ArticlePDF Available

The good, the bad and the ugly: ENSO-driven oceanographic variability and its influence on seabird diet and reproductive performance at the Houtman Abrolhos, Eastern Indian Ocean

Authors:

Abstract and Figures

Each spring/summer more than a million pairs of seabirds breed at the Houtman Abrolhos, Western Australia, the most significant seabird breeding site in the Eastern Indian Ocean. The southward-flowing Leeuwin Current is the dominant oceanographic feature influencing the region, in conjunction with the northward-flowing Leeuwin Undercurrent and the Capes Current. Seabirds at the Houtman Abrolhos are reliant wholly upon marine sources of food, and several species feed predominately upon larval ichthyoplankton species, the availability of which has been found to play a pivotal role in their reproduction. We conducted a comparative study of the timing of breeding, breeding participation and reproductive success over a 17-year period of four tropical pelagic seabird species in relation to the regional oceanographic conditions affecting the Leeuwin Current. Three tern species, the Lesser Noddy Anous tenuirostris, Brown Noddy A. stolidus and Sooty Tern Sterna fuscata, and the Wedge-tailed Shearwater Puffinus pacificus comprised our study species at the Houtman Abrolhos between 1991 and 2007. The diet of these species was also investigated between 1991 and 2000. Life-history traits determined the response of these seabirds to fluctuations in marine resources caused by variation in the flow of the Leeuwin Current. During El Niño Southern Oscillation (ENSO) events, reproductive effort and output were severely reduced in all species, which coincided with reduced volumes of key prey species in regurgitates. Between 1991 and 2000, ENSO driven changes in the Leeuwin Current resulted in lower participation rates and reduced breeding success or catastrophic breeding failure for all four seabird species and delayed timing of breeding for the tern species. Between 2000 and 2007, the relationship between each ENSO event and a subsequent poor reproductive season was not as strong. Increasing years of poor breeding performance were recorded outside El Niño periods, accompanied by a significant seasonal delay in the onset of breeding in the three tern species. Based on our seabird observations, we postulate that in recent years the high number of ENSO events have resulted in a regime shift in offshore and oceanic planktonic food chains off central Western Australia. The use of seabirds as an upper-trophic-level indicator of change in marine productivity as a result of variability in the Leeuwin Current system is discussed. The possibility that tropical seabirds in the region are adapting to a new suite of prey, dictated by a Leeuwin Current system which is influenced by more frequent ENSO events, is also discussed.
Content may be subject to copyright.
Surman & Nicholson: ENSO-driven oceanographic variability at the Houtman Abrolhos 129
Marine Ornithology 37: 129–138 (2009)
129
INTRODUCTION
The effects of prey availability on the reproductive performance of
seabirds has been well documented for seasonal climates (Monaghan
et al. 1992, Kitaysky et al. 1999, Rindorf et al. 2000, Furness
& Tasker 2000); the effects of less predictable patterns of prey
availability upon tropical seabirds are still being explored (Ramos
2000, Ramos et al. 2002, Surman & Wooller 2003, Jaquemet et
al. 2007, Monticelli et al. 2007). Seabirds respond negatively to
reduced prey availability, and this effect is compounded in surface
feeders (as compared with pursuit divers) and in seabirds with
low diet diversity (Gaughan et al. 2002). Typical responses are
reduced breeding success (Ramos 2000, Rindorf et al. 2000) and
reduced breeding attempts (Crawford & Dyer 1995). The localised
and regional environmental factors that affect prey availability,
as measured by changes to local seabird breeding phenology and
effort, are just beginning to be identified and understood (Ashmole
& Ashmole 1968, Diamond 1983, Schreiber & Schreiber 1986,
Gaughan et al. 2002, Ramos et al. 2002, Frederiksen et al. 2004,
Jaquemet et al. 2007, Monticelli et al. 2007).
The influence of the El Niño Southern Oscillation (ENSO) on
ocean currents, sea levels and fisheries off the Western Australian
coast has been well documented (Pearce & Walker 1991, Pearce
1997). There is evidence that the ENSO profoundly influences
the availability of prey to the region’s tropical pelagic seabirds,
resulting in reduced breeding participation and reduced breeding
success, probably by inducing changes in the oceanographic
conditions that drive marine productivity (Surman 1998, Dunlop et
al. 2002, Gaughan et al. 2002, Nicholson 2002).
The Houtman Abrolhos is located at 28°S and contains one of
the southernmost extensions of tropical marine flora and fauna in
the Indian Ocean, principally because of the flow of the Leeuwin
THE GOOD, THE BAD AND THE UGLY: ENSO-DRIVEN
OCEANOGRAPHIC VARIABILITY AND ITS INFLUENCE ON SEABIRD
DIET AND REPRODUCTIVE PERFORMANCE AT THE
HOUTMAN ABROLHOS, EASTERN INDIAN OCEAN
CHRISTOPHER ANDREW SURMAN1 & LISA WENDY NICHOLSON
1Halfmoon Biosciences, 604 Ocean Beach Road, Denmark, 6333, Western Australia
(halfmoon.biosciences@westnet.com.au)
Received 13 August 2008, accepted 4 February 2009
SUMMARY
SURMAN, C.A. & NICHOLSON, L.W. 2009. The good, the bad and the ugly: ENSO driven oceanographic variability and its influence on
seabird diet and reproductive performance at the Houtman Abrolhos, Eastern Indian Ocean. Marine Ornithology 37: 129–138.
Each spring/summer more than a million pairs of seabirds breed at the Houtman Abrolhos, Western Australia, the most significant seabird
breeding site in the Eastern Indian Ocean. The southward-flowing Leeuwin Current is the dominant oceanographic feature influencing the
region, in conjunction with the northward-flowing Leeuwin Undercurrent and the Capes Current. Seabirds at the Houtman Abrolhos are
reliant wholly upon marine sources of food, and several species feed predominately upon larval ichthyoplankton species, the availability of
which has been found to play a pivotal role in their reproduction. We conducted a comparative study of the timing of breeding, breeding
participation and reproductive success over a 17-year period of four tropical pelagic seabird species in relation to the regional oceanographic
conditions affecting the Leeuwin Current. Three tern species, the Lesser Noddy Anous tenuirostris, Brown Noddy A. stolidus and Sooty Tern
Sterna fuscata, and the Wedge-tailed Shearwater Puffinus pacificus comprised our study species at the Houtman Abrolhos between 1991 and
2007. The diet of these species was also investigated between 1991 and 2000. Life-history traits determined the response of these seabirds
to fluctuations in marine resources caused by variation in the flow of the Leeuwin Current. During El Niño Southern Oscillation (ENSO)
events, reproductive effort and output were severely reduced in all species, which coincided with reduced volumes of key prey species
in regurgitates. Between 1991 and 2000, ENSO driven changes in the Leeuwin Current resulted in lower participation rates and reduced
breeding success or catastrophic breeding failure for all four seabird species and delayed timing of breeding for the tern species. Between
2000 and 2007, the relationship between each ENSO event and a subsequent poor reproductive season was not as strong. Increasing years
of poor breeding performance were recorded outside El Niño periods, accompanied by a significant seasonal delay in the onset of breeding
in the three tern species. Based on our seabird observations, we postulate that in recent years the high number of ENSO events have resulted
in a regime shift in offshore and oceanic planktonic food chains off central Western Australia. The use of seabirds as an upper-trophic-level
indicator of change in marine productivity as a result of variability in the Leeuwin Current system is discussed. The possibility that tropical
seabirds in the region are adapting to a new suite of prey, dictated by a Leeuwin Current system which is influenced by more frequent ENSO
events, is also discussed.
Key words: ENSO, Leeuwin Current, Seabirds, Houtman Abrolhos, Diet, Reproductive Performance
130 Surman & Nicholson: ENSO-driven oceanographic variability at the Houtman Abrolhos
Marine Ornithology 37: 129–138 (2009)
Current, which is a body of warm (>24°C), low-salinity (<35‰)
water that flows southward along the continental shelf of Western
Australia in a broad and shallow band (200 km wide by 50 m depth)
at speeds of up to 2 km h–1 (Cresswell 1990, Pearce & Walker
1991). The strength of the flow of the Leeuwin Current varies both
annually in its cycle and seasonally. It flows principally in winter,
between April and July, and its strength and temperature are affected
by ENSO events. In ENSO years, the Leeuwin Current is weaker,
resulting in cooler, more saline water along the outer continental
shelf (Pearce & Walker 1991). Higher sea levels and warmer
sea temperatures occur along the Western Australian coastline
during non-ENSO years, resulting in the Leeuwin Current having
a stronger southward flow (Cresswell et al. 1989). Associated
with the Leeuwin Current are the northward-flowing Leeuwin
Undercurrent at the subsurface, and the northward-flowing Capes
Current on the continental shelf during the summer months
(Pattiaratchi 2005). The dynamics of the Leeuwin Current system
in the delivery and distribution of seabird ichthyoplankton prey at
the Houtman Abrolhos is not clearly understood.
Most studies agree that seabird breeding is specifically timed to
coincide with maximal food supply (Ashmole & Ashmole 1968,
Diamond 1983, Ainley et al. 1986, Ramos et al. 2002, Frederiksen
et al. 2004, Jaquemet et al. 2007, Monticelli et al. 2007). This
timing varies with latitude, with high-latitude species having a
narrower period of opportunity to breed because of climate and
seasonal food availability, resulting in predictable breeding times,
high synchrony and relatively short breeding seasons. At lower
latitudes, however, breeding at some locations may commence at
any time, and timing is determined only by the availability of food
in reach of the colony. Because of the influence of the Leeuwin
Current system, the waters surrounding the Houtman Abrolhos
provide breeding seabirds with a mixture of tropical and temperate
prey sources (Surman & Wooller 2003) and a breeding timetable
that is highly synchronous but that, for many species, can shift in
timing from year to year. The seabirds that breed at the Houtman
Abrolhos are considered predominantly tropical, and they breed
annually during the austral spring/summer (Surman 1998). This
location is the central breeding area for the eastern Indian Ocean
metapopulations of the Lesser Noddy Anous tenuirostris, Brown
Noddy A. stolidus, Sooty Tern Sterna fuscata and Wedge-tailed
Shearwater Puffinus pacificus (Surman 1997). The terns species
share breeding times of between 75 and 85 days (Surman 1998);
the Wedge-tailed Shearwater has a much longer breeding time of
120 days (Serventy et al. 1971) and, as a consequence, has the least
ability to shift timing of breeding. The Lesser Noddy is a resident
species at this location; the other three species migrate or disperse
northwards during the austral winter (Surman 1997).
This paper aims to determine the influence of ENSO-induced
variability in the Leeuwin Current system upon the reproductive
performance of four seabird species breeding at the Houtman
Abrolhos, and the relationship between current flow, food supply
and reproductive performance. We hypothesized that, at the
Houtman Abrolhos, seabirds were breeding later and were being
less successful as time progressed during the 17-year study.
STUDY AREA AND METHODS
The study was conducted on Pelsaert Island (28°56S, 113°5830ʺE),
the southernmost and third largest of an archipelago of 192 islands,
islets and rocks (the Houtman Abrolhos), 60 km off the mid-
western coast of Australia (Fig. 1). Pelsaert Island (120 ha) is 12 km
long, only 50–500 m wide and very low (<3 m Above High Water
Springs), comprising coral rubble, limestone and sand.
Study species
On Pelsaert Island, the Lesser Noddy (35 000 pairs) nests in White
Mangrove Avicennia marina trees, the Brown Noddy (130 000 pairs)
nests on low Nitre Nitraria billardierei and Samphire Halosarcia
halocnemoides bushes, the Sooty Tern (260 000 pairs) nests on
bare ground below herbaceous Atriplex cinerea and Nitre bushes,
and the Wedge-tailed Shearwater (75 000 pairs) digs burrows up
to one metre long in sandy areas throughout the southern portion
of the island (Surman & Wooller 1995, 2000). All four species
breed during the austral spring/summer. At the Houtman Abrolhos,
Pelsaert Island contains 99% of Brown Noddies and 70% of Lesser
Noddies breeding along the Western Australian coastline. It also
contains the largest colony of Sooty Terns in the Eastern Indian
Ocean and significant colonies of Wedge-tailed Shearwaters.
Breeding phenology, participation and success
Nest sites of each species were selected at random and permanently
marked. Nest contents were recorded weekly over the spring/
Fig. 1. Timing of breeding (Julian date, where 27 October = 300) as
represented by the date of first egg (closed diamonds) and mean lay
date (open squares) in the (A) Sooty Tern Sterna fuscata, (B) Brown
Noddy Anous stolidus and (C) Lesser Noddy A. tenuirostris, with
trend line for the mean lay date for Lesser and Brown Noddies and
date of first egg for Sooty Terns. The sea levels at Geraldton (open
squares) and Fremantle (closed squares) and at the bottom the
Southern Oscillation Index (SOI) are shown for the same period. El
Niño–Southern Oscillation events are associated with negative SOI
values and lower sea levels.
(A)
240
260
280
300
320
340
360
(B)
220
240
260
280
300
320
340
360
(C)
220
240
260
280
300
320
340
360
Linear
(Fre mant le)
Gerald to n
Sea level
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Freman tle
-30
-20
-10
0
10
20
30
Julian DateJulian DateJulian DateSea Level (m)SOI
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
SOI
Surman & Nicholson: ENSO-driven oceanographic variability at the Houtman Abrolhos 131
Marine Ornithology 37: 129–138 (2009)
summer period (September to May) between 1991 and 2001, and
in more recent years (2002–2007) during several shorter visits
between October and January of each year. There are some data
gaps in these latter years, which were excluded from analyses. The
burrow contents of the Wedge-tailed Shearwater were determined
using an electronic burrowscope. Breeding participation was
recorded as the percentage of all nest sites or burrows that contained
an active breeding attempt during the breeding season.
Laying chronology was determined using lay dates of known-age
eggs; the laying dates of other eggs was estimated by backdating,
using egg water-loss techniques (Wooller & Dunlop 1980, Surman
& Wooller 1995). Eggs known to be re-laid were excluded from
calculations of the mean date of laying for each species. Chick age
was estimated from growth curves described in Surman (1997).
As a measure of reproductive performance, we used breeding
success or the proportion of all active breeding attempts that
survived to produce a fledgling. We arbitrarily assigned those
seasons when breeding success was less than 15% overall for each
study species as “bad,” and those above 15% as “good.”
In more recent years, we have noticed that the commencement of
breeding was later for the three terns studied. To test the hypothesis
that, at the Houtman Abrolhos, seabirds were breeding later and
were being less successful as time progressed, we divided our data
into the periods 1991–1999 inclusive and 2000–2007 inclusive.
Environmental parameters
We measured ENSO events using the Southern Oscillation Index
(SOI), which is based on the difference in atmospheric pressure
between Tahiti, in the Pacific, and Darwin in northern Australia. The
SOI is expressed as positive values (La Niña) or negative values (El
Niño) (Pearce & Walker 1991). The mean monthly SOI was obtained
from the Australian Bureau of Meteorology (www.bom.gov.au/).
The strength of flow of the Leeuwin Current was determined by
using the mean monthly sea level at Fremantle and Geraldton,
Western Australia, and these data were obtained from the National
Tidal Facility, Flinders University, Australia. Sea level is measured
by remote tidal gauges in centimetres and is presented in units of
metres. Mean monthly Sea Surface Temperature (SST, in degrees
Celsius) was obtained using the Reynold’s SST database for the
latitude band at 28.5°S, between 113.5°E and 114.5°E. Data were
kindly provided by Alan Pearce.
Dietary samples
Adult birds recently returned from a foraging trip were captured
at the nest-site by hand or using a small net, whereupon most
would spontaneously regurgitate. They were marked individually
with leg bands to ensure that no individual was sampled more
than once during any single breeding season. A minimum of
10–20 regurgitations were collected from each tern species in each
sampling month, corresponding to the incubation, small nestling
and large nestling stages in their breeding cycle. Regurgitations
were collected from shearwaters during the nestling stages of their
breeding cycle only, because earlier in the season, they were less
likely to spontaneously regurgitate. Sampling was not undertaken
in the period 2002–2007.
Regurgitates were preserved in 70% ethanol, then rinsed and
vacuum filtered to allow the percentage volumes of identifiable
material to be estimated before obtaining the wet mass of each
sample. We used a volumetric estimate that determined the surface
area of a 14-cm diameter Petri dish covered by each prey type. Prey
items were identified using keys (Last et al. 1983, Leis & Rennis
1983, Smith & Heemstra 1986, Leis & Trnski 1989, Gommon et
al. 1994) and from reference specimens in the Western Australian
Museum. The detailed methodology is described in Surman and
Wooller (2003).
Statistical analysis
Data was analysed using the Statistica statistical software package
(StatSoft, Tulsa, OK, USA) and was based on statistical methods
described in Zar (1996). We arcsine-transformed data expressed
as percentages before undertaking analysis. Long-term trends were
investigated by pooling data from the study years 1991–1999 and
2000–2007 inclusive.
RESULTS
Timing of breeding
Wedge-tailed Shearwaters commenced breeding on or near
17 November each year, with no discernable variability. The mean
lay date for this species varied by a single day around 23 November
in each year studied. The commencement of breeding was much
more variable for the three tern species.
Fig. 1 shows the date of first eggs and the mean lay date for Sooty
Terns, Lesser Noddies and Brown Noddies at Pelsaert Island. Later
breeding coincided with stronger ENSO events (negative SOI values)
and lower sea levels. All three tern species were delayed significantly
during the 1996/97 and 1997/98 austral summers, with mean lay
dates 60 days later (Sooty Tern), 65 days later (Lesser Noddy) and 98
days later (Brown Noddy) than in the preceding years. This change
was not significantly related to Fremantle sea levels between 1991
and 1998 (F1,6 = 3.55, P = 0.11); however, between 1991 and 1998,
the mean lay dates for both the Lesser (y = 0.02x + 18.05, R2 = 0.67,
F1,6 = 12.18, P = 0.012) and the Brown Noddy (y = 0.02x + 19.02,
R2 = 0.74, F1,6 = 11.73, P = 0.026) were correlated with the SST
in January and March of the breeding year. These species returned
to early breeding during the1999/00 and 2000/01 seasons, which
coincided with the SOI becoming positive once again. The data
suggest that the commencement of breeding in the three tern species
had become later over the past 17 years (Fig. 1). The variability in
the onset of breeding was greatest in Brown Noddies, with a range of
103 days. In years when breeding was delayed until November, there
was a shorter period between the first egg laid and the mean lay date
for Lesser and Brown Noddies alike.
Between 1991 and 1999, the mean dates of the first egg laid on
Pelsaert Island for Lesser and Brown Noddies were 11 September
and 22 September respectively; for the Sooty Tern, the mean date
was 16 October. Between 2001 and 2007, this date was delayed to
15 October for Lesser Noddies, 28 October for Brown Noddies, and
7 November for Sooty Terns. The differences were significantly for
the two noddy species (Lesser: t15 = –2.52, P = 0.01; Brown: t =
–1.99, df = 12, P = 0.03), but not for the Sooty Tern (t11 = –1.68,
P = 0.06). The mean lay date had also increased by 24 and 30 days
in Lesser and Brown Noddies respectively, which was significantly
132 Surman & Nicholson: ENSO-driven oceanographic variability at the Houtman Abrolhos
Marine Ornithology 37: 129–138 (2009)
later for the Lesser Noddy (t = –2.17, df = 15, P = 0.02), but not for
the Brown Noddy (t13 = –1.31, P = 0.11, Table 1).
Breeding participation
From 1991 to 2007, delayed breeding was accompanied by a
decline in the proportion of birds participating in that year for all
tern species (Fig. 2). Poor breeding participation was most evident
during the lead-up to, and continuing into, the 1997/98 and 2002/03
ENSO events for all four species. Wedge-tailed Shearwaters did not
attempt to breed at all during the 1997/98 ENSO event. The Lesser
Noddy appeared to be the least affected of the four species by these
events (Fig. 2) and, overall, had the least fluctuation in breeding
participation (from 39.6% to 54.7%) when the mean participation
rates for bad years and good years were compared (Table 2). Mean
Sooty Tern breeding participation varied the most between bad and
good years, from 0.3% to 49.5% (t9 = 4.82, P < 0.001). Mean Brown
Noddy breeding participation also varied significantly (t9 = –3.16,
P = 0.008), but mean breeding participation in both Lesser Noddies
(t12 = –1.49, P = 0.10) and Wedge-tailed Shearwaters (t7 = 2.02,
P = 0.07) was not significantly lower in “bad” years, even though
it was reduced (Table 2). Brown Noddy, Sooty Tern and Wedge-
tailed Shearwater participation was not significantly correlated
with sea level or SST; however, the number of participating Lesser
Noddies was significantly related to the SST between January
and March of the breeding year during 1991–1998 (y = –0.02x
+ 24.14, R2 = 0.78, P = 0.004). There was also a significant linear
relationship between Lesser Noddy participation and Fremantle sea
levels (y = 0.002x + 0.786, R2 = 0.47, P = 0.03).
Breeding success
Reproductive success was greatly reduced during ENSO years
(Fig. 3). The mean SOI in reproductively bad years was –48.4,
compared with 45.6 in reproductively good years. The sea level
in Geraldton was 0.55 m and 0.58 m respectively (Table 3). In the
Wedge-tailed Shearwater, reproductive output was zero for both the
1997/98 and 2007/08 austral summers on Pelsaert Island. For the
two noddy species, success was very low in the 1996/97, 1997/98,
2002/03 and 2004/05 seasons, coinciding with ENSO events. In the
2007/08 season, the Brown Noddy fared poorly, although the Lesser
Noddy had more than 40% breeding success in the same year. A
robust indicator of breeding success was not obtained for the Sooty
Tern, because it was difficult to follow young to fledging age;
however, no young were raised in the 1996/97, 1997/98, 2002/03
and 2004/05 breeding seasons, which coincided with ENSO events.
The Lesser and Brown Noddy both showed significant differences
in mean reproductive output between bad years and good years.
Lesser Noddies had a mean breeding success rate of 9.6% in poor
years, compared with 40.2% in good years (t11 = 4.36, P = 0.001);
Brown Noddies had a mean rate of <1% in poor years compared
with 18.0% in good years (t10 = 3.18, P = 0.009, Table 2).
Similarly, Wedge-tailed Shearwaters had a significantly lower
mean breeding success rate of 9.4% in “bad” years, compared with
58.8% in “good” years (t7 = 6.63, P < 0.001, Table 2).
Diet and reproductive performance
Dietary analysis of 3500 regurgitations (48 000 items) revealed that
Lesser Noddies consumed 37 species of prey, but with three main
ichthyoplankton species being in the majority, namely, Beaked
Salmon Gonorynchus greyi, Hawaiian Bellowfish Macroramphosus
scolopax and Black-spotted Goatfish Parupeneus signatus (Table 4;
for details, see Surman & Wooller 2003). Brown Noddies were found
to consume slightly fewer species (n = 32) and relied predominately
on Beaked Salmon and cephalopods. Similarly, Sooty Terns
consumed 34 species, predominately cephalopods, larval goatfish,
Beaked Salmon and lanternfishes (Myctophidae). The Wedge-
tailed Shearwater fed on 15 species, mainly cephalopods, adult
Scaly Mackerel Sardinella lemuru and small quantities of larval
Beaked Salmon (Table 4).
TABLE 1
The mean lay date and the date of first eggs for the three tern
species studied on Pelsaert Island, Houtman Abrolhos
Species 1991–1999 2000–2007
Lesser Noddy Anous tenuirostris
First egg 11 Sep 15 Oct
Mean lay date 9 Oct 2 Nov
Brown Noddy A. stolidus
First egg 22 Sep 28 Oct
Mean lay date 12 Oct 11 Nov
Sooty Tern Sterna fuscata
First egg 16 Oct 7 Nov
Mean lay date 24 Oct
Fig. 2. Breeding participation (percentage) for (A) Sooty Terns
Sterna fuscata (solid squares) and Wedge-tailed Shearwaters Puffinus
pacificus (open squares) and (B) Lesser Noddies Anous tenuirostris
(solid diamonds) and Brown Noddies A. stolidus (open squares). The
sea levels at Geraldton (open squares) and Fremantle (closed squares)
and at the bottom the Southern Oscillation Index (SOI) for the same
period are shown. El Niño–Southern Oscillation events are associated
with negative SOI values and lower sea levels.
(A)
0
10
20
30
40
50
60
70
80
90
100
(B)
0
10
20
30
40
50
60
70
80
90
100
-30
-20
-10
0
10
20
30
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Breeding Participation (%)Sea Level (m)SOI
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Gerald to n Sea level
Freman tle
SOI
Surman & Nicholson: ENSO-driven oceanographic variability at the Houtman Abrolhos 133
Marine Ornithology 37: 129–138 (2009)
Low availability of Beaked Salmon later in the breeding season,
measured by volume in the diets of Brown and Lesser Noddies,
appeared to occur during strong ENSO years. Fig. 4 shows the
relationship between the volume of Beaked Salmon in regurgitates
of Lesser and Brown Noddies compared with their breeding success
in the seasons 1991 to 2000. Both noddy species displayed lower
volumes of Beaked Salmon during the severe 1997/98 ENSO event;
however, the Lesser Noddy appeared to be less reliant upon this
prey item, because it had higher breeding success than the Brown
Noddy did over the same period. Despite this trend, the volume of
Beaked Salmon was significantly correlated with breeding success
in the Lesser Noddy (y = 0.54x +26.41, R2 = 0.44, P = 0.03), but
there was not such a clear relationship in the Brown Noddy.
The Brown Noddy population suffered a complete reproductive
failure in those years when the proportion of Beaked Salmon
fell below 50% by volume (Fig. 4). In general, Lesser Noddy
regurgitates were nearly always characterized by Black-spotted
Goatfish and Beaked Salmon, whilst Brown Noddy regurgitates
were dominated by Beaked Salmon and cephalopods. The volume
of Beaked Salmon was lower during most “bad” seasons than in
“good” seasons for both noddy species, although this difference
was significant only in the Lesser Noddy (t7 = 3.83, P = 0.003).
Conversely, the volume of Black-spotted Goatfish was significantly
higher in “bad” years in the Lesser Noddy (t5 = 3.11, P = 0.01,
Table 2, Fig. 5).
Seasonal prey availability, ENSO and breeding success for
Lesser and Brown Noddies
Because the diet and foraging grounds of the two noddy species
exhibited substantial overlap (Surman & Wooller 2003), and
because both exhibited some reliance upon the same prey item, ten
years of dietary data were pooled for Brown and Lesser Noddies
on Pelsaert Island. “Good” breeding success for the noddy species
appeared to be linked to the availability of large volumes of Beaked
TABLE 2
The breakdown of timing, performance and diet for those years
with sufficient data for each species studied on Pelsaert Island, Houtman Abrolhos
Species
Seasona
Bad
(n=4)
Good
(n=7)
Lesser Noddy
Participation (%) 39.6 54.7
First egg date (day) 306 255
Mean lay date (day) 318 284
Breeding success (%) 9.6 40.2
Beaked Salmon Gonorynchus greyi (% volume) 34.7 53.9
Black-spotted Goatfish Parupeneus signatus (% volume) 46.5 20.4
Brown Noddy
Participation (%) 22.9 66.9
First egg date (day) 319 266
Mean lay date (day) 344 275
Breeding success (%) 0.9 18.0
Beaked Salmon G. greyi (% volume) 48.0 73.2
Black-spotted Goatfish Parupeneus signatus (% volume) 2.9 0.9
Sooty Tern Sterna fuscata
Participation (%) 0.3 67.5
First egg date (day) 318 280
Mean lay date (day) 330 291
Wedge-tailed Shearwater Puffinus pacificus
Participation (%) 21.6 46.9
First egg date (day) 322 322
Breeding success (%) 9.4 58.8
a Years were pooled into reproductively bad or good years to illustrate the other breeding characteristics for each species in those years.
We assigned seasons as “bad” when breeding success was less than 15%.
134 Surman & Nicholson: ENSO-driven oceanographic variability at the Houtman Abrolhos
Marine Ornithology 37: 129–138 (2009)
Salmon during the early (September/October) portion of the
breeding season (Fig. 5). In reproductively good seasons, the volume
of Beaked Salmon collected from regurgitates in the early portion
of the breeding season was very high (70%), dropping off later to
about 50%. During bad seasons, the volume of Beaked Salmon was
very low (18%) early in the breeding season, rising gradually to
40% in the latter months (November–February, Fig. 5).
The proportion of Black-spotted Goatfish was higher in poorer
years for both noddy species (Table 2, Fig. 5), with the volume in
regurgitates increasing as the season progressed in both “good” and
“bad” years. The occurrence of other major prey species remained
TABLE 3
The breakdown of oceanographic parameters
in bad and good seasons
Measure Season
Bad Good
Total annual Southern
Oscillation Index
–48.3 45.6
Sea level Geraldton (m) 0.55 0.58
Sea level Fremantle (m) 0.75 0.76
SST @ 114.5E (°C) 22.2 22.4
a Years were pooled into reproductively bad or good years to
illustrate the other breeding characteristics for each species in
those years.
TABLE 4
A summary of the main prey items identified in regurgitates from the Lesser Noddy Anous tenuirostris, Brown Noddy A. stolidus,
Sooty Tern Sterna fuscata and Wedge-tailed Shearwater Puffinus pacificus from Pelsaert Island, Houtman Abrolhos between 1991
and 2001 (after Surman and Wooller 2003)
Prey Lesser
Noddy
Brown
Noddy
Sooty
Tern
Wedge-tailed
Shearwater
Beaked Salmon Gonorynchus greyi (% by vol.) 45.5 68.8 6.4 3.1
Black-spotted Goatfish Parupeneus signatus (% by vol.) 27.4 1.2 7.3 0.5
Hawaiian Bellowfish Macroramphosus scolopax (% by vol.) 7.7 7.6 2.8 0.1
Lanternfish (Myctophidae) (% by vol.) 1.4 2.5 5.4 2.9
Scaly Mackerel Sardinella lemuru (% by vol.) 0.0 0.0 0.0 11.1
Squid (% by vol.) 2.4 11.5 65.4 59.8
Total items (n) 1303 768 449 ?
proportionately the same in both “good” and “bad” years, which
would suggest that their delivery to the Houtman Abrolhos region
was independent of ENSO events.
DISCUSSION
Our data suggest that the reproductive performance of the four
tropical pelagic seabird species studied was affected by large-
scale (ENSO) and regional variation in the Leeuwin Current
system. During the 17 years of this study, ENSO-driven changes
in oceanographic conditions (as measured by SOI and sea-level
changes) were correlated with lowered breeding participation and
success for all species, and delayed breeding phenology in the
three tern species. For the Lesser and Brown Noddy, initiation of
breeding was significantly later in years 2000–2007 as compared
with the earlier period of this study (1991–1999). In the later years
of this study, as the frequency of ENSO events increased, poorer
reproductive performance began to occur for all four species in
non-ENSO years also. The Wedge-tailed Shearwater’s timing
of breeding remained unaffected, which reflects its life-history
strategy of a long breeding schedule and dispersal during the non-
breeding season, precluding delayed breeding as a strategy when
faced with a limited window of opportunity (Nicholson 2002)
Fig. 3. Breeding success (percentage) in the (A) Wedge-tailed
Shearwater Puffinus pacificus and the (B) Lesser Noddy Anous
tenuirostris (solid diamonds) and Brown Noddy (open squares). A
robust measure was not obtained for the Sooty Tern Sterna fuscata.
The sea levels at Geraldton (open squares) and Fremantle (closed
squares) and at the bottom the Southern Oscillation Index (SOI) for
the same period are shown. El Niño–Southern Oscillation events are
associated with negative SOI values and lower sea levels.
-30
-20
-10
0
10
20
30
0.4
0.5
0.6
0.7
0.8
0.9
1.0
(A)
0
10
20
30
40
50
60
70
80
(B)
0
10
20
30
40
50
60
70
Breeding Success (%)Sea Level (m)SOI
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Sea level
Gerald to n
Freman tle
SOI
Surman & Nicholson: ENSO-driven oceanographic variability at the Houtman Abrolhos 135
Marine Ornithology 37: 129–138 (2009)
ENSO events also had a strong influence on the reproduction of
the Brown Noddy breeding at the Seychelles, Western Indian
Ocean, with the years of poorest breeding success occurring in
El Niño and La Niña years (Ramos et al. 2006). Roseate Terns
Sterna dougallii at the same location experienced delayed onset of
breeding and decreased hatching success as a result of ENSO events
(Ramos et al. 2002) and were affected by temporal variability in
oceanographic conditions (Monticelli et al. 2007).
It has been suggested that seabird breeding phenology is adjusted
to the timing of emergence of a species’ main prey, which is
anticipated through localized changes in SST during the pre-
breeding phase (Frederiksen et al. 2004, Monticelli et al. 2007).
There was a relationship of the mean SST between January and
March and the mean lay date for Brown and Lesser Noddies in
the following season at the Houtman Abrolhos during the earlier
years of this study. In addition, the breeding participation for
Lesser Noddies was also significantly related to the SST and to
Fremantle sea levels in the same years. The Lesser Noddy is the
only resident species in this study, and so it would be expected
that Lesser Noddies may rely on more local environmental cues
than do the other migratory species. In a study of seabird breeding
phenology in the North Sea, it was found that widely dispersing
species showed responses to large-scale environmental cues,
whereas a resident species responded to more local oceanographic
cues, including SST (Frederiksen et al. 2004). Another study found
that individual environmental cues in the western Indian Ocean
would mediate different responses in Roseate Tern reproduction,
with large-scale climatic fluctuations (such as ENSO) affecting
timing of breeding, and local SST affecting breeding participation
(Ramos et al. 2002).
It would seem that residential knowledge of the pre-breeding
environment at the Houtman Abrolhos conferred the advantage of
higher breeding success to Lesser Noddies, even in years affected
by ENSO events. Variation in the strength of flow of the Leeuwin
Current appeared to influence the timing of delivery of Beaked
Salmon larvae to within the seabird foraging ranges at the Houtman
Abrolhos. The reduced volume of Beaked Salmon in regurgitates of
the two noddy species during ENSO years suggested that there was
a reduction in the availability of this prey.
The Lesser Noddy was far less reliant upon larval Beaked Salmon
than was the migratory Brown Noddy, which consumed the highest
proportion of this prey of all the species studied. The Lesser Noddy
was buffered by its ability to take advantage of the smaller larval
Black-spotted Goatfish later in the season; the Brown Noddy required
a much higher proportion of Beaked Salmon throughout the season
and, on average, faired worst in ENSO years when this prey was
less available. The Brown Noddy population suffered a complete
reproductive failure in those years when the proportion of Beaked
Salmon fell below 50% by volume, which coincided with low SOI
values (ENSO years). Surman and Wooller (2003) found that whilst
this prey occurred in a similar proportion of samples from both
noddy species, Brown Noddy regurgitates contained significantly
more Beaked Salmon in reproductively “good” years, both in terms
of number and volume consumed. The Brown Noddy was the
species most affected by the lower availability of Beaked Salmon
during ENSO years, and it may be that Brown Noddies are less
able to switch prey. Given that the two noddy species have overlap
in their foraging range and prey composition, it is possible that the
cost–benefit of capturing and feeding on the smaller goatfish was not
advantageous to the larger Brown Noddy (Gaughan et al. 2002).
Importantly, Beaked Salmon appeared to be a requirement at high
volumes in the diet of Brown Noddies during the pre-laying period
of the season in September/October. Overall, Beaked Salmon
comprised 73% of the diet in good years, but 48% in bad years,
although the poorer years were skewed by the appearance of
Beaked Salmon in the regurgitates of Brown Noddies later in the
year than would normally be expected. Similarly, Crawford and
Dyer (1995) observed changes in the numbers of breeding attempts,
and the numbers of chicks that were raised, of four seabird species
in relation to changes in the abundance of the Cape Anchovy
Engraulis capensis in their diet. In addition, Ramos (2000)
observed a reduction in the volume of the principal prey, goatfish,
in the diet of Roseate Terns breeding in the western Indian Ocean
during reproductively poor years. Interestingly, at this eastern
Indian Ocean breeding site, there were higher volumes of goatfish
in regurgitates of both the Lesser Noddy and Brown Noddy in
reproductively poor years, perhaps supplementing the decline in the
presence of Beaked Salmon.
Although both Sooty Terns and Wedge-tailed Shearwaters fed
predominately on adult squid, a significant portion of their dietary
intake also included Beaked Salmon and Black-spotted Goatfish.
The Sooty Tern and Wedge-tailed Shearwater also experienced
breeding failures in ENSO years, even though Beaked Salmon was
not their main prey. The oceanographic factors that may have led to
a change in the distribution or abundance of Beaked Salmon near
Fig. 4. The volume of Beaked Salmon Gonorynchus greyi found
in regurgitates (open squares, left axis) and the breeding success
(closed diamonds, right axis) of the Lesser Noddy and Brown
Noddy over nine years that dietary data was collected at Pelsaert
Island, Houtman Abrolhos.
1991 1993 1994 1995 1996 1997 1998 1999 2000
1991 1993 1994 1995 1996
Year
Year
1997 1998 1999 2000
Lesser Noddy
Brown Noddy
Volume of Beaked SalmonVolume of Beaked Salmon
Breeding Success (%)Breeding Success (%)
100
90
80
70
60
50
40
30
20
10
0
70
60
50
40
30
20
10
0
136 Surman & Nicholson: ENSO-driven oceanographic variability at the Houtman Abrolhos
Marine Ornithology 37: 129–138 (2009)
to the Houtman Abrolhos in these years could represent a general
reduction in marine productivity across the region.
Beaked Salmon is a southwestern benthic species (Hutchins &
Swainston 1986), and the delivery of its larvae appeared to be higher
when the Leeuwin Current was stronger, during non-ENSO years.
This cause and effect is counterintuitive, because the larvae were then
being delivered northwards against a stronger-flowing southwards
current. However, mixing of the Leeuwin Current and the Capes
Current at the Geelvink Channel may have facilitated the delivery of
Beaked Salmon larvae within range of the two species south and west
of the Abrolhos, where previously-identified foraging “hotspots”
of both noddy species occurred, at least to the shelf edge (Surman
& Wooller 2003). The two species have seldom been observed
foraging in the Geelvink Channel itself (Surman pers. obs.). The
Capes Current flows most strongly between October and February,
principally along inshore shelf waters of Western Australia, at a time
that the Leeuwin Current flows most weakly and during the breeding
season of the four study species. The long (3 months) larval duration
of Beaked Salmon may facilitate its northward delivery via the Capes
Current. During ENSO years, there is less potential for upwelling in
the area, which may reduce the distribution of Beaked Salmon larvae.
The orientation of eddies regularly observed west of the Houtman
Abrolhos may also play a role in defining the distribution of Beaked
Salmon larvae during ENSO years.
Whilst the Leeuwin Current has likely facilitated the seabird
colonisation of the Houtman Abrolhos by tropical seabirds, the
dynamics of how the current delivers prey or suppresses or enhances
the arrival of prey from southern regions remains unclear and, from
the data presented here, appears to be changing. Oceanographic
productivity data is still inadequate in the region off south-western
Australia, but the observations reported here suggest that there is
zonal shift in productivity on the shelf edge during ENSO conditions.
A similar phenomenon has been reported on Australia’s eastern
boundary (Prince 2001, Ramos et al. 2002, Ramos et al. 2006,
Monticelli et al. 2007). Further studies of seabird productivity
at the Houtman Abrolhos could benefit from assessments of the
influence of other environmental indices upon breeding performance.
Measurements of the Indian Ocean Dipole, which affects SST, winds
Fig. 5. Pooled seasonal dietary data (open squares = reproductively successful years; closed squares = reproductively poor years) for Lesser
Anous tenuirostris and Brown Noddies A. stolidus at Pelsaert Island, Houtman Abrolhos, for the four key prey items consumed by these
species: (A) squids, (B) Hawaiian Bellowfish Macroramphosus scolopax, (C) Black-spotted Goatfish Parupeneus signatus and (D) Beaked
Salmon Gonorynchus greyi. Dietary data was pooled into early season (September–October, n = 782), mid-season (November–December,
n = 1226), and late season (January–February, n = 394).
Squids
40
35
30
25
20
15
10
5
0
35
30
25
20
15
10
5
0
12
10
8
6
4
2
0
70
60
50
40
30
20
10
0
Hawaiian Bellowfish
Black-spotted Goatfish Beaked Salmon
Volume (%)
Volume (%)
Volume (%)
Volume (%)
Time of Year
Sep/Oct Nov/Dec Jan/Feb
Time of Year
Sep/Oct Nov/Dec Jan/Feb
Time of Year
Sep/Oct Nov/Dec Jan/Feb
Time of Year
Sep/Oct Nov/Dec Jan/Feb
Surman & Nicholson: ENSO-driven oceanographic variability at the Houtman Abrolhos 137
Marine Ornithology 37: 129–138 (2009)
and precipitation in the Indian Ocean, and estimates of chlorophyll-a
concentration (utilizing Sea-viewing Wide Field-of-view Sensor)
to assess changes in primary productivity around the Houtman
Abrolhos (as described and utilized in Monticelli et al. 2007 could
provide further information about the dynamics of prey delivery in
this region of the Indian Ocean.
The long-term distribution of these tern metapopulations may
be driven by the increasing frequency of the ENSO (Allan et al.
1996). The data presented in this study on the timing of breeding
of the Lesser Noddy, Brown Noddy and Sooty Tern illustrate that
the breeding season of these species at the Houtman Abrolhos is
undergoing a gradual seasonal shift. During a study of seabirds
in the North Sea, Frederiksen et al. (2004) have highlighted the
likelihood that global climate change will cause shifts in the timing
of peak food availability, with successful adaptation of animals
requiring that they be able to adjust the time at which they initiate
breeding. With the increasing likelihood of later commencement
of breeding continuing for tern species at the Houtman Abrolhos,
unless a concurrent shift in peak marine productivity occurs,
reproductive output is likely to continue to be poor. Later breeding
often correlates with lower breeding success (Dunlop & Jenkins
1994, Gaughan et al. 2002, Ramos et al. 2002, Ramos et al. 2006,
Monticelli et al. 2007). The implications for the core breeding
populations of seabirds at the Houtman Abrolhos are severe. One
response to poor food availability is to relocate elsewhere. The
establishment of recent seabird colonies southwards along the
Western Australian coastline has been well documented (Dunlop
& Wooller 1990, Dunlop 2009). Little suitable island habitat exists
south of the Houtman Abrolhos to continue to support the southern
range extension in tropical seabirds (Dunlop 2009). Global climate
change is likely to present some severe challenges for seabirds at
the Houtman Abrolhos. Coupled with the effects of increasing sea
levels upon island habitats, seabirds at the Abrolhos must adapt to
what appears to be shifting levels of prey availability as a result
of changes in productivity in the oceanic region adjacent to the
Houtman Abrolhos.
ACKNOWLEDGEMENTS
Many people assisted in the field over the years, but we particularly
thank Ray and Helen Surman, Steven Oswald, Andy Russell, Brent
Wise, Mark Garkaklis, Michael Nikulinsky, Michael Featherstone,
Nick Nicholson, Janine Boreland and Bill Fogg for their
companionship in the field. We also thank our son Shae for putting
up with his parents’ pursuits amongst the mangroves. Transport
was kindly provided by the crews of Phillip King, Le Chelle, Eco
Abrolhos, Force Five, Southern Lady II and ReefSeeker. We are
indebted to Mal and Wendy MaCrae, and the Newtons, Basiles
and Franceschis of Burnett Island. We would like to thank the
Western Australian Department of Environment and Conservation
(DEC) for their permission to conduct research at the Houtman
Abrolhos. Accommodation was kindly provided by the Department
of Fisheries, Barry Humfrey and Murray Davidson. This research
was in part funded by a Murdoch University Scholarship, the
Fisheries Research and Development Corporation and the Natural
Heritage Trust. Since 2006, funding has been allocated through
the DEC managed project, “Back from the Brink,” with funding
provided through the Northern Agricultural Catchments Council
from State and Commonwealth funds. We also thank DEC’s
Geraldton office for supporting the continuation of this research,
particularly Anthony Desmond for his encouragement. We also
thank J.N. Dunlop for his support over the years and his help as a
fellow seabird botherer. Flinders University National Tidal Facility
kindly provided sea level data, and Alan Pearce of CSIRO, WA,
kindly provided the Reynolds SST data. An earlier draft of this
manuscript benefited greatly from the constructive comments of
Sèbastian Jaquemet and one other anonymous reviewer.
REFERENCES
AINLEY, D.G., CARTER, H.R., ANDERSON, D.W., BRIGGS,
K.T., COULTER, M.C., CRUZ, F., CRUZ, J.B., VALLE, C.A.,
FEFER, S.I., HATCH, S.A., SCHREIBER, E.A., SCHREIBER,
R.W. & SMITH, N.G. 1986. Effects of the 1982–83 El Niño-
Southern Oscillation on Pacific Ocean bird populations. Acta XIX
Congressus Internationalis Ornithologici 2: 1747–1758.
ALLAN, R.J., LINDESAY, J. & PARKER, D.E. 1996. El Niño
Southern Oscillation and climatic variability. Collingwood,
Australia: CSIRO Publishing. 405 pp.
ASHMOLE, M.J. & ASHMOLE, N.P. 1968. The use of food samples
from seabirds in the study of seasonal variation in the surface
fauna of tropical oceanic areas. Pacific Science 22: 1–10.
CRAWFORD, R.J. & DYER, B.M. 1995. Responses of seabird
species to a fluctuating availability of Cape Anchovy Engraulis
capensis off South Africa. Ibis 137: 329–339.
CRESSWELL, G.R., BOLAND, F.M., PETERSEN, J.L. & WELLS,
G.S. 1989. Continental shelf currents near the Abrolhos Islands,
Western Australia. Australian Journal of Marine and
Freshwater Research 40: 113–128.
CRESSWELL, G.R. 1990. The Leeuwin Current. Corella 14:
113–118.
DIAMOND, A.W. 1983. Feeding overlap in some tropical and
temperate seabird communities. Studies in Avian Biology 8:
24–46.
DUNLOP, J.N. 2009. The population dynamics of tropical seabirds
establishing frontier colonies on islands off south-western
Australia. Marine Ornithology 37: 99–106.
DUNLOP, J.N. & WOOLLER, R.D. 1990. The breeding seabirds
of south-western Australia. Trends in species, populations and
colonies. Corella 34: 107–112.
DUNLOP, J.N. & JENKINS, J. 1994. Population dynamics of the
Bridled Tern Sterna anaethetus colony on Penguin Island,
south-western Australia. Corella 18: 33–36.
DUNLOP, J.N., LONG, P., STEJSKAL, I. & SURMAN, C. 2002.
Inter-annual variations in breeding participation at four Western
Australian colonies of the Wedge-tailed Shearwater Puffinus
pacificus. Marine Ornithology 30: 13–18.
FREDERIKSEN, M., HARRIS, M.P., DAUNT, F., ROTHERY,
P. & WANLESS, S. 2004. Scale-dependent climate signals drive
breeding phenology of three seabird species. Global Change
Biology 10: 214–1221.
FURNESS, R.W. & TASKER, M.L. 2000. Seabird–fishery
interactions: quantifying the sensitivity of seabirds to reductions
in sandeel abundance, and identification of key areas for
sensitive seabirds in the North Sea. Marine Ecology Progress
Series 202: 253–264.
GAUGHAN, D., SURMAN, C.A., MORAN, M. & BURBIDGE,
A. 2002. Feeding ecology of seabirds nesting at the Abrolhos
Islands, Western Australia. Project 1998/203. Perth, Australia:
Fisheries Research and Development Corporation. 140 pp.
GOMMON, M.F., GLOVER, J.C.M. & KUITER, R.H. (Eds). 1994.
The fishes of Australia’s South Coast. Adelaide, Australia: State
Print. 992 pp.
138 Surman & Nicholson: ENSO-driven oceanographic variability at the Houtman Abrolhos
Marine Ornithology 37: 129–138 (2009)
HUTCHINS, B. & SWAINSTON, R. 1986. Sea fishes of southern
Australia: complete field guide for anglers and divers. Perth,
Australia: Swainston Publishing. 180 pp.
JAQUEMET, S., LE CORRE, M. & QUARTLEY, G.D. 2007.
Ocean control of the breeding regime of the Sooty Tern in the
southwest Indian Ocean. Deep Sea Research I 54: 130–142.
KITAYSKY, A.S., WINGFIELD, J.C. & PIATT, J.F. 1999.
Dynamics of food availability, body condition and physiological
stress response in breeding Black-legged Kittiwakes. Functional
Ecology 13: 577–584.
LAST, P.R., SCOTT, C.O.G. & TALBOT, P.H. 1983. Fishes of
Tasmania. Hobart, Australia: Tasmanian Fisheries Authority.
563 pp.
LEIS, J.M. & RENNIS, D.S. 1983. The larvae of Indo-Pacific coral
reef fishes. Sydney, Australia: New South Wales University
Press. 269 pp.
LEIS, J.M. & TRNSKI, T. 1989. The larvae of Indo-Pacific
shorefishes. Sydney, Australia: New South Wales University
Press. 371 pp.
MONAGHAN, P., UTTLEY, J.D. & BURNS, M.D. 1992. Effects
of changes in food availability on reproductive effort in Arctic
Terns. Ardea 80: 71–81.
MONTICELLI, D., RAMOS, J.A. & QUARTLEY, G.D. 2007.
Effects of annual changes in primary productivity and ocean indices
on breeding performance of tropical roseate terns in the western
Indian Ocean. Marine Ecology Progress Series 351: 273–286.
NICHOLSON, L.W. 2002. Breeding strategies and community
structure in an assemblage of tropical seabirds on the Lowendal
Islands, Western Australia [PhD thesis]. Perth, Australia:
Murdoch University. 327 pp.
PATTIARATCHI, C. 2005. Variability in the Leeuwin Current.
Climate Note 10/05 in Indian Ocean Climate I
PEARCE, A.F. 1997. The Leeuwin Current and the Houtman
Abrolhos Islands. In: Wells, F.E. (Ed). The marine flora and
fauna of the Houtman Abrolhos Islands, Western Australia.
Volume 1, Perth, Australia: Western Australian Museum.
pp. 11–46.
PEARCE, A.F. & WALKER, D.I. (Eds). 1991. The Leeuwin
Current: an influence on the coastal climate and marine life of
Western Australia. Journal of the Royal Society of Western
Australia 74: 1–140.
PRINCE, J.D. 2001. Ecosystem of the South East Fishery (Australia),
and fisher lore. Marine and Freshwater Research 52: 431–449.
RAMOS, J.A. 2000. Characteristics of foraging habitats and chick
food provisioning by tropical Roseate Terns. Condor 102:
795–803.
RAMOS, J.A., MAUL, A.M., AYRTON, V., BULLOCK, I.,
HUNTER, J., BOWLER, J., CASTLE, G., MILETO, R.
& PACHECO, C. 2002. Influence of local and large-scale
weather events and timing of breeding on tropical Roseate Tern
reproductive parameters. Marine Ecology Progress Series
243: 271–279.
RAMOS, J.A., MAUL, A.M., BOWLER, J., WOOD, L.,
THREADGOLD, R., JOHNSON, S., BIRCH, D. & WALKER,
S. 2006. Annual variation in laying date and breeding success of
Brown Noddies on Aride Island, Seychelles. Emu 106: 81–86.
RINDORF, A., WANLESS, S. & HARRIS, M.P. 2000. Effects of
changes in sandeel availability on the reproductive output of
seabirds. Marine Ecology Progress Series 202: 241–252.
SCHREIBER, R.W. & SCHREIBER, E.A. 1986. Christmas Island
(Pacific Ocean) seabirds and the El Niño Southern Oscillation
(ENSO): 1984 perspectives. In: Medmaravis and Monbailliu, X.
(Eds). Mediterranean marine avifauana: population studies and
conservation. NATO ASI series. Vol. G12. Berlin, Germany:
Springer Verlag. pp. 397–408.
SERVENTY, D.L., SERVENTY, V.N. & WARHAM, J. 1971. The
Handbook of Australian sea-birds. Sydney, Australia: Reed.
255 pp.
SMITH, M.M. & HEEMSTRA, P.C. 1986. Smith’s sea fishes.
Johannesburg, South Africa: Macmillan South Africa. 1047 pp.
SURMAN, C.A. 1997. A comparative study of the breeding
and feeding ecology of three sympatric tropical terns on the
Houtman Abrolhos, Western Australia [PhD thesis]. Perth,
Australia: Murdoch University. 224 pp.
SURMAN, C.A. 1998. Seabird breeding schedules at the Pelsaert
Group of islands, Houtman Abrolhos, Western Australia between
1993 and 1998. Records of the Western Australian Museum
19: 209–215.
SURMAN, C.A. & WOOLLER, R.D. 1995. The breeding biology
of the Lesser Noddy on Pelsaert Island, Western Australia. Emu
95: 47–53.
SURMAN, C.A & WOOLLER, R.D. 2000. Nestling escape
behaviour in tree, bush and ground-nesting tropical terns. Ibis
142: 320–322.
SURMAN, C.A. & WOOLLER, R.D. 2003. Comparative foraging
ecology of sympatric terns at a sub-tropical, eastern-Indian
Ocean island. Journal of Zoology, London 259: 219–230.
WOOLLER, R.D. & DUNLOP, J.N. 1980. The use of simple
measurements to determine the age of Silver Gull eggs.
Australian Wildlife Research 7: 113–115.
ZAR, J.H. 1996. Biostatistical analysis. Third edition. Upper Saddle
River, New Jersey: Prentice Hall. 927 pp.
... Whether these numbers reflect a long or short-term decline is unclear. A number of factors may be contributing to population decline, including mangrove dieback (Surman and Nicholson 2009a), ENSO-driven reproductive failures (Surman and Nicholson 2009b) and longer term changes in local oceanography (Surman et al. 2012). Surman et al. (2012) found that there were significant oceanographic influences upon the timing of breeding of Lesser Noddies, with later breeding between 2001 and 2010 when compared with the previous ten years (1991)(1992)(1993)(1994)(1995)(1996)(1997)(1998)(1999)(2000). ...
... The strong link between ENSO events and breeding failures in seabirds appears to be decoupling with poor breeding seasons being also observed in non-ENSO years (Surman 1998;Surman and Nicholson 2009b). It would appear that the environmental trigger for the arrival of prey, perhaps as a result of a delayed autumn sea surface temperate (SST) peak (Caputi et al. 2009), has shifted and has led to consistently later breeding. ...
Article
Full-text available
Estimated numbers of breeding pairs of the Lesser Noddy (Houtman Abrolhos) Anous tenuirostris melanops are reported from 14 surveys made between 1986 and 2014. Numbers on the three breeding islands, Pelsaert, Wooded and Morley, have fluctuated between years and within colonies. Total estimated numbers of breeding pairs for the Houtman Abrolhos population have varied from c. 77 000 in 1986 to c. 48 000 in 1993 and 2007; however, in most cases a lower number in one year has been followed by higher numbers in succeeding years. There has been decline in numbers since 1986, with both 2013 and 2014 both having relatively low estimates. Lesser Noddies nest only in Grey Mangrove Avicennia marina low forests and there has been considerable mangrove dieback during the past two decades, with Morley Island being particularly affected. The subspecies clearly meets IUCN Red List criteria for Vulnerable, and if a precautionary approach is taken to evaluation, could meet criteria for Endangered.
... Adams and Brown (1989) suggested that eastern rockhopper penguins are inefficient feeders on fish, but repeated instances of high occurrence of fish in their diet show them to be opportunistic and efficient feeders on available forage resources. The presence of Z. spinifer in their diet in 1996/97 and 1997/98 coincided with the large-scale climatic perturbation caused by the ENSO event of 1997/98 (Surman and Nicholson 2009). Interestingly, in 1997 eastern rockhopper penguins fledged heavier chicks than usual at Marion Island (Crawford et al. 2003b). ...
... Clausen and Pütz (2002) showed an increase of different fish species in the diet of eastern rockhopper penguins in the Falkland Islands during the same period. ENSO-driven oceanographic variability has been linked to changes in seabirds' prey availability, breeding success (Chastel et al. 1993;Surman and Nicholson 2009) and survival (Hodder and Graybill 1985). However, the relationship between fledgling mass, oceanographic conditions and the diet of the two penguin species was not investigated in this study. ...
Article
Populations of the macaroni penguin Eudyptes chrysolophus and the eastern rockhopper penguin E. filholi breeding at Marion Island (Prince Edward Islands) in the sub-Antarctic Indian Ocean decreased from 1994 to 2018. We examined their diets when rearing chicks during this period. There was substantial overlap in the diets of the two species, with crustaceans, dominated by the euphausiids Thysanoessa vicina and Euphausia vallentini, making up >80% of the diets by number and >60% by mass over the study period. The lanternfishes Krefftichthys anderssoni and Protomyctophum tenisoni were the most-commonly consumed fish in the diet of breeding macaroni penguins and dominated in their diet in three of the 25 years (1994, 1998 and 2014); they were also the most commonly consumed fish by eastern rockhopper penguins. Macaroni penguins consumed more of the amphipod crustacean Themisto gaudichaudii and the lanternfish Electrona carlsbergi as well as a greater diversity of and larger-sized fish than did eastern rockhopper penguins. The horsefish Zanclorhynchus spinifer was found in substantial amounts in the diet of eastern rockhopper penguins in 1996 and 1997. Despite annual variations in relative prey contributions to the diets, there were no significant long-term changes in the diet of either penguin species over the study period or when compared with an earlier assessment in 1982. We conclude that changes in the relative proportions of prey in the diets of these penguin species during breeding are unlikely to account for the recent declines in these penguin populations.
... Numerous stressors, including climate change, threaten seabirds. While the impact of warming oceans, hotter overall temperatures, and extreme events has been widely documented for seabirds on their foraging grounds (Piatt et al., 2020;Surman & Nicholson, 2009), not all seabirds are affected similarly, and we often lack data on key aspects coast, the islands of the Recherche Archipelago typically experience longer fire-free periods than similar habitats on the adjacent mainland (e.g., Cape Le Grand and Cape Arid national parks; DEC 2012; Fig. 1). For example, it is estimated that the 1972/73 bushfire on Middle Island (which burnt ~ 50% of the island) was the first fire on the island for 170 years (Brown et al., 1984), and the 2002 fire on Mondrain Island was the first in 58 years (Pearson et al., 2004). ...
Article
Traditional burning regimes have long been employed to enhance biodiversity and mitigate high-intensity wildfires. The link between changes in the distribution, success, and timing of breeding in seabirds and climatic and oceanographic variation in the marine environment has been established, with migratory seabirds less able to respond to climate variability than resident species. While climate-driven changes can also occur on seabird breeding islands, few data are available regarding potential impacts. Here we investigate the frequency and severity of bushfires on seabird breeding islands in Western Australia, regarding the 2020 fire on Figure of Eight Island in the Recherche Archipelago. A lack of quantitative, historical surveys limited our ability to quantify the number of shearwaters lost in this event. However, a review of available data suggests thousands of birds die due to burning every one or two years across the Archipelago. On Figure of Eight, shearwater burrow occupancy and density were low 12 months post-burn (0.25 and 0.02 ± 0.03, respectively), with minimal evidence of recovery (very few burrows detected) 23 months post-burn. We discuss opportunities to develop an adaptive, community-based program for reinstating collaborative, cultural methods of fire management and monitoring regimes on seabird breeding islands in Australia.
... Global climatic cycles such as the El Niño Southern Oscillation (ENSO) have documented effects on seabirds, altering their breeding and foraging success, often dramatically (Schreiber & Schreiber, 1984;Surman & Nicholson, 2009). With global climate change, more intense peaks in climatic cycles such as El Niño and La Niña events are predicted (Timmermann et al., 1999), along with increased storm intensity and frequency (Easterling et al., 2000;Trenberth, 1999;Walsh & Ryan, 2000). ...
Technical Report
Full-text available
Threats to seabird species breeding and visiting the northern Aotearoa New Zealand region.
... Over the past few decades, increasingly substantial declines in both the breeding success and population size of many seabird species have occurred globally (Schreiber & Schreiber 1984, Croxall et al. 1999, Wanless et al. 2005, Jones et al. 2018. Considerable evidence exists that changes in sea surface temperatures (SSTs) are impacting seabird populations (Smithers et al. 2003, Surman & Nicholson 2009a, Cannell et al. 2012, Jones et al. 2018. Observations of widespread starvation and/or reduction in the growth rate of chicks in response to increased SST (Schreiber & Schreiber 1984, Congdon et al. 2007, Velarde & Ezcurra 2018 suggest that climatic variables may have a strong influence on adult dietary intake and therefore chick provisioning rates in seabird populations (Smithers et al. 2003). ...
Article
The causes of intraspecific variation in diet and isotopic niche width can provide important insights into the local food resource requirements for a population. This information is particularly important for highly philopatric colonially nesting species, where local competition for food resources may be high. We investigated the relative influence of environmental, temporal and spatial attributes on intraspecific variation in the diet of colonially nesting brown boobies Sula leucogaster using both regurgitant samples and stable isotope analysis of blood and feathers. Diet analyses revealed that Indian anchovy Stolephorus indicus was the predominant prey species in brown booby diet. Despite the predominance of Indian anchovy, there was significant populationlevel intraspecific variation in diet. Our results supported the intersexual competition hypothesis, with female diet not only exhibiting greater species richness, but non-breeding females likely to feed in a different habitat. The isotopic niche also varied according to life-history stage, with individuals utilising different food sources between the breeding and non-breeding seasons. Additionally, there were significant inter-annual differences in diet composition associated with warmer sea surface temperatures. Furthermore, we identified sub-colony differences in the nonbreeding diet. The different patterns of food intake represent those typical of a central place forager: during the non-breeding season (when adults were not area-restricted due to breeding activity), the width of the isotopic niche of both sexes increased. This study has revealed multiple causes of intraspecific variation in diet and isotopic niche and highlights the need for comprehensive dietary analyses to manage seabird populations effectively within specific locations.
... For example, our findings are similar for shearwaters (Puffinus carneipes) from southwest WA, in that a stronger El Niño resulted in a weaker LC which changed the distribution of prey which, in turn, negatively impacts the seabirds (Bond & Lavers, 2014). Similarly, off central WA, during El Niño events, there were reduced volumes of prey species documented in pelagic seabird regurgitates, suggesting a general reduction in marine productivity, which ultimately affects the reproductive output of the birds (Surman & Nicholson, 2009). Elsewhere, El Niño events indirectly affect delphinid movement patterns by changing the distribution of prey, for example in bottlenose dolphins (T. ...
Article
Full-text available
Large-scale climate modes such as El Niño Southern Oscillation (ENSO) influence population dynamics in many species, including marine top predators. However, few quantitative studies have investigated the influence of large-scale variability on resident marine top predator populations. We examined the effect of climate variability on the abundance and temporary emigration of a resident bottlenose dolphin (Tursiops aduncus) population off Bunbury, Western Australia (WA). This population has been studied intensively over six consecutive years (2007–2013), yielding a robust dataset that captures seasonal variations in both abundance and movement patterns. In WA, ENSO affects the strength of the Leeuwin Current (LC), the dominant oceanographic feature in the region. The strength and variability of the LC affects marine ecosystems and distribution of top predator prey. We investigated the relationship between dolphin abundance and ENSO, Southern Annular Mode, austral season, rainfall, sea surface salinity and sea surface temperature (SST). Linear models indicated that dolphin abundance was significantly affected by ENSO, and that the magnitude of the effect was dependent upon season. Dolphin abundance was lowest during winter 2009, when dolphins had high temporary emigration rates out of the study area. This coincided with the single El Niño event that occurred throughout the study period. Coupled with this event, there was a negative anomaly in SST and an above average rainfall. These conditions may have affected the distribution of dolphin prey, resulting in the temporary emigration of dolphins out of the study area in search of adequate prey. This study demonstrated the local effects of large-scale climatic variations on the short-term response of a resident, coastal delphinid species. With a projected global increase in frequency and intensity of extreme climatic events, resident marine top predators may not only have to contend with increasing coastal anthropogenic activities, but also have to adapt to large-scale climatic changes.
... In addition, their timing of breeding and reproductive performance exhibits high annual variability. This is linked to El Nino-Southern Oscillation influences upon the Leeuwin Current and delivery of their planktivorous fish prey (Surman & Wooller 2003, Surman et al. 2012, Surman & Nicholson 2009a). ...
Article
Full-text available
We present the first tracking data of the foraging behaviour of a small tern, the 100 g Lesser Noddy Anous tenuirostris from the eastern Indian Ocean. Using small geologging devices (hereafter GPS), the 17 individuals tracked foraged 4.8–112 km from Pelsaert Island, Houtman Abrolhos, Western Australia. All tracking devices were recovered, and all the tracked individuals continued to breed normally. The mean trip distance was 79.5 km (standard error [SE] 9.8 km), with a mean trip length of 5 h 39 min (SE 39 min), at a mean travel speed of 12.6 km/h (SE 0.6 km/h). In the summer, breeding Lesser Noddies foraged diurnally from 04h00 to 20h40, returning to their colony at night. Individuals tracked in November spent significantly more time foraging and commuted further afield than those tracked in December. Lesser Noddies foraged in the west-southwest sector from the main colonies on Pelsaert Island. The Lesser Noddy at Pelsaert Island is the lightest seabird (104 g) tracked to date using GPS devices. Monitoring of noddies before and after tracking, in conjunction with rapid device attachment, deployment and recovery, indicated that these lightweight GPSs provided a successful tracking device for small seabirds.
... Based on data from the Birds Australia atlas projects, these shifts are already evident in some Australian species, such as Scaly-breasted Lorikeet Trichoglossus chlorolepidotus, Pacific Baza Aviceda subcristata and White-headed Pigeon Columba leucomela which have experienced reduced reporting rates at their northern range boundaries and increased reporting rates at their southern boundaries over time (Silcocks and Sanderson 2007). Tropical seabirds in Western Australia have also been observed to be expanding their range southwards (Dunlop 2009; Figure 1) and altering their breeding timing (Surman and Nicholson 2009). ...
Article
Increasingly, space use by foraging seabirds is being used as an indicator of ocean condition to inform projected planning for climate change, fisheries management and marine protected areas. We tracked foraging common noddies (Anous stolidus) from a colony in the East Indian Ocean using back-mounted solar GPS trackers during incubation and chick rearing to evaluate their suitability as biomonitors of ocean condition, and the overlap of flight tracks with marine protected area boundaries. This is the first study to track this species in its eastern distribution and across different stages of the breeding cycle. Six birds were tracked for 89 days in total, describing 10 089 km of flight. Birds made significantly longer trips during chick rearing, which may reflect reduced availability of prey. The tracking period coincided with a particularly strong ENSO event, which may have impacted foraging behaviour, but the foraging area was found to be at least 10 000 km². Foraging was associated with the end points of outward trips that were generally at the edge of the continental shelf, or within proximity of canyon-like bathymetric features or current structures on the shelf. Birds foraged over the shelf during incubation, suggesting a greater reliance on food web structures associated with Leeuwin Current structures. Home ranges and movement tracks showed limited overlap with proposed marine park boundaries, but are promising as indicators of ocean productivity, suggesting that their role in the design of marine reserve networks in the future should be maximised.
Article
We isolate and characterise 16 polymorphic microsatellite loci for the super-abundant, pan-tropical sooty tern (Onychoprion fuscatus), facilitating population genetic studies. In 70 samples from two breeding colonies, the total number of alleles per locus ranged between 5 and 21, observed heterozygosity ranged from 0.143 to 0.942, while estimated null allele frequency varied from −0.131 to 0.273. Polymerase chain reaction (PCR) conditions were optimised across loci, enabling multiplexing and rapid multilocus genotyping. These 16 loci will be useful for future studies of genetic diversity and population structure, and can be used as a proxy through which to assess ecosystem function and change. We additionally test cross-species amplification in the brown (Anous stolidus) and lesser (A. tenuirostris) noddies, illustrating a use of these microsatellites in other related Sternidae species.
Article
Full-text available
We analysed the effects of local- (sea-surface temperature [SST] and windspeed) and large- (multivariate El Nino index) scale weather conditions and timing of breeding on reproductive parameters of tropical roseate terns Sterna dougallii on Aride Island, Seychelles, using up to 17 years of data. The size of the breeding population and initiation of breeding were negatively and positively correlated, respectively, with both SST and the multivariate El Nino index for the laying season (May-June). It is the first time that an El Nino index obtained for the Pacific Ocean is shown to be correlated with reproductive parameters of seabirds in the Indian Ocean. Hatching success decreased significantly with later initiation of breeding. Virtually no chicks fledged when breeding started in June (40 % of the years monitored). We suggest that oceanographic conditions over a relatively large scale have an influence on tern arrival date to the breeding grounds and that SST abound the breeding colony influences the number of birds that attempt to breed. Despite the influence of factors such as predatory fish on food availability, this influence appears to be overridden by the importance of weather events and oceanographic conditions, which are likely to determine marine productivity. This study suggests that ecosystem-level phenomena appear to be important in shaping the population dynamics of tropical roseate terns.
Article
Full-text available
Sandeels Ammodytes marinus are important food for many breeding seabirds in the North Sea, and are harvested in large quantities by an industrial fishery. There is very little evidence of the fishery reducing availability of sandeels to breeding seabirds, but there is concern that fishery managers should take account of the needs of breeding seabirds. Here we present a quantitative index of the sensitivity of different seabird species' breeding success to reduced abundance of sandeels. The index is based on seabird size, cost of foraging, potential foraging range, ability to dive, amount of 'spare' time in the daily budget, and ability to switch diet. Testing the index with empirical data from Shetland during periods of reduced sandeel abundance shows a close correlation between seabird breeding performance and predictions from the index. Mapping the distributions around the North Sea of seabirds with breeding success highly sensitive to sandeel abundance shows that the majority of sensitive seabirds breed in Shetland and Orkney. industrial fishing in those regions should be closely controlled to avoid depleting the local sandeel stocks on which seabirds depend. This analysis considers only impacts on seabird breeding. There is a need for analysis of possible influences on other aspects of seabird demography.
Article
I studied tropical Roseate Terns (Sterna dougallii) on Aride Island, Seychelles, between 1997–1999. Productivity in 1998 was 0.58 fledglings/breeding pair, and in 1999 no young fledged. Roseate Terns on Aride concentrated their foraging along the coastline exposed to prevailing winds, with flock size over this area being significantly correlated with amount of food offered to chicks. In 1998, Lesser Noddies (Anous tenuirostris) were present in 91% of the Roseate Tern flocks, but in 1999 occurred in only 32%. During the 1998 successful breeding season, Roseate Terns were associated with dense flocks of Lesser Noddies over predatory fish, whereas during the 1999 failure season most Roseate Tern flocks were either monospecific or mixed with Fairy Terns (Gygis alba), and without predatory fish. The mean flock size of Roseate Terns (82 vs. 6 birds) and the rate of foraging attempts (8.3 vs. 2.8 attempts min−1) were significantly greater in association with predatory fish. Mullidae (Parupeneus or Mulloidichthys) were the primary prey taken by Roseate Terns, and alternative sources of food were apparently scarce. The high daily variations in the amount of food brought to chicks, intermediate periods of low food delivery, and an apparent seasonal decline in the amount of food brought to the colony suggest that food is unpredictable on a daily and seasonal basis. Absence of predatory fish may explain complete breeding failures and periods of low food delivery, but the importance of other factors is unknown. Information on the ecology and movement patterns of predatory fish around Aride Island is needed to assist the conservation of the Roseate Terns.
Book
The book begins with an overview set within an historical context followed by a synthesis of developments in theories and dynamics that have generated the present modelling approaches to forecasting the phenomenon. The second half of the book provides composites illustrating the near-global physical impacts of ENSO phases, and brief descriptions of findings from the growing number of studies examining biological, chemical and ecological impacts. The book is completed by a time sequence showing simultaneous global MSLP and SST anomalies from 1871-1994. The CD-ROM provides monthly MSLP and SST anomalies from 1871-1994.
Article
Significant changes have been observed since 1900 in the distribution and abundance of populations of at least eight tropical seabird species off south-western Australia, south of the Houtman Abrolhos Islands. The observed changes have involved a southward shift in breeding distribution or the rapid growth of colonies located on or beyond previous limits. The rate of change appears to have accelerated over the last three decades. A study that encompassed the entire establishment period of one "frontier colony" was used as a framework for a demographic transition model extrapolated to other frontier colonies. A wide range of historical observations, together with more detailed recent studies, are brought together to propose a general hypothesis to explain the observed changes in tropical seabird population dynamics in the region.
Article
The Lesser Noddy Anous tenuirostris (melanops) is a small, dark-plumaged, tropical tern which breeds only on three islands in the Houtman Abrolhos, off Western Australia, where it nests in mangrove trees. On Pelsaert Island, in 1991, the single egg clutches were laid over a 102-day period from late August to early December, peaking in September. The incubation period averaged 34 days and the fledging period 40 days. Overall, 60% of eggs hatched and 47% gave rise to free-flying young. Storms caused extensive egg loss from the more exposed nests; some pairs laid replacement eggs. Birds laying earlier in the season tended to use more sheltered sites and, consequently, had a higher reproductive success than those laying later and in more exposed areas.