Session

Phytosociological maps are crucial for biodiversity preservation within the EU, but conventional methods for their update are labor-intensive and slow. Currently, remote sensing data and machine learning algorithms provide substantial opportunities for more effective vegetation mapping. However, their effectiveness depends greatly on the availability and quality of reference data. This research explores the use of drone technology combined with the Braun-Blanquet phytosociological method to improve the quality of reference data and decrease the time and resources needed for field surveys. To achieve these objectives, we standardized drone procedures to obtain high-resolution images of forest and grassland vegetation, aiming to replicate traditional field surveys. The study covered the forest and grassland areas of Mount Conero and Mount Valmontagnana, both significant for their biodiversity within the Natura 2000 network. Our approach uses standardized methods to identify and capture the main variations in vegetation within a chosen area. To guarantee detailed observations, we utilize drones equipped with optical telephoto lenses specifically selected for their ability to capture details even from moderate distances. In both forest and grassland scenarios, drone imagery was analyzed to identify plant species and assess their abundance, emulating the Braun-Blanquet cover-abundance scale. Our study applied multivariate statistical methods, including the Mantel test and supervised k-means classification, to compare drone survey data with traditional methods. Results showed strong correlations between drone-derived data and traditional field surveys in both forest and grassland settings. Principal component analysis showed similar floristic gradients in both datasets, further confirming the effectiveness of the drone approach. These findings highlight the potential of standardized drone procedures to provide reliable reference data, offering a cost-effective alternative to traditional ground-truthing and paving the way for more precise and updated phytosociological maps. Integrating drone technology with traditional phytosociological methods represents a valuable approach for supporting ongoing monitoring and conservation of forest and grassland biodiversity.