Skip to content

Latest commit

 

History

History
64 lines (49 loc) · 2.54 KB

README.md

File metadata and controls

64 lines (49 loc) · 2.54 KB
Raw

pyglider

Python tools for interacting with ocean glider data

Installation:

  • install conda/miniconda
  • clone the repository onto your computer and cd into the directory
  • from inside the cloned repository directory run
    conda env create -f environment.yml, it will read environment.yml and automatically install the dependencies, and create environment pyglider, (though you can change the name by editing environment.yml)

Editable installation

If you want to be able to edit the files in pyglider/pyglider then install as above, and then do pip install -e .. That will re-install pyglider with links to the local directory, so you can edit the library files.
If you do so consider making a pull-request with your changes!

Getting Started

The work flow (for now) for a new deployment data set from a Slocum glider is

  1. offload the data from the glider computer or get from the dockserver
  2. copy data from offload/Science/SENTLOGS/*.EBD, offload/Science/LOGS/*.EBD, offload/Main_board/LOGS/*.DBD, to ./binary/ of your new deployment presumably in a different directory than example
  3. copy the data from the caches to yourdeployment/cac/; these are found in offload/Science/STATE/CACHE/*.CAC and offload/Main_board/STATE/CACHE/*.CAC.
  4. Edit rosie_713_sensors.txt (and probably name after your own glider) for the data streams from the raw files you want. The name of all the variables are in the *.CAC files.
  5. Edit deployment.yaml with appropriate metadata for your glider and deployment. There are a lot of fields, but most should not change after the first deployment unless the sensor payload changes on the glider.

Why?

There are other glider workflows, but the only ones I know of are in Matlab, and there is no reason for that.

The only disadvantage of this code is that the dcoding step from binary to the raw netcdf is quite slow for the Slocum binary data. That can/should be improved by using a C library instead of python, but really, its not that slow.

Resources