14. Transect 7, Strata 1-10, Winter 1997-99 Ugamak Island, Alaska
104
Fig. A15. Pacific cod stomach samples from SMMOCI longline surveys and
Steller sea lion scat samples, June 1996 at Ugamak Island, Alaska
105
Fig. A16. Pacific cod stomach contents from SMMOCI longline surveys and
Steller sea lion scat samples, June-July 1997 at Ugamak Island, Alaska.
106
Fig. A17. Pacific cod stomach contents from SMMOCI longline surveys and
Steller sea lion scat samples, June-July 1998 at Ugamak Island, Alaska.
107
Fig. A18. Pacific cod stomach contents from SMMOCI longline surveys and
Steller sea lion scat samples, March 1997 at Ugamak Island, Alaska
108
89
Fig. A19. Pacific cod stomach contents from SMMOCI longline surveys and
Steller sea lion scat samples, March 1998 from Ugamak Island, Alaska
109
Fig. A20. Halibut stomach contents from SMMOCI longline surveys and Steller
sea lion Scat samples, June-July 1996 at Ugamak Island, Alaska
110
Fig. A21. Halibut stomach contents from SMMOCI longline surveys and Steller
sea lion scat samples, July-Aug. 1997 at Ugamak Island, Alaska
111
Fig. A22. Halibut stomach contents from SMMOCI longline surveys and Steller
sea lion scat samples, June-July 1998 at Ugamak Island, Alaska
112
Fig. A23. Halibut stomach contents from SMMOCI longline surveys and Steller
sea lion scat samples, March 1997 at Ugamak Island, Alaska
113
Fig. A24. Halibut stomach contents from SMMOCI longline surveys and Steller
sea lion scat samples, March 1998 at Ugamak Island, Alaska
114
Fig. A25. Bottom Trawl CPUE at Ugamak Island, Alaska, June-July 1995-97 115
Table A1. Location of bottom trawls at Ugamak Island, AK, June-July 1995-98
116
Table A2. Summary CPUE and fish taxa during quantitative beam trawl surveys
at Ugamak Island, AK, summer 1995-97.
119
90
0.016
1995
1996
0.014
1997
1998
Average density (kg/m2)
0.012
0.01
0.008
0.006
0.004
0.002
0
0-10
11-20
21-30
31-40
41-50
51-60
61-70
71-80
81-90
Depth Strata (m)
Fig. A1. Transect 1, Strata 1-9, Summer 1995-98, Ugamak Island, Alaska
91
0.007
1995
1996
0.006
1997
1998
Average density (kg/m2)
0.005
0.004
0.003
0.002
0.001
0
0-10
11-20
21-30
31-40
41-50
51-60
61-70
Depth strata (m)
Fig. A2. Transect 2, Strata 1-9 Summer 1995-98 Ugamak Island, Alaska.
92
71-80
81-90
0.006
1995
1996
0.005
1997
Average density (kg/m2)
1998
0.004
0.003
0.002
0.001
0
0-10
11-20
21-30
31-40
41-50
51-60
61-70
71-80
81-90
91-100
101-110
Depth strata (m)
Fig. A3. Transect 3, Strata 1-11, Summer 1995-98 Ugamak Island, Alaska
93
0.01
1995
0.009
1996
1997
0.008
1998
Average density (kg/m2)
0.007
0.006
0.005
0.004
0.003
0.002
0.001
0
0-10
11-20
21-30
31-40
41-50
51-60
61-70
71-80
Depth strata (m)
Fig. A4. Transect 4, Strata 1-10, Summer 1995-98 Ugamak Island, Alaska.
94
81-90
91-100
0.006
1995
1996
0.005
1997
Average density (kg/m2)
1998
0.004
0.003
0.002
0.001
0
0-10
11-20
21-30
31-40
41-50
51-60
61-70
Depth strata (m)
Fig. A5. Transect 5, Strata 1-7, Summer 1995-98, Ugamak Island, Alaska
95
0.012
1995
1996
0.01
1997
Average density (kg/m2)
1998
0.008
0.006
0.004
0.002
0
0-10
11-20
21-30
31-40
41-50
51-60
Depth strata (m)
Fig. A6. Transect 6, Strata 1-8 (m), Summer 1995-98 Ugamak Island, Alaska.
96
61-70
71-80
0.009
1995
0.008
1996
1997
0.007
Average density (kg/m2)
1998
0.006
0.005
0.004
0.003
0.002
0.001
0
0-10
11-20
21-30
31-40
41-50
51-60
61-70
71-80
81-90
Depth strata (m)
Fig. A7. Transect 7, Strata 1 - 9 (m), Summer 1995-98 Ugamak Island, Alaska
97
0.009
1997
1998
0.008
1999
Average density (kg/m2)
0.007
0.006
0.005
0.004
0.003
0.002
0.001
0
0-10
11-20
21-30
31-40
Depth strata (m)
Fig. A8. Transect 1, Strata 1-5, Winter 1997-99 Ugamak Island, Alaska
98
41-50
0.007
1997
0.006
1998
1999
Average density (kg/m2)
0.005
0.004
0.003
0.002
0.001
0
0-10
11-20
21-30
31-40
41-50
51-60
61-70
71-80
Depth strata (m)
Fig. A9. Transect 2, Strata 1-8, Winter 1997-99 Ugamak Island, Alaska.
99
0.01
1997
0.009
1998
0.008
1999
Average density (kg/m2)
0.007
0.006
0.005
0.004
0.003
0.002
0.001
0
0-10
11-20
21-30
31-40
41-50
51-60
61-70
Depth strata (m)
Fig. A10. Transect 3, Strata 1-8, Winter 1997-99 Ugamak Island, Alaska.
100
71-80
81-90
0.04
1997
0.035
1998
0.03
Average density (kg/m2)
1999
0.025
0.02
0.015
0.01
0.005
0
0-10
11-20
21-30
31-40
41-50
51-60
61-70
71-80
81-90
91-100
101-110
Depth strata (m)
Fig. A11. Transect 4, Strata 1-11, Winter 1997-99 Ugamak Island, Alaska
101
0.007
1997
1998
0.006
1999
Average density (kg/m2)
0.005
0.004
0.003
0.002
0.001
0
0-10
11-20
21-30
31-40
41-50
51-60
Depth strata (m)
Fig. A12. Transect 5, Strata 1-8, Winter 1997-99 Ugamak Island, Alaska
102
61-70
71-80
0.01
1997
0.009
1998
1999
0.008
Average density (kg/m2)
0.007
0.006
0.005
0.004
0.003
0.002
0.001
0
0-10
11-20
21-30
31-40
41-50
51-60
61-70
71-80
Depth strata (m)
Fig. A13. Transect 6 Strata 1-8, Winter 1997-99 Ugamak Island, Alaska
103
0.04
1997
1998
0.035
1999
Average density (kg/m2)
0.03
0.025
0.02
0.015
0.01
0.005
0
0-10
11-20
21-30
31-40
41-50
51-60
61-70
71-80
Depth strata (m)
Fig. A14. Transect 7, Strata 1-10, Winter 1997-99 Ugamak Island, Alaska
104
81-90
91-100
80
% FO Pcod
70
60
50
40
30
20
10
M
ol
lu
sc
a
C
ru
st
O
ac
th
er
ea
In
ve
rte
br
at
es
H
al
ib
ut
C
ep
ha
lo
po
d
Fl
at
fis
he
s
S
an
df
is
h
H
er
rin
A
g
tk
aM
ac
ke
O
re
th
l
er
de
m
er
sa
S
l
m
.S
ch
oo
lin
g
S
al
m
on
P
co
d
P
ol
lo
ck
0
G
ad
id
ae
% Frequency of Occurrence
%FOScats
Fig. A15. Pacific cod stomach samples from SMMOCI longline surveys and Steller sea lion
scat samples, June 1996 at Ugamak Island, Alaska
105
100
% FO Pcod
90
% FO Scats
%Frequency of Occurrence
80
70
60
50
40
30
20
10
M
ol
lu
sc
a
C
ru
O
st
th
ac
er
ea
In
ve
rte
br
at
es
H
al
ib
ut
C
ep
ha
lo
po
d
Fl
at
fis
he
s
S
an
df
is
h
A
tk
aM
ac
ke
re
O
th
l
er
de
m
er
sa
S
l
m
.S
ch
oo
lin
g
H
er
rin
g
S
al
m
on
P
co
d
P
ol
lo
ck
G
ad
id
ae
0
Fig. A16 Pacific cod stomach contents from SMMOCI longline surveys and Steller sea lion
scat samples, June-July 1997 at Ugamak Island, Alaska.
106
100
% FO Pcod
90
% FO Scats
70
60
50
40
30
20
10
M
ol
lu
sc
a
H
al
ib
ut
C
ep
ha
lo
po
d
C
ru
O
st
th
ac
er
ea
In
ve
rte
br
at
es
S
an
df
is
h
Fl
at
fis
he
s
H
er
rin
A
g
tk
aM
ac
ke
O
re
th
l
er
de
m
er
sa
S
l
m
.S
ch
oo
lin
g
S
al
m
on
P
co
d
P
ol
lo
ck
0
G
ad
id
ae
% Frequency of Occurrence
80
Fig. A17. Pacific cod stomach contents from SMMOCI longline surveys and Steller sea lion
scat samples, June-July 1998 at Ugamak Island, Alaska.
107
% FO Pcod
80
% FO Scats
% Frequency of Occurrence
70
60
50
40
30
20
10
M
ol
lu
sc
a
H
al
ib
ut
C
ep
ha
lo
po
d
C
ru
O
st
th
ac
er
ea
In
ve
rte
br
at
es
S
an
df
is
h
Fl
at
fis
he
s
H
er
rin
A
g
tk
aM
ac
ke
O
re
th
l
er
de
m
er
sa
S
l
m
.S
ch
oo
lin
g
S
al
m
on
P
co
d
P
ol
lo
ck
G
ad
id
ae
0
Fig. A18. Pacific cod stomach contents from SMMOCI longline surveys and Steller sea lion
scat samples, March 1997 at Ugamak Island, Alaska
108
80
% FO Pcod
70
% FO Scats
50
40
30
20
10
M
ol
lu
sc
a
C
ru
st
O
ac
th
er
ea
In
ve
rte
br
at
es
H
al
ib
ut
C
ep
ha
lo
po
d
Fl
at
fis
he
s
S
an
df
is
h
H
er
rin
A
g
tk
aM
ac
ke
re
O
th
l
er
de
m
er
sa
S
l
m
.S
ch
oo
lin
g
S
al
m
on
P
co
d
P
ol
lo
ck
0
G
ad
id
ae
% Frequency of Occurrence
60
Fig. A19. Pacific cod stomach contents from SMMOCI longline surveys and Steller sea lion
scat samples, March 1998 from Ugamak Island, Alaska
109
80
% FO Halibut
70
% Frequency of Occurrence
%FOScats
60
50
40
30
20
10
H
al
ib
C
ut
ep
ha
lo
po
d
C
O
ru
th
st
er
ac
In
ea
ve
rte
br
at
es
M
ol
lu
sc
a
S
al
m
on
H
er
A
rin
tk
g
aM
ac
O
ke
th
re
er
l
de
m
e
S
rs
m
al
.S
ch
oo
lin
g
S
an
df
is
h
Fl
at
fis
he
s
P
co
d
P
ol
lo
ck
G
ad
id
ae
0
Fig. A20. Halibut stomach contents from SMMOCI longline surveys and Steller sea lion
scat samples, June-July 1996 at Ugamak Island, Alaska
110
50
% FO Halibut
45
% FO Scats
35
30
25
20
15
10
5
S
al
m
on
H
er
A
ri
tk
aM ng
ac
O
ke
th
re
er
l
de
m
er
S
m
sa
.S
l
ch
oo
lin
g
S
an
df
is
h
Fl
at
fis
he
s
H
al
ib
C
ut
ep
ha
lo
po
d
C
O
ru
th
st
er
ac
In
e
ve
rte a
br
at
es
M
ol
lu
sc
a
P
co
d
0
G
ad
id
ae
P
ol
lo
ck
% Frequency of Occurrence
40
Fig. A21. Halibut stomach contents from SMMOCI longline surveys and Steller sea lion
scat samples, July-Aug. 1997 at Ugamak Island, Alaska
111
80
% FO Halibut
70
% FO Scats
% Frequency of Occurrence
60
50
40
30
20
10
H
al
ib
C
ut
ep
ha
lo
po
d
C
O
ru
th
st
er
ac
In
ea
ve
rte
br
at
es
M
ol
lu
sc
a
S
al
m
on
H
er
A
ri
tk
aM ng
ac
O
ke
th
re
er
l
de
m
er
S
m
sa
.S
l
ch
oo
lin
g
S
an
df
is
h
Fl
at
fis
he
s
P
co
d
G
ad
id
ae
P
ol
lo
ck
0
Fig. A22. Halibut stomach contents from SMMOCI longline surveys and Steller sea lion
scat samples, June-July 1998 at Ugamak Island, Alaska
112
70
% FO Halibut
60
50
40
30
20
10
M
ol
lu
sc
a
H
al
ib
ut
C
ep
ha
lo
po
d
C
ru
O
st
th
ac
er
ea
In
ve
rte
br
at
es
S
an
df
is
h
Fl
at
fis
he
s
H
er
rin
A
g
tk
aM
ac
ke
O
re
th
l
er
de
m
er
sa
S
l
m
.S
ch
oo
lin
g
S
al
m
on
P
co
d
P
ol
lo
ck
0
G
ad
id
ae
% Frequency of Occurrence
% FO Scats
Fig. A23. Halibut stomach contents from SMMOCI longline surveys and Steller sea lion
scat samples, March 1997 at Ugamak Island, Alaska
113
80
% FO Halibut
70
% FO Scats
% Frequency of Occurrence
60
50
40
30
20
10
M
ol
lu
sc
a
H
al
ib
ut
C
ep
ha
lo
po
d
C
ru
O
st
th
ac
er
ea
In
ve
rte
br
at
es
S
an
df
is
h
Fl
at
fis
he
s
H
er
rin
A
g
tk
aM
ac
ke
O
re
th
l
er
de
m
er
sa
S
l
m
.S
ch
oo
lin
g
S
al
m
on
P
co
d
P
ol
lo
ck
G
ad
id
ae
0
Fig. A24. Halibut stomach contents from SMMOCI longline surveys and Steller sea lion
scat samples, March 1998 at Ugamak Island, Alaska
114
Buffalo sculpin
Poachers (3)
Other sculpins (3)
Artedius spp.
Bigmouth sculpin
Unidentified roundfish
Roundfish, other
Flatfishes, other
Pleuronectes spp.
Arrowtooth flounder
Rock sole
Pacific halibut
Pacific sand lance
Ronquils (3)
Lumpsuckers and snailfishes
Species Group
Triglops spp,. (4)
Myoxocephalus spp.
Northern sculpin
Gymoncanthus sp.
Rockfishes
Irish Lords (2)
Walleye Pollock
Pacific Cod
Gadidae
800
700
1995
600
1996
500
1997
CPUE 400
300
200
100
0
Fig. A25. Bottom Trawl CPUE at Ugamak Island, Alaska, June-July 1995-97
115
Table A1. Location of bottom trawls at Ugamak Island, AK, June-July 1995-98 (Flag =quantitative tows (1),
non-quantitative tows (2), and bad tows that were not kept (3); dashes = no data)
Station
1
1
1
3
4
4
5
5
6
6
8
8
8
10
11
1
2
3
4
5
5
5
6
7
8
9
10
11
12
13
14
15
116
Tow
1
2
3
1
BT4-1
BT4-2
BT5-1
BT5-2
BT6-1
BT6-2
BT8-1
BT8-2
BT8-3
BT10-1
BT11-1
1
1
1
1
1
2
3
1
1
1
1
1
1
1
1
1
1
Date
6/25/95
6/25/95
6/25/95
6/25/95
6/25/95
6/25/95
6/25/95
6/25/95
6/25/95
6/25/95
6/26/95
6/27/95
6/27/95
6/27/95
6/27/95
6/27/96
6/27/96
6/27/96
6/27/96
6/27/96
6/27/96
6/27/96
6/27/96
6/27/96
6/28/96
6/28/96
6/28/96
6/28/96
6/28/96
6/29/96
6/29/96
6/29/96
Time
8:25:00
9:55:00
10:25:00
12:05:00
14:35:00
15:08:00
16:18:00
16:55:00
19:14:00
20:45:00
0:10:00
1:20:00
1:57:00
8:28:00
9:30:00
9:34:00
10:36:00
11:53:00
13:00:00
15:38:00
16:32:00
17:22:00
20:23:00
21:14:00
11:18:00
12:35:00
21:08:00
22:19:00
23:15:00
0:01:00
1:28
3:08:00
Beginning
Lat. (N)
54 12.43
54 12.17
54 12.39
54 11.26
54 10.31
54 10.29
54 11.05
54 11.10
54 11.16
54 11.18
54 07.01
54 07.14
54 07.21
54 10.03
54 11.87
54 12.31
54 12.3
54 12.18
54 12.19
54 11.56
54 11.35
54 11.38
54 12.30
54 10.59
54 11.29
54 11.29
54 12.21
54 12.40
54 12.29
54 11.75
54 10.54
54 07.19
Long. (W)
164 47.62
164 47.61
164 47.56
164 47.13
164 47.12
164 46.36
164 45.41
164 45.13
164 46.35
164 45.98
164 43.82
164 44.49
164 45.09
164 47.38
164 48.51
164 47.51
164 47.75
164 47.71
164 47.40
164 45.80
164 46.05
164 45.90
164 47.19
164 46.56
164 45.36
164 45.37
164 47.47
164 47.26
164 47.45
164 47.80
164 46.63
164 45.09
Ending
Lat. (N)
54 12.24
54 12.06
54 12.24
54 11.28
-54 10.74
54 11.04
54 10.86
54 11.16
54 11.22
54 07.14
54 07.32
54 07.32
54 09.90
-54 12.08
54 12.04
54 12.47
54 11.90
54 11.77
54 11.53
54 11.60
54 12.22
54 10.78
54 11.38
54 11.27
54 12.45
54 12.20
54 12.10
54 11.60
54 10.25
54 07.21
Long. (W)
164 47.76
164 47.58
164 47.52
164 47.04
-164 46.68
164 45.24
164 45.72
164 45.54
164 46.02
164 44.46
164 45.30
164 45.36
164 47.16
-164 47.44
164 47.69
164 47.62
164 47.51
164 45.98
164 46.39
164 46.20
164 47.18
164 46.94
164 45.52
164 46.01
164 47.63
164 47.50
164 47.80
164 47.70
164 46.85
164 45.60
Dur.
(min)
10.0
4.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
5.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
3.0
10.0
8.5
10.0
12.0
10.0
10.0
19.0
10.0
10.0
Depth
(m)
min.
25.0
33.0
27.0
51.0
65.0
65.0
68.0
66.0
61.0
62.0
82.0
82.0
88.0
57.0
31.0
28.0
29.0
35.0
30.0
50.0
60.0
59.0
27.0
67.0
66.0
63.0
26.0
24.0
30.0
35.0
67.0
89.0
Flag
1
2
1
3
1
1
2
1
3
16
17
18
39
39
40
40
40
41
42
43
44
45
45
46
47
48
49
50
53
53
54
54
55
56
57
58
59
59
983100101
983100201
983100202
983100301
983100401
983100402
983100501
983100502
983100601
983100602
1
1
1
1
2
1
2
3
1
1
1
1
1
2
1
1
1
1
1
1
2
1
2
1
1
1
1
1
2
6/29/96
6/29/96
6/30/96
7/7/97
7/7/97
7/7/97
7/7/97
7/7/97
7/7/97
7/7/97
7/7/97
7/7/97
7/7/97
7/7/97
7/7/97
7/7/97
7/7/97
7/7/97
7/7/97
7/12/97
7/12/97
7/12/97
7/12/97
7/12/97
7/12/97
7/12/97
7/12/97
7/12/97
7/12/97
7/7/98
7/8/98
7/10/98
7/8/98
7/8/98
7/8/98
7/8/98
7/8/98
7/9/98
7/10/98
4:28:00
5:47:00
1:37:00
0:09:00
0:42:00
1:34:00
2:34:00
2:54:00
3:25:00
4:24:00
5:36:00
15:55:00
17:05:00
17:31:00
17:49:00
19:02:00
20:16:00
22:18:00
23:14:00
13:59:00
14:35:00
15:40:00
15:55:00
18:56:00
19:43:00
20:57:00
21:58:00
22:51:00
23:18:00
23:09:00
0:13:00
8:22:00
1:23:00
2:57:00
3:30:00
4:48:00
5:05:00
8:50:00
9:33:00
54 07.35
54 10.60
54 10.90
54 11.77
54 11.87
54 11.13
54 11.13
54 11.15
54 11.02
54 10.49
54 07.19
-54 12.22
54 12.25
54 12.28
54 12.11
54 12.50
54 07.18
54 07.46
54 12.22
54 12.22
54 10.89
54 11.06
54 11.12
54 11.10
54 10.58
54 10.07
54 07.23
54 07.23
54 07.22
54 07.25
54 07.28
54 07.31
54 10.50
54 10.51
54 11.23
54 11.21
54 12.30
54 12.35
164 45.45
164 45.30
164 45.10
164 46.31
164 46.29
164 46.02
164 45.99
164 46.01
164 45.30
164 46.47
164 44.96
-164 47.24
164 47.20
164 47.51
164 47.38
164 47.54
164 45.48
164 45.46
164 47.57
164 47.52
164 48.14
164 48.04
164 46.16
164 45.00
164 46.64
164 47.36
164 45.23
164 45.26
164 44.98
164 45.27
164 45.43
164 45.37
164 47.04
164 46.90
164 46.28
164 46.07
164 47.40
164 47.58
54 07.42
54 10.22
54 10.80
54 11.95
54 11.92
54 10.98
54 10.92
54 10.90
54 10.85
54 10.65
54 07.36
-54 12.14
164 45.90
164 45.30
164 45.12
164 46.40
164 46.29
164 45.95
164 45.90
164 45.94
164 45.46
164 46.56
164 45.31
-164 47.26
54 12.12
54 12.03
54 12.44
54 07.19
54 07.26
54 12.41
54 12.41
--54 11.14
54 11.03
54 10.51
54 10.28
-54 07.38
54 07.21
54 07.38
54 07.10
54 07.51
54 10.60
54 10.56
--54 12.15
54 12.28
164 47.48
164 47.37
164 47.51
164 44.96
164 45.09
164 47.30
164 47.60
--164 46.36
164 45.36
164 47.07
164 47.55
-164 45.48
164 45.36
164 45.61
164 44.90
164 45.68
164 46.61
164 46.49
--164 47.42
164 47.41
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.0
5.0
10.0
0.0
10.0
10.0
10.0
0.1
10.0
10.0
10.0
10.0
10.0
10.0
5.0
0.3
10.0
10.0
10.0
10.0
10.0
10.0
10.0
10.3
10.0
10.0
10.0
5.0
5.0
84.0
67.0
68.0
54.0
53.0
66.0
65.5
66.5
70.5
68.5
90.5
36.5
33.0
33.5
40.5
25.5
90.0
87.0
32.5
32.5
48.0
43.0
63.5
68.0
68.0
60.0
90.0
90.0
88.0
87.0
84.0
86.0
67.0
67.0
59.0
61.0
29.0
27.0
1
1
3
1
3
1
3
3
1
1
117
983100701
983100801
983100901
983101001
983101101
983101201
983101301
983101302
983101501
118
7/9/98
7/9/98
7/9/98
7/9/98
7/9/98
7/9/98
7/9/98
7/9/98
7/10/98
9:28:00
10:03:00
11:26:00
13:06:00
14:34:00
15:19:00
17:36:00
17:55:00
7:12:00
54 12.21
54 12.14
54 12.02
54 11.74
54 11.13
54 11.01
54 12.08
54 12.12
54 11.05
164 47.86
164 47.21
164 46.02
164 45.60
164 46.22
164 46.21
164 47.69
164 47.78
164 45.28
54 12.20
54 12.06
54 11.07
-54 11.25
54 11.20
-54 12.24
54 11.27
164 47.98
164 47.12
164 45.70
-164 46.27
164 46.25
-164 47.74
164 44.92
5.0
4.0
5.5
1.0
5.0
7.3
0.3
4.0
10.0
30.0
38.0
47.0
58.0
59.0
60.0
35.0
34.0
66.0
1
2
3
3
3
1
3
3
1
Table A2. Summary CPUE and fish taxa during quantitative beam trawl surveys
at Ugamak Island, AK, summer 1995-97. (CPUE = catch per unit effort)
1995
1996
1997
Fish cumulative CPUE
(# fish/1000 m2)
Range
Average + Std Dev
1.4 - 154.1
25.9 ± 46.1
.76 - 782.3
49.9 ± 155.7
1.8 - 211.9
38.0 ± 54.7
# Total Fish taxa
23
28
30
Range
Average + Std Dev
1 - 12
6.3 ± 3.6
4 - 15
7.0 ± 3.2
1 -10
5.3 ± 2.8
# Tows
12
17
17
# Fish taxa/tow
Tow depth (m)
Range
Average + Std Dev
6 - 90
55.9 ± 26.4
24 - 90
52.9 ± 22.0
26 - 91
58.5 ± 21.4
119