Characterizing carbon stocks in aboveground biomass (AGB) is particularly challenging in Mediterranean ecosystems due to their strong spatial heterogeneity. This is the case of Mediterranean oak woodlands of the Iberian Peninsula, which are structurally and ecologically heterogeneous ecosystems shaped by long-term human coevolution. Originally dominated by continuous oak forests, these systems have been progressively converted by human management into agrosilvopastoral systems known as dehesas in Spain and montados in Portugal. In the present they are a mosaic landscape with variable oak density depending on land management. Consequently, AGB patterns in these land-use systems reflect the combined influence of human management and environmental conditions. In this study, we evaluated the potential of Landsat-8 imagery to estimate different components of AGB, including woody biomass (AGB_wood), foliage and shrubs (AGB_foliage), and total perennial biomass (AGB_total). We combined field data from 47 experimental plots with two cloud-free summer Landsat-8 images using a spatially structured cross-validation approach. Our results show that AGB_foliage (R² = 0.36, rRMSE = 0.37) and AGB_total (R² = 0.35, rRMSE = 0.41) are more accurately predicted than AGB_wood (R² = 0.16, rRMSE = 1.02), mainly due to pixel-mixing effects associated with the coarse spatial resolution of Landsat data in open woodlands. Red, NIR, and SWIR bands were identified as the most informative predictors across biomass components. Instead of producing a highly accurate static estimate of AGB, our results demonstrate that Landsat-8 imagery can be useful for tracking changes in AGB with moderate accuracy, offering a practical approach to monitor carbon stock changes driven by management, disturbance, and climate variability. Landsat-8 imagery can also be combined with inventory data to improve the accuracy of carbon stock estimates.

This research was financed by the REMAS project (co-financed by the Interreg Sudoe Programme through the European Regional Development Fund (ERDF), grant number “SOE3/P4/E0954” and PRR/IAPMEI/Aviso 02/C05-i01/2022, CIRCULARTECH (Contribute to solving a problem that humanity has faced since the industrial revolution: waste) – Carbon WISE