In colony-forming seabirds, individuals nesting at the center of a colony generally exhibit higher parental survival and reproductive success than those nesting at the periphery, a pattern known as the peripheral-central distribution (Coulson, 1968; Tenaza, 1971; Aebischer and Coulson, 1990; Piro and Schmitz Ornés, 2021). Variations in fitness-related traits across nesting locations may be influenced by internal parental factors, such as age (Aubry et al., 2009), breeding experience (Vergara and Aguirre, 2006), and foraging ability (Spurr, 1975), as well as external factors, including predation risk (Vine, 1971; Brodin, 2001; Brown and Kotler, 2004; Piro and Schmitz Ornés, 2021), physical environmental conditions (Montevecchi, 1978; Rounds et al., 2004; Minias, 2014; Pagenaud et al., 2022), and social interaction levels (Birkhead, 1977; Kokko et al., 2004).

Seabirds may select nesting sites according to these factors, often shifting from subcolonies or peripheral areas to more significant central sites (Serrano and Tella, 2007). As a result, older, more experienced, and reproductively successful individuals are more likely to occupy the central regions, whereas less experienced individuals tend to nest peripherally (Aebischer and Coulson, 1990; Bosch and Sol, 1998; Kim and Monaghan, 2005; Indykiewicz et al., 2019). However, some seabird species exhibit strong nesting site fidelity, returning to the same location in each breeding season (Pearce, 2007; Piro and Schmitz Ornés, 2021), potentially because of their genetic preferences for specific habitats (Rodway and Regehr, 1999).

Although colony position and associated factors can influence reproductive outcomes (but not always; see Ryder and Ryder (1981), Shaw (1985)), the long-term strategies of iteroparous seabirds may buffer against the negative impacts of occasional breeding failures. Temporary breeding failures in peripheral nests may not substantially reduce overall fitness because skipping a breeding attempt allows individuals to conserve resources, including physiological conditions, that can be allocated to future breeding opportunities. To further clarify the peripheral-central nest-site distribution, comparing the physiological loads of parental seabirds, in conjunction with their behavior, between the peripheral and central nests is essential.

One promising approach involves examining oxidative stress, which has become a key indicator of the physiological costs associated with wildlife behavior (Koyama et al., 2021). It arises from an imbalance between pro-oxidants, particularly reactive oxygen species (ROS), and antioxidant defenses (Monaghan et al., 2008). While ROS plays a crucial role in pathogen elimination, their excessive accumulation can damage DNA, proteins, and lipids, leading to fatigue due to impaired mitochondrial function and a shift toward anaerobic metabolism. Conversely, antioxidants mitigate oxidative damage through endogenous enzymes and dietary compounds (Dröge, 2002). Elevated ROS levels are generally associated with physiological strains, whereas antioxidant levels indicate the capacity for recovery and resilience (Ahmadi et al., 2006). In avian studies, female European starlings (Sturnus vulgaris) experimentally subjected to increased breeding costs exhibit reduced physiological functions, including increased oxidative stress (Fowler and Williams, 2017). Furthermore, individuals may adjust their antioxidant mechanisms in response to anticipated conditions, potentially balancing endogenous and dietary antioxidant activities based on past environmental experiences and expected intake (Noguera et al., 2015). Oxidative stress has also been observed to exert delayed effects on life-history traits linked to survival (Noguera et al., 2012) and is often positively correlated with reproductive effort (Christe et al., 2012; Fletcher et al., 2013); however, some studies have reported no such correlation (Nussey et al., 2009; Markó et al., 2011).

In this study, we compared the physiological loads and foraging movements of black-tailed gulls (Larus crassirostris) incubating on Kabushima Island, where a fenced, sea-surrounded protected center (Central Group, CG) was contrasted with a more exposed periphery accessible to humans and predators (Peripheral Group, PG) (Figure 1). Although defining clear boundaries between the central and peripheral areas is challenging, the fenced structure of this colony allowed us to identify distinct groups consistent with the peripheral-central distribution. We recorded foraging movements using biologging and measured physiological loads based on oxidation levels and antioxidant capacity. Given that sexual differences in foraging behavior (Kazama et al., 2018) and/or antioxidant capacity (Lin et al., 2022) may reflect differences in breeding investment or physical condition, we also compared clutch size and body size, traits often correlated with and indicative of reproductive effort, between the CG and PG. Clutch size, which represents reproductive demand, may also influence oxidative stress because it is related to the heat requirements of seabirds (Biebach, 1984; Moreno et al., 1991; Moreno and Sanz, 1994), including black-tailed gulls (Niizuma et al., 2005). Because black-tailed gulls on Kabushima Island are known to exhibit strong nest site fidelity (Narita and Narita, 2004), we hypothesized that PG individuals may consistently experience lower reproductive success due to interrupted breeding or reduced reproductive investment while simultaneously gaining physiological benefits that enhance self-maintenance and future breeding potential.

The average clutch size for the five PG gulls (3 females and 2 males) was 2.4 ± 0.54 SD (range: 2–3). No chicks hatched in four of these nests, whereas in one nest, two chicks hatched; however, the first chick disappeared at 5 days post-hatching, and the second chick died at 4 days of age. Although PG parents remained near the nest after these losses, no additional eggs were laid. For the CG, 20 gulls (5 females and 15 males) were caught during the incubation period, with an average clutch size of 2.25 ± 0.79 SD (range: 1–4), an average of 1.86 ± 0.91 SD (range: 0–3) hatched chicks per clutch, and an average of 1.00 ± 0.77 SD (range: 0–3) fledglings (chicks that survived to 30 days of age). In two CG nests, all eggs failed to hatch.

The PG birds were at least 10 years old, including one born in 2006 and another in 2013 (both identified by metal rings). The other three were captured in 2007 and 2008 and identified using plastic rings. The age of the CG birds ranged from 5 to 24.

The average trip duration and maximum distance traveled were 1.48 ± 3.20 h and 15.72 ± 15.70 km for PG, and 3.26 ± 2.89 h and 24.28 ± 18.89 km for CG (Figures 2, 3). The mean number of daily foraging trips was 3.79 ± 1.24 for CG and 2.75 ± 1.18 for PG, respectively. The mean oxidation levels (d-ROMs) and antioxidant capacity (BAP) were 41.8 ± 25.7 U. CARR and 1,431.0 ± 435.0 µM for PG, and 69.6 ± 25.2 U. CARR and 1,293.0 ± 326.0 µM for CG (Figure 3). In addition, behavioral and oxidative stress measurements of the two CG gulls whose eggs failed to hatch did not deviate from the range observed in the CG group.

The Bayesian t-test results showed significant differences in trip duration between the PG and CG, with a Bayes Factor (BF) of 20.97, indicating a strong effect. BF = 4.60 suggested a moderate difference between the groups for the maximum distance traveled. The oxidation level showed weak evidence for a difference with BF = 1.82, whereas the antioxidant capacity (BF = 0.54) showed no difference between the groups. The clutch size also showed no evidence for a difference, with a BF of 0.45. Additionally, the Bayesian t-test results for the number of measurement days, the total number of trips, and the average number of trips per day yielded the following BF values: 2.52 for the number of measurement days, providing weak evidence for a difference between the groups, 148.30 for the total number of trips, providing strong evidence for a difference; and 21.90 for the average number of trips per day, also providing strong evidence for a difference.

The PGs could not raise chicks because of several factors, including non-hatching, egg loss, and chick mortality, a pattern similar to that observed in previous years for most parents in peripheral areas. It is unclear whether non-hatching results from unfertilized eggs or inadequate embryonic development. This is because the PG area was easily accessible to tourists and feral cats, frequent disturbances caused parents to leave their nests. These disturbances are likely to reduce hatching success, either through direct nest attacks or lowered nest temperatures linked to shortened incubation periods (Di Giovanni et al., 2022; Clemencin et al., 2024), as well as increased stress resulting from proximity to disturbances (Ellenberg et al., 2007).

Regardless of hatching success, the antioxidant capacity was similar between the PG and CG, indicating that the intra-species variation in antioxidant capacity was slight. This finding suggests that black-tailed gulls maintain comparable levels of non-enzymatic endogenous antioxidants (e.g., vitamins A and E, uric acid; Sharma et al., 2018) and exogenous antioxidants (e.g., coenzyme Q10; Zozina et al., 2018) and carotenoids (Milani et al., 2017). In contrast, differences in foraging movements and oxidative stress levels suggest that the primary defense against oxidative stress during the breeding season likely depends on endogenous enzymes such as glutathione peroxidase. Similar to other bird species (McWilliams et al., 2021), glutathione peroxidase may serve as the first line of defense against ROS. Consequently, if the observed levels of foraging and breeding efforts remain within the capacity of endogenous enzymatic systems, the immediate need for additional exogenous antioxidants may be minimal.

Compared to the CG, the PG exhibited significantly shorter trip durations, a smaller maximum distance reached (strongly correlated with the total distance), and a lower frequency of foraging trips. Black-tailed gulls on Kabushima Island utilize a variety of feeding grounds and consume a broad diet beyond fish and shellfish (Yoda et al., 2012; Mizutani et al., 2021). Their regurgitated food items include freshwater fish and insects (Narita and Narita, 2004). Although the exact types and quantities of food consumed in this study remain unknown, the foraging range of PG overlapped substantially with that of CG, and no differences in antioxidant levels were observed between the groups. This suggests that both the PG and CG consumed similar types and amounts of food. However, the observed differences in foraging behavior likely influenced energy expenditure, leading to slight differences in ROS production and oxidation levels.

Although antioxidant levels were similar between individuals, PG exhibited lower oxidation levels, possibly due to nearby, shorter foraging trips and a lower daily foraging frequency. Gull species spend much of their flying time flapping their wings, a highly energy-intensive activity (Bishop, 2005; Mizrahy-Rewald et al., 2022) that likely increases energy consumption and ROS production. Additionally, the incubation period imposes significant physiological costs on the parent birds. For example, in common eiders (Somateria mollissima), high incubation demands lead to weight loss and decreased immune function due to reduced food intake, resulting in decreased fertility in the following year and long-term fitness costs (Hanssen et al., 2005). In contrast, the failed hatching observed in PG suggests inadequate incubation, which would have led to lower incubation costs. Although breeding costs are generally buffered during the non-breeding season (Senner et al., 2014; Briedis et al., 2018; Gow et al., 2019), consistently reduced breeding investment may minimize negative carryover effects in subsequent seasons.

PG individuals experienced consistently low reproductive success over multiple years; however, it is intriguing that they do not change their nesting locations. They may lack the competitive ability to secure favorable nest sites. Instead, the low oxidative stress associated with less demanding foraging trips may offset the high reproductive costs borne by CG individuals, reflecting an underlying life-history trade-off. Our results indicate that a peripheral-central distribution generates heterogeneity in reproductive success within a colony. However, the peripheral groups may gain fitness benefits by occasionally foregoing reproduction. Future research should investigate behavioral differences during the incubation, chick-rearing, and non-breeding periods, as well as their long-term physiological costs. Such investigations would enable a more comprehensive evaluation of rearing costs within the broader framework of life-history trade-offs. From a conservation perspective, completely fencing off all black-tailed gull nesting areas to protect PG groups from predators and human disturbances may be impractical. Instead, phased measures, such as deterrents against easily accessible small terrestrial mammals and visitor restrictions, combined with countermeasures such as fencing, which are considered critical to success, are expected to improve the hatching success rate of PGs.

One limitation of this study is the absence of continuously recorded physiological data. Blood, a vital biomarker source, is widely used in diagnostics and research (Toner and Irimia, 2005), underscoring the significant demand for real-time blood sampling loggers. Wearable automatic blood-sampling devices designed to minimize pain and distress are being developed for use in captivity settings (Li et al., 2009; Li et al., 2015; Hopper, 2016). While the long-term collection of blood from wild flying animals remains challenging, advances in non-invasive techniques, such as continuous glucose monitoring via interstitial fluid (Mathew et al., 2024), offer promising alternatives. Deploying such technologies in seabirds could revolutionize biologging, enabling researchers to explore how species balance reproduction and survival, respond to environmental disturbances, and allocate energy across life-history stages.