Using Data on Vegetation Structure to Initialize and Test Ecosystem Models (Invited)

Scientists have long recognized the importance of vegetation structure in forest dynamics, but relevant data and models have generally not been available for large-scale applications. Now future satellite missions utilizing lidar and radar technologies have the potential to provide globally consistent data on vegetation structure, and new models are being developed to track the dynamics of vegetation 3-D structure over large scales. Used together, these advances have the potential to revolutionize the initialization and testing of terrestrial models, and thereby reduce key uncertainties and improve model predictions of carbon cycle dynamics. Here, using both theoretical models and an advanced mechanistic ecosystem model, we investigate the use, and potential use, of data on various forest structural metrics for initializing and testing terrestrial models. Focus is on quantifying the potential utility of various metrics, and the resolution and accuracy with which these measurements need to be made.