Please use github.com/ucyo/xsuite

This module helps create chains of cdo commands for easy manipulation of climate data.

• Method execution is lazy and gets processed only on function call Chain.execute().
• Input supports Unix style pathname pattern search.
  • The Input will be first run with glob and checked if several files match.
    ❗ If that is the case a temporary file will be created.
• Output can be a file on disc, an netCDF4.Dataset or (not) masked numpy.ndarray.

python3.5 -m pip install cdochain --pre

--- or ---

git clone https://github.com/OnionNinja/cdochain.git
cd cdochain
python3.5 setup.py install

from cdochain import chaining as cch

input = './tests/testdata/sresa1b_ncar_ccsm3-example.nc'
output = './enso34-mm.nc'
data = cch.Chain(ifile=input, ofile=output)
enso34 = data.sellonlatbox(190,240,-5,5).monmean()
out = enso34.execute()

This module implements method chaining for the Climate Data Operators (CDO) tool from the Max Planck Institute for Meteorology. Let us start:

from cdochain import chaining as cch

For initialisation one has to define input, output, and may define several options.

Now we have to define the files we want to work on:

• To use one file

input = './tests/testdata/sresa1b_ncar_ccsm3-example.nc'

• To use several files you can give a Unix style pattern

input = './tests/testdata/*.nc'

This creates a temporary file ❗

For defining the output we have several options.

• To output a file on disc:

data = cch.Chain(ifile=input, ofile='/path/to/output.nc')

• To output an netcdf4.Dataset object:

data = cch.Chain(ifile=input, ofile='netCDF4')

• To output an numpy.ndarray object:

data = cch.Chain(ifile=input, ofile='array:<var>')  # numpy.ndarray
# or
data = cch.Chain(ifile=input, ofile='maarray:<var>')  # masked numpy.ndarray

<var> defines the variable to be extracted and saved to numpy.ndarray.

As for options one can use the same as described on the CDO website. The default is options='-O -f nc'.

The operations defined in CDO can now be used on the data element.

analysis = data.sellonlatbox(190,240,-5,5).sellevidx(1).mermean()
fn = analysis.execute()

Have fun