Video transects of the sea floor collected by the U.S. Geological Survey on Stellwagen Bank during three surveys aboard the R/V Auk, September 2020 to August 2021

Seabed video imagery was collected using two battery-operated, high-definition (1920 x 1080 pixels) GoPro digital video cameras and a SeaViewer high-definition (1920 x 720 pixels) video camera mounted on the SEABOSS. See the cross-references for a description of the SEABOSS (Blackwood and Parolski, 2001; Valentine and others, 2000). The SeaViewer camera had a live feed to the surface. It was mounted a little higher on the frame than the GoPros and was used to observe obstacles, like gill nets and lobster gear, in order to avoid entanglement. One of the GoPro cameras was selected as the primary source of video. In case of camera failure, the additional GoPro and SeaViewer cameras provided backup video.
As the SEABOSS drifted over the seabed, the cameras imaged the seabed at a rate of 60 frames per second from an altitude of ~1 meter. Lighting was provided by two battery-operated Keldan LED dive lights. Each light provided a light flux of up to 9000 lumens at a correlated color temperature of 5000 kelvins. Two parallel red lasers mounted 20 centimeters apart provided a scale in the video images.
GoPro video imagery includes the launch of the SEABOSS, transit through the water column, and the drift over the seabed. The duration of 168 individual video files ranges 2 to 13.5 minutes. Video files are archived in .MP4 format and range in size from 0.2 to 3.0 gigabytes.
These video files are not available as part of this data release. Video files are archived at the USGS Woods Hole Coastal and Marine Science Center, 384 Woods Hole Road, Woods Hole, MA 02543.
The position (latitude and longitude) and time of the start of the video drift was recorded from the navigation receiver by hand in the cruise log book (see the dataset cruise_logs_2020-2021 from the larger work citation of this data release). In some cases the time of the deployment of the SEABOSS is recorded, but in all cases, the time when the system reached the seabed, and when it left the seabed are recorded - which corresponds to the video drift transects. A video transect represents where the SEABOSS drifted and collected video of the seabed, not where the entire video was recorded at a station, which also includes the launch of the SEABOSS and descent through the water column. Latitude and longitude positions recorded by hand in the log book were later adjusted to match the data recorded in the GPS navigation file based on time.
Data were collected from 2020 to 2021 with the process date representing the most recent acquisition year.

The original raw GGA navigation was run through an AWK script (awkauk_rndissue, on a machine running Cygwin - www.cygwin.com) to reformat the latitudes and longitudes into decimal degrees and output the data into a comma-delimited text file with the following columns of information: latitude (decimal degrees out to five decimal places), longitude (decimal degrees out to five decimal places), hour, minute, second, Julian day, year. The Julian day was determined by the original date in the filename of the navigation file as the GGA string does not contain any date information. The script made sure to round the latitude and longitude values to five decimal places in the same way that Excel would round the values. The AWK script also made sure that there were no duplicate records in terms of time, such as two navigation points recorded at 16:38:23. Depending on the boat drift rate, navigation fixes carried out to only five decimal places could have duplicate fixes. The individual navigation files were concatenated into a single file for each field activity in order of acquisition. This process was repeated for all three field activities. This process step and the remaining process steps were performed by the same person - VeeAnn A. Cross.

Another AWK script (awkdrift_pullstartend) was run on the station summary CSV file to extract the following columns of information: station, Julian day of video start, UTC time of video start, Julian day of video end, UTC time of video end, and year. This same process was repeated for all three field activities.

A Python script (driftlog_gui_v6.py) run from Python IDLE version 2.7.18 on Windows 10 was used to extract all the navigation points from the navigation file based on the start and end time of the video drift and combine these into polylines. Ideally, there is a match of the same time (with a corresponding navigation fix) in the GPS file to the start and end video times recorded in the log book. However, if this situation does not exist, then the closest (in time) navigation record to the start and end time of the video transect is used, along with all the navigation points in between those two times. Since location in the log book were updated based on the navigation files, these values should match. These extracted points were then combined into a polyline shapefile. The resulting shapefile has a polyline for every video drift on a single cruise. Based on the navigation system, the coordinate system of the dataset was set to geographic coordinates, WGS 84.

The three individual polyline shapefiles were then combined in QGIS 3.16.7 using the Merge Vector Layers tool. The output dataset was set to a shapefile. Two fields created by QGIS as a result of the merge (layer and path) were removed as unnecessary information.