Change of support for heavy-tailed zero inflated data: application to spatially aggregated ecological data for fine-scale species distribution inference - Archive ouverte HAL Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2024

Change of support for heavy-tailed zero inflated data: application to spatially aggregated ecological data for fine-scale species distribution inference

Résumé

In environmental sciences, available data are often available at a coarse resolution. This can result in a mismatch between the data resolution and the resolution at which process inferences should be made. This misalignment, usually named as change of support (COS) issue, can lead to biased inferences if not addressed in the models. Yet, solutions to the COS issue have only been proposed for a limited number of simple observational processes (e.g. Poisson or Gaussian processes), which narrows their range of application. Motivated by a fisheries science case study, we introduce a hierarchical method addressing COS issue for zero-inflated data with highly skewed tails. Such data are common in environmental sciences but are not handled by existing COS methods. Our approach requires to know the spatial locations of point-level data and considers that aggregated available data are convolutions of these point-level data. We assess the accuracy of our method through a simulation study, describing different scenarios of COS. Subsequently, we apply our model to a motivating case study, focusing on the distribution of the common sole in the Bay of Biscay. Our findings illustrate that our approach provides better estimates and predictions than the ad hoc methods used to geoprocess aggregated data and refine their resolution.
Fichier principal
Vignette du fichier
JRSSC_COS.pdf (4.17 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-04626952 , version 1 (01-07-2024)

Licence

Identifiants

  • HAL Id : hal-04626952 , version 1

Citer

Baptiste Alglave, Bastien Mourguiart, Kasper Kristensen, Etienne Rivot, Mathieu Woillez, et al.. Change of support for heavy-tailed zero inflated data: application to spatially aggregated ecological data for fine-scale species distribution inference. 2024. ⟨hal-04626952⟩
0 Consultations
0 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More