Direct and indirect impacts of climate change and land use change over biodiversity: a case of study with the brown bear in Europe

Climate and land use changes are the main drivers of biodiversity loss and species distribution range contractions. Previous studies have primarily explored the direct effects of these drivers on biodiversity while ignoring the indirect effects that come through other species. Models ignoring these mechanisms are prone to erroneous predictions of how global change impacts biodiversity, which impedes our ability to make effective management decision for biodiversity conservation. Here, we focus on the brown bear (Ursus arctos), a well-studied omnivore species that has trophic interactions with many different taxa. We aim understand how direct and indirect impacts of climate and land use change would affect the future distribution and vulnerability of different populations across Europe. We first conducted a literature search in Web of Science and Google Scholar to gather all available brown bear diet studies in Europe, based on faecal or stomach content. From the 32 identified studies, we selected 16 in which food items were described at the species level. In total, we compiled information on ~100 species that are generally consumed by brown bears. We constructed a spatial dataset assigning diet to different populations. We download well-defined and located occurrence data for all species found in the diet of the from brown bear from GBIF. To build reliable species distribution models for all these species, we used biomod2 in R that provides multiple algorithms and allows generating ensemble of species predictions per species. We control for the potential spatial bias in GBIF in the selection of pseudo-absence data following recommendation by Phillips et al. 2006. Single species distribution models for all species found in the diet were finally built in function of current climate and land use data. Using different scenarios for climate and land use change, we projected the future distributions of the brown bear diet species. Finally, using these projections and scenarios, we calculated the potential current and future suitable habitats of brown bear in Europe in function of its diet, climate and land use. The range dynamic of the brown bear was best explained by considering both direct and indirect impacts. Accounting for diet (indirect effect) buffered the direct impact of climate and land use change. Our results highlight the importance of taking a food-web ecosystem approach to evaluate the impact of global changes and predict modifications in the distribution and vulnerability of species.peerReviewe