UNIVERSITY OF HAWAI'I LIBRARY

DIET DYNAMICS AND TROPHIC RELATIONS OF LAYSAN AND

BLACK-FOOTED ALBATROSSES ASSOCIATED WITH PELAGIC

LONGLINE FISHING

A THES IS SUBM ITTED TO THE GRADUATE DIVISION OF THE

UN IV ERS ITY OF HAWAI'IIN PARTIAL FULFILLMENT OF TI-IE REQUIREMENTS FOR THE DEGREE OF

MASTER OF SCIENCE

IN

ZOOLOGY (ECOLOGY. EVOLUTION AND CONSERVATION 1310LOGY)

AUG UST 2008

By Jeremy R. Bisson

Thes is Comm ittee:

David Duffy. Chairperson Sheila Conant

Brian Popp

We certify that we have read this thesis and that, in our opinion, it

is satisfactory in scope and quality as a thesis for the degree of

Master of Science in Zoology (Ecology, Evolution, and

Conservation Biology).

THESIS COMMITIEE

ii

Abstract

Commercial fishing is a source of food for many types of seabirds, including albatrosses, but the importance of commercial fishing may vary between fishing methods, locations and affected species. Longline fishing in the north Pacific has expanded and affects Laysan Albatrosses and Black-footed Albatrosses through incidental bycatch, but the importance oflongline fishing to the diets of these albatrosses is unknown. I analyzed fishery observer data to determine whether there are differences in fishery association behavior between Laysan and Black-footed Albatrosses and whether they differ in their amount of catch scavenging effort. I analyzed the digestive tract contents of tuna and swordfish longline-associated albatrosses to determine the fishery component ofthe albatross diets. Last, I analyzed the stable nitrogen isotopic compositions oflongline­associated birds and birds from colonies, as well as prey, to determine whether there might be broader, time-averaged effects on the albatross diets. Black-footed Albatrosses were more abundant around vessels than were Laysan Albatrosses and swordfish longline fishing attracted more birds than did tuna longline fishing. Black-footed Albatrosses also scavenged more swordfish catch, but not necessarily in greater proportion than expected. Laysan Albatrosses ate more fish bait and these two species also differed in their naturally acquired diets, with Laysan Albatrosses consuming more T. borealis. Albatross lil~ values differed between species and Black-footed Albatrosses differed between fishing methods, but it is unclear whether these differences in nitrogen isotopic compositions were fishery related because of overlap in the lil~ values among fishery and naturally acquired prey. Differential fishery resource use was evident, even when the albatrosses were exploiting the same resource, indicating that these species should not be given equal treatment in management decisions. Managers should consider the effect of longline fishing on the diets of Lay san and Black-focited albatrosses separately when making decisions that may change fishing practice. Black-footed Albatrosses should receive the most attention because they are the most affected species, while swordfish operations appear to have the greatest affect on the albatross diets.

iii

List of Figures

Figure 1: Relative Abundances of Lays an and Black-footed Albatrosses Associated with

Swordfish Operations during November 2004 - January 2006 ........................................ 16

Figure 2: Relative Abundances of Lay san and Black-footed Albatrosses Associated with

Tuna Operations during November 2004 - January 2006 ................................................. 17

Figure 3: Monthly Total Number and Percent of Total Landed Dead Swordfish

Scavenged by Albatrosses during June 2004 - March 2006 ............................................. 18

Figure 4: Monthly Summary of Scavenging by Laysan and Black-footed Albatrosses on

Swordfish Catch during December 2004 - March 2006 ................................................... 19

List of Tables

Table 1: Prey Items of 52 Longline-SaIvaged Laysan Albatrosses .................................. 21

Table 2: Prey Items of26 Longline-Salvaged Black-footed Albatrosses ......................... 24

Table 3: Laysan and Black-footed Albatross /)I~ values for Albatrosses Salvaged from

Longline Operations and Hawaiian Breeding Colonies ••..................•......••....................... 28

Table 4: Albatross Prey /)I~ values ................................................................................. 29

Iv

Table of Contents

Abstract .............................................................................................................................. iii List of Figures .................................................................................................................... iv List of Tables ..................................................................................................................... iv Chapter 1: Introduction and Background ............................................................................ 1

Laysan and Black-footed Albatrosses ............................................................................. 2 North Pacific Albatross Diets .......................................................................................... 3 Hawaii Based Pelagic Longline Fishing .......................................................................... 4 Potential Pelagic Longline Related Resources for Albatrosses ....................................... 5

Chapter 2: Methods ............................................................................................................. 7 Introduction ..................................................................................................................... 7 Sampling .......................................................................................................................... 7 Age and Sex of Long line Associated Albatrosses ........................................................... 8 Relative Abundances of Long line Associated Albatrosses ............................................. 8 Albatross Scavenging on Longline Catch ....................................................................... 9 Digestive Tract Content Analyses ................................................................................. 10 Diet Overlap Analysis ................................................................................................... 11 Nitrogen Isotope Analyses ............................................................................................. 12

Albatross Analyses ..................................................................................................... 14 Prey Analyses ............................................................................................................. 15

Chapter 3: Results ............................................................................................................. 15 Relative abundance of Longline Associated Albatrosses .............................................. 15 Albatross Scavenging on Longline Catch ..................................................................... 17 Laysan Albatross Digestive Tract Contents .................................................................. 19 Black-footed Albatross Digestive Tract Contents ......................................................... 23 Diet Overlap Analysis ................................................................................................... 26 Nitrogen Isotope Analyses ............................................................................................. 26

Albatross Analyses ..................................................................................................... 26 Prey Analyses ............................................................................................................. 28

Chapter 4: Discussion ....................................................................................................... 30 Chapter 5: Conclusion ....................................................................................................... 35 Acknowledgements ........................................................................................................... 37 References ......................................................................................................................... 39

v

Chapter 1: Introdnction and Background

High seas squid and large-mesh drift-net fishing operations of the North Pacific once

supplied an important food resource for Laysan Albatrosses (Phoebastria immutabilis) and Black

footed Albatrosses (P. nigripes) until these fisheries were banned in 1991 (Gould et aI. 1997).

Since drift net fishing ceased, longline hook operations have continued throughout the oceanic

habitats of Laysan and Black-footed albatrosses. but the importance of scavenging longline

fisheries waste to Laysan and Black-footed albatrosses is poorly understood.

Understanding how commercial fishing practices affect seabird diets is important to

managing both seabirds and fisheries because fishing can either positively or negatively affect

seabird populations (Furness 2002, Gonzalez-Solis 2003, James and Stahl 2000, Pedrocchi 2002,

Roland et aI. 2008, Votier 2004). This thesis focuses on determining the importance of pelagic

longline fishing operations to the diets of Lays an and Black-footed albatrosses, to assist fishery

managers in improving management decisions regarding pelagic iongline fishing that might

affect albatrosses.

The first question I addressed was whether Laysan and Black-footed albatrosses

associated with longline operations have diets that are fishery acquired or natural. I addressed

this question through digestive tract content and diet overlap analyses. Results were compared

with those of Gould et al. (1997), Harrison el al. (1983) and Pitman et al. (2004) to determine if

diets may have changed over time. Next, I examined whether tuna and swordfish longline

operations may have differing effects on Laysan and Black-footed trophic relations. Stable

nitrogen isotope analyses were used to estimate the trophic positions of the albatrosses and their

prey, and ultimately to determine the potential effects of these two different types of fishing

operations on the bird's diet. The last question I addressed was whether there were differences in

fishery-associated behavior between Laysan and Black-footed albatrosses. Fisheries observer

data were used to determine ifthere are inter-specific differences in the relative abundances and

catch-scavenging behavior between Laysan and Black-footed albatrosses.

Laysan and Black-footed Albatrosses

Laysan and Black-footed albatrosses are distributed almost exclusively within the North

Pacific. Both species have colonies on islands and atolls throughout the subtropical North Pacific

from Japan to Mexico (McDermond and Morgan 1993, Pitman 1988, Rice and Kenyon 1962a).

Black-footed Albatrosses breed from October to mid-June and Laysan Albatrosses breed from

late October to mid-July (Rice and Kenyon 1962b).

At sea, Laysan and Black-footed albatrosses disperse throughout the North Pacific,

though Black-footed Albatrosses generally are found more easterly (Gould and Piatt 1993).

During the summer, non-breeding Laysan Albatrosses are mostly found in pelagic waters along

the Aleutian Chain and in the Gulf of Alaska, while Black-footed Albatrosses concentrate in the

California Current along the continental shelf. During the incubation and brooding periods,

Laysan and Black-footed albatrosses forage predominantly in the subtropical waters surrounding

the breeding colonies, but brooding Laysan Albatrosses also forage from subtropical waters and

in the transition zone chlorophyll front (TZCF) (Fernandez et al. 2001, Hyrenbach et af. 2002).

The TZCF is a feature of all major oceans where the high chlorophyll-containing subarctic

waters submerge beneath the low chlorophyll containing subtropical waters. This creates

concentrations of prey that attract apex predators from several classes of marine animals

(Polovina et aI. 2001).

An expansion of foraging range coincides with the rearing period, which frees adults

from guarding chicks, with Black-footed Albatrosses making longer trips to the continental shelf

2

of North America, and Laysan Albatrosses making longer trips north to the Aleutian Islands and

the Gulf of Alaska (Fernandez et al. 2001, Hyrenbach et af. 2002). Hyrenbach and Dotson

(2003) found that post-breeding Black-footed Albatrosses spent considerable time in the TZCF.

The presence of Japanese Black-footed Albatrosses salvaged from Hawaii longline operations

(included in this study) indicates that their foraging range overlaps with the foraging range of

Hawaiian Black-footed Albatrosses (Walsh and Edwards 2005).

North Pacific Albatross Diets

Previous research on the diets of Laysan and Black-footed albatrosses used a variety of

methods including analyses of forced regurgitated samples from breeding colonies, digestive

tract content analysis, stable isotope analysis and voluntarily regurgitated pellet content analysis.

The first significant study used samples acquired through forced regurgitation of chicks shortly

after they were fed (84%), and from adults returning from foraging trips (16%) at several

northwestern Hawaiian Islands during 1978-1980 (Harrison et al. 1983). Black-footed

Albatrosses fed predominantly on flying fish eggs and squid. Squid were most important to

Laysan Albatrosses, especially from the family Ommastraphidae (Harrison et af. 1983).

Gould et af. (1997) used digestive tract content and nitrogen isotope analyses to

investigate the diets of non-breeding Laysan and Black-footed albatrosses salvaged in 1990 and

1991 from high seas drift net fishing. Both species obtained a significant portion of their diet by

feeding on squid, fish and offal made available from this fishery. Based on an index of relative

importance (IR!), Gould et af. (1997) found that the Laysan Albatross diet consisted primarily of

cephalopods (%IRI = 76.0) with neon flying squid (Ommastrephes burtrumz) (%IRI = 73.8), the

target species ofthis fishery, the most important food. Fish (% IRI = 19.4) was the next most

3

important food, with Pacific pomfret (Bramajaponica) (%IRI 10.5), a non-target but fishery­

related species, the third most important

Gould et af. (1997) found that Black-footed Albatrosses also consumed more

cephalopods (%IRI = 89.3), and neon flying squid (%IRI = 83.9) was the most important food.

Fish (% IRI = 7.7) was next most important, with Pacific pomfret (%IRI = 7.3) third most

important. Nitrogen isotope analyses of albatrosses and their prey revealed that the exploitation

of fishery related food sources increased the trophic positions of both species. The fishery­

acquired diet consisting primarily of neon flying squid had much higher nitrogen isotope ratios

than their naturally-acquired diet. Black-footed Albatrosses were determined to feed on average

at a whole trophically level higher than Laysan Albatrosses (Gould et af. 1997).

Pitman et af. (2004) investigated the cephalopod portion of the diet of Lay san Albatrosses

breeding at Guadalupe Island. Mexico, through an examination of pellets regurgitated by chicks

and collected in June 2000. They found that the birds fed on 22 species of cephalopods from 12

families. Based on the percentages of identified beaks, these included Histioteuthis hoyfei

(2Q.4%), Taonius borealis (13.8%), Gonatus pyros (I 1.2%) and Galiteuthis? spp. A (11.0%).

Because these species are primarily found in deep water, Pitman et af. (2004) suggested that

Laysan Albatrosses from Guadalupe Island fed more heavily on diurnally scavenged squid then

on live squid.

Hawaii Based Pelagic Longline Fishing

Pelagic longline fishing operations based out of Hawaii target either tuna (Thunnus spp.)

or swordfish (Xiphius gfadius). Fishing occurs entirely in deep water, often outside the US

Exclusive Economic Zone. The main component of a longline operation is the horizontal

4

mainline which is suspended below the surface by floats. The length of the mainline varies from

one to 60 miles. Branchlines drop from the sagging mainline between the floats, and reach

various depths depending on where along the mainline they are attached. Branchlines are usually

weighted and have either a wire or monofilament leader between the hook and weight. Typically

700 to 3,000 hooks are deployed each set, and vessels usually only set gear once per day. After

gear deployment, the lines are left for several hours to allow enough time for fish to find the bait.

Then, the mainline is retrieved onto a hydraulic reel and branchlines are retrieved manually.

Catch is predominantly pelagic billfish, tuna and sharks.

Both Hawaii and California swordfish operations use Argentine squid (ntex argentinus)

for bait and light sticks as additional fish attractants, gear is set shallow (without the aid of a line

shooter) and set during dusk or at night, then hauled in the next morning (Xi et aI. 1997). Tuna

operations occur in oligotrophic subtropical waters both north and south of Hawaii so they

almost invariably fish farther south than do swordfish operations. Bait used by tuna operations

throughout the study time period were Pacific saury (Cololabus saira) and sardines (Sardinops

sagax). Hooks are set deep, aided by a line shooter that adds sag. Approximately 20-30 hooks are

set between floats (Beverley et al. 2003). Tuna gear is usually set in the morning and hauled

back through the night (Xi et al. 1997).

Potential Pelagic Longline Related Resources for Albatrosses

The types and amount of food made available by pelagic longline operations for

albatrosses are diverse and differ significantly between tuna and swordfish operations.

Albatrosses feed on unusable discarded bait that is too small or mangled. They steal bait off

hooks while the hooks are located at or just below the surface, whether the gear is being set or

5

hauled. If they miss, they still have a chance at obtaining a piece of discarded bait. Bait stealing

while gear is being hauled, requires tremendous skill and timing because the bait usually gets

stolen off hooks that are being dragged behind the vessel while it is moving and while the lines

are actively being pulled in by crews. Throughout the haul, offal is made available to albatrosses.

Offal from swordfish targeting vessels primarily consists of the heads and guts of

swordfish. Swordfish heads, when bills are removed, tend to float (B. McNamera, pers. comm.

2005). Albatrosses can also feed on lipid-rich swordfish livers and gonads, and possibly on

swordfish stomachs that often contain very large squid such as diamondback squid

(Thysanoteuthis rhombus) or neon flying squid (pers. obs.). Little offal was made available to

albatrosses by tuna-targeting vessels because tuna were not gutted.

Non-target fish of little commercial value are discarded and available to albatrosses

attracted to both tuna and swordfish operations. However, the availability of targeted fish is

limited to the amount of time the caught animal spends in the water before being retrieved. Dead

swordfish which have a propensity to float are almost exclusively available to albatrosses

associated with swordfish operations. On the other hand, tuna do not tend to float and are not

known to be scavenged by seabirds. The likelihood that an albatross will feed on a dead

swordfish probably increases when swordfish suffer damage prior to death, such as large shark

bites, cookie cutter shark (Isistius spp.) bites, or squid damage. This is because swordfish skin is

tough and would be difficult for albatrosses to penetrate on their own. While there are

differences in the types of food provided to albatrosses by tuna and swordfish longline

operations, it is unclear whether these differences matter to albatrosses.

6

Chapter 2: Methods

Introduction

I used fisheries observer data, digestive tract content analysis and stable isotope analyses

to investigate the influence of pelagic longline fishing on albatross diets. Data collected by

fisheries observers were analyzed to determine the relative abundances of albatrosses associated

with pelagic longline operations, the relative frequency oflongline-related resource use and if

there is a difference in longline related resource use between Laysan and Black-footed

albatrosses. I used digestive tract content analysis to determine what the albatrosses had recently

eaten and I used nitrogen isotopic analysis to determine whether there might be broader

differences in diet that potentially could be influenced by pelagic longline fishing.

Sampling

Albatrosses used in this study were salvaged from North Pacific pelagic longline fishing

operations based out of Hawaii and from several breeding colonies in Hawaii where accidental

deaths occurred. These birds were used for digestive tract content analysis, muscle nitrogen and

carbon isotopic analysis. age and sex determination. Hawaii longline fishery observers

successfully salvaged 21 Laysan and 23 Black-footed albatrosses from pelagic longline fishing

vessels which were targeting tuna, during 2002-2003. California longline fishery observers

successfully salvaged three Laysan and nine Black-footed albatrosses from pelagic longline

fishing vessels targeting swordfish during 2002-2003. A research experiment testing seabird

deterrents during two pelagic longline cruises targeting tuna salvaged 18 Laysan Albatrosses and

two Black-footed Albatrosses during 2002, and 29 Laysan Albatrosses and two Black-footed

Albatrosses during 2003. The U.S. Fish and Wildlife Service (USFWS) donated six salvaged

7

Laysan Albatrosses from Midway Atoll National Wildlife Refuge, one salvaged Black-footed

Albatross from French Frigate Shoals, Pacific Remote Islands National Wildlife Refuge

Complex, and 28 salvaged Laysan Albatrosses from Kilauea Point National Wildlife Refuge.

The Hawaii State Department of Natural Resources Division of Forestry and Wildlife donated

muscle collected from 20 salvaged Laysan Albatrosses and one Black-footed Albatross salvaged

from breeding colonies at Kure Atoll.

Age and Sex of Longline Associated Albatrosses

The age classification of albatrosses was based on the degree of involution of the Bursa

of Fabricius (Broughton 1994). Specifically, each albatross was determined to belong to one of

three age classes depending on the size of its bursa offabricius: <50 mm2 = breeding-age; >75

mm2 and < 500 mm2 = pre-breeding; and >600 mm2 = newly-fledged. Albatrosses were sexed

based on examination of gonads. Too few albatrosses were examined to effectively determine

whether a sex bias existed. However, these data could be used to determine if there are

differences in diet between sexes.

Relative Abundances of Long line Associated Albatrosses

The NOAA Fisheries Service places fisheries observers on pelagic longline vessels to

monitor fishing operations for protected species interactions. Observers were tasked with

collecting relative abundance data to assess the effectiveness of seabird deterrents. I used these

data to determine if there were differences in the relative abundances of the two species of

albatrosses associated with swordfish and tuna operations.

8

Starting in 2004 and continuing through January 2006, NOAA fisheries observers

watched the full first hour of setting operations for seabird interactions. Additionally, the

observers scanned and recorded seabirds for 5 minutes at the beginning of setting operations and

again after 30 minutes (whether the birds were interacting with fishing operations or not). Five­

minute seabird observation periods were repeated at the beginning of every hour during the haul.

Data recorded by observers during scan counts included low, best and high estimates for all

observed seabird species. The relative abundances of Lay san and Black-footed albatrosses

associated with tuna and swordfish operations were based on the average ofthe maximum best

estimate for each species per longline set. Paired t-tests were performed to determine whether

there were differences in the relative abundances between Laysan and Black-footed albatrosses

for swordfish and tuna operations (11 =0.05). Also, paired t-tests were used to determine whether

there were differences in the relative abundances ofLaysan and Black-footed albatrosses

associated with swordfish or tuna operations (11 =0.05).

Albatross Scavenging on LongJine Catch

Catch data collected by NOAA Fisheries Service observers deployed on swordfish

targeting longline vessels were used to determine the amount of albatross depredation on

long line catch. Observers watching entire hauling operations recorded damage that occurred to

each swordfish landed, including damage caused by albatrosses. Evidence of albatrosses

scavenging on swordfish was based on the size and shape of the bite marks on the fish left by the

albatrosses. Observers were trained to differentiate damage caused by albatrosses from sharks,

squid and marine mammals. The proportions of swordfish with bird damage were summarized

monthly during June 2004 to March 2006.

9

NOAA Fisheries Service observers also recorded observations of albatrosses actively

feeding on swordfish catch, and in most cases, the species of albatross observed was recorded.

The numbers of albatrosses observed scavenging catch and identified to species were

summarized monthly from December 2004 through March 2006. A paired t-test was performed

on these data to determine ifthere are differences in the numbers of albatrosses scavenging catch

between Laysan and Black-footed albatrosses (a =0.05).

Digestive Tract Content Analyses

The stomachs from undamaged longline-salvaged Laysan and Black-footed albatrosses

were examined for digestive tract contents. Colony-salvaged albatrosses were examined but

excluded from digestive tract content analysis because nearly all of them had empty stomachs.

Food items were identified, weighed and sorted taxonomically. Inorganic items were noted but

excluded from diet analysis. The identification of squid was based on comparisons of lower

beaks with reference collections and keys, using characteristics oflower beaks (Clark 1986).

While some species have upper beaks with diagnostic characteristics that could aid

identification, the ability to use them is not consistent among alI species so their use would have

bias results. Upper beaks were used as a means of counting individuals. Examples of lower beaks

were selected and sent to William Walker (NOAA Fisheries Service National Marine Mammal

Laboratory, Seattle, WA) for confirmation of the identification of common species and for

assistance with the identification of rare and difficult species, and ultimately for the development

of a reference collection. Whole fish, eggs and skeletons were identified using various keys

(Clark 1986, Clothier 1950, Collette et al. 1983, Tyler 1980, Wisner 1974), with assistance from

10

Bruce Mundy (NOAA Fisheries Pacific Islands Fishery Science Center). All otoliths, were

identified by W. Walker (NOAA Fisheries National Marine Mammal Laboratory).

Counts offood items were based on the minimum number of individuals possible from

the body parts and their fragments present to maintain consistency. For instance, if three lower

squid beak halves were found in a stomach, I assumed a minimum of at least two individuals

even though there may have been three. However, if there were also three whole upper beaks

within the same digestive tract, then I assumed there were two individuals of the species

identified by the lower beak halves, as well as one individual of an unidentified species. All

unidentifiable material, including plastics and stomach oil, were discarded and excluded from

analyses. For each taxonomically distinct group that occurred, the percent frequency of

occurrence (F), percent wet mass (M), and the percent number of items (N) were calculated.

From these, an index of relative importance (lRI), following a modification by Gould et al.

(1997), was calculated using the formula: IRI = %F (%N + %M) (pinkas et al. 1971). For each

food group, %IRI was calculated by dividing the IRI for the group by the total1RI.

Diet Overlap Analysis

Diet overlap was investigated using the Schoener (1970) index on the %1RI for each food

item, and separately, for fish %IRI and cephalopod %IRI, using the formula: 100(1-Y.~ I Pxi-Pyi

I) with Pxi and Pyi being the %IRI for each food item of Lay san and Black-footed albatrosses

respectively. Differences between the %IRI for each food type were ranked to determine the

food types that contributed the most to dietary differences. Wallace (1981) found that the

Schoener Index was the best diet overlap index when there is an absence ofinformation about

the availability of resources. Wallace (1981) also recommended Schoener Index be used in

II

conjunction with the average of volume percentages because other measures such as frequency

and number percentages tend to reduce the importance of volume or weight percentages, which

he suggested are better measures of the caloric importance of diet. However, albatross digestive

tracts frequently contain numerous squid beaks and otoliths. While squid beaks and otoliths

alone do not contribute much to volume or weight, they do represent food items with significant

calories. An IRI which takes into account number, weight and frequency information is likely to

best approximate diet for estimating overlap between the two albatross species.

Nitrogen Isotope Analyses

Stable nitrogen and carbon isotopic analyses may indicate trophic differences in diet

because there is enrichment in 1~ by-3.0-3.59'oo, and in 13C by -0.7-1%0 ofan animal's tissues

over its diet, which can correspond to an increase in trophic level (Deniro and Epstein 1978,

198 I). However, stable nitrogen isotope analyses also indicate feeding in different regions

because of regional or latitudinal differences in lil~ and 1i13C values at the bases offood chains

(Minami et al. 1995, Takai et al. 2000).

All samples for isotopic analysis were oven dried for 72 hours at 60· C, homogenized by

hand using a mortar and pestle and all but 20 of the albatross muscle samples were lipid­

extracted using automated solvent extraction (ASE 200, Dionex Corporation). Methods followed

manufacturer recommendations. Briefly, hexane was used as the solvent and the machine was set

for two cycles at a pressure of 1200 psi, six minutes of 100·C step, five minutes of static step,

40% of the volume was flushed and then purged for one minute. Lipids were initially removed

for carbon isotopic analysis oftissues because lipids are depleted in 13C and tissues having

variable amounts of lipids will cause reduced accuracy and precision of Ii 13C values (post et al.

12

2007). However, examination of carbon isotopic variations detennined that the carbon isotope

data could only distinguish the same diet differences found between albatross groups with the

largest differences in nitrogen isotopic composition, so they provided little additional

infonnation and will be excluded from further discussion. Post et al. (2007) found a small

(0.25%0 mean difference) but significant effect oflipid extraction on animallll~ values, which

correlated with % lipid, but concluded that this effect is not much greater than typical analytical

error. On the other hand, Pinnegar and Polunin (\999) found lipid extraction did not significantly

change fish muscle Cil~ values. The results of these studies likely differ because Post et al.

(2007) used both whole animal and muscle tissues in their analyses with the intent of capturing

extremes of lipid content, while Pinnegar and Polunin (1999) used fish muscle which contains

very little fat when compared with whole animals. So, nitrogen isotopic analyses oflipid-

extracted albatross muscle were combined with albatross muscle non-lipid-extracted samples

with the expectation that lipid-extraction would not significantly affect results. Fish muscle and

egg samples were lipid extracted using the same methods used for the birds. Other tissue types

including squid muscle, squid beaks and insect bodies were prepared without lipid extraction

because they have relatively low lipid content. Between 0.3-0.4 Jlg of each sample were analyzed

using a Carla Erba NC2500 Elemental AnalyzerlFinnigan MAT ConFloIIIFinnigan MAT DeltaS

mass spectrometer located at the University of Hawaii Stable Isotope Biogeochemistry

Laboratory. The standards for nitrogen and carbon were air and V-PDB respectively. Stable

nitrogen isotope ratios are reported in Ci-notation:

a. Cil~ or 1l13C = {(R..mpIJRstandard)-I} x loJ b. R.t.nm.ro = l~llN ratio for AIR (0.0036765) and 13C/12C ratio for V-PDB (0.0020672)

13

Muscle tissue from one adult Laysan Albatross from Kure Atoll was analyzed repeatedly to

detennine precision (n=8, Mean 61~ = 11.75%0, SD = 0.02%0). Seven juvenile Laysan

Albatrosses from Kure Atoll were used to detennine within population precision (n=7, Mean

61~ = 13.020/00, SD = 0.63%0). All ofthese birds were salvaged during July 2004 and had

acquired all of their hard feathers. In addition, glycine in which the nitrogen isotopic composition

had been previously detennined (61~ = 11.25%0 vs. AIR) was used as an internal reference

throughout this study to ensure accuracy (n=33, Mean 61~ = 11.34%0, SD = 0.17%0).

Collectively, these data can be used to estimate the analytical error of the results of this study.

Conservatively, the analytical accuracy and precision of the isotopic analyses is taken as:::: 0.20/00.

The variability in the 61~ values of juvenile Laysan Albatrosses from Kure Atoll is more than

three times the uncertainty in the glycine internal reference compound and is thus interpreted as

environmental rather than analytical uncertainty.

Albatross Analyses

Albatross pectoral muscle issues were obtained from frozen salvaged albatrosses for

nitrogen isotopic analysis. All of the albatrosses except one were breeding age adults so analyses

were only conducted on adults. Differences between sexes in diet as indicated by a difference in

61~ values were analyzed for Laysan and Black-footed albatrosses that were salvaged from tuna

targeting longline operations using student t-tests (11 =0.05). Insufficient sample size prevented

analysis of differences between sexes in the albatrosses salvaged from swordfish targeting

operations. I tested for differences in 61~ values between Laysan and Black-footed albatrosses

salvaged from tuna targeting longline vessel (with sexes pooled if not found to be significantly

different in the previous test) using a one-way ANOVA (11 = 0.05). I tested for differences in

61~ values between Black-footed Albatrosses salvaged from tuna and swordfish targeting

14

vessels using a student t-test (a = 0.05). Insufficient sample size prevented analysis of the effect

of fishing operation type on Laysan Albatrosses. I tested for differences in lil~ values among

Laysan Albatrosses salvaged from different colonies using an ANOVA (a = 0.05) and a Tukey's

pairwise comparison to determine whether the colonies differed from each other individually.

Use of this test assumes the data are normally distributed and have equal variances.

Prey Analyses

Representative samples of albatross prey were collected for nitrogen isotopic analyses

from collections from albatross stomachs and included whole fish, whole squid and beaks, as

well as whole fish and squid collected by pelagic longline fisheries observers while at the fishing

grounds. Representative samples offish bait were purchased from a local supplier of bait to

vessels. Argentine squid was purchased from a local grocer. Beaks and muscle from several

individuals of Onyclwteuthis borealis-japonicus were analyzed simultaneously to estimate

natural differences in 1l1~ values between beaks and muscle. This step was taken because only

beaks ofthe potentially important squid species T. borealis were available, and knowing if

systematic differences in 1l1~ values between beak and muscle of O. borealis-japonicus exist

might help estimate the potential effect of T. borealis to the albatross diets.

Chapter 3: Results

Relative abundance of Longline Associated Albatrosses

The relative abundances of both Laysan and Black-footed albatrosses associated with

swordfish operations were significantly greater than at tuna operations (dr- 14, p<0.01) (dr- 14,

15

p<O.OI) (fig 1,2). The relative abundances of Black-footed Albatrosses were significantly greater

than Laysan Albatrosses associated with either tuna or swordfish operations (df.= 22, p<O.OI)

(df.= 16, p<O.O 1) (fig 1,2). Both species were more abundant around longline operations during

winter months (fig 1,2).

Figure 1: Relative Abundances of Laysan and Black-footed Albatrosses Associated with Swordfish Operations during November 2004 - January 2006.

25 • Black-footed o Laysan .... 20 2i

-

(/) (!)

15 III -

e +-' (II 10 :9_

e:( (!)

EC/) 5 ::J (!) c

E= .- 0> ~ C 0 2.3

-

-

~I,L_ -I I I I I I J I I I I I I

16

Figure 2: Relative Abundances of Laysan and Black-footed Albatrosses Associated with Tuna Operations during November 2004 - January 2006.

'-CI.l C. !II CI.l

~ e

2

~ 1

- -« CI.l EC/) :::l CI.l E p. .- C> X c: CO 0 0 ~-I

-

-

- L~~ L _ JL ___ LL~ I I I I I I I I I I J I I I I

Albatross Scavenging on Longline Catch

• Black-footed DLaysan

The highest percent of swordfish with albatross damage occurred in Jaouary (2005) but

the month with the most damaged swordfish was March (2005) (fig 3). Observations of Black-

footed Albatrosses feeding on dead swordfish catch outnumbered observations of Lay san

Albatrosses feeding on dead swordfish (Fig 4) (df-=II. p<O.OI).

17

Figure 3: Monthly Total Number and Percent of Total Landed Dead Swordfish Seavenged by Albatrosses during June 2004 - March 2006.

200 5.1

Percent of Total Dead Swordfish

6.1 4.0

100 2.5 2.8

1.8

~ 5.72.4

5.6 05 4.6 ~~O~~D 0 ~ ~ o

7.18.0

1.800 1:3~O~ 0

18

Figure 4: Monthly Summary of Scavenging by Laysan and Black-footed Albatrosses on Swordfish Catch during December 2004 - March 2006.

90 DLaysan "0

80 • Black-footed ~ :Jl 70 .c 60 0 VlJ: 50 (I),s

III III 40 0 0 il § 30

JJ .c 0)

20 J ______ JJ __ « c -'6 III (I) 10 "0 (I) I-U- 0

Laysan Albatross Digestive Tract Contents

I examined the digestive tracts of 52 longline-salvaged Laysan Albatrosses. A total of

245 prey items were identified. belonging to six fish families, 12 cephalopod families and two

other invertebrate families (Table I). Cephalopods were the most important food for Laysan

Albatrosses and ranked highest in percent number of items (%N = 71.8), percent frequency of

occurrence (%F =98.1) and percent IRI (%00 =65.6), and second for percent mass (%M = 27.0).

Fish were the next most important group with the highest ranking in percent mass (%M =72.9)

and second highest ranking in percent number of items (%N = 23.3), percent frequency of

occurrence (%F =57.7) and percent index of relative importance (%IRI = 34.2). Invertebrates

19

other than cephalopods (%IRI = 0.25) did not contribute significantly to the diet. T. borealis

(%IRI = 33.43) was the highest ranking food. Unidentified squid were next (%IRI = 23.10).

followed by S. sagax (%IRI = 16.53). C. saira (%IRI = 10.90). unidentified fish (%IRI = 5.78)

and 1. argentinus (%IRI = 3.67).

20

Table 1: Prey Items of 52 Longline-Salvaged Laysan Albatrosses

Percent Percent Percent Number Percent Index of

Frequency of of Items Mass Relative Pre;r: Grou~ Occurrence {245} {1,424 gl Im~ortance

Fish Unidentified Fish 25.00 5.71 4.02 5.78 Clupeidae Sardlnops saga:< 15.38 4.08 41.14 16.53 Myctopbidae Unidentified Myctophidae 5.77 1.22 0.05 0.18 Ceraloscopelos warmlngll 3.85 0.82 0.14 0.09 Lampanyctus spp. 3.85 0.82 0.09 0.08 Dlaphus spp. 1.92 0.41 0.00 0.02 Bolinlchlilys spp. 5.77 1.22 0.04 0.17 Molidae Ronzanla laevls 1.92 0.82 1.34 0.10 Seomberesoeldae Cololabus salra 15.38 5.31 24.52 10.90 Exoeoetidae Exocoetus vol/lans 1.92 0.41 0.00 0.02 Unidentified Exocoetidae 3.85 1.22 0.20 0.i3 Unidentified Exocoetidae (eggs) 3.85 0.82 0.00 0.07 CaproJdae AnI/gonia eos 1.92 0.41 1.39 0.08

Cephalopods Teuthid, unidentified 48.08 15.92 4.30 23.10 Cranebladae raanlus borealis 51.92 26.94 0.16 33.43 Leachla dlslocala 1.92 0.41 0.00 0.02 Gal/leulhls spp. 3.85 0.82 0.04 0.08 Megalacranchla spp. 1.92 0.41 0.00 0.02 Hlstloteu thldae Hlsllaleulhls spp: 19.23 6.12 0.04 2.82 Hlslloleulhls hoylel 5.77 1.63 0.10 0.24 Hlslloleulhls corona 1.92 0.41 0.00 0.02 Ommastrepbldae Unid. Ommastrephidae 5.77 2.04 1.94 0.55 Slhenoleulhls oualanlensls 1.92 0.41 0.03 0.02 Ommmostrephes bartramll 1.92 0.41 0.49 0.04 lllex argentinus 7.69 1.63 18.49 3.68 Enoplotenthldae Anclstrochelrus lemeurl 1.92 0.41 0.00 0.02

21

Percent Percent Percent Number Percent Index of

Frequency of of Items Mass Relative Prel: Groul! Occurrence {24~ {1,424 !U Iml!ortance Cblroteuthldae Chlroteuthls spp. 3.85 1.22 0.00 O.ll Chlroteuthls calyx 3.85 0.82 0.00 0.07 Cyeloteutbidae Cycloteuthls a1dmushkinl 1.92 0.41 0.01 0.02 Psyeroteutbldae Discoteuthis spp. 1.92 0,41 0.00 0.02 Mastigoteutbldae Unidentified Mastigoteuthid 3.85 1.63 0.01 0.15 Mastlgoteuthls spp. 3.85 0.82 0.00 0.07 Oetopoteutbldae Octopoteuthls dele tron 1.92 0.41 1.17 0,07

Onyeboteutbldae Onychoteuthls spp. 3.85 1.63 0.00 0.15 Moroteuthls lonnbergl 1.92 0.41 0.18 0,03

Gouatldae Gonatus spp. 5.77 1.63 0,0] 0.22 Berryteuthls anonychus 5.77 4.49 0.00 0.62 AlIoposldae AI/oposus mol/Is 1.92 0.41 0.00 0.02

Other Invertebrates Unidentified invertebrate 3.85 0.82 0.00 0.07

Crustacea Unidentified shrimp 1.92 0.41 0.05 0.02

Inseeta Gerldae Halobates serlceus 1.92 3.67 0.00 0.17

22

Black-footed Albatross Digestive Tract Contents

I examined the digestive tracts of26 longline-salvaged Black-footed Albatrosses. The

digestive tracts of 12 other longline-salvaged Black-footed Albatrosses were damaged by hooks

and were excluded from digestive tract content analysis. A total of 158 prey items were

identified belonging to 4 fish families. 9 cephalopod families and 2 other invertebrate families

(Table 2). Cephalopods were the most important food for Black-footed Albatrosses and ranked

highest in percent number ofitems (% N = 67.1). percent frequency of occurrence (% F = 96.2)

and percent IRI (% IRl = 71.4). and second in percent mass (% M = 20.8). Fish were the next

most important group with the highest ranking in percent mass (% M = 79.2) and second in

percent number ofitems (% N = 13.3). percent frequency of occurrence (% F =50.0) and percent

index of relative importance % IRl = 23.7}. Other invertebrates (% IRl = 4.98) were relatively

insignificant in their contribution to the diet. Unidentified squid (% IRI = 33.43) were the highest

ranking food. T. borealis (% IRI = 23.56) was next highest, followed by unidentified

Histioteuthis (% IRI = 10.16). flying fish eggs (unid. Exocoetidae) (% IRI = 8.69). C. saira (%

IRl = 6.46) unidentified fish (% IRI = 4.47) and S. sagax (% IRI =3.64).

23

Table 2: Prey Items of 26 Longline-Salvaged Black-footed Albatrosses

Percent Percent Percent Number Percent Index of

Frequency of of Items Mass Relative Pre~ Groul! Occurrence ~158} {445 g2 Iml!ortance

Fish Unidentified Fish 23.08 3.80 4.45 4.47 Clupeldae Sardlnops sagm: 7.69 1.27 18.86 3.64 Myetophldae Bollnlchlhys spp. 3.85 0.63 0.00 0.06 Seomberesocldae Cololahus salTa 7.69 1.90 33.83 6.46 Exocoetldae GypselUTUs spp. 3.85 0.63 0.00 0.06 Unidentified Exocoetidae 3.85 0.63 0.52 0.10 Unidentified Exocoetidae (eggs) 15.38 2.53 21.50 8.69 Stemoptycbldae Argyropeilcas spp. 3.85 0.63 0.05 0.06 Gooatostomatldae Unid. Gonatastomatidae 3.85 0.63 0.00 0.06 Coogerldae Congerspp. 3.85 0.63 0.00 0.Q6

Cephalopods Teuthid, unidentified 61.54 18.99 3.52 32.56 Craocbladae Taonlus boreails 46.15 21.52 0.19 23.56 Leachla dis/ocala 11.54 1.90 0.00 0.52 Hlstloteutbldae Hlslloleuthls spp. 38.46 8.23 3.01 10.16 Hlslloteulhls hoylel 3.85 2.53 0.02 0.23 Ommastrephldae Unidentified Ommastrephidae 3.85 0.63 0.00 0.06 JIIex argent/nus 3.85 0.63 13.25 1.26 Eooploteutbldae Anclslrochelrus lesueurl 3.85 0.63 0.00 0.06 Cblroteutbldae Ch/roteuthls calyx 3.85 0.63 0.02 0.06 Mastlgoteutbldae Unidentified Mastigoteuthid 7.69 1.90 om 0.34 Mastlgoteuthls spp. 7.69 1.27 om 0.23 Oetopoteutbldae Octopoleuthls dele Iron 7.69 1.90 0.02 0.35

24

Percent Percent Percent Number Percent Index of

Frequency of ofitems Mass Relative Pre~ Groul! Occurrence {I5S} !445 gl Iml!ortance Onycbotentbldae Onycholeulhls spp. 11.54 2.53 0.09 0.71 Moroleulhls ionnbergl 7.69 1.27 0.58 0.33 Gonatldae Gonatus spp. 15.38 2.53 0.05 0.93

Other Invertebrates Crustacea Lepadldae Lepesspp. 3.85 1.90 0.00 0.17 Insecta Gerldae Halobales serleeus 11.54 17.72 0.01 4.81

25

Diet Overlap Analysis

There was a Schoener Index diet overlap of 66.8 % between Laysan and Black-footed

albatrosses. The largest contributions to dietary differences in descending order were due to

Laysan Albatrosses consuming more S. sagax, T. borealis and unidentified squid, and Black­

footed Albatrosses consuming more flying fish eggs and unidentified Histioteuthls spp.

Nitrogen Isotope Analyses

Albatross Analyses

The lil~ mean and standard error (SE) of albatrosses grouped by species, by sex and by the type

of fishery with which they were associated when caught, or the colony where they died, are

reported in Table 3. Breeding age male Laysan Albatrosses salvaged from tuna operations were

enriched in I~ by 0.4%0 relative to breeding age female Laysan Albatrosses salvaged from tuna

vessels (df= 47, p=O.OI). No significant difference in lil~ values was found between male and

female Black-footed Albatrosses salvaged from tuna operations (df= 15, p=0.44), so male and

female Black-footed Albatrosses are pooled in subsequent analyses. Pooled male and female

breeding age Black-footed Albatrosses associated with tuna operations were enriched in I~ by

1.1 ro. relative to breeding age male Laysan Albatrosses associated with tuna operations and

1.5%0 relative to female breeding age tuna-associated Laysan Albatrosses (df= 72, p< 0.01).

Black-footed Albatrosses associated with swordfish operations were enriched in ISN by 1.6%0

relative to tuna-associated Black-footed Albatrosses. Analysis of variance indicated that the

lil~ values of breeding age Laysan Albatrosses salvaged from the various colonies differed

significantly (df= 42, p<O.O I), but Tukey's pairwise comparisons indicated that only Kure Atoll

26

and Kauai birds differed from each other, and that the birds from Midway were intermediate and

overlapped the confidence intervals of the 1i1~ values of both Kauai and Kure Atoll.

27

Table 3: Laysan and Black-footed Albatross Ii[~ values for Albatrosses Salvaged from Lon~line 0l!erations and Hawaiian Breedin~ Colonies Species Source Sex n Mean SE

Laysan Albatross Tuna Male 32 13.4 0.1

Laysan Albatross Tuna Female 21 13.0 0.1

Laysan Albatross Swordfish Pooled 3 14.3 0.6

Laysan Albatross Kure Atoll unknown 13 12.2 0.2

Laysan Albatross Midway unknown 4 12.5 0.5

Laysan Albatross Kauai unknown 26 13.0 0.1

Black-footed Albatross Tuna Male 9 14.4 0.2

Black-footed Albatross Tuna Female 13 14.5 0.3

Black-footed Albatross Tuna Female I 15.7 • (pre-breeding)

Black-footed Albatross Swordfish Pooled 9 16.0 0.3

Black-footed Albatross French Frigate Shoals unknown 14.4 *

Prey Analyses

The /il~ values of 47 actual and potential food items were analyzed (Table 4). S. sagax

had the highest mean /il~ value of the fish bait used during this study. Lanternfishes and flying

fish eggs had the lowest /il~ values for natural prey. Octopoteuthis deletron (a squid) had the

highest average /ilsN values for natural prey. Combined beak and muscle analyses of the squid

Onychoteuthis showed that muscle was enriched in I~ 3.6%0 relative to beak.

28

Table 4: Albatross Pr!:r 1l1~ values

Prey type n Mean SE (Mean) Minimum Maximum

Ceraloscope/us 3.5 • 3.5 3.5

Scomber japon/cus 15.3 • 15.3 15.3

Exocoetus voillam 7 9.5 0.8 6.9 11.7

Prognicl/rys gI/be rll 2 10.9 1.2 9.7 12.1

Flying fish eggs 2 7.8 0.1 7.7 7.8

Ha/obales serlceus 2 7.2 0.3 6.9 7.5

JIIex argent/nus 9.9 • 9.9 9.9

Oclopoleuthis dele Iron 11.9 • 11.9 11.9

Onycholewhls (muscle) 4 6.6 0.04 6.5 6.7

Onycholeuthls (beak) 4 3.0 0.4 2.3 4.2

C%/abus salra 6 8.0 0.3 7.4 9.0

Sardlnops saga:< 6 14.0 0.2 13.2 14.5

Mega/ocranchl (beak) 8.4 • 8.4 8.4

Taonlus borealis (beak) 14 10.1 0.7 5.8 15.0

29

Chapter 4: Discussion

Fishery association analyses found that Black-footed Albatrosses were more abundant

around Hawaii based longline operations than were Laysan Albatrosses. Also, swordfish

operations attracted more albatrosses than did tuna operations. These results are consistent with

reports that Black-footed Albatrosses had higher incidental catch rates in the fishery and that

North Pacific albatrosses are more commonly associated with swordfish operations than with

tuna operations (Cousins et af. 200 I). The results are also consistent with the conclusion of

Gould et af. (1997) that Black-footed Albatrosses might be more susceptible to scavenging from

fishing operations because of their greater dependence on scavenging. Fisher (2007) found that

Black-footed Albatrosses use TZCF water in the winter, where and when swordfish operations

occur, while Laysan Albatrosses used subtropical waters more to the south of the TZCF. The

diets of Black-footed Albatrosses would be most affected by pelagic longlines since Black­

footed Albatrosses were relatively more abundant than Laysan Albatrosses around vessels.

Similarly, swordfish operations would have a greater effect on the diets of both species than tuna

operations.

Results from digestive tract content and observer data analysis indicated that Laysan and

Black-footed albatrosses acquired different diets when associated with pelagic longline fishing

operations. SpecificalIy, Laysan Albatrosses consumed more fish bait, while Black-footed

Albatrosses scavenged off swordfish carcasses. However, Black-footed Albatrosses were also

more abundant than Laysan Albatrosses around boats, so it is unclear whether Black-footed

Albatrosses proportionally scavenged more swordfish carcasses.

30

Longline-associated Laysan and Black-footed albatrosses also differed in the most

important naturally-acquired portions of their diets. T. borealis was the most important naturally­

acquired food for both species but was consumed in greater proportion by Laysan Albatrosses,

while Histioteuthis was consumed in greater proportion by Black-footed Albatrosses.

Differences in foraging habitat or foraging behavior could both explain differences in the

naturally acquired diets between longline-associated Laysan and Black-footed albatrosses but

Ballance et aZ. (2001) suggested diet differences are generally best explained by differences in

foraging habitat. If diet differences are best explained by differences in foraging habitat rather

than foraging behavior, then there must be differences in prey availability between different

habitats. Numerous tracking studies found that Laysan and Black-footed albatrosses forage in

different habitats (Fernandez et al. 2001, Fisher 2007, Hyrenbach et aZ. 2002). However, it is

unclear whether differences in foraging habitat are the reason for differences in the naturally­

acquired Laysan and Black-footed albatross diets because the distributions of T. borealis are

Histioteuthis are unknown.

Comparisons of this study's results with those of Gould et al. (1997) and Pitman et al.

(2004) show that there is either temporal or spatial variation in the naturally acquired portions of

the albatross diets. In this study, which examined diets during the breeding season, T. borealis

were more important to both albatross species than in the birds studied by Gould et al. (1997)

approximately a decade earlier, during the post breeding season. However, it is unclear whether

the differences occurred due to seasonal, annual or decadal differences in the availability of T.

borealis.

Compared with Pitman et al. (2004), this study found that T. borealis were more

important to Laysan Albatrosses from Hawaii than to Laysan Albatrosses from Mexico during

31

roughly the same period. However, the importance of T. borealis and Histioteuthis to Hawaiian

Black-footed Albatrosses was similar to that of Mexican Laysan Albatrosses. Possibly, these

differences were due to inter-specific differences in foraging habitat between Hawaiian and

Mexican Laysan Albatrosses, and similarities in foraging habitat between Hawaiian Black-footed

and Mexican Laysan albatrosses. Diet studies combined with tracking studies may resolve this.

Differences in foraging habitat might explain differences in natura1ly acquired food but

they cannot explain differences in fishery derived diets because fishery associated albatrosses

coexist on the same resource. Differences in fishery-acquired diets could be explained by

temporal differences in foraging behavior. One potential difference in foraging behavior

suggested by Harrison et al. (1983) was that Laysan Albatrosses might forage more nocturnally

than Black-footed Albatrosses, but Fernandez and Anderson (2000) found no differences in

nocturnal foraging between these species.

Another possible explanation of why Laysan Albatrosses acquired more fish bait is that

they might be more skilled at acquiring bait. Acquiring bait probably requires more agility than

scavenging from other fishery related food resources because bait is small, compared with

swordfish catch, and albatrosses must compete to pick it up quickly either during retrieval (while

it is skimming behind the boat) or after it is discarded. One possible reason why some birds may

be more agile than others is differential wing loading, whereby birds with lower wing loading

have lower flight speeds but greater maneuverability (Warham 1997). Albatrosses with greater

wing loading appear to use foraging habitats characterized by higher wind speeds than

albatrosses with lower wing loading (Phillips et al. 2004, Schafer et al. 2001). Suryan (2006)

found approximately a 10.6% difference in wing loading between Laysan and Black-footed

albatrosses but despite this, wind velocity did not appear to affect foraging location choices.

32

Wing loading may not explain segregation offoraging habitat between Laysan and Black-footed

albatrosses but it could explain differences in bait acquisition. An analysis of the effectiveness of

the albatrosses at bait stealing could provide an answer to this question but the necessary data is

currently unavailable.

Nitrogen isotopic analyses found Black-footed Albatrosses had significant intra-specific

differences in diet between albatrosses associated with swordfish and tuna operations, as well as

inter-specific differences between Laysan and Black-footed albatrosses associated with tuna

operations. Swordfish-associated Black-footed Albatrosses had the greatest 1i1~ values.

Considering the potential effect that ,swordfish consumption might have on the 1i1~ values of

Black-footed Albatrosses, it appears that swordfish operations might be more influential to their

diets. However, there was significant overlap in the 1i1~ values offood types acquired naturally

and from the fishery, indicating that these results cannot effectively connect differences in

nitrogen isotopic compositions to differential fishery resource use. Instead, there could be other

potential explanations, such as feeding at different trophic levels naturally, or feeding in different

regions having different characteristic 1i1~ values at the base of the food chain.

Comparison ofisotopic results from drift-net and longline operations indicate that Laysan

and Black-footed 1i1~ values appeared to have increased since the demise of drift-net fishing

(Gould el al. 1997). This was unexpected as Gould el al. (1997) found that drift-net fishing

increased the albatross trophic positions, with Black-footed Albatrosses experiencing the greatest

increase. Possible explanations for the increase include shifting naturally or fishery acquired

diets to consuming greater amounts of food types with higher 1i I ~ values such as swordfish or T.

borealis, or consuming less food with lower 1i1~ values, such as flying fish eggs or myctophids.

Increased consumption of T. borealis supports this explanation. However, swordfish would have

33

had to be consumed by the albatross in a manner not detectible by digestive tract analyses. This

could occur if swordfish scavenging did not lead to capture and sampling. Consumption ofless

food with lower 61~ values is an unlikely explanation because both this study and Gould et 01.

(1997) found such species were a relatively insignificant part of the naturally acquired portions

of the fishery associated albatross diets.

Another possible explanation is that the albatross lil~ values varied either seasonally,

annually or decadally due to shift in the isotopic composition at the base ofthe food chain. This

would compare well with the results of Hobson et 01. (2007) who found that the Waved

Albatrosses (Phoebastria irrorata) had seasonal differences in 61~ values which were higher

earlier in the breeding season. Seasonal, annual or decadal differences in 61~ values of Laysan

and Black-footed albatrosses could occur due to fishery influences which may vary seasonally

between fishing method, location and effort. However, seasonal, annual or decadal differences in

lil~ values could also be caused by changes in the availability ofnatumlly acquired prey

because of climatic cycles such as the El Nino and Pacific Decadal Oscillations (DuffY 1993,

Hare and Francis 1994, Mantua and Hare 2000), feeding on the same prey but from different

regions with different nitrogen isotopic compositions at the base of the food chain, or the effects

of climate cycles on the nitrogen dynamics and their effect on plankton 61~ values (Altabet et

01. 1999, Dore et 01. 2002, Mullin et 01. 1984). For instance, Dore et 01. (2002) found

considemble seasonal and inter-annual fluctuations in the 61~ values of particulate nitrogen

resulting from increased nitrogen fixation by diazotrophic phytoplankton during the summer.

Inter-annual fluctuations appeared to be affected by El Nino conditions which tended to increase

nitrogen fixation (Dore et 01.2002). Also, Altabet et aI. (1999) found higher 61~ values of

nitrate in coastal areas then which occur in oceanic areas. So, birds feeding in coastal areas, as do

34

Black-footed Albatrosses, subsequently ought to have higher lil~ values. Possible factors which

may change the diet of albatrosses include all or some ofthese influences. For instance, Roland

et aI. (2008) found that Black-browed Albatrosses were influenced by both climate and fishing,

but fishing was more important to albatross productivity. While no direct link of climate or

fishing to the reproductive ability of Lays an or Black-footed albatross has been found, the results

of this study show that the potential mechanism for such a relationship exists.

Samples from the breeding colonies indicate that Laysan Albatrosses differed in diet at

least between Kure Atoll and Kauai. These results are similar to those of Green et al. (I 998} who

found that Light-Mantled Sooty Albatrosses from different islands also have different diets.

Possibly, these results indicate regional diet differences in the availability of prey, as Laysan

Albatross colonies are spread out over thousands of miles and breeding adults forage relatively

close to the colony during critical nesting periods. Fasting could elevate 81~ values, but for this

to cause differences, the birds from different colonies would have to fast for different lengths of

time (Hobson et al. 1993). Differential fishery resource availability could also explain these

differences, and the higher lil~ values ofKauai birds, which breed closest to the fishery center,

further support this explanation. Though this study could not pinpoint the cause, these data

suggest that Laysan Albatrosses among different colonies may share different fates, especially in

terms of diet, and some colonies could potentially be more vulnerable to the effects of fishing

than others.

Chapter 5: Conclusion

This study showed that pelagic longline fishing in the North Pacific can affect the diets of

Laysan and Black-footed albatrosses in different ways. Because of this, fisheries managers

3S

should consider how changes in fishing practice due to changes in fishery regulations affect the

diets of each albatross species independently. Such decisions might relate to increasing or

decreasing fishing effort, modification of fishing location, bait use and offal discarding. Black­

footed Albatrosses are more influenced by pelagic longline fishing than Laysan Albatrosses and

swordfish operations are the greatest source of diet influence. Effects from swordfish operations

and effects on Black-footed Albatrosses should be given the most consideration in decision

making.

Though fishing is known to improve the productivity of albatrosses, it also can cause

tremendous harm due to high rates of incidental mortality. Safer fishing practices are available

and should be employed whenever albatross bycatch is an issue. Caution should be taken when

reducing effort by fishing operations using albatross-safe practices, because the albatrosses might

compensate for the loss of this food source by scavenging from more dangerous operations.

Pelagic longline fishing is a global phenomenon, so the potential influence of pelagic

longline fishing on the diets of albatrosses should be considered globally. Similarly, differences

in the ways different albatross species are affected should be acknowledged. Fishery and seabird

refuge managers should not expect the effects of fishing in other regions to necessarily affect

albatrosses the same as they did in this study. Furthermore, the benefits of pelagic longline

fishing to Laysan and Black-footed albatrosses need to be weighed against the costs of by catch

to the albatrosses. Decisions should be made with the long term survival of the albatrosses in

mind, allowing the coexistence of both the albatrosses and the longline fishing.

36

Acknowledgements

This study could not have been completed without the help of many generous

organizations and individuals. I would first like to thank the Pelagic Fisheries Research Program

and the Ecology, Evolution and Conservation Biology for funding my research and travel. A

contribution that I can never be thankful enough for came from the NOAA fisheries observers

from both the Pacific Islands and Southwest regions who spent months of their lives at sea while

continually experiencing hazardous situations, but still had the dedication to ensure that the

albatrosses used in this study were salvaged with precise data. Next, I would like to thank the all

the biologists and researchers who salvaged or provided the logistics to salvage the remaining

albatrosses, including Nigel Brothers, Eric Gilman, the Hawaii Longline Association, Pacific

Ocean Producers, Hawaii Department of Land and Natural Resources and the U.S. Fish and

Wildlife Service.

A unique group of individuals with the strongest of stomachs helped with gut analyses,

especially with dissections and sorting, weighing and identifYing prey. These generous and

talented individuals included my wife, Jami Bisson, as well as Lindsay Young, Norma Bustos,

Jodi Copes, Kelly Esh, Dusty Marshall, Richard Kupfer and Janis Matsunaga.

I am forever indebted to William Walker, Bruce Mundy and Richard Young who

contributed their time and expertise in helping me with prey identification. I am especially

thankful to the students and lab managers at the UH Isotope Biogeochemistry Lab for help with

stable isotope analyses, including Brittany Graham, Richard Wallsgrove, Terri Rust and Jami

Tanimoto. Thanks are also deserved to Ray Clark for help with proposals and Karla Gore for

editorial review.

37

I wish to thank my thesis committee, David DuffY, Sheila Conant and Brian Popp, for

their valuable instruction, time helping me plan this work, reviewing my documents and

especially for understanding my time management difficulties due to full time employment. I

thank my employers, NOAA Fisheries and the Bureau of Land Management for allowing me to

maintain a flexible schedule so I could balance my work with the needs of this project. Last, I

would like to thank Jami again and the rest of my family for their sacrifices that were necessary

to allow me to fmish.

38

References

Altabet, M.A., C. Pilskaln, R. Thunell, C. Pride, D. Sigman, F. Chavez, and R. Francois. 1999. The nitrogen isotope biogeochemistry of sinking particles from the margin of the Eastern North Pacific. Deep Sea Res., Part I, 46:655- 679.

Ballance, L.T., D.G. Ainley, and G.L. Hunt, Jr. 2001. Seabird Foraging Ecology. Pages 2636-2644 in J.H. Steele, SA Thorpe and K.K. Turekian (eds.) Encyclopedia of Ocean Sciences, vol 5. Academic Press, London.

Beverly, S.L. Chapman and W. Sokimi. 2003. Horizontallongline fishing methods and techniques - A manual for fisherman. Secretariat of the Pacific Community, Noumean, New Caledonia 130 pp.

Broughton, J.M. 1994. Size of the bursa offabricius in relation to gonad size and age in Laysan and Black-footed albatrosses. Condor 96:203-207.

Clarke, M.R. 1986. A handbook for the identification of cephalopod beaks. Clarendon Press, Oxford 273 pp.

Clothier, C.R. 1950. A key to some Southern California fishes based on vertebral characters. Fishes. California Department ofNatoral Resources Division ofFish and Game Fish Bulletin 79.

Collette, B.B., G.E. McGowen, N.V. Parin, and S. Mito. 1984. Beloniformes: development and relationships. Pages 335-354 in H. G. Moser (eds.), Ontogeny and systematics of fishes. American Society ofIchthyologists and Herpetologists. Special Publication no. I.

Cousins, K. 2001. The Black-footed Albatross Population Biology Workshop: A step to understanding the impacts of longline fishing on seabird populations. Pages 95-I 14 in E.F. Melvin and J.K. Parrish (eds.), Seabird bycatch: Trends, roadblocks, and solutions. Proceedings ofan international symposium. University of Alaska Sea Grant, AK-SG-OI-01, Fairbanks.

DeNiro, M.J. and S. Epstein. 1978. Influence of diet on the distribution of carbon isotopes in animals. Geoehimica et Cosmochimica Acta 42:495-506.

DeNiro, M.J. and S. Epstein. 1981. Influence of diet on the distribution of nitrogen isotopes in animals. Geochimica et Cosmochimica Acta 45:34 I -35 I.

Dore, J .E., J .R. Brum, L. Tupas, and D.M. Kar\. 2002. Seasonal and interannual variability in source of nitrogen supporting export in the oligotrophic subtropical North Pacific Ocean. Limnology and Oceanography 47:1595-1607.

DuffY, D.C. 1993. Stalking the Southern Oscillation: environmental uncertainty, climate change, and North Pacific seabirds. In Vermeer, K., K.T. Briggs, K. H. Morgan, and D. Siegel-

39

Causy. (eds.) The status, ecology, and conservation of marine birds of the North Pacific. Canadian Wildlife Service, Special Publication, Ottowa, Ontario.

Fernandez, P. and OJ. Anderson. 2000. Nocturnal and diurnal foraging activity of Hawaiian albatrosses detected with a new immersion monitor. Condor 102:577-584.

Fernandez, P., OJ. Anderson, P.R. Sievert, and K.P. Huyvaert. 2001. Foraging destinations on low latitude albatross (Phoebastria) species. Journal of Zoology, London 254:391-404.

Furness, R.W. 2002. Impacts of fisheries on seabird communities. Scientia Marina 67:33-45.

Gilman E., N. Brothers and D.R. Kobayashi. 2005. Principles and approaches to abate seabird by-catch in longline fisheries. Fish and Fisheries 6:35-39.

Gonzalez-Solis, J. 2003. Impact of fisheries on activity, diet and predatory interactions between Yellow-legged and Audouin's gulls breeding at the Chafarinas Islands. Scientia Marina 67:83-88.

Gould, P., P. Ostrom, and W. Walker. 1997. Trophic relationships of albatrosses associated with squid and large-mesh drift-net fisheries in the North Pacific Ocean. Canadian Journal of Zoology 75:549-562.

Gould, P .1. and J.F. Piatt. 1993. Seabirds ofthe central North Pacific. Pages 27-38 in K. Vermeer, K.T. Briggs, K.H. Morgan, D. Siegel-Causey (eds), The status, ecology, and conservation of marine birds of the North Pacific. Canadian Wildlife Service, Special Publication, Ottowa, Ontario.

Green, K.K.R. Kerry, T. Disney, and M.R. Clarke. 1998. Dietary studies of Light-mantled Sooty Albatrosses (Phoebetria Palpebra/a) From Macquarie and Heard Islands. Marine Ornithology 26: 19-26.

Hare, S.R. and R.C. Francis. 1994. Climate change and salmon production in the Northeast Pacific Ocean, p. 357-372. In R. J. Beamish (ed.) Climate Change and Northern Fish Populations. Canadian Special Publication Fisheries Aquatic Science 121 pp.

Harrison, C.S., T.S. Hida, and M.P. Seki. 1983. Hawaiian seabird feeding ecology. Wildlife Monographs 85:1-71.

Hollowed, A.B., K.M. Bailey, and W.S. Wooster. 1987. Patterns ofrecruitrnent of marine fishes in the Northeast Pacific Ocean. Biological Oceanography 5:99-131.

Hobson KA, R.T. Alisauskas and R.G. Clark. 1993. Stable-nitrogen isotope enrichment in avian tissues due to fasting and nutritional stress: implications for isotopic analyses of diet. Condor 95:388-394.

40

Hobson K.A., J.F. Piatt, and J. Pitocchelli 1994. Using stable isotopes to determine seabird trophic relationships. The Journal of Animal Ecology 63:786-798.

Hobson, K.A. DJ. Anderson, and J.A. Awkerman. 2007. Isotopic (1l1~ and 1l13C) evidence for intersexual foraging differences and temporal variation in habitat use in Waved Albatrosses. Canadian Journal of Zoo logy 85:273-279.

Hyrenbach, K.D., P. Fernandez, and D.J. Anderson. 2002. Oceanographic habitats of two Sympatric North Pacific albatrosses during the breeding season. Marine Ecological Progress Series 233:283-301.

Hyrenbach, K.D. and R.C. Dotson 2003. Assessing the susceptibility offemale Black-footed Albatross (Phoebastria nigripes) to longline fisheries during their post-breeding dispersal: an integrated approach. Biological Conservation 112:391-404.

James, G.D. and J. Stahl. 2000. Diet of southern Buller's Albatross (Dtomedea bulleri bullen) and the importance of fishery discards during chick rearing. New Zealand Journal of Marine and Freshwater Research 34:435-454.

Mantua, NJ. and S.R. Hare 2002. The Pacific Decadal Oscillation. Journal of Oceanography 58:35-44.

McDermond, D.K. and K.H. Morgan. 1993. Status and conservation of North Pacific albatrosses. Pages 70-81 in K. Vermeer, K.T. Briggs, K.H. Morgan, D.Siegel-Causey (eds), The status, ecology, and conservation of marine birds of the North Pacific. Canadian Wildlife Service, Special Publication, Ottowa, Ontario.

Minami, H., M. Minagawa and O. Haruo. 1995. Changes in stable carbon and nitrogen isotope ratios in Sooty and Short-tailed shearwaters during their northward migration. Condor 97:565-574.

Mullin, M.M., G.H. Rau and R.W. Eppley. 1984. Stable nitrogen isotopes of zooplankton: Some geographic and tempora! variations in the North Pacific. Limnology and Oceanography 29:1267-1273.

Pedrocchi, V., D. Oro, J. Gonzalez-Solis, X. Ruiz, and L. Jover. 2002. Differences in diet between the two largest breeding colonies of Audouin's Gulls: the effects of fishery activities. Scientia Marina 66:313-320.

Phillips, R.A., J.D. Silk, B. Phalan, P. Catry and J.P. Croxall. 2004. Seasonal sexual segregation in two Thalassarche albatross species: competitive exclusion, reproductive role specialization or foraging niche divergence? Proceedings of the Royal Society London 271:1283-1291.

41

Pinkas, L., M.S. Oliphant and I.L.K. Iverson. 1971. Food habits of albacore, bluefin tuna and bonito in California waters. California Department of Fish and Game. Fish Bulletin 152 105 pp.

Pinnegar, J.K. and N.V.C. Polunin. 1999. Differential fractionation ofli l3C and 151~ among fish tissues: implications for the study of trophic interactions. Functional Ecology 13:225-231.

Pitman, R.L. 1988. Laysan Albatross breeding in the eastern Pacific - and a comment. Pacific Seabird Group Bulletin 15:52.

Pitman, R.L., W.A. Walker, W.T. Everett and J.P. Gallo-Reynoso. 2004. Population status, foods and foraging ofLaysan Albatrosses (Phoebastria immutabilis) nesting on Guadalupe Island, Mexico. Marine Ornithology 32:159-165.

Polovina, J.J.E. Howell, D.R. Kobayashi, M.P. Seki 2001. The transition zone chlorophyll front, a dynamic global feature defining migration and forage habitat for marine resources. Progress in Oceanography 49: 469-483.

Post, D.M., C.A. Layman, D.A. Arrington, G. Takimoto, C.G. Montana, and J. Quattrochi. 2007. Getting to the fat of the matter: models, methods and assumptions for dealing with lipids in stable isotope analyses. Oecologia, 152:179-189.

Rice, D.W. and K.W. Kenyon. 1962a. Breeding distribution, history, and populations of North Pacific albatrosses. Auk 79:365-386.

Rice, D.W. and K.W. Kenyon. 1962b. Breeding cycles and behavior of Lay san and Black-footed albatrosses. Auk 79:517-567.

Rolland Y., C. Barbraud, H. Weimerskirch. 2008. Combined effects of fisheries and climate on a migratory long-lived marine predator. Journal of Applied Ecology 45:4-13.

Schoener, T.W. 1970. Non-synchronous spatial overall oflizards in patchy habitats. Ecology 51 :408-418.

Shaffer, S.A., Weimerskirch, H. & Costa, D.P. 200 I. Functional significance of sexual dimorphism in Wandering Albatrosses (Diomedea exulans). Functional Ecology 15:203-210.

Suryan R.M. 2006. Comparative foraging ecology of five species of Pacific seabirds: Multi-scale analyses of marine habitat use. Unpublished Ph.D. Dissertation, Oregon State University, Corvallis, OR 193 pp.

Takai N, S. Onaka, Y. Ikeda, A. Yatsu, H. Kidokoro, W. Sakamoto. 2000. Geographical variations in carbon and nitrogen stable isotope ratios in squid. Journal of the Marine Biological Association of the United Kingdom 80: 675-684.

42

Tyler, J.C. 1980. Osteology, phylogeny, and higher classification of the fishes of the Order Plectognathi (Tetraodontiformes). U.S. Deparbnent of Commerce, National Oceanic and Atmospheric Administration, NOAA Technical Report NMFS Circular 434 422 pp.

Votier, S.C., R.W. Furness, S. Bearhop, J.E. Crane, R.W.G. Caldow, P. Catry, K. Ensor, K.C. Hamer, A.V. Hudson, E. Kalmbach, N.r. Klomp, S. Pfeiffer, R.A. PhiIIips, I. Prieto, and D.R. Thompson. 2004. Changes in fisheries discard rates and seabird communities. Nature 427:727-730.

Wallace Jr., R.K. 198 I. An assessment of diet-overlap indexes. Transactions of the American Fisheries Society I 10:72-76.

Walsh, H.E. and S.V. Edwards. 2005. Conservation genetics and Pacific fisheries bycatch: Mitochondrial differentiation and population assignment in Black-footed Albatrosses (Phoebastria nigripes). Conservation Genetics 6:289-295.

Warham, J. 1977. Wingloadings, wing shapes, and flight capabilities of Procellariformes. New Zealand Journal of Zoology 4:73-83.

Wisner, R.L. 1974. The taxonomy and distribution of Lantern fishes (Family Myctophidae) of the Eastern Pacific Ocean. Navy Ocean Research and Development Activity. NORDA Report 3. 299 pp.

Xi, He, K.A. Bigelow, C.H. Boggs. 1997. Cluster analysis of long line sets and fishing strategies within the Hawaii-based fishery. Fisheries Research 31:147-158.

43