Remote sensing is an essential tool for observing, quantifying, and monitoring forest ecosystems across vast spatial and temporal scales. Advances in sensor technology, data availability, and computational methods enable detailed observations from terrestrial, mobile, UAV, airborne, and space-borne platforms. Optical, multispectral, and hyperspectral imagery, alongside SAR, LiDAR, photogrammetry, and Structure-from-Motion (SfM), provide complementary data on forest composition, structure, function, and dynamics across natural, managed, and urban environments.
The availability of multi-source observations creates unprecedented opportunities but also raises significant scientific challenges. Because forest information is scale-dependent, no single sensor can capture all ecological characteristics. Integrating observations across spatial, temporal, and spectral scales is therefore essential for developing robust, transferable, and scalable approaches. Furthermore, in-situ measurements remain crucial for the calibration, validation, uncertainty assessment, and physical interpretation of remote sensing products.
Remote sensing supports qualitative and quantitative assessments, from mapping forest cover, species, and vegetation characteristics to estimating tree height, DBH, biomass, timber volume, canopy structure, and density. In urban environments, these tools support tree inventories and green infrastructure assessment. At larger scales, remote sensing enables the detection of critical disturbances like fire, drought, and insect outbreaks, supporting sustainable management, carbon monitoring, and ecological modeling.
This session invites research on integrating heterogeneous remote sensing and in-situ observations. We welcome contributions on multi-sensor/multi-platform data fusion, field observations, and methods for transferring information across scales. Submissions may address terrestrial laser scanning, mobile mapping, UAV, airborne, and satellite observations, as well as GeoAI and AI-based approaches (machine learning, deep learning, foundation models, transfer learning, and domain adaptation). We emphasize methodological developments and operational applications, with interest in uncertainty, validation, scalability, transferability, generalization, interpretability, and reproducibility.