Classic or hybrid? The performance of next generation ecological models to study the response of Southern Ocean species to changing environmental conditions.

• Published in: Diversity and Distributions
• Main Authors: Guillaumot, Charlène, Belmaker, Jonathan, Buba, Yehezkel, Fourcy, Damien, Dubois, Philippe, Danis, Bruno, Le Moan, Eline, Saucède, Thomas
• Other Authors: Laboratoire de Biologie Marine (LBM), Université libre de Bruxelles (ULB), Biogéosciences UMR 6282 (BGS), Université de Bourgogne (UB)-Centre National de la Recherche Scientifique (CNRS), School of Zoology, Faculty of Life Sciences Israel, Tel Aviv University (TAU), Dynamique et durabilité des écosystèmes : de la source à l’océan (DECOD), Institut Français de Recherche pour l'Exploitation de la Mer (IFREMER)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE), Work supported by a ‘Fonds pour la formation à la Recherche dans l'Industrie et l'Agriculture’ (FRIA) and ‘Bourse Fondation de la mer’ grants.
• Format: Article in Journal/Newspaper
• Language: English
• Published: HAL CCSD 2022
• Subjects: Bayesian inference; data-poor systems; integrated approaches; Kerguelen Islands; sea urchin; species distribution modelling; [SDE.BE]Environmental Sciences/Biodiversity and Ecology; Antarctic; Baie du Morbihan; Kerguelen; Southern Ocean; Antarc*
• Online Access: https://hal.science/hal-03833265; https://doi.org/10.1111/ddi.13617

17 pages International audience Aim In the context of intensifying threats of climate change on marine communities, ecological models are widely applied for conservation strategies, though polar studies remain scarce given the limited number of datasets available. Correlative (e.g. species distribution models, SDM) and mechanistic (e.g. dynamic energy budget models, DEB) modelling approaches are usually used independently in studies. Using both approaches in integrative, hybrid models could help to better estimate the species potential ecological niche, as mechanistic and correlative models complement each other very well, giving more insights into species potential response to fast-changing environmental conditions. Location The study focusses on the Baie du Morbihan, a silled basin located in the east of the Kerguelen Islands (sub-Antarctic). Methods A hybrid, correlative-mechanistic model was implemented to predict the response of the endemic sea urchin Abatus cordatus (Verrill, 1876). We compared the performances of classic and integrated approaches to predict A. cordatus distribution according to two dates representing seasonal contrasts. Two integrated approaches were studied and performed by either (1) including the spatial projection of the DEB model as an input layer inside the SDM ('integrated SDM-DEB') or (2) using a Bayesian procedure to use DEB model outputs as priors of the SDM ('integrated Bayesian' approach). Results Results show higher performances of 'integrated Bayesian' approaches to evaluate A. cordatus potential ecological niche compared with 'classic' and 'integrated SDM-DEB' methods. The influence of environmental conditions on model predictions is further captured with these Bayesian procedures and better highlights the environmental influence on the species-predicted distribution. Model performance is good for the different simulations, and uncertainty in predictions is well-highlighted. Main conclusions The good performances of 'integrated Bayesian' approaches to estimate species ...