LiDAR patch metrics for object-based clustering of forest types in a tropical rainforest

摘要: Abstract Tropical rainforests support a large proportion of the Earth’s plant and animal species within restricted global distribution, play an important role in regulating climate. However, existing knowledge forest types or habitats is relatively poor there are uncertainties quantification carbon stock these forests. Airborne Laser Scanning, using LiDAR, has advantages over other remote sensing techniques for describing three-dimensional structure With respect to habitat requirements different species, can be defined by canopy height, cover vertical arrangement biomass. In this study, patches were identified based on classification hierarchical merging LiDAR-derived Canopy Height Model tropical rainforest Sumatra, Indonesia. Attributes used as inputs k-medoids clustering. The clusters then analysed comparing them with field. There was significant association between field, which arang forests mixed agro-forests contributed most. topographic attributes determine whether structural classes, potentially types, related topography. tallest occurred at significantly higher elevations (>850 m) steeper slopes (>26°) than clusters. These likely remnants undisturbed primary conservation studies estimation. This study showed that LiDAR data map structure, but similarities floristic composition human use histories limit separability determined