Interrelationships between breeding frequency,timing and outcome in King Penguins Aptenodytes patagonicus: are King Penguins biennial breeders?

King Penguin Aptenodytes patagonicus chick growth is interrupted by a winter fast which extends the length of the breeding cycle (laying to chick independence) to 14–16 months, so that continuous annual successful breeding appears to be impossible. The 3–month laying period imposes further constraints with respect to timing of breeding attempts in relation to the onset of fasting. By the frequent resighting of individually marked birds at Marion Island, we examined the frequency of breeding and the relationships between timing, outcome and frequency in the same and in consecutive years. A total of 3101 adult King Penguins were banded between 1984 and 1991, yielding continuous breeding histories spanning a maximum of 5 years. Most penguins attempted to breed in consecutive years, although the likelihood of taking a year off increased with the number of consecutive attempts. In any one season, about 19% of potentially breeding adults did not breed. Early breeders were more likely to succeed than late breeders and comprised 84% of breeding attempts of known timing. Successful birds in one season usually bred late in the following season, whereas failed attempts were usually followed by an early attempt. Non-breeding was more likely to occur after a successful outcome than a failed one, and after a year off 93% of birds bred early. It is incorrect to refer to the King Penguin as a biennial breeder.     

The breeding biology of Guillemots Uria aalge on the Isle of May over a six year period

The breeding of Guillemots was studied in five areas of different breeding density and habitat type on the Isle of May in 1981-86. Prior to 1981 numbers were increasing at 5 6° per annum but during the study the rate of increase slowed down and from 1983 to 1986 numbers were fairly constant. Adult survival was high, with a mean minimum annual adult survival of 930% (s.e. = 03). Observations in 1986 suggested that the percentage return of colour-marked immature birds was low, with only l-6% and 5.5% of second and third year birds being seen. We suggest that poor recruitment was responsible for the levelling off in numbers at the colony.
The timing of laying was constant from year to year in 1981-85 but was later in 1986. It was significantly and inversely related to sea temperature the previous March. There was a consistent ranking in median laying dates amongst the areas, with area 1 (the highest density of birds) always earliest. However, there was no significant difference in synchrony between the areas. Overall breeding success was high (0–71-0-82 young fledged per pair). There was no consistent ranking of breeding success with breeding density, habitat type or laying synchrony.
The only aspect of Guillemot biology which changed significantly was the daily food intake of a chick which approximately halved during the study period. However, this reduction in food intake had no detectable effect on either the weight of chicks with wing lengths greater than 60 mm or the amount of time off-duty breeders spent at the site. Both of these parameters were still consistent with conditions being favourable in 1986.     

Winter distribution and behaviour of gentoo penguins<Emphasis Type="Italic"> Pygoscelis papua</Emphasis> at South Georgia

Knowledge of the spatial and temporal dynamics of foraging penguins is important to our understanding of the Southern Ocean marine ecosystem. We use satellite tracking to provide the first data on the distribution and behaviour of gentoo penguins (Pygoscelis papua) during the winter at South Georgia. Five penguins tracked from Bird Island remained close inshore, and although they did not return to the initial tagging site, they did appear to return to land each evening. They made diurnal trips to sea of similar distance from land as those during the breeding season, even though the constraints of chick rearing were absent. Despite potential greater flexibility in their responses to variations in prey availability in winter, the penguins still returned to land each night. This may reflect benefits from conserving energy by resting on land, possibly facilitating information exchange and avoiding predation. The distribution and behaviour of gentoo penguins during the winter enables efficient exploitation of a dynamic, patchy prey resource and may ultimately determine the timing of return to the colony, and onset of breeding in the following season.     

Annual variation in breeding biology of gentoo penguins, Pygoscelis papua, at Bird Island, South Georgia

The breeding biology of the gentoo penguin, Pygoscelis papua , was studied over a three-year period (1986–1988) at Bird Island, South Georgia, with particular reference to birds of known age or breeding experience. Laying date varied significantly between all three years, being three weeks later in 1987, when the breeding population decreased markedly. Factors involved in the timing of breeding are discussed. Within years egg-laying was highly synchronous: 95% of clutches were initiated in 14·5 days or less. The incubation period was 35 days and the laying interval, between the two eggs, 3·3–3·4 days. Chicks creched when 25–30 days old, and this varied between years, possibly related to food supply and chick growth. Chicks left the colony for the first time between 75 and 85 days of age. The breeding population at Bird Island decreased by 20% and increased by 84% in successive years during the study period. Breeding success (chicks fledged per egg laid) varied between 0·33 and 0·65 within colonies, but for the whole island was very consistent over the three years: 0·45, 0·51 and 0·47. Overall, colony differences were not correlated between years. Disturbance from Antarctic fur seals, Arctocephalus gazella , is suggested as the cause of consistently lower breeding success at one colony. Mean egg weight varied annually, and with age of the breeding bird, nest location and, in one year, with laying date. Young, first-time breeders laid smaller eggs and had lower breeding success compared to older, experienced birds, similar to other seabirds. However, they differed from other species in laying on average earlier than older birds. The relationship between age, egg weight, laying date and breeding success is discussed in relation to predation and seasonal food supply.     

Effect of breeding performance on the distribution and activity budgets of a predominantly resident population of black‐browed albatrosses

Pelagic seabirds breeding at high latitudes generally split their annual cycle between reproduction, migration, and wintering. During the breeding season, they are constrained in their foraging range due to reproduction while during winter months, and they often undertake long‐distance migrations. Black‐browed albatrosses (Thalassarche melanophris) nesting in the Falkland archipelago remain within 700 km from their breeding colonies all year‐round and can therefore be considered as resident. Accordingly, at‐sea activity patterns are expected to be adjusted to the absence of migration. Likewise, breeding performance is expected to affect foraging, flying, and floating activities, as failed individuals are relieved from reproduction earlier than successful ones. Using geolocators coupled with a saltwater immersion sensor, we detailed the spatial distribution and temporal dynamics of at‐sea activity budgets of successful and failed breeding black‐browed albatrosses nesting in New Island, Falklands archipelago, over the breeding and subsequent nonbreeding season. The 90% monthly kernel distribution of failed and successful breeders suggested no spatial segregation. Both groups followed the same dynamics of foraging effort both during daylight and darkness all year, except during chick‐rearing, when successful breeders foraged more intensively. Failed and successful breeders started decreasing flying activities during daylight at the same time, 2–3 weeks after hatching period, but failed breeders reached their maximum floating activity during late chick‐rearing, 2 months before successful breeders. Moon cycle had a significant effect on activity budgets during darkness, with individuals generally more active during full moon. Our results highlight that successful breeders buffer potential reproductive costs during the nonbreeding season, and this provides a better understanding of how individuals adjust their spatial distribution and activity budgets according to their breeding performance in absence of migration.     

Foraging strategy of the little auk Alle alle throughout breeding season – switch from unimodal to bimodal pattern

Energy and time allocation differs between incubation and chick‐rearing periods, which may lead to an adjustment in the foraging behaviour of parent birds. Here, we investigated the foraging behaviour of a small alcid, the little auk Alle alle during incubation and compared it with the chick‐rearing period in West Spitsbergen, using the miniature GPS (in Hornsund) and temperature loggers (in Magdalenefjorden). GPS‐tracking of 11 individuals revealed that during incubation little auks foraged 8–55 (median 46) km from the colony covering 19–239 (median 120) km during one foraging trip. Distance from the colony to foraging areas was similar during incubation and chick‐rearing period. During incubation 89% of foraging positions were located in the zone over shallower parts of the shelf (isobaths up to 200–300 m) with sea surface temperature below 2.5°C. Those environmental conditions are preferred by Arctic zooplankton community. Thus, little auks in the Hornsund area restrict their foraging (both during the incubation and chick‐rearing period) to the area under influence of cold, Arctic‐origin water masses where its most preferred prey, copepod Calanus glacialis is most abundant. The temperature logger data (from 4 individuals) indicate that in contrast to the chick‐rearing period, when parent birds alternated short and long trips, during the incubation they performed only long trips. Adopting such a flexible foraging strategy allows little auks to alter their foraging strategy to meet different energy and time demands during the two main stages of the breeding.     

Strong temporal consistency in the individual foraging behaviour of Imperial Shags Phalacrocorax atriceps

Individual consistency in foraging behaviour can generate behavioural variability within populations and may, ultimately, lead to species diversification. However, individual‐based long‐term behavioural studies are particularly scarce in seabird species. Between 2008 and 2011, breeding Imperial Shags Phalacrocorax atriceps at the Punta León colony, Argentina, were tracked with GPS devices to evaluate behavioural consistency during their foraging trips. Within a breeding season, individuals were highly consistent in the maximum distances they reached from the shore and the colony, as well as in the time invested in flight and diving across consecutive days during early chick rearing. In addition, each individual had its specific foraging area distinct from the foraging area of other individuals. Comparing between early and late chick rearing in the same season, individuals were consistent, to a lesser degree, in the maximum distance they reached from the colony and the shore, increasing in consistency later on in the season. Within the season, females were more consistent than males in the maximum distance they moved from the colony and the shore, the sexes segregated in their foraging areas and individual females were segregated from one another. Twenty‐eight individuals tracked in different breeding seasons were marginally consistent in their trip durations and maximum distance reached from shore across seasons. Among seasons, foraging locations differed between sexes and among individual females. Individuals from this colony exhibited consistency over time in several aspects of foraging behaviour, which may be due to a combination of individual characteristics such as learning abilities, breeding experience or health, as well as targeted prey type and stability of the environment at this location.     

Stage-dependent foraging in breeding black-legged kittiwakes Rissa tridactyla: distinguishing behavioural responses to intrinsic and extrinsic factors

Different phases of the annual cycle in birds and mammals are often associated with characteristic and recurrent foraging behaviours. The extent to which stage‐dependent changes in foraging behaviour are caused by intrinsic or extrinsic factors is unclear. We controlled for the effects of extrinsic factors by synchronising groups of incubating and chick‐rearing black‐legged kittiwakes Rissa tridactyla. Synchrony amongst incubators and rearers was achieved experimentally by switching eggs between nests. Behavioural responses to the treatment varied between the sexes. Male kittiwakes with prolonged‐incubation made fewer foraging trips but of greater duration compared to those rearing chicks resulting in no change in the time spent on trips between the two groups. Females with prolonged‐incubation carried out fewer trips than those rearing chicks but trip duration did not differ between the two stages which resulted in prolonged‐incubating birds spending a lower percentage time on trips. In contrast, foraging ranges did not differ between prolonged‐incubation and chick‐rearing birds for either sex. This suggests that extrinsic factors, such as food availability and distribution determine kittiwake foraging locations and ranges, whereas intrinsic factors, reflected in parental duties, constrain nest attendance. Female prolonged‐incubators invested lower levels of parental effort, in terms of daily energy expenditure, compared to chick‐rearers whereas males did not show stage‐related differences in energy expenditure. This provides evidence that incubation could be an energetically cheaper stage although under normal conditions this difference may be masked by temporal variation in environmental factors. We conclude that while conditions differ between the incubation and chick rearing stages for kittiwakes at this colony, they are not the main factors prompting changes in stage‐related foraging patterns. Intrinsic factors such as sex differences, or behaviours required for each stage of the annual cycle, rather than extrinsic factors related to seasonal environments, are likely to be the main proximate cause of recurring changes in behaviour between breeding stages.     

Contrasting bottom-up effects of warming ocean on two king penguin populations

Breeding success is often correlated with climate, but the underlying bottom-up mechanisms remain elusive—particularly in marine environments. Consequently, conservation plans of many species often consider climate change as a unilateral threat, ignoring that even nearby populations can show contradicting trends with climate. Better understanding the relationship between climate and environment at different scales can help us interpret local differences in population trends, ultimately providing better tools to evaluate the global response of a species to threats such as global warming. We studied a growing king penguin population nesting at Kerguelen island (Southern Indian Ocean), hosting one of the largest colonies in the world. We used a unique dataset of foraging, breeding success, and climate data spanning over 25 years to examine the links between climate, marine environment, and breeding success at this colony. The results were then compared to the neighboring population of Crozet, which experienced the steepest decline for this species over the past few decades. At Crozet, penguins experienced lower breeding success in warmer years due to productive currents shifting away from the colony, affecting foraging behavior during chick rearing. At Kerguelen, while chick mass and survival experienced extreme variation from year to year, the annual variation was not associated with the position of the currents, which varied very little compared to the situation in Crozet. Rather than being affected by prey distribution shifts, we found evidence that chick provisioning in Kerguelen might be influenced by prey abundance, which seem to rather increase in warmer conditions. Furthermore, warmer air temperature in winter increased chick survival rate, likely due to reduced thermoregulation cost. Investigating the mechanisms between climate and fitness allowed us to predict two different fates for these populations regarding ongoing global warming.     

Parental feeding in the Red-throated Diver Gavia stellata

We describe parental feeding by Red-throated Divers Gavia stellata from 14.2–44.7 h of observation on each of eight pairs, coordinated with simultaneous observations at potential fishing lakes, in southwest Sweden. All pairs bred on small tarns (0.8–10 ha), normally devoid offish, and foraged in larger freshwater lakes 1.9–7.7 km away from the breeding site. Most pairs directed the majority of the foraging trips to the same lake. On average, parents made seven foraging trips per day per young. Both parents contributed to feeding activities and, in broods of two young, each chick obtained an equal amount of food. Cyprinid fish (mainly Rutilus rutilus ) and salmonids (mainly Coregonus albula ) dominated. Qualitative predictions from central-place foraging theory concerning relationships between the time for a foraging trip and the size of prey delivered by a single-prey loader were not corroborated. The amounts of fish biomass delivered were in agreement with or lower than predictions from models on daily energy requirements; this result is related to the decreased abundance of fish in acidified freshwater lakes and the decline of Red-throated Diver populations in south Sweden.     

The effects of food abundance on breeding performance and adult time budgets of Guillemots Uria aalge

The breeding performance, food fed to chicks and adult time budgets of Guillemots Uria aalge were examined in a year of high and a year of low food availabiIity. There was no difference between the 2 years in reproductive success, although the rate of chick feeding, chick weight and fledging success were greater in the year of high food availability. On average, chick prey items were larger in the poor food year, but this was insufficient to compensate for the lower feeding frequency. Chick feeding frequency did not differ between days in the good year but did increase later in the season in the poor food year. Compared with the high food availability year, adult Guillemots in the year of low food availability spent much less time resting at the breeding colony. and their foraging trips were twice as long. Foraging birds tended to make several successive trips before resuming brooding duties from their mates when food supplies were good, but in the low food availability year single trips were the norm. These results demonstrate that predators experiencing reduced food supply may mitigate the effects on their reproductive output by shifting their time allocation such that more time is available for foraging.     

Breeding ecology of the gentoo penguin <Emphasis Type="Italic">Pygoscelis papua</Emphasis> at Kerguelen Archipelago

The gentoo penguin (Pygoscelis papua) is one of the most widespread penguin species and has been proven to be highly plastic in many aspects of its ecology. However, data from their sub-Antarctic range suggest an unexplained decline of their populations over the last 10–20 years, stressing the need for additional knowledge on their breeding ecology and demography. The present study provides insights into the breeding ecology of the gentoo penguin at a major breeding site, Kerguelen Archipelago, over three breeding seasons (1987, 2002 and 2003). Similarly to other northern populations, gentoo penguins breeding at Kerguelen exhibited winter laying, slow provisioning rate, slow growth rate associated with an extended rearing period and relatively low breeding success compared to southern populations. Our study also revealed interannual differences in the timing of laying and growth parameters as well as unusual sex differences in parental investment. Despite their high plasticity, there are indications that gentoo penguins at the northern edge of their range might work at the upper limit of their capacities. Sub-Antarctic populations would, therefore, be more sensitive to environmental changes than more southerly ones and need to be closely monitored.     

Diets and feeding of Fulmars Fulmarus glacialis during the breeding season:a comparison between St Kilda and Shetland colonies

Diets and nest attendance of Fulmars at Foula, Shetland and St Kilda, Outer Hebrides were examined during the breeding season. At Foula, sandeels formed the bulk of the diet, with fish offal and pelagic zooplankton of minor importance. At St Kilda, 71% of regurgitates consisted of pelagic zooplankton, probably captured at night. Dietary overlap between the two colonies was 14% by species composition. Foraging trips from Foula shortly after chick hatching generally lasted for less than 10 h, while trips from St Kilda late in chick rearing often lasted more than 24 h. The diurnal pattern of feeding suggested mainly nocturnal foraging from St Kilda but both diurnal and nocturnal foraging from Foula.     

Arthropods in diet of Little Bustards Tetrax tetrax during the breeding season in western France

Capsule Seasonal, age- and sex-related variations in the diet were studied during March to October.

Aims To determine which arthropod orders contributed most during the laying–rearing period and those important in determining breeding success of the species in an intensive agricultural habitat.

Methods Faeces (n = 388, 345 g) and three gut samples were dissected to determine which arthropod orders contributed most to the diet, especially during the laying–rearing period.

Results Coleoptera were the most numerous prey followed by Orthoptera, Dermaptera and Hymenoptera. No difference was observed in the diet of adult males and females. Although adults ate arthropods throughout the breeding season, plant material contributed 97–99% of faecal contents by weight in each month. Chicks, however, only ate arthropods, notably Coleoptera and Orthoptera, at least until 2–3 weeks old.

Conclusion Increasing arthropod availability would be a useful management tool for maintaining endangered Little Bustard populations by potentially increasing chick survival.     

Endocrine correlates of parental care in an Antarctic winter breeding seabird, the emperor penguin, Aptenodytes forsteri

Plasma levels of luteinizing hormone (LH) and prolactin associated with parental behavior were measured in the Antarctic winter breeding emperor penguin, Aptenodytes forsteri. Males exclusively incubate the egg while females exclusively brood the nonhomeothermic young; both sexes alternate in rearing the homeothermic young. Birds were sampled on arrival from the sea through egg laying, incubation, and brooding. All parent birds lost their chicks at the end of the brooding period due to harsh weather but sampling continued. In females, LH titers dropped after egg laying but levels were restored when the birds returned from the sea to brood the chicks and were not depressed by high prolactin levels. Plasma prolactin remained low in males captured on arrival and kept until the free-living males finished incubation. In breeders, prolactin secretion increased during the prelaying period when day length decreased. Prolactin levels stayed elevated in males during incubation and in brooding females returning after a 2-month absence at sea. Prolactin values were higher in brooding females than in males ending incubation or returning in late brooding. These levels did not drop after chick loss, and the sexual difference in prolactin values was maintained after breeding failure. In emperor penguins, increased prolactin secretion appears to be triggered around the time of egg laying and continues, driven by an endogenous mechanism, through incubation and brooding until rearing is completed. Prolactin secretion independent of external stimuli may have evolved in pelagic seabirds to maintain parental care despite long absences at sea from the breeding colony.     

Does the foraging strategy of adult Short‐tailed Shearwaters cause obesity in their chicks?

The chick provisioning behaviour of Short-tailed Shearwaters Puffinus tenuirostris breeding at the northern edge of their distribution on Montague Island, New South Wales, was examined in February and March 1997. The duration of individual foraging trips of parents, weight changes of adults and chicks, and meal sizes delivered to chicks were determined. It was found that individual parents mixed a long foraging trip to Antarctic waters (14.4±2.0 days) with one to three short foraging trips (1.36±0.7 days, mode=1 day). Adults gained body mass on long trips and lost weight on short trips. The size of meals fed to the chicks was significantly greater after a long trip (161±21 g) than after a short trip (135±28 g), although short trips increased the overall chick feeding frequency. The variable number of short trips made by adult Short-tailed Shearwaters and the relationship between short trips and adult body condition were consistent with current life-history theory: adults do not sacrifice their own body condition to increase food delivered to their chicks.
Modelling revealed that this dual foraging strategy inevitable leads to chicks enduring long intervals between meals. These long intervals may have led to the evolution of an over-feeding strategy by parents and the nestling obesity reported in this shearwater. The durations of the long trips from Montague Island were significantly greater than those for Short-tailed Shearwaters breeding at the centre of their distribution in Tasmania, although there was no significant difference in the length of short trips. A commitment to feed regularly in Antarctic waters may explain why the breeding distribution of this species does not extend much further north.     

Time and travelling costs during chick‐rearing in relation to habitat quality in Little Owls Athene noctua

In many bird species, parents adjust their home‐ranges during chick‐rearing to the availability and distribution of food resources, balancing the benefits of energy intake against the costs of travelling. Over recent decades, European agricultural landscapes have changed radically, resulting in the degradation of habitats and reductions in food resources for farmland birds. Lower foraging success and longer foraging trip distances that result from these changes are often assumed to reduce the reproductive performance of parents, although the mechanisms are not well understood. We tested the behavioural response of chick‐rearing Little Owls Athene noctua to variation in habitat diversity in an agricultural landscape. We equipped females with GPS loggers and received adequate range‐use data for 19 individuals (6063–14 439 locations per bird). In habitats dominated by homogeneous cropland habitats, home‐ranges were over 12 ha in size, whereas in highly diverse habitats they were below 2 ha. Large home‐ranges were associated with increased flight activity (117% of that of birds in small home‐ranges) and distances travelled per night (152%), increased duration of foraging trips (169%) covering larger distances (246%), and reduced nest visiting rates (81%). The study therefore provides strong correlative evidence that Little Owls breeding in monotonous farmland habitats expend more time and energy for a lower benefit in terms of feeding rates than do birds in more heterogeneous landscapes. As nestling food supply is the main determinant of chick survival, these results suggest a strong impact of farmland characteristics on local demographic rates. We suggest that preserving and creating islands of high habitat diversity within uniform open agricultural landscapes should be a key target in the conservation of Little Owl populations.     

Simultaneous pituitary-gonadal recrudescence in two Corsican populations of male blue tits with asynchronous breeding dates

Animal populations living in geographically variable environments respond to different selection pressures. The adaptive character of the responses to environmental information determines the degree of synchrony of the breeding period with local optimal conditions. An example is provided by two populations of Mediterranean blue tits (Parus caeruleus) in Corsica, breeding in different habitats, with a 1-month difference in the onset of egg laying. This difference in the onset of lay is supposed to be adaptive because, although chicks from both populations are raised mostly on caterpillars, the timing of the appearance of caterpillars is earlier for populations of tits associated with deciduous oak trees than those associated with evergreen oak trees. Here, we show that, despite the difference in the timing of egg laying, males from these two populations start seasonal hypothalamo-hypophysial-testicular development at approximately the same time, in late winter. Specifically, the vernal recrudescence of brain GnRH-I perikarya and fibers, testes volume and song activity began around the same dates and proceeded at the same pace in late winter in both populations. Plasma testosterone and LH levels displayed seasonal variations that were shifted by less than 2 weeks compared to the 1-month difference in egg laying periods. We hypothesize that the strong selection pressures on these two populations to adapt the timing of their breeding seasons to their local environment may have acted mostly on the female egg laying dates, and not so much on the initiation and rate of seasonal recrudescence of the hypothalamo-hypophysial-testicular activity in males.     

Factors determining the laying date of the Waved Albatross Diomedea irrorata

The Waved Albatross laying period is between mid-April and the end of June. This paper investigates the factors determining when, during this period, a female lays. It is shown that birds breeding for the first time lay relatively late. For older birds, the laying date depends on date and outcome of the previous year's attempt: females that bred successfully in the previous year lay slightly later than in that year, while birds that failed lay earlier. A corresponding difference exists in the dates of arrival at the breeding grounds of successful and failed males. It is argued that one calendar year is not quite sufficient for pairs that have reared a chick successfully to regain breeding condition, while failed breeders can choose the laying date more freely.
Under catastrophic El Niño weather conditions in 1983, all albatrosses arrived late at the breeding grounds, but still the birds that had failed in 1982 were clearly earlier than those that had reared a chick.     

Inter-annual variation in provisioning behaviour of Southern Rockhopper Penguins Eudyptes chrysocome chrysocome at Staten Island, Argentina

Understanding the relationship between reproductive performance and food availability requires knowledge about many different variables, including such factors as the length of incubation shifts, provisioning rates and patterns, as well as how variability in these factors affects reproductive output. To examine some of the most important aspects of parental investment, we studied the provisioning behaviour and patterns of adult Southern Rockhopper Penguins Eudyptes chrysocome chrysocome breeding at Staten Island, Argentina. We investigated foraging trip duration, provisioning rates and chick survival using adult foraging patterns. Our results show that Rockhopper Penguins had clear sex-specific differences in their provisioning behaviour. Females provision chicks throughout chick rearing. By contrast, males provision chicks only during the crèche stage and at a slightly lower rate than females during this period. Foraging trips increased in length as the breeding season progressed. Rockhopper Penguins from Staten Island performed longer trips throughout the breeding season than do other species of Eudyptes at several other locations. Our results also show differences in parental investment between years that were related to differences in chick survival. We suggest that this was most likely to be related to female rather than male foraging behaviour as only females showed inter-annual differences in their provisioning rates.