Assessment of Oak Groves Conservation Statuses in Natura 2000 Sacs with Single Photon Lidar and Sentinel-2 Data

Among the main objectives of Natura 2000 Network sites management plans is monitoring their conservation status under a reasonable cost and with high temporal frequency. The aim this study to assess ability single-photon light detection ranging (LiDAR) technology (14 points per m2) Sentinel-2 data classify oak forests in four special areas Navarra Province (Spain) that comprise three habitats. To capture variability within habitats, we first performed random stratified sampling based on measured field, canopy cover, terrain slope height. Thereafter, compared two metric selection approaches, namely Kruskal–Wallis Dunn tests, machine learning classification methods, forest (RF) support vector (SVM), statuses using LiDAR data. best-fit model, which included only metrics, was obtained method, an overall accuracy after validation 83.01%, 75.51%, 88.25% for Quercus robur (9160), pyrenaica (9230), faginea (9240) respectively. models include six structural diversity indices (LiDAR height evenness index, LHEI, LHDI) cover (FCC) being most relevant ones. inclusion NDVI index from image did not improve significantly. This approach demonstrates its value classifying subsequently mapping groves other habitat at regional scale, could serve more effective biodiversity