Rubber plantations in Southeast Asia contribute significantly to deforestation and habitat degradation when replacing natural forests. Nevertheless, well-managed plantations can support sustainability objectives such as climate change mitigation, renewable energy development, and sustainable land management. Aboveground biomass (AGB) is a critical component in assessing carbon storage and potential emissions. This study aimed to develop an optimal statistical model for estimating AGB in rubber plantations using vegetation indices derived from Landsat 8 OLI-TIRS imagery. Linear and non-linear regression analyses were conducted, incorporating vegetation and landscape properties to evaluate their influence on AGB estimation. The resulting model provides a basis for spatial AGB mapping, including differentiation by plantation age and landscape category. Statistical results demonstrated strong agreement between estimated and observed AGB values, with differences below 1%. Non-linear regression outperformed linear approaches, as evidenced by a fivefold increase in the coefficient of determination (R²). The findings highlight the effectiveness of remote sensing-based vegetation indices for accurate AGB estimation in rubber plantations.