Reassessing global change research priorities in mediterranean terrestrial ecosystems: how far have we come and where do we go from here?

Aim Mediterranean terrestrial ecosystems serve as reference laboratories for the investigation of global change because of their transitional climate, the high spatiotemporal variability of their environmental conditions, a rich and unique biodiversity and a wide range of socio-economic conditions. As scientific development and environmental pressures increase, it is increasingly necessary to evaluate recent progress and to challenge research priorities in the face of global change. Location Mediterranean terrestrial ecosystems. Methods This article revisits the research priorities proposed in a 1998 assessment. Results A new set of research priorities is proposed: (1) to establish the role of the landscape mosaic on fire-spread; (2) to further research the combined effect of different drivers on pest expansion; (3) to address the interaction between drivers of global change and recent forest management practices; (4) to obtain more realistic information on the impacts of global change and ecosystem services; (5) to assess forest mortality events associated with climatic extremes; (6) to focus global change research on identifying and managing vulnerable areas; (7) to use the functional traits concept to study resilience after disturbance; (8) to study the relationship between genotypic and phenotypic diversity as a source of forest resilience; (9) to understand the balance between C storage and water resources; (10) to analyse the interplay between landscape-scale processes and biodiversity conservation; (11) to refine models by including interactions between drivers and socio-economic contexts; (12) to understand forest–atmosphere feedbacks; (13) to represent key mechanisms linking plant hydraulics with landscape hydrology. Main conclusions (1) The interactive nature of different global change drivers remains poorly understood. (2) There is a critical need for the rapid development of regional- and global-scale models that are more tightly connected with largescale experiments, data networks and management practice. (3) More attention should be directed to drought-related forest decline and the current relevance of historical land use.