Pré-Publication, Document De Travail Année : 2025

Empirical Orthogonal Functions for ecology

Résumé

Spatio-temporal data are ubiquitous in ecology, involving large spatial domains, long time periods, and multiple variables. To take advantage of these data for ecological analysis, it is crucial to rely on both efficient and ecologically interpretable statistical methods. Empirical Orthogonal Functions (EOFs) are a widely used method for reducing the dimensionality of spatio-temporal data. While EOFs are common in meteorology and climate studies, their use in ecology is still emerging. This paper introduces the basics of EOFs and presents new methods to specifically address key ecological questions. Since EOFs lack explicit ecological constraints, their results may be difficult to interpret. We propose a method to rotate EOFs using a temporal ecological variable, improving interpretability by better representing ecological patterns. We also explore the use of EOFs in multivariate analysis to identify shared spatial and temporal patterns in community and ecosystem dynamics. We illustrate these methods on two case studies including NDVI satellite data over France and spatio-temporal predictions from fish species distribution models. Our goal is to provide a clear and concise roadmap for applying EOFs to ecological problems. We emphasize practical guidance through code snippets and two worked examples, making the approach accessible to a broad audience of ecologists.
Fichier principal
Vignette du fichier
main_manuscript.pdf (5) Télécharger le fichier
SM1.pdf (101) Télécharger le fichier
SM2.pdf (153) Télécharger le fichier
SM3.pdf (1) Télécharger le fichier
SM4.pdf (5) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Origine Fichiers produits par l'(les) auteur(s)
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04693871 , version 1 (16-09-2024)
hal-04693871 , version 2 (11-02-2025)

Identifiants

  • HAL Id : hal-04693871 , version 2

Citer

Baptiste Alglave, Benjamin Dufée, Said Obakrim, James T Thorson. Empirical Orthogonal Functions for ecology. 2025. ⟨hal-04693871v2⟩
142 Consultations
96 Téléchargements

Partager

More