 This paper examines the use of various machine learning algorithms to accurately estimate the above-ground biomass, AGB, of four different forest types in Yunnan province, China. The authors found that random forest, RF, was the best model for all four forest types, with an overall R2 value of 0.503 and an average RMSE of 52.335 megagrams Ha-1. Additionally, they determined that the most important variables for each type of forest differed significantly, with DEM being the most important variable for coniferous forests, evergreen broadleaves, and mixed forests, while the vegetation index was the most important variable for deciduous broadleaves. This article was authored by Tianbao Huang, Guanglong O, Yongwu, and others.