CODAS FAQ

---

How do I access ADCP data (CODAS, VmDAS LTA or ENX) using Matlab?

Check this link

How can I tell if the Ashtech messages are good?

Check this link

What is CODAS processing (what does it do?)

• See the overview
• In addition to the present documentation, there is a somewhat outdated but very detailed manual in postscript form.

What software do I need to get started?

You will need several different programs, as well as software from the UH ADCP group.

• Download and setup instructions for programs from this web site are detailed here
• Links to Python, Infozip, and Ghostscript are here. You might want to get m_map, but it is optional)
• You will also need matlab, a commercial program.

My computer is all set up, but I am new to ADCP data processing. Where should I start?

We recommend you start by learning about CODAS processing with the python script “quick_adcp.py”, run on “pingdata” (averaged data from an RDI NB150) or LTA data (averaged data acquired by VmDAS). You must be on line or have qdemo_lta.zip installed.

You should start here

Specifically

• Under “Quick_adcp.py Instructions and guides” choose the “practice demos” link and look for the column with LTA example.

• Get the data, read the links, process the data:

  • Look at the example directory and notes for more information.

How hard will this be? Will it take long to learn?

It depends on your background, but the short answer is: ‘yes, it will take some tine to learn, but less time than it would take to write your own code’.

Longer answer:

That depends on many factors, including

• how well you know your computer (the commandline interface, how to configure the paths and environment variables)

• how well you know matlab (to explore the data directly)

• your familiarity with the instrument (what is measured, what is stored in the data)

• your familiarity with the processing steps (what should be done, how to recognize problems)

• how good your dataset is (does it have clock problems? navigation problems?)

• the manner in which the heading correction (to the gyro) will be calculated (pingdata with ue4 is easy, LTA is hard)

  For a unix-savvy matlab-aware individual working with pingdata, it could be as little as a few days to gain some experience with the processing.

  For person on a Windows machine who has never dealt with the C:\ command line, who does not know matlab, and who as never seen ADCP data before, it would be longer, potentially weeks not days.

I am familiar with CODAS processing. How do I get started on quick_adcp.py?

For someone already familiar with CODAS processing steps, you can

• look at the overview in this file,
• look at a more detailed quick_adcp.py processing tutorial
• explore command-line help in quick_adcp.py itself (“quick_adcp.py - -help”).

Can I get this documentation for off-line use?

Yes, the zipfile “doc.zip” from this location contains all the UH Currents group CODAS and UHDAS documentation. Follow these instructions to ensure that CODAS setup and processing documentation is accessible on your computer.

If you are about to leave, and you want to be able to run CODAS processing or look at data while off line, you might consider grabbing all the zip files listed, installing Python, and definitely making sure you have Matlab installed.

What are my options for plots of the ocean currents during a cruise?

That depends on the instrument and data acquisition system used on your cruise. If you have UHDAS, this is what you can expect. Otherwise, you probably have VmDAS, and can either process the data or get a quick look using matlab read.m .

I have a DVD full of UHDAS data from a cruise. What is on it? How can I look at the data?

This link gives an introduction to the data in a UHDAS cruise directory and points to useful tools for reading the data

Please read about CODAS conventions (eg. decimal day is zero-based)

Note

The database has undergone preliminar processing, but steps remain to ensure there are no calibration errors, gaps, or bad profiles.

I have a DVD full of UHDAS data from a cruise. How do I finish processing the data?

Automated CODAS processing has already been run. You now need to deal with the last parts of the processing manually.

If you have never processed a dataset with CODAS before, and if you want confidence that your final product is good, you should start by reading about quick_adcp.py and running through the LTA demo. The best place to start is the quick_adcp.py overview.

Once you have gone through the LTA demo and have some confidence with the tools, take a look at these documents...

• CODAS conventions
• cruise directory contents
• overview of UHDAS + CODAS at sea, specifically the Processing Component

... and then, to reprocess or finish processing the data from a cruise, look at the UHDAS Data Processing Demo and follow the instructions closest to your case.

Note

The UHDAS demo is contained in qdemo_uhdas.zip . You should install this on your computer, along with all the rest of the appropriate zip files. NEVER ALTER ANYTHING in the programs directory, but you can copy files and directories from q_demos/uhdas to practice the demo.

I have a dataset from a Workhorse already processed but I learned I may be able to see more bins at the bottom. How do I do that?

Yes, there has been a bug in the code wherein the blanking over the bottom due to side-lobe reflection was hardwired to expect beams with 30degrees. This has been updated in the Mercurial repositories as of May 1, 2009. If your code is older than that, you must update your CODAS installation. As of this writing, the new executables have not all been compiled yet. This is an excellent opportunity to install CODAS software using Mercurial repositories. See this link for instructions.

An ADCP with 30deg beams can only use 85% of the water column when the bottom is in range, but with 20deg beams the useful data range is more like 95% of the total.

It is not hard to reset the bottom-masking flags. Explanations and prescriptions are here