National-scale tree species mapping with deep learning reveals forest management insights in Germany

Yang Mu, Jianhua Guo, Muhammad Shahzad, Xiao Xiang Zhu

Abstract: Highlights•We develop a novel ForestFormer model to address tree species mapping challenges.•We generate a Germany-wide tree species map with 83.94% overall accuracy.•We analyze species resistance, biodiversity, and key distribution drivers.•The actionable insights can be integrated into national forest management strategies.