Answers are grounded in this paper and cite their sources.
Ethiopia's lack of periodic forest and carbon stock inventory data hinders sustainable forest management and limits access to climate funds. This study explores aboveground biomass (AGB) estimation in the Kofale plantation forests in southeast Ethiopia using Sentinel-2 Multispectral Instrument (S2 MSI) remote sensing data combined with field inventory measurements. A total of 36 sample plots were established, and trees with a diameter at breast height (DBH) ≥ 5 cm were measured for field-based biomass estimation, which was analyzed using allometric equations in R software (version 4.0.5). From the remotely sensed data, 24 independent variables were analyzed to identify the best predictors for aboveground biomass (AGB) estimation. Multi-linear regression with a stepwise selection method was employed to refine the predictor variables, adjusting forward selection (prem = 0.05) and backward elimination (penter = 0.1). The AGB of plantation forests was measured and analyzed, yielding an average field AGB of 64.63 t/ha, with values ranging from 10.95 t/ha to 172.13 t/ha. We developed seven predictive models using remotely sensed data and selected the best model based on performance statistics, including RMSE, AIC, R 2 , and Adjusted-R 2 . The optimal model, constructed from selected vegetation indices red edge normalized difference vegetation index (RE_NDVI3), red edge ratio vegetation index, normalized difference vegetation index (NDVI), excessive greenness index, and ratio vegetation index achieved an RMSE of 26.83 t/ha and an Adjusted-R 2 of 0.701. Among these indices, the NDVI showed a negative correlation with field above-ground biomass due to shadowing effects and age variations. Validation using the leave-one-out cross-validation technique yielded an RMSE of 20.2 t/ha and an R 2 of 0.76, indicating a strong correlation (R = 0.86) between estimated and field-measured biomass. The study also mapped AGB values, ranging from 2.6 to 149.25 t/ha, with variations attributable to species-specific traits, leaf structure, and spectral reflectance. However, discrepancies were observed, likely caused by limitations in species-specific allometric equations, shadowing effects of larger trees, and spectral interference from other species. The study shows Sentinel-2 MSI data improve biomass estimation in plantation forests. Therefore, stakeholders should prioritize plantation establishment and management to maximize their significant biomass and carbon storage potential. • The plantation forests have a comparable forest structure and no prior research in the study area was conducted, the study filled a gap in the understory saturation problem. • Plantation forests have potential biomass and carbon storage potentials. • Estimated aboveground biomass was strongly correlated with the field aboveground biomass (R = 0.86).
No takes yet. Share an insight, caveat, or question.
Gudisa et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: