Climate change might affect marine top predators by altering availability and nutritional quality of their prey. Climate effects vary on a regional basis, and our understanding of the relationships between fluctuations in climate and food resources in sub-arctic regions with seasonal ice cover is limited. We studied the effects of inter-annual climate variability (as reflected in the Pacific Decadal Oscillation, PDO, and the timing of the winter ice retreat) on zooplankton-planktivorous predator linkages in one of the most productive regions of the southeastern Bering Sea, the 'Green Belt'. We examined changes in diets (species composition of chick meals and stable isotope signatures of adult blood) and relative food availability (as reflected in blood plasma concentrations of the stress hormone corticosterone) of planktivorous least auklets Aethia pusilla breeding on St. George I. (shelf-break) and St. Paul I. (shelf) during 2003 to 2005 and 2008 to 2009. We found that isotopic signatures of blood and composition of chick meals differed between the colonies. The proportion of energy-rich oceanic copepods Neocalanus spp. in the diet declined during warm years (high PDO and early ice retreat) on St. Paul but not on St. George. However, inter-annual and seasonal dynamics of corticosterone were similar between the colonies, and auklets experienced higher nutritional stress during warm compared to cold years. Our results suggest that the influx of prey-bearing water masses from the ocean basin and the retention time of oceanic copepods on the shelf are the main factors affecting composition of prey and its availability to auklets. We conclude that anticipated climate warming will negatively affect food availability of planktivorous predators in the Green Belt region of the southeastern Bering Sea.

Summary 1. Life‐history theory predicts a trade‐off between costs of current reproduction and future survival of individuals. Studies of short‐lived animals in general support this prediction. However, the effect of nutritional stress during reproduction on survival of long‐lived animals is poorly understood. 2. We examined the link between nutritional stress, fecundity and return to a breeding colony (hereafter ‘survival’) of black‐legged kittiwakes (Rissa tridactyla) at two colonies with contrasting patterns in adult survival, fecundity, and numerical trends. 3. We tested the observational (at Duck and Gull Is., Cook Inlet, Northern Gulf of Alaska) and experimental (at Middleton I., Gulf of Alaska) relationships between variations in the secretion of the stress hormone corticosterone (CORT) and food abundance. Then, we examined the relationships between nutritional stress (as reflected in CORT), reproduction, and survival of individuals. 4. On average, CORT was higher in kittiwakes breeding on Duck I. (declining, low fecundity, high survival) compared to those breeding on Gull I. (increasing, high fecundity, low survival). 5. At both colonies, CORT was directly negatively correlated with food abundance quantified at sea. Experimental feeding of individuals ad libitum resulted in a reduction of CORT in birds breeding on Middleton I. These results suggest that CORT is a reliable measure of food availability and defines nutritional stress (stress) in kittiwakes. 6. On Gull I., where survival is low (86%), production of young declined as stress increased. On Duck I., where survival is high (93%), parents always failed in raising young, though they experienced a wide range of stress levels. 7. Survival of individuals is linked to their CORT levels during reproduction. High levels of CORT predicted disappearance of individuals from both colonies. 8. The results support the hypothesis that nutritional stress during reproduction affects both survival and reproduction in long‐lived animals. However, even within a species the ways in which survival and reproduction trade‐off against each other may vary among populations. Results suggest that reproductive consequences of nutritional stress might differ between declining and increasing populations, which should be tested. We conclude that severity of nutritional stress during reproduction is one of the major factors defining population processes in kittiwakes.

Seabirds are thought to time breeding to match the seasonal peak of food availability with peak chick energetic demands, but warming ocean temperatures have altered the timing of spring events, creating the potential for mismatches. The resilience of seabird populations to climate change depends on their ability to anticipate changes in the timing and magnitude of peak food availability and 'fine-tune' efforts to match ('Anticipation Hypothesis'). The degree that inter-annual variation in seabird timing of breeding and reproductive performance represents anticipated food availability versus energetic constraints ('Constraint Hypothesis') is poorly understood. We examined the relative merits of the Constraint and Anticipation Hypotheses by testing 2 predictions of the Constraint Hypothesis: (1) seabird timing of breeding is related to food availability prior to egg laying rather than the date of peak food availability, (2) initial reproductive output (e.g. laying success, clutch size) is related to pre-lay food availability rather than anticipated chick-rearing food availability. We analyzed breeding biology data of common murres Uria aalge and black-legged kittiwakes Rissa tridactyla and 2 proxies of the seasonal dynamics of their food availability (near-shore forage fish abundance and sea-surface temperature) at 2 colonies in Lower Cook Inlet, Alaska, USA, from 1996 to 1999. Our results support the Constraint Hypothesis: (1) for both species, egg laying was later in years with warmer sea-surface temperature and lower food availability prior to egg laying, but was not related to the date of peak food availability, (2) pre-egg laying food availability explained variation in kittiwake laying success and clutch size. Murre reproductive success was best explained by food availability during chick rearing.

Apex predators can provide valuable information about effects of climate variability on trophodynamics in the Bering Sea. We used corticosterone (the primary avian stress hormone, “CORT”) as a proxy of changes in prey availability for planktivorous and piscivorous seabirds. CORT secretion reflects energy balance in breeding individuals and can be used to monitor changes in the marine environment that alter food availability. We tested whether CORT in planktivorous least auklets (Aethia pusilla) and piscivorous thick-billed murres (Uria lomvia) on two Pribilof Islands differed spatially or temporally in 2003 and 2004. During June–September of each year, we sampled birds breeding on St. Paul and St. George Islands. We found that seasonal dynamics of CORT varied between years. Although the seasonal dynamics were similar between islands in a year, in 2004 there were inter-island differences in CORT levels for both species. In 2003, CORT in murres was low throughout the season, suggesting that their prey availability (primarily forage fishes) was consistently good. In 2004, CORT levels suggested that overall food availability was poor for murres and declined as the season progressed. For auklets, inter-annual variability was associated with contrasting intra-seasonal patterns in the availability of zooplankton prey (primarily Neocalanus spp.). In 2003, high early-season (June) CORT indicated that auklets were food-limited, but CORT decreased in July, suggesting that the availability of zooplankton improved as the season progressed. In 2004, CORT was low early in the season (June), but an abrupt increase late in the season (July) suggests that food became scarce. Our data indicate that environmental differences between 2003 and 2004 affected prey availability for planktivorous and piscivorous alcids in opposite ways. These results suggest that in the shelf regions of the Bering Sea, the populations of apex predators feeding on zooplankton may be affected by environmental fluctuations differently than those feeding on forage fish.

Linking diving and foraging behavior of small seabirds with the fine-scale characteristics of water masses has been challenging largely due to sampling constraints. We examined the diving behavior of 12 chick-rearing thick-billed murres (Uria lomvia) at St. George Island, southeastern Bering Sea, in relation to sea-surface temperature (SST) and thermocline depth using ventrally attached depth–temperature–acceleration data loggers. Our results from summer 2004 showed that murres swam in water masses ranging from well-mixed (SST 7–9°C, estimated distance of 14km from the breeding colony) to well-stratified (SST 9–12°C, estimated distance of 30–50km). Murres dove deeper (modal depth: 60–70m) in the mixed water mass than in the stratified water, where most dives were to just below the thermocline depth (modal depth: 20–30m). We suggest that the thermocline is important in shaping dive profiles of thick-billed murres, possibly through its effect on the vertical distribution of both zooplankton and fish prey.

Received 7 January 2008, accepted 27 August 2008 SUMMARY JODICE, P.G.R., ROBY, D.D., TURCO, K.R., SURYAN, R.M., IRONS, D.B., PIATT, J.F., SHULTZ, M.T., ROSENEAU, D.G. & KETTLE, A.B. 2008. Growth rates of Black-legged Kittiwake Rissa tridactyla chicks in relation to delivery rate, size, and energy density of meals. Marine Ornithology 36: 107–114. The growth rate of chicks is a parameter often used to examine the effects of changes in food availability on reproductive performance of seabirds. We analyzed the relationship of growth rates in Black-legged Kittiwake Rissa tridactyla chicks at three colonies in Prince William Sound, Alaska, and three colonies in Lower Cook Inlet, Alaska, to two components of nestling meals that represent diet quantity (meal delivery rate and meal size) and one that represents diet quality (energy density). During this study, kittiwakes appeared to rely on an array of relatively high-lipid prey. Growth rates of alpha and singleton chicks were positively correlated with annual productivity and were best explained by meal delivery rate and meal size; growth rates of beta chicks were not correlated with productivity and were best explained by meal delivery rate alone. The lack of a positive effect of energy density on growth rates was in contrast to results from a companion study that found a moderately positive effect of energy density on kittiwake productivity. Taken together, results from these two studies demonstrate the need to measure both diet quantity and quality to reliably assess the relationships between diet composition, provisioning rates, growth rates, and productivity in seabirds.

Flexible time budgets allow individual animals to buffer the effects of variable food availability by allocating more time to foraging when food density decreases. This trait should be especially important for marine predators that forage on patchy and ephemeral food resources. We examined flexible time allocation by a long-lived marine predator, the Common Murre (Uria aalge), using data collected in a five-year study at three colonies in Alaska (USA) with contrasting environmental conditions. Annual hydroacoustic surveys revealed an order-of-magnitude variation in food density among the 15 colony-years of study. We used data on parental time budgets and local prey density to test predictions from two hypotheses: Hypothesis A, the colony attendance of seabirds varies nonlinearly with food density; and Hypothesis B, flexible time allocation of parent murres buffers chicks against variable food availability. Hypothesis A was supported; colony attendance by murres was positively correlated with food over a limited range of poor-to-moderate food densities, but independent of food over a broader range of higher densities. This is the first empirical evidence for a nonlinear response of a marine predator's time budget to changes in prey density. Predictions from Hypothesis B were largely supported: (1) chick-feeding rates were fairly constant over a wide range of densities and only dropped below 3.5 meals per day at the low end of prey density, and (2) there was a nonlinear relationship between chick-feeding rates and time spent at the colony, with chick-feeding rates only declining after time at the colony by the nonbrooding parent was reduced to a minimum. The ability of parents to adjust their foraging time by more than 2 h/d explains why they were able to maintain chick-feeding rates of more than 3.5 meals/d across a 10-fold range in local food density.

In his seminal paper about using seabirds as indicators of marine food supplies, Cairns (1987, Biol Oceanogr 5:261-271) predicted that (1) parameters of seabird biology and behavior would vary in curvilinear fashion with changes in food supply, (2) the threshold of prey density over which birds responded would be different for each parameter, and (3) different seabird species would respond differently to variation in food availability depending on foraging behavior and ability to adjust time budgets. We tested these predictions using data collected at colonies of common murre Uria aalge and black-legged kittiwake Rissa tridactyla in Cook Inlet, Alaska. (1) Of 22 seabird responses fitted with linear and non-linear functions, 16 responses exhibited significant curvilinear shapes, and Akaike's information criterion (AIC) analysis indicated that curvilinear functions provided the best-fitting model for 12 of those. (2) However, there were few differences among parameters in their threshold to prey density, presumably because most responses ultimately depend upon a single threshold for prey acquisition at sea. (3) There were similarities and some differences in how species responded to variability in prey density. Both murres and kittiwakes minimized variability (CV < 15%) in their own body condition and growth of chicks in the face of high annual variability (CV = 69%) in local prey density. Whereas kittiwake breeding success (CV = 63%, r(2) = 0.89) reflected prey variability, murre breeding success did not (CV = 29%, r(2) < 6.00). It appears that murres were able to buffer breeding success by reallocating discretionary 'loafing' time to foraging effort in response (r(2) = 0.64) to declining prey density. Kittiwakes had little or no discretionary time, so fledging success was a more direct function of local prey density. Implications of these results for using 'seabirds as indicators' are discussed.

Food availability comprises a complex interaction of factors that integrates abundance, taxonomic composition, accessibility, and quality of the prey base. The relationship between food availability and reproductive performance can be assessed via the nutritional stress (NSH) and junkfood (JFH) hypotheses. With respect to reproductive success, NSH posits that a deficiency in any of the aforementioned metrics can have a deleterious effect on a population via poor reproductive success. JFH, a component of NSH, posits specifically that it is a decline in the quality of food (i.e. energy density and lipid content) that leads to poor reproductive success. We assessed each in relation to reproductive success in a piscivorous seabird, the black-legged kittiwake Rissa tridactyla. We measured productivity, taxonomic composition, frequency, size, and quality of meals delivered to nestlings from 1996 to 1999 at 6 colonies in Alaska, USA, 3 each in Prince William Sound and Lower Cook Inlet. Productivity varied widely among colony-years. Pacific herring Clupea pallasi, sand lance Ammodytes hexapterus, and capelin Mallotus villosus comprised ca. 80 % of the diet among colonyyears, and each was characterized by relatively high energy density. Diet quality for kittiwakes in this region therefore remained uniformly high during this study. Meal delivery rate and meal size were quite variable among colony-years, however, and best explained the variability in productivity. Parent kittiwakes appeared to select prey that were energy dense and that maximized the biomass provisioned to broods. While these results fail to support JFH, they do provide substantial support for NSH.