Processing steps 1 and 2 produced a preliminary bathymetry data set. An error in the University of New Brunswick (Clarke, 1998, see cross reference) processing software that resulted in the multibeam observations not being properly projected onto a plane was discovered in 2003. A second error in the gridding algorithm was discovered in 2007, requiring a shift of 1/2 grid cell to northwest. This processing step uses the corrected software. Processing to further edit the data, correct for tides, and to produce the final grid of the data included:
1. Correct errors in soundings due to sound refraction, caused by variations in sound velocity profile, using the SwathEd refraction tool. These artifacts can be recognized in a cross-swath profile of a relatively flat patch of sea floor. When viewing the swath data across a profile, the sea floor will appear to have a "frown" or "smile" when in fact the data should be flat across the profile. Insufficient and/or erroneous sound velocity information, which is usually due to widely spaced or non-existent velocity profiles within an area, results in an under or over-estimate of water depth which increases with distance from the center of the swath. For a discussion of how this effect can be recognized in a swath bathymetric data file, see
<
http://www.omg.unb.ca/AAAS/UNB_Seafloor_Mapping.html>.
2. Remove erroneous soundings that were not edited in the field using the SwathEd program.
3. Reference the bathymetric data to mean lower low water using the observed tidal data at Sandy Hook, NJ (NOAA tide station 8531680). The measured elevations were adjusted for fluctuations in sea level during the survey by subtracting tidal elevations predicted by a tidal model and low-frequency sea level observed at the National Oceanic and Atmospheric Administration Sandy Hook tide station located at 40 degrees 28 minutes N., 74 degrees 0.6 minutes W. The tidal model utilized nine constituents derived from a 4-month bottom pressure record obtained at Station A, located at 40 degrees 23.4 minutes N., 73 degrees 47.1 minutes W. in 38 m water depth about 2.7 km east of the HARS, during the winter of 1999-2000 (Butman, Alexander, and others, 2003; see cross references). An estimate of the error due to sea level remaining in the multibeam observations after the sea level correction is about 3 cm.
Using the adjusted tidal elevations, create a binary tide file to be used in merging the tidal heights with the bathymetric soundings.
Command line: binTide -year YYYY asciiTideFile BinaryTideFile
The program mergeTide brings the swath soundings to the MLLW vertical tidal datum:
Command line (tides): mergeTide -tide BinaryTideFile filename.merged
4. Correct near-nadir beams for depth offset caused by error in EEPROM programming (see Logical Consistency Report)
Command line: deHump -depscale 0.0105 filename.merged
5. Create a new 3-meter grid of the edited bathymetric soundings from each line file (filename.merged) using the SwathEd routine weigh_grid.
Command line: weigh_grid -fresh_start -omg -tide -coeffs -mindep -2 -maxdep -800 -beam_mask -beam_weight -custom_weight EM1000_Weights -butter -power 2 -cutoff 6 -lambda 1.5 gridFile filename.merged
6. Convert binary bathymetric grid to ESRI ASCII raster format:
Command line: r4toASCII gridFile.r4
This creates a file called gridFile.asc.