Methods to detect spatial biases in tracking studies caused by differential representativeness of individuals, populations and time - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Diversity and Distributions Année : 2023

Methods to detect spatial biases in tracking studies caused by differential representativeness of individuals, populations and time

Virginia Morera‐pujol
Paulo Catry
  • Fonction : Auteur
Maria Magalhães
  • Fonction : Auteur
Clara Péron
José Manuel Reyes‐gonzález
  • Fonction : Auteur
José Pedro Granadeiro
Teresa Militão
Maria P Dias
  • Fonction : Auteur
Daniel Oro
  • Fonction : Auteur
Giacomo Dell'Omo
  • Fonction : Auteur
Martina Müller
  • Fonction : Auteur
Vitor H Paiva
  • Fonction : Auteur
Benjamin Metzger
  • Fonction : Auteur
Verónica Neves
  • Fonction : Auteur
Joan Navarro
  • Fonction : Auteur
Georgios Karris
  • Fonction : Auteur
Stavros Xirouchakis
Jacopo G Cecere
  • Fonction : Auteur
Antonio Zamora‐lópez
  • Fonction : Auteur
Manuela G Forero
  • Fonction : Auteur
Ridha Ouni
  • Fonction : Auteur
Mohamed Salah Romdhane
  • Fonction : Auteur
Fernanda de Felipe
  • Fonction : Auteur
Zuzana Zajková
  • Fonction : Auteur
Marta Cruz‐flores
Jacob González‐solís
  • Fonction : Auteur
Raül Ramos

Résumé

Aim: Over the last decades, the study of movement through tracking data has grown exceeding the expectations of movement ecologists. This has posed new challenges, specifically when using individual tracking data to infer higher-level distributions (e.g. population and species). Sources of variability such as individual site fidelity (ISF), environmental stochasticity over time, and space-use variability across species ranges must be considered, and their effects identified and corrected, to produce accurate estimates of spatial distribution using tracking data. Innovation: We developed R functions to detect the effect of these sources of variability in the distribution of animal groups when inferred from individual tracking data. These procedures can be adapted for their use in most tracking datasets and tracking techniques. We demonstrated our procedures with simulated datasets and showed their applicability on a real-world dataset containing 1346 year-round migratory trips from 805 individuals of three closely related seabird species breeding in 34 colonies in the Mediterranean Sea and the Atlantic Ocean, spanning 10 years. We detected an effect of ISF in one of the colonies, but no effect of the environmental stochasticity on the distribution of birds for any of the species. We also identified among-colony variability in nonbreeding space use for one species, with significant effects of population size and longitude. Main conclusions: This work provides a useful, much-needed tool for researchers using animal tracking data to model species distributions or establish conservation measures. This methodology may be applied in studies using individual tracking data to accurately infer the distribution of a population or species and support the delineation of important areas for conservation based on tracking data. This step, designed to precede any analysis, has become increasingly relevant with the proliferation of studies using large tracking datasets that has accompanied the globalization process in science driving collaborations and tracking data sharing initiatives.
Fichier principal
Vignette du fichier
Diversity and Distributions - 2022 - Morera_E2_80_90Pujol - Methods to detect spatial biases in tracking studies caused by.pdf (11.76 Mo) Télécharger le fichier
Origine : Fichiers éditeurs autorisés sur une archive ouverte

Dates et versions

hal-04254302 , version 1 (23-10-2023)

Identifiants

Citer

Virginia Morera‐pujol, Paulo Catry, Maria Magalhães, Clara Péron, José Manuel Reyes‐gonzález, et al.. Methods to detect spatial biases in tracking studies caused by differential representativeness of individuals, populations and time. Diversity and Distributions, 2023, 29 (1), pp.19 - 38. ⟨10.1111/ddi.13642⟩. ⟨hal-04254302⟩
12 Consultations
6 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More