Discriminating climate, land-cover and random effects on species range dynamics

• Autor(a) principal: Taheri, S.
• Data de Publicação: 2021
• Outros Autores: García-Callejas, D., Bastos Araújo, Miguel
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10174/34914; https://doi.org/10.1111/gcb.15483

Species are reportedly shifting their distributions poleward and upward in several parts of the world in response to climate change. The extent to which other factors might play a role driving these changes is still unclear. Land-cover change is a major cause of distributional changes, but it cannot be discarded that distributional dynamics might be at times caused by other mechanisms (e.g. dispersal, ecological drift). Using observed changes in the distribution of 82 breeding birds in Great Britain between three time periods 1968–72 (t1), 1988–91 (t2) and 2007–2011 (t3), we examine whether observed bird range shifts between t1-t2 and t1-t3 are best explained by climate change or land-cover change, or whether they are not distinguishable from what would be expected by chance. We found that range shifts across the rear edge of northerly distributed species in Great Britain are best explained by climate change, while shifts across the leading edge of southerly distributed species are best explained by changes in land-cover. In contrast, at the northern and southern edges of Great Britain, range dynamics could not be distinguished from that expected by chance. The latter observation could be a consequence of boundary effects limiting the direction and magnitude of range changes, stochastic demographic mechanisms neither associated with climate nor land-cover change or with complex interactions among factors. Our results reinforce the view that comprehensive assessments of climate change effects on species range shifts need to examine alternative drivers of change on equal footing and that null models can help assess whether observed patterns could have arisen by chance alone.