Overcoming challenges in trait-based global modeling

Scientific investment in predictive power must be commensurate with the crucial role of Earth system models in crafting environmental policy. Biological feedbacks are a major source of uncertainty in predictions of global temperature and ecosystem resilience. Trait-based concepts offer promise in reducing uncertainty, but we need to advance fundamental theory, identify emergent patterns, and scale up knowledge about traits to make quantitative predictions. Meeting these challenges will require training a new generation of researchers to cross the empirical-modeling divide. We must also invest in new scientific networks and model infrastructure to engage a broader community in tackling trait-based prediction.