Effects of Gill-Net Fishing on Marine Birds
in a Biological Hotspot in the Northwest Atlantic
GAIL K. DAVOREN
Cognitive and Behavioural Ecology Programme, Departments of Biology & Psychology, Memorial University of Newfoundland,
St. John’s, Newfoundland A1B 3X9, Canada, email [email protected]
Abstract: Marine biological hotspots, or areas where high abundances of species overlap in space and time,
are ecologically important areas because energy flow through marine food webs, a key ecosystem process, is
maximized in these areas. I investigated whether top predators aggregated at persistent spawning sites of a key
forage fish species, capelin (Mallotus villosus), on the NE coast of Newfoundland during July and August 2000–
2003. By examining the distributional patterns of top predators through ship-based surveys at multiple spatial
and temporal scales, I found that the biomasses of birds—dominated by Common Murres (Uria aalge)—and
mammals—dominated by whale species—were concentrated along the coast, with a biological hotspot forming
near two persistent spawning sites of capelin in all years. The formation of this hotspot was well defined in
space and time from middle of July to middle of August, likely coinciding with the spawning chronology of
capelin. Within this hotspot, there was a high spatial and temporal overlap of Common Murres and gill nets
set to capture Atlantic cod (Gadus morhua). This resulted in breeding murres becoming entangled in gill nets
while feeding on spawning capelin. Despite an acknowledged uncertainty of bycatch mortality, estimates for
the larger regional-scale area (1936–4973 murres/year; 0.2–0.6% of the breeding population) underestimated
mortality relative to estimates within the hotspot (3053–14054 murres/year; 0.4–1.7%). Although fishing effort
for Atlantic cod has declined substantially since the groundfish moratorium in 1992, chronic, unnatural, and
additive mortality through bycatch continues in coastal Newfoundland. Restricted use of gill nets within this
and other biological hotspots during the capelin spawning period appears to be a straightforward application
of the “ecological and biologically significant area” management framework in Canada’s Oceans Act. This
protection would minimize murre bycatch and maintain ecosystem integrity.
Keywords: biological hotspot, bycatch, capelin, Common Murre, gill net, northwest Atlantic, protected area
Efectos de las Redes Agalleras sobre Aves Marinas en un Sitio Biológicamente Importante en el Noroeste del
Atlántico
Resumen: Los sitios marinos biológicamente importantes, o áreas en las que altas abundancias de especies
se traslapan en espacio y tiempo, son ecológicamente importantes porque el flujo de energı́a a través de
las redes alimenticias marinas, un proceso ecológico clave, se maximizan en esas áreas. Investigué si los
depredadores superiores se agregan en los sitios de desove de una especie presa clave, el capelán (Mallotus
villosus), en la costa noreste de Terranova durante julio y agosto 2000 – 2003. Mediante el examen de los
patrones de distribución de los depredadores superiores con base en muestreos realizados en barco en diferentes
escalas espaciales y temporales, encontré que la biomasa de aves – dominada por Uria aalge – y mamı́feros –
dominados por especies de ballenas – estaba concentrada a lo largo de la costa, cerca de un sitio biológicamente
importante en formación cerca de dos sitios de desove de capelán en todos los años. La formación de este sitio de
importancia fue bien definida en espacio y tiempo desde mediados de julio a mediados de agosto, coincidiendo
probablemente con la cronologı́a de desove del capelán. Dentro de este sitio, hubo un traslape espacial y
temporal significativo de Uria aalge y redes agalleras colocadas para capturar bacalao del Atlántico (Gadus
Paper submitted July 7, 2006; revised manuscript accepted December 5, 2006.
1032
Conservation Biology Volume 21, No. 4, 1032–1045
C 2007 Society for Conservation Biology
DOI: 10.1111/j.1523-1739.2007.00694.x
Davoren
Gill-Net Fisheries and Common Murre Mortality
1033
morhua). Esto resultó en que individuos de U. aalge quedaron atrapados en las redes agalleras mientras se
alimentaban de capelanes. A pesar de la incertidumbre de la mortalidad de la captura incidental, estimaciones
para un área a escala regional (1936–4973 araos/año; 0–2 – 0.6% de la población reproductora) subestimaron
la mortalidad en relación con estimaciones dentro del sitio de importancia (3053–14054 araos/año; 0.4 – 7%).
Aunque el esfuerzo de pesca de bacalao ha disminuido sustancialmente desde la moratoria de 1992, en la costa
de Terranova sigue ocurriendo una mortalidad aditiva, crónica y anormal debido a la captura incidental.
La restricción en el uso de redes agalleras dentro de este y otros sitios biológicamente importantes durante
el perı́odo de desove de capelán parece ser una aplicación directa del marco de referencia en la gestión de
un “área ecológica y biológicamente significativa” del Acta de Océanos de Canadá. Esta protección deberı́a
minimizar la captura incidental de araos y mantener la integridad del ecosistema.
Palabras Clave: área protegida, capelán, captura incidental, noroeste del Atlántico, red agallera, sitio
biológicamente importante, Uria aalge
Introduction
There is an imminent need for conservation efforts in marine ecosystems. Global-scale declines in the biodiversity
(Worm et al. 2005) and biomass of large vertebrate predators (Myers & Worm 2003) are increasing, and marine
communities are being altered worldwide (e.g., Frank et
al. 2005). Identifying areas of high species richness ( biodiversity hotspots) has been suggested as a method of
pinpointing priority areas for conservation on a global
scale in both terrestrial and marine ecosystems (Myers
et al. 2000; Worm et al. 2005). Similarly, the concept of
no-take marine reserves or protected areas has surged
(Hyrenbach et al. 2000; Pauly et al. 2002; Hooker & Gerber 2004). Regional-scale investigations often reveal localized areas where high abundances of organisms are spatially concentrated ( biological hotspots), typically reflecting areas of high marine productivity (Hooker & Gerber
2004). Due to the fluid boundaries of water masses and
difficulty in identifying spatially discrete habitats for protection (Soulé & Orians 2001), biological hotspots may
be appropriate areas for protection in marine systems
(Hooker & Gerber 2004) because they are highly vulnerable to industrial threats, such as oil pollution and fisheries
activities.
When high densities of commercial and noncommercial species overlap spatially and temporally with fishing
activities, high bycatch of nontarget species can result
(e.g., Carter & Sealy 1984; Piatt et al. 1984; Takekawa et
al. 1990). Gill nets trap and kill large numbers of both
target and nontarget organisms. High entanglement rates
have been reported for marine mammals (e.g., Johnson et
al. 2005) and pursuit-diving seabirds, especially murres of
the genus Uria (Piatt & Nettleship 1987). Bycatch events
can be highly episodic (e.g., Piatt et al. 1984; Vader et al.
1990), likely due to the overlap of aggregations of these
species and fishing effort at certain times of the year in
specific areas. Estimates of bycatch mortality can be substantially weakened by including events with no bycatch
(Ainley et al. 1981; Strann et al. 1991) that result when
fishing effort and animal densities do not overlap. Integrating this source of error with other data inadequacies,
such as underreporting of bycatch by fishers either intentionally or unintentionally, makes quantitative estimates
of bycatch mortality somewhat speculative. Nonetheless,
quantification of bycatch mortality is critical for conservation and management purposes.
In many northern marine ecosystems, top predators
rely on capelin (Mallotus villosus), a small, schooling fish
species (Carscadden & Vilhjálmsson 2002). This focal forage fish lies at the core of marine food webs, providing
essential linkages for energy transfer among trophic levels. Atlantic cod (Gadus morhua) was the main predator
of capelin on the Newfoundland shelf prior to its population collapse and the closure of the eastern Canadian
ground fishery (1992). Inshore cod fisheries, consisting of
a small commercial and a sentinel fishery, have operated
since the mid-1990s on a smaller scale (Lilly et al. 2003).
Both fisheries use gill nets predominantly, with the commercial fishery accounting for > 90% of the total landings
of cod per year (Lilly et al. 2003).
Recently, off-beach, or demersal, spawning sites of
capelin were discovered on the northeast coast of Newfoundland (Davoren et al. 2006). Persistently large aggregations of capelin have occurred near these spawning
sites over a number of years (2000–2003, Davoren et al.
2006; Fig. 1: Gull Island). My primary objective was to
determine whether high abundances of top predators
overlap with these persistent aggregations of spawning
capelin, as suggested by the spatial concentration of Common Murres in 2000 (Davoren et al. 2003, 2004), resulting
in the formation of biological hotspots. I also examined
whether gillnetting for cod overlaps with these persistent
aggregations and whether this leads to entanglement of
animals in nets. Lastly, I compared quantitative estimates
of bycatch mortality associated with these persistent aggregations ( based on mean bycatch rates where fishing
effort and high abundances of organisms overlapped)
with regional-scale estimates of mean bycatch rates and
fishing effort. This investigation is a critical first step in
Conservation Biology
Volume 21, No. 4, 2007
1034
Gill-Net Fisheries and Common Murre Mortality
Davoren
delineating ecologically and biologically significant areas
of management concern.
Methods
Study Area and Survey Design
This study was conducted during 2000–2003 on the
NE coast of Newfoundland (49◦ 00 N-49◦ 45 N, 53◦ 30 W54◦ 00 W; Fig. 1) near the Funk Island Seabird Ecological Reserve (60 km off the coast [49◦ 45 N, 53◦ 11 W]),
where the primary nesting seabird is the Common Murre
(412,000 breeding pairs; Chardine et al. 2003). Many
other seabird species breed on Funk Island, most of which
have high proportions of capelin in their diets (Davoren
& Montevecchi 2003a). These species include Northern
Gannets, Atlantic Puffins, Black-legged Kittiwakes, Thickbilled Murres (Uria lomvia), Razorbills (Alca torda), Herring, and Great Black-backed Gulls. (If not provided in
text, scientific names of birds are in Table 1.) Another
nearby colony on Cabot Island (49◦ 10 N, 53◦ 22 W) has
3000 breeding pairs of Common Murres (Cairns et al.
1989).
To determine the distributional patterns of capelin, marine birds, and mammals in the study area, I conducted a
regional-scale survey from mid through late July (range:
6–8 days/year) in 2000–2003 (Fig. 1). Surveys were carried out in daylight aboard the 23-m Canadian Coast Guard
research vessel Shamook, which operated 12 hours/day.
I surveyed nine eastwest (across shelf ) hydroacoustic survey lines with 9-km north–south spacing. I also surveyed
one coastal survey line (Fig. 1). The route deviated slightly
among years, owing to different wind direction and speed
conditions, but the same paths were followed each year.
The route was established to cover high-density aggregations of marine fish, birds, and mammals (Davoren et al.
2003). During the survey the number of birds and mammals were continuously recorded, along with simultaneous hydroacoustic recordings of fish biomass (Davoren et
al. 2006). I periodically interrupted the survey to collect
dead animals when aggregations were encountered and
to sample fish aggregations, indicated by acoustic signals,
with a modified shrimp trawl (96.5% capelin in all years;
Davoren et al. 2006).
After the regional-scale survey in each year, I used
the remaining ship time (range: 5–7 days/year) and time
aboard a 15-m commercial fishing vessel, the Lady Easton II (2001, 2–8 August; 2002, 3–10 August; 2003, 2–
20 August), to conduct similar but shorter surveys for
capelin, birds, and mammals in fine-scale areas where
high abundances of organisms were aggregated during
the regional-scale survey (Fig. 1: lines 1, 5, 6, 7, and Gull
Island). I visited each fine-scale area over a number days
to investigate the persistence of these aggregations both
within and among years of this study (Davoren et al. 2003,
2006).
Conservation Biology
Volume 21, No. 4, 2007
Figure 1. A map of Newfoundland indicating (a) the
study area on the northeast coast (rectangle) and ( b)
the regional-scale survey (white dashed line);
fine-scale surveys ( black dashed line); two demersal
spawning sites of capelin (star); and Funk and Cabot
island seabird colonies with gray-scale depth contours.
At-Sea Distributional Patterns
During the regional-scale survey, birds were counted with
standard strip-transect methods (method Ib, Tasker et al.
1984). One observer made continuous counts of seabirds
from the bridge (2–3 m high) out to 300 m in a 90◦ arc
from the tip of the bow to the port side of the vessel.
Marine mammals, including seals, whales, and dolphins,
were counted inside this strip transect, as were dead
animals and floats on the ocean surface, indicating the
Davoren
presence of gill nets. Large aggregations of marine birds,
mammals, dead animals, and gill-net floats outside the
strip transect were recorded separately, and total estimates were rounded to multiples of 10. Gill-net floats
were only systematically counted during surveys in 2001–
2003. All counts, behavioral descriptions (on water, feeding, flying [with direction], dead), and whether the
species was observed inside or outside the strip transect were recorded on a laptop computer. The laptop
interfaced with the navigational system of the vessel, and
counting software (D. Senciall, Birds & Beasty Counter,
1998, Fisheries and Oceans Canada, version 1.0) appended a position (latitude and longitude) and Greenwhich Mean Time (GMT) to each entry. A navigational
software package (Bioplot, version 2.0, 1991, BioSonics,
Seattle) continuously recorded the ship’s position and
GMT, providing a time-stamped cruise track for every 100
m of each survey line. I used the date and GMT to merge
count data with the cruise track.
I conducted the fine-scale surveys in a similar manner
along lines 1, 5, 6, 7, and in the Gull Island area in 2001
(Fig. 1), but I only repeated the Gull Island survey line
in 2002 and 2003. The route of this survey line varied
slightly among days because the ship often had to alter
course to avoid entangling high concentrations of gill-net
floats.
Data Analysis: At-Sea Distributional Patterns
All birds observed on the water within the strip transect
were used in the analysis to indicate foraging habitat use.
I excluded from density and biomass estimates counts of
birds flying because they were likely commuting and not
foraging in the area. Exceptions were made for species
that forage during flight, including Northern Gannets and
Leach’s Storm Petrels. In these cases counts of flying birds
were included. The total count of each bird and mammal species per 100 meters was multiplied by the average mass of this species from published sources. Bird
and mammal biomass per 100 meters were then averaged
separately for 5-km segments of each regional-scale survey with a Matlab routine. The total count of dead animals
and gill-net floats per 100 meters were summed for 5-km
segments of each regional-scale survey and for 2.5-km
segments of each fine-scale survey. Average biomass and
total counts per 2.5 or 5 km were mapped with Surfer 8.0
(Golden Software) to show distributional patterns. Finally,
I calculated the percentage each species contributed to
the overall biomass in each regional-scale survey. Because
the Common Murre was the most numerous breeding
bird and the only species found dead in the study area,
the number of murres per 100 m was also averaged for 5km segments of each regional-scale survey and for 2.5-km
segments of each fine-scale survey. All means are reported
as ± standard error.
Gill-Net Fisheries and Common Murre Mortality
1035
Collection and Laboratory Analysis of Dead Animals
Upon an encounter with dead murres, I interrupted both
the regional-scale and fine-scale surveys to collect specimens with a long-poled dipnet. Each bird was individually
wrapped in a labeled plastic bag and frozen immediately.
Specimens were later thawed and examined for external
indications of the cause of death (e.g., penetrating injuries, severe feather abrasions). I recorded breeding status of the birds, as revealed by the presence or absence
of a brood patch. Through dissection, internal indications
of the cause of death (including internal hemorrhaging,
subcutaneous bruising, fluid in the lungs, and broken
ribs; Darby & Dawson 2000) were recorded for each bird.
The sex of each bird was determined by examining the
gonads.
The stomach, including the proventriculus and gizzard,
was removed and placed in a glass dissecting tray. Digested material was washed thoroughly into the glass tray.
I placed the tray on a black garbage bag and placed a directional light source overhead to illuminate fish otoliths.
I removed all otoliths and glued each into an individual
slot on a tray. Species identification of otoliths was the
primary method used to determine the species composition of murre stomachs. Undigested material (i.e., heads,
tails) was also used for species identification, and parts
of individuals were combined into whole specimens to
count the number of fish in each stomach (i.e., 1 head +
1 tail = 1 fish). Capelin specimens or parts were identified to sex and maturity stage for females based on the
presence of roe and for males based on the morphological characteristics of the anal and pectoral fins and the
presence of spawning ridges.
Estimates of Dead Birds
Regional-scale estimates of bycatch mortality of Common Murres per year were based on mean bycatch rates
of murres and total fishing effort of gill nets over the
study area. Data obtained from the sentinel fishery (Fisheries and Oceans Canada, unpublished data) included the
mean catch rate of Atlantic cod (kilograms per net) during 2000–2003 and the number of murres caught in gill
nets reported by fishers (2002 only) in the study area.
Each 100-m-long gill net used in the sentinel fishery was
anchored to the seabed in depths usually <50 m and
stretched 2 m high with mesh sizes of 140 and 83 mm.
To estimate the total fishing effort for cod (kilometers of
gill net deployed) in the study area during July–August
of each year, I divided the total landings of cod from gill
nets of all fisheries (commercial, sentinel, bycatch) between 1 July–31 August of 2000 (Lilly et al. 2001), 2001,
and 2002 (Lilly et al. 2003) by the mean catch rate of Atlantic cod per kilometer from the sentinel fishery in each
year. To estimate the total number of murres caught in the
study area per year, I multiplied the total length of gill nets
Conservation Biology
Volume 21, No. 4, 2007
1036
Gill-Net Fisheries and Common Murre Mortality
Davoren
Table 1. The mean density of marine bird and mammal species encountered in the study area (Newfoundland’s northeast coast) during the
regional-scale survey and the percentage that each species contributed to the total biomass in each year from 2000–2003.
Total number/ km2
Species
Mass (kg)
Birds
total
Common Murre, Uria aalge
0.99
Northern Gannet, Morus bassanus
3.20
Atlantic Puffin, Fratercula acrtica
0.46
Sooty Shearwater, Puffinus griseus
0.79
Leach’s Storm-Petrel, Oceanodroma leucorhoa
0.05
Herring Gull, Larus argentatus
1.12
Great Black-backed Gull, Larus marinus
1.68
Black-legged Kittiwake, Rissa tridactyla
0.44
Greater Shearwater, Puffinus gravis
0.79
Northern Fulmar, Fulmarus glacialis
0.80
Marine mammals
total
harp seal, Phoca groenlandica
130
humpback whale, Megaptera novaeangliae
31,000
minke whale, Balaenoptera acutorostrata
5600
harbour seal, Phoca vitulina concolor
100
fin whale, Balaenoptera physalus
38,500
deployed in each year by the mean bycatch of murres per
kilometer reported in the sentinel fishery in 2002.
Estimates of bycatch mortality of Common Murres
within each fine-scale area were based on mean murre
and gill-net float density per kilometer observed during
fine-scale surveys and the mean bycatch rates of murres
reported during the sentinel fishery in the Gull Island
area in 2002. I scaled the bycatch rate of murres to murre
density with the number of murres caught in two sentinel gill nets on 6 August 2002 (n = 3) and the mean
density of murres (6.5 ± 1.5 murres/km) within 2.5 km
of this bycatch event over 3 days (4, 6, and 8 August
2002). Integrating these numbers with the mean density
of murres and gill-net floats per kilometer in each survey
and the total length of each survey, I estimated the total
number of murres caught in gill-nets per survey. Fishing
effort and bycatch rates of murres were calculated per
net because the length of each gill net observed during
surveys was unknown. To estimate the number of murres
caught in gill nets within each fine-scale area per year, I
multiplied the estimated mean number of murres caught
in gill nets within each area in each year by the number
of days that the breeding season of murres (1 June – 31
August) overlapped with the commercial fishery (2001,
9 July–30 November; 2002, 30 July–13 October).
Results
Species Composition in the Study Area
Over all years bird biomass was dominated by Common
Murres (72%), followed by Northern Gannets (14%), and
to a lesser extent Atlantic Puffins (7%) and Sooty Shear-
Conservation Biology
Volume 21, No. 4, 2007
Biomass (%)
2000
2001
2002
2003
2000 2001 2002 2003
10.1
7.7
1.1
0.6
0.1
0.4
0.1
0.03
0.04
0.02
0.01
21.3
15.9
2.2
0.9
1.6
0.2
0.01
0
0.2
0.1
0.1
5.2
3.8
0.6
0.8
0
0.02
0.01
0
0
0
0
8.5
4.9
2.3
0.8
0.01
0.02
0.3
0.2
0.02
0
0
76.9
10.7
5.8
1.1
4.0
0.6
0.3
0.4
0.2
0.1
74.8
10.5
4.2
7.5
1.1
0
0
0.8
0.5
0.6
72.8
11.3
15.3
0
0.4
0.2
0
0
0
0
58.0
27.1
8.8
0.1
0.2
3.2
2.3
0.2
0
0
0.2
0.1
0.03
0.02
0.01
0.01
0.06
0
0.04
0.02
0
0
0
0
0
0
0
0
0.01
0
64.4
0.004 14.9
0.01
11.5
0
6.3
0
2.9
0
60.3
39.7
0
0
–
–
–
–
–
0
28.6
71.4
0
0
waters (4%; Table 1). Other species that were incidentally observed in one survey year and, thus, were not
included in the total biomass calculations include the
Common Tern (Sterna hirundo), Arctic Tern (Sterna
paradisaea), Black Guillemot (Cepphus grylle), Manx
Shearwater (Puffinus puffinus), Pomarine Jaeger (Stercorarius pomarinus), Parasitic Jaeger (Stercorarius parasiticus), Long-tailed Jaeger (Stercorarius longicaudus),
Great Skua (Catharacta skua), and South Polar Skua
(Catharacta maccormicki). The mammal biomass was
dominated by humpback whales (26%) and minke whales
(22%) in most years, with the exception of the 2000
survey when three small groups of harp seals (n = 30,
20, 14 individuals) were observed near the offshore limits of the study area (Table 1). Other species that were
incidentally observed in one survey year include the
killer whale (Orcinus orca), Atlantic white-sided dolphin
(Lagenorhynchus acutus), harbor porpoise (Phocoena
phocoena), and sei whale (Balaenoptera borealis). Overall, 21 bird and 9 mammal species were observed over the
4 years.
Regional-Scale Distributional Patterns of Bird
and Mammal Biomass
In all years the biomass of birds on the water during midto late July was concentrated close to Funk Island, likely
reflecting birds resting near the colony. Biomass also was
concentrated close to shore and declined with increasing distance from shore (Fig. 2). This relationship, however, was less obvious in 2002, when the coastal survey
line was not conducted due to inclement weather and
time constraints aboard the ship (Fig. 2c). Marine mammal
biomass was scattered throughout the study area, but was
Davoren
also concentrated along the coastal survey line in most
years except in 2002, when mammals were not observed
during the survey. The distribution of Common Murres
on the water reflected the distribution patterns of bird
biomass during the survey in all years (Fig. 3) because
the majority of the biomass was composed of murres (Table 1). In all years Common Murres were the only dead
animals encountered during the study period. Scattered
dead individuals were observed throughout the survey in
all years; however, large aggregations of dead murres were
observed near Gull Island in all years but 2003 (Fig. 3).
Fine-Scale Distributional Patterns of Common Murres
and Gill-Net Floats
High densities of Common Murres were observed in all
fine-scale areas, with lower densities along line 1 and
variable densities among days (Table 2). High concentrations of gill-net floats were always observed along the Gull
Gill-Net Fisheries and Common Murre Mortality
1037
Island survey line, with the exception of a survey earlier
in the season (19 July 2001) when nets were not observed
(Table 2). In contrast, the density of gill-net floats on lines
1 and 7 was much lower, and floats were never observed
on lines 5 and 6 (Table 2). High densities of murres and
gill-net floats along the Gull Island survey line resulted in
a high spatial and temporal overlap in murre and gill-net
distributions (Fig. 4).
During 2000–2002 four large aggregations of dead murres were observed near Gull Island (27 July 2000; 26 July
2001; 4, 8 August 2002). It was impossible to quantify
the number of dead murres in each aggregation because
of varying sea conditions and because birds floated both
chest up (easily detected) and chest down (difficult to
detect). Based on encounter rates with dead murres and
visual assessments during collection, I estimated that at
least hundreds of dead murres comprised each aggregation on each of the four encounters.
Figure 2. Biomass of marine birds and mammals during the regional-scale survey in (a) 2000, ( b) 2001, (c)
2002, and (d) 2003. Dashed line is the survey route.
Conservation Biology
Volume 21, No. 4, 2007
1038
Gill-Net Fisheries and Common Murre Mortality
Davoren
Figure 2. continued.
The location of dead murres overlapped spatially and
temporally with the high densities of gill-net floats and
murres near Gull Island (Fig. 4). A similar association
could not be determined in 2000 and 2003 because gill
nets were not systematically counted in 2000 and few gill
nets were observed in 2003. The gill-net set locations of
sentinel fishers, which were used to calculate the rates
of bird bycatch within the regional-scale study area, did
not overlap with the high concentrations of gill nets and
murres in the Gull Island area (Fig. 4), indicating that
the majority of gill nets observed in this area were associated with the commercial fishery. Overall, persistent
concentrations of gill nets and murres, coinciding with
large numbers of dead murres, were observed near Gull
Island from late July to mid-August throughout the years
of this study.
Conservation Biology
Volume 21, No. 4, 2007
Stomach Contents of Dead Birds
Owing to time constraints aboard the ship, only a small
sample of dead Common Murres (n = 67) was collected
on four occasions in 2000–2002 (Table 3). Carcasses were
all in good physical condition with fresh eyes. This suggests that these birds died on the day they were collected
because otherwise they would have been degraded by
scavenging birds (e.g., gull species). Their good condition
and high concentration also suggest that birds originated
from a nearby source. External examination of dead birds
did not reveal a cause of death. Internal examination, however, revealed that all birds had extensive subcutaneous
bruising in the neck region and/or internal hemorrhaging
in the thoracic or abdominal cavity. This combined with
water-saturated plumage and the close proximity of gill
Davoren
Gill-Net Fisheries and Common Murre Mortality
1039
Figure 3. Density of Common Murres observed during the regional-scale survey and the density of dead Common
Murres observed during both the regional-scale and fine-scale surveys in (a) 2000, ( b) 2001, (c) 2002, and (d)
2003. Dashed line is the survey route.
nets (Fig. 4) is strong evidence that the cause of death
was drowning due to entanglement in gill nets (Darby &
Dawson 2000).
The testes and ovaries of most dead birds were enlarged
or differentiated and most (93%) had brood patches, suggesting that they were breeders. The majority of birds
had prey in their stomachs (64%), but birds with brood
patches with empty stomachs might have been collecting
fish for their chick (Table 3). This suggests that most birds
were actively feeding prior to death. Over all years the
proportion of empty stomachs did not vary significantly
among collection days for either sex (males: χ23,40 = 6.906,
p >0.05; females: χ23,27 = 6.171, p > 0.05; Table 3). All
otoliths collected from stomachs (n = 247) were identified as capelin, indicating that the unidentified fish parts
in the stomach and gizzard were capelin. All undigested
Conservation Biology
Volume 21, No. 4, 2007
1040
Gill-Net Fisheries and Common Murre Mortality
Davoren
Table 2. Fine-scale estimates (± SE) of the total bycatch mortality of breeding Common Murres from Funk and Cabot islands in localized areas.
Survey
line/date
No.
No.
No.
Survey
No.
murres/km nets/km dead/kma length (km) dead/surveyb
Gull Island–2001
19-July
18.9 ± 7.0
27-July
24.9 ± 7.8
04-August
8.6 ± 3.3
05-August
29.3 ± 11.6
06-August
16.2 ± 6.7
08-August
15.7 ± 7.4
Gull Island–2002
04-August
12.9 ± 5.3
06-August
10.8 ± 3.3
08-August
14.6 ± 6.1
09-August
3.3 ± 1.2
10-August
8.8 ± 1.8
Gull Island–2003
09-August
2.5 ± 1.0
Line 1–2001
04-August
2.8 ± 0.8
06-August
1.1 ± 0.4
08-August
3.9 ± 0.8
Line 7–2001
21-July
36.7 ± 12.1
26-July
25.7 ± 5.4
02-August
30.1 ± 10.1
03-August
15.8 ± 5.0
04-August
36.1 ± 10.0
05-August
13.7 ± 5.1
Line 5–2001
22-July
44.6 ± 16.5
23-July
11.4 ± 3.0
24-July
4.6 ± 1.2
03-August
18.2 ± 8.7
05-August
2.2 ± 0.8
06-August
3.5 ± 1.4
Line 6–2001
21-July
198.5 ± 32.4
23-July
12.1 ± 2.7
04-August
11.6 ± 2.4
06-August
48.4 ± 16.5
No.
dead/day
No.
dead/yearc
260.3 ± 114.2 14053.7 ± 6166.7
0
0
0.4 ± 0.2 2.1 ± 0.6
2.6 ± 1.3 5.2 ± 0.7
2.4 ± 1.2 16.4 ± 0.7
1.0 ± 0.6 3.9 ± 0.7
4.2 ± 2.2 15.3 ± 0.7
37.5
35
35
37.5
30
37.5
–
74.6 ± 21.8
180.6 ± 23.1
614.9 ± 25.1
116.1 ± 22.0
575.3 ± 27.8
1.4 ± 1.6
1.4 ± 1.4
1.7 ± 1.0
1.2 ± 0.2
1.3 ± 0.6
4.2 ± 1.2
3.5 ± 1.1
5.7 ± 0.8
0.9 ± 0.5
2.6 ± 0.6
20
20
35
35
35
83.4 ± 24.7
69.5 ± 21.4
200.1 ± 26.5
32.0 ± 16.5
92.1 ± 19.5
2.2 ± 2.2
1.2 ± 1.1
27.5
0.1 ± 0.1
0.2 ± 0.2
0.2 ± 0.1
0.0 ± 1.1
0.1 ± 1.1
0.2 ± 0.7
18
17
17
0.7 ± 19.1
1.0 ± 18.8
3.6 ± 11.1
0
0
0.1 ± 0.1
0.5 ± 0.5
0.6 ± 0.6
0.2 ± 0.2
0
0
0.6 ± 1.1
1.9 ± 1.0
5.0 ± 1.1
0.7 ± 0.8
22
23
23
21
20
22
–
–
13.9 ± 24.4
40.1 ± 20.8
100.0 ± 21.3
15.8 ± 18.2
0
0
0
0
0
0
0
0
0
0
0
0
16
20
14
30
16
15
–
–
–
–
–
–
0
0
0
0
0
0
0
0
15
15
14
13
–
–
–
–
34.2 ± 30.3
% breeding
populationd
1.7 ± 0.7
95.4 ± 28.1
3052.7 ± 899.6
0.4 ± 0.1
34.2
–
1.8 ± 0.9
94.8 ± 51.1
0.01 ± 0.01
34.0 ± 16.2
1834.3 ± 873.4
0.2 ± 0.1
–
–
–
–
–
–
a Based on the bycatch rate of Common Murres to murre density, which was the number of murres (n = 3) caught in two sentinel fishery gill
nets on 6 August 2002 and the mean density of murres (6.5 ± 1.5 murres/km) within 2.5 km of this bycatch event over 3 days (4, 6, and 8
August 2002). Estimates of the total number of murres caught in gill nets per kilometer were calculated by integrating these numbers with the
mean density of murres and gill-net floats per kilometer in each survey.
b Derived by multiplying the total number of murres caught in gill nets per kilometer by the total length of each survey.
c Derived by multiplying the mean number of murres caught in gill nets within each area in each year by the number of days the breeding
season of murres (1 June–31 August) overlapped with the commercial fishery (2001, 9 July–30 November; 2002, 30 July–13 October).
d Funk Island breeding population was 824,000 individuals (Chardine et al. 2003); Cabot Island had 6,000 individuals (Cairns et al. 1989).
heads and tails were identified as capelin. An exception
was the presence of mysids in the stomachs of two murres. Of the 193 fish counted in stomachs, 91 were identified as capelin of unknown sex and maturity, and the
remaining 102 were identified as capelin of known sex
and maturity (spent females: n = 16; gravid females: n
= 56; mature males: n = 30; Table 3). The frequency
of the sex and maturity categories observed in stomachs
varied significantly among years (χ24,102 = 40.691, p <
0.001), with primarily gravid female and mature male
capelin observed. This suggests that murres were feeding
Conservation Biology
Volume 21, No. 4, 2007
on spawning shoals of capelin immediately prior to their
death.
Estimates of Dead Birds
Regional-scale estimates of Common Murre bycatch per
year, based on gill-net landings from all sources (commercial, sentinel, and bycatch) and the mean bycatch rate
in the sentinel fishery over the entire study area in 2002
(2.0 ± 0.7 murres/km), ranged from 1,936 to 4,973 individuals or 0.2–0.6% of the breeding populations of Funk
Davoren
Gill-Net Fisheries and Common Murre Mortality
1041
Figure 4. Density of Common Murres, number of dead murres, and number of gill nets per day observed during
the Gull Island fine-scale survey in (a) 2001 and ( b) 2002. Dashed line is the fine-scale survey route.
and Cabot islands in 2000–2002 (Table 4). In contrast,
fine-scale estimates of Common Murre bycatch within
the Gull Island area per year, based on bycatch rate estimates scaled to the density of murres and gill-net floats,
ranged from 3,053 to 14,054 individuals or 0.4–1.7% of
the breeding populations in 2001–2002, with fewer individuals (range: 95–1834) estimated in other fine-scale areas (Table 2). During years when regional and fine-scale
bycatch estimates were possible (2001–2002), bycatch
estimates in fine-scale areas exceeded regional-scale estimates.
Discussion
The biomass of birds and mammals was primarily concentrated along the coast, which generally reflects the
distributional patterns of capelin biomass (Davoren et al.
Conservation Biology
Volume 21, No. 4, 2007
1042
Gill-Net Fisheries and Common Murre Mortality
Davoren
Table 3. The percentage of nonbreeding Common Murres and murres with empty stomachs and the frequency of occurrence of a variety of sex and
maturity categories of capelin and other species found in the stomachs of dead female and male Common Murres collected at sea off the northeast
coast of Newfoundland in 2000–2002.a
Year/date
2000
28-Jul
28-Jul
2001
27-Jul
27-Jul
2002
4-Aug
4-Aug
8-Aug
8-Aug
a All
n
%
nonbreeders
% empty
stomachs
Total no.
of fish
Unknown
capelin
Spent
female capelin
Gravid
capelin
Male
capelin
Otherb
female
male
7
8
14
13
14
25
40
28
13
8
0
2
26
14
1
4
0
0
female
male
10
25
0
8
40
36
23
48
14
20
2
8
0
6
7
14
0
34
female
male
female
male
5
5
5
2
0
0
0
50
20
0
80
100
17
35
2
–
4
30
2
2
2
0
10
0
0
1
3
0
0
0
0
Sex
fish were identified as capelin, based on the species identification of n = 247 otoliths collected in the stomachs.
includes mysids.
b Other
2006). An aggregation of high abundances of top predators and capelin (i.e., a biological hotspot) was persistently observed near the demersal spawning sites of
capelin close to Gull Island within and among all years
of this study. Within this hotspot there was a high
spatial and temporal overlap of the distributional patterns of gill nets and top predators, especially Common
Murres. Concentrated fishing effort within this hotspot
resulted in the localized bycatch mortality of an estimated 3053–14054 murres/year due to their entanglement in gill nets while feeding on spawning capelin.
Estimates of bycatch mortality of murres within this
hotspot exceeded regional-scale estimates, indicating
that underestimates of mortality can arise from incorporating areas where fishing effort and animal density do not overlap (Ainley et al. 1981; Strann et al.
1991).
Bycatch of Common Murres
Common Murres have many attributes that make them
highly vulnerable to entanglement in gill nets relative to
other marine animals. The foraging habit of pursuit diving
generally results in an increased probability of encountering nets set below the surface (e.g., Ainley et al. 1981;
Piatt & Nettleship 1987). Alternately, surface feeders and
plunge divers may only be susceptible to entanglement in
nets set near the ocean surface (e.g., Northern Gannets
in salmon drift nets; Piatt & Nettleship 1987). The divedepth capabilities of the large-bodied murre exceed those
of other smaller-bodied pursuit-diving species (Watanuki
& Burger 1999), thereby allowing murres to encounter
nets set at a wider range of depths. As observed in this
study, even when many pursuit-diving species are found
together in an area where maximum dive depths can
Table 4. Regional-scale estimates of the total fishing effort for Atlantic cod in all gill-net fisheries (commercial, sentinel, and bycatch) and of the
mortality of breeding Common Murres in gill nets from Funk and Cabot islands during July and August 2000–2002.
Year
Parameters
A
B
C = B/A
D
E = (C × D) × 93%
F = E/A 830,000
a Mean
Parameter definition
mean catch rate of Atlantic cod (kg/km)a
landings (t) of Atlantic cod by gill netsb
fishing effort (number of km of gill nets fished)
mean (± SE) bycatch of Common Murre (birds/km)c
number (± SE) of breeding Common Murres killed
percentage (± SE) of breeding population killedd
2000
2001
2002
391.9
416.4
1063
2.0 ± 0.7
1936 ± 678
0.2 ± 0.1
214.1
584.4
2729
2.0 ± 0.7
4973 ± 1742
0.6 ± 0.2
120.9
186.2
1541
2.0 ± 0.7
2807 ± 983
0.3 ± 0.1
weight of cod caught per kilometer of gill net during the sentinel fishery in study area during 2002 (Rick Stead, Fisheries and Oceans
Canada, St. John’s, Newfoundland).
b Total reported landings (t) of Atlantic cod in study area from gill nets from all sources (commercial, sentinel, and bycatch) for July and August
in 2000 (Lilly et al. 2001), 2001, and 2002 (Lilly et al. 2003).
c Mean bycatch of Common Murres per kilometer during the sentinel fishery in study area during 2002 (Rick Stead, Fisheries and Oceans
Canada, St. John’s, Newfoundland).
d Funk Island breeding population was 824,000 individuals (Chardine et al. 2003); Cabot Island had 6,000 individuals (Cairns et al. 1989).
Conservation Biology
Volume 21, No. 4, 2007
Davoren
be reached by all species, other species (e.g., puffins)
tend to be less susceptible to entanglement than murres
despite similar relative abundances (Piatt & Nettleship
1987; Strann et al. 1991). This increased vulnerability
is likely related to the foraging behavior of murres (Piatt & Nettleship 1987). Preference for prey species with
clumped distributional patterns results in large foraging
aggregations formed at sea (Ainley et al. 2002; Davoren
et al. 2002, 2003), making murres highly vulnerable to
bycatch mortality in fisheries directed at their prey or a
commercially fished species with the same diet (Carter
& Sealy 1984; Piatt et al. 1984; Melvin et al. 1999; this
study).
The spatial and temporal overlap of high densities of
pursuit-diving birds and fishing effort generally results in
high bycatch rates (Carter & Sealy 1984; Piatt et al. 1984;
Takekawa et al. 1990), with rates being indicative of patterns of distribution and abundance at sea (Ainley et al.
1981; Piatt & Nettleship 1987). I found bycatch mortality
estimates were low in fine-scale areas where murre densities were high but the densities of gill nets were low (lines
1, 5, 6, 7). In addition, capelin biomass in the study area
was lower in 2002 relative to 2001 (Davoren et al. 2006),
which coincided with lower bird and mammal biomass,
lower landings of cod in the commercial fishery (Lilly et
al. 2003), and subsequently a lower estimated bycatch
mortality of murres in 2002.
The accuracy of quantitative bycatch estimates is limited by the types of data available. Regional-scale estimates generally incorporate mean bycatch rates (number
of animals caught per unit of fishing effort) and the total
fishing effort in a particular region. The underlying assumption is that bycatch rates are constant in space and
time, suggesting that the fine-scale spatial and temporal
overlap of fishing effort and animal densities does not vary
throughout a region. If this overlap is variable, as observed
in this study, mean bycatch rates at the regional-scale will
be substantially diluted by the inclusion of events with no
bycatch from areas where there is no overlap relative to
fine-scale estimates from areas where there is high overlap. I argue that my fine-scale estimates more accurately
represent the yearly bycatch mortality of Common Murres in the study area because the data are more detailed
at the finer scale and the overlap of murre and gill-net
densities were low throughout the larger study area. In
addition, fine-scale bycatch estimates per day supported
my visual estimates of at least hundreds of dead murres
per aggregation on days when aggregations of dead murres and surveys overlapped (27 July 2001; 4 and 8 August
2002; Table 2). The large error associated with bycatch estimates resulted from error surrounding mean densities
of murres and gill-net floats, which generally showed a
patchy distribution at sea (Fig. 4). Despite the uncertainty
associated with these estimates, this is an important first
step in quantifying the current bycatch mortality of Common Murres on the northeast coast of Newfoundland.
Gill-Net Fisheries and Common Murre Mortality
1043
Gill-Net Impact on the Common Murre Population
in Newfoundland
The gill-net fishing effort for cod in coastal Newfoundland
has declined dramatically since the groundfish moratorium in 1992; thus, seabird bycatch is considered significantly reduced (Ainley et al. 2002). Prior to the moratorium (1981–1984), a large-scale investigation of seabird
bycatch in coastal gill-net fisheries near four major breeding aggregations quantified ∼ 22,000 murres per year (Piatt & Nettleship 1987). I estimated similar numbers in one
fine-scale area in this study. Similarly, incidents of murre
bycatch continue to be reported in other coastal areas
during the breeding season ( Wilhelm et al. 2003). Overall this indicates that gill-net fishing in coastal areas continues to result in substantial mortality, especially when
fishing efforts are concentrated in biological hotspots.
To maintain a stable population of Common Murres,
mortality of adults should not exceed 6–12% of the breeding population (Piatt et al. 1984). This range is based on
several life-history characteristics (Ainley et al. 2002) and
indicates that reduced annual survivorship of breeding
adults will have the highest impact on the population.
The mortality of parents likely results in the death of the
chick as well, lowering the potential for future recruitment. Overall, the level of chronic, unnatural, and additive mortality every year from bycatch estimated in this
study (0.4–1.7%) would reduce the growth potential of
murre populations and result in increased vulnerability of
this species to population declines, as argued by Wiese et
al. (2004), even though this species is considered highly
abundant.
For several years bycatch rates of breeding murres
near colonies have led to substantial population declines
of murres elsewhere. For instance, an estimated 10,000
breeding murres killed per year in gill nets for 7 years
caused a 53% decline in a breeding population of 229,080
individuals in California (Takekawa et al. 1990), and an estimated 13,000 breeding murres killed per year in gill nets
for 25 years caused a 95% decline in a breeding population
of 250,000 individuals in Norway (Strann et al. 1991).
Owing to the difficulties of censusing large colonies
of seabirds, it is hard to assess whether bycatch mortality has had an impact on population size until there are
substantial declines (Chardine 1998). Population trends
available for Common Murres at Funk Island suggest that
this colony is stable (Chardine et al. 2003), whereas other
colonies are increasing (Robertson et al. 2006). This indicates that the Funk Island colony is either at carrying
capacity or that bycatch mortality, combined with other
demographic parameters (e.g., Davoren & Montevecchi
2003a,b), is high enough to limit population growth. This
restricted growth potential will likely affect the entire
northwestern Atlantic population of murres because 75%
of the population breeds on Funk Island (Cairns et al.
1989).
Conservation Biology
Volume 21, No. 4, 2007
1044
Gill-Net Fisheries and Common Murre Mortality
Conservation and Management Implications
High bycatch mortality of nontarget species occurs in areas where high abundances of marine organisms overlap
with fishing grounds (e.g., Carter & Sealy 1984; Piatt et al.
1984; Takekawa et al. 1990; this study). Under Canada’s
Oceans Act a framework exists to protect “ecologically
and biologically significant areas” (EBSA) or areas that
play a major role in maintaining the structure and function of ecosystems (DFO 2004). Protection is geared toward nonrepresentative oceanic areas, where organisms
are highly aggregated for an important part of their annual
cycle. Use of these areas should enhance potential population growth of targeted species or the maintenance of
important ecosystem processes (DFO 2004). The persistence of capelin at coastal spawning sites during the summer shapes the foraging migrations of large predatory fish
(e.g., Atlantic cod; Rose 1993) and whales (e.g., Whitehead & Cascadden 1985) and the breeding chronology
and success of seabirds (Davoren & Montevecchi 2003b).
The formation of biological hotspots at these spawning
sites concentrates predator–prey interactions, maximizing energy flow through the marine food web, a key
ecosystem process. These characteristics make the Gull
Island area an excellent candidate for protection under
the EBSA framework. Reduced seabird bycatch has been
accomplished in other regions either by altering fishing
gear and techniques (Melvin et al. 1999) or by reducing
the spatial and temporal overlap of fisheries activities with
bird distribution (e.g., Takekawa et al. 1990).
I recommend that the use of gill nets within biological hotspots associated with demersal spawning sites of
capelin be restricted during the period when capelin are
spawning (mid-July–mid-August; Penton 2006) because
these persistent aggregations of spawning capelin likely
drive the concentration of top predators within these
areas. This should ensure that murre bycatch is minimized and that energy flow through this marine food
web is maintained. Even though the commercial cod
fishery closed in 2003, gill nets with similar mesh sizes
are still used in coastal Newfoundland to harvest other
species (e.g., winter flounder [Pseudopleuronectes americanus], lumpfish [Cyclopterus lumpus]), resulting in
continued bycatch of murres and Atlantic cod, currently
designated as an endangered species in Canada (Hutchings 2003). As in other fishery-regulated areas, the potential and efficacy of this protection will depend critically
on the level of support from local community members,
especially fishers, for successful protection (Hooker &
Gerber 2004).
Acknowledgments
I gratefully acknowledge A.D. Murphy for directing, operating, and managing all technical equipment and electronic data aboard the Shamook and Lady Easton II. I
Conservation Biology
Volume 21, No. 4, 2007
Davoren
also thank C. Burke, E. MacDonald, G. Redmond, and the
captain and crews of the both vessels for assistance with
field work. Thanks to P. Eustace for identifying the species
of all fish otoliths. I especially thank R. Stead for preparation of the sentinel fishery data. Funding was provided
by Natural Science and Engineering Research Council
of Canada (NSERC) postgraduate scholarship, Mountain
Equipment Co-op, Royal Bank Marine Studies Fund, The
National Chapter of Canada Imperial Order Daughters of
the Empire War Memorial Scholarship, Orville Erickson
Memorial Fund, and Canadian Federation of University
Women to G.K.D. Operating funds were also provided by
NSERC Discovery and NSERC ship-time grants to William
A. Montevecchi and Fisheries and Oceans vessel support
to John T. Anderson.
Literature Cited
Ainley, D. G., D. N. Nettleship, H. R. Carter, and A. E. Storey. 2002. Common Murre Uria aalge. Pages 1–144 in A. Poole and F. Gill, editors.
The birds of North America. Number 666. American Ornithologists’
Union, Philadelphia, Pennsylvania.
Ainley, D. G., A. R. DeGange, L. L. Jones, and R. J. Beach. 1981. Mortality
of seabirds in high-seas salmon gill nets. Fishery Bulletin 79:800–
806.
Cairns, D. K., W. A. Montevecchi, and W. Threlfall. 1989. Researcher’s
guide to Newfoundland seabird colonies. Occasional paper in biology, number 14. Memorial University of Newfoundland, St. John’s,
Newfoundland and Labrador.
Carscadden, J. E., and H. Vilhjálmsson. 2002. Capelin—What are they
good for? ICES Journal of Marine Science 59:863–869.
Carter, H. R., and S. G. Sealy. 1984. Marbled Murrelet mortality due to
gill-net fishing in Barkley Sound, British Columbia. Pages 212–220
in D. N. Nettleship, G. A. Sanger, and P. F. Springer, editors. Marine
birds: their feeding ecology and commercial fisheries relationships.
Canadian Wildlife Service Special Publication, Ottawa.
Chardine, J. W. 1998. Review of the seabird bycatch problem in Acrtic Canada. Pages 9–14 in Conservation of Acrtic flora and fauna.
Technical report number 1. Circumpolar Seabird Working Group,
Ottawa, Ontario.
Chardine, J. W., G. J. Robertson, P. C. Ryan, and B. Turner. 2003. Abundance and distribution of Common Murres breeding at Funk Island,
Newfoundland, 1972 and 2000. Technical report series 404. Canadian Wildlife Service, Atlantic Region, Sackville, New Brunswick,
Canada.
Darby, J. T., and S. M. Dawson. 2000. Bycatch of yellow-eyed penguins
(Megadyptes antipodes) in gillnets in New Zealand waters 1979–
1997. Biological Conservation 93:327–332.
Davoren, G. K., and W. A. Montevecchi. 2003a. Consequences of foraging trip duration on provisioning behavior and fledging condition
of Common Murres. Journal of Avian Biology 34:44–53.
Davoren, G. K., and W. A. Montevecchi. 2003b. Signals from seabirds
indicate changing fish stocks. Marine Ecology Progress Series
258:253–261.
Davoren, G. K., J. T. Anderson, and W. A. Montevecchi. 2006. Shoal
behavior and maturity relations of spawning capelin (Mallotus villosus) off Newfoundland: demersal spawning and diel vertical movement patterns. Canadian Journal of Fisheries and Aquatic Sciences
63:268–284.
Davoren, G. K., W. A. Montevecchi, and J. T. Anderson. 2002. Scaledependent associations of predators and prey:constraints imposed
by flightlessness of Common Murres. Marine Ecology Progress Series
245:259–272.
Davoren
Davoren, G. K., W. A. Montevecchi, and J. T. Anderson. 2003. Search
strategies of a pursuit-diving marine bird and the persistence of prey
patches. Ecological Monographs 73:463–481.
Davoren, G. K., W. A. Montevecchi, and J. T. Anderson. 2004. The influence of fish behavior on search strategies of Common Murres
Uria aalge in the Northwest Atlantic. Marine Ornithology 31:123–
131.
Department of Fisheries and Oceans. 2004. Identification of ecologically and biologically significant areas. Scientific Advisory Secretariat ecosystem status report 2004/006. Department of Fisheries
and Oceans Canada, Ottawa, Ontario.
Frank, K. T., B. Petri, J. S. Choi, and W. C. Leggett. 2005. Trophic cascades
in a formerly cod-dominated ecosystem. Science 308:1621–1623.
Hooker, S. K., and L. R. Gerber. 2004. Marine reserves as a tool
for ecosystem-based management: the potential importance of
megafauna. BioScience 54:27–39.
Hutchings, J. A. 2003. Update COSEWIC status report on Atlantic cod
Gadus morhua in Canada. Committee on the Status of Endangered
Wildlife in Canada, Ottawa, Ontario.
Hyrenbach, K. D., K. A. Forney, and P. K. Dayton. 2000. Marine protected
areas and ocean basin management. Aquatic Conservation 10:437–
458.
Johnson, A., G. Salvador, J. Kenney, J. Robbins, S. Landry, P. Clapham,
and S. Kraus. 2005. Fishing gear involved in entanglements of
right and humpback whales. Marine Mammal Science 21:635–
645.
Lilly, G. R., P. A. Shelton, J. Brattey, N. G. Cadigan, B. P. Healey, E. F.
Murphy, and D. E. Stansbury. 2001. An assessment of the cod stock
in NAFO Divisions 2J+3KL. Scientific Advisory Secretariat research
document 2001/044. Department of Fisheries and Oceans Canada,
Ottawa, Ontario.
Lilly, G. R., P. A. Shelton, J. Brattey, N. G. Cadigan, B. P. Healey, E. F.
Murphy, D. E. Stansbury, and N. Chen. 2003. An assessment of the
cod stock in NAFO Divisions 2J+3KL in February 2003. Scientific
Advisory Secretariat research document 2003/023. Department of
Fisheries and Oceans Canada, Ottawa, Ontario.
Melvin, E. F., J. K. Parrish, and L. L. Conquest. 1999. Novel tools to
reduce seabird bycatch in coastal gillnet fisheries. Conservation Biology 13:1386–1397.
Myers, R. A., and B. Worm. 2003 Rapid worldwide depletion of predatory fish communities. Nature 423:280–283.
Myers, N., R. A. Mittermeler, C. F. Mittermeler, G. A. B. da Fonseca, and J.
Kent. 2000. Biodiversity hotspots for conservation priorities. Nature
403:853–858.
Pauly D., V. Christensen, S. Guenette, T. J. Pitcher, U. R. Sumaila, C. J.
Walters, R. Watson, and D. Zeller. 2002. Towards sustainability in
world fisheries. Nature 418:689–695.
Gill-Net Fisheries and Common Murre Mortality
1045
Penton, P. 2006. A comparison of beach and demersal spawning in
capelin (Mallotus villosus) on the Northeast coast of Newfoundland.
M.S. thesis. University of Manitoba, Winnipeg, Manitoba, Canada.
Piatt, J. F., and D. N. Nettleship. 1987. Incidental catch of marine birds
and mammals in fishing nets off Newfoundland, Canada. Marine
Pollution Bulletin 18:344–349.
Piatt, J. F., D. N. Nettleship, and W. T. Threlfall. 1984. Net mortality of
Common Murres Uria aalge and Atlantic Puffins Fratercula acrtica
in Newfoundland, 1951–1981. Pages 196–206 in D. N. Nettleship,
G. A. Sanger, and P. F. Springer, editors. Marine birds: their feeding
ecology and commercial fisheries relationships. Special publication.
Canadian Wildlife Service, Ottawa.
Robertson, G. J., A. E. Storey, and S. I. Wilhelm. 2006. Local survival
rates of Common Murres breeding in Witless Bay, Newfoundland.
Journal of Wildlife Management 70:584–587.
Rose, G. A. 1993. Cold spawning on a migration highway in the northwest Atlantic. Nature 366:458–461.
Soulé, M. E., and G. H. Orians. 2001. Conservation biology: research
priorities for the next decade. Island Press, Washington, D.C.
Strann, K.-B., W. Vader, and R. Barrett. 1991. Auk mortality in fishing
nets in north Norway. Seabird 13:22–29.
Takekawa, J. E., H. R. Carter, and T. E. Harvey. 1990. Decline of the
Common Murre in central California, 1980–1986. Studies in Avian
Biology 14:149–163.
Tasker, M. L., P. Hope Jones, T. Dixon, and B. F. Blake. 1984. Counting
seabirds at sea from ships: a review of methods employed and a
suggestion for a standardized approach. Auk 101:567–577.
Vader W., R. T. Barrett, K. E. Erikstad, and K.-B. Strann K-B. 1990. Differential responses of common and thick-billed murres to a crash in the
capelin stock in the southern Barents Sea. Studies in Avian Biology
14:175–180.
Watanuki, Y., and A. E. Burger. 1999. Body mass and dive duration
in alcids and penguins. Canadian Journal of Zoology 77:1838–
1842.
Whitehead, J., and J. E. Carscadden. 1985. Predicting inshore whale
abundance – whales and capelin off the Newfoundland coast. Canadian Journal of Fisheries and Aquatic Sciences 42:976–981.
Wiese, F. K., G. J. Robertson, and A. J. Gaston. 2004. Impacts of chronic
marine oil pollution and the murre hunt in Newfoundland on thickbilled murre Uria lomvia populations in the eastern Canadian Acrtic. Biological Conservation 116:205–216.
Wilhelm, S. A., G. J. Robertson, P. A. Taylor, S. G. Gilliland, and D. L. Pinsent. 2003. Stomach contents of breeding Common Murres caught
in gillnets off Newfoundland. Waterbirds 26:376–378.
Worm, B., M. Sandow, A. Oschlies, H. K. Lotze, and R. A. Myers. 2005.
Global patterns of predator diversity in the open oceans. Science
309:1365–1369.
Conservation Biology
Volume 21, No. 4, 2007