loading VMDAS data (BB and OS) into CODAS
We now have programs to load BB and OS data collected by VMDAS into
codas. This is an expansion of our first attempt to get all this code
in one place for other people to use: the
documentation and
editing tools have improved, but we are still assuming you
already know how to load pingdata into CODAS and process it. Please
be aware that this code is still under development; at the moment,
loading LTA files appears to work, but the software is not yet
ready for loading ENS files.
The zip archive "vmdas_demos.zip" contains three "demos"
(mini-tutorials) for VMDAS data. The "os_ltademo" outlines how to
load OS LTA files into codas. (There is a parallel directory for
BB data, "bb_ltademo"). The LTA demo also works for STA files, as
they are written in the same format by VMDAS. The third directory
(os_ensdemo, for ENS data) is not ready for use.
These demos are set up using
quick_adcp.prl and utilze "refsm.m", the new variable-depth reference
layer scheme. You can still use "smoothr": it is more robust, but does not handle changes in water depth well. If you are feeling conservative or run into problems with refsm, try using smoothr (and pleas let me know what problems you had!).
General notes and caveats:
- Although there is no mention of headings or rotation in the demo, we
still recommend you use gyro heading as your standard heading and correct
the headings in post-processing using an gps-derived heading such as the
Ashtech 3DF heading.
- These loading programs are not tested in a PC environment, so if you
are doing this on Microsoft machine, you may encounter some bugs. Let
us know what you find and we'll try to fix it.
- Refsm blindly interpolates across missing fixes. A log file
(nav/refsm_interp.log) keeps track of
time differences between interpolated fixes. If you have 300-second averages
and see more than a 2-3 consecutive entries with 298-302 seconds in the last
column, these are consecutive fixes that were interpolated. It is advisable to
use smoothr (rather than refsm) under these circumstances.
ENS (single-ping) data and CODAS
- The "ens_demo" outlines how to load (i.e average and load) the
single-ping ENS files into codas. At the moment (June 2002) there are
some problems getting the data into CODAS, but that should be fixed "soon".
(stay tuned)
- Editing is not particularly sophisticated, though it should be adequate.
Improving editing is a priority. Be aware that loading ENS data can
become a very time-consuming process.
To try these, you'll want to get or update vmdas_demos.zip, cmat.zip, and
the appropriate binary files (binsol.zip, binlnx.zip for solaris or linux,
bindj.zip or binmgw.zip for PC). For the matlab directory in particular,
you should replace your current version with the new version because of
reorganization of directories: if you simply "upgrade", you may end up using
the wrong files. A new version of adcptree written in perl (adcptree.prl)
is provided in the same directory as adcptree, and will start a new
cruise directory for you with the appropriate
new files copied into it
That's all for now. If you find something that doesn't work, or
doesn't work as advertised, please let me know . I've tried to sprinkle
READMEs around in appropriate places.