In the 21st century a multitude of global change forces, unprecedented in human history, are radically altering species distributions across the globe. In order to plan biodiversity conservation in the face of this threat, ecologists have developed tools to model the distributions of species individually, and forecast locations to which species might shift their ranges in the future. However we are still far from confident in our predictions, as a result of uncertainty in how accurate our models are, or whether we are modelling the processes that are important for range-shifts, and because there are still many elements of global change that we have not begun to predict.

I will discuss my recent work that addresses these uncertainties. I will present analyses of the climate paths along which species will shift as climate changes, revealing that fluctuations in climate change can prevent range-shifts by causing gaps in climate paths, even in the absence of geographic barriers. I will discuss how we can re-evaluate species distribution models using a little-used source of independent data – species distributions where they have become naturalised outside of their native region.

This approach indicates that predictive accuracy of distribution-based forecasts is low – far below the level of confidence required to inform conservation policy. This work fundamentally challenges the status quo, but also provides guidance for improving forecast accuracy. Finally, I address a powerful threat to biodiversity – biological invasions – the risk from which has not yet been modelled at a global level. I demonstrate that we can integrate available data to evaluate global risk levels. The developing world, as yet largely unscathed by biological invasions, is highly threatened in the 21st century, as are some biodiversity hotspots and islands, in which the impacts of invasions are likely to be severe.

Seminario organizado por el Departamento de Biodiversidad y Biología Evolutiva del Museo Nacional de Ciencias Naturales (MNCN), CSIC, y retransmitido a través de Cienciatk por el Servicio de Audiovisuales-Mediateca, MNCN-CSIC.