Our study explored parent-offspring allocation strategies in Leach’s storm-petrels, both by describing natural variation in parental feeding behavior and by investigating how manipulating chick access to food affected parental feeding behavior. In the unmanipulated year, we found that chicks were fed by a single parent about half of the nights, by both of the parents a quarter of the nights, and not fed by either parent on the remaining quarter of nights. However, in the manipulated year when chicks had their access to food limited by half, while the proportion of single parent feed nights remained similar to the unmanipulated year, the proportion of double parent feed nights increased substantially and the proportion of nights that no parents fed fell close to zero. Supplementing the chicks with additional food did not affect parental feeding behavior. These data provides strong support for a flexible-level allocation strategy in this long-lived seabird.
While parents of restricted chicks altered their feeding behavior by increasing the frequency of their feeding visits, there was no change to the size of the meal delivered. Specifically, the size of meals delivered to chicks did not differ between the unmanipulated year, and the manipulated year among the three treatment groups. Taken together, while the parents of restricted chicks did not provide larger individual meals, by altering their feeding frequency, they provided a larger total amount of food to the chicks over the course of the manipulation. Other studies in pelagic seabirds agree with our finding of a flexible-level feeding strategy (4, 7, 25–27) For example, a study exploring natural variation in Dovekie (Alle alle) feeding behavior found that parents changed the number of feeding visits they made depending on food quality and abundance (16). Another study which used a manipulation to restrict the amount of food Scopoli's shearwater (Calonectris diomedea) chicks received over a six day period reported that parents began delivering larger quantities of food to these restricted chicks (11).
While our results agree with other papers which found support for a flexible-level parental allocation strategy, they do conflict with another group of studies which support the fixed-level parental allocation strategy (10, 14, 22, 28). Interestingly, some of these other studies were done on the same population of Leach’s storm-petrels nesting on Kent Island. For example, Ricklefs (22), who performed a similar restriction manipulation to the current study, found that storm-petrel parents did not respond to chick undernourishment by increasing feeding rate. And, in another study on Scopoli's Shearwater (Calonectris diomedea), Hammer and Hill (28) reported that neither meal size nor feeding frequency were related to chick body condition.
One possible way to reconcile these two seemingly conflicting bodies of literature is that both fixed and flexible-level allocation strategies exist, but within populations these strategies may shift back-and-forth in a context-dependent fashion (7). Seabirds live in an unpredictable environment which can vary widely both within (29–31) and among years (16, 32). This uncertain environment may favor breeding strategies where parental allocation can change and adjust according to the prevailing environmental conditions and food availability (8). Interestingly, studies that focused on shorter periods of parental feeding behavior, or for only one season were more likely to find support for a fixed-level allocation strategy. While studies that were across seasons were more likely to find support for a flexible-level allocation strategy. Given the stochastic nature of seabird environments, studies focusing on a longer period may be more likely to capture that variability, and thereby a shift in parental feeding strategies. Unlike previous studies that followed parental feeding behavior for relatively short periods of one to four days (11, 26, 27, 33), or for up to a week (4, 16, 22, 28, 34), our study period lasted for 38 days in each of the two seasons. Interestingly, 2006 had a higher proportion of single feed nights, while 2007 had a higher proportion of double feed nights. This suggests a potential difference in food quality or abundance between the two years which resulted in parents altering their feeding behavior to feed more often in 2007. This is not surprising given the broad interannual variation in food quality in marine systems in general, and in our Leach’s storm-petrel study system in particular (35). The increase in nights where both parents visited the burrow suggests that 2007 may have been a more challenging year compared to 2006. In agreement with this, chicks had larger 24-hour mass losses on days after a night where no parent visited the burrow to feed chicks, suggesting those chicks may have had to use a larger proportion of their food reserves to fuel their metabolism.
Our finding that parents modified feeding frequency, but not meal size is reported in other studies as well (16, 36). This suggests that parents may be feeding until their food payload reaches a physical limit (10). Alternatively, parents may only be willing to feed chicks so much of their food store to protect their own nutritional demands. Leach’s storm-petrels, like other Procellariiforms, have high adult survival, and the single chick reared each year represents a small portion of their lifetime reproductive success, so parents prioritize investment to safeguard themselves over the needs of their chicks (14). For example, a study on Antarctic Petrels (Thalassoica antarctica) showed parents who had a poor body condition fed less food to offspring food (4). In another example, a recent study on Little Auks (Alle alle) by Kidwa et al (33)) found that parents dosed with corticosterone, a hormone that is reliable indicator of poor condition (37), would feed their chicks less food compared to parents with unmanipulated corticosterone levels (33). Taken together, this suggests that Procellariiform parents base decisions on allocations to their offspring by assessing their needs. In support of this, studies that supplemented chicks daily with large amounts of food found that parents delivered less food to their offspring and, in turn, kept more for themselves (24, 34). While we did not find this in our study, the food supplemented to chicks, was only 2g, and more so, chicks were only supplemented following nights that parents did not feed. This was intended to ensure no mass loss occurred, but this limited amount does not appear to have been enough to alter parental feeding behavior.
Because the amount of food an adult can carry is limited, if parents shift to a more flexible-allocation strategy by feeding more frequently they can increase food delivery to chicks while still keeping the same proportion of the foraged food for themselves. However, over the course of a foraging bout, parents must also use some of the food they gather while foraging to fuel the demanding cost of flight (38). Thus, the food hauled back to their offspring does not represent the total sum of the food gathered during the foraging trip, and during longer foraging trips a larger total amount of food is gathered (39). Following this, other studies in Procellariformes have observed a dual foraging strategy, where adults will travel short distances to forage food for their young, and longer distances to forage for themselves (15, 39). These extended foraging trips enable parents to acquire, process, and excrete food needed for self-maintenance completely before obtaining more food to deliver to their offspring (40, 41). While this dual feeding strategy has not been directly observed in Leach’s storm-petrels, if parents are traveling shorter distances this would allow for more frequent visits to burrows. However, this also suggests that the parents of restricted chicks may have been relatively resource limited themselves, which could result in some physiological costs. Future work should attempt to quantify these costs and determine if they affect lifetime reproductive success.