Orthomosaics of Pea Island National Wildlife Refuge DUNEX Site, North Carolina in September and October 2021

Digital Surface Model was created in Agisoft Metashape v. 1.8.1 using the following general steps (see Over and others, 2021 for a more detailed explanation of methodology): 1. Imagery (located in larger work citation) and GCPs (located in larger work citation) were brought in using the NAD83(2011)/UTM Zone 18N NAVD88 datums. 2. GCPs were manually inspected on the images and blurry images were removed. GCPs that were moved during surveys, in less than three images, or were consistently in blurry images were made into checkpoints. 3. Accuracies for the GCPs were set 0.01 m or 0.10 m depending on quality of the images. Photos from each survey were aligned in 4D at high accuracy (the pixels are not subsampled for increased processing speed) with a keypoint limit of 40,000 and unlimited tie points. 4. The alignment process uses the positions and matching pixels between images to create point clouds and put the imagery into a relative spatial context. Metashape software refined and optimized the camera positions and lens model in a least squares sense to minimize the reprojection error or the distance between the measured points and the software created points. Project supported gradual selection and optimization of: Ru = 11, Pa = 6, and Re = 0.3. Individual survey GCP and product errors are calculated separately and reported in the Entity and Attribute Information. 5. A dense point cloud was generated for each survey in a separate Metashape chunk using high quality (images are not subsampled) and a low frequency filtering algorithm. The dense point cloud was then edited to remove noise by filtering by point confidence before generating an interpolated digital elevation model (the software calls all models elevation models, but technically it is a surface model because it includes vegetation/canopy returns, a terrain model would be only bare earth returns). 5. The digital surface model was used to build an orthomosaic, which was then exported from the software at a resolution of 10 cm and in NAD83(2011)/UTM Zone 18N (ESPG:6348).

Orthomosaics were turned into Cloud Optimized GeoTIFFs (COG) using gdal_translate with the following command:for %i in (.\*.tif) do gdal_translate %i .\cog\%~ni_cog.tif -co COMPRESS=LZW -co COPY_SRC_OVERVIEWS=YES -co BIGTIFF=YES -co TILED=YES (v. 3.1.4 accessed October 20, 2020 https://gdal.org/). Where i was the name of each geoTIFF section and a folder named cog is in the directory.