| Summary: | Interests in exploring Cold Water Corals (CWC) ecosystems witnessed a dramatic increase in the last decades, after the realisation that their habitats are threatened by ocean warming and acidification. However, they are still largely overlooked by the scientific community in deep and harsh environments like the Southern Ocean. Recent advances in species distribution models (SDM) have allowed forecasting species distribution patterns and assessing climate change impacts at different spatial scales. Several limitations related to the accuracy of species presences, the lack of reliable absence data and the limited spatial resolution of environmental factors, have restricted the widespread utilisation of these approaches in polar areas. In this work, real presence-absence records of 13 species were gathered from research expeditions and literature and combined with model-generated pseudo-absences, to cover the study area. Moreover, a final set of 14 high-resolution environmental variables was pre-selected and nine species distribution modelling algorithms were merged with means of the ensemble forecasting platform 'biomod2' to model the habitat suitability for azooxanthallate scleractinian corals, in the Weddell Sea. 'Biomod2' is implemented in 'R' and is a freeware, open source package. Response of scleractinian distribution to the future climate change was also investigated, based on two future scenarios of the bottom sea temperature. Present ensemble prediction maps accurately captured the potential ecological niches of the modelled species (good to excellent true skill statistic (TSS) and area under the receiver operating characteristic curve (AUC) evaluation measures). In the Weddell Sea, scleractinian distribution is limited to the continental shelf and slope areas with preference to small scale features (i.e., seamounts), which have been identified as having a high probability of supporting cold-water coral habitat. The most important factors in determining CWC habitat suitability were distance to coast and ...
|
|---|