Python tools for processing output data from the Parallel Ocean Program (POP)

Strong as Popeye the Sailor Man

git clone https://github.com/j08lue/poppy.git
cd poppy
python setup.py install

The poppy package contains modules for various data postprocessing tasks. Of particular interest might be the metrics package, that contains routines for extracting large-scale ocean metrics such as the maximum AMOC strength or meridional heat transport from an arbitrary number of files. The metrics are stored as Pandas dataframes in HDF5 (if Pandas is available) or Pickled NumPy arrays and can be easily plotted together.

The scripts directory contains mainly command-line interfaces for the different metrics functions, e.g. to plot the AMOC strength evolution directly from the model output files:

cd <CASENAME>/ocn/hist
plot_amoc_timeseries.py *.pop.h.????-??.nc

The nclook.py script offers a quicklook into netCDF files. If you create an alias like

alias nclookipy='ipython -i /path/to/nclook.py'

you can call e.g.

nclookipy *.pop.h.0100-??.nc

and have the files opened and ready to explore as a multi-file xray or netCDF4 Dataset.

The tests only cover a small part of the module, but their passing indicates functionality of the most important bits. Try

git clone https://github.com/j08lue/poppy.git
cd poppy
python -m unittest discover tests

Please see the setup.py for dependencies. Written in python2 but works well with 2to3. One method to make this work is to rename the poppy folder to e.g. poppy_py2, and then calling

2to3 poppy_py2 -o poppy -W -n

It should then be possible to install the package with python setup.py install.