spatial_coords_with_time.py

from shapely.geometry import Point

import ocgis


path = '/path/to/data.nc'
lat_variable = 'XLAT'
lon_variable = 'XLONG'
lat_dimension = 'south_north'
lon_dimension = 'west_east'
data_variables = ['T2']


# Load the field from disk and modify it before passing it to operations. Operations can take a request dataset
# a field.
field = ocgis.RequestDataset(path, variable=data_variables).get()
# Remove the time dimension from the spatial coordinate variables.
for cv in [lat_variable, lon_variable]:
    var = field[cv]
    new_dimensions = var.dimensions[1:]
    var.reshape(new_dimensions)
# Update the field's metadata so it knows where the x/y coordinate variables are.
field.set_x(field[lon_variable], dimension=lon_dimension)
field.set_y(field[lat_variable], dimension=lat_dimension)
# Create and set the coordinate system.
lcc = ocgis.CRS(
    proj4='+proj=lcc +a=6370000.0m +b=6370000.0m +lon_0=97.00w +lat_1=33.00n +lat_2=45.00n +lat_0=40.00n +units=m +no_defs',
    name='rpolcc')
field.set_crs(lcc)

# Run operations and write to netCDF.
geom = [{'geom': Point(20, 2), 'crs': lcc}]
ops = ocgis.OcgOperations(dataset=field, geom=geom, output_format='nc')
ret = ops.execute()