Orthomosaic imagery for Whiskeytown Lake and surrounding area, 2019-06-03

Aerial imagery was collected by the 111th Group, Inc. on 2019-06-03 using a Nikon D-850 mounted in a nadir orientation. For this dataset, the images were acquired between 22:09 and 23:34 Universal Coordinated Time (UTC) (15:09 to 16:34 Pacific Daylight Time (PDT)). The imagery was acquired from an approximate altitude of 610 meters (2,000 feet) above ground level, to produce a nominal ground sample distance (pixel size) of 5 centimeters (2 inches). A total of 1,881 images were acquired in the survey area, with an approximate 70 percent along-line image overlap (forelap) and 50 percent side-to-side overlap (sidelap). An onboard dual-frequency GPS receiver was used to record the precise shutter time and approximate position of each image.

Ground control was established after the image acquisition by measuring the positions of approximately 43 photo-identifiable marks or objects using survey-grade post-processed-kinematic (PPK) GPS equipment. All measurements were referenced to a static base station operating on a temporary benchmark within the survey area. The position of the base station was established using the National Geodetic Survey Online Positioning User Service (OPUS).

Structure from Motion (SfM) processing techniques were used to create the Digital Surface Model and Orthomosaic in the Agisoft Photoscan/Metashape software package using the following work flow: 1. Imagery was geotagged using a combination of post-processed-kinematic (PPK)GPS positions and DGPS positions from the onboard single-frequency GPS unit. 2. Initial image alignment was performed with the following parameters - Accuracy: 'High'; Pair selection: 'reference', 'generic'; Key point limit: 0 (unlimited); Tie point limit: 0 (unlimited). 3. Optimization and lens calibration parameters f, cx, cy, k1, k2, k3, p1, and p2 were included. 4. Ground control points (GCPs) previously measured with PPK GPS were manually identified and placed in the imagery. 5. Sparse point cloud error reduction was performed using an iterative gradual selection and camera optimization process with the following parameters: Reconstruction Uncertainty: 10; Projection Accuracy: 3; Reconstruction Error: 0.3. 6. A dense point cloud was created using the 'high' accuracy setting, with 'aggressive' depth filtering. 7. Ground and low-noise points were identified using the 'Classify Ground Points' tool with the following parameters: Max. Angle: 15 degrees; Max. Distance: 0.5 meters; Cell Size: 5 meters. Due to the prevalence of vegetation and tree cover in most areas, it is not expected that this step removed all of the vegetation points. The dense point cloud was exported to LAZ format. 8. An initial Digital Surface Model (DSM) with a native resolution of 0.116 meters per pixel was created using the ground classified points from the dense point cloud. An exterior boundary was digitized and used as a clipping mask to exclude obvious edge artifacts and large areas of interpolation. The DSM was exported to a GeoTIFF format with a 0.250-meter pixel resolution. 9. An RGB orthomosaic with a native resolution of 0.058 meters per pixel was created using the DSM as the orthorectification surface. The orthomosaic was exported to a GeoTIFF format at two resolutions: 0.060-meter pixels, and 0.250-meter pixels. 10. The RGB orthomosaic was converted to a cloud optimized GeoTIFF format for compatibility with cloud storage services using the GDAL software package. The orthomosaic was scaled to byte values and compressed using JPEG compression, with a quality of 90. The higher-resolution orthomosaic was divided into an east and west tile to reduce the file size, while the lower-resolution version was kept as one file.

Edited metadata to add keywords section with USGS persistent identifier as theme keyword. No data were changed.