Sea-floor videos and location of bottom video tracklines collected in Long Island Sound, Connecticut and New York, in fall 2017 and spring 2018 by the U.S. Geological Survey, University of Connecticut, and University of New Haven during field activities 2017-056-FA and 2018-018-FA (MP4 video files and polyline shapefile)

Step 1: Collected data.
Two marine geological surveys were conducted in Long Island Sound, Connecticut and New York, in fall 2017 and spring 2018. The R/V Connecticut occupied one of the target sites and the SEABOSS was deployed off the vessel's A-frame on the stern of the ship. The SEABOSS was equipped with a modified Van Veen grab sampler, a Nikon D300 digital still camera with a Photosea strobe, two video cameras (one forward-looking so that a shipboard operator could monitor for proper tow depth and obstacles, and one downward-looking, a Kongsberg Simrad OE1365 in this setup, that overlapped with the field of view of the still camera) with a topside feed, a GoPro HERO4 Black camera recording backup video, and lights to illuminate the sea floor for video and photograph collection. The elements of this particular SEABOSS were held within a stainless-steel frame that measured 1.15 x 1.15 meters. The frame had a stabilizer fin that oriented the system as it drifted over the seabed. Two red lasers were set 20 centimeters apart (both as they are mounted on the SEABOSS frame and as seen in photographs and video on the seabed) for scale measurements. The red laser dots can usually be seen in the sea-floor images and videos depending on the bottom type and distance to the sea floor. A third laser is positioned at an angle so that when it intersects the other lasers, the SEABOSS is at the optimum height (approximately 75 centimeters) off the bottom for a photograph. The winch operator lowered the SEABOSS until the sea floor was observed in the topside live video feed. For those sites that were primarily targeted for a sediment grab, the vessel and SEABOSS then drifted with wind and current for up to a few minutes to ensure a decent image with a clear view of the sea floor was acquired; for those sites that were targeted for both a video transect of the sea floor and a sediment grab, the vessel was navigated along a planned transect for up to an hour. A scientist monitored the real-time bottom video and acquired bottom photographs at points of interest by remotely triggering the Nikon camera shutter. Bottom video was also recorded during the drift from the downward-looking Kongsberg video camera directly to hard drives using an Odyssey7 video recorder. Then, at most sites the winch operator lowered the Van Veen grab sampler until it rested on the sea floor. When the system was raised, the Van Veen grab sampler closed and collected a sample as it was lifted off the sea floor. The sampler was recovered to the deck of the survey vessel where a subsample was taken for grain-size analysis at the sediment laboratory at the USGS Woods Hole Coastal and Marine Science Center. A total of 210 sites were occupied aboard the R/V Connecticut with the SEABOSS: 93 sites were occupied in fall 2017 during field activity 2017-056-FA, and 117 sites were occupied in spring 2018 during field activity 2018-018-FA. Bottom videos were acquired at all 210 sites.

Step 2: Acquired and processed navigation.
During the surveys, WAAS-enabled GPS navigation from a Garmin GPSMAP 76C receiver was logged through a DataBridge data logger and ArcMap GPS. The GPS was set to receive fixes at a 2-second interval in geographic coordinates (WGS 84). Dates and times were recorded in Coordinated Universal Time (UTC). Log files were saved for each Julian day in text format. An AWK script (parse_gprmc17056.awk for the fall 2017 log files and parse_gprmc18018.awk for the spring 2018 log files) was used to parse the GPRMC navigation string from the log files for each survey and create ASCII Comma Separated Values (CSV) text files. The output files were merged for each survey and then reformatted using an AWK script (nav_time_reformat.awk), creating a processed navigation CSV text file for each sampling survey.

Step 3: Processed video files.
A shell script (do_concat_and_TCBurn.sda) was run on the original video files to join the video clips for each site (the Odyssey7 splits clips into less than 4 GB segments) and transcode the video from MOV to MP4 format. The script also created a text file for each survey with the date, start time, and duration of each video recording. The timecode from the videos used to create the text files was unreliable, so the start times in the text files were checked and modified using the GPS time overlay from each video's first frame. The videos were renamed to include LISMaRC (which stands for Long Island Sound Mapping and Research Collaborative), season, year, and date and start time in the ISO 8601 standard (YYYYMMDD T [time separator] HHMMSS Z [Zulu/UTC time]) in the filename.

Step 4: Prepared navigation for video trackline script.
Before creating the video tracklines shapefile, the navigation files were checked for erroneous and missing fixes and prepared for the video trackline script. To prepare the navigation data, first, a column of the original source filename was deleted from each processed navigation CSV text file (DataBridgeNav_GPRMC_ALL.csv for fall 2017 and 2018018_AllDataBridgeNav_parsed.csv for spring 2018) and the time field was formatted to a fixed-length number with six digits in Microsoft Excel 2016 for Mac, which added leading zeros as necessary. Rows with no coordinates were then deleted, and typos in the time and date were corrected as needed. Next, shapefiles were created from each CSV file in Esri ArcGIS (version 10.3.1) and used to identify erroneous fixes, which were deleted from the CSV text files. The shapefiles were also used to identify gaps in the navigation. For the fall 2017 data, missing navigation was interpolated in Esri ArcGIS to get 2-second navigation for site NB32 (clip 448) where the navigation dropped out for 154 seconds on Julian day 333. To do this, a line segment was created between the last and first known fixes around this gap (between GPS times 22:49:38 and 22:52:12). The Construct Points editing tool was used to create 153 evenly spaced points along the line segment so that a point was created for each second. The coordinates of the points were calculated using the Calculate Geometry tool (Property=X Coordinate of Point and Y Coordinate of Point; Use coordinate system of the data source=WGS 1984; Unit=Decimal Degrees). The interpolated points with even seconds were selected and exported as a CSV file (points with even seconds were extracted so that the navigation was interpolated at a 2-second interval), the coordinates were rounded to seven decimal places in Microsoft Excel 2016 for Mac, and the points were added to the processed navigation CSV text file. For the spring 2018 survey, the navigation was also logged using ArcMap GPS, so shapefiles of the ArcMap GPS logs were used to fill in gaps. Any feature from the ArcMap GPS logs not within 0.1 meters of the original navigation was selected, exported, and added to the navigation CSV file. Concurrent fixes were deleted, and any gaps 6 seconds or larger were filled in using data from the ArcMap GPS logs if available (these points intersect other fixes and were not originally exported). Finally, the fields in the navigation CSV text files (now named DataBridgeNav_GPRMC_ALL_for_videos.csv for fall 2017 and 2018018_AllDataBridgeNav_parsed_w_ArcMap_GPS_Logs_for_videos.csv for spring 2018) were reordered and formatted to be used with the video trackline script. This process step and the subsequent process steps were performed by the same person, Emily Huntley.

Step 5: Created a CSV file of the bottom video trackline points.
A Jupyter Notebook Python script (Video_trackline_prep_WORKING_v2.ipynb) was run for each survey to create a CSV file of the bottom video trackline points by extracting the navigation data for each video drift using information from the start times/durations text files. The script reads the video start time and duration from the text files, calculates the video end time, extracts the navigation points that fall within those start and end times, and exports the navigation points to a CSV file.

Step 6: Created the final bottom video tracklines shapefile.
Point shapefiles were created for each survey using the bottom video trackline points CSV files in Esri ArcGIS (version 10.3.1). The Points to Line tool was then run for each survey with the video trackline points as the input features and the video filenames as the line field to create a polyline shapefile of the video tracklines. Two videos acquired in fall 2017 (clips 424 and 494) were less than 1 second and were not mapped. XTools Pro (version 12.0) for Esri ArcGIS was used to rename, reorganize, and add new fields (Table Operations - Table Restructure) to the polyline shapefiles, including an attribute for the site number of the video trackline (FIELD_NO), start time of the bottom video drift in UTC (STARTTIME), end time of the video drift in UTC (ENDTIME), Julian day of collection (JD), date of collection (DATE), year of collection (YEAR), trackline length in meters (LENGTH_M), camera used (CAMERA), survey ID (FA_ID), sampling device used to collect the video (DEVICE_ID), survey vessel (VEHICLE_ID), and University of Connecticut's NURTEC site name for the spring 2018 survey (COMMENTS). XTools Pro changed some of the bottom video trackline features from singlepart to multipart features if they overlapped or intersected themselves. To correct this, the Spatial Join tool was run for each survey with the original bottom video tracklines shapefile as the input features and the updated tracklines shapefile as the join features using the intersect match option to add the updated attributes to the original singlepart features. Unnecessary fields created when running the Spatial Join tool were deleted (i.e., Join_Count and TARGET_FID). Next, the bottom video tracklines shapefiles for each survey were joined with the trackline points shapefiles to add the start and end times of the video drifts (STARTTIME and ENDTIME, respectively), Julian day of collection (JD), and date of collection (DATE). Then, the trackline length (LENGTH_M) was calculated using the Calculate Geometry tool (Property=Length; Use coordinate system of the data frame=WGS 1984 UTM Zone 18N; Unit=Meters). The bottom video tracklines shapefiles for each survey were then joined with the survey logs to assign the site number (FIELD_NO) of each video trackline and, for the spring 2018 data, NURTEC site name (COMMENTS). Three tracklines from fall 2017 were for videos used for camera calibration (clips 418, 443, and 451) and were removed. The trackline shapefiles for each survey were combined using the Merge tool and reordered using the Sort tool to sort the video tracklines in chronological order (YEAR, JD, and STARTTIME sort fields in ascending order). Finally, the bottom video line names (LINENAME) were updated to match the new video filenames, which include LISMaRC (which stands for Long Island Sound Mapping and Research Collaborative), season, year, and date and start time in the ISO 8601 standard in the filename.