Aim: Seasonally migratory species generate large movements of organisms and biomass between distant breeding and non-breeding grounds. However, our understanding of how migratory species shape global networks of interconnected communities (meta-communities) remains limited. Migratory links between communities can be measured in different ways (e.g., species occurrence, abundance or biomass), each providing complementary information by modulating the relative importance of species in meta-communities. We aim at investigating to what extent measuring migratory links using species occurrence, abundance or biomass can reveal alternative structures (i.e., topology) in a meta-community linking an Arctic breeding ground to remote non-breeding grounds. Location: We use as a study case the High-Arctic vertebrate community of Bylot Island (Nunavut, Canada), along with ecoregions of North and South America, Europe and Africa. Time period: Present. Major taxa studied: Terrestrial Arctic birds (30 species) and mammals (5 species). Methods: We first consider species occurrence at the non-breeding grounds to define migratory links within the meta-community. Secondly, we measure the number of individuals and the amount of biomass traveling along those links. We finally compare the meta-community structure under each scenario using a migration network representation. Results: Patterns of species occurrence, abundance and biomass reveal that temperate ecoregions of South and especially North America maintain strong ecological connections with the vertebrate community of Bylot Island. However, the structural role of species within the network can vary substantially depending on how migratory links are measured (i.e., contrasting topological anomalies). Using abundance or biomass to measure migratory links results in a finer partitioning of the network into modules compared to using species occurrence alone. Main conclusions: We highlight that using different metrics of migratory links reveals unique, yet complementary structural features of meta-communities. These findings contribute to assessing the vulnerability of communities to perturbations occurring in distant but connected environments through migration.

Funding provided by: Canada First Research Excellence FundROR ID: https://ror.org/01ktx4s83Award Number: Funding provided by: Center for Northern StudiesROR ID: https://ror.org/01q8ytn75Award Number: Funding provided by: Centre de la science de la biodiversité du QuébecCrossref Funder Registry ID: Award Number: Funding provided by: Environment and Climate Change CanadaROR ID: https://ror.org/026ny0e17Award Number: Funding provided by: Fonds de Recherche du Québec – Nature et TechnologiesROR ID: https://ror.org/00b9f9778Award Number: Funding provided by: Garfield Weston FoundationROR ID: https://ror.org/00rm5fa18Award Number: Funding provided by: Aboriginal Affairs Northern Dev CanadaROR ID: https://ror.org/05pgxdd43Award Number: Funding provided by: Government of CanadaROR ID: https://ror.org/010q4q527Award Number: Funding provided by: Natural Resources CanadaROR ID: https://ror.org/05hepy730Award Number: Funding provided by: Natural Sciences and Engineering Research CouncilROR ID: https://ror.org/01h531d29Award Number: Funding provided by: ArcticNetROR ID: https://ror.org/01tca3t44Award Number: Funding provided by: Nunavut Wildlife Management BoardROR ID: https://ror.org/056r65c22Award Number: Funding provided by: Parks CanadaROR ID: https://ror.org/008sy4716Award Number: Funding provided by: Polar Knowledge CanadaROR ID: https://ror.org/00rfash91Award Number: Funding provided by: Université du Québec à RimouskiROR ID: https://ror.org/049jtt335Award Number: Funding provided by: Université LavalROR ID: https://ror.org/04sjchr03Award Number:

Species abundance and biomass: The mean summer abundance and biomass of all vertebrate species in the Bylot Island community were estimated from the long-term ecological monitoring program (see Moisan et al., 2025). Species non-breeding range: We used the previously defined non-breeding ranges of all migratory and partially migratory species of Bylot Island (see Moisan et al., 2023) derived from tracking of individuals (6 species; Gilchrist, Mosbech and Sonne, 2004; Therrien, Gauthier and Bêty, 2012; Robillard, Gauthier, Therrien and Bêty, 2018; Léandri‐Breton, Lamarre and Bêty, 2019; LeTourneux et al., 2021; Seyer, Gauthier, Bêty, Therrien and Lecomte, 2021; Lamarre et al., 2021) and refined species range maps (BirdLife International and Handbook of the Birds of the World, 2019; Fink et al., 2021). Ecoregions of the world: We used the classifications of terrestrial (Olson et al., 2001), marine (Spalding et al., 2007) and freshwater ecoregions (only for large bodies of water; Abell et al., 2008).