TY  - EJOU
AU  - Tsvetkova, Mihaela 
AU  - Chanev, Milen 
AU  - Filchev, Lachezar 

TI  - Improving the Estimation of the Main Norway Spruce Forest (<i>Picea abies</i> (L.) Karst.) Parameters from Sentinel-2 Satellite Data
T2  - Revue Internationale de Géomatique

PY  - 2026
VL  - 35
IS  - 1
SN  - 2116-7060

AB  - This study addresses the challenges of traditional forest inventory methods for Norway spruce (<i>Picea abies</i> (L.) Karst.) by leveraging Sentinel-2 multispectral data to derive critical forest parameters, including biomass, stand density, and site class. Remote sensing offers scalable solutions for large-scale monitoring, yet topographic variability and spectral saturation limit the use of empirical vegetation index (VI)-based approaches. The methodology analyzed 43 Norway spruce subcompartments in Bulgaria’s Parangalitsa Reserve using a 2017 Sentinel-2 L2A scene, calculating 24 vegetation indices (e.g., Canopy Chlorophyll Content Index (CCCI), Forest Cover Index (FCI1/FCI2), Normalized Difference Water Index (NDWI) and three biophysical parameters Leaf Area Index, Fraction of Absorbed Photosynthetically Active Radiation, and Fraction of green Vegetation Cover (LAI, FAPAR, FCover). Statistical correlations between spectral indices and inventory data were stratified by slope aspect (north-east vs. south-west) to account for microclimatic influences. Results revealed that the CCCI index showed strong positive correlations with age, height, and stock on NE slopes (r = 0.5–0.6), while FCI1/FCI2 achieved high correlations with site class on SW slopes (r = 0.9–0.93). Negative correlations between FCover/LAI and structural parameters on SW slopes highlighted water stress and shadowing effects. Slope-aspect stratification improved VI-parameter correlations by 12%–18%, demonstrating the necessity of topographic calibration. Overall, the results demonstrate that slope-aspect stratification is the key factor improving the reliability of vegetation index-based estimation of forest structural parameters in complex mountainous terrain. The study validates Sentinel-2’s operational potential for Norway spruce monitoring but emphasizes the need for aspect-specific models to address spectral saturation and environmental variability. These findings advance precision forestry by integrating topographic modulation into remote sensing workflows, enabling scalable forest health assessments and informing climate-resilient management strategies.
KW  - Sentinel-2; remote sensing; vegetation indices; forest parameters; growing stock volume; site index

DO  - 10.32604/rig.2026.079622