Drone-Based Rattan Inventory: Comparison with Ground-Based Surveys in a Tropical Forest

Rattans are important non-timber forest products that contribute significantly to the livelihoods of many rural communities across tropical Asia and Africa. However, increasing pressures from overexploitation and deforestation threaten the sustainability of these resources. Conventional field inventories are often labour-intensive and limited in spatial coverage, creating the need for more efficient monitoring approaches. This study evaluated the potential of drone imagery for detecting and inventorying rattan populations in the Hein agroforest located in the South-East Côte d’Ivoire, in tropical West Africa. Four plots containing native rattans were surveyed using both conventional ground inventories and drone-based image analysis. UAV images were processed into orthomosaics using Agisoft Metashape Pro and visually interpreted in QGIS through a grid-based approach. Density estimates obtained from both methods were compared using the Wilcoxon signed-rank test, while their relationship and agreement were assessed using Spearman correlation and Bland–Altman analysis, respectively. Five rattan species were identified at the species level within the study area. Results showed a significant difference between ground and drone-derived density estimates (V = 153, p < 0.001), with the drone method generally underestimating densities (mean bias = −25.22 ramets/ha). Nevertheless, a strong positive correlation was observed between methods (ρ = 0.762, p < 0.001). These findings highlight the potential of UAV imagery as a complementary tool for sustainable rattan monitoring and management.