Forests are dynamic systems increasingly affected by natural and anthropogenic disturbances, including wildfire, drought, storms, insects, disease, logging and land-use change. These disturbances influence individual trees, forest structure and composition, carbon storage, biodiversity, and ecosystem functioning, while forest recovery varies across spatial and temporal scales. Understanding these dynamics across scales is critical for assessing ecosystem resilience, quantifying forests’ contributions to the global carbon cycle, and supporting sustainable forest management under a changing climate.
This session highlights advances in remote sensing and artificial intelligence (AI) for detecting, characterizing and predicting forest disturbance, recovery and resilience across scales. We welcome contributions using optical, SAR, LiDAR, hyperspectral and thermal observations, together with time-series analysis, machine learning, ecological modeling, and data fusion. We particularly encourage studies that: (1) identify early warning signals of tree- and forest-level changes; (2) detect and quantify disturbance and recovery at tree, plot, forest, and landscape scales; (3) characterize changes in tree and forest structure, biomass, carbon, and function; (4) integrate observations across spatial and temporal scales; and (5) combine field measurements, process-oriented models and multi-source Earth observations to assess and simulate forest dynamics and resilience. Contributions using emerging satellite missions and novel AI approaches to connect tree-level processes with forest- and landscape-scale patterns are especially welcome.