Digital Surface Model of the Lost River Fault, Idaho, from 1966 Aerial Photographs

This data release contains a point cloud and raster image of a digital surface model (DSM) that covers a portion of the Lost River fault in Idaho that ruptured in the 1983 M6.9 Borah Peak earthquake. The DSM provides a high-resolution, pre-earthquake record of the landscape from 1966. It is constructed using structure-from-motion photogrammetry software and aerial photographs taken in 1966. The processing report, list of photographs, and list of ground control points (GCPs) used to georeference the DSM are also included.
 
Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.
 
Photographs:
The photographs used were taken on July 17, 1966, by the U.S. Geological Survey and are available in high-resolution to download from https://earthexplorer.usgs.gov/ (project VBNS0). The mostly cloud-free images were acquired at 1:40,000 scale from an altitude of 20,000 ft with a focal length of 152.71 mm on a Fairchild cartographic camera type T-11 No. 54-016 equipped with a Bausch & Lomb Planigon lens No. XF6778. The camera calibration report and photograph metadata are available from https://earthexplorer.usgs.gov/. More information on the Aerial Photo Single Frames archive is available here: https://www.usgs.gov/centers/eros/science/usgs-eros-archive-aerial-phot…
 
Processing Methods:
The 234 photographs used to make the DSM were processed in Agisoft Metashape version 2.0.2. Photograph borders were masked prior to being aligned on the “high” setting. Photographs do not contain geospatial information. GCPs were added to georeference the dataset. The elevation of the GCPs must be from prior to the earthquake to avoid adding artifacts to the dataset because the earthquake caused vertical deformation of the ground surface. To achieve this, we located 38 National Geodetic Survey (NGS) benchmarks that were measured prior to the 1983 earthquake. We obtained the raw leveling data from the NGS because benchmark orthometric height is reported in a reference system that has been corrected for earthquake deformation. Additionally, because benchmarks are too small to be viewed directly on the images, we located them in Google Earth and used their location in Google Earth to identify their locations on the images. We then excluded benchmarks with high error because benchmarks are not always located on the ground surface, leaving 18 benchmarks to georeference the model. Geospatial error in the final model is 1.7 m in X, 1.8 m in Y, and 1.5 m in Z. Depth maps were constructed on the “high quality” setting with aggressive filtering, and a 3D model and point cloud were made with the “high quality” settings. The resulting point cloud contains >616,000,000 points. The DSM constructed from the point cloud in Agisoft Metashape covers approximately 1600 km2 with ~0.4 pts/m and ~1.8 m/pixel and no interpolation. See the attached processing report for all parameter values and processing settings.
 
The topography represented by the raster tif and point cloud laz files may contain holes, wayward points, and other noise resulting from using legacy data (the 1966 images).
 
Description of Data:
1966_v6_hidensecld_PC_NAD83-2011UTM12.laz
Point cloud containing location (X, Y, and Z) and color (red, green, blue) for each point in NAD83(2011) / UTM zone 12N (EPSG:6341) coordinate system.
 
1966_v6_hidensecld_DSM_NAD83-2011UTM12.tif
Digital Surface Model (DSM) raster file of the 1966 topography model in NAD83(2011) / UTM zone 12N (EPSG:6341) coordinate system. Elevation values in meters.
 
Agisoft_processing_report_1966_v6_hidensecld_20240607.pdf
The processing report automatically generated by Agisoft Metashape software. Contains all processing parameters used to generate the 1966 topography model.
 
photographs.csv
List of photographs used to make the 1966 topography model.
Photograph: The “Photo ID” of the image as listed in the “Aerial Photo Single Frames” dataset of EarthExplorer, available at https://earthexplorer.usgs.gov/
 
ground_control_points.csv
List of ground control points used to georeference the 1966 topography model.
Label: The benchmark label, called the “Permanent ID” or “PID” at NGS
Enable: Whether or not the GCP is used in the georectification. 1 = used; 0 = not used.
X_Longitude: Longitude in decimal degrees in NAD83 HARN (EPSG:4152) coordinate system.
Y_Latitude: Latitude in decimal degrees in NAD83 HARN (EPSG:4152) coordinate system.
Z_Altitude: Elevation in meters
Error_m: Total Root Mean Square Error (RMSE) in meters
X_error_m: Error in the x coordinate in meters
Y_error_m: Error in the y coordinate in meters
Z_error_m: Error in the z coordinate in meters
X_est_lon:  Estimated location of the X coordinate in decimal degrees in NAD83 HARN coordinate system
Y_est_lat: Estimated location of the Y coordinate in decimal degrees in the NAD83 HARN coordinate system
Z_est_m: Estimated location of the Z coordinate in meters