Prolonged foraging trips and egg desertion in little penguins (Eudyptula minor)

New Zealand Journal of Zoology ISSN: 0301-4223 (Print) 1175-8821 (Online) Journal homepage: http://www.tandfonline.com/loi/tnzz20 Prolonged foraging trips and egg desertion in little penguins (Eudyptula minor) Mihoko Numata, Lloyd S. Davis & Martin Renner To cite this article: Mihoko Numata, Lloyd S. Davis & Martin Renner (2000) Prolonged foraging trips and egg desertion in little penguins (Eudyptula minor), New Zealand Journal of Zoology, 27:4, 277-289, DOI: 10.1080/03014223.2000.9518236 To link to this article: https://doi.org/10.1080/03014223.2000.9518236 Published online: 30 Mar 2010. Submit your article to this journal Article views: 306 View related articles Citing articles: 23 View citing articles Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalinformation?journalcode=tnzz20

New Zealand Journal of Zoology, 2000, Vol. 27: 277-289 0301-4223/00/2704-0277 $7.00/0 The Royal Society of New Zealand 2000 277 Prolonged foraging trips and egg desertion in little penguins (Eudyptula minor) MIHOKO NUMATA LLOYD. S. DAVIS MARTIN RENNER Department of Zoology University of Otago P.O. Box 56 Dunedin, New Zealand Abstract Penguins share incubation duties between the male and female of a pair. The one left on the nest must fast while its partner is foraging at sea. Nest attendance patterns during the incubation period reflect the way time is allocated between the conflicting demands of incubation and foraging. We conducted daily nest checks for little penguins (Eudyptula minor) at two localities in New Zealand (Motuara Island in the Marlborough Sounds and Oamaru, Otago) in the 1998/99 breeding season and found significant variation in nest attendance between the two areas. Penguins breeding at Motuara Island made significantly longer foraging trips and were in poorer body condition than those breeding at Oamaru. Foraging trip durations were significantly correlated with the body condition of birds at both areas: the poorer the condition, the longer the foraging trip durations. The increased risk of egg desertion observed at Motuara Island is probably a consequence of the prolonged foraging trips that limit the time available for incubation. Keywords Eudyptula minor, E. m. variabilis; E. m. minor, little penguin; incubation; nest attendance; foraging trip; body condition; egg desertion Z99054 Received 22 December 1999; accepted 29 May 2000 INTRODUCTION Life history theory predicts that an individual's investment in current reproductive effort may be curtailed to maximise its lifetime productivity (Roff 1992; Stearns 1992). If the cost of reproduction is too great, an animal may abandon a breeding attempt (Aebischer & Wanless 1992; Chastel et al. 1993). For example, a fasting incubating bird may desert the nest to start feeding before it starves to death (Groscolas 1990). With the exception of the emperor penguin (Aptenodytes forsteri), penguins share incubation duties between the male and female of a pair. A fasting penguin on the nest is relieved when its foraging partner returns from sea. In the offshore-feeding Adelie penguin (Pygoscelis adeliae), incubation spells can last for about two weeks and unusually long foraging trips may cause nest desertions (Davis 1982, 1988). Such desertions are not uncommon in offshore-feeding penguins (Lishman 1985; Groscolas 1990; Olsson 1997; Croxall & Davis 1999) and flying pelagic seabird species (e.g., Chaurand & Weimerskirch 1994a, b; Lorentsen & Rov 1995). Co-ordination of nest relief between partners throughout the incubation period is a prerequisite to ensure successful breeding (Croxall & Ricketts 1983). Co-ordination of nest relief involves two 'decisions': first, the foraging bird has to decide when to return, and second, the incubating bird has to decide when to leave the nest. The body condition of a bird may play an important role in determining these behaviours (McNamara & Houston 1996). Birds in poor condition may need to spend a longer time foraging to recover (e.g., Weimerskirch 1995; Tveraa et al. 1997) or may be able to fast for a shorter period (e.g., Yorio & Boersma 1994). Little penguins (Eudyptula minor) are generally regarded as inshore feeders (Croxall & Davis 1999). They are widespread along the coast of New Zealand and southern Australia (Williams 1995). Radiotracking studies of foraging little penguins (E. m. novaehollandiae) breeding on Phillip Island in Victoria, Australia, showed that for most of the year Published online 30 Mar 2010

278 New Zealand Journal of Zoology, 2000, Vol. 27 Motuara Island Fig. 1 Location map of the sampling sites. OBPC: Oamaru Blue Penguin Colony; OCPR: Oamaru Creek Penguin Refuge. New Zealand most foraging trips last only one day (Weavers 1992; Collins et al. 1999), although one-day trips were relatively less frequent during incubation (Collins et al. 1999). However, little penguins (E. m. variabilis) breeding on Motuara Island in the Marlborough Sounds, New Zealand, went on much longer foraging trips. The mean durations recorded were 4.7 and 7.2 days during the incubation period in 1995 and 1996, respectively (Renner 1998). Renner (1998) found the longer the foraging trip duration of the parents, the higher the probability of egg desertion. He estimated egg desertion rates of the penguins on Motuara Island to be 18 and 37 % in 1995 and 1996, respectively. Chick survival of the penguins observed on the island was 36 % (Renner 1998). This is as low as low breeding success recorded elsewhere (c.f. Reilly& Cullen 1981; Fortescue 1995; Perriman & McKinlay 1995; Rogers et al. 1995; Perriman 1997; Bullen 1997; D. Houston pers. comm.). On Motuara Island, failure of foraging parents to return in time to feed their newly-hatched chicks due to the long foraging duration contributed to the low chick survival (Renner 1998). In contrast, the breeding success of little penguins (E. m. minor) at Oamaru in Otago, New Zealand, is among the highest ever recorded. A weekly monitoring programme at Oamaru extending over the last five years has shown an average fledging rate of 89% and low egg desertion rates of 0-7% (D. Houston unpublished data). This suggests that

Numata et al. Duration of foraging and egg desertion in little penguins 279 nest relief failures are rarer at Oamaru than elsewhere. The aim of this study was to examine the association between foraging trip durations and egg desertions in little penguins. We made the following predictions and tested related hypotheses: (i) penguins breeding on Motuara Island would make longer foraging trips than those breeding at Oamaru; (ii) one-day foraging trips would be more frequent at Oamaru than at Motuara Island; and (iii) risk of egg desertion would be higher at Motuara Island than at Oamaru. We also examined the correlation between the body condition of the penguins and their foraging trip durations. The results may give us an explanation for the variable breeding performance of the species from locality to locality and/or from year to year. STUDY AREAS AND METHODS Study areas and animals This study was conducted at two colonies of little penguins in New Zealand: Motuara Island (4 l 06'S, 174 16'E) in the Marlborough Sounds; and Oamaru (45 06'S, 170 58'E) in Otago (Fig. 1), during the breeding season in 1998/99. MN conducted all the sampling at both study areas (Table 1). Motuara Island is 59 ha with its highest peak at 128 m above sea level. Once modified intensively for farming, it became a scenic reserve in 1926. It has been free from introduced predators since 1990. About 300 pairs of little penguins are thought to breed throughout the island (Renner 1998). We restricted sampling sites mainly to the south-western side of the island after Renner (1998). Penguin nests were searched and checked daily for nest attendance and status of eggs during daylight hours. Nests were found in earth burrows, rock crevices, hollow trees, under vegetation or in nest boxes and scattered from near sea level to the top of the island. We disregarded nests where the egg(s) were already abandoned when found for the first time, as such eggs could have been laid in previous breeding seasons. A total of 74 incubating individuals occupying 39 nests were sampled on Motuara Island. Nesting penguins were marked using TIRIS transponders implanted subcutaneously in the upper back. In some cases only one bird of a pair was marked with a transponder and its partner was marked with a number painted with white acrylic paint on its flippers and forehead. Birds were weighed to the nearest 10 g using a Pesola spring balance. Flipper length (from the middle of elbow joint to the tip, dorsally) was measured to the nearest 1 mm using a stainless steel ruler. Bill length and bill depth (at the nares, Warham 1975) were measured to the nearest 0.1 mm with vernier calipers. Nest contents were checked either by lifting Table 1 Sampling scheme of this study. Sampling period Areas Number of pairs CC date known CC date unknown Mean CC date 10Aug-19Sep 24 Sep-2 Dec 5 Dec-5 Jan OBPC OCPR Motuara I. OBPC OCPR 1 2 25 1 3 23*1 11 5 29 Sep* 2 (8 Oct* 3 ) 17 Oct* 4 (28 Nov* 5 ) 14 Oct* 6 (6 Dec* 7 ) OBPC: Oamaru Blue Penguin Colony. OCPR: Oamaru Creek Penguin Refuge. CC: Clutch completion. * 1: Of these, nine pairs were observed only during the chick-rearing period. CC date of these nine pairs were estimated as 37 days before observed or estimated hatching. *2: Estimated mean for all breeding pairs observed in this study (n = 48). *3: Observed mean for 25 pairs used in the clutch survival analysis. *4: Estimated mean for all breeding pairs recorded in OBPC weekly monitoring (n = 108). *5: Estimated mean for 12 pairs used in the clutch survival analysis. *6: Estimated mean for all breeding pairs recorded in OBPC weekly monitoring (n = 83). *7: Estimated mean for 8 pairs used in the clutch survival analysis.

280 New Zealand Journal of Zoology, 2000, Vol. 27 the incubating bird up with a pole or feeling by hand underneath the abdomen of the bird. The Oamaru study area consisted of two sampling sites: the Oamaru Blue Penguin Colony (OBPC) and the Oamaru Creek Penguin Refuge (OCPR). OBPC was established in late 1992 for a tourist operation providing supervised penguin viewing. OCPR was established as a control site with no tourist impact (Houston 1997). Both sites were partitioned off from public access. The OBPC staff have monitored penguins occupying wooden nest boxes placed into earth mounds at these sites since 1993. The monitoring work involves marking of all the penguins encountered with individually numbered metal flipper bands, and recording of nest attendance and nest contents. This has revealed no adverse impact of tourist operation on the breeding of the penguins (D. Houston unpublished data). Cats (Felis catus) and Norway rats (Rattus norvegicus) were occasionally sighted at these sites. However, no evidence of predation was found during this study. For this study, a total of 64 individuals occupying 34 nest boxes for incubation were sampled. All die penguins sampled had already been numbered with a flipper band. Body mass and other measurements were taken in the same manner as were done at Motuara Island. When the lids of the nest boxes were opened, most of the nesting birds raised the upper part of their bodies and any eggs were visible. This enabled nests to be checked without touching the birds. However, birds sometimes had to be handled to read the ID number on the flipper band. The necessity for doing this was much reduced from December by dabbing red enamel paint on the forehead of one of the pair. The presence/ absence of the colour and the visible part of the ID number satisfied identification most of the time. Penguins with an enlarged external cloaca shortly after egg laying were sexed as females and their partners as males (Boersma & Davies 1987). To sex penguins found after egg laying, we used a discriminant function developed by Renner & Davis (1999) for little penguins at Motuara Island, and another one developed by Hocken & Mackley (pers. comm.) for penguins at Oamaru. We followed subspecific division of the species described by Kinsky&Falla(1976). Definitions and data analysis Nest attendance patterns Incubation was the period between laying of the second egg (= clutch completion) and hatching of the first chick. We defined an incubation spell as a period during which a penguin was seen on the nest when it was inspected, and foraging trip duration as the period during which a penguin was absent from the nest. We termed a spell in which the incubating bird left the nest before its foraging partner returned as a nest relief failure spell (NRF spell). The last foraging trip of the parent present at hatching of the first chick was called the last foraging trip (LFT). The LFTs of the species were expected to be mostly one day long as reported by Renner (1998) and Collins et al. (1999). We calculated mean duration of incubation spells (excluding NRF spells) and of foraging trips (excluding LFTs) for each individual, and also proportion of one-day foraging trips per individual (excluding LFTs). Differences in the means of these variables between Oamaru and Motuara Island, and between the sexes, were tested by using two-way ANOVAs. Durations of NRF spells were known for four and 16 penguins from Oamaru and Motuara Island, respectively. We tested differences between the mean duration of NRF spells and the mean duration of incubation spells of other individuals using Mests. Here paired comparisons between NRF spells and other spells for the same individuals would have been more appropriate, but the lack of sufficient data resulted in these unpaired tests without dependency. Durations of LFTs were known for 16 and seven penguins from Oamaru and Motuara Island, respectively. Frequency of one-day foraging trips in LFTs were compared for each study area with the mean proportion of one-day foraging trips per individual as expected values by log-likelihood (G) tests (Sokal & Rohlf 1995). We disregarded spells/trips if we failed to determine the exact duration of the spells/trips. Data about the duration of incubation spells/foraging trips were log-transformed and proportions were arcsine-transformed to satisfy the homoscedasticity assumption of Mests or ANOVAs. Body condition of penguins Body mass has been used as an index of body condition (Johnson et al. 1985) and related to both breeding performance (Jones & Ward 1976; Blums et al. 1997; Olsson 1997) and survival (Lack 1966, pp. 276-277) of birds. Since fat has been estimated to be the major source (55 %) of energy in penguins (Croxall 1982; Costa et al. 1986), fat reserve can be a more direct measure of body condition than body mass (Rogers 1991). However, measuring or estimating fat content in a bird is difficult to conduct in the field (e.g., Rogers 1991). Body mass often correlates well with fat reserves (e.g., white-fronted geese Anser albifrons; Johnson et al. 1985). A ratio of body mass against a linear measure of body size

Numata et al. Duration of foraging and egg desertion in little penguins 281 is known to improve the correlation between the condition index and fat reserves (e.g., Odum et al. 1964; Owen & Cook 1977). We used a body mass/ flipper length ratio index. We assumed no physical growth in sampled penguins throughout this study. We obtained daily body mass loss data during the incubation spell from 27 penguins (13 males and 14 females) at Oamaru and 22 penguins (11 males and 11 females) at Motuara Island by weighing them at intervals of 2-7 days, except one female weighed daily for two consecutive days at Oamaru. To estimate body mass of a bird before/after weighing, we used an equation taken from Croxall (1982): PF,= W o e kt, where W o is the body mass at the time of weighing, W t is the body mass at t days before/after weighing, and k is the proportion of body mass lost per day. For those birds from which daily mass loss values were not available, we used the mean k value. This way of estimating body mass allowed us to calculate the body mass and the condition index of a bird on the last day of an incubation spell (= preforaging body mass and pre-foraging condition index) without unnecessary disturbance from weighing the birds too often. Post-foraging body mass and condition indices were obtained for 35 penguins (18 males and 17 females) at Oamaru and 31 birds (15 males and 16 females) at Motuara Island with known duration of foraging trips. We subtracted post-foraging body mass from pre-foraging body mass to calculate daily body mass gain, and tested differences in the mean daily mass gains between the two areas and sexes using twoway ANOVAs. For those females that we weighed between the laying of the first and the second eggs, mass of the second egg (50-60 g) was subtracted from the body mass of the female. Seven penguins (four males and three females) from Oamaru and seven penguins (two males and five females) from Motuara Island were weighed at the return from foraging in the first half and again in the last half of the incubation period with an interval of 2-3 weeks. We used these data to test for difference in body mass changes over the entire incubation period between the two areas using a one-way ANOVA. Clutch survival during the incubation period Survival analysis was used to assess the risk of nest desertion resulting from nest relief failures against other risks. We estimated 'clutch survival' by treating each clutch of two eggs as a sample unit using a Kaplan-Meier estimator (Kaplan & Meier 1958) in StatView 4.51 (Abacus Concepts, Inc., Berkeley, USA). Cumulative clutch survival rates are reported for 37 days, the mean duration of the incubation period observed in this study (n = 9). We used the fate of 25 clutches with known clutch completion date at Motuara Island. At Oamaru, clutch completion date was known for seven pairs (Table 1). Another 16 pairs were observed daily at Oamaru from the first/second week of incubation (the weekly OBPC monitoring recorded no clutch loss or failure in the week preceding the start of these observations). Clutch completion dates of these were estimated to ± 3.5 days from the weekly OBPC monitoring (M. Home pers. comm.). If a nest hatched at least one chick, the clutch was defined as successful and 'censored' (Cox & Oakes 1984) on the day of hatching of the first chick. Differences in estimated clutch survival between Oamaru and Motuara Island were examined using logrank tests (Kalbfleisch & Prentice 1980, pp. 17-18). We classified a clutch as deserted if a period of absence of both parents from the nest lasted for at least five consecutive days. Little penguin eggs can remain viable when unattended for up to five days on Phillip Island in Australia (P. Dann pers. comm.). Failure time for the clutch desertions was defined as the first day of the period of absence of both parents. Other types of clutch failure observed at both study areas were egg loss (broken, missing or rolled out of the nest) and intrusion of conspecifics and, on Motuara Island, flooding by rainfall. We estimated overall clutch survival from all types of failures but three flooded nests on Motuara Island were censored on the day of flooding, because the risk of flooding at Oamaru was considered zero with no flooded nests recorded over the last five years (D. Houston unpublished data). When the failure time differed for two eggs in one clutch, we used the latter failure time. Two nest desertions were presumably caused by our handling of the nesting birds at Motuara Island. The individuals (one male and one female) each left its nest on the next day after being handled, before its foraging partner returned. These clutches were censored on the first day of desertion. RESULTS Nest attendance patterns The mean duration of incubation spells was significantly longer at Motuara Island than at Oamaru {F X1 2 = 26.74, P < 0.001) with no significant difference between the sexes (F 172 = 0.75, P > 0.5) (Table 2; Fig. 2). Likewise, the mean duration of foraging trips was significantly longer at Motuara Island than at Oamaru (F ] 73 = 35.52, P < 0.001) with no significant

282 New Zealand Journal of Zoology, 2000, Vol. 27 Oamaru Motuara I. Example Oamaru*i Example Motuara l.*2 Clutch complstion Male :emale Female Male Female Time (days) Hatching i i i i i r i r i 0 5 10 15 20 25 30 35 37 77A V77J Y/A V7A f 77A V7777A Y7777/ 77777A 7777A V7777, V7777/. /A Male \yy77. 777/ 3 UV777 <7/A i f '777777A 77777777/ 7777/ V777777, K77777777, Fig. 2 Schematic diagrams (top two) and observed examples (bottom two) of nest attendance patterns in little penguins at Oamaru and Motuara Island during the incubation period in the 1998/99 breeding season. Shaded areas show periods of attendance. * 1: The paler shades in the example from Oamaru correspond to the period when daily observation was lacking.*2: The example from Motuara Island had 39 days of incubation. Table 2 Duration of the foraging trips/incubation spells and body mass change in little penguins at Motuara Island and Oamaru during the incubation period in the 1998/99 breeding season (mean ± SEM). Sample sizes in parentheses. Foraging trip duration (days) Incubation spell duration (days) NRF spell duration (days) Body mass * (g) Condition index * (g/vnm) Mass loss/day (%) Mass gain/trip (%) Mass gain/day (%) Mass gain over the entire incubation period (%) Motuara Oamaru Males Females Males Females 6.6 ±0.6 (19) 6.2 + 0.6(23) 8.9 ±0.6 (8) 1070 ±17.0 (36) 11.38 ±0.17 (36) 3.3 ±0.5 (11) 19.4 ±0.5 (6) 2.5 ±0.6 (6) 3.0 ±0.3 (2) 6.3 ±0.6 (18) 5.2 ±0.5 (15) 8.6 ± 1.1 (8) 971 ±13.8 (38) 10.60 ±0.15 (38) 2.9 ±0.4 (11) 20.0 ± 0.2 (7) 2.6 ± 0.4 (7) 1.3 ±0.1 (5) 2.8 ± 0.2(20) 2.9 ±0.3 (18) 9.0 ±2.9 (3) 1200 ±15.5 (35) 12.83 ±0.16 (35) 3.0 ±0.04 (14) 10.6±2.4(16) 2.9 ±0.7 (16) 12.6 ±1.8 (4) 3.6 ±0.5 (20) 3.2 ±0.4 (20) 13(1) 1076 ±19.4(35) 11.95 ±0.25 (35) 2.9 ±0.4 (13) 13.0 ±2.5 (16) 3.3 ±0.5 (16) 12.8 ±0.5 (3) * Mean during the incubation period. For those birds that we weighed more than once, mean of the all weights/indices per individual was used. difference between the sexes (F il3 = 0.34, P > 0.5). The mean proportion of one-day foraging trips was significantly higher at Oamaru than at Motuara Island (F,, 73 = 12.19, P < 0.001) with no significant difference between the sexes (F 173 = 0.01, P > 0.5) (Fig. 3). At Oamaru, 68 % of the LFTs were one day long. This was significantly higher than the mean proportion at Oamaru (39%, n = 40) (G = 4.790, P < 0.05). On Motuara Island, 24% of LFTs lasted one day; this did not vary significantly from the mean proportion at Motuara Island (14%, n = 37) (G = 0.752, P> 0.05). On Motuara Island, the mean duration of NRF spells was significantly longer than the mean duration of incubation spells of other individual females (t = 2.187,18 df, P < 0.05) and the difference was nearly significant in males (t = 1.925, 23 df, P = 0.067) (Table 2). At Oamaru, very few NRF spells were observed but the mean duration of NRF spells was, with the unbalanced sample size, significantly longer than the mean duration of incubation spells of other individual males (t = 4.011,19 df, P < 0.001) (Table 2). No statistical comparison was possible for females at Oamaru. Changes in body condition of penguins during the incubation period The mean body mass (F, 140 = 50.42, P < 0.001) and condition index (F, 140 = 56.77, P < 0.001) during incubation was significantly lower in penguins at Motuara Island than in penguins at Oamaru (Table 2). Males were heavier (F, 140 = 46.08, P < 0.001)

Numata et al. Duration of foraging and egg desertion in little penguins 283 u <u 0) 30-20- 10-30 0.5 Oamaru (n = 40) Motuara Island (n = 37) correlated with foraging trip durations except in females from Oamaru (Fig. 4). The earliest breeding female at the Oamaru study area made a 12-day foraging trip, the longest observed at this site, and it gained only 60 g or 6% of its mass. Exclusion of this outlier resulted in the correlation coefficient as high as in other groups. The foraging trip durations were negatively correlated with pre-foraging body conditions at both sites and the poorer the condition, the longer the foraging trip durations (Fig. 5). The percentage of body mass gained per day did not vary significantly between the two areas (F x ^ = 0.48, P > 0.4) with no significant difference between the sexes (F 1>41 = 0.14, P > 0.7) (Table 2). The mean daily body mass loss while on the nest and the mean daily body mass gain at sea indicated that penguins at Motuara Island tended to lose mass over the incubation period, while those at Oamaru tended to maintain their mass (Table 2). Even when looking at the change in body mass over the entire incubation period, body mass gain was significantly greater at Oamaru than at Motuara Island (one-way ANOVA, F l>n = 9.572, P< 0.05). o c d> 3 IX 0> 20-10- 0 0.5 1 Proportion of one-day trips Clutch survival during the incubation period At Motuara Island, five of the six clutch desertions occurred in the first week of incubation, and the last one in the second week. The cumulative clutch survival from desertions was 0.712 (n = 25). This tended to be lower than that at Oamaru (0.909, n = 23) (log-rank test, % 2 = 3.391, P = 0.066), where two out of the 23 clutches were deserted (Fig. 6). Clutch survival from all types of failures was estimated to be 0.326 at Motuara Island. This was significantly lower than at Oamaru (0.652; log-rank test, % 2 = 4.521, P< 0.05). Fig. 3 Histogram of proportion of one-day foraging trips in little penguins breeding at Oamaru and Motuara Island during the incubation period in the 1998/99 breeding and had a higher condition index (F i ^o = 20.00, P < 0.001) than females at both study areas. The percentage of body mass lost per day during the incubation spell did not vary significantly between penguins from the two areas (F[4i = 0.51, P > 0.4) with no significant difference between the sexes (F,4 5 = 4.031, P> 0.05). After foraging at sea, penguins returned with significantly increased body mass. The proportion of body mass gained per trip was positively DISCUSSION Prolonged foraging trips and higher risk of egg desertion Foraging trip duration is generally positively correlated with foraging trip distance (Bethge et al. 1997). Our results revealed a significant variation in the nest attendance patterns of little penguins at Oamaru and Motuara Island. Foraging trip durations were longer at Motuara Island than at Oamaru. This suggests that little penguins from Motuara Island travel, at least in the study period, further away from the colony to forage than those from Oamaru. Another possible explanation is that food sources may be less plentiful around Motuara Island compared to Oamaru. The

284 New Zealand Journal of Zoology, 2000, Vol. 27 Fig. 4 Correlation between foraging trip duration and proportion of body mass gained per trip in little penguins breeding at Oamaru and Motuara Island during the incubation period in the 1998/99 breeding season. Males: filled circles and solid line; females: blank circles and dashed line. o o. o 0-.9-0.5-i o 0.4-0. 0.3-0.2- S 0.1- -0.1- Oamaru Males: n = 18, r = 0.786, P < 0.001 Females: n = 16, r = 0.807, P < 0.001 8 10 12 14.9-0.5-, a a> c '5 O) (0 (0 re E c o E oao 0- S. 0.4-0.3-0.2-0.1-0.1- Motuara 2 4 6 8 10 12 14 Trip duration (days) Island Males: n = 15, r = 0.737, P < 0.005 Females: n = 16, r = 0.618, P < 0.01 long foraging trips observed on Motuara Island probably increased nest relief failures and also the risk of clutch desertions. There may be a critical level of body fat and/or protein reserves triggering the start of re-feeding in incubating penguins (Groscolas 1990; Olsson 1997). Increased clutch desertions caused by delayed nest relief have been observed in another inshore-feeding species, the gentoo penguin (Pygoscelispapua) (Bost & Jouventin 1990). The mean foraging trip duration of penguins from Motuara Island recorded in our study was within the range of previously reported figures for the same area (4.7 and 7.1 days in 1995 and 1996, respectively; Renner 1998). Renner (1998) concluded that the unusually long foraging trips at Motuara Island were partly a response to the reduced availability of pilchard (Sardinops neopilchardus) that died off in the Marlborough Sounds during the 1995 breeding season (Smith et al. 1996). Increased time spent foraging at sea in the face of food shortage has been reported for little penguins (Weavers 1992; Hobday 1992) and also several other seabirds and marine mammals (e.g., Croxall et al. 1988; Le Maho et al. 1993). Long foraging trips of penguins from Motuara Island were common prior to the pilchard die-off

Numata et al. Duration of foraging and egg desertion in little penguins 285 Fig. 5 Correlation between preforaging condition index and foraging trip duration in littlepenguins breeding at Oamaru and Motuara Island during the incubation period in the 1998/99 breeding season. Males: filled circles and solid line; females: blank circles and dashed line. (A c o "D a. 14-12- 10 8 6 4- Oamaru Males: n = 18, r = -0.586, P < 0.01 Females: n = 16, r = -0.494, P < 0.05 2-0- 9 10 11 12 13 14 12- f 10- c 8^ o I 6- o Q. 4 - Motuara Island Males: n = 15, r = -0.527, P < 0.05 Females: n = 16, r = -0.489, P < 0.05 2-0- 8 9 10 11 12 13 14 Condition index (g/mm) (Renner 1998), suggesting that local food sources for penguins were generally limited. The different timing of sampling at the two study areas could have had an influence on our results. Intra-year variations in the breeding success of little penguins have been reported from Phillip Island in Australia (Reilly & Cullen 1981) and Motuara Island (Renner 1998). This might be a consequence of variability in the availability of prey within a season (Hobday 1992). There was a strong El Nino event in the year 1997/ 98 (Basher 1998) followed by a La Nina in 1998/ 99. The changes in sea surface temperature due to these events are likely to affect the relative abundance offish in New Zealand waters (Basher 1998). The effects are considered to vary for different species of fish, as breeding success of fish is often regulated by sea temperature that varies for different species (Robertson 1980; Basher 1998). In the 1998/ 99 breeding season, the onset of egg laying at Oamaru (23 August) was 4-15 weeks later than in the previous five seasons (D. Houston pers. comm.). The estimated mean clutch completion date on Motuara Island in this study (29 September) was 4-6 weeks later than in the 1995 and 1996 breeding seasons (Renner 1998). Such delays may be related

286 New Zealand Journal of Zoology, 2000, Vol. 27 0.8 " 1.0-0.6-.2 0.4 - I O 0.2-0- 10 15 20 25 Time (days) 30 35 Fig. 6 Cumulative clutch survival from desertions in little penguins breeding at Oamaru (dashed line) and Motuara Island (solid line) in the 1998/99 breeding season. to limited food supply earlier in the year, as suggested by Cullen et al. (1992). The low body mass gain while foraging recorded for the earliest female breeder of the Oamaru study area supports this assumption. El Nino/La Nina events could have caused reduced availability of prey near the colony. Prolonged foraging trips and body condition of penguins Weavers (1992) suggested that little penguins in poor body condition tended to spend a longer time foraging than those in good condition. It is likely that penguins extend foraging trips to recover depleted body condition. Our results support this assumption and suggest that penguins in poor condition are able to fast for a shorter time than those in good condition, make longer foraging trips and, thus, are more likely to desert their eggs. Body condition was higher in penguins from Oamaru than in those from Motuara Island. At Oamaru, penguins seemed to have accumulated body mass towards the end of the incubation period. Such good body condition probably resulted in the high frequency of one-day foraging trips during incubation and also the increase of one-day foraging trips near hatching. Increase in body mass after egg-laying and towards the end of the incubation period has been observed in several Procellariiformes (Prince et al. 1981; Croxall & Ricketts 1983; Hatch 1990; Lorentsen & Rev 1995). The presumably longer distance travelled by penguins from Motuara Island could require use of additional energy reserves, preventing the accumulation of body mass over the incubation period. Renner (1998) and Collins et al. (1999) reported an increase in one-day foraging trips preceding hatching in little penguins. This is likely to be the same behaviour shown by offshore-feeding Adelie penguins (Davis 1988; Davis & Miller 1990) and also Procellariiformes (Johnstone & Davis 1990; Warham 1990; Lorentsen & R0v 1995); birds curtail foraging trips near the end of the incubation period to ensure they are back at the nest in time to feed newly-hatched chicks. This is consistent with the suggestion that penguins possess a hormonallybased internal timer that triggers the return of foraging birds just prior to hatching (Davis et al. 1995). Little penguin chicks begin to die three days after hatching if not fed (Renner 1998). Adult little penguins tested in captivity digested food within 16 hours (Gales 1987), indicating that parents in attendance of the nest for more than one day may not have enough food left in their stomach to feed chicks. Considerations A consideration of our results is a lack of adequate replications. The variable behaviour of the species observed at the two study areas could potentially be related not only to the availability of food sources, but also to other biotic and abiotic factors. For example, the significant difference in body mass of the penguins from Oamaru and Motuara Island could merely reflect morphological variations between the two subspecies. The validity of the body mass/flipper length ratio index used in our study as a measure of fat reserves needs to be examined further. On Phillip Island in Australia, Dann et al. (1995) did not detect any adverse effect caused by repeated weighing of little penguins taken at every stage of breeding over several years. Our repeated handling of penguins may have induced two clutch desertions at Motuara Island. At Oamaru, on the other hand, the hatching success (proportion of eggs that hatched over the total number of eggs laid) for the 23 breeding pairs observed daily by us was 61.4% (n = 44 eggs). This did not vary significantly from that for another 174 clutches laid by 15 8 pairs at Oamaru prior to our daily observation and monitored weekly by OBPC staff (72.9%, n = 340 eggs; D. Houston unpublished data) (G = 2.40, P > 0.1). The repeated exposure to handling for penguins at Oamaru could have resulted in increased tolerance of the animals to such disturbance. Of the 35 penguins at Motuara Island that failed to hatch chicks, 25 were never seen again through the rest of the study period (> 3 weeks), while the

Numata et al. Duration of foraging and egg desertion in little penguins 287 eight individuals that failed to hatch chicks at Oamaru were re-sighted within three weeks. There are at least two possible explanations for such a difference in re-sighting. First, the mortality of penguins from Motuara Island may be higher than at Oamaru due to the poorer body condition or the longer time spent at sea and exposed to marine predators. Second, we might have failed to locate penguins if they moved to nest sites outside our sampling site. Further investigations are necessary to follow the movements of penguins after breeding failure, especially at Motuara Island. ACKNOWLEDGMENTS We would like to thank Dave Houston of Department of Conservation (DoC), Oamaru, Mandy Home of OBPC, Bill Cash of DoC, Picton, and Les and Zoe Battersby of Dolphin Watch Marlborough, Picton, for their support in our fieldwork. Tony Hocken kindly provided the discriminant function for sexing of penguins breeding at Oamaru. Murray Cresswell advised me on the use of body mass index. We thank Kazumi Hataguchi, Yoshiko Yamaga, Satoru Ogawachi, Yoshimi Okayama, Rebecca Rawson, John and Barbara McGann and Fumiya and Shihoko Kamakura for their long-term field assistance. Animals were handled according to University of Otago Animal Ethics Approval number 64-98. Equipment and financial support were provided by a University of Otago Research Grant to LSD and by the Department of Zoology, University of Otago. REFERENCES Aebischer, N. J.; Wanless, S. 1992: Relationships between colony size, adult nonbreeding and environmental conditions for shags Phalacrocorax aristotelis on Isle of May, Scotland. Bird Study 39: 43-52. Basher, R. E. 1998: The 1997/98 El Nino event: impacts, responses and outlook for New Zealand. Review prepared for Ministry of Research, Science and Technology, Report No. 73. Bethge, P.; Nicol, S.; Culik, B. M; Wilson, R. P. 1997: Diving behaviour and energetics in breeding little penguins (Eudyptula minor). Journal of Zoology, London 242: 483-502. Blums, P.; Mednis, A.; Clark, R. G. 1997: Effects of incubation body mass on reproductive success and survival of two European diving ducks: a test of the nutrient limitation hypothesis. The Condor 99:916-925. Boersma, P. D.; Davies, E. M. 1987: Sexing monomorphic birds by vent measurements. The Auk 104: 779-783. Bost, C.-A.; Jouventin, P. 1990: Evolutionary ecology of gentoo penguins (Pygoscelis papua). In: Davis, L. S.; Darby, J. T. ed. Penguin biology. San Diego, California, Academic Press. Pp. 85-1 12. Bullen,L. S. 1997: The breeding biology of the little blue penguin Eudyptula minor on Somes/Matiu Island. MSc thesis, Victoria University of Wellington, New Zealand. Chastel, O.; Weimerskirch, H.; Jouventin, P. 1993: High annual variability in reproductive success and survival of an Antarctic seabird, the snow petrel Pagodroma nivea: a 27-year study. Oecologia 94: 278-285. 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. Journal of Animal Ecology 63: 275-282. Collins, M; Cullen, J. M.; Dann, P. 1999: Seasonal and annual foraging movements of little penguins from Phillip Island, Victoria. Wildlife Research 26: 705-721. Costa, D. P.; Dann, P.; Disher, W. 1986: Energy requirements of free ranging little penguin, Eudyptula minor. Comparative Biochemistry and Physiology 85A: 135-138. Cox, D. R.; Oakes, D. 1984: Analysis of survival data. London, Chapman & Hall. Croxall, J. P. 1982: Energy costs of incubation and moult in petrels and penguins. Journal of Animal Ecology 51: 177-194. Croxall, J. P.; Davis, L. S. 1999: Penguins: paradoxes and patterns. Marine Ornithology 27: 1-12. Croxall, J. P.; McCann, T. S.; Prince, P. A.; Rothery, P. 1988: Reproductive performance of seabirds and seals at South Georgia and Signy Island, South Orkney Islands, 1976-1987: implications for southern ocean monitoring studies. In: Sahrhage, D. ed. Antarctic Ocean and resources variability. Berlin, Heidelberg, Springer-Verlag. Pp. 261-285. Croxall, J. P.; Ricketts, C. 1983: Energy costs of incubation in the wandering albatross Diomedea exulans. Ibis 125: 33-39. Cullen, J. M.; Montague, T. L.; Hull, C. 1992: Food of little penguins Eudyptula minor in Victoria: comparison of three localities between 1985 and 1988. Emu 91: 318-341.

288 New Zealand Journal of Zoology, 2000, Vol. 27 Dann, P.; Cullen, J. M.; Jessop, R. 1995: Cost of reproduction in little penguins. In: Dann, P.; Norman, I.; Reilly, P. ed. The penguins: ecology and management. Chipping Norton, NSW, Surrey Beatty & Sons. Pp. 39-55. Davis, L. S. 1982: Timing of nest relief and its effect on breeding success in adelie penguins (Pygoscelis adeliae). The Condor 84: 178-183. Davis, L. S. 1988: Coordination of incubation routines and mate choice in adelie penguins (Pygoscelis adeliae). The Auk 105: 428-432. Davis, L. S.; Cockrem, J. F.; Miller, G. D.; Court, G. S. 1995: An incubation timer for seabirds: progesterone and its relationship to hatching in adelie penguins. Emu 95: 245-251. Davis, L. S.; Miller, G. D. 1990: Foraging patterns of adélie penguins during the incubation period. In: Kerry, K. R.; Hempel, G. ed. Antarctic ecosystems: ecological change and conservation. Berlin, Heidelberg, Springer-Verlag. Pp. 203-207. Fortescue, M. E. 1995: Biology of the little penguin Eudyptula minor on Bowen Island and at other Australian colonies. In: Dann, P.; Norman, I.; Reilly, P. ed. The penguins: ecology and management. Chipping Norton, NSW, Surrey Beatty & Sons. Pp. 364-392. Gales, R. P. 1987: Validation of the stomach-flushing technique for obtaining stomach contents of penguins. Ibis 129: 335-343. Groscolas, R. 1990: Metabolic adaptations to fasting in emperor and king penguins. In: Davis, L. S.; Darby, J. T. ed. Penguin biology. San Diego, California, Academic Press. Pp. 269-296. Hatch, S. A. 1990: Time allocation by northern fulmars Fulmarus glacialis during the breeding season. Ornis Scandinavia 21: 89-98. Hobday, D. 1992: Abundance and distribution of pilchard and Australian anchovy as prey species for the little penguin Eudyptula minor at Phillip Island, Victoria. Emu 91: 342-354. Houston, D. 1997: Blue penguins at Oamaru: a history. Unpublished report presented at the Oamaru blue penguin seminar September 1997. Johnson, D. H.; Krapu, G. L.; Reinecke, K. J.; Jorde, D. G. 1985: An evaluation of condition indices for birds. Journal of Wildlife Management 49: 569-575. Johnstone, R. M.; Davis, L. S. 1990: Incubation routines and foraging-trip regulation in the grey-faced petrel Pterodroma macroptera gouldi. Ibis 132: 14-20. Jones, P.; Ward, P. 1976: The level of reserve protein as the proximate factor controlling the timing of breeding and clutch-size in the red-billed quelea Quelea quelea. Ibis 118: 547-574. Kalbfleisch, J. D.; Prentice, R. L. 1980: The statistical analysis of failure time data. New York, John Wiley & Sons. Kaplan, E. L.; Meier, P. 1958: Nonparametric estimation from incomplete observations. Journal of the American Statistical Association 53: 457-481. Kinsky, F. C; Falla, R. A. 1976: A subspecific revision of the Austrasian blue penguin (Eudyptula minor) in the New Zealand area. National Museum of New Zealand Records I: 105-126. Lack, D. 1966: Population studies of birds. Oxford, UK, Clarendon Press. Le Maho, Y.; Gendner, J-P.; Challet, E.; Bost, C-A.; Billes, J.; Verdon, C; Plumere, C; Robin, J.-P.; Handrich, Y. 1993: Undisturbed breeding penguins as indicators of changes in marine resources. Marine Ecology Progress Series 95: 1-6. Lishman, G. S. 1985: The comparative breeding biology of adelie and chinstrap penguins Pygoscelis adeliae and P. antarctica at Signy Island, South Orkney Islands. Ibis 127: 84-99. Lorentsen, S.-H.; Rev, N. 1995: Incubation and brooding performance of the Antarctic petrel Thalassoica antarctica at Svarthamaren, Dronning Maud Land. Ibis 137: 345-351. McNamara, J. M.; Houston, A. I. 1996: State-dependent life histories. Nature 380: 215-221. Odum, E. P.; Rogers, D. T.; Hicks, D. L. 1964: Homeostasis of the nonfat components of migrating birds. Science 143: 1037-1039. Olsson, O. 1997: Clutch abandonment: a state-dependent decision in king penguins. Journal of Avian Biology 28: 264-267. Owen, M.; Cook, W. A. 1977: Variations in body weight, wing length and condition of mallards (Anas platyrhynchos) and their relationship to environmental changes. Journal of Zoology, London 183: 377-395. Perriman, L. 1997: Blue penguins (Eudyptula minor) at Taiaroa Head and the Otago Peninsula, 1993-95. Department of Conservation. Science for conservation No. 59. Perriman, L.; McKinlay, B. 1995: The blue penguin (Eudyptula minor) at Taiaroa Head, Otago, 1992-1993. Department of Conservation. Science & Research Series No. 86. Prince, P. A.; Ricketts, C; Thomas, G. 1981: Weight loss in incubating albatrosses and its implications for their energy and food requirements. Condor 83: 238-242. Reilly, P. N.; Cullen, J. M. 1981: The little penguin Eudyptula minor in Victoria. 2: Breeding. Emu 81: 1-19.

Numata et al. Duration of foraging and egg desertion in little penguins 289 Renner, M. 1998: Survival of little penguin chicks. MSc thesis, University of Otago, New Zealand. Renner, M.; Davis, L. S. 1999: Sexing little penguins Eudyptula minor from Cook Strait, New Zealand using discriminant function analysis. Emu 99: 74-79. Robertson, D. A. 1980: Hydrology and the quantitative distribution of planktonic eggs of some marine fishes of the Otago coast, south-eastern New Zealand. New Zealand Ministry of Agriculture and Fisheries. New Zealand Fisheries Research Bulletin vol. 21. Roff, D. A. 1992: The evolution of life histories. New York, Chapman and Hall. Rogers, C. M. 1991: An evaluation of the method of estimating body fat in birds by quantifying visible subcutaneous fat. Journal of Field Ornithology 62: 349-356. Rogers, T.; Eldershaw, G.; Walraven, E. 1995: Reproductive success of little penguins, Eudyptula minor, on Lion Island, New South Wales. Wildlife Research 22: 709-715. Smith, P. J.; Holdsworth, J.; Anderson, C; Hine, P. M; Allen, D.; Gibbs, W.; McKenzie, L.; Taylor, P. R.; Blackwell, R. H.; Williamson, S. H. 1996: Pilchard death in New Zealand, 1995. NIWA. New Zealand Fisheries data report No. 70. Sokal, R. R.; Rohlf, F. J. 1995: Biometry: the principles and practice of statistics in biological research. Third ed. New York, W. H. Freeman and Company. Stearns, S. C. 1992: The evolution of life histories. New York, Oxford University Press. Tveraa, T.; Lorentsen, S.; Saither, B. 1997: Regulation of foraging trips and costs of incubation shifts in the Antarctic petrel (Thalassoica antarctica). Behavioral Ecology 8: 465-469. Warhain, J. 1975: The crested penguins. In: Stonehouse, B. ed. The biology of penguins. London, Macmillan Press. Pp. 189-269. Warham, J. 1990: The petrels: their ecology and breeding systems. London, Academic Press. Weavers, B. W. 1992: Seasonal foraging ranges and travels at sea of little penguins Eudyptula minor, determined by radiotracking. Emu 91: 302-31 7. Weimerskirch, H. 1995: Regulation of foraging trips and incubation routine in male and female wandering albatrosses. Oecologia 102: 37-43. Williams, T. D. ed. 1995: The penguins: Spheniscidue. Oxford, Oxford University Press. Yorio, P.; Boersma, P. D. 1994: Causes of nest desertion during incubation in the Magellanic penguin (Spheniscus magellanicus). The Condor 96: 1076-1083.