New Feature: Reading netcdf files

[carleyjmartin]

New Feature

Name: read_netcdf

module: in superdarn.py

package: superdarn

Scope

Will read netcdf files produced at APL into dictionary to use with pyDARN.

Code to work from:

"""
Demo read SuperDARN netCDF files available from https://superdarn.jhuapl.edu/download/
nc_utils available from github.com/alexchartier/nc_utils
Alex Chartier, 2023
"""
try:
    import nc_utils
except:
    print('Get nc_utils from github.com/alexchartier/nc_utils')
import sys

def main(fit_fn, grid_fn):
    # Read the whole file using load_nc() (provides access to header info)
    grid_file = nc_utils.load_nc(grid_fn)
    fit_file = nc_utils.load_nc(fit_fn)

    # Read the variables (a little easier to use)
    grid_vars = nc_utils.ncread_vars(grid_fn)
    fit_vars = nc_utils.ncread_vars(fit_fn)

    """ Print the details of the fit file """
    print('\n*****')
    print(fit_fn)
    print(fit_file.description)  # Note the grid_file (and other python objects) can be interrogated on the commandline using dir()
    print('Lat: %s' % fit_file.lat)
    print('Lon: %s' % fit_file.lon)
    print('Boresight: %s\n' % fit_file.boresight)
    for k, v in fit_vars.items():
        print('%s: %i entries, e.g. %1.1f' % (k, len(v), v[0]) )
    print('*****\n')

    """ Print the details of the grid file """
    print('\n*****')
    print(grid_fn)
    print(grid_file.description)  # Note the grid_file (and other python objects) can be interrogated on the commandline using dir()
    print('Lat: %s' % grid_file.lat)
    print('Lon: %s' % grid_file.lon)
    print('Boresight: %s\n' % grid_file.boresight)
    for k, v in grid_vars.items():
        print('%s: %i entries, e.g. %1.1f' % (k, len(v), v[0]) )
    print('*****\n')


if __name__ == '__main__':
    args = sys.argv
    demo_fit_fn = "~/Downloads/19940701.gbr.v3.0.despeckled.nc"
    demo_grid_fn = "~/Downloads/20150317.cly.v3.0.grid.nc"
    assert len(args) == 3, 'Should have 2 args, e.g.:\n python3 read_nc.py %s %s' % (demo_fit_fn, demo_grid_fn)
    fit_fn = sys.argv[1]
    grid_fn = sys.argv[2]
    main(fit_fn, grid_fn)

[carleyjmartin]

Requested again for reading and writing to netcdf. Not stale.

[JWiker]

Requested again for reading and writing to netcdf. Not stale.

@carleyjmartin this one got moved to the "on hold" part of my task list because I heard the PIs were going to decide on a formal netCDF4 product that would be part of a future RST release, and that would replace the current standalone APL netCDF product. Any idea what the status of that is?

[carleyjmartin]

Not a clue, but I can nudge and see if we can find out!

Also sorry, the comment above was for me so I don't close this issue in the future, not a passive aggressive 'what's going on with this' comment to you!

[JWiker]

No worries - I'll be interested to hear if there's been any movement on that front...it seems like more and more people are interested in working with netCDF data.

[aburrell]

A lightweight way to do this might be a to/from xarray feature for the main classes, and then let xarray handle the netCDF production. There are downsides to that (from my experience mainly time handling), but then you have the upside of not having to specifically support netCDF.