A challenge in scaling up population dynamics using remote sensing is dealing with uncertainty in the identity of individuals between datasets collected at different times. Without a model for this uncertainty, we risk biasing the estimates of growth, recruitment and mortality that are key to understanding population and how they vary over space and through time. We are utilizing statistical linkage models -- approaches first developed in the social sciences to track individual people between incomplete databases -- to estimate forest dynamics for a landscape in the Colorado Rockies using repeat LiDAR scans.
