Multi-Sensor Aboveground Biomass Estimation in the Broadleaved Hyrcanian Forest of Iran

A combination of Sentinel-2 (S2) and Sentinel-1 (S1) datasets using the k-Nearest Neighbor algorithm achieved the highest accuracy for estimating aboveground biomass (AGB) in the broadleaved Hyrcanian Forest, with an R² of 0.57 and a relative root mean square error (rRMSE) of 14.68%. This approach outperformed individual datasets, demonstrating that integrating optical data (Landsat-8 or Sentinel-2) with synthetic aperture radar (SAR) data improves AGB estimation in this UNESCO-listed forest ecosystem. The study also identified FAPAR and red-edge NDVI as the most correlated Sentinel-2 metrics for biomass prediction.

Ronoud, G., Fatehi, P., Darvishsefat, A. A., Tomppo, E., Praks, J., & Schaepman, M. E. (2021). Multi-Sensor Aboveground Biomass Estimation in the Broadleaved Hyrcanian Forest of Iran. Canadian Journal of Remote Sensing, 47(6), 818–834. https://doi.org/10.1080/07038992.2021.1968811