Multibeam backscatter data collected in the vicinity of Buffalo Reef, Michigan, within Lake Superior during USGS Field Activity 2018-043-FA using a dual-head Reson T20-P multibeam echosounder (8-bit GeoTIFF, UTM Zone 16N, NAD 83, 2-m resolution)

PROCESSING STEP 1: CARIS HIPS DATA PROCESSING.
Multibeam bathymetry processing within CARIS HIPS (version 10.4) during the survey consisted of the following flow:
1) Vessel configuration files were created in CARIS for the port ("M"=main, or primary) and starboard ("S"=secondary) sonars (RVRafael_DualT20P_M.hvf, and RVRafael_DualT20P_S.hvf) which includes, linear and angular installation offsets for each T20-P unit as well as vendor specified uncertainty values for each of the survey sensors.
2) A CARIS HIPS project (version 10.4) was created with projection information set to Universal Transverse Mercator (UTM) Zone 16N, WGS 84. Separate HIPS projects were created for the Port (M), and Starboard (S) line files using the two vessel configuration files in #1 above.
3) Each Reson s7k file (M and S) was imported to the new CARIS projects (M and S respectively) using the Import/Conversion Wizard.
4) Delayed heave data from raw POS MV files were used to update HIPS survey lines using the import auxiliary data function.
5) Navigation was reviewed and edited as needed using the Navigation Editor tool.
6) Sound velocity correction was applied using the CARIS algorithm, a master SVP file containing all the sound velocity profiles collected during the survey and specifying the nearest in distance method, delayed heave source, and use surface sound speed.
7) Data were merged selecting no tide and the delayed heave source.
8) 2-m resolution Swath Angle Weighted (SWATH) surfaces were created to incorporate all the files (M and S) as they were processed, and the SWATH surfaces were reviewed for inconsistencies and anomalies.
9) The swath and subset editors were used to remove spurious points through manual editing and filter application, and the refraction editor was used to adjust sound speed values in areas where velocimeter data did not adequately correct depth profiles, which were obviously influenced by local anomalies in speed of sound through the water column.
10) Survey lines adjusted for refraction anomalies were remerged, and the respective SWATH surfaces were recomputed to reflect the changes. Processing during the survey was primarily focused on QA/QC during acquisition. Editing processes did require trial and error, and were at times iterative. The contact person for this and all subsequent processing steps below is Brian Andrews.

PROCESSING STEP 2: APPLY POST PROCESSED SBET FILES AND EDIT SOUNDINGS.
Post-survey processing within CARIS HIPS (version 10.4) consisted of the following flow: 1) Post-processed navigation, vessel attitude, and GPS height data from POSPac Smoothed Best Estimate of Trajectory (SBET) files, and post-processed RMS attitude error data from POSPac smrmsg files were used to update HIPS survey lines using the import auxiliary data function.
2) Navigation source was set to Applanix SBET, and navigation was reviewed and edited as needed using the Navigation Editor tool.
3) GPS tide was computed using delayed heave data, the vessel water line, and a single datum value of 0 m (vertically referencing the data to the WGS 84 Ellipsoid).
4) Sound velocity correction was reapplied using the CARIS algorithm, the master SVP file containing all the sound velocity profiles collected during the survey and specifying the nearest in time method, delayed heave source, and use surface sound speed.
5) Data were remerged selecting the GPS tide and delayed heave sources.
6) 2-m resolution SWATH surfaces using both the M and S files were created using the Swath Angle method and a maximum footprint of 9.
7) Additional editing was conducted using the swath and subset editors to minimize inconsistencies and artifacts, and the SWATH surface was recomputed to reflect the changes. Finally, small "no data" holidays were filled using the "Fill Raster Holiday" tool and a 5x5 cell filter using data from a minimum of 5 neighboring cells.

PROCESSSING STEP 3: QPS FMGT PROCESSING
A backscatter mosaic were created using QPS FMGT (v. 7.7.5) processing software. Raw s7k files containing the time series values, were paired with the HIPS line files that contained the depth data to create a two meter resolution time series backscatter mosaic that followed the steps below.
1) Create new FMGT project for the Buffalo Reef survey area using UTM 16N, WGS 84 coordinate system.
2) Set processing parameters for T20-P Sonar (defaults)
3) Set filter method to flat 300
4) Add source/paired files using the s7k for the backscatter files, and the "processed.depth" files from the HIPS line file directory. All the s7k and HIPS files were added by Julian day. Once the line files were added to the project a two meter draft mosaic was created using the "time-series" as the backscatter source for each Julian day for review.
5) A final two-meter mosaic was created and visually reviewed for inconsistencies or anomalies. Individual line files were moved to up or down in the mosaic order to increase quality if needed. The "Backscatter Adjustment" tool was used to increase or decrease intensity of individual lines as required to match adjacent lines.
6) The resulting mosaic was exported out of FMGT as an 8-bit gray-scale GeoTIFF.

Added keywords section with USGS persistent identifier as theme keyword.