Seeing Trees from Drones: The Role of Leaf Phenology Transition in Mapping Species Distribution in Species-Rich Montane Forests

The complex topography of subtropical montane forests favors the coexistence diverse plant species, making these species-rich a high priority for biodiversity monitoring, prediction, and conservation. Mapping tree species distribution accurately in areas is an essential basis research often challenging due to their structure. Remote sensing has widely been used mapping but relatively little attention paid forests. In this study, capability high-resolution UAV remote imagery six standing dead trees, canopy gaps was tested forest at elevation 816~1165 m eastern China. Spectral, spatial geometrical, textural features specific phenological period when obvious color differences among leaves different were extracted, four object-based classification algorithms (K-nearest neighbor (KNN), regression (CART), support vector machine (SVM), random (RF)) classification. We found that: (1) using low-cost RGB leaf great application potential with terrain. (2) Plant spectral senescence contributed significantly classification, while contribution limited. highest accuracy 83% KNN combination geometrical features. (3) Topographical complexity had significant impact on distribution. generally higher steep areas, especially low slope area.