Accurately monitoring natural resources is vital for sustainable forest management, but it is challenging due to the high variability of the populations and environmental conditions (e.g., land use changes, naturalized species, wind disturbances). Recently, the rapid expansion and increasing availability of open-source databases (e.g., aerial photographs, LiDAR point cloud, long-term measurements from national forest inventories) offer opportunities to characterize site conditions and quantify uncertainty of growth and yield for diverse forests. In this presentation, common challenges in forest modeling and inventory will be discussed (e.g., non-sampled or sparsely sampled domains, subject-specific predictions across a large number of groups). Approaches to address these challenges will also be presented, such as the calibration of growth and yield models and the integration of remotely sensed data into forest inventory (e.g., small area estimation). Examples will be drawn from loblolly pine plantations in the southeastern United States, as well as mixed-species tropical forests in the Caribbean and Pacific Islands.