Assisted tree migration can reduce but not avert the decline of forest ecosystem services in Europe

European forests are facing multiple natural and anthropogenic pressures that are expected to become more severe in the next decades. Tree diversity is projected to decline in many areas across the continent. How this will affect the provision of forest services remains an open question, whose answer depends, among others, on the practical and theoretical challenges of incorporating assisted migration into climate adaptation strategies. Here, we tackle the issue by combining a large dataset of tree species occurrences, future climatic projections, and data on tree functional traits and tree-specific forest services into a novel modelling framework. We estimate that, by the end of the century and under a natural dispersal scenario, the provision of forest services would decrease on average by 15% in Europe (for RCP 4.5; 23% for RCP 8.5), and up to 52% (70% for RCP 8.5) in the Mediterranean. To explore if and how management could reduce the projected losses, we simulated a suite of alternative assisted migration strategies aimed at identifying, for each locality, the tree species communities offering the best compromise in terms of resilience to climate change and delivery of specific combinations of ecosystem services. Such strategies could reduce losses of services by 10% (15%) on average in Europe, and even increase service availability in the Alpine and Boreal regions but not in the Mediterranean, where losses will remain as high as 33% (54% for RCP 8.5). Our findings highlight how science-driven management strategies could be vital to reduce an otherwise dramatic, European-wide decline of forest services. Our results are qualitatively robust to different assumptions on future carbon emissions and related climate trajectories. That is, our simulated assisted migration strategies identify similar tree species communities under different pathways (RCP 4.5 vs RCP 8.5). This makes our approach a powerful tool for forest management, as it generates advice that is valid regardless of whether, and to what extent, human society will steer away from business-as-usual emission trajectories.