Tropical Forest Tree Species Classification Using Meter-Scale Image Data

Accurate identification of tropical tree species is critical for studies of forest habitat, composition, biomass, and ultimately a better understanding of the role trees play in climate variability through carbon uptake. The aim of this study was to derive an accurate classification of a tropical forest study site in Costa Rica using high-resolution imagery. A series of corrections for look and view angle, and the utilization of the DigitalGlobe atmospheric compensation procedure (AComp) provided the study with an accurate surface reflectivity dataset from WorldView-3 imagery. A rule-set object-oriented classification schema defined trees in the study area using ENVI -defined tree canopies through a segmentation of the multispectral image. The results show that select WorldView-3 bands, and WorldView-3-specific vegetation indices, can produce an accurate species-level tree classification in a complex tropical forest.