This dataset contains high-resolution (5 cm/pixel) orthomosaics and digital elevation models (DEMs) from unoccupied aerial vehicle (UAV) surveys of biogenic structures in the German Wadden Sea. Two Pacific oyster reefs (Kaiserbalje, Nordland) and one blue mussel bed (Nordstrand) were monitored between 2020 and 2022. The data, processed via structure from motion (SfM) and georeferenced, are provided as raster files (*.tiff), ready for GIS analysis. The Random Forest (RF) classification shapefiles support the mapping of biogenic structures. This dataset facilitates research on biogenic structure growth, sediment dynamics, and geomorphological processes in intertidal environments

Data Collection: This dataset was collected using UAV surveys with a DJI Phantom 4 Pro equipped with an RGB camera. Flights were conducted at a maximum altitude of 40 m, achieving a ground sampling distance (GSD) of 1.2 cm/pixel. Surveys were performed in 2020 and 2022 at three sites in the German Wadden Sea: two Pacific oyster reefs (Kaiserbalje and Nordland) and one blue mussel bed (Nordstrand). Each survey occurred around low tide and incorporated ground control and check points, whose coordinates were measured using a Stonex S900A dGPS in Real-Time Kinematic (RTK) mode, with horizontal and vertical accuracies of 0.008 m and 0.015 m, respectively.Data Processing: The UAV imagery was processed into georeferenced digital elevation models (DEMs) and orthomosaics (.tiff) using Structure from Motion (SfM) photogrammetry. These datasets are ready for GIS analysis. A Random Forest (RF) machine learning algorithm was applied for classification, using structural and spectral features as training data, resulting in shapefiles that map biogenic structures.