Induced breeding failure alters movements, migratory phenology, and opportunities for pathogen spread in an urban gull population

Abstract

Background: Annual-cycle movements of wildlife are driven by a combination of intrinsic and extrinsic factors. In urban systems, management strategies to reduce human-wildlife interactions could also alter wildlife movement and distribution, with potential effects on key ecological processes such as pathogen spread.

Methods: To better understand how management actions interact with existing spatial dynamics to mediate wildlife movement patterns, we experimentally subjected urban-nesting yellow-legged gulls to induced breeding failure via egg-oiling. We then followed their movements using bird-borne GPS transmitters throughout the treatment season as well as the following annual cycle and compared them to the movements of tracked gulls whose nests were not oiled, while also accounting for individual and temporal factors known to influence movement patterns including sex, body size, and breeding stage.

Results: Gulls with oiled nests had smaller breeding-season home ranges, spent more time at breeding sites, made fewer foraging trips, and traveled shorter distances than gulls with non-oiled nests during the treatment season but not during the following breeding season. Gulls were partially migratory, with individuals showing a variety of migratory strategies from year-round residency to long-distance migration to inland urban centers. Although egg-oiling delayed the onset of post-breeding migration, individual migration strategies remaining consistent between years regardless of treatment. Antibody titres against three common pathogens varied among pathogens but not by migration distances or individual characteristics.

Conclusions: Our results show that induced breeding failure via egg-oiling may have unintended short-term consequences including smaller home range areas, altered habitat use, delayed migration, and longer breeding-site residency, suggesting that management actions aimed to reduce breeding success could increase opportunities for human-wildlife conflict and spread of spatially heterogeneous pathogens at local scales. At the landscape scale, the migration patterns and wintering distribution of yellow-legged gulls are unlikely to be affected by egg-oiling. However, long-distance inland migrations of a portion of the population present a novel pathway for pathogen transmission between and among marine habitats and terrestrial human, livestock, and wildlife populations.

Keywords: Larus michahellis; Foraging; Frioul; Human-wildlife conflict; Mediterranean; Yellow-legged gull.

Fig. 1

Changes in foraging trip metrics with significant year-nest oiling interactions (coefficients of interaction terms in GLMM did not overlap 0; Table 1) between 2021 (Y1) and 2022 (Y2) for tracked yellow-legged gulls nesting on Frioul, Marseille, France: a) distance, b) duration, c) trips per day, d) proportion of locations in urban habitat, e) core home range (50% KDE), and f) full home range (90% KDE). Shaded bands indicate 95% confidence intervals of trendlines. The oiled group (Y1: n = 14; Y2: n = 13) received egg-oiling treatment in Y1 but not Y2, while the non-oiled group (Y1: n = 13; Y2: n = 7) did not receive egg-oiling treatment in either year

Fig. 2

Breeding-season kernel density home ranges for yellow-legged gulls on Frioul, Marseille, France, 2021 (Y1; a) oiled, n = 14; b) non-oiled, n = 13) – 2022 (Y2; c) oiled n = 13, d) non-oiled, n = 7). Darker polygons show the cumulative footprint of all individual 50% contour areas, and lighter colors show 90% contours. Yellow stars indicate the breeding colony location. The interaction of year with oiling status was a significant predictor of both 50% and 90% contour areas (see Table 1)

Fig. 3

Proportion of time spent at the nest by day of year for tagged yellow-legged gulls on Frioul, Marseille, France during the breeding seasons of 2021 (Y1; solid lines and darker points) − 2022 (2022; dashed lines and lighter points). Shaded bands indicate 95% confidence intervals of trendlines. The oiled group (Y1: n = 14; Y2: n = 13) were subject to egg-oiling treatment in Y1 but not Y2, while the non-oiled group (Y1: n = 13; Y2: n = 7) did not receive egg-oiling treatment in either year. The interaction of year with oiling status was a significant predictor of colony attendance (see Table 1)

Fig. 4

Migration routes of yellow-legged gulls breeding on Frioul, Marseille, France, 2021 (Y1)– 2022 (Y2). Gulls with oiled nests (solid lines) received nest oiling treatment in Y1 and not in Y2. Gulls with non-oiled nests (dashed lines) did not receive nest oiling treatment in either study year. Blue lines represent riparian corridors and grey dots show the locations of major cities (> 30,000 inhabitants). Insets show main wintering areas near a) Paris, b) Tours/Orleans, and c) La Rochelle/Bordeaux, with areas classified as urban (based on Copernicus Land Monitoring Service 2018 Corine Land Cover classes) shown in grey

Fig. 5

Difference in a) departure date, b) return date, and c) migration distance for migratory yellow-legged gulls from oiled (n = 6) and non-oiled (n = 5) nests on Frioul, Marseille, France, 2021 (Y1)-2022 (Y2). Shaded bands indicate 95% confidence intervals of trendlines. The oiled group were subject to egg-oiling treatment in Y1 but not Y2, while the non-oiled group did not receive egg-oiling treatment in either year. The interaction of year with oiling status was a significant predictor of departure date, but not of return date or migration distance (see Table S2)

Fig. 6

Antibody concentrations against a) avian influenza virus (AIV), b) Toxoplasma gondii (TOX), and c) infectious bronchitis virus (IBV) for resident (blue, n = 9), local (pink, n = 4) and migratory (yellow, n = 7) yellow-legged gulls on Frioul, Marseille, France, 2021. Titers are reported as a proportion of the range between positive and negative controls