UHDAS Heading and Averaging

Setting Heading Sources

The heading correction comes from the device specified. This heading device is specified in two places. It is best if they are the same, to avoid confusion. The two locations are:

• in the file cruise_cfg.m (probably named something like

  km0601_cfg.m, probably located in a subdirectory ‘config’ of the actual processing directory. The line looks like this:

hcorr_inst =  'ashtech';   %(gyro, posmv, ashtech, seapath)

• in the control file for quick_adcp.py, where the line looks like:

--hcorr_inst  ashtech # choose : gyro, ashtech, posmv, seapath

• NOTES

  • The value in cruise_cfg.m is the default; the value in quick_adcp.py’s control file overrides it. It’s easiest to make them the same.
  • specifying ‘gyro’ disables the heading correction, i.e. the heading used is simply that of the gyro

Apply Heading Correction

There are two mechanisms of applying the heading correction. Those are specified in the quick_adcp.py control file by using one of the two following

-- ping_headcorr
-- post_headcorr

If neither is chosen, or if the heading correction instrument is “gyro” then no heading correction is performed.

ping_headcorr

This is the mechanism used during real-time at-sea acquisition and processing. It is not recommended for batch processing. This mechanism calculates a heading correction for each independent 5-minute ensemble. The value of the correction is stored in the database as a watertrack and bottomtrack misalignment, and the velocities are rotated before being stored. Files written in ‘cal/rotate’ with prefix ‘ens_hcorr’* indicate values stored IN the ensembles. Quick_adcp.py “rotate” stage does not apply a time-dependent heading correction. The files are left on the disk as a record of what was applied IN the ensembles

post_headcorr

This is the mechanism recommended for use in batch processing. Files are generated in the ‘cal/rotate’ subdirectory that show the heading correction to be applied. These are not applied until the “rotate” step, allowing the user to reprocess the dataset and clean the angle file prior to use (eg. interpolate across gaps). These files all start with “hcorr”.

Details of files in ‘cal/rotate’:

In both cases (ping_headcorr and post_headcorr)

• The rotate.log file will show a post-processing correction using the file only if that method used it.

• the heading correction is in the ancillary_2 data structure

  of the database

• the database should act the same with either method: * the gyro (primary) heading stored in ancillary_1 ‘heading’ * heading correction applied is stored in ancillary_2 * measured velocities have been rotated by the value in ancillary_2

• using “unrotate” with “rotate rotate.cnt” will revert to

  zero heading correction

• lst_hdg now includes a column with the ancillary_2 heading correction

• files are created in cal/rotate as a record

specific to –ping_headcorr

file                   purpose         details
------                 -------         --------
ens_hcorr.ang          APPLIED         heading corrections present
                                       IN the ensembles
ens_hcorr.asc          diagnostics     heading corrections and their quality
ens_hcorr_err.txt      diagnostics     comment for each 5-minute average
hcorr_stats.txt        diagnostics     running statistics from ens_hcorr.asc
print_hcorrstats_tmp.m matlab program  (makes hcorr_stats.txt)
hcorr_plot.png         diagnostics     at sea: running plot of ens_hcorr.asc
                                      heading correction and quality

NOTE: see this link for important details about ens_hcorr.txt

specific to –post_headcorr

file                   purpose         details
------                 -------         --------
ens_hcorr.ang          diagnostics     heading corrections calculated
                                       on a per-ensemble basis
ens_hcorr.asc          diagnostics     heading corrections and their quality
ens_hcorr_err.txt      diagnostics     comment for each 5-minute average
hcorr.asc              intermediate    1-minute heading corrections with QC
hcorr.ang              APPLIED         derived from hcorr.asc
hcorr_stats.txt        diagnostics     statistics of whole dataset
print_hcorrstats_tmp.m matlab program  (makes hcorr_stats.txt)
plot_hcorrstats_tmp.m  matlab program  (makes hcorr.ps)