A metapopulation approach to predict species range shifts under different climate change and landscape connectivity scenarios

• Autor(a) principal: Mestre, Frederico
• Data de Publicação: 2017
• Outros Autores: Mira, António, Beja, Pedro, Pita, Ricardo
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10174/22707; https://doi.org/10.1016/j.ecolmodel.2017.06.013
• topic: Dispersal; Metapopulation; Climate change; Landsacape change; Ecological niche modelling

Forecasting future species distributions under climate change scenarios using Ecological Niche Models(ENM) is common practice. Typically, these projections do not account for landscape connectivity andspecies dispersal abilities. When they do account for these factors, they are based on either rather sim-plistic or overly complex and data-hungry approaches. Here we apply a new approach for predictingspecies range shifts under different climate change and landscape connectivity scenarios that balancesdata requirements and output quality. The approach builds on the metapopulation concept to producea dispersal model based on repeated simulations of stochastic extinction-colonization dynamics acrossmultiple landscapes of variable connectivity. The model is then combined with an ENM to produce morerealistic predictions of species range shifts under environmental change. Using the near-threatened Cabr-era vole (Microtus cabrerae) as a model species and considering two contrasting climate change scenarios(B2 and A1b) and three scenarios of increasing landscape connectivity, we confirmed that model predic-tions based solely on ENM overestimated future range sizes (2050 and 2080) in relation to predictionsincorporating both future climates and landscape connectivity constraints. This supports the idea thatlandscape change critically affects species range shifts in addition to climate change, and that models dis-regarding landscape connectivity tend to produce overly optimistic predictions, particularly for specieswith low dispersal abilities. We suggest that our empirically-based simulation modelling approach pro-vides a useful framework to improve range shift predictions for a broad range of species, which is essentialfor the conservation planning of metapopulations under climate and landscape change.