NCCOS Mapping: Comprehensive Seafloor Substrate Mapping and Model Validation in the New York Bight from 2017-10-20 to 2019-03-27 (NCEI Accession 0204737)

Multibeam sonar data (seabed elevation) and acoustic imagery data (backscatter) were collected using a 400khz Reson 7125 multibeam echosounders (MBES) on the NOAA Ship Ferdinand Hassler (2013) and the NOAA Ship Nancy Foster (2017-2018). The Foster also collected Em710 data during the 2017 survey simultaneously with the Reson 7125 at 100 khz to analyze sub surface sediment textures, and 400 khz Em2040 multibeam which replaced the 7125 in 2018. The Hassler survey (H12629) was collected, processed, and archived by the NOAA Office of Coast Survey (OCS) and downloaded from NCEI (https://www.ngdc.noaa.gov/nos/H12001-H14000/H12629.html). The multibeam datasets from the Foster were preliminarily processed by NCCOS who conducted initial processing to corrected for motion, sound speed, and other major artifacts using CARIS HIPS and SIPS workflow. The final data from the Foster 2017 survey were processed and reviewed to ensure IHO S-44 specs by SolMar Hydro, using the Smoothed, Best Estimated Trajectory (SBETs) work flow from the Real Time Kinematic Marinestar GPS system. The 2018 data was collected to resurvey areas from the 2017 survey where data quality was poor due to bad weather and seastate, however that survey was affected by bad weather as well and was not used in the habitat mapping analysis or to supersede existing survey data. The backscatter from the multibeam datasets were mosaicked using the Fledermaus Geocoder Toolbox (FMGT) software. Once the bathymetric and backscatter datasets were finalized, a suite of morphometrics were generated using the Bathymetry and Reflectivity-based Estimator for Seafloor Segmentation (BRESS) software, and from an R script created by NCCOS. The morphometrics were used as predictors for the Boosted Regression Tree (BRT) predictions of the benthic substrate, biotic cover, and geoform components using a pixel based classification method. The Foster 2017 mission also collected Simrad EK60 split beam echo sounder (SBES) data to analyze fish and other biota in the water column. These datasets were cleaned and converted into a CSV file using Echoview software then processed into a point density data using an R script. The point density data could then be projected into ArcMap as a shapefile and was then interpolated into a raster for fish distribution analysis. NCCOS returned to the NYWEA on the R/V Tiki XIV in 2018 to collect ground validation and accuracy assessment data using the Seaviewer system and Gopro Hero 4 underwater cameras and sediment grabs using a modified van Veen sampler. The video and contents from the sediment grabs were used to correct the habitat predictions and estimate biogenic cover. Sediment grab samples were sent to TDI-Brooks to be sieved and separated into the Wentworth phi and grainsize classes. The sediment grab data was also projected onto maps of grainsize distribution using Bayesian Kriging. Once the bathymetric models and grab sample data were finalized for the NYWEA priority area, that information were compiled into a regional substrate and hardbottom model using BRT and Max-Ent, respectively. For more details, see Battista et al. (2019).