Estimating above-ground biomass (AGB) through the utilization of optical remote sensing and synthetic aperture radar (SAR) data presents a practical approach for the long-term monitoring of forest quality in a biosphere reserve. The significance of forest biomass lies in its capacity for substantial carbon storage and its contribution to mitigating global climate change. This study introduces an integrated methodology that relies on multi-source remote sensing data and machine learning algorithms to quantify above-ground biomass (AGB) and analyze the spatial distribution of diverse forest types within the Western Nghe An Biosphere Reserve in Vietnam. A total of 169 sample plots werecollected during on-site surveys conducted in 2022. Out of these, 118 plots were employed for machine learning modeling to estimate AGB, while the remaining 51 plots were reserved for result validation. The model's performance was evaluated and confirmed using metrics such as the coefficient of determination (R2), root mean square error (RMSE), and mean absolute error (MAE). The results demonstrate the effectiveness of the model, with RMSE and MAE exhibiting errors below 30 Mg/ha and an R2 value of approximately 0.81 for AGB estimation. Through a comprehensive analysis of remote sensing data and machine learning models, this study provides a fresh and insightful perspective on AGB estimation models based on multi-source remote sensing technology for tropical forests within the World Biosphere Reserve.