aalto1 untyped-item.component.html
Predicting habitat suitability for Asian elephants in non-analog ecosystems with Bayesian models
Loading...
Access rights
openAccess
publishedVersion
URL
Journal Title
Journal ISSN
Volume Title
A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
This publication is imported from Aalto University research portal.
View publication in the Research portal (opens in new window)
View/Open full text file from the Research portal (opens in new window)
View publication in the Research portal (opens in new window)
View/Open full text file from the Research portal (opens in new window)
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.
Date
Department
Major/Subject
Mcode
Degree programme
Language
en
Pages
15
Series
Ecological Informatics, Volume 82, pp. 1-15
Abstract
Rewilding is an ambitious approach to conservation aiming at restoring and protecting natural processes. As the world is rapidly transitioning into conditions that have not been observed before, we need to be able to extrapolate and predict how natural processes would act under new conditions. Species distribution models have a good potential to inform rewilding decisions by the predictive modeling of potential species presence under various habitat conditions. A critical requirement when utilizing these models is to be able to express the uncertainty in the environment or its predictions. This study demonstrates the use of Bayesian statistical models to address this challenge. As a case study, we explore Bayesian logistic regression and Bayesian generalized additive models in order to predict suitable habitats for Asian elephants (Elephas maximus) until the year 2070 under the worst case working scenario of climate change. In this comparative study predictions of habitat suitability are solely based on climatic conditions. The results of the two Bayesian models are compared to two benchmark models, maximum-likelihood estimated logistic regression and random forest. We analyze and discuss trade-offs, relative advantages, and limitations of these modeling choices. The results of our analysis suggest that one configuration of Bayesian logistic regression gives the most robust predictions in this setting, which tend to correspond with the distribution of woodland biomes broadly similar to those in the species' historical range.
Description
Publisher Copyright: © 2023
Other note
Citation
Noda, R, Mechenich, M F, Saarinen, J, Vehtari, A & Žliobaitė, I 2024, 'Predicting habitat suitability for Asian elephants in non-analog ecosystems with Bayesian models', Ecological Informatics, vol. 82, 102658, pp. 1-15. https://doi.org/10.1016/j.ecoinf.2024.102658