"A thorough understanding of feedbacks between the terrestrial biosphere and the climate system is pivotal for any climate change mitigation strategy. Such feedbacks could significantly affect and potentially accelerate climate change. Terrestrial biosphere models are increasingly coupled interactively to ocean and atmosphere models with the goal to quantify these feedbacks. However, state-of-the-art terrestrial biosphere models fall short of the current understanding of important ecophysiological controls of land surface processes, as evidenced by the increasing amount of ecosystem observations that have become available recently. This project aims at bridging the gap between observational science and large scale biosphere modelling by developing representations of relevant ecophysiological processes at a level of detail suitable for an Earth system model. Two of the most important fields that urgently need to be better represented will be addressed, namely the biological control of canopy conductance and plant-soil interactions. Both substantially influence land surface fluxes, ecosystem productivity and long-term carbon sequestration. The project will use innovative techniques to develop novel process representations of canopy conductance by combining different types of ecosystem observations such as eddy-covariance based flux observations and plant trait data collected by the IGBP fast track initiative on plant functional types. Novel representations of soil organic matter dynamics, including a substrate control, and plant nutrient uptake pathways will be developed and evaluated based on recent laboratory and ecosystem manipulation experiments. The improved model will be used to revise predictions of state-of-the-art terrestrial biosphere models for present-day and future conditions. The project will thus provide a tool to better quantify potential future interactions of the terrestrial biosphere and the climate system for the use in coupled Earth system models."