bibliography 2

ORIGINAL PAPER

Aleks Terauds Æ Rosemary Gales

Provisioning strategies and growth patterns of Light-mantled Sooty Albatrosses Phoebetria palpebrata on Macquarie Island

Received: 12 January 2006 / Revised: 9 March 2006 / Accepted: 10 March 2006 / Published online: 20 April 2006 � Springer-Verlag 2006

Abstract Provisioning regimes and growth of Light- mantled Sooty Albatrosses (LMSA) (Phoebetria palpe- brata) were investigated on subantarctic Macquarie Is- land using an automatic tracking system and automatic weighing nests. The nests were deployed under five chicks in the post-brood provisioning period in 2000 and 2001. Adults typically utilised a cyclical foraging strategy con- sisting of a long foraging trip followed by three to four shorter trips. Chicks received an average (±SE) of 37.5±2.3 kg of food in the post-brood provisioning per- iod and were fed every 1.6±0.1 days with a mean meal size of 520±10 g. Chicks grew at a rate of 61.3±1.0 g day-1 to a peak mass of 4.4±0.1 kg. Mean chick fledging mass was 3.0±0.1 kg. LMSA on Mac- quarie Island fed their chicksmore frequently and showed a lower mean trip duration than conspecifics at South Georgia, which is likely related to proximity of productive Antarctic shelf waters, differences in prey availability and competition with other Procellariiformes.

Introduction

Although the growth of Procellariiforme chicks has been studied over several decades (Ricklefs 1968a; Ricklefs 1973; Ricketts and Prince 1981; Ricklefs 1983; Ricketts and Prince 1984), comprehensive analyses of provision- ing regimes were not possible until the late 1980s and 1990s with the application of sophisticated automatic

tracking systems for parents and in some studies, auto- matic weighing platforms with artificial nest bowls for chicks (Prince and Walton 1984; Reid et al. 1999; Huin et al. 2000; Hedd et al. 2002). The use of remote tracking equipment coupled with the automatic weighing nests (or frequent manual weighing) allows the contribution of each parent to be quantified, and several studies have shown this aspect to be important in comprehensively describing the provisioning regime of a Procellariformes (Chaurand and Weimerskirch 1994b; Weimerskirch et al. 1994a; Hedd et al. 2002).

Of particular importance is the regulation of parental foraging trips during chick rearing. Many Procellarii- formes utilise a combination of long trips and short trips during this period of the breeding cycle. Some species alternate long and short foraging trips (e.g. Blue Petrels Halobaena caerulea and Thin-billed Prions P. belcheri) while others have been shown to make repeated cycles of several short trips followed by a long trip (e.g. Wan- dering Albatrosses Diomedea exulans, Indian yellow- nosed Albatrosses Thalassarche carteri, Campbell Albatrosses T. impavida, Antarctic Prions Pachyptila desolata and White chinned Petrels Procellaria ae- quinoctialis) (Chaurand and Weimerskirch 1994a, b; Weimerskirch et al. 1994a, 1995, 1999; Catard et al. 2000; Cherel et al. 2002). Rather than using an alter- nating strategy, some species exhibit primarily short neritic (e.g. Black-browed Albatrosses T. melanophrys, Shy Albatrosses T. cauta) or primarily long oceanic (e.g. Grey-headed Albatrosses T. chrysostoma) foraging re- gimes (Huin et al. 2000; Waugh et al. 2000; Hedd et al. 2002; Terauds et al. 2006).

Several studies have shown that the decision to undertake a long or short foraging trip appears related to the compromise between maintaining adult body condition and fulfilling the nutritional requirements of the chick. Drent and Daan (1980) described this conundrum in detail and suggested that there was a threshold body condition that adults would not go be- low in order to sustain a chick, and this has been dem- onstrated by manipulative experiments on several

A. Terauds University of Tasmania, Churchill Avenue, Sandy Bay, TAS 7005, Australia

R. Gales (&) Æ A. Terauds Department of Primary Industries, Water and Environment, Biodiversity Conservation Branch, PO Box 44, Hobart, TAS 7000, Australia E-mail: [email protected]

Polar Biol (2006) 29: 917–926 DOI 10.1007/s00300-006-0133-6

Procellariiform species (Chaurand and Weimerskirch 1994a; Weimerskirch et al. 2000). For Procellariiformes, frequent short trips are considered to increase the energy flow to the chick at the expense of the body condition of the parent, while longer trips allow the parent to increase their own mass and condition (Weimerskirch et al. 1994a). During chick brooding, Light-mantled Sooty Albatrosses (LMSA) from Iles Crozet exhibited a bi- modal pattern typical of long and short trips (We- imerskirch 1998), but the foraging strategies of this species after the brood-guard are not well understood.

On Macquarie Island, LMSA have an estimated annual breeding population of 1,000–1,150 pairs (Gales 1998) and are biennial breeders, typically fledging a chick every 2–3 years when successful (Weimerskirch et al. 1986). The aims of the current study were to examine the parental provisioning re- gimes of LMSA on Macquarie Island during the post- brood chick rearing period and describe them in rela- tion to the foraging strategies. The foraging regime and chick provisioning of LMSA from South Georgia has been recently described (Phillips et al. 2005) and earlier studies at breeding sites in the Indian Ocean have described chick growth based on frequent manual weighing (Berruti 1979; Thomas et al. 1983; We- imerskirch et al. 1986). This study complements these earlier data and provides a sound basis for comparison with strategies utilised at other breeding sites, partic- ularly South Georgia.

Methods

Species and study site

Macquarie Island (54�30¢S, 158�55¢E) is a small isolated subantarctic island situated approximately 1,500 km south east of Tasmania, just north of the Antarctic convergence. Typically, LMSA eggs are laid in October and November, hatching in December and January with most chicks fledging in May and June. Following hatching, parents brood the chicks with short alternating shifts for approximately three weeks and chicks usually fledge at around 5 months of age (Thomas et al. 1983). The study was conducted over two austral summers in field seasons 1999/2000 (hereafter referred to as 2000) and 2000/2001 (2001).

Parental foraging regime

An automatic tracking system (ATS) consisting of a VHF receiver connected to a datalogger was used to record the attendance of parents of chicks on automatic weighing nests. An omni-directional antenna was used to detect the transmitters from 200 m. Transmitters were attached using cable ties on a leg mounted velcro band during early-mid incubation. The Sirtrack transmitters (Havelock North, New Zealand), each had a unique

frequency between 150 and 151 MHz, and weighed 7.6– 7.7 g (less than 0.5 % body mass).

Ten transmitters were deployed in January 2000 in conjunction with the automatic nests and nine were deployed in January 2001. Five additional breeding pairs (four in 2000 and one in 2001) were monitored to act as controls. The receiver was set up to scan continuously, taking 3-4 min to scan all frequencies. It is unlikely that any returning birds were missed using this sampling re- gime as no adults were observed to spend less than four minutes at the nest, and all feeding events could be attributed to individual birds. The logger recorded the presence or absence of a transmitter every 30 min. The ATS equipment was left in place until after chicks fledged in 2000 but was removed prior to chick fledging in late March 2001 of the second field season. The nests were rebuilt in the latter season and each chick fledged successfully from these nests. Transmitters were re- moved from breeding birds during subsequent breeding attempts.

The ATS data were used to investigate the adult foraging regime (trip duration, and frequency of visits) during the post-brood provisioning period and in com- bination with the automatic nest data, calculate how much food each adult delivered at each visit. Based on the bi-modal distribution of the foraging trips and the position of the trough (see Results, Fig. 1), foraging trips were classified as short (<4 days) or long (‡4 days).

Chick provisioning

Five automatic weighing nests (Francis Scientific Instruments, Cambridge, UK) were installed underneath LMSA chicks between 24 and 28 January in 2000 and 2001, 2–5 days after the chick was left unattended for the first time. The nests consisted of a fibreglass bowl and base that housed the weighing platform and were tested and calibrated (±10 g) with known weights prior to installation. Electronic output from the nests was

Pe rc

en ta

ge o

f tr

ip s

0

5

10

15

20

25

30

35

40

0 1 2 3 4 5 6 7 8 9 10+

Length of trip (days)

Fig. 1 Frequency of foraging trips of Light-mantled Sooty Alba- trosses provisioning chicks on automatic weighing nests in 2000 (black) and 2001 (grey)

918

logged every 10 min and data were downloaded from the data-loggers via a laptop computer at three to 9 day intervals. The nests were removed in late May 2000 after the chicks had fledged in the first field season and in late March in the 2001 field season. All feeding events could be attributed to a specific parent. If the transmitter on one of a pair failed then any subsequent feeding event not associated with the return of the parent with the working transmitter was attributed to the partner. One nest malfunctioned in 2000 and data from this nest are excluded from the provisioning analyses.

Over 16,000 mass records were logged from deploy- ment to fledging for each nest in the first field season. These data were simplified for the purposes of examining chick growth rates. The chick mass at 04:00 on each day of the deployment was used as this mass has been shown to be least affected by recent feeding events (Huin et al. 2000; Hedd et al. 2002, this study). This smoothing procedure also filtered out most of the ‘noise’ that was present in the raw data.

Provisioning rates were examined over two tempo- ral scales; firstly the mean growth rates over 10 day periods were calculated and secondly growth rates were examined in the period up to the attainment of peak mass (2000 and 2001) and also during the period after peak mass to fledging (2000 only). Standard equations from the Richards’ family of curves (Logis- tic, von Bertalanffy, Gompertz) were fitted to the data (Richards 1959; Ricklefs 1968a; Brown and Rothery 1993) as was the growth equation created by (Huin and Prince 2000). The Gompertz growth curve had the best fit to all data analysed (highest r2) and was used to estimate growth parameters. Correlations in the data were examined using Spearman’s rho (rs) or Pearson correlation tests (rp) and all means cited in- clude standard errors.

Results

Parental provisioning regimes

The foraging regime of all pairs was characterised by a cyclic attendance pattern comprised of a long foraging trip (4–20 days, mean=7.07±0.22) followed by several short foraging trips (0.2–4 days, mean=1.36±0.05), followed by another long foraging trip to begin the cycle again. Parents undertook 7–12 cycles over the entire post brood provisioning period and during this time between 14–33 short trips were undertaken. The fre- quency distribution of all trip lengths reflects the pre- dominance of short trips over long trips (Fig. 1). The mean length of long foraging trips for each individual bird over the entire post-brood provisioning period ranged from 6.2 to 9.2 days whilst the mean length of the short foraging trips ranged from 1.1 to 1.8 days. There were no differences in shift lengths between years (t494=1.27, P=0.20).

The cyclic pattern of long and short shifts was also reflected in the amount of food delivered to chicks and correlation analyses showed that the mass of each meal was significantly correlated with the length of time for which the bird had been foraging (n=496, rp=0.371, P<0.001). Birds returning from long trips delivered a significantly larger meal (mean=637±10 g) than birds returning from shorter trips (mean=470±9 g, t494=9.51, P<0.001). As each individual has different statistical weight in the above analysis there is possibly a problem with pseudorepliation. Therefore a second more complex model was fitted where meal size was defined as the dependent variable, foraging trip length as an explanatory variable and the individual identity of each bird was entered as a random effect. This model also provided evidence that changes in trip length are related to significant changes in the meal size (ANOVA, F311=1.68, P<0.001). The proportion of visits made by each parent was strongly correlated with the proportion of the total food provided by that bird (n=9, rp=0.87, P<0.01), indicating that those birds which made more frequent trips did not do so at the expense of overall food mass delivered through the season (and vice versa). Individuals from each pair usually provided similar amounts of food throughout the foraging regime with only two pairs showing a difference of greater than 20% in the amount of food contributed by each parent.

Chick provisioning

Meal mass

Mean meal mass for the entire provisioning period was similar in both years of nest deployment with a mean of 520±10 g (range 510–540 g, Table 1.). There was simi- lar consistency in the mean meals mass during each 10- day period (470–550 g, Fig. 2a); however, there was some evidence that mean meal mass declined as the chick grew older (Spearman’s rho, n=15, rs=-0.55, P=0.03). There was considerable variation in the indi- vidual meal sizes delivered and single meals ranged in mass from 160 to 1,050 g.

Chick fasting interval

The overall mean chick fasting interval ranged from 1.3 to 1.9 days (mean=1.6±0.1 days). The interval be- tween feeding events tended to increase as the chick grew older (Fig. 2b, Spearman’s rho, n=15, rs=0.66, P<0.01). This increase was relatively gradual and not consistent throughout the provisioning period. The intervals between feeding events were variable with chicks occasionally receiving successive meals within hours of each other, or at the other extreme, fasting for up to 10 days when both parents were on long foraging trips. Chicks were fed most frequently between 30 and 60 days of age.

919

Chick provisioning rate

As mean meal mass only declined slightly throughout the provisioning period, and feeding frequency declined, the overall provisioning rate (g day-1) also declined with age (Fig. 2c, Spearman’s rho, n=15, rs=-0.76, P=0.001). A correlation was also observed between the provisioning rates and the concurrent mean fasting intervals (n=15, rs=-0.54, P=0.04). Mean provisioning rates in the first 60 days of nest deployment were similar in both years (paired-t, t4=1.3, P=0.3) and there was no relationship between provisioning rate and the final fledging mass of the chick (n=4, rs=-0.8, P=0.2).

Mass-specific provisioning rate

To assess whether chick mass influenced the provision- ing rate, mass specific provisioning rates were calculated in g kg-0.75/day (Fig. 2d). The mass-specific provisioning rate decreased sharply between the ages of 20 and 80 days (both years), which is also the period of the most rapid chick growth. Although this parameter was strongly correlated with chick age (Spearman’s rho, n=15, rs=-0.92, P<0.001), mass-specific provisioning appeared to plateau at just over 100 g kg-0.75/day post- 80 days of age, and remained at this level until the chick fledged.

Chick growth

Overall

All chicks in these analyses fledged in May and June of 2000. The mean mass of the chicks when the nests were installed was 1.4±0.1 kg (n=10). Chicks were fed be- tween 30 and 42 kg of food during the entire post brood provisioning period (mean=37.5±2.3 kg) for an overall increase in body mass of between 1 and 2 kg (mean=1.6±0.1). All records of birds returning to the

nest were associated with a feeding event. The mean final fledging mass of these chicks was 3.0±0.1 kg and they reached peak mass (4.4±0.1 kg) at 77.5±2.0 days of age. The mass of most chicks began to plateau after 50– 70 days (Fig. 3). Chicks were fed less frequently as they approached fledging (see also previous analyses of age and chick fasting interval).

Temporal variation in growth rates

The mean growth rates for corresponding 10-day periods in 2000 and 2001 were similar (paired t test, t5=-0.4, P=0.8); however, the standard errors indicate there were considerable differences between chicks (Fig. 4). Indi- vidual growth rates were generally positive and high when the chick was between 40 and 60 days of age, fol- lowing which they decreased and often became negative as the mass of the chick began to plateau and gradually decline. These growth rate trends were consistent with the reduction in feeding frequency observed in the latter half of the provisioning period. The overall mean growth rate in all the 10-day periods leading up to the attainment of peak body mass was 47.8±7.3 g day-1 (n=21); post peak mass this rate was �15.09±7.1 g day-1 (n=25) and the mean overall growth rate for all 10-day periods was 13.6±6.9 g day-1 (n=46). There was a strong correlation between the growth rate in each 10-day period and the corresponding provisioning rate (pooled data, n=40, rp=0.5, P=0.001).

Growth curves

Gompertz growth curves were fitted to six data sets and the mean parameters calculated are shown in Table 2. Growth to peak mass datasets were used to facilitate comparison between years and there were no differences found in asymptotic mass (U tests, Z=-0.5, P=0.6),growth constants (U test, Z=1.6, P=0.1) or age at maximum growth (U test, Z=-0.1, P=0.9). Using

Table 1 Provisioning parameters of Light-mantled Sooty Albatross chicks from automatic weighing nests deployed in 2000 (n=4) and 2001(n=5)

Nest Age installed (days)

Age removed (days)a

Duration (days)

No. of visits

Mean fasting interval (days)

Total amount food (kg)

Mean feed (kg)

Overall provisioning rate (g/day)

Mass at fledging (kg)

2000 (Nests in place until fledging) Mean ± SE 23.6±0.9 139.4±1.7 116.3±2.3 72.3±3.8 1.61±0.11 37.5±2.3 0.52±0.01 322.4±19.1 3.0±0.1 2000 (To day 89 to allow comparison with 2001) Mean ± SE 23.6±0.9 – 65.9±0.8 47.6±1.2 1.39±0.06 25.2±1.5 0.52±0.01 382±21.4 2001 (Nests in place until chicks 85–89 days old) Mean ± SE 26.6±0.7 85.7±0.5 59.2±0.3 41.6±2.0 1.43±0.12 21.2±1.0 0.51±0.03 358.5±16.2 Statistical tests Provisioning to peak mass t (df=7) 1.72 0.20 2.20 0.22 0.89 2000 versus 2001 P value 0.14 0.85 0.08 0.83 0.40

aAge at fledging in 2000

920

data from Thomas et al. (1983) the body mass at hatching was assumed to be 180 g and this value was used when fitting all growth curves. Fitting a general Gompertz growth curve to the pooled data sets from all

chicks (entire provisioning period and to peak mass) provided a general model of growth for this species (excluding the mass recession period). The r2 value for this curve was high (0.81) suggesting a good fit to the data and estimated parameters indicated chicks grew at their fastest rate at 31.7±4.3 days of age and reaching an asymptotic mass of 4.69±0.77 kg. An attempt was also made to fit separate curves (e.g. reverse Gom- pertz—Huin and Prince 2000) to the mass recession period; however, none of these curves fitted the data adequately (as indicated by low r2 values between 0.1 and 0.2).

Growth from t10 to t90

Growth was also described using the time it took for chicks to grow from 10% (t10) to 90% (t90) of asymp- totic mass. The only significant difference between years was the average daily mass gain suggesting that there may have been some differences in growth rates between years (Table 3). The mean growth rate in 2000 was higher than that observed in 2001; however, when examined as a percentage of asymptotic mass, growth rates were not significantly different between years (Table 3).

0

800

600

400

200

(a)

(d)

(c)

(b)

1.5

1.0

0.5

0

0

100

200

2.5

2.0

300

0

500

400

0

100

200

300

M ea

n m

ea l m

as s

(g )

M ea

n in

te rv

al b

et w

ee n

fe ed

s (d

ay s)

M

ea n

ch ic

k pr

ov is

io ni

ng r

at e

(g d

ay –

1 )

M ea

n m

as s

sp ec

if ic

p ro

vi si

on in

g ra

te

(g k

g –

0. 75

p er

d ay

)

Age of chick (days)

30 40 50 60 70 80 90 100 110 120 130

Fig. 2 Temporal variation in a mean meal size, b mean interval between feeds, c mean chick provisioning rate and d mean mass specific provisioning rate during the Light-mantled Sooty Albatross provisioning regime in 2000 (black) and 2001 (grey). Error bars represent standard errors

M as

s SE

( kg

)

8

9

9

9

9 9 4 4

4 4

4

Age of chick (days) 20 13030 40 50 60 10070 80 90 110 140120

4

3

2

1

9

Fig. 3 Composite growth curves of Light-mantled Sooty Albatross chicks on Macquarie Island (pooled data from 2000 to 2001—sam- ple sizes shown on points)

M ea

n gr

ow th

r at

e (

g da

y– 1 )

Age of chick (days)

60

120

60

0

20 13030 40 50 60 10070 80 90 110 140120

Fig. 4 Mean growth rates (±SE) of Light-mantled Sooty Alba- tross chicks (pooled data from 2000 to 2001)

921

Impact of the automatic weighing nests on provisioning regimes

The provisioning regimes of breeding birds with chicks on automatic weighing nests and control breeders with transmitters were compared. The feeding frequency (mean number visits per day) of each bird in the first 60 days post brood was calculated and used as the basis of these analyses. There were no significant differences in the feeding frequency of breeding birds with chicks on automatic weighing nests (0.47±0.03 visits per day) and those on natural nests (0.44±0.04) in 2000 (U test, Z=- 1.2, P=0.24). In addition, the control chick (known hatching date) fledged at 138 days of age in 2000, which was very similar to the mean age of chicks on automatic weighing nests (140±1.2 days). Due to transmitter fail- ure only one breeding bird was used as a control in 2001; however, the feeding frequency of this bird was very similar to the mean feeding frequency of the breeders with chicks on automatic weighing nests (0.43 visits day- 1 cf. 0.42±0.1 visits day-1).

Discussion

Parental foraging regimes

During post-brood chick rearing LMSA parents undertook a cyclical foraging strategy of long trips (4– 20 days, mean=7.1±0.2) followed by several shorter trips (0.2–4 days mean=1.4±0.05). Several procellarii- form species utilise this strategy, and in a review of five species, Weimerskirch et al. (1994a) suggested that the short trips allow a parent to provide the chicks with increased amount of food in a short period of time. However, these and other studies found that short trips are energetically expensive for the parent and that long trips are necessary to allow the parents to regain con- dition and continue provisioning the chick successfully (Weimerskirch et al. 1994b; Weimerskirch et al. 1999; Catard et al. 2000). It seems likely that LMSA on Macquarie Island are also using this strategy to main- tain condition whilst still satisfying the nutrient requirements of the chick.

Chick provisioning and growth

Light-mantled Sooty Albatross chicks showed a rela- tively typical procellariiform growth pattern with a rapid exponential increase in mass observed after the brood- guard until approximately 60 days of age, followed by a period of slower growth as chicks approached their peak mass. Even though Gompertz growth curves fitted data from the whole provisioning regime well, when fitted to the attainment of peak mass, a significantly better fit was observed. This probably reflects the inability of a posi- tive Gompertz curve to accurately model the decline in mass during the mass recession period.T

a b le

2 G ro w th

p a ra m et er s o b ta in ed

fr o m

o b se rv ed

d a ta

a n d G o m p er tz

cu rv es

fo r p o st

b ro o d p ro v is io n in g p er io d L ig h t- m a n tl ed

S o o ty

A lb a tr o ss

ch ic k s in

2 0 0 0 a n d 2 0 0 1

M a ss

a t

d ep lo y m en t

(k g )

A g e a t

d ep lo y m en t

(d a y s)

A sy m to p ic

m a ss

(A )a

P ea k

m a ss

(k g )

(M m a x )

A g e a t

p ea k

m a ss

(d a y s)

(t m a x )

G ro w th

ra te

to p ea k m a ss

(g d a y -1 ) (k

p )

G ro w th

co n st a n t

k 1 (· 1 ,0 0 0 )a

A g e

m a x im

u m

g ro w th

(d a y s)

(t i) a

F le d g e

m a ss

b

(M e n d )

A g e

fl ed g e

(d a y s)

O v er a ll

g ro w th

ra te

(g d a y -1 )

(k t)

M ea n

d ec li n in g

g ro w th

ra te

(k d )

G o m p er tz

r2

E n ti re

p o st

b ro o d p ro v is io n in g p er io d 2 0 0 0 (n =

5 )

M ea n ±

S E

1 .4 1 ±

0 .1

2 3 .6 ±

0 .9

3 .6 8 ±

0 .0 2 4 .5 9 ±

0 .1 9 8 0 .3 ±

8 .8

6 0 .4 ±

7 .1

7 8 .3 ±

5 .4

2 8 .3 ±

1 .5

3 .0 ±

0 .1

1 4 0 .8 ±

3 .2

1 3 .8 ±

1 .2

-2 6 .9 ± 5 .8

0 .8 0 ±

0 .0 2

P o o le d d a ta

3 .6 4 ±

0 .0 1

7 7 .9 ±

2 .4

2 7 .8 ±

0 .3

0 .6 9

T o P ea k w ei g h t 2 0 0 0 (n =

5 )

M ea n ±

S E

1 .4 1 ±

0 .1

2 3 .6 ±

0 .9

4 .7 6 ±

0 .5 5 4 .5 9 ±

0 .1 9 8 0 .3 ±

8 .8

6 0 .4 ±

7 .1

4 7 .2 ±

5 .3

3 2 .4 ±

2 .7

3 .6 3 ±

0 .3 2

3 7 .0 ±

3 .8

0 .8 7 ±

0 .0 2

T o p ea k w ei g h t 2 0 0 1 (n =

5 )

M ea n ±

S E

1 .3 9 ±

0 .2 0

2 6 .6 ±

0 .7

4 .6 1 ±

0 .2 8 4 .2 6 ±

0 .1 0 7 7 .3 ±

3 .8

5 9 .5 ±

5 .9

3 7 .0 ±

4 .3

3 1 .7 ±

3 .3

3 .7 6 ±

0 .2 5

4 1 .1 ±

0 .4 .7

0 .8 0 ±

0 .0 4

T o p ea k w ei g h t (2 0 0 0 a n d 2 0 0 1 , n =

1 0 )

M ea n ±

S E

1 .4 0 ±

0 .2 0

2 6 .5 ±

0 .5

4 .6 9 ±

0 .3

4 .3 6 ±

0 .1 2 7 5 .9 ±

0 .9

6 1 .3 ±

1 .0

4 2 .1 ±

1 .1

3 2 .0 ±

0 .8

3 .7 0 ±

0 .2 5

3 9 .0 ±

1 .0

0 .8 4 ±

0 .2 0

P o o le d d a ta

(b o th

y ea rs )

4 .6 9 ±

0 .7 7

3 6 .6 ±

1 .4

3 1 .7 ±

4 .3

0 .8 1

a A s ca lc u la te d b y in d iv id u a l G o m p er tz

eq u a ti o n s

b O r w ei g h t a t en d o f d ep lo y m en t p er io d if n es t re m o v ed

b ef o re

fl ed g in g

922

Mass recession in fledgling birds is a well-known phenomenon (Ricklefs 1968b) but few studies have quantified data from this growth period, particularly for Procellariiformes. Huin and Prince (2000) combined a positive and negative Gompertz curve and showed that it could be used to adequately describe growth throughout the whole provisioning period in Black- browed and Grey-headed Albatrosses from South Georgia. This curve did not fit the growth data from the LMSA chicks (r2=0.1–0.2) and this may be due to the significant fluctuations in mass throughout this period due to the combination of episodic and intense feeding events that characterised this period of the provisioning regime.

Comparison with other species

Generally, larger size in petrels and albatrosses has been associated with a slower growth rate (Pennycuick et al. 1984). The medium sized Shy Albatross, for example, grows more quickly than the large Wandering albatross and more slowly than the smaller species of albatross (Table 4). The different growth rates (both normal and mass-specific) of the similarly sized LMSA, Black-bro- wed and Grey-headed Albatrosses are likely to be re- lated to the provisioning rate and the energy content of the food delivered, as Phillips et al. (2003) showed that there is no fundamental differences in the mass-specific metabolic rates of these three species during linear growth. Age-specific growth rate was also correlated with provisioning rate in the LMSA on Macquarie Is- land similar to that observed with Black-browed and Grey-headed Albatrosses at South Georgia. (Huin et al. 2000).

The foraging strategy utilised during the provisioning regime is clearly an important factor in the amount and frequency of food delivered to each chick. Grey-headed Albatrosses from Campbell and Marion Islands that forage for squid primarily in oceanic waters (Waugh et al. 2000; Nel et al. 2001) fed their chick less frequently than Black-browed and Shy Albatrosses, which tend to forage over more neritic waters during their provisioning regimes (Weimerskirch et al. 1997; Huin et al. 2000;

Waugh et al. 2000; Hedd et al. 2002). Macquarie Island LMSA undertook relatively long trips (ie. up to 20 days in length) as part of their cyclic alternating foraging strategy; however, on average fed chicks only slightly less frequently than Black-browed and Grey-headed Albatrosses breeding on South Georgia (Huin et al. 2000).

The mean meal mass delivered to the chicks of dif- ferent species varies (Table 4) and appears related to the mass of the adults, foraging strategy and feeding fre- quency. LMSA also showed a low chick-provisioning rate relative to the other smaller albatross species and this is mainly attributable to the comparatively low feeding frequency. Excluding the large Wandering and Royal (D. sandfordi) Albatrosses; Shy Albatrosses, the most neritic and frequent feeders, delivered the smallest meals to chicks on average (Hedd et al. 2002), whilst Grey-headed Albatrosses from South Georgia, which regularly forage in pelagic waters, delivered the largest meals to chicks (Huin et al. 2000). LMSA on Macquarie Island, which undertake a combination of both long and short foraging trips , delivered a mean meal size that was intermediate between that of the neritic and oceanic feeders. LMSA from Macquarie Island also delivered larger meals to chicks after longer foraging trips.

Temporal trends in provisioning parameters also varied between species. The mean meal mass delivered to Black-browed and Shy Albatross chicks increased up to the age of around 80–100 days (Huin and Prince 2000; Hedd et al. 2002) while the mean meal size delivered to Grey-headed Albatross chicks increased more rapidly, peaking at around 50–59 days of age and then remaining stable until the chicks were around 100 days old before declining (Huin et al. 2000). In contrast to both these trends, the mean meal size delivered to LMSA on Macquarie Island remained stable for the entire post- brood provisioning period.

Comparison with other sites

Weimerskirch (1998) also described a similar cyclical foraging strategy for LMSA on Iles Crozet and noted that it was probably related to the distribution of dif-

Table 3 Growth of Light-mantled Sooty Albatross chicks expressed as the time taken to grow from 10% (t10) to 90% (t90) of asymptotic mass

Year/statistic Asymptotic mass (kg)a Age (days) Mean daily mass gain

t10 t90 t10–90 g As percentage of asymptotic mass

2000 Mean±SE 4.76±0.55 7.5±0.2 67.9±2.1 60.3±1.9 62.2±0.8 1.4±0.6 2001 Mean±SE 4.61±0.28 8.0±0.3 71.6±2.8 63.7±2.5 51.1±0.7 1.2±0.4 t (df=8) 1.8 �1.1 �1.1 �1.1 9.8 2.2 P value 0.1 0.3 0.3 0.3 <0.001 0.1 Overall mean±SE 4.69±0.3 7.8±0.2 69.8±1.8 62±1.6 56.6±1.9 1.3±0.04

aAs calculated by individual Gompertz equations

923

T a b le

4 C o m p a ri so n o f p ro v is io n in g a n d g ro w th

p a ra m et er s fo r ei g h t sp ec ie s o f a lb a tr o ss

a t se v en

b re ed in g lo ca ti o n s, li st ed

in d es ce n d in g o rd er

o f m a ss

a t fl ed g in g

S p ec ie s a n d

b re ed in g si te

A d u lt

m a ss

(k g )

P ea k

M a ss

(k g )b

F le d g e

p er io d F le d g in g m a ss

(p er ce n ta g e

o f d ec re a se )

F ee d in g

in te rv a l

(d a y s)

c

M ea n

m ea l

m a ss

(g )

P ro v is io n in g

ra te

(g d a y -1 )c

G ro w th

co n st a n t (k )t 1 0 – t 9 0 (d a y s) M ea n d a il y m a ss

g a in

t 1 0 – t 9 0

S o u rc es

(g ) P er ce n ta g e o f a sy m p to ti c m a ss

W a n d er in g A lb a tr o ss

S o u th

G eo rg ia

6 – 1 1 a

1 2 .9

f 2 7 9 f

1 1 .1

(1 4 )f

2 .5

f 9 6 0

2 6 4 d

0 .0 2 6

1 6 4

5 6 .0 0 .5

B er ro w

et a l. (2 0 0 0 ),

T ic k el l (1 9 6 8 )

J. P . C ro x a ll

(u n p u b li sh ed

d a ta )

(c it ed

in T h o m a s et

a l. 1 9 8 3 )

Il es

C ro ze t

1 3 .4

2 6 2

1 0 .0

(2 5 )

2 .8

9 1 0 – 1 ,3 3 0 –

0 .0 2 6

1 4 5

7 4 .5 0 .4

W ei m er sk ir ch

a n d L y s (2 0 0 0 ),

W ei m er sk ir ch

et a l. (1 9 8 6 )

N o rt h er n R o y a l A lb a tr o ss

N ew

Z ea la n d

1 0 .9

2 3 6

8 .8

(1 9 )

– –

– –

1 0 5

8 3 .0 0 .8

R ic h d a le

(1 9 5 2 )

S h y A lb a tr o ss

T a sm

a n ia

3 .4 – 3 .8

a 5 .5

1 2 7

5 .0

(9 )

0 .9

3 7 2

3 9 9

0 .0 3 5

8 0

5 8 .3 1

H ed d et

a l. (2 0 0 2 )g

B la ck -b ro w ed

A lb a tr o ss

S o u th

G eo rg ia

4 .7

e 1 1 6

3 .5

(2 5 )

– –

– 0 .0 4 8

6 4 h

7 0 .0 1 .4

R ic k et ts

a n d P ri n ce

(1 9 8 1 ),

T ic k el l a n d P in d er

(1 9 7 5 )

S o u th

G eo rg ia

3 .5

4 .6

1 1 6

– 1 .2

5 6 9

5 4 2

0 .0 4 2

– –

– H u in

et a l. (2 0 0 0 )g

G re y -h ea d ed

A lb a tr o ss

S o u th

G eo rg ia

4 .7

1 4 1

3 .4

(2 8 )

– –

– 0 .0 4 2

7 3 h

6 3 .0 1 .3

R ic k et ts

a n d P ri n ce

(1 9 8 1 ),

T ic k el l a n d P in d er

(1 9 7 5 )

S o u th

G eo rg ia

3 .5

4 .5

e 1 4 1

– 1 .3

6 1 6

5 3 9

0 .0 3 7

– –

– H u in

et a l. (2 0 0 0 )g

Y el lo w -n o se d A lb a tr o ss

A m st er d a m

Is la n d 2 .5

a 3 .5

1 1 5

– –

– –

– 5 4

5 0 .0 1 .5

W ei m er sk ir ch

et a l. (1 9 8 6 ),

Jo u v en ti n (1 9 8 3 )

L ig h t- m a n tl ed

S o o ty

A lb a tr o ss

M a cq u a ri e Is la n d

4 .4

1 4 1

3 .0

(3 2 )

1 .5

5 1 7

3 4 3

0 .0 3 7

6 0

5 7 .0 1 .3

T h is st u d y

S o u th

G eo rg ia

2 .9

3 .4

1 4 1

2 .6

(2 5 )

2 .0 2

5 4 7

- 0 .0 4 3

6 2

4 5 .0 1 .3

T h o m a s et

a l. (1 9 8 3 ),

P h il li p s et

a l. (2 0 0 5 )g

Il es

C ro ze t

– 1 5 7

– 2 .9

– –

– 6 7

4 2 .0 � 1 .3

W ei m er sk ir ch

et a l. (1 9 8 6 )

S o o ty

A lb a tr o ss

Il es

C ro ze t

2 .4 – 2 .7

a 3

1 6 4

– –

– –

– 6 9

3 7

1 .3 2

W ei m er sk ir ch

et a l. (1 9 8 6 )

a M a rc h a n t a n d H ig g in s (1 9 9 0 )

b M ea n p ea k m a ss

ex ce p t w h er e in d ic a te d

c F ro m

en d o f b ro o d in g to

p ea k m a ss

d F ro m

en d o f b ro o d in g to

fl ed g in g

e F ro m

g ro w th

cu rv e eq u a ti o n s

f C a lc u la te d fr o m

d a ta

in B er ro w

et a l. (2 0 0 0 )

g S tu d ie s th a t o b ta in ed

d a ta

u si n g a u to m a ti c w ei g h in g n es ts

h F ro m

G o m p er tz

p a ra m et er s in

R ic k et ts

a n d P ri n ce

(1 9 8 1 )

924

ferent prey items and energy requirements during dif- ferent stages of the breeding cycle. It is likely that similar factors are influencing attendance patterns on Macqua- rie Island, and driving the cyclical pattern of long and short trips.

In contrast to Macquarie Island, LMSA parents feeding chicks at SouthGeorgia did not appear to utilise a cyclic foraging strategy (Phillips et al. 2005). It is likely that the relatively close proximity of productive Antarctic waters to SouthGeorgia, in conjunction with competition from other albatrosses and petrels, precluded the LMSA from effectively utilisingmore shelf waters closer to South Georgia. The lack of any consistent age related pattern in feeding frequency at South Georgia (Phillips et al. 2005) also suggests that adults from this site are able to main- tain their body condition without needing to undertake a cyclical foraging strategy.

Whilst it has been shown that LMSA from Mac- quarie Island also utilise Antarctic waters during incu- bation (Weimerskirch and Robertson 1994), the predominance of shorter trips during provisioning sug- gests that they are also utilising waters close to Mac- quarie Island during the post brood period. The smaller populations of other seabirds that may compete for similar resources, most notably the absence of White- chinned petrels, and the increased distance to Antarctic shelf waters, are likely to facilitate successful foraging by the LMSA in waters closer to their breeding island, in contrast to those breeding on South Georgia.

Acknowledgements The Australian Department of Environment and Heritage (formerly Environment Australia), the Antarctic Scientific Advisory Committee (ASAC Project 751) and an Aus- tralian Postgraduate Award through the University of Tasmania jointly funded this project. The Australian Antarctic Division and the Department of Primary, Industries, Water and Environment provided logistic support. We would also like to thank J. Hamill and R. Alderman for their assistance in the field on Macquarie Island. A. Richardson, J.-C. Stahl and other reviewer commented on earlier versions of this manuscript and provided useful advice.

Reference

Arata J, JC Xavier (2003) The diet of Black-browed Albatrosses at the Diego Ramirez Islands. Chile Polar Biol 26:638–647

Berrow SD, Humpidge R, Croxall JP (2000) Influence of adult breeding experience on growth and provisioning of Wandering albatross Diomedea exulans chicks at South Georgia. Ibis 142(2):199–207

Berruti A (1979) The breeding biologies of the Sooty Albatrosses Phoebetria fusca and P. palpebrata. Emu 79(4):161–175

Brown D, Rothery P (1993) Models in biology: mathematics, sta- tistics and computing. Wiley, Chichester

Catard A, Weimerskirch H, Cherel Y (2000) Exploitation of distant Antarctic waters and close shelf-break waters by white-chinned petrels rearing chicks. Mar Ecol Prog Ser:249–261

Chaurand T, Weimerskirch H (1994a) Incubation routine, body mass regulation and egg neglect in the Blue petrel Halobaena caerulea. Ibis 136:285–290

Chaurand T, Weimerskirch H (1994b) The regular alternation of short and long foraging trips in the Blue petrel Halobaena caerulea: a previously undescribed strategy of food provisioning in a pelagic seabird. J Anim Ecol 63:275–282

Cherel Y, Bocher P, De Broyer C, Hobson KA (2002) Food and feeding ecology of the sympatric thin-billed Pachyptila belcheri and Antarctic P. desolata prions at Iles Kerguelen, Southern Indian Ocean. Mar Ecol Prog Ser 228:263–281

Drent RH, Daan S (1980) The prudent parent: energetic adjust- ments in avian breeding. Ardea 68:225–252

Gales R (1998) Albatross populations: status and threats In: Robertson G, Gales R (eds) Albatross biology and conserva- tion. Surrey Beatty and Sons, Chipping Norton, pp 20–45

Hedd A, Gales R, Brothers N (2002) Provisioning and growth rates of Shy Albatrosses (T. cauta) breeding in Australia. Condor 104:12–29

Huin N, Prince PA (2000) Chick growth in albatrosses: curve fitting with a twist. J Avian Biol 31:418–425

Huin N, Prince PA, Briggs DR (2000) Chick provisioning rates and growth in Black-browed Albatross Diomedea melanophrys and Grey-headed Albatross D. chrysostoma at Bird Island, South Georgia. Ibis 142:550–565

Marchant S, Higgins PJ (1990) (Coordinators) Handbook of Australian, New Zealand and Antarctic Birds. Oxford Uni- versity Press, Oxford

Nel DC, Lutjeharms JRE, Pakhomov EA, Ansorge IJ, Ryan PG, Klages NTW (2001) Exploitation of mesoscale oceanographic features by Grey-headed Albatross T. chrysostoma in the southern Indian Ocean. Mar Ecol Prog Ser 217:15–26

Pennycuick CJ, Croxall JP, Prince PA (1984) Scaling of foraging radius and growth rate in petrels and albatrosses. Ornis Scan 15:145–154

Prince PA, Walton DWH (1984) Automated measurement of meal sizes and feeding frequency in albatrosses. JApplBiol 21:789–794

Phillips RA, Green JA, Phalan B, Croxall JP, Butler PJ (2003) Chick metabolic rate and growth in three species of albatross: a comparative study. Comp Biochem Physiol Part A Mol Integr Physiol 135:185–193

Phillips RA, Silk JRD, Croxall JP (2005) Foraging and provi- sioning strategies of the Light-mantled Sooty Albatross at South Georgia: competition and co-existence with sympatric marine predators. Mar Ecol Prog Ser 285:259–270

Prince PA, Croxall JP, Trathan PN, Wood AG (1998) The pelagic distribution of South Georgia Albatrosses and their relation- ships with fisheries. In: Robertson G, Gales R (eds) Albatross biology and conservation. Surrey Beatty and Sons, Chipping Norton, pp 137–167

Reid K, Liddle GM, Prince PA, Croxall JP (1999) Measurement of chick provisioning in Antarctic Prions Pachyptila desolata using and automated weighing system. J Avian Biol 30:127–134

Richards FJ (1959) A flexible growth function for empirical use. J Exp Bot 20:290–300

Richdale LE (1952) Post-egg period in albatrosses. Biol Monogr 4:1–66

Ricketts C, Prince PA (1981) Comparison of growth in albatrosses. Ornis Scan 12:120–124

Ricketts C, Prince PA (1984) Estimation by use of field weighings of metabolic rate and food conversion efficiency in albatross chicks. Auk 101:790–795

Ricklefs RE (1968a) Patterns of growth in birds. Ibis 110:419–451 RicklefsRE (1968b)Weight recession in nestlingbirds.Auk85:30–35 Ricklefs RE (1973) Patterns of growth in birds 2. growth rate and

mode of development. Ibis 115:177–201 Ricklefs RE (1983) Some considerations of the reproductive ener-

getics of pelagic birds. Stud Avian Biol 8:84–91 Terauds A, Gales R, Baker GB, Alderman R (2006) Foraging areas

of Black Browed and Grey-headed Albatrosses fromMacquarie Island in relation to Marine Protected Areas. Aquat Conserv– Mar Freshw Ecosyst 16:133–146

Tickell WLN (1968) The biology of the Great albatrosses Diomedea exulans and D. epomophora. In: Austin OL (ed) Antarctic Bird Studies. American Geophysical Union, Washington, pp 1–55

Tickell WLN, Pinder R (1975) Breeding biology of the Black- browed albatross Diomedea melanophrys and Grey-headed albatross D. chrysotoma at Bird Island, South Georgia. Ibis 117(4):433–451

925

Thomas G, Croxall JP, Prince PA (1983) Breeding biology of the Light–mantled Sooty Albatross (Phoebetria palpebrata) at South Georgia. J Zool (Lond) 199:123–135

Waugh SM, Weimerskirch H, Cherel Y, Prince PA (2000) Con- trasting strategies of provisioning and chick growth in two symmetrically breeding albatrosses at Campbell Island, New Zealand. Condor 102:804–813

Weimerskirch H (1998) Foraging strategies of Indian Ocean Albatrosses and their relationships with fisheries. In: Robertson G, Gales R (eds) Albatross biology and conservation. Surrey Beatty and Sons, Chipping Norton, pp 168–179

Weimerskirch H, Lys P (2000) Seasonal changes in the provisioning behaviour and mass of male and female Wandering albatrosses in relation to the growth of their chick. Polar Biol 23(11):733– 744

Weimerskirch H, Robertson G (1994) Satellite tracking of Light- mantled sooty albatrosses. Polar Biol 14(2):123–126

Weimerskirch H, Jouventin P, Stahl JC (1986) Comparative ecol- ogy of the six albatross species breeding on the Crozet Islands. Ibis 128(2):195–213

Weimerskirch H, Chastel O, Ackermann L, Chaurand T, Cuenotchaillet F, Hindermeyer X, Judas J (1994a) Alternate long and short foraging trips in pelagic seabird parents. Anim Behav 47:472–476

Weimerskirch H, Doncaster CP, Cuenotchaillet F (1994b) Pelagic seabirds and the marine environment—foraging patterns of Wandering Albatrosses in relation to prey availability and distribution. Proc R Soc Lond B Biol Sci 255:91–97

Weimerskirch H, Chastel O, Ackermann L (1995) Adjustment of parental effort to manipulated foraging ability in a pelagic seabird, the Thin-billed Prion Pachyptila belcheri. Behav Ecol Sociobiol 36:11–16

Weimerskirch H, Mougey T, Hindermeyer X (1997) Foraging and provisioning strategies of Black-browed Albatrosses in relation to the requirements of the chick—natural variation and exper- imental study. Behav Ecol 8:635–643

Weimerskirch H, Fradet G, Cherel Y (1999) Natural and experi- mental changes in chick provisioning in a long-lived seabird, the Antarctic Prion. J Avian Biol 30:165–174

Weimerskirch H, Prince PA, Zimmermann L (2000) Chick provi- sioning by the yellow-nosed Albatross Diomedea chlororhyn- chos: response of foraging effort to experimentally increased costs and demands. Ibis 142:103–110

926