Drone and ground-truth data collection, image annotation and machine learning: A protocol for coastal habitat mapping and classification.

Aerial drone imaging is an efficient tool for mapping and monitoring of coastal habitats at high spatial and temporal resolution. Specifically, drone imaging allows for time- and cost-efficient mapping covering larger areas than traditional mapping and monitoring techniques, while also providing more detailed information than those from airplanes and satellites, enabling for example to differentiate various types of coastal vegetation. Here, we present a systematic method for shallow water habitat classification based on drone imagery. The method includes:•Collection of drone images and creation of orthomosaics.•Gathering ground-truth data in the field to guide the image annotation and to validate the final map product.•Annotation of drone images into - potentially hierarchical - habitat classes and training of machine learning algorithms for habitat classification.As a case study, we present a field campaign that employed these methods to map a coastal site dominated by seagrass, seaweed and kelp, in addition to sediments and rock. Such detailed but efficient mapping and classification can aid to understand and sustainably manage ecologically and valuable marine ecosystems.

Quantifying seaweed and seagrass beach deposits using high-resolution UAV imagery.

Macroalgae and seagrass wash ashore by tidal waters and episodic events and create an ocean-to-land transport of carbon and nutrients. On land, these deposits (beach wrack) are consumed by macrofauna, remineralized by microorganisms, or washed back to the sea, during which recycling of carbon and nitrogen affect the biochemical cycles in coastal zones. Manual quantification of beach wracks is time-consuming and often difficult due to complex topography and remote locations. Here, we present a novel method using Unoccupied Aerial Vehicle (UAV) photogrammetry combined with in situ measurements of carbon and nitrogen contents of wrack to quantify marine carbon and nutrient deposits in beach zones. The UAV method was tested against placed cubes ranging from 125 to 88,218 cm3 and demonstrated a high accuracy (R2 > 0.99) for volume acquisition when compared to manual measurements. Also, the UAV-based assessments of the cross-sectional area of beach deposits demonstrated a high accuracy when compared to manual and high-precision GNSS (Global Navigation Satellite System) measurements without significant differences between the methods. This demonstrated that UAVs can provide detailed spatial maps, three-dimensional (3D) surface models, and accurate volumetric assessments of beach wrack deposits. In three case studies, combined with carbon and nitrogen measures, total organic carbon and nitrogen deposits in beach wracks were quantified ranging from 4.3 to 9.7 and from 0.3 to 0.5 kg per meter coastline, respectively. In conclusion, this UAV method demonstrated an effective tool to quantify ecosystem carbon and nitrogen deposits relevant to ecosystem assessments and quantification of blue carbon stocks. The method is optimal when the terrain below beach wrack deposits is known, as in the case with before-and-after or repeated surveys. Further, UAVs display strong time- and cost-effective advantages over manual methods which is amplified with increasing project scale. We propose it as a valuable method for multiple scientific and commercial applications related to environmental monitoring and management, including marine resource exploration and exploitation.