Forested biomes contain the largest terrestrial carbon pools and currently buffer global warming by tempering the rise in atmospheric CO2 concentrations. However, a limited understanding of forests’ responses to climate change contributes to widely divergent estimates of their future carbon sink capacities. To reduce these uncertainties, more extensive empirical data are needed to quantify and link forest growth dynamics with climate variability over large spatial scales and over inter-annual to centennial time-scales. Annual growth rings of trees are a natural archive that fulfill these needs.

 In this talk I will detail progress in improving our understanding of i) carbon cycle-climate feedbacks that are expected to amplify anthropogenic warming and ii) the physiological responses of forests to increased CO2 concentrations. I will furthermore show approaches using tree-ring archives to validate (or falsify) components of Earth System Models: we for example find that vegetation models underestimate the summer temperature sensitivity of forest growth in Scandinavia, while overestimating the precipitation sensitivity in the Mediterranean region. I will conclude with some discussion of research priorities and strategies to provide improved quantification of carbon cycle processes during the next decade.